Plight of the Power Sector in India Inefficiency, Reform and Political Economy 818635302X

Citation preview

PLIGHT OF THE POWER SECTOR IN INDIA Inefficiency, Reform and Political Economy

K. P. KANNAN N. VIJAYAMOHANAN PILLAI

Centre for Development Studies Ulloor Trivandrum - 695 011 Kerala India

September 2001

CONTENTS Acknowledgements Preface Part I National Scenario: Chapter 1 Introduction Chapter 2 SEBs and Their Saga of Inefficiency Physical Performance Chapter 3 SEBs and Their Saga of Inefficiency Financial Performance Part II Power Sector in Kerala - A Diagnostic Analysis

Chapter 4 Electricity Demand Analysis and Forecasting Chapter 5 Time and Cost Overruns of the Power Projects in Kerala Chapter 6 T and D Loss in Kerala Power System Chapter 7 Pricing of Electricity - Theory and Practice and Some Exercises for the Kerala System Part III The Surge of Reforms

Chapter 8 Experiences from Abroad Chapter 9 Reforms in the Power Sector in India

Part IV The Political Economy

Chapter 10 The Political Economy of Public Utilities A Study of the Indian Power Sector Chapter 11 Conclusion

Glossary Bibliography Index

2

IN LIEU OF A PROLOGUE With the degeneration and collapse of the once-powerful socialist alternative, most of the contemporary states in the world appear to be engaged in resurrecting a sort of free market capitalism. This signifies a disastrous return to social Darwinism, devoid of the social-ethical commitments of building up a society embedded in collective care. The famous Thatcherite TINA (There Is No Alternative) proclamation sums up the situation. Gone with the Second World is the uncompromised balance of the Third World countries in terms of their bargaining power for development assistance. With the help of the IMF and World Bank, development finance is now injected for a fresh flourishing of the market in distinct social sectors of the poor countries including education, health, power, transport, water supply, and so on, that were once in the public domain on account of their vital significance for social welfare. Their supply involved substantial subsidies to enable the majority of the population to access these basic amenities in a raw life. Liberalism that exacts a price on every transaction at its market value thus stands to exclude the majority from accessing these amenities under its reign. This not only reinforces the existing processes of social exclusion, but also generates new ones. And this is the dangerous implication of the new era for countries like India.

An unfortunate historical coincidence that strengthened the TINA premise came from the infamous dysfunctioning and the decay of the public sector. This in terms of operational inadequacy and financial infirmity had its origin in the x-inefficiency that permeated the whole pyramid of the functioning of the public sector. This in turn was situated in the political economy of uncompensating populism and unmitigated corruption. A prime example of this comes from the public utilities in India known as State Electricity Boards under the control of the State governments. The present study is an attempt to address and analyse these problems. We must clarify at this juncture that our study covers only the State power sector, and not the Central power sector under the control of the Central government. We are aware that the problems of the latter are of a

3

different nature and not entirely related to inefficiency. In fact the Central power sector in general has a record of high level performance efficiency in stark contrast to the State sector.

A work of this dimension would not have been possible but for the contributions from a collective body of the informed social consciousness in one way or another.

Thus we are indebted to a number of persons for their support, comments and criticism in completing this study. First and foremost stands Professor K. N. Raj, commanding our respectful thanks for his initial suggestion and encouragement to undertake this study. His support has been a source of inspiration in keeping our spirits high and overcoming difficult phases in the completion of the study. Amulya Kumar Reddy, an eminent energy expert, has also been instrumental in encouraging us to proceed with the study.

As the study progressed, a number of seminars were given at the Centre on various issues relating to the power sector. In addition, K. P. Kannan was invited by the Institute of Environmental Studies of the Free University of Amsterdam to present the political economy issues in a workshop held in The Hague (The Netherlands) in May 2001. We also distributed drafts of a number of chapters to some of our colleagues as well as professional experts. Discussions were also held with planners, administrators and scholars. The comments we received from these persons have contributed immensely in revising the drafts. We would therefore like to make a special mention of the following persons:

M. N. V. Nair, former Dean of IIM, Bangalore; J. Vasudevan, Additional Secretary, Ministry of Power, Government of India; T. L. Sankar, former Principal, Administrative Staff College of India, Hyderabad; V. Ramachandran, Vice Chairman, Kerala State Planning Board; U. Sankar, Madras School of Economics, Chennai; K. S. Menon, Management Consultant; N. J. Kurien and Sonkar, Planning Commission, New Delhi; Rakesh Mohan, former Director General, NCAER, New Delhi; S. B. K. Menon,

4

former Chief Engineer, Kerala State Electricity Board; R. V. G. Menon, Director, Integrated Rural Technology Centre, Palakkad; C. H. Hanumantha Rao and S. Mahendra Dev, Centre for Economic and Social Studies, Hyderabad.

At the CDS: K. K. Subrahmanian; Chandan Mukherjee; K. Pushpangadan; Indrani Chakraborty; Sunil Mani; Achin Chakraborty; D. Narayana; K. Ravi Raman and V. Santhakumar.

Arising out of the work on this study, we have published (besides a number of working papers) a couple of articles (Economic and Political Weekly, January 13 and 20, 2001; and The Hindu daily, January 11, 2001). Some more are expected to be published in the near future. It should also be noted that the first author remained in charge of the project in general while the study was solely conceived, structured and carried out and the Report was written by Vijayamohanan Pillai.

The financial support for the study was provided by the Centre. We would like to record our thanks to the Director Chandan Mukherjee, Chief Administrative Officer Soman Nair, Finance Officer V. Ramesh Kumar, Librarian V. Ramakrishnan and his colleagues, and J. Muraleedharan Nair and his colleagues at the Computer Centre for their support. A special mention must be made of the efficient administrative support provided to us by the Senior Assistant Administrative Officer K. Muraleedharan, D. Girija and M. Krishnan Kutty throughout the period of the study, and also by the Publication Officer Tilak Baker. We would also like to thank our families, especially, Shobha and Asalatha, for their cooperation and understanding during the course of this study. And Vijayamohanan Pillai would like to record his special thanks to Rju for smiling away his excuses for his absences from her little kingdom.

K. P. Kannan N. Vijayamohanan Pillai Thiruvananthapuram, 15 May 2002.

5

“Turning and turning in the widening gyre The falcon cannot hear the falconer; Things fall apart; the centre cannot hold; Mere anarchy is loosed upon the world, The blood-dimmed tide is loosed, and everywhere The ceremony of innocence is drowned; The best lack all conviction, while the worst Are full of passionate intensity.

Surely some revelation is at hand; Surely the Second Coming is at hand.”

- W. B. Yeats (‘The Second Coming’)

The world appears to be engaged in an experiment in reinventing history in terms of a resurrection of Mercantilism in its most dogmatic practice of Individualism – a disastrous return to Darwinism, devoid of the social-ethical commitments of Collectivism. It is significant to recognise the favourable historical setting that tuned in the teleology of this return: the unfortunate degeneration and demise of the oncepowerful socialist alternative. The famous Thatcherite TINA (There Is No Alternative) proclamation sums up the situation. Gone with the Second World is the uncompromised balance of the Third World countries in terms of their bargaining power for development finance. Milking of two cows at will is now a proverb only. The only cow that remains now has thus got an undeniable monopoly of dictating, in return for its service, conditionalities meant to revive the erstwhile moribund liberalism in the void of an alternative. In this light must we examine the IMF-World Bank reinventing the wheel of Mercantilism. Development finance is now injected for a fresh flourishing of the market in distinct sectors of the poor countries such as education, health, power, transport, water

6

supply, and so on, that were once in the domain of the public enterprise on account of their vital significance for social welfare. Their supply involved substantial subsidies to enable the majority of the population to access these basic amenities in a raw life. Liberalism that exacts a price on every transaction at its market value thus stands to exclude the majority from accessing these amenities under its reign. And this is the dangerous implication of the new era for India too.

An unfortunate historical coincidence that strengthened the TINA premise came from the infamous dysfunctioning and the decay of the public sector. This in terms of operational inadequacy and financial infirmity had its origin in the x-inefficiency that permeated the whole pyramid of the functioning and in the political economy of uncompensating populism and unmitigated corruption. It goes without saying that economic incentives in conjunction with contractual coercion would take care of xinefficiency problem that in general crops up in the vacuum of a sense of oneness and ownness. On the other hand, measures that make up populism, despite its derogatory connotation, as it is aimed at enlarging the popular support base of the political leadership of the government at a social cost, have an undisputable content of social welfare and thus stand above any rational censure. These measures have a desirable practical value provided they involve a corresponding compensatory mechanism. The free/subsidised power supply to targeted sections of consumers or rural electrification is a case in point here. If the government could financially compensate the Electricity Board (EB) in full and in time for the loss involved in such subsidisation, imposed on the Board as a part of the socio-economic policies (read pork barrel politics) of the government, the system could remain healthy. The political economy of corruption, the scope of which is much larger under the new liberalism, requires, on the other hand, an ever-watchful civic platform of checks and balances as a counter-force of public praxis against the rentseeking pursuits of the political leadership at social cost.

The whole upshot of our argument is thus to suggest that the unfortunate sage of dysfunctionings and the subsequent decay of the public sector are just internal to the respective system, and hence there do remain sufficient quarters for remedial exercises,

7

meant to remove the problems that stand in the way of its improved performance, rather than subjecting it to a radical surgery of restructuring involving deleterious consequences. This in turn emphasises the original role of the State in dispensing welfare concerns and development commitments, especially in a poor country like ours. Unfortunately, however, the informed circle remains indifferent to the grave consequences of the new initiations of marketisation. This is all the more significant in the context of reforms in the power sector of India. And it is this background that has provoked our look into the plight of the Indian power sector in terms of the accumulated inefficiency and the consequent initiatives of reform process as well as the political economy of that plight. We do believe that our ambitious attempt at such a comprehensive diagnosis would sow some sparks in the otherwise frigid informed circle.

8

PART I

NATIONAL SCENARIO True to the spirit of a social-democratic State, India had originally evolved her power development policy, and shouldered that responsibility, in line with the State’s professed commitment to honouring and ensuring social security equations. Though the State Electricity Boards (SEBs) were statutorily required to function as autonomous service-cum-commercial corporations, they became in effect agents of the Governments to subserve the socio-economic policies of the State, and hence never felt the requirement to break even or to contribute to capacity expansion programs. This unaccountability culture in turn led to gross inefficiency at all levels – technical, institutional and organizational, as well as financial. And the cost escalation from such pampered inefficiency remained above the revenue realized from an irrational subsidized pricing practice. With losses mounting up, the field was getting cleared for some new entrants of ideas and practices, that the so-called ‘fiscal crisis’ at the turn of the nineties ushered in subsequently. Thus has commenced an era of reforms and restructuring of power sector in India, at the initiation of the World bank that has also lit up an informed atmosphere of debates and discourses. However, little light has been thrown on the significant aspects of inefficiency costs involved in the SEBs’ forced functioning that allegedly finally warranted the reforms. This part of the study is a modest attempt at this.

Here, inter

alia, we have estimated, on some very plausible assumptions, the avoidable cost of inefficiency at a few amenable levels and found it to represent about one-third of the reported cost of electricity supply in India in 1997-98 ! And this is regardless of a number of other possible inefficiency sources at all levels of performance.

This part is divided into three chapters, beginning with an introductory one. Chapter 2 deals with physical performance focusing on such aspects as technical inefficiency, T & D losses and its possible underestimation as well as some aspects of institutional and organisational inefficiency. Chapter 3 deals with the supply cost of electricity, tariff and revenue as well as financial performance.

9

CHAPTER 1

INTRODUCTION

“Agne naya supatha raye asman…” -

Isavasya Upanisad1 (18)

1. The Energy Scenario

The significance of energy as the cosmic elan vital has manifested itself through the pride of place conferred on fire in both antique mythology and philosophy. Rightly, the discovery of fire brought the human race on to the path toward civilisation. It was the control of fire that facilitated the development of the hominids from the Old Stone Age hunters of the tropical forests into the first village-dwelling farmers of the Neolithic about 7000 BC. And fire has had an essential role at every stage in the subsequent growth of civilisation, made possible by the development of technology and science, through a continual increase in the amount of energy available in terms of fire brought under human control, the latest being the nuclear one, the very source of cosmic energy.

The progression of energy sources have been from human and animal power of primitive society to fossil fuels and radioactive elements, and water, wind and solar power of the industrial society.2 As widespread industrialisation has picked up,

10

consumption of fuels has also followed in tandem. The consumption of commercial energy3 in developing countries has been rising rapidly, especially recently, and is projected to dominate energy markets world-wide.

Energy consumption in the low

income economies during 1965-1980 grew at an average annual rate of 8.2 per cent and during 1980-1990 at 5.5 per cent, against the world average rate of 4.1 and 2.5 per cent respectively and the high income countries’ average rate of 3.1 and 1.4 per cent respectively (World Bank 1992: Table 5). Although commercial energy consumption in developing countries tripled between 1970 and 1990, despite oil price shocks and financial crises, there is still a disproportionate share of the world’s energy consumption in favour of the developed economies: the US (23 per cent), Japan (5 per cent), and other high income countries (22 per cent), with only 15 per cent of the world’s population, consumed half of the world’s commercial energy in 1997 (World Bank 2000 a: Table 3.7). On a per capita basis, these wealthy countries used more than eight times as much commercial energy as did the low income countries (see Table 1). It should be noted that in most of the developing countries, consumption of traditional fuels (firewood, charcoal, bagasse, and animal and vegetable wastes) still commands a considerable proportion in total energy consumption.

More than half the world’s coal consumption and 30 per cent of fossil fuel consumption go to generate electricity4

(World Bank 1992: 116). A number of

characteristics make electricity a very attractive form of energy as compared with others – for example, its practically instantaneous transportation, cleanliness at the user’s end, ease of control, and economy. And per capita electricity consumption of a country is usually taken as one of the indicators of the standard of living of its population. During 1980-1997, electric power generation world-wide rose by 170 per cent (at an average annual growth rate of 4.2 per cent); it increased by 155 per cent in the high income countries (at 2.9 per cent per annum) and by more than 330 per cent in the low income countries (at 8.4 per cent annually) (Table 1). Despite the impressive increase in electricity generation in the developing economies, their people still consume much less of it; in 1997, per capita electricity consumption in the low income countries was just

11

about 5 per cent of that of the high income countries, and about 20 per cent of the world average (Table 1).

2. The Electric Power System

As in the case of many other products, three distinct functions are involved in the supply of electricity in its usable form to the consumers: generation (i.e., production), transmission (i.e., transportation to markets), and distribution (i.e., retail sales), though with two basic differences – i) electricity cannot be stored in its original form; and hence ii) it should be generated the moment it is demanded, which is made possible by its unique ability to be transported instantaneously, as electricity moves at a speed approaching that of light. The instantaneous co-ordination of these three functions in turn necessitates a vertically integrated organisation of the electricity supply industry.

Electricity in large quantities required to supply a power system is produced in generating stations (power plants). The generating units in stationary plants convert energy from falling water (hydroelectric plants), coal, natural gas, oil and nuclear fuels (thermal plants) into electric energy. In a hydro-power plant, potential energy of water stored in a dam at a height is converted into mechanical energy by use of a water turbine and a generator connected to the water turbine converts it into electrical energy. In a conventional thermal station, heat energy in the fuel (coal, gas or oil) is liberated by combustion in the boiler furnace and is used to convert water into steam. The steam energy is then converted into mechanical energy of a rotating shaft by some form of steam engine, either reciprocating (as in a diesel engine) or turbine,5 and this energy is converted by a generator into electric energy.6 In a nuclear plant, energy from either fission or fusion of nuclei of atomic fuel is released in the form of heat in the reactor, which converts water into steam; and the steam energy finally generates electric energy as in any conventional thermal plant. A three-phase alternating current (AC)7 system is almost universally adopted. The standard frequency in Britain as well as in India is 50 cycles per second; in the US, 60

12

cycles per second. The AC system has an advantage that voltages can easily be stepped up or down by the use of transformers only; and this facilitates transmission and distribution of power.8

The high voltage transmission line is the second important component in electricity supply systems. Both the invariable location-specificity of the hydro-power plants and the economies in developing pit-head-based thermal stations necessitates extensive network of transmission lines9 for conveying large amounts of electricity from generation sources to bulk sub-stations. There are also economies as well as gains of mutual assistance in case of emergency in inter-connecting different power stations and systems in an electric grid or power pool with centralised control of the operation that also necessitates transmission networks. Electricity is transmitted at a voltage much higher than the distribution voltage, using step-up transformers, in order to reduce energy losses in transit. This power is received at the bulk sub-stations, where transformers step down the high voltage to medium voltage to supply the feeders of the primary distribution network, that carry power to various distribution sub-stations.

At the distribution sub-stations, transformers further step down the voltage of the primary feeders to low voltage for the distribution network, that supply power to customers. The nominal transmission lines in vogue in India are of 400 kilo Volt (kV), 230/220 kV, 110 kV and 66 kV. A relatively new approach to high voltage long distance transmission is high voltage direct current (HVDC),10 which offers the advantages of less costly lines, lower transit losses, and insensitivity to many system problems that restrict AC systems. Its disadvantage is the need for costly equipment for converting the receiving end DC into AC for distribution to end-use point. The distribution network is normally three-phase four-wire and the standard voltage in India is 400/230 Volts.

3. The Energy Sector in India

Energy consumption in India is growing by leaps and bounds. The commercial energy production and consumption in India almost doubled during 1980 to 1997, at

13

average annual compound growth rate of nearly 4 per cent (World Bank 2000 a: Table 3.7). On a per capita basis, however, India lags much behind many of her peers, and even behind the average of low income countries (see Table 1). In 1997, the commercial energy consumption per capita in India11 was nearly one-half of that of China, one-third of the World average, and less than one-tenth of that of the high-income countries. A marked feature of energy consumption in India is the fast decreasing fraction of traditional fuel use in total energy use, which is still much higher than the World average (i.e., about 21 per cent against 7 per cent).

Domestic coal has emerged as the mainstay of the primary commercial energy available in India, accounting for over half of it (51 per cent in 1999-2000), due mainly to its growing use in the electric power and industry sectors (CMIE 2001: 1-2). Growing transport demand and higher share of road transport have led to increasing consumption of oil products, the petroleum sector now accounting for about 35 per cent of the energy consumption in India, nearly two-thirds of which are imported, as India’s hydrocarbon reserves are limited. Impressive growth has been registered by natural gas consumption in India, though its share is still very low (7.5 per cent in 1999-2000 against only about 1 per cent in the early 1980s). The share of hydropower in the total availability of commercial energy on the other hand has suffered a decline from about 4 per cent in the early 1980s to 2.2 per cent in 1999-2000 (ibid.), much (about three-fourths) of its potential estimated at 84044 MW still remaining untapped.

The total commercial energy availability in India increased from nearly 100 million tonnes of oil equivalent (mtoe) in 1980-81 to 306 mtoe in 1999-2000 at an average annual compound growth rate of about 6 per cent. The share of import in total availability was almost around 25 per cent. Net availability for consumption by different sectors however suffered from higher conversion losses, growing at an average rate of 8 percent per year. The ratio of losses to gross availability was always on the rise, from 25.8 per cent in the early 1980s to 33 per cent in 1997-98, indicating higher inefficiencies at conversion centres (power plants, refineries, etc.) and higher losses in distribution (CMIE 2000: 2).

14

Industry and transport sectors account for the major share (48 and 24 per cent respectively in 1997-98) in commercial energy consumption in India, while the residential and agricultural sectors consume about 10 and 5 per cent respectively. The very low residential share in turn points to the poignant fact that the majority of India’s population still remain deprived of the benefits of the standards of a modern life. For example, around 75 per cent of the rural populace in India in the early 1990s used firewood and chips for cooking and kerosene for lighting (Government of India 1992 a). Even in urban India, firewood and chips accounted for the highest share (37.4 per cent) of the households in the primary energy sources for cooking, and 27 per cent of the households used kerosene for lighting.

4. Indian Power Sector

As per 1991 Census, only 42 per cent of the Indian households had electricity facility, with about 71 per cent in the rural and 24 per cent in the urban areas remaining unelectrified. This in turn is to reflect in a very low per capita electricity consumption, one of the leading indicators of the living standard of a people. In 1997, it was just 363 kWh against 714 kWh of China, 1634 kWh of Argentina, 2011 kWh of Chile, 2352 kWh of Malaysia and 1459 kWh of Mexico (Table 1), and was nearly one-tenth of the World average.

Power development, placed in the concurrent list of the Indian Constitution, is a joint responsibility of the regional states and the Centre. There was pervasive enthusiasm for power development in the states during the 1960s (generating capacity growing at an annual growth rate of over 12 per cent per annum) and the 1970s (with an annual growth rate of nearly 10 per cent). During the next decade, this regional zeal, however, further evaporated to drive the growth rate down to 6.6 per cent per annum, and commensurately there was a significant shift in the weight of capacity addition from the states to the Central sector, the latter registering an annual growth rate of about 18 per cent in the 1980s. This in fact helped the whole system to keep somewhat the tempo of power

15

development, at an annual growth rate of 8.6 per cent, despite the deceleration in the states. However, a further shift in the policy of capacity addition towards the private sector in line with the liberalisation drives of the 1990s has left this whole decade ‘wasted’, with very little investment actually coming on stream in all the three sectors, Central, state and private. Consequently, during 1990-91 to 1999-2000, the total power generating capacity increased (from 66086 MW to 97846 MW) at an annual rate of just 4.5 per cent. It should be added that only 53.8 per cent of the Eighth Plan target of 30538 MW could be achieved – the target achievement of the three sectors being Central: 65 per cent, state: 46 per cent and private: 45 per cent.

On the whole, the growth of the system from a mere 1.7 thousand MW of installed capacity (IC) in 1950 to 100136 MW by December 2000 (nearly 60-fold increase, at an annual growth rate of 8.5 per cent) appears quite impressive. So does the growth of energy generated also – nearly 95-times increase from 5.1 billion kWh in 1950-51 to 480.7 billion kWh in 1999-2000 (at an annual rate of about 10 per cent). Despite this seemingly tremendous growth, the per capita consumption of electricity in India still remains one of the lowest in the world, as we have already seen. The capacity deficiency of the Indian power system, lagging far behind the growing demand, has plunged the country into a chronic shortage situation – with an energy deficit of 6.2 per cent and a peaking shortage of 12.4 per cent in 1999-2000; these were 11.5 and 18 per cent respectively by the end of the 8th Plan (1996-97).

A proper diagnosis of a shortage situation (excess of demand over available supply) requires a close look into the relevant correlates on both the sides: demand and supply. That the demand actually pressing against the system capacity is not out of any extravaganza facilitated by cheap availability should be primarily ensured. That is, the demand we should consider is one of ‘rational’ use; and unfortunately this is not the actual fact in our country. There remains much to be desired in respect of energy conservation, with tremendous scope at every end-use point. For example, the almost free electrical energy, whenever available, is just lavishly burnt out in the agricultural sector. And we know well that even the highly educated and socially conscious among us are not

16

free from the careless forgetting that leaves lights and fans ‘switched on’ even when not in need in offices and other public places, though some time we are keen to conserve energy at home. It goes without saying that the present energy crisis could to a good extent be solved by conscious efforts of energy conservation, the significance of which was acknowledged in India long back.12 The case for energy conservation is recognised to rest ‘on the solid fact that conservation measures are cost-effective, require investments which are much smaller than what would be needed to produce additional energy equivalent to the energy they save and above all, have very short gestation periods, unlike energy supply projects which take years to plan and implement.’ (Government of India 1986: 7). Despite the importance given and measures undertaken, the social behavioural pattern has not been much affected to effect any perceptible result in energy conservation, through better housekeeping, proper maintenance and better controls of instruments, and adaptation of latest technologies.13

Equally disappointing, but of more costly consequences, behavioural pattern of an inefficiency reigns on the supply side also. In no plan period the target in IC could actually be achieved. Avoidable inordinate time overruns leading to excessive cost escalations have been a fundamental trait of the project execution machinery. While the efforts on timely capacity additions have thus been half-baked, the available capacity itself on the other hand has remained half-tapped. For example, the capacity utilisation in terms of energy generated per kW of IC in 1999-2000 was only 4913 kWh/kW or 56 per cent. It is not surprising to find that had the capacity utilisation rate been a minimum of 60 per cent in that year, there would not have been any energy deficit at all!

While capacity utilisation has remained at an improvable low level, a good part of this short-supply has also been frittered away in transit before reaching the end-user. The T & D loss of Indian power sector stood at 25 per cent in 1997-98 and rose to 26 per cent in the next year (Government of India 2001: 175), against 8 per cent of China, 9 per cent of Chile, Thailand and Malaysia, 8 per cent of the World average and 6 per cent of the advanced countries in 1997 (Table 1). In addition to inadequate investment in T & D capacity expansion, defective metering and high pilferage in connivance with the Board

17

officials and under political patronage have contributed to such substantial energy loss. It goes without saying again that the present shortage situation could well be avoided, were at least the T & D system alone operated efficiently. And note that the reported T & D loss percentage has been a grossly underestimated one; as agricultural consumption of electricity in most of the states remains unmetered, it has been a practice of the concerned SEBs to ‘dump’ much of the T & D loss (read energy theft) into this sector in an effort to conceal their operational inefficiency.

The ilfare of the utility has extended to its financial field also. Apart from the impacts of the technical inefficiencies, the government policies have also adversely affected the financial health of the power system. Since the responsibility of power development was originally shouldered by the government, as in the case of other infrastructure sectors with high capital intensity and long lead time, the power sector has been expected to subserve the social, political and economic policies of the State. The patronising policies of the governments have created avenues for excessive employment, especially at the non-technical, establishment and administration level, involving unwarranted cost increases. Moreover, the little energy available after the low capacity utilisation and very high T & D loss itself has been sold out at subsidised rates, irrespective of considerations of costs, under compulsions of government policies. For instance, the gross subsidy involved in electricity sales was Rs. 7449 crores in 1991-92 (accounting for 1.1 per cent of GDP) and Rs. 33814 crores in 1999-2000 (accounting for 1.7 per cent of GDP). The gross subsidy of the state power sector as a percentage of gross fiscal deficit of state governments was about 36 per cent in 1999-2000 (Government of India 2001: 175). And the commercial losses rose from Rs. 4117 crores in 1991-92 to Rs. 24920 crores in 1999-2000 – a six-fold increase.

Again, it has not been in the practice of the SEBs to collect regularly even this revenue realised in such subsidised sales. Thus the receivables to the SEBs have been mounting up, often representing more than 4 months’ sales revenue being locked up with the consumers at any point of time, against the maximum allowable norm of two months’ sales revenue. And all these inefficiencies have left the system with little internal

18

resources for sustainable development. Partly the SEBs were reared in such an unaccountability culture – as the SEBs were to subserve the socio-economic policies of the State and hence expected not to view every aspect of developmental activities exclusively from the viewpoint of profit (as highlighted by the Venkataraman Committee of 1964), there was no compulsive requirement (till the 1978 amendment of the Section 59 of the Electricity (supply) Act of 1948) for the SEBs to break even, or to contribute internal resources to expansion programmes. In addition to Plan outlays allocated to the power sector, government subventions were also on the way in. And the SEBs have never come out of that spell of unaccountable, non-commercial performance.

In this backdrop of a cumulative internal inefficiencies, there is then no wonder that the apparent ‘fiscal crisis’ at the turn of the 1990s should have ushered in ‘the much desired’ reform process in the power sector too. The reforms and restructuring of the Indian power sector, commenced at the initiation of the World Bank has also lit up an informed atmosphere of debates and discourses. However, little light has been thrown on the significant aspects of inefficiency costs involved in the SEBs’ forced functioning that allegedly finally warranted the reforms. The present study starts from this point. It is an attempt at an evaluation of the functioning and developments in the power sector of India, with particular reference to its South-western state of Kerala in a broad objective background of impartial perspective. The special reference to Kerala is warranted in that in the development literature, Kerala is now well known for its remarkable achievements in human development despite low income. Here is a paradox of high level social development coupled with very low provision of infrastructure facilities, especially power supply. Nevertheless, our observations on Kerala apply to most other SEBs, though in varying degrees.

5. The Kerala Scenario

Kerala presents a unique paradox of a level of development comparable with that of the affluent countries and a very low per capita electricity consumption. It has been consistently much lower than the all-India average (278 units against 334 units in 1998-

19

99) and that of the neighbouring states. Ironically, it appears that power development in the state has not received the required thrust as in other states; consequently, the system has always been kept as a smaller one, with the IC rising from 37.5 MW in 1950-51 to 2391 MW in 1999-2000, more than 70 per cent of which is hydro. An undue preference for hydropower has characterised the energy development planning (if any, worth that name) in Kerala, against an optimal planning for a healthy mix of thermal and hydel capacity, on the grounds that the state has no known source of fossil fuels and her resource endowment for power generation is limited to the ‘vast’14 hydro potential. However, even this potential has not been brought under a consistent planning mechanism of materialisation regularly and in time. For one thing, almost all the projects taken up have run over longer time lags, involving exorbitant opportunity costs, extracted cleverly by collusions among the contractors, militant labour, KSEB officials and the politicians in power. Moreover, there has appeared since the Sabarigiri project in the 1960s an undesirable and deleterious tendency of large project preference resulting in wide divergence between capacity and potential; for example, the generation potential of the Sabarigiri project is only 51 per cent of its IC, and that of the Idukki project, a mere 35 per cent! Such wasteful mismatch has eaten up enormous resources and time. Only recently has the KSEB shown some interest in taking up small projects, most of which, however, are entangled as usual in time and cost overruns.

The defective and myopic power development planning in Kerala will be evident when one finds recurring long periods of investment inactivity in the sector. ‘Kerala’s is the only power system in the country with the unfortunate record of not having added anything to its installed capacity’ (as Government of Kerala 1984: 27 laments) for a long period of 9 years since 1976-77 and again for 7 years since 1986-87. During the 1990s, the addition to the state sector (i.e., by the Kerala State Electricity Board (KSEB)) was only about 500 MW! Such inadequacy has been much more in the T & D sector such that it has often so happened that when a power plant is ready for commissioning, evacuation lines are not in place. The traditional feud between the civil and electrical wings of the KSEB has added much to the woe; the warring functionaries of the same organisation have in effect been sinking the ship itself.

20

A favourable condition that has come in handy for the KSEB to keep on the small system characteristic stems from the domestic sector dominant consumption profile, almost unique to the Kerala power sector.15 The domestic sector accounts for about 50 per cent of the total electricity consumption in the state, while industry for just 33 per cent and agriculture about 4 per cent. The consumption share of the industrial sector has steadily been on the decline over time, reflecting the low industrialisation drives in the state. A growing power system presupposes wide and vibrant industrialisation, not just urbanisation as in Kerala. And this has been an important determinant to arrest the momentum of growth in power demand and keep the system smaller in the state. But the functional inefficiencies have cumulated to such an extent that system capacity has for a long time been unable to catch up even with the small increase in the domestic power consumption. 6. The Present Study

The present study on the plight of the power sector in India examines three aspects: the operational inefficiency in the State sector (i.e., State Electricity Boards), the consequent reform drives and the political economy involved in these two aspects. The organisation of the study is as follows:

This introductory chapter in Part I is followed by an evaluation of the physical and financial performances of the SEBs in Chapters 2 and 3.

In Part II, we attempt at a diagnostic analysis of the problems of the power sector focusing on Kerala. The problems studied are demand estimation (Chapter 4), time and cost overruns (Chapter 5), transmission and distribution (T & D) loss (Chapter 6) and pricing (Chapter 7).

21

In Part III, we examine the surge of power sector reforms and start the study with a review of international experiences (Chapter 8), followed by discussions on the reform measures in India (Chapter 9) and in Kerala (Chapter 10).

Part IV deals with the problems in the Indian power sector from a political economy perspective in order to understand the steady decline in efficiency. The final Chapter (12) sums up the discussion.

Below we present a brief presentation of the main ideas in each chapter.

We highlight in Chapter 2 the saga of inefficiency in the State Electricity Boards (SEBs) in respect of their physical performance, focusing on such aspects as technical inefficiency, transmission and distribution (T & D) losses, and its possible underestimation as well as some aspects of institutional and organisational inefficiency. It is shown that much of the capacity/energy deficit we experience today could be easily avoided with some achievable improvement in capacity utilisation and T & D operation.

In Chapter 3 we take up an evaluation of the financial performance of the SEBs; some aspects of cost of electricity supply, tariff and revenue realised are analysed. We also estimate, on some very plausible assumptions, the avoidable cost of inefficiency at a few amenable levels and find it to represent about one-third of the reported cost of electricity supply in India in 1997-98! And this is regardless of a number of other possible inefficiency sources at all levels of performance.

Based on these observations, we argue that the present system predicament is due to problems which are just internal to the system. This then implies that there do remain sufficient quarters for remedial exercises, meant to remove the problems that stand in the way of the SEBs’ improved performance. In other words, what the system badly requires is essence-specific reforms, not structure-specific ones.

22

A situation of capacity/energy deficit arises whenever the demand exceeds the available supply. This calls for detailed studies on the different aspects of demand for and supply of electricity. In Chapter 4 of Part II, we deal with some methodological issues in electricity demand analysis and forecasting.

A large number of studies have come up in India on electricity demand, toeing the same methodology as applied elsewhere. However, we find these studies analytically insufficient and methodologically unsound at least on three grounds. i) As these studies did not care for model adequacy diagnostic checking, indispensably required to verify the empirical validity of the residual whiteness assumptions underlying the very model, their results might be misleading. This criticism in fact applies to all regression analysis in general. (ii) As the time series regression approach of these studies did not account for possible non-stationarity (i.e., unit root integratedness) in the series, their significant results might be just the misleading result of spurious regression. They also failed to take advantage of an integrated analytical framework for a meaningful long-run equilibrium and short-run ‘causality’ in a cointegrating space of error correction. (iii) These studies, by adopting a methodology suitable to a developed power system in advanced economies, sought to correlate the less correlatables in the context of an underdeveloped power system in a less developed economy. All explanations of association of electricity consumption in a hopeless situation of chronic shortage and unreliability with its generally accepted ‘causatives’ (as in the developed systems) of population, per capita income, average revenue, etc., all in their aggregate time series, might not hold much water here. The empirical results in this chapter prove our scepticism at least in the context of Kerala power system. The chapter is divided into four sections. Following an introduction, section two presents a brief theoretical discussion on forecasting and demand analysis, important tests for model adequacy, and unit root problem. In the third section are presented our empirical results and the last one concludes the chapter with an attempt to forecast electricity demand in Kerala in a simple, objective and theoretically

23

sufficient manner. An appendix, including discussions of the theoretical underpinnings of our approach, is provided at the end of the chapter.

On the supply side of shortage, one of the important causatives of inadequate capacity is the time and cost overruns of the power projects. This aspect is analysed in chapter 5 with case studies of a large number of (20) projects, commissioned and under construction, in Kerala. Most of the projects commissioned in Kerala have undergone time overruns of 8 to 13 years over and above the normal construction period, with more than 2 to 8 times cost escalation.

Cost overruns of these projects are compared using different criteria, absolute and relative ones. The avoidable costs to the society involved in such time overruns, in addition to the cost escalation, are also analysed in terms of the additional energy realisable, were the projects commissioned in time, as well as the additional sales revenue thereof. Among the causes of delays, we find the recurring labour militancy as the single factor that puts the highest cost burden and we present some case studies of interest.

One of the primary causatives of energy shortage, T & D loss in India in general is on the high side by world standard. Kerala’s is not an exception. Though the T & D loss in Kerala was around 12 per cent in the 1970s, it went up even to 29 per cent in 1987-88, and at present is a little below 20 per cent mark.

Chapter 6 starts with a general technical note on T & D loss and goes on discussing the possible causes of the very high T & D loss the Kerala system has been suffering from, viz., inadequate investment in transmission capacity as well as its belated realisation, defective metering, energy theft, etc. The next section attempts to estimate the financial costs of the T & D loss; we also estimate the investment required to bring the loss factor down to minimum 15 percentage point and compare it with the possible savings from the loss reduction. We find that there is a 32 per cent net saving over the capital cost.

24

Chapter 7 is an almost exhaustive chapter on electricity pricing. It starts with a discussion of different pricing methods – cost-plus and marginal cost pricing. Also discussed in this context are the two types of regulated utility pricing in practice, viz., fixed price and cost-plus mechanisms. The next section critically examines the power tariff policy in India. This is followed by a discussion on the tariff policy and financial viability of the KSEB.

An Expert Committee constituted by the Kerala State Planning Board under the Chairmanship of K. P. Rao, to review the tariff structure of KSEB, has recommended a number of corrective and constructive measures in its Report of 1998. We examine the tariff rates estimated by the KSEB on the basis of the guidelines for tariff formulation recommended by the K. P. Rao Committee and find that the tariff practice being followed by the KSEB is not at all based on a full cost recovery principle.

We then estimate the electricity rates based on full cost pricing policy, using the parameters under two scenarios, one under the Expert Committee’s tariff policy recommendation and the other under the existing tariff policy.

We also set out in detail the estimation of electricity rates based on marginal cost pricing method, using the capacity costs of a representative hydro power plant and a representative coal-based thermal plant, as well as the average incremental costs of T & D, at different discount rates. The chapter ends with some objective comments on marginal cost pricing, both in theory and practice.

The critical examination of the past performance of the power sector naturally leads us to the reform/restructuring drives being contemplated/carried out in the Indian power sector at present. We begin our discussion on this aspect from a global perspective. In Chapter 8 of Part III, we bring together the experiences from abroad of regulation/restructuring of the power sector.

25

The chapter starts with a brief critical examination of the background of the power sector reforms: the Thatcherite experiments in the UK, the deregulation bids in the US, the World Bank’s Structural Adjustment Loans, etc. This is followed by a detailed discussion on the agenda of power sector reforms – the conditions for bringing in competition in the power sector (through unbundling and possible privatisation), and regulation (through rate of return or price-cap mechanisms). In the third section are given the experiences from abroad – England, USA, Canada, Scandinavia, and France in the developed world; Chile, Argentina, South-East Asia and China in the less developed world.

The chapter ends with a critical analysis of the lessons from the experiences – improved power supply position, inconclusive evidence on efficiency, higher capital costs and price rises, labour market disturbances (higher incidence of unemployment), and environmental issues. An important finding is that public sector orientation of power supply is very much active in some of the developed and developing countries. In many others there is a public-private collaboration, whereas in some others private sector has a dominant position.

Chapter 9 deals with the issue of power sector reforms in India. It starts with the organisation and regulation aspects of the power sector. After a brief discussion on the power sector performance and background for reforms, a detailed review of the power sector reform processes in India is presented – private sector participation in generation, mega power projects programmes, reforms in the transmission sector, setting up of Electricity Regulatory Commissions and enacting a comprehensive Act, Electricity Bill 2000, proposed to replace the existing three Acts that govern the power sector now. The chapter ends with a critical assessment of the responses to the opening up policy – the main problems that dissuade the independent power producers from coming forward are discussed: problems of litigation/renegotiation leading to delays, financing problems, and problems in obtaining fuel linkage agreements and clearances.

26

In Chapter 10, we turn to the reform process in the Kerala power sector. The chapter begins with a detailed analysis of the plight of the power famine that terribly haunts Kerala today. We find that the system growth in Kerala has never been up to the mark of its potential requirement. Unaccountability and negligence in respect of system planning and management, even in the short run, led to deterioration in the system performance – both physical and financial. Like other SEBs in India, the KSEB too has thus been marked for reforms. However, unlike in some other States in India, the politically surcharged atmosphere of Kerala puts up a big NO to unbundling and privatisation of power sector in the State. In this context, we take up a critical analysis of the core issue: is structural reform the panacea? Based on our earlier observations, we argue that the Board should be functionally efficient and financially sufficient to meet all its requirements; however, the problems it faces now are just internal to the system, not structural; this then means that the system predicament can be remedied by removing the problems that stand in the way of the SEB’s improved performance. Reports on restructuring experiences in some other States indicate that such irreversible steps could be disastrous. We conclude the chapter with a number of suggestions based on our analysis.

Finally in Chapter 11 of the part IV, we come to the core organisational issue: the political economy.

Intervention by governments in the capacity of political authority in the economic sphere finds its justification in its historically necessitated role of a regulator on mainly two fronts: (i) to correct market failures in the provision of public goods and in the presence of externalities and natural monopoly, and (ii) to ensure the translation of the lofty ideal of a ‘welfare State’. A convergence of the two fronts is achieved in the establishment of a public sector in a particular historical context. It is argued in a simple neo-classical framework that if the benefits of productive efficiency outweighs the costs of allocative inefficiency, then the society will have a welfare gain from maintaining the natural monopoly organisation of a public utility. Furthermore, if such monopoly power (price) can be brought down to a competitive level, say by means of its nationalisation,

27

then there will be both an equity gain and an efficiency gain. However, the vast scope for administering discretionary powers by the political and bureaucratic control processes involves substantial costs of rent-seeking activities in a non-Smithian cultural regime of self-interest maximisation. This characterises every turn of the capitalist survival strategy, be it the old public sectorisation or the new private sectorisation. In this chapter we discuss these aspects in the context of the development of the Indian power sector with special reference to Kerala.

The chapter begins with viewing the presence of natural monopoly as a public choice problem, followed by a brief discussion of theories on the political process, especially private interest theories. We find that the self-interest alignments in the context of a public utility in a developing country like India are interpretable in a modified framework of principal-agent rent seeking relationship in the context of an intermediate regime. A brief review on rent seeking also is included. In the next section the development of welfare State concept is discussed.

The Indian context is first presented in the background of the development of a public sector at a commanding height. The ilfare of the electric utility is examined in the modified principal-agent relationship framework. This enables us to estimate the costs of rent seeking involved in the administration of the Power sector, especially in Kerala.

Next we take up the implications of the private sectorisation bids in the Indian power sector. Corruption and disasters involved in privatisation is discussed. We conclude the chapter with some viable suggestions on the strength of the finding that there operates no TINA (‘There is no alternative’) factor to warrant privatisation in the Indian power sector. The chapter includes several case studies also.

A detailed summary of the findings along with our concluding observations is given in the last chapter. -----------------------------------------------

28

Table 1 World Energy Scenario GNP GDP per Commercial Electric Power per capita unit of Energy use Consumption Production T & D loss (Dollars) energy use** per capita * per capita (kWh) growth rate @ % + 1997 1997 1997 1980 - 97 1997 1998 Argentina Australia Bangladesh Canada Chile China France India Indonesia Iran Iraq Israel Italy Japan Korea, Republic Malaysia Mexico Myanmar Nepal New Zealand Pakistan Philippines Russian Federation Singapore South Africa Sri Lanka Thailand United Kingdom United States

8970 20300 350 20020 4810 750 24940 430 680 1770 na 15940 20250 32380 330 3600 3970 na 210 14700 480 1050 2300 30060 2880 810 2200 21400 29340

6.9 4 6.8 3 5.7 3.3 5 4.2 4.5 3 na 5.8 7.3 6 3.9 4 5.1 na 3.7 4 3.9 7.2 1.7 2.9 3.3 7.6 4.7 5.3 3.6

1730 5484 191 7930 1574 907 4224 479 693 1777 1240 3014 2839 4084 3834 2237 1501 296 321 4435 442 520 4019 8661 2636 386 1319 3863 8076

1634 8307 76 15829 2011 714 6060 363 329 1163 1353 5069 4315 7241 4847 2352 1459 57 39 8380 333 432 3981 7944 3800 227 1360 5241 11822

3.5 4 10.1 2.5 6.6 8.6 4.2 8.6 13.5 8.8 5.4 6.2 2.1 3.9 12.1 10.9 5.5 6.7 11.1 2.8 8.9 4.1 -0.5 8.3 3.8 6.5 12.5 1.4 2.9

17 6 15 4 9 8 6 18 12 22 na 9 7 4 4 9 14 35 28 11 24 17 10 4 8 17 9 7 6

Low Income Countries Excluding India & China Middle Income Countries Low & Mid. Income Countries East Asia & Pacific Europe & Central asia Latin America & Caribean Middle East & North Africa South Asia Sub-Saharan Africa High Income Countries Europe EMU World

520 380 2950 1250 990 2190 3940 2050 430 480 25510 na 4890

na na na na na 2.2 na 3.3 na na na 5.5 na

646 500 1830 1005 942 2689 1181 1353 443 695 5369 3767 1692

448 222 1928 896 771 2692 1402 1158 324 446 8238 5344 2053

8.4 7.8 6.6 7.1 9.3 8.3 5 7.7 8.6 3.5 2.9 2.4 4.2

12 18 12 12 8 12 16 13 18 10 6 6 8

29

** = Purchasing power parity dollars per kg oil equivalent. + = World Bank Estmates * =Commercial energy use per capita is in kg of oil equivalent @ = Average annual percentage growth rate of production during 1980 to 1997 Sources: (i) World Bank (2000 a: Tables 3.7, 3.8 and 5.10); (ii) World Bank (2000 b: Table 1). Notes:

NOTES 1

“O fire, by a bright path to prosperity lead us…”

2

The principal fossil fuels are coal, lignite, petroleum, and natural gas. Non-fuel sources of energy include waste, water, wind, geothermal deposits, biomass and solar heat. 3

Commercial energy production refers to commercial forms of primary energy – petroleum (crude oil, natural gas liquids, and oil from non-conventional sources), natural gas, and solid fuels (coal, lignite, and other derived fuels) – and primary electricity, all converted into oil equivalents. 4

Note that this has significant implications for environmental pollution.

5

For example, a gas turbine works on the principle of compression and expansion of a gas, normally air. A gas turbine essentially differs from other forms of heat engine in that the expansion and compression take place continuously across rotating parts, not by reciprocating motion as in a diesel engine. 6

With gas or oil as the initial source of energy, an alternative means of conversion is provided by the use of internal combustion engines, in which the (gaseous) fuel is mixed with air to form a combustible mixture in the cylinder and fired by spark ignition. 7

In an AC system, the direction of flow of electric current alternates, i.e., is revered may times a second. A cycle contains two reversals, viz., from one direction to the opposite, and then back to the original direction. 8

Power and energy should be distinguished. Power is the rate of doing work, measured in units of work (energy) per unit time. Its unit is the Joule per second (J/s) or Watt (W). In electrical engineering, power is the plant capacity or rating, expressed in kilowatts (kW) or megawatts (MW). Energy is the capacity for doing work; its unit is the Joule (J). Electrical energy is one form of energy. A 100 W electric light bulb consumes (100W x 60 x 60 seconds =) 360,000 J of electrical energy in one hour. Since 1 Joule = 1 Watt second, this is equivalent to 100 watt-hour (Wh) of electrical energy. In electrical engineering, energy is expressed in kilo-Watt-hour (kWh), also known as unit. 9

It has been found that transportation of coal is less economical than transmission of an equivalent amount of electric power over long distances. 10

Starting in the late 1950s with a 65-mile 100 kV system in Sweden, HVDC is in greater vogue worldwide with higher voltage. 11

The per capita energy use in India in 1999 was 277 kilograms of oil equivalent against China’s 602 kg and the World average of 1428 kg (CMIE 2001: 1).

30

12

Government of India (1986: 11) in this context cites Japan as ‘an outstanding example’ of energy conservation – ‘it succeeded during the years 1973 to 1983 in reducing the amount of primary energy needed to produce the same amount of GNP by a factor of over 30 per cent.’ 13

The Eighth five-year plan of India envisaged a demand reduction of 4,600 MW through demand side management (DSM) measures (i.e., of energy conservation and efficiency of use at the consumer end). A study undertaken at Indira Gandhi Institute of Development Research (IGIDR), Mumbai, (Parikh et al. 1994) has shown that a demand saving of 2235 MW (excluding cogeneration) is possible over a 20-year plan period ending with 2012 AD at a cost of Rs. 10,907 million (in 1993-94 figures), and an energy saving in the terminal year of 6,835 GigaWatthour (ibid.: xxiv). Extending the results to all-India level, Reddy (1995) shows a saving potential of 25,000 MW over a period of 15 years and 100 billion kWh in the 15 th year (quoted in Parikh 2001). A DSM study of low tension consumers in Kerala (Unnikrishnan et al. 1997) gives a peak load saving of 1094 MW and an energy saving of 1048 MU. Many of the DSM options, once adopted by the consumers, help him recover their investment in two to three years. However, a number of institutional barriers to adoption (such as information availability, cognitive skills, risk and uncertainty and capital market accessibility; see Pillai 1999 b) stand in the way of reaping the full potential. 14

This is estimated to be 4500 MW, of which 40 per cent has already been exploited; but the development of the remaining potential is constrained by environmental problems. It should be noted that there are other states (for example, the neighbouring Tamil Nadu) having significant (actually much higher than Kerala’s) hydel potential, but opting for a healthy hydro-thermal mix in power development. 15

Manipur and Tripura are the only other two states in India with a domestic sector dominant composition of power consumption.

31

1

PART I

PLIGHT OF THE POWER SECTOR IN INDIA - SEBs AND THEIR SAGA OF INEFFICIENCY True to the spirit of a social-democratic State, India had originally evolved her power development policy, and shouldered that responsibility, in line with the State’s professed commitment to honouring and ensuring social security equations. Though the State Electricity Boards (SEBs) were statutorily required to function as autonomous service-cum-commercial corporations, they became in effect agents of the Governments to subserve the socio-economic policies of the State, and hence never felt the requirement to break even or to contribute to capacity expansion programs. This unaccountability culture in turn led to gross inefficiency at all levels – technical, institutional and organizational, as well as financial. And the cost escalation from such pampered inefficiency remained above the revenue realized from an irrational subsidized pricing practice. With losses mounting up, the field was getting cleared for some new entrants of ideas and practices, that the so-called ‘fiscal crisis’ at the turn of the nineties ushered in subsequently. Thus has commenced an era of reforms and restructuring of power sector in India, at the initiation of the World bank that has also lit up an informed atmosphere of debates and discourses. However, little light has been thrown on the significant aspects of inefficiency costs involved in the SEBs’ forced functioning that allegedly finally warranted the reforms. This part of the study is a modest attempt at this.

Here, inter alia, we have

estimated, on some very plausible assumptions, the avoidable cost of inefficiency at a few amenable levels and found it to represent about one-third of the reported cost of electricity supply in India in 1997-98 ! And this is regardless of a number of other possible inefficiency sources at all levels of performance.

This part is divided into two chapters. Chapter 2 deals with physical performance focusing on such aspects as technical inefficiency, T & D losses and its possible underestimation as well as some aspects of institutional and organisational inefficiency. Chapter 3 deals with the supply cost of electricity, tariff and revenue as well as financial performance.

2

CHAPTER 2 PHYSICAL PERFORMANCE OF THE SEBs "Quite obviously it came up through the waste, Rejects through ignorance or apathy That passage back. The problem must be faced; And life go on….." - Roy Fuller ("The Image")

1. INTRODUCTION

Power development is placed in the concurrent list of the Indian Constitution, as a joint responsibility of both the States and the Centre. In the First Five Year Plan (FYP), 19.02 per cent of the total Plan outlay was earmarked for power development. Even though the power sector outlay has steadily increased since then in absolute terms, its percentage share fell to 8.89 per cent in the Second FYP, and then rose in the subsequent Plans to reach the all-time maximum of 20.13 per cent in the 6th FYP, only to fall again in the 7th (19.04 per cent) and 8th (18.33 per cent) Plans. The total installed capacity (IC) grew at an average annual compound growth rate of 8.65 per cent during the last four decades from 3,223.11 mega watt (MW) in 1957-58 to 89,090 MW in 1997-98. The share of hydel in total capacity plummeted to 24.6 per cent from 37.7 per cent and that of thermal (including nuclear) went up to 76.4 per cent from 62.3 per cent. Out of the total IC in 1997-98, 63.3 per cent was owned by the States, 30.7 per cent by the Centre, and 6 per cent was in the private sector. Actual generation increased during these four decades at a rate of 9.45 per cent p. a., from 11,369.14 million units (MU; 1 unit = 1 kWh) in 1957-58 to 4,20,405 MU in 1997-98, and total sales of electricity at a rate of 9.0 per cent p. a., from 9,345 MU to 2,93,479 MU respectively.

This seemingly impressive growth, however, conceals much of the innate inadequacies of the system; its deficient capacity, lagging far behind the growing demand, has plunged the country into a chronic shortage situation – with an energy deficit of 11.5 per cent and a peak load deficit of 18 per cent by the end of the 8th Plan (1996-97). Still worse, the per capita consumption of electricity in India has been one of the lowest in the world. The immediate victims of the widening load-capacity gap have been the quality and reliability of the power supplied; for example, the Kerala system operates under low voltage and low frequency (some times up to 47.5 Hz, instead of 50 Hz) to reduce load further in addition to regular power cuts and load shedding, that have become the rule of the day.

3

The cumulative effect of a legion of compounded forces has been at work behind this plight of shortages. For one thing, in no Plan period the target in IC could actually be achieved, the cumulative slippage between the target and the achievement remaining well over 20 per cent. Poor capacity utilization has substantially corroded the system performance. Capacity utilization in terms of energy generated per KW of IC grew over the last four decades in India at an average annual compound rate of just 0.73 per cent from 3,527.38 KWh/KW in 1957-58 (utilization of 40.27 per cent) to 4,718.9 KWh/KW (53.87 per cent) in 199798. Still much more dismal is the condition of capacity utilization in terms of energy sold per KW of IC – with a growth rate of only 0.32 per cent p. a., from 2,899.37 KWh/KW (33.10 per cent utilization) to 3,294.19 KWh/KW (37.61 per cent) over the same period. The growth over the last four decades of energy generated and sold indicates an elesticity of energy sales with respect to energy generated of just 0.843. This highlights high levels of auxiliary consumption and extremely high transmission and distribution (T&D) losses. Adding to these infirmities of inadequacies have been the financial failures from a host of other factors – irrational pricing practices and over-manning, sponsored by political pampering of subsidies at the cost of efficiency, and an infamously flourishing ‘X-inefficiency culture’.

The financial morbidity of the SEBs has allegedly not only decelerated capacity addition in the States, but damped down the private sector sentiments also. As the agenda notes of the recent conference of Power Ministers noted, “Fresh attempts at generation projects by Independent Power Producers (IPP) have reached a dead end with escrow capacity having been more or less exhausted in the country” (Government of India 2000: 1). The only alternative of Central sector investment too stands to suffer from the mounting receivables from the SEBs; (the SEBs owe NTPC a cumulative sum of Rs. 12428 crores – ibid.: 1). The failure in adequate additions in capacity by all the sectors thus continues, and serves as an impetus to inviting World Bank initiation into reforms.

In what follows, we attempt to look into the above aspects for a possible explanation of what in their behaviour trajectories have warranted reforms in the power sector in India in general and Kerala in particular. The references to Kerala situation in this paper arises out of a larger study intended to diagnose the problem faced by the power sector in the State. Nevertheless, our observation on Kerala apply to most other SEBs, though in varying degrees..

4

2. PHYSICAL PERFORMANCE

Inadequate Capacity Additions. The apparently impressive growth in installed capacity at the aggregate national level, however, is not distributed evenly across regions. During the seventies, marked by pervasive enthusiasm for power development, regional disparity in the growth of IC was significantly evident; 9 States, out of the 19 considered, had a growth rate higher than the national average of 7.5 per cent p. a., and 5 States, less than 5 per cent. (Table 1). Even Kerala, which was a power surplus State during this period, was in the intermediate group of States. While power development of higher growth profile entails uneven distribution, power shortage converges growth rates to the minimum and thus ensures an equation among them, as has been evident since the eighties across the States. None of the States has had a growth rate even to touch the immediate vicinity of the national average of 6.6 per cent, all crowding in around a minimum. This also signifies the shift in the weight of capacity addition from the States to the Central sector. In fact, the share of the Central sector in the ownership of the total IC increased from 9.8 per cent in 1970-71 to 22.3 per cent in 1990-91 and then to 30.7 per cent in 1997-98, and the share of the States fell from about 80 per cent in 1970-71 to 63.3 per cent in 1997-98. Thus, with every one percentage point fall in the States’ share, the Central share increased by about 11 percentage point. Indeed the Central sector IC growth rate (11.52 per cent) was about twice the States’ sector one (5.97 per cent), with the former now necessarily catering to the needs of the latter. Even then, by the end of the 8th Plan (in 1996-97), the country as a whole stood to suffer from a peak power deficit of 18 per cent, with little change over the Plan period, and from an increased energy deficit of 11.5 per cent. In 1997-98, these deficits were respectively 11.3 and 8.1 per cent. In most of the States, the situation has been on the worse. Though the Central and State governments do continue to be confident of the dream of a power surplus nation coming into reality by 2012, a time-run-out assertion, considering the present tempo of the progress in many States in the highly unconducive atmosphere complicated by political, social, and ecological issues and conducts, it just seems to be an excusable quarter for another time-run-out. Significant in this respect has been the avoidable disturbing trend in the power sector investment in terms of an unwarranted bias against cheap hydro-power, the hydro-thermal mix being 1:3 by 1997-98; i.e., hydropower accounts for only about 25 per cent of the total IC in 1997-98. This in turn implies an untapped potential of conventional hydro resources to the tune of about 74 per cent (out of the total 84,000 MW estimated at 60 per cent load factor) in the country.

In the popular perception, the temptation would be to

5

blame the organized ecological concerns farrowing the high cost thermal power1. We are not sure that this alone would explain the lack of enthusiasm in exploiting the hydro potential in the country.

Technical Inefficiency Side by side with this inadequate timely capacity additions has been the inescapable long-run experience of under-utilization of the existing capacity itself in the country. An unavoidable reason for an apparent under-utilization of capacity stems from the gradual growth of power demand against the periodic burst of increase in capacity due to its indivisibility. Thus normally with every capacity addition, its utilization rate immediately dips down, as was the case in most of the States during the seventies. But in a power deficit situation, with inadequate capacity addition against an ever-increasing demand, utilization of the available capacity is necessarily expected to be higher, if not maximum. The actual experience, however, has been far from this possibility. In 1997-98 (even in the face of deficit), only 54 per cent of the existing IC in India was utilized (Table 2). As many as 11 (out of 19) SEBs had a use factor much less than this all-India average, including Kerala and Tamil Nadu in the South, and only four (as well as the Central Sector with 63 per cent) had a rate higher than 60 per cent. It should be noted that for a hydro-power dominant system, such as in Himachal Pradesh, Meghalaya, Kerala, and, to some extent now, Karnataka, utilization efficiency should be evaluated with respect to firm power capacity (the always available and dependable capacity corresponding to the minimum stream flow and storage) rather than with respect to IC. Thus taking into account the hydel firm power capacity of 714.5 MW of Kerala in 1997-98, the actual capacity utilization comes out to be 6,308.46 KWh/KW or about 72 per cent. However, a distressing question here concerns about the wide gap of ‘waste’ between the IC and the dependable power of the hydro-plants; the latter being just 42.3 per cent of the hydel

IC in Kerala in 1997-98. Considerable timely efforts on firm power

augmentation projects are called for here, besides those on the usual IC additions. One important causative factor of such low capacity utilization is the poor technical efficiency, reinforced by an inability to attain and assimilate significant technological progress over time. Technical efficiency in generation in general is determined by plant availability (which in turn is determined by forced outages), by plant load factor (PLF), as also by auxiliary consumption. Forced outages occur when a unit is thrown out of service due to unexpected causes such as breakdown, equipment malfunction, etc., and are usually of a random nature. These outages generally befall on the operation side in generators, boilers, turbines, and their auxiliaries. There are also electrical and mechanical forced outages, due to poor quality of fuel, wet coal being supplied, and lack of timely and proper maintenance practices that cause Grid system faults, which are always avoidable. Units are also shut down at times for planned preventive maintenance, intended to ensure their proper running conditions, and also due to lack of adequate system load and of water in reservoir in the

6

case of hydro plants. Considerations of plant availability factor and PLF are usually associated with analyses of technical efficiency only of thermal power plants. Hydro plants are generally expected to be much less prone to forced outages than thermal plants, and their availability is expected to be open always and at maximum subject to firm power capacity constraints. However, the hydro plants in Kerala stand an exception to this expected rule, and also smart for higher forced outage rates (FORs) and loss of load probability (Pillai 1991, 1999). The FORs of the hydro system in Kerala (41 units of 11 plants) on an average were as high as 17.71, 22.59, and 13.12 per cent respectively for the three years of 1982-83 to 1984-85. In 1996-97, it was 8.96 per cent, while the all-India average for thermal plants was 12.8 per cent. The planned maintenance rate of the hydro-power system in Kerala on an average was 12.88 per cent in the same year, and the reserve shut down rate, 11.87 per cent, the latter being largely due to lack of water in storage. The thermal systems of the other Southern States had much lower FORs. Bihar, Assam, Uttar Pradesh, Haryana, and Orissa (till 1994-95) are some of the States with very high FORs and hence much lower availability of capacity (Table 3). The availability factor is defined as unity less planned maintenance rate (PMR) less forced outage rate (FOR); i.e., availability = 1 – (PMR + FOR ).2 In 1997-98, the availability of thermal plants in India in general was nearly 80 per cent, with 8 SEBs having availability higher than this average, including all the three neighbours of Kerala in the South, Andhra Pradesh being the topper (since 1995-96 onwards). The availability of the Kerala hydro-power system is estimated at 78.16 per cent only for 1996-97, reflecting the undesirably higher extent of outages. Bihar’s has been the worst affected SEB for a long time in this respect; Assam follows suit. Delhi, Haryana, Uttar Pradesh, and West Bengal are all in the red (Table 3). Load factor is generally defined as the ratio of average load to maximum (or peak) load. More exactly, it is also defined as the ratio of energy consumed (average load) in a given period to energy which would have been consumed, had the maximum demand been maintained throughout that period. Extended thus to a generating unit, plant load factor (PLF) then refers to the ratio of the actual generation of that plant to its maximum possible generation during a period (one year). Remember that even if the plant is available with a high probability, it may have at times to be backed down due to lack of adequate system load (reserve shut down), and hence the actual generation of the plant may fall short of availability. PLF is then defined in this vein also as availability less reserve shut down rate. Thus the difference between availability and PLF represents a safety margin, buffer, or

7

reserve margin, with a demand- cushioning effect. A PLF very close to availability might be misconstrued as reflecting better capacity utilization; such over-exertion, however,would definitely tell upon the life of the plant, and increase its ‘down’ chances. Hence, along with a higher availability, an adequately high reserve margin also is desirably sought for. PLF also is influenced by factors like age of the generating plant, quality of coal, and its timely and adequate availability, shortcomings in energy evacuation, and equipment deficiencies. While the plant availability remained about 75 to 79 per cent in the 8th Plan period, the average PLF of the thermal plants had a distinct improvement from 55.3 per cent in 1991-92 to 64.7 per cent in 1997-98. In that year, the PLF in the Central sector was nearly 71 per cent, and in the Private sector, 71.1 per cent, while the all-SEBs average was only 60.9 per cent, ranging from 16.1 per cent of Bihar to 82 per cent of Andhra Pradesh. Of the other two neighbours of Kerala, Karnataka had a PLF of 75.2 per cent and Tamil Nadu, 68.1 per cent. Kerala hydro-power system had an estimated PLF of 66.29 per cent in 199697. When compared with availability, most of these rates are satisfactorily tolerable, revealing at the same time the outages that affect availability as the important culprit in low levels of capacity utilization in India. It should also be pointed out that the power plants in the State sector are in general much older than in the Central or Private sector, and the state of maintenance of these units also remains very poor3. A significant determinant of the higher PLF in the Central sector has been an increasing share, in their total IC, (now about 75 per cent) of 500 – 200 MW capacity plants, with fluidized bed boiler (FBB) designs suited to the Indian coal quality, whereas in the State sector such larger capacity plants constitute less than 60 per cent of the total

IC only. Plants of lower capacity (120 MW and below), with an

inappropriate boiler design (Czech), that cannot handle Indian coal of high ash content, make up only 20 per cent of the total IC in the Central sector, but as much as about 40 per cent in the State sector, out of which almost 16 per cent make up plants with less than 90 MW capacity (Table 4). In fact, there have been attempts that attribute the increasing trend in the PLF in the Indian power sector in general since the eighties to the introduction of larger capacity plants – 200 MW introduced in the late seventies, and 500 MW in the mid-eighties4. Some cases, however, invalidate this ‘size matters’ claim – for example, in Punjab, Maharashtra, Karnataka and Tamil Nadu (in the early nineties in the last two cases), where larger capacity plants (more than 200 MW) predominate, the PLF trend was not satisfactory, whereas Andhra Pradesh fared far better with much lower share of larger plants than others5. This reveals some still untapped quarters of improvement available in many States. Side by side with the introduction of new vintage plants of higher technical efficiency, proper and timely maintenance of plants to ensure their healthy life also is indispensable. It has been recognized that in many cases investments in long term rehabilitation and re-powering of old plants fructify more promisingly than in installing new generation capacity.

8

In addition to this technical inefficiency in energy generation is the higher level of auxiliary consumption at generation end that eats into the energy available for transmission. Auxiliary consumption in the power station depends upon its layout, operation conditions, automisation, and design of various equipment. Though taken to be of the order of 3 to 5 per cent in a modern thermal plant and 0.5 per cent in a hydro plant, auxiliary consumption in India has been nearly 10 per cent over the years. Reported as a weighted average of thermal and hydel plants in the State sector, it remained in the range of around 7 per cent in the 8th Plan period. Bihar, Orissa, and West Bengal have had always much higher auxiliary consumption – more than 10 per cent (Table 3). In Kerala, the trend in auxiliary consumption has of late been on the rise, away from the erstwhile satisfactory plane; it is expected to be so, as more and more thermal plants come into operation.

T & D Losses The energy sent out, net of auxiliary consumption, then fritters away in transmission and distribution (T & D) network6 to such a substantial extent that by the time it reaches the sales point, it would often be only a smaller fraction of the net generation. Over 82 billion units of electricity were lost in T & D in various States in India in 1997-98. The losses increased from 19.8 per cent in 1992-93 to 23 per cent in 1996-97, and then declined marginally to 21.8 per cent in the next year (Table 5). These are very high by international standards – compared with less than 10 per cent in most of the developed economies and with less than 15 per cent in many developing countries such as China (7 per cent), Thailand (10 per cent), Argentina (12 per cent), and Chile (11 per cent) (Rao, et. al 1998-99: 42). In almost all the States the losses remain very high, from 15.2 per cent in Maharashtra to 47.5 per cent in Jammu & Kashmir in 1997-98. Delhi stands next to Jammu & Kashmir with 43 per cent; then Orissa (39 per cent), Haryana (32.2 per cent), Andhra Pradesh (25 per cent), Assam (24 per cent), Bihar, Uttar Pradesh, and Rajasthan (23 per cent each). T & D losses in Kerala was in a satisfactorily comparable position till some two decades back, the losses having been less than 15 per cent. However, it increased to substantial extent in the following years, averaging about 24 per cent during 1982-83 to 1996-97. In 1997-98, it was 17.87 per cent, while for Andhra Pradesh, it was 25 per cent, for Karnataka, 18.4 per cent, and for Tamil Nadu, 17 per cent.

The neglect of the T & D sector, especially the transmission sector, in terms of adequate investments in capacity and maintenance, and the lack of systematic T & D planning over the years are the major technical factors contributing to the high level of T & D losses. Defective metering, unmetered supply and pilferage are the main non-technical factors. There has been over the years a

9

pronounced bias in investment in favour of augmenting generation capacity to the utter neglect of the 1:1 norm in investment in generation and T & D sectors. Despite the increased funds allocation given to T & D sector in the recent past, out of the belated recognition of the compounded effects of neglect, under-utilization or diversion of funds (meant especially for transmission capacity augmentation) into generation and/or distribution sector still plagues the system. Increase in demand by an increasing number of consumers vis-à-vis inadequate T & D capacity has resulted in heavy overload on the system, causing substantial line losses. During the period 1970-71 to 1996-97, the number of consumers increased by 7.26 per cent per annum, and IC, though restricted, by about 7 per cent, while the annual growth in transmission lines was 4.55 per cent and distribution (low tension, LT) lines, 6.15 per cent. The ratio of the length of transmission lines to the length of distribution lines dropped from 7.73 per cent in 1970-71 to 5.2 per cent in 1996-97; in 1990-91, it was only 4.78 per cent. Evidently, the imbalance between the two has been on the rise, worsening the overload problem. Where domestic load is more spread out, as in Kerala, large-capacity distribution transformers demand large lengths of LT line, resulting in increased line losses. Larger number of small transformers is more desirable in such situations; this is possible only with substantial increase in 11 kV (or above) lines. However, the 11 kV lines to LT lines ratio which was about 1:1 in 1951 in Kerala, for example, has now fallen to 1:5 (in 1997-98). The ratio of the length of 15/11 kV lines to that of LT lines was 1:2 in 1996-97 in India. At the same time, transformation losses are higher for small-capacity transformers; in a 200 kVA transformer, it has been found, iron loss is 0.28 per cent, and copper loss is 1.67 per cent, while in a 25 kVA transformer, the losses are respectively 0.75 per cent and 3.5 per cent (Shah, Dalal, and Patel 1985). Three-phase lines instead of the common onephase line would also reduce the T & D loss considerably (by more than one-sixth). The nominal transmission (extra high voltage, EHV) lines in vogue in India are of high-voltage direct current (HVDC), 400 kV, 230/220 kV, 110 kV, and 66 kV. HVDC lines have been so far introduced by the Andhra Pradesh SEB (37 circuit km.) and the Central sector in Northern region (1630 ckt. km.) only, and 400 kV lines by the SEBs of Punjab, Uttat Pradesh, West Bengal, Madhya Pradesh, Maharashtra, Gujarat, Karnataka, Bihar, and Orissa, besides the Central Sector. The low voltage 132/110/90 kV lines predominate the transmission sector in the proportion of 400 kV : 230/220 kV : 132/110/90 kV = 0.36 : 0.81: 1 (as in 1996-97). Similarly, the low tension (LT) distribution lines predominate over the high voltage ones in the proportion of 33/22 kV : 15/11 kV : 6.6/3.3/2.2 kV : LT = 0.077:0.49:0.0015:1. The proportion of LT lines to EHV lines is just 1 : 0.07. In Kerala, the proportion of 220 kV : 110 kV : 66kV : 11 kV : LT in 1997-98 was 0.013 : 0.019 : 0.0195 : 0.195 : 1, and the proportion of LT lines to EHV lines was 1 : 0.032, more than double the all-India average. Since a predominantly low voltage network characterizes the Indian power sector in general, higher technical

10

line losses and poor quality of electricity at user ends are an inescapable fact. In fact, the low-voltagelow-frequency profile common in many States is an easy option of escape route for mitigating the power deficit, which would get aggravated with any attempt to raise the voltage level in the basic system without adequate additions to generating capacity. In this respect, the tie up of a State Grid with a Regional Grid that operates at low system frequency due to overload further reduces quality. For example, the Southern Grid, with which Kerala system is tied up, runs at a low frequency up to even 47.5 Hz instead of the normal 50 Hz. The low voltage conditions in turn lead to the use of stepup transformers or voltage stabilizers by consumers, which in turn induces high inductive load and further worsens the conditions.

Even the SEBs that report lower losses (e.g., Maharashtra) have to improve further to attain standards of efficient systems abroad; yet a large potential for energy and capacity savings is available if all SEBs could bring losses down at least to these levels e.g., of Maharashtra). Let us assume such a situation – that T & D losses are only 15 per cent of the energy available in India. Then in 1997-98, the losses would be only 59,443.13 MU, instead of the actual 82,462.9 MU, giving a potential saving in energy of 23,019.78 MU and in revenue of Rs. 42,466.88 million, at an average rate of Rs. 1.845 per unit. This brings out the immense cost of the avoidable inefficiency in the Indian T & D sector – a revenue loss of around Rs. 4,000 crores every year ! Moreover, the energy thus lost in excess of the notional 15 per cent in fact represents a generating capacity of about 4,380 MW at 60 per cent load factor. It means that if the T & D system in India could maintain the energy loss at least at 15 per cent per annum, it could then help dispense with the need for adding about 4,000 MW to the installed capacity, saving immensely in investment and working capital costs. That these savings were in addition to the potential increase in sales revenue by around Rs. 4,000 crores per year speaks volumes for the gravity of the problem. Now just reflect upon a drop in T & D losses to the ideal 10 per cent norm. It must, however, be noted here that the non-technical energy losses due to theft, etc., cannot be converted into energy and capacity savings, but can only be included in revenue savings. Though theft of electricity has been made a cognizable offence since 1986 under the Indian Electricity Act, 19107, this has had no effect on the theft problem. Some of the SEBs are reported to conduct checks and detect cases of theft or misuse of electricity. Some estimates of energy lost in pilferage/misuse are also available – e.g., in Karnataka, as much as 16.3 thousand units of electricity are estimated to have lost per case of theft/misuse detected in 1996-97, and in Gujarat, only 0.43 units per case detected. In Kerala, the loss was estimated at 1,842.3 units per case detected, in Maharashtra, as much as 6,895.1 units per case, and in Punjab, 3,065.4 units per case (Table 5). On an average, in 1996-97, an estimated quantum of about 1,332 units of electricity was lost per case of theft/misuse

11

detected in 13 States. Though under-estimates, these figures do represent a big drain on the SEBs’ revenue stream. The estimated revenue loss for the 13 States in 1996-97 in this respect amounted to Rs. 100.19 crores at an average rate of Rs. 1.63/unit, and for Punjab alone, Rs. 29.63 crores, at Rs. 1.36/unit. Data are unavailable/withheld on the estimates of energy loss in theft in some SEBs, where in fact pilferage is a major problem, for example, Delhi, with no rural electrification commitment that involves high T & D loss.

T & D Losses – An Underestimate.

There is little doubt that even these high figures of T & D losses are only underestimates that find a suitable cover-up in the overestimates of agricultural consumption. In most of the States, agricultural consumption is largely unmetered, and the SEBs, in their eager to record reduced transit losses, find this situation a convenient ‘dump’ for a good part of the unaccounted-for energy. We can have a rough estimate of such diversion. The energy consumption per energized pump-set in the agricultural sector, (that accounts for about 30 per cent of total electricity consumption), of India in 1997-98 was 7,492.4 units. In Tamil Nadu, where most of the agricultural consumption, accounting for about 27 per cent of total consumption, is metered, consumption per energized pump-set in that year was only 4,471.05 units. In 1996-97, average electricity consumption per energized pump-set in India was 7,264.72 units and in Tamil Nadu, 4,425.46 units per set. It may not be unreasonable then to assume that the power consumption in general in the agricultural sector in India is around 4,000 – 4,500 units per energized pump-set. This in turn implies that about 40 per cent of what is branded as agricultural consumption, estimated as a residual after setting the target for T & D loss (reduction), accounts for unaccounted-for energy. The estimate is of course, a rough one, as it ignores the differences in capacity, efficiency, and duration of use of the pump-sets on the farm across the country: still it drives home the essential point of the cover-up. Comparing energy consumption per kW of connected load (CL) in the agricultural sector would be a better method, though it too suffers from the problems of differences in efficiency and duration of use, etc. We have, however, tried out that also. The average electricity consumption per kW of CL in the agricultural sector in India in 1996-97 was 1,866.36 units/kW, and that in Tamil Nadu, 1,287.2 units/kW, indicating that, by this definition, a little over 30 per cent of what is reported as agricultural consumption in India represents unaccounted-for energy. If we consider consumption per agricultural consumer, it was 7,444.21 units in India in 1996-97, and 4,711.71 units in Tamil Nadu, showing that about 37 per cent of the reported agricultural consumption in India must be included in the unaccounted-for energy category. Thus there is no gainsaying the fact that the so-called agricultural consumption in India is an over-estimate

12

by as much as 30 to 40 per cent – an easy cover up of the large quantum of energy losses. The actual T & D loss in India inclusive of this then amounts to about 31 to 29 per cent, instead of the reported 21.8 per cent in 1997-98. Now, considering an actual 30 per cent T & D loss (including unaccounted for energy) and proceeding with the assumption of the Indian power supply system attaining a standard level of 15 per cent loss only, we find, in 1997-98, a potential saving in energy to the tune of 59,443 MU and in revenue of Rs. 10,966 crores at an average rate of Rs. 1.85 per unit. The potential energy saving represents a generating capacity of nearly 11,310 MW at 60 per cent load factor. So much is the cost of inefficiency in one aspect (T & D) of the electricity supply in India !

Power Purchase

A good part of the net generation, itself falling short of demand, thus being lost in transit, power purchase from other States and Central sector perforce increases more than is required otherwise. In 1997-98, energy import by SEBs ranged from 16.5 per cent of the total energy sales in Meghalaya to as much as 164 per cent in Orissa. Bihar (109 per cent), Karnataka (116 per cent), Delhi (147.5 per cent), Jammu & Kashmir (156.5 per cent), West Bengal (89 per cent) and Assam (79 per cent) were the other major importers (Table 6). The appalling situation of having to resort to energy purchase much in excess of cent per cent, as in the case of the above five SEBs, means that their auxiliary consumption and other losses of energy far exceeded their own generation to cut down even the costly purchase itself. For an instance, in the case of Delhi in 1997-98, total energy sold was only 67.8 per cent of the energy imported; in other words, about 32 per cent of the energy purchased plus the whole of its own generation were lost ! In the same year in Orissa, total energy sales were only 61 per cent of the energy purchase, and the losses, the whole generation plus 39 per cent of the purchase ! Other (14) States were able to convert in varying degrees their own generation into sales revenue; in West Bengal, only 10.2 per cent of the energy generated went into sales stream (plus the whole purchase, the remaining having been lost); and in Meghalaya, as much as 76.5 per cent, in 1997-98. Kerala had to import about 55 per cent of energy needed to meet her consumers’ demand, in addition to about 67 per cent of her own generation in that year. That also means about 33 per cent of energy generated was lost in auxiliary consumption and in transit.

Energy Consumption

Thus the Indian power sector, characterized by inadequate capacity, its under-utilization, and high level of losses, remains poor in its supply. Being one of the world’s lowest, per capita

13

consumption of electricity in India was only 283 units in 1993 as against 2,761 units of Venezuela, 1,627 units of Chile, 1,479 units of Uruguay, 1,463 units of Brazil, 1,438 units of Argentina, and 1,072 units of Mexico (Council of Power Utilities 1997). In 1996-97, it just reached 338 units, with the Western region having the maximum of 521 units and the North eastern region, the lowest, 107 units. In the Southern region, Kerala has always had the lowest per capita consumption, always lower than the all-India average also. One of the reasons for this , besides the restricted energy supply, is the high density of population per sq. Km. in Kerala, which is more than double of all-India average. The same low level profile is seen for Kerala in terms of electricity consumption per connected consumer also – it was only 1.41 thousand units in 1996-97 as against the Southern region average of 2.09 thousand units and the all-India average of 2.95 thousand units. Bihar enjoyed the highest average consumption level per consumer of 5.4 thousand units and Nagaland had to be contented with only 1.04 thousand units. India stand poor in terms of the average connected load (CL) also, with only 1.97 kW per consumer in 1996-97. While as many as 11 States had higher average CL than the all-India average, no State in the Southern region came closer to this, with Kerala having only 1.23 kW of CL per consumer .

A significant change over time in the composition of electricity consumption by customer categories is discernible in almost all the States in terms of increasing share in total consumption of the domestic and agricultural consumers at the cost of industrial as also commercial consumers. At the all-India level, share of the domestic sector increased from 10.8 per cent in 1970-71 to 18.4 per cent in 1998-99, and of the agriculture from about 10 per cent to 30 per cent during the same period, whereas the share of the industry dropped from 61.6 per cent to 33.7 per cent and of the commercial sector from 7.2 per cent to about 5 per cent during this period. Kerala witnessed the most dramatic behaviour in these trends – an 11-fold increase in the share of domestic sector to account for nearly 50 per cent of the total electricity consumption in the State, and a 50 per cent fall in that of industry to account for about 33 per cent of total consumption. No other State in India (barring Manipur and Tripura) has such a domestic-sector-dominant composition of power consumption. Kerala also is one among the very few States (Orissa, Uttar Pradesh, and West Bengal) where the commercial sector prospered to some extent. In most of the other States, agriculture and/or industry account for the major share in total power consumption (Table 7), with agriculture enjoying the highest share in total consumption, with an average of nearly 40 per cent, in Andhra Pradesh, Gujarat, Haryana, Karnataka, Madhya Pradesh, and Uttar Pradesh. However, the increase in the share of agricultural consumption should be taken with a pinch of salt, since in most of the States, as already explained, it just represents the residual, that remains after accounting for all other sectors’ consumption and the ‘targeted’ losses.

14

That the increase in the share of the agricultural consumption in Tamil Nadu, where it is mostly metered, was from 24.8 per cent in 1970-71 to just 27 per cent in 1998-99 does in fact lend enough strength to this contention. Moreover, the very high T & D loss percentage reported, for example, for Delhi, where agricultural consumption is very minimal and thus offers no convenient ‘dump’ for unaccounted-for energy, also supports our argument8.

In addition to this low level of electricity consumption (even per customer), electricity supply industry in India is characterized by low level of accessibility – by 1991 Census, only about 42 per cent of the households in India had electricity facility, with wide rural (38.5 per cent) – urban (75.8 per cent) disparity. The lowest accessibility was in Bihar, with only 12.6 per cent of the households having been electrified and as many as 94 per cent of the rural households remaining unelectrified (Table 8). In Kerala, the percentage of households electrified, according to 1991 Census, was only 48 per cent, even though she achieved the target of cent per cent village electrification long back (in May 1979). However, as per a recent survey (Zachariah, et al. 1999: 198), conducted in 1998, 74 per cent of the households have electricity facility in Kerala. This explains partly the rapid increase in the share of domestic consumption in Kerala. The Prasad Working Group on Energy, appointed by the Government of India (1979) opined long back that village electrification is very deceptive as an index of rural electrification. That in many States, electricity still remains inaccessible to more than 20 to 25 per cent of the urban households, even though the cost of providing connection in the urban areas is minimal, is a pointer to the sluggish growth of this industry.

Institutional and Organizational Inefficiency

Besides these taut constraints upon the technical efficiency of the power sector in India are the institutional and organizational factors. Though the Electricity (Supply) Act, 1948, requires the SEBs to function as an autonomous corporation, their actual position is as good as that of a State government department. Excessive interference in the affairs of the SEBs by the State governments, in their careerist pursuit of patronizing the social security concerns, has resulted, for one example, in over-employment in the SEBs, especially and more unwarrentedly, in administration section. The number of employees per MU of energy sold in India in 1990-91 was about 5 (implying a labour productivity of 0.2 MU per employee), while it was 0.2 (or 5 MU per employee or 25 times higher than that in India) in Chile, Norway, and USA, about 0.6 (or 1.7 MU per employee) in New Zealand, Argentina, and UK, and less than 2.5 (or 0.4 MU per employee) in some developing countries such as China, Philippines, and Indonesia (Rao et al. 1998-99: 42-43). Though the ratio declined marginally to

15

3.6 in 1996-97, still higher than the standards abroad, wide disparity prevails across the States, from 41.4 in Arunachal Pradesh to 1.9 in Gujarat (Table 9). Kerala had a ratio (3.8) somewhat corresponding to the all-India average, and slightly higher than her neighbours. The States like Andhra Pradesh, Maharashtra, Madhya Pradesh, etc., where there is a significant component of thermal generation, which entails substantially more manpower than required for hydro-generation, had higher labour productivity than Kerala with a pure hydro system. The over-manning problem is acute in the Special Category States of Jammu & Kashmir, Himachal Pradesh, and the North-Eastern States. Number

of

employees

dropped to 11.2 in 1996-97.

per Kerala

thousand had

consumers in India was 13.3 in 1992-93, which

the lowest ratio during

all these years with 6.1

in 1992-93 and 5.5 in 1996-97. Karnataka with 5.9 in 1996-97 stood next to Kerala. Arunachal Pradesh had the highest ratio in this respect also, with all other Special Category States having higher ratios.

Another institutional factor breeding inefficiency has been the lack of professional management with commitment, accountability, inclination and initiative in decision making. A steady enervating erosion of competitive management values has sapped the institutional texture to the bottom, giving rise to all-round X-inefficiency. For one thing, continuity of management by top personnel at the policy making level has been a perpetual loss. In most of the SEBs, the average tenure of Chairmen and Chief Engineers is very limited – for an example, four Chairmen of KSEB in 197374 had tenures less than one year, out of which one of them had less than three months (Government of Kerala, 1984: 41). The new Chairman of the KSEB, who has recently taken charge, is the fourth in four years. Similarly, there were five incumbents on the chair of Chief Engineer (Planning) of the KSEB in a period of six years during 1978 to 1984 (Government of Kerala 1984: 41). The story still continues and is the same with other SEBs also. The appointments being mostly on seniority basis, by the time a person reaches the top chair, he would be on the verge of superannuation, that retards his commitment and involvement in serious policy making.

Committees after committees have

recommended that appointments be made based on selection, and that the selected person with proven ability and integrity should have at least 2 to 5 years further service for superannuation (Government of Kerala, 1984: 41– 42; Government of Kerala, 1997: 57–58).

Moreover, the socio-political dynamics in different States have led to a situation of wide-spread corrupt practices of nepotism, all at the cost of merit, ability, and efficiency. A general lethargic indisposition for accountability booms under such umbrellas of patronage. “Certainly improved worker selection could improve productivity at the plant level. To the extent that people are not

16

working at what they are most proficient at, productivity should rise as a consequence of superior selection methods” (Leibenstein, 1976: 38). Leibenstein’s analysis of internal motivation to efficiency starts from the premise that contracts for labour supply within the firm are incomplete, they do not include a specification of the job, so the efficiency of the labour depends on the motivation to effort, which by all counts is constrained by his preference for less effort, confined in an ‘inert area’. This problem is more acute in the public sector of many developing countries, where loose contract, if at all any, guarantees job security till superannuation, whatsoever be the output of his effort. “Since there are no professional job descriptions, personnel are often assigned to areas for which they are not competent…..People are hired against general specifications and not specific job needs….Employees do not have a clear understanding of their responsibilities. Positions do not have performance objectives…(and) clearly defined selection criteria for recruitment purposes.” (Government of Kerala, 1998: 5.4). Besides the superior selection procedure, linking the terms of job continuity and remuneration to productivity would certainly yield a sea of change.

-----------------------------------------------

17

TABLE 1: Growth of Installed Capacity Installed Capacity (MW)

Andhra Pradesh Assam Bihar Delhi Gujarat Haryana Himachal Pradesh Jammu & Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Meghalaya Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal Central Sector DVC State Departments Local Bodies Private Sector All India

1970-71

1980-81

1997-98

608 180 499 252 907 504 51 40 878 547 727 2119 68* 564 680 541 1966 1351 1212 1441 1062 974 267 1488 14709

2240 228 941 276 2197 1141 129 206 1470 1012 1631 3992 131 923 1536 810 2329 3612 1726 2198 1422 1481 276 1382 30214

5764.2 616.7 1988.4 653.6 4883.2 1780.3 299.5 365.8 3434.5 1775.8 3875.9 8289.8 188.8 1693.0 2465.1 1369.8 5763 6168.8 2904 27379.5 .. .. .. 5337.0 89090.0

Annual Average Compound Growth Rate (%) 1970-81 1980-98 1970-98 13.929 2.392 6.549 0.914 9.250 8.514 9.724 17.810 5.289 6.350 8.416 6.538 14.013* 5.049 8.490 4.119 1.709 10.334 3.599 4.313

5.717 6.028 4.499 5.202 4.810 2.651 5.080 3.435 5.119 3.366 5.224 4.392 2.173 3.633 2.822 3.139 5.474 3.199 3.108 15.994

8.687 4.666 5.254 3.593 6.433 4.785 6.776 8.542 5.182 4.461 6.395 5.182 4.751* 4.155 4.886 3.501 4.064 5.786 3.289 11.522

-0.736 7.464

8.272 6.568

4.844 6.899

Note: * = For (with respect to) 1975-76; All India IC (1997-98) includes that for EDs, BBMB and others (Islands). Source: For 1970-71 and 1980-81 and for Haryana, from CMIE, Energy, March-April, 1999; 1997-98 from Planning Commission (GOI), Annual Report on Working of SEBs & EDs, April1999; for Kerala, KSEB, Power System Statistics.

18

TABLE 2: Growth of Energy Generation Energy Generated (MU) Utilization (KWH/KW) Utilization Rate (%) ACGR of Generation (%) 1970-71 1980-81 1997-98 1970-71 1980-81 1997-98 1970-71 1980-81 1997-98 1970-81 1980-98 1970-98 Andhra Pradesh Assam Bihar Delhi Gujarat Haryana Himachal Pradesh Jammu & Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Meghalaya Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal Central Sector All India

2937 7319 369 465 1372 2281 1027 1313 4176 9363 1848 4289 62 245 168 768 4754 6392 2126 5242 2754 5952 9134 17664 181* 353 1766 3137 2365 6483 1509 3393 5638 7372 5725 10190 4056 5563 5399 8450 55828 110844

Note: * = For (with respect to) 1975-76; Source: As in Table 1.

27674 4830.6 1074 2050.0 3666 2749.5 2509 4075.4 25102 4604.2 3773 3666.7 1274 1215.7 976 4200.0 17032 5414.6 5188.7 3890.2 20125 3788.2 41466 4310.5 597 2661.8* 5729 3131.2 13007 3477.9 7694 2789.3 23079 2867.8 23690 4237.6 10540 3346.5 151189 3746.7 420405 3795.5

3267.4 2039.5 2424.0 4757.2 4261.7 3759.0 1899.2 3728.2 4348.3 5182.0 3649.3 4424.8 2694.7 3398.7 4220.7 4188.9 3165.3 2821.2 3223.1 3844.4 3668.7

4801.0 1741.5 1843.7 3838.7 5140.5 2119.3 4253.8 2668.1 4959.1 2921.9 5192.3 5002.1 3162.1 3383.9 5276.5 5616.9 4004.7 3840.3 3629.5 5522.0 4718.9

55.14 23.40 31.39 46.52 52.56 41.86 13.88 47.95 61.81 44.41 43.24 49.21 30.39* 35.74 39.70 31.84 32.74 48.37 38.20 42.77 43.33

37.30 23.28 27.67 54.31 48.65 42.91 21.68 42.56 49.64 59.16 41.66 50.51 30.76 38.80 48.18 47.82 36.13 32.20 36.79 43.89 41.88

ACGR = Annual Average Compound Growth Rate.

54.81 19.88 21.05 43.82 58.68 24.19 48.56 30.46 56.61 33.36 59.27 57.10 36.10 38.63 60.23 64.12 45.72 43.84 41.43 63.04 53.87

9.56 2.34 5.21 2.49 8.41 8.78 14.73 16.41 3.00 9.44 8.01 6.82 14.29* 5.91 10.61 8.44 2.72 5.94 3.21 4.58 7.10

8.14 5.05 2.83 3.88 5.97 -0.75 10.18 1.42 5.93 -0.06 7.43 5.15 3.14 3.61 4.18 4.93 6.94 5.09 3.83 18.49 8.16

8.66 4.04 3.71 3.36 6.87 2.68 11.85 6.73 4.84 3.36 7.64 5.76 5.57* 4.45 6.52 6.22 5.36 5.40 3.60 13.14 7.76

19

TABLE 3: Some of the Technical Performance Indicators Forced Outage Rate (%) Load Factor (%) Auxiliary Consumption (%) 1991-92 1997-98 ACGR (%) 1991-92 1997-98 ACGR (%) 1992-93 1997-98 ACGR (%) Andhra Pradesh 11.28 4.30 -14.85 62.1 82.0 4.74 5.36 6.53 4.03 Assam 34.39 48.50 5.90 24.6 21.3 -2.37 9.81 7.84 -4.38 Bihar 27.57 48.30 9.80 21.3 16.1 -4.56 12.87 14.00 1.70 Delhi (DESU) 16.50 19.08 3.09 57.2 47.2 -3.15 8.44 8.26 -0.43 Gujarat 10.62 7.80 -5.01 56.9 65.6 2.40 10.39 9.47 -1.84 Haryana 30.94 22.70 -5.03 45.8 49.4 1.27 5.33 6.00 2.40 Himachal Pradesh : : : : : : 0.35 0.48 6.52 Jammu & Kashmir : : : : : : 1.00 1.00 0.00 Karnataka 8.81 5.30 -8.12 59.1 75.2 4.10 0.12 1.82 72.26 Kerala 4.19 8.96* 16.42* 73.6* 66.3* -1.73 0.50 0.61 4.06 Madhya Pradesh 16.51 10.40 -7.41 49.2 66.0 5.02 9.26 8.63 -1.40 Maharashtra 13.67 9.40 -6.05 61.3 68.3 1.82 7.91 7.38 -1.38 Meghalaya : : : : : : 0.42 0.34 -4.14 Orissa 26.98 5.20 -24.00 30.0 65.3 13.84 2.89 10.52 29.49 Punjab 7.07 4.70 -6.58 52.8 69.1 4.59 4.20 4.97 3.42 Rajasthan 14.82 5.00 -16.56 65.7 80.5 3.44 7.33 7.47 0.38 Tamil Nadu 13.80 8.20 -8.31 55.7 68.1 3.41 6.06 7.24 3.62 Utter Pradesh 27.67 23.20 -2.89 44.3 48.8 1.63 8.74 7.18 -3.86 West Bengal 13.90 22.00 7.95 30.8 40.0 4.45 11.14 10.8 -0.62 Central Sector 13.50 10.60 -3.95 64.5 70.4 1.47 Private Sector 8.60 2.90 -16.57 64.5 71.1 1.64 All India 15.19 12.10 -3.72 55.3 64.7 2.65 6.91 7.06 0.43 Note: * = for (with respect to) 1996-97; ACGR = Annual Average Compound Growth Rate (%). Sources: Planning Commission, (GOI), April 1999; Kerala's forced outage rates and load factors are estimated from KSEB, System Operations and Power System Statistics (various years) respectively. State

20

TABLE 4: Capacity-wise Distribution of Thermal Plants - 1994-95 Percentage Distribution of Plants by MW Size > 200 - 210 140 -150 115 - 120 105 – 110 90 -110 < 90 Total Andhra Pradesh Assam Bihar Gujarat Haryana Jammu & Kashmir Karnataka Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal All SEBs Electricity Depts. Central Sector Private Sector Total

59.1 0 0 57.8 25.2 0 83.1 66.5 72.4 0 74.1 74.1 79.9 52.3 60.7 59.2 0 74.5 38.4 62.4

0 0 0 7.1 0 0 0 0 4.4 0 0 0 0 0 0 1.8 0 5.2 5.5 3.2

Source: Rao et al. (1998-99), Table 6.

0 0 0 18.2 0 0 0 15.3 9.5 0 0 0 0 0 19.8 7.3 0 8.2 4.4 7.4

20.7 0 66.2 0 52.8 0 0 0 6.9 46.8 25.9 25.9 14 21.6 0 13.8 10.5 3.4 10.6 9.8

0 0 7.7 0 0 0 0 0 0 0 0 0 0 7.4 0 1.6 0 4.1 0 2.4

20.2 100 26.1 16.9 22 100 16.9 18.2 6.8 53.2 0 0 6.1 18.7 19.5 16.3 89.5 4.6 41.1 14.8

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

21 TABLE 5: T & D Loss as Percentage of Electricity Available ACGR (%)

Theft / Misuse Cases (1996-97) Number Loss (MU) 25681 61.81 .. .. 579 NF 675 0.38 NF NF .. .. 46241 0.02 NF NF 1529 0.13 NF NF 184 3.00 1813 3.34 245185 150.68 6142 42.35 .. .. 654 0.72 .. .. .. .. 845 3.16 70924 217.41 53218 113.99 .. .. 774 NF .. .. NF NF 8249 17.72 462693 614.71

1970-71 1980-81 1990-91 1997-98 1970-98 Andhra Pradesh 25.42 22.6 22.4 25.0 -0.06 Arunachal Pradesh NA 24.0 20.0 30.0 1.32* Assam 17.68 19.3 24.1 24.0 1.14 Bihar 22.85 22.1 21.1 23.0 0.02 Delhi (DVB) 11.07 18.4 24.9 43.0 5.15 Goa 18.09 25.7 25.0 26.0 1.35 Gujarat 14.52 19.8 23.7 18.0 0.80 Haryana 27.94 22.6 27.5 32.2 0.53 Himachal Pradesh 12.23 19.3 21.5 17.4 1.31 Jammu & Kashmir 21.66 48.1 42.3 47.5 2.95 Karnataka 14.62 24.6 20.1 18.4 0.86 Kerala 12.80 14.2 21.6 17.9 1.24 Madhya Pradesh 14.69 22.3 24.9 19.0 0.96 Maharashtra 13.67 16.2 18.1 15.2 0.39 Manipur NA 55.6 28.0 21.5 (-)5.44* Meghalaya NA 9.1 11.8 16.9 3.71* Mizoram NA 22.2 29.6 26.0 0.93* Nagaland NA 23.1 26.1 29.0 1.35* Orissa 6.15 19.2 25.3 39.0 7.08 Punjab 22.38 19.6 19.0 18.0 (-)0.80 Rajasthan 13.11 26.6 25.9 23.0 2.10 Sikkim NA 22.9 24.5 20.0 (-)0.79* Tamil Nadu 17.67 19.1 18.7 17.0 -0.14 Tripura NA 31.5 29.6 29.8 (-)0.33* Uttar Pradesh 24.49 15.6 26.9 23.0 -0.23 West Bengal 10.18 13.7 21.8 19.7 2.48 All India 17.55 20.6 22.9 21.8 0.81 Note: * = with respect to 1980-81; NF = Not Furnished (to CEA). Source:1970-71, 1980-81 and 1990-91, CMIE, Energy, March - April 1999, and Basic Statistics Relating to the Indian Economy, (different volumes); 1997-98 and 1998-99 from Planning Commission (GOI), April, 1999; for Kerala, KSEB, Power System Statistics; last two columns, from CEA, Public Electricity Supply, General Review, 1996-97.

22

TABLE 6: Growth of Energy Sales Share of Energy Purchased Share of Own Energy Sold in ACGR of ACGR of ACGR of in Energy Sold (%) Energy Generated (%) Sales (%) Purchase Own Energy Energy Sold (MU) (%) Sold (%) 1970-71 1980-81 1997-98 1970-71 1980-81 1997-98 1970-71 1980-81 1997-98 1970-98 1970-98 1970-98 Andhra Pradesh 2195 5223 26831 42.96 12.81 41.95 42.63 62.22 56.49 9.69 Assam 298 640 2204 0 259.22 78.81 46.61 -219.14 43.48 7.69 Bihar 1886 4103 7510 65.16 72.24 108.69 47.89 49.93 -17.80 5.25 Delhi 1157 2308 8495 83.41 95.84 147.51 18.70 7.31 -160.86 7.66 Gujarat 3326 7573 29515 34.67 12.43 48.23 52.04 70.83 60.87 8.42 Haryana 939 2793 9050 132.27 99.21 70.06 -16.40 0.51 71.81 8.75 Himachal Pradesh 165 433 2897 77.58 66.97 69.52 59.68 58.37 69.31 11.20 Jammu & Kashmir 170 515 2897 31.18 43.69 156.45 69.64 37.76 -98.16 10.41 Karnataka 4052 5575 16814 1.78 19.52 115.87 83.72 70.20 -15.66 5.41 Kerala 1859 4499 7716 1.13 0.97 54.91 86.44 85.00 67.05 5.41 Madhya Pradesh 2328 4583 24234 50.00 44.25 52.88 42.27 42.93 56.74 9.06 Maharashtra 7660 14399 46260 16.41 6.59 32.33 70.10 76.14 75.50 6.89 Meghalaya 162 318 547 0 0 16.47 89.50 90.08 76.53 4.61 Orissa 1617 2624 6684 16.64 16.39 163.93 76.33 69.94 -74.59 5.40 Punjab 2192 5197 18342 120.48 85.59 31.76 -18.99 11.55 71.90 8.19 Rajasthan 982 3002 15053 111.71 135.11 64.47 -7.62 -31.06 69.51 10.64 Tamil Nadu 5261 8881 26569 20.47 44.07 38.62 74.21 67.38 70.66 6.18 Uttar Pradesh 4509 8142 28829 13.13 4.20 49.60 68.42 76.55 61.33 7.11 West Bengal 4228 5747 9790 23.18 28.87 89.02 80.08 73.49 10.20 3.16 All India 43724 82409 293479 7.04 Note: * = with respect to 1984-85; # = with respect to 1975-76 ACGR = Annual Average Compound Growth Rate; Own Energy Sold = Sales – Purchase. Source: As in Table 1.

9.60 12.67# 7.26 9.96 9.76 6.22 10.75 17.21 23.05 21.29 9.29 9.61 24.91* 14.72 2.97 8.41 8.71 12.52 8.43

9.77 1.68 (-) (-) 7.49 (+) 12.47 (-) (-) 2.39 8.83 6.05 3.92 (-) (+) (+) 5.17 4.97 -4.01

23

TABLE 7: Sector-wise Consumption of Electricity (Per Cent) Households Commercial Agriculture Industry 1970-71 1998-99 ACGR 1970-71 1998-99 ACGR 1970-71 1998-99 ACGR 1970-71 1998-99 (%) (%) (%) 8.16 19.00 3.06 9.94 5.00 -2.42 18.61 34.0 2.18 50.69 33.67 13.31 24.01 2.13 17.75 9.15 -2.34 .. 2.5 55.97 28.32 4.83 12.73 3.52 5.09 5.07 -0.01 3.66 18.5 5.96 74.71 53.88 28.26 36.51 0.92 21.0 5.63 -4.59 0.59 1.0 1.90 39.06 37.39 NA 23.60 NA 7.59 NA 1.2 NA 58.95 7.31 8.82 0.67 3.25 2.46 -0.99 12.19 39.4 4.28 72.28 38.33 0.62 19.76 13.16 5.61 3.24 -1.94 33.56 43.5 0.93 53.76 22.1.0 16.95 16.27 -0.15 8.47 4.55 -2.19 0.85 0.5 -1.88 16.95 40.92 37.65 24.04 -1.59 8.82 4.98 -2.02 5.88 17.6 3.99 35.29 13.07 7.74 21.18 3.66 3.16 2.71 -0.55 6.05 48.0 7.68 76.05 21.56 4.34 48.95 9.04 3.53 8.45 3.17 2.20 4.42 2.52 65.93 32.59 7.86 18.54 3.11 4.14 2.95 -1.20 3.45 37.5 8.89 78.12 27.07 9.57 11.58 0.68 7.15 2.95 -3.11 4.67 32.5 7.17 69.44 36.62 NA 50.15 NA 6.53 NA 3.0 NA 15.43 NA 18.41 NA 9.30 NA 0.4 NA 13.95 2.31 23.03 8.56 3.56 5.25 1.40 0.69 3.5 5.97 88.18 56.89 6.52 18.23 3.74 5.10 3.86 -0.99 21.93 32.8 1.45 65.50 40.53 7.80 15.24 2.42 6.43 4.90 -0.97 11.91 31.3 3.51 64.91 37.99 6.53 16.11 3.28 7.99 6.96 -0.49 24.78 27.0 0.31 56.96 40.90 NA 33.04 NA 7.02 NA 23.1 NA 13.16 8.63 27.28 4.20 2.94 7.81 3.55 16.83 33.4 2.48 64.01 22.66 15.63 22.25 1.27 5.66 10.11 2.09 0.51 15.7 13.02 80.32 27.03 10.78 18.41 1.93 7.20 4.86 -1.39 9.90 30.0 4.04 61.56 33.65

ACGR (%) -1.45 -2.40 -1.16 -0.16

Andhra Pradesh Assam Bihar Delhi (DESU) Goa Gujarat Haryana Himachal Pradesh Jammu & Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Manipur Meghalaya Orissa Punjab Rajasthan Tamil Nadu Tripura Uttar Pradesh West Bengal All India Note : ACGR = Annual Average Compound Growth Rate (%). Sources: For 1970-71, CMIE, Energy, March - April, 1999; for 1998-99, Planning Commission, (GOI), April 1999; Kerala, KSEB, Power System Statistics

-2.24 -3.12 3.20 -3.49 -4.40 -2.48 -3.71 -2.26

-1.55 -1.70 -1.89 -1.18 -3.64 -3.81 -2.13

For

24

TABLE 8: Percentage of Villages and Households Electrified, Pumpsets Energized and Number of Consumers % Villages Electrified (March,1998)

% Pumpsets Energized (March,1998)

No. of Consumers (1996-97) (Million)

Connected Load (1996-97) ('000 MW)

Percentage Households having Electricity Facility (1991 Census) Rural Urban Total 37.5 73.31 46.30 33.88 80.96 40.85 12.44 63.21 18.74 5.57 58.77 12.57 81.82 88.77 84.69 56.43 82.96 65.93 63.2 89.13 70.35 85.86 96.24 87.01 .. .. .. 41.75 76.27 52.47 41.95 67.65 48.43 34.49 72.52 43.30 58.45 86.07 69.40 41.73 75.45 50.92 16.34 83.04 29.16 35.47 85.50 59.20 47.16 75.58 53.42 17.45 62.11 23.54 76.98 94.60 82.31 22.44 76.67 35.03 57.12 92.37 60.66 44.49 76.80 54.74 28.50 80.43 36.93 10.96 67.76 21.91 17.75 70.19 32.90

Andhra Pradesh 97.03 114.04 9.87 17.25 Arunachal Pradesh 66.35 0.08 0.05 Assam 86.47 1.84 0.71 1.70 Bihar 70.82 27.03 1.97 5.74 Goa 93.26 0.33 8.73 Gujarat 99.02 88.21 7.64 16.02 Haryana 100.21 95.21 3.29 6.63 Himachal Pradesh 98.98 50.98 1.23 1.83 Jammu & Kashmir 97.50 37.47 0.68 0.72 Karnataka 98.65 127.31 7.54 11.57 Kerala 100.00 109.79 4.92 6.09 Madhya Pradesh 95.24 94.54 7.96 10.43 Maharashtra 102.69 119.51 13.13 27.3 Manipur 95.33 0.45 0.13 0.12 Meghalaya 50.88 0.65 0.11 0.23 Mizoram 94.87 0.00 0.08 0.09 Nagaland 103.69 1.76 0.12 0.54 Orissa 72.23 14.41 1.23 2.98 Punjab 100.70 105.02 4.67 11.37 Rajasthan 97.95 94.18 4.61 9.26 Sikkim 92.05 0.00 0.05 0.04 Tamil Nadu 99.94 107.28 11.19 19.67 Tripura 100.00 17.64 0.14 0.18 Uttar Pradesh 78.11 32.92 6.48 13.99 West Bengal 77.11 20.88 4.06 7.95 All States 86.65 81.75 . .. U. T.s 97.06 73.74 . .. All India 86.67 81.72 94.94 187.05 30.54 Sources: Planning Commission, Annual Report on the Working of SEBs & EDs, Various Issues; and 1991 Census.

75.78

42.37

25

TABLE 9: Some of the Performance Indicators

Employees per MU of Employees per Thousand Energy Sold Consumers 1992-93 1996-97 ACGR (%) 1992-93 1996-97 ACGR (%) Andhra Pradesh 3.7 3.3 -2.82 9.0 7.7 -3.83 Arunachal Pradesh 71.4 41.4 -12.74 50.0 45.6 -2.28 Assam 14.6 10.3 -8.35 38.8 28.4 -7.50 Bihar 7.6 5.5 -7.77 26.0 19.2 -7.30 Delhi (DESU) 3.3 3.5 1.48 14.2 12.6* - 2.94* Goa 6.0 4.6 -6.43 12.7 11.6 -2.24 Gujarat 2.5 1.9 -6.63 8.3 7.9 -1.23 Haryana 5.2 5.3 0.48 15.8 14.6 -1.96 Himachal Pradesh 6.8 5.2 -6.49 11.0 11.5 1.12 Jammu & Kashmir 10.8 9.2 -3.93 26.1 26.8 0.66 Karnataka 4.1 2.9 -8.29 7.0 5.9 -4.18 Kerala 4.1 3.8 -1.88 6.1 5.5 -2.56 Madhya Pradesh 4.9 3.7 -6.78 13.8 11.9 -3.64 Maharashtra 3.5 2.6 -7.16 12.0 9.7 -5.18 Manipur 28.9 20.1 -8.68 52.0 44.3 -3.93 Meghalaya 10.8 9.5 -3.16 51.1 39.3 -6.35 Mizoram 17.2 9.0 -14.95 21.0 16.1 -6.43 Nagaland 38.4 29.0 -6.78 40.8 33.5 -4.81 Orissa 6.1 5.5 -2.56 30.0 23.5 -5.92 Pondicherry 2.2 2.1 -1.16 11.0 10.0 -2.35 Punjab 5.0 4.1 -4.84 17.9 15.8 -3.07 Rajasthan 5.3 4.0 -6.79 16.1 12.4 -6.32 Sikkim 24.3 15.7 -10.35 41.0 27.0 -9.92 Tamil Nadu 5.0 3.5 -8.53 10.4 8.2 -5.77 Tripura 30.4 18.8 -11.32 46.0 34.8 -6.74 Uttar Pradesh 4.4 3.5 -5.56 18.7 14.7 -5.84 West Bengal 6.7 3.9 -12.65 22.3 14.7 -9.89 All India 4.6 3.6 -5.94 13.3 11.2 -4.21 Note: * = for (with respect to) 1995-96. Source: Planning Commission, Annual Report on the Working of SEBs & EDs, Various Issues.

26

Notes

1

The Ninth Plan envisages a prioritization of hydro-electric power development, targeting an addition of 9,820 MW of hydro capacity, in order to rectify the prevailing imbalance in the hydro-thermal mix. It should be noted that inadequate hydel support in the Western and Eastern regions adversely affects the performance of the thermal plants, as they are uneconomically used to provide only the peaking power, thus having to be backed down during the off-peak hours. 2

There can also be some partial outages due to internal constraints of the deficiency in achieving full rating of the units either in equipment or in auxiliaries, and also/or due to external constraints such as shortage of fuel and coolant or absence of adequate power evacuating capacity. 3

It is estimated that there are about 117 thermal units of 11,000 MW (out of a total thermal capacity of about 59,000 MW) that have already completed more than 20 years of their useful design life (of 25 years); about 50 per cent of these stations operate at less than 45 per cent PLF. Similarly, there are about 35 hydro power stations that have been in operation for over 30 years in excess of their useful operating life (Government of India 2000: 19-20). 4

The fluidized bed boiler design of these larger plants provides much higher efficiency of combustion than the conventional manual or stroker firing, thus reducing the quantity of fuel required. Moreover, it maintains a low fuel bed temperature preventing the formation of lumps of molten ash, a regular problem with the combustion of Indian coal of high ash content. Note that in this light the increased PLF may be taken as not so much of better performance as of partial adoption of a technological progress 5

Acceptance and adoption of PLF as a general criterion of plant performance efficiency can have adverse effects in certain circumstances, as when units are to back down for want of adequate system load. In fact, the practice of linking employee bonus schemes to the PLF attained by the corresponding plants is identified as one of the factors contributing to aggravating grid indiscipline. It should be noted that availability is the internationally accepted measure of plant efficiency. 6

Both the invariable location-specificity of the hydro-power plants and the economies in developing pithead-based thermal stations (as transportation of coal has been found to be less economical than transmission of an equivalent amount of electric power over long distances) necessitate extensive network of transmission lines. Also there are economies in interconnecting different power stations as well as systems in an electric Grid. 7

The Indian Electricity Act, 1910 was amended (in 1986) through Sections 39 and 39A to make theft of energy and its abetment a cognizable offence with deterrent punishment of up to three years imprisonment. 8

That the agricultural consumption of power in India is highly doctored, due to its being a ‘residual’ in estimation, is now a widely acknowledged fact; in the States, where the power sector has been restructured, the regulators, in recognition of this ‘misclassification’, have revised upwards the T & D loss percentages – for example, in Orissa, from 23 per cent before restructuring to 51 per cent post-reform; in Andhra Pradesh, from 25 per cent to 45 per cent, in Haryana, from 32 per cent to 40 per cent, and in Rajasthan, from 26 per cent to 43 per cent (Government of India 2000: 35; also see Morris 2000, and Rao 2000).

1

CHAPTER 3

FINANCIAL PERFORMANCE OF THE SEBs

“Light is sorrow, son, Darkness, pleasant one.” - Akkitham (Malayalam poet) 1. INTRODUCTION In the last chapter, physical performance o the state power sector was evaluated, and the inadequacy and inefficiency involved were brought out, in relation to their possible causatives. The next section of this chapter analyses the cost structure of electricity supply in India; and cost savings realizable from some reasonable improvements in efficiency at certain accessible levels of technoeconomic performance are also estimated. After a brief discussion in Section 3 on the tariff structure, we take up in Section 4 an appraisal of the financial performance of the State power sector and light up the likely implications involved in inefficiency. The final part concludes the discussion with a rather cynical note on the power sector reforms vis-à-vis the form-substance dialectics. 2. COST ANALYSIS

All these inefficiencies must come out in inflated proportion in the cost of electricity supply. For all the SEBs in India, the unit cost of supply of electricity in 1974-75 was 22.5 paise per unit, which increased to 41.9 paise per unit in 1980-81 (at an annual average compound growth rate of 10.9 per cent), and further to 108.6 paise per unit in 1990-91 (at an annual rate of 10 per cent). The nineties saw sharp rise in the unit cost of supply, from 116.8 paise/unit in 1991-92 to 227.89 paise/unit in 1997-98 (at a rate of 11.8 per cent p. a.). It is expected to reach Rs. 2.43/unit in 1998-99 (an increase of 6.6 per cent). In 1997-98, the unit cost varied from Rs. 1.60/unit in Himachal Pradesh to Rs. 4.23/unit in Assam. Two important factors that cause such wide variation in unit supply cost in general are (i) the source of power, whether hydro or thermal, and (ii) the coverage of electrification of villages and households. The pure hydro-power systems of Himachal Pradesh and Meghalaya have lower unit cost; however, the higher unit cost of the pure thermal systems of Delhi and Assam is not due to higher fuel cost, but due to higher power purchase cost. Though Kerala is still a hydro-power dominant system, her unit cost of supply (Rs. 1.92/unit) exceeds that of

2

Karnataka (Rs. 1.89/unit), now a thermal-power dominant (72 per cent) system, on account of the increased share of imported (thermal) power. If we take into account this aspect also, i.e., the sources of total energy sold out, Kerala power system would become a predominantly thermal (about 80 per cent) one. During the seventies, the average cost of electricity supply in Kerala had an annual average compound growth rate of 3.8 per cent, during the eighties, 11.8 per cent, and during 1991-92 to 1997-98, 15.4 per cent, reflecting largely the increasing impact of power purchase cost.

The major components of electricity supply cost are (i) the revenue expenditure, consisting of expenditure on fuel, power purchase, operation and maintenance (O & M), establishment and administration (E & A), and on other miscellanies; and (ii) the fixed costs, including depreciation and interest payable to institutional creditors and to the concerned State Governments.

Fuel Cost

Fuel cost has accounted for about 25 per cent of the total supply cost since 1992-93. Gujarat, Tamil Nadu, Maharashtra, and Punjab have had fuel cost share higher than the all-India average, while the pure thermal systems of Assam and Delhi, much lower (Table 1). Fuel cost depends, besides other factors, on the specific consumption of coal and oil, and the transportation costs of these fuels. The specific coal consumption of the thermal plants of the SEBs has been about 0.74 to 0.78 kg/unit since 1992-93. A number of SEBs, including that of Andhra Pradesh, Bihar, Haryana, Madhya Pradesh, Maharashtra, Orissa, and Uttar Pradesh, have had consistently higher than 0.8 kg/unit of coal consumption during this period. The specific secondary oil consumption in the coalbased thermal plants increased steeply from 7.8 ml/unit in 1992-93 to 10.8 ml/unit in 1995-96, and then dropped to reach 9.9 ml/unit in 1997-98. In the late seventies and the early eighties, it was over 12 ml/unit. The average specific oil consumption in Bihar, Haryana, and Assam has been higher than the all-India average, while that in Andhra Pradesh, Tamil Nadu, Maharashtra, Rajasthan, Madhya Pradesh (recently) and Punjab, much lower. The cost of coal per unit of electricity generation increased from 53.4 paise/unit in 1992-93 to 89.4 paise/unit in 1997-98 (at an annual growth rate of 10.9 per cent), and that of secondary oil from 3.7 paise/unit to 7.3 paise/unit (at an annual rate of 14.6 per cent) during this period (Government of India, 1999: Annexures 4.9 – 4.12). The States, viz., Gujarat, Tamil Nadu, Haryana, Punjab and Rajasthan, located farther away from coal fields have to bear higher cost of coal per unit of generation, thus having higher share of fuel component in their unit cost.

3

Much remains to be desired and improved in the aspect of overall thermal efficiency of the steam power plants also. 25 out of the 77 steam power stations considered in the country, accounting for 19.4 per cent of their total IC (of 50,115.48 MW) in 1996-97, reported an overall thermal efficiency below 25 per cent, and an average capacity utilization of 3,975 kWh/kW or 45.4 per cent. 19 steam stations, representing 24.2 per cent of the IC, had an overall thermal efficiency in the range of 25 to 30 per cent, and an average capacity utilization of 4,377.6 kWh/kW or about 50 per cent; and the remaining 33, with 56.4 per cent of IC, had, above 30 per cent overall thermal efficiency, and an average capacity utilization of 6,164.4 kWh/kW or 70.4 per cent. The average utilization rate for all the 77 stations was 5,307.6 kWh/kW or 60.6 per cent (Government of India, 1996-97: Table No. 45). Thus about 57 per cent of these steam stations, accounting for about 44 per cent of their total IC, were utilized for less than 50 per cent of the time, and all the stations together, about 60 per cent.

Costs of Power Purchase

Expenditure on power purchase is the largest component of the total cost of electricity supply. It increased from 27.9 per cent of the unit cost of supply of electricity in 1992-93 to 36.2 per cent in 1997-98 (at an annual growth rate of 5.3 per cent). Bihar, Delhi, Haryana, Jammu and Kashmir, Madhya Pradesh, Karnataka, Orissa and West Bengal have much higher proportion of power import cost than the all-India average; as much as 74 per cent of the unit cost of supply in Delhi was accounted for by power purchase in 1997-98, and nearly 70 per cent in Orissa. The average rate of payment for power purchase steadily increased from 76 paise/unit in 1992-93 to Rs. 1.39/unit in 1997-98 (at an annual rate of 12.7 per cent). In 1997-98, the total cost of power purchased by all the SEBs and Electricity Departments (EDs) was Rs. 24,187.4 crores. A good part of this huge cost in fact represents the price paid for the inefficiency in the T & D system. We can have an estimate of this inefficiency that stands to inflate the unit cost of electricity supply. The net generation of electricity by all the SEBs and EDs in 1997-98 is estimated to be 2,28,020.3 MU. If we assume that the T & D loss could be kept at a minimum of 15 per cent, then the energy that must be available for a sale of 2,93,478.9 MU in that year would be 3,45,269.3 MU, thus necessitating an import of 1,17,248.99 MU (about 40 per cent of the total sales) only, instead of the reported 1,74,373.9 MU (about 60 per cent of the sales), giving a saving in power purchase of 57,124.9 MU, or in power purchase cost of Rs. 7,924 crores at an average power purchase rate of Rs. 1.39/unit, or a saving of 27 paise per unit sold1. This would reduce the unit cost of electricity supply to Rs. 2.01

4

per unit sold, against the reported Rs. 2.28/unit. Thus the cost of inefficiency in the T & D system alone comes out to be about 27 paise per unit of electricity sold ! In the case of Kerala, this is 5.16 paise per unit sold, and the unit cost of supply would then be only Rs. 1.86/unit instead of the given Rs. 1.92/unit. Remember, Kerala reported a T & D loss of 17.87 per cent only in 1997-98. On the other hand, for Delhi, the cost of inefficiency comes to 94.36 paise/unit sold, and the unit cost of supply, Rs. 2.57/unit, instead of 3.51/unit !

The burden of power purchase could still be lessened if the SEBs and EDs were able to improve their operational efficiency and thus increase their net generation. The above analysis was based on the actual figures on an average of a PLF of about 50 per cent (i.e., a utilization of 71.3 per cent at 70 per cent availability), and about 7 per cent of auxiliary consumption for all the SEBs and EDs. At 80 per cent availability, the thermal power generation in 1997-98 in the State sector implies a utilization of 69.4 per cent (and hence a PLF of 55.5 per cent); and at an (assumed) availability (dependable firm power) of 60 per cent, the hydro power generation implies a utilization of 61.45 per cent (a PLF of about 37 per cent). Now, it would be only reasonable to assume a PLF of 70 per cent (that may imply a utilization of 87.5 per cent at 80 per cent availability) for the thermal plants in the State’s sector2. Similarly, let the hydro power stations have a PLF of 47.5 per cent (that may imply a utilization of nearly 80 per cent at 60 per cent availability). This increased operational efficiency would reduce the power purchase (of 1997-98) by 1,05,186.4 MU, assuming 7 per cent auxiliary and 15 per cent T & D consumption. This represents a saving in power purchase cost of Rs. 14,590.4 crores or 49.72 paise per unit sold, and the unit cost of electricity supply would be only Rs. 1.78 per unit ! For Delhi, such operational efficiency improvement would reduce the unit supply cost by as much as 116.91 paise per unit sold to Rs. 2.34/unit, and for Kerala, by 40.32 paise per unit to Rs. 1.52/unit, assuming 0.61 per cent of auxiliary consumption as reported !

Needless to repeat, adequate and timely capacity additions could further improve the situation. If, for example, Kerala could achieve her targets of commissioning of power plants as anticipated during the 7th, 8th and 9th Plans (as detailed in Government of Kerala 1984: Statement 2), she could still continue to enjoy being a net exporter, rather than be, as at present, one of the States worst affected by power shortage. For one thing, on the surface, the failure ‘was apparently attributable to the complacency created out of the comfortable power position prevailing in Kerala until the recent failure of monsoon and the consequent power cut’ (Government of Kerala 1984: 26). Deep-rooted, however, a number of factors have in accumulation wreaked havoc on the system. Some ‘classical’ examples of project time-overruns may rightly be credited to Kerala – Kallada (15

5

MW), Kakkad (50 MW), and Lower Periyar (180 MW), as also some minor projects (all hydro power projects), to have been commissioned during the 7th Plan, could finally be put on line in the mid-90s only. The time overrun (over and above the originally scheduled commissioning date, once the works started) in the case of Idamalayar project was 9 years, Kakkad, 13 years, Kallada, 5 years, Lower Periyar, 6 years, and the mini projects, Peppara, 6 years, and Madupatty, 9 years. The consequent cost overrun was Idamalayar: 285 per cent, Kakkad: 725 per cent, Kallada: 53 per cent, Lower periyar: 299 per cent, Peppara: 74 per cent, and Madupatty: 64 per cent (Government of Kerala, Economic Review, different issues). Ideally, a revised cost estimate should sufficiently cover the general price rise. Then what remains in the revised cost escalation of a project over and above the general price inflationary influences is a matter of serious consideration; it may represent an over-estimation due to uncertainty or an element of deliberate attempt at wasteful mismanagement of resources. In the case of most of the projects with time overruns in Kerala, the revised cost estimates significantly exceeded the general inflationary impact, signifying the effect of some ‘unexplained’ factors (of deliberate or otherwise mismanagement) on cost escalation. For example, for Kakkad, about 264 per cent of the cost escalation remained to be explained by factors other than general inflation, and for Lower Periyar, about 105 per cent.

The time overruns of power projects involves manifold and thus heavy costs – besides incurring the cost escalation of the projects and the power purchase costs, the system also is forced to forgo additional sales revenue obtainable. Thus the cost of inefficiency at the planning and execution level also is very high. “The basic reason for the power crisis engulfing the State (Kerala) today is mainly …..the failure, of the Electricity Board, in planning and in the timely execution of the power projects.” (Government of Kerala 1997: 9) A host of factors are at work here – changes in the technical design and feasibility report, original cost estimates being based on inadequate or incomplete data and unrealistic assumptions, inefficient management, inadequate geological and technical investigations of the projects in their initial stages, vague and ambiguous specifications and conditions of contract, delays due to sluggish decision making at various stages of construction, lack of availability of materials or of transportation facilities, high mobility of planning and supervisory staff between projects during their construction, militant trade union interference, excessive ecological concerns, unwarranted court interventions for aggrieved contractors, and above all, vitiating corruption, and indifference of the public.

There is yet another factor. Power purchase agreements (PPA) often contain booby traps of forced purchase provisions3; in order to respect the PPA, the SEBs are sometimes compelled to back

6

down their own cheap generators. An apt case in point here is the PPA between the NTPC and the KSEB in respect of the power from the Kayamkulam thermal station. It being completely a ‘State project’, the KSEB should, by PPA, purchase all the energy generated here. Though the average capacity utilization factor is set at the usual norm of 68.5 per cent, the Kayamkulam project is often operated at the full 100 per cent capacity factor, and the Board is thus forced to take in the whole lot. This in turn results in backing down some of the hydro power plants, with the cheapest generation cost of only 14 paise/unit4. The Chief Engineer (Thermal, O & M) of the KSEB has estimated that the Board could save Rs. 250.56 crores every year, if it needed to purchase only the normal generation (at 68.5 per cent capacity factor) from the NTPC project (Malayala Manorama daily, June 14, 1999).

The Board, on the other hand, seeks to save its face by scheduling these hydro power plants for repair, maintenance or renovation works, even during the monsoon, when water, if not fully utilized, would spill over from the small reservoirs of these plants. For example, (according to an estimate of the Board), during the monsoon of 1998, water worth Rs. 3.6 crores of energy was lost from the Peringalkuthu reservoir, as one unit of 8 MW capacity there had been in outage for more than 1½ years, though it required only some minor repairs. Similarly, in Sengulam power station, one unit had been in outage for more than one year, and water worth Rs. 12 crores of energy was lost. Neriamangalam station also had the same fate in that year (Mathrbhoomi daily, June 21, 1999).

At present, the daily electricity consumption in Kerala is about 33 MU, and on Sundays it goes up to 35 MU. The generators closed down in the name of maintenance or repair works cost about 4 MU per day, that is about Rs. 80 lakhs a day. (The loss would be even higher if estimated at the rate of the power purchase, which could have been averted, rather than at the selling rate.) At Idamalayar (2 x 37.5 MW), one of the two generators had been in outage since February 2000, requiring only some minor repairs. The reservoir of the power station had, by mid-April, water sufficient for 105 MU of energy; and the only unit in service, if run at its maximum capacity during April and May, can consume water equivalent to only 50 MU, which in turn means spilling over of substantial quantum of water during the imminent monsoon. Kuttiady power station of 75 MW capacity has been out of service for a long time in the name of extension works for augmenting capacity by another 50 MW. During the monsoon Kuttiady is operated at its maximum, often beyond its capacity in order to utilize fully the monsoon bounty, which otherwise would spill over. The unutilized water, equivalent to 6 MU of energy (a revenue of about Rs. 12 million at the sales rate), that the small dam contains at present, would go to waste during the monsoon spill over

7

(Mathrbhoomi daily, April1 6, 2000). The turnkey project work, started in 1996, should have been completed in 3 years. The Government has now allowed six more months for completion (sanctioning the demanded cost overruns to the contractor, a controversial Canadian firm), which means substantial loss of water during this monsoon also. The KSEB has estimated a daily loss of Rs. 36 lakhs, and a total loss of more than Rs. 55 crores during the six months extension period (Malayala Manorama, April 26, 2000).

In addition to letting some of the cheap hydro plants remain closed down in the name of maintenance or repairs, KSEB also operates its own thermal plants at minimum capacity factor. One unit of the Brahmapuram thermal plant also is in outage and the other units are run below 20 per cent of the installed capacity, while the Kozhikode thermal plant units are operated below 30 per cent. The operating costs of these plants are reported to be very high, much more than Rs. 5/unit. Though these stations are supposed to use low sulfur heavy stock (LSHS) as fuels, at present they are run on naphtha, like the Kayamkulam (NTPC) and Cochin (BSES) plants. It should be noted that nowhere in the world is naphtha, which is highly volatile and very costly, used for power generation. It was a reckless and disastrous choice for Kerala to have all her thermal plants, both commissioned and under construction or consideration, based on naphtha. And way back in 1996, the Planning Commission opposed the use of naphtha as feedstock for new power projects, considering the high generation cost per unit and the huge foreign exchange outflow that naphtha imports might entail. Kozhikode thermal plant is confronted with a threat of environmental problems also; it has not yet obtained the green clearance from the Pollution Control Board. If the Board objects, the plant will have to be closed down ! The KSEB purchases about 6.5 MU of electricity every day from the NTPC Kayamkulam thermal station at an exorbitant cost of Rs. 4.5/unit (the generation cost is reported to be Rs. 5.5/unit), while its hydro-power costs it only about 20 paise per unit (Mathrbhoomi daily, April 16, 2000). Though the NTPC is willing to convert this ‘State project’ into a regional one, in the event of which the purchase price would come down (as its PLF increases), the State Government is dragging its feet in view of the prospects of getting liquefied natural gas (LNG) in case the proposed LNG terminal at Kochi is implemented soon. Power from the Nuclear Power Corporation costs only Rs. 1.89/unit for Kerala, while that from the Neyveli Lignite Corporation costs Rs. 1.04/unit (First phase) and Rs. 1.87/unit (Second phase); Kerala also gets power from the NTPC Ramagundam project at Rs. 1.57/unit only (The New Indian Express, March 22, 2000). Thus there do remain quite feasible possibilities for adopting fruitful means to reduce the power purchase cost to a substantial extent – by converting temporarily the NTPC ‘State project’ into a regional one

8

till the LNG terminal at Kochi is materialised, and at the same time by ensuring an enhanced share of power from the Central pool and its regular and constant delivery.

O & M and E & A Costs

The proportion of O & M costs in the unit electricity supply cost had a marginal decrease from 4.7 per cent in 1992-93 to 4.5 per cent in 1997-98. In general, the hydro-power systems of Himachal Pradesh and Meghalaya have much higher (above 10 per cent) share of O & M costs, while Assam, Gujarat, Jammu & Kashmir, Punjab, and Tamil Nadu, much lower (about 2 to 3 per cent).

Establishment and administration (E & A) charges consist mainly of the wages and salaries of staff. Its share in unit supply cost declined from 15.2 per cent in 1992-93 to about 12.3 per cent in 1997-98. Himachal Pradesh, Kerala, and Meghalaya have very high share of E & A costs, often more than 30 per cent, while Andhra Pradesh, Delhi, Gujarat, Jammu & Kashmir, Rajasthan, Maharashtra and Uttar Pradesh show much lower share than the all-SEBs average. As already explained, over-manning especially in non-technical sections on account of the employmentproviding patronage of the Governments has been another source of inefficiency. The labor productivity in the States’ power sector in 1997-98 is estimated to be about 3.4 employees per MU of electricity sold, as against less than 2.5 in many developing countries5. We can have an estimate of this inefficiency too. The total E & A charges in 1997-98 come out to be Rs. 8,009 crores, at a unit E & A cost of 27.29 paise/unit sold, for an estimated number of 9,86,537 employees, giving an average E & A expense of Rs. 81.18 thousand per employee per year. If labor productivity increases to, say, 2 employees per MU of electricity sold, (i.e., with the given quantum of sale), then number of employees would be reduced to 5,86,958, and the E & A costs, to Rs. 4,765.1 crores. This gives a unit E & A charge of power supply of 16.24 paise per unit sold, and a unit cost saving of 11.05 paise/unit, which is the cost of inefficiency involved in over-employment. For Kerala, with an average E & A cost of Rs. 1.44 lakhs per employee per year in 1997-98 (about 1.8 times the State sector average), the cost saving is turned out to be 19.78 paise/unit sold, reducing the unit E & A cost to 28.88 paise/unit from 48.66 paise/unit. For Delhi, it is 10.46 paise/unit, a reduction from 31.33 paise/unit to 20.87 paise/unit.

9

Fixed Costs

The share of fixed costs, viz., depreciation and interest payments in average cost of electricity supply declined from 25 per cent in 1992-93 to 21.7 per cent in 1997-98. Interest charges have always commanded a bigger share out of this – much more than 10 per cent. While the share of depreciation rose from 7.6 per cent in 1992-93 to 9.2 per cent in 1994-95 and then fell to 8.3 per cent in 1997-98, that of interest steadily declined from 17.5 per cent in 1992-93 to 13.5 per cent in 1997-98. Kerala, Madhya Pradesh, Maharashtra, and Uttar Pradesh have much higher stakes in depreciation, around 10 per cent; Delhi and Himachal Pradesh, on the other hand, the least, less than 5 per cent. The share of depreciation in unit cost of Kerala remains around 6 per cent.

Very high interest charges are a big problem for many States – Assam, Himachal Pradesh, Kerala, Meghalaya, and Uttar Pradesh have higher share of interest in supply cost, more than 20 per cent. Delhi maintains the lowest share position here also – nearly 4 per cent only. Note that depreciation is an important item contributing to internal resources generated, while interest charges are a real drain, and hence increased share of the latter in total cost signals financial weakness of dependence. In fact, the share of interest in supply cost could be significantly reduced in a number of potent ways. SEBs in general do not repay the State Government loans and the interest thereon. These interest charges due to State Government are usually carried forward every year, and the accumulated charges stand to make the balance sheet dip totally into the red. The situation could be eased by converting part of Government loans into equity. Originally, the Electricity (Supply) Act did not provide for an equity component, and the entire capital of SEB consisted only of loans from the State government and from institutional lenders. In general, a debt-equity ratio of 1:1 is maintained in all capital-intensive industries, including the Central power sector. Hence in 1978, the E(S) Act was amended to enable the State Governments to provide for equity by converting part of their loans into equity. However, SEBs in general are reluctant to take up this provision seriously, lest the Board’s profits, likely to be exhibited consequent upon the introduction of equity capital, sould be liable to income tax. Yet this inhibition condones letting the unpaid/unpayable interest charges inflate the supply cost. To the extent that this part remains unpaid, the supply cost thus calculated turns out to be an over-estimate. Recently, the Kerala Government has decided to convert Rs. 1,552 crores due to it on account of accumulated loan and interest, projected to reach Rs. 2,280 crores by 1998-99, into equity capital, stipulating that the Board, like the independent power producers (IPPs), earn a return of 16 per cent on capital employed. However, the Board still continues its practice of carrying forward the interest charges on Government loans and including

10

the annual interest charges in total expenditure, without allowing for any reduction possible on account of equity introduction. Ideally, a 1:1 debt-equity ratio accounting practice would reduce the interest charges by one-half, such that, for instance, in 1997-98, the unit interest cost in the State sector would be reduced to 14.85 paise per unit sold, and the overall unit supply cost, to Rs. 2.13/unit sold. For Kerala, the benefit of reduction in unit interest cost would be 23.13 paise/unit sold6.

The Cost of Inefficiencies

That the cost of electricity supply in the State sector is an over-estimate inflated by inefficiencies at all points is a foregone conclusion. Allowing for some improvement in operational, T & D, and man-power planning efficiencies, as discussed earlier, would reduce the unit cost of supply of all-SEBs substantially, by 60.77 paise per unit sold, to Rs. 1.67/unit from Rs. 2.28/unit in 1997-98. For Kerala, the unit cost saving is 60.10 paise/unit, giving a unit supply cost of Rs. 1.32/unit instead of the reported Rs. 1.92/unit, and for Delhi, 127.37 paise/unit, the unit supply cost reducing to Rs. 2.24/unit from Rs. 3.51/unit. With a 1:1 debt-equity capital base, the unit electricity supply cost would still go down for all-SEBs to Rs. 1.52/unit sold, and to Kerala, Rs. 1.09/unit. The unit cost of inefficiency in the State sector is about 33.2 per cent of the reported unit cost of electricity supply, and in Kerala, about 43.3 per cent, and in Delhi, 36.3 per cent. And this is regardless of the unquantifiable cost of inefficiency at all other levels ! Now the pertinent question is: Should the consumer be made to pay for this inefficiency ?

It should, however, be stressed that this conclusion is in the accounting cost sense, and not in the economic, opportunity, cost sense. The latter, for instance, demands that the opportunity cost of land, given virtually free to the SEBs by the State Governments, also be included in the total cost of supply. Moreover, the straight line depreciation method, followed for accounting by the SEBs, can by no means reflect economic depreciation in considerations of the actually required replacement cost. 3. TARIFF AND REVENUE REALIZATION

In general, increasing block rate tariff that penalizes higher consumption levels because of capacity shortage is in practice in India. Hence the average tariff (or more precisely average revenue, AR, as it is reported) at the aggregate level cannot be the price confronting the customer in his

11

decision making options; rather it can be only a supply price to the utility. The average price for sales of electricity by the SEBs was 18.8 paise/unit in 1974-75, 32.3 paise/unit in 1980-81 (growing at an annual average compound rate of 9.45 per cent), which increased to 81.8 paise/unit in 1990-91 (at an annual rate of 9.74 per cent). During the nineties, AR increased steeply from 89.1 paise/unit in 1991-92 to 184.5 paise/unit in 1997-98 (at an annual rate of 12.9 per cent). It is expected to grow further by 7.25 per cent to 197.85 paise/unit in 1998-99. During the seventies and eighties, the growth in AR was slightly less than that in AC, but in the nineties, the former exceeded the latter. Larger inter-State variations mark this trend. In 1980-81, Bihar, Gujarat, Madhya Pradesh, and West Bengal had higher AR (more than 40 paise/unit); and Meghalaya had the lowest, 22.6 paise/unit. In 1990-91, Maharashtra and West Bengal had AR greater than 100 paise/unit, and Delhi, Assam, and Rajasthan, greater than 90 paise/unit; Jammu & Kashmir had the lowest, about 36 paise/unit. In 1997-98, Assam, Bihar, Maharashtra, and Orissa reported AR estimates greater than Rs. 2/unit, and Jammu & Kashmir still maintained the lowest – 39.3 paise/unit. During the seventies, the AR realized from electricity sales in Kerala registered an annual growth rate of 11.6 per cent, during the eighties, about 8 per cent, and during 1991-92 to 1997-98, 13.3 per cent. Since the eighties, growth in AR in Kerala has been lagging behind that in AC.

Though the SEBs are empowered by the E(S) Act to determine prices with the State Governments expected to have only an advisory role, it is the latter that effectively take decisions. The socio-political compulsions of distributional solicitude of the Governments have resulted in significant distortions in setting tariffs for various consumer categories in line with the cost involved in supplying each group. Thus the cost of providing electricity to low voltage (LV) consumers (domestic, agriculture, commercial, etc.) is much higher on account of the additional cost of extensive distribution network, and more importantly, of higher distribution loss of energy, than the high voltage (HV) and extra high voltage (EHV) industries. However, the agricultural and domestic consumers enjoy a privilege of heavily subsidized supply of electricity at the cost of others. The AR realized from these two sectors is significantly lower than the overall AR, while that from the commercial customers, industry and railway traction is much higher. Agricultural consumption is charged at the lowest. In 1997-98, the AR from this sector was 27.7 paise/unit and from the domestic sector, Rs. 1.34/unit, against the overall AR of Rs. 1.85/unit. On the other hand, commercial customers paid on an average Rs. 3.33/unit and the industrial customers, Rs. 2.85/unit. The AR realized from the railway traction was the highest, Rs. 3.75/unit in that year. During 199293 to 1997-98, the overall AR realized grew at an annual average compound rate of 11.9 per cent, while the AR from the industrial sector, at a rate of 10.7 per cent, that from the agricultural sector, at

12

11.5 per cent, domestic sector, 11.6 per cent, railway traction, 12.6 per cent, and commercial sector, 15.1 per cent (Table 2).

Subsidized Power Supply

There are wide inter-State variations in the structure of subsidized supply of electricity. A consensus decision was taken at a conference of State Power Ministers in January 1993 to charge at least 50 paise/unit for agricultural power consumption. The consensus was repeated in 1996 also and a Common Minimum Action Plan for Power was put out in December 1996. This tariff was to rise, within three years, to 50 per cent of the unit cost of generation. But only a few States have implemented the minimum tariff policy – for example, Orissa and Haryana, where the sector has been restructured. Kerala realized about 55 paise/unit of AR from the agricultural sector in 1997-98; in the previous year, it was only 29 paise/unit. Andhra Pradesh, Gujarat, Jammu & Kashmir, Maharashtra, Rajasthan, Uttar Pradesh, and West Bengal provide for agricultural consumption at rates less than 50 paise/unit, and Tamil Nadu and Punjab, virtually free. In Karnataka and Madhya Pradesh also power supply to agriculture is free, but they have specified certain thresholds of connected load (10 HP in Karnataka, and 5 HP in Madhya Pradesh) above which some rates are charged. In Maharashtra, a paradoxically discriminatory tariff structure is meted out to the agriculture sector – (i) metered tariffs for irrigation pumps used in food crops fields that consume relatively much less electricity, and (ii) unmetered flat-rate tariff, based on the horse power, for pumps in water-intensive cash crops fields that consume a lot more electricity !

The power of

‘sugar politics’ overwhelms any economic logic in the allocation and use of such scarce resources as water and power.

Domestic consumers are favoured in Jammu & Kashmir, Himachal Pradesh, Kerala, Madhya Pradesh, Meghalaya, and West Bengal, the lowest rate being in Jammu & Kashmir, 32 paise/unit. All other States charge more than 100 paise/unit for domestic consumption, the AR realized from this sector in Punjab, Gujarat and Haryana being greater than their overall AR. On the other hand, the AR realized from industrial and commercial sectors were in general more than double that from domestic sector and more than 10-times that from agricultural sector in 1997-98. It was so in most of the States also. In Kerala, the proportion of the AR in the domestic, industrial and commercial sectors in 1997-98 was 1 : 2.06 : 3.6, while for all-SEBs, it was 1 : 2.13 : 2.5, for Karnataka, 1 : 2.56 : 4.23, for Tamil Nadu, 1 : 2.2 : 2.76 and for Andhra Pradesh, 1 : 1.98 : 2.2.

13

The inefficiency due to Government interference in price determination7 favouring the agricultural and domestic sectors has much to do with the financial performance of the SEBs. While electricity sales to these two sectors accounted for nearly one-half of the total sales, revenue realized from them was only about one-sixth of the total sales revenue of the SEBs in the recent years.

4. FINANCIAL PERFORMANCE

Commercial Losses

That revenue realized from sales must be sufficient at least to recover costs of supply is the basic prerequisite for the health of any industry. Starting from this premise and comparing revenue realized with cost incurred in the power sector serve the purpose of highlighting the parlous financial position of the SEBs, which in turn is used to justify the clamour and claim for reforms. The revenue-cost ratio went down recently to as low as 76 per cent (in 1995-96), i.e., the sales revenue was enough just to recover 76 per cent of the supply cost. The cost recovery ratio has slightly improved since then. In 1974-75, it was 83.4 per cent, which decreased to 77 per cent in 1980-81 (at an annual average decay rate of 1.3 per cent), and further to 75.3 per cent in 1990-91 (at a rate of 0.23 per cent). It tried to regain in the next two years a little of what it had lost and reached up to 82.2 per cent in 1992-93, but only to climb down in the following years.Among the 19 States considered, Maharashtra has had almost always the highest cost recovery ratio – greater than 90 per cent. In fact, in the early 90s, the ratio was nearly 100 per cent, and in 1997-98, it was estimated to be about 98 per cent. Himachal Pradesh and Tamil Nadu also had a ratio greater than 90 per cent in the recent past. Himachal Pradesh has the unique distinction of being the only State having had a sales-revenue that actually exceeded the cost in one year (1996-97). Assam, Haryana, Jammu & Kashmir, Punjab, and Uttar Pradesh had less than 70 per cent cost recovery ratio in most of the years, and J & K had the lowest, less than 20 per cent (see Table 3).

The cost-revenue deviation or commercial loss (see Table 13) of the SEBs (without subsidy) increased from Rs. 4,560 crores (implying a rate of return (RoR) of (–) 12.7 per cent) in 1992-93 to Rs. 10,684 crores (RoR of (–) 18 per cent) in 1997-98 (at an annual growth rate of 18.6 per cent), and is projected to increase to Rs. 12,323 crores (RoR of (–) 18.7 per cent) in 1998-99 (at a growth rate of 15.3 per cent). The 1983 amendment to the Section 59 of the E(S) Act, 1948, requires the SEBs to ensure that the total revenues in any year of account shall, after meeting all expenses

14

properly chargeable to revenues, including operating, maintenance, and management expenses, depreciation and interest payable, as also taxes, if any, leave such surplus as not less than 3 per cent, or such higher percentages, as the State governments may specify, of the value of fixed assets of the Board in service at the beginning of such year.

Thus the goal of tariff-making has become

predetermined. Yet, a tariff mechanism in line with the basic tenets of tariff-setting still remains to be properly evolved in order to achieve this set goal. At present, the tariff structure includes capacity (demand) and energy charge components for large consumers, and consumption slabs for small consumers. It should also reasonably incorporate the distinct cost elements of fixed capacity costs, variable energy costs and customer-related costs on equipment, metering, billing and collection, in the spirit of Hopkinson rate structure8. Despite the set goal of at least 3 per cent RoR, a marked deterioration has been observed in the trend of the RoR of the SEBs in general. Such commercial loss suggests that if the total revenue earned by the SEBs had been enough to cover the total costs, an additional amount, say, of Rs. 10,684 crores would have been available in 1997-98 for reinvestment in the power sector. That an accumulated amount of Rs. 45,177 crores would have been available with the SEBs during the 6 years from 1992-93 for ploughing back in the sector, had the total cost been recouped, brings out the extent of the colossal loss the SEBs suffer over time. Achieving a minimum 3 per cent RoR would have mobilized additional revenue of Rs. 12,099.2 crores in 1997-98, and a break-even RoR, Rs. 10,487.5 crores. Universal adoption of the minimum 50 paise/unit tariff for agricultural sales would have generated additional resource of Rs. 2,254.2 crores in the same year. Such additional revenue could have comfortably been used for capacity expansion and for improving the performance of the existing assets. This would have also reduced the burden of the State governments’ having to provide the SEBs with subvention. That all these would have been possible every year leaves one sickened and cynical at the morbid sector. Maharashtra’s was the only SEB that earned a profit in 1997-98 (Rs. 111.8 crores) and in the case of other SEBs, the commercial loss ranged from Rs. 18.9 crores for Himachal Pradesh to Rs. 1,735.8 crores for Uttar Pradesh. Gujarat and Punjab also had a loss of more than Rs. 1000 crores, and as many as 7 other SEBs, more than Rs. 500 crores (see Table 4). In the early nineties also Maharashtra reported profit. It should be pointed out that in general, the SEBs carry forward accumulated losses and hence even if a particular year turns out profit, the cumulative reserves may be negative. For example, the Kerala SEB earned net profit in 1989-90 and again continuously during 1992-96, technically reporting the statutory requirement of 3 per cent rate of return. However, in all these years, KSEB suffered cumulative losses of no small magnitude.

15

Though subvention from the State Governments has improved the situation, the RoR has still remained negative, the commercial loss, for example in 1997-98, coming down to Rs. 6,977.8 crores, and the RoR, to (-) 11.7 per cent. Subvention has secured positive RoR of about 3 per cent for Karnataka all these years in the nineties. Gujarat, Himachal Pradesh, Orissa, Rajasthan and Tamil Nadu also reported profit and positive RoR for some of the years. As many as 6 SEBs suffered losses greater than Rs 500 crores in 1997-98 even with the support of subsidy, Punjab leading the list with a loss of Rs. 1,346 crores. Implementing the proposed national minimum agricultural tariff of 50 paise/unit across all the States also would not have saved the SEBs out of the red. The RoR in 1997-98, for instance, would still have been negative, (-) 10.5 per cent.

Losses Due To Subsidized Power Supply

A major factor that determines the level of commercial loss is the differential pricing policy. Loss results if the Government subsidy payments and cross subsidy from other sectors are not enough to neutralize the effective subsidies given to agriculture and domestic consumers. Effective subsidy (cross subsidy) is defined as (AC – ARi) Qi , where AC is the average cost of power supply, ARi is the average revenue realized from the ith sector and Qi is the total power sold to that sector. If this expression is positive, it is taken as a subsidy to the ith sector, and if it is negative, as a cross subsidy from the ith sector. On this basis, it can be seen that the effective subsidy to agriculture in general increased from Rs. 7,335 crores in 1992-93 to Rs. 17,531 crores in 1997-98 (at an annual average growth rate of 19 per cent), and that to domestic sector from Rs. 2,035 crores to Rs. 4,685 crores during the same period (at a rate of 18.15 per cent per year). The subsidy given by the State Governments, on the other hand, increased from Rs. 3,182 crores to Rs. 3,706 crores only (at an annual rate of only 3.1 per cent), and the cross subsidies from the other sectors (industry, commercial and railway traction) from Rs. 3,911 crores to Rs. 11,289.3 crores during this period (at an annual rate of 23.62 per cent). These two together could neutralize only about 57 to 83 per cent of the effective subsidies provided during this period. Note that not all States compensate the SEBs for the subsidized electricity sales to agriculture and domestic consumers. For example, in 1997-98, 8 State Governments (out of 19) did not provide any compensation at all, and in the case of Assam, it was too nominal. Some of the State governments, on the other hand, write off the interest due to them also in compensation for the subsidized sales. Moreover, the tilt in the compensating mechanism has been to tax the other sectors heavily and tap the maximum cross subsidies; in 199798, the State Government subsidy constituted only 17 per cent of the total effective subsidy to agriculture and domestic sectors, while the cross subsidy accounted for about 51 per cent of it. Such

16

over-burdening would have very serious impact on the competitive and healthy operation of these sectors and drive them on to set up their own captive generation.

Agriculture has accounted for around 80 per cent of the total effective subsidies. Andhra Pradesh, Gujarat, Haryana, Karnataka, Madhya Pradesh, Maharashtra, Punjab, Rajasthan, Tamil Nadu, and Uttar Pradesh, where agricultural consumption is higher, suffer substantial losses due to subsidized power sales (Tables 5 A and B). Remember in Punjab and Tamil Nadu, the subsidy is cent per cent, and in Karnataka and Madhya Pradesh, it is nearly so. Even the introduction of the national minimum agricultural tariff of 50 paise/unit would still leave a significant gap uncovered. In 1997-98, for example, this gap was of the order of Rs. 15,277.2 crores. Similarly, subsidized domestic power consumption has been responsible for very high losses in Delhi, Kerala, Madhya Pradesh, Maharashtra, and Uttar Pradesh. Kerala had a cost-recovery ratio in 1997-98 of 40.6 per cent in the domestic sector, 28.4 per cent in agriculture, 83.7 per cent in industry, and 145.7 per cent in the commercial sector (see Table 2). Since agriculture accounts for only 4.4 per cent of the total power consumption, subsidy-bred loss cannot be very high from this sector. However, the domestic and industrial sectors’ consumption constitutes about 82 per cent of the total, and thus imposes a heavy burden of loss due to subsidy, especially the domestic sector, which consumes nearly 50 per cent of the total power. It should be pointed out that about 180 thousand households, consuming less than 20 units in a month, are given free power in Kerala. Moreover, in a bid to attract industries, Kerala also allows reduced rates for new industries for the first 5 years, while most of the other SEBs charge the ruling rate, with the subsidy going directly to the beneficiaries from the State Governments. However, the very rationale for such industrial subsidy in Kerala has soon got defeated, as almost all of these new (metal, chemical, etc.) industries sprung up in the State under the subsidy umbrella have been capital- and energy-intensive with very limited prospects for creating employment opportunities, the prime objective of the subsidy scheme. An estimate has put the loss to the KSEB at more than Rs. 30 crores a year on account of the subsidized power sales to these new industries that employ less than 800 workers in all ! Unlike in most of other States, it is the commercial sector alone that is made to bear the burden of cross subsidization in Kerala.

The Receivables and the Dues

To ensure financial health of SEBs, it should be ensured in turn that the prescribed tariffs are adequate for the purpose and are reviewed periodically and revised, whenever necessary, consistent with the trend of the operational parameters, input costs, etc. In addition, and more importantly, it

17

should also be ensured that the sales revenue these tariffs yield is collected regularly in time and the outstanding dues are kept to the minimum possible. As revenue arrears accumulate, the very purpose of tariff revision gets defeated; and sadly this is so in almost all the States. The uncovered revenue dues outstanding against different consumers in the State power sector was always on the increase over time, for example, from Rs. 6,720 crores in 1992-93 to Rs. 11,535 crores in 1996-97, growing at an annual rate of 14.5 per cent. Accounting for about 26 to 36 per cent of the annual sales turnover, these arrears represent about 4 months’ sales revenue being locked up with the consumers at any point of time, against the maximum allowable norm of two months’ sales revenue. States like Bihar and Jammu & Kashmir have revenue arrears of up to 183 per cent (in 1996-97) and 228 per cent (in 1995-96) respectively of their annual sales, equivalent to about 22 and 27 months’ sales revenue respectively, while it is at the lowest in Tamil Nadu with 2 to 4 per cent (i.e., 7 to 15 days’ sales) only. Assam, Delhi, Uttar Pradesh, and West Bengal had often more than 50 per cent turnover of revenue arrears, i.e., more than 6 months’ sales, while for Kerala, 23 to 43 per cent, i.e., about 3 to 5 months’ sales (see Table 6). In 1997-98, it was about 41 per cent for Kerala, nearly 5 months’ sales. Besides these receivables against electricity supply, there are other sundry debtors also, which, for example, in Kerala amounted to Rs. 326 crores in 1996-97 and Rs. 319 crores in 1997-98, about 47.1 and 32.8 per cent of, or 6 and 4 months’, sales revenue respectively. Regular and timely collection of all receivables could increase the liquidity available with the SEBs and arrest the excessive loan-tropism. For instance, if all the SEBs could limit the revenue arrears receivable to nearly two months’ sales norm, additional revenue collected of Rs. 4,490 crores would be available with them in 1996-97, which in turn means that they could dispense with additional loans of the order of about Rs. 4,500 crores in that year or be relieved of some of the old loans. In other words, this is the cost of inefficiency in the management of sundry debtors in 1996-97. For 1995-96, this amounts to Rs. 7,567 crores. That every year such huge cost of liquidity restriction is left to be incurred explains the financial accountability of the SEBs.

While on one side the receivables to the SEBs mount up, so are their own outstanding dues to the major Central power sector undertakings such as National Thermal Power Corporation (NTPC), National Hydro Power Corporation (NHPC), Damodar Valley Corporation (DVC), Power Finance Corporation (PFC), etc. (see Table 6). These arrears (with surcharge) to be paid by the SEBs are reported as on September 30, 1998, to be over Rs. 16,800 crores. Some of the SEBs are overwhelmed by these dues – Uttar Pradesh (over Rs. 3,480 crores), Bihar (over Rs. 3,010 crores), and West Bengal (over Rs. 1,600 crores) (Governmant of India 1999: 89). On the other hand, some

18

of the Central sector undertakings are left poor by substantial amounts of receivables – NTPC: Rs. 8,847 crores, NHPC: nearly Rs. 2,357 crores, and REC: nearly Rs. 2,727 crores.

Commercial Losses Due To Inefficiencies

Now let us analyze this situation a little more objectively in terms of the inefficiency estimates we have obtained earlier. To start with, remember that with some, quite reasonably achievable, improvement in the operational, T & D, and manpower deployment efficiencies, as well as with 1:1 debt-equity capital structure, the all-SEBs’ 1997-98 unit cost of electricity supply has been found to fall to Rs. 1.52/unit. A summary of unit cost savings from efficiency improvement, for all-India and Kerala, is given in Table 7. Compare this with the AR realized from sales of Rs. 1.85/unit in that year. This would yield an additional revenue of about Rs. 9,459 crores over and above the total cost of electricity supply – a commercial profit! Similarly, Kerala could earn a profit of Rs. 121.06 crores and Delhi, Rs. 349.72 crores ! To this extent then the reported commercial loss of the SEBs, attributed to the so-called unit-cost-unrecoverable AR, turns out to be nothing but inefficiency-caused loss. If we allow for the expenses capitalized, then the total cost in the accounting sense would still decline and commercial profit increase. And the vociferous arguments and assertions for steep rises in tariff rates, proposed to be required to contain the increasing supply costs in order to save the SEBs from the red, reduces to calculated camoufaging of pampered inefficiency.

However, this is not meant to justify the present unscientific tariff setting. A rational tariff structuring should, among others, aim to help the SEB earn a reasonable return over and above the total costs, that differ at different voltage levels, once the effect of distribution loss factor also is accounted for. Thus, for example, at the LV distribution level, sufficient weight in terms of the actual distribution loss experienced should be put on the cost of supply of electricity that includes revenue expenditure and the fixed costs. This then gives, for instance, an average tariff of Rs. 2.06/unit for 1997-98, with our estimated supply cost of Rs. 1.52/unit, a mark-up of 15 per cent and a T & D loss of 15 per cent. This is greater than the AR realized in 1997-98 by 21.51 paise/unit. For Kerala, such an average tariff estimated would be Rs. 1.47/unit, higher by 22.75 paise/unit than the actual AR realized, and for Delhi, Rs. 3.03/unit, higher by 37.86 paise/unit. While all the LV consumers are logically expected to bear this charge, the HV-EHV industrial consumers need to pay much less, as supply of electricity to them involves lower unit cost of supply as well as T & D loss.

19

Unlike such historical (accounting) cost method generally practiced, a rational tariff policy would require charging the consumers for the actual cost of service to them. The average price structured in such a truly cost-reflecting tariff would be the long run marginal cost to the system.

Cost of the ‘Cover Up’

Indeed, subsidization also involves problems of inefficiency. However, the reported loss due to subsidized power sales to agriculture in India is a substantially over-estimated one, in view of our earlier explanation on leaving agricultural energy consumption as a residual estimate. We have found that about 30 to 40 per cent of what is usually reported as agricultural power consumption in fact represents unaccounted-for energy. Now assuming, quite reasonably, that the actual agricultural consumption is only 65 per cent of the reported one, we can estimate the commercial ‘loss’, due to subsidized sales of 57,707 MU of electricity (instead of the reported 88,780 MU) to agriculture at a unit cost-revenue margin of 197.47 paise/unit in 1997-98, to be Rs. 11,395.4 crores, instead of the given Rs. 17,531.3 crores. The total effective subsidy provided to both agriculture and domestic sector would then be Rs. 16,080.4 crores, and accounting for cross subsidy and subsidy from the State Government, the ‘loss’ due to subsidized power sale would turn out to be only Rs. 1,084.3 crores, instead of the reported Rs. 7,220.6 crores. Thus, a good part of the huge amount of ‘subsidy’ claimed to be provided to agriculture, the slogan of which in turn is used unfairly to enhance the populist image of the Governments, does in fact represent the cost of inefficiency in not operating and maintaining the T & D system properly.

But the story is not yet complete; we have not counted the course of the 35 per cent of the agricultural consumption liberated as above from misclassification. Let this be available for sales instead of being thieved away. And with the heroic assumption that consumption is satisfied at the given level of 2,93,478.9 MU (in 1997-98), and that the operational efficiency, as explained earlier, has already brought down power purchase requirement to (1,74,373.9 – 1,05,186.4 =) 69,187.5 MU, we can have a further reduction in power purchase cost, using the 35 per cent recovered energy equivalent to 31,073 MU, to the tune of Rs. 4,310.14 crores or 14.69 paise per unit sold, that represents the cost of the ‘cover up’ (of energy theft by misreporting it as agricultural consumption). This will reduce the ‘efficient’ unit cost of power supply further to Rs. 1.38/unit, from the reported cost of Rs. 2.28/unit, with a cost inefficiency9 of about 40 per cent ! Comparing this with the AR realized in 1997-98 would yield a commercial profit of Rs. 13,770 crores !

20

As an alternative scenario (in our flight of imagination), let this 35 per cent retrieved energy be made available for additional sale to, say, industrial and commercial sectors. Then at a costrevenue margin of 103.65 paise per unit in 1997-98, this would bring in an additional cross-subsidy of Rs. 3,220.7 crores, taking the total cross subsidy plus State Government subvention to Rs. 18,216.4 crores, in excess of the ‘actual’ effective subsidy (we obtained above) to agriculture and domestic sectors of Rs. 16,080.4 crores. There is thus a commercial profit (Rs. 2,136 crores) due to (cross) subsidization ! This means that the cost, from this perspective, of the ‘cover up’ alone comes out to be Rs. (7,220.6 + 2,136 =) 9,356.6 crores ! The question now echoes: Should the cost of this inefficiency be transferred on to the public ?

Though the real subsidy reaching the agriculture sector flows into the vast fields of big kulaks, the powerful and embedded socio-political sentiments guard and guarantee the practice of backing the backbone of the economy. Many studies have given the lie to the illusion of power subsidy to agriculture; for example, it has been found that in Maharashtra (with the discriminatory tariff) the primary beneficiaries of subsidized power in agriculture are the 5 per cent of affluent farmers growing water-intensive cash crops such as sugarcane, not the majority of poor farmers of food crops (Sant and Dixit 1996). The general profile may not be different from this case. This subsidized unmetered power consumption, though heavily cross-subsidized by the commercial and industrial customers, has, however, given the Governments an easy and costless access to vast vote banks, but at the cost of the financial health of the SEBs in the absence of any comparable compensation.

Unlike agricultural sector, however, little economic justification is found in subsidizing the domestic sector as a whole (supply to which typically imposes higher costs on the system in terms of peak time requirements, extensive distribution network, and losses). The fact that in general the unelectrified households belong to the poorest of the society questions the justification, if any, of such subsidy to this sector, that too across the board. Social and welfare regards would require special treatment to low income groups by means of a ‘life-line tariff’ applied to the lowest consumption slab only. Cross subsidization required should be tapped from other consumers in the same (domestic) sector, such that the sector as a whole remains subsidy-cost-free. Internal Resources

21

It should be stressed that the performance of the SEBs was largely determined for a long time by the assertions and defenses of their statutorily intended promotional role in power development. The SEBs were to subserve the socio-economic policies of the State and hence expected not to view every aspect of developmental activities exclusively from the point of view of profit or return, as highlighted by the Venkataraman Committee of 1964. Thus there was no compulsive requirement, till the late seventies (till the 1978 amendment of the Section 59 of the E(S) Act, 1948), for the SEBs to break even, as also even to provide for full depreciation and/or interest payable on Government loans, both of which could, under the Statute, be provided for only if there were adequate surpluses after meeting all other obligations. Thus there seemed to be no idea, let alone requirement, of the SEBs contributing internal resources to expansion programs. The SEBs have not yet come out of that spell of unaccountable, non-commercial performance, and in general continue to have negative internal resources.

Net internal resource (NIR) refers to the surplus left with the SEBs after meeting revenue expenditure and loan repayment obligations. It thus includes operating surplus, depreciation and subvention from State Government. In line with the tradition, the NIR of the SEBs slided down from Rs (–) 162 crores in 1992-93 to Rs (–) 373 crores in 1997-98. Maharashtra was the only State to report positive NIR in all these years, and Bihar, Delhi, and West Bengal, negative NIR. Note that the actual resource generation might be much less, as revenue outstandings are high and on the increase. Further, redemption of capital loans might eat substantially into the surplus.

The NIR available with the SEBs could be increased if they were allowed to retain with them the State electricity duty (SED), collected by the SEBs and passed on to the respective State exchequer. The SED collections increased from Rs. 1,131 crores in 1992-93 to Rs. 2,365 crores in 1997-98. Gujarat has had the highest SED collection – about 37 per cent of the total in 1997-98; followed by Madhya Pradesh (16 per cent), Maharashtra (12.8 per cent), Karnataka (5.8 per cent ) and Punjab (4.6 per cent). The average incidence of SED on the sale of electricity was in the range of 5 to 8 paise/unit in the 90s, or nearly 5 per cent of the estimated overall tariff for electricity sales. Provision for retention of SED with the SEBs would have left them with substantial positive NIR in all these years, except in 1997-9810.

Before concluding, let us reiterate that lapses in financial discipline and accountability penalize the system heavily. For an instance from Kerala, consider the following observations by the Accountant General on the KSEB:

22

“(i)

Loss due to investment of borrowed funds on short-term deposits: Rs. 27.55 lakhs;

(ii)

Loss due to payment of penal interest towards non-submission of statement to banks: Rs. 13.64 lakhs;

(iii)

Loss due to failure of the Board to detect the wrong transfer of funds: Rs. 3.85 lakhs;

(iv)

Loss due to payment of penal interest, liquidated damages, etc., due to belated payment of principal and interest to LIC of India: Rs. 74.99 lakhs.” (quoted in Government of Kerala 1997: 56)

5. CONCLUSION

As in the case of other infrastructure facilities with high capital intensity and long gestation period, the responsibility of power development also was originally shouldered by the Government. And in turn, the power sector was expected to subserve the social, political and economic policies of the State, even though the SEBs were required by the E(S) Act, 1948, to function as autonomous corporations. The patronizing policies of the State resulted in excessive employment, especially at the non-technical, administrative level, involving unwarranted cost increases and in irrational pricing practices for subsidized power sales, irrespective of considerations of costs, leading to substantial losses. In addition to Plan outlays allocated to the power sector, Government subventions were also on the way in, such that the SEBs never felt the pressing requirement to break even or to contribute to capacity expansion programs. The unaccountability culture, thus engendered and encouraged, permeated the whole institutional texture, and the consequent gross inefficiency contagioned the system. The rot set in. Losses mounted up, and prospects counted down. And then one fine day, the Government awakened to the bitter truth that its coffer could no longer contain such losses, and exhorted and enjoined the SEBs to mend their ways and mind their means. Then followed the pandemonium, the chaos that is to precede any restructruring. By that time, however, the lot had been cast.

The whole system could be spared from such avoidable chaos, if the Government interference were kept to a minimum and the SEBs were let to function as autonomous commercialcum-service corporations, as required by the E(S) Act. We have seen that if some minimum, affordable standards of efficiency were maintained at the technical, and institutional/organizational levels in the functioning of the SEBs, considerable cost savings could be achieved and this, coupled with a rational pricing practice, could win the system a very comfortable position. It could work

23

even otherwise; if the Government fully compensated the SEBs for its induced inefficiencies regularly and in time, the industry could still sustain its survivability11. The compensation system has failed on both the fronts – the timely submission of the accounts by the SEBs and the timely payment by the Government. Here is an instance: ‘The rural electrification subsidy receivable from the Government of Kerala for the loss incurred by the KSEB due to Rural Electrification operations during 1985-86 to 1993-94 was estimated and submitted to the Government for sanctioning the release of subsidy’ only by 1996-97 (KSEB, Annual Statement of Accounts 1996-97 and 1997-98). When will these accounts now hatch ?

The utter negligence and neglect of the means to ensure minimum T & D loss has been another contaminated fallout of the Government-sponsored inefficiency. Unmetered drawal of electricity is rampant in several urban areas, in connivance with the Board staff, or by errant consumers enjoying protective patronage. The Union Power Minister has recently dubbed this unaccountable-for energy as “theft and dacoity losses”, amounting to about Rs. 15,000 crores every year. ‘He gave the example of Orissa, where the private sector companies that have taken over distribution of electricity are finding it difficult even to install meters, what to speak of collecting the dues. “AES of USA is having to employ goon gangs to install meters”, the Minister said.’ (The Hindu Business line, March 31, 2000). Isn’t the reform process initiated in Orissa then a reflection of the defeated political will at the hands of a Frankenstein ?

Compounding all these is the infamous X-inefficiency at all levels of ‘work culture’, that has deteriorated to such an abyss that it remains devoid of any accountability, a legacy of the original sin of service-only-orientation forced on the SEBs. It may not be unreasonable to state that this is in fact basic and central to many problems in the electricity supply industry in India. And inefficiency continues to rot the system, the inefficiency bred and fed by a host of factors at technical, institutional and organizational, financial as well as socio-political policy levels. However, the most relieving aspect of this system predicament is that the problems are just internal to the system, as we have shown above12. This then implies that there do remain sufficient quarters for remedial exercises, meant to remove the problems that stand in the way of the SEBs’ improved performance. In other words, what the system badly requires is essence-specific reforms, not structure-specific ones.

The parlous financial position of the SEBs has come in handy for the institutional lenders including the World Bank to press for structure-specific reforms. The attraction of soft loans offered

24

as a package with reforms and of the selling out of public sector assets have cornered and captured the political theory of corruption that governs the prodigal governments. The result resembles an irreversible, disastrous Alexanderian solution to the Gordian knot – the so-called reforms now under way in a number of States which in practice will remain incapable of addressing the real problems internal to the system. Will a forced change of the form transmute the substance also ? -----------------------------------------------

25

TABLE 1: Cost Structure of SEBs (1997-98) (Paise/unit sold)

SEB Andhra Pradesh Assam Bihar Delhi (DESU) Gujarat Haryana Himachal Pradesh Jammu & Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Meghalaya Orissa Punjab Rajasthan Tamil Nadu Utter Pradesh West Bengal Average

Fuel

Power Purchase

O&M

59.33 50.72 15.67 35.52 101.58 53.36 0 8.11 7.01 4.18 44.81 67.28 0 0 75.60 52.14 89.20 48.46 35.56 56.19

68.00 127.78 172.74 259.72 77.01 99.63 55.00 181.02 103.77 70.62 71.43 56.74 10.46 187.67 42.19 83.60 46.11 66.00 142.81 82.42

7.65 11.71 14.40 10.74 6.34 11.88 21.23 6.46 8.09 5.63 8.91 12.10 22.83 9.65 6.66 8.19 5.23 11.00 8.95 10.02

Establt. / Miscellaneous Admin. Expenditure 20.71 86.74 0 31.33 19.89 39.10 44.53 25.70 31.92 48.66 32.89 25.03 85.41 33.20 41.05 27.00 42.09 31.31 26.81 27.29

3.08 4.40 47.41 0 0 0 0 0 7.14 6.96 4.14 1.41 0 0.22 1.36 26.78 0.68 0.33 3.88 4.18

Depreciation

Interest

Total

16.62 30.01 23.96 14.00 16.74 16.92 6.44 16.11 13.09 9.83 21.98 20.96 11.88 23.32 17.39 16.24 14.96 27.74 12.93 18.09

42.02 111.84 21.10 0 13.82 23.2 32.81 50.00 17.58 46.26 32.71 23.28 36.14 17.93 41.43 45.79 18.78 54.73 23.73 29.70

217.41 423.20 295.28 351.31 235.38 244.09 160.01 287.40 188.60 192.14 216.87 206.8 166.72 271.99 225.68 259.74 217.05 239.57 254.67 227.89

Source: Planning Commission, Annual Report on the Working of SEBs & EDs, April 1999; for Kerala, KSEB, Annual Statement of Accounts, 1997-98..

26

TABLE 2: Customer Category-wise Average Tariff, 1997-98 (Paise/unit)

Domestic

Commer- Agricult- Industrial cial ural Andhra Pradesh 167.0 367.8 17.4 330.3 Assam 145.9 297.0 166.8 209.4 Bihar 109.3 211.4 14.0 275.9 Delhi (DESU) 258.7 237.0 116.0 299.4 Gujarat 200.0 371.0 20.0 336.8 Haryana 238.0 338.0 55.0 338.0 Himachal Pradesh 60.0 210.0 50.0 182.8 Jammu & Kashmir 32.0 58.0 12.5 47.3 Karnataka 118.3 500.3 87.1 302.9 Kerala 77.99 279.9 54.6 160.85 Madhya Pradesh 74.5 337.5 10.0 401.8 Maharashtra 162.8 281.9 25.5 345.3 Meghalaya 83.8 160.0 50.0 163.9 Orissa 123.8 235.0 65.8 256.2 Punjab 148.2 276.5 0 241.2 Rajasthan 144.0 300.0 38.0 304.7 Tamil Nadu 146.2 403.0 0 321.4 Utter Pradesh 108.9 354.0 44.7 335.5 West Bengal 94.3 214.0 29.7 246.7 Average 133.9 333.3 27.7 284.8 Note: For Delhi, Annual plan estimates for 1998-99. Source: As in Table 10.

Average 188.5 216.6 210.7 270.7 193.0 178.8 162.3 39.3 163.7 124.6 176.2 213.8 130.0 218.0 139.7 194.9 197.1 171.5 194.0 184.5

A v e r a g e T a r I f f / C o s t R a t i o (%) Domestic Commercial Agricultural Industrial Overall 76.81 169.17 8.00 151.92 86.70 34.48 70.18 39.41 49.48 51.18 37.02 71.59 4.74 93.44 71.36 73.64 67.46 33.02 85.22 77.05 84.97 157.62 8.50 143.09 82.00 97.51 138.47 22.53 138.47 73.25 37.50 131.24 31.25 114.24 101.43 11.13 20.18 4.35 16.46 13.67 62.73 265.27 46.18 160.60 86.80 40.59 145.66 28.43 83.71 64.85 34.35 155.62 4.61 185.27 81.25 78.72 136.32 12.33 166.97 103.38 50.26 95.97 29.99 98.31 77.98 45.52 86.40 24.19 94.19 80.15 65.67 122.52 0 106.88 61.90 55.44 115.50 14.63 117.31 75.04 67.36 185.67 0.00 148.08 90.81 45.46 147.76 18.66 140.04 71.59 37.03 84.03 11.66 96.87 76.18 58.76 146.25 12.15 124.97 80.96

27

TABLE 3: Average Costs and Revenues of SEBs (Paise/unit)

Average Cost

1980-81 Average Tariff / Tariff Cost (%)

Average Cost

1990-91 Average Tariff

Andhra Pradesh 37.80 37.30 98.68 78.72 74.49 Assam 76.46 38.58 50.46 249.59 98.84 Bihar 67.68 43.21 63.84 168.97 88.56 Delhi (DVB) .. .. .. 137.89 99.10 Gujarat 46.01 40.10 87.15 110.11 78.00 Haryana 40.31 29.52 73.23 103.66 66.63 Himachal Pradesh 59.71 29.98 50.21 94.77 79.13 Jammu & Kashmir .. .. .. 125.59 35.92 Karnataka 26.26 28.22 107.46 82.55 81.34 24.40 Kerala 22.38 109.03 68.17 52.58 Madhya Pradesh 52.44 40.17 76.60 116.44 84.86 Maharashtra 36.52 28.41 77.79 107.44 103.06 Meghalaya 27.99 22.64 80.89 137.28 59.21 Orissa 37.57 31.99 85.15 71.43 67.89 Punjab 36.46 23.29 63.88 106.79 54.87 Rajasthan 39.12 28.01 71.60 114.59 92.91 Tamil Nadu 43.82 30.42 69.42 114.32 86.53 Utter Pradesh 56.33 35.38 62.81 110.04 73.09 West Bengal 49.00 41.34 84.37 157.19 104.19 Average 41.90 32.30 77.09 108.59 81.80 Note: ACGR = Annual Average Compound Growth Rate (%). Source: As in Table10.

Tariff / Cost (%) 94.63 39.60 52.41 71.87 70.84 64.28 83.50 28.60 98.53 77.13 72.88 95.92 43.13 95.04 51.38 81.08 75.69 66.42 66.28 75.33

ACGR of 1997-98 ACGR of Tariff / Average Average Tariff / Tariff / Cost Ratio Cost Cost Tariff Cost (%) (1990-98) (1980-98) (1980-91) -0.42 217.44 188.5 86.69 -1.24 -0.76 -2.39 423.17 216.6 51.19 3.73 0.08 -1.95 295.28 210.7 71.36 4.51 0.66 .. 365.02 265.1 72.63 0.15 -2.05 235.37 193.0 82.00 2.11 -0.36 -1.30 244.09 178.8 73.25 1.88 0.00 5.22 160.17 162.3 101.33 2.80 4.22 .. 290.47 39.3 13.53 -10.14 -0.86 188.59 163.7 86.80 -1.79 -1.25 -3.40 199.04 126.9 63.76 -2.68 -3.11 -0.50 218.12 176.2 80.78 1.48 0.31 2.12 218.24 213.8 97.97 0.30 1.37 -6.09 168.19 130.0 77.29 8.69 -0.27 1.11 272.03 218.0 80.14 -2.41 -0.36 -2.15 226.64 139.7 61.64 2.63 -0.21 1.25 259.57 194.9 75.09 -1.09 0.28 0.87 217.06 197.1 90.80 2.63 1.59 0.56 239.59 171.5 71.58 1.07 0.77 -2.38 254.70 194.0 76.17 2.01 -0.60 -0.23 227.89 184.48 80.95 1.03 0.29

28

TABLE 4:Commercial Losses Due to Differential Pricing Policy (Rs. Crores) With Subsidy Without Subsidy 1992-93 1997-98 ACGR 1992-93 1997-98 ACGR (%) (%) Andhra Pradesh 4.3 503.4 159.2 4.3 503.4 159.2 Assam 205.4 440.5 16.5 205.4 440.6 16.5 Bihar 279.6 370.2 5.8 279.6 370.2 5.8 Delhi (DVB) 207.3 759.9 29.7 207.3 759.9 29.7 Gujarat 100.0p 770.0 (-) 519.0 1270.0 19.6 Haryana 368.3 275.6 -5.6 403.6 525.6 5.4 Himachal Pradesh 1.6p 18.9 (-) 1.7p 18.9 (-) Jammu & Kashmir 224.5 608.6 22.1 224.5 608.6 22.1 Karnataka 32.2p 60.7p 13.5 19.4 308.5 73.9 Kerala 65.3 218.8 27.4 65.4 370.8 41.5 Madhya Pradesh 112.9 322.1 23.3 492.9 697.1 7.2 Maharashtra 161.6p 111.8p -7.1 161.6 p 111.8p -7.1 Meghalaya 1.9 10.1 39.7 8.4 19.1 17.9 Orissa 26.0p 257.9 (-) 85.4 300.9 28.6 Punjab 626.3 1346.0 16.5 626.3 1346.0 16.5 Rajasthan 22.1p 506.6 (-) 259.5 506.6 14.3 Tamil Nadu 92.4p 194.8 (-) 257.6 469.8 12.8 Utter Pradesh 807.5 63.8 -39.8 807.5 1735.8 16.5 West Bengal 257.5 483.1 13.4 257.5 544.1 16.1 Total 2724.9 6977.8 20.7 4560.3 10684.2 18.6 Note: p = Profit; ACGR = Annual Average Compound Growth Rate (%). Source: As in Table 10.

29

TABLE 5 (A): Commercial Losses Due to Differential Pricing Policy (Rs.Crores) L o s s

d u e t o E f f e c t i v e s u b s i d y t o Agriculture Domestic Sector At Current Tariff * At 50 Ps. per Unit * 1992-93 1997-98 ACGR 1992-93 1997-98 ACGR 1992-93 1997-98 ACGR (%) (%) (%) Andhra Pradesh 725.9 1815.5 20.12 395.2 1519.7 30.91 40.4 255.8 44.65 Assam 6.2 14.1 17.86 6.2 14.1 17.86 37.2 118.9 26.16 Bihar 267.6 390.8 7.87 207.3 340.8 10.45 40.0 177.9 34.78 Delhi (DVB) 8.6 25.4p 24.18 7.8 25.4p 26.63 297.4 422.7p 7.28 Gujarat 1055.4 2503.9 18.86 751.8 2155.1 23.44 88.0 92.1 0.91 Haryana 456.8 742.7 10.21 342.9 742.7 16.72 95.6 11.0 -35.11 Himachal Pradesh 1.0 1.4 6.96 0.8 1.4 11.84 13.7 45.1 26.91 Jammu & Kashmir 27.5 116.7 33.52 20.7 101.0 37.30 49.5 156.4 25.87 Karnataka 496.5 825.2 10.69 249.8 825.2 27.00 22.5 220.7 57.88 Kerala 0 51.7 .. 0 51.7 .. 60.5 344.3 41.59 Madhya Pradesh 421.1 1854.2 34.51 329.3 1497.8 35.39 252.7 592.9 18.60 Maharashtra 1030.9 2875.8 22.77 741.0 2510.2 27.64 151.6 295.5 14.28 Meghalaya 0 0.2 .. 0 0.2 .. 1.4 7.6 40.26 Orissa 20.7 48.3 18.47 14.9 48.3 26.52 80.4 228.3 23.21 Punjab 687.1 1473.1 16.48 444.9 1148.1 20.88 54.3 252.1 35.94 Rajasthan 347.5 1040.9 24.53 285.9 984.5 28.06 183.3 264.8 7.63 Tamil Nadu 642.5 1561.7 19.44 384.5 1202.0 25.60 141.8 304.6 16.52 Utter Pradesh 1035.4 1890 12.79 878.9 1838.6 15.91 331.8 987.7 24.38 West Bengal 104.2 324.9 25.54 81.7 295.6 29.33 92.9 329.3 28.80 Total 7334.9 17531.3 19.04 5143.4 15277.2 24.33 2034.9 4685.0 18.15 Note: p = Plan estimate for 1998-99; ACGR = Annual Average Compound Growth Rate (%). * - excluding the subsidy for rural electrification given by the State Govts. Source: Planning Commission, (GOI), Annual Report on the Working of SEBs and EDs, April 1999.

30

TABLE 5 (B): Structure of Subsidization (Rs.Crores)

Andhra Pradesh Assam Bihar Delhi (DVB) Gujarat Haryana Himachal Pradesh Jammu & Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Meghalaya Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal Total

Total Subsidy Subsidy Received from Cross Subsidy from (to Agriculture + Domestic) State Govt. Other Sectors 1992-93 1997-98 ACGR 1992-93 1997-98 ACGR 1992-93 1997-98 ACGR (%) (%) (%) 766.3 2071.3 22.00 0 0 .. 668.0 1321.4 14.62 43.4 133.0 25.10 0.1# 0.1 0 -148.0 -381.3 (-) 307.6 568.7 13.08 0 0 .. -66.0 267.9 (+) 306.0 448.1 7.93 0 0 .. 71.0 -595.5p (-) 1143.4 2596.0 17.82 619.0 500.0 -4.18 319.0 1334.3 33.14 552.4 753.7 6.41 35.0 250.0 48.17 23.0 269.0 63.53 14.7 46.5 25.90 0 0 .. -6.7 128.3 (+) 77.0 273.1 28.81 0 0 .. -118.0 -355.8 (-) 519.0 1045.9 15.04 51.6 369.2 48.23 475.0 1241.6 21.19 60.5 396.0 45.61 8.6 # 152.0 77.61 -1.5 167.4 (+) 673.8 2447.1 29.43 380.1 375.0 -0.27 310.0 1457.8 36.29 1182.5 3171.3 21.81 0 0 .. 1135.0 3028.2 21.69 1.4 7.8 40.99 6.5 9.0 6.72 0.3 -4.0 (-) 101.1 276.6 22.30 1390.0 43.0 -50.10 -166.7 -84.7 (+) 741.4 1725.2 18.40 0 0 .. 35.0 146.9 33.23 530.8 1305.7 19.72 281.6 560.5* 14.76 87.0 342.2 31.51 784.3 1866.3 18.93 350.1 275.0 -4.71 461.0 1341.0 23.81 1367.2 2877.7 16.05 1237.0 1672.0 6.21 920.0 990.1 1.48 197.1 654.2 27.12 68.1 61.0 -2.18 -86.4 78.9 (+) 9369.8 22216.3 18.85 3182.0 3706.4 3.10 3911.0 11289.3 23.62

Note: * for 1996-97; # = for 1994-95; p = Plan estimate for 1998-99. Source: As in Table 14 (A).

31

TABLE 6: Outstanding Dues to Central Enterprises and Revenue Arrears (Rs. Crores)

Outstanding Dues as by as by ACGR 1995 1997 (%) Andhra Pradesh 158 196.4 Assam 189 -10.8 Bihar 1033 509.9 Delhi (DESU) 522 NA Gujarat 173 238.0 Haryana 641 NA Himachal Pradesh 35 11.7 Jammu & Kashmir 323 69.1 Karnataka 54 74.9 Kerala 320 31.1 Madhya Pradesh 717 357.5 Maharashtra 431 36.4 Meghalaya 88 1.1 Orissa 380 NA Punjab 550 153.9 Rajasthan 500 0 Tamil Nadu 1671 NA Utter Pradesh 2054 2349.4 West Bengal 627 NA Total 10465 Note: $ = for (with respect to) 1995-96. Source: As in Table 14 (A).

Revenue Arrears Receivable 1992-93 1996-97 ACGR As % of Sales Revenue (%) 1992-93 1996-97 ACGR (%) 11.49 345.9 566.7 13.14 18.9 17.2 -2.33 (-) 85.2 235.2 28.90 44.3 55.8 5.94 -29.74 613.0 2317.6 39.44 92.3 182.5 18.58 .. 499.2 2192.0 $ 63.75$ 46.6 140.5 $ 44.47$ 17.29 590.0 1031.0 14.97 31.8 24.9 -5.93 .. 479.2 516.5 1.89 76.6 36.7 -16.80 -42.18 34.2 59.5 14.85 20.0 16.6 -4.55 -53.75 81.9 140.5 14.45 151.7 226.9 10.59 17.77 409.2 760.6 16.76 34.0 35.6 1.16 -68.83 99.8 211.6 20.67 23.1 31.5 8.06 -29.39 394.2 1045.2 27.61 20.4 24.9 5.11 -70.94 1125.1 1840.0 $ 17.82 26.1 26.2 $ 0.13$ -88.82 NA 66.7 $ .. .. 122.3 $ .. .. 184.3 NA .. 44.0 .. .. -47.10 127.4 248.5 18.18 12.5 10.2 -4.96 (-) 211.5 455.8 21.16 17.7 19.2 2.05 .. 70.0 94.0 7.65 3.8 2.1 -13.78 6.95 1171.6 3005.2 26.55 48.5 77.7 12.50 .. 198.6 847.2 43.71 58.6 57.9 -0.30 6720.3 11534.9 14.46 30.3 26.2 -3.57 ACGR = Annual Average Compound Growth Rate (%)

32

TABLE 16: Unit Cost Savings From Efficiency Improvement (Paise/Unit) INDIA Ps/unit % 1. Reported unit cost of power supply in 1997-98

KERALA Ps/unit %

227.89

100.00

192.14

100.00

49.72

21.82

40.32

20.98

11.05

4.85

19.78

10.29

14.85

6.52

23.13

12.04

75.62 152.27 184.50

33.18 66.82 80.96 (121.17)

83.23 108.91 124.60

43.32 56.68 64.85 (114.41)

2. Cost savings obtainable (i) in power purchase, from operational efficiency improvement (ii) in establishment & administration, from reduction in over-manning (iii) in interest payments, from introduction of 1:1 debt-equity ratio 3. Total cost savings possible 4. 'Efficient' unit cost of power supply 5. Average revenue realized in 1997-98 6. Unit commercial profit realizable 7. Electricity sold in 1997-98 (MU) 8. Commercial profit realizable (Rs. Million)

32.23 293479 94588.25

15.69 7715.50 1210.60

Note: Figures in brackets are ratio of average revenue to ‘efficient’ unit cost.

33

Notes

1

The savings in T & D reduction could as well lead to an increase in energy sales (to, say, the industrial sector, that suffers the most from power shortage) and thus in revenue, instead of helping to cut down energy import and thus supply cost. 2

In 1997-98, remember, 4 SEBs had more than 88 per cent availability and 3 SEBs had more than 75 per cent PLF.

3

This is because the PPA is in general designed for a base load plant only, which is permitted to generate at full load whenever possible. The 1992 Notification, issued in the wake of the 1991 opening up policy, does endorse such a costly design. This commitment requires backing down of the existing cheaper power stations during off-peak periods and monsoon season, causing uneconomic plant dispatch, i.e., low unit cost power being replaced by high cost power (also see World Bank 1995: 84, and D’Sa, et al. 1999). 4

Maharashtra SEB also faces a similar problem of ‘systemic inefficiency’ due to uneconomic ‘merit’ order, thanks to its PPA with Enron. Honouring the PPA (at Rs. 4.50 per unit of Enron power) costs the MSEB a good part of the cheaper power from Tata Electric Power as well as from its own thermal power plants with costs around a fourth of Enron power (The Hindu Business Line, July 11, 2000). 5

Rajadhyaksha Committee on Power (1980) observes on this aspect: “Besides low tariffs, the causes of the poor financial performance are the low operating efficiencies, high capital cost of projects due to long delays in construction and high overheads – mainly the result of heavy overstaffing. Although precise comparisons are not possible, the average employees per MW of installed capacity in India is 7 compared to 1.2 in the USA, 1.5 in Japan and 1.7 in the UK. Within the country, the expenditure on salaries varies from 12 per cent to 40 per cent of the total income of the SEBs. Much of this overstaffing is due to SEBs being compelled under political pressures to take on people they do not need.” (Government of India 1980: 53). 6

Another solution comes from raising the available internal resources, through, say, prompt collection of revenue arrears that could substantially reduce loan requirements. This aspect we will consider shortly. 7

It should be pointed out here that courts have upheld the validity of the power of the Government to fix tariff rates (e.g., 1988 (1) K. L. T. 727; 1987 (1) KLT 777; 1978 KLT 613; AIR 1960 SC, 610; 1984 SC 170 etc.). “It is true that the Board is the primary authority to fix electricity tariff rates. But, there is a statutory power reserved in favour of the Government under Section 22-B to issue, when conditions exist, necessary orders to ensure equitable distribution of electrical energy. When the power is so exercised by the Government, it can also fix the tariff rates, for, the fixation of tariff rates is incidental to the power to regulate supply, distribution, and consumption and use of electrical energy and is also part of the regulatory process of equitable distribution of electrical energy. The Government is free to make their own classification of consumers for fixation of different rates of electricity tariff and they are not bound by the specification, categorisation, designation or division made by the Board for purposes of levying electricity charges.” (1988 (1) KLT 727, Social S. G. of Assisi Sisters vs. KSEB, para. 7) 8

Hopkinson rate, popularly known as maximum demand tariff or two-part tariff, includes a (fixed) demand charge per period based on maximum demand and a variable charge based on actual energy consumption. The English engineer Dr. John Hopkinson is considered the grandfather of electricity rate making. 9

Note that this aspect of inefficiency we have not included in Table 16.

10

Note that the legal validity of SED is in fact under question. The Kerala High Court has opined: “….the surcharge imposed adds to the revenue of the State and surcharge order is a fiscal measure intended to augment the financial resources of the State…It was argued that under Section 63 of the Electricity (Supply) Act, the State Government may make subventions to the Board and this surcharge is deemed to enable the Government to make subventions. Section 63 does not authorise the Government to raise its revenue from the consumers of electric energy to enable it to make subventions to the Board. Under Section 63, there is no obligation on the Government to make any subvention and the grant is “entirely on the bounty of the Governhment”…If the action was that the Board should be benefitted by this surcharge, there was no necessity for the Government to collect the same and them make subvention to the Board.” [1988 (2) KLT 680, Chakolas Spinning and Weaving Mills Ltd. Vs. KSEB, paras 9, 24 and 25].

34

11

This, however, rests on the assumption that the SEBs do not tend to make unfair use of the compensation facility by laying their own inefficiencies in the exchequer’s net. 12

Committees after Committees have already identified these problems and prescribed remedies, the timely adoption and execution of which would have spared the system from the present predicament.

PART II

ON KERALA POWER SYSTEM

In this part, we attempt at a diagnostic analysis of the problems of the Kerala power system on both the demand and supply fronts. Chapter 4 deals with demand estimation, while the remaining three chapters, with the problems on the supply side: time and cost overruns of the power projects in Kerala (Chapter 5), transmission and distribution loss (Chapter 6) and electricity pricing (Chapter 7).

CHAPTER 4

ELECTRICITY DEMAND ANALYSIS AND FORECASTING

“Puranam ityeva na sadhu sarvam; Na capi kavyam navam ity avadyam. Santah pariksyanyatarad bhajante; Mudah parapratyayaneya budhih.”1 - Kalidasa (Malavikagnimitra I, 2). 1. Introduction

Electricity has become a vital input to the wellbeing of any society and the demand for it from an ever-expanding set of diverse needs is growing at an increasing rate. This in turn places increasing demands on scarce resources of capital investment, material means, and man-power. Forecasting of electricity consumption needs has thus become a significant element of utmost necessity of the planning exercise in the power sector. More specifically, the advent of the ‘energy crisis’ has made crucial the need for accurate projection of electricity demand.

A large number of studies have come up in India too, toeing the same methodology as applied elsewhere. These studies, however, are analytically insufficient and methodologically unsound at least on three grounds.

i) As these studies did not care for model adequacy

2

diagnostic checking, indispensably required to verify the empirical validity of the residual whiteness assumptions underlying the very model, their results might be misleading. This criticism in fact applies to all regression analysis in general. (ii) As the time series regression approach of these studies did not account for possible non-stationarity (i.e., unit root integratedness) in the series, their significant results might be just the misleading result of spurious regression. They also failed to take advantage of an integrated analytical framework for a meaningful long-run equilibrium and short-run ‘causality’3 in a cointegrating space of error correction. (iii) These studies, by adopting a methodology suitable to a developed power system in advanced economies, sought to correlate the less correlatables in the context of an underdeveloped power system in a less developed economy. All explanations of association of electricity consumption in a hopeless situation of chronic shortage and unreliability with its

2

generally accepted ‘causatives’ (as in the developed systems) of population, per capita income, average revenue, etc., all in their aggregate time series, might not hold much water here.

In what follows our empirical results prove our secepticism at least in the context of Kerala power system. This chapter is divided into four sections. Following this introduction, the next section presents a brief theoretical discussion on forecasting and demand analysis, important tests for model adequacy, and unit root problem. In the third section are presented our empirical results and the last one concludes the chapter with an attempt to forecast electricity demand in Kerala in a simple, objective and theoretically sufficient manner. An appendix, including discussions of the theoretical underpinnings of our approach, is provided at the end of the chapter.

2. Electricity Demand Analysis and Forecasting

The forecasting methods used for electricity demand in general may be divided into formalised and non-formalised methods.

Non-formalised methods such as some variants of Delphi (‘jury of executive opinion’) method are in general used for forecasts for more distant periods of time during which some changes in the structure of the power sector must be considered.

In the case of the formalised forecasting methods, two approaches may be distinguished in their scope:

i)

an input-output approach in which we try to penetrate the internal structure, and to examine the internal and external linkages of the observed object and to explain its response to input impulses; and

ii)

a statistical approach in which the object is treated as a ‘black box’ whose internal workings are unknown.

More common are the statistical approaches that take the object as a ‘black box’ and try to explain its mechanism on the basis of the interconnections of the individual elements of the 3

observed path of the system. Here the analysis of extrapolation (or non-‘causal’) and onedimensional or multi-dimensional regressions (or ‘causal’ methods) are used. The latter seek to explain the behaviour of the variables and its ‘determination’ in a relationship framework, while the former non-‘causal’ methods are solely concerned with forecasting.

Extrapolation methods, based on the assumption that the past patterns repeat in the future, thus utilise time series data to identify past pattern in the observations and then to project it into the future. Past patterns in time series data are recognised in two ways – one is based on the trend of the series, i.e., the general movement of the series in a particular direction. In such trend extrapolation, the general behaviour of the variable over time (as presented in the time series data) is determined and is then projected into the future. Thus time is the argument of trend functions. A number of functions are used for trend extrapolation, viz., linear, quadratic, exponential and so on. In the second method of extrapolation, viz., auto-regressive model, one or more previous values of the observations themselves are the arguments; the order of the function is determined by the number of previous observations used as arguments.

Simple extrapolation of historical growth rates had presented reasonably accurate results for decades (Tyrrell 1974). In fact, the post-World War II demand for electricity in the United States had been recognised to have a consistent exponential growth (uniformly in all the sectors: residential, commercial and industrial) and this is a well documented phenomenon (Tansil and Moyers 1974). Later on, however, it was felt questionable whether these trends would remain unchanged in the future, whether the simple extrapolation technique would provide accurate predictions of the future, in the face of the changes observed to occur in many of the underlying economic factors. This scepticism was well-confirmed by the findings of Chapman, Tyrrell and Mount (1972) who explained electricity demand growth as an econometric function of four ‘causal’ factors: population, per capita personal income and the prices of electricity and natural gas, and compared electricity demand projections obtained from this model with the extrapolated estimates of the government and industry.

This pioneer econometric model has since been refined to reflect more accurately the behaviour of each class of electricity consumers, by, say, incorporating price of electric appliances as an additional argument, formulating variable elasticity model (e.g., to account for spatial heterogeneity), instead of constant elasticity one, and by employing more consistent 4

Power Consumption Forecasting in India

At the all-India level, forecasts of electricity requirement and demand are made by the Planning Commission and by the Annual Electric Power Surveys (APS) convened by the Ministry of Energy, the Central Electricity Authority (CEA) being the Secretariat to the APS. The two bodies do have extensive discussion and this usually leads to a reconciliation of results. Still, subtle differences exist between the methodologies employed by them.

The Planning Commission estimates electricity demand as part of its macro-economic analysis for all the sectors of the economy. Industrial power demand is estimated for a selected set of ‘major’ (i.e., very power intensive) industries by applying consumption norms to production targets. The rest of the industrial sector is assumed to consume some proportion of the power consumption of these major industries. Railway and irrigation requirements are also projected, based on targets and consumption norms. For all other sectors – domestic, commercial, public lighting, water works, and miscellaneous – power consumption is estimated using trend extrapolations or regression analysis that relates sectoral growth rates to electricity requirements. The Planning Commission also uses input-output model to check the consistency of the macro level estimates.

On the other hand, the projection by the APS of power consumption of the industry starts with a detailed survey of major industries (that demand 1 MW or more of power) on their estimated requirements. For all other sectors, almost the same methods are used by the APS as by the Planning Commission, many of the coefficients of output-electricity relationship being identical. However, the APS relies more on trend extrapolation than on ‘causal’ demand analysis; further, the APS exercises are carried out first State-wise, then region-wise and finally at the aggregate national level. In fact, the APS State-wise forecasts form the basis for power requirement estimates of the SEBs and State governments. The APS position is of immense significance for the States, since the State level power 5

sector investment programmes are attuned to these forecasts and stand to influence the case for Central Plan assistance.

The APS, while providing disaggregated data, unlike the Planning Commission, suffers from its length of preparation and the considerable cost involved in organising a detailed survey of so many units throughout India. Furthermore, it has been found that the APS may often be upwardly biased. The APS forecasts exceed the demand met by between 20 and 80 per cent, and the divergence generally increases in the later years, as might be expected. Thus the energy consumption forecasts for Kerala by the successive APS since the 12th APS, for 1994 are in the order of 12466, 9328, 9409, and 8567 million units (MU) respectively (by the 12th, 13th, 14th, and 15th, the latest, APS). It should, however, be pointed out that it does no good to compare these forecasts with the actual demand met (about 7027.7 MU of energy internally sold) in Kerala, fraught with severe power cuts and load shedding. One way to account for such upward divergence is to regard it as reflecting the unsuppressed demand more faithfully than the realised demand.

estimation techniques (e.g., instrument variable estimation versus the familiar ordinary least squares). Such demand analysis has facilitated to estimate the ‘effect’ of a variable on electricity demand in terms of elasticity measures; if the model is specified in logarithms, the coefficient of an argument directly gives the demand elasticity with respect to that variable.

To account for the dynamic characteristic of demand, a lagged dependent variable is usually used as a regressor in the log-linear model with a partial adjustment mechanism (Koyck distributed lag with geometrically declining weights). This specification facilitates to distinguish between short run and long run elasticities. Thus while the coefficient of the price variable in this model represents the short run price elasticity of demand, the long run price elasticity is obtained by dividing the short run coefficient by one less the coefficient of the lagged dependent variable used as a regressor, i.e., by the rate of adjustment. However, the presence of the lagged dependent variable, as already noted, makes the ordinary least squares (OLS) estimator inconsistent due to the possible correlation between the lagged endogenous variable and the 6

Demand Forecasts for Kerala Demand projections for Kerala based on the 12th, 13th, 14th and 15th APS results are in consideration now in the State. A steady decrease in the peak demand/energy consumption requirements is discernible in each of these forecasts that is attributed to some restrictions and revisions in the trends relative to the base year, (reflecting the increasing quanta of suppressed demand due to lack of generation capability). The State has accepted the 14th APS as ‘more dependable’ (Government of Kerala, Report of the Steering Committee on Energy and Power, Ninth Five Year Plan, 1997-2002, State Planning Board, Thiruvananthapuram, Feb., 1997, p. 12); whereas the Balanandan Committee (to study the development of electricity in Kerala) finds the 15th APS ‘as the better estimates for future planning’ (Report, Feb., 1997, p.37).

Considering the divergences in these

forecasts of the APS for Kerala (as shown above, for example, for 1994), the State Planning Board constituted a working group to study the demand forecasts for Kerala. The committee used a log-linear model and growth rates of 4.72, 10, and 15 per cent for the HT and EHT industries to arrive at three different demand projections. The domestic demand projections in all these exercises were based on the growth of population as per the Census report. For the other sectors, the projections were made based on the trends (using semi-log scale). It should be noted that the energy demand forecast for 1994 by the Committee is only 8945 MU.

International Energy Initiative (IEI), Bangalore, has put forward a Development Focused End-Use Oriented, Service directed methodology (DEFENDUS) for estimating demand and supply of energy in an energy system, and an exercise based on this has been done for the KSEB. This methodology, with its twin focus of developed living standard and improved end-use efficiency, seeks to estimate demand for a particular energy source/carrier in a given year based on two variables – the number of energy users and their actual energy requirement in any base year as well as the expected changes in the subsequent years. The total energy demand is then equal to the aggregate demand of all the categories of users for every end-use.

7

random variable, as well as the serial correlation among the successive values of the latter. Hence the significance of instrument variable estimation.

It should be pointed out that the use of such models is connected with the prognosis of the independent variables. This in turn may involve macro-econometric modelling4.

No econometric study of electricity demand had dealt with the decreasing block pricing in a completely satisfactory way, and the estimates of price (as also income) elasticities probably contained biases of indeterminate sign and magnitude as a consequence. Taylor's suggestions (1975) to deal with this problem were two-fold:

(a) Multi part tariffs require the inclusion of marginal price and intramarginal expenditure as arguments in the demand function, and

(b) The prices employed should be derived from actual rate schedules.

All the studies (in the US) had used either ex post average prices or prices derived from Typical Electric Bills, an annual publication of the US Federal Power Commission. In response to Taylor's suggestions, most of the studies since then have utilised 'the wisdom of employing electricity prices from actual rate schedules'. Several studies have also sought to improve modelling of the dynamics of electricity demand through inclusion of stocks of electricity consuming appliances in the demand function, and also the possibilities of inter-fuel substitution. Some other studies have utilised date on individual households, small geographical areas, or the area served by a utility, in a bid to utilise a data set of higher quality than that provided by data at the state, or national level, as well as to avoid (or at least reduce) aggregation bias in estimates of price and income elasticities.

Model Adequacy Diagnosis

In addition to the usual parameter significance tests, demand analysis and forecasting models are evaluated for their simulation potential also. The simulation error measures, signifying the deviation of the simulated variable from its actual time path, which we consider in this study are the Theil inequality coefficient (TIC) and its 3 components. 8

TIC is a very useful simulation statistic related to root mean square error (RMSE), which is the square root of the mean of the squared deviations between the simulated and the actual values, and applied to the evaluation of historical simulations or ex post forecasts. It is given by the ratio of the RMSE to the sum of the square roots of the mean squared values of the simulated and the actual data series, such that it will always fall between 0 and 1. If TIC =0, the simulated and the actual series coincide for all t and there is a perfect fit. If TIC = 1, on the other hand, the predictive performance of the model is the worst. The TIC is decomposed into 3 components, bias proportion (BP), variance proportion (VP), and covariance proportion (CP), with BP + VP + CP = 1. The BP is an indication of systematic error, since it measures the extent to which the mean values of the simulated and actual series deviate from each other. Whatever be the value of the TIC, we would hope to obtain a BP much closer to zero for a good fit. The VP indicates the ability of the model to replicate the degree of variability in the variable under study. If VP is large, it means that the actual series has fluctuated considerably while the simulated series shows little fluctuation, or vice versa, which is quite undesirable. We would hope to see minimum variability between the two. The CP measures the unsystematic error, i.e., it represents the remaining error after deviations from average values and average variabilities have been accounted for. Since it is unreasonable to expect simulations perfectly correlated with actual series, this component of error is less problematic. In fact, it is generally accepted that for any value of TIC > 0, the ideal distribution of inequality over the 3 components is BP = VP = 0, and CP = 1.

An important stage, however, that is to precede hypothesis testing in forecast modelling is model adequacy diagnostic checking, one of the three concerns in this paper. The fitted model is said to be adequate if it explains the data set adequately, i.e., if the residual does not contain (or conceal) any ‘explainable non-randomness’ left from the (‘explained’) model. It is assumed that the error term in the model is a normally distributed white noise5 (with zero mean, constant (finite) variance, no serial (auto) correlation and no (cross) correlation with the explanatory variables). Since the ordinary least squares (OLS) estimators are linear functions of the error term, (under its normality assumption) they themselves are normally distributed. This normality assumption is essential for deriving the probability (sampling) distributions of the OLS estimators and facilitates hypothesis testing, using t and F statistics, which follow t and F distributions only under normality assumption, in finite samples. Hence a diagnostic checking 9

on normality assumption must be carried out before proceeding with hypothesis (significance) tests. The normality test we report here is described in Doornik and Hansen (1994); it tests whether the skewness and kurtosis of the OLS residuals correspond to those of a normal distribution. A reasonably high probability (p-) value, associated with a small test statistic value, indicates non-rejection of the normality assumption. It should also be noted here that the mean of the OLS residuals is zero by construction when an intercept is included in the model.

The no serial correlation assumption may be tested by checking whether the residual autocorrelation coefficients are statistically zero compared with standard deviation limits. Alternatively, we can test the joint hypothesis that all the autocorrelation coefficients (for a given lag) are statistically zero, using the residual correlogram (‘portmanteau’) statistic, viz., Ljung-Box (1978) statistic6. Too large a value of the ‘portmanteau’ statistic can be viewed as evidence against model adequacy, or conversely, a large p-value confirms model adequacy. However, as residual autocorrelations are biased towards zero, when lagged dependent variable is included as regressor in the model, this (as well as Durbin-Watson, DW) statistic is not reliable. The correct procedure in such conditions is to use Lagrange Multiplier (LM) test as residual correlogram; the F-form LM test, suggested by Harvey (1981), is the recommended diagnostic test of no residual autocorrelation. Durbin h test for first-order serial correlation is a LM test. It should also be noted that a low DW statistic need not be due to autoregressive errors, warranting correction for first-order autoregression7 (AR(1)). Mis-specifications in time series data can also induce serious serial correlation among the residuals, to be reflected in low DW statistic. The RESET (Regression Specification Test, due to Ramsey 1969) tests the null of no functional form mis-specification, which would be rejected if the test statistic is too high.

In addition to these, the assumption of no-heteroscedastic errors should also be checked, using, say, White’s (1980) general heteroscedasticity (F-) test; a small p-value (associated with large F-value) rejects the null of no heteroscedasticity in errors. Often the observed serial correlation in errors may be due to what is called autoregressive conditional heteroscedasticity (ARCH) effect, that makes the residual variance at time t depend on past squared disturbances (Engle 1982). Hence it is advisable that one test for the ARCH effect too before accepting the routine statistics at face value. We can also test for the instability of the parameters in the model through a joint (F-) statistic, large values of which reveal parameter non-constancy and indicate

10

a fragile model with some structural breaks (Hansen 1992). Note that the indicated significance is valid only in the absence of non-stationary regressors.

Unit Root Problem

This is the second of our concerns.

Demand analysis/forecasting based on time series regression approach, however, is beset with a fundamental problem of possible non-stationarity (or unit root integratedness) in the time series data. A time series, which is theoretically a particular realisation (i.e., a sample) of a stochastic process, is said to be stationary, if its characteristics are finite and invariant with respect to time. This simply means that the mean, variance and autocovariances of the series are all constants. In this case, the time series can be described in terms of a regression model with fixed coefficients, estimated from past data. If, on the other hand, the series is non-stationary with non-finite and time-varying parameters, then its regression modelling with fixed coefficients is not possible.

Results from regressions with non-stationary variables can be very much misleading. Granger and Newbold (1974) found that the regression coefficient estimated from two series generated by independent random walk processes was statistically significant, with very high 2

R , but very low DW statistic (indicating high autocorrelation in residuals). When the regression 2

was run in first differences of the series, the R was close to zero and the DW statistic close to 2, thus proving that there was no relationship between the series and the significant results 2

obtained earlier was spurious. Hence they suggested that the event R  DW meant ‘spurious regression’ and the series should therefore be examined for association by running regression in the first differences of the variables. Plosser and Schwert (1978) gave further empirical evidences in favour of first differencing in regression models.

In view of spurious regression with non-stationary variables, the usual conventional time series (Box-Jenkins) analysis of proceeding with suitably differenced, stationary, variables has gained much attraction. However, soon this temptation and trend fell under fire; solving the nonstationarity problem via differencing was equated to ‘throwing the baby out with the bath 11

water’, because differencing results in ‘valuable long-run information being lost’. Most of the economic relationships are stated in theory as long-term relationships between variables in their levels, not in their differences. We need to conserve and utilise in analysis this long-run information contained in the level variables, and at the same time, we have to be on the watch for spurious regression of integrated variables. Both these seemingly irreconcilable objectives could be achieved by means of cointegration mechanism.

In short, if, in a regression relationship between yt and xt, one of them is an integrated (stochastic) process (and the other deterministic), we have a case of spurious regression; if both variables are deterministic, the regression results are valid; but if both the variables are integrated processes, then the regression is spurious, unless the variables are cointegrated.

The results of the present study signifies that the earlier works both in professional and in academic circles on electricity/energy demand analysis and forecasting, without accounting for non-stationary, integrated, behaviour of the time series they used, must have involved misleading results of spurious regression and of inconsistent and less efficient estimates. That these econometric practices lacked analytical soundness and intellectual integrity is evident in the utter neglect of model adequacy diagnostic checking, the indispensable primary stage in significance evaluation of any regression mapping. Just taking for granted the assumptions underlying a model, without an examination of its empirical significance, using available techniques, amounts to gross negligence, if not sheer gloss. At least these two fundamental flaws, viz., not caring for model adequacy diagnosis and not allowing for non-stationarity in the time series data, detract the whole value from these studies.

To start with, Pachauri (1977) and Tyner (1978), through regression technique, have found very strong association between energy consumption and economic development in India, and the latter has gone to the extent of attempting to identify ‘causation’ between the two. A large number of regression analysis of electricity demand (forecasting models and ‘causative’ models, using population or number of consumers, per capita state income or domestic product or sectoral income, average sales revenue, etc.) have mushroomed in the luxuriant academic/professional fields. The Fuel Policy Committee of India (1974), Banerjee (1979), 12

World Bank (1979), Parikh (1981) and Pillai (1981) are some of the forerunners here, in addition to the regular exercises by Planning Commission, CEA and SEBs.

Demand analysis, in contrast to simple trend extrapolation, appears to be especially more relevant and significant in the context of Kerala, with a high standard of living, reinforced by a flourishing construction sector and a market for electrical and electronic goods. The so-called ‘Gulf boom’ of increasing remittances of the non-resident Keralites from the Gulf has triggered an unprecedented growth of the housing sector and encouraged an increasing demand for electricity intensive appliances in Kerala especially since the mid-seventies. Number of houses in the electrified group must also have increased (in absolute terms) as a result of the social security schemes of the government.8 Though the serious power shortage situation has however entailed restrictions on providing new connections since 1982-83, energy consumption intensity in relation to number of customers as well as connected load has been on the increase. Therefore, we also take up an analysis of electricity demand in the State in the framework of a ‘causality’ model.

3. Analysis

Model Adequacy Diagnostic Checking

We start with an analysis of the time series data on the internal consumption of electricity (in million units, MU) in the Kerala system from 1957-58 to 1998-99 in the framework of the common extrapolation models9. We are not considering the growth curves, which are more appropriate for the demand for durable goods with an acceptable market saturation level. Table 1 reports the OLS estimates of the parameters along with other statistics of these models – the four trend extrapolation models (linear, quadratic, k-transformation and semi-log or exponential) and the two first order autoregressive [AR(1)] ones. The ‘ktransformation’ model has been turned out to be either defined or significant only for the values of k = 0.3, k = 0.4 and k = 0.5, out of a range of values tried; we report only the results for k = 0.5. 2

All the models appear to have highly significant fit, based on the conventional tests (R , F- and t- values), to the immediate satisfaction of an average researcher. The estimated measures of simulation error – TIC and its three components – also offer pleasant results. By 13

these measures, it appears that all the models in general have very good fitting performance, with very low TIC, along with an almost zero BP in most cases and a close to zero VP. In fact, this close correspondence between the actual and the fitted is an indication of non-stationarity of the series (Doornik and Hendry 1997: 33); but an average investigator, unaccounting for this, might be easily misled by the seemingly significant results. Here lies the significance of model adequacy tests. 2

Now see the danger signal of ‘spurious regression’ (R  DW) blazing in most of these models, where diagnostic tests for model adequacy fail to recognise them. Thus, although for all the four trend extrapolations, the normality assumption of the residuals cannot be rejected10, the important stationarity conditions all stand violated. The very low DW statistics for these four models indicate possible positive first-order serial correlation among the residuals that leaves the estimated standard errors unreliable. But this is not the only problem; the LM statistic is highly significant, such that the null of no residual autocorrelation gets rejected with almost certainty in all the four cases11. So does the null of no heteroscedasticity in errors for the linear and semi-log models. Thus in these two models, the observed residual autocorrelation may be due to the ARCH effect also12. In the k-transformation model too this is so (i.e., the error variance is serially correlated), but at 10 per cent significance level only, while in the quadratic trend model, at 25 per cent level. The joint parameter stability statistic is large enough to reject the null hypothesis of parameter constancy and of a strong model in all these cases; and so is the RESET (F-) statistic such that the null of no functional form mis-specification too is rejected (except the k-transformation model, for which these statistics could not be estimated); the observed autocorrelation can be due to mis-specified functions also.

The effects of first order autoregressive [AR(1)] correction on linear, quadratic and semilog trend models13 are also reported in Table 1. The first two models fail to recover in this exercise. The parameter instability persists; the data remain functionally mis-specified, and the residuals come out to be non-normal, serially correlated and heteroscedastic. The semi-log model, on the other hand, has tremendously improved, with no mis-specification. The errors are now statistically normal, serially uncorrelated and homoscedastic; the null hypothesis of joint parameter constancy cannot be rejected. The series appears to be almost stationary after the ‘quasi-differencing’, involved in the AR(1) correction of the logarithmically transformed series. And the model may pass safely for the next stage of hypothesis testing14. 14

The two autoregressive extrapolation models offer opposite behaviour patterns, though very low RESET statistics refute mis-specification in both the cases. Note that the coefficient of the lagged dependent variable used as the regressor in both the cases is almost unity! The residuals from the simple autoregressive equation are distributed highly leptokurtic, such that the normality test fails. The Durbin-h statistic for the simple AR model turns out to be 0.533, which is much less than the normal critical value of 1.645 at 5 per cent significance level, indicating non-rejection of the null of no first order serial correlation. However, the LM test confirms the presence of overall serial correlation in the errors, which are also heteroscedastic, as the White and ARCH tests indicate. The model is also fragile with joint parameter nonconstancy. The logarithmic autoregressive model, on the other hand, passes all the tests – the parameters are not unstable; and the residuals are normal, uncorreelated15, and nonheteroscedastic also.

Having thus proved that unwarranted application of extrapolation models for forecasting without model adequacy tests leads to misleading results, we now turn to examine the general practice of time series econometric analysis of electricity demand. It goes without saying that in the backdrop of a high standard of living, the distinctly evolved influential matrix of socioeconomic factors must have a significant say in determining electricity consumption in Kerala – for one thing, consider the spread effect of the ‘Gulf boom’, blooming the construction sector and the markets for electrical and electronic appliances. Hence the significance of a demand analysis.

In Table 2 we report the results of the econometric analysis of electricity demand (internal consumption) in Kerala for the period 1960-61 to 1998-99. The ‘causal’ factors considered are the ones usually used in such studies – per capita state income (at 1980-81 prices), number of consumers (in the place of population), and real average revenue (average sales revenue deflated by wholesale price index number for electricity, base: 1981-82; as a proxy for average price). The results are mixed for the two types of models (simple and logarithmic) considered, though all the first four models suffer from parameter instability16. Surprisingly, the logarithmic model (Model 2) is haunted by ‘spurious regression’ effect, which persists even in the presence of a time trend, included to ‘detrend’ the variables (Model 4). In the simple Model 1, DW test result is inconclusive, but there is no presence of it when a time 15

trend is included (Model 3). Except this one, all the other three models are functionally misspecified also. Normality and White homoscedasticity assumptions are violated for the simple models (1 and 3) without and with trend, though there is no ARCH effect; and by the LM tests, residuals from all the models are autocorrelated.

A final model (Model 5), including a one-period lagged dependent variable in the logarithmic mould, comes out to be non-fragile with normally distributed and homoscedastic residuals. Note that this is the usually used partial adjustment (short run consumption) model17, 2

appearing here with appealingly significant R and t-values. However, the large Durbin-h statistic strongly rejects the null of no serial correlation18; the LM (F-) test also confirms this. And the no-mis-specification null too remains rejected. Note that the lagged dependent variable in this model, unlike in the above autoregressive cases, has not (as it should have) biased DW towards 2.

The upshot of the whole exercise brings into light an important aspect in model building, in terms of the significant results of the diagnosis for model adequacy of the two extrapolation models: semi-log trend model with AR(1) correction and first-order logarithmic AR model. That important aspect is that both the models involve logarithmic transformation and ‘quasi’ differencing of the consumption series that could induce to some extent stationarity in the nonstationary series19. And this induced stationarity is reflected here through the whiteness of the residuals.

It should be noted that while the above two models pass the diagnosis, its failure marks the multivariate models, which might otherwise pass all hypothesis and simulation significance tests and mislead a researcher.

Our intention of this presentation has been to bring it home that a non-judicious handling 2

of regression techniques (considering only the significance of R and t-values, as also the simulation error measures) for time series analysis/forecasting could be misleading. Most macroeconomic time series being non-stationary, a fixed-coefficients model building endeavour is just undesirable. Successful infusion of stationarity into non-stationary series, however, depends on the right choice of the appropriate method – detrending or differencing. And this in turn depends upon the factual recognition of the true nature of these series; i.e., whether they 16

belong to TSP class or DSP class. In any analysis based on time series, an identification exercise for the series must then precede the model building stage, because of possible problems of misleading results particularly of under-differencing20 (i.e., modeling a DS series as a TS series). This is a possible problem with trend extrapolation models. For instance, electricity consumption in Kerala being a DS series with a unit root (this will be proved later on), its underdifferencing in the above trend models results in misleading results of spurious regression. In this light should we consider the common practice of estimating trend growth rate from semi-log (exponential) trend model. If model adequacy tests are significant after first order AR 2

correction, carried out in view of R  DW, the coefficient of t may be interpreted as the trend growth rate.

In this context we propose another useful model – a partial adjustment (short run) growth rate model, regressing logarithmic consumption on its own first order lagged term and time (Table 1, Model 10). Most of the results from this model are the same as those from the semilog trend model with AR(1) correction (model 9), such that the two models are equivalent, since the presence of the lagged dependent variable as a regressor has the same quasidifferencing effect (in Model 10) as AR(1) correction (in Model 9); the two parameter estimates are equal (0.798). The advantage of using Model 10 is that it gives a short run growth rate (coefficient of t = 0.0136), a coefficient of long run adjustment (0.798) and a long run growth rate (0.0136/(1 – 0.798) = 0.0676), the same trend growth rate from Model 9.

In line with our interpretation of residual based cointegration test as a model adequacy diagnostic checking, we have applied unit root (DF) tests to the residuals from these three significant models (7, 9 and 10), and found no unit root in the noise functions, thus reconfirming their whiteness21. These models are thus adequate. The logarithmic AR(1) model (7) is a random walk with drift; the intercept (0.231) gives the approximate growth over the previous period.

Our linear and quadratic extrapolation models with and without AR(1) correction, as well as the econometric models 3 and 4 (models with time trend) are good illustrations of underdifferencing. In the absence of a detailed model adequacy diagnostic checking, the ‘high significance’ of these models would have fascinated and thus misled an average researcher; and so it has been, unfortunately, in the case of almost all the previous studies on electricity/energy consumption in India. 17

Unit Root Tests and Cointegration Analysis

We now therefore turn to the starting point of our time series analysis, viz., the identification stage: finding out whether our series belong to TSP or DSP class. The series we analyse are: (internal) electricity consumption (C in MU) in Kerala, number of consumers (N), average price (revenue) (AR, paise per unit) (all during 1957-58 to 1998-99) and per capita State income (PCI, in constant Rs., during 1960-61 to 1998-99). All the variables are in logarithms; logarithmic transformation is expected to reduce the effects of a time-varying variance in a series and make it stationary22 (Holden, et al. 1990: 64).

Following Nelson and Plosser (1982), we base upon the Bhargava-type formulation of two ADF test models, our conclusion and interpretation of the unit root test results under the null and alternative hypothesis – one with a trend (including constant) and the other with a constant only. In the ADF test model, the specification of the lag length assumes that the residual (ut) is white noise. Hence the optimum lag length (2 for C and PCI and 3 for N and AR in levels and 1 for all in differences) is selected so as to achieve empirical white noise residuals23, satisfying normality, stationarity and homoscedasticity assumptions (Table 3). The selected lag was favoured by Akaike information criterion also. The univariate ADF unit root test results are reported in Table 4.

The DW-statistic for the level of a variable (yt) is a simple indicator of its integrated property, and therefore we also report the DW-statistics for the concerned level variables. If yt is a random walk (with or without drift), DW will be close to zero, and if it is white noise, DW will be around two24. The DW-statistics obtained of the levels of (the logs of) C, N, and PCI are close to zero and that of AR is also small, indicating the integratedness of these variables. The univariate ADF test results also show that the unit root null cannot be rejected in all the cases – that is, all the series we consider do belong to (drifting) DSP class. We further check for another unit root in the series. The DW-statistics for the first-differences of (the logs of) C and AR are around two, suggestive of their whiteness, but those for N and PCI are small, giving some signs of integratedness. The ADF tests, however, fail to find any more unit root, and hence we maintain that all these series are I(1), not I(2). 18

Is this inference influenced by the effects on the ADF test statistic of structural change in the series? To find out whether any significant structural change has tended to taint the test statistic in favour of non-rejection of the unit root null in each case, we apply Perron’s (1989) unit root test in the presence of structural breaks. Graphical analyses25 identify three possible breaks of ‘crashes’ (and subsequent ‘growth rises’) in the time series of electricity consumption in 1983-84, 1987-88 and in 1996-97, and a ‘growth leap’ in the series of customers’ number in 1979-80 and in the series of per capita income in 1985-86. The average price series appear very much erratic and fail to help us recognise any trend break in its temporal behaviour.

The infamous power famine inflicted on the pure hydro-power system of Kerala by a series of drought since the turn of the 80s in league with the defective capacity expansion planning explains the ‘crashes’ in the power consumption series. At the same time, demand has been on the rise at an increasing rate reinforced by an ever-growing number of new connections as well as connected load. In 1983-84, consumption fell by about 7.2 per cent over 1981-82, and then rose by 25 per cent in the next year; a fall of 4 per cent in consumption in 1987-88 over 1985-86 was followed by an increase of 21 per cent in the next year, and a fall of 5.3 per cent in 1996-97 over the previous year, by a rise of 10 per cent in 1997-98. The growth in the number of consumers got an accelerated fillip with the commissioning of the Idukki (Stage I) power project in 1976-77, and by 1979-80 the growth trend started to shoot up, but only to lose some momentum during the shortage period. In (constant) per capita income series, an insignificant growth is discernible after 1964-65 for a few years; from 1970-71, the series appears stagnant for about one and a half decade, and then from 1985-86, significant growth pushes the series up forcefully – a manifestation of the ‘gulf boom’ in a liberalised economic atmosphere.

In the face of such apparent breaks in these series, we subject them to Perron (1989)’s unit root test and the results are presented in Table 5. The optimum lag is identified such as to achieve white noise residuals here also. The coefficients of t, DU and DT in the Perron’s unit root test regression models turn out to be insignificant in the case of consumption with break years of 1983-84 and 1987-88. These coefficients are significant for the consumption series with a break in 1996-97 (at 10 per cent level only), for customers' number with break in 1979-80, and for per capita income series with break in 1985-86. However, considering the estimated Perron’s

19

test statistic, in no case is it significant even at the 10 per cent level, reconfirming the presence of unit root in these series.

The series thus being integrated of the same order, i.e., I(1), we next turn to check whether the power consumption series has a long-term relationship with other variables under consideration, that is, whether there exists an economically meaningful cointegrating vector (cv) among these variables, using

the two commonly used cointegration tests namely, the

(Augmented) Engle-Granger (AEG, Engle and Granger 1987) test and Johansen and Juselius (1990) test. Since the cointegration test results are sensitive to the lag length of the VAR model (Hall 1991), optimum lag length for cointegration test is determined on the basis of the residual mis-specification tests of the VAR model. For a lag length of 2, the VAR model residuals have been found to be strictly white noise (Table 3).

As a first step, we compare the CRDW statistic of 0.635, obtained from a logarithmic model26 of electricity consumption (C) with number of consumers (N), per capita State income (PCI) and average price (AR), with the approximate critical value of 0.641 at 5 per cent significance level, and fail to reject the null of no cointegration among the variables (even 2

though the R is close to unity, which is an indication of cointegration). Next we go to the AEG procedure to examine whether the residuals from this relationship are stationary, I(0). The results up to 2 lags are reported in Table 6. Here too the non-rejection of the null of no cointegration (or of I(1) residuals) persists even at 10 per cent significance level for all the lags up to 2. Hence, for reconfirmation we turn to the JJ method, which provides more robust results when there are more than two variables (Gonzalo 1994). The JJ cointegration test results are given in Table 7, where we use the maximum eigenvalue and trace statistics with small sample correction (Reimers 1992). Starting with the null hypothesis of no cointegartion (r = 0) among the variables, we find that both the corrected maximum eigenvalue and trace statistics27 are well below the respective 95 per cent critical values, further confirming non-rejection of the null of no cointegration among these variables at 5 per cent level of significance; i.e., there are no common stochastic trends and the system contains four unit roots. Hence we conclude that the cointegrating regression is spurious: the regression residual is an I(1) process and there is no equilibrium in the levels of the variables (Phillips 1986). Hence the analysis should now be proceeded with on their differences. 20

We continue with these testing procedures to see if there exists any significant relationship for C with different possible combinations of the three ‘causal’ variables. Thus, for instance, we consider the logarithmic model of C with N and PCI; the CRDW statistic is 0.618, less than the critical value at 5 per cent significance level; all the AEG test statistics up to lag 2 are also less (in absolute value) than the respective critical values, even at 10 per cent level (Table 6). The JJ test also fails to reject the null of no cointegration among the three variables now considered (Table 7). Continuing with other combinations, we find that there exists statistically no relationship at all for C with any of the three proposed ‘causal’ variables28.

Causality in Growth Models

The result that there exists no meaningful cointegrating vector of interest among the variables considered (that any linear combination of these integrated variables still remains integrated) deprives us of taking advantage of a valid error correction representation29, and thus analysing the relationship among the variables in their levels, without losing valuable long run information. This leaves us with the only option of differencing the set of variables, proved to belong to DSP class, prior to further analysis. Differencing, as already noted above, is recommended for integrated series (Granger and Newbold 1974); taking differences of logarithmic series is approximately equivalent to using rates of growth of the series. Hence the significance of growth rate models, expressing relationship among variables in terms of their growth rates, that is first differences of their logarithms.

All the four I(1) series in our consideration are therefore first-differenced and the resultant stationary series (as proved by ADF tests earlier) of growth rates come in for possible choice as candidates in a growth rate model. This selection is carried out in terms of the significance of the variables (in growth rates) in a temporal lead-lag relationship, to find, through pair-wise Granger-non-‘causality’ tests, whether the growth in N, AR and PCI are the leading indicators of the growth in C. Remember that capacity expansion planning is based on possible growth in demand from a growing number of consumers in conjunction with price and income. The results are reported in Table 8. In none of the cases we can reject the null hypothesis of pair-wise Granger-non-‘causality’. That is, the annual growth rates of electricity consumption are not granger-‘caused’ by those of any of the three variables, each considered in turn. Similarly, there has been no significant temporal feedback from annual growth rates of 21

electricity consumption to those of any of the other variables considered30. Since the Grangernon-‘causality’ test is very sensitive to the number of lagged terms included in the model, it is recommended that more rather than fewer lags should be used. Hence we have considered lags up to 10, obtaining the same result of non-rejection of the null31.

The Less Correlatables Dissected

A third strain in our scepticism about the earlier studies in general relates to their efforts of correlating the less correlatables. In an underdeveloped power system like ours, plagued with long-run constraints of inadequate and unreliable supply, electricity consumption remains an input too insignificant to our economic life to be analysed in the framework of some macroeconometric ‘causality’ models, as is usually done in the context of advanced systems. The second part of our analysis in terms of cointegration and Granger-‘causality’ confirms this at least in the case of Kerala power system. Electricity consumption in the State, coupled with the usually selected ‘causatives’ of number of consumers, per capita income and average price, all being non-stationary variables (of the same order of integration), fails to be explained in a cointegrating space in any combination. All the linear combinations examined turn out to be still non-stationary. Further analysis for identifying some temporal lead-lag relationship (Granger‘causality’) among them in terms of their annual growth rates, found to be stationary, again draws blank. These two unusual results are a potent pointer to the badly constrained electricity consumption in an underdeveloped system, devoid of its inherent growth mechanism (even from the number of customers itself). The general evolution of the economy may have dragged it up along some of its trend.

The surprising result that none of the three variables considered is eligible to be included in the growth rate model of electricity consumption in Kerala also leaves us finally with no further scope for multivariate time series regression analysis of demand, despite the seemingly significant scope for electricity demand analysis in Kerala, having a high standard of living. However, these results do make some sense in an underdeveloped power system like ours, plagued with substantial supply bottlenecks. Our scepticism on applying regression method directly to non-stationary series should also descend upon the common practice of attempting to correlate the less correlatables. Estimating GDP-electricity use elasticity in industrialised 22

countries where electricity service contributes significantly to everyday life has become a standard tool of simple analysis for some obviously general conclusions. However, in a comprehensive international comparative study of bivariate ‘causality’ between energy use and GNP of five countries, Yu and Choi (1985) found no ‘causality’ in the US, the UK and Poland, but observed a unidirectional ‘causality’ from energy consumption to GNP in the Philippines and a reverse ‘causality’ from GNP to energy consumption in South Korea. Recently, Cheng (1995) detected in a multi-variate framework no ‘causality’ from energy consumption along with capital to economic growth in the US. In another study (Cheng 1997) for the Latin American countries, he found ‘causality’ from energy use to economic development in Brazil, but not in Mexico and Venezuela. In a most recent study for India (Cheng 1999; the first of its kind in his knowledge), Cheng found, in a multi-variate model, no ‘causality’ from energy consumption to economic development, ‘which in general is consistent with many previous studies of other countries’ (ibid.: 47), but saw a reverse ‘causality’ from GNP to energy use, using Hsiao’s version of the Granger-‘causality’ method.

We argue, however, that it may be unfair to map such elasticity/‘causality’ methodology on to an alien range in an underdeveloped power system where the contribution of the service of electricity is insignificant. This is so even in the industrial sector in India, where power remains too insignificant an input32, highly substitutable by capital and/or labour, primarily because of inadequate and unreliable supply, which has become a long-run experience.

The methodology is questionable even in the industrialised sector power demand analysis in a less industrialised region like Kerala, that too with very limited number of electricity-intensive industries. Adoption of this methodology here then amounts to correlating the national/State domestic product or industrial product exclusively with an insignificant input, in violation of the ethics of a consistent and logical analytical exercise, and results in gross specification error. Moreover, there are a large number of small scale and cottage industries that use practically little electricity but together contribute significantly to the industrial product. About 41 per cent of the net State domestic product in 1997-98 that originated in the manufacturing sector in Kerala was contributed by the unregistered firms, most of which use little electricity33. It was found in 1994-95 that about 66 per cent of the enterprises in rural India and about 52 per cent in urban India did not use any energy in their manufacturing process (Government of India 1998 b: ii). Only 7.9 per cent of the rural firms and 30.4 per cent of the 23

urban firms in India (and 11.9 and 17.7 per cent respectively in Kerala) are reported to have used some electrical energy in their production process in that year (ibid.: 35-36). The contribution of the unorganised manufacturing sector, on the other hand, in terms of gross value added to the national economy in 1994-95 was estimated at Rs. 32,274.89 crores, out of which 41 per cent came from the rural sector; and that to the Kerala State’s economy was at Rs. 646.64 crores, with 72.1 per cent from the rural enterprises (ibid.: 57-64).

There is in this respect another aspect also. Kerala experienced one of the worst drought and the consequent power famine in 1983-84, with year-long imposition of 10 to 100 per cent power cut on industries. However, it had surprisingly no negative effect on the growth of industrial output. The contribution of the manufacturing sector to net State domestic product at current prices rose by 5.4 per cent over 1982-83, and that at constant prices showed a marginal increase of 0.2 per cent in the registered manufacturing sector and a fall of 12.3 per cent in the unregistered sector – this fall in the unregistered sector continued for the following years almost till the turn of the nineties, including normal periods, indicating the influence of some other factors. Though power cut was in force in the three years from 1986-87 to 1988-89, the contribution at constant prices of the manufacturing sector, though declined by 10.5 per cent in 1986-87, shot up in the following years (at 14.4 and 12.8 per cent respectively); at current prices, however, it was steadily on the rise. The registered sector had the very same pattern. The growth trend without a break continued in the following years too. In 1996-97, even 35 to 100 per cent power cut had no adverse effect on manufacturing (both the sectors) contribution (at both constant and current prices). As already explained, all the earlier studies in India on electricity demand34 have invariably used as ‘causal’ regressors national/State (per capita) income, average sales revenue and population. The first two variables in aggregate values conceal everything of the characteristics of the units into which the analysis is paradoxically intended to make a look. It goes without saying that the time series regression with these variables, even if valid in a cointegrating space, yields only the macro level elasticities over time. Consumption elasticities in the true sense, i.e., across different income categories and tariff blocks, get suppressed in this aggregation. Moreover, the use of time series data simply ignores the possibility of changes of the intercept (that on an average accounts for the influence of factors other than those considered in the model) and of the slope of the line (that reflects the average intensity of energy 24

consumption with respect to that variable); using dummy variables to account for significant structural changes might result in increasing loss of degrees of freedom. Choice of a suitable deflator also poses problems.

Average sales revenue as a proxy for average price is pregnant with a danger of measurement error as well. Proper estimation of price elasticity of demand requires, (in conformity with the suggestion made by Taylor long back, 1975, 1977), the use of data on actual rate paid by a cross-section of consumers, rather than the aggregate average revenue (to the utility) over time. Average revenue might be an indicator of the supply price in the aggregate, but never a representative of the demand price, the particular price a customer is faced with at a decision making juncture, especially in the context of the block rate tariff system. Average tariff rate might be a better alternative here.

This too in its aggregation, however, conceals an important implication of block rate structure, that makes the price of electricity itself a function of consumption, since in the increasing block rate tariff prevalent in Kerala/India, the price to a customer rises as the volume of consumption increases. This in turn entails a simultaneous equations system for electricity demand and price35 across customer categories in different blocks of tariff. Temporal effects through changes in intercept and slope can be checked and explained, if pooled time series cross section data are used in this model. Thus a better alternative is electricity demand analysis based on pooled data, subject to appropriate unit root tests. However, such data-base is not at all available in India; and at best we can have only a cross sectional primary survey for study.

Even here the very elasticity of electricity demand is open to serious questioning. The price elasticity of demand loses its relevance in an underdeveloped power system such as ours. Demand for electricity remains largely unresponsive or less responsive to its price as it has almost become a necessity for the basic need of lighting for the habitual customers. In fact some studies have shown that even the domestic customers are willing to pay much higher prices for uninterrupted supply (Upadhyay 1996, 2000).

At the same time, energy consumption

commands a substantially lower budget share due both to lower unit price and to low consumption level. Thus for example, the share of fuel and power in the total private final consumption expenditure in the domestic market (at current prices) in India in 1997-98 was just 3.29 per cent; and electricity consumption accounted for only 0.68 per cent in the total. In 198025

81 and 1990-91, the share of fuel and power was 4.64 and 4.52 per cent respectively, and that of electricity, 0.40 and 0.62 per cent respectively36. The growth in electricity consumption has not been strong enough to facilitate a pronounced rate of substitution for other fuels, especially, the traditional one, kerosene oil. The percentage share of electricity in the private final consumption expenditure on total fuel and power grew from 8.63 per cent in 1980-81 to 20.57 per cent in 1997-98, marking an average annual compound growth rate of 5.24 per cent, while that of kerosene oil fell from 15.22 per cent to 10.5 per cent only over the same period at a decay rate of (-) 2.16 per cent per annum. This in turn suggests a very weak marginal rate of substitution of electricity for kerosene oil (or elasticity) of just about (-) 0.40; i.e., one percentage increase in the share of electricity consumption expenditure could on an average substitute for (or induce a fall of) 0.4 percentage in the share of kerosene oil consumption expenditure. In short, electricity could not yet make an effective inroad upon the economic life in India in general to the extent it should have done.

For a more concrete example, let us consider the case of the connected consumers themselves. The per capita electricity consumption of the connected domestic customers (that made up about 75 per cent of the total customers) in India in 1995-96 was 772.32 units at an average rate of Ps. 95.94 per unit (for 18 State Electricity Boards), thus giving in general an average per capita electricity consumption expenditure of Rs. 740. 97 (or, Rs. 61.75 per month) – only 7.04 per cent of the per capita income (of Rs. 10524.8) of that year. In the case of Kerala State, the electricity consumption per (electrified) domestic consumer in 1997-98 was 953.72 units at an average rate of Ps. 76.96 per unit, that indicates an average domestic consumption expenditure of Rs. 733.99 (or Rs. 61.66 per month) on electricity. The domestic sector that made up about 75 per cent of the total customers nearly consumed 50 per cent of the electricity sold in Kerala in that year. In general, the consumption of electricity per connected consumer in Kerala in 1997-98 was 1480.71 units at an average rate of Ps. 123.74 per unit, giving an average electricity consumption expenditure of Rs. 1832.16 (or, 152.68 per month) – only 15.35 per cent of the per capita income (of Rs. 11936) of that year. Similarly, as a substantial share of residential and commercial electricity consumption goes to serve the basic need of lighting which is fairly unresponsive to income rather than to more income elastic, luxury end uses, power demand remains less income elastic also to this extent. Moreover, the whole edifice of demand analysis crumbles to dust in an encounter with power cuts and load shedding, that

26

restrict actual consumption to availability rather than to actual requirement which is the long run experience of Kerala.

We thus see that the economic relationship demand is hypothesised to have with (per capita) income and unit price is weak and hence unwarranted in the case of an underdeveloped power system such as in India/Kerala. The same is true for the role of the demographic variable viz., population too in power demand analysis. As annual population figures are only interpolated ones, they might contain a systematic pattern, causing residual serial correlation that might not be there in the original data; using these data thus involves analytical problems. Moreover, since in a less developed power system, electricity connection remains inaccessible to a large section of the population37, number of consumers, instead of population, must be accepted as a more direct and right determinant. The growth of demand for power is generally assumed to be determined by the growth of number of (connected) consumers and that of intensity of their power consumption (i.e., electricity consumption per customer), as also the interaction between these two factors38.

Till the turn of the Eighties, Kerala had apparently been a power surplus state, exporting power to neighbouring states. Since the drought year of 1982-83, unprecedented power shortage has become a part of life in the state. Recurring drought coupled with inadequate installed capacity has thus unleashed a reign of power cuts and load shedding, constraining the actual demand down39. Reliance on past demand data for forecasting purposes thus becomes grossly erroneous and highly questionable. If some measurement of these shortages is possible to be made, the constrained demand can be adjusted accordingly to arrive at a probable measure of unsuppressed demand, which in turn can be used as data base for forecasting, subject to the unit root constraints (unless it contains any induced pattern). One method is to assume first that when restrictions are imposed on consumers, their level of consumption is held at some fraction of their consumption during an earlier base period. Then the shortfall in supply equal to these percentage restrictions can be found and inflated by a factor that reflects suppressed growth in demand since the base period and the impact of unscheduled load shedding. This in turn can be used to adjust the suppressed demand data (World Bank 1979: 13). Another method uses as demand inflative factor, the fraction of customers affected by load shedding during peak period and thus deprived of chance to contribute to peak period demand. The main problem with all such methods is the non-availability of accurate data and information. 27

3. Electricity Demand Projection for Kerala

Demand forecasting in such contexts becomes highly uncertain. To the extent that the demand forecast has nothing to do with the capacity expansion planning on a bounded budget as well as with the actual materialised capacity additions in the system, the very exercise becomes futile, except as some routine liturgy. The widening gaps between the actual consumption and the forecast levels (even with the revised lower ones of the 15th APS or of the KSEB-State Planning Board), in the last few years in Kerala prove this point. Accurate demand forecasting is relevant as well as essential only in a growing system under an efficient management directed by a government of determined political will. This notwithstanding, forecasts under such circumstances, however, do serve a good purpose of quantifying (through the gap between forecast and the actual) the unsatisfied demand, the extent of the shortage.

As a corollary to the above implications of our results, it is high time we questioned the inappropriate, pedantic, practice of linear regression mapping for trend extrapolation not only for the unit root problems, but also when similar results can be generated by means of much simpler methods, for example, growth rate based projections40. Useful short-term projection can be had from simple annual growth rates (percentage deviation over previous year) of electricity consumption. The method can be modified by accounting for the effect on consumption of possible growth of the direct causatives such as number of consumers and connected load. Below we suggest one of such models: lnCt = CN rN + lnCt1

where

CN

= lnC/lnN is the elasticity of consumption (C) with respect to number of

consumers (N) or consumption intensity factor and rN is the growth rate of N. The above relation is in fact an identity only41. The model can be modified to include the effect of connected load (of electrical appliances) also by rewriting CN as

CN = CLLN, where CL is the consumption

28

intensity with respect to connected load and

LN

is the load intensity of the customers.

Moreover, the expression resembles the ‘explained’ part of a random walk with drift.

Consumption intensity of the power customers in general in the State was quite elastic (much more than unity) in normal years. It even went up to more than 2.5 during the two years of 1966-68 when the Board became liberal in giving new connections following the commissioning of the Sabarigiri project, and more than three in 1984-85 and 1988-89 immediately after the ‘crashes’ of 1982-84 and 1986-88. These years saw great leaps in electricity consumption (the growth rates being between 20 – 33 per cent over the previous year) of a fast-growing number of customers. However, as energy export picked up, the consumption elasticity fell below unity; the ‘informal’ constraints on internal electricity use, covertly imposed in order to boost export show, continued till the draught year of 1982-83; growth in new connections was also checked during most of these years. Once the export frenzy has subsided, consumption now grows subject only to the combined constraints of inadequate capacity and monsoon failure, eased to some extent by heavy imports. And the consumption elasticity in the recent years has been well above 1.5, the growth in new connections being about 7 per cent.

In forecasting electricity demand using past data, especially in Kerala, one should be very much wary of the supply-constrained low-range of the past consumption series. Two periods in this time series can be distinguished when the inadequate supply constrained the demand to a minimum – one due to the generally recognised effect of the prolonged and severe drought since 1982-83, coupled with the inadequate generating capacity, and the other due to the less recognised effect of the covertly boosted energy export especially since the mid-1970s.42 Though export I perceptible quantum started after the commissioning of Sabarigiri project in 1966 and 1967, it was not at the cost of the internal consumption of energy, as evidenced by the growth of the parameters over the 11 years from 1965-66 to 1976-77, when Idukki project (phase 1) was put on line: number of consumers and connected load as well as internal energy consumption in the State all grew at an annual compound rate of about 10 per cent during this period. Generation grew at a rate of 12.8 per cent, and total energy sales at 12.3 per cent; and energy loss at 8.1 per cent per annum.

After the commissioning of the Idukki project (phase 1), the scenario fast deteriorated; energy export picked up in an attempt to boost the ‘power surplus’ image of the State; the 29

system contrived to constrain the internal energy consumption to a minimum. It grew at an annual rate of just 6.4 per cent during 1976-77 to 1981-82 (the eve of the drought period) against a growth rate of 13 per cent registered in the number of consumers and 9 per cent in connected load. Generation increased at a rate of 11.9 per cent and total sales at 11.7 per cent, while energy loss at a higher rate of 13.8 per cent per annum.

Though the export drives ceased with the onslaught of the drought since 1982-83, the supply constraints on consumption have still remained, now due to the inadequate generating capacity and hydro-fuel supply. Consumption has over the 17 years grown since 1981-82 along with the number of consumers and connected load at an annual compound rate of 6.9 per cent, while generation at just 1.9 per cent, heavy imports compensating for the fall in generation.

Thus the past electricity consumption series in Kerala contains the effects of two periods: a period of almost normal, unconstrained, consumption till 1976-77, and a period of supply-constrained one thereafter; the latter period remains further sub-divided into two: the initial period of export drives and the later period of drought. Similarly, the series of the number of customers (as well as of the connected load), as a direct determinant of electricity consumption, also remains normal till 1981-82 and constrained thereafter. Hence, any attempt to project and forecast electricity demand in Kerala using time series data must take into account these facts that determine the behaviour of the series.

Based on these observations and using the above relationship, we attempt here to project electricity consumption in Kerala in two scenarios: constrained and unconstrained ones from 1977-78 onwards.

We assume a consumption elasticity of unity in both the models. In the

constrained consumption model, number of customers is assumed to grow (rN) at the historical rates as obtained from the actual time series till 1998-99, and at 7 per cent per annum thereafter. In the unconstrained model, on the other hand, rN is taken from the actual series up to 1981-82 only, and an annual growth rate of 10 percent is assumed thereafter. The two scenarios, along with the actual consumption series and the 14th and 15th APS forecasts are given in the Table below. Notice that our constrained projection series is only marginally different from the 14th APS series, except for a few of the later years, during which our estimates are close to the 15th APS series; from 1995-96 onwards, our constrained series lies between the 14th and 15th APS

30

series. It is a very significant result, considering the much simple method we have used to estimate it.

Actual and Estimated Electricity Consumption in Kerala

Electricity Consumption (in Million Units) Actual

Projections Constrained

Unconstrained 14th APS

15th APS

1976-77

2137.3

2137.3

2137.3

1977-78

2331.3

2325.60

2325.60

1978-79

2419.3

2552.34

2552.34

1979-80

2384.4

2938.11

2938.11

1980-81

2767.4

3496.69

3496.69

1981-82

2911.5

4101.46

4101.46

1982-83

2839.98

4452.78

4532.81

1983-84

2703

4662.19

5009.53

1984-85

3376

5032.52

5536.39

1985-86

3776

5455.87

6118.66

1986-87

3697

5955.45

6762.16

1987-88

3626.5

6340.65

7473.35

1988-89

4387

6735.06

8259.32

1989-90

4794

7347.20

9127.96

6701

1990-91

5331.86

7965.07

10087.96

7326

1991-92

5596

8558.28

11148.92

8008

1992-93

5838.55

9113.87

12321.46

8686

1993-94

6234.16

9648.83

13617.32

9409

8567

1994-95

7027.69

10279.15

15049.47

10169

9383

1995-96

7414.62

10923.35

16632.24

10998

10116

1996-97

7020.74

11490.33

18381.46

11893

10921

1997-98

7716.23

12181.76

20314.66

12861

11770

1998-99

8960

13225.77

22451.17

13908

12666

14184.75

24812.38

15040

13617

1999-2000

31

2000-01

15213.26

27421.92

16264

14632

2001-02

16316.35

30305.91

17588

15756

2002-03

17499.41

33493.21

19020

16890

2003-04

18768.27

37015.72

20565

18106

2004-05

20129.12

40908.70

22243

19410

2005-06

21588.64

45211.10

24053

20808

2006-07

23154.00

49966.00

26011

22099

2007-08

24832.85

55220.97

28228

23557

2008-09

26633.44

61028.61

30418

25112

2009-10

28564.58

67447.04

32894

26770

2010-11

30635.74

74540.51

28537

The energy consumption estimates in the unconstrained scenario, on the other hand, is much higher than the 14th APS series. For 1998-99, the 14th APS estimate (13908 MU) is only 62 per cent of our unconstrained estimate of consumption (22451 MU), and for 2009-10 (32894 MU), about 49 per cent only (67447 MU). Evidently, the two consumption estimates (constrained and unconstrained ones) may be taken as the lower and upper limits of the ‘actual market’ demand. Assigning appropriate weights to the two scenarios, we can have a weighted mean series to represent this market demand.

Now, assuming 60 per cent load factor, the constrained consumption estimate for 19992000 implies a maximum demand of 2698.8 MW, which, accounting for 18 per cent loss factor (as at present), entails an available capacity of about 3291 MW. The total installed capacity of the State in that year is reported to be 2391.18 MW only (Government of Kerala 2000: 91). --------------32

Table 1 Estimation Results of the Forecast Models Period: 1957-58 to 1998-99 1. Linear Trend: Consumption = f (Time) Estimate t-value Adj. R2 F-value DW statistic Constant -899.54 -3.76 0.895 348.62 0.157 Time 181.02 18.67 Parameter instability: 3.159** Simulation Error Analysis TIC BP VP CP 0.099 4.90E-17 0.027 0.973 Residual Analysis 2 Skewness Kurtosis SD Normality ( ) 3.35 (0.1870) 0.681 3.25 752.21 Autoregression (F) Heteroscedasticity (F) ARCH (F) RESET (F) 30.03 9.11 12.95 18.38 (0) (0.0006) (0) (0) 2. Quadratic Trend: Consumption = f (Time, Time Squared)

Constant Time 2 Time

2

Normality ( ) 0.631 (0.7294) Autoregression (F) 10.94 (0)

Estimate t-value Adj. R2 F-value DW statistic 725.61 4.68 0.982 1088.1 0.706 -40.595 -2.44 Parameter instability: 2.371** 5.1540 13.750 Simulation Error Analysis TIC BP VP CP 0.041 8.05E-16 0.004 0.996 Residual Analysis Skewness Kurtosis SD -0.056 3.59 311.11 Heteroscedasticity (F) ARCH (F) RESET (F) ++ 1.39 10.70 (0.2548) (0.0002)

3. Semi-Log Trend: ln Consumption = f (Time) Estimate t-value Adj. R2 F-value DW statistic Constant 6.082 141.94 0.977 1773.7 0.317 Time 0.073 42.12 Parameter instability: 3.437* Simulation Error Analysis TIC BP VP CP 0.0086 2.10E-12 0.0056 0.994 Residual Analysis 2 Skewness Kurtosis SD Normality ( ) 1.44 (0.4864) 0.119 2.12 0.135 Autoregression (F) Heteroscedasticity (F) ARCH (F) RESET (F) 10.34 10.14 5.69 13.28 (0) (0.0003) (0) (0) 33

4. k-transformation (k = 0.5): Consumption = f (Time)

Constant Time

2

Normality ( ) 2.08 (0.3541) Autoregression (F) +

Estimate t-value 7.26 3.67 1.937 31.530 Simulation Error Analysis TIC BP 0.05 0.017 Residual Analysis Skewness Kurtosis -0.29 3.92 Heteroscedasticity (F) ARCH (F) + 2.16 (0.0845)

Adj. R2 0.973

F-value 1479.5

VP 0.0074

CP 0.976

DW statistic 0.487

SD 377.14 RESET (F) +

5. First-order Auto-regressive: Ct = f (One-period lagged Ct) Estimate 15.296 1.068

Constant Ct-1

2

Normality ( ) 17.50 (0.0002) Autoregression (F) 7.25 (0.0001)

t-value 0.22 55.560

Simulation Error Analysis TIC BP 0.034 1.33E-14 Residual Analysis Skewness Kurtosis -0.387 6.11 Heteroscedasticity (F) ARCH (F) 10.82 2.78 (0.0002) (0.0355)

6. Logarithmic Auto-regressive: ln Ct = f (ln Ct-1) Estimate t-value Constant 0.231 2.24 0.9799 72.99 ln Ct-1

2

Normality ( ) 1.37 (0.5039) Autoregression (F) 1.45 (0.2322)

Simulation Error Analysis TIC BP 0.005 9.99E-14 Residual Analysis Skewness Kurtosis 0.411 2.64 Heteroscedasticity (F) ARCH (F) 0.215 0.336 (0.8073) (0.8872) 34

Adj. R2 F-value DW statistic 0.987 3087.1 1.837 Durbin h: 0.533 Parameter instability: 1.149* VP 0.003

CP 0.997

SD 260.94 RESET (F) 0.076 (0.9132)

Adj. R2 F-value DW statistic 0.993 5327.46 1.873 Durbin h: 0.414 Parameter instability: 0.216 VP 0.002

CP 0.998

SD 0.075 RESET (F) 0.369 (0.6936)

7. Linear Trend with AR(1) correction: Consumption = f (Time) Estimate t-value Adj. R2 F-value DW statistic Constant -1983.66 -0.19 0.987 1510.03 1.810 Time -100.570 -0.25 Parameter instability: 1.545* AR(1) 1.045 16.40 Simulation Error Analysis TIC BP VP CP 0.034 4.95E-11 0.003 0.997 Residual Analysis 2 Skewness Kurtosis SD Normality ( ) 14.81 (0.0006) -0.334 5.87 260.4 Autoregression (F) Heteroscedasticity (F) ARCH (F) RESET (F) 6.95 8.78 3.64 3.35 (0.0002) (0.0007) (0.0109) (0.0457) 8. Quadratic Trend with AR(1) Correction: C = f (Time, Time Squared) Estimate t-value Adj. R2 F-value DW statistic Constant 1413.96 1.80 0.989 1162.47 1.625 Time -110.12 -1.55 Parameter instability: 2.185* 2 6.65 4.72 Time AR(1) 0.709 4.99 Simulation Error Analysis TIC BP VP CP 0.031 5.60E-13 0.003 0.997 Residual Analysis 2 Skewness Kurtosis SD Normality ( ) 11.34 (0.0034) -0.481 5.39 239.37 Autoregression (F) Heteroscedasticity (F) ARCH (F) RESET (F) 5.76 ++ 4.99 3.67 (0.0007) (0.0019) (0.0350) 9. Semi-log Trend with AR(1) correction: ln C = f (Time) Estimate t-value Adj. R2 F-value DW statistic Constant 6.25 39.54 0.993 2927.36 1.766 Time 0.068 12.82 Parameter instability: 0.810 AR(1) 0.798 9.540 Simulation Error Analysis TIC BP VP CP 0.004 1.13E-13 0.002 0.998 Residual Analysis 2 Skewness Kurtosis SD Normality ( ) 0.095 3.05 0.07 0.066 (0.9674) Autoregression (F) Heteroscedasticity (F) ARCH (F) RESET (F) 1.36 0.363 0.202 0.091 (0.2655) (0.6977) (0.9591) (0.9132)

35

10. Partial Adjustment (Short-run Growth Rate) Model: ln Ct = f (ln Ct-1,Time) Estimate t-value Adj. R2 F-value DW statistic Constant 1.31 2.61 0.993 2927.36 1.766 Time 0.0136 2.20 Durbin h: 0.887 0.798 9.54 Parameter instability: 0.810 ln Ct-1 Simulation Error Analysis TIC BP VP CP 0.004 1.13E-13 0.002 0.998 Residual Analysis 2 Skewness Kurtosis SD Normality ( ) 0.095 3.05 0.07 0.066 (0.9674) Autoregression (F) Heteroscedasticity (F) ARCH (F) RESET (F) 1.36 1.77 0.202 2.36 (0.2655) (0.1562) (0.9591) (0.1082) Note: 1. *' and '**' indicate statistical significance at 5 and 1 per cent respectively. 2. + = not available in non-linear least squares 3. ++ = near singular matrix 4. Figures in brackets are the corresponding p-values. 5. C = Electricity Consumption in the State (Million Units) 6. Adj. R-squared = Adjusted R-squared; 7. TIC = Theil inequality coefficient 8. ln = Natural log 9. BP = Bias proportion; VP = Variance proportion 10. CP = Covariance proportion 11. AR(1) = Estimate of first order auto-regression coefficient 12. Parameter instability = Joint (F-) test statistic for parameter constancy 13. ARCH (F) = Autoregressive Conditional Heteroscedasticity (F) statistic (5 lags) 14. RESET (F) = Regression Specification Test (F) statistic

36

Table 2 Multi-variate Econometric Models Period: 1960-61 to 1998-99 Model 1. Ct = f(Nt, PCIt, ARt)

Constant Nt PCIt ARt

2

Normality ( ) 14.19 (0.0008) Autoregression (F) 3.08 (0.0233)

Estimate t-value Adj. R2 F-value DW statistic -129.66 -0.23 0.993 1686.1 1.306 1.175 18.53 Parameter instability: 1.591* 1.127 4.23 -30.53 -2.30 Simulation Error Analysis TIC BP VP CP 0.024 1.64E-14 0.002 0.998 Residual Analysis Skewness Kurtosis SD -1.161 4.83 191.94 Heteroscedasticity (F) ARCH (F) RESET (F) 2.64 0.443 3.44 (0.0340) (0.8145) (0.0434)

Model 2. ln Ct = f(ln Nt, ln PCIt, ln ARt)

Constant ln Nt ln PCIt ln ARt

2

Normality ( ) 1.006 (0.6048) Autoregression (F) 6.08 (0.0005)

Estimate t-value Adj. R2 F-value DW statistic 0.237 0.27 0.990 1266.21 0.635 0.683 29.61 Parameter instability: 2.501** 0.539 4.71 -0.398 -3.26 Simulation Error Analysis TIC BP VP CP 0.005 1.03E-13 0.002 0.998 Residual Analysis Skewness Kurtosis SD -0.048 2.22 0.077 Heteroscedasticity (F) ARCH (F) RESET (F) 0.695 0.791 6.95 (0.6552) (0.5652) (0.0029)

37

Model 3. Ct = f(Nt, PCIt, ARt, Time)

Constant Nt PCIt ARt Time

2

Normality ( ) 232.32 (0.0) Autoregression (F) 3.44 (0.0147)

Estimate t-value Adj. R2 F-value DW statistic -1261.12 -1.92 0.994 1505.83 1.609 0.876 7.12 Parameter instability: 1.837* 1.567 5.37 -18.287 -1.41 30.180 2.76 Simulation Error Analysis TIC BP VP CP 0.022 7.54E-15 0.001 0.999 Residual Analysis Skewness Kurtosis SD -1.22 5.91 173.47 Heteroscedasticity (F) ARCH (F) RESET (F) 3.03 1.10 0.457 (0.0129) (0.3821) (0.6373)

Model 4. ln Ct = f(ln Nt, ln PCIt, ln ARt, Time) Estimate t-value Adj. R2 F-value DW statistic Constant -4.52 -1.73 0.991 1023.51 0.748 1.040 5.53 Parameter instability: 2.155** ln Nt 0.916 4.07 ln PCIt -0.295 -2.29 ln ARt Time -0.040 -1.92 Simulation Error Analysis TIC BP VP CP 0.005 2.56E-13 0.002 0.998 Residual Analysis Mean SD Skewness Kurtosis JB-Normality 3.67E-08 0.073 -0.165 2.46 0.652 (0.7219) Autoregression (F) Heteroscedasticity (F) ARCH (F) RESET (F) 4.88 0.756 1.05 7.76 (0.0023) (0.6430) (0.4115) (0.0017)

38

Model 5. ln Ct = f(ln Nt, ln PCIt, ln ARt, ln Ct-1) Estimate t-value Adj. R2 F-value DW statistic Constant -0.004 -0.01 0.994 1476.77 1.37 0.301 3.45 Durbin h: 2.788 ln Nt 0.300 2.83 Parameter instability: 1.445 ln PCIt -0.217 -2.05 ln ARt 0.537 4.50 ln Ct-1 Simulation Error Analysis TIC BP VP CP 0.004 3.69E-13 0.001 0.999 Residual Analysis 2 Skewness Kurtosis SD Normality ( ) 0.1.65 (0.4379) -0.110 3.26 0.060 Autoregression (F) Heteroscedasticity (F) ARCH (F) RESET (F) 2.77 0.461 0.0004 5.92 (0.0774) (0.8718) (0.9833) (0.0206) Note: 1. *' and '**' indicate statistical significance at 5 and 1 per cent respectively. 2. C = Electricity Consumption in the State (Million Units) 3. N = Number of Electricity Consumers 4. PCI = Per Capita State Income (at 1980-81 prices) 5. AR = Average Price (Revenue) (at 1981-82 prices) 6. Figures in brackets are the corresponding p-values.

39

Table 3 Residual Analysis

C 1. ADF unit root tests - Levels (Model 2) Standard Deviation 0.064 Skewness 0.054 Excess kurtosis 0.292 2 0.4146 Normality ( ) p-value Autocorln LM (F) p-value 0.7832 Heteroscedasticity White (F) p-value 0.3250 ARCH (F) p-value 0.8302

(Variables in logarithms) N AR

2. ADF unit root tests - Differences (Model 1) Standard Deviation 0.072 Skewness 0.518 Excess kurtosis 0.419 2 0.2827 Normality ( ) p-value Autocorln LM (F) p-value 0.6162 Heteroscedasticity White (F) p-value 0.7554 ARCH (F) p-value 0.3148

PCI

0.024 0.343 0.823 0.1323 0.3342

0.088 0.427 -0.298 0.4566 0.4446

0.029 -0.005 0.345 0.2589 0.5325

0.4613 0.1887

0.4303 0.3847

0.2938 0.4882

0.027 0.598 0.271 0.2528 0.4216

0.096 0.433 -0.194 0.4631 0.1230

0.030 -0.112 0.070 0.6039 0.2229

0.2437 0.3189

0.9030 0.8155

0.7851 0.7904

3. VAR Model (for 2 lags) Skewness -0.026 0.317 1.505 Excess kurtosis 0.138 1.218 -0.2440 2 0.8099 0.4368 0.4243 Normality ( ) p-value Autocorln LM (F) p-value 0.3787 0.0892 0.2167 Heteroscedasticity White (F) p-value 0.3819 0.3996 0.3350 ARCH (F) p-value 0.9169 0.4505 0.1594 2 Vector normality  = 5.189 (0.1744) Vector autocorrelation F = 1.315 (0.1744) Vector heteroscedasticity F = 0.476 (0.9991) Note: Autocorln = Autocorrelation Figures in brackets are the corresponding p-values.

40

-1.122 0.053 0.1075 0.8150 0.9856 0.4559

Table 4 Results of Unit Root Tests DW -Statistic ADF test statistics for Variables Model 1 Model 2 (with Inference variable (in (in log) (with trend + constant) log) constant) I Levels 1. C 0.0176 (0.632) -2.146 -2.554 DS with drift 2. N 0.0088 (0.632) -2.825 -0.381 DS with drift 3. AR 0.919 -2.363 -2.603 DS with drift (0.632) 4. PCI 0.0378 1.514 -0.038 DS with drift (0.659) II First Differences 1. C 1.727 -5.614** -6.042** Stationary (0.645) 2. N 0.862 -3.459* -4.957** Stationary (0.645) 3. AR 2.199 -7.349** -7.244** Stationary (0.645) 4. PCI 1.292 -3.029* -3.684* Stationary (0.673) Note: 1. '*' and '**' indicate statistical significance at 5 and 1per cent respectively. 2. Inference for the levels is based on Model 2 and for the differences on Model 1. 3. C = Electricity Consumption in the State (Million Units) 4. N = Number of Electricity Consumers 5. PCI = Per Capita State Income (at 1980-81 prices) 6. AR = Average Price (Revenue) (at 1981-82 prices) 7. Figures in brackets are approximate critical values at 5 per cent significance level (Sargan and Bhargava 1983:Table 1).

41

Table 5 Perron's Unit Root Test in the Presence of Structural Break 1. Consumption (Ct) Break year: 1983-84; Lags 2; TB/T = 0.61. Trend (t) DU DT Estimate 0.0173 -0.197 0.002 t-value 1.374 -0.472 0.391 Residual Analysis SD Skewness Kurtosis 0.063 0.247 3.216 Autoregression (F) Heteroscedasticity (F) 0.347 (0.7097) 1.487 (0.2123)

Critical value 10% -3.95

Ct-1 -0.257 -1.837 2

Normality ( ) 1.512 (0.4695) ARCH (F) RESET (F) 0.031(0.8618) 0.010 (0.9206)

2. Consumption (Ct)Break year: 1987-88; Lag 1; TB/T = 0.71. Trend (t) DU DT Critical Ct-1 Estimate 0.016 -0.011 5.20E-05 -0.223 value 10% t-value 1.711 -0.018 0.008 -2.022 -3.86 Residual Analysis 2 SD Skewness Kurtosis Normality ( ) 0.070 -0.043 3.159 1.223 (0.5427) Autoregression (F) Heteroscedasticity (F) ARCH (F) RESET (F) 2.385 (0.1083) 0.775 (0.6407) 0.134 (0.7172) 0.278 (0.6017) 3. Consumption (Ct) Break year: 1996-97; Lag 1; TB/T = 0.95. Trend (t) DU DT Estimate 0.014 -8.792 0.090 t-value 2.011 -1.742 1.738 Residual Analysis SD Skewness Kurtosis 0.067 0.081 3.554 Autoregression (F) Heteroscedasticity (F) 1.637 (0.2104) 1.066 (0.4211)

Critical value 10% -3.46

Ct-1 -0.208 -2.191 2

Normality ( ) 2.954 (0.2283) ARCH (F) RESET (F) 0.802 (0.3771) 0.315 (0.5783)

4. No.of consumers (Nt); Break year: 1979-80; Lags 5; TB/T = 0.50. Trend (t) DU DT Estimate 0.021 0.708 -0.008 t-value 1.944 2.653 -2.482 Residual Analysis SD Skewness Kurtosis 0.019 0.238 3.977 Autoregression (F) Heteroscedasticity (F) 1.623 (0.2183) 0.431 (0.9312) 42

Critical value 10% -3.96

Nt-1 -0.222 -2.411 2

Normality ( ) 5.111 (0.0776) ARCH (F) RESET (F) 0.272 (0.6069) 0.119 (0.7331)

5.Per capita income (PCIt) Break year: 1985-86; Lags 3; TB/T = 0.66 Trend (t) DU DT Critical PCIt-1 Estimate 0.0024 -1.901 0.022 -0.467 value 10% t-value 1.64 -3.52 3.60 -3.209 -3.86 Residual Analysis 2 SD Skewness Kurtosis Normality ( ) 0.022 -0.821 3.664 4.45 (0.1079) Autoregression (F) Heteroscedasticity (F) ARCH (F) RESET (F) 0.116 (0.8909) 0.375 (0.9555) 0.238 (0.6299) 1.893 (0.1806) Note: 1. Critical values are from Perron (1989 Table VI B). 2. TB/T = ratio of pre-break sample size to total sample size. 3. C = Electricity Consumption in the State (Million Units) 4. N = Number of Electricity Consumers 5. PCI = Per Capita State Income (at 1980-81 prices) 6. Figures in brackets are the corresponding p-values.

43

Table 6 Cointegration Analysis Cointegration of C with

CRDW Statistic

1. N, PCI, AR

0.635 (0.641)

2. N, PCI

0.618 (0.661)

3. N, AR

0.448 (0.661)

4. AR, PCI

0.109 (0.661)

5. PCI

0.072 (0.681)

6. AR

0.072 (0.681)

7. N

0.483 (0.681)

Augmented Engle-Granger Test Lag ADF Critical Value Statistic (10 %) 0 -2.909 -4.034 1 -2.381 -4.040 2 -2.497 -4.046 0 -2.926 -3.618 1 -2.910 -3.623 2 -2.785 -3.627 0 -1.903 -3.606 1 -1.678 -3.610 2 -1.503 -3.614 0 -1.801 -3.618 1 -1.256 -3.623 2 -1.697 -3.627 0 -1.738 -3.157 1 -1.125 -3.160 2 -1.656 -3.164 0 -1.360 -3.149 1 -1.149 -3.152 2 -1.305 -3.154 0 -2.133 -3.149 1 -2.421 -3.152 2 -1.834 -3.154

Note: 1. All variables are in logarithms. 2. CRDW = Cointegrating Regression Durbin-Watson statistic; figures in brackets are approximate critical values at 5 % significance level (Sargan and Bhargava 1983: Table 1). 3. C = Electricity Consumption in the State (Million Units) 4. N = Number of Electricity Consumers 5. PCI = Per Capita State Income (at 1980-81 prices) 6. AR = Average Price (Revenue) (at 1981-82 prices)

44

Table 7 Johansen and Juselius (J J) Cointegration Tests 1. Variables (in log): C, N, AR, PCI. Eigenvalues: 0.529, 0.226, 0.147, 3.8E-06 Null Ho: r=0 r1 r2 r3

Maximum Eigenvalue Test Trace Test Alternative Statistic+ 95 % CV Alternative Statistic+ 95 % CV r=1 21.81 27.1 33.83 47.2 r1 r=2 7.42 21.0 12.01 29.7 r2 r=3 4.60 14.1 4.60 15.4 r3 r=4 0.0001 3.8 0.0001 3.8 r 4

2. Variables (in log): C, N, PCI. Eigenvalues: 0.223, 0.174, 0.0015 Null Maximum Eigenvalue Test Trace Test Ho: Alternative Statistic+ 95 % CV Alternative Statistic+ 95 % CV r=0 r=1 7.83 21.0 13.80 29.7 r1 r=2 5.92 14.1 5.96 15.4 r1 r2 r = 3 0.046 3.8 0.046 3.8 r2 r3 3. Variables (in log): C, AR, PCI. Eigenvalues: 0.399, 0.137, 2.79E-07 Null Maximum Eigenvalue Test Trace Test Ho: Alternative Statistic+ 95 % CV Alternative Statistic+ 95 % CV r=0 r=1 15.83 21.0 20.40 29.7 r1 r=2 4.57 14.1 4.57 15.4 r1 r2 r = 3 8.6E-06 3.8 8.6E-06 3.8 r2 r3 4. Variables (in log): C, N, AR. Eigenvalues: 0.351, 0.172, 0.068. Null Maximum Eigenvalue Test Trace Test Ho: Alternative Statistic+ 95 % CV Alternative Statistic+ 95 % CV r=0 r=1 13.40 21.0 21.44 29.7 r1 r=2 5.86 14.1 8.04 15.4 r1 r2 r = 3 2.18 3.8 2.18 3.8 r2 r3 5. Variables (in log): C, AR. Eigenvalues: 0.301, 0.026. Null Maximum Eigenvalue Test Trace Test Ho: Alternative Statistic+ 95 % CV Alternative Statistic+ 95 % CV r=0 r=1 11.83 14.1 12.69 15.4 r1 r=2 0.85 3.8 0.85 3.8 r1 r2

45

6. Variables (in log): C, PCI. Eigenvalues: 0.145, 9.74E-04 Null Maximum Eigenvalue Test Trace Test Ho: Alternative Statistic+ 95 % CV Alternative Statistic+ 95 % CV r=0 r=1 5.18 14.1 5.21 15.4 r1 r=2 0.032 3.8 0.032 3.8 r1 r2 7. Variables (in log): C, N. Eigenvalues: 0.169, 0.116 Null Maximum Eigenvalue Test Trace Test Ho: Alternative Statistic+ 95 % CV Alternative Statistic+ 95 % CV r=0 r=1 6.11 14.1 10.17 15.4 r1 r=2 4.07* 3.8 4.07* 3.8 r1 r2 Note: + = Test statistics are with small sample correction. * = Significant at 5 % level; CV = Critical value.

Table 8 Pair-wise Granger Non-'Causality' Tests Null Hypothesis 1. r(C) is not Granger-'caused' by Lags r(N) 1 0.975 2 0.631 3 0.990 4 0.962 5 0.967 6 0.984

p-values r(PCI) 0.650 0.939 0.934 0.888 0.864 0.349

r(AR) 0.743 0.166 0.350 0.323 0.417 0.577

2. r(C) does not Granger-'cause' Lags r(N) r(PCI) 1 0.190 0.391 2 0.270 0.414 3 0.289 0.803 4 0.049 0.716 5 0.096 0.461 6 0.322 0.154 Note: r refers to growth rates (I.e., first differences of logarithmic series)

46

r(AR) 0.234 0.165 0.068 0.158 0.233 0.418

Electricity Consumption in Kerala

Number of Electricity Consumers in Kerala

10000

6000

9000 5000 8000

Number of Customers

6000

5000 4000

4000

3000

2000 3000 2000 1000 1000 0

1957- 1959- 1961- 1964- 1966- 1968- 1970- 1972- 1975- 1977- 1979- 1981- 1984- 1986- 1988- 1990- 1992- 1995- 199758 60 62 65 67 69 71 73 76 78 80 82 85 87 89 91 93 96 98

C 363

423.3 618.2 682.6 905

1338 1531 1729 2015 2331 2384 2912 3376 3697 4387 5332 5839 7415 7716

1957- 1959- 1961- 1963- 1965- 1968- 1970- 1972- 1974- 1976- 1978- 1980- 1982- 1984- 1986- 1989- 1991- 1993- 1995- 199758 60 62 64 66 69 71 73 75 77 79 81 83 85 87 90 92 94 96 98

N 106.2 149.8 198.8 242.1 325.1 439.7 544 711.2 825.4 988.6 1172 1569 1969 2217 2606 3192 3698 4154 4686 5211

Average Price (Revenue) of Electricity in Kerala (at 1981-82 Prices)

Per Capita State Income (at 1980-81 Prices)

40

3000

35 2500 30 2000 25 Costant Rs.

0

Constant Paise per Unit

Million Units

7000

20

1500

15 1000 10 500 5

0

1957- 1959- 1961- 1963- 1965- 1968- 1970- 1972- 1974- 1976- 1978- 1980- 1982- 1984- 1986- 1989- 1991- 1993- 1995- 199758 60 62 64 66 69 71 73 75 77 79 81 83 85 87 90 92 94 96 98

RAR 24.93 26.61 23.49 24.93 29.09 22.99 24.13 25.63 22.97 23.34 21.1 24.74 23.48 23.99 31.43 28.46 26.93 25.78 25.13 26.71

0

1960- 1962- 1964- 1966- 1968- 1970- 1972- 1974- 1976- 1978- 1980- 1981- 1983- 1985- 1987- 1989- 1991- 1993- 1995- 199761 63 65 67 69 71 73 75 77 79 81 82 84 86 88 90 92 94 96 98

PCI 1261 1266 1286 1304 1391 1450 1501 1451 1443 1461 1508 1469 1406 1507 1482 1705 1826 2114 2349 2519

Notes 1

“Everything is not good because it is old; no literature should be treated as unworthy simply because it is new. Great men accept the one or the other after due examination. [Only] the fool has his understanding misled by the beliefs of others.” 2

In fact this important stage in regression analysis is entirely overlooked and skipped in general. It is just assumed that the residual whiteness assumption is satisfied by the model considered, without empirically verifying for its non-violation, except in some ARIMA modelling. 3

Econometric ‘causality’ is a contentious term. Can econometrics explain ‘causality’ (in the sense the word is generally understood), instead of mere ‘association’ among variables? For example, Edward Leamer and others prefer ‘precedence’ to ‘causality’, in the context of Granger-‘causality’ that explains temporal lead-lag relationship between two variables. On Granger-‘causality’ Pagan (1989) remarks: ‘…….it was one of the most unfortunate turnings for econometrics in the last two decades, and it has probably generated more nonsense results…’ Hence our use of quotation marks enclosing ‘causality’. 4

A number of computer software packages of energy planning models are available at present for energy demand forecasts, such as LEAP (Long range Energy Alternative Planning), BEEAM-TEESE (Brookhaven Energy Economy Assessment model-TERI Economy Simulation and Evaluation), MEDEE-S, ELGEM, etc. A sequence u(t) , t  0, is white noise process if it possesses a constant spectral density function. Thus a white noise process is a stationary process which has a zero mean and constant variance and is uncorrelated over time. It is therefore necessarily second-order (i.e., covariance-) stationary, and if ut is normally distributed, it is strictly stationary as well, since in this case higher-order moments are all functions of the first two. Also see Granger and Newbold, 1977: 51. 5

6

This corresponds to Box-Pierce Q statistic (Box and Pierce 1970), but with a degrees of freedom correction (Ljung and Box 1978), and has more powerful small sample properties than the Box-Pierce Q statisitic. 7

Hendry and Doornik (1999) remark: “…most tests also have some power to detect other alternatives, so rejecting the null does not entail accepting the alternative, and in many instances, accepting the alternative would be a non sequitur” (p.187). “Perhaps the greatest non sequitur in the history of econometrics is the assumption that autocorrelated residuals entail autoregressive errors, as is entailed in ‘correcting serial correlation using CochraneOrcutt’” (p. 131). 8 See IRTC and IEI, Exercises for Integrated Resource Planning for Kerala: End-Use Analysis – An Empirical Study: Technical Report I – Electricity, 1996, Chap. 3, p.33. 9

Remember the data we use here are the ones actually constrained by power shortages – power cuts and load shedding. It goes without saying that reliance on and use of these data for forecasting purposes just involves high risk of errors of underestimation. 10

As indicated by the high p-values associated with the low normality test statistic values – thus the residuals are distributed with statistically small skewness and excess kurtosis. 11

LM test and White (F-) test are not available for the non-linear k-transformation model; AR(1) correction also is not possible for this model. And for the quadratic model, White (F-) test could not be computed due to near singularity of the matrix. 12

There is a reverse possibility also, residual autocorrelation causing ARCH effect (Engle, Hendry and Trumbull 1985), and this may be due to the difficulty in interpreting results when several tests reject together. 13

AR(1) correction is not possible for the non-linear k-transformation model. Also see foot note 7.

14

It should be pointed out, however, that it may not be appropriate to consider again DW statistic for the efficacy of the AR(1) correction; see the note by Kenneth White in his SHAZAM (p. 86). 15

Durbin-h in this case is 0.414, much less than the 5 per cent normal critical value.

16

It should be noted that the indicated significance is only valid in the absence of non-stationary regressors, which is not the case here. 17

Note that all the long-run elasticities implied by the model, if valid, are much less than unity.

18

Remember that these autoregressive errors, in the presence of lagged dependent variable as regressor, leave the OLS estimators inconsistent. 19

It should be noted that the parameter estimate of the logarithmic AR(1) model being close to unity and the residuals being white noise, we have a random walk model with drift. The unit root tests will also prove this. 20

Note that our findings compare well with those of Nelson and Kang (1984) who discuss misleading results resulting from estimating relationships among under-differenced series. 21

Applying unit root (DF) test to the residuals from Model 7, (logarithmic AR(1) model), the t-statistic obtained is –5.741 against the critical value of –3.607 at one per cent significance level, that thus rejects the null of unit root in the noise term. Similarly, for the semi-log trend model (9) with AR(1) correction and the short run consumption model (10), the t-statistic estimated is –5.670 versus the critical value of –3.607 at one per cent level, (same estimate for both the models, as they are equivalent), reconfirming the stationarity of the residuals. 22

Nelson and Plosser (1982: 141) state that ‘the tendency of economic time series to exhibit variation that increases in mean and dispersion in proportion to absolute level motivates the transformation to natural logs and the assumption that trends are linear in the transformed data’. 23

The residuals from the models are strictly white noise for these lags with levels and with differences. Note that the unit root null might be rejected for some other lags, since the results of univariate ADF testing are sensitive to the lag length in the regression model for the tests. Hence the significance of a choice of optimum lag length, that is to satisfy the residual whiteness assumption. 24

This DW-statistic for the level of a variable is not to be confused with the cointegrating regression Durbin-Watson (CRDW) statistic of the residuals; see foot note 52. 25

Cow tests for structural stability carried out on the logarithms of the series confirm the following breaks: 1983-84 and 1987-88 in consumption; 1979-80 in number of consumers; and 1985-86 in per capita income. However, since the Cow test is meant for only stationary variables, its results cannot be relied upon in our case, and they are not reported. 26

See the multivariate econometric model 2.

27

The first row (null of r = 0) maximum eigenvalue and trace statistics are respectively 27.83 and 43.16, and the former is significant, though marginally, at 5 pe rcent level, but the latter is not. Dickey, et al. (1991) recommend the maximum eigenvalue test as more reliable than the trace test especially in small samples. This then suggests that there exists one cointegrating vector (cv) of long-run relationship among the four variables, if we disregard small sample bias. The relationship of interest in our case is that of electricity consumption (C) with other variables. The estimated cv of this relationship (with normalised coefficients representing long-run elasticities) is given by C = 0.734 N – 2.941 AR – 0.474 PCI, where all the variables are in logarithms. In view of the wrong sign of PCI (as well as the very high elasticity of AR against actual experiences), we fail to give a consistent economic meaning to this cv, and conclude against identifying the relationship of interest. Any other relationship among the variables implied in the existence of a cv is of no interest to us now. In Table 8, in the last model of C with N, note that both the statistics in the second row, Ho: r  1, are significant at 5 per cent level. However, since the first row, Ho: r = 0, cannot be rejected, we cannot consider the second row. That is, if the first (row) statistic is not significant, then r is selected as zero (Doornik and Hendry 1997: 224-225). 28

29

Remember that by the Granger Representation Theorem (Granger 1983), if a set of variables are cointegrated, then there exists an error correction representation (and vice versa).

49

30

Note the null hypothesis that elctricity consumption (C) growth rates do not Granger-‘cause’ N growth rates can be rejected at 5 and 10 per cent significance levels respectively for lags 4 and 5, and the same is so for AR growth rates at 7 per cent level for lag 3. However, in view of the persistence of non-rejection for all other lags, we cannot consider such isolated results. 31

The results are reported for lags up to 6 for space limitation.

32

For example, in 1995-96, the percentage share of fuels, electricity, and lubricants consumed in the ASI factory sector of India in the value of total inputs was 9.56 per cent only and that in value of products, 7.81 per cent; in Kerala, these were respectively 5.77 and 4.72 per cent only (Government of India 1998 a: 85-86). During the 90s (1991-92 to 1997-98), power and fuel expenses of the whole manufacturing sector in India remained at about 6 per cent of the net sales and at about 7.5 per cent of the total production costs (CMIE 1999). 33

The percentage share of the unregistered firms in the manufacturing sector’s contribution to net domestic product (at current prices) in India in 1997-98 was 35.4 per cent; in 1970-71, 1980-81 and 1990-91, it was respectively 46.7, 46.3 and 39.1 per cent. 34

An apt example of mechanical adoption and use of econometrics against its grain usual in the academic circles is Pillai (1981)’s Cobb-Douglas production function approach to Kerala’s hydro-electric power system, with capital and labor as ‘variable’ inputs. It is common sense that labor is not at all a variable factor of production in hydroelectric power generation, it being a part of sunk capital. 35

The latter (price) equation need not be confused with the usual supply function; many studies (for example, Halvorsen 1975) assume electricity supply in this context as fixed. However, the unique technical characteristic of electricity that it cannot be stored in its original form and hence must be generated the moment it is demanded stands to do away with the usual demand-supply distinction. This also makes the question of identification irrelevant. The earlier studies in India on electricity demand analysis have ignored the question of identification, as pointed out by Dr. Indrani Chakraborty; no reason is provided as to why the equation estimated as for demand may not be a supply function. It should however be noted that a distinction between demand (= supply) and capacity provision is possible here except in power shortage situations. 36

Government of India, National Accounts Statistics, different issues.

37

Nearly 50 per cent of the households in Kerala (and nearly 60 per cent in the rural areas) remains unelectrified (as per 1991 Census). This problem also haunts the regressor of per capita State income that includes the share of the unelectrified households also. 38

See Pillai (1981: 81 – 82); Henderson (1975) uses sectoral output in the place of number of consumers. Another immediate factor of influence is connected load, the total of the rating (in kilowatts) of all the electricity using appliances installed on a consumer’s premises. This also may be considered along with the relevant intensity of energy consumption (electricity consumption per kilowatt (KW) of connected load) and the interaction between the two. However, number of customers (N) is more immediate and direct than connected load (CL) in determining energy demand, as not only is N in fact the causative of CL, but also a customer may not use all his electric devices simultaneously or continuously; there are times, on the other hand, when all the consumers together exert demand pressure on the system. There is yet another significant reason. A growing power system is expected to become more and more electricity intensive in that its CL grows faster than N (so that the electricity intensification factor, i.e., connected load per customer (CL/N), increases over time). Despite the restrictions imposed on providing new connections since 1982-83, the domestic, commercial, and LT industrial consumers in Kerala have behaved in the expected line, becoming more electricity intensive (in terms of appliances installations), but the HT-EHT industry and ‘others’ (agriculture, public services, licensees, etc.) have not. This surprising tendency of a faster decaying electricity intensity in the State’s HT-EHT industrial and agricultural sectors has overshadowed the normal growth in the other sectors and been reflected in the aggregate, the growth of CL trailing behind that of N. 39

Even during the ‘surplus’ period, it can be seen, the internal consumption was constrained in order to boost the KSEB’s export extravaganza. 40

Note that the projections from semi-log (i.e., exponential) trend extrapolation model and simple and logarithmic AR(1) models are in fact (constant) growth rate based ones.

50

This follows since rN = lnN. Note that the relation also amounts to one period (compound) growth expression: Ct = Ct1(1+r), where r is the compound growth rate of consumption and (1+r) = exp(rC), where rC = lnC. 41

42

These will be discussed in detail in the penultimate chapter.

51

CHAPTER 5

TIME AND COST OVER-RUNS OF THE POWER PROJECTS IN KERALA

‘Time and tide wait for none.’ 1. INTRODUCTION

This paper on time and cost overruns of the power sector projects in Kerala is a part of a larger study on 'The Plight of the Power Sector in India: Inefficiency, Reforms and Political Economy', and discusses the costs of inefficiency in the particular context of the Kerala power sector at the project implementation stage. In an earlier paper (Kannan and Pillai 2001 a) we have discussed the cost of inefficiency involved in general in the Indian power sector at the various stages of operation. Here we take up an analysis of the cost of inefficiency involved in time and cost overruns in the power projects in Kerala. This is of very significance in the present context of arguments by the government in favour of private sector participation in power generating capacity addition, under the pretext of a resources crunch. The government is said to be under a tight constraint of severe funds scarcity and hence incapable of undertaking new projects for power development. However, we will find that this argument is flimsy to the extent that the government is actually over-spending on each of the projects undertaken. Each project involves immense cost overrun. Had the government been able to implement each project efficiently within the normally expected constraints of time and cost, then it could have saved huge resources and hence undertaken a large number of additional projects. It is not that the government has no resources meant for power development, because it is actually over-spending; the problem is in the inefficiency of management, coupled with

1

the political economy of corruption. The present paper, in six sections, has the limited objective of bringing into light this aspect. Following this introduction, section 2 provides a brief discussion of the individual projects, falling under the time and cost overruns and the third section, their comparative analysis. The costs of delays are examined in section 4, and the possible causes in section 5. The last part briefly discusses the political economy of corruption involved in the time and cost overruns of the power projects in Kerala, and concludes the study.

Delays in project implementation and the attendant cost escalation have been a regular feature in the electric power sector in Kerala. Normally the construction of a major hydro-power plant is expected to be completed within 8 – 10 years and that of a mini hydel project in 2 – 3 years. However, the Kerala experience baffles all the common senses in this respect, with longer time and higher cost over-runs in the case of both major and mini hydel projects. A `classic’ example is the Kakkad hydro-electric project of 50 mega watt (mw) installed capacity; the project was sanctioned as long back as in 1976 with an original cost estimate of Rs. 1860 lakhs; this project was proudly presented that time as the least cost hydro-electric project in the State! It was scheduled to be commissioned in 1986; but it took 23 years for the Kerala power system to tap the energy potential of this project (major construction works on it started only in 1979), at an estimated cost of Rs. 153.5 crores, about 725 per cent above the original one!

The Kakkad story is not an isolated exception, but forms only a part of an unending serial of over-runs in Kerala power system. The prestigious major project of Idukki also was an unfortunate victim of time and cost over-run, mainly due to labour disputes, the prime villain in every instance. Idukki Stage I project (3 units of 390 MW) could not be commissioned in the Fourth Five Year Plan (1969-74) as scheduled and had a long gestation period due to labour problems, until it was finally commissioned in 1976. When idukki Stage II project (3 units of 390 MW) was put on line in 1986, after a time over-run of about 8 years, it had a cost escalation of 115 per cent over the original estimate.

2

Similarly, the next project, Idamalayar (of 75 MW, started way back in 1970 and commissioned in 1987), suffered a time over-run of about 9 years and a cost increase of 285 per cent. Two major firm power augmentation schemes, Sabarigiri Augmentation and Idukki Stage III, too had the same fate. Started in 1972 and 1975 respectively, the works on these projects could not be completed till the turn of the 90s. A cost over-run of nearly 780 per cent (the highest ever among the projects in the Kerala system!) and a time over-run of 10 years go to the discredit of Sabarigiri Augmentation scheme, beyond any common sense accounts. And a cost increase of about 270 per cent with a time over-run of about 10 years lie behind the Idukki Stage III project.

2. PROJECT-WISE ANALYSIS

Data on time and cost overruns of 16 other hydro-power projects are available for analysis, the data having been collected from the various volumes of Economic Review of Kerala State since 1985. These projects are 1) Kakkad, 2) Kallada, 3) Lower Periyar, 4) Pooyankutty, 5) Malampuzha, 6) Madupetty, 7) Malankara, 8) Chimony, 9) Peppara, 10) Azhutha Diversion, 11) Kuttiar Diversion, 12) Poringalkuthu Left Bank Extension, 13) Vadakkeppuzha Diversion, 14) Vazhikkadavu Diversion, 15) Kuttyadi Tail Race and 16) Kuttiady Extension. The details of these projects are given in Table 1 (and also in the Appendix).

1. Kakkad

This project, considered the least cost hydro-electric project in Kerala, is to use the tail race waters of Sabarigiri power house (PH), together with the inflow of two tributaries of Kakkad river, viz., Moozhiyar and Veluthodu streams over a gross head of 132.6 m. for power generation of 262 million units (mu) with an installed capacity of 50 mw.

Though the project was sanctioned by the Planning Commission way back in 1976 at an estimated cost of Rs. 1860 lakhs, the construction activities were started only

3

by 1978-79 due to paucity of funds. Even after the work was started, the progress was tardy. For one example, the total length of the inter-connecting tunnel driven as by the close of 1986-87 was only 886 meters (out of 3036 meters). The poor performance was mainly due to labour disputes for over a year from 3-10-1985 to 29-10-1986. Though the work was resumed on 30-10-1986, it was interrupted on 6-2-1987 due to a rock fall inside the tunnel. In the case of the power tunnel, some progress was achieved only in 1986-87; two earlier contracts with poor performance had to be terminated here.

The time and cost overrun story of this project has already been mentioned. Over a period of 23 years, with a time overrun of about 13 years as in 1999, when it was finally commissioned, the cost escalation of this project was 725 per cent above the original estimate. That is, the actual cost was more than 8 times the original cost estimate. It should be noted that a project is sanctioned at the costs that exist at the time when the project is submitted. The cost estimate is likely to increase over time on account of price inflation. Though the original cost estimate is presumed to include an allowance for possible price inflation, often the actual experience can deviate from the assumptions. Hence it is natural to consider and identify that part of the cost escalation that is due to price inflation which can by no means be attributed to improper project formulation and/or implementation. However, what remains in the cost overrun over and above the effect of price increase is a matter of concern demanding explanations in terms of real factors involved in faulty planning and execution. For each of the projects, we have estimated the price inflation in terms of WPI for all commodities experienced during the project implementation period, in order to differentiate the effects of price inflation and of other factors on capital expenditure.

The WPI for all commodities registered as in 1999 an increase of only 461.2 per cent over 1976 (when the Kakkad project was sanctioned). This implies that the cost escalation is about 1.5 times the general price inflation (as given by the WPI for all commodities). Thus it is clear that price inflation alone is not responsible for cost overrun; about 260 per cent of the increase in the cost estimate can be attributed to factors other than price inflation, which can evidently be treated as a waste of resources.

4

2. Kallada.

This project envisages construction of a dam toe power station of 15 MW installed capacity and generation of 53 MU of power from the existing Kallada irrigation project. Though the contracts for civil works were settled in April 1985, and works began immediately, frequent releases of water through the irrigation outlets of the dam flooded the work areas, preventing the progress of works.

The project was sanctioned in 1981 with an original cost estimate of Rs. 1180 lakhs and was commissioned in 1993-94 at a (revised) cost of Rs. 1802 lakhs, representing 52.71 per cent increase. It was to be commissioned at the earliest by 1989, but had to undergo a time overrun of 5 years. During the same period, the WPI (all commodities) rose by 148 per cent; and the revised cost estimate of this project appears not to have been inflated to that extent.

3. Lower Periyar.

This is a tail race cum run-off-river scheme in the lower reaches of Periyar river downstream of Neriamangalam power house. The scheme envisages the utilisation of the waters of Neriamangalam power station, the spill from the Kallarkutty dam and the available yield from the Perinjankutty catchment and the catchment areas below the dams at Kallarkutty, Idukki and Cheruthoni river, over an average gross head of 302.63 m. for power generation, with an installed capacity of 180 MW and annual generation of 493 MU.

Started in 1983 with an original estimate of Rs. 8843 lakhs, this project was commissioned in 1997 and its revised cost estimate as in 1999-2000 stood at Rs. 353 crores. Over these 14 years (including a time overrun of 6 years), the cost estimate saw an

5

increase of about 300 per cent against an increase in the WPI (all commodities) by 194 per cent. Thus after accounting for the full impact of price inflation on the capital cost of the project, about 111 per cent increase needs to be explained by other factors of wasteful management.

4. Pooyankutty.

The scheme envisages construction of a 148 m. high concrete dam across river Pooyankutty and a surface power station with two units of 120 MW each; thus with an installed capacity of 240 MW and annual generation of 645 MU. The scheme was approved by the Planning Commission as far back as in August 1986. However, the central government’s sanction of forest clearance is still to be received. The state government and the KSEB are reported to have fulfilled all the formalities for the issuance of sanction, including the proposals for compensatory afforestation as required under the Forest Conservation Act of 1980. So far only minor preliminary works have been done.

A 14 years incubation for a project proposal is ample evidence of the lethargy and non-commitment on the part of the planners. During this period, the cost estimate was revised upward by 228 per cent, from Rs. 250 crores to Rs. 820 crores, far exceeding (by 45 per cent) the general price inflation (182.7 per cent) during this period.

5. Malampuzha.

A mini hydel project of 2.5 MW with an annual generation of 5.6 MU, this scheme envisages construction of a power station on the downstream side of the existing irrigation dam (owned by the State PWD) to utilise the irrigation release. Started in 1987 and expected to be on-line by 1989, this mini project is now expected to be commissioned `in the near future’. After 12 years with a time overrun of about 10 years as in 1999-2000, the capital cost was revised from the original Rs. 295 lakhs to Rs. 679

6

lakhs – an increase of about 130 per cent. Over the same period the WPI (all commodities) registered an increase of 169 per cent.

6. Madupetty.

Another mini hydel project of 2 MW with 6.4 MU of annual generation, this scheme aims at construction of a dam toe power house at the existing Madupetty dam for power generation using the water released from the Pallivasal hydro-electric project. Started in 1987 and expected to yield its energy by 1989, this mini project was at long last fully commissioned by January 1998 after a time overrun of about 9 years. The cost estimate was revised from the original Rs. 292 lakhs to Rs. 775 lakhs by 1995, which, however, came down to Rs. 478 lakhs by 1998, providing a good example for the reliability of estimation procedure of the KSEB; in the case of most of the projects this is so. The cost increase in this case is 64 per cent, against an increase in the general price level by 145 per cent.

7. Malankara.

Another small hydro-electric (HE) project with an installed capacity of 10.5 MW and annual generation of 65 MU, this scheme envisages the construction of a dam toe power station at the Malankara dam of the Muvattupuzha valley irrigation project (under construction by the State PWD). The project will utilise the tail water releases from the Moolamattom power house of Idukki hydroelectric project together with the inflow from 153 square km. free catchment less the irrigation requirements.

Started in 1987 and expected to generate power by 1990, this project has by now (1999-2000) registered a time overrun of about 9 years; its capital cost was revised over the period from Rs. 780 lakhs to Rs. 43.36 crores by 1997 and then to Rs. 41.13 crores by 1998 and 1999-2000, thus undergoing a phenomenal increase of 427.3 per cent against a wholesale price rise of 169 per cent. Reminiscent of the mammoth inflationary influence of the `other factors’ on the capital cost of Kakkad project, in this case the other

7

factors of sheer waste and overestimation account for as high an increase as about 258 per cent in the capital cost, that calls for another careful diagnosis.

8. Chimony.

Another mini hydel project, this scheme envisages installation of a generating unit of 2.5 MW in a dam toe power station at Chimony irrigation dam (under construction by the State PWD). It is expected that 6.5 MU of energy can be economically generated during the period from December to May.

Started in 1987 and originally scheduled to be commissioned in 1990, this project fell by 1993 a prey to a dispute between the contractor of the electrical works and the KSEB and all the works were paralysed thanks to a stay order from the High Court obtained by the contractor. By 1993, the capital cost was revised from its original level of Rs. 314 lakhs to Rs. 425 lakhs, representing an increase of 35.35 per cent against a general price rise of 72 per cent over the same period.

9. Peppara

This small project was proposed to benefit Thiruvananthapuram city by making use of the drinking water supply released from Peppara dam (owned by the Kerala Water Authority) through a dam toe power house of an installed capacity of 3 MW and an annual generation of 11.5 MU.

Again a 1987 project supposed to have the normal gestation period of 3 years, it was finally commissioned only in 1996, with a time overrun of 6 years and a cost escalation of 73.7 per cent over the original estimate of Rs. 392 lakhs, against a general price rise of 118.3 per cent during this period. Note that the cost estimate was earlier revised to Rs. 850 lakhs in 1995 and then reduced to Rs. 625 lakhs in 1998 only to raise again to Rs. 671 lakhs in 1999 – another apt example for the haphazard planning mechanism.

8

10. Poringalkuthu Left Bank Extension

This scheme is to construct a second power station with an installed capacity of 16 MW and an annual generation of 38 MU for better utilisation of the water release from the existing scheme (Poringalkuthu power house). Its works were started in 1989 and it was expected to be commissioned in 1992-93. After a time overrun of about 6 years, it was commissioned in 1999; the original cost estimate of Rs. 902 lakhs rose by about 374 per cent to reach Rs. 42.7 crores. Comparing this with the rise in the WPI (all commodities) over the same period by 113 per cent, about 261 per cent of the increase in the cost estimate is found to be attributable to `other factors’ of wasteful management and over-estimation.

11. Kuttiyadi Tail Race

This project proposes to utilize the regulated discharge from the existing Kakkayam power station of Kuttiyadi HE project for power generation in a station to be located further downstream. The proposed installed capacity is 2.5 MW and the annual generation 15 MU.

The project was started in 1989, and expected to be commissioned in 1992-93. By 1999-2000, with a time overrun of 7 years, the estimated cost rose by 225 per cent from the original Rs. 397 lakhs. The general price rise during this period was by 132 per cent, indicating an increase of about 93 per cent in the cost estimate due to `other factors’, over and above the influence of price inflation. Note that the revised estimate in 1997 was Rs. 14.48 crores (265 per cent above the original) and in 1998, Rs. 13.38 crores!

12. Azhutha Diversion

9

This scheme envisages diversion of waters from about 16,8389 sq. km. catchment of the upper reaches of Azhutha river, a major tributary of river Pamba to Idukki reservoir for increasing the power potential of Idukki power project by 57 MU. The scheme will provide diversion of about 57.6 mm3 of water on an average per annum.

The work on this project was started in 1987, anticipating it to be commissioned in 1991. After a time overrun of about 6 years, it was partially commissioned in June 1998. By 1999-2000, the original cost estimate was revised upward from Rs. 290 lakhs to Rs. 14.46 crores, an increase of nearly 400 per cent, against the rise in the WPI (all commodities) by 145 per cent. Thus the factors other than price rise appear to account for about 254 per cent increase in the cost estimate.

13. Kuttiar Diversion

This scheme envisages diversion of waters from a catchment of 10.4 sq. km. of Kuttiyar river (a tributary of Muvattupuzha river) to Idukki reservoir to raise the power potential of Idukki power project by 36.6 MU.

Started in 1988 with an original cost estimate of Rs. 214 lakhs, this project was to be completed at the earliest by 1990-91. At present it is expected to be commissioned in the near future, with a cost escalation by 343.5 per cent to Rs. 949 lakhs over a time overrun of about 8 years as in 1999-2000. This is against a general price rise by 132 per cent during the same period. Thus about 211 per cent increase in the cost estimate of this project remains to be explained in terms of `other factors’.

14. Vadakkeppuzha Diversion

This scheme envisages diversion from 3.43 sq. km. catchment of Vadakkeppuzha, a tributary of Muvattupuzha river and 0.625 sq. km. catchment of Pothumattom stream, also of Muvattupuzha basin, to Idukki reservoir to augment the firm generation of Idukki project by 12.3 MU.

10

When the project work was started in 1989, it was proposed to be completed by 1991-92. However, even after a time overrun of 8 years as in 1999-2000, the commissioning date remains ‘not fixed’, and the original cost estimate of Rs. 131 lakhs rose by 292 per cent to Rs. 514 lakhs against a rise in WPI (all commodities) by 132 per cent over the same period, leaving 160 per cent increase in the cost estimate to be accounted for by `other factors’. Note that the cost estimate was revised upward to Rs. 786 lakhs in 1997-98 and then downward to Rs. 705 lakhs in the next year only to be drastically cut down again to Rs. 514 lakhs in 1999-2000.

15. Vazhikkadavu Diversion

This scheme envisages diversion of waters from 6 sq. km. of catchment of Vazhikkadavu to the Idukki Reservoir by a diversion tunnel to increase the firm power of Idukki project by 24 MU.

Started in 1989, this project was expected to be completed by 1992-93. However, even after a time overrun of about 7 years as in 1999, it too remains with an uncertain commissioning date. The original cost estimate had to be revised by a phenomenal 760 per cent, dwarfing even the classical Kakkad phenomenon, from Rs. 186 lakhs to Rs. 15.99 crores against a general price inflation by 132 per cent over the same period. Thus an increase to the tune of about 628 per cent in the cost estimate remains as due to the influence of `other factors’ – a shocking example of mismanagement in the preparation of project proposal and cost estimation, that too in the case of only a diversion project, meant to increase water availability only.

16. Kuttiady Extension

The storage capacity of the existing Kuttiady reservoir being highly inadequate, full utilisation of the inflow is not possible now. Hence, under this extension scheme, capacity addition (one unit of 50 MW; 75 MU) to the existing power station is proposed.

11

Though the project was cleared by the Planning Commission in January 1992, major works on it started only in February 1994, and it was originally expected to be commissioned in 1995-96. After a time over-run of 4 years, it was finally commissioned in 2000, with a cost overrun of 544 per cent over the original estimate of Rs. 30.73 crores (that went to Rs. 198 crores), against a general price inflation of just 73 per cent, leaving an unbelievable waste gap of 471 per cent!

3. A COMPARATIVE ANALYSIS

As already explained, the estimation of the capital cost of a project is made based on the price level prevailing at the time when the project proposal is made; and hence there is a time-element of error involved in it representing under-estimation in the face of inflation. Cost estimate is often revised upwards to take account of this, especially when the price level is rising rapidly and/or the time-overrun involves an element of uncertainty as to the completion of the project. Ideally, a revised cost estimate should sufficiently cover the general price rise. And hence what remains in the revised cost escalation of a project over and above the general price inflationary influences is a matter for serious consideration; it may represent an over-estimation due to uncertainty or an element of deliberate attempt at wasteful mismanagement of resources.

Of the 20 projects we have considered above, barring 7 projects, all others have significantly very high remainder in their revised cost estimates in excess of the general inflationary impact (Table 2). The 7 projects are Idukki II, Idamalayar, Kallada, Malampuzha, Madupetty, Chimony, and Peppara. In the case of Chimony project, the work of which had to be suspended due to a dispute with the contractor that brought in Court intervention, the inadequate coverage of the general price inflation in the revised cost estimate might be a case of under-estimation. In the case of a number of projects (for example, Kakkad, almost all the mini projects and some of the diversion projects), the cost estimates have been revised every year in a very haphazard manner, some time upward and then downward, indicating an inconsistent planning mechanism.

12

It should be noted that apart from the `Classical’ case of Kakkad project, it is the mini hydel projects and diversion schemes that have become comparatively more prone to time and cost overrun. The mammoth cost escalations in the case of Malankara mini HE project and Vazhikkadavu diversion are a phenomenal swell in some element of error that has crept in the project design and estimation. The other things appear to have influential sway over most of the other projects also.

In general, these 20 projects of the last 3 decades account for time overruns ranging between 62.5 per cent (Kallada) and 500 per cent (Malampuzha) of the expected period of construction, and cost overruns ranging between 52.7 per cent (Kallada) and 777 per cent (Sabarigiri Augmentation), of the original cost estimate (excluding Chimony).

For a more objective comparison, we can analyse the capital cost per kWh of potential energy of these projects (Table 2). Among the power plants considered, the capital cost per unit of electricity was the lowest for Idukki II Stage with 68 Paise per unit and among the augmentation schemes, for Idukki III Stage with only 40 Paise per unit. The highest cost escalation of Sabarigiri augmentation project has spread very thinly over the large units of its energy potential, resulting in a capital cost of only 90 Paise per unit. Idamalayar stands with a capital cost of Rs. 2.81 per kWh of energy. On the other extreme, one’s common sense may be baffled at the mammoth capital cost of Rs. 26.4per unit as per the latest estimate in the case of Kuttiady extension project. Energy from the still unborn Pooyankutty project too is priced out very high at Rs. 12.7 per unit! Malampuzha (Rs. 12.1 per unit) and Kuttiady Tail Race (Rs. 8.6 per unit) are also planned to be high cost energy generators. Note that the capital cost of energy from Kakkad, the classical example for time and cost over-run, is Rs. 5.9 per unit.

It will be enlightening to compare these figures with the original capital cost of Enron project (Dabhol power project phase I) in Maharashtra much criticised as ushering in an era of stupendously high-cost energy in India. Its original capital cost of Rs. 4.48 crores per MW of capacity at the normal load factor of 68.5 per cent implies a unit capital

13

cost of Rs. 7.5 per kWh. The Kuttiady extension project undertaken with a Canadian loan and contracted for its completion with a Canadian firm (SNC Lavalin) involves a capital cost, which is about 3.5 times the controversial original cost of the Enron project! It should be remembered that Enron’s was a new project, while only an extension work was done at Kuttiady. It is highly significant to note that the Kuttiady extension work contract was awarded to the Canadian firm by a leftist government in the State that is credited with an assertive anathema against foreign capital, especially the Enron, but has time and again stood in defence of the Canadian firm, sanctioning all their demands of time and cost over-runs. Now compare the other projects also.

4. THE COST OF DELAYS The delay in commissioning a power project invariably involves different elements of avoidable costs to the society. The most immediate one is the cost escalation itself. A direct cost of over-runs is in terms of the additional energy realisable, were the project commissioned in time, as well as the additional sales revenue thereof. The increased availability of power could reduce the requirement of costly energy import, thus effecting some cost savings in it. In addition to these is the indirect cost of unsatisfied demand corresponding to the additional energy realisable.

In this section we make an attempt to quantify the cost of time over-runs of the projects under study in terms of additional energy and revenue that could be realised if these projects were commissioned in time. The results are shown in table 3.

We start with the year 1983-84, by which time, it is assumed, the four earlier projects, Idukki II and III Stages, Idamalayar, and Sabarigiri Augmentation could be brought on line, so that the available firm generation capacity in 1983-84 would be 5554 MU, instead of the actual 3726 MU. Given the firm power capacity utilisation (98 per cent) and loss (26 per cent) structure in the system, this then yields additional generation

14

of 1788 MU and additional sales of 1327 MU, which at an average rate of 35.2 Paise per unit would realise an additional revenue of Rs. 46.7 crores in that year. Additional revenue obtainable in 1984-85 comes out at Rs. 53.8 crores. The total revenue thus realisable during these 17 years from 1983-84 to 1999-2000 is estimated at Rs. 886.3 crores, or Rs. 52 crores per year! This then represents one cost of avoidable time overruns of these 19 projects (excluding the non-starter Pooyankutty project) in Kerala (Table 3). It is very distressing to think of such a situation that the cash-strapped KSEB has been forced to forego a revenue of about Rs. 52 crores a year on average due to delays in getting the on-going projects commissioned in time.

Such additional generation that could be effected through timely completion of projects could reduce to a good extent the costly dependence on energy imports.

Timely completion of these projects could avoid the substantial burden of capital cost escalation also (Table 4). Such savings factor highlights the fact that when capital cost is escalated more than what is planned, it results in a loss of its alternative uses. Considering the resources constraint of the Government, if these resources were used more efficiently, then the resultant increased availability of these resources to the Government could be used for taking up more projects. To the extent that such actual cost escalation reflects inefficient resources utilisation, the savings in capital cost, that ould have been obtained in the absence of cost overruns, also represents a capital waste involved. For example, suppose that Kakkad hydro-electric project could be commissioned in time in 1986 itself, 8 years after its construction works started. Accounting for the general price inflation during this period, the capital cost of this project by 1986 would be at the most only Rs. 39.66 crores, saving as much as Rs. 113.86 crores, almost enough to construct 3 more similar plants, or to add to the system capacity by another 140 MW at the nominal cost of Kakkad project! Thus the capital waste involved in this case is equivalent to 3 more similar plants (Table 4) or an installed capacity of 140 MW! Timely completion of lower Periyar project could save as much as Rs. 189 crores, enough for a similar project of more than 200 MW capacity! The second highest savings, after Lower Periyar project, could come from Kuttiady extension project

15

to the tune of Rs. 158.3 crores, almost enough for four similar or Kakkad-type projects! As already noted, Kallada project (the only exception), even with 5 years over-run, has not eaten up resources beyond the limits set by general price inflation.

Timely

completion of all other 18 projects (excluding the non-starter Pooyankutty) could yield a mammoth saving in capital cost of Rs. 644.03 crores, almost enough for 16 Kakkad-type projects with 800 MW capacity! Since so much capital resources have gone wasted, this 800 MW (or Rs. 644 crores) represents the capital waste involved in the faulty planning and implementation of power projects in Kerala. That is, the capital waste factor involved is 16 (i.e., 16 Kakkad-type projects)! And the KSEB still reeling down in the red, the government lets such waste and mismanagement pass.

It is in this light then that we should examine the so called financial 'inability' of the SEBs (and the governments) to finance power development in general. The basic argument put up in defence of inviting private sector participation in power development has come out of the resources crunch experienced by the governments. However, this defence is turned out to be flimsy in the face of the act that there is over capitalisation in actual practice in the case of each project the government has undertaken; the government could, through efficient performance, save substantial resources, which could in turn be used for taking up additional projects. Behind this inability works the political economy of corruption.

The gravity of the problem of over-runs can be gauged by considering the combined effect of both the time and cost over-runs, a measure of which, called ‘capital x time waste factor’ (also see Morris 1990), is obtained as the difference between the actual capital x time (CaTa) and the originally planned capital x time (CoTo) measures as a percentage of the latter (where Ca and Co are the actual (or latest) and originally planned estimates of capital cost and Ta and To are the corresponding period of commissioning). In estimating this resources waste factor, we assume that expenditure over the course of a project takes place uniformly. Thus in the case of the Kakkad project, the originally planned resources were Rs. 18.6 crores x 10 years = Rs. 186 crore years, but the actual resources spent were Rs. 153.52 crores x 23 years = Rs. 3530.96 crore years, such that

16

there was a capital x time waste of Rs. 3344. 96 crore years or 1798.4 per cent of the originally planned resources. Thus it shows that as a result of time and cost overruns, this project has eaten up about 1800 per cent more capital x time than what was originally expected. In other words, if the Kakkad project could be completed on time as per plans, then the KSEB could increase the quantum of similar projects by about 1800 per cent with the same resources it actually spent for a single project.

The capital x time waste factor for the 19 projects (excluding the non-starter Pooyankutty) ranges from 148 per cent for Kallada project to 2766 per cent for Vazhikkadavu diversion! (Table 4). There are as many as 9 projects (6 of which are mini or diversion projects) having more than 1000 per cent waste factor. That on average, each project has eaten up extra resources worth 1100 per cent just shows in general the enormous waste of capital x time resources in power project implementation in Kerala.

Kuttiady power project had been out of service for a long time now in the name of extension works going on there. The extension programme with a time over-run of more than four years and a stupendously exorbitant capital cost of Rs. 26.4 per kWh of energy potential, also involved substantial revenue loss for the parent project due to its closure. The firm generation potential of Kuttiady power station is about 270 MU or 0.74 MU a day, equivalent to a sales revenue of about Rs. 15 lakhs a day. If the extension scheme were commissioned in time (i.e., in 1995-96), it could fetch sales revenue of about Rs. 7.1 lakhs a day. During the last 5 years, the total loss of sales revenue alone comes out to be Rs. 399 crores in this case!

5. CAUSES OF DELAYS

A host of causatives are at work behind the delays – changes in the technical design and feasibility reports, original cost estimates being based on inadequate or incomplete data and unrealistic assumptions, inefficient management, inadequate geological and technical investigations of the projects at the outset, vague and ambiguous specifications and conditions of contract, sluggish decision making at various stages of

17

construction, lack of availability of materials or of transportation facilities, infighting and ego clashes among different groups of the bureaucracy and technocracy of the KSEB, unwarranted transfer of planning and supervisory staff between projects during their construction, a lack of vision about the power needs of the State, labour disputes, court

Causes of Delays The principal causes of delays in the case of hydro-power plants, inter alia, have been listed by the Committee on Shortfall in Generation During the Third Five Year Plan under the Chairmanship of Sri. K. P. S. Nair as follows: 1. Inadequate investigation before finalising technical project report. 2. Major change in the scope of work like (a) change in the location of dam; (b) change in design of dam foundation; (c) change in design of Water Conductor System; (d) change in location of power station and switch yard; (e) change in generator capacity. 3. Delay due to inter-State aspects. 4. Delay in issue of authorisation by Central/State authorities. 5. Delay in foreign exchange tie ups. 6. Change in key personnel in the course of advance planning and execution. 7. Delay in procurement of equipment due to (a) late issue and late finalisation of tenders; (b) procedural delays in processing through DGS&D; (c) processing of foreign exchange release by Government of India (GOI). 8. Delay in procurement of construction equipment. 9. Shortage of cement and steel, welding rods, explosives, etc. 10. Shortage of spare parts for construction equipment. 11. Late arrival of erection specialists. 12. Delay in delivery of equipment due to failure of supplier to keep up schedule; 13. Difficulties in transporting equipment to site (a) in moving over dimensional packages on railway due to restrictions imposed by bridges, tunnels, etc.; (b) due to difficult terrain and lack of proper access roads. 14. Unprecedented rains and floods. 15. Land acquisition and rehabilitation.’

18

interventions for aggrieved contractors, and so on (Kannan and Pillai 2001 a). Nurturing all these is a lack of political will to finish the work on schedule, borne and bred of course by high level corruption and an indifferent public.

Recurring labour militancy is recognised in general as the single factor that puts the highest cost burden in this respect. And it cannot be otherwise in a politically surcharged atmosphere of highly pampered unionism of diverse hues peculiar to Kerala. Not a single project in Kerala (including the prestigious major project of Idukki) has been left unhaunted by the spectre of tools-downing militancy. The construction work of the Idukki project was much pompously inaugurated by the then chief minister, EMS Namboothiripad, on 10 February 1966; and the very next day started a labour strike, that finally culminated

The Cost of Labour Militancy There are two distressing examples from the recent history of power development in Kerala of the damages caused to the overall power and economic development of the State by the irrational behaviour of organised militant labour. The first is the example of Idukki Stage I, a 390 MW project, which could not be commissioned in the Fourth Five Year Plan (1969-74) and had a long gestation period because of frequent strikes and interruptions of work by labour. This project could be ultimately commissioned only in 1976. The Electricity Board suffered the consequences of delays caused in commissioning this project by way of escalation in costs and revenue foregone as a result of longer gestation period 8 years ago. At the time Idukki – I was commissioned in 1976, there were a large number of consumers in all sectors of the State’s economy waiting for power connections. Public memory being proverbially short, people have foregotten the great damage caused to the economy of the State by the long delay in the commissioning of Idukki, We would, however, like to recapitulate a recent experience of Idamalayar hydro-electric project, which unfortunately is yet to be commissioned (at the time this report is being got ready) because of unreasonable and irrational labour militancy. …………… The strike by the employees in this project started within three months of the commencement of work on the construction of the dam. The first strike was on 8-12-1976. There were a number of strikes between 8-12-1976 and 5-9-1979 by employees working in dam construction, but these strikes were settled without

19

much loss of time. But there was a long strike which increased the gestation period of the project by 6 months and 15 days (excluding monsoon off) which commenced on 6-6-1979 (ninth month of construction) and ended only on 25-31980. The direct financial loss on this account is estimated to Rs. 125 lakhs and it has also escalated the cost of the project by Rs. 142.5 lakhs. During the period between 7-5-1977 and 18-1- 1983, there were a number of strikes in the power house resulting in a total financial loss of Rs. 15 lakhs. The two strikes in the tunnel work of this project were something unique perhaps without parallel in the history of power development anywhere in the world. Initially the employees engaged in the tunnel work struck work between 9-6-1980 and 20-11-1980 increasing the gestation period by 5 months. But the most crucial strike which affected the project and postponed its commissioning was started on 10-4-1981 and continued till 10-6-1983 thereby postponing the completion of the project by 2 years and 2 months. The employees involved in the strike were only 110. The financial commitment for settling the strike was about Rs. 125 lakhs………….. The major issue causing this strike was the demand by the contractor’s employees engaged in this project for an assurance that they would be absorbed as permanent employees of the Electricity Board. We understand that a number of these workers were working as contractor’s labour in earlier hydro-electric projects in Idukki and elsewhere. But we cannot appreciate how this would give any moral or legal rights to these employees to claim permanent employment in the Electricity Board. It is difficult to quantify the losses to the community due to the 3 strikes (one in the dam construction and two in the tunnel work) extending over a total period of three years and one month. Cosidering that the total installed capacity of the hydro system in Kerala is only 1011.5 MW an addition of 75 MW three years earlier would have cushioned to some extent the power famine in Kerala especially in the year 1982-83. The losses to the Electricity Board as a result of the strike during dam construction has been estimated to be Rs. 267.5 lakhs. The losses due to the delay in completing the tunnel is estimated to be Rs. 30.98 crores out of which Rs. 29.31 crores is loss of revenue due to delay in commissioning of the project and Rs. 1.67 crores is due to escalation in costs and revision of schedules. The total loss incurred by the project as a result of the delay of three years and one month (1125 days) is Rs. 33.65 crores. The loss per day of delay works out to slightly less than Rs. 3 lakhs. This state of affairs did not stir the conscience of the people of Kerala who remained apathetic. A project being delayed for such a long time and every day’s delay costing Rs. 3 lakhs to the taxpayer did not receive adequate publicity in the Press or political platforms. That this could happen in a State with a vigilant press and politically conscious people is a tragedy.

20

We feel that an in-depth study by one of the all India management institutions into this strike, especially how and why it was allowed to continue for over three years and how and why the public opinion in the most literate State of the Country was silent, would be very useful to draw appropriate lessons for the future. We strongly recommend issuing an ordinance followed by enactment of appropriate legislation prohibiting strikes under any circumstances in all power projects under construction. …….Those who take part in such strikes and their leaders should get a minimum punishment of compulsory imprisonment for a specific period prescribed in such a law. In addition, all those who participate in such strikes should be debarred from being eligible for appointments under Government or any other institution owned or controlled by Government. …….. - Government of Kerala 1984: 57-61. in the death of two workers in police firing! It might be a cruel irony that the project (Stage I) could be completed and commissioned only under the coercive ‘normality’ during the infamous period of national emergency!

Idamalayar project was one of the most unfortunate victims of recurring and long-inertial periods of labour unrest. Some stories, as told in the Report of the High Level Committee (1984) of Government of Kerala, are given in the box below.

Kakkad project had a long tale of unending woes of corruption and trade union militancy. When construction works started, serious defects in design were found out. Initially the whole construction works were awarded to one contractor who had no prequalification but was preferred by the then concerned minister. The contractor was too inexperienced and inefficient to yield any progress in works for quite a long time, and the KSEB was forced to terminate the contract in June 1981and select fresh ones. The construction works on the interconnecting tunnel was started in 1980 at an estimated cost of Rs. 5.59 crores. Soon the workers went on strike, as the contractor refused to pay the ruling wage rate. In June 1981, another company was entrusted with the work, but still there was no progress; hence the work was divided and given to three contractors on

21

The Kakkad Saga of Leakages The Kakkad hydroelectric project of 50 MW, that took more than 20 years for completion, has been under the jinx on a number of fronts - excessive time and cost overruns and faulty planning and construction. One of the most infamous example in this connection was the costly effect of an engineering defect in the power tunnel construction that went on from the two opposite sides (with the good intention of expediting the work), but never meeting together. The two tunnels dug from opposite sides just went in parallel! Leakages in the power tunnel has been another recurrent problem. A major leak was detected in the concrete lining of the tunnel gate at Adit-5 of this 13 kmlong power tunnel just two months prior to the commissioning of the project in 1999. The KSEB had to spend Rs. 15 lakhs to repair the damaged portion using Epoxy mixture (The Hindu daily, 2 September 2001). And very recently, the project had to be shut down (on August 28, 2001) following the detection of a major leak through the same Adit-5. There are reports of widespread allegations of a corrupt nexus between certain KSEB quarters and the contractor lobby, attempting to create more and more work avenues in one or another way (ibid.). What is missing in general, however, is an expected social concern over the security of the tunnel and the dangerous consequences. And it must be so in an environment vicious of the political economy of rent seeking and the public indifference to it. condition that the work should be completed within 41 months, and the cost went up to Rs. 11 crores. In due course, three more contractors joined, yet by March 1988, only 30 per cent of the work could be completed! Rightly, it was also a situation where too many cooks were spoiling the broth. Reports show that in all there were 16 contractors entrusted with the work in different phases (The New Indian Express, June 20, 2001).

The tortoise continued its pace, but not on any race. Interrupted very often by agitations, the tunnel construction went on and on from two opposite sides, but it never met together; the two tunnels from opposite sides just ran in parallel! An excellent engineering feat!

Finally after 21.5 years, the tortoise reached its destination, eating away more than Rs. 150 crores.

22

The World Bank aided Lower Periyar project, visualised in the 1970s and cleared by the Planning Commission in 1983, also tells almost the same story of delays. The public sector National Power Construction Corporation (NPCC), that took up the civil works, just wasted more than 4 years without any progress. Finally this contract was terminated in 1993 in an out-of-court settlement and the private sector Hindustan Construction Corporation (HCC) entered the scene. The same company (HCC) had taken up the tunnel

The Lower Periyar Ecological Jinx The Lower Periyar power project involves some serious long-run fallout on environment not considered properly. The project causes a 15 km break in the course of the Periyar river, at least during summer, as it is diverted through the tunnel from Pambla to Karimanal, the power station site. The river, already tamed considerably by the Idukki project, thus ‘dies’ at Pambla and ‘resurrects’ at Karimanal, where the tail-race water from the power station gives life back to the river! works (in February 1984), with the deadline set on 26 October 1989. Later on HCC requested for time extension citing reasons as beyond their control, and the deadline was extended to 30 June 1992. Just one month prior to this date, HCC submitted to the KSEB a memorandum giving details of delays as follows – initial troubles: 5 months; labour problems: 10 months and 29 days; climatical problems: 10 months and 6 days; and obstructions/impediments on the part of the KSEB: 15 months! The company demanded for an additional payment of Rs. 16.33 crores to cover the increased costs due to this time over-run. They had already been allowed a cost increase of about Rs. 61.8 crores, against the original estimate of Rs. 23 crores. A committee, constituted to look into the fresh demand, recommended, surprisingly, a payment of Rs. 8.5 crores with an immediate disbursement of Rs. 2.5 crores to HCC. The alleged bias towards HCC of the committee, that never cared for the loss to the KSEB amounting to Rs. 117 crores due to the 47 months time over-run, made headlines in the media and the clamour echoed in the legislative assembly for days. The company moved the High Court and the matter went up to the Supreme Court; finally the KSEB had to eat its heart out! It should be added

23

that the World Bank, that had given aid to the project initially, but got reportedly frustrated over the time and cost over-runs, backed out long back.

On Inefficiency, Again! Anyone familiar with the history of the Lower Periyar project can narrate any number of instances the KSEB’s inefficiency and lack of seriousness in getting the job done on time. For instance, when the steel rope of the surge shaft’s gate snapped a few months [before its due date of commissioning], it took close to two months for the Board to retrieve the equipment from the power shaft’s well and to replace the rope. This could have been done in a few days, had the Board acted promptly. Instead of retrieving the equipment and thus speeding up the commissioning, the Board’s attention was focussed on fixing the responsibility for the disaster and find scapegoats….According to a rough estimate, the loss of a day’s power generation at Lower Periyar was Rs. 14 lakhs. Any responsible authority would have fixed the generating system first, and fixed the responsibility later…… One of the main impediments to the project becoming fully functional is said to be the delay in the arrival of the hoists of the five radial gates of the dam at Pampla. The Allahabad-based Thriveni Structurals, a public sector undertaking, was given the order for these equipment long back, in 1988. It failed several times to honour the commitment. Insiders allege that the KSEB miserably failed in forcing Thriveni to stick to its schedule or find alternatives [in time]. -The Hindu, 23 October 1997. Another jinxed project is Malampuzha, one of the first projects planned in the State to generate electricity from water let out from an irrigation dam. The contract for the design, supply and installation works were awarded to a private firm which allegedly had no previous experience in such projects. The civil work was done by the KSEB.

Though the company started the erection work in 1992, it took as many as four years to attempt at a trial run. However, during the trial run, some defects were noticed in the butterfly valve. In 1997, another trial run was tried, but again during the run, a valve disc got broken. And the story still continues…..

24

Chimony, locked in a High Court stay obtained by the contractor since 1993, on the other hand, is altogether left out from the KSEB reports now!

It is significant to note in this respect that the KSEB used to present, in its Annual Administration Report, a detailed status report on the progress of each project, which, however, has been missing for quite some time now. Absence of such transparency makes difficult any examination on the causes of delay.

It should not, however, be construed that every power project in Kerala necessarily falls under the jinx of delay. The NTPC thermal project at Kayamkulam could be completed and test fired on 1 November 1998, four months ahead of the schedule. Similarly, the first private sector hydroelectric plant at Maniyar (12 MW) could be completed and commissioned within 15 months in 1994, by the Carborandom Universal Company. In this light, it goes without saying that something is rotten behind the KSEB projects – and it is nothing but the dead political will, dead of corrupt politicians and indifferent public.

6. THE POLITICAL ECONOMY OF CORRUPTION

A detailed discussion of this aspect is provided in Kannan and Pillai (2001 b); below we sketch out the most relevant ones.

In a neo-classical representation of political process, the relationships among the public, government and utility may be aptly analysed in the light of a three-tier hierarchical model of principal-agent problem. The problem consists in the default and breach of trust (i.e., moral hazard and adverse selection, Arrow 1985), likely on account of the conflicting objectives of self-interest maximisation of the concerned parties and the uncertainty or information asymmetry involved in the relationship. In its simple version, it is assumed that in a regulatory governance structure, the principal's (i.e., the public's) objective is to maximise some measure of social welfare, while the agent (the government as supervisor) and the sub-agent (utility) aim to maximise the returns of their

25

respective rent seeking pursuits. In a complex structure of relationships, the principal may be viewed as a composite set of sectional interests against the background of the general welfare objective; each class in this composite set, such as the contractors, construction workers, bureaucracy, politicians and others, follows its own designs of predatory rent seeking that dominate, in a particular context, the common objective. Such a structuring facilitates to analyse the political economy of corruption involved in the time and cost overruns in the power projects in Kerala.

Apart from the usual 'sales' procedures of construction contracts and materials purchase orders carried out by means of a collusion between the supervisor (government) and the sub-agent (bureaucracy in the utility), favouring certain contractors, the practice of allowing for time overruns of projects and sanctioning the associated cost escalations involves a ‘wide spectrum collusion’ among the domineering class interests in the composite principal set, viz., the political party in power (i.e., government), bureaucracy, contractors and trade unions. As already highlighted, recurring unrestricted labour militancy is recognised in general as the single factor that puts the heaviest burden on the pace of the construction works of power projects in Kerala, largely dictated by partypolitical rivalry rather than genuine labour demands, as for example, in the construction of Idukki hydro-electric project, to begin with. The time overruns out of the striking militancy upon one or another pecuniary pretext essentially go into the contractors’ demand for cost escalation, that is soon endorsed by the Board and sanctioned by the government.i Such rent-sharing is a widely recognised official practice in the powerirrigation sectors. The glaring laxity on the part of the government in fulfilling its committed responsibility for enforcing its authority on the contractors and workers to bind them within the contractual terms they agreed to take up to honour is a clear indication of its corrupt collusion. As mentioned above, in Kerala, the time and cost overruns have afflicted only the State power projects; the public sector NTPC thermal and the private sector hydro projects in the State having been completed well within their scheduled times. In this light, then, the cost escalation sanctioned for each late-run project may rightly be taken to represent the cost of corruption involved in construction contract sales in the power sector of the State. Accounting for the general price inflation during

26

the normal construction period, this amounts to Rs. 644 crores or Rs. 35.8 crores per project!1 Unbelievably, it represents on an average about 60 per cent of the actual project cost! In some cases it is well above 70 per cent; for example, Sabarigiri Augmentation (75 per cent), Kakkad (74 per cent), Malankara (76 per cent), Poringalkuthu left bank extension (71 per cent), Kuttiady extension (80 per cent) and the diversion projects of Azutha (71 per cent), Kuttiar (73 per cent) and Vazhikkadavu (84 per cent). This is all shared among the four parties involved, at the cost of the helpless majority in the ‘principal’ set of tax payers.

Such lucrative rent sharing collusion has unfortunately become firmly institutionalised in the political process of the country. A highly individualistic selfinterest domineering ethos have come to stay across the social texture only to strengthen this political economy of corruption. It is not that the principal, the public at large, is unaware of all these murky dealings and developments; but they largely remain apathetic, even after enlightened enough in one or another way by the Press, true to the rotten spirit of an individualistic utilitarian society, lying moribund but never dying. This in fact questions at least to some extent the validity of the neo-classical apology of imperfect information as leading to the principal-agent problem. What is at heart of the malady is a lack of a sense of oneness, resulting in the void of an effective platform of checks and balances, that would have avoided problems arising from moral hazards and adverse selection. And this should point towards the significance of a soul-cleansing cultural revolution, reminiscent of that of the era of liberalism.

This may, however, appear a highly idealistic long-term objective. We do recognise the exertion of significant public praxis by a few concerned citizens and their organisations for immediate, palliative results. Strengthening and extending such praxis can go a long way towards imposing the public will for common interests on the political process. For example, there are measures that can effectively be applied to restrain time 1

Excluding the hydro projects of Kallada and Pooyankutty, and the two diesel power plants. If we stick to the strict assumption that the original project cost estimate allow for possible inflation during construction period, such that the estimate be as on the completion date, then the corruption charges involved would be very much higher.

27

and cost overruns in the public projects: the construction contracts be so structured as to provide for making the contractors liable for stringent penalties in case of nonperformance such as time overrun. The previous LDF state government (1996-2001) was reported to have made some steps in this direction in the case of the Athirappally hydroelectric project by initiating to institute in the contract penalty provisions for delay something of the first kind in the history of the KSEB, if implemented. And it is such ifs that govern the direction and tempo of our development. ----------------------------------------------------

28

Table 1 Profile of Time and Cost Overruns of the Projects

Project

Energy Potential (MU) 1007 376 125 320 262 53 493 5.6 6.4 65 6.5 11.5 645 57 74 36.6 12 24 15 75

Year of Starting

Originally Scheduled Year of Completion 1978 1981 1980 1978 1986 1989 1991 1989 1989 1990 1990 1990 1991-92 1992-93 1990-91 1991-92 1992-93 1992-93 1995-96

Idukki II Stage 1970 Idukki III Stage 1975 Sabarigiri Augmentation 1972 Idamalayar 1970 Kakkad 1976 Kallada 1981 Lower Periyar 1983 Malampuzha 1987 Madupetty 1987 Malankara 1987 Chimony 1987 Peppara 1987 Pooyankutty 1986 Azhutha Diversion 1987 Poringalkuth LB Extn 1989 Kuttiar Diversion 1988 Vadakkepuzha Diversion 1989 Vazhikkadavu Diversion 1989 Kuttiady Tail Race 1989 Kuttiady Extension 1992 Note: * = by 1993. Source: Government of Kerala, Economic Review (various years).

Year of commissioning 1985-86 1991 1990-91 1987 1999 1993-94 1997-98 ? 1998 ? ? 1996 ? 1998 1999 ? ? ? ? 2000

Estimated cost (Rs. Lakhs) Original Actual 3168 6800 410 1511 128 1122 2340 9003 1860 15352 1180 1802 8843 35304 295 679 292 478 780 4113 314 425 392 681 25000 82000 290 1446 902 4273 214 949 131 514 186 1599 397 1292 3073 19800

29

Table 2 Cost Escalation of Power projects in Kerala (as in 1999-2000)

Projects

Time Overrun (Years) (%)

Idukki II Stage 8 Idukki III Stage 10 Sabarigiri Augmentation 10 Idamalayar 9 Kakkad 13 Kallada 5 Lower Periyar 6 Malampuzha 10 Madupetty 9 Malankara 9 Chimony 9 Peppara 6 Pooyankutty 15 Azhutha Diversion 6 Poringalkuth LB Extn 6 Kuttiar Diversion 8 Vadakkepuzha Diversion 8 Vazhikkadavu Diversion 7 Kuttiady Tail Race 7 Kuttiady Extension 4 Note: * = by 1993; NAP = Not Applicable.

100 166.67 125 112.5 130 62.5 75 500 450 300 300 200 120 150 400 400 233.3 233.3 100

Cost Overrun (Rs. Lakhs) (%) 3632 1101 994 6663 13492 622 26461 384 186 3333 111* 289 57000 1156 3371 735 383 1413 895 16727

114.65 268.54 776.56 284.74 725.38 52.71 299.23 130.17 63.70 427.31 35.35* 73.72 228.00 398.62 373.73 343.46 292.37 759.68 225.44 544.32

WPI (All Commodities) Increase (%) 258.31 237.89 403.15 305.92 461.15 147.80 193.68 168.53 144.55 168.53 71.96 118.32 182.67 144.55 112.67 132.42 132.42 132.42 132.42 73.03

30

Table 3 Extra Energy and Revenue Realisable from Timely Completion of Projects

Year 1983-84 1984-85 1985-86 1986-87 1987-88 1988-89 1989-90 1990-91 1991-92 1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-2000 Total

Firm Energy (MU) Actual Realisable 3725.73 3725.73 4397.33 5053.13 5053.13 5053.13 5053.13 5554.13 5554.13 5554.13 5607.13 5607.13 5607.53 5619.03 6118.43 6249.43 6586.43

5554.13 5554.13 5816.13 5816.13 5816.13 5816.13 5881.13 6000.73 6562.73 6675.73 6675.73 6675.73 6751.13 6751.13 6751.13 6751.13 6751.13

Generation (MU) Actual Realisable 3643.4 4884.9 5357.1 4642 4093.1 4548 5075 5491 5326 6189 5822.3 6572.3 6662 5502.9 5188.7 7601.6 7655.57

5431.38 7282.02 7085.79 5342.70 4711.07 5234.52 5906.42 5932.32 6293.00 7438.77 6932.08 7824.62 8020.34 6611.38 5725.63 8212.07 7846.84

Extra Energy Saleable (MU) 1326.69 1797.91 1298.12 508.85 439.11 521.75 648.62 346.28 756.60 987.34 886.60 1001.38 1086.20 887.38 440.70 501.95 158.10

Extra Revenue Realisable (Rs. Crores) 46.67 53.83 39.94 24.55 24.44 29.58 34.65 18.35 45.40 73.01 72.75 86.80 100.93 84.84 56.91 67.52 26.16

13593.58

886.33

31

Table 4 Capital Cost Savings Capital Cost (Rs) per kWh of Savings in Capital Waste Energy Potential Capital Cost Factor** Original Actual (Rs. Lakhs) Idukki II Stage 0.31 0.68 910.20 0.15 Idukki III Stage 0.11 0.40 844.33 1.27 Sabarigiri Augmentation 0.10 0.90 838.42 2.96 Idamalayar 0.73 2.81 4652.58 1.07 Kakkad 0.71 5.86 11386.17 2.87 Kallada 2.23 3.40 -153.26 Lower Periyar 1.79 7.16 18940.91 1.16 Malampuzha 5.27 12.13 339.78 1.00 Madupetty 4.56 7.47 142.23 0.42 Malankara 1.20 6.33 3124.06 3.16 Chimony 4.83 6.54* 26.89* 0.07 Peppara 3.41 5.92 184.00 0.37 Pooyankutty 3.88 12.71 NAP Azhutha Diversion 0.51 2.54 1027.80 2.46 Poringalkuth LB Extn 1.22 5.77 3028.05 2.43 Kuttiar Diversion 0.58 2.59 696.27 2.75 Vadakkepuzha Diversion 1.09 4.28 349.72 2.13 Vazhikkadavu Diversion 0.78 6.66 1342.28 5.23 Kuttiady Tail Race 2.65 8.61 744.06 1.36 Kuttiady Extension 4.10 26.40 15825.39 3.98 Note: * = by 1993; NAP = Not Applicable; ** = Equivalent to number of Kakkad-type projects Projects

Capital x Time Waste Factor (%) 329.29 882.76 1872.27 717.58 1798.37 148.16 598.65 1281.02 800.34 2009.23 441.40* 421.17 NAP 996.97 1084.31 2117.29 1861.83 2765.59 984.80 1188.64

32

Appendix TIME AND COST OVERRUN OF POWER PROJECTS IN KERALA KAKKAD KALLADA LOWER PERIYAR Year of Starting 1976 1981 1983 Original Cost Estimate, Rs. Lakhs 1860 1180 8843 Energy Potential, mu 262 53 493 Original Expected Year of Commissioning 8 Years 8 Years 8 Years

Year 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999

KAKKAD KALLADA Revised Cost Revised Year of Revised Cost Estimate Commissioning Estimate 4117 1986-88 1180 5500 1990 1389 5500 Jul-90 1389 6941 Sep-91 1389 7012 Sep-92 1389 7012 1993-94 1389 7012 1993-94 1389 8800 1995-96 1437 9869 1995-96 1606 10935 1995-96 1802 15080 1996-97 15080 1997-98 14599 Jun-98 14599 March 1999 15275 Commissioned on 14.10.1999

LOWER PERIYAR Revised Year of Revised Cost Revised Year of Commissioning Estimate Commissioning 1989 10050 1990 1987-88 14209 1990 Aug-89 14209 1990 Dec-90 10900 Sep-91 Dec-90 14000 Sep-92 1992-93 14000 1993-94 14000 1994-95 1993-94 18000 1994-95 1993-94 26000 1995-96 Commissioned 27300 1995-96 in 1993-94 27300 1996-97 27300 1997-98 29899 Commissioned 29899 In October 1997 35768

33

MALAMPUZHA MADUPETTY Year of Starting Original Cost Estimate, Rs. Lakhs Energy Potential, mu Original Expected Year of Commissioning MALAMPUZHA Year

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1988 1999 2000

1987 295 5.6 1989 MADUPETTY

Revised Cost Revised Year of Revised Cost Estimate Estimate Commissioning Estimate 345 Nov-90 332 345 Dec-90 332 345 1992-93 332 345 1992-93 332 425 1993-94 365 425 1993-94 435 486 1994-95 453 675 1995-96 775 675 1997-98 775 611 Mar-98 546 608 1998-99 478 677 1999-2000 478 679 2000-01

1987 292 6.4 1989

MALANKAR A 1987 780 65 1990

MALANKAR A Revised Year of Revised Cost Revised Year of Commissioning Estimate Nov-90 780 Mar-91 997 1992-93 997 1992-93 997 1993-94 1600 1993-94 1600 1995-96 1665 1995-96 1298 1997-98 1298 Commissioned 4336 on 16.1.1998 4157 4113 4113

Commissioning 1991-92 1992-93 1993-94 1993-94 1995-96 1995-96 1996-97 1996-97 1997-98 1998-99 March 2000 2001-02 2001-02

34

Year of Starting Original Cost Estimate, Rs. Lakhs Energy Potential, MU Original Expected Year of Commissioning

Year 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000

CHIMONY 1987 314 6.5 1990

CHIMONY PEPPARA Revised Cost Revised Year of Revised Cost Estimate Estimate Commissioning Estimate 360 1991-92 567 360 1991-92 567 360 1992-93 567 360 1992-93 567 425 1994-95 580 425 1995-96 580 580 850 850 850 625 671 681

PEPPARA 1987 392 11.5 1990

POOYANKUTTY - I 1986 25000 645 8 Years

POOYANKUTTY - I Revised Year of Revised Cost Revised Year of Commissioning Estimate 1991-92 25000 1991-92 25000 1992-93 25000 1992-93 25000 1993-94 25000 1994-95 25000 1994-95 59000 1995-96 59000 Commissioned 59000 in June 1996 59000 82000 82000 82000

Commissioning 1992-93 During IX Plan

2003-04 8 Years 8 Years 8 Years

35

AZHUTHA DIVERSION Year of Starting 1987 Original Cost Estimate, Rs. Lakhs 290 Energy Potential, mu 57 Original Expected Year of Commissioning 1991

Year 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000

PORINGALKUTHU LEFT BANK EXTN.

1989 902 74 1992-93

KUTTIAR DIVERSION 1988 214 36.6 4 Years

PORINGALKUTHU LEFT BANK EXTN. KUTTIAR DIVERSION AZHUTHA DIVERSION Revised Cost Revised Year of Revised Cost Revised Year of Revised Cost Revised Year of Estimate Estimate Commissioning Estimate Commissioning Estimate Commissioning 300 Aug-91 214 1992-93 370 Oct-91 902 1992-93 214 1992-93 370 1992 902 1993-94 214 1993-94 370 1993 902 1993-94 214 1993-94 420 1994-95 2192 1995-96 254 1994-95 600 1994-95 2600 1996-97 660 1995-96 784 1995-96 2334 1996-97 496 1996-97 850 1995-96 2490 1996-97 755 1996-97 850 1997-98 2490 1997-98 755 1998-99 1399 March 1999 3689 June 1998 814 1998-99 1399 March 1999 3669 1999 836 May 1999 1461 Partially 4318 1999 949 1999-2000 1446 commissioned 4273 Commissioned 949 2001-02 in June 1998 in 1999

36

VADAKKEPPUZHA VAZHIKKADAVU DIVERSION

Year of Starting Original Cost Estimate, Rs. Lakhs Energy Potential, MU Original Expected Year of Commissioning VADAKKEPPUZHA DIVERSION

Year 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000

1989 131 12 1991-92

DIVERSION

1989 186 24 1992-93

KUTTIADY TAIL RACE 1989 397 15 1992-93

KUTTIADY EXTENSION 1994 3073 75 1995-96

VAZHIKKADAVU DIVERSION

Revised Cost Revised Year of Revised Cost Estimate Estimate Commissioning Estimate 131 1991-92 185 131 1991-92 185 131 1991-92 185 160 1994-95 200 160 1994-95 359 160 1996-97 419 385 1996-97 1595 385 1996-97 1595 385 1996-97 1595 786 May 2000 2000 705 Not fixed 1564 514 2001-02 1599

KUTTIADY TAIL RACE Revised Year of Revised Cost Revised Year of Commissioning Estimate 1992-93 397 1992-93 1993-94 1995-96 1996-97 1996-97 660 1996-97 1000 1998-99 1000 1998-99 1448 May 2000 1338 Not fixed 1123 2001-02 1292

Commissioning 1992-93

1995-96 1995-96 1997-98 31.7.1999 1999 2000-01 2001-02

37

Cost escalation (%) Kuttiady Extension

Kuttiady Tail Race

Vazhikkadavu Diversion

Vadakkepuzha Diversion

Kuttiar Diversion

Poringalkuth LB Extn

Azhutha Diversion

Pooyankutty

Peppara

Chimony

Malankara

Madupetty

Malampuzha

Lower Periyar

Kallada

Kakkad

Idamalayar

Sabarigiri Augmentation

Idukki III Stage

Idukki II Stage

Cost Escalation of Power Projects Vs. General Price Inflation

900

800

700

600

(%) 500

400

300

200

100

0

Inflation (%) 38

39

CHAPTER 6

T & D LOSS IN KERALA POWER SYSTEM

1. General – Technical Notes The supply of electricity in its usable form to the consumers involves three distinct functions, viz., generation, transmission and distribution corresponding to production, transportation to market and retail distribution of many other products. However, unlike other products, transportation (i.e., transmission) of electricity involves a transformation. Thus the power generated at a plant is increased to high voltage (HV) level at source by step-up transformers and is fed into the busbars in the transmission substation. Here the voltage is further increased to extra-high voltage (EHV) and sent to the substation nearer to the load centre, where the voltage is reduced by step-down transformers, after which the power is fed, through the HV transmission lines, into the distribution substations located close to the customers. Here another voltage reduction occurs and the power flows out of the distribution substation busbars through primary feeders to various distribution transformers where the voltage is further reduced. Finally the power is delivered at the user-end through the secondary distribution line.

The total amount of power (kW) and energy (kWh) generated at the source will be greater than the corresponding values actually consumed at the user-end by an amount equal to the losses in the system. Transmission and distribution (T&D) network losses occur in the lines, substations and transformers. In general, power losses can be attributed to

1) power station auxiliary consumption, 2) transformation and transmission consumption, 3) sub-transformation and sub-transmission consumption, 4) distribution consumption, and 5) other losses.

Since the power losses of sub-heads (1) – (4) are necessarily for the processes of generation, transmission and distribution of power in the true sense, they constitute the functional energy consumption rather than losses in the system, while sub-head (5) represents the real losses.

The power station auxiliary consumption depends upon its layout, operational conditions, automisation and design of various sets of equipment. The optimum level of such consumption is about 6 – 7 per cent in a modern thermal plant and 0.5 per cent in a hydro-plant.

The transformation, transmission and distribution consumption is dependent upon the design of the T&D system, substation equipment, operational conditions and quality of materials and maintenance efficiency. In general this consists of line (transmission and distribution) and transformer (transformation) losses (i.e., consumption). The line losses 2

are the I R losses where I stands for current (in amperes) and R for resistance of the conductor (in ohms). Thus line consumption increases with current and resistance. That is why power is transmitted at very high voltage; with high voltage transmission, the 2

current, and therefore, the I R losses and voltage drop are lower than with that at lower voltage. The other component, transformer consumption, is the sum of iron and copper losses (consumption) in the transformers. The iron loss of a transformer is independent of its load, depending only upon the characteristics of the transformer itself. Thus so long as the transformer is energised, it absorbs a constant amount of power. Copper losses, on the other hand, are, for any given copper loss at full load, proportional to the square of the load (in kW), the proportionality constant depending on the physical properties and voltage level of the section. Thus both line and transformer consumption vary over time and space. In general, however, a reasonable level of T&D consumption is of the order of 10 – 15 per cent in developing countries.

Thus, in reality, the difference between the power generated less the power required for the functioning of the power system minus the actual consumption by the

2

users can be termed power losses, the sub-head (5). These can further be classified into design losses, operating losses and unaccountable losses.

Design losses result in sub-heads (1) – (4), from the system design adopted on a compromising basis due to non-availability of materials, equipment, finance, etc.

Operational losses are due to under-loading or over-loading conditions of the power system in its developmental stage only.

These two losses fall in the category of technical losses, while last one, unaccounted losses, due to defective metering and degenerate billing system, unmetered supplies and theft, etc., represent the non-technical losses.

To Reduce Power Loss

As the losses vary directly as the resistance and square of the current, it is necessary to keep both these quantities at the minimum possible, depending upon the economic viability and technical feasibility to achieve the most minimum loss conditions. The resistance may be reduced by lines of smaller lengths having bigger size conductor, by proper planning of distribution system; and the current may be reduced by improving poor power factor likely to occur due to inductive loads, line inductance and magnetising currents of transformers by the application of shunt capacitors, etc.

-Government of India 1974.

The efficiency of a T&D system can be measured in terms of the ratio of the receiving end power to the sending end power; the smaller the difference between the power at the two ends (i.e., the transit loss), the nearer this ratio to unity, the maximum efficiency. Thus one less this efficiency ratio gives the loss factor that may be taken to represent the inefficiency of the system.

3

2. T&D Losses: The Kerala Scenario

Kerala once enjoyed a covetably respectable status with an apparently optimal level of T&D losses of around 12 per cent especially during the 1970s except for a few years; in 1977-78 it was as low as 11.4 per cent of generation (less auxiliary consumption). However, the loss percentage started shooting up during the 1980s and reached a phenomenal zenith of 28.95 per cent in 1987-88, exactly a decade after the covetable nadir point of 11.4 per cent. After that, it began declining, first a little precipitously to 22 per cent within two years and then very gradually, as if it reached a plateau, to 20.05 per cent in 1995-96 (see the chart). The T&D loss factor of Kerala has been well above that of her neighbours: thus in 1993-94, the T&D percentage loss of A.P. was 20.21, of Karnataka, 19.49, and of Tamil Nadu, 16.99, while that of Kerala was 20.52, above the Southern region average of 18.94, but just below the all-India average of 21.41per cent.

The unprecedented increase in T&D loss that Kerala had to experience after the turn of the 1980s has been a hot subject of concern. The inefficiency in the T&D system as reflected in the high loss factor has soon been identified as caused mainly by inadequate investment in transmission as well as its belated realisation. The Economic Review (Kerala) started lamenting on this plight in the transmission sector as far back as in 1980 itself.

3. The Possible Causes

Like any other power system in its developmental stage in India, Kerala too concentrated much of her attention and effort on power generation and earmarked larger financial outlays for the development of generation facilities. The consequent neglect of the transmission system which has later on resulted in raising its ugly head of the cumulative reactionary effect, is evident from the fact that out of about Rs. 384 crores invested in power development up to 1977-78, the end of the fifth plan period, in Kerala, as much as Rs. 237 crores (61.6 per cent) went into generation, and the transmission,

4

distribution and rural electrification works were left starving for adequate resources. As on 31.03.1978 generation accounted for about 60 per cent of the total fixed assets (capital outlay of Rs. 300.69 crores) of the power system in Kerala, with transmission, high tension (HT) distribution and medium and low tension(M<) distribution sharing the balance among themselves in the percentages of 14.2, 6.95 and 18.8 respectively.

Coupled with this were a host of other problems leading to avoidable time overrun. Forest Conservation Act (1980), and organised mass resistance from the public against cutting of trees stood in the way of laying of transmission lines and left many projects on construction of substations to remain on paper only. Thus several of the substations and lines originally scheduled for construction during the fourth and fifth five year plans (during 1970s) remained incomplete even during the 1980s (Economic Review, 1980.1981 and 1982).

Thus the transmission network of the State could not be built in a proper way enough to meet the load. And the deficiencies manifested themselves in the inadequate transformer and line capacity and led to frequent interruptions, poor voltage conditions, high percentage of losses, etc. The painful realisation of this reality lent the much desired direction to the restructuring of the investment programmes in the power development of Kerala, and the subsequent plans apportioned more resources to the needy T&D sectors, starting with the annual plans of 1978-79 and 1979-80 that earmarked respectively 64 and 72 per cent of the total plan outlays to the development of T&D.

However, this belated effort seems to have been too negligible to make any dent in the deficiencies accumulated in the T&D system. The growth in transmission capacity has not been able to catch up with that in the load. That transmission of power at very high voltage is more economical in the sense of less line losses is a well-known fact, yet its concrete recognition in Kerala is conspicuous by its absence. The recently laid 400 kV transmission line to import power from Udumelpet to Thrissur substation is the only

5

Illegal Extraction of Energy ……The large difference now noticed between the quantum of energy generated (and purchased) and that sold cannot be wholly attributed to normal losses in transmission. There is no doubt that this is partly due to illegal extraction of energy by consumers and others. With greater awareness of modern technological advances, unscrupulous people are likely to employ different methods for bypassing meters and making them ‘dead’ while power is flowing in the system. According to an estimate given to me by the Executive Engineer of the Research Division, Pallom, 10 to 12 per cent of electrical energy is lost by pilferage and wastage….. [He] also informed me that a study was conducted on the methods of preventing theft and that his proposals were submitted to the Chief Engineer (Electrical) more than a year back [and shelved away for ever, as usual!]. There seem to exist some effective methods by which theft of energy could be reduced to a large extent. My own impression is that this aspect of the working of the system has not received the attention it deserves. -Government of Kerala 1967: 20-21. transmission line of highest capacity in Kerala. The progress of the laying of 220 kV lines, that appeared only by 1966-69 and are still confined to a limited region only, has been too tardy; though it increased from 317 circuit kilometre by the beginning of the fourth plan (1969) to 1897.5 circuit (ct.) km in 1998-99. It actually fell drastically in terms of its capacity per mw of maximum demand from 1.14 ct. km. per mw to 1.0 ct. km. per mw during the same period. And this is with the much-restricted (internal) maximum demand. If we could take into account the unrestricted maximum demand, the 220 kV lines capacity would be found to be too negligibly small to meet any load at all.

110 and 66 kV lines appear to have been the rule and vogue in the transmission sector in Kerala, though the focus is fast shifting from the latter to the higher capacity lines. However, the capacity of even these lines to meet the increasing maximum demand has been steadily on the decline – the capacity of the 110 kV lines falling from 4.5 to 1.5 ct. km per mw of maximum demand during 1961 – 99 period, and that of 66 kV lines from 16.98 to 1.4 ct. km per mw over 1956 – 99 period, considering the restricted maximum demand.

6

Even though severely restricted in many ways, demand for power has always been on a surging trend. The internal average load defined in terms of generation plus import less export increased from 168 MU in 1950 – 51 to 11197.13 MU in 1998-99 – an increase of about 67 times with an annual compound growth rate of 11.7 per cent. The number of consumers increased from a mere 0.28 lakhs in 1951 to 56.4 lakhs by 1998-99 – a 200-fold increase, with an annual compound growth rate of 15 per cent, and the connected load from 70 MW to 7276 MW over the same period, registering a 104-fold increase with 13 per cent annual compound growth rate. This expansion has been

Delays in Transmission Sector It is most essential that side by side with construction of new power projects, development of transmission and distribution as also expansion thereof are carried out in order that, by the time the generating stations are established and power becomes available, there will be as little time lag as possible in the utilisation of power…….. [This, however, has not been so in Kerala; take for example the Sabarigiri episode.] Unfortunately, the contracts for transmission line towers and construction of the important lines were awarded to a person who had no previous experience in transmission works. The Board had perhaps taken this decision in view of the large difference between his tenders and the next lowest……In spite of the rigid conditions and heavy penalty imposed, the contractor could not execute the works in time and it was with the greatest difficulty and engaging other agencies also that even the 220 kV lines from Sabarigiri to Pallom was completed just in time for transmission of power from Sabarigiri. In the case of most of the other lines also, the Board had to take steps to terminate the contracts and entrust the work to other agencies……… The real cause for delay in implementation of the transmission programme has been due to not having a definite time schedule and no target dates for completion of the various works. Even now, there does not appear to be any definite dates for completion of pending items of work under transmission…. -Government of Kerala 1968: 3, 13-14, 15.

7

reportedly facilitated by the ‘special drive’ launched by the KSEB since 1980-81 for clearing the pending applications for new electric connections. Coupled with this has been the substantial effect of the effort on the rural electrification front. All the villages in the State as per the 1971 census have been electrified by as early as 1979-80. The cumulative result of all these may be seen to have reflected itself in the wide expansion of the distribution network; the capacity of the LT lines has had a 146-times increase during 1951-99 period (about 14 per cent compound growth rate annually). In relation to the maximum demand, the distribution capacity increased from about 40 km per MW to 76.3 km per MW of maximum demand over the same period.

It should be remembered that this higher and increasing level of LT lines per mw of demand itself is a pointer towards the inadequate HT and extra high tension (EHT) systems. In general, it is pointed out, the capacity of the distribution transformers must keep pace with increasing load, the number of ct. km of HT lines will rise more slowly, and that of LT lines will increase more slowly still, thanks to some sort of scale economies. However, this is true for a system already having T&D capacity sufficient to meet the normal level of demand; the Kerala T&D network, born and reared with an with an inherent deficiency and the consequent inefficiency, falls far wide of this ideal standard. The system needed bigger push to be cleared of the accumulated deficiency reminiscent of the Aegian stable. All the exercises planned and devised to improve the system have more or less failed in their mission thanks, as will be seen shortly, mainly to the under-utilisation of resources meant for the deficient transmission sector or their massive diversion in to the distribution sector, as well as to the socio-political impediments we have already mentioned. Thus with the rising demand, the distribution network expanded and deepened at the cost of the transmission capacity, making it overloaded.

It is true the priority in the structure of the plan investment on power development has been shifted from generation to T&D since the 1978-80 annual plans. However, the transmission sector has again been left unattended and poor; because of the everincreasing demand for new connections, the major share of the resources set apart for

8

T&D had to be diverted to the distribution system (Economic Review, 1980, p. 72). The distribution network has grown at the cost of the transmission system leaving the latter still more deficient and incapable against an ever-surging load.

Another twist of events that has plagued the transmission capacity in the State runs in terms of the wide deviations among the plan investment proposal, approved allocation and the actual expenditure in the different sectors of the power system in the State. Thus in the case of the VI plan (1980-85) the plan proposal by the KSEB was in the proportion of 1.2:1:1.6 for the three sectors, viz., generation (Rs. 131 crores), transmission (Rs. 106.5 crores) and distribution including rural electrification works (Rs. 167 crores) respectively, whereas the approved allocation by the planning commission to each of these sectors was of the order respectively of only 70 per cent (Rs. 91.66 crores), 75 per cent (Rs. 80 crores) and 82 per cent (Rs. 136.72 crores) of the plan proposal. However, the actual expenditure, as usual, turned out to be much higher in the generation sector (Rs. 138.35 crores, higher than even the plan proposal), and very much less in the transmission sector (Rs. 67.61 crores, about 85 per cent of the allocation and only 64 per cent of the plan proposal). In the distribution sector, the actual expenditure was Rs. 108.67 crores, about 80 per cent of the proposed outlay. (Sector-wise break up of data is not readily available for the VII plan). Such awful tendency of avoidable under-utilisation and lapse of the resources allocated for a starving area as is the transmission sector in Kerala still continues, along with the easy choice and practice of diversion of funds meant for the development of transmission capacity into other fields, especially the distribution system in the face of the expansionary connection demands. Thus in the annual plan of 1990-91, the plan proposal for the two sectors, transmission and distribution including rural electrification works, were respectively Rs. 63 crores and Rs. 33 crores, but the actual expenditure fell short (by about 12 per cent of the proposal) in the transmission sector (to Rs. 55.49 crores, about 88 per cent) while it upcrossed in the other sector (Rs. 42.9 crores, 130 per cent). The story still continues. While in the first year (1992-93) of the VIII plan (1992-97), the deviation of the actual expenditure (Rs. 68.31 crores) was by (-) 8.3 per cent in transmission and by (+) 7.7 per cent in distribution (actual: Rs. 70.37 crores against proposal: Rs. 65.35 crores), the deviation in

9

the fourth year (1995-96) was by (-) 41 per cent (expenditure: Rs. 83.43 crores; approval: Rs. 141.6 crores) in the transmission sector; in distribution also, the actual (Rs. 121.77 crores) fell short of the proposal (Rs. 208.40 crores, about 58 per cent).

Similarly, in 1998-99, the expenditure in the power sector (Rs. 754.4 crores) far exceeded the outlay (Rs. 650 crores). Reminiscent of the initial period of power development, generation sector claimed a major chunk of this, quite rightly in view of the capacity shortage in the State. Expenditure (Rs. 390.5 crores) in this sector was nearly twice the outlay (Rs. 214.3 crores). In the distribution sector too expenditure (Rs. 200 crores) exceeded the outlay (Rs. 186.9 crores). But, for transmission, as usual, expenditure (Rs. 75 crores only) fell short of the limited outlay of only Rs. 90 crores by 17 per cent.

No wonder the tilt in the structure of the capital outlay in favour of generation (which, as we have already seen, was 60 per cent as on 31.03.1978) has gradually changed its orientation towards distribution. After a decade, as on 31.03. 1988, distribution (EHT, HT and LT) accounted for about 48.3 per cent of the total fixed assets of Rs. 548.61 crores in the State power sector, generation for 35.1 per cent (Rs. 192.80 crores) and transmission for only 16.6 per cent (Rs. 90.88 crores). The share in fixed assets of generation steadily declined and that of transmission shot up in the next few years such that as on 31.03. 1996, a historical balance between the two was eventually achieved at about 27 per cent. Distribution amassed the remaining major share of 46 per cent of the total capital outlay of Rs. 1508.62 crores. Transmission grew at a phenomenal compound growth rate of 20.6 per cent per annum during this 8-year period to reach this higher level of share, up by 10 per cent over 1978. By 1995-96, the structure of the capital outlay in the three sectors viz., generation, transmission and distribution was in the proportion of 1 : 1: 1.72 against 2.12 : 1: 2.90 by 1987-88 and 4.23 : 1 : 1.81 by 197778.

However, this balance was lost by the end of 1997-98 and the system reverted to the old lopsided structure of capital assets. Generation now accounted for about 40 per

10

cent of the total fixed assets of Rs. 2275.14 crores (with hydel and thermal sharing it in the ratio of 3:2). The share of distribution (Rs. 863.8 crores) fell to about 38 per cent and that of transmission (Rs. 495.64 crores) still down to about 22 per cent only.

If the earlier tempo of attention and effort forced out of the deep concern and regret over the plight from the higher level of losses were maintained intact and the system were simultaneously cleared of the usual harmful practices of under-utilisation and diversion of the apportioned resources, the cumulative effect must have appeared in the desired direction by now. However, considering the severe generation capacity shortage in the State now and the consequent concentration of attention on that sector, the situation in the T&D sector is likely to worsen still further. This in turn stands to mop up much of the cumulative effect of the lumpy investment in the transmission sector in the State carried out during the early 90s. Implied in these circumstances is a painfully unavoidable fact of a reversion of the system to higher levels of energy losses. The situation might be eased to some extent if stringent action is taken against the avenues of non-technical losses.

The fall of the loss percentage from 28.95 to 17.87 marks an annual decay rate of (-) 4.7 per cent. The system loss percentage is estimated, as we know, in relation to the average load, which during this period grew at a compound rate of 6.38 per cent per year. The fixed assets in transmission that grew at a compound rate of 18.5 per cent per year over the same period, thus maintained, relative to the average load, an annual compound growth rate of 11.38 per cent, giving an elasticity of power loss with respect to capital cost in transmission (both in relation to average load) equal to (-) 0.41. That one unit increase in transmission capital outlay resulted in only 0.4 unit reduction in the loss factor during this period substantiates the earlier conclusions i) that there is substantial loss of power on non-technical front, and ii) that much of the cumulative effect of the lumpy investment in the transmission sector of Kerala carried out in the 1990s so far is yet to appear or is getting neutralised.

11

Another important aspect to be considered is the unavoidable requisite for very long transmission lines associated with the dependence solely on hydro-power thanks to its very peculiar confined location. As many as 8 power stations in Kerala, with an installed capacity of 1101.5 MW (about 73.2 per cent of the total of 1505.5 MW as in 1995-96) are in the Periyar river basin in the south-central part of the state. And another two of 315 MW capacity (about 21 per cent of the total) are in the southern region. The northern Kerala is served by one and only one hydro power station (Kuttiyadi) of only 75 MW installed capacity, that too with seasonal operation only. Though a diesel power plant of 128 MW was commissioned by November 1999 in Kozhikode, it is often run below 30 per cent of its capacity, owing mainly to the high operating cost. Thus the demand in this region is met largely by the power drawn from the southern region through long transmission lines and the imported power from the neighbouring States. The transmission of power to this region is mostly through 110 kV and 66 kV lines with the consequence of huge line losses resulting in poor voltage and creating technical problems in the inter-grid operation often ending up in system instability. At present a 220 kV transmission lines system from Idukki power station through Thrissur substation to Kasaragod in the north is under construction; but one is forced to wonder why Kerala cannot dream of transmission lines of higher capacity, when the order of the day even in other parts of India is of capacity at least of 400 kV.

That the deficiency in the transmission capacity accumulated out of neglect or faulty system planning and execution has contributed to the higher loss factor remains a foregone conclusion, yet the factor of unaccounted losses can by no means be ignored. Various steps have been initiated on all-India level since start of the VII plan to reduce T&D losses. As a significant part of it, theft of energy has been made a cognisable offence under the amended provisions of section 39 of the Indian Electricity Act, 1910, which provides for stringent punishment of all offenders. In 1996-97, 1813 theft or misuse cases were reported from Kerala, with a loss of 3.34 MU of energy, worth, on an average, about Rs. 32 lakhs. It goes without saying that the actual loss can be very much higher.

12

The Department of Power of the Government of India introduced in 1989 an incentive scheme for the reduction of T&D losses, and even awards are being given out. It is widely reported that instead of providing technical remedies to such high T&D losses, the State Electricity Boards often carry out greater manipulations to win these awards. The system loss is estimated from the difference between the available energy and the total sales as a percentage of available energy. This depends heavily on the total sale of energy, including that to agricultural sector, which is not metered in most of the States in India. As explained elsewhere, this provides the SEBs a convenient ‘dumping ground’ to throw in a good part of the T&D loss in an effort to whitewash their performance. Agriculture, however, accounts for only 4.4 per cent of total electricity consumption in Kerala, which is also metered, and hence cannot offer an easy leverage for large scale manipulation. It is unofficially reported that around 9 lakhs connection meters in Kerala power system remain defective or inoperative at all. Added to this is theft of energy in co-operation with the Board officials.

As already explained, T & D losses in Kerala was in a satisfactorily comparable position till some two decades back, the losses having been less than 15 per cent. However, it increased to a substantial extent in the following years, averaging about 24 per cent during 1982-83 to 1996-97, with a reported peak at 29 per cent in 1987-88. In the recent years, the loss is reported to be declining to 17 per cent. It should be stressed here that even these high figures are only underestimates, put out by the SEBs in their eagerness to record reduced transit losses, as mentioned above. Since a modicum of metering is done only at the consumption end (that too barring agricultural consumption in most of the states), the SEBs find it convenient to put up a comfortable loss rate and dump the remaining part of the unaccountable-for energy in the manipulated figures of generation and auxiliary and agricultural consumption. A detailed investigation, conducted by the Integrated Rural Technology Centre (IRTC), into the distribution losses in a typical rural electrical major section, viz., Kongad of Palakkad distribution circle in Kerala, has found the losses to be of the order of 35 per cent from the point of transformers, about 85 per cent of which could be accounted for by technical losses, with 30 per cent of the energy meters being faulty (quoted in CDS 2001: 57-58). Along with

13

such high technical losses thrive losses through theft and unauthorised drawal; in 199798, the anti-theft squad of the KSEB detected cases of theft of energy worth Rs. 1.21 crores, and in 1998-99, worth Rs. 1.04 crores (Malayala Manorama daily 23 August 2000). Again during March 7 – 23, 2002, the vigilance squad of the KSEB conducted raids in the premises of the major consumers and detected energy theft to the tune of Rs. 1.33 crores across the state. It is now estimated that energy og 180 crore units are stolen every year in the state, worth Rs. 400 crores. It was also reported that a study by the Energy Management Centre, Thiruvananthapuram, found that energy loss in the state is 35 per cent, half of which is theft (Mathrubhoomi daily, April 1, 2002).

4. Financial Costs

Energy lost is money lost. In a more rigorously formal sense, energy loss involves not only direct costs, but indirect costs also associated with the inconveniences and other problems of (under) unsatisfied demand. Energy loss in excess of the usual unavoidable system consumption is an utter waste of scarce resources, immediately calling for corrective measures. This excess loss, as we have already noted, may be due either to technical causes or to non-technical ones. Enough light has already been shed on the tremendous effect of the technical causes in terms of the deficient transmission capacity in relation to the increasing load leading to heavy energy losses in the power system in Kerala. The non-technical, unaccounted losses due to theft, defective metering, etc., reasonably presumed to be unavoidably prevalent in any developing countries can by no means be neglected.

The loss of energy in Kerala as a percentage of the system available energy (generation less auxiliary consumption plus import) had, till the turn of the Eighties, been around 15 per cent, the maximum reasonable level of system consumption. In 1982-83 the loss percentage abruptly shot up to 22.8 per cent over the previous year’s 15.5 per cent, and the upward trend continued till 1987-88 to reach the phenomenal peak of 28.95

14

per cent. It then started to fall to come down to 20.05 per cent in 1995-96, and then to 17.75 per cent in 1998-99.

The Costly, Avoidable, Import With the commissioning of the Sabarigiri project, the Kerala grid has adequate generating capacity for meeting all the internal demands. Due to the delay in completion of the transmission lines, however, full utilisation of the potential created is held up. ….The purchase of energy after the commissioning of Sabarigiri generators in June 1966 was certainly an avoidable transaction. The Chief Engineer (Electrical) informs me that, if there were enough load connected to our system, the additional energy that could have been generated during 196667 was about 746 MU. Thus when the Kerala grid had the capacity to generate more than the quantity of energy required in Kerala, we have had to spend a large amount on the purchase of energy from the neighbouring States. [In 1966-67, energy purchase was 77.99 MU, worth Rs. 56.88 lakhs.] This situation has been brought about by lack of co-ordinated approach towards completion of the T&D systems side by side with the commissioning of projects. -Government of Kerala 1967: 52-53. If the Kerala’s power system could have somehow managed to maintain the 15 per cent system consumption of energy in transit, it would have resulted in a higher level of savings in energy and revenue. In 1982-83, if the system loss were 15 per cent only, with an energy level available for sale of 4545 MU, it would take away only 682 MU as losses against the actual loss of 1037 MU; this could then lead to a potential saving in energy to the tone of 355 MU and in revenue of Rs. 9 crores at the rate of 25.36 Paise per unit in 1982-83. Similar calculations for the following 16 years up to 1998-99 yields a cumulative possible saving in energy of the order of 8000 MU and in revenue of Rs. 525 crores Table 1). In other words, the inefficiency of the Kerala’s power system primarily in terms of deficient transmission capacity and inefficient operation has been eating away about 470 MU of energy worth Rs. 31 crores (at current prices) every year during the last 17 years. The energy thus lost in excess of the notional 15 per cent in fact represents a capacity of about 90 MW at 60 per cent load factor. It simply means that if the system were efficient enough to maintain the energy loss at its maximum reasonable level of 15

15

per cent per annum, it could help the system dispense with the need for adding about 90 MW to the total installed capacity, saving highly in investment and working expenses. That this saving was in addition to the potential increase in revenue by 31 crores per year highlights the immensity of the problem of the transmission loss in Kerala.

Now, assuming that the entire energy loss is technical, we can have an estimate of the investment required to bring the loss factor down to 15 percentage, and compare it with the possible savings from the loss reduction.

The system energy loss percentage dropped from its all-time high of 28.95 per cent in 1987-88 to 17.87 per cent in 1997-98, thus effecting a loss reduction of 1.01 percentage point per year. The fixed assets in the State transmission sector had about 4.5 times increase during the period from 1987-88 to 1997-98 from Rs. 90.9 crores to Rs. 495.64 crores, with an annual investment to the tone of Rs. 36.8 crores. Assuming a direct cause-effect function between the two, it appears that one percentage point reduction in T&D loss required about Rs. 36.5 crores worth of investment in the transmission sector, which lies evidently on the high side, as the major chunks of this investment, as we have already seen, occurred only during the 90s, the effect of which is yet to be felt in reducing the loss percentage. Calculations at this rate, however, shows that 13.95 percentage point reduction in energy loss, from 28.95 per cent to the reasonable 15 per cent level, entails an investment of about Rs. 510 crores. And we know the maximum possible cumulative savings in sales revenue that could have been achieved, had the ideal 15 per cent loss level been maintained over the years, amount to Rs.525 crores. The savings in capacity costs also should be added to this. The capital cost of a 120 mw hydro-electric power plant is reported to be equal to Rs. 180 crores, or Rs. 15000 per kW (CMIE 1996). At this rate, the savings in capital cost (alone, i.e., regardless of the O&M costs and working capital) for a hydro-electric plant of installed capacity of 100 MW (accounting for about 10 per cent reserve margin over the 90 MW ‘averted’ capacity) turn out to be Rs. 150 crores. Thus the total savings effected amount to Rs. 675 crores. Comparing this with the investment that would have been required to keep the loss factor at 15 per cent yields a savings-investment ratio of 1.32 : 1, i.e., 32 per

16

cent net saving over the capital cost. Remember all these are at current prices; however, the result would not be much different in qualitative terms with the present-valued calculations. Of course, our assumption that the loss factor, as reported by the KSEB, is entirely technical which amounts to winking at the substantial loss due to defective metering and theft, involves an over-estimation of the investment required, and hence an under-estimation of the level of savings possible. That the actual savings could be much more than the rough estimate obtained above, were the T & D sector in Kerala run efficiently, is thus a foregone conclusion.

Table 1 Savings Realisable from T and D Loss Reduction Loss Factor

Available

Total Sales (MU)

Energy (MU) with 15%

Actual

Loss 1982-83

0.228

4544.6

3862.91

3507.90

Savings Realisable Energy

Revenue

(MU)

(Rs. Crores)

355.01

9.00

17

1983-84

0.258

3766.25

3201.313

2793.00

408.31

14.36

1984-85

0.25

4938.9

4198.065

3705.00

493.07

14.76

1985-86

0.249

5558.82

4724.997

4172.00

553.00

17.02

1986-87

0.274

5126.59

4357.602

3716.00

641.60

30.96

1987-88

0.2895

5077.74

4316.079

3626.60

689.48

38.38

1988-89

0.2402

5748.8

4886.48

4387.00

499.48

28.32

1989-90

0.22

6208.75

5277.438

4859.07

418.37

22.35

1990-91

0.2157

6768.83

5753.506

5332.00

421.51

22.34

1991-92

0.2177

7155.8

6082.43

5596.00

486.43

29.19

1992-93

0.21

7406.54

6295.559

5838.50

457.06

33.80

1993-94

0.201

7825.17

6651.395

6234.16

417.23

34.24

1994-95

0.2008

8794.35

7475.198

7027.69

447.51

38.79

1995-96

0.2005

9278.2

7886.47

7414.62

471.85

43.84

1996-97

0.1996

8776.28

7459.838

7020.77

439.07

41.98

1997-98

0.1787

9395.1

7985.835

7715.49

270.34

33.68

1998-99

0.1775

11167.18

9492.103

8959.61

532.49

71.63

8001.81

524.64

Total

18

Generation, Loss and Import of Electricity in Kerala 8000

7000

6000

Million Units

5000 Generation import Loss

4000

3000

2000

1000

0

Generation import Loss

1957- 1960- 1963- 1965- 1968- 1971- 1974- 1977- 1979- 1982- 1985- 1988- 1991- 1993- 199658 61 64 66 69 72 75 78 80 83 86 89 92 94 97 441

591

763

33.5

56.8

79

841.9 1623 2293 2659 116

19

205

10

4471 5119 105

50.9

4488 5357 4548 83

227.7 1227

5326 5822

5503

1856 2036

3298

101.07 115.96 165.87 193.50 261.08 409.12 433.94 518.47 816.39 1036.1 1384.1 1380.8 1557.8 1572.8 1751.7 19

T and D Loss Percentage in Kerala Power System 35

30

Percentage

25

20

15

10

5

0

1957- 1959- 1961- 1964- 1966- 1968- 1970- 1972- 1975- 1977- 1979- 1981- 1984- 1986- 1988- 1990- 1992- 1995- 199758 60 62 65 67 69 71 73 76 78 80 82 85 87 89 91 93 96 98

Loss (%) 21.3

18.6

14.4

21.8

20.1

15.9

12.8

13.1

16.1

11.4

15.9

15.5

25

27.4 24.02 21.57

21

20.05 17.87 20

CHAPTER 7

PRICING OF ELECTRICITY -THEORY AND PRACTICE AND SOME EXERCISES FOR THE KERALA SYSTEM

1. INTRODUCTION

For building a rational tariff structure for an electric utility, close attention would need to be given to cost accounts. There is more to cost accountancy than the mere provision of data upon which tariff can be logically based. A good costing system will enable the policy makers to detect unfavourable trends and take corrective actions. Also there is more to tariff making than the mere means of recovering, on a reasonably fair basis, sufficient revenue from the consumers to find the cost of operation of the utility.

By a judicious mixture of incentives and deterrents, a good tariff can also provide a tool in the hands of the management, not only for encouraging the growth of the enterprise and thereby reaping the benefit of `economies of scale', but also for steering the growth towards set objectives. A well-constructed tariff can discourage waste and foster the efficient use of national resources.

When forming rates, it is often difficult to be strictly scientific and to operate on commercial lines. Frequently it may be necessary to waive scientific arguments in favour of over-riding practical considerations. To wit, it may often be necessary to fix rates according to what a market can bear rather than what it should bear.

Electricity charges serve two purposes:

i)

to raise revenue to meet the cost of operating the utility, and

1

ii)

to serve as a tool in implementing policy.

Broadly there are three lines of revenue policy which may be pursued:

i)

A tariff aimed at raising (on an average) just sufficient revenues to enable the utility to pay its way. This might involve temporary surplus to achieve a precise balance every year, which would call for some form of equalization fund.

ii)

Subsidization would be involved when a segment of the public to be served is believed to be unable to pay enough for the services offered and where those are regarded as sufficiently important, either as means of achieving indirect economic benefit. Then the tariff may be so formed as to raise revenues to meet a part of the cost of operations, the balance being met by a subsidy. In any case, a form of cross-subsidization is generally unavoidable even in a reasonably scientific structure.

iii)

A tariff may be so framed as to raise more revenue than is required to meet the cost of operating the utility, thus producing a profit for purposes of providing a reasonable return to the investor and for funding social services.

Abstracting from other (life-line) policy considerations, we concentrate here only on the cost-revenue aspects of pricing. If total costs are to be covered by totasl revenue realisable, then, a linear pricing principle in a single product case requires that price equal average cost (AC). Though accounting practices in general favour and follow AC pricing (in its commonly accepted cost-plus version), it is less attractive 4conomically, as it involves a welfare (dead-weight) loss in the context of a natural monopoly of decreasing costs. An economic pricing procedure, on the other hand, proposes that price equal marginal cost (MC), as it efficiently allocates resources. Below we discuss these two procedures in the case of electricity supply.

2. Cost-Plus or Full-Cost or Mark-Up Pricing

2

A cost-oriented pricing procedure widely used in manufacturing and retail trades, this approach is to estimate the full cost per unit of the product or service and then add to it a profit margin (mark-up over full cost) to arrive at the selling price. This profit margin may be fixed simply by a rule of thumb at what the firms individually consider a ‘fair’ or ‘just’ percentage. An important assumption in the full cost pricing is that both variable and fixed components of cost at unit output level are known. Sometimes it may not be very much easy to allocate fixed cost on product basis which may render the procedure of price fixation inaccurate. This difficulty can be removed if prices are computed on the basis of total cost and not unit cost of the product. That is, first the total cost of production is estimated to which the profit margin is added; this gives the expected revenue, dividing which by the volume of output gives the unit price.

The full cost pricing procedure is easy to follow. Moreover, this approach ensures full cost recovery, i.e., not selling at a loss, if sales are correctly estimated. These advantages make this pricing procedure quite popular in practice. However, it is fraught with a number of serious limitations as a result of which it is taken as just a crude or ad hoc pricing method. One important limitation of this procedure is that it is based on a concept of cost that is frequently not relevant for the pricing decision. It is argued that the incremental or variable cost rather than the full cost of a product should be used for fixing its price. This argument has led to ‘variable or incremental cost pricing’ by modifying the full cost pricing method. Here the variable cost of production is taken as the basis for price fixation. The average variable cost of production includes direct material and labor costs and all other expenses which vary with the level of output such as fuel and power costs, additional maintenance and supervision costs, etc. The aggregate of all such variable costs sets the minimum price of the product. Anything above this is a contribution to fixed cost plus profit. This method avoids some of the errors in full costing arising from arbitrary allocations of overheads and widens the range of prices acceptable as profitable. The contribution margin over the variable cost would be obviously much higher than the profit margin over the full cost because it has to cover both fixed cost and the desired profit margin. Thus fixed costs are excluded from the

3

basis on which the price depends, but are taken into account through the contribution margin.

Pricing in Regulated Utilities

There are two types of regulated utility pricing in practice now: fixed price and cost-plus mechanisms.

Under a fixed price regulatory mechanism, the regulator defines and devises ex ante a set of prices (or a weighted average of prices for different services) for the services of a regulated utility. The mechanism also involves automatic price adjustment formulas tied to exogenous variables such as general inflation, possible productivity growth or fuel price rise. A critical feature of this pricing is that the prices so defined are not directly linked to the regulated firm’s realized costs or profits.

Under cost-plus mechanism or rate of return (RoR) regulation, prices are set in public hearings based on an assessment of the capital and operating costs required to generate and supply electricity, by setting a rate of return on capital employed for the industry. The required revenue for as firm's targeted rate of return in a particular period from projected (or alternatively from its historical costs) is given by

R = OE + D + T + (RB* RoR), where R = revenue required, OE = operating expenses, D = depreciation expense, T = tax expense, RB = rate base, and RoR = rate of return.

The undesirable tendency to invest large scale capital (‘gold plating’) has been attributed to this form of regulation. Averch and Johnson (1962) argued that when the Regulator sets the allowed RoR above the capital cost, the utility will tend to use more capital than if it were unregulated and choose an inefficiently high capital-labour ratio for its level of output. In a framework of the principal-agent theory, RoR regulation is hypothesised to cause a managerial slack (x-inefficiency), possibly owing to the absence

4

of competition. These deficiencies have led to the proposals of a number of incentivebased regulation methods such as price cap, revenue cap, sliding scale, partial cost adjustment, yardstick competition, targeted incentive and some hybrid methods. In the RoR framework, the US has later on developed a more sophisticated approach involving not only the assessment of capital required, but also determination of a fair RoR on this rate base in an inter-industry comparison and allowing for cost components such as fuel cost to be passed through.

The base prices once set are not adjusted (or only partially adjusted) automatically for changes in costs over time, that is, prices remain fixed until they are reviewed by the Regulator again. The regulatory lag (the period between regulatory reviews), often last for several years. This regulatory lag effectively turns cost-plus regulation also into a fixed price regulation system with a cost-based ratchet adjustment every few years.

Price cap regulatory system that is in vogue in many privatized infrastructure sectors belongs to the fixed price regulatory regime. Price capping, recommended for British Telecom by Stephen Littlechild (1983) (who later became British Electricity Regulator), sets prices indexed on the rate of inflation, usually measured by the retail price index (RPI), which are then reduced by the possible productivity increases (X), (due to internal efficiency or technological improvement), as assessed by the Regulator and then increased by allowances for fuel cost rises (Y), if any. Thus for each period, the price ceiling Pt of a firm is estimated based on the previous period's price ceiling Pt-1, adjusted by RPI minus the effiicency factor X and by a correction factor Y; that is Pt = Pt-1 (1 + RPI - X)  Y.

In the absence of fuel cost rises, prices cannot be increased by more than the RPI less X, the Regulator’s assessment of possible productivity, i.e., (RPI – X) per cent. That is, in real terms, prices come down by X per cent. There is an implicit incentive for efficiency in this price cap in that if the utility can outperform X (that is, increase the actual productivity more than the Regulator’s assessment), it can increase its profits also

5

thanks to the regulatory lag. The English electricity utility and the electric utilities in other countries that have adopted the English competition model (for example, Argentina) follow this method.

The fixed price feature of the price cap system, however, means that profits from productivity improvement reaped by the firm during a certain period (during the ‘fixed price’ contract) will not result in automatic downward adjustments to permitted prices. Hence the value of ‘X’ is periodically renegotiated and then this may deter cost-reducing innovations especially towards the end of a regulatory period even if opportunities are open. In general, if the firm has to decide on an investment level prior to the regulator fixing a price, under-investment would be the predicted outcome (Vickers and Yarrow 1988: 90). The firm manages to operate with an inefficiently low capital stock in the hope of the Regulator setting a welfare-maximizing MC price, given the (short run) capacity of the firm. This result, however, would not follow, if the firm expected the Regulator to set a price reflecting long run marginal costs (LRMC). But such a regulatory mechanism remains unviable as it requires the Regulator to have costless access to unbounded information on cost and demand conditions.

3. Marginal Cost Pricing

All the above rates are based on cost accounting principles - recovery of the financial costs involved in supplying electricity at the end-use level. These private financial costs incurred by the utility are quite inadequate as indicators of economic costs that should be at the basis of any price-structure. Here lies the significance of the marginal cost (MC) pricing, which is generally recommended as an appropriate policy for public enterprise and regulated industries on the grounds that it is the pricing policy which maximizes social welfare. MC is the increase in cost resulting from a small increase in the rate of output of a good or service; and as such MC pricing in electricity means a tariff structure which is such that the cost to any consumer of changing the level or pattern of his consumption equals the cost to the electric utility of his doing so. MC pricing is an optimal pricing structure in that it maximizes the social welfare; the MC of a

6

good shows the value of the resources absorbed in producing the marginal unit of output, and hence the value of the other goods and services which the economy could have produced with those resources. Thus MC pricing implies economic costs to the society as compared with the private financial costs reflected in the accounting cost pricing. At the same time MC pricing is compatible with the historical cost accounting.

In what follows we propose a method of structuring MC pricing applicable to an electric utility in a typical developing country (Pillai 1991).

Three broad categories of MCs may be identified at the outset: capacity costs, energy costs, and consumer costs. Marginal capacity costs are basically the costs of investment in generation, transmission and distribution facilities to supply additional kilowatts. Marginal energy costs are the fuel and operating costs required to provide additional units (kilowatt-hours) from a thermal plant, and in a hydro-electric system, a part of the investment cost associated with storage may be related to energy, in addition to the operation and maintenance (O & M) costs. Marginal customer costs are the incremental costs directly attributable to customers, including costs of hook-up, metering and billing.

The long run MC (LRMC) of capacity refers to the change in system capacity costs C associated with a sustained increment D in the long run peak demand. Thus the LRMC of generation would be C / D, where the increment of demand D is marginal both of time and of megawatts. Of the absence of sophisticated computerized simulation models of capacity expansion, a common accessible method is to have recourse to a static interpretation of LRMC. By this method, the required LRMC of C

generating capacity (LRMCG ) may be approximated by the cost of expanding the peaking capacity by 1 kilowatt (kw). This may be necessitated by the cost of a kilowatt installed capacity, annuitised over the expected lifetime, and adjusted for the percentage reserve margin (PRM) and appropriate percentage loss typically caused by station use, that is, auxiliary consumption as a percentage of generation (PAUX). Thus,

7

C

LRMCG = (Annuitised cost per kw)  (1 + PRM) / (1 – PAUX).

The annuitised value of a lump-sum investment is defined as the fixed annual payment over the lifetime of the investment, whose present value at the given discount rate is exactly equal to the original expenditure. Annuitising the investment is conceptually equivalent to expanding the capital outlay by one year.

Next, the LRMC of transmission and distribution (T & D) is calculated. Generally, all costs of investment of T & D – except customer costs – are allocated to incremental capacity, because the designs of these facilities are determined principally by the peak kilowatts that they carry rather than the kilowatt-hours (kwh). Particularly at the distribution level, however, the size of a given feeder may depend on the local demand peak, which may not occur within or coincide with the system peak period. At this stage structuring of the MC by different voltage level such as extra high, high, medium and low (EHV, HV, MV, and LV) is introduced. Since customers at each voltage level are charged only upstream costs, capacity costs at each voltage level must be identified.

In order to estimate the LRMC of T&D, the simplest method of the average incremental cost (AIC) is used. If in year t MWt and It are the increase of demand served (relative to the previous year) and the investment cost respectively, then,

T

AIC 

I L T



t

( 1  r )t

t 0

 MW

t

( 1  r )t

t L

where r is the discount rate (e.g., the opportunity cost of capital), T is the planning horizon and L is the average time delay between the investment and commissioning dates for new facilities. An alternative method of determining marginal T&D costs would be to use historical data to fit regression equations (e.g., Transmission Costs =  +  Peak Demand).

However, such past relations may not hold true of the future also as the system

8

expands and the estimate may not be dependable, especially more so of an underdeveloped system like ours.

Once the AIC of EHV and HV transmission has been computed and annuitised over the lifetime of the plant that yields the LRMCHV, then the price to be charged at the C

HV level (PHV) would be equal to the total LRMC of capacity at that level, LRMCHV , given by C

C

PHV (= LRMCHV ) = LRMCG / (1 – LHV) + LRMCHV ,

where LHV is the percentage of incoming peak power that is lost of the EHV and HV network.

This procedure may be repeated at the MV and LV levels. Thus the price to the MV consumers (PMV) is C

PMV = (LRMCMV ) = PHV / (1 – LMV) + LRMCMV,

where LRMCMV is the marginal T&D costs at the MV level, for example, the AIC of distribution substations and primary feeders, and LMV is the percentage of incoming peak power lost at the MV level.

Coming to the second cost category, viz., marginal energy cost, this refers to the running costs of the marginal plant in the merit order. To be more specific, this would be the fuel and operating costs of the additional plant in the capacity expansion plan to meet the demand increment D. These costs have to be adjusted further by the appropriate loss factors at each voltage level the same way as marginal capacity costs.

Recurrent customer costs, the third cost category, stem from meter reading, billing, administrative and other expenses. Though a part of the investment and operating

9

costs of the distribution system, these may be imposed as a flat charge on a repeated basis, of addition to the usual kw and kwh charges.

The above cost considerations and the consequent pricing structure have been under conditions of the system peak, the capacity expansion having been necessitated by the peak demand pressing against the limited capacity. The prices thus derived are then applicable only to the peak load customers, those responsible for the additional capacity costs. It follows, therefore, that the off-peak consumers (or the customers during the offpeak load period) should be charged only the running costs of the plant in operation during that period.

Controlling the system peak is an important objective of any rational tariff structure. In view of the significance of the time of load incidence, it is a good policy for a utility to offer incentives to consumers for taking energy at times other than at the system peak so as to improve the system load factor, and thus to have the more effective use of the generating plant and EHV lines. A two part tariff with an `entry fee’ or capacity (or demand) charge and an energy rate would intrinsically encourage the improvement of a consumer’s load factor, but if his maximum demand does not coincide with the system peak load, this individual improvement will not necessarily be reflected in improved system load factor.

The introduction of peak load pricing, as discussed above, involves installation, with the consent of the consumer, time switches or equivalent devices which will either disconnect the consumers concerned during peak load hours if the consumption exceeds the allowed limit, or preferably, cause the off-peak period consumption to be registered by a separate set of meters from those which record the usual (off-peak) consumption. The off-peak energy would then be charged at a lower rate than the on-peak energy; and this lower rate is expected to induce the consumer to alter the pattern of his consumption towards improved load factor and reduced consumption during the peak period.

10

Where such metering is not possible, the peak and off-peak rates may be averaged out and the MC pricing may still be applied in a two- (or multi-) part tariff framework without regard to the demand variability.

4. Power Tariff Policy in India

Having outlined the broad principle governing tariff policy, it is necessary to highlight the regulatory provisions under the Electricity (Supply) Act, 1948. Section 59 of this Act lays down the general principles of the Board's finances. The priorities of liabilities, if in any year the revenue receipts are not adequate, are laid down in Section 67 of the Act. By the Eighth Schedule to the Electricity (S) Act, principles have been laid down for determination of costs of production of electricity at generating stations and the Ninth Schedule delineates the principles governing allocation of costs of production at generating stations.

A number of Committees appointed by the Government of India have gone into the working of the SEBs resulting in norms being laid down by the Venkataraman Committee, the World Bank and subsequently, the Rajadhyaksha Committee.

The E(S) Amendment Act, 1983, has amended the provisions of Section 59 of the Act. The amendments have the effect of stipulating that the Board shall, after taking credit for any subvention from the State Government under Section 63, carry on its operations under this Act and adjust its tariffs so as to ensure that the total revenues in any year of account shall, after meeting all expenses properly chargeable to revenues, including operating, maintenance and management expenses, taxes (if any) on income and profits, depreciation and interest payable on all debentures, bonds and loans, leave such surplus as is not less than 3 %, or such higher percentages as the State Government may, by notification in the Official Gazette, specify in this behalf, of the value of fixed assets of the Board in service at the beginning of such year (the italicised portion relates to the new amendments). The Amending Act has also defined by an Explanation that for the purpose of this subsection "value of the fixed assets of the Board in service at the

11

beginning of the year" means the original cost of such fixed assets as reduced by the aggregate of the cumulative depreciation in respect of such assets calculated in accordance with the provisions of this Act and consumers' contributions for service lines. With the basic objective having been set, viz., to earn a surplus of not less than 3 % after meeting all operating expenditure, the goal of tariff making has been predetermined. Nevertheless, to achieve this set goal, a tariff mechanism which is in line with the basic tenets of tariff-making would still need to be properly evolved. The Venkataraman Committee that recommended this 3 % return also noted that this would amount to a return of 11 % actually, taking into account the electricity tax / duty levied by State Governments, i.e., interest 6 %, net return 3 %, general reserve 0.5 % and electricity duty 1.5 %. It goes without saying that these rates, laid down some 35 years back, require revisions.

In spite of the clear provisions of the E(S) Act as above, no State Government has issued any directives to the concerned SEB, though all the SEBs have taken this 3 % net return as a sacrosanct obsession into their heads and been acting as if it were the sole tariff objective, which is certainly not of keeping with the letter and spirit of the Act and its policy directives on tariff formulation. Moreover, a 3 % net return as at present is scarcely enough to generate resources required even for debt servicing of institutional loans, let alone the requirements for expansion.

5. Power tariff for the KSEB Tariff Policy and Financial Viability of the KSEB

During 1996-97, the average electricity tariff of KSEB was 92 Paise per unit. The tariffs have since been raised with effect from February, 1997, with the average for 199798 being about 126 Paise per unit. It has also been approved to raise the tariff by 10 % annually from February every year, though the domestic sector has so far been spared from its actual practice.

12

Having a predominantly hydro-power system with much lower generation costs, KSEB has enjoyed a singularly advantageous position of the country. The unit cost of power supply was only 144.31 Paise in Kerala in 1997-98 against an all-India average of 217.9 Paise per unit. Yet another distinct advantage favouring KSEB is the much lower energy sale to agricultural pump-sets (about 3.8 % of total sales, against 29 to 45 % in other States), the subsidised sector that yields very low tariff per unit. Expectedly, this should have resulted in a better average rate of realisation per unit sold. However, KSEB’s revenue per unit sold is by and large the lowest of the country (the only exception being Jammu and Kashmir); for example, in 1996-97, the average tariff for sale of electricity (including supply to agricultural pump-sets) in Kerala was 103.2 Paise per unit, while the all-India average was 157.7 Paise per unit. Excluding the supply to agriculture, the difference will become much larger. Such low average tariff appears to be quite of line with the advantageous position of KSEB having a low unit supply cost thanks to very favourable hydro-thermal mix of energy and good monsoons. However, the problem lies in the unfavourable and unviable financial obligations of having to cope with an average tariff much lower than a unit cost of supply. And this has much to do with the financial viability of KSEB.

Though KSEB has been earning about 3 % rate of return on its capital base since 1993 as per Section 59 of E(S) Act, its accumulated reserves have always been negative as shown in Table 1. Even with the required 3 % return achieved, the internal resources generated during 1993-97 have been totally inadequate to meet the debt redemption obligations even in respect of non-Government institutional loans

(Table 2). The

resources actually generated might in fact be less than the reported figures owing to a number of factors such as revenue out-standings, diversion towards working capital, etc. Even if we ignore these factors, the reported internal resources remain inadequate even to meet debt redemption obligations to institutional lenders. If repayments have indeed been made as shown of the Table, the only way was to borrow more. Surely, KSEB is in the clutches of a debt-trap, losing its credit worthiness. The future also appears gloomy; it has been projected that even if tariffs are revised regularly to achieve 3 % surplus, the

13

internal resource generation would be enough only to meet 20 to 60 % of the institutional loans liability.

The position regarding the Government loans and interest thereon is still worse. Repayable loans are advanced by the State Government to the Board, currently at an interest of 17 % per annum. The State Government loans are never repaid nor the interest. As on 31.03.1997, the board had the following out-standings: State Government loans: Rs. 1130 crores., Interest thereon: Rs. 582 crores. And loans from institutional lenders: Rs. 1397 crores. Neither the Board is in a healthy position to repay the loans and the interest thereon to the Government, nor the Government is willing to liquidate them. Consequently, these huge amounts get reflected of the Board’s balance sheets as its outstanding liabilities and seriously affect its credit worthiness. This financial crisis stands as a serious deterrent to institutional / international lenders. If at all they come forward, the loans package they advance is likely to involve conditions of grave consequences such as pressure for `power sector reforms’ in the form of unbundling, restructuring / privatisation of the Board, etc., leading to its total elimination, as it so occurred in some States (Orissa, Haryana and Andhra Pradesh) and in some other sectors (DPEP in Kerala). All these together make up an urgent and important case for making KSEB financially viable and vibrant.

It should be remembered that to achieve financial viability, periodical increase in tariff is just one among many policy measures to be adopted synchronously. The inefficiency of the management of the Board that contributes to a good extent to the malady should not be passed on to the consumers by way of increased tariff. Many Committees set up at the Central and State level, including the Rajadhyaksha Committee, Balanandhan Committee, etc., have time and again cautioned against such inefficient practices.

An Expert Committee constituted by the State Planning Board under the chairmanship of Sri. K. P. Rao, to review the tariff structure of KSEB, has recommended

14

a number of corrective and constructive measures in its Report submitted in May 1998. The suggestions include, inter alia,

a) Restructuring the capital of the Board, including waiver of accumulated interest and conversion of loans into equity of 1:1 proportion. b) Prescribing a mandatory return of not less than 16 % on equity funds / internal resources (reserves) of the Board deployed for the expansion programmes. c) Laying down of sound tariff policies. d) Formation of tariff Regulatory Commission as an independent body to prescribe, regulate and monitor tariffs against tariff objectives.

The Tariff Recommended by the K. P. Rao Expert Committee

As suggested by the Committee, KSEB has estimated the tariffs that would be needed on the basis of the above guidelines for tariff formulation recommended by the Committee, taking into account a plan size of Rs. 4380 crs. for 1997 – 2002 (Ninth Plan period). These rates are reproduced of Table 3 along with the tariff needed to generate a 3 % return as per present tariff practice, keeping the other parameters unchanged. It can be seen that the proposed tariff in 1997-98 is higher by 21 Ps. than the tariff under the existing practice to achieve 3 % RoR. However, the difference steadily declines over the years and in 2000-2001, the proposed rate becomes lower (by 2 Ps.) than the existing alternative and in the next year, again lower by 3 Ps. The Committee asserts that there need not therefore be any apprehension that the tariff policy recommended would lead to steep rise in tariffs only because of the new policy; on the other hand, on a longer time frame, the tariffs as per new policy suggested leads to slightly lower tariffs than what would be applicable under the existing policy.

The KSEB's projections of the operating parameters as under the Ninth Plan (1997-2002) in the two scenarios are given in Table 6. Unit cost of supply at the distribution end and the percentage mark-up implicit in the tariff rates in the two scenarios (as estimated by us) are also reported in this Table. Reading these data invokes

15

some confusion as to the methodology adopted by the KSEB in computing the proposed tariff. "The precise tariffs and the tariff schedules which are confidential in nature" (Report of the K. P. Rao Expert Committee, 7.6.5. p. 38) are not reproduced in the Report and hence we are left with an option of working it out all using the parameters given in the data. It should be carefully noted that though the Committee has asked the KSEB "to estimate the Tariff that would be needed on the basis of the above guidelines for tariff formulation" (Committee Report, 7.4.1. p. 32) i.e., by assuming "not less than 16 % as cost of equity funds for computing tariffs" (Committee Report, 7.3.1. p. 30), the estimated tariff rates yield different RoR on equity capital as computed by the KSEB itself and given in the Report. It ranges between 13.93 % in 1999-2000 to a mere 2.06 % in the next year. This single entry appears to belie the estimation procedure of the KSEB based on the recommendations. The RoRs on the net fixed assets (NFA) employed at the beginning of the year, implicit in the proposed tariff rates are estimated by us and given in the Table. It ranges between 8.03 % in 1998-99 and just 0.67 % in 2000-01. In the last year of the Ninth plan period (2001-02), the RoR on NFA is 2.95 % only. Thus the tariff rates proposed for the last two years appear to be out of line, more particularly, the one for 2000-01. The proposed tariff rate is lower by 2 Ps. than the unit cost of supply (estimated by us) in this year (2000-01) and is higher by just 7 Ps. in the next year. Thus the implicit mark-up (over total cost) of the proposed rates in 2000-01 becomes negative. The tariff rates with 16 % RoR on equity capital as estimated by us based on the parameters given for the 5-year period are also reported in the above Table. This is estimated as total expenses including interest charges plus 16 per cent of the State government equity capital divided by total sales units. That the proposed sales rate for 2000-01 is too low is very much in evidence now, with a negative implicit mark-up over total costs. However, it should be noted that the RoRs on equity after capitalisation of interest and expenses are also given in the Report, which are sufficiently high being 20, 25, 29 17 and 20 per cent respectively for the 5 years. The RoRs on NFA after capitalisation of interest and expenses as worked out by us are respectively 22.6, 21.1, 16.0, 6.7 and 7.4 per cent respectively. The fall in the rates in the fourth year is evident.

16

The confusion surrounding the computation procedure of the tariff of the KSEB appears naturally in the case of the rates estimated under the existing tariff practice also. Absence of equity in the capital base results in increased interest payment obligations to the State government in this context and hence the total expenses become higher than in the first scenario. Obviously, a cost-recovery based pricing system must yield higher tariffs in such situation than in the equity-involved low cost scenario. However, the estimated sales rates are lower for the first 3 years of the plan period under the high-cost existing tariff practice than under the low-cost proposed tariff practice. The rates are much lower than the unit cost of supply. Cost recovering average tariff required to obtain 0 per cent RoR (break-even pricing) and the stipulated 3 per cent RoR are estimated by us on a full cost pricing basis and reported in the Table. Note that the break-even prices are lower than the unit supply cost due to the presence of the miscellaneous receipts in the total revenue. The RoR on NFA after capitalisation of interest and expenses, given in the Report, is a flat 3 per cent for all the 5 years. One thing is certain now: the tariff practice being followed by the KSEB is not at all based on a full cost recovery principle; rather they might be following some variant of the variable cost pricing approach.

Mark-up Prices

The electricity sales rates based on full cost pricing policy, using the parameters under the two scenarios (under the Expert Committee’s tariff policy recommendation and under the existing tariff policy) are given in Table 7. The sales rates at 10 per cent markup under the recommendation scenario in the first 2 (or 3) years of the IX plan period are very much comparable with those recommended by the Committee. The rates at 10 per cent mark-up under the existing tariff policy are higher than these rates as the total costs are also higher. (Remember the recommended rates are higher than the rates under the existing pricing policy for the first 3 years of the period.) we do not have enough data on variable cost categories to try the incremental cost pricing approach; nevertheless, for illustration we take the risk of computing the rates based on variable cost pricing policy, considering the data given in the Report on expenses on fuel, power purchase, O & M and establishment expenditure, though the last includes fixed cost component also; i.e.,

17

we leave out from total costs those on depreciation and interest charges and other expenses. These rates at 60 per cent mark-up over the ‘variable’ costs we consider are comparable with the rates under the recommended policy for the first 2 (or 3) years, and the rates at 40 per cent mark-up are comparable with the first 2 (or 3) years’ rates under the existing tariff policy. As a matter of fact, the mark-ups over these ‘variable’ costs, implicit in the recommended tariff are 0.6, 0.6, 0.45, 0.25 and 0.22 respectively for the 5 years and those in the existing practice rates are 0.4, 0.4, 0.35, 0.28 and 0.22 respectively.

Marginal Cost Prices The detailed calculations of energy rates based on MC pricing method are given in Table 8. The increasing dependence on power purchased from the Central sector thermal stations makes the Kerala power system more and more a mixed, hydro-thermal system. Hence in calculating the MC of energy supplied, we need to consider a power system with both hydro and thermal plants – MC based price estimation usually proceeds by taking into account the financial and technical characteristics of a representative plant. The data on these characteristics of a hydro-power plant and of a coal-based thermal power plant to represent a mixed hydro-thermal system are taken from the Draft Report on Least Cost Planning Exercise for KSEB (International Energy Initiative, Bangalore, March, 1998), as reliable and representative data are unavailable from the Kerala power system especially thanks to the common inflationary problems of time and cost overruns of power projects.

The computation of MC per unit at generation end of a representative hydropower plant is given in detail in Tables 8(a) and 8(b). The capital cost of the 120 mw capacity plant is Rs. 180 crores. at 1996 prices. The construction period of a hydro-plant is usually taken as 10 years and in the present exercise we assume that the plant would be commissioned in 1999, and hence the construction work starts in 1990. The capital cost of Rs. 180 crores is distributed over this 10-year period in accordance with the percentage apportioning of the capital cost of the Lower Periyar hydro-power plant in Kerala over its 10-year construction period as given in its Project Report. Table 8(a) presents the

18

distribution of the capital outlay at 1996 prices at varying discount rates of 12 – 20 per cent. Thus the present-valued capital cost of this representative hydro-plant at the time of its commissioning in 1999 is Rs. 203.7 crores at 12 per cent discount rate and Rs. 224.4 crores at 20 per cent rate.

Table 8(b) proceeds to compute the MC per unit of energy of this plant at different discount rates. Working capital at 5.75 per cent of the present-valued capital cost is added to the latter to get the total capital cost which is then annualised for 25 years’ plant-life. The annualised capital cost is divided by the plant capacity to obtain the capital cost per kW capacity per year, which is further adjusted for 20 per cent reserve margin to account for demand exigencies. Annual O&M costs at 11.5 per cent of the original capital per kW are discounted as in 1999 (when the plant would be commissioned into operation) with base 1996 and added to the annualised capital cost per kW. Dividing this by the net kWh per kW of the plant capacity at a capacity factor of 68.5 per cent and after accounting for 5 per cent auxiliary consumption yields the MC of the hydro-plant capacity per unit energy. It ranges between 69.7 Ps. per unit and 101.7 Ps. per unit at 12 – 20 per cent discount rates. (The percentages of working capital and O&M costs, capacity factor, etc., are also taken from the Draft Report mentioned above.)

Similarly, the distribution of the capacity cost of a representative coal-based thermal power project, following the distribution of the capital cost over the construction period of 6 years of the Kayamkulam thermal power project Stage I, as given in its first Project Report, is presented in Table 8(c). The cost of this 350 mw capacity thermal plant is Rs. 1310 crores at 1996 prices. The plant is assumed to be commissioned in 1999, having started its construction work in 1994. The present-valued capacity cost ranges from Rs. 1250.4 crores to 1229.2 crores at the discount rates of 12 – 20 per cent. The annual costs include O&M costs (2.5 per cent of capacity costs) and fuel (coal and oil) costs [see Table 8(d)]. The annual costs per kw are present-valued as in 1999 (with base 1996) and added to the annualised total capital cost per kw (including working capital at 4.83 per cent). Dividing this total cost per kw per year by net kwh per kw at a capacity factor of 68.5 per cent after accounting for 10 per cent auxiliary consumption, we get the

19

MC of a representative coal-based thermal plant per unit of energy. It ranges between 166.6 Ps and 214.5 Ps per unit at 12 – 20 per cent discount rates.

Now these MCs per unit of the two representative hydro-thermal plants should be averaged, taking appropriate weights. It is projected by the KSEB (given in the K. P. Rao Expert Committee Report) that by 1999 the hydro-thermal mix in energy supply in Kerala would be in the proportion of 60 (hydro) and 40 (thermal). Based on these weights, the average MC of a unit energy at generation end of this representative hydrothermal system as in 1999 is estimated for different discount rates [Table 8(g)]. It ranges between 108.5 Ps per unit and 146.8 Ps per unit at 12 – 20 per cent discount rates.

To this generation cost must be added the transmission and distribution (T & D) cost. Constrained with data limitation here, we are forced to consider only a 5-year period distribution of capital cost in T & D during the IX plan (given in K. P. Rao Expert committee Report and in Annual Plan Estimates of KSEB). Computation of average incremental cost (AIC) of T & D needs capital cost vis-a-vis incremental demand during the period in question. Demand data at transmission as well as distribution end are obtained from the 15th Power survey. The incremental demand at T & D ends during the IX plan period is discounted at various rates and given in Table 8(e). Thus the presentvalued (with base 1996) incremental demand at transmission end ranges between 626.3 mw at 12 per cent and 514.5 mw at 20 per cent discount rates and at distribution end, between 582.6 mw and 478.4 mw respectively. The corresponding distribution of the discounted capital outlay in T & D during the IX plan period is given in Table 8(f). The discounted (with base 1996) capital cost in transmission ranges from Rs. 713.3 crores to Rs. 587 crores at 12 – 20 per cent discount rates and that in distribution from Rs. 356.6 crores to Rs. 293.5 crores respectively.

The AIC in transmission is obtained by dividing the total discounted capital cost in transmission by the total discounted incremental demand at transmission end [Table 8(g)]. This is about 130 Ps. per unit. Similarly, the AIC in distribution is estimated by dividing the total discounted capital cost in distribution by the total present-valued

20

incremental demand at distribution end, and this is about 70 Ps per unit. Now adding the three cost categories of generation, transmission and distribution, we get the MC of electricity per unit; this ranges from 308.4 Ps. to 347 Ps. per unit at 12 – 20 per cent discount rates.

It should be noted that the AICs of T&D are a little biased in that the T&D facilities should have been operational by 1999 along with the generation facility, rather than by 2002; i.e., the AICs should have been discounted as in 1999 with base 1996. Nevertheless, as the AIC is in the nature of an average increment per unit demand per year and our aim is just an exposition of the method of estimation rather than a proposal of tariff in practice, the bias may be neglected.

As already explained, the actual tariff differs from this MC per unit in that the former is arrived at after adjusting the latter for the transit loss of energy at various voltage level. This exercise, however, is not carried out here due to paucity of reliable data.

Despite its international acceptance in most of the developed and in some of the developing nations and a large number of World Bank sponsored studies and suggestions, there have been few takers of LRMC pricing in practice in India. For example, the K. P. Rao Expert Committee remarks on such a suggestion: ‘While this is a sound approach from an economist’s point of view, such an approach has not been adopted in any sector in the country and adoption of such a principle in isolation only in the Power sector could lead to problems. For example, others like Coal suppliers/ Railways could ask for a similar treatment. Besides, there can be some debate on the parameters to be adopted. Also the transparency of the computations is slightly less compared to costs / tariffs derived from real costs associated with the operations.’ (Report, 7.2.1. p. 29). However, the Committee in the next breath opines: ‘There is no reason why this approach should not be adopted, particularly for industrial HT consumers, industrial LT consumers as well as commercial consumers. Today the tariff as (on) 1.10. 96 for industrial HT in KSEB are about ½ to 1/3 of what is charged in neighbouring States and thus too low…. The existing

21

tariffs are also much less than what it costs to get additional power from any existing source or from IPPs. Even granting that sectors such as domestic and agriculture may need some concessions in tariff, it would be imperative that sectors such as industrial HT, industrial LT and commercial tariffs are fixed on the basis of LRMC.’ (Report, 7.4.2.6. p. 35).

In the U. K., France and Scandinavian countries electricity tariff making essentially follows MC pricing principles with incentives for peak period withdrawal. The modalities of MC pricing based tariff implementation in these countries are in general based on two-part tariff. This is being followed even by countries such as Hong Kong, Taiwan, Republic of Korea, Singapore, Malaysia, Thailand and Indonesia. U. S. regulatory bodies and utilities had generally hesitated for a long time to give up the conventional accounting cost approaches to rate making and to adopt LRMC pricing. The Public Utilities Regulatory Policies Act of 1978 was an important step forward in helping to rationalise electricity tariffs in the U. S. It may be interpreted in the spirit of the LRMC methodology.

The experience of countries such as New Zealand have shown that while electricity price is set on MC principles, it should also cover network connection costs and overheads of maintenance, an aspect already mentioned. Such pricing practices involve two schemes – in the short-run with spot pricing and contracts with customers, often expressed as MW-Km

rates and in the long-run announcing a tariff based on

expected future prices depending on the region to be supplied. Thus, in regions where reinforcement of network is imminent, a high price is charged, while in regions where plenty of spare capacity exists, a low price is recovered.

22

TABLE 1 COMMERCIAL LOSSES OF KSEB Year ending March,

Profit/Loss for the year Rs. Crores. -14.34 11.84 -21.69 -35.69 18.42 24.12 21.80 22.76

Accumulated reserves Rs. Crores. -76.75 -64.91 -86.60 -122.29 -103.87 -79.75 -57.95 -35.19

1989 1990 1991 1992 1993 1994 1995 1996 (Provisional) Source: Report of the Expert Committee to Review the Tariff structure of KSEB, May, 1998.

Rate of return on capital base % -4.64 3.26 -5.96 -9.86 3.00 3.67 3.03 3.00

TABLE 2 INTERNAL RESOURCES GENERATED AND LOANS REPAID 1. Depreciation 2. Surplus 3. Internal Resources 4. Loans Repaid i) To State Government ii) To Institutional Lenders 5. Total Repayment Source: As in Table 1

(Rs. in Crores) 1992-93 1993-94 1994-95 1995-96 1996-97 27.92 39.09 47.17 56.29 62.76 18.42 24.11 21.80 22.76 -17.53 46.34 63.20 68.97 79.05 45.23 0.35 45.33 45.68

0.38 48.72 49.10

16.21 75.81 92.02

0.32 96.05 96.37

0.08 139.57 139.65

23

TABLE 3

KSEB'S PROJECTION ON LOANS FROM STATE GOVT. UNDER THE EXISTING TARIFF POLICY

Year 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-2000 2000-2001 2001-2002

Opening Balance 538.27 622.58 690.05 767.88 848.85 1130.59 1430.71 1775.35 1862.85

Loan Received 84.69 83.68 81.05 81.05 281.74 300.12 344.64 87.5 296.89

(Rs. Crs.)

Loan Repaid 0.38 16.21 3.22 0.08

Closing Balance 622.58 690.05 767.88 848.85 1130.59 1430.71 1775.35 1862.85 2159.74

Source: As above.

TABLE 4

KSEB'S PROJECTION ON INTEREST DUE TO STATE GOVT. ON GOVT. LOANS (Rs. Crs.) Year Opening Interest Interest Closing Balance Due Paid Balance 1992-93 203.45 63.37 0.13 266.69 1993-94 266.69 58.9 0.06 325.53 1994-95 325.53 69.9 0.03 395.4 1995-96 395.4 87.52 0.01 482.91 1996-97 482.91 98.8 581.7 1997-98 581.7 109.82 691.52 1998-99 691.52 157.76 849.28 1999-2000 849.28 258.71 1107.99 2000-2001 1107.99 477 1584.99 2001-2002 1584.99 584.69 2169.67 Source: As above.

TABLE 5 KSEB's PROJECTION OF TARIFFS Year

Under the new policy Recommendations 1997-98 164.48 1998-99 188.33 1999-2000 227.71 2000-2001 248.42 2001-2002 303.05 Source: As above.

(Paise per unit) Under the existing practices 143.19 164.78 207.24 250.35 305.72

24

TABLE 6 PROJECTION OF OPERATING PARAMETERS OF KSEB UNDER THE NINTH PLAN (1997-2002) 1997-98 1. Installed Capacity (mw) Hydel Diesel Others (Wind, etc.) Total 2. Gross Generation (mu) Hydel Diesel Others (Wind, etc.) Total 3. Auxiliary Consumption (mu) 4. Net Generation (mu) 5. Power Purchased (mu) 6. Total Energy Available (mu) 7. T & D Losses (%) 8. Energy Sale (mu) 9. Expenses (Rs. Crs.) a) Fuel b) Power Purchase c) Operation & Maintenance d) Establishment Expenditure e) Depreciation f) Other Miscellaneous Expenses Total

1998-99 1999-2000 2000-2001 2001-2002

1757 100 2 1859

1757 100 2 1859

1837 228 2 2067

1951.25 228 2 2181.25

1951.25 228 2 2181.25

5967 25 3 5995 31.34 5963.67 3235 9198.67 18 7542.91

6551 450 3 7004 59.76 6944.25 3747 10691.3 20 8553.00

6849 867 3 7719 86.27 7632.74 5047 12679.74 20 10143.79

6849 1204 3 8056 106.49 7949.52 7515 15464.52 20 12371.61

6849 1204 3 8056 106.49 7949.52 10865 18814.52 20 15051.61

5.57 407.72 52.30 320.17 79.85 51.51 917.12

80.60 516.68 57.53 346.56 98.81 56.66 1156.84

162.41 907.57 63.28 435.95 143.72 62.33 1775.26

230.87 1726.01 69.61 407.61 199.76 68.56 2702.42

242.42 2994.76 76.57 443.05 226.91 75.42 4059.13

UNDER THE POLICY RECOMMENDATIONS OF THE EXPERT COMMITTEE (Rs. Crores.)

10. Average Sales Rate (Ps./unit) 11. Sales Revenue 12. Miscellaneous Receipts 13. Total Revenue 14. Interest Payable To Institutional Creditors To State Government Total 15. Total Expenses 16. Net (Retained) Surplus (+)/Deficit (-) 17. Value of Net Fixed Assets Employed at the Beginning of the Year 18. Rate of Return on N. F. A. (%) 19. State Government Equity Capital 20. RoR on Equity Capital (%)

164.68 1242.16 44.45 1286.61

188.33 1610.75 48.90 1659.65

227.71 2309.89 53.78 2363.67

248.42 3073.42 59.16 3132.58

303.05 4561.45 65.08 4626.53

211.89 8.50 220.39 1137.51 149.10

287.37 17.00 304.37 1461.20 198.45

342.86 25.50 368.36 2143.62 220.05

362.63 34.00 396.63 3099.05 33.53

357.61 42.50 400.11 4459.23 167.30

1872.52 7.963 1480.02 10.07

2470.24 8.034 1530.02 12.97

3593.12 6.124 1580.02 13.93

4993.92 0.671 1630.02 2.06

5672.77 2.949 1680.02 9.96

25

21. Net Surplus/Deficit after Captalization of Interest and Expenses 301.29 389.51 454.59 22. Return on Equity 0.20 0.25 0.29 23. Value of NFA 1334.29 1846.66 2840.48 24. RoR on NFA (%)* 22.6 21.1 16.00 25. Unit Cost of Supply (Ps./kwh) * 150.81 170.84 211.32 26. Implicit Mark-Up of Sales Rates (%) * 9.20 10.24 7.75

273.7 0.17 4088.48 6.70 250.5 -0.83

335.93 0.20 4517.66 7.40 296.26 2.29

27. a) Sales Rates with 16 % RoR on Equity Capital (Ps./kwh) * b) Implicit Mark-up of this Rate (%) *

271.58 8.416

314.12 6.028

250.35 3097.20 59.16 3156.36

305.72 4601.63 65.08 4666.71

423.94 264.08 688.02 3390.44 -234.08 -4.687

438.85 261.97 700.82 4759.95 -93.24 -1.644

122.65 3.00 274.05

135.53 3.00 316.24

269.27 281.38

311.92 323.23

182.20 20.818

199.46 16.754

236.25 11.793

UNDER THE EXISTING TARIFF PRACTICE

(Rs. Crs.)

28. Average Sales Rate (Ps./unit) 143.19 164.78 207.24 29. Sales Revenue 1080.05 1409.33 2102.20 30. Miscellaneous Receipts 44.45 48.90 53.78 31. Total Revenue 1124.50 1458.23 2155.98 32. Interest Payable To Institutional Creditors 211.89 301.71 376.43 To State Government 173.75 223.80 261.41 Total 385.64 525.51 637.84 33. Total Expenses 1302.76 1682.35 2413.10 34. Net (Retained) Surplus (+)/ Deficit(-) -178.26 -224.12 -257.12 35. Rate of Return on N. F. A. (%) -9.520 -9.073 -7.156 36. Net Surplus/Deficit after Capitalization of Interest and Expenses 40.03 55.40 85.21 37. RoR on NFA (%) 3.00 3.00 3.00 38. Unit Cost of Supply (Ps./kwh) * 172.71 196.70 237.89 39. Average Tariff (Ps./kwh) Required to Obtain a) 0 % RoR * 166.82 190.98 232.59 b) 3 % RoR * 174.27 199.64 243.21 Note: * = Estimated by us.

26

TABLE 7 ELECTRICITY TARIFF BASED ON COST-PLUS (MARK-UP) PRICING POLICY (USING THE PARAMETERS UNDER THE EXPERT COMMITTEE'S TARIFF POLICY RECOMMENDATION) 1. Energy Sales (mu) 2. Total Expenses (Rs. Crores) 3. Sales Rate (Ps./unit) At 10 % Mark-Up At 15 % Mark-Up 4. Unit Cost of Supply (Ps./kWh) 5. Sales Revenue Receipts with 10 % Mark-Up Rates (Rs. Crores) 6. Miscellaneous Revenue (Rs. Crores) 7. Total Revenue Receipts (Rs. Crores) 8. Net Surplus (Rs. Crores) 9. Value of Net Fixed Assets Employed at the Beginning of the Year (Rs. Crs.) 10. Rate of Return on N. F. A. (%) 11. State Government Equity Capital 12. RoR on Equity Capital (%)

1997-98 1998-99 1999-2000 2000-2001 2001-2002 7542.91 8553.00 10143.79 12371.61 15051.61 1137.51 1461.20 2143.62 3099.05 4459.23 165.89 173.43 150.81

187.92 196.47 170.84

232.46 243.02 211.32

275.55 288.07 250.50

325.89 340.70 296.26

1251.26 44.45 1295.71 158.20

1607.32 48.9 1656.22 195.02

2357.98 53.78 2411.76 268.14

3408.96 59.16 3468.12 369.07

4905.15 65.08 4970.23 511.00

1872.52 8.449 1480.02 10.689

2470.24 7.895 1530.02 12.746

3593.12 7.463 1580.02 16.971

4993.92 7.390 1630.02 22.642

5672.77 9.008 1680.02 30.416

USING THE PARAMETERS UNDER THE EXISTING TARIFF POLICY 13. Total Expenses (Rs. Crores) 1302.76 1682.35 2413.10 3390.44 14. Sales Rate at 10 % Mark-Up (Ps./kWh) 189.98 216.37 261.68 301.46 15. Sales Receipts (Rs. Crores) 1433.04 1850.59 2654.41 3729.48 16. Total Revenue Receipts (Rs. Crores) 1477.49 1899.49 2708.19 3788.644 17. Net surplus (Rs. Crores) 174.73 217.14 295.09 398.20 18. Rate of Return on N. F. A. (%) 9.331 8.790 8.213 7.974

4759.95 347.87 5235.95 5301.025 541.08 9.538

TARIFF BASED ON 'VARIABLE' COST PRICING POLCY 19. Total 'Variable' Costs (Fuel, Power Purchase, O & M and Establishment Expenditure) (Rs. Crores) 785.76 1001.37 1569.21 2434.10 3756.80 20. Sales Rate (Ps./unit) at 60 % mark-up 166.68 187.33 247.51 314.8 399.35 at 40 % mark-up 145.84 163.91 216.58 275.45 349.43

27

Table 8(a) Marginal Cost pricing A Representative Hydro-Power Plant Distribution of Capital Outlay at Varying discount Rates (Rs. Crores) Year

Capital

Present-Valued Capital Outlay at Discount Rates of

Outlay (1996 prices)

12%

14%

15%

16%

17%

18%

20%

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999

1.53 4.574 10.714 17.874 25.956 34.038 44.784 32.058 6.308 2.164

3.020 8.061 16.859 25.112 32.559 38.123 44.784 28.623 5.029 1.540

3.358 8.807 18.096 26.481 33.732 38.803 44.784 28.121 4.854 1.461

3.539 9.200 18.739 27.184 34.327 39.144 44.784 27.877 4.770 1.423

3.728 9.607 19.399 27.899 34.926 39.484 44.784 27.636 4.688 1.386

3.925 10.028 20.077 28.627 35.531 39.824 44.784 27.400 4.608 1.351

4.130 10.464 20.772 29.368 36.141 40.165 44.784 27.168 4.530 1.317

4.569 11.382 22.217 30.886 37.377 40.846 44.784 26.715 4.381 1.252

Total

180

203.709

208.497

210.986

213.538

216.156

218.839

224.407

28

TABLE 8 (b) MARGINAL COST PRICING - A HYDRO-POWER PLANT Total Capital Cost at 1996 Prices Capacity of the Plant Constuction Time Plant Life Period Capacity Factor Kwh/kw Auxiliary Consumption Net kwh/kw

= Rs. 180 Crs.

= 120 mw. = 10 years = 25 years = 68.5 % = 8760 x 0.685 = 6000.6 =5% = 5700.57

COMPUTATION OF MARGINAL COST PER UNIT (KWH) ELECTRICITY At Different Discount Rates of Present-valued capital cost (Rs.Crs.) Working capital (5.75 %) (Rs.Crs.) Total capital cost (Rs. Crs.) Annuity factor for 25 years plant-life Annualized capital cost (Rs. Crs.) Cost per kw capacity (Rs. per year) Cost adjusted for 20 % reserve margin O & M costs (11.5 %) (Rs./kw/year) Discounted O & M costs as in 1999 with base 1996 (Rs./kw/year) Total cost (Rs./kw/year) Cost per unit (Ps./kwh)

12%

14%

15%

16%

17%

18%

20%

203.71 11.71 215.42 0.1275 27.47 2288.86 2746.63 1725

208.50 11.99 220.49 0.1455 32.08 2673.36 3208.03 1725

210.99 12.13 223.12 0.1547 34.52 2876.35 3451.62 1725

213.54 12.28 225.82 0.1640 37.04 3086.40 3703.67 1725

216.16 12.43 228.58 0.1734 39.64 3303.50 3964.20 1725

218.84 12.58 231.42 0.1829 42.33 3527.62 4233.15 1725

224.41 12.90 237.31 0.2021 47.96 3997.07 4796.49 1725

1227.82 3974.45 69.72

1164.33 4372.36 76.70

1134.22 4585.83 80.45

1105.13 4808.81 84.36

1077.04 5041.24 88.43

1049.89 5283.04 92.68

998.26 5794.75 101.65

29

TABLE 8 (c) MARGINAL COST PRICING A REPRESENTATIVE COAL-BASED THERMAL POWER PLANT DISTRIBUTION OF CAPITAL OUTLAY AT VARYING DISCOUNT RATES (Rs. Crores) Year

Capital Outlay (1996 prices)

1994 1995 1996 1997 1998 1999

88.425 206.325 399.55 288.2 216.15 111.35

Total

1310

Present-Valued Capital Outlay at Discount Rates of 12% 110.920 231.084 399.550 257.321 172.313 79.257

14% 114.917 235.211 399.550 252.807 166.320 75.158

15% 116.942 237.274 399.550 250.609 163.440 73.214

16% 118.985 239.337 399.550 248.448 160.635 71.337

17%

18%

20%

121.045 241.400 399.550 246.325 157.901 69.524

123.123 243.464 399.550 244.237 155.236 67.771

127.332 247.590 399.550 240.167 150.104 64.439

1250.446 1243.963 1241.029 1238.292 1235.744

1233.38

1229.181

30

Table 8 (d) MARGINAL COST PRICING - A COAL-BASED THERMAL POWER PLANT Total Capital Cost at 1996 Prices = Rs. 1310 Crs. Capacity of the Plant = 350 mw. Constuction Time = 6 years Plant Life Period = 25 years Capacity Factor = 68.5 % Kwh/kw = 8760 x 0.685 = 6000.6 Auxiliary Consumption = 10 % Net kwh/kw = 5400.54 FUEL COST Coal Amount of coal required per kwh (kg) = 0.532 Amount of coal required per kw (kg) (0.53 x 8760 x 0.685) = 3192 Price of coal (Rs./tonne) = 1000 Cost of coal per kw (Rs./kw) = 3192 Oil Amount of oil required per kwh (ml) = 12 Amount of oil required per kw (ml) = 72000 Price of oil (Rs./kl) = 6396 Cost of oil per kw (Rs./kw) = 461 Total fuel cost (Rs./kw) = 3653 O & M costs (2.5 %) (Rs./kw/year) =935.71 Total annual costs (Rs./kw/year) = 4588.71 COMPUTATION OF MARGINAL COST PER UNIT (KWH) ELECTRICITY At Different Discount Rates of Present-valued capital cost (Rs.Crs.) Working capital (4.83 %) (Rs.Crs.) Total capital cost (Rs. Crs.) Annuity factor for 25 years plant-life Annualized capital cost (Rs. Crs.) Cost per kw capacity (Rs. per year/kw) Cost adjusted for 20 % reserve margin Discounted annual costs as in 1999 with base 1996 (Rs./kw/year) Total cost (Rs./kw/year) Cost per unit (Ps./kwh)

12%

14%

15%

16%

17%

18%

20%

1250.50 60.40 1310.89 0.1275 167.14 4775.40 5730.48

1243.96 60.08 1304.05 0.1455 189.74 5421.05 6505.26

1241.03 59.94 1300.97 0.1547 201.26 5750.27 6900.32

1238.29 59.81 1298.10 0.1640 212.91 6083.00 7299.60

1235.74 59.69 1295.43 0.1734 224.66 6418.80 7702.56

1233.38 59.57 1292.95 0.1829 236.51 6757.29 8108.75

1229.18 59.37 1288.55 0.2021 260.44 7441.15 8929.38

3266.15 8996.64 166.59

3097.25 9602.51 177.81

3017.15 9917.47 183.64

2939.79 2865.06 2792.83 2655.50 10239.39 10567.61 10901.58 11584.88 189.60 195.68 201.86 214.51

31

TABLE 8 (e) PRESENT-VALUED INCREMENTAL DEMAND (MW) AT DIFFERENT DISCOUNT RATES

Year 1996-97 1997-98 19998-99 1999-2000 2000-2001 2001-2002

Year 1996-97 1997-98 19998-99 1999-2000 2000-2001 2001-2002

*Gross Demand 2187 2368 2559 2763 2983 3226

*Gross Demand 2187 2368 2559 2763 2983 3226

Demand at Transmission End ** 1858.95 2012.8 2175.15 2348.55 2535.55 2742.1

Incremental Demand

Demand at Distribution End *** 1703.71 1844.71 1997.94 2157.73 2332.66 2525.97

Incremental Demand

153.85 162.35 173.4 187 206.55 Total

141 153.23 159.79 174.93 193.31 Total

Incremental Demand at Transmission End at Different Discount rates of 12% 14% 15% 16% 17% 18%

137.366 129.424 123.423 118.842 117.202 626.257

134.956 124.923 117.040 110.719 107.276 594.914

133.783 122.760 114.013 106.918 102.692 580.166

132.629 120.652 111.090 103.278 98.341 565.991

131.496 118.599 108.266 99.793 94.210 552.363

130.381 116.597 105.537 96.453 90.285 539.253

Incremental Demand at Distribution End at Different Discount rates of 12% 14% 15% 16% 17% 18%

125.893 122.154 113.735 111.171 109.689 582.643

123.684 117.906 107.854 103.573 100.399 553.415

122.609 115.864 105.065 100.017 96.109 539.663

121.552 113.875 102.371 96.612 92.037 526.447

120.513 111.937 99.768 93.351 88.171 513.740

119.492 110.047 97.253 90.227 84.498 501.517

20%

128.208 112.743 100.347 90.181 83.008 514.488

20%

117.500 106.410 92.471 84.361 77.687 478.428

Notes: * = As per 15th Power Survey ** = After accounting for 15 % auxiliary consumption and transit losses *** = As per 15th Power Survey after accounting for T & D losses

32

TABLE 8 (f) PRESENT-VALUED CAPITAL OUTLAY IN TRANSMISSION AND DISTRIBUTION DURING THE IX PLAN (Rs. Crs.) Capital Outlay in Year Transmission 1997-98 150 19998-99 200 1999-2000 250 2000-2001 220 2001-2002 180 Total 1000

Year 1997-98 19998-99 1999-2000 2000-2001 2001-2002 Total

Capital Outlay in Distribution 75 100 125 110 90 500

Discounted Capital Outlay at Different Discount Rates of 12% 14% 15% 16% 17%

133.929 159.439 177.945 139.814 102.137 713.263

131.579 153.894 168.743 130.258 93.486 677.959

130.435 151.229 164.379 125.786 89.492 661.320

129.310 148.633 160.164 121.504 85.700 645.312

128.205 146.103 156.093 117.403 82.100 629.903

18%

20%

127.119 143.637 152.158 113.474 78.680 615.066

125.000 138.889 144.676 106.096 72.338 586.998

Discounted Capital Outlay at Different Discount Rates of 12%

14%

15%

16%

17%

18%

20%

66.964 79.719 88.973 69.907 51.068 356.632

65.789 76.947 84.371 65.129 46.743 338.980

65.217 75.614 82.190 62.893 44.746 330.660

64.655 74.316 80.082 60.752 42.850 322.656

64.103 73.051 78.046 58.702 41.050 314.952

63.559 71.818 76.079 56.737 39.340 307.533

62.500 69.444 72.338 53.048 36.169 293.499

33

TABLE 8 (g) COMPUTATION OF MARGINAL COST PER UNIT OF ELECTRICITY AT DIFFERENT DISCOUNT RATES Discount Rates of 12% 14% 15% 16% 17% Marginal Cost of Generation (including Working Capital and O & M Cost) (Ps./kwh) a) Hydro-power 69.72 76.70 80.45 84.36 88.43 b) Thermal power 166.59 177.81 183.64 189.60 195.68 Average * 108.47 117.14 121.73 126.46 131.33 Transmission Incremental cost (Rs. Crs) 713.263 677.959 661.32 641.312 629.903 Incremental demand (mw) 626.257 594.914 580.166 565.991 552.363 Incremental demand (mu) 5486.011 5211.447 5082.254 4958.081 4838.700 AIC (Ps./kwh) 130.01 130.09 130.12 129.35 130.18 Distribution Incremental cost (Rs. Crs) 356.632 338.98 330.66 322.656 314.952 Incremental demand (mw) 582.643 553.415 539.663 526.447 513.74 Incremental demand (mu) 5103.953 4847.915 4727.448 4611.676 4500.362 AIC (Ps./kwh) 69.87 69.92 69.94 69.97 69.98 Marginal cost of electricity (ps./kwh) 308.36 317.16 321.79 325.77 331.49

18%

20%

92.68 201.86 136.35

101.65 214.51 146.79

615.066 539.253 4723.856 130.20

586.998 514.488 4506.915 130.24

307.533 501.517 4393.289 70.00

293.499 478.428 4191.029 70.03

336.56

347.07

Note: * = Average, assuming 60 per cent hydro-power and 40 per cent thermal by 1999 - 2000.

34

TABLE 9 MARGINAL COST PRICING KAKKAD HYDRO-POWER PLANT DISTRIBUTION OF CAPITAL OUTLAY AT VARYING DISCOUNT RATES (Rs. Lakhs) Year

Capital Outlay Present-Valued Capital Outlay at Discount Rates of (original) at 1996 12% 14% 15% 16% 17% 18% 20% Prices Upto 1984-85 1295 3392.23 13216.047 16343.441 18149.279 20136.369 22321.102 24721.190 30245.463 1985-86 251 629.7 2190.443 2661.258 2929.611 3222.342 3541.426 3888.981 4678.724 1986-87 1045.1 2477.68 7695.298 9185.308 10023.597 10930.124 11909.783 12967.770 15341.141 1987-88 514.9 1128.05 3128.171 3668.361 3968.340 4289.930 4634.481 5003.414 5820.490 1988-89 570 1162.16 2877.465 3315.162 3555.074 3810.043 4080.871 4368.396 4997.075 1989-90 0 0 0 0 0 0 0 0 0 1990-91 1366 2352.18 4642.786 5162.971 5440.735 5730.843 6033.728 6349.837 7023.572 19991-92 1182 1789.5 3153.710 3445.529 3599.324 3758.561 3923.386 4093.943 4452.849 1992-93 1022 1405.86 2212.148 2374.442 2458.858 2545.505 2634.423 2725.650 2915.191 1993-94 939 1192.13 1674.857 1766.193 1813.081 1860.791 1909.331 1958.708 2060.001 1994-95 1350 1546.09 1939.415 2009.299 2044.704 2080.419 2116.443 2152.776 2226.370 1995-96 346.24 368.25 412.440 419.805 423.488 427.170 430.853 434.535 441.900 1996-97 938.22 938.22 938.220 938.220 938.220 938.220 938.220 938.220 938.220 1997-98 1569.54 1497.2 1336.786 1313.333 1301.913 1290.690 1279.658 1268.814 1247.667 1998-99 2210 2108.14 1680.596 1622.145 1594.057 1566.691 1540.025 1514.033 1463.986 Total 14599.00 21987.39 47098.38 54225.47 58240.28 62587.70 67293.73 72386.27 83852.65 Note: Data are not available for 1989-90. The yearly distribution of outlay is estimated from cumulative outlay in each year given in Economic Review, Govt. Of Kerala, various volumes.

35

TABLE 10 MARGINAL COST PRICING - KAKKAD HYDRO-POWER PLANT Total Capital Cost at 1996 Prices = Rs. 21987.39 lakhs. Capacity of the Plant = 50 mw. Constuction Time = about 20 years Plant Life Period = 25 years Capacity Factor = 68.5 % Kwh/kw = 8760 x 0.685 = 6000.6 Auxiliary Consumption =5% Net kwh/kw = 5700.57 COMPUTATION OF MARGINAL COST PER UNIT (KWH) OF ELECTRICITY At Different Discount Rates of 12% 14% 15% 16% Present-valued capital cost (Rs.Crs.) 470.98 542.25 582.40 625.88 Working capital (5.75 %) (Rs.Crs.) 27.08 31.18 33.49 35.99 Total capital cost (Rs. Crs.) 498.07 573.43 615.89 661.86 Annuity factor for 25 years plant-life 0.1275 0.1455 0.1547 0.1640 Annualized capital cost (Rs. Crs.) 63.50 83.43 95.28 108.55 Cost per kw capacity (Rs. per year) 12700.66 16686.76 19055.59 21710.84 Cost adjusted for 20 % reserve margin 15240.80 20024.11 22866.71 26053.01 O & M costs (11.5 %) (Rs./kw/year) 3357.77 3357.77 3357.77 3357.77 Discounted O & M costs as in 1999 with base 1996 (Rs./kw/year) 2389.99 2266.40 2207.79 2151.18 Total cost (Rs./kw/year) 17630.79 22290.51 25074.50 28204.19 Cost per unit (Ps./kwh) 309.28 391.02 439.86 494.76

17% 672.94 38.69 711.63 0.1734 123.41 24682.70 29619.24 3357.77

18% 723.86 41.62 765.48 0.1829 140.02 28004.31 33605.18 3357.77

20% 838.53 48.22 886.74 0.2021 179.23 35845.42 43014.51 3357.77

2096.49 31715.74 556.36

2043.64 35648.82 625.36

1943.15 44957.66 788.65

36

PART III

THE SURGE OF REFORMS

The following two chapters examine the main aspects of the regulation/restructuring of the power sector, starting with a review of the international experiences (Chapter 8), followed by discussions on the reform measures in India in general and in Kerala in particular (Chapter 9).

1

CHAPTER 8 EXPERIENCES FROM ABROAD

"The supply of gas and of water, electric lighting and the establishment of tramways must be confined to very few contractors. They involve interference with the street, and with the rights and privileges of individuals. They cannot, therefore, be thrown open to free competitions, but must be committed, under stringent conditions and regulations, to the fewest hands. As it is difficult…. satisfactorily to reconcile the rights and interests of the public with the claims of an individual, or a company seeking, as its natural and legitimate object, the largest attainable private gain, it is most desirable that, in all these cases, the municipality should control the supply, in order that the general interest of the whole population may be the only object pursued."

-

Chamberlain (1894).

1. Power Sector Reforms: The Background

Since electricity cannot be readily stored, but must be supplied continuously in adjustment with varying demand, it is technically required that power sector is to be a vertically integrated natural monopoly of generation, transmission, and distribution. Scale economies and specific assets (those with high costs for consumers to switch suppliers), particularly in distribution, provide the natural monopoly with considerable market power. Hence the significance of control or regulation of this sector. In most of the countries, therefore, the government has wielded the monopoly power of the electricity industry.

However, the post-war period public enterprise experiments had an unfortunate but avoidable history of dysfunctionings in general in a background of high level corruption and porkbarrel politics.. Subsequently, the Thatcherite experiments, on a ‘There is no alternative (TINA)’ logic, brought the public sector down from the ‘commanding heights’ to a tragic, premature end. The reforms in the electricity supply industry (ESI) of England and Wales has been ranked as one

2

of the most ambitious and the most appealing attempts anywhere to introduce competition into a normally vertically integrated monopoly, This has had far reaching consequences of spread effect across the globe, ushering in an era of re-emerged liberalism.

Further fillips have come from the Reagonomics, and the historic fall of the Communist Bloc in the eighties. Still further fuels have been poured by the premier international financial agencies in the course of their apotropaic mission of delivering the developing nations from the apparent fiscal crisis that their ‘infatuation’ with the public sector slogans eventually enticed them into. And the siege is on !

While power sector reforms in most of the developed nations have largely been an emulation endeavour of the famous ‘big experiments’ of the UK with the radical liberalism, the causative spur for reforms or restructuring in almost all the developing nations has been whipped up by their survival surge out of the infamous fiscal crisis. Chile pioneered the movement, much before the UK, initiating the reform process as back as in 1978 with the help of the World Bank. The ‘global fiscal crisis’ gave the World Bank the much sought after leverage in devising a conditional loan policy. In fact the euphemism of structural adjustment entered into the international parlance in 1980 with the introduction of the World Bank’s structural adjustment loan (SAL) as a new type of credit. This was to provide quick-dispersing loans to finance general imports over a period of years conditional on an agreed set of measures intended to strengthening the balance of payments (BoP) while maintaining a development momentum. During the 1980s, however, the Bank’s original emphasis on the BoP gradually faded away, while the stress upon the ‘economy-wide programs of reforms’ got much more strengthened. Later on the Bank switched emphasis from SAL to sectoral adjustment loans with narrower policy objectives, though the general policy thrust is similar. For added momentum, the International Monetary Fund (IMF) also rose to the occasion in 1986 with a new Structural Adjustment Facility, intended to provide medium-term BoP assistance to low income countries facing protracted BoP crisis in return for a program of policy measures dictated by the IMF. It was augmented at the end of 1987 by an Enhanced Structural Adjustment Facility with considerably greater resources.

3

Structural adjustment, according to these international financial agencies, refers to reform of policies and institutions – microeconomic (such as taxes), macroeconomic (such as fiscal imbalances), and institutional (such as public sector inefficiencies). The intended structural changes improve resource allocation, increase economic efficiency, expand growth potential and increase resilience to further shocks (World Bank 1998 a: 11; IMF 1989: box 4).

Most of the countries in Africa, Latin America (Chile, Argentina, Brazil, Uruguay, etc.) and Asia (New Zealand, Malaysia, the Philippines, Thailand, etc.) are ‘progressing’ on such reform path that leads especially to a deregulated privatised electricity industry. In all these countries electricity supply is now no longer the statutory monopoly of the State-owned public utilities; independent power producers (IPP) have already made their powerful appearance. Power ministries or authorities have been converted into corporations (even in the ‘socialist’ China and Vietnam). In Pakistan and some other countries, public power generation assets have been sold to private investors. In Malaysia, the National Electricity Board was privatised in 1990, and renamed as Tenaga Nasional Berhad (TNB); its assets have been listed in the local financial market and shares sold to the public. The generation sector has opened up for multiple players. The Philippines stands out among these countries by vigorously promoting private power projects with the 1991 crash program of IPP entry in response to the major and persistent power shortages. In Thailand, the Electricity Generating Authority of Thailand (EGAT) gave way to its commercial subsidiary Electricity Generating Company (EGC), IPPs, and small power producers (SPPs) in 1992. Similarly, in Indonesia, the national power utility, Perusahaan Umum Listric Negara (PLN) split itself into two commercial generating companies in 1994 and allowed IPP entry.

The power sector reform process in China also had its origin in the capital crisis of the late 1970s that resulted in cutbacks in investment not only in heavy industry but also in power industry. The process began with the decentralisation of administrative responsibility to provincial and regional authorities, with provisions for changes in financing sources from budget allocations to loans. The search by the provincial and local governments for alternative sources of financing had its first result when in 1980 the Guangdong provincial government started discussions with the China Power and Light Co. Ltd. of Hong Kong (China Light) on a joint venture (JV) development of a nuclear power plant in the province. It took five years for the final signing of the JV

4

agreement, that marked the beginning of foreign direct investment (FDI) in China’s power sector. During the period from 1984 to 1998, there were 34 financial closures of power projects with private FDI in China, worth US$ 14.8 billion. Of this amount, total debt financing accounted for US$ 12.3 billion (83 per cent) and equity participation for 17 per cent. These projects represented 26 thousand MW of installed capacity (IC), equivalent to about 10 per cent of China’s total IC.

The privatisation bid in the Chinese power sector took place in 1993. The initiative came from the Shenyang municipal government in Liaoning province when it raised US$ 100 million by transferring 55 per cent of its ownership and management rights in an existing 2x200 MW coalfired power plant to a private Hong Kong firm under a 20-year co-operation agreement. The Electricity Law of December 1995 explicitly stipulates that the generation sector is open to investment by all economic entities and individuals, both domestic and foreign. The FDI laws of June 1995 permit private ownership without shareholding limitation of all generation plants or companies (except hydropower plants with IC of more than 250 MW), but prohibit private investment in T & D facilities. Private participation is allowed by means of both partial or total divestiture of existing assets and direct investment in new projects.

Most of the former Communist Bloc countries have also started large scale restructuring: the Czech Republic, Hungary, Poland, and Croatia, who are also keen to join the European Union (EU). Romania and Slovak Republic have, however, so far desisted from committing themselves to the ultimate goal of electricity privatisation. Among the OECD countries, the UK, Canada, Germany, Spain, and Australia have made ‘great strides’ towards privatisation, while liberalisation gathers momentum across the US, Ireland, Sweden, Norway, and New Zealand. The only electricity industry in the public sector that still stands impervious to the sweeping waves of the so called reform reagents is the Electrcite de France (EDF) in the hands of the French Government. Within the EDF, the electric sector is totally integrated – generation, transmission and distribution. The vertical integration includes also the supply of primary energy supplies viz., water and coal. Countrywide, the state’s monopoly consists of 85 per cent of the generation, 100 per cent of the transmission, and 95 per cent of the distribution. Private electricity production is subject to heavy constraints – it is forbidden to sell or to supply electricity to a third party. The producers, having

5

surplus power generation, net of their own consumption, are connected to EDF’s network, and EDF is bound to buy the private production.

2. Power Sector Reforms: The Agenda

It is argued that technological progress coupled with institutional adaptability has offered enough rooms for competition in a normally vertically integrated monopoly such as ESI. The new arrangement is seen to enhance the prospects for the consumers to entertain the much needed multiple choice over the supply of electricity service. This in turn is seen to drive the producers and suppliers to innovate and increase efficiency leading to greater productivity, implied in competitiveness, and then to higher standard of living. i)

Competition

Competition presupposes liberalisation; all the economic agents (buyers and sellers) should be free on their part of decision-making. Customers should have free access to choice over suppliers based on their willingness to pay for the services. This requires elimination of monopoly elements through a regime for third party access (TPA) to the monopoly network infrastructure. That is, the suppliers should be free to enter the market – they should have easy access to the transmission and distribution (T & D) network in order to carry on their business of buying power from (any) generator and selling it to (any) customer; similarly for the bulk consumers to buy power directly from (any) generator. Such right to ‘wheeling’ is indispensable in deregulated ESI; ‘wheeling’ refers to the function of a utility owning transmission networks to take in and pass along (i.e., deliver) electricity of (or generated by) another utility.1 Moreover, they should also have the ‘freedom to fail’ and hence to exit out of the market. All the countries that have embarked upon power sector reforms/restructuring have instituted such a TPA regime.

Success of liberalisation, on the other hand, depends on the corresponding restructuring of the whole sector. Restructuring refers to changes in the ownership or internal operation of an electric utility aiming to effect a separation of the generation, transmission and distribution functions which is a prerequisite for competition. Remember, the ESI is an interrelated continuum

6

of three activity links: generation, transmission, and distribution. The last, distribution business, in turn, in view of retail competition, is separated into two components: a ‘wires’ or pure electricity transportation business and a ‘supply’ or retailing business that involves purchasing electricity from the generators and reselling it to retail customers in the wholesale electricity market. (‘Supply’ is the word used in the UK Electricity Act and in companies’ licenses to denote the selling of electricity to consumers.) Now the first and the last in the vertical bundling, i.e., generation and supply, can be easily split up and subjected to competition. However, the T & D system still remains a natural monopoly, defeating the ideal of an easy split up and liberalisation. (Some countries have allowed duplication of electric wires to induce competition on the distribution front too, as also of cables in telecommunication sector.) In this context, if a generator or a supplier facing competition in his respective field, is actually integrated with the natural monopoly sector of T & D, then he will be able to extract monopoly rents from the natural monopoly sector. In order to avoid this detrimental possibility that can effectively bar TPA to T & D network, it is essential to establish arms-length relationship among the ownership interest of all these sectors. That is, the continued interrelation of generation with transmission and that of distribution with supply should be completely eliminated through unbundling, i.e., vertical separation, of the T & D system from other sectors. This can be accomplished in three ways: i) structural unbundling, ii) functional unbundling, and iii) accounts unbundling.

The first, full structural separation, that splits up the vertically integrated industry into several legal entities having no significant common ownership or operations, is the best way to ensure the separation of the segments that can be subjected to competition from the monopoly segments.

In contrast to this, functional unbundling, while permitting to rein in the ownership of both the competitive (i.e., generation) and the monopoly (i.e., transmission) segments, assigns the operation of each sector to separate management structures. The disaggregated entities, though managed independently, are not legally separate companies. This arrangement, however, calls for tight and vigilant regulation in order to safeguard the TPA rights of those entrants having no direct financial interests in the monopoly network.

7

Accounts unbundling is the weakest form of separation. By this the accounts of each of the different businesses of the whole company are ‘ring-fenced.’ Ring-fencing requires that transactions between business units be on a commercial basis; goods and services provided by one to the other be charged for. This is expected to force cross-subsidies into the open, as costs and revenues that legitimately accrue to each business need to be identified, and thus to render the incumbent utility unable to pursue discrimination in favour of itself against competitors. In practice, however, this method is not fully viable; such complete transparency on the part of the company cannot be expected; so is the vigilance on the part of the Regulator also. Moreover, there still remains the possibility (and temptation) for collusion, i.e., transfer of commercially valuable information between business units.

In addition to such restructuring based on unbundling of the sectors, liquidity and transparency are also required to ensure competition in a liberalised environment. A properly liquid commodity market is characterised by the interactions of a large number of economic agents, each with no substantial market power to influence the market price. This is possible only if the market is transparent, i.e., only if there is simultaneous and equal access to full market information for all market participants. Thus competition facilitated by a non-discriminatory TPA regime requires as a prerequisite cost-oriented, unsubsidised and transparent TPA tariffs and conditions. Tariff transparency guarantees no discrimination, and this transparency can be effected only if all the electricity suppliers have equal and easy access to information on tariffs and network capacity such that they come to face the same tariff under same conditions. The cost-orientation of tariffs is normally based on long-run marginal cost (LRMC) of network infrastructure with timedifferential designs in line with temporal variations in demand for access, as well as with regulated tariff caps that can bring in increased efficiency.

In order to ensure the existence of sufficient numbers of buyers and sellers to allow the market to be liquid, all forms or degrees of market concentration or dominance (as the generation entities of the integrated monopoly have) should be eliminated. This requires that the generation and the supply entities be sufficiently disaggregated into several entities of insignificant influence. Once this is effected, the consequent market liquidity, along with market transparency, ‘kickstarts’ and sustains competition.

8

Presently there are two approaches to competition in ESI – i) expanded wholesale competition (EWC, i.e., competition at the wholesale level); and ii) retail competition (RC, i.e., competition at the retail level). The former facilitates opening up of the wholesale market such that all generators could sell power to any local distributors and other wholesale customers. In other words, under the EWC, the three functions of the utilities would be unbundled such that local distributors could purchase transmission services from other utilities as well as generation services from other generators. Customers would however still purchase power from the local distributors as they do now. Thus EWC involves deregulation only in the generation sector, along with a nondiscriminatory TPA regime in the transmission segment.

The second, RC, is more radical. Here customer has the right to purchase electricity from any distributor of his choice or directly from any generator. The local line company wheels power to the customer. Thus in this arrangement, both the upstream generation and the downstream retail sale functions are deregulated and open for competition.

Similarly, there are two market structures considered for the ESI: i) bilateral contracting, and ii) Poolco. In the first market, electricity transactions take place under specific supply contracts between two parties: generators on one side and distributors, marketers or final consumers on the other. In the other market, short term power trades are co-ordinated by a centralised spot market.2 Each of these market structures can be implemented with any of the two competition types.

Since electricity cannot be stored, electrical loads must be balanced such that generation equals total consumption. Under bilateral contracting, the load-balancing function is performed by a new entity, the Independent System Operator (ISO). The ISO is not allowed to own any generating utility so as to ensure its impartiality toward all the generators it is responsible for coordinating. Usually, the transmission grid company owning no generating facilities takes on the ISO’s role. The ISO’s counterpart in the spot market is the Poolco, responsible for the operation of a centralised spot market for electricity. Thus spot market is the quintessential manifestation of competition, facilitated by liquidity and transparency. Immediate delivery, implied in spot market

9

concept, on the other hand, requires adequate and truly accessible T & D network infrastructure. Hence the significance of a TPA regime.

In addition to spot market operations, transactions can also take place outside the Poolco domain. Suppliers or customers can have long term bilateral contracts (called self-nominated contracts) with generators; and this may often result in imbalances between the actual and the contracted for deliveries. Consequently, a method of pricing these imbalances becomes essential; and this is facilitated by Power Pool or Exchange. The mechanism for identifying a price for these imbalances, however, differs in different liberalised electricity markets. For example, in Norway, imbalances are settled net of contracted amounts; in Victoria (Australia), the entire supply of electricity is considered an imbalance; and in the English electricity market, traders use Contracts for Differences (CfDs) to achieve a Pool price replacement. CfDs are a variant of swap contracts which are more convenient as they are settled by transferring the difference between the contract price and market price. Generators and distributors enter into contracts that specify a fixed price for electricity for some time period. Under these CfDs, sellers would compensate buyers when the spot price exceeds the contract price and buyers would compensate sellers when the spot price is less than the contract price. Thus a generator receives, besides the normal pool price for any sales, a sum equal to the difference between the specified strike price and the pool price, multiplied by the specified number of units contracted. The effect is as if the generator had sold electricity forward at the strike price, while the supplier had purchased forward at the same strike price; this thus provides a hedge for both parties, and at the same time ensures that the pool price serves its role of dispatching stations in the correct merit order. Though CfDs are financial instruments, they have the same economic effect as a contract for physical delivery.

A significant offshoot of the establishment of spot market has been the futures market for financial instruments based on spot price. Futures markets have evolved and flourished as they serve two main functions: price discovery and risk transfer. As futures contracts are standardised and traded on a central Exchange, prices, determined through open and competitive bidding among a large number of market players, can be readily observed and compared, in contrast to prices settled on negotiations between two parties. Futures contracts provide an independent transparent pricing signal for the market that serves as a pricing index for other contracts. The futures price of

10

a commodity may be the best indicator of its expected spot price in the future. Besides facilitating such price discovery, the futures market also offers means for risk transfer through hedging or speculation. Below we give an account of the radical restructuring of the ESI meant to initiate competition in the sector in some of the leading reformer countries.

England

Though the process of the restructuring of the whole UK economy started in 1957 and acquired strength and character during the Thatcherite period of late 1970s, the power sector’s reforms were delayed till late 1980s. The White paper on privatisation of the English electricity supply industry (“Privatising Electricity”, February 1988) initially proposed that all the nuclear stations together with 60 per cent of the conventional stations would be placed in one large company, National Power, with the rump of the conventional stations in PowerGen. Despite substantial financial support provided through a fossil fuel levy on conventional generation, nuclear power proved unsaleable. At a very late stage, all the Central Electricity Generating Board (CEGB)’s nuclear stations were withdrawn from the sale and transferred to Nuclear Electric, to remain in government ownership, and National Power and PowerGen were to operate under license and work in a deregulated sector with competitive supply/market conditions. The revised proposal became law as the Electricity Act in July 1989. Transmission was transferred as a regulated natural monopoly to the National Grid Company (NGC), created in the 1989 reforms as a separately owned and operated company, prohibited as part of its license conditions from owning or operating any generation business. The CEGB was thus divided into four companies (NGC, PowerGen, National Power, and Nuclear Electric), which were vested as public limited companies on March 31, 1990, so were made the 12 Area Boards, now known as the Regional Electricity Companies (RECs) ), which received electricity from the CEGB under the Bulk Supply Tariff (BST), and then distributed and sold this electricity to its regional consumers at its own tariffs.. The NGC was transferred to the joint ownership of the RECs, and the RECs were sold to the public in December 1990; but in 1995, the NGC was demerged from the RECs and became an independent company. Sixty per cent of National Power and PowerGen were subsequently sold to the public in March 1991 and the remainder four years later. Of the twelve RECs in the England

11

and Wales, eight were taken over by US electricity companies, two by UK-based water companies, one by Scottish Power and another merged with Scottish Hydro. Some have changed ownership a second time. Several RECs have developed active second-tier supply businesses also (i.e., supply outside one’s own authorised area). Most of the RECs have also entered generation in partnership with new entrants. One REC, Eastern, has acquired substantial generation capacity which National Power and PowerGen had to divest following widespread concerns about their market power, and it is now one of the largest generators in England and Wales. Most RECs are now active in the supply of gas as well as electricity and some also have telecommunications licences.

The new structure introduced in March 1990 thus divided the process of electricity supply into four activities: generation, transmission, distribution and supply. Generation accounts for around two-thirds of the industry’s costs, transmission for 10 per cent, distribution for 20 per cent, and supply for the remaining 5 per cent. Supply is further subdivided into sales to a franchise market of smaller consumers, restricted to the local REC, and a non-franchise market of large consumers, which can be served by any company acting as a private, or second-tier, supplier.

Each REC’s distribution business is an effective regional monopoly, but, subject to controls on the prices it can charge and the quality of service to provide. Supply businesses on the hand are progressively liberalised, facing increasing degrees of competition. When the ESI was privatised, it was provided that all customers would have the right to choose their own retail supplier, with market opening in three phases: 1990 (large customers), 1994 (medium customers) and 1998 (domestic consumers). However, the retail competition was fully implemented only by May 1999. By the end of September 2000, over 33 per cent of domestic customers registered to change suppliers.

U.S.A. In the USA, the Federal Energy Regulatory Commission (FERC) Order 888 of April 1996 required the power industry to separate wholesale transmission from generation functions and to offer open non-discriminatory access to transmission networks for wholesale buyers and sellers, under the same tariff as applied to their own wholesale sales and purchases of electricity. Accordingly 13 States have adopted plans to implement electricity industry restructuring.

12

California, prompted in part because of the highest electricity prices in the country, was the first state to deregulate the electricity market, following the Deregulation Bill of September 1996. Its three main utilities, Pacific Gas and Electric (PG&E), Southern California Edison (SCE) and the state's big shareholder-owned company, San Diego Gas and Electric sold their generating assets, but retained ownership and responsibility for the maintenance of the T & D assets. Most of these purchasers were headquartered in other states, such as Texas' Dynegy and North Carolina's Duke Power. An Independent System Operator (ISO) has been created that allows transmission-owning utilities to keep legal title to their transmission facilities, while the operational control is taken over by the ISO. Thus ISO is responsible for short-term co-ordination (such as day-ahead scheduling and hourly re-dispatch), prices for use of the transmission grid (which are to ensure incentive consistent with a competitive market and least-cost use of the grid), and administering a system of tradable transmission congestion contracts. The ISO is forbidden to have any ownership interest in generation, thus securing its unbundling from transmission to ensure impartiality in its responsibility for co-ordinating all the generators. Reforms in a limited number of States also aim at separation of transmission and supply. Similar ISOs have been set up in other parts of the U. S. also. The US electricity futures market started in March 1996.

However, contrary to expectations, California’s deregulation experiments has now ended up in a disastrous power crisis, with abnormal rate hikes, continuing black outs and brown outs, and perilous condition of the state’s two main electric utilities. The crisis has come to such a pass that the state has had to step in to underwrite the electricity purchases of these utilities and also to buy power directly from suppliers. The general demand now is for ‘re-regulation’ of the sector with an active state presence in power.

Scandinavia A first step in the deregulation process, following the Energy law that came into effect on January 1, 1991, in Norway, was to separate the high voltage transmission system from Statkraft, the large state-owned generating enterprise. Statnett is the state-owned enterprise in charge of the national grid. Statkraft has been reorganised and corporatised, it now being a pure generating company without any public obligations or system responsibility. Also the principle of common carriage and third-party access (TPA), on a non-discriminatory basis to all electricity transmission

13

grids (national, regional and local) by all connected consumers, has been established. Charges for use-of-system are published on a transparent basis. In principle, a household can buy power directly in the spot market (now called daily market, as it is cleared daily) for its own consumption. Vertically integrated companies have been required to ring-fence generation from distribution through separate divisions, allowed to operate as independent entities in the market with separate budgets and accounts. Some distribution companies have been corporatised, but not privatised.

Similarly in Sweden, the transmission function of Vattenfall (Swedish State Power Board) was divested to Svenska Kraftnat (SwedeGrid), created in 1991 as the State-owned company, responsible for the national high voltage transmission grid. The company is independent of generators and distributors. It is interesting to note that this unbundling took place in Sweden before the new electricity framework was implemented in 1996. However, the failure in organisational separation for regional and local transmission has presented problems for properly monitoring allocation costs to transmission activities. Svenska Kraftnat is statutorily obliged to connect buyers and sellers to the network infrastructure on reasonable and open terms.

Finland is following Sweden; a new electricity act has been accepted by the Parliament in 1995, similar to the proposed Swedish act. This act removes all barriers to entry into the electricity market. A main result is that Swedish Vattenfall has taken a substantial share of the Finnish market from the dominating Finnish generator, Imatran Voima. Denmark, characterised by no state ownership in ESI (all assets are owned directly or indirectly by retail distributors that in turn are owned by municipalities or co-operatives), also is considering to follow Sweden and Finland.

Chile In 1978, the Chilean government started a drastic restructuring of the electricity system which had been based on two publicly owned integrated companies, ENDESA and CHILECTRA. The drastic restructuring of the sector was achieved by separating generation and transmission from local electricity distribution. Thus, the distribution side of ENDESA was broken into several distribution companies, each with coherent geographic and economic units, and they were subsequently privatised. Similarly, CHILECTRA was broken into 3 units, one of generating and two of distribution.

14

Argentina The restructuring of the economy of Argentina was a key element in the stabilisation policy adopted after the hyperinflation years of 1989 and 1990. The privatisation process occurred within a short time frame in early 1990s. Till 1992, ESI in Argentina was in the hands of federal and state authorities. A number of major problems were facing the sector such as: (a) public inefficiencies, (b) breaks in the payment chain leading to multiple inefficiencies in supply system, (c) tariffs being not related to economic costs resulting in inefficient use, (d) supply and transport restrictions, and (e) blackouts. This situation compelled the government to plan for major restructuring and in 1992, the transformation took place and a legal framework similar to the one used in the UK was adopted that divided the vertically integrated electricity business into three distinct activities of generation, transmission, and distribution. Under the new law, the following agencies were set up:

CAMMESA – to manage the wholesale electricity market players transparently, using high technology that facilitates efficient energy generation and sale;

ENRE (National Energy Regulatory Board) – to function as the regulator for the electricity sector; and

WEM (Wholesale Electricity Market) – to consist of all generators, transmitters, distributors, and large users as well as other participants like traders and brokers.

Generation was separated from transmission before privatisation, and the larger generation companies were split up. About 80 per cent of the generation capacity have been privatised. Generators have freedom to enter into contracts with distributors and large consumers. Transmitters consist of the national company Transener (mostly 350 KV), a private company, and six regional and two independent transmitters at 220 and 132 KV level. Transmission companies are obliged to allow non-discriminatory access to capacity to any participant in the spot market,

15

and similarly, distributors must allow third parties to use their system in exchange for a regulated tariff. 70 per cent of the distribution has been privatised and there are about 28 companies in the distribution sector. Traders are responsible for commercialisation of generation, demand, royalties, and exports/imports. Bulk customers (100 kW and above) are allowed to directly access the wholesale (spot) market; this scheme is administered by a separate organisation representing Government, generation and distribution companies, and bulk consumers.

In August 1997, an agreement between Argentina and Brazil was signed to integrate the two countries’ power markets with guaranteed free competition among generators, banning all State subsidies, and requiring pricing be based purely on costs.

The privatisation and regulatory restructuring of the current administration resembles the Chilean electricity law, in that it creates a competitive wholesale market and regulates transmission and distribution. However, there are two important differences between the Chilean and the Argentinean regulatory reform processes. First, Chile’s restructuring of the sector was more measured. It started with a regulatory reform that forced the state-owned enterprises to behave according to the new regulatory rules; and the second stage was privatisation, which was done at a relatively slow pace. But privatisation in Argentina was thrust upon the power sector within a very short time frame. Second, Chile’s electricity companies are widely owned by both employees and the public at large (a large portion through institutional investors). As a result, there is strong political support for maintaining the financial viability of the companies. Moreover, the electricity law is such that if the companies and the administration differ radically on major price readjustments, there is a process for conflict resolution that limits administrative discretion. On the other hand, Argentina’s centralised decision making in the hands of the administration reduces to a large extent the credibility of any regulatory legal statute, raising questions about the credibility of Argentina’s current reform.3 ii)

Regulation

The significance of regulation in liberalised electricity market lies in its active role in developing and then facilitating the continuation of competition, starting with regulating access to

16

the network. The Regulator should have the authority to make decisions regarding what is required to create and keep a level playing field for market agents as also to maximise benefits to customers.

More particularly, an independent specialist Regulator should oversee tariff setting in respect of access to network infrastructure and other specific and detailed market entry criteria. Two types of regulatory mechanisms are there in practice now: fixed price and cost-plus mechanisms.

Under a fixed price regulatory mechanism, the regulator defines and devises ex ante a set of prices (or a weighted average of prices for different services) that the regulated firm can charge the customers for its services. The mechanism also involves automatic price adjustment formulas tied to exogenous variables such as general inflation, possible productivity growth or fuel price rise. The critical feature of this mechanism is that the prices so defined are not directly linked to the regulated firm’s realised costs or profits.

Under cost-plus mechanism or rate of return (RoR) regulation, prices are set in public hearings based on an assessment of the capital and operating costs required to generate and supply electricity, by setting a rate of return on capital employed for the industry. The undesirable tendency to invest large scale capital (‘gold plating’) has been attributed to this form of regulation. Averch and Johnson (1962) argued that when the Regulator sets the allowed RoR above the capital cost, the utility will tend to use more capital than if it were unregulated and choose an inefficiently high capital-labour ratio for its level of output. It is because of the inadequate and asymmetric information for the Regulator on the company that makes it difficult for him to set the rate at the most efficient level. A number of empirical tests of the Averch-Johnson hypothesis, based on data before 1973 of the US utilities, have demonstrated a bias toward capital. However, this could be explained by other factors, such as risk-averse behaviour of the utilities by providing for higher reserve margins. Moreover, ‘gold-plating’ can lead to beneficial environmental effects, such as investing in pollution control capital. The US has later on developed a more sophisticated approach involving not only the assessment of capital required, but also determination of a fair RoR on this

17

rate base in an inter-industry comparison and allowing for cost components such as fuel cost to be passed through.

The base prices once set are not adjusted automatically (or only partially adjusted) for changes in costs over time, that is, prices remain fixed until they are reviewed by the Regulator again. The regulatory lag (the period between regulatory reviews), often last for several years. This regulatory lag effectively turns cost-plus regulation also into a fixed price regulation system with a cost-based ratchet adjustment every few years.

18

TECHNICAL INNOVATION

Private power production offers incentives for technical innovation that are absent from rate of return regulation. Under the fixed price formulas contained in private power contracts, any cost-reducing technology adopted by the supplier adds to profit. The private producer also takes the risk that the innovation adopted will not perform as expected. Regulated firms, on the other hand, must pass through to ratepayers all production economies realised through innovation. The economies do not contribute to profit. When the outcome of adopting new technology is not favourable, the regulated firm may also be denied cost recovery by the regulator. These incentives may combine to bias the regulated firm away from risky new technology, while leaving the private firm neutral.

There was a significant record of new technology adoption by the private power industry during the 1980s. The first steam-injected gas turbine (STIG) was installed by a California paper mill subject to stringent requirements for air pollution control. The adoption of circulating fluidised bed (CFB) coal combustion was also substantially more widespread in the private power section than among regulated or government-owned utilities. In the area of renewable energy, the solar thermal technology of Luz International was commercialised through private power contracts. Although none of this evidence conclusively proves that more innovation will occur under private power than under a regime of regulation, it is clearly suggestive. - Gilbert and Kahn (1996 a, p. 204)

In recognition of the weaknesses of cost of service regulation, alternative, incentive-based regulatory methods have come up. The regulatory gap (i.e., the period between when a utility’s costs fall and when the regulator orders a corresponding rate reduction) in the case of RoR regulation in fact serves as an informal built in incentive in that the utility can reduce its costs and profit for some time before the next hearing.

Price cap regulatory system that is in many respects an extended version of regulatory lag and is in vogue in many privatised infrastructure sectors belongs to the fixed price regulatory 19

regime. Price capping, recommended for British Telecom by Stephen Littlechild (1983) (who later became British Electricity Regulator), sets prices indexed on the rate of inflation, usually measured by the retail price index (RPI), which are then reduced by the possible productivity increases (X), (due to internal efficiency or technological improvement), as assessed by the Regulator and then increased by allowances for fuel cost rises (Y), if any. In the absence of fuel cost rises, prices cannot be increased by more than the RPI less X, the Regulator’s assessment of possible productivity, i.e., ‘RPI – X’. That is, in real terms, prices come down by X per cent. There is an implicit incentive for efficiency in this price cap in that if the utility can outperform X (that is, increase the actual productivity more than the Regulator’s assessment), it can increase its profits also.

The fixed price feature of the price cap system, however, means that profits from productivity improvement reaped by the firm during a certain period (during the ‘fixed price’ contract) will not result in automatic downward adjustments to permitted prices. Hence the value of ‘X’ is periodically renegotiated and then this may deter cost-reducing innovations especially towards the end of a regulatory period even if opportunities are open. In general, if the firm has to decide on an investment level prior to the regulator fixing a price, under-investment would be the predicted outcome (Vickers and Yarrow 1988: 90). The firm manages to operate with an inefficiently low capital stock in the hope of the Regulator setting a welfare-maximising MC price, given the (short run) capacity of the firm. This result, however, would not follow, if the firm expected the Regulator to set a price reflecting long run marginal costs (LRMC). But such a regulatory mechanism remains unviable as it requires the Regulator to have costless access to unbounded information on cost and demand conditions.

Considerations of sunk costs and specific assets also pose problems to the Regulator, as he is to find some way of committing himself not to ‘hold up’ the firm by setting prices too low to yield reasonably sufficient return. RoR regulation, on the other hand, has some solution to such problems. However, there remains a problem with both the regulatory mechanisms, i.e., the Regulator finds it difficult to set the cap (on price or RoR) at the correct level for want of adequate information. Moreover, encouraging efficient spatial and time of day price signals proves to be very difficult under price regulation. Though this was intended to be ‘regulation with a light hand’,

20

price regulation in several industries in the UK has, in effect, become tighter and more detailed over time, and RoR considerations have been evidently of prime importance at points of regulatory review.

Revenue cap regulation sets the total allowable annual income for the transmission grid owner, regional grid owners and distributors. This would cover the total cost of the grid activities plus a return on invested capital. Norway, for example, uses such a system.

A pro-competitive governance structure of regulation presupposes transparency. Chances for any clandestine negotiations between the Regulator and the regulated should be effectively averted. This will enhance the credibility of the results of regulation. This requires setting up a transparent formal procedure for the Regulator’s decision-making and clean, published reasons for his decisions once made. This is also a prerequisite for accountability. All the concerned parties affected by the decisions of the Regulator should be able to seek a review of the decisions; this process should be clear, detailed and set out in advance, such that it will dispense with the need for judicial review of Regulatory decisions involving significant delays and uncertainty.

Below we give an account of the regulatory mechanisms in the ESI of some of the leading reformer countries.

England Regulation of the sector is carried out by the Office for Electricity Regulation (OFFER). The British Regulator is constrained by systematic checks and balances: any decisions that the utility opposes must be cleared by both the Monopolies and Mergers Commission and the Secretary of State for Trade and Industry. In June 1999, the two regulatory offices of gas (Ofgas) and electricity (OFFER) were merged under the Office of Gas and Electricity Markets (Ofgem).

The ‘RPI – X’ cap is used in the UK for transmission and distribution (T & D) network infrastructure revenues. REC supply price controls allow the pass-through of generation costs which are competitively determined and T & D costs, which are directly regulated. Price discrimination is strictly prohibited in both the CfDs and the supply markets. The OFFER

21

periodically reviews and specifies the price caps and oversees non-discriminatory practices in the sector. PRICE-CAP Vs. RoR REGULATION

Economists have increasingly criticised RoR regulation for its inefficiency……Politicians and the public have criticised price regulation for its lack of fairness in the distribution of rents between consumers, shareholders and managers…….

….It is interesting that in Britain regulation of the gas and electricity networks is moving more towards RoR regulation of the level of total revenue, while encouraging the network operator to propose efficient relative prices to improve the structure of the tariffs, subject to regulatory approval system that requires vertical separation to avoid an anti-competitive pattern of changes………

Certainly, price regulation is increasingly replacing RoR regulation in the US for telecoms, while the World Bank is a strong advocate elsewhere. While it is clear that price regulation is superior for telecoms where it may only be needed in the transition to deregulation, it is less clear that permanent price regulation with periodic reviews is superior for core network monopolies like water, gas and electric distribution, balancing the better incentives of price regulation against the lower perceived investor risk and cost of capital of RoR regulation.

- Newbery (1998).

USA Rate setting in the U. S. electric power sector in the post war period has been dominated by the traditional cost-of-service or rate-of-return model, in which prices for electricity are set at levels sufficient to cover the full costs. The individual State’s Public Utility Commission (PUC) sets a rate of return on capital employed in the industry, assessed primarily on the capital required to generate and supply electricity. This ‘revenue requirements’ approach has begun to yield to rate mechanisms that provide incentives for efficient production, but this shift has occurred only recently and most utility revenues are still cost-based. It is common for the revenue requirement to 22

be determined by two separate kinds of hearings, one oriented to fuel costs (originated in response to the world oil price shocks in the 1970s) and one oriented to non-fuel costs including overhead and administrative costs, certain kinds of operation and maintenance costs, and the fixed costs of capital investment. Issues in a ‘fuel cost adjustment’ hearing involve the utilities’ efforts to minimise operating costs by efficient fuel contracting and maximising the opportunities for wholesale purchase.

FERC retains exclusive jurisdiction over the rates, terms and conditions of interstate transmission of power and interstate open access policies and terms. The emerging competitive market for wholesale electricity service dictates the FERC to place more reliance on market forces that are to displace cost-based regulation gradually. State regulatory authorities would still retain their power to regulate generation asset costs, siting of generation and transmission facilities, and decisions on bundled retail service.

In California, a rate cap ensured that no customer’s rate would be higher than that on 10 June 1996 up till 2002. Small customers (with a maximum peak demand of less than 20 kW) are to receive a cumulative rate reduction of 20 per cent by 1 April 2002. This retail price cap against the soaring wholesale price of electricity is in part alleged to have caused the supply utilities to incur huge losses, leading to the present power crisis.

Canada Two key features distinguish the Canadian system of electricity generation from that of the U. S. First is the preponderance of publicly owned systems and second, the lack of statutory regulatory agency (SRA) supervision. The pace of restructuring and regulation has been slow in Canada mainly due to lower electricity costs from predominantly hydel and nuclear stations, absence of federal regulation(regulation being largely a provincial matter), dominance of provincially owned utilities, and local issues.

The province of Quebec has started electricity regulation recently. The regulator, Regie de l'energie was established in 1997 to regulate natural gas distribution and electricity transmission

23

and supply monopolies, examine complaints from customers, and to provide a transparent, equitable, independent, and impartial decision process.

In the state of Alberta, the government has restructured the power market without appointing a regulator. The Electric Utilities Act of January 1, 1996 created a power pool consisting of all the electricity bought and sold in Alberta, along with energy imported and exported through the province. The pool works like a commodity exchange, provides a place for distribution companies and markets to buy power and for power producers and marketers to sell their output. The power pool itself does not buy and sell electricity; rather, it operates the market, receiving offers and bids from pool participants and establishing an hourly market price for electricity by matching supply with demand.

In Ontario province, regulatory supervision of the choice of technology, quantum of capacity, and construction program is very complex; numerous agencies are involved, both provincial and federal. The government has recently appointed a committee to suggest the pattern of regulation and restructuring of the energy sector (including power sector). There have been much discussion to privatise Ontario Hydro, the second largest Canadian electricity utility, that supplies wholesale power to over 300 municipally owned distribution utilities, and proposals of a number of alternatives – increased third party access, and horizontal and vertical unbundling – whereby competition would be encouraged.

Scandinavia The formal, government-enforced regulations have historically been fairly weak in the Scandinavian countries in general. Instead, the industry is to a large extent characterised by publicly owned dominant firm leadership, self-enforced club-regulation, and yardstick competition. The Scandinavian countries being highly co-operative societies, the development of the electric power sector is mainly the result of negotiations, co-operation, and self-enforced regulation among the major market agents. This holds, for example, for power exchange, transmission services, reserve capacity, and the like. In particular, there has been a close coordination of planning for long-term capacity expansion among the main power companies.

24

Until recently, the concepts of regulation or deregulation hardly existed in the Scandinavian vocabulary and even less so in connection with the electricity supply industry. This reflects reality in that formal price, rate of return, or cost of service regulation has never been imposed on that industry in Scandinavia. Among the two major regimes for electricity markets, regulated private ownership and public enterprise, Scandinavia falls somewhere in between with its club-regulated markets. The key notions in regulation have been self-regulation, price leadership, and yardstick competition. Most public service obligations (such as those concerning merit order dispatch, realtime dispatch, frequency and voltage regulation, and reserve capacity) have been solved through the club organisations, especially the grid clubs and the pool clubs. To avoid free rider problems, membership in these clubs has been tied to obligations to share the costs of system responsibility.

Norway employs a system of revenue cap regulation that sets the total allowable annual income for each of the monopolies – Statnett, the regional grid owners, and the distributors. This would cover the total cost of the grid operation and a return on the invested capital. The latter was set by the Norwegian Water Resources and Energy Administration (NVE), the main regulatory body. In 1997, it was set at 8.3 per cent. NVE also defines a maximum and minimum RoR as well as an efficiency factor. The grid owner must demonstrate to NVE the cost-effectiveness of new investment before it can be added to the rate base.

Chile Much of the success of this restructuring process in the Chilean ESI is based on the nature of the regulatory regime, which is quite transparent, developed following the creation of the National Energy Commission (CNE) in 1978. It has the responsibility for developing and coordinating investment plans, policies, and regulation for the sector. The two basic functions of the CNE are: 1) it determines the regulated prices, which have to be approved by Minister of Economics, though the administration can interfere with only major retail or toll price realignments; and 2) it is to guarantee the co-ordination of the several independent generation, transmission, and distribution companies in the interconnected system. Bulk customers (2 MW and above) are given free choice of supplier.

25

Chile uses benchmark RoR regulation (in both the electricity and telecommunications sectors), which bases regulatory decisions on the performance of a notional best practice firm. The Chilean system specifies in detail how the benchmarks are to be calculated. It is administratively very complex, as it depends on advanced applications of regulatory economics.

Argentina Argentina uses ‘RPI – X’ regulation to transmission and distribution tariffs, with ‘X’ (the Regulator’s assessment of possible productivity increases) set equal to zero at the time of privatisation. Transmission tariffs are adjusted every six months based on a weighted average of the US producer and consumer price indices, and reviewed every five years. Distribution rates are adjusted every six months based on a US price index. Large users can opt out of this tariff scheme and negotiate their own energy supply directly by paying a distribution toll fee, regulated by ENRE, for the right to use the distribution network infrastructure.

South-East Asia (The Philippines, Indonesia, and Thailand) Tariff reform was an integral part of the sectoral restructuring in all these countries. Entry of IPPs resulted in high cost of power, which, however, the system was able to absorb without much aberration, thanks, especially in the Philippines, to a phased program, initiated earlier, of rationalisation of power tariffs and elimination of subsidies. Besides the tariff reform, the following are some of the other notable features of the Philippines power sector reforms: (i) All the earlier projects had a cost plus basis for tariffs, with specified rate of return; however, later on, contracts were awarded on the basis of lowest tariff price, without limiting the return on investment, thus inducing the promoters to minimise costs both during construction and operation. (ii) Private sector entry in the Philippines was regulated through a special law (Republic Act 6957), that stipulated a detailed bidding and evaluation procedure with the proviso that the winning bidder will automatically be granted the franchise.

China China has adopted a catalogue tariff system since the 1960s in view of providing preferential treatment to heavy industry, chemical industry and agriculture in terms of both allocation and price of power. It has in general eight main categories of consumers (viz.,

26

residential, lighting (non-residential), industrial-commercial, heavy industry, chemical industry, commercial, agriculture and irrigation) with three voltage classifications, making 24 basic categories. To each of these categories is assigned a basic (catalogue) price, to which are added a range of charges and fees (such as local government surcharge, construction fee, connection fee, and so on) to derive the final end-user price.

The actual, precise, definition of each category varies among the provinces. And each supply company has its own level of catalogue prices and its own system for structuring the final end-user prices. Though the supply company publishes prices, the price-setting process lies in the domain of the Price Administration Department (PAD) of the State Planning Commission (renamed as State Development and Planning Commission (SDPC) following the 1998 restructuring of the government), whose primary concern is the control of inflation. The PAD sets inflation targets for each province, and within these constraints negotiations on consumer tariffs are carried out. The key players are the local and provincial governments, State Power Corporation of China (SPCC), State Economic and Trade Commission (SETC), as well as the provincial power companies (PPCs) and their local supply companies. However, the Commission may have its own priorities, for example, of raising prices in order to generate funds for investment through either raising specific fees or allowing for profit.

China has recently adopted a ‘new plant-new price’ tariff policy such that part of the power sector investment should come from high tariff charged to consumers of incremental power. This ‘new plant-new (i.e., higher) price’ policy along with the ‘old power-old (i.e., low) price’ regime divides the sector between existing capacity, serving the existing customers and new capacity that is to meet incremental demand. The new price serves to meet the higher costs of power purchase from the (mostly privately owned) new plants such that the state power supply companies need not absorb any loss. In many provinces, the ‘commercial’ category was introduced in the 1990s in line with the ‘new price’ policy as the ‘new’ profitable enterprises such as hotels, banks, large shops and department stores and other new non-state firms can afford to pay more (and require no subsidy). Startling discrepancies exist between the ‘new’ and the ‘old’ prices. For example, in Beijing the ‘new’ commercial customer is paying twice the final tariff of the ‘old’ industrial customer, who in turn is paying 50 per cent more than the residential customer.

27

The present tariff policy based on the Electricity Law of 1995 and the ‘new plant-new price’ regulations stipulate that tariffs allow recovery of costs including taxes and reasonable profits. The formal tariff approval process has two parts: setting initial tariff levels and annual tariff adjustments. The tariff levels are derived from annual financial statements with a focus on debt repayments. The government has adopted a system of annual tariff revisions, rather than indexation, (though preferred by the private developers), in order to ensure that price rises are ‘socially’ acceptable.

3. Lessons From The Experiences

Many of the countries have not yet completed the liberalisation process; hence a fullfledged evaluation may not be in order now. However, at least the short-run manifestations of the results/consequences of privatisation and liberalisation have already made their effects felt. One of the most pronounced effects of liberalisation of the power sector has been an improved power supply position in those countries where power shortage as a major problem was one of the primary causatives for the opening up of the power sector to IPPs (as in the Philippines, for example). However, in some other countries (e.g., Indonesia and Thailand) this has resulted in over capacity and the consequent excess obligation to make capacity payments (World Bank 2000 c). The evidences on efficiency, the singularly acclaimed hallmark of competition, however, are not conclusive. The results are mixed as regards the investment efficiency (long-run) and operating efficiency (short-/medium-run).

‘Where privatisation has occurred with the liberalisation of an energy sector, the cost of capital used in decision-making is likely to increase. The cost of capital for such new projects increases because of the reduced availability not only of cash-flow to repay investments guaranteed by taxpayers and captive customers, but also of favourable interest rates which a government [on the other hand] can obtain for financing projects.’ (ICC 1998: 76). Such an increased investment cost regime necessarily leads to higher (than socially optimal !) marginal cost and tariff of electricity. For instance, most of the fast track projects of the IPPs especially with predominantly foreign funding, Dabhol Power Project being an initial typical case, accepted

28

wholeheartedly by India, have had capital costs significantly higher than the potential indigenous alternatives (say, of National Thermal Power Corporation, NTPC). (For example, the capital cost of Dabhol Power Project (Phase I) was Rs. 4.48 crores per MW, whereas that of the NTPC projects might have been in the range of Rs. 3 crores to Rs. 3.6 crores per MW. The IPPs with foreign direct investment (FDI) imposed as much as 40 per cent service cost on foreign resources, while the upper bound for the cost of foreign resources could have been barely 15 per cent; see Morris, 1996). Most of the IPPs are financed through bilateral credit, which perforce involves tying of equipment, and overpricing. Experiences of other developing countries on liberalisation path may not be much different and an inductive generalisation of the Indian experience is quite warranted. For example, in Indonesia, tariffs in the PPAs, established through negotiation between PLN and project companies (including consortium partner companies owned by or with close links to high government officials), have been in the range of 6 to 8 US cents per unit, far higher tha PLN’s retail tariff rates of about 3 US cents per unit (World Bank 2000 c). Moreover, indexing the PPA tariffs to the US dollar in most of these developing countries serves as a heavy drain of the limited forex reserves, and the ‘fiscal crisis’ has dealt severe blows to these economies through this provision. The long-run least cost criterion of (investment) efficiency is under sure shadows in these countries. In the case of liberalising developed countries, for example, the UK, the ‘main argument for expecting improved investment efficiency is that a privatised industry is likely to introduce proven technology more quickly, while avoiding the temptation to invent the nextgeneration technology (such as Advanced Gas-cooled Reactors, AGRs)’, that saves investment costs (Newbery and Green, 1996, p. 70). But nothing bars a public sector utility too to follow suit.

29

FROM NEW ZEALAND Initially, a major motivation for the government to support privatisation was, rightly or wrongly, the size of the debt reduction that could be achieved by the sale. ECNZ is a major corporation, by New Zealand standards, and sale of such an asset would have a significant impact on public sector debt….….government and the public need to be convinced that the benefits of further reform will outweigh the costs. Successive studies have found this hard to prove, particularly in view of the relatively large overhead required to establish sophisticated market mechanisms in such a small economy. Privatisation is not expected to provide further improvements in operational efficiency or co-ordination, both of which appear to be more than adequately furnished in the status quo, but to improve the environment for competitive investment and to prevent an otherwise inevitable regression of the ECNZ to the inefficiencies of its former self…..

On the downside, concern has arisen, particularly after recent experience of shortages, that a multitude of generators will encounter difficulties in co-ordinating their activities or that they may deliberately ‘play games’ in ways that raise prices, increase costs, reduce energy efficiency, or cause outages. Studies confirm that such behaviour can be expected unless a fairly high level of contracting, break up, or both is established and can impose costs on the economy by distorting economic dispatch. On the other hand, long-run prices will ultimately be disciplined by entry, and short-run gaming can be interpreted as legitimate market behaviour designed to establish an appropriate level of contracting in the market….. - Culy, Read, and Wright (1996: 362, 364)

A significant component of operating efficiency concerns using the right inputs of fuel and labour in the right amounts, i.e., reducing the costs of generation. In the UK, ‘The effect of privatisation on the generators is that they have nearly halved their staff in the first three years, as well as closing research laboratories. There are clearly strong incentives to reduce the cost of fuel' (but 'this has primarily affected British Coal’) (op. cit.: 74). In the investor-owned utilities in the US, the same pattern of a decrease in the level of staff is registered, with a reduction of about 100 000 employees during 1986-95 period. Their total operating and maintenance (O&M) costs have 30

decreased by nearly 25 per cent, mainly, however, due to the decline in the fuel costs, though there has been some reduction in non-fuel O&M costs. (ICC 1998: 53-54). The internal restructuring of the Electricity Corporation of New Zealand (ECNZ) resulted in reduction of staff numbers directly employed from around 6000 in 1987 to fewer than 3200 by 1992 (Culy, Read, and Wright 1996: 333). Though the reduction in the number of the employed has been accused of adversely affecting customer service, there is not enough data to support a conclusion either way.

Many studies (starting with, for example, Stigler and Friedland 1962) in the US have found that the electricity prices of the publicly owned utilities are much lower than of the private utilities, the prices of the regulated privately owned electric utilities being lower than of the unregulated ones (Smiley and Greene 1983; Moore 1970; Peltzman 1971). Some studies (for example, Pescatrice and Trapani 1980) have found the costs incurred by private utilities as much higher than the costs of public utilities.

The recent energy crisis, particularly the Californian power crisis, has however, lent an inflationary trend to the power price in general. The worst hit has been California, whose electricity prices had been much higher than other states’ prices in the US, on account mainly of cost overruns of nuclear power and the expensive alternative ‘green’ power. Deregulation was conceived of inducing competition among numerous sellers and reducing prices. However, the wholesale electricity peak hours spot prices in California, which was much less than $ 50 per megawatt-hour before 1998, began to shoot up by 2000 and crossed $ 300-mark by 2000 end (The Economist, January 20, 20001: 57). On May 8th, 2001, it reached the all-time high of $ 560 per megawatt-hour (The Economist, May 12, 2001: 39). With the retail prices being kept capped, these rate hikes have not affected the customers, but the suppliers have been driven to the red. Under pressure from the suppliers, however, there was initially a 10 per cent rise in the retail rates, and later on by more than 40 per cent (mainly for business customers). Finally as there appeared signs that California's power crisis might be spreading to neighbouring states, the top electricity regulator of the country, the FERC, tightened its grip; in June 2001, it unanimously voted to impose a form of price control on wholesale electricity prices in 11 western states, including California.

31

INFRASTRUCTURE SECTOR REFORM IN DEVELOPING COUNTRIES

In an effort to improve performance, many developing countries have implemented or are contemplating major infrastructure sector reform programs. Reforms generally focus on three related issues: privatisation, restructuring to promote competition, and regulatory reform. But while privatisation, competition, and deregulation are the standard prescriptions for improving the performance of infrastructure sectors, these terms can be misleading. Important segments of most infrastructure sectors continue to be natural monopolies, and competition in the market cannot be relied on to yield satisfactory performance. Moreover, competition for the market – through concession or franchise contracts – must confront problems resulting from significant sunk costs, asset specificity, and incomplete contracts. Many segments in which competition in the market is a reasonable option for allocating resources are likely to be imperfectly competitive and characterised by some degree of market power in both the short and long run. Moreover, the effectiveness of competition will depend on policies governing the initial structure of the competitive segments, the conditions of entry into the market, and the price and non-price terms and conditions of access to bottleneck monopoly network facilities for competing suppliers.

These considerations imply that simplistic privatisation and ‘complete’ deregulation policies are unlikely to be realistic or effective policy options in most infrastructure sectors.

- Joskow (1998).

The labour markets of the developing countries on the liberalisation front have however been the worst affected. Phenomenal increase in economy-wide unemployment has been the pinching price of reform in these countries (for example, Argentina, Chile, etc.; some of the European countries too have faced the same problem, for instance, Sweden), in terms of big cuts into the over-manning inefficiency in the public sector, coupled with a complete employment freeze in this sector following fiscal tightening measures by the Governments. Whereas in the developed countries such as the UK, the US, Australia, etc., the reform-induced unemployment rise has subsided in due course, it has remained unabated in the other countries on account mainly 32

of structural potential differences. The developed countries have already been endowed with a developed and hence highly flexible and absorbent private sector; moreover, the erstwhile public sector, once restructured and opened up, has also been able to respond in quick positivity to the changes by accommodating more private initiatives

with copious openings. Such structural

resilience and adaptive response, promoting ‘inward’ investment, cannot be expected in a developing economy with an immature and imbecile private sector; and the drastic cuts into (over)employment in the restructured sectors in these countries just go to swell the tidal waves of unemployment. The reform has been an ill-timed imposition, ignoring the logical necessity of development of some prerequisites for it, at a high cost.

A perverse effect of the private initiative to increase cost efficiency in restructured ESI has been the strong incentives and tendencies to retire the older, labour-intensive, coal-fired power plants, especially in the UK. The real motive behind this has, however, been to raise prices by restricting supply (Newbery and Green 1996: 74). Price-raising machinations are innate to practical private enterprise logic, through, say, collusion. Competitive efficiency presupposes a transparent, liquid market. On the other hand, ‘In an imperfectly competitive industry, the opportunities for raising prices above operating cost is greater.’ (op. cit., p. 71). In the English ESI, for example, the combined market share of generation output of the privatised National Power and PowerGen, among about 35 generation companies, has been about 50 per cent; such large size (market concentration) has given these two generators ‘the ability to bid at prices well above marginal cost, with potentially serious dead-weight losses as a result’ (op. cit.: 73; also Green 1999), giving the lie to the ideal of competition by means of transparency and liquidity. This experience has taught, for example, Argentina, to divide the generation sector of its ESI into a number of generation companies of comparable size before privatisation. However, such horizontal split may not often be technically feasible and economically viable, much of which depends on the size distribution of plants also. On the other hand, market liquidity as well as transparency might be secured under conditions of a contestable contract market, that might drive the contract prices of the generators down to the LRMC of entrants (Newbery and Green 1996: 73), if such conditions are achievable.

33

That price tending towards LRMC would be higher after privatisation is a foregone conclusion, and experiences across the globe corroborate this. Moreover, MC pricing favours the large industrial customers and penalises the small retail consumers (in sharp contrast to the actual practice in many developing countries). Thus electricity prices have declined in real terms much more for the industrial customers, for instance, in the UK, the US, and in the Nordic market. Real price reductions for domestic customers in these developed countries have however been less, partly because of the lack of economies of bulk purchase unlike in the case of larger consumers, and because of the expensive meter charge requirements (ICC 1998: Ch. 4). In most of the developing countries, electricity prices were heavily subsidised. For example, the World Bank Report on Long Term Issues in India’s Power Sector (1991) estimated that the average customer tariffs in India were less than half of LRMC. Liberalisation lets loose this spiral up across the LRMC cap, as the Philippines experience confirms. The entry of IPPs there has resulted in very high cost of power. We have already seen that the entry of IPPs with FDI (e.g., in India, and in other developing countries too) into the power sector generally results in higher (than socially efficient) marginal capacity costs and hence prices. Rapid liberalisation measures in developing countries often involve some ineluctable tendencies of ex cathedra assertions and attempts to legitimise such monopoly pricing practices in the hope of attracting IPP entry, and at the same time to impose cold turkey measures of cost-oriented sharp tariff rises on the customers, unlike the phased program of tariff revisions in the Philippines. Apt examples in point are the controversial behaviour of the Telecom Regulatory Authority of India in tariff revision on the morrow of the privatisation of the sector, and the recent electricity tariff hike in AP in line with MC pricing principles that has sent the worst shock waves across a hitherto pampered sector of residential customers. In some of the countries, this is often characterised and justified in terms of a trade off between cheaper power and more investment for the benefits of quality of service. In China too the ‘consumers are heavily affected by the reforms in the shape of higher prices, but are rewarded by improved quality of service’ (Andrews-Speed et al. 1999: 446).

One of the festering problems associated with power sector reform stems from the evaluation and solution of ‘stranded costs’. The situation refers to the prospect that, as the ESI becomes more competitive, the assets of some utilities may become obsolete and they may not be able to earn enough to recover the costs of these investments. In other words, with competition and

34

MARKET POWER AND ELECTRICITY PRICE

In practice, it soon became clear that the two major generators, National Power and PowerGen, had sufficient market power in the Pool to raise prices and make supernormal profits. Between 1991 and June 1995, the regulator issued seven reports on the Pool, each of which raised concerns about the companies’ position. Although Pool prices are highly visible (they are published in the Financial Times each day), and have attracted much attention as an indicator of problems in the electricity market, their short-term financial importance for electricity companies and most consumers is limited. This is because although practically all wholesale electricity trades must pass through the Pool, and are made at Pool prices, the great majority are hedged via CfDs…….

…….The cost was passed on to small consumers, who were within the Regional Electricity Companies (RECs) monopoly franchise until 1998. …… From 1998 onwards, however, all consumers will be allowed to choose their supplier, which will limit the RECs’ ability to pass high costs on to their domestic consumers. - Green (1999).

entry of new, low-cost generators, electricity prices tend to fall, and the older high-cost generators or suppliers are left unable to recover the cost of their plants or contracts. The unrecovered costs thus become stranded. Wide variation exists in the estimates of stranded costs, because of uncertainty of information. For example, the stranded-cost losses of the US ESI are estimated to range from $ 10 billion to $ 200 billion (Brennan et al. 1996: 100). The real issue is regarding its resolution: should the investors be forced to swallow these losses, or should these be transferred to the customers in higher retail prices? Stranded costs are said to have played a role in the Californian power crisis: under pressure from the big and politically powerful utilities, the state agreed to value the assets built before deregulation much more generously and passed on a part of the costs to the new entrants (as well as to customers). Thus the new entrants have become severely handicapped to compete on price (The Economist, January 20, 2001: 58).

35

Another important source of concern is the impact of market reforms on environment. The blind profit-fetish of the private enterprise has always been at the target end of accusals of reckless corrosion of environment. Generation of electricity produces at least four forms of air pollution: (i) particulate patter (soot, dust, dirt, aerosols), (ii) sulphur dioxide (SO2), primarily responsible for acid rain, (iii) nitrogen dioxide (NO2, or in general nitrogen oxides, NOx), and (iv) greenhouse gases, especially carbon dioxide (CO2), mainly responsible for contributing to the greenhouse effect (global warming, a general increase in the temperature at the earth’s surface).4 However, evidences indicate that environmental effects of reforms are case-specific, and hence can be positive or negative, depending on the market circumstances. Much depends in turn on the types of electricity generation plant and fuel mix. In fact, the development of combined cycle gas turbine5 (CCGT) technology, that has facilitated privatisation of generation sector, has demonstrated natural gas as a viable replacement generation fuel, more thermal efficient than coal, and less environment-detrimental (through decreased production of harmful emissions). For example, it can almost eliminate emissions of sulphur dioxide, and reduce considerably the emissions of oxides of nitrogen and carbon dioxide. Moreover, natural gas-fired plants produce no sludge or ash, thus averting problems of landfill availability and groundwater contamination. However, the problems associated with the discharge of warm water mixed with effluents back into the sea/lake/river still persist. The rise in temperature at the discharge point will affect the flora and fauna of the receiving waters. In the UK, natural gas has replaced coal as generation fuel to an extent of 30 per cent of the UK’s electricity generation. However, where non-fossil fuels such as hydro and nuclear predominate the generation fuel mix, competition could result in increased use of fossil fuels for generation, if found economically more viable, with the consequent increase in greenhouse gas emissions. For example, Lee and Darani (1995) found that restructuring could lead to substantial rise in nitrogen oxide emissions: a 78 billion kWh increase in generation from existing coal facilities (two-thirds of which are assumed to be replacing generation from lesspolluting plants and one-third of which is assumed to be supplying new demand) could lead to a 500,000 ton increase in nitrogen oxide emissions. Burtraw and Palmer (1996) suggested that the amount of additional generation from existing coal-fired plants stimulated by restructuring could be substantially larger than Lee and Darani’s estimate, depending in part on the rate of growth of transmission capacity.

36

CALIFORNIA’S POWER CRISIS th

On January 16 [2001], the Californian state assembly passed a bill giving the state a central role in the local electricity market. This, in effect, turned the clock back on the deregulation of California’s power industry begun in 1996 amid grand promises of reduced rates for consumers, and bigger markets for power companies. But in fact, the state had few options. On the same day, two of California’s largest utilities had their debts reduced to junk by the leading credit agencies after one of them, Southern California Edison (SCE), announced that it would not be paying $ 596 million due to creditors, in order to “preserve cash”.

That undermined the ability of SCE and of Pacific Gas & Electric (PG & E), the other big utility in the state, to buy power on credit, and pushed them to the brink of bankruptcy. On the same day, a “stage 3” emergency was declared, the highest level of alert, called only when power reserves fall below 1.5 % of demand. On January 17 th, one-hour black-outs rolled round the area of northern California served by PG & E. And Governor Gray Davis declared a state of emergency, authorising the state water department to buy power.

This is a dreadful mess for a state that is held up around the world as a model of innovation and dynamic markets, and that was the first in America to pursue deregulation. - The Economist (January 20, 2001: 57)

The greatest lessons of power sector reforms come from the Californian experience in the US and the Enron experience in India. (The latter we will discuss in another chapter.) The first and one of the most ambitious deregulation plans in the US, the Californian experiment has, however, proved fatal to the interest of the public and the state, and has at last resulted in ‘re-regulation’ by the government, all within five years. The Californian deregulation had in effect replaced a stable price system ensured through government regulation with a ‘free’ market of violently fluctuating prices influenced by a group of profiteering out-of-the-state generators. California with 50,000 MW capacity could have well met a 31,000 MW peak load, but 11,500 MW capacity went out of 37

service due to unplanned outages6 (Reddy 2001). Moreover, California could not add to its installed capacity during the last decade thanks to the toughest environmental laws and the ‘not in my back yard’ (NIMBY) syndrome among its public. The setting was ideal for the unscrupulous profiteers and the ‘contrived’ supply scarcity pushed the prices up without limit. The soaring wholesale prices7 (e.g., reaching $ 1.40 per unit) against a capped retail price (at $ 0.066 per unit) plunged the supply utilities into the red (ibid.). As their credit ratings plummeted, generators refused to sell them power. Losses drove the state’s two main utilities close to collapse. PG&E, the largest utility serving the northern part of the state, filed for bankruptcy protection on April 6, 2001. On April 9, government agreed to buy part of the transmission grid owned by the other major utility, SCE, for $ 2.8 billion (in an effort to restore its financial health through injection of cash). It should be added that cities like Los Angeles with publicly owned utilities that did not opt for deregulation has remained unaffected.

However, the power crisis led to an active role of the FERC in 're-regulation'. In June 2001, it unanimously decided to impose price cap control on wholesale electricity prices in the whole western electricity grid (including California). It should be noted that the FERC had already put partial price caps on California, applicable only during emergencies. The new regulations extended these caps to all periods and throughout the western grid. And it just shows that all is not well with the deregulation bids. NOTES 1

The right to wheeling was first established in 1979 by the US Supreme Court in Otter Tail Power, Inc., vs. United

States. The Court found that Otter Tail could not refuse to wheel federally marketed power to the municipally owned distribution utility in Elbow Lake, Minnesota. 2

‘A spot market is a market where sellers have their commodities on hand and the goods are delivered immediately –

on the spot. Spot markets are distinguished from contract (or forward or futures) markets, where the commodities are not physically present, but, instead, are bought and sold via contracts that specify a delivery date in the future and fix a price. Buying on the spot market is buying instantaneously; on the contract market, it is over the long term. The difference between the two is, for instance, the difference between buying a magazine at a news-stand (spot) or through a subscription (contract).’ (Brennan et al. 1996: 50).

38

3

It should be noted that Argentina’s economic reforms have lately plunged the country into severe crises, leading to

riots. The people have had to pay dearly for the choking debt trap, brought about by the economic maladministration by the successive governments resorting to increasing debt and raising taxes. 4

Particulate matter, besides affecting visibility and exposed surfaces, can create or intensify breathing and heart

problems and lead to cancer. The smaller particles cause the most damage. Sulphur dioxide is a gas that may affect heart and lungs in ways similar to particulates. Moreover, it may damage trees and lead to acid rain, that harms lakes and streams and corrodes exposed materials (e.g., outsides of buildings). It is suspected that sulphuric acid and nitric acids may be produced from the sulphur dioxide and nitrogen oxides present in polluted air. Nitrogen dioxide is a brownish gas with adverse effects similar to those of sulphur dioxide. Besides, it can in the sunlight contribute to the formation of ground-level ozone (or smog), that causes respiratory problems and crop losses. Nitrogen dioxide (NO 2) is produced when nitric oxide (NO) emitted from power plants combines with oxygen in the air. In general, discussions of nitrogen-based air pollution refer to nitric oxide (NO) and nitrogen dioxide (NO 2) together as nitrogen oxides (NOx). Nitrogen oxide emissions presently merit the most immediate policy attention. 5

A combined cycle power plant is one that utilises the combined cycle technique for increasing substantially the

conversion efficiency of using gas for electricity generation. The technique involves first using the gas to fuel a combustion turbine (or gas turbine, which uses the expansions of burnt gases to turn the turbine for power generation), and then recovering the waste (exhaust) heat to generate steam for application to a conventional steam turbine for further power generation. 6

One of the significant factors that led to the crisis was the ban on long term contracts between the supply utilities and

generators (‘forward power’). Paul Joskow notes the consequences: ‘If a generator has a long term contract, then the financial incentive is to generate steady power in order to maximise sales. If you have no contractual promises and there is a new price every day on a spot market, then your incentive is to withhold production to maximise price.’ (quoted in Easterbrook 2001: 44). So it happened in California. Behind the artificial supply scarcity following ‘massive plant outages’, there was an encouraging factor too: the independent power generating companies owned by the out-of-state profiteers were not bound by the same legal obligations that governed the regulated utilities. 7

The electricity spot market price, which was $ 30 per megawatt-hour in 1998, went up as high as $ 1500 on some

days thereafter. It is reported that the phenomenal price rise resulted in some ‘economic absurdities’ also. Some of the big factories having long term low-priced power contracts found that they could make more money now by simply closing down and marketing their unused power allotment! For instance, The Kaiser Aluminium plant in Mead in Washington seized this opportunity and closed down the plant in November 2000, and began to sell the whole electricity, contracted from the Bonneville Power Administration at $ 22.50 per megawatt-hour for the plant consumption, back to the same organisation at $ 555 per megawatt-hour! (ibid.)

39

CHAPTER 9 REFORMS IN THE POWER SECTOR IN INDIA

1. Power Sector in India: Organisation

Now a centenarian, electricity supply industry (ESI) in India showered its first, though feeble, light in 1883 on the historic city of Surat. The first major attempt at urban electrification, however, started only in 1899 in Calcutta when the Calcutta Electric Supply Corporation (CESC) commissioned a large thermal power station. The first hydroelectric plant came into operation in 1902 in Mysore, and some time later the Tata Hydro Electric lighted Bombay. By the time of Independence, ESI had been largely in the hands of small private companies, besides some municipalities and government electricity departments, all confined to some urban centres. The industry was governed by the Indian Electricity Act (IEA), 1910, which was regulatory in character. Soon after Independence, Electricity (Supply) Act (E(S) Act), 1948 was enacted and under the Act, the State Electricity Boards (SEBs) and the Central Electricity Authority (CEA) were created: the SEBs, purported to be autonomous corporate bodies in the public sector, were entrusted with monopoly rights for power development in the States, and the CEA, with the responsibility for the overall policy and co-ordination at the national level for power development. The Authority, inter alia, conducts the techno-economic appraisal of the project reports in respect of setting up of generating stations in the country and issues techno-economic clearance (TEC) for projects. The West Bengal Electricity Board was the first (on 1 – 5 – 1956) and the Kerala State Electricity Board was the second (on 31 – 3- 1957) SEB to be established. The E(S) Act, in fact led to the ‘public sectorisation’ of the Indian ESI which policy was later on formalised by the Industrial Policy Resolution, 1956, that reserved the production of power for the public sector. Thus as a matter of policy, the licenses of the private companies which expired were not renewed, and their businesses were taken over by the SEBs, save five companies, licensed under the IEA, 1910 (in Bombay (2 companies), Calcutta, Ahmedabad, and Surat). Power is placed in the concurrent list of the Indian Constitution, with the States having the primary responsibility for power development in their areas of jurisdiction.

1

In the early 1960s, taking into account the uneven distribution of resources in different States for power development, it was felt imperative to reap the advantages of integrated operation of power systems at the regional level. Thus the country was divided into five regions and Regional Electricity Boards (REBs) were established in 1964, to coordinate the integrated operation of the power system in these regions, and the Regional Load Dispatch Centres (RLDCs), to monitor that of the Regional Grids. The REBs were given statutory status in 1991 through amendment in the E(S) Act, 1948, to strengthen grid management and enforce grid discipline. In November 1996, REBs were given the authority to decide on plant dispatch, that is, to decide which plants should be on line to meet demand, and which should be backed down in case of a fall in demand, on the basis of the merit order operation clause.

The recognition of the inadequate resources availability in the States led to the setting up in the Central power sector of National Thermal Power Corporation (NTPC, 1975), National Hydro-Electric Power Corporation (NHPC, 1975), and North Eastern Electric Power Corporation (NEEPCO, 1976). In addition to these, nuclear power plants are also in the Central sector, under charge of the Atomic Energy Commission and Nuclear Power Corporation under the Department of Atomic Energy.

A national grid that facilitates inter-regional transmission of power was the next step towards a logical conclusion. Section 27A of the Indian Electricity Act, 1910, provides for a Central Transmission Utility (CTU) to undertake transmission of energy through inter-State transmission system and to discharge all functions of planning and co-ordination relating to inter-State transmission system with State Transmission Utilities, Central Government, State Governments, generating companies

etc.

Thus in 1989, the Power Grid Corporation of India Ltd. (PGCIL) came into operation with the Central sector transmission lines as the Central Transmission Utility. The transmission assets of NTPC, NHPC and Neyveli Lignite Corporation (NLC) were transferred to this new entity; in 1994 PGCIL took over the charge of RLDCs also.

2. Regulation

2

Up till the 1980s, the public sector in India held ‘the commanding heights of the economy’. But the turn of the 90s ushered in an unprecedented surge of apostasy, in the ineluctable culmination of an avoidable chain of domestic economic and political developments, smoothly facilitated by a conducive global environment of liberalism. There on a fine morning, India found herself taking ablutions in (the sprinkle from) the mighty waves of reform sweeping across the globe. Till then the governance structure of the Indian electricity supply industry (ESI) was one of the public sector regulation, with the three bundled functions of electricity generation, transmission and distribution (including supply) being regulated by the two Acts – the Indian Electricity Act, 1910 and Electricity (Supply) Act. 1948, together with the amendments, and supported by the Indian Electricity Rules (IER), 1956. The IEA, 1910 provided for the issue of licenses to supply electricity and outlined procedures to regulate the licensees, while the E(S)A, 1948 had as its objective, rationalisation of the power development at the State level through the SEBs and at the national level through the CEA. IER, 1956 laid down technical standards for power supply, construction of T & D lines and safety standards for electrical installations. The principles of financial performance of SEBs contained in Section 59 of the E(S)A, 1948 were amended with effect from June 3, 1978 such that the SEBs were to adjust their tariff from time to time, after taking credit for any subvention from the State Government, in order to ensure that the total revenue shall, after meeting all expenses properly chargeable as the State

Government may

to

revenues,

leave

such

surplus

specify. The amendment enabled State Government, if it deemed

expedient to do so, to notify the SEB as a body corporate; and the desirability of SEBs converting a part of the outstanding loans into equity also was recognised.

3

The World Bank Interprets… India’s pre-1991 planned development strategy helped the country escape from the massive illiteracy, recurrent famines, fertility rates of about seven children per woman, and secular stagnation prevailing before Independence. However, it also was the source of severe financial imbalances which are yet to be corrected. It isolated the country from the rest of the world with the result that from 2 per cent in the 1950s, India’s share of world trade had declined to less than half of one per cent in the late 1980s. It forced Indian consumers to pay higher prices for goods of lower quality and deprived the country from the benefits of foreign direct investment and modern technology. It discouraged production for exports, created recurrent shortages of foreign exchange, and made the balance of payments extremely vulnerable to external circumstances. Most important of all, it held back the country’s growth and thus the pace at which poverty could have been reduced. As argued by India’s own eminent economists, among them Bhagwati (1993), low productivity, rather than inadequate savings, explains the weak growth performance of the past decades. -World Bank 1996: 3

4

Guidelines Laid Down By The High Level (Venkataraman) Committee (i)

The Electricity Boards as statutory bodies intended to play a promotional role in power development will have to subserve the socio-economic policies of the State and therefore cannot view every one of developmental activities exclusively from the point of view of profits or return.

(ii)

Programs like rural electrification may not be profitable looking purely from a commercial angle, but electrical undertakings will have to implement this program in the national interest.

(iii)

Electricity Boards are in effect commercial-cum-service organisations.

(iv)

In the present state of our national economy, it may not be appropriate to expect the SEBs to function with immediate effect on a financial level comparable to similar undertakings in advanced countries. The financial position of SEBs is bound to improve with the development of the country’s economy.

(v)

In public utility concerns too great a burden should not be placed on the consumer by fixing targets so high for replacement of assets as well as for loan redemption.

(vi)

The Electricity Boards should attempt a revenue sufficient to cover (a) operation and maintenance charges, (b) depreciation, (c) general reserve, and (d) interest charges on loan capital.

(vii)

In a number of SEBs, the proportion of construction capital out of total capital outlay is very high and it may not be fair to expect these boards to pay interest charges on the construction capital during the gestation period. They may be allowed to capitalise these interest charges in accordance with the well known commercial principles.

The Committee felt that the net return to be earned by the SEBs after meeting operation and maintenance charges, contribution to depreciation and general reserve, and interest on loan capital might be fixed at 3 per cent of the capital base. This would amount to a return of 11 per cent taking into account the electricity tax/duty levied by State Governments i.e., interest 6 per cent, net return 3 per cent, general reserve ½ per cent, and electricity duty 1 ½ per cent. - As summarised in Government of Kerala 1984: 33 – 34.

3. Performance and the Background for Reforms

5

Installed capacity in the Indian ESI, which was only 1564 MW in 1950-51, increased to 89090 MW in 1997-98, marking an annual compound growth rate of about 9 per cent; and electricity generation increased from 5100 million units (MU) in 1950-51 to 258534 MU in 1997-98, at nearly 9 per cent The Plight Two Decades Back Under the Electricity Supply Act, which regulates the operation of the SEBs, the Boards were not till recently specifically required to earn a return on the capital they use. A number of committees, of which particular mention should be made of the Venkataraman Committee, 1964, examined the working of the SEBs and recommended a gross return of 9.5 per cent (excluding electricity duty) on capital employed after providing for operating expenses and depreciation. However, when the statute was amended in 1978, although it was provided that Boards should earn a positive return, no specific figure was mentioned. In actual practice, however, the Boards are often regarded as promotional agencies to be used to subsidise different classes of consumers and with little or no control over their tariff policy. As a result, on the whole, the returns specified by the Venkataraman Committee have not been realised and on the contrary, large arrears of interest are due to the State Governments on the loans given by them to the SEBs… Besides low tariffs, the causes of the poor financial performance are the low operating efficiencies, high capital cost of projects due to long delays in construction and high overheads – mainly the result of heavy overstaffing. Although precise comparisons are not possible, the average employees per MW of installed capacity in India is 7, compared to 1.2 in the USA, 1.5 in Japan, and 1.7 in the UK. Within the country, the expenditure on salaries varies from 12 per cent to 40 per cent of the total income of the SEBs. Much of this overstaffing is due to SEBs being compelled under political pressures to take on people they do not need. The result of all this is that many of the Boards are wholly dependent upon the State Government even for meeting their normal operating expenses making it even more difficult for them to function as the autonomous bodies which they were set up to be. The weaknesses in the management of the utilities, in particular the SEBs, … arise partly out of the desire of some State Governments to exert a high degree of day to day control on the operations of the Boards, and partly due to management culture, inherited from the bureaucratic style of functioning, that most SEBs had when they were Government Departments. - Report of the (Rajadhyaksha) Committee on Power 1980: 4.

6

growth rate. The per capita consumption of electricity, which was less than 15 units at Independence, rose to 338 units in 1996-97 at about 7 per cent growth rate. Among other growth indicators, the percentage of villages electrified increased from 0.54 in 1950-51 to 86.1 (of the total 579,000) in 1996-97, and irrigation pumpsets energised numbered 11.6 million out of a possible 14.5 million. However, these apparent achievements appear trifle in relation to real requirements. Serious power shortages have been plaguing the country for a long time; at the commencement of the Eighth Plan (1991-92), India faced a peaking shortage of about 19 per cent and energy shortage of about 8 per cent, and the situation remained almost so by the end of the Plan period. This chronic shortage situation has been the inevitable outcome of a cumulative decline in capacity addition in the power sector, explained by the compounded effects of an increasingly inadequate investment tempo and the inordinate, but avoidable, delays in project completion. Investment deficiency and inefficiency have thrived in both the segments of the sector – Central and State (as also private).

The deficiency syndrome has had behind it a long history of abuses and aberrations of public sector management dictums and dictates. The SEBs were required, in line with the letters sans spirit of the Venkataraman Committee Report of 1964, to subserve the socio-economic policies of the State and hence not to view power development exclusively from the perspectives of profits or return, as also not to put a heavy tariff burden on the consumers for purposes of replacement of assets and loan

7

The Poor, Dependent SEBs…. In accordance with the provisions of the E(S) Act, 1948 (till it was amended in 1978), prior to meeting the liability on account of interest charges payable to State Government, the SEBs were required to make specific transfers every year at the prescribed rates to the general reserve and depreciation funds. These ‘internal resources” were supplemented by items such as voluntary loan contributions from consumers, deposits from contractors, security deposits, employee provident fund contributions, etc. However, according to the amendment brought out in 1978 to the E(S) Act, 1948, the Board are now required to meet their interest liabilities prior to transferring funds to depreciation. No revenues are to be transferred to the general reserve fund. These would, in effect, reduce the funds available to them as internal resources…. [I]t is evident that the capital expenditure of the Boards is financed largely by borrowings from the State Governments and other institutional and internal resources represent a relatively small share of this investment and this share has been falling…. As a result, the SEBs have, over the years, become excessively dependent on the State Governments. This is one of the important factors that has led to the dilution of the autonomy of the SEBs. - Report of the (Rajadhyaksha) Committee on Power 1980: 70. redemption. Thus there was no need whatsoever on the part of the SEBs, at least till late 70s, to earn a return on their capital and to contribute internal resources to capacity expansion. This unaccountable indulgence of non-commercial performance easily got embodied into the functional ethics of the SEBs, and they became more and more dependent on the State Governments even for meeting their normal operating expenses at the dearer cost of autonomy. Thus on the one hand, the SEBs remained cash-strapped and on the other, the conventional source of funding (i.e., the Government) unreasonably began to dry up. There was a steady deceleration over time in the Plan provisions to the power sector, leading to cumulative investment deficiency. And then to crown the worst, there descended before the sector an impasse out of the infamous fiscal crisis at the dawn of the 90s.

Confronted with the consequences of the Gulf war in a political flux of frequent changes in Government, India plunged into a deep balance of payments (BoP) crisis in the second half of 1990-91. As India’s credit rating in international capital markets nose-dived, access to external capital borrowing narrowed and substantial capital flight occurred out of the country. In June 1991, 8

despite a severe squeeze on imports, India wavered on the verge of a default on external debt obligations for the first time in her history. However, she survived the crisis, and she emerged unscathed, but with a new flag of ideological allegiance in her hand. She emerged enlightened “that the economy needed substantial reforms if the crisis was to be fully overcome”, and that “both the BoP problems which were building up over t he

past

few years

and

the

persistent

inflationary pressure were the result of large budgetary fiscal deficits which characterised the economy year after year….A reversal of the trend of fiscal expansionism was essential to restore macroeconomic balance in the economy” (Government of India, Economic Survey, 1991-92: 11). The fiscal correction that followed the awakening meant still meagre provisions to an already starving power sector, designed on the premise that further investments required in the sector should be financed from internal resources. A system traditionally attuned to unaccountability and hence functionally sick and financially wreck, the SEBs thus left in the lurch by the Governments to fend for itself, had then only one way open before them – that leading to the private sector. But the domestic private sector remained meek and weak, ergo, the door was to be opened to the global agents. And the siege then started – the siege of power sector reforms!

In addition to the domestic compulsions, the move was facilitated by a most harmonious international environment, exuberant with examples and their emulation elsewhere of ‘big experiments’ with liberalism as in the UK under the Thatcherite privatisation regime, and with deregulation, as in some parts of the USA. The eventful fall of the socialist bloc and the attendant

9

The World Bank Compares…. ……India did not have the inflation, external debt, and social inequities so severe as in Latin America – and was thus able to stabilise the economy more rapidly and at a lower social cost. Unlike former centrally planned economies in Eastern Europe and elsewhere in Asia, India already had an important private sector and all the institutions of a free market economy. India was thus able to avoid the costly industrial and financial closures and restructurings, so frequent and so painful in most of the former Socialist economies of Europe and Central Asia, and which have considerably delayed the supply response to reforms. On the other hand, because India’s macroeconomic crisis was considerably less traumatic than in Latin America, it has been much harder to reach political consensus on the need to reduce fiscal imbalances to the levels achieved for instance by Latin American, East Asian, and Western European countries. And fiscal imbalances remain the single most important threat to India’s long term growth. Similarly, notwithstanding five rounds of trade reforms, India’s trade protection remains among the world’s highest. Likewise and in spite of five years of liberalisation, excessive regulation remains a problem particularly in the financial sector, agriculture and agro-industry. In addition, in the financial sector, the public sector continues to be the major shareholder of India’s largest banks, insurance companies, and contractual savings institutions raising questions on how truly autonomous these institutions can be. Finally, the development of India’s human resources has been slow in comparison with countries in East Asia or Eastern Europe. -World Bank 1996: 5. resurrection of private enterprise drives added to this international enthusiasm. Yes, the whole world appeared in unison to go on a pilgrimage of private sectorisation. (The only powerful exception of the French ESI, still in the hands of the French Government, however, was conspicuous by its exclusion from this picture.)

There has been a universal unanimity in cognising and recognising these causative and promotional strains in the background of (power sector) reforms. However, a significant catalytic element in the whole process has been left unaccounted for in all the reviews of reforms initiatives – the role of the international financial agencies, viz., the World Bank and the IMF, which in fact has been so active that a ‘leftist’ interpretation of the reform process could identify these institutions as the prime mover. Any study of the power sector reforms would be incomplete and biased without a study of this aspect in its proper juxtaposition with others.

10

The soft loan facility, termed Structural Adjustment Loan (SAL), introduced by the World Bank in 1980, in the wake of the Chilean reforms initiative following severe financial crisis, was the forerunner of the reforms-facilitating financial aid. The SALs were designed to provide quickdispersing soft loans to meet BoP crisis over a period of years in return for an agreed set of measures of structural adjustments in the economy. The timing of the provision of assistance was in perfect harmony with the requirements felt pinching across the developing world, as there emerged a global financial crunch that began to haunt each country in the South in vengeful turns. The fundsfamished, but prodigal, Governments in no time seized the easy lease in full endorsement of the pedantic rendition of the wreck in terms of their conventional infatuation with the public sector. This has had an added advantage too – it has also opened up yet another source of finance to the Governments – through the sell out of public sector assets. In the face of such encouraging response to the Pied Piper during the 1980s, the Bank's original emphasis on the BoP gradually faded with a corresponding increase in the stress on ‘economy-wide program of reforms’ (Killick, 1993, p. 69). Later on the Bank devised another facility, more viable and effective one, viz., sectoral adjustment loan that aims at piecemeal reforms processes. Recently (in September 1997, especially in the wake of Haryana’s reform efforts in power sector, vide infra), the Bank started to grant Adaptable Program Loans (APL), involving a series of loans intended to provide phased and sustained support for the loanee’s long-term reform programmes. A number of financial agencies are there now in the market involved in sectoral reforms facilitating loan programs; e.g., IMF set up in 1986 the Structural Adjustment Facility, augmented at the end of 1987 by Enhanced Structural Adjustment Facility, with considerably greater resources.

4. Indian Power Sector on the Reform Path

As already explained, the capacity-deficient Indian ESI had the rude shock when confronted with the fiscal crisis begotten revelation that the conventional budgetary funds support for capital augmentation programs had dried up. The ill-ridden performance of the ESI had already left it penniless and penurious. The prospects of international aid also appeared dim and grim. The World Bank had (in 1989) stated that requests from ESIs in developing countries added up to $100 billion per year against an availability of only about $20 billion from multilateral sources (quoted in D’Sa, et al. 1999). The predicament thus posed had also its ready-made solution prominently decked on 11

its cap – the private sector. But the Indian capital market, remaining in some infantilism, was too feeble and frail to support the sector and hence, the significance of the foreign sector. It was also hoped that there would be a side-benefit in respect of efficiency which remained at an unacceptably low level. This efficiency was thought to be improved through the oft-claimed better management and higher technical performance of the private sector.

Key Features of the Power Sector Reform Policy Introduced Starting 1991  Private sector companies may build, own, and operate generating stations of any size and type (except nuclear).  Foreign equity is permitted in generation companies.  A post-tax return on equity of 16 per cent at a plant load factor (PLF) of 68.5 per cent is guaranteed, based on a two-part tariff formula, which covers both fixed and variable costs.  Additional returns (of 10 to 12 percentage points) on equity allowed where the PLF exceeds 68.5 per cent.  Free repatriation of dividends and of interest on foreign equity and loans.  A five-year tax holiday for new generation and distribution companies.  Protection from exchange rate fluctuations.  Depreciation rates on plant and machinery have been increased.  Custom duty on imports of equipment has been reduced by 20 per cent.  A private power generator can sell power to anyone with the permission of the concerned State Government. - World Bank 1995: 83; Box 3.2. Private Sector Participation in Generation

The generation sector of the vertically integrated natural monopoly of ESI had become increasingly recognised as having the potential to accommodate competition and thus it was the natural starting point for introducing private participation, both domestic and foreign. (It should be noted that private sector participation in the power sector has been allowed since long, but they had accounted for only an insignificant share in the sector. For example, the installed capacity in the private sector in 1993-94 was only about 4 per cent of the total (and in 1997-98, about 6 per cent)). This was accomplished through the October 1991 amendment to 1910 and 1948 Acts that for the first time introduced the concept of generating company as a distinct entity (The Electricity Law 12

(Amendment) Act, 1991) – the first amendment with structural implications for the ESI in India. Under this Amendment Act, private companies can now build, own, and operate power stations subject to certain terms and conditions detailed subsequently in the Notification of March 13, 1992, from the Ministry of Power and Non-Conventional Energy Sources, Department of Power. The independent power producers (IPPs) were expected to negotiate power purchase agreements (PPAs), with the concerned SEBs, that would reflect those terms and conditions. In addition to an Investment Promotion Cell (IPC), a High Powered Board (under the Chairmanship of the Minister of Power) was also set up to facilitate project implementation by serving as “a single point forum for faster clearance of the proposals received within a definite time frame” (quoted in World Bank 1995: 84)

The IPP entry on the basis of negotiation (memorandum of understandings, MoU) with the tariff determined on cost-plus formula had the inherently inevitable rate padding tendency that led to higher tariffs. This belated (and hence costly) realisation resulted in the January 1995 policy that provided for entry of IPPs on the basis of competitive bids only, administered by the States (for their purchases) and by the Centre (for mega projects, for example). Guidelines were issued by the Ministry of Power for such competitive bidding processes. In October 1995, the scope for private investment was further enlarged inviting private sector participation in the renovation and modernisation (R&M) and life extension (LE) of existing power plants. (The Phase – I R&M program had in fact been launched way back in 1984 for 34 thermal power stations in the country, at a cost of Rs. 500 crores that had yielded an additional generation of about 10, 000 MU, against a target of 7000 MU, by 1991-92.) About 117 thermal units of about 11000 MW capacity (out of the total thermal capacity of about 59000 MW), that have been in operation for more than 20 years, are now estimated to require LE works; a capacity of about 1200 MW, under long shut down, can also be brought under LE programme. Similarly, 35 hydro-power stations, in operation for over 30 years in excess of their useful operating life, are also identified to require major R & M works which are cost-effective, environmental friendly and require shorter lead time. R&M and LE yield additional capacity at a cost of only 15 to 25 per cent of the cost of equivalent new capacity (Government of India, 2000, p. 19).

13

In addition, a number of measures were taken in quick succession intended to facilitate capacity expansion programs in the power sector. The MoP suggested that the States encourage entry of captive power units into the system, by offering private investors an appropriate tariff for the purchase of surplus power by the Grid and third party access (TPA) for direct sale of power to the other industrial units. In addition, to avoid delays in the installation of small private power plants (PPPs), the threshold for requiring CEA’s TEC of individual proposals was raised in December 1995 from Rs. 1 billion to Rs. 4 billion for generating stations to be developed through a process of competitive bidding. In February 1996, the MoP advocated the use of barge-mounted power plants as a possible option for coastal States, again aimed at encouraging private investment.

The Government has recently reviewed the policy on automatic approval of foreign equity participation in power sector (both in generation and in T&D), and revised the limit from 74 per cent to 100 per cent of equity participation in cases where project cost does not exceed Rs. 15000 million. Again, for speedy environmental clearance, the Ministry of Environment and Forests has agreed to delegate powers to the States regarding environmental clearances for cogeneration projects and captive plants up to 250 MW, coal-based plants using fluidised technology up to 500 MW, power plants on conventional technology up to 250 MW, and gas/naphtha-based plants up to 500 MW. In November 1998, the limit in respect of various categories of power projects beyond which the concurrence of the CEA would be required was enhanced. The 1991 Policy had envisaged that not more than 40 per cent of the total outlay for the private sector units might be raised from Indian public financial institutions. The Government has recently removed the ceiling for the extent of domestic debt, subject to the adoption of a norm by Indian public financial institutions whereby a higher domestic debt component would be allowed for projects based on indigenous equipment.

During 1998-99, the Central Government announced a Policy on hydro-Power Development with a view to exploiting at a faster rate the vast hydro-power potential available in the country. Subsequently guidelines were issued that simplify the procedure for TEC by the CEA, reducing the normative availability factor for hydro-power stations from 90 per cent to 85 per cent, and allowing the sale rate of secondary energy at the same rate as applicable for primary energy in order to provide an additional incentive for attracting investment in hydro projects.

14

For the purpose of financing the power sector, new arrangements have also been made. These include setting up of the Infrastructure Development Finance Company IDFC), broadening the scope of the public sector Power Finance Corporation (PFC), allowing an active role for the PFC in negotiating loans from international banks and foreign capital markets, constitution of a Power Development Fund by the power ministry for speedy implementation and execution of power projects as also to finance feasibility studies for setting up power plants, mooting a Power Trading Company (PTC) to purchase power from power-surplus regions and sell it to power-deficient regions, launching of ‘Infrastructure Bonds’ to channel household savings into the power sector, and involving provident funds as a potentially important source of funding.

Mega Power Plants

The initial guidelines for setting up mega power projects (MPPs), defined then as those projects of 1500 MW capacity or more and supplying power to more than one State, were issued in November 1995. The policy was reformulated in October 1998 as applicable to the construction and operation of hydro-electric power plants of at least 500 MW and thermal plants of at least 1000 MW under duty-free incentives. A Standing Independent Group (SIG) was constituted by the Government in November 1997 as the apex body to establish the parameters for the negotiation of MPPs and to oversee their implementation. Such MPPs, supplying power to more than one State, are proposed to be set up in the public as well as private sectors. Hence it was considered necessary to develop a single power purchasing entity; thus a Power Trading Corporation (PTC) was also formulated, with majority equity participation by PGCIL along with NTPC, Power Finance Corporation (PFC), and other financial institutions. PTC is secured financially by a letter of credit and recourse to the States’ share of Central Plan allocation. The large IPPs can sell energy directly to a ‘cluster’ of large consumers or to the PTC, rather than to the cash-strapped SEBs at a risk of possible defaults. Considerations of tariff orders are required to be made as far as possible on the basis of competitive bidding. In order to ensure that domestic bidders are not affected adversely, a price preference of 15 per cent is given to the projects under the public sector, and deemed export benefits under the exim policy are given to domestic bidders from both the public and private sectors. The additional benefits accruing to the MPPs are: i) the import of capital equipment for these projects are free of customs duty; ii) the income tax holiday regime would continue for a block 15

of 10 years within the first 15 years; and iii) the State Governments have to exempt supplies to MPPs from sales tax and local levies.

Representatives of SEBs oppose the idea of the mega projects bypassing the SEBs and attracting large customers. The IPPs feel that this policy would be a hindrance to smaller projects, and prefer that the concessions extended to mega projects be extended to all IPPs.

Liquid-Fuel-based Power Projects

The significance of liquid fuel-based power plants lies in their being an economically viable option for quick capacity addition in view of the long gestation period for coal- or lignite-based power projects as well as hydro-electric projects. Combustion-turbine-based combined cycle plants using liquid/gas fuels (such as heavy petroleum stock (HPS), low sulphur heavy stock (LSHS), heavy furnace oil (HFO), and natural gas, but not high speed diesel (HSD)), were allowed to be set up in areas where transportation of coal is either costly or infeasible.

The policy of using naphtha for power generation being both uneconomical and shortsighted, the Commerce Ministry had (in August 1996) refused licences for its import for power plants. Later on, however, new guidelines for fuel linkages were formulated in December 1996, including the provision for permitting the use of naphtha based on considerations of the concerned State’s peaking power shortage. Further modifications to the liquid fuel policy were made in view of the significance of the role of non-traditional fuels such as condensates and orimulsion in new projects. Open Generalised Licence (OGL) facility has been extended to these traditional fuels also; LSHS is already on OGL.

It was decided in July 1998 that the existing ceiling of 12000 MW (allocated to various States) would need to apply only to naphtha; this would be in addition to the fuel oil/low sulphur heavy stock (FO/LSHS) linkages already given. The States are therefore now free to contract for new projects based on FO beyond the existing linkage, on the condition that such FO-based power plants are to use integrated gasification combined cycle (IGCC) technology or any other technology that would confine sulphur emissions within the stipulated levels. 16

Transmission Sector

Though a number of amendments were added to the 1910 and 1948 Acts, they were of only clarificatory nature. In contrast, the first Amendment Act (The Electricity Law (Amendment) act, 1991) with structural implications for the ESI in India was designed in 1991 to introduce the concept of generating company as a distinct entity. In 1998, similarly, the Electricity (Amendment) Act, 1998 treated transmission as a separate entity of business which could be licensed and which could thus facilitate private participation. In this new light, the Power Grid Corporation of India Ltd. (PGCIL) that owns and operates the Central sector transmission assets was notified as the Central Transmission Undertaking (CTU). This enabled it to gain explicit legal control of the grid management system in the country. Similar entities at the State level were notified as State Transmission Undertakings (STUs).Among the main functions of these undertakings are identification of transmission lines, issue of specifications and selection of private party ready to participate in the sector. Private sector participation in transmission sector investment is proposed to be limited to BOOT (build-own-operate-transfer) basis projects under the supervision and control of the PGCIL. Private sector entry is welcomed through two routes: (i) cent per cent equity, or (ii) joint venture with the PGCIL. In the latter case, 26 per cent stake would be held by the PGCIL and the remaining by the private partner or a consortium. The PGCIL would be the authority to identify such projects at national and regional levels as well as to select private investors and to recommend them to the Central Electricity Regulatory Commission (CERC) for issuance of licences; at the State level, STUs would be the corresponding authority to recommend to the State Electricity Regulatory Commission (SERC) (vide infra). The licences thus issued would be for a period of 30 years for the private investor’s BOOT project. The PGCIL, if it is so prepared, can take the line on rental basis; in the case of joint venture, the tariff would be based on cost-plus basis. It was also decided that four transmission links, viz., Meramundali – Jeypore, Madurai – Edamon – Thiruvananthapuram, Purnea – Muzaffarpur, and Muzaffarpur – Gorakpur links, would be given as test cases to the private sector.

Electricity Regulatory Commissions

17

It was unequivocally emphasised that the future development of the ESI in India depended on two factors: (i) improved operating efficiency of the SEBs, and (ii) their financial viability. To achieve this objective, it was then proposed and decided that the ESI be restructured and unbundled wherever possible for effective private participation, assumed to usher in competition and efficiency. Since private power sector presupposes regulation, it was further acknowledged that unbundling could not effectively take place unless regulators were appointed first.

Thus the Indian Electricity Act of 1910 and the electricity (Supply) Act of 1948 were amended in 1996 to enable the setting up of State and Central level electricity regulatory commissions. Each State and Union Territory was to set up an independent State Electricity Regulatory Commission (SERC) to deal with tariff fixation, that is, to determine the tariff for wholesale or retail sale of electricity and for the use of transmission facilities. Later on the GOI issued an ordinance which was later converted into an act in 1998 (The Electricity Regulatory Commission (ERC) Act, 1998), to enable the appointment of regulators at the national and state level. At the Centre, a Central Electricity Regulatory Commission (CERC) was set up (on 24 July 1998) to deal with all state-level appeals and inter-state power flows. Such commissions had already been set up in Orissa in 1996 and in Haryana in 1998 under state legislation. With the concurrence of the GOI, Andhra Pradesh passed a separate Regulatory and Restructuring Act in 1999, in line with the Orissa and Haryana acts. Due to the federal nature of our Constitution, the central government had decided that though it would pass an Electricity Regulatory Commission Act, it would not impose a restructuring model on any state by central legislation, and that it would only issue guidelines and model acts for the consideration of the states.

Since April 1, 1999, The CEA has entrusted the CERC with the task of regulating power tariffs of central government power utilities, inter-state generating companies, and inter-state transmission tariffs. One of the important objectives of CERC is to improve operations in the power sector, by means of measures such as increased efficiency, large investments in the T & D systems, time-of-day pricing, and power flow from surplus to deficit regions. Further, the central government or the CERC can grant a transmission license to anyone to construct, maintain, and operate any inter-state transmission system under the direction, control, and supervision of the central transmission utility. 18

Till the ERC Act came into effect, the provision for the norms for tariff fixation by the Central Government in respect of the electricity sale by a generating company to the SEB, lay in Section 43 (A) (2) of the Electricity (Supply) Act, 1948. Once SERCs are constituted in States, this provision need to be disapplied by the Central Government (under Section 51 of the ERC Act, 1998). However, this provision can be omitted only when the SERC have come up with the required and relevant tariff regulations, lest there be a void in the matters of norms for tariff fixation. The Central Government has omitted this provision of E(S) Act, 1948, in the case of Orissa and Haryana, whose regulatory commissions have assumed tariff fixation powers; recently APERC also has assumed such powers.

Setting up of SERCs has also become one of the preconditions for the beneficiary States under the revised Mega Power Policy.

19

Functions of the Central Electricity Regulatory Commission Some of the important functions of the CERC are:  Regulate the tariff of generating companies owned or controlled by the Central Government.  Regulate the tariff of generating companies other than those controlled by the Central Government, if such generating companies enter into, or otherwise have a composite scheme for, generation and sale of electricity in more than one State.  Frame guidelines in matters related to electricity tariff.  Aid and advise the Government in the formulation of a tariff policy which shall be fair to the consumers and facilitate mobilisation of resources.  Regulate the inter-State transmission of energy including tariff of transmission lines.  Arbitrate or adjudicate upon disputes involving such generating companies or transmission undertakings as indicated above,  Promote competition, efficiency, and economy in the activities of the ESI.  Associate with the environmental regulatory agencies to develop appropriate policies and procedures for environmental regulation. Section 17 (1) of the Electricity Regulatory Commission Act empowers the State Government to set up SERC, whose main functions are:  Determine and regulate the tariff for electricity, wholesale, bulk, grid, or retail.  Determine the tariff payable for use of the transmission facilities.  Regulate power purchase and procurement process of transmission undertakings and distribution undertakings.  Promote competition, efficiency, and economy in the activities of the ESI. - Government of India 1998 c. A number of SEBs are on the reform/restructuring path. So far sixteen States (Andhra Pradesh, Arunachal Pradesh, Delhi, Goa, Gujarat, Haryana, Karnataka, Kerala, Madhya Pradesh, Maharashtra, Orissa, Punjab, Rajasthan, Tamil Nadu, Uttar Pradesh and West Bengal,) have either constituted or notified the constitution of SERC (see the Appendix to this Chapter). The SEBs of Orissa, Haryana, Andhra Pradesh, Karnataka, and Uttar Pradesh have already been unbundled/corporatised. The first move towards such reform process was initiated in Orissa, even before the formulation of the CERC at the Centre. Orissa Electricity Regulatory Commission was the first of its kind in the country, designed as an independent regulatory commission to regulate the power sector in the State. Again Orissa is the only State to have fully privatised the distribution 20

business of its ESI in the State. The World Bank has sanctioned a loan of 350 million dollars to Orissa for its power sector reforms.

Power Sector Restructuring in Orissa Orissa is the first State in India to have started restructuring of the power sector (in 1996), through the enactment of the Orissa Reforms Act, 1995. The erstwhile vertically integrated utility of the Orissa SEB was unbundled into separate corporations – Grid Corporation of Orissa (GRIDCO) for transmission and distribution, and Orissa Hydro Power Corporation (OHPC) for hydel generation. Subsequently, four wholly owned subsidiary companies of GRIDCO were carved out for distribution, and later on these subsidiary companies were privatized by the sale of 51 per cent of the share of GRIDCO’s equity holding. BSES took over three companies (NESCO, WESCO, and SOUTHCO) in the north, west and south zones, and AES Corporation of USA, the central zone (CESCO). The Orissa Power Generation Corporation (OPGC) has been disinvested to the extent of 49 per cent. The transition process involved valuation, apportioning and adjustments of assets and liabilities. Adjustment of subsidies and electricity charges, totalling Rs. 340 crores, payable to OSEB/GRIDCO against the upvalued amount of Rs. 1194 crores, cast a heavy strain on the finances of GRIDCO. Moreover, a major proportion of past losses and overdue liabilities were retained by GRIDCO with a view to successful privatisation of the distribution companies. The four distribution companies were assigned only the project related liabilities totalling Rs. 630 crores, while GRIDCO retained liabilities totalling Rs. 1950 crores. In addition, GRIDCO issued Rs. 253 crore worth of shares and Rs. 400 crore worth zero coupon bonds to the State Government. All these left GRIDCO heavily cash-strapped and forced to default to generating companies and other suppliers (Government of India 2000: 9). Hence the significance of the financial assistance from the institutional lenders, especially the World Bank through its Adaptable Program Loans. In fact, Orissa has been implementing the model of power sector restructuring as conceived by the World Bank (Dixit, Sant and Weigle 1998). The Haryana SEB was unbundled into two separate entities on 14 August 1998 – the Haryana Power Generation Corporation for generation, and the Haryana Vidyut Prasaran Nigam for transmission. For distribution, the State has been divided into two zones viz., north (managed by the Uttar Vidyut Vitaran Nigam) and south (managed by the Dakshin Vidyut Vitaran Nigam).

21

Andhra Pradesh has received Presidential assent to its Electricity Reform Bill, 1998, which has led to the formulation of the SERC. Under the provisions of the Andhra Pradesh Electricity Reforms Act, two corporations, viz., the Andhra Pradesh Power Generation Corporation Limited The World Bank Assistance to Haryana The World Bank has agreed to support Haryana’s efforts to the extent of US $ 600 million over a period of eight to ten years through a series of Adaptable Program Loans (APL), a new lending instrument approved in September 1997. The new instrument involves a series of loans through which the Bank provides phased and sustained support for a borrower’s long-term reform program. On January 15, 1998, an initial APL of US $ 60 million was approved to finance critically needed investments, enhance the credibility of Haryana’s reform agenda and demonstrate that “something is happening in the power sector”. Subsequent APLs will be processed as Haryana meets the milestones in its reform agenda, based on progress in implementing the investment program and the financing requirements. The APL approach has allowed the Bank to start its support to Haryana at an early stage of the reform process and to express a long-term commitment to support reforms. The Haryana Program Approach will provide much more flexibility to adapt the Bank’s assistance to evolving conditions. UKDFID and USAID have decided to provide technical assistance for the reform program, and several other aid agencies such as OECF, KfW, and CIDA (Canada) also have expressed interest. A financially sound distribution system is expected to attract more private generation. -World Bank 1998 b: 21; Box 2. (AP GENCO) and the Transmission Corporation of Andhra Pradesh Limited (AP TRANCO) have replaced the Andhra Pradesh SEB. It has been decided that distribution in the State will be divided into five zones, and 51 per cent of the stake in distribution will be offered to the private sector. It should be remembered that the AP Electricity Regulatory Commission has recently hiked the tariff to such an extent that the whole State has been paralysed for many days together with public agitation. As per a report, the revised power cost in AP is estimated to be about 16 cents, compared with a world average power cost of only seven cents (The Hindu Business Line, 29 July 2000).

Karnataka is the first State in India to have separated generation of power from transmission and distribution by setting up the Karnataka Power Corporation Limited (KPCL) as far back as in 1970. Transmission business of the Karnataka SEB has recently been corporatised (Karnataka Power Transmission Corporation Limited, KPTCL), and Karnataka proposes to incorporate 22

distribution companies by the end of 2000 and to privatise them by December 2001. Under the current proposal, already cleared by the State Government, 51 per cent of the equity is to be provided to the private sector promoters/bidders, and the remaining 49 per cent, to other intending equity holders including financial institutions, with a small stake for the KPTCL. KPCL has recently decided to insure its assets as a prelude to restructuring and the subsequent disinvestment (either splitting of KPCL into separate thermal and hydel power generation companies or converting KPCL into a single State-owned holding company with equity stakes in both these ventures separately) – with this move, KPCL will be the first public sector company in the country to insure its assets.

The Uttar Pradesh Government has also a plan to privatise distribution in Kanpur as a test case, as it has already done in NOIDA. The Government of Kerala has set up profit centres for generation, transmission and distribution, and opened up the generation sector to IPPs. The Maharashtra Government has recently decided to trifurcate MSEB into three companies for power generation, transmission and distribution; the Government has also announced its “no privatisation” decision in a bid to placate the MSEB employees who went on strike against the trifurcation decision.

Rajasthan was one of the first States to take up restructuring initiation. Rajasthan SEB was proposed to be corporatised to become a bulk purchaser of power and a transmission company. The model envisaged that distribution function would be progressively privatised. Initially two zones were suggested to be given over to private management through joint ventures. However, Rajasthan now seems to have gone back on most of these proposals.

Accelerated Power Development Programme

A new Central assistance scheme, viz., Accelerated Power Development Programme (APDP), has come into force for leveraging reforms in the power sector in the States. APDP will finance projects relating to: (i) Renovation & Modernization / Life Extension / Uprating of old power plants (thermal and hydel); and (ii) Upgradation of sub-transmission & distribution network (below 33 KV or 66 KV) including energy accounting & metering. Priority is given to projects from those States who commit themselves to a time bound programme of reforms in terms of 23

(i) setting up State Electricity Regulatory Commission (SERC) and making it operational as envisaged under the law, and the State power utilities sending the first proposal for fixation of tariff to the SERC. (ii) creating separate profit centres / restructuring generation/transmission/distribution to make the system accountable; dividing the state into a number of zones for the purpose of distribution and privatisation of each zone or alternatively giving responsibility of electricity distribution to Panchayats/Users' Association/Co-operatives/Franchisees, in case it is found that improvement in public sector management is not feasible. (iii) completing cent per cent of metering in a planned manner. It is also stipulated that APDP funds shall also be available to the States which otherwise achieve high level of operational efficiency and financial viability. The fund under APDP is scheme specific, provided to the State Government as a special Central Assistance over and above the normal Central Plan Allocation. The State Government should release this fund to the State Power utility under the same terms and conditions as they receive from the Central Government, within a week of the said amount being credited to the State Government account and send confirmation to the GOI; otherwise it is treated as diversion of fund.

No project will receive assistance both under Accelerated Power Development Programme (APDP) and Accelerated Generation & Supply Programme (AG&SP) of the Power Finance Corporation (PFC). Energy audit, accounting and system studies, however, can be financed through AG&SP under the Model Distribution Scheme. Renovation and modernisation/life extension projects costing less than Rs.100 crores will be financed under APDP, and those costing more than Rs.100 crores will be financed under AG&SP of PFC. It is also provided that cent per cent metering only within the identified distribution circles will be financed under APDP. Metering upto 11 kV out-going feeders & high tension (HT) consumers and cent per cent metering outside the distribution circles will be financed by PFC under AG&SP.

Electricity Bill, 2000

24

Recently, the Central Government has introduced a new Act, proposed to replace the existing three Acts, that govern the power sector. The three Acts are the Indian Electricity Act, 1910, Indian Electricity (Supply) Act, 1948, and the recently enacted Electricity Regulatory Commissions Act, 1998, that together constitute the legal foundation for the ESI in India at present. Electricity Bill, 2000 will now replace all these three Acts.

A very significant provision in the Bill is that all the SEBs of present constitution will ‘wither away’ within six months of the new Act coming into effect. This has since then led to a very heated controversy. However, it is open to a State Government to set up its own SEB if it wants the existing

25

The Important Milestones in the Electricity Bill, 2000        







 

    

State Regulatory Commissions shall be established in all States within three months of the new law coming into force. All activities in the electricity sector, including captive generation, shall be based on authorisation given by Regulatory Commissions. Corporatised State transmission utilities shall be established within 120 days of the coming into force of the new Act. All SEBs as at present constituted shall cease to exist within six months of the new Act coming into force. Existing approvals, licences, etc. will be valid for a period of one year from the date of enforcement of the new Act. Area distributors shall assume responsibilities of distribution in respect of areas within one year of the coming into force of the new Act. Pooling arrangements to facilitate establishment of a spot market for electricity to come into force from the anniversary of the new Act. All applications for authorisation shall be disposed of by the Regulatory Commission within 120 days of receipt of the application after following the due process. On expiry of this time limit, the authorisation shall be deemed to be granted. Projects now no longer require techno-economic clearance by CEA; but CEA shall make its recommendations to the competent commission within 30 days of receipt of a copy of the application for authorisation. Powers are conferred on Regulatory Commissions to enforce the terms and conditions of authorisation as well as to revoke the authorisation at any time after following the due process to pre-empt any monopoly of the authorised persons. The tariff orders are to be issued by Commissions within 120 days of receipt of the application, failing which the applications shall be deemed to have been approved. Tariff could be left to be determined by market forces at appropriate time in future through Government direction . Legitimising deposit of security with area distributor before commencement of supply and discontinuance of supply if the security is insufficient; exemption from deposit of security is allowed in case of pre-paid meters. Provision is made for discontinuance of supply (including bulk supply) on negligence to pay for electricity, after giving seven days’ notice. Requirement of consent of State Governments is done away with for interState flow of power. Strict enforcement of metering requirements, as stipulated by the CEA. Provisions are also made for pre-paid metering. Stipulation for standards of performance for supply and strict enforcement of the standards. Strict measures against anti-competition practices. - Economic and Political Weekly, May 6, 2000: 1594. 26

system to continue. In fact, once the E(S) Act, 1948 is done away with, the very existence of SEBs, the establishment of which was the objective of the Act, come to natural end. Now the State Government needs to reconstitute the SEB, if it so desires, through its own legal provision. At the same time, on the other side appears the fact that once the SEBs cease to exist following restructuring, then the very E(S) Act of 1948 becomes redundant. Hence the significance of the replacement.

The Bill envisages time-bound radical restructuring in terms of unbundling and corporatisation. All the States have to establish State Regulatory Commissions, authorised to supervise, direct and control all the activities in the ESI. This implies that the Government interference in the day to day affairs of the sector is minimised, though the Government is still allowed to wield significant powers. The Bill also seeks to establish spot market for electricity through pooling arrangements. This necessitates setting up of an independent system operator (ISO) for transmission such that the ‘wires business’ becomes one of national dimension rather than of inter-State dimension. There is a threat, however, lurking in such development in that the concurrent, federal, nature of authority of the States on the ESI may soon be superseded by centralised, unitary, authority.

The Bill also plans, in a bid to facilitate a level playing field for transmission sector participants (transmitters), to restrict the role of the Central Transmission Utility – PGCIL – to that of power grid management only, divesting it of the other role of being also a player in the transmission business. This is in the wake of a long-standing feud between PGCIL and the CERC over a directive issued by the CERC to the Power Grid in 1999 to operate its business and perform its role as a grid manager as separate autonomous business units. The earlier version of the Electricity Bill vested the regulatory control in the transmission sector with the PGCIL; however, the recent version (sixth draft) has restored (in line with the Electricity Amendment Act, 1998) the power to the CERC with some minor modifications.

27

A major criticism levelled against the Bill has been that there is not enough emphasis on rural electrification in the Bill and that the Bill actually over-emphasises commercialisation of power rather than making available power to everybody.

The Bill as such has caused much flutter and protest; many States (for example, Kerala) and SEB employees suspect the Central move as an attempt to usurp the State’s authority on the ESI, and impose restructuring where the State is unwilling. The Bill has fallen in a debate-revision circle and now the sixth draft of the Bill is expected to be placed on table in the Parliament in this monsoon session itself.

5. Responses to the Opening Up Policy

Quite contrary to the confident expectation in 1991-92, however, the private sector has not come forward to contribute sufficiently to bridging the gap between power demand and supply. Although the CEA has provided techno-economic clearance (TEC) to nearly 30000 MW and inprinciple clearance to another 9578 MW, only 5370 MW has been commissioned during the past ten years (as in January 2001, as shown below). It should be stressed that a whole decade has gone to

Status of Private Power Projects (PPPs) (since 1991; as in January 2001)

Number Capacity (MW)

28





  

Projects techno-economically cleared by the CEA Thermal Hydro Total Projects with in-principle clearance by the CEA Thermal Hydro Total PPPs commissioned so far PPPs under construction Detailed project reports under examination by the CEA Thermal Hydro Total

52 5 57

27859.5 1516 29375.5

16 4 20 25* 17*

7208.9 2369 9577.9 5369.75 5149

8 1 9

2553.58 70 2623.58

Note: * = including those projects which do not require the techno-economic clearance of CEA and also including licensees. Source: http://powermin.nic.in/ waste by waiting for private sector participation in power capacity addition programme: the installed capacity (IC) during the last decade grew at an annual average rate of 4.5 per cent only, while the growth rate in the 1970s as well as in the 1980s was 7.5 per cent and 8.2 per cent respectively. As in March 2001, the private sector accounted for just nearly 10 per cent (9926.65 MW) of the total IC in the Indian power sector (101657 MW), while state sector for 60 per cent (60860 MW).

A host of problems are there confronting the IPPs and dissuading them from coming forward. Some of them are as follows:

(i)

Litigation/re-negotiation leading to delays

Several reasons, including high costs, environmental impacts, perception of financial irregularities, etc., have raised voices of protests against some power plants IPPs have shown interest in. Litigation and re-negotiation of Power Purchase Agreements (PPAs) have caused long delays in project completion resulting in unwarranted time and cost overruns and in backing out of some of the IPPs. The recent Cogentrix episode in Karnataka and the Enron controversy in Maharashtra are the fit cases in point here. 29

While on the one side the Government is talking of unveiling second generation reforms in the power sector, a couple of foreign IPPs, on the other hand, have already decided to pull out from various power projects, indicating that even the first generation reforms continue to be plagued with restrictive policies. Cogentrix in November 1999 and Electricite de France in July 2000 announced their decision to move out of the two counterguaranteed projects, Mangalore Power Project (in Karnataka) and Bhadrawati Power Project (in Maharashtra) respectively. Reports suggest that major foreign IPPs like National Power, ABB, PowerGen, and Siemens are either slowing down activities or reducing their presence in India. Most of these foreign IPPs, frustrated by restrictive policies, procedures and litigations leading to delays, have now found much greener pastures in the US and Europe, as the markets there have started invitingly booming (Financial Express, July 12, 2000).

(ii)

Financing problems

In fact, the low returns and low payback periods involved in the infrastructure projects has traditionally confined these projects almost exclusively in the public sector, even though the advent of the gas-turbine projects of smaller sizes and shorter lead times has improved profitability perception. Despite the Government’s offer of high returns on independent power projects, they still involve considerable delays in planning and execution. IPPs also face difficulties in obtaining necessary funds, as the financial institutions are unlikely to agree to loans with a maturity of periods longer than 3 years, to match the tenor of their deposit liabilities.

Moreover, the financial sickness of the SEBs has been largely identified as the primary reason that inhibits private sector participation in the power sector. The SEBs being the sole purchaser of the power that the IPPs generate, the latter naturally hesitate to take the risk of not being paid by the financially poor SEBs. Hence they have sought for counter-guarantees

30

from the Central Government, which have subsequently been sanctioned to some of the fasttrack projects.

The Impediments to Private Investment The fundamental obstacle to private sector investment in the power sector is the weak financial position of the SEBs……. At this time, the SEBs are not financially viable clients for potential private power producers. The SEBs generally are prevented by their respective State Government from charging commercially viable tariffs; they are not allowed to cut power from non-paying customs, and many of them are institutionally too weak to contain power theft in rural and urban areas. As a result, the SEBs’ financial condition is one of the country’s most serious structural constraints to the reduction of the public sector deficit. Other impediments to private investment include lack of clarity in the private power policy (though some of the issues have been addressed in the amendments), fuel supply risks emanating from Government monopoly control over coal and railway transport, lack of legally enforceable fuel supply and transportation contracts, weak financial situation of the State Governments to backstop SEB’s obligations, and lack of clarity in CEA’s review and approval process. -World Bank 1996: 101.

One of the two alternatives considered and sanctioned in this respect has been the right of an IPP to sell power directly to any customer using the monopoly network services at a certain wheeling charge (right to third party access).

The other alternative has been an escrow account into which the concerned SEB’s revenues are deposited and to which the IPP would have first-access in case of a default by the SEB.

It is recommended that the amount in the escrow account should be 9.25 times the monthly tariff payable by the SEBs; the escrow account should be charged exclusively in favour of an IPP with a provision to assign the same to the lenders of the IPP; and the escrow account should be established before financial closure (Mukherjee 1998). But the problems still remain. Financial institutions are reportedly tending to limit their loans to IPPs to the order of the SEB’s ‘escrowable’ capacity. But several States, on the other hand, have already signed a large number of power purchase agreements (PPAs) with an aggregate capacity 31

higher than could be supported by escrows. “….independent estimates suggest that the country as a whole only has about 8000 MW of escrowable capacity” (World Bank 1998 b: 22).

(Recently, the Ministry of Power has announced that a proposal is under its consideration to accord sovereign guarantees for select large capacity power generation projects that do not enjoy a State guarantee. The guarantee will be given to PTC, which will be purchasing power from the project and selling it onwards to the SEBs. – The Hindu Business Line, July 27, 2000)

According to the current (1998) guidelines, the following criteria must be met when IPPs obtain funds for a project:

(a) the promoter’s share in a private sector power project must be at least 11 per cent of the total outlay;

(b) the company is allowed a debt-equity ratio of 4:1; (c) up to 40 per cent of the total outlay can be raised from Indian financial institutions and banks; and

(d) no single FI/bank can lend more than 25 per cent of its net worth to an individual company or project, nor more than 15 per cent of its total outstanding loan and guarantee portfolio to a single industry.

Furthermore, the IPPs have to wait for other arrangements such as fuel supply agreements to be finalised.

(iii)

Obtaining fuel linkage agreements

Fuel linkage agreements (including licenses for importing fuels – coal, naphtha, diesel, and LNG or higher grade Indian coal) have , at times, been difficult to obtain. 32

Currently, fuel linkage agreements have to be made on the basis of the Techno-Economic Clearance (TEC). The latter is not awarded till the environmental clearance is obtained, which in turn is dependent on the type of fuel to be used. In addition, the rules on the use of some of the fuels have not been clear or have been changed. This indecision has delayed several projects. Moreover, the charges the IPPs have to pay have appeared for them to be too high, as these include charges for commitment, import-handling, service, and so on. Added to these are the many contentious conditions in the Fuel Supply Agreement (FSA) of 1997.

(iv)

Obtaining clearances

A number of clearances – statutory and non-statutory – are there to be obtained for starting a power project. The statutory clearances include cost estimate clearance, TEC from CEA, water availability clearance from the CWC/State government, pollution clearance from the State/Central Pollution Control Board (SPCB/CPCB), forest and environment clearance and rehabilitation and resettlement clearance from the Ministry of Environment and Forests, and the SEB/State government clearance. The non-statutory clearances include land availability from the State government, fuel linkages from the departments of coal and petroleum and natural gas, transportation of fuel from these departments, and the ministries of railways, and shipping and surface transport, and financing from CEA/Department of Power/Department of Economic Affairs/Financial Institutions. All these can result in considerable time and cost overruns. The role originally meant for Investment Promotion Cell (IPC), that of a one-stop window for project developers, has not actually been materialised, because of the split of responsibilities between the Central and State Governments, as also of the Cell’s inadequate institutional set up. “The number of clearances and the number of relevant bodies involved has limited the ability of the IPC to ensure that all clearances are granted expeditiously” (World Bank 1995: 85). It should be noted, however, that since August 1996, power projects with investment of Rs. 1000 crores have been exempted from CEA and environmental clearances. Several ‘fast-track’ projects, however, are above this limit.

33

6. Power Sector Reforms in Kerala

The waves of power sector reforms that have swept the world over and some parts of India as well had left only moderate imprints till recently in Kerala’s power sector. It was acknowledged in the previous state government’s electric power policy of 1998 (the first of its kind in Kerala!) that the huge capital investment required in the power sector imposes heavy burden on the KSEB with its weaker financial standing. During the 9th Plan period (1997-2002) projects in the three sectors of generation, transmission, and distribution involve about Rs. 4380 crores, of which only Rs. 350.06 crores can be had from the internal resource generation of the KSEB, provided tariffs are revised regularly, and Rs. 735.51 crores from the State government as loans, leaving the KSEB to rely heavily on the financial institutions (FIs) for the remaining resources. If tariffs are not regularly revised or arrears in revenue collections build up, the borrowings will have to be more. Given the financial status of the Board and its track record, it is found doubtful if external loans of this order can be raised. The situation is thus made ripe for some attempts on reforms.

The E. Balanandan Committee to Study the Development of Electricity in Kerala (1997) recommended to set up a government-owned company viz., The Kerala Power Development and Finance Corporation Ltd., to develop, finance, and manage generation of electricity, construction and installation of power stations and transmission lines in Kerala1. The Task Force on Policy Issues Relating to Power Sector and Power Sector Reforms (1997) and the Expert (under K. P. Rao) Committee to Review the Tariff Structure of the KSEB (1998), both constituted by the State Planning Board, provided detailed discussion on reforms processes that to a great extent reflects the present Government’s ideological prejudice and political compulsions.

In the State’s electric power policy, it was clearly stated that the (then Marxist Party-led) government had no intention of unbundling or of privatizing the SEB. The suggestion to corporatise the three divisions of generation, transmission, and distribution was also rejected. However, it was acknowledged that there should be significant changes in the structure and approach of the Board. The Task Force had, inter alia, stressed that a major change in the work culture in KSEB be required to eliminate the inefficiency inherent in it at present and recommended that as a first step in this direction, the three operations under its control, viz., generation, transmission and distribution, 34

be compartmentalised and made as profit centres, fully accountable for the results. This arrangement was From the Power Policy of Kerala Government, 1998 Ensuring financial viability through improvement in operational efficiency and cultivating good relationship with customers by avoiding activities that leave them dissatisfied – these two objectives should be realised by the Board, along with power self-sufficiency by 2000 AD. More focus should be placed on consumer-service areas and a development-service oriented and energetic work culture should be cultivated in the Board. It should be strictly enforced that regular and timely meter reading, billing and revenue collection be ensured. The spot billing procedure at present in force in 67 sections would be gradually extended to other sections. Suitable measures would be taken to solve the problems/difficulties of the customers in the present system of payment of electricity charges. A large number of three-phase meters and single phase meters remain defective, causing substantial revenue loss to the Board. A phased program of replacing them with electronic meters would be implemented. Further replacement would be the responsibility of the customer himself. With the new connections, it would be the responsibility of the customer to buy meter. The Government would also take effective measures, besides those by the Board, to check power theft, illegal drawal, etc. More number of anti-power theft squads would be organised. Special squads to watch the HT – EHT customers would be deployed. Vigilance department would be strengthened. …. ……… …… ……… ………. The power sector in Kerala is entering through this policy into a long and difficult action program that would lend a new direction to development ventures and thus ensure liberation from power crisis for ever. The success of this action program depends upon the collective cooperation of the people, workers, officers, and capital investors. And the Government is sure that the people, forgetting all differences, would stand united in this save-Kerala effort. expected to facilitate the relative efficiencies in each sector and enable KSEB to take corrective actions more effectively (quoted in Government of Kerala 1998: Annexure 2)

The then government accordingly initiated necessary steps to restructure the functioning of the Board in terms of ‘profit centres’ at the levels of generation, transmission, and distribution; three regional profit centres with head quarters at Thiruvananthapuram, Eranakulam, and Kozhikode also were established. These regional centres would have the control over the electricity supply in the 35

State. The profit centres would have wide autonomous powers in decision making in several areas including capital investment, resource generation, appointments of personnel and so on.

Though the government promised all the help and cooperation to the IPPs, only two projects (one mini hydro power project of 12 MW at Maniyar owned by Tata Tea Estate, and a thermal project at Kochi of 160 MW under the ownership of BSES and KSIDC) have so far been commissioned in the private sector. In 1997, the then government proposed some ambitious plans2 to set up power projects in the private and public sectors within 5 years, with a total IC of 5041 MW (including the BSES, NTPC and the KSEB’s own thermal projects, works on which had already started that time). However, the fate of these projects, other than those mentioned above, is still not known. Despite the professed commitments and colourful plans, the required firm political will and sense of responsibility to value common good above everything else is conspicuously missing in our Governments. The recent so-called ‘Kannur-Ennore’ controversy is an apt example in point here.3

The present (Congress Party-led) state government, however, decided to swim along with the current, by joining the group of other 14 states in the country already engaged in radical power sector reforms at the terms and conditions of the central government; in August 2001, Kerala signed a memorandum of understanding (MoU) with the Union power ministry, expressing its willingness to undertake power sector reforms. As per the MoU, the KSEB is now to be run on commercial lines and also to securitise all its dues to the central public sector undertakings (CPSUs). Such securitisation implies that the KSEB ensure that CPSU outstandings never cross the limit of two months' billing. And in return for its commitments, the state would be provided by the central government with funds from the Accelerated Power Development Programme (APDP) for renovation and modernisation of thermal and hydro plants of the state and for improvement of subtransmission

36

A big NO to Unbundling and Privatisation Restructuring is being considered [now in India] mainly because the SEBs are operating at loss and are not in a position to meet the electricity demand and are also considered inefficient with high T&D losses and KSEB is no exception. At the same time, the factors that have led the SEBs into this situation, which are quite well known, have not been removed and no attempt has been made in that direction. …..The Task Force is of the view that before contemplating any restructuring, which becomes irreversible, it is to be examined whether the present sickness can be remedied without any drastic surgery by removing the problems that cause the sickness. It is to be stressed that the problems that may arise consequent to restructuring could be more severe than the existing ones, particularly when a large and complex organisation like KSEB is unbundled and split into several units….. Electric utility, unlike other engineering industries, requires perfect and total coordination between generation and T&D. A composite organization is suited best for this purpose. Financial assistance is likely to be extended to KSEB from banks and financial institutions including international agencies in case its balance sheet is healthy, which is possible only if it is permitted to follow a rational and sound tariff policy. The Task Force noted that the utilities, by tradition and practice, are a natural monopoly and there can never be a competitive situation vis-a-vis the consumer. This is for the reason that the consumer has no option to choose from more than one source providing the utility service. The Task Force is of the view that if there were to be a monopolistic situation, Government monopoly, which is subject to Government control keeping in view the social objective, is far more preferable than a private monopoly where commercial or profit making interests prevail over other considerations. The Task Force, accordingly, strongly recommends against privatisation of transmission and distribution activities. (Executive Summary of the Report of the Task Force on Policy Issues Relating to Power Sector and Power Sector Reforms; quoted in Government of Kerala 1998: Annexure 2)

and distribution and metering in the identified circles in the state. The MoU requires the state government to 'desegregate' the KSEB to make it accountable in respect of its functions of generation, transmission and distribution; accordingly, the KSEB was divided into three 'independent profit centres' having separate administrative set up and accounts in April 2002. It is also proposed to set up an independent State Electricity Regulatory Commission and to file tariff

37

petitions. It should be stressed that the developments indicate that the power sector privatisation in the state, so far a radical anathema to it, might not be unlikely. 7. Is Structural Reform the Panacea?

Both the task Force on Power Sector reforms (1997) and the Expert Committee to Review the Tariff structure of the KSEB (1998) have “strongly cautioned that hasty decisions in this respect would lead to irreversible actions which could lead to many unforeseen problems. Besides, it does not help to resolve the problems faced by the Board of inadequate tariffs, internal resources and liquidity.” (Government of Kerala 1998:6). The present Government of Kerala, with professed leftist leanings, seems to explicitly endorse this view. Apart from this “disastrous irreversibility” premise, the very logic of the power sector reform process stands helplessly vulnerable to multiple points of weaknesses, much of this facet, however, remaining outside the plane of informed debate. Here we take up some of them.

As explained above, private sector participation had been solicited on account of the fiscal crisis begotten funds scarcity. But this very fiscal crisis, obvious to any open eyes, has been due to the Government’s inability to raise the revenue receipts and/or to reduce revenue expenditures. Instead, the axe has fallen on the capital expenditures, at the cost of development; and these savings, in a reverse logic of necessity, have begun to contribute to the revenue account4 – such is the public finance management of our Governments! Still worse, it is the developmental expenditures in both the accounts5 that have suffered the most, again in a perverse logic. This stands to ridicule all the blab of financial discipline raised in the face of the so-called fiscal crisis. As explained elsewhere in this Report, the crisis, under the tutelage of the World Bank, awakened the Government to the prescription of identifying fiscal stability of the economy with very low level of fiscal deficit. This in turn implied strict measures of financial discipline through severe expenditure cuts. But, the guillotine descended

on

the

wrong

heads – of developmental/capital expenditures, while

profligacy stood to fatten the non-developmental/revenue expenditures, leaving the fiscal deficit, the alleged prime mover of crisis, without any perceptible change, even in the face of increasing capital account surplus, achieved through capital expenditure cuts ! If so, if it was not for translating any effect on to the crisis-breeding deficit, then one naturally tends to doubt the genuineness of all the 38

fuss and justification of all the initiatives. Indeed, there seems to have been some snag in it. And it is to be seen in the effects of a combination of three forces – the two indigenous factors of the political economy of corruption and of hypocrisy in company with the exogenous World Bank hegemony.

The much-coloured ‘fiscal crisis’ of balance of payments shortage of 1990-91 came in handy for the World Bank to dictate conditions of ‘economy-wide structural adjustments’ or ‘reforms’ in return for soft loans provided for tidying over the shortage problem (ballooned into a ‘crisis'). The prescriptive measure of fiscal deficit reduction had a built-in effect of increasing external dependency and thus submission. Here the World Bank rose to the occasion and exhorted, besides imposing the structural adjustment loans, that the Government relieve itself of the financial crunch by reducing its role to a facilitator only, instead of being as hitherto a provider. Selling out public sector assets, accordingly, yielded two birds at a stroke – relief from public sector management burden and substantial funds. The offer of sift loans and the option of public sector divestment were powerful enough to lure the political economy of corruption, while the populist sentiments pampered by the religion of hypocrisy, and Governmental profligacy6 dared not to touch revenue expenditures. And the price was paid from the capital/developmental account. The logical culmination of such a situation was the cultivated perception that the Government was left with no money. Now the responsibility for developmental investments naturally devolved upon the private sector, making the World Bank approach easier. But the domestic private sector remaining not so strong, the door was to be opened to the foreign capital. This dynamics should have served as a frame of reference in any informed debate on interpretations and implications of the so-called reforms move in India. Unfortunately, however, this has not been so, so far.

The funds scarcity proposition is still weaker on another potential score also. Sadly enough, very few eyes have been open to the folly of the fiscal deficit = instability equation setting economics of the World Bank. This might be true in a Keynesian set up of an advanced economy where aggregate demand and effective demand coincide, leaving an inflammable situation for additional finance unaccompanied by additional output. On the other hand, in a less developed economy of poor majority with low purchasing power, which in turn means over-production or equivalently, under-utilization of capacity, pump-priming serves only to boost the economy. 39

However, the main thrust of our point here is that despite the World Bank compulsion, deficit financing still continues in India as before, but now only for revenue expenditures; this is in addition to fiscal incentives through tax reductions. At the same time, large cuts in capital expenditures also are effected; and this situation has been capable of fuelling inflationary flames in the World Bank economics sense, though inflation in India remains (at least in the official claims) in the manageable reach only. Such a particular situation of an insensitive inflationary mechanism7 in force in India, however, has been sadly missed by many

40

“Bogey of Resources Constraint” Why then has the State chosen to embark upon a policy that is so damaging to the nation’s economic interest ? The answer no doubt has to do with the fiscal crisis of the State, but as we will argue certain ideological position and the interaction of this ideology with the crisis is important. The fiscal crisis is by now well known. It arises due to its inability to cut revenue expenditures, or to raise the price of infrastructural services and products including electricity, and its total impotence to mobilise additional resources via taxes on the rich. As was expected by many economists in 1991-92, it has only cut capital expenditures. This has accelerated so that today the savings on capital account contribute to the revenue account ! In power, budgetary contributions have sharply fallen, and in short, to use the popular jargon there is simply no money with the government. While a tightening of the budgetary resources to its parastatals is expected in a period of structural adjustment, and becomes necessary to create the pressure for cost consciousness and commercial orientation, or even to soften them up for privatisation, in the power sector it would be entirely disastrous. Power is the most critical input for agricultural and industrial production. The Government’s own statements amply confirm thet it is well awrae that the “marginal productivity” of power in the rest of the economy is far greater tyhan the cost of power. (Somewhat cynically, for the xenophiles, this would amply justify the foreign IPPs even if they are more expensive. And it does because, even IPPs, set up entirely with foreign capital and all equipment sourced from outside, is better than power cuts and shortages.) This means that power development ought to be the topmost economic priority of the State. It also means that there is an opportunity for deficit financing of power projects, so that the required additions to capacity to match demand need not suffer for want of resources. Only a dogmatic monetarist position would insist on identifying the finances for power development with required savings for the economy as a whole. Deficit financing in the case of power (if tight implementation schedules can be adhered to) need not be inflationary given the extremely high marginal product of power in industry and agriculture. With the extra power availability, if output can go up significantly, then the resources would be self-financed for the economy as a whole, via the increased income generation. In other words, of the twin considerations (or near term objectives) in structural adjustment – expenditure switching and expenditure reduction – in an economy where the productive sector is fundamentally constrained by a critical supply side bottleneck (here power), and where removal of the bottleneck does not involve long gestation, the right policy would be major expenditure switching. Overall expenditure reduction has to be tempered to accommodate the expenditure increase in overcoming the critical bottleneck….. Morris 1996.

41

intellectual eyes and the Government too. Thus the Government, if found itself still comfortable with deficit spending, should have, as is truly expected, drawn rein on revenue expenditures, and effected that fiscal financing for developmental purposes, which would necessarily have averted the dependency problem, and along with it the painful chaos of present ‘reforms’.

However, this would be only a partial solution. The real resolution should have emanated from the Electricity Board itself with an active spur from the Government. The Board should have been functionally efficient and financially sufficient to meet all its requirements. There is no inviolable destiny or curse that public sector be inefficient. A large number of living examples do shatter this myth (though it still reigns supreme over a large terrain of social consciousness). Kerala herself has enjoyed a golden era in respect of FACT under an able management. In power sector, Maharashtra SEB had been consistently performing efficiently for a long time till the entry of Enron. In the Central sector, the NTPC has won laurels for its top performance8. The NTPC, accounting for about 25 per cent of India’s total power generation with an IC of about 20 per cent, is the World’s sixth largest thermal power generator and second most efficient, according to a survey by Datamonitor of the UK (based on 1998 performance data). Given a conducive environment for a committed management, the Electricity Board could have fared better true to its guiding principles of a commercial-cum-service organisation, as interpreted by almost all the Committees. But the socio-political populist compulsions of the Governments could not honour and ensure its statutory status of an autonomous corporation (as required by the Electricity (Supply) Act of 1948), and they found in it a cornucopia for their immediate gains through subsidised tariff, heavy rural electrification, and employment generation on one hand, and corruption on the other. Had the Government compensated the Board for all the populist favours then and there, at least its balance sheet would not have run into the red. And all these have never been unknown to any one, and Committees after Committees have echoed in vain the same tone. However, the Government could by no means simply forgo this easy but powerful vehicle that it was using for translating populist baits into its own immediate gains and the forced conversion of the Board into a Government department prevailed9. It is the weight of this compound of corruption and hypocrisy that in fact restrains some of the State Governments, with ideological assertions coloured in populism, from openly supporting the reform moves.

42

A legitimate question might crop up now: Why did some Governments then decide to forgo this cornucopia ? The answer must be clear in terms of the political economy of corruption on a large scale of favouritism and kickbacks in reaching agreements with private parties, besides the lure of soft loans from different agencies, made possible in the wake of the reforms10. Kickback rule has become an integral part of private sector participation in power sector explicitly ever since the Enron

43

The Golden Touch of the NTPC An operations evaluation report of the World Bank last year [1993] on the NTPC gave a lot of credit for its outstanding performance from its inception in the seventies, to the leadership provided by Mr. [D.V.] Kapur, [its founder-chairman], and the organisational structure that he built…. Some months ago, Kapur circulated a paper within the higher echelons of the Indian government outlining how the expertise in NTPC could be employed to raise the efficiency of electricity generation. The route Kapur suggested is the one the NTPC followed when it was asked to take over and improve the efficiency of an ailing 600 MW plant near Delhi, at Badarpur, in 1978. After an initial survey to identify Badarpur’s problems, he delegated one or two engineers to work with the engineering staff at Badarpur, and get them to formulate plans for their own improvement. In NTPC he set up a five-man team of senior engineers to back-stop the team of Badarpur. Most important, the services of an unusually strong corporate planning office set up by Kapur at NTPC were placed at Badarpur’s disposal. The system worked. In two years, through retraining of supervisory staff and a variety of minor repairs, plugging of leakage and other detailed improvements that cost very little money, the PLF [plant load factor] of the plant was raised from 33 to 56 per cent. In two more years, through major repairs and technological upgrading, the PLF was raised to 70 per cent. The benefits of Badarpur experiment are still percolating through the power sector. In 1993, a newly appointed chairman of the West Bengal SEB was able to bring about a dramatic improvement in the performance of the plants operated by it so much so that West Bengal began to export electricity to the Northern Grid, which serves Delhi. The chairman was one of the members of the original team that set Badarpur right. In the same way, a thermal power plant set up by the UPSEB was turned around dramatically by the NTPC after it took it over, by sending as its general manager another engineer who had been seconded to then Badarpur plant in the seventies. [Hence] the scope for improvement is vast…….What D. V. Kapur’s paper has shown is how these improvements can be made and the time period – roughly four years – over which they will bear fruit. This is still roughly half time that new plants being proposed by the private sector will take to come on stream. …. - Jha 1994. controversy11. The tendency has been to allow the kickbacks to be included in the capital cost such that exorbitant marginal capacity cost is thrust upon the system12. The proposal for introducing marginal cost pricing regime should be necessarily debated in this light: Should the society be burdened with such inflated marginal capital costs in the guise of ‘efficient’ prices ? And as we 44

know, this corruption-push is only one frequency band in the wide spectrum of the cost-inflation. It is not fair for a ‘welfare State’ to yield to the tendency to put all its inefficiency upon the public, branding it as socially efficient costs, though it might be in line with its religion of hypocrisy. This will unfortunately lead to an undesirable exclusion of the entire poor from access to light, that even the 50 years of populism could not bring to them. In fact, one of the serious concerns raised in the context of reform exercises is regarding the rural access to electricity, which the 1948 E(S) Act stood to guarantee. It is a well known fact, confirmed by many a survey, that the unelectrified households in general belong to the poor of the society – an ironical reflection of the Government commitment. Given the highly skewed income/assets distribution set-up in our country, then, the so called reforms with its intended functional structure of market orientation (manipulated by private interests of profit- and rent-seeking of all hues that now never coincide with social interests – gone are the reconciling days of Adam Smith!) would stand to darken the still dark alleys of the poor section. While the Government is too eager to shirk its fundamental social responsibility of subsidising the poor, under the chastisement of the World Bank and its indigenous pedants, all these parties involved very conveniently forget that ‘subsidy’ is not a Third World phenomenon only.

Another important aspect thus apparently winked at should also be highlighted. The much taunted investment incapacity of the SEBs has been the prime leverage in justification for the private sector participation (PSP) in the power sector. However, the SEBs being the major (if not sole) purchaser of power from the IPPs, the fear of payment default tends to strike at the very root of the PSP program, and this in turn necessitates that the SEBs be financially healthy to provide an escrow cover for the purchase. This circularity argument just nullifies the very PSP logic; if the SEB can afford to buy power from the IPPs (that too at higher prices13), then why cannot it afford to have its own generation facilities (at lower costs)?

That the power sector has problems galore as the Pandora’s box of reforms is opened goes without saying. As explained elsewhere, the terms and conditions of power purchase agreement (PPA) for the IPP’s ‘must run’ base load plants adversely affects the merit order operation of the power system, thus causing systemic inefficiency. The higher capital costs and the consequent higher tariff rates result in exclusion of the majority of the poor from the ‘market’. Moreover, the inescapable problems involved in the irreversible restructuring/dismantling of a complex 45

organization, as far as the experience of the Orissa experiment proves, might lead one to doubt whether these problems are not an exorbitant cost to compensate the original problems that the restructuring was supposed to tackle. Quite disheartening are the reports, on the power sector management health (or even its survival itself), from Orissa, where the World Bank model has

The Plight of the SEBs KSEB also faces the problems of its counterparts in the country. There has been frequent criticism that it is not functioning as an autonomous corporation as required by the Electricity (Supply) Act, 1948. The Committee on Power (Rajadhyaksha Committee) have made a number of recommendations relating to the restructuring of SEBs and made detailed suggestions regarding the composition of the Board, manner of selection of top personnel, improvement of management systems and procedures and information and control systems. ….Implementation of these recommendations has been considered very essential having regard to the increasing role of the Electricity Boards and their major responsibility for implementing Power Programs. Most of the State Governments including the Government of Kerala have not yet accepted the recommendations of the Rajadhyaksha Committee. We recommend that Government of Kerala take immediate decision in favour of accepting these recommendations and implementing them. -Government of Kerala 1984: 40. The SEBs were created as autonomous bodies and were statutorily entusted with the task of undertaking planning, generation and supply of power to the ultimate consumers, in the most economic manner. They were also required to operate with due autonomy, fix tariffs adequately as per dictates of the E(S) act, so as to be financially viable. Events over the last few decades have adversely affected the operations, particularly of the State Electricity Boards. Many of the factors that have contributed to the present financial crisis of the Boards have long been identified but the underlying causes have not been remedied. The result has been that there is a pressure from international lending agencies such as World Bank to increasingly divest the SEBs of their rightful functions, unbundle services and throw them open to the private sector and even abolish SEBs altogether. - Government of Kerala 1998: 23. fulfilled its full mission in terms of achieving unbundling and privatisation. Both the State and the Central Governments have been injecting heavy doses of finance in frantic attempts to rescue 46

the system from imminent collapse, while the international Chief Surgeon has just backed out after the “The Curate’s Egg” Despite[the] preoccupation, there is no justification for privatisation; it is supported only with handwaving. If efficient functioning of the power sector is the objective, there is no discussion of privatisation versus competition to achieve efficiency. There is no judgement on whether privatisation or competition is fundamental and primary. The EC report [the report of the high level committee on escrow cover (EC) to independent power producers, chaired by Deepak Parekh] shows no awareness of the important case of Norway where there is effective competition without privatisation. The report also does not prove that privatisation will lead automatically to competition and therefore efficiency. A recent article in Frontline on the Orissa privatisation shows that private distributors may not be interested in efficiency if it does not increase returns. The EC report has a faith in market forces that is touching, but naïve. It seems unaware of the colossal failures of demand forecasting by the private sector in the United States leading to overestimation, construction of mega (nuclear) plants and dead investments. But its belief in the market is not consistent. Where it should be arguing for a level playing field for market competition, it demands command and control policies. For example, it recommends privatisation of existing generation capacity…It also asks for new generation to be assigned to the private sector instead of asking for a level playing field in which there will be a competition with the public sector. …….A most unwelcome, perhaps even dangerous, aspect of the EC report is the contemptuous way it has dismissed integrated resource planning. It has implied that once the energy system is privatised, the market takes over and the government vacates the scene, everything will be perfect in the private sector. In fact, notwithstanding all the advantages of the market as an allocator of manpower, materials, technology and investments, the market is notorious for its failure to safeguard equity and environment in the long run. Its horizon is limited to the balance sheet. So, the State has an important role. - Reddy, D’Sa and Murthy 2000. initial incisions, requiring the domestic surgeons to do what they can to complete the operation! While the Governments are too eager to let out any signals of a wrong turn during the course of such a drastic surgery, one would wonder why these Governments could not apply a little of this wisdom and sincerity during the previous phase.

47

Orissa is the only reforming State where distribution sector (also) is privatised; and the move is on in this direction in other reforming States of Karnataka, AP, and Haryana also. It is generally recognised that distribution is the weakest link in the whole structure of power supply system.

48

The Pandora’s Box Problems Faced by the SEBs Unacceptable PPA terms – not viable for the SEBs: According to the terms specified in some of the PPAs, the country would have to pay an exorbitant price for foreign participation. Several harmful features are listed below. Assured high PLF: Plants were to be assured of electricity sales at PLFs of  68.5 per cent, these high PLFs being buttressed by power purchase agreements (PPAs). This commitment implies that during the daily off-peak hours and the monsoon season, the existing plants would have to be backed down, resulting in uneconomic plant dispatch (that is, lower cost-per-unit power would be replaced by higher cost power). Considering that several existing thermal plants that can operate at higher than 68.5 per cent are backed down in periods of good reservoir inflows in the southern region, the situation would only be worsened. Further, if the real ailment of the power sector is a shortage of peaking power rather than energy, then the addition of base-load power stations is not likely to provide a solution. High return on equity: A relatively high ( 11 per cent) rate of return (ROI) was promised to the investor, at a capacity utilization of 68.5 per cent. This return would be increased if the utilization exceeded this level. More importantly, these returns were to be guaranteed by the Central Government if the SEBs were unable to pay. High capital costs of private plants: The capital costs of some projects (as per their PPAs) were much higher than those known to be incurred both abroad and in India where international competitive bidding did not take place. For example, the capital cost of Phase I of the Enron (Dabhol Power Company) LNG-based plant was Rs. 4.23 crore/MW or $ 1366/kW (=Rs. 2942.6crore for 695 MW). [This is obtained after deducting the development fee of Rs. 86.4 crore from the actually reported Rs. 3029 crore.] In comparison, in the US, a basic 300 MW coal-based steam-electric plant (about 30 per cent more expensive than an LNG-based plant) required about $ 1100/kW in 1990 prices, which works out to about Rs. 3.4 crore/MW @ Rs. 31/$, and NTPC’s 645 MW gas-based Kawas project (commissioned in September 1993) at Rs. 2.4 crore per MW. In addition, there were payments in the deal for equipment/consultancy/recurring expenses to affiliates of the owner-firms. All this led to critical comments and some re-negotiation. The Enron project (Phase I) cost was reduced, as was the Cogentrix 1000 MW project cost (from Rs. 4387 crore to Rs. 3950 crore). However, a part of the reduction in costs is claimed by critics to be cosmetic: for instance, the Enron-LNG facility appears under operating rather than capital cost, and customs duty reductions have been reflected as capital cost reduction. In cases where the prices of equipment are falling, adherence to the PPA prices would be uneconomical for the power purchaser (the SEB).

High tariffs: In addition to the capital component, the variable costs chargeable during the life of the projects are expected to rise, allowing the escalation of the costs of

49

various components – fees (such as Management Fee, Testing Fee, and Commissioning Fee), insurance charges, ‘tax incremental charges’, etc., to be passed on to the purchaser. Unfavourable financing: The rates of interest payable on dollar and rupee debt have been fixed as on the date of financial closure and securement of counter-guarantees), the perceived lender risks and the corresponding rates of interest are relatively high. However, as the project progresses, the risk falls and the debt could be refinanced (that is, interest rates can be lowered through re-negotiation). Despite this, the utility is still bound by the fixed rates. Further problems Technical losses and improvement of the T & D system: Increasing the generation capacity is necessary but not sufficient for supplying electricity to consumers; the T&D system has to be extended and maintained to ensure the efficient evacuation of power from the generation sites. Without improved T & D facilities, the technical inefficiencies will continue. A separate trading enterprise for T & D (for example, GRIDCO in Orissa) that needs to collect a certain ROI would entail much higher tariff rates which some consumers may be unable to bear. Commercial losses on the T & D system: The losses incurred along the distribution system due to theft of electricity have not been addressed by introducing more generators into the system. In fact, the SEBs’ financial position would worsen if electricity purchased at higher prices (the costs-plus-return formula) were not paid for by the users. Privatisation of the T & D system: Private participation in the T & D of the electricity system has also presented problems. The evaluation of assets in cases of transfer to new owners has to be carefully worked out. For joint venture undertakings between an SEB and a private firm/consortia of firms, the SEB is liable to lose control. In addition, the SEBs sometimes define the requirements for transmission contracts such that there are very few companies capable of fulfilling the criteria as defined, so that negotiation is even more difficult. Non-subsidised electricity: The consumers (mainly domestic and agricultural) currently provided electricity at subsidised rates would be unable to handle “user-cost recovery”, that is, to pay cost-reflective tariffs. Further, if only these consumers are left to the SEBs, their financial position would be far worse than at present. Fuel imports: In spite of the availability of indigenous sources of electricity (hydropower, coal, bio-mass), foreign power producers tend to opt for imported fuel. The larger the number of foreign power producers in the field, the gretaer will be the country’s dependence on imported fuel for power generation, worsening its debt levels still further. - D’Sa, Murthy and Reddy 1999.

50

The massive leakage from this inefficient outlet in the form of subsidised sales and distribution loss, including technical loss and theft, illegal drawal, etc., under protective patronage, have been steadily sapping the SEBs, thus taking them to a no-return point of forced reforms. Plugging such leakage thus constitutes the urgent remedy for all the problems. And a general perception in the informed circle endorses immediate privatisation of the distribution sector projected as the only way out (for example, see Morris 2000). Tackling such leakage in many rural/suburban areas involves “a law and order dimension as well” (Governemnt of India 2996: 59), and a populist Government, so far in thehabit of winking at (if not abetting) such criminal errancy, finds it difficult to come out on the front. The Government saves its face by leaving everything to the private sector. Thus the private distribution company in Orissa, “the AES of USA is having to employ goon gangs to install meters”, and to collect the dues (The Hindu Business Line, March 31, 2000). See how easy the problem is solved! A blatant sell-out of governmental obligations14!

It is not that there is no alternative to such suicidal sell-out. There have been some informed suggestions on setting up cooperatives at local levels and entrusting them or the local bodies themselves with distribution responsibilities. For example, the Task Force constituted by the State Planning Board on policy issues relating to power sector and power sector reforms cites the good examples of Hukkeri Cooperative in Karnatake and Trissur Municipality in Kerala. The former is one among the 38 cooperatives in the country set up as conceived by the Rural Electrification Corporation. Power is supplied to these cooperatives at tariffs below the standard bulk rates such as to enable them to operate with a surplus. In Trissur Municipality area, a licencee under the control of the Municipality is engaged in electricity distribution in a very satisfactory manner. A number of countries have such alternative arrangements functioning efficiently15.

All this should not be misconstrued, let us reiterate, as an unreasonable justification for the persistence of avoidable inefficiency in the performance of the SEBs. As we have shown elsewhere, the inefficiency problems are only internal to the system. There do remain rooms for remedial exercises meant to remove these problems inhibiting the SEBs’ improved performance. That is, what the system requires is only an essence-specific (internal) reform – a reformed work culture

51

under the leadership of an enlightened, committed, professional management and Government should flourish and further – not a disastrous structural reform, as is fetishistically made out now.

52

From Orissa With Pains Four years after the Orissa Government began experimenting with reforms in the power sector under the gaze of the World Bank, Gridco, the State-owned transmission company, crippled by a massive debt-servicing burden, is on the verge of being taken to the infirmary – the Board for Industrial and Financial Reconstruction (BIFR). Thanks to the consistent step-motherly treatment meted out to it by the reformers, Gridco’s net worth has been eroded completely. As on March 31, 2000, it will have outstanding loans of Rs. 2714.5 crores payable to financial institutions, the public and the World Bank, and dues amounting to Rs. 1160.4 crores on account of power purchased from the generating agencies. The collapse of the entire reform program appears imminent unless Gridco is bailed out with a massive financial restructuring package. Further, the performance of the four recently privatised distribution companies (distcos) has left a lot to be desired, raising fundamental questions about the methodology adopted for the reform experiment, if not its very rationale. Considering that Orissa adhered to the blueprint for reform and restructuring drawn up by experts from the World Bank, which is extending a $ 350 million loan to the State, it would not be unreasonable to expect the Bank to devise and fund a course correction. Yet, the Bank appears to have left it to the Government of India and its agencies to evolve bailout measures. A letter dated December 21, 1999 by Edwin Lim, the Bank’s Country Director, addressed to the Orissa’s Chief Secretary, requires “assistance from the GoI and its agencies beyond levels envisioned in current drafts” to rescue Gridco. The Centre promptly complied with a financial restructuring package lest the setback should send negative signals to other reforming States……… Where exactly did the experiment go wrong ? The unbundling of generation from T & D in 1996 left the latter with assets valued at Rs. 1183 crores at depreciated replacement cost. But since the liabilities were much higher, the assets were upvalued to Rs. 2395.8 crores, just to match the liabilities rather than on any reasoned ground. The State Government owed Rs. 340 crores to the SEB in the form of arrears of annual subsidy and outstanding bills for power consumed. These were adjusted in the ‘upvaluation’ process. The other adjustments included the issue of equity worth Rs. 253 crores and zero coupon bonds worth Rs. 400 crores by Gridco to the Government. The ‘upvaluation’, which was recommended by Bank-appointed/approved consultants (who included such names as KPMG and Price Waterhouse (Coopers)) was justified on the grounds that Gridco needed a large capital base to absorb the substantial debt funds required for the upgradation of the T & D system. The alacrity with which the State Government agreed to this proposal stemmed from its own financial circumstances. Thus a lethal combination of the Bank’s wisdom and the State Government’s expediency left Gridco penniless in the face of massive debt-servicing burden. Gridco was doomed from Day One. But the bank’s wisdom extended further. It had ordained that henceforth the unbundled entities would charge tariffs based on their revalued asset base. This

53

54

meant that the price of power sold by the generators – the Orissa Power Generation Corporation (OPGC) and the Orissa Hydro Power Corporation (OHPC) – would reflect their new asset value. It would be calculated at 16 per cent return on capital as per Schedule VI of the E(S) Act of 1948. From just 18 paise a kWh the previous day, the OHPC increased the tariff to 38 paise from Day One of unbundling (April 1. 1996), which further went up to 49 paise from April 1997. Gridco’s annual bill for power purchase from the OHPC shot up from Rs. 60 crores in the Board system to Rs. 140 crores from April 1, 1996 and to Rs. 180 crores from April 1, 1997. In addition, it had to contract a 30-year power purchase agreement (PPA) with the OPGC for its newly commissioned Units I & II of Ib Valley. …… Gridco’s power purchase bill from all sources quadrupled in four years to Rs. 1200 crores a year. While Gridco had to commit to purchasing power from the generators at predetermined prices, it did not have the corresponding freedom to charge its customers – the four distribution companies – prices that would reflect its revalued asset base and cover the cost of the power it purchased from the generating agencies. In a blatantly asymmetrical arrangement, the Orissa Electricity Regulatory Commission (OERC) was given the remit to clear the tariffs charged by Gridco – the retail consumers to begin with and the four distcos subsequently – but the prices that the generating companies charged Gridco under the PPAs already signed were not subject to regulatory scrutiny. Therefore, when Gridco proposed a tariff that would cover its current costs which included power purchase, employee costs and operation and maintenance expenses) and debtservicing, the regulator disallowed taking into account the latter on the grounds that past liabilities of Gridco could not be loaded onto its tariff. And rightly so. Otherwise, it would have led to steep tariff increases. Further, it would have been iniquitous to make present consumers pay for liabilities arising out of past operations, especially in a situation where only half the consumers actually paid for what they consumed. The World Bank made a facile miscalculation in assuming that Gridco would charge tariffs reflecting its revalued asset base. And worse, it assumed that the regulator would go along with its view regardless of its economic and social implications. The Staff Appraisal Report had assumed an average tariff increase of 16 per cent in 1996-97 and 18 per cent in 1997-98, whereas the regulator allowed much less. Therefore, Gridco had to reckon with steep increases in the cost of the power it purchased but had to settle for tariffs that would cover only its current liabilities. [However, in Andhra Pradesh, the regulator allowed steep rises in tariff, that has sent shock waves among the consumers and led to widespread agitation.] The other important reason for the precarious financial health of Gridco relates to its decision to take on the liabilities of the distcos, on the advice of its consultants who include Price Waterhouse (Coopers). Preparatory to the privatisation of the four distribution zones, their accounts were separated from those of Gridco, and four companies were incorporated. The four distribution companies together had liabilities of over Rs. 2000 crores. But the Bank suggested that Gridco transfer only 55

part of these liabilities and take on itself the rest. This was in order to make the distribution companies attractive to prospective buyers. Accordingly, Gridco transferred only Rs. 650 crores of the liabilities relating to capital expenditure to the distcos. The remaining Rs. 1950 crores were loaded on to Gridco’s balance sheet, threatening its very survival. In 1999, 51 per cent of the shares of the distcos themselves were sold to the highest bidders at prices that would reflect their earning capacity rather than asset value. Therefore, the regulator allowed retail tariffs that would cover the cost of bulk power purchased by the distcos from Gridco plus a 16 per cent return on the investments made in the purchase of the companies, with an allowance for T & D losses of 35 per cent. It is another matter that the distcos found their actual losses to be more (in the region of 50 per cent) and therefore found the tariffs allowed by the regulator inadequate. ….. Mahalingam 2000

Before concluding, it should be stated that there definitely has appeared a silver lining: thanks to controversial power projects, there has been wide public debate as well as informed discussion, though greater transparency in decision making, greater public participation (especially from the civil society), and greater information dissemination are still wanting. -----------------------------------------

APPENDIX State-wise Reforms and Restructuring 1. ANDHRA PRADESH  State Reforms Act came into force with effect from 1.2.1999  APSEB was unbundled into Andhra Pradesh Generation Company Ltd. (APGENCO) and Andhra Pradesh Transmission Company Ltd. (APTRANSCO for transmission and distribution)  Andhra Pradesh Electricity Regulatory Commission was operational w.e.f. 3.4.1999.  Obtained World Bank loan of US $ 210 million under the Adaptable Programme Loan (APL) 1 w.e.f. 22.3.1999 for reforms & restructuring. 56

 And also DFID's 28 million UK Pound as technical co-operation grant.  CIDA is giving technical assistance of Canadian dollar 4 million. 2. ARUNACHAL PRADESH  Decision has been taken for undertaking reforms.  SERC was constituted. 3. ASSAM  A tariff rationalisation study report, sponsored by Power Finance Corporation (PFC), has been submitted by the consultants. 4. CHANDIGARH  The Task Force, constituted under the chairmanship of Special Secretary, Union Ministry of Power for privatisation of distribution in the UT, has held a series of meetings. It has been decided that a consultant would be appointed for assisting in the implementation of privatisation process, including preparation of bid documents, legal documentation, bid evaluation, negotiation with bidders etc. PFC would provide technical and financial assistance for selection and appointment of the consultant. 5. DELHI  The State Government has come out with a strategy paper on reforms and restructuring proposed as remedial measures to the problems afflicting the Delhi power sector after conducting a diagnostic study.  A single member SERC was appointed w.e.f. 10.12.1999.  Delhi Electricity Reforms Bill has been approved by Ministry of Power. 6. GOA  The State Government is proceeding with the restructuring process for which PFC has sanctioned a grant of Rs. 1 crore and concessional loan of Rs. 4.5 crore for the consultant's services. 57

 Notification for setting up of SERC has been issued. 7. GUJARAT  State Reforms Bill is being finalised.  Restructuring programme has emphasised on metering of all categories of consumers and imposing a cap on agricultural subsidy.  SERC became functional w. e. f. 10.3.1999. It has conducted public hearings and invited consultants for making the preliminary presentations and discussions for undertaking tariff and reform related studies.  Negotiations are going on with ADB for loan. 8. HARYANA  State Reforms Act came into force w.e.f. 14.8.1998  SERC became operational w.e.f. 17.8.1998  Haryana SEB was unbundled into Haryana Vidyut Prasaran Nigam Ltd., a Trans Co. ( HVPNL) and Haryana Power Corporation Ltd. on 14.8.1998.  Two Government owned distribution companies, viz. Uttar Haryana Bijli Vitaran Nigam Ltd. (UHBVNL) and Dakshin Haryana Bijli Vitaran Nigam Ltd. (DHBVNL) have been established. These two companies are expected to operate as subsidiaries of HVPNL, until they become independent licensees.  World Bank loan of US $ 600 million is available under succeeding APLs. The works under the first APL of US $ 60 million have been completed.  DFID's technical co-operation grant of UK Pound 15 million is available for reforms works. 9. JAMMU AND KASHMIR  The State Government appointed ASCI as consultants for conducting the reforms studies and to formulate long term perspective plan for 20 years under a grant provided by PFC.  ASCI submitted the draft final report to Government of J&K in March, 2000. 10. KARNATAKA

58

 State Electricity Reforms Act came into effect from 1.6.1999.  Two new companies namely Karnataka Power Transmission Corporation Ltd. (KPTCL) and Visvesvaraya Vidyut Nigama Ltd., a GENCO, (VVNL) were incorporated and came into existence as on 1.8.1999.  KPTCL has carved out five Regional Business Centres (RBC) for five identified zones.  SERC has been functional since 15.11.1999.  State Government has signed a Memorandum of Agreement (MOA) on 12.2.2000 with the Ministry of Power, Government of India (GOI), charting out the actions to be taken towards power sector reforms in a structured and time bound manner.  Steps have been taken for the completion of privatisation of distribution, one of the main points of MOA.  As per the MOA, GOI has committed support among other things for reduction in T&D losses, strengthening and improving the transmission network and enabling supply of additional power to Karnataka, Rural Electrification Programme, Structural Adjustment, new generating capacity, etc. 11. KERALA  State Government signed a Memorandum of Agreement (MOA) with the Ministry of Power, Government of India, in August 2001, agreeing to power sector reforms.  Central government has agreed to sanction Rs. 150 crores for electricity sector reforms.  KSEB was divided into three profit centres for generation, transmission and distribution in April 2002.  Distribution would be further split into three profit centres.  Notification for setting up SERC has been issued.  Discussions are going on with ADB (and possibly other agencies) for assistance.

12. MADHYA PRADESH  Ministry of Power has conveyed its no objection on State Reforms Bill.  SERC has been operational since 30.1.1999.

59

 The proposed reform model envisages setting up of three Power Generation Companies, one Power Trading Company, one Power Transmission Company and nine Power Distribution Companies.  Measures are proposed for 100% metering, reduction of T&D losses, realisation of outstanding revenue etc.  Discussions are going on with ADB for assistance.  State Government signed a Memorandum of Agreement (MoA) on 16.5.2000 with the Ministry of Power, Government. of India (GOI), charting out the action to be taken towards power sector reforms in a structured and time bound manner.  100% electrification of villages and hamlets, metering of all supplies by December,2001 and at least 75% of the cost of supply of electricity to be charged from all categories of consumers (subject to SERC's decision) are some of the important provisions in the MoA.  As per the MoA, GOI has committed support among other things for reduction in T&D losses, strengthening and improving the transmission network and enabling supply of additional power to the State, Rural Electrification Programme, Structural Adjustment, hydro power development, etc. 13. MAHARASHTRA  State Government is committed to reforms with technical and financial assistance of PFC.  Actions initiated for appointment of consultants for undertaking tariff and reform related studies.  SERC has been functional w.e.f. 6.10.1999.  MSEBhas expressed its intention to carry out study for formation of a Joint Venture Company for distribution of Electricity in Bhiwandi area in the district of Thane. 14. MANIPUR  State Government has taken decision to undertake reforms and set up Regulatory Commission.

15. ORISSA  The first state to undertake reforms back in 1996 through State Reforms Act. Also the first state to privatise distribution w.e.f. 1.4.1999. 60

 Disinvestment has taken place in OPGC, a State Government owned company.  Process started to convert OHPC, a State Government owned hydro power generation company into a Joint Venture Company with 51% disinvestment for private partners.  OERC issued third tariff order in December, 1999, and revised tariff order w.e.f. 1.2.2000. 16. PUNJAB  Consultants have submitted the final report of the tariff rationalisation study financed by PFC.  Constitution of SERC was notified on 31.3.1999. 17. RAJASTHAN  State Reforms Act was notified on 10.1.2000.  SERC has been functional since 2.1.2000. 18. TAMIL NADU  State Ernst and Young was appointed as consultants for undertaking the reforms and restructuring study, and the agency has submitted their mid term report.  Two members of SERC have joined. 19. UTTAR PRADESH  State Reforms Act was notified on 15.1.2000..  SERC is functional.  As per the decision of the Government of Uttar Pradesh (GoUP), the activities of generation, transmission and distribution of erstwhile UPSEB have been transferred to:  Uttar Pradesh Rajya Vidyut Utpadan Nigam Ltd. (UPRVUNL)  Uttar Pradesh Jal Vidyut Nigam Ltd. (UPJVNL)  Uttar Pradesh Power Corporation Ltd. (UPPCL) - UPPCL took over the transmission and distribution functions of erstwhile UPSEB.  The activities, assets and staff of erstwhile UPSEB have been transferred to the new companies. 61

 A Memorandum of Understanding (MoU) between Government of India (GOI) and Government of Uttar Pradesh (GoUP) was signed on 25.2.2000 charting out the actions to be taken towards reforms in which GOI has committed to support the GoUP in R&M, transmission works, reforms studies, joint venture hydro projects, rural electrification and by way of additional central power allocation.  GoUP has taken a bold step of writing-off of Rs. 19,000 crores of liabilities of erstwhile UPSEB with a view to starting the new companies, created after unbundling, with healthy balance sheets.  Distribution business of Kanpur has been handed over to the Kanpur Electricity Supply Company (KESCO).  Uttar Pradesh Electricity Regulatory Commission (UPERC) has conducted five open house discussions for formulation of a tariff procedure.  World Bank has sanctioned loan of US $ 150 million for power sector reforms.  UPERC is to get an assistance of US$ 150, 000 from the Public-Private Infrastructure Advisory Facility (PPIAF), a facility established and funded by a number of bilateral and multilateral agencies and international agencies. 20. WEST BENGAL  SERC has been functional since 10.3.1999.  Re-organisation Committee set up to study the State Power Sector has submitted its recommendations to State Govt.  The State Government has set up State Rural Energy Development Corporation (WBREDC) as an independent company under the Companies Act to manage distribution for rural and agricultural consumer segments with assistance of Rural Energy Co-operatives.  Consultants have submitted the final report of the tariff rationalisation study, financed by PFC.  Four Task Forces have been formed to initiate the implementation of reform programme. The areas covered are: (a) Human Resources Planning, (b) Identification, Valuation and Transfer of Assets, (c) Identification and segregation of urban and rural feeders and zones and earmarking operational areas of WBREDC and WBSEB UD System and (d) Metering, Billing, Collection, Electricity Accounting and System Loss Reduction of WBSEB -------------------------------------------------------------------------------------

Source: Based on http://powermin.nic.in

Notes

62

1

This reminds one of a phoenix, rising from the ashes (read: heaps of dust of neglect and negligence) of a 1984 study by a group of well-wishing KSEB engineers entitled ‘A Decade Plan to Make Kerala Self-Sufficient in Electricity Generation up to AD 2000’. Following the example of the Karnataka Power Corporation set up way back in 1970 by the Karnataka SEB, these engineers suggested to form a holding company with a share capital of about Rs. 400 million for power generation in Kerala; the idea was to lessen the financial burden on the KSEB of power development and thus to improve the power supply situation. Detailed plans on a number of hydro power projects to be undertaken by the corporation during the next 20 years were included in the proposal. Thanks to the far more politically conscious trade unions in Kerala, however, the study was simply shelved away by the Government, that too during the ‘unprecedented” power crisis period! That was the Kerala model of power development! 2

The State Government signed PPA for six proposed projects with a total capacity of 2175 MW to generate 15378 MU on a tentative cost of Rs. 6529 crores with Siasen Energy Ltd., Wise Ltd.., New Delhi, Kasargod Power Corporation, Finolec cables, Mumbai, and Kumar Energy Corporation. Besides, MoUs were signed with three companies, BPL Ltd., KPP Nambiar Associates and EDC International, Bangalore, for setting up power plants in Chimeni in Kannur and Manakara in Palakkad with a total IC of 1330 MW at a tentative cost of Rs. 4523 crores to generate about 8000 MU. All these projects except one were to use Naphtha as fuel. In addition, in the public sector, a 500 MW power project was proposed by Kochi Refineries Ltd., using residual fuel, at a tentative cost of Rs. 2000 crores; 10 KSIDC-sponsored schemes using naphtha with a total capacity of 300 MW, and two diesel plants with a total capacity of 24 MW, to be jointly set up by Kerala Infrastructure Development Corporation (Kinfra) and Kerala Electricals Ltd., were also under consideration. The tentative investment on the State public sector projects was Rs. 972 crores. KSEB, on its part, proposed small thermal units in substations and 48 small hydro projects with a total capacity of 312 MW at an estimated cost of Rs. 675 crores. And all these ambitious plans were to add to the system an IC of 5041 MW (to generate 26371 MU) at a tentative cost of Rs. 15549 crores. (The Hindu Business Line, September 29, 2000). Like all ambitious plans, these too still lie on paper. 3

See end note No. 17 in the next chapter (‘The Political Economy of Public Utilities’) for details on this episode. 4

Thus the revenue account has always been in the red, the deficit often shooting up at stupendous rates, for example, in 1993-94, the Central revenue account deficit grew by more than 83 per cent over the previous year, and in 1997-98, by more than 30 per cent. On the other hand, the capital account has been made to register surplus since 1990-91, by cutting capital expenditure drastically relative to receipts; in 1993-94, the Central capital account surplus increased about four-fold over the previous year, and in 1997-98, about 2.25 times ! 5

In the revenue account, the developmental expenditures fell from about 55 per cent in 1980-81 to about 49 per cent in the late nineties, and in the capital account, from about 39 per cent to around 30 per cent respectively. 6

Over the high-pitched clamour for financial austerity, loom large the ever-increasing ‘jumbo-size’ Cabinets and the attendant lackeys both at the Centre and in the States, squandering public money at 63

will. In fact, the introduction and institution of Panchayat Raj governance serves only to decentralise such official profligacy and corruption with wider nets. Added to this is the populist extravaganza such as, for example, the recent freebies (free telephone facilities) from the Telecom Minister to all his Department employees ! And still the Government has no money for the most important power sector investments ! 7

It should be noted here that inflation in India in general seems to have been to a good extent Government-sponsored, through administrative price hikes and their spread effects (Pillai, 1995). The almost non-significance of Keynesian or Monetarist inflation in Indian economy, thus, needs a careful analysis. 8

“The NTPC was founded in 1976, and was pioneer in India in developing well conceived and documented systems and procedures for construction of power plants in record time and thereafter operating the plant at record PLF. It is therefore not surprising that NTPC annually added almost 1030 MW of new capacity in the first decade (1980-92) of its operation at a difficult period of the economy. It also had the distinction of achieving record annual new capacity additions of 2410 MW in 1987-88 and an average of more than 2000 MW new capacity additions in two consecutive years thereafter.” (Business Standard, September 22, 2000) 9

Again it should not be misconstrued that a Government department per se is fated to be inefficient. It is the inefficiency, in terms of lethargy, incompetence, non-commitment and what not, of the powers that be that is reflected through the department. 10

There have been allegations of corruption against the present leftist Government in appointing a Canadian firm as consultants in power sector matters in Kerala, in return for a Canadian loan. The same firm was given the contract for Kuttiady extension works which the firm subcontracted to some other local contractors! The works, started in 1996 and expected to be completed within 3 years are still on, the power station still remaining shut down! The Government has, however, allowed time extension and also sanctioned the demanded cost overruns to the Canadian contractor! The Canadian consultants have also been given extension with hefty payments in fees! Again, the KSEB awarded the maintenance works of Panniar and Sengulam projects to the same Canadian company, ignoring the recommendation of a panel (headed by E. Balanandan) that it should not be given to this firm (The New Indian Express, 24 September 2000). 11

Morris remarks in a footnote: “We cannot ignore the role of corruption and kickbacks in the preference for foreign projects. Nearly all high level government officials in informal discussions agree that kickbacks in foreign contracts have become the norm. Only the percentage involved varies…” (Morris 1996:1210). 12

Enron’s original Dabhol Power Project (Phase I) reported a capital cost of Rs. 4.48 crore per MW, whereas an indigenous NTPC project of similar type cost about Rs. 3 crore per MW that time. Enron’s cost was higher by about Rs. one crore per MW than that of a large number of IPPs for which MoUs were signed around 1995 (see Morris 1996). The recent kickback controversy kicked up in connection with the ministerial shelving away of a private (Kannur) power project in Kerala itself is a powerful example of the corruption potential of this area. The situation appears even more dangerously grim when one finds that this comes from a (self-styled) leftist Government. 64

13

The high capital cost includes, besides the back-door payments, high returns, to the tune of 16 per cent, on capital also. There have been criticisms from the SEB-circle itself that while the IPPs and even the Central sector generating projects are allowed 16 per cent on equity (in addition to highly attractive incentives), the SEBs are severely constrained by the Governments even in matters of earning the stipulated 3 per cent return. K.P. Rao Committee recommends that the SEB be allowed 16 per cent return to provide “a level playing ground”” for it (Government of Kerala 1998:30). It is an irony that while the IPPs are allowed entry on the plea that it (this situation) vis-à-vis SEB increase competition and hence efficiency, the SEB still remains constrained as a Government department, without having a free, level playing field! 14

In this regard, one would be reminded of a recent Supreme Court verdict in another context (Dr. Rajkumar kidnap case) that if a Government cannot tackle a problem with a firm political will and iron hand, wherever and whenever required, cannot be a Government de jure and should bow out of office. 15

It is reported that the National Rural Electric Cooperatives Association (NRECA) of USA is engaged in helping to form small cooperatives of consumers in villages and to transfer rights of distribution and transmission of electricity to them. This experiment has been a big success in Bangladesh and Costa Rica in recent times, and previously in the US also. In Bangladesh, the NRECA has helped to form 50 cooperatives serving 2.6 million metres. It has registered collection of nearly 97 per cent of billing. The growth rate also is impressive – some 1000 connections are added every day! (Business Standard, April 21, 2000).

65

PART IV

THE POLITICAL ECONOMY

In this part, we attempt at an analysis of the political economy of the Indian power sector with special reference to Kerala in the light of a generic model of the political economy of public utilities we develop in the first part of the paper. The model seeks to explain the political economy of the rent seeking drives in a non-Smithian imperfect regime of self-interest maximisation, with a regulatory structure of the public utility, described in a framework of the principal-agent relationship. In contrast to the usual neoclassical monolithic representation of principal and agent, we characterise each entity in a MarxianKaleckian vein, as a composite set of conflicting sectional interests. This helps us develop a comprehensive perspective of the politico-economic implications of the relationship among the public, government and utility.

Based on this generic model, we seek to analyse, in the second part of the paper, the political economy of the power sector in India, with emphasis on Kerala. We also attempt, wherever possible, to estimate the costs of corruption involved in the administration of the power sector.

CHAPTER 10 THE POLITICAL ECONOMY OF PUBLIC UTILITIES A Study of the Indian Power Sector

[The Jury acquitted Clodius of the charge of violating the rites of Bona Dea, wherein Cicero had given evidence against the alibi Clodius had set up.] “The Jury” sneered Clodius, “did not give you credit on your oath.” “Yes,” retorted Cicero, “25 out of the 56 did; the remaining 31 refused credit, for they took the bribe in advance!” - quoted in Thakur (1979:166).

1. Introduction

In most countries, intervention by governments in the economic sphere finds its justification on mainly two fronts: (i) to correct market failures in the provision of public goods and in the presence of externalities and natural monopoly, and (ii) to ensure the translation of the lofty ideal of a ‘welfare State’. Often a convergence of the two fronts is sought to be achieved, as in India, in the establishment of a public sector in a particular historical context. It is argued in a simple neo-classical framework that if the benefits of productive efficiency outweighs the costs of allocative inefficiency, then the society will have a welfare gain from maintaining the natural monopoly organisation of a public utility. Furthermore, if such monopoly power (price) can be brought down to a competitive level, say by means of its nationalisation, then there will be both an equity gain and an efficiency gain. This is the theory. However, the practice could be very different from the theory, as is often the case. For example, the vast scope for administering discretionary powers by the political and bureaucratic control processes involves substantial costs of rent-seeking activities in a non-Smithian cultural regime of self-interest maximisation.1 In the non-Marxian (more precisely neo-classical) representation of political process, the relationships among the public, government and utility may be aptly analysed in the light of a model of principal-agent problem. The problem consists in the default and breach of trust, likely on account of the 2

conflicting objectives of self-interest maximisation of the concerned parties and the uncertainty or information asymmetry involved in the relationship. In contrast to the usual neo-classical monolithic characterisation of the principal and the agent(s), a kind of Marxian structuring of each of the entities (especially the principal, the ruling class) in terms of composite sets of sectional rent seeking interests in the functional domains (of agriculture, industry, trade, and labour) may yield more insights into the relationships. These different sub-sets in the principal set vie with each other in appropriating the benefits of the utility, through their representatives in political power. The very same agents of diverse interests, on the other hand, strive to stand as a cohesive group to ensure the long run end of both the continuity of their own regime and the survival of the system, while catering to the short-term contingencies of clashing sub-class interests. The long-term common agenda of capitalist survival in turn requires pacifying class strains, the general cause of crises. And this could be ensured in general by way of a captivating welfare State slogan, sought to be materialised in terms of State intervention and nationalisation. Such drives often went to the extent of equating nationalisation with socialism; and the political process was projected to be managed by ‘representative governments’. Such ‘intermediate regimes’ (to use the Kaleckian term) were in fact an instrument for securing the class interests of capitalist empowerment, by ensuring both economic development and social security equations, true to the professed welfare State slogans.

The apparent fall of socialism and its emulations elsewhere has, however, opened up an impressive interpretation of the viability of economics: that there is no alternative to the capitalist mode of general welfare. Thus has started the new notion of liberalisation-based world welfare to sweep across the countries, replacing nationalisation by private sectorisation as a now seemingly viable stratagem in a historical necessity of capitalist survival. In the new dispensation of the on-going realignment of sub-class interests in India as elsewhere, the industrial and trade interests have risen to assert themselves. And in the political economy, corruption has scaled new heights in the implementation of privatisation drives, in addition to the old transactions in awarding concessions and contracts.

In what follows, we discuss these aspects in the context of the development of the Indian power sector. While the provocation for this paper arose out of our just concluded study of the power sector in Kerala, the discussion of its political economy here is firmly set in the national context. Though the reference to Kerala experience is largely illustrative, there is a significant dimension to it. In the development literature, Kerala is now well known for its remarkable achievements in human development despite low income, reflecting the continuing inability of the state to translate the social development into commensurate economic development. What the discussion in this paper points out is that the political

3

economy of government intervention in Kerala in the economic sector (as opposed to such social sectors as school education and health care) is not very different from that in most other parts of India.

The following discussion is divided into three parts. In the first part, we present a generic analysis of the political economy of public utilities in a new synthetic methodological framework. The present stage in the development of history is interpreted in a Marxian perspective as characterised by a series of seemingly feasible capitalist survival strategies. The class character of the State, on the other hand, is found more or less to obey the Kaleckian proposition of the ‘intermediate regime’, but largely dictated by the currently viable alignment of the sub-class interests in the common capitalist class. And in the superstructure of political economy of regulation, a neo-classical theorisation of rent seeking drives in an imperfectly co-ordinated domain of self-interest maximisation within a principal-agent relationship framework is adopted to obtain more insights. Such a synthetic analysis appears to provide a comprehensive and consistent explanation of the issues under study.

Leaning on this generic background, the second part discusses the plausible implications in the political economy of the Indian power sector, with special reference to Kerala. An attempt has also been made to estimate, wherever possible, the costs of corruption involved in the administration of this public utility in terms of the purely avoidable but allowed cost escalation. Finally, the third part gives our concluding remarks.

2. A Generic Analysis

The economic theory of public choice runs in the justification of market failure in the provision of public goods and in the presence of externalities and natural monopoly. Where market fails in the absence of preferences revelation, the political process steps in to obtain such revelation; ballot voting replaces rupee voting. Consumers as voters find it in their interest to vote such that the political outcome approximates their own preferences and choices. Such voting on collective tax and expenditure decisions reveals their choices in the determination of provisions of goods and services that price system cannot supply. Tax functions as a price here.

The natural monopoly problem Our concern centres on the market failure (and the subsequent political intervention) from the existence of natural monopoly. Traditionally, public utilities (such as gas, electricity, telephone, water, cable TV and waste treatment facilities) are defined in terms of the technical features giving rise to 4

natural monopoly position. A natural monopoly used to be interpreted as a single product, decreasing cost industry.2 However, most utilities are multi-product: electric utilities distribute high and low voltage power as well as peak and off-peak power, telephone industries provide local as well as long-distance call facilities, etc. In this context, a natural monopoly is defined under the cost conditions when ‘the cost of a sum of any m output vectors is less than the sum of the costs of producing them separately’ (Baumol 1977: 809).3 If this condition is satisfied, then the least cost method of producing the whole vectors of output is with a single firm. Hence the natural monopoly position.

Electric utility is unique in that its product is non-storable and must be generated and supplied the moment it is demanded. This technical characteristic in turn makes the industry essentially a vertically integrated monopoly with the co-ordination of all the three basic functional processes of generation (production), transmission (transportation to markets) and distribution (supply to final users) for reaping the full advantages of an integrated network system. This in turn gives rise to economies of scale and the resultant natural monopoly status.

The natural monopoly justification (in terms of productive efficiency) has however the danger of violating allocative efficiency criterion, as prices are set above marginal cost (MC). Ensuring allocative efficiency requires competition in the market that drives price down to MC. But too many firms flooding the market leads to productive inefficiency. This in turn opens up the fundamental problem of public utility economics, as to the choice of an appropriate institutional arrangement of governance structure that can manage to make use of these economies, but without the excesses of monopoly power, creating dead weight loss. The practical solutions to this problem were originally addressed in terms of two kinds of government intervention: (i) outright nationalisation (state monopoly) as in most of the developing countries, France and the UK till the end of the eighties; and (ii) regulation of the private monopoly as in most of the USA, where the former is considered an anathema.

The significance of nationalisation appears obviously overwhelming, once we recognise the scale economies associated with a public utility and accept the consequent natural monopoly position of it. This is easy to show, following Williamson (1968), in terms of a simple neo-classical analysis of a trade off between market power and scale economies. In Fig. 1, a natural monopoly, enjoying scale economies, is shown to be able to supply at a lower average cost ACm than a competitive, or more realistically, an oligopoly, utility at an average cost of ACc. If, under such conditions, the cost savings, given by the rectangle PcBDE, exceed the dead weight loss in consumer surplus due to the monopoly price (Pm), given by the triangle ABC, there will be a welfare gain from accepting the monopoly organisation of the utility. If, by nationalising the public utility, the monopoly power can be eliminated and the price reduced 5

to the competitive (or oligopolistic) level of Pc, then there would be both an equity gain and an efficiency gain, in addition to profit.

Nationalisation still involves possibilities of further price reductions and

P

Pm

A

Pc

B

E

C

ACc

D

ACm

AR MR

Q

Qm

Figure 1. Benefits of the monopoly organisation of electric utility and its nationalisation

increased gains. As we will see below, the significance of such public sectorisation is further fulfilled in materialising a welfare state.

Private interest theories Market economic theory of political intervention presupposes a demand for and a supply of it (Stigler 1971); a collective demand for government intervention to reduce the abuse of monopoly power leads to an appropriately chosen institutional structure for governance of monopoly. Broadly, two strains of themes have permeated the theoretical discourses of political economists in this regard: ‘public interest’ and ‘private interest’ theories. In the public interest framework, the social choice manifests itself in a political process that seeks to protect public interest or, more precisely, to maximise social welfare through an appropriate regulatory mechanism.

On the other extreme is a large spectrum of private interest theories of manifold hues and shades. Here the mechanism of political process presents itself as the manifestation of the combined effect of rent 6

seeking activities of various interest groups. Paralleling homo economicus, we have here homo civicus – the agents are assumed to be rational in their choices in utility maximisation. The interest groups seek to improve their own wellbeing by capturing and manipulating the administrative channels of the State’s resources of coercive power. The behaviour of legislators is determined by their desire to stay on in power and hence legislation is designed to maximise their political support. Since legislation involves redistribution of wealth among the subjects, interest groups compete to capture government by offering political support in exchange for favourable legislation4 (Stigler 1971; Peltzman 1976).

Bureaucracy, the administrative agency of the government to implement its policies, as an interest group to capture the government also is considered. It is assumed that this organisation’s production is not sold in the market place (Downs 1967) or equivalently, not sold at per unit prices, such that the compensation of the bureaucrats does not depend upon the success of the organisation, defined in terms of surplus of revenue over cost (Niskanen 1971; 1973). Hence it is in the interests of the bureaucrats to seek to maximise their own utility. They are capable, because of the one-sided advantage of better information they possess, of capturing their superiors (the politicians in power) to obtain budgets larger than those normally granted by the politicians in government if they were properly informed.

In another viewpoint, government itself is assumed to be a monopoly, treated as a leviathan maximising its own surplus (Buchanan 1975: Ch. 9). This involves a systematic bias in the fiscal system resulting in an over-expansion in the budget, born of the self-interested bureaucrats (Niscanen 1971) and politicians in power (Mackay and Weaver 1978), seeking to maximise their budgets so as to gain in income, influence or power, and to have larger staffs and perquisites. On the other end of this theoretical band is a view that takes governments as ‘inhabited by self-interested individuals’, i.e., as composite structures, instead of a monolithic one: ‘…these centres compete with one another in the production and supply of the goods and services demanded by citizens’ (Breton 1996: 13-17).

The very raison d’etre of self-interest prompted pressure groups is warranted by the fact that government intervention causes income redistribution through legislation, and that public policies in general are just the response to the rent seeking activities5 of private interest groups trying to change income redistribution in their favour. Rent seeking appears in the context of a contrived monopoly rent – a monopoly artificially created by imposing restrictions on potential competitors. Competition for the contrived monopoly rents in effect transfer them to the politicians in power who manage to create the monopoly by restricting potential competition through legislation. The lure of this power involving a rent from monopoly a la government in turn is the source of competition among politicians, i.e., rent seeking activities by them. Similarly, bureaucracy, the executive agency of the government, also enjoys monopoly 7

rents; hence rent seeking among prospective candidates to reach and secure powerful positions in the hierarchy (for example, excessive expenditures on (special) education to prepare and on other arrangements for civil service examinations: Tullock 1980).

The principal-agent problem An important contribution to the study of regulatory behaviour comes from the focus on the significance of information especially in a principal-agent relationship framework. In its simplest form of a vertical relationship, government, representing the public, is seen as the principal and the utility as the agent in its employ or under its authority. In an extended form, (e.g., of a three-tier hierarchy) the public stand as principal and government (regulator) as agent (i.e., supervisor) who contracts with a further agent, the public utility, to supply the vector of services. In its barest terms it is assumed that in a regulatory governance structure, the principal’s objective is to maximise some measure of social welfare, while the agent (utility) aims to maximise profit. Information asymmetry against the principal explains the raison d’etre of the agent who is better informed or better skilled. The divergence in objectives and the uncertainty or information asymmetry result in two effects6: moral hazard (principal being affected by ‘hidden actions’ by agent) and adverse selection (principal being affected by ‘hidden information’ agent has at his command) (Arrow 1985). Hence the principal should structure his contract (compensation scheme) with such incentive designs as to encourage the agent to expend the expected effort that will compensate the information asymmetry the principal faces in his maximisation objective.7

A new look into the principal-agent problem These neo-classical teleological representations of the political process underscores the positions of the players (people and government including politicians and bureaucracy) in distinct, disjoint, sets, ‘external’ to each other. Such ‘externalisation’ of the government from the people becomes sharper in a Marxist framework of the functional role of State. Where the society is divided into classes and subclasses, clashes of self-interests inevitably lead to demoralisation and alienation. Feelings of alienation permeate the whole texture of consciousness, individual and social, and this in fact explains all the cases of indiscipline in functionings and insincerity in responsibility and the consequent stress out of the mistrust that underlie the principal-agent problem. Neo-classical analysis, as is its wont, never turns its microscopic eye-piece to these fundamental strains. This is equally capable of explaining the much bitter experiences of alienation under the very ‘socialist’ regimes, where the agent, entrusted with the task of ensuring the ‘dictatorship of the proletariat’ remained entirely ‘external’ to and above the principal, the proletariat. A lack of a sense of oneness led to a principal agent problem, much sharper in the void of a civic platform of checks and balances, that would have avoided problems arising from moral hazards and

8

adverse selection. This should point towards the significance of an all-embracing cultural revolution preceding the political take over of the State.

A critical examination and exposition of the principal-agent model in this light in the context of an Indian-type public utility is in order now. In contrast to the usual monolithic characterisation of principal and agent(s), we find these entities as composite sets of sectional interests. The State being the objectification of the common capitalist class interests, this ruling class, at this particular historical juncture, stands as the principal. In actual realisation, however, the common interests have only long-run significance, if at all possible, in the face of the short-run contingencies of intense conflicts of intra-class interests (competition being a widely used euphemism) in the capture of the benefits of the efforts of its agent, the utility, through the supervisor, the government. The sub-sets in the common capitalist class represent the industrial, agricultural and commercial interests, aided by their own peripheral allies in the organised labour. It is not easy for the supervisor to wade through the chaos of such inter-/intra-class interests clashes. The politicians, who wield the State power qua government, have their own primary objectives of interests also, viz., their own survival and continuity in power, overcoming all the ruses of the rivals, to secure the long-run accumulation of the attendant benefits. This very objective in turn relies heavily on indulging the interests of the influential vote banks. In a predominantly rural economy, such as in India, large vote premiums are easily cornered by being in league with the farming capitalists, who can herd at their beck and call vast ballot blocks. This then explains the much pampered practice of holding the umbrella of subsidy policy over the agriculture sector in these countries; for example, the free or near free supply of power. This then becomes an easy option as compared with what we would like to call a developmental option through a policy of capital formation in the rural sector by way of an effective investment strategy (e.g., development of land and water resources), leading to structural transformation and growth of the rural economy.

Such milking of a public utility by a section of the composite set of the principal is at the cost of the others in the set. Cross-subsidisation constitutes a source of tension among the sub-classes of the principal and the consequent feelings of alienation. The supervisor i.e. the government, however, is able to buy peace by appeasing, through subsidised power sales, all these sections together, including the allies in the organised labour, in their capacity as domestic customers. Cross subsidisation in this respect is extended at the cost of some sections of the population; the net losers in this game, however, are the poor (non-electrified households). And the role of the State as a coercive instrument for exploitation is fully fulfilled here. This conclusion still holds even if we enlarge the major set of principal to encompass ‘the whole public’, in contrast to the above Marxist proposition. In this case the supervisor colludes with the beneficiary section of the principal and milks the public utility for them at the cost of the remaining poor, 9

who are seldom in a position to exercise their choice of ballot use independently. The supervisor requires such conditions only. Such actions of the supervisor are often given a veil of legitimacy through a notional subsidy offered to the weaker sections such as small farmers and rural households.

Nationalisation to fulfil the welfare State Both the early experience of the erstwhile Soviet Union as well as the influence of Keynesian economics gave nationalisation pride of place in State policy such that it soon came to be equated with socialism. The whole of Western Europe was swept over by a wave of nationalisation immediately after World War II, that set a precedent to a large number of countries elsewhere. Politically, such a ‘socialism through legislation’, in contrast to socialisation, was in fact a response and a counter to the socialistic (more precisely, quasi socialistic) conversion of Eastern Europe under the extended military umbrella of the then Soviet Union, though temporarily. It should be noted that these East European countries economically at best represented a symbiosis of plan and market, as envisaged by the European liberal socialists such as Oskar Lange and Maurice Dobb and this lent nationalisation in other countries an equal footing with socialism to the satisfaction of their elite proletariat and intelligentsia.8

The neo-classical theorisation of political process, while overlooking these undercurrents, was, however, confined to analysing their reflections on the surface. Thus the public sectorisation and the welfare State practices were explained (e.g., Mishra 1984) in terms of a number of forces that cumulated to exert an upward ‘ratchet’ effect on welfare programmes and expenditures such as political competition for votes, lack of cost constraints on voters’ behaviour due to the low salience of taxes, pressure of interest groups outside the State, notably trade unions and professionals, and the operation of budget maximising [politicians and] bureaucracies within the State.

In this background it was then natural for the developing countries, most of them rising from the colonial subjugation, to mould their independent economic life in the same fold of a welfare State. There was an added factor covertly accepted that too necessitated State control of economics in these countries – the lack of a developed capital market and the supply of entrepreneurs in the domestic private sector. Implied in this official recognition was a hidden agenda of treating the public sector as a temporary midwife for the development of private sector, and history has proved this later on. India has been one of the pioneers along this U-path.

The’ intermediate regimes’ What is the specific class character of the State that works behind this U-path development strategy? 10

The class that rose to power in many underdeveloped countries upon Independence after the World War II was identified by Kalecki (1967) as the ‘intermediate class; (that is between the capitalists and the workers), or more specifically, the lower-middle class and rich peasantry. A few specific conditions facilitated such an ‘intermediate dictatorship’, in contrast to the dictatorship of bourgeoisie and of proletariat. For one thing, the lower-middle class was ‘very numerous’, and the ‘big business’, comparatively weak and ‘predominantly foreign controlled’ (ibid.). This ensured the rise to power of the representatives of the intermediate class without being captured by the big business as was usual. For their survival in power, they sought to eliminate the ‘comprodor’ elements (i.e., to ‘gain a measure of independence from foreign capital’), as well as the ‘remnants of the feudal system’ (the latter by carrying out land reforms, meant to weaken the alliance of the big business with the landlords). This in turn necessitated ‘continuous economic growth’. Since ‘the native upper middle class’ was too weak ‘to perform the role of ‘dynamic entrepreneurs’ on a large scale’, the State came forward with the basic investment for economic development, in line with the universally accepted principle of ‘State economic interventionism’ of that era. Another favourable condition was the availability of credit finance for economic development. The intermediate regimes, taking advantage of the competition between the first (advanced capitalist) and second (socialist) worlds, behaved like ‘the proverbial clever calves that suck two cows’.

In this context it was in the interest of the intermediate class to promote economic development through the commanding position of the public sector. Ideally, ‘State capitalism concentrates investment on the expansion of the productive potential of the country’, that provides a luxuriant atmosphere for the small firms (of the ruling class) to thrive, without fear of concentration and centralisation of capital, under the watchful patronage of the State. The consequent economic growth opens up more and more avenues of employment ‘for ambitious young men of the numerous ruling class’. Such a sustainable growth dynamics in turn is logically to materialise a welfare State, put up as the end of the intermediate regime.

As interpreted by Raj (1973), the historical reality has however failed the Kaleckian ideal of dynamic State entrepreneurship, thanks solely to the conflicting self-interests of the numerous classes and their fringe allies in the ruling coalition. Political survival has entailed indulging the demands of some or other classes for subsidised supply of State sector outputs, resulting in accruing very little investible surplus or incurring losses. Moreover, the government also has been meant to support and cater to the pecuniary demands of ‘a large and growing body of salary earners whose contribution to economic growth may be negligible’ involving ‘further drain on the investible surpluses available to the State’.

11

Thus even though the intermediate regime was politically viable, its survival has been at the expense of the economic viability of its dynamic entrepreneurship.

The subsequent developments out of such fabricated dysfunctioning of the public sector in the intermediate regimes, leading in another conducive global environment to private sectorisation, offer a number of implications. Most basically, we find that the new order has facilitated the growth of the native upper-middle class (into independent big business), and the expansion of the small firms. Some of the lower-middle class and rich peasantry (the latter through diversification) also have been able to climb up to higher rungs of the capitalist echelon. It goes without saying that mergers and concentration have largely characterised the development of capital in these regimes, along with a peaceful coexistence by small capital too. Thus the intermediate regime had a hidden agenda in line with the rule of development of capital. This was so at least in the case of India, where, as Kalecki (1967) himself admits, big business, at the time of Independence, ‘was much stronger’, and could dictate its terms too. Simply put, intermediate regime was an explicit agreement among different, both small and big, strata of the capitalist class in alliance with organised labour for facilitating the survival and development of capitalism. Though the numerical strength of the intermediate class elevated its representatives to power, some live strings still attached them to both the extremes – big business and labour. Once the native capitalism has matured into imperialism (that exports capital in contrast to commodities), it along with the global capital in an environment of coexistence has now become able to capture the intermediate regime and to reinstate ‘classical capitalism’ (envisaged as possible by Kalecki 1967). This explains the U-turn strategy.

3. The Indian Power Sector

The welfare State concept had its fervent devotees in India too. If Gandhiji talked of ‘Ramarajya’, then Nehru talked of a ‘socialistic society’. Both the Congress Party and the splinter groups of social democrats talked of nothing else. The coexistence of market and plan continued to be the official economic principle of Independent India under the Congress, along with an import substitution model of development for building a strong industrial base, both for basic and intermediate goods and for heavy machine building. And the famous 1955 (60th) Avadi session of the Congress party adopted a ‘socialistic pattern of society’ for India, through planning, ‘where the principal means of production are under social ownership or control, production is progressively speeded up and there is equitable distribution of the national wealth’ (from the Resolution, quoted in Zaidi and Zaidi 1981: 52). Following this guideline, the Industrial Policy Resolution of 1956 shifted the primary responsibility for development on the public sector and demarcated and reserved the core and strategic areas exclusively for the State. 12

In intermediate regimes the State machinery has had to indulge not only the sectional interests of very numerous classes, but also those of the very class representatives (as an external set) wielding power. Nationalisation, for example in India, was at times more a profitable political agenda (in addition to an additional outlet of rent seeking) than an economically viable requirement. As in the case of other infrastructure facilities with high capital intensity and long gestation period, that deterred large scale initiative of private enterprise, the responsibility of power development also was thus originally shouldered by the State in India. The sector, rightly expected to subserve the social, political and economic policies of the State, soon became in effect the translatory channel for the populist policies of the political party in power in the various provincial states in the pursuit of votes.

Ilfare of the electric utility In such a context of composite interests, accumulation of disfunctionings has been a natural outcome in the power sector. For one thing, subsidised power, in the name of swift industrialisation, has gone to turn the factory machines, at the behest of the capitalists, who could, say, finance election campaigns. The true price, however, of such industrialisation drive, for example, in Kerala, was very high, as most of the industries sprung up in the state under the subsidy umbrella were capital- and energyintensive, with very limited prospects for creating employment opportunities, one of the professed objectives of the drive (Kannan and Pillai 2001).9 Highly pampered by the subsidised power sale also are the domestic customers, the influential section among the electorate. By seventies the logic of power subsidy was also extended to the agricultural sector to placate the powerful lobby of rich farmers. As already noted, the fact that in general the non-electrified households and fragmented farms belong to the poorest of the society questions the justification of the welfare content of such across the board subsidy to the powerful groups.

The appeasement strategy on the part of the supervisor (government) in favour of the influential section of the principal (the public in general) has a downstream extension also towards the bureaucracy of the agent (the public utility). This is especially ensured by the powerful trade unions, the loyal affiliates of one or another of the supervisor, and hence the collusion is a given fact. The appeasement appears more prominently in overmanning, especially and unwarrantedly in establishment and administration (E & A), in a populist bid of employment generation. The consequent increase in E & A costs stands in turn to inflate the supply costs of electricity and penalises directly the customers and further indirectly the poor tax payers. In Kerala, such collusion has had increased effect, as we already noted in an earlier paper (ibid.), not only on the extent of overmanning, but also on the average annual earning per employee, which was Rs. 1.44 lakhs in 1997-98, about 1.8 times the state-sector average in 13

India. That is, the cost of such collusion was about 80 per cent higher in the power sector of Kerala with a highly militant labour.

Alignments of sectional interests We have enlarged much upon the effects of the various sectional interests in the composite set of principal vis-à-vis a monolithic supervisor. That electricity supply is included in the concurrent list of the Constitution in India renders the supervisor also a major set of sectional interests, those of the Central as well as of the regional state. Such a regional distinction of the interests of the supervisor has assumed greater significance only recently, in the context of the accelerated Central drives for power sector restructuring in India, as there was little scope for conflict of interests till the end of 1980s. The power sector reform programmes in the version of the Central government are not at all acceptable to many of the state governments, at least for the populist slogans against any restructuring of the sacrosanct public sector. In fact, it is in view of such conflict of regional interests in the functionings of the supervisor, that proposals are floated now for transferring electricity supply to the Central list in the Constitution.10 The Electricity Bill – 2000 may be taken as a first big leap towards that direction.

It is not surprising to find, in the present phase of the capitalist survival strategy, that what underlies the Central government prescription of power sector restructuring is a realignment of sectional self-interests. The vociferous demand and stringent stipulation by the Central government for phasing out the subsidy regime altogether is a clear indication of this. The domestic industrial capital in league with the global capital is in the process of capturing the supervisory organ of the State. In the power sector, the proposed restructuring absolutely favours the industrial capital – besides the benefits from the removal of the burden of cross-subsidisation, it can also reap cheap power upon marginal cost pricing practice, as supply to high voltage industries involves much less T & D network costs and loss factor costs than that to low voltage domestic and rural sectors. Putting an end to the costly subsidised power sale might appear as an attractive proposition of reason and justice on the surface, as it breaks the unfair nexus between the supervisor and the rich rural lobby. Though it seems desirable from that viewpoint, the process, however, would have long-run detrimental distributional effects as far as the public development issues are concerned, as we will see later on. Moreover, the new configuration of collusion must also arouse a natural doubt as to the rationality of the supervisor relinquishing his sure rural vote base. But the much impressive TINA (There Is No Alternative) factor involved in the reform proposals, coupled with the diverse vote purchase mechanisms is expected to safely take care of this problem.

14

The corruption channels: energy theft Along with the force of vote-premium aligns high-powered corruption. As the bureaucracy (of the public utility) enters, the scene gets complicated with implications of collusion of different configurations. One major combination is among the Board officials and erring customers, enjoying political patronage, for uninterrupted theft of power flagrantly practised, for example, in the very capital city of the country. Corruption greases a smooth relationship here, so long as its cost (bribery, in-kind transfer, favouritism) remains less than the value of power drawn behind the meter. The effect of collusion transcends the direct end of transfer and spreads in the whole functional veins of the Board; it credits, without verifying the authenticity of its own procedures, the power thus lost, to the farm sector, where consumption is mostly unmetered. As we have already found (ibid.), about 30 to 40 per cent of what is usually reported as agricultural power consumption in fact represents power lost in such illegal ‘sale’. Then assuming, quite reasonably, that the actual agricultural consumption is only 65 per cent of the reported one, we have already estimated that in 1997-98, the energy thieved away in connivance with all the State Electricity Boards (SEBs) amounted at least to 31073 MU, equivalent to Rs. 5733 crores, at a sales rate of Rs. 1.85 per unit! This, though an underestimate, gives in effect an annual cost of corruption at only one (i.e., sale) end in the Indian power sector.

A further heroic assumption here is that the reported T&D loss is the actual one, a good part of which in fact is ‘theft and dacoity loss’. Thus the actual value of corruption at the sales end must be much higher than our rough estimate. For example, with the assumption of an actual 15 per cent T&D loss in 1997-98 (including technical and inefficiency loss due to inadequate transmission capacity), the total cost of corruption at the sales end in the Indian power sector comes out to be a staggering Rs. 10705 crores!11 In the case of Kerala power sector, where agricultural power consumption is mostly metered and accounts for only about 4 per cent, such illegal ‘sale’ of power is included directly in the T&D loss. Assuming an actual 15 per cent T&D loss in 1997-98 in Kerala against the reported 17.9 per cent, we find that the cost of such corruption amounts to Rs. 33.7 crores.12

A more sophisticated way of ‘theft’ is to get the officials to allow huge energy bills to mount up and then to write them off as ‘bad debts’.13 In our study mentioned earlier we have seen that the revenue arrears outstanding against different consumers for all SEBs in 1996-97 was Rs. 11,535 crores, accounting for over four months’ sales revenue, against the maximum allowable norm of two months’ sales revenue. The excess of outstandings over the admissible norm may then be taken as an approximate measure of the cost of corruption involved at this end of energy ‘theft’ in the form of deliberate nonpayment by customers of electricity charges in connivance with the officials. This amounted in 1996-97 to Rs. 4220 crores, equivalent, as we have estimated earlier, to the additional revenue at hand if all the 15

SEBs could limit their revenue arrears to two months’ sales norm, and in 1995-96, to Rs. 7364 crores! In Kerala, corruption on this front cost Rs. 175 crores in 1995-96, and Rs. 198 crores and Rs. 252 crores in the next two years! The ‘bad debts’ written off during these three years by the KSEB were Rs. 11.8 crores, Rs. 12.5 crores, and Rs. 14.8 crores respectively.

Corruption at the high up The government (i.e., the political party in power) also falls within the collusion circle to the extent that it condones such theft without striking its coercive authority properly, lest that ruin the other ‘side contracts’ it has with the erring Board and customers. The government seeks not only to maximise its vote base to secure its survivability through populist administration in the best interests of the capitalist State, but also to gain income for its individual coalition partners. Ministry formation is in fact a ‘rentsharing’ side contract in proportion to the bargaining strength expressed at the time of reaching the coalition contract. Every ministry or department has its own illegal inlet of income or ‘sale counter’ for the concerned minister and his coterie; thus industrial concessions are sold at a price; so is a new college or liquor shop licence. In the power and irrigation sectors, construction contracts and purchase orders are conferred at a price.14 By the ‘side contract’ of collusion, the Board bureaucracy may share in the price along with its supervisor, the particular ministry in the government, or gain other favours of larger budgets. There have come up a number of allegations of corruption involving ministers15 and bureaucracy in the Kerala power sector. Some of them have recently been convicted also. For example, a former Minister along with his power secretary and some top officials of the KSEB were convicted in a case involving award of construction contract.16 There have been allegations of corruption against other power minister(s) in recent times too; for example, on the contract with the Bharat Heavy Electricals Ltd. (BHEL) in the case of Kozhikode and Kasargode diesel power plants, and with a Canadian firm in the case of an extension scheme (Kuttiady hydro-power project). The infamous Kannur-Ennore episode is another apt case in point here.17

Purchase of materials and machinery, especially power generating equipment, involve large scale corruption, the scope of which has widened since 1992 (post liberalisation period) with the stipulation for bilateral credit options that necessarily involves purchases from foreign equipment suppliers, as bilateral credit is inevitably tied.18 An Enquiry Commission in Kerala has indicted another former power minister and his officials in the case relating to financial irregularities, involving a loss to the KSEB of Rs. 75 crores in the award of contracts in the case of the Brahmapuram diesel power plant; a vigilance probe is in progress into this case.19 At national level, the infamous Jain hawala revelations have indicted a large 16

number of Central government officials, about half of them being from the power sector -–the NHPC (Dulhasti project) and the NTPC (many bilaterally funded projects) actively involved.20

Wide-spectrum corruption As ‘rents’ increase, the bandwidth of collusion also widens to include the contractors and the trade unions, in addition to the government and its agent. An apt example of such ‘wide spectrum collusion’ is the large corruption involved in allowing for time overruns of projects and sanctioning the associated cost escalations. Recurring unrestricted labour militancy is recognised in general as the single factor that puts the heaviest burden on the pace of the construction works of power projects in Kerala, largely dictated by party-political rivalry rather than genuine labour demands, as for example, in the construction of Idukki hydro-electric project, to begin with. The time overruns out of the striking militancy upon one or another pecuniary pretext essentially go into the contractors’ demand for cost escalation, that is soon endorsed by the Board and sanctioned by the government.21 Such rent-sharing is a widely recognised official practice in the power-irrigation sectors. The glaring laxity on the part of the government in fulfilling its committed responsibility for enforcing its authority on the contractors and workers to bind them within the contractual terms they agreed to take up to honour is a clear indication of its corrupt collusion. In Kerala, the time and cost overruns have afflicted only the State power projects; the public sector NTPC thermal and the private sector hydro projects in the State having been completed well within their scheduled times. In this light, then, the cost escalation sanctioned for each late-run project may rightly be taken to represent the cost of corruption involved in construction contract sales in the power sector of the State. Accounting for the general price inflation during the normal construction period, this amounts to Rs. 644 crores or Rs. 36 crores per project!22 Unbelievably, it represents on an average about 60 per cent of the actual project cost! In some cases it is well above 75 per cent. This is all shared among the four parties involved, at the cost of the helpless majority in the ‘principal’ set of tax payers.

At the top of these ‘milky’ projects is an extension project (Kuttiady hydropower project phase 1, works on which started in 1994),23 with a corruption cost of nearly 80 per cent of the actual project cost. In this case, it should be noted that so much capital cost inflation was allowed by the government not for a new project, but on an extension project for 50 MW only! A new hydropower project as per current estimates is expected to cost around Rs. 2.5 crores per MW, whereas this extension project has cost Rs. 3.96 crores per MW! (Government of Kerala 2000). The contractor for the project, SNC Lavalin International Inc., a Canadian firm, has been involved in a number of controversies and a corruption case against them is under vigilance probe at present. It is worth mentioning that even while being under the clouds of a corruption case, the very same foreign contractor was awarded the modernisation works of 3 old hydropower plants, viz., Pallivasal, Panniar and Sengulam, and that too through a MoU only, without 17

calling for international tenders as per guidelines! Current estimates put the costs of such modernisation works at Rs. 1.25 crores per MW, whereas the contract to SNC was given at a cost of Rs. 2.42 crores a MW (Rs. 280.5 crores for 115.5 MW of the three plants) (Malayala Manorama daily 7 February 2001). Assuming the validity of these estimates, a new hydro-plant of more than 110 MW could be constructed at this cost!

Corruption-inflated capital costs Corruption lodged in inflated costs of power projects24 is not an India-specific phenomenon, though allegedly materialised since the entry of Enron. The Dabhol power project of Enron has cost about US$ 2830 million (US$ 1.4 million per MW) as compared with a cost of US$ 1200 million (US$ 0.64 million per MW) for a similar plant, the 1875 MW Teesside project of Enron in England, (Mehta 2000: 98) i.e., more than twice! This works out to be Rs. 4.48 crores per MW, much higher than the NTPC’s 645 MW gas-based Kawas project (implemented in November 1993 at Rs. 2.32 crores a MW), which comes in effect to Rs. 3.56 crores per MW only at an assumed inflation rate of 10 per cent by March 1997 (Morris 1996: fn.2). The National Working Group on Power Sector (1994: 14, table 5.3), a motley organisation of left economists, trade unionists and former heads of SEBs, has shown in a detailed study that an Indian alternative combined cycle gas turbine (CCGT) plant will cost only Rs. 3.05 crores per MW and a coal-based one, much costlier than a gas-based plant, Rs. 3.13 crores per MW only. The calculations of the Central Electricity Authority (CEA) have put the total capital costs of an Enron-type plant at Rs. 1.91 crores a MW (December 1997 completed costs). However, the Ministry of Power (MoP) has, since the Enron entry, been justifying higher capital costs of power projects in India; for instance, while the CEA has estimated the cost of the Bakreshwar thermal plant at Rs. 2.91 crores per MW, the MoP has put it at Rs. 4.36 crores per MW.25 The ministry has, moreover, put out a list of projects with final costs of Rs. 4 crores to Rs. 5 crores per MW, and has thus sought to justify clearing private sector projects costing Rs. 3.28 crores to Rs. 5.09 crores a MW (The Economic Times (editorial) 22 March 1994).

The projects in Kerala too have become heavily loaded with inflated capital costs, as shown above. The very high capital costs allowed to SNC Lavalin by the government through a MoU only in the case of both Kuttiady extension (phase 1) and the Pallivasal-Panniar-Sengulam modernisation schemes should now be compared with the capital costs quoted in an international tender bidding for Kuttiady extension phase 2 project. Among the four companies left in the fray, the lowest bid has come from a consortium of two Indian companies L&T – BHEL) at Rs. 164 crores, while the highest from the SNC Lavalin at Rs. 324.4 crores! (Malayala Manorama daily 7 February 2001). It should be noted that the Board’s own estimated capital cost for the project, recognised by the government itself, is Rs. 220.5 18

crores, i.e., Rs. 2.21 crores per MW! The 1991 project report of the KSEB has estimated the cost of machinery at Rs. 170 crores, while BHEL has promised, in its letter to the chief engineer on 2 February 1998, to supply the items at Rs. 51.5 crores! In terms of machinery cost itself thus there is a gain of Rs. 118.5 crores, more than worth another two sets of machinery, in addition to the obvious benefits of encouraging indigenous production and supply.

Corruption bursting Corruption remains untraceable if the bond of collusion is sticky and stable. The probability of its disclosure or detection is a function of the ‘incentive’ a participant has to confess or to report. Though it may look unlikely, the benefit of being approver inviting less penalties or a vindictive tactics, both possible in a prisoners’ dilemma framework, may induce a party.26 But the more potent tool rests with a government itself for digging out, if it wills, evidences of corruption engaged in by the previous rival government – the old files and records with the ministry and the Board would speak volumes for the dubious circumstances27 the then authorities created for possible corruption. It was in this manner, as already explained, the active roles of two former power ministers in Kerala in large scale corruption could be brought into light by judicial enquiry commissions. Though characterised as an unscrupulous vindictive move to ‘finish the rival’, such competitive exhumation, set in eruption in Indian politics recently, in place of the earlier tacit collusive condonation, is in fact a welcome sign. Such much needed competition between governments, in line with competition in the economic sphere induced by liberalisation that is assumed to do away with unfair practices, can act as a powerful deterrent to corruption, by filling the rivals with a fear that exhumation might lead to political extinction, though it can also lead to ingenious ways to whet efficiency in corruption practices. At the same time, this involves dangers of a deliberate kill, as a government, through its coercive power, can ‘create’ evidences of corruption against a rival for its short-term gains.

It should, however, be added here that the too-slow pacing of the Indian judicature along with its multi-tiered appeal-provision hierarchy28 sets the convicts free in effect and leaves the system, more often than not, farcically ineffective as a deterrent to corruption as well as crime.

Moreover, corruption bursting a la exhumation will not work, if the new government finds scope for rent-extraction from and further rent sharing with the briber-contractor of the previous regime; the need to protect the briber protects the previous bribee too. The active presence of SNC Lavalin in the Kerala power sector even now despite a vigilance case against it is a case in point. SNC came to Kerala on 2 April 1993 as consultant to a hydropower project (Lower Periyar) upon the instructions of the World Bank, the loaner to the project. Sickened at the prolonging time overrun of this project, the Bank stopped 19

its assistance and quit the scene in December 1994, which should have in effect automatically terminated the contract with SNC. However, it was reported that the technical and finance members of the KSEB, without waiting for the full Board decision, extended the contract and paid SNC US$ 43 lakhs against the World Bank recommended payment of US$ 30 lakhs. A vigilance probe was ordered later on into this financial irregularity that had caused a loss of US$ 13 lakhs to the Board. But the new government also was soon ‘captured’ by SNC, as evidenced by the later developments of fresh contracts ‘procurements’ by them – Kuttiady extension and Pallivasal-Panniar-Sengulam modernisation projects. And the vigilance probe moves at its legendary snail’s pace, now having covered more than four years of no result!

Private Sectorisation It is common knowledge that the strong waves of liberalisation started to sweep across the world along with the fall of socialism. Public sectorisation had given capitalism a new lease of life in the face of threats from a flourishing socialism. But the disintegration of the socialist bloc, the raging discontent that resulted in massive popular uprisings against the (quasi) socialist regimes in the Eastern Europe and the costly inefficiency of unaccountability that characterised the public sector in general were all detracting from the vitality of socialist slogans and had the makings of a new twist which the stagnant history badly needed. The capitalist survival now required a new strategy of global expansion, facilitated by a variant of laissez faire. The Thatcherite drives of private sectorisation in the UK, projected under a colourful TINA banner, and the deregulation bids in the US were powerful political over-fishing in the troubled water of a vacuum of reliable pro-people alternative. In this powerful sweep, the public sector in India, that was apparently qualified as having fallen from the ‘commanding heights’ to the ‘demanding depths’ of inefficiency, too was soon marked for market. Conveniently concealed in this hasty decision making were the obvious evidences of the socio-economic development India had been able to achieve through the ‘commanding’ distributive channel of a public sector.29 The official machinery was only keen to magnify the dark specks of avoidable functional inefficiency of a number of public sector enterprises (PSEs) in a concerted bid to justify privatisation. Even an influential section of the informed atmosphere of India appeared alarmed over the apparent low productivity of the public sector (for example, Bhagwati 1993), without caring for locating its sources for possible cures, and the functional inefficiency, still avoidable or easily curable, was identified with structural/organisational deviations. This made it easier to reach a foregone conclusion in favour of restructuring. Thus, starting with the sixth Plan (in the early 1980s), private sectorisation in India got the full thrust at the cost of public sector with the eighth Plan in the era of liberalisation, privatisation and globalisation, the new form of the cyclical survival tactics of the capital.

20

Corruption in privatisation In addition to major contracts and concessions, liberalisation has opened up another avenue of corruption, that is, privatisation. This has been the single largest route of payments that has pushed the transition economies (the countries of the former Soviet Union) to the highest level of corruption in the world.30 China too is not an exception in this respect of its reform drives.31 An explanation of corruption in transition views privatisation and liberalisation as market oriented reforms that allow rent seeking and corrupt practices to proliferate. Corruption in transition is the result in particular of the incipient nature of restraining legal and other regulatory institutions (see e.g., Weisskopf 1992). However, this ahistorical explanations, ignoring the roots of corruption (the Czarist Russia was one of the most corrupt nations, Massie 1980) fails to reflect upon the conditions in the transition economies in the period immediately before and during the break down of the socialist system that acted as an ideal medium for the growth of ‘monetary corruption’. The New Institutional Economics views corruption in transition as a continuum from the past, thus recognising the legacy of corruption under the socialist system. Since institutional changes occur slowly and incrementally, history may be taken as a predictor of the continuing patterns of corruption during the transition (Feige 1997). Though historical path dependency is an important framework of explanation of corruption, the experiences of the transition economies had an immense and immediate (‘big bang’) onset of political and institutional changes that were by no means evolutionary nor incremental. The changes imposed through ex cathedra proclamations in fact disregarded the need for a conscious development of a rule of law to substitute, while unleashing the reins of order of the old regime. And the resultant chaos filled the vacuum with full corruption.

The transition in India on both the occasions (of initial public sectorisation and the later liberalisation) was however within the confines (of a modicum) of rule of law.32 But the historical path dependency of corruption still stuck in the inevitable loopholes of rules.33 Unlike the Western industrialised nations, India (and other developing countries) could not pass through a character-smelting cultural revolution in the progressive phase of the development of capitalism. Just as colonialism in these countries had found it profitable to prop up the corrupt cultural vestiges of the old feudal system, so did capitalism too. Thus corruption continued as if determined by a historical necessity for the State.

In the initial period of moulding a socialistic pattern of society under a sacrosanct planning system in India, almost every economic activity had some contact with control and regulation. Where control and regulation were tighter and grew in complexity, political and bureaucratic discretion in administering controls naturally involved an increasing scope for rent seeking (Government of India 1964: 7-8). Though Gunnar Myrdal (1968: 942-943) believed ‘on the basis of scanty evidence’ that India, ‘where a moralistic attitude is especially apparent’, might ‘…on the balance, be judged to have somewhat less corruption than 21

any other country in South Asia’, Santhanam Committee (on Prevention of Corruption 1964) found corruption as an increasing function of economic controls in the Indian planning system.34 Krueger (1974), who formalised the notion of rent seeking, estimated the annual welfare costs of rent seeking on account of price and quantity controls in India to be about 7.3 per cent of the national income of 1964, ‘judged large relative to India’s problems in attempting to raise her savings rate.’ (p.294). Following the same ‘procedure of approximating rent seeking costs by the value of rents created by controls’ in the external sector, capital market, goods market (including agriculture) and labour market, Mohammad and Whalley (1984) however found the cost of rent seeking to be approximately 30 to 45 per cent of the national income in 1980-81.

To the extent that politicisation of economic activities through control regimes results in vast scope for corruption, its antidote is sought in depoliticisation. Thus liberalisation, while putting an end to administrative discretion35 of control raj over the private sector in principle closes down the associated avenues of corruption too. Where control is over ownership rights (the state sector), depoliticisation of economic activities entails privatisation, i.e., conversion of control rights, involving discretionary political and official actions, into private, market-driven choices, supposedly free of corruption. But an important question, not at all addressed in this respect is: are the benefits from elimination of costs of control rights greater than the costs of profit-driven private choices? Here we concur with the liberal socialists who would still say: ‘Officials subject to democratic control seem preferable to private corporation executives who practically are responsible to nobody.' (Lange and Taylor 1938: 109-110). This is especially so in a country like India where the private sector is characterised by an absence of transparency in its functioning, let alone that of its susceptibility to social control. The infamous dysfunctionings of the capital market with the corresponding predatory behaviour of its actors lends sufficient credence to such a view.

Privatisation, transfer of control rights, is expected to reduce corruption, but the privatisation transaction itself can be corrupt in the same way as in the award of concessions and contracts. The prospective buyers may vie and pay for getting included on the list of pre-qualified bidders as well as for restricting the number of other bidders. Rose-Ackerman (1999: 35-38) illustrates three more corrupt practices in this respect. (1) In the absence of a scientific method of valuation of assets of the state enterprise marked for privatisation, the uncertainties of the process can facilitate scope for insider plays. The favoured buyer can easily procure information not available to others, or much earlier or reserve special treatment in the bidding process. He can even get the assessment process corrupted in his favour by having assessors of his choice get the bid and do the work. (2) With no assets evaluation criterion to rule, corrupt officials can under-value a state enterprise in return for pay off. The firm may be presented 22

as unhealthy and its prospects, feeble such that the favoured buyer can outbid others. (3) The prospective buyers would be keen and ready to pay more to retain whatever monopoly power was available to the state firm. ‘To an impecunious state and its bidders, assuring monopoly power is in the interests of both. Thus the conflict between revenue maximisation and market competition arises for all privatisation deals. If a state gives lip service to competitive principles, however, it may be unable to endorse monopolisation openly. Corrupt back-channel deals can then accomplish that objective…’(ibid.: 37).

Privatisation of electricity sector in the Indian context is obviously ominous of disaster. The assets of SEBs are highly under-valued; the gloomy presentation of a sick SEB would further cut into its value. Howsoever professedly meticulous the assets valuation rule(s), privatisation would thus amount to a cheap sell-out. The very high corruption potential would just add to this woe. The whole assets, accumulated by two generations of tax payers over a period of half-a-century, would thus be lost for a one-time paltry payment to the then government to squander.

Moreover, privatisation of the electric utility necessarily involves the problem of retention of monopoly power of some degree, as history amply shows. Manzetti (1994) argues, among other cases, that the privatisation of electricity industry in Chile involved such (unfair) deals that could generate monopoly rents for the winners. For another example, the two major generators in the English electricity supply industry, viz., National Power and PowerGen, had enjoyed sufficient market power in the Pool to raise prices and make supernormal profits (Green 1999). Rent seeking costs, related to such monopoly power retention processes, as explained earlier, are necessarily accommodated in higher market prices. In the English electricity supply industry too, as elsewhere, the increased cost was passed on to small consumers (ibid.).

The drive for power sector reform in India has been opening up a vast field for corruption in which the international lenders too have been eager to claim their stakes. Such experience comes with its rude shock from Orissa itself where the World Bank has been a major party to misappropriate and squander a good part of its structural adjustment loan to the state in the name of consultancy fee, service charges, and so on (see box 1). The Government has been forced to opt for foreign firms, instead of capable indigenous firms, as consultants in the reform programme, in violation of guidelines. Crores of rupees have been drained away into the consultants’ coffers, of course with a part of it re-channelled into some domestic pockets also. The same is the case in almost all the States, whether or not the Government in power is keen on implementing any reforms at all. Even in Kerala, that is dead against the so-called power sector reforms, there have been much heated

23

allegations of corruption in respect of appointing a Canadian firm (SNC Lavalin) as consultants on ‘power sector reform-related policy matters’.

Yet another disastrous consequences comes from the fact that a private enterprise system necessarily works on exclusion principle. The vast scope for lodging all sorts of large scale rent seeking costs in over-capitalisation stands pretty well to inflate supply costs that can exclude a sizeable proportion of consumers with limited purchasing power. Higher incidence of exclusion would be one of the deleterious social costs of private sectorisation in a poor country like ours, leading to increasing or excessive inequality, both individual and regional. As argued by Galbraith (1998), though in another context, the process, beyond a certain indefinable threshold, may become cumulative and unstable, and is likely to result in a loss of community and social coherence. And all this is in addition to the wasteful expenditures and transfer of resources. Moreover, such capital cost inflation confounds the very problem (viz., allocative inefficiency) presumably intended to be solved through privatisation.

There is no TINA force! As nationalisation of natural monopoly ensures both productive as well as allocative efficiency and equity, a vertically integrated monopoly organisation of electric utility in the public sector remains a foregone conclusion. However, an atmosphere of warring sectional interests out to capture benefits along with a conducive regulatory policy of populism has contributed to a mismanagement syndrome in the case of most of the SEBs (Kannan and Pillai 2001). Their functional inadequacies and financial infirmities, though entirely avoidable, have come in handy for a mis-characterisation of the whole sector: the costly dysfunctionings are unreasonably identified with economic inefficiency, which in turn is associated with the standard notion of some market structure devoid of competition. As already explained, this inevitably makes restructuring in favour of privatisation seemingly desirable. Behind this work informed attempts unfortunately organised to focus solely on aspects of allocative efficiency to justify the move. For example, there are strong arguments that technological advancements (such as combined cycle gas turbine (CCGT) plants of smaller size and shorter gestation periods) render the natural monopoly in generation sector irrelevant and hence competition for allocative efficiency is possible in that sector – both competition for market (initially in setting up plants, given a corruptionfree franchise bidding mechanism) and competition in market (later on during operation, given a highly efficient ‘tatonnement’ agency) are postulated to be possible. The distribution sector, though purely a local monopoly, also is proposed to be compatible with competition for market. However, the invariable location specificity of plants other than CCGT ones and the asset specificity in the transmission-distribution sector still leave the system predominantly a natural monopoly and its nationalisation does ensure increased gains in both equity and efficiency.36 It is at the cost of these 24

gains and with higher (transaction) costs of co-ordination and regulation that the hypothesised competition is being sought.

The cunning generalisation of the experience of performance disorders of some of the PSEs has been at the cost of the name of other well-functioning ones. In the power sector itself, the National Thermal Power Corporation (NTPC) continues to be a star performer by world standard.

The

Maharashtra SEB (MSEB) had been adjudged as a model for other SEBs in both physical and financial performance till the entry of the Enron through the openings of liberalisation. The MSEB’s encounter with the Enron illustrates the potential disaster involved in the new policy (see boxes 2 and 3). At the same time, this invalidates the already unfounded claims for liberalisation as stemming out of a TINA force of economic inefficiency; in fact the Enron a la liberalisation has been instrumental in inducing systemic inefficiency into the MSEB. Moreover, the glaring examples of the PSEs with golden track records have already refuted such TINA force argument. And this becomes evermore obvious as the Indian government is feverishly engaged in selling out only the profit making PSEs, for example, the Bharath Aluminium Company (Balco). If privatisation is thus resorted to not on account of economic inefficiency and out of a relevant TINA force, then, naturally a possible explanation is to be found in the vast scope for corruption in it.

4. Concluding Remarks As already explained, the socio-economic development India has achieved over the last half-acentury owes entirely to the commanding heights of the public sector, even with all its failings and corruption channels. If the vested interests were restrained and government intervention moderated in its intentions, the sector could bear more fruits, as expected in a poor developing country like ours. When public sector plays a crucial role (including in power supply) even in countries like France, Canada and Scandinavia, in defiance of any TINA force, its significance in poor countries goes without any argumentation. The primary concern in these countries should then be to restore its relevance to this sector, not to restructure it. In its self-rejuvenation, the public sector can adopt and use certain functional/behavioural traits from the private sector in a competitive environment of coexistence.

The private sector stands to sustain on account of its assignment-specific accountability of each and every agent in terms of measurable productivity. There is little scope for free rider plays in such arrangement of return-related hire-fire rules. This along with an appropriate mix of interactive policies of coercive compulsion and incentive compensation can yield an efficient outcome to the desired degree, despite some covert principal-agent problems. To the extent that such productivity-linked 25

labour contracts based on superior selection procedure have nothing to do with the nature of property rights, there can be nothing to bar its application in the public sector too, barring the absence of a strong political will on the part of both the public and the government. This is the sure recipe for an efficient public sector. After all, if the government can relinquish its authority over public sector in favour of private parties, there is no reason why it cannot do that over its costly populist policies in favour of an efficient public sector for the public good.

In concluding, let us re-stress the role of effective Government intervention in the interest of common good. The emergence of governmental authority in the history of the development of the social relations of the mankind signified the significance of common good over individual interests, though later on the institutional intention got tainted by the power of private property rights. The enlightened rulers of the ancients were expected to identify their own individual interests with common interests and to rule accordingly. At a progressive stage of the development of social history, even ‘the invisible hand’ of laissez faire could be thought of having yielded, though initially only, the greatest social benefit through individual pursuit of own interests. However, at a reactionary stage, as we seem to witness now, the laissez faire of private interests would only conflict and collide with each other under the ‘animal spirits’ of a natural selection rule. Hence the need for government intervention. This assumes added significance especially in a less developed economy of majority poor. However, such ethical commitments dry up under the hypocritical archetypes, ingrained in the Indian subconscious mind, in league with the political economy of corruption. The future holds promises only with the rise of an enlightened society out of a soul-cleansing cultural revolution, reminiscent of that of the era of liberalism, having ‘a system of politics and administration marked by a high degree of personal integrity’ (Myrdal 1968: 957).37 Along with the Santhanam Committee, we would like to add: ‘We are convinced that ensuring absolute integrity on the part of Ministers at the Centre and the states is an indispensable condition for the establishment of a tradition of purity in public services. …..In the long run, the fight against corruption will succeed only to the extent to which a favourable social climate is created. When such a climate is created and corruption becomes abhorrent to the minds of the public and the public servants and social controls become effective, other administrative, disciplinary and punitive measures may become unimportant and may be relaxed and reduced to a minimum.’ (Government of India 1964: 101-102). This underscores the imperative for a vigilant civil society, fully conscious of and committed to its duties and rights, to act as a watchdog in the common interest. However, the emergence of such a civil society cannot be spontaneous, but has to be striven for by conscious public praxis in toto. Although we recognise the exertion of such public praxis by a few concerned citizens and their organisations, the challenge is so enormous that it calls for much greater intensification of efforts so as to eliminate, at the least, the scope for rent seeking. 26

“Praja sukhe sukham rajna: Prajanam ca hite hitam, Naatma priyam hitam rajna: Prajanam tu priyam hitam.” -

Arthasastra38 (1.19.34)

----------------------------------------------------------

27

Box 1

International Penetration of Corruption

According to the latest report of the Comptroller and Auditor-General (CAG) of India, crores of rupees loaned by the international lenders went down the drain as foreign consultants were engaged in the country’s maiden power sector reform launched in Orissa. The consultants were appointed in violation of guidelines and no attempt was made to engage domestic firms for the purpose…. The report says that during the selection process, World Bank’s senior energy economist virtually put pressure on the Government to opt for foreign firms, particularly KPMG, UK and Arthur Anderson, USA, and sent the list for approval. The State Government agreed to the World Bank (WB) official’s suggestion without enquiring into the firms’ experience and capabilities…. The WB staff, in violation of the Bank’s own guidelines and without any request from the Government, also reviewed suo moto the proposals submitted by the short-listed consultants and took Rs. 2.2 lakhs as service charges. A consortium of consultants led by KPMG was finally chosen with whom the State Government entered into an agreement. The consortium’s work in the first stage was to formulate basic strategies for carrying out the reform program which were to be identified and implemented in the second stage for which it was paid Rs. 41.97 crores. Despite extension of the deadline from 44 to 300 months, it could not complete its work, which got extended to a third stage, costing the Government an additional Rs. 72.96 crores. The State Government and WB did not have “a realistic perception of the requirement of the consultants’ time”, the report observes. Though the consultants were supposed to review and help the State negotiate power purchase agreements (PPAs) and related contracts for the privatisation of Orissa power generation corporation, they claimed it as an additional work and took an extra payment of Rs. 75 lakhs. They requisitioned, without asking the Government, a UK based firm, NGC, and billed an extra Rs. 35 lakhs. The NGC’s system of load despatch, however, was ultimately found not suitable for Orissa! Worse still, as per the CAG report, was the inability of the State Government to draft a legislation on reforms. The work was entrusted to McKenna Company for Rs. 56 lakhs. Due to some delay in establishing the project office and providing the necessary office equipment, the consultants claimed, and were paid, ‘idle time’ valued at Rs. 35 lakhs!……….. -Based on The Times of India, July 17, 2000.

28

Box 2

Enter the Enron…….

Close on the heels of a tour to the USA in May/June 1992 of a team of government of India officials for inviting private power producers to India, in the wake of the power sector liberalisation move, an official team of Enron Corporation and General Electric Company (GE) landed in New Delhi on 15 June 1992. They selected Maharashtra as their possible investment field as MSEB was the only healthy Board in India. So the team arrived in Bombay on the evening of 17 June, and the following two days they visited over half a dozen potential sites in the state. And on 20 June, on the third day itself, the MSEB signed a Memorandum of Understanding (MoU) with Enron and GE, by which MSEB would buy electricity for 20 years at a price of US cents 7.3 per unit (Rs. 2.34 per unit at then exchange rates) from a LNG-run plant of about 2000 MW capacity to be built, owned and operated by Enron near Dabhol. ‘There is no explanation on record or otherwise as to why a decision that involved the largest series of payments in India’s history was taken do quickly’ (Mehta 2000: 21). It should be noted that the World Bank was not in favour of this lop-sided contract. In its report of 30 April 1993, sent to the finance ministry, the Bank, in reply to the MSEB’s request for financial assistance for the project, concluded ‘that the project is not viable’. In fact, Maharashtra, like the Northern parts of India, had only peak load shortages of power and hence required suitable peak load plants only. But Enron plant was designed to be a base-load plant, which meant ‘surplus power in the off-peak periods’, requiring closing down of some of the Board’s cheaper plants. ‘LNG generation at a variable cost of about paise 150/kwh would displace coal-based power costing paise 30/kwh’! Again, taking at its face value the unit cost of energy of US cents 7.3 per unit as reported by the Enron, the Bank found that the ‘resulting retail revenue from LNG power would …. (be)…. Rs. 4.6/kwh in 1998 prices’, thus imposing high-cost power on the consumers. Moreover, ‘implementation of the project would place a significant long-term claim on India’s foreign exchange reserves’ with the ‘estimated annual fuel cost’ being ‘about US$ 500 million, subject to escalation’. Both the state government and the Board tried their best to refute the Bank’s conclusions and to justify the project, but the Bank remained unmoved. In its final reply on 26 July 1993, the World Bank severely criticised the MSEB’s attempt to justify the project on the grounds of a ‘more pessimistic’ projection of a decline in its own efficiency, without taking measures ‘to reverse this projected deterioration’, and declined to finance the project. Central Electricity Authority (CEA) is the ultimate authority in India to accord or refuse to accord techno-economic clearance of a power project after examining the tariff and cost of the project proposed and its consequences on the system. Though the MSEB and the government of Maharashtra (GoM) of the Congress party under Sharad Pawar, wanted to go ahead with the project without waiting for the CEA procedures, the CEA intimated to the government of India (GoI) of the same party (on 7 August 1992) that its ‘concurrence under Section 31’ was ‘statutory’ and could not ‘be dispensed with’. Then Enron submitted a technical report of the project that in fact did not contain any information on many essential parameters for evaluation such as cost components of the project, rate of interest, etc. The officially advertised capital cost of Rs. 9053 crores (about US$ 2830 million) gives Rs. 4.48 crores per MW (US$ 1.4 million) for the Dabhol plant. The CEA from its study however found that the reasonable capital cost of a similar combined cycle gas turbine plant would be Rs. 1.91 crores per MW (completed costs by December 1997). It is interesting to find that after five years, the ministry of power (MoP) officially conceded in September 1997 to the CEA estimate. The CEA study also found that about 408 MW of MSEB’s generating capacity, costing 50 paise to 80 paise a unit would have to be backed down in the first year in order to accommodate 695 MW of Enron's power at the MSEB-calculated rate of Rs. 3.47 29

a unit, that confirmed the World Bank’s warning about the Enron’s effect on MSEB’s economic merit order system. Enron simply refused to cooperate with the CEA by furnishing relevant information it had demanded for examination. In its final reply to the CEA on 11 October 1993, the company stated: ‘It is important to note….that capital costs are irrelevant to CEA’ [!] and ‘Your request for more detailed project costs of equipment/system/works other than those provided in the capital cost summary cannot be supported and is not deemed necessary.’ [!] Neither the government nor its bureaucracy was of any support to the CEA; and the MoP informed the CEA on 11 November 1993 that ‘Finance secretary observed that the question of cost of power had been looked into and it had been found that it was more or less in line with other projects being put up in Maharashtra.’ This in effect amounted to requiring the CEA not to look into the financial aspects of the project. The CEA therefore decided the next day itself ‘that given this background, the completed cost would not be considered by CEA at a later stage’, and that the project be given ‘clearance to technical aspects’ only. While controversy still surrounds the actual date of technical clearance by the CEA, the power purchase agreement (PPA) was signed by the MSEB and Enron on 8 December 1993. The project was to be implemented in two phases: only the first phase of 695 MW (to be run initially on naphtha) was binding on the MSEB as per the PPA; the Board had thus option not to buy power from the second phase of more than 1450 MW. The payments due from the MSEB to Enron as per the PPA were guaranteed by the GoM (on 10 February 1994) and counter-guaranteed by the GoI (on 16 September 1994). According to clause 11.2 (f) of the PPA, the MSEB ought to pay the current month’s bill on a two-part tariff on or before the 25th day of the following month. In case of a default or delay in payment by the MSEB, Enron’s first recourse is the irrevocable letter of credit against the MSEB’s receivables. If the MSEB fails to pay, then the GoM is bound under its guarantee to make payments within 7 days of invoking of the state guarantee. If that too fails, then Enron can invoke the GoI’s sovereign guarantee, in the event of which the Centre is entitled to reappropriate the amount, paid by it to Enron, through state devolutions that would in effect amount to cutting of assistance to social sectors in the state. Meanwhile, the project had flared up much controversy and the Swadeshi Jagran Manch (SJM) under the political umbrella of the Siv Sena (SS)-BJP opposition in the state was spearheading public agitation against the project. The SS-BJP combine called the project a betrayal of the nation involving large scale corruption and promised ‘to throw the project into the Arabian sea’. This single issue brought them to power in the next election in 1995. Immediately after the election and before the alliance took over the reins of power, the caretaker Congress government of Sharad Pawar, as asked by Enron, waived the conditions precedent in the PPA, making the PPA contractually binding and enforceable. A cabinet subcommittee constituted by the new GoM on 3 May 1995 found that the previous government had ‘committed a grave impropriety’ by conducting ‘private negotiations on a one to one basis with Enron’ under ‘circumstances which made the Enron/MSEB arrangement on Dabhol lack transparency’. The committee reached the ‘irresistible conclusion’ on the issue of ‘whether undue favours and concessions’ had been granted to the project, that ‘several unseen factors and forces seem to have worked to get Enron what it wanted’. The GoM then decided on 1 August 1995 ‘to scrap phase 1 and cancel phase 2 of the project’. The government also filed a suit against Daphol power company (DPC) and MSEB in the Bombay High Court seeking cancellation of the PPA ‘on grounds of fraud, corruption and misrepresentation’. 30

The estrangement, however, lasted only for two months. Rebecca Mark, Enron’s CEO, met, on 7 November 1995, Bal Thackrey, ‘the self-proclaimed hand holding the remote control’ of the SS-BJP government of Maharashtra; and the very next day, the government in a volte face announced ‘renegotiations’, despite its umpteen promises of ‘no negotiations’. The government appointed a committee in the name of ‘renegotiations’, but having a mandate to ‘revive both phases of the project’, which the committee fulfilled in 11 days. The renegotiated agreement provided for an expansion of phase 1 capacity from 695 MW to 740 MW at no additional cost. Then on the Republic day of 1996, the GoM formally announced the revival of both phases of the project, thereby increasing the obligations of the MSEB to draw full power from the project, i.e., from both the phases! And to crown it all, the ‘swadeshi’ government of the BJP coalition at the Centre cleared the sovereign guarantee to the ‘new’ project on the last day of its legendary 13 day reign in May 1996, just before dropping the reins of government in the face of a vote of confidence in the Lok Sabha. Interestingly, the new PPA was signed after 3 months only, in August 1996, by the MSEB and DPC, for the supply of over 2000 MW of electricity to the MSEB for a period of 20 years. ‘The payments due on the renegotiated contract constitute one of the largest contracts (civilian or military) in world history, and the single largest contract in this country’s history. Payments amount to about US$ 1300 million in the first year of phase 2 of the project going on line. Total payments amount to about US$ 35,000 million (Rs. 125,000 crores) over the life of the contract….. a conservative low-end estimate of the net present value (NPV) of this stream of payments is about US$ 17 billion (Rs. 70,000 crores) to about US$ 23 to 25 billion (Rs. 1,00,000 to Rs. 1,06,000 crores) at the middle end.’ (Mehta 2000: 3). According to the MSEB’s own calculations now, when the phase 2 project also is put on line, the power bill from the Enron to MSEB would be Rs. 7000 crores a year, against the Board’s anticipated revenue of Rs. 12,000 crores; i.e., the MSEB would have to pay almost 60 per cent of its revenue for about 20 per cent of its power purchase! A termination of the project would cost the MSEB Rs. 35,000 crores by way of compensation to Enron! (The New Indian Express daily, 2 January 2001). Today the MSEB pays a capacity charge, that must be paid regardless of the quantum of purchase as per the two-part tariff, of Rs. 93 crores per month; this will go up to Rs. 230 crores a month, when the phase 2 project also comes on stream! (The Hindu daily (Business section), 21 December 2000).

31

Box 3

The Enron Effects

The post-Enron period experiences have, however, painfully proved the critics right. The developments have put all the parties, government of Maharashtra (GoM), Maharashtra State Electricity Board (MSEB) as well as government of India (GoI) and the Enron, in a not unforeseeable quandary. The first phase of 740 MW of Dabhol power company (DPC), in which Enron has 65 per cent interest, was put on line in May 1999 at an actual unit tariff of Rs. 3.37 per kWh. This subsequently rose to Rs. 8.81 per unit in July 2000, and has remained well above Rs. 7 per unit. The total payments from the MSEB to Enron up to December 1999 were Rs. 1006.21 crores (Rs. 125.8 crores per month on average), and up to October 2000 was Rs. 2596 crores (Rs. 152.7 crores per month). MSEB soon felt this monthly drain too depleting, and all hell broke down there. The once-profitable and efficient MSEB found itself reduced to an inefficient loser. On the other hand, DPC, as per audited results, generated a cash profit of about Rs. 480 crores during the first 11 months of operation (The Sunday Times April 29, 2001). And the Enron project has come back in the hot news again. Since June 2000, MSEB had failed to pay its monthly bills to Enron on their respective due dates. It could not pay fully its November 2000 bill for Rs. 198 crores even by January 2001. According to clause 11.2(f) of the PPA, the MSEB must pay the current month’s bill on or before the 25th day of the following month. As the outstandings accumulated, MSEB requested GoM for a subsidy of Rs. 130 crores per month to help it to meet its obligations to Enron presently, and a monthly subsidy of Rs. 440 crores starting with the commissioning of Enron’s phase 2 project. MSEB’s inability to pay the bills led to Enron’s resorting to the threetired payment security mechanism one by one. First it invoked the letter of credit and then the state government guarantee and finally on 6 February 2001, the sovereign guarantee itself to recover Rs. 79 crores dues for November 2000. The temperature of the state of payments crisis was soaring up and the governments were losing face and hearts as well. The Credit Rating Information Services of India Ltd. (Crisil) downgraded the GoM by four notches at one stroke to speculative grade. Finally on February 12, the MSEB cleared the entire November 2000 bill, out of ‘funds made available by the state government’, without touching the dues of Rs. 152 crores of December 2000 bill for which also Enron had subsequently invoked the state government guarantee. And in early March 2001, Enron again invoked the GoI’s counter guarantee for a second time to recover the outstanding dues of Rs. 102 crores towards the December bill. The Maharashtra Electricity Regulatory Commission (MERC) had already directed the MSEB to stick to the economic merit order ranking of power plants: to buy or generate the least in terms of the costliest power. Thus for the year 2000-01, the average realisation for the MSEB was projected at Rs. 2.80 per unit and energy costing more than this had to be purchased the least. In the case of Dabhol power, the variable cost exceeded this average realisation mark in August 2000 and thereafter, leading to the least off-take from the Dabhol plant. In effect, the MERC pegged the quantum of power to be drawn from Enron at 3044 MU for 2000-01, corresponding to 50 per cent of its PLF. The latest developments have, however, been more sordid. As per the PPA, if the Dabhol plant fails to generate its maximum capacity within three hours of notice, the Board can claim penalty for the same. And so did the MSEB, slapping on March 1, a penalty of Rs. 400 crores for a default on January 28; again MSEB recently decided for another Rs. 400 crores penalty for a similar breakdown on February 13. According to official reports, such breakdowns are primarily due to the sub-standard project equipment used by the DPC. However, the 32

company found it in its interest to deal with the situation by invoking, on April 9, the ‘political force majeure’ clause of its PPA with the MSEB, which meant ‘unforeseeable circumstances that prevent someone from fulfilling a contract’. The notice indicated that the concerted, deliberate and politically motivated actions of the GoM, GoI and the MSEB had or potentially would have a material and adverse effect on the DPC’s ability to perform its obligations under the PPA. Significantly, the DPC invoked the force majeure clause just one day before the Godbole Energy Review Committee (constituted by the present GoM to scrutinise the Enron deal) submitted its report, which pointed out, inter alia, how Enron had engineered to fleece the MSEB through the PPA ‘facilitated by the failure of governance’ at every step of the decisionmaking process related to the DPC matters. The Committee’s calculations show that the DPC was overcharging the MSEB by Rs. 930 crores per year, and compare this with its claimed equity of Rs. 3,500 crores! Evidently, the Enron wanted to evade and escape the hard impact of the findings. Swift moves and counter-moves have marked each day since then. On April 25, the DPC Board of Directors that met in London voted to opt out of the project, and authorised the DPC’s managing director to issue a notice of intent to terminate its PPA with the MSEB. The GoM on its part mandated a new Committee, again headed by Dr. Godbole, to renegotiate a restructuring of the PPA, and on May 5, the Company agreed to meet the state panel for renegotiations. However, the DPC issued on May 19 a preliminary termination notice in view of the failure of the MSEB to pay two months’ dues and of the rebate (penalty) claim made by the Board. And the MSEB, authorised by the GoM in turn, retaliated on May 24 by issuing a notice to the DPC, to rescind the PPA, citing misrepresentation by the latter on the performance capability of the Phase I plant – the DPC had failed at least on three occasions to deliver power at nameplate capacities within three hours of demand by the MSEB. And on May 29, the MSEB decided not to buy power from the DPC any more. Subsequently, the DPC had to shut down the 740 MW phase I project and stop the construction works on the 1444 MW Phase II plant. Considering the Enron’s explicit desire to pull out of the Daphol project, the Godbole renegotiation committee decided to wind up by August 17. It should be noted here that as a way out of the present impasse of payments crisis, the GoM and MSEB had initially suggested that the Centre buy the entire output from the DPC directly or through NTPC, ‘pool its higher costs with what is generated elsewhere and sell it to other deficit states’ (The Hindu daily, 8 February 2001). Simply it amounts to spreading the heavy burden of Maharashtra’s own sin all over the other deficit states! Meanwhile, Enron had initiated arbitration proceedings, after the central government had refused to honour the counter guarantee to the project on advice of the GoM. In a swift reaction, the MSEB referred the dispute to the MERC, and the DPC moved first the Mumbai High Court (HC) and then the Supreme Court (SC) challenging the invocation of the jurisdiction of the MERC by the MSEB. Both the courts restrained the DPC from continuing international arbitration proceedings, pending resolution of the MERC jurisdiction issue. The apex court asked the Mumbai HC to decide whether the MERC had he jurisdiction to adjudicate the dispute, and the HC began the hearing on September 20. And the previous day, the GoM had ordered a judicial probe into all aspects of the project. Finding the game getting tougher and trying, the DPC, in a letter to the prime minister in August, repeated its offers to sell its equity in the project to the government for $ 1.2 billion on a no-profit no-loss basis. Later on the offer came down to settle for nearly $ 800 million (The Financial Express, 11 November 2001). Now the group of the domestic financial institutions (FIs), led by the Industrial Development Bank of India, that had funded or guaranteed 72 per cent of the debt component of the project, to the tune of Rs. 6199 crores, 33

came out from their indifferent slumber and suggested, as a solution to the Daphol problem, that the project be divided into 3 blocks of roughly 700 MW each; of these, one each could be purchased by the MSEB and the NTPC, and the third block by a private party, who must be granted a MSEB distribution circle. The MSEB, however, rejected the proposal forthwith. From the private sector, the Mumbai-based utilities, viz., the Tata Power Company and the BSES Ltd., had meanwhile emerged evincing interest in buying the stakes. And finally, however, as time ticked away, the DPC served on November 5 a notice of asset transfer to the MSEB, marking the penultimate step towards the termination of the project, the final termination notice (FTN) to be served any time after November 19. The Indian FIs swiftly moved the Mumbai HC, which on November 9 restrained the DPC from issuing a FTN to the MSEB and passed orders for preserving the securities of the FIs. On the international market, the wind was blowing hard against Enron Corp, the parent of the DPC. Enron shares, which traded at as high as $ 90.56 on August 23, 2000 (during the last summer’s energy crisis when energy stocks soared), plummeted to a 10-year low of $ 7 on November 7, 2001. The energy marketer, Dynegy Inc. was about to take over Enron Corp (a company that a year back was valued at nearly $ 80 billion) for $ 9.5 billion in stock (a value of $ 10.41 per share). However, the Dynegy backed out of the deal alleging misrepresentation to it by Enron of facts of its financial health; and subsequently Enron had to seek refuge under the clauses of bankruptcy, thus signalling the demise of an inflated giant. Noteworthy in this respect are the following strands of fall-out from these sordid affairs: 1. The MSEB requested the GoM to keep on hold the two other independent mega power projects planned in the state, viz., the Reliance’s 447 MW Patalganga project and the Ispat Group’s 1084 MW Bhadravati project. Like the Enron project, these too have a dollar payment component to their cost. Moreover, the GoM made it clear that there would be no Enron-type agreements in future: ‘We do not want to be trapped into the clutches of MNCs and in future there will not be Enron-type agreements’, so declared the Maharashtra chief minister Vilas Rao Deshmukh (The Hindu 9 February 2001). 2.

Another important Enron effect is the GoI’s decision to drop its proposal to offer sovereign guarantees to the three other mega plants – the 3960 MW Reliance-SEAP promoted Hirma project, the 1800 MW LNG-fired Ennore project and the 2000 MW Pipavav project. (The government is in two minds now about the need at all for setting up the Pipavav project in Gujarat, following the CEA’s findings that once the Hirma mega project is on stream six years from now, the off-take of power from the Pipavav project would be only at 60 per cent of plant capacity. This means either down-sizing the Pipavav project capacity or scrapping the project altogether.) The government, instead, offered the option of direct sale of power to earmarked consumers in the purchasing states. This arrangement obviates the need for the government to accord sovereign guarantee, necessitated primarily on account of the financially fragile monopsonistic nature of the power market (i.e., SEB). The Electricity Bill, 2000 also allows for direct sale of power to consumers. This provision, as argued elsewhere, by facilitating the weaning away of a significant sure source of revenue (major industries) from the SEBs would in fact only worsen their position further.

The lessons from the Enron effects are obvious. Dabhol power project in effect offered a monopoly market for the products of its own owners – a LNG market to Enron, a gas supplier, operating oil/gas fields; an electrical machinery market to General Electricals, a supplier of power capital equipment; and a construction market to Bechtel, a project engineer firm. These three ‘suppliers’ possess the majority of stakes in the DPC. Once the market was captured, they 34

managed to interpret the government notification on return on equity in their favour to corner more than 30 per cent guaranteed return also (Srinivasan 1996). Inefficient capital overexpansion (Averch-Johnson effect/‘gold-plating’) is the empirically proved behaviour under such conditions of guaranteed returns. With a guaranteed high return (the highest in the world), what incentives will a company get to minimise capital expenditure? It led both to accommodating all rent seeking costs in the project costs and also to over-capitalisation, making power consumption in India exorbitantly expensive. The over-capitalisation behaviour, on the other hand, is induced further by the guaranteed provision for payments in dollar denomination. Given a historically weak, submissive, rupee against dollar, nobody with a little sane commercial sense would venture into a contract wherein the payment for inputs (gas, machinery, etc.) were denominated in dollars, while the output (electricity) was to be sold out at rupees. When the project was contracted in 1993, the dollar, on its rising trend, was quoting at Rs. 32.50, whereas at present it is around Rs. 46.50! All the Indian authorities concerned were not unwise to this imminent danger that would make Indian electricity costlier. For example, the MSEB’s letter (of 8 July 1993) to the energy secretary of the GoM, considering the effects of a range of scenarios of escalations of exchange rate, capacity charge, oil price, gas price, etc., shows full awareness of this on their part. Despite this exercise, the MSEB wrote (on 30 September 1994) to the energy secretary (GoM) that the Board ‘had already examined the tariff structure of the Dabhol Power Company and accordingly intimated the Central Electricity Authority …. [on] September 17, 1993….that the tariff offered by M/s Enron is lower than that calculated on the basis of two part tariff notified by Government of India.’ Nowhere else in the world is naphtha, the highly volatile and costly fuel, used for power generation. Maharashtra had at that time only peak power shortages, which could have been met through import from the Central pool. The import price of naphtha in January 1999 was Rs. 5,200 a tonne, which shot up to Rs. 16,000 a tonne by December 1999; today it is around Rs. 19,000 a tonne (Malayala Manorama daily 8 February 2001). The unique Indian adoption of naphtha for power generation was the offshoot of a wrong forecast by the petroleum ministry of the domestic availability of naphtha, expected to be in high surplus. The power ministry based its entire liquid fuel policy on this forecast, and suggested to Enron (as well as to NTPC) to turn to naphtha; Enron had originally planned to use distillate in the first phase (The Hindu Business line daily, 2 January 2001). Contrary to the forecast, the domestic production of naphtha fell short, and at the same time, the world price soared steeply up.

Notes 1

Remember in a Smithian regime individual pursuit of self-interest maximisation by itself results in public interest maximisation. 2

Equivalently, C(q)  C(q),  1 , for any q  0, where C(q) is the total cost of producing q.

3

This is the condition of strict subadditivity of the cost function. More formally, strict global subadditivity of costs for the 1 m 1 m 1 m multi-product cost function C(q) implies that for any m output vectors, q , …., q , C(q + ….+q )  C(q ) + ….+ C(q ), i i i th where q = (q1 , …., qn ) is the i output vector.

35

4

The Chicago school view of interest group politics started with Stigler’s (1971) discussion on the capture of regulatory authorities by industries they were meant to regulate. Both Stigler and his colleague Peltzman (1976) have built on the work of Olson (1965). Becker’s model also is in the same grain except that he suppresses the role of legislators, driven by the selfinterest of maximising political support. ‘Politicians, political parties, and voters …. are assumed mainly to transmit the pressure of active groups’ (Becker 1983: 372). 5

Tullock (1967) started the analytical discussion of rent seeking in economics as a socially costly pursuit for income transfers, though it was Kreuger (1974) who, unaware of Tullock’s work, characterised the behaviour in the present terminology, in a theoretical framework where rent seeking in the face of restrictive policies in international trade results in social costs. Posner (1975) was the first to develop a formal model of and to empirically estimate the social costs of rent seeking, and the first survey of the theory came form Tollison (1982). In the theoretical arena of international trade, rent seeking appears in the form of tariff evasion (Bhagwati and Hansen 1973), competition for premium-fetching import licences (Krueger 1974), lobbying for protectionist trade tariffs (Brock and Magee 1978) and for tariff revenue resulting from the adoption of protectionist tariffs (Bhagwati and Srinivasan 1980), and so on. On the other hand, rent seeking in the public choice theory represents competition (e.g., lobbying) for monopoly rents (Tullock 1967) and for rent-creating regulation (Stigler 1971), ‘rent-sharing’ by trade unions through collective bargaining with rent-earning regulated firms (Rose 1987), lobbying for and against a tax-financed transfer, rent-seeking contract (for money, votes, etc.) by politicians (Peltzman 1976) or just ‘rent-taking’ by them or rent extraction – by threatening to extract private rents (e.g., through taxation) and then refraining from doing so at a payment (McChesney 1987, 1997), and so on. Rent seeking can thus be extended to government and bureaucracy too. 6

The term ‘principal-agent problem’ appeared first in Ross (1973). The earlier discussion on principal-agent problem in the framework of imperfect monitoring and imperfect information appeared in Stiglitz (1975), Mirrlees (1976), Harris and Raviv (1978), Holmstrom (1979) and others. For excellent surveys, see Hart and Holmstrom (1987), Levinthal (1988) and Holmstrom and Tirole 1989). It should be noted that the principal-agent model was originally employed to analyse insurance, sharecropping, physician-patient relation, law enforcement, etc. It was only with the development of the model in the framework of imperfect monitoring and imperfect information by Stiglitz and others that the model was applied to analyse bureaucracies and hierarchies of organisations. 7

The significance of the scope for collusive behaviour in the principal-agent environment is explored by Tirole (1986) and

Laffont (1988, 1990). 8

Hence Samuelson (1970: 771) wrote: ‘through direct public services and through transfer-payment programs, the modern mixed economy is in effect a gigantic system of mutual insurance against the worst economic disasters of life.’ 9

The KSEB has estimated that just two new industrial units (in steel smelting and electro-chemicals) are drawing more than 8.7 MU every month at concessional rates. The tariff difference alone gives them a benefit of about Rs. 30 crores a year, while the employment generated is only nearly 350. That is, the KSEB is forced to give a subsidy of about Rs. 8 lakhs per year for each job thus created (Menon 1999:464). 10

For example, recently in a presentation to the union power ministry on ‘measures to attract foreign direct investment in power sector in India’, leading foreign financial institutions have required the Central government to take measures to move the power sector from the concurrent list to the Central list as in the case of the Telecom sector (The Hindu daily 27 March 2001). 11

Note that the total commercial loss in the sale of 293479 MU at a cost-tariff deviation of 43.4 paise per unit in that year comes out to be Rs. 12740 crores only, against this revenue loss on account of corruption of Rs. 10705 crores. 12

It should be noted that the reported loss percentage of the Kerala power system is a gross underestimate. It is now generally recognised that the energy loss in the state is not less than 35 per cent, half of which is theft. Though theft of power has been made a cognisable offence since 1986 under the amended Electricity Act, 1910, collusion stands to nullify its effect. All the SEBs do have anti-theft squads that conduct regular but superficial homilies of checks and detect some pilferage cases of lesser fry, just to justify the survival of the squads. In 1997-98, the anti-theft squad of the KSEB detected cases of theft of energy worth Rs. 1.21 crores, in 1998-99, worth Rs. 1.04 crores and in 1999-2000, worth only Rs. 80.42 lakhs. This steep fall in detection trend, despite an officially recognised rise in theft losses, points to the need for vigilance over the vigilante squad itself! The Board as well as the government keeps the eyes closed and ejects some regular warnings and orders (just to justify their own presence!) such as the one recently put out again in vain by which each of the squads at centres of Thiruvananthapuram, Kochi and Kozhikode is required to detect cases of theft of at least 50 lakhs units a year (Malayala Manorama daily 23 August 2000). That major thefts do take place is confirmed by the reports of the vigilace squad of the KSEB having conducted raids in the premises of the major consumers and found energy thefts to the tune of Rs. 1.33 crores across the state over a period of just 17 days, during March 7 – 23, 2002 (Mathrbhoomi daily: April 1, 2002). 36

13

For example, Mehta (2000:10) cites a case from Maharashtra. The Mula Pravara Cooperative Society is reported to have outstanding dues of power bills to the Maharashtra State Electricity Board (MSEB) to the tune of about Rs. 250 crores, accumulated over a decade. The scion of the society had defected to the Shiv Sena and was a minister in the Shiv-Sena- BJP government last time. The MSEB still accounts for this amount as ‘receivables’ and writes off as fraction each year as bad debts. 14

The Santhanam Committee in its Report on Prevention of Corruption remarked long back: ‘We were told by a large number of witnesses that in all contracts of construction, purchases, sales, and other regular business on behalf of the Government, a regular percentage is paid by the parties to the transaction, and this shared in agreed proportions among the various officials concerned. We were told that in the constructions of the Public Works Department, seven to eleven per cent was usually paid in this manner and this was shared by persons of the rank of Executive Engineer and below down to the Ministry, and occasionally even the Superintending Engineer might have a share.’ (Government of India 1964: 10) 15

The Santhanam Committee noted that ‘There is a widespread impression that failure of integrity is not uncommon among Ministers and that some Ministers who have held office during the last 16 years have enriched themselves illegitimately..’ (Government of India 1964:101). 16

This was the first time that a court convicted a minister for corruption in Kerala. The starting point was the detection of a leak in the power tunnel of the Idamalayar hydro-power project (that took 17 years for completion by 1987 with a cost escalation of 387 per cent). Under pressure from the opposition parties in the State Assembly, the then Congress-led Ministry instituted, on 21 December 1985, an enquiry commission (The Justice Sukumaran Enquiry Commission) to examine the allegations of corruption involving the then power minister, power secretary, high officials of the KSEB, and the contractors. The Commission, on the basis of detailed enquiry, indicted all of them on charges of corruption involving huge losses to the exchequer. After a due process of judicial trial, they all were convicted also. Subsequently, there have been allegations, followed by enquiry commissions, relating to the award of contracts in the construction of other power projects also (for example, Brahmapuram diesel power plant near Kochi). It should be added that the same minister along with the officials were convicted in another case also – the Graphite case that concerns illegal diversion for resale of NTPC power to six companies in Karnataka including the Bangalore-based industry, Graphite India Ltd., when Kerala was reeling under unprecedented power shortage during the mid-eighties. These private companies are alleged (in the FIR) to have profited to the tune of Rs. 70 lakhs by way of getting the cheaper Kerala power. 17

The MoU for a 513 MW combined cycle power project at Kannur at an estimated cost of Rs. 1500 crores was signed in February 1995 by the KSEB and the KPP Nambiar and Associates. It was one of the nine mega projects cleared by the High Power Committee (at the Centre) in 1995, when a Congress-led Government was in power in Kerala. The power purchase agreement (PPA) was signed on March 14, next year, and by the end of 1997, the Kannur project was accorded technoeconomic clearance (TEC) by the CEA. But the project was an ill-starter. The new left Government in Kerala could not tolerate the Enron co-sponsorship of the project and hence rejected the State clearance to the project. However, after some dilly dallying, the Government agreed to clear the project(the Chairman of the company being a close relative of the Chief Minister!) provided it found a new co-developer acceptable to the State Government. Thus a new Kannur Project was then recommended by the State Government with the El Paso Energy International of the US as the co-promoter. Kannur power project was one of the three projects in the power sector (including the NTPC-Birla sponsored 1886 MW Ennore power project in Tamil Nadu with 100 per cent foreign (US) participation) identified by the Union Government to be presented at the IndoUS summit in Washington to attract US investment to India during the recent visit by the Indian Prime Minister there. But the State Electricity Minister called the joint secretary in the Union Power Department, on the eve of the PM’s visit to the US and said, “We have decided in favour of Ennore and not Kannur” (The New Indian Express daily, September 20, 2000). The Kannur project is pictured as the most recent victim of inner party factional frictions as well as unrequited kickback demands (The New Indian Express, September 28, 2000). The Chairman of the company himself has recently come out and reported to the Press of the kickback demands for Rs. 75 crores by the son of a political bigwig controlling the government. The El Paso co-sponsorship of the project also has been recently rejected by the government. 18

This has been at the tragic cost of the domestic producers of power generating equipment, especially Bharat Heavy Electricals Ltd. (BHEL), which, being up to mark technologically and much cheaper, has gained good markets in some of the developing countries. Till the late seventies, the BHEL had won nearly all international tenders floated in India for power equipment (Mehta 2000: 15). However, its share in the capacity addition in thermal sets in India has recently come down to the order of 50 to 60 per cent: ‘BHEL itself has repeatedly stated that the single most important handicap it faces in India is its inability to arrange for kickbacks’ (Morris 1996: fn. 15, 29). 19

The Brahmapuram project, the first thermal power plant of the KSEB (and the second diesel power plant in the country, the first being the Yelahanka project in Karnataka) of 100 MW capacity was implemented with French assistance of Rs. 160 crores loan under the Indo-French protocol. The Justice V. Bhaskaran Nambiar Enquiry Commission has found that the purchase contract with a French company SEMT Pilistik for five generators (each of 20 MW) without calling for international tenders as per guidelines resulted in a gain to the French company (and a loss to the KSEB) to the tune of Rs, 71 crores, that the contract 37

with Geo Tech for land reclamation and levelling led to a loss of Rs. 2.96 crores, and that the contract with Tata Project for generator erection works involved a loss of Rs. 1.38 crores. The files in the government and the Board, examined by the Commission, were clear proofs for the high powered collusion that had gone out of their way to favour the contractors (various dailies 12 April 1999). Diesel, a highly expensive fuel like naphtha, is not at all permitted by the Central government in power generation. The KSEB has, however, now two diesel plants – at Brahmapuram (100 MW) and at Kozhikode (120 MW). These two plants have often been run much below capacity, sometimes at not more than 30 per cent, because of the high operating costs, much more than Rs. 5 per unit, greater than twice the cost of imported energy from the Central pool. It is estimated that to run a generator at Brahmapuram for 24 hours requires about 120 kilo litres of diesel worth Rs. 9 lakhs, whereas its energy fetches less than Rs. 5 lakhs only. This then involves a loss of about Rs. 20 lakhs per day to the Board from the Brahmapuram plant alone. Hence it has been decided to run the diesel plants only during the peak load hours. The same peak load could, however, be met by cheaper power from the Central share with a judicious import policy. A simple lesson from this then is that the KSEB could safely have dispensed altogether with the costly diesel plants and saved substantial resources. 20

In this context, Morris (1996: fn. 29) writes: ‘From other sources we know that for a small favour of a year or so’s extension, a former chairman of the NTPC had to arrange for a contribution of a crore of rupees to the Congress [party] kitty. Thus even the task-oriented enterprises within the PS [public sector] necessarily have to accommodate corruption. (The state of affairs in enterprises that have veered too far from their primary task can well be imagined.)’ 21

A high Level Committee even went to the extent of recommending to the government of Kerala long back to enact appropriate legislation prohibiting strikes under any circumstances in all power projects under construction’, especially citing a ‘classical’ example of three strikes by the construction workers of a project (Idamalayar hydro-power project, works on which started in 1970 but completed only in 1987) that extended over a total period of three years and one month, causing a loss of Rs. 33.65 crores to the KSEB (Government of Kerala 1984: 57-61). 22

Excluding the hydro projects of Kallada and Pooyankutty, and the two diesel power plants. If we stick to the strict assumption that the original project cost estimate allow for possible inflation during construction period, such that the estimate be as on the completion date, then the corruption charges involved would be very much higher. 23

Interestingly, in an advertisement by the Public Relations department of the KSEB announcing the commissioning of the Kuttiady extension project claimed it to be ‘the first project to be completed on time in a span of 25 years’ (various dailies 26 January 2001), whereas in fact the project, major works on which were started in February 1994, was originally scheduled to be commissioned in 1995-96, thus involving a time overrun of about 5 years and a cost escalation of more than Rs. 160 crores. Again, the advertisement put the project cost at Rs. 160 crores, while the latest cost estimate is given at Rs. 198 crores by the Economic Review 2000. Even at the advertised capital cost, the energy potential of 129 MU from the project cost as much as Rs. 12.4 per unit! 24

Rose-Ackerman (1999:28-29) quotes a number of cases of corruption-inflated project costs: for example, Itaipu dam on the Brazilian border. In the 1970s, two German companies reportedly paid bribes of 20 per cent of the value of construction contracts for a steel mill in Indonesia to a state government official. In Germany, in the mid 1990s, bribes played a major role in awarding contracts to build Terminal 2 at Frankfort Airport; according to the public prosecutor, corruption led to an increase in the air fares of about 20 to 30 per cent. In Italy, the costs of several major public construction projects reportedly fell steeply after the anti-corruption investigations of the early 1990s. Overall successful bids on public tenders were reported to be 40 to 50 per cent lower in 1997 than five years back. 25

Another possible explanation for this divergence is that while the CEA has been giving original cost estimates, the MoP has been reporting the final estimated costs, after allowing for overruns. But as we have already argued, overruns are often deliberate and conceal corrupt collusion. 26

Examples come galore in India in this respect – the revelations of Harshad Mehta, the bull instrumental for the great stock market crash in India, about his crores worth bribes to the Congress party kitty through the then prime minister (PM) during the early 1990s, confession of a Jharkhand Mukthi Morcha MP of having been bribed by Congress party by crores of rupees for voting in favour of its PM in the face of a non confidence motion in the Lok Sabha (the very same PM) was convicted in this case), the recent allegations by the chairman of Kannur Power Projects about the demand for a bribe of Rs. 75 crores for sanctioning the power project by a Marxist bigwig’s son wielding full power in the power ministry of Kerala, etc. 27

The absence of relevant files would also prove the cases (!) as it has so happened in the case of the Graphite energy resale cases in Kerala. There has not been a single official paper in the Board or in the ministry regarding the energy export to Karnataka; the matter came into light when the Karnataka SEB wrote to the KSEB about the energy sales later on! 28

A former chief secretary of Kerala, when convicted recently in a corruption case, reportedly simply quipped: ‘No problem; there are higher level courts!’ 38

29

While enlarging upon the need for Indian economy’s restructuring, the World Bank (1996: 3) had to recognise, though in passing, that ‘India’s pre-1991 planned development strategy helped the country escape from the massive illiteracy, recurrent famines, fertility rates of about 7 children per woman, and secular stagnation prevailing before Independence.’ 30

See, for a comparative analysis of corruption levels across countries, World Bank (1997). Corruption in the trasition economies is explained among others by Boycko et al. (1995), Feige (1997), Kaufman and Siegelbaum (1997), Malia (1995) and Weisskopf (1992). 31

According to Quinglian (2000), what has occurred in China since 1978 as a result of what she calls ‘the marketisation of power’ has been nothing but a ‘socialist free lunch’ by which the politically powerful in China have used their still awesome administrative and personal power to plunder the former state-owned economy and ‘laugh all the way to the bank’! 32

To be precise, the changes in India were not political and institutional, but were in the ruling economic principles.

33

See Thakur (1979) for a detailed discussion on corruption in the ancient India. Myrdal (1968 Chap. 20) gives a good discussion on corruption in modern India, extensively quoting from Santhanam Committee Report (1964). 34

For one example, the Committee (Government of India 1964:18) found that import licences in India were worth 100 to 500 per cent of their face value! 35

‘Where there is power and discretion, there is always the possibility of abuse, more so when the power and discretion have to be exercised in the context of scarcity and controls and pressure to spend public money.’ (Government of India 1964: 9) 36

Remember, in a country like India, rich with hydro-power potential, a judicious hydro-thermal plant mix in generation capacity, along with considerations of high-cost gas power vis-à-vis cheap and clean hydro-power can ensure this for a long time. 37

The Santhanam Committee (1964) long back recognised that ministers and legislators must be above suspicion and proposed codes of conduct for these two categories of politicians and special procedures for complaints against them. Accordingly, on 29 October 1964 itself, the Government of India released the text of a code of conduct for ministers both at the Centre and in the states. The code required disclosure by a person taking office as minister of the details of his and his family’s assets and liabilities as well as business interests. He was also required to sever all connections with the conduct of any business. However, the scepticism expressed at that time itself on the loyalty on the part of the intended persons to the codes of conduct has proved right. Myrdal in 1968 itself wrote: ‘Later, the eagerness for reform seems to have died down. The reports are that corruption in India has recently been increasing.’ (p. 956, fn. 2) 38

“In the happiness of the subjects lies the happiness of the king; in the welfare of the subjects is the welfare of the king; what is desirable and beneficial to the subjects and not his personal desires and ambitions, is desirable and beneficial for the king.”

39

CHAPTER 11 CONCLUSION “… but, the end, in truth, proved to be only a beginning.” - Plato (The Republic Book II) “And to make an end is to make a beginning. The end is where we start from.” -

T. S. Eliot (‘Little Gidding’ v)

1. Introduction As in the case of any other product, supply of electricity also involves three distinct functions of production (generation), transportation to market (transmission) and retail supply (distribution), the only difference being that electricity is non-storable in its usable form and hence must be generated the moment it is demanded for. This in turn requires instantaneous co-ordination and integration of the three vertical functions which is technically facilitated by the continuous, instant flow of electricity from the generator to the end-use equipment at a velocity approaching that of light. Thus an electric utility is distinctly characterised by the technical significance of vertical integration.

In addition to this technical condition for centralisation is an economic requirement for the integrated functioning of the electric utility. This emanates from its natural monopoly status, granted by its characteristic cost complementarity that occurs in the presence of economies of scale and scope. The economies of ‘non-convexities’ (i.e., the economies of overhead costs, Clark 1923) in turn are related to the asset specificity that characterises the electric utility. Asset specificity refers to the relationship-specific investment, for example, in the transmission and distribution sectors, which, once sunk, has little value in alternative uses (i.e., other than the intended one). The large scale transmission (and the associated primary distribution) asset specificity arises in the context of the site specificity of the hydropower plants and the mine-mouth coal plants. The size constraint, in favour of large plants in the generation sector, also involves

economies.

The consequent vertically integrated natural monopoly position of the

electric utility thus ensures productive efficiency in the sense that the cost of supply is minimised by having a single firm supply electricity (under which condition the cost function is said to be sub-additive). Securing such productive efficiency, however, can be disastrous if the monopoly is in the hands of private profiteers with the functional behaviour of setting the output below optimum and the price above marginal cost, causing dead weight loss. Such allocative inefficiency may be avoided in principle in a competitive market of many firms, which, on the other hand, will violate the productive efficiency criterion, favouring a single firm in the industry. Nationalisation of the natural monopoly in the general interests of the society can resolve this dilemma and ensure both allocative and productive efficiency and equity. This is the economic rationale as well as the welfare justification for the public utility in power supply.

This in-principle organisational superiority notwithstanding, the public sector in general has remained in practice at the receiving end of a number of dysfunctionings. A justificatory setting has also been in the making here. The market is given a far-flung recognition as providing sufficient signals for efficient performance of the economic agents, despite the fact that the actual situation seldom simulates this perfection. Juxtaposed with such an ideal picture, the non-market behaviour of a State sector easily falls under an impression of inevitable doom of inefficiency. Unfortunately, history has yielded enough corroborating substance to this tendency, turning it out into an almost universal truth, in terms of the infamous dysfunctionings under a number of quasisocialist systems especially in the Eastern Europe. Added to this have been the costly consequences of pork-barrel politics in the third world countries of making a fetish of socialist institutional forms sans essence – with unscrupulous corruption ingrained in the whole body politic, moral hazard and the consequent x-inefficiency syndrome have cankered all the functional commitments. In the absence of productivity consciousness, internal resource generation meant for further capacity expansion in the State sector has often drawn blank, requiring heavy State financing. The cumulative effects of all these functional irregularities have reached such a pass as to take the public sector for granted

2

as structurally inefficient. This dynamics of destiny have had enough room for its full play in the Indian power sector too.

As in the case of other infrastructure facilities with high capital intensity and long gestation period, that stood to deter any entrepreneurial initiative by a nascent private sector, the responsibility of power development also was originally shouldered by the State in India. Power is a concurrent subject under the Indian Constitution, its development being a joint responsibility of the central and state governments. However, since 1956 till the 1970s, the subject had almost exclusively been confined to the State sector with an accelerated growth. In the 1980s, on the other hand, the state sector enthusiasm dried up, and the weight of capacity addition shifted from the state to the central sector. And the private sector has been assigned a major role in power generating capacity expansion since the turn of the 1990s. This variable trajectory of investment behaviour illustrates the effects of the ilfare of the power sector in India in terms of the financing capacity for development.

Even though the State Electricity Boards (SEBs), established for the rationalisation of power development at the state level, were statutorily required by the Electricity (Supply) Act of 1948 to function as autonomous corporations, they were in effect regarded as promotional agencies, expected to subserve the social, political and economic policies of the governments. (The central sector utilities, on the other hand, are corporations, like the private ones, under the Companies Act.) The patronising policies of the State resulted in overstaffing, especially at the non-technical, administrative level, involving unwarranted cost increases and in irrational pricing practices for subsidised power sales in the name of industrialisation, agricultural development as well as domestic sector distributional considerations, all tainted with political motives. The State’s commitment on rural electrification, as translated through the SEBs on their responsibility also imposed heavy cost burdens. Although the State was required to fully compensate the SEBs for its induced inefficiencies in terms of subsidised power sales and rural electrification duties, the compensation was irregular and inadequate, causing substantial cumulative losses to the Boards.

3

The government intervention further extended to the very day to day organisational affairs of the SEBs. The socio-political dynamics of in different states resulted in a situation of widespread corrupt practices of nepotism, at the cost of merit, ability and efficiency. Political considerations dictated the appointment and the tenure of office of the top management personnel, and this retarded their commitment and involvement in serious independent policy making. A general lethargic indisposition for accountability, stemming from a steady enervating erosion of competitive management values infected the institutional texture to the bottom. This was also due to the management culture, in the bureaucratic ways of functioning, inherited by the SEBs as they had been carved out of the earlier government departments. In such contexts, selfinterest maximisation drives in a favourable climate of information asymmetry and incomplete contracts stood to stimulate moral hazard effects – that is, default and breach of trust in doing one’s duty committed, giving rise to a sort of x-inefficiency. The unaccountability culture in its accumulated scale worked along with the direct government intervention behind the dysfunctionings of the SEBs.

It should be stressed that the performance of the SEBs was largely determined for a long time by the assertions and defenses of their statutorily intended promotional role in power development. The SEBs were to subserve the socio-economic policies of the State and hence expected not to view every aspect of developmental activities exclusively from the point of view of profit or return, as highlighted by the Venkataraman Committee of 1964. Thus there was no compulsive requirement, till the late seventies (till the 1978 amendment of the Section 59 of the E(S) Act, 1948), for the SEBs to break even, as also even to provide for full depreciation and/or interest payable on Government loans, both of which could, under the Statute, be provided for only if there were adequate surpluses after meeting all other obligations. Thus there seemed to be no idea, let alone requirement, of the SEBs contributing internal resources to expansion programmes. The SEBs have not yet come out of that spell of unaccountable, non-commercial performance, and in general continue to have negative internal resources.

4

A number of committees, for example, the Venkataraman Committee of 1964, examined the functioning of the SEBs and recommended a net return of 3 per cent on capital base, after providing for operation and maintenance charges, contribution to depreciation and general reserves, and interest on loan. However, the 1978 amendment to the principles of financial performance of SEBs contained in Section 59 of the 1948 Act did not stipulate a specific figure of return to be earned, but merely provided that the Boards should earn a positive return ‘after taking credit for any subvention from the state government’, and after meeting all expenses properly chargeable to revenues, including operating, maintenance and management expenses, taxes on income and profits, depreciation and interest charges. The amendment also recognised the desirability of the SEBs having part of their capital as equity and allowed any state government to notify the SEB as a body corporate. However, the government wisdom could not digest such desirabilities, lest it should forgo a political cornucopia of populism. And the SEBs remained in the same old spell of unaccountable, non-commercial performance, leaving little internal resources for expansion.

All these developments necessarily had the making of an apparent harbinger of radical mutation. The inability of both the state and central power sectors to finance further capacity expansion led ipso facto to the foregone conclusion of large scale private participation as the sole solution to the worsening power shortage situation. Both the local and global settings were changing in colour in favour of a triumphant return of liberalism, following the fall of socialism. The new private sectorisation drives had an apparently formidable backing of a now widely justified ‘there-is-no-alternative (TINA)’ logic. In this receeptive background of functional inadequacies and financial infirmities of the SEBs, though entirely avoidable, came in handy for a mis-characterisation of the whole sector: the costly disfunctionings were unreasonably identified with economic inefficiency, associated with the standard notion of some market structure devoid of competition. Thus mistaking functional inefficiency for structural/organisational deviations made it easier to put up a foregone conclusion in favour of a need for restructuring the power sector. And there opened up in the Indian power sector a vast vista to the ambitious global capital to emulate Enron-tragedy.

5

It is this background that has provoked our study into the dynamics of the plight of the Indian power sector in terms of inefficiency and the consequent initiatives of reform process as well as the political economy of that plight.

It is found that much of the capacity/energy deficit India is in general confronted with today could be easily avoided with some achievable improvement in capacity utilisation and T & D operation. Committees after Committees have in fact pointed out this simple but valuable fact of the effects of the ‘inert areas’ in the functionings, thanks to some innate moral hazard tendencies. If it could have been nipped in the bud, it would not have posed an ‘Aegean stable’ in its accumulation, apparently calling for a drastic surgery of cleansing. We find that the avoidable cost of inefficiency at a few amenable levels of functioning represent about one-third of the reported cost of electricity supply in India in 1997-98, and more than 40 per cent in Kerala! In fact this alone could have yielded substantial commercial profit in the power sector, given the ‘low’ average revenue realised. And this is regardless of a number of other possible inefficiency sources at all levels of performance. For example, the financial inefficiency induced by the government intervention for subsidised power sale and rural electrification commitment, poor performance in revenue collection, etc.

The picture of the settings of this plight becomes clearer as we come closer from the general to the particular case of Kerala power system.

A scientific planning process is fundamental to any power system operation. Hence the significance of a study into the adequacy of the present planning process in the context of the widening gap between demand and supply. This in turn necessitates to examine the accuracy and rationale of the mechanism of power demand forecasting on one hand, and, given the demand for projections, the adequacy of the capacity addition planning as well as the operational efficiency, on the other hand.

6

An objective forecasting mechanism, capturing the full implications of the socioeconomic reality, has been conspicuous by its inadequacy in our country. Time trend projection is in general employed for power demand estimation, without caring for any model adequacy diagnostic checking nor accounting for possible non-stationarity in the time series data base, even in the academic circles. In some specific cases, estimated industrial growth rates are utilised, but without examining the general validity of correlation or causation between the variables. In fact, there are a large number of small scale and cottage industries that use practically little electricity, but together contribute significantly to the industrial product. India has for a long time been in the grip of severe power shortage, and the use of time series data on supply-constrained demand for future demand estimation is fundamentally inadequate and inappropriate. Moreover, the low level socio-economic development of India may also invalidate the demand analysis technique in its usual framework of price-income-population correlatives. On the other hand, Kerala with a high standard of living in a substantially developed social environment presents significant scope for electricity demand analysis, but in the chronic situation of power cuts and load shedding, this too loses any sense of reality. Our econometric exercises prove that demand analysis can in no way explain the objective situation in the power sector of Kerala.

The actual performance of the sector in fact has been such as to render the very exercise of power demand forecasting futile, except as some routine ritual; the demand forecast has never had anything to do with the capacity expansion planning prepared on a bounded budget as well as with the actual materialised capacity additions in the system in any state in India. Kerala has been a typical example, with little recognition of the need for a comprehensive development-based perspective planning mechanism that is to ensure smooth system growth in terms of adequate capacity expansion not only of hydropower but also of thermal backing to the system in order both to meet the exigency of monsoon failure and to strengthen the reliability parameters. Even though an optimal merit order operation requires a favourable hydro-thermal mix in generation capacity, the KSEB had for a long time evinced an unwarranted aversion to setting up thermal plants in Kerala, and had even ridiculed and rejected at least two times the central government’s

7

offers of thermal plants. As a result of the defective and myopic power development planning, the Kerala system has remained much smaller under a favourable condition of a domestic-sector-dominant consumption profile.

While the planning process for capacity expansion has had only nominal relevance in force, its execution has been failing all expectations. Not a single project, even the micro ones, has been left unhaunted by the spectre of excessive time overruns, involving exorbitant opportunity costs, extracted cleverly by collusions among the contractors, militant labour, KSEB officials and the politicians in power. If these projects had been completed and commissioned regularly in time, it would have saved substantial resources, yielded additional revenue and eased the shortage problem to a good extent. However, such time overruns have been generally accepted as a convenient camouflage for large scale corruption at the high up. Thus both the processes of planning and execution have had a track record of inadequacy and functional failure, and the very same behavioural characteristic has extended to the operational field also. Though hydro plants are generally expected to be much less prone to forced outages than thermal plants, those in Kerala stand an exception and have registered higher forced outage rates and loss of load probability. Even if the plants are available, their service is subject to firm power capacity constraints, given normal monsoon. The technocratic tendency for unrealistic assumptions have however led to an undesirable situation of over-capitalisation in terms of wide divergence between installed capacity and firm power capacity in Kerala – another flagrant failure of planning. Then the worse occurs in the transit; a good proportion of the energy sent out from the inadequate firm power capacity gets lost behind the meter – again in corrupt collusion with the Board officials.

The detailed discussion unfolds the fact that the present system predicament is due to problems which are just internal to the system. To this extent, then, there do remain sufficient quarters for remedial exercises, meant to remove the impediments to the SEBs’ improved performance. That is, what the system today requires is essence-specific (internal) reforms, not structural mutation as unfortunately made out now, and acknowledged even in the informed circles, quite surprisingly. On the other hand, the

8

strict insistence on and the straight involvement in hasty policy changes in favour of the liberalist exercises of the governments are easily understandable. The political economy is witnessing extensive openings for corruption to scale new heights in the implementation of privatisation drives, in addition to the usual transactions in awarding concessions and contracts available in the old regime. The State’s function is conveniently confined to its teleological mission of administering its coercive authority in the defence of property rights that facilitate market mechanism. The welfare concerns and the development commitments of the State, undertaken in emulation of and as a counter to socialism, have dried up along with its fall in the resultant vacuum of a competitive alternative. As governments conveniently relieve themselves of the socialistic involvements, the tax income extracted from the public in the name of the State’s services now becomes available to them for expanding their own budget on the face of an apparently indifferent public. And unfortunately, the informed circle remains indifferent to the crime, instead of instituting itself as a counter-force on a civic platform of checks and balances. Public praxis should be concerted on such platforms and directed to displace the capitalist teleology of the State with a new socialist ontology for the public interest. Viable measures are still available for a resurrection of an alternative. For example, power sector could well be rehabilitated at minimum cost within its structure of public sector itself, instead of being subjected to the most painful surgery of irreversible restructuring. Below we enlist a number of feasible suggestions in the context of the power sector in Kerala which are more or less applicable to other SEBs also.

2. Some Practical Way-Outs

All our analyses have revealed that the problems haunting the Kerala (or Indian) power sector are only internal to the system, and hence there do remain sufficient quarters for remedial exercises, meant to remove the problems that stand in the way of the KSEB’s improved performance. These may be classified into short-term measures of crisis management and long-term steps of power sector development, as follows. (Also see the Appendix to this Chapter for an overview of the remedial prescriptions in the context of the Kerala power system.)

9

A. Short-term remedies

1. At present, the Kerala state has a total installed capacity of 2391.2 MW. Of which the KSEB accounts for a capacity of 1979.1 MW (including two diesel plants and a wind farm). However, the firm power capacity of the hydro (and wind) power system is only 753.85 MW, enough to meet a demand for 18 to 20 MU a day, against an actual (constrained) daily consumption of about 40 MU (including T & D loss). The contribution of the two high-cost diesel power plants of the KSEB is only 1 to 2 MU. The remaining is accounted for by energy purchase – often more than 30 per cent of the gross available power. The excessive dependence on energy purchase and the consequent cost burden could be reduced in a number of ways of operational efficiency and commitment on the part of both the KSEB and the government. These are:

1.1 Take immediate and necessary steps to complete and commission at the earliest all the projects entangled in time overruns. These are minor projects like Malampuzha, Malankara, Chimony and Kuttiady Tail Race, and the diversion schemes of Kuttiar, Vadakkepuzha and Vazhikkadavu, works on which were started in the late 1980s. Once completed, they will add to the system 165 MU of energy potential.

1.2 Start construction works on the already approved projects like Athirappally, Kuttiady additional extension, Neriamangalam extension and others as well as the 14 micro hydel schemes under the Chinese collaboration, with a total installed capacity of nearly 400 MW. Caution should be exercised against any room for possible time and cost overruns; the construction contracts should be so structured as to provide for making the contractors liable for stringent penalties in case of non-performance such as time overrun. The LDF government (1996-2001) is reported to have made some steps in this direction in the case of the Athirappally project by initiating to institute in the contract penalty provisions for delay – something of the first kind in the history of the KSEB. This should be strictly adhered to and extended to all other projects. The savings in time and other resources will also be enormous. This ready-to-start

10

project should therefore be implemented forthwith along with the above-mentioned expansion projects.

1.3 The mini hydel project, Chimony, works on which were suspended following a High Court stay order obtained by the contractor of the electrical works in 1993, should also be saved at the earliest by moving the court for vacating the stay order. This, in our view, is an instance of ransoming the larger public interest for some personal motives.

1.4 Uprating, renovation and modernisation of all the old projects, especially Pallivasal, Sengulam, Poringalkuthu, Neriamangalam and Sabarigiri, that have been under consideration for a long time should immediately be taken up and pushed through for completion at the earliest. Similarly, uprating and extension of small plants (Kuttiady, etc.) to utilise surplussing

water during rainy season will also increase energy

availability. Measures should be taken to clear the silts accumulated in these reservoirs that limit their capacity (e.g., Idukki, Pallivasal, etc.) and to prevent further silting (afforestation, etc.).

1.5 The KSEB has at present two diesel plants (at Brahmapuram and Kozhikode) which are in general utilised incredibly far below capacity (often in the range of only 10 to 40 per cent) on account of the much higher cost of generation. Nonetheless, the Board is fast setting up another diesel plant at Kasaragode, which is going to have the same fate of underutilisation. It is high time that the KSEB refrained from such imprudent practices of wasteful planning and mismanagement at least in view of scarce resources.

1.6 Purchases from the NTPC’s Kayamkulam thermal project is a high-cost burden at low plant load factor (PLF) owing to its ‘state project’ status. Converting it into a regional one could go a long way to increasing the PLF and thus reducing the purchase price substantially. The government is reported to have taken up this matter for negotiation

11

with NTPC and the neighbouring systems. However, an enhanced share of power from the Central pool and its regular and constant delivery should also be ensured.

B. Medium/Long-Term Measures

1. The only resource for power generation generously available to Kerala is hydraulic energy. About 40 per cent of the estimated 4500 MW of hydro potential of the state has already been tapped. The development of the remaining sites is however beset with clearance difficulties out of environmental concerns. Those projects which do not face objections on environmental grounds, for instance, Mananthavady and Kerala Bhavani, that remain locked up in objections in terms of inter-state disputes, should be taken up for clearance. And both the government and the KSEB should refrain from such unwise wild goose chase as that incurring wasteful expenditures on environmentally sensitive projects like Pooyankutty.

2. Having no known sources of fossil fuels, Kerala state is to depend on imported fuels for thermal power development, which is also constrained by non-availability of suitable sites for major thermal plants, especially of coal, thanks to fragile nature of coastline and high density of population, confining the choices to plants based on cleaner fuels. Kerala should strive for an early access to LNG grid and LNG-based power plants. The state government should lobby for the early implementation of the proposed LNG terminal at Kochi and the laying of pipelines to Kayamkulam in the South to benefit the NTPC project there and to Malabar in the North.

3. In addition, the following solutions merit attention for the improved performance of the state power system:

3.1 Improved technical efficiency: 3.1.1

Though hydro-plants are in general less prone to forced outages (FO), some of the plants in Kerala are afflicted by very high FOs (e.g., Panniar, etc.). Regular and timely planned maintenance along with full and proper repairs well within time

12

will ensure higher availability. KSEB is to revamp its standards, system and organisation for proper maintenance of its plants.

3.1.2

The ‘wasteful’ gap between installed capacity and firm power or dependable capacity in Kerala is now about 43 per cent. This gap should be bridged (for full capacity utilisation) by enhancing the firm power capacity through augmenting water supply to the existing reservoirs. In planning and implementing future projects, care should be taken to avoid such over-capitalisation; augmentation schemes should be planned and executed simultaneously with the parent project. This could save considerable resources.

3.1.3

Check the tide of time and cost overruns. Study on 18 hydro projects in Kerala reveals that the timely completion and commissioning of these projects as per original schedule would have saved at least Rs, 35 crores per project (excluding the implicit interest cost) that went into cost escalation and additional sales revenue of at least Rs. 52 crores per year that was lost in time overrun. The resources thus saved could have been utilised for further capacity expansion and it would have eased the so-called resources crunch of the government that is unfairly used to woo the private sector. The government should see to it that the future projects are completed in time, without time and cost overrun. Future construction contracts should be so structured as to stipulate the condition that legally binds the contractor to compensate the Board for any delay.

3.2 Improved T & D Efficiency: 3.2.1

Reducing T&D loss to 15 % from the present 18 % can save about Rs. 34 crores in additional annual sales revenue to the KSEB in the medium term. The long term objective should be to reduce the T & D loss to 10 per cent. This could be achieved by enhancing transmission capacity, replacing defective meters, effectively checking theft of energy and regular maintenance of the network Antitheft squad should be earmarked compulsory ‘quota’ and strictly policed over to prevent collusion; administer incentive schemes also.

13

3.3 Efficiency in Management: 3.3.1

Computerised scientific inventory control should be introduced on the basis of a thorough study of the existing system, which is highly haphazard in management of all aspects. Energy audit is another area that calls for urgent attention.

3.3.2

The KSEB is to gear up itself to undertake cost efficient measures. Cost of energy supply can be substantially reduced on a number of fronts. Burden of power purchase cost can be lessened by improved operational efficiency – higher availability (reduce forced outage rates), larger inflow (more augmentation schemes), minimum T & D loss and auxiliary consumption, etc. About 20 % reduction in per unit cost can be expected on this count.

3.3.3

Over-manning in establishment & administration (E & A) along with a shortage of technically skilled personnel is a bane of any power system, and sadly this is the case of the KSEB. Trimming over-manning in establishment & administration can reduce per unit supply cost by about 20 %. At the same time, there must be a mechanism to uprate and up to date the technical skill of the personnel.

3.3.4

Allowing 1:1 debt-equity ratio can check accumulation of unpaid interest charges (due to government) and make the balance sheet look healthy in this respect (on the stipulation that a reasonable return is ensured on the equity). This can bring about a reduction of about 12 % in the unit cost of supply through reduced interest charges.

These measures, if properly applied, can be shown to bring a commercial profit of Rs. 121 crores (in 1997-98) to the KSEB, instead of a commercial loss of Rs. 521 crores, at the ruling average revenue. Given the vast scope for cost minimisation, clamours for tariff increases lose their justification.

14

3.3.5

Still, along with such cost-effective measures, there is an urgent need to apply scientific tariff structuring on the basis of such efficient costs. It should also be ensured that lifeline tariffing with prompt government compensation mechanism also is structured in such a way that the additional costs from such equity-based subsidisation never interfere with operational efficiency.

3.3.6

Just tariffing is not sufficient. It should be ensured that the sales revenue these tariffs yield is collected regularly in time. The receivables from the sale of power in Kerala in 1997-98 represented 41 % of the sales revenue, i.e., about 5 months’ sales revenue being locked up with the consumers against the maximum allowable norm of two months’ sales revenue. If these arrears could be collected in time, it could be used for further capacity expansion without going in for additional loans, after simply writing off a fraction of the arrears every year as bad debts. This necessitates universal spot billing, and putting an end to the collusion of the KSEB officials with the errant customers. For this, the ‘collection centres’ should be strictly made accountable for their monthly ‘quota’ through both coercive and incentive stipulations. Similarly, there has been an uninterrupted practice of non-payment of energy bills by almost all the government institutions; this practice should be strictly stopped and all the dues collected in time.

3.4 Organisational efficiency: 3.4.1

This is in fact central to the improved functioning of the KSEB. Convert the Board into an autonomous corporation as under the Electricity (Supply) Act, 1948. Such a mere restructuring by nomenclature is not enough. The fundamental requirement is autonomy; governmental intervention should be done away with fully in its day to day affairs, including appointments. At the same time, the Board should subserve the welfare policies of the state for which the government should compensate it promptly (rural electrification and subsidised power to weaker sections). An independent regulatory authority should be formed to co-ordinate, direct and watch all the functionings in a transparent manner, with checks and

15

balances on a platform of public hearing. This will help dispel the impression being created by sectional interests that the proposed regulatory body is intended to be only a tariff-fixing/raising machinery. Ensuring the efficient functioning of the board should be the objective of the regulatory body. In such an ensured efficiency environment, cost increases, if any, can be justified only on account of factors external to the functioning of the board.

3.4.2

Continuity of management by top personnel at the policy making level is another important factor. Appointments at higher levels be made based on selection, and the selected official with proven ability and integrity should have at least 2 to 5 years further service for superannuation.

3.4.3

It goes without saying that there is an urgent need to stem the rot in work culture (X-inefficiency) through superior selection procedures, linking the terms of job continuity and remuneration to productivity and

accountability clauses and

periodical evaluation of the performance. To achieve these, a package of incentives for performance and disincentives for non-performance be instituted.

3. Summing Up

Administering such efficiency improvement measures in the technical, organisational and financial management could certainly win the system a comfortable footing on its own. And this could in turn dispense with the need for the irreversible and disastrous restructuring that is not only incapable of solving the real problems, but fraught with dangerous implications threatening the very social coherence. One of the disastrous consequences of private sectorisation comes from the fact that a private enterprise system necessarily works on exclusion principle. The vast scope for lodging all sorts of large scale rent seeking costs in over-capitalisation results in inflated supply costs that can exclude a sizeable proportion of consumers with limited purchasing power. Higher incidence of exclusion would be one of the deleterious social costs of private sectorisation in a poor country like ours, leading to increasing or

16

excessive inequality, both individual and regional, and is likely to result in a loss of community and social coherence.

However, the beneficial significance of the private sector participation in power generation should have a say in designing the development plans. Entry of the private power producers (PPPs) should be facilitated in a transparent competitive environment to ensure the selection of least-cost options by the watchful regulatory authority, who should further structure the provisions for the conduct of the chosen PPPs. On the other hand, no private interests should have any hold on the sales end; that is, by no means the distribution sector be privatised. Collective interests could be safeguarded and fulfilled for common objectives of development with equity only by the vehicle of public sector provisions. The present dispensation of distribution of power can be restructured in a more direct administration of collective responsibility at local levels through co-operatives and municipalities. This will in fact enhance and sustain the significance of decentralisation of democratic power, with the direct involvement of the local consumers in distribution activities. And such direct involvement alone can rid the society of feelings of alienation that lie at the heart of moral hazard effects and the consequent x-inefficiency. It is in facilitating this that a State realises its meaning. By shirking its social responsibilities, a State ceases to exist in essence. What is required in this light is a strong political will to stand up to and tackle the problems which are only internal.

“The Committee firmly believes that, given the national and especially the political will to surmount the difficulties that lie ahead, the country has ample managerial, technical and physical resources to accomplish the task of ridding the nation of the endemic power shortages which have plagued it for the last two decades. What is even more important, these resources applied to the twin tasks of conservation and development of new energy sources could do what a few nations today seem capable of doing surviving the energy crisis that is engulfing the world.” -

Report of the (Rajadhyaksha) Committee on Power (1980: 5)

17

An overview of the Prescriptions (in the context of the Kerala power system) Measures

Expected gains

A. Short Term 1. Complete

the

projects

already Additional energy potential of 165 MU

under construction (minor and mini = reduced power purchase cost and hydro-power projects and diversion increased sales revenue. schemes)

2. Start construction of the already Additional capacity of 400 MW = approved projects.

reduced

power

purchase

cost

and

increased sales revenue.

3. Uprating,

renovation,

and Cost

modernisation of old projects.

4. No more diesel plants.

reduction

and

sales

revenue

increase.

Resources savings and cost reduction.

B. Medium/Long Term

1. Take up only feasible projects other Resources savings and timely capacity than environmentally sensitive ones

addition.

2. Lobby for the early implementation Substantial cost reduction, as naphtha is of the proposed LNG terminal at replaced by LNG. Kochi to benefit NTPC and other thermal projects in the state.

3. Check time and cost over-runs.

Substantial

resources

savings

(e.g.,

about Rs. 644 crores from 18 hydro projects constructed/under construction

18

so far, and additional sales revenue (e.g., Rs. 52 crores per year, were these projects completed in time).

4. Improve technical efficiency -

Increased generation reduces power

(a) revamp standards, system, and

purchase and thus supply cost; e.g., with

organisation for proper

50 % PLF of hydro plants, 70 % PLF of

maintenance of the plants to

thermal plants, 0.64 % (as reported)

increase technical avaialability.

auxiliary consumption and 15 % T&D

(b) Take up more augmentation

loss, the import cost savings in 1999-

schemes to increase firm power

2000 would be 32 Paise per unit sold

capacity in line with installed

(14 % of unit supply cost) and in 1997-

capacity.

98, nearly 50 paise per unit (26 %).

5. Improve T&D efficiency.

Reduced power import and thus supply cost; e.g., with 15 % T&D loss in 19992000, power purchase cost reduces by 6.2 Paise per unit sold and with 10 % loss, by 18 Paise per unit sold.

6. Improve management efficiency:

Cost reduction.

(a) Computerised scientific inventory control (b) Energy accounting and auditing (c) Professional management. (d) Trim

over-manning

in Reducing

over-manning(i.e.,

to

the

establishment and administration standard 2 employees per MU sold) (E&A).

would reduce E&A (and thus supply) cost (as in 1997-98) by 20 Paise per unit sold (about 10 %).

19

(e) Allow 1:1 debt-equity ratio.

This would reduce interest charges and thus unit supply cost (as in 1997-98) by 23 Paise/unit sold (about 12 %), and by 33 Paise/unit, as in 1999-2000, (about 14 %).

7. Scientific tariff structuring.

Inefficiency

not

transferred

onto

customers.

8. Regular and prompt collection of revenue

No

resources

crunch;

e.g.,

the

receivables against electricity supply in 1997-98 was 41 % of the sales revenue, and other sundry debtors, 33 %.

9. Improve organisational efficiency:

The core problem of inefficiency is

(a) Superior selection procedures

solved.

(b) Continuity of management by top personnel (c) Strengthening professionalism of management (d) Higher work culture (e) Productivity- and accountabilitylinked job continuity and remuneration

20

GLOSSARY Availability: Availability refers to the probability that the unit is available for operation. This is a combined index expressing maintenance and forced outages requirements of a given unit. Availability = (available hours/period hours) x 100, where the denominator denotes the number of hours in the period involved (a year) and the numerator, the number of hours in the period during which the unit is available for service, whether or not it is actually in service (i.e., service hours + reserve shut down hours). Average Load: The average rate at which customers take energy from an electric system during some long specified time period. Average load = Generation + Import  Export. Base Load: The minimum load of an average electric utility. Base-Load Generation: Generation of electricity that occurs by generating plant operating most of the time to meet the invariable base of demand (or load). The plant thus operates essentially at a high load factor. Cogeneration: A system in which both the heat and power needs of a work are met, utilising for power generation, exhaust heat from turbines and diesel engines. Such combined heat and power generation is used on a large scale in the oil, paper, chemical and sugar industries. Combined Cycle Power Plant: A power plant that utilises the combined cycle technique for raising the thermal efficiency of using gas for electric power generation. The technique involves first using the gas to fuel a combustion turbine (or gas turbine which uses the expansions of burnt gases to turn the turbines for power generation) and then recovering the waste (exhaust) heat to generate steam for application to a conventional steam turbine for further power generation. Combustion Turbine: A machine that, like a jet engine, uses the expansions of burnt gases to turn a turbine for electricity generation; also referred to as ‘gas turbines’. The combustion turbine is designed to convert the heat energy of fuel into mechanical energy.

Connected Load: The sum of the capacities or ratings (in kilowatts) of all the electric power consuming apparatuses installed on a consumer’s premises and connected to a supplying system. Distillate Fuel: The lighter fuels obtained from fractions boiling above gasoline in the distillation of petroleum. The most important are kerosene, furnace oils and diesel. . Distribution: The last of the three integrated technical functions (viz., generation, transmission, and distribution) of an electricity supply industry, distribution refers to the low-voltage reticulation of electricity for retail sale, accomplished by an extensive network of low voltage (or low tension) power lines. Electrical Energy: The energy associated with electric charges and their movements measured in watt-seconds (joules) or in kilowatt-hours (kWh). Electric Power: The rate (per unit time) of doing work measured in watts. In electricity parlance, it is the rate (per unit time) of supplying or using electrical energy. Energy Audit: A survey of energy usage in an organisation or system, and an examination of efficient usage of energy. It requires the application of both engineering and economic tools. Firm Power: The power of a hydroelectric plant or system that is always available and dependable for carrying load; it corresponds to the minimum stream flow with due consideration to the effects of pondage and the load factor. Fluidised Bed: A bed of solid fuel particles through which air or gas is blown upwards at a velocity high enough to buoy the particles such that the bed has the appearance of a boiling liquid and shows many fluid-like properties. Fluidised Bed Boiler: A new type of boiler designed to reduce combustion product pollutant and reduce boiler size. Forced Outage: Forced outage occurs when a unit is thrown out of service due to unexpected causes such as break down, equipment malfunction, etc. These are usually of a random nature.

2

Forced Outage Rate (FOR): The failure pattern of a given generating unit is usually described by its FOR, computed as the percentage of the unit’s down time relative to the total service plus down time. FOR = (Forced outage hours/forced outage hours plus service hours) x 100, where forced outage hours refers to the time during which the unit is unavailable due to a forced outage, and the service hours, to the time during which the unit is actually in operation. Generation: Large scale production of electrical energy, using generators: i.e., machines for producing electrical energy from mechanical energy. There are two primary sources for driving generators – hydro and thermal. Hydro-electric power is derived from generators turned by falling water. Most other electrical energy is obtained from generators driven by steam produced either by a nuclear reactor or by burning fossil fuels – viz., coal, oil and natural gas. Generation is the primary function of a vertically integrated electricity supply industry, the other two functions being transmission and distribution. Hydropower Plant: An electric power plant in which the energy of falling water is converted into electricity by turning a turbine generator. Internal Energy Sales: Average Load  Auxiliary Consumption  Losses. Kilo-Watt (kW): A measure of electrical power, or the ability to do work; 1000 watts = 1 kW. Though a basic measure, kW is often too small to be useful for practical purposes; hence, the use of mega watt. Kilo-Watt-Hour (kWh) or Unit: The basic measure of electrical energy which is ordinarily used for retail billing; one kilowatt applied for a period of one hour: 1000 watts times 1 hour. Load: In electricity parlance, load is equivalent to demand. Load Factor (LF): The ratio, expressed as a percentage, of the average load on a system or one of its units (or plants, in which case it is known as plant load factor) to the peak load. The annual average total cost of generating one kWh (or unit) of electricity at a given power station depends on the amount of electricity produced, i.e., on the load factor of the station. As LF increases, cost per kWh of generated energy decreases, as total cost is now spread over a larger number of units.

3

Loss of Load Probability (LOLP): A measure to evaluate the reliability performance of a given electric power system, LOLP gives the expected accumulated amount of in a given period, usually a year, during which the system will experience a shortage of energy. It is expressed in terms of days per year, hours per day, or as a percentage of time. When expressed as a fraction of time, the LOLP can be thought of as the probability that there will be a shortage of power of any magnitude in the given period; hence the name. Mega-Watt (MW): A measure of electric power, or the ability to do work; 1 million watts = 1000 kW = 1 MW. Merit Order Loading: The system requirement at any time is met by loading the units for generation in order of their average production costs, which serves as an indicator of a unit’s merit order. The most efficient unit, i.e., the unit with the lowest average production cost, is loaded first and operated at its rated capacity; the next most efficient unit is then loaded for generation at its rated capacity, and so on until demand is satisfied. This is arranged by a central control (load dispatcher) which instructs the stations to start up and shut down as required in order of fuel cost.

Natural Monopoly: An industry with significant long-run economies of scale; public utilities possess, to some degree, the characteristics of a natural monopoly. Nowadays natural monopoly is defined in terms of economies of scope as well as economies of scale, the former referring to cost reduction involved in joint production. Outage: That state of a plant unit when it is not available for generation due to some event directly associated with it. An outage may be a forced outage or a scheduled outage. Peak Load or Maximum Demand: The highest rate of simultaneous consumption of electricity by consumers connected to a common electricity supply system, measured over a half-hour period. Peak Load Generation: Generation that occurs periodically during the course of a day to meet the peak load placed on the system.

4

Plant Load Factor (PLF): The ratio of the actual generation of a power plant to its maximum possible generation during a period (one year);thus represents a capacity factor; also PLF = availability less reserve shut down rate. Public Utility: A public utility conventionally refers to an essential service, such as electricity, gas, water, telephone, railway, etc., that is available to all people in general. The traditional technical characteristic that distinguishes a public utility runs in terms of economies of scale. Reliability: The ability of the system to meet the demand for power at any given point in time. Reserve Margin: An excess of generating capacity designed to ensure the continuation of reliable electricity supply in the event of a sudden loss of a generator. Thus the required installed capacity of a power system is determined after allowing for a certain percent reserve margin with reference to the expected peak load. Large interconnected systems can achieve cost reductions by pooling reserve requirements. Reserve Shut Down Hours: Period (in hours) during which a plant is backed down due to lack of adequate system load (and/or, in the case of hydro-plants, due to lack of adequate water in the reservoir). Scheduled (or Planned or Maintenance) Outage: Scheduled outage occurs when a plant unit is deliberately taken out of service at a selected time, usually for the purpose of preventive maintenance or overhaul or repair, intended to ensure their proper running conditions. Service (or On Line) Hours: Number of hours a plant is actually operated during a period (one year), which may be divided into service hours, outage hours (both forced and planned), and reserve shut down hours. Steam (or Thermal) Plant: A generating unit that uses the passage of (superheated) steam across a turbine to generate electricity. The steam might be produced by burning a fossil fuel such as oil or coal, by nuclear fission, or as a result of cogeneration. Substation: An electrical installation where the power delivered by transmission circuits is stepped down to voltages suitable for use in industrial and residential areas. On transmission systems, such a facility is called bulk-power substation; at or near

5

factories or mines, industrial substation, and in residential and commercial areas, distribution substation. Transmission: The second of the three integrated technical functions of an electricity supply industry, transmission refers to the conveyance of electricity at high voltage from generating stations to substations for subsequent retail distribution, or high voltage industrial uses. Transmission is accomplished by an extensive network of high voltage power lines. Voltages higher than those produced by power plant generators are required when transferring current over long distances in order to reduce power losses that result from the resistance of transmission lines. Step-up transformers are employed at the generating stations to increase the transmission voltage. At the substations step-down transformers reduce the voltage to the levels suitable for distribution. Utilization Rate: The ratio of the actual energy produced by a plant unit to the maximum possible energy that could have been produced during the same available period.

6

BIBLIOGRAPHY

1. Ahluwalia, Isher Judge (1998). ‘Comment on “Regulatory Priority for Infrastructure Reform in Developing Countries”, by Paul L. Joskow’, in Annual world Bank Conference on Development Economics, 1998, pp. 224 – 227, World Bank, Washington DC. 2. Andrews-Speed, P., et al. (1999), ‘Do the Power Sector reforms in China reflect the Interests of Consumers?’, The China Quarterly, No. 158, June: 431-446. 3. Arrow, K.J. (1985), ’The Economics of Agency’, in Pratt, J.W. and Zeckhauser, R.J. (eds.) Principals and Agents: The Structure of Business, Harvard Business School Press, Boston: 37-51. 4. Averch, H., and Johnson, L. (1962). ‘Behaviour of the Firm Under Regulatory Constraint’, American Economic Review, Vol. 52, December, pp. 1053 – 69. 5. Baijal, Pradip (1999). ‘Restructuring Power Sector in India – A Base Paper’, Economic and Political Weekly, Sept. 25. 6. Banerjee, A., Dolado, J.J., Galbraith, J.W. and Hendry, D.F. (1993), Cointegration, Error Correction and the Econometric Analysis of NonStationary Data, Oxford University Press, Oxford. 7. Banerjee, A., Dolado, J.J., Hendry, D.F. and Smith, G. (1986), ‘Exploring Equilibrium Relationships in Econometrics through Static Models: Some Monte Carlo Evidence’, Oxford Bulletin of Economics and Statistics, 48: 253-277. 8. Banerjee, Nirmala (1979), Demand for Electricity, Centre for Studies in Social Sciences, Calcutta, K. P. Bagchi and Co., Calcutta. 9. Bardsen, G. (1989), ‘The Estimation of Long-Run Coefficients from Error Correction Models’, Oxford Bulletin of Economics and Statistics, 51: 345-350. 10. Barnett, W.A., Hendry, D.F., Hylleberg, S., Terasvirta, T., Tjostheim, D., and Wurtz, A. (2000), Non-linear Econometric Modelling in Time series, Cambridge University Press, Cambridge. 11. Baumol, W. J. (1977), ‘On the Proper Tests for Natural Monopoly in a Multiproduct Industry’, American Economic Review, 67 (5), December: 809-822. 12. Becker, Gary S. (1983), ‘A Theory of Competition among Pressure Groups for Political Influence’, Quarterly Journal of Economics, 98 August: 371-400.

13. Betancourt, (1987).‘Capital Utilization’, The New Palgrave A Dictionary of Economics. pp. 368-370. Macmillan, Cambridge, Massachusetts. 14. Bhagwati, J.N. (1982), ‘Directly Unproductive Profit-Seeking Activities’, Journal of Political Economy, 90 (5), October: 988-1002. 15. Bhagwati, J.N. (1993), India in Transition, Clarendon Press, Oxford. 16. Bhagwati, J.N. and Hansen, B. (1973), ‘A Theoretical Analysis of Smuggling’, Quarterly Journal of Economics, 87, May: 172-187. 17. Bhagwat, J.N. and Srinivasan, T.N. (1980), ‘Revenue-Seeking: A Generalisation of the Theory of Tariffs’, Journal of Political Economy, 88 (6) December: 1069-1087. 18. Bhargava, A. (1986), ‘On the Theory of Testing for Unit Roots in Observed Time Series’, Review of Economic Studies, 53: 137-160. 19. Box, G. E. P. and Jenkins, G. M. (1970), Time Series Analysis, Holden-day, San-Francisco. 20. Box, G.E.P. and Pierce, D.A. (1970), ‘Distribution of Residual Autocorrelations in Autoregressive-Integrated-Moving Average Time Series Models’, Journal of the American Statistical Association, 65: 1509-1526. 21. Boycko, M., Shleifer, A. and Vishny, R.W. (1995), Privatising Russia, MIT Press, Cambridge, M.A. 22. Brennan T. J. et al. (1996), A Shock to the System: Restructuring America’s Electricity Industry, Resources for the Future, Washington DC. 23. Breton, A. (1996), Competitive Governments, Cambridge University Press, New York. 24. Brock, W.A. and Magee, S. P. (1978), ‘The Economics of Special Interest Politics: The Case of the Tariff’, American Economic Review Papers and Proceedings, 68, May: 246-250. 25. Buchanan, J.M. (1975), The Limits of Liberty, University of Chicago Press, Chicago. 26. Burtraw, Dallas and Palmer, Karen (1996), ‘Electricity Restructuring and Regional Air Pollution’, Draft Working Paper, Resources for the Future, May. 27. Centre for Development Studies (2001) Report 5: 2000-2001, Kerala Research Programme on Local Level Development, Thiruvananthapuram.

2

28. Chamberlain, J. (1894), 'Municipal Government: Past, Present and Future', The New Review, 61, June. 29. Chan, K.H., Hayya, J.C. and Ord, J.K. (1977), ‘A Note on Trend Removal Methods: The Case of Polynomial Versus Variate Differencing’, Econometrica, 45: 737-744. 30. Chapman, D., Tyrrell, T. and Mount, T., (1972), ‘Electricity Demand Growth and the Energy Crisis’, Science, (November 17), 178: 703-708. 31. Cheng, B.S. (1995), ‘An Investigation of Cointegration and Causality Between Energy Consumption and Economic Growth’, The Journal of Energy and Development, 21: 73-84. 32. Cheng, B.S. (1997), ‘Energy Consumption and Economic Growth in Brazil, Mexico and Venizuela: A Time Series Analysis’, Applied Economic Letters, 4: 476-674. 33. Cheng, B.S. (1999), ‘Causality Between Energy Consumption and Economic Growth in India: AnApplication of Cointegration and Error Correction Modelling’, The Indian Economic Review, 34(1): 39-49. 34. Clark, John Maurice (1923), Studies in the Economics of Overhead Costs, University of Chicago Press, Chicago. 35. CMIE (1988). Basic Statistics Relating to the Indian Economy; Vol. 2: States. September. Bombay. 36. CMIE (1996), Report on the Current Energy Scene in India, Mumbai. 37. CMIE (1999), Industry – Financial Aggregates and Ratios, Company Finance, June, Mumbai. 38. CMIE (2000), Energy, March, Mumbai. 39. CMIE (2001), Energy, April, Mumbai 40. Council of Power Utilities (1997). Salient Features of Electricity Generation, Capacity and Consumption in Some Countries, January. New Delhi. 41. Crafts, N.F.R., Leybourne, S.J., and Mills, T.C. (1989 a), ‘Trends and Cycles in British Industrial Production, 1700-1913’, Journal of the Royal Statistical Society, Series A, 152: 43-60. 42. Crafts, N.F.R., Leybourne, S.J., and Mills, T.C. (1989 b), ‘Economic Growth in Nineteenth Century Britain: Comparisons with Europe in the Context of Gerschenkron’s Hypothesis’, Warwick Economic Research Paper 308. 3

43. Crew, Michael A. and Kleindorfer, Paul R. (1979), Public Utility Economics, St. Martin's Press, New York. 44. Culy, J.G., Read, E. G., and Wright, B. D. (1996). ‘The Evolution of New Zealand’s Electricity Supply Structure’, in Gilbert and Kahn (1996 b). 45. Davidson, J.E.H., Hendry, D.F., Srba, F. and Yeo, S. (1978), ‘Econometric Modeling of the Aggregate Time Series Relationship Between Consumers’ Expenditure and Income in the United Kingdom’, Economic Journal, 88: 661692. 46. De Jong, D.N., Nankervis, J.C., Savin, N.E. and Whiteman, C.H. (1992), ‘The Power Problems of Unit Root Tests for Time Series with Autoregressive Errors’, Journal of Econometrics, 53: 323-343. 47. De Jong, D.N. and Whiteman, C.H. (1991 a), ‘The Temporal Stability of Dividends and Stock Prices: Evidence from the Likelihood Function’, American Economic Review, 81: 600-617. 48. De Jong, D.N. and Whiteman, C.H. (1991 b), ‘trends and Random Walks in Macroeconomic Time Series: A Reconsideration Based on the Likelihood Principle’, Journal of Monetary Economics, 28: 221-254. 49. Deshpande, M. V. (1966), Elements of Electrical Power Station Design, Sir Isaac Pitman and Sons Ltd., London. 50. Dickey, D.A. (1976), Estimation and Asymptotic Testing for Non-Stationary Time Series, Ph.D. Dissertation, Iowa State University. 51. Dickey, D.A., Bell, W.R. and Miller, R.B. (1986), ‘Unit Roots in Time Series Models: Tests and Implications’, American Statistician, 40: 12-26. 52. Dickey, D.A. and Fuller, W.A. (1979), ‘Distribution of the Estimators of Autoregressive Time Series with a Unit Root’, Journal of the American Statistical Association, 74: 427-431. 53. Dickey, D.A. and Fuller, W.A. (1981), ‘Likelihood Ratio Statistics for Autoregressive Time Series with a Unit Root’, Econometrica, 49: 1057-1072. 54. Dickey, D.A. and Pantula, S. (1987), ‘Determining the Order of Differencing in Autoregressive Processes’, Journal of Business and Economic Statistics, 5: 455461.

4

55. Dickey, D.A., Jansen, D.W. and Thornton, D.L. (1991), ‘A Primer on Cointegration with Application to Money and Income’, Review, Federal Reserve Bank of St. Louis, 3: 58-78. 56. Diebold, F.X. and Nerlove, M. (1990), ‘Unit Roots in Economic Time Series: A Selective Survey’ in Rhodes and Fomby (Eds.) (1990). 57. Dixit, S., Sant, G., and Wagle, S. (1998). ‘WB – Orissa Model of Power Sector Reform: Cure Worse Than Disease’, Economic and Political Weekly, April 25, pp. 944 – 949. 58. Dolado, J.J., Jenkinson, T. and Sosvilla-Rivero, S. (1990), ‘Cointegration and Unit Roots’, Journal of Economic Surveys, 4: 249-273. 59. Doornik, J.A. and Hansen, H. (1994), ‘A Practical Test for Univariate and Multivariate Normality’, Discussion Paper, Nuffield College. 60. Doornik, J.A., and Hendry, D.F. (1997), Modeling Dynamic Systems Using PcFiml 9.0 for Windows, International Thomson Business Press, London. 61. Downs, A. (1967), Inside Bureaucracy, Little, Brown, Boston. 62. D’Sa, Antonette, Murthy, K. V. N., and Reddy, A. K. N. (1999). ‘India’s Power Sector Liberalization: An Overview’, Economic and Political Weekly, June 5, pp. 1427-1434. 63. Easterbrook, G. (2001), ‘Brown and Out’, Span, 42(3), May-June: 42-45. 64. Engle, R.F. (1982), ‘Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation’, Econometrica, 50: 987-1007. 65. Engle, R.F. (1983), ‘Estimates of the Variance of US Inflation Based on the ARCH Model’, Journal of Money, Credit and Banking, 15: 286-301. 66. Engle, R.F., Hendry, D.F. and Richard, Econometrica, 51: 277-304.

J.F. (1983), ‘Exogeneity’,

67. Engle, R.F., Hendry, D.F. and Trumbull, D. (1985), ‘Small Sample Properties of ARCH Estimators and Tests’, Canadian Journal of Economics, 43: 66-93. 68. Engle, R.F., and Granger, C.W.J. (1987), ‘Cointegration and Error Correction: Representation, Estimation and Testing’, Econometrica, 55(2): 251-276. 69. Engle, R.F. and Yoo, B.S. (1987), ‘Forecasting and Testing in Cointegrated Systems’, Journal of Econometrics, 35: 143-159.

5

70. Feige, E.L. (1997), ‘Underground Activity and Institutional Change: Productive, Protective and Predatory Behaviour in Transition Economies’, in Transforming PostCommunist Political Economies, National Research Council, National Academy Press, Washington DC. 71. Frisch, R. and Waugh, F.V. (1933), ‘Partial Time Regressions as Compared with Individual Trends’, Econometrica, 1:387-401. 72. Fuller, W.A. (1976), Introduction to Statistical Time Series, Wiley, New York. 73. Fuller, W.A. (1985), ‘Non-Stationary Autoregressive Time Series’, in E.J. Hannan, P.R. Krishnaiah and M.M. Rao (eds.) (1985), Handbook of Statistics, Vol. 5: Time series in the Time Domain, 1-24, North-Holland, Amsterdam. 74. Galbraith, J.K. (1998), Created Unequal: The Crisis in American Pay, The Free Press, New York. 75. Geweke, J. (1982), ‘Measurement of Linear Dependence and feedback Between Time Series’, Journal of the American Statistical Association, 79: 304-324. 76. Gilbert, Richard J., and Kahn, Edward P. (1996 a). ‘Competition and Institutional Change in US Electric Power Regulation’, in Gilbert and Kahn (1996 b), pp. 179 – 230. 77. Gilbert, Richard J. and Kahn, Edward P. (1996 b) (ed.) ‘International Comparisons of Electricity Regulation’, Cambridge University Press. 78. Gonzalo, J. (1994), ‘Five Alternative Methods of Estimating Long-Run Relationships’, Journal of Econometrics, 60: 203-233. 79. Government of India (1964 a), Report of the Santhanam Committee on Prevention of Corruption, Ministry of Home Affairs, New Delhi. 80. Government of India (1964 b). Report of the High Level Committee on Power. Ministry of irrigation and Power, New Delhi. 81. Government of India. (1974), Report of the Fuel Policy Committee of India, New Delhi. 82. Government of India (1974), Line Loss Reduction in Primary and Secondary Distribution, Research Scheme on Power, Review No. 4, Central Board of Irrigation and Power, New Delhi. 83. Government of India (1980). Report of the (Rajadhyaksha) Committee on Power. Ministry of Energy, Department of Power. New Delhi.

6

84. Government of India (1986), Energy Conservation: Challenges and Opportunities, Advisory Board on Energy, August, New Delhi. 85. Government of India (1992 a), Sarvekshna, Vol. 16 (1), July – September. 86. Government of India (1992 b). Economic Survey – 1991-92. 87. Government of India (1996). The India Infrastructure Report. Vol. III, Sector Reports. Expert Group on the Commercialisation of Infrastructure Projects. For Ministry of Finance, NCAER, New Delhi. 88. Government of India (1996-97). Public Electricity Supply, All India Statistics, General Review, CEA, New Delhi. 89. Government of India (1998 a), Statistical Abstract – India, CSO, New Delhi. 90. Government of India (1998 b), Unorganised Manufacturing Enterprises in India, NSS 51st Round, 1994-95, NSSO, New Delhi. 91. Government of India (1998 c). The Electricity Regulatory Commissions Act, 2 July. New Delhi. 92. Government of India (1999 a). Annual Report of the Working of SEBs and EDs, April, Planning Commission, New Delhi. 93. Government of India (1999 b). Annual Report 1998-99, Ministry of Power. New Delhi. 94. Governemnt of India (2000). Conference of Power Ministers: Agenda Notes. 26 February, Ministry of Power, New Delhi. 95. Government of India (2001), Economic Survey 2000-2001, Ministry of Finance, New Delhi. 96. Government of Kerala, Economic Review, (different volumes), State Planning Board, Thiruvananthapuram. 97. Government of Kerala (1967). Report of the Kerala State Electricity Board Finances Enquiry Commission (Single-Member Commission of P.S. Padmanabhan). October. Government Press, Trivandrum. 98. Government of Kerala (1968). Report of Enquiry by One Man Commission (K.P.S. Nair) on Transmission and Distribution of the KSEB. August. Government Press, Trivandrum.

7

99. Government of Kerala (1978) Report of the Steering Committee on Energy: Plan Proposals – 1978 – 83. August. State Planning Board, Trivandrum. 100. Government of Kerala (1984), Report of the High Level Committee on Industry, Trade and Power, Vol. III, Report on Power Development, May, State Planning Board, Trivandrum. 101. Government of Kerala (1997). Report of the Committee to Study the Development of Electricity (Chairman: E. Balanandan). February. 102. Government of Kerala (1998 a). Report of the (K. P. Rao) Expert Committee to Review the Tariff Structure of KSEB, May 1998. 103. Government of Kerala (1998 b). Organizational Self-Assessment: Reports of the Working Groups – Discussion Paper, KSEB and EISP, Thiruvananthapuram. 104. Government of Kerala (2000), Economic Review – 2000, Kerala State Planning Board, Thiruvananthapuram. 105. Granger, C.W.J. (1969), ‘Investigating Causal Relations by Econometric Models and Cross-Spectral Methods’, Econometrica, 37: 424-438. 106. Granger, C.W.J. (1981), ‘Some Properties of Time Series Data and Their Use in Econometric Model Specification’, Journal of econometrics, 16(1): 121130. 107. Granger, C.W.J. (1983), ‘Cointegrated Variables and Error Correcting Models’, UCSD Discussion Paper, 83-13. 108. Granger, C.W.J. and Newbold, P. (1974),’Spurious Regressions in Econometrics’, Journal of Econometrics 2: 111-120. 109. Granger, C.W.J. and Newbold, P. (1977), Forecasting Economic Time Series, Academic Press, New York. 110. Green, Richard (1998), ‘England and Wales: A Competitive Electricity Market?’, Working Paper Series No. pwp-60, POWER, UCEI, Berkeley. 111. Green, Richard (1999). ‘The Electricity Contract Market in England and Wales’, The Journal of Industrial Economics, Vol. XLVII No. 1, March, pp. 107 – 124. 112. Griffiths, W.E., Hill, R.C. and Judge, G.G. (1993), Learning and Practicing Econometrics, John Wiley & Sons, New York. 113. Gupta, B. R. (1983). Generation of Electrical Energy, Eurasia Publishing House, New Delhi.

8

114. Hall, S.G. (1991), ‘The Effect of Varying Length of VAR Models on the Maximum Likelihood Estimates of Cointegrating Vectors’, Scottish Journal of Political Economy, 38: 317-323. 115. Halvorsen, R. (1975), ‘Residential Demand for Electric Energy’, Review of Economics and Statistics, 57: 12-18. 116. Hansen, B.E., (1992), ‘Testing for Parameter Instability in Linear Models’, Journal of Policy Modelling, 14: 517-533. 117. Harris, M. and Raviv, A. (1978), ‘Some Results on Incentive Contracts with Applications to Education and Employment, Health Insurance and Law Enforcement’, American Economic Review, 68(1), March: 20-30. 118. Hart, O.D. and Holmstrom, B.R. (1987), ‘The Theory of Contracts’, in Bewley, T.F. (ed.) Advances in Economic Theory, Fifth World Congress, cambridge University Press, New York, 71-155. 119. Harvey, A.C. (1981), The Econometric Analysis of Time Series, Philip Allan, Deddington. 120. Hasza, D.P. and Fuller, W.A. (1979), ‘Estimation for Autoregressive Processes with Unit Roots’, Annals of Statistics, 7: 1106-1120. 121. Henderson, P. D., (1875), India: The Energy Sector, Oxford University Press. 122. Hendry, D.F. and Doornik J.A. (1999), Empirical Econometric Modelling using PcGive, Vol 1, Timberlake Consultants Ltd., Kent. 123. Hendry, D.F., Pagan, A.R. and Sargan, J.P. (1984), ‘Dynamic Specification’ in Z. Grilliches and M.D. Intrilligator (eds.) Handbook of Econometrics II, North-Holland, Amsterdam, 1023-1100. 124. Herman, F. K. (1974), Electricity Tariffs, University of Stellenbosch, South Africa. 125. Holden, K., Peel, D.A., and Thompson, J.L. (1990), Economic Forecasting: An Introduction, Cambridge University Press. 126. Holland, P.W. (1986), ‘Statistics and Causal Inference’, Journal of the American Statistical Association, 81: 945-960. 127. Holmstrom, B.R. (1979), ‘Moral Hazard and Observability’, Bell Journal of Economics, 10(1) Spring: 74-91.

9

128. Holmstrom, B.R. and Tirole, J. (1989), ‘The Theory of the Firm’, in Schmalensce, R. and Willig, R.D. (eds.) Handbook of Industrial Organisation, Vol. 1 North Holland, Amsterdam: 61-133. 129.

IMF (1989) Annual Report.

130. International Chamber of Commerce (ICC) (1998) Privatisation of the Energy sector, ICC Publishing S. A., Paris.

Liberalisation and

131. International Energy Initiative, (March 1998), Least Cost Planning Exercise for Kerala State Electricity Board: Draft Report, Bangalore. 132. IRTC and IEI (1996), Exercises for Integrated Resource Planning for Kerala: End-Use Analysis – An Empirical Study: Technical Report I – Electricity. 133. Jha, Prem Shankar (1994). “More Power: A Quicker, Cheaper Way”. Economic Times, April 29. 134. Johansen, S. (1988), ‘Statistical Analysis of Cointegration Vectors’, Journal of Economic Dynamics and Control, 12: 231-254. 135. Johansen,S., (1992), ‘Testing Weak Exogeneity and the Order of Cointegration in U.K. Money Demand Data’, Journal of Policy Modeling, 14(3): 313-334. 136. Johansen, S. and Juselius, K. (1990), ‘Maximum Likelihood Estimation and Inference on Cointegration – with Application to the Demand for Money’, Oxford Bulletin of Economics and Statistics, 52: 169-210. 137. Joskow, Paul L. (1998). ‘Regulatory priorities for Infrastructure Sector Reform in Developing Countries’, in Annual world Bank Conference on Development Economics, 1998, pp. 191 – 223, World Bank, Washington DC. 138. Kalecki, M. (1967), 'Observations on Social and Economic Aspects of 'Intermediate Regimes'', Coexistence, 4 (1): 1-5. Reprinted in Selected Essays on the Economic Growth of the Socialist and the Mixed Economy, Cambridge University Press, 1972. 139. Kannan, K.P. and Pillai, N. Vijayamohanan (2001), ‘Plight of the Power Sector in India’, Economic and Political Weekly, Vol. 36 (2 & 3), January 13 & 20, 130-139; 234-246. 140. Kaufman, D. and Siegelbaum, P. (1997), Privatisation and Corruption in Transition Economies’, Journal of International Affairs, 50 (2): 419-458.

10

141. Kerala State Electricity Board, Power System Statistics (different volumes), Thiruvananthapuram. 142. Kerala State Electricity Board, Annual Administration Report (different volumes), Thiruvananthapuram. 143. Kerala State Electricity Board, Annual Statement of Accounts (different volumes), Thiruvananthapuram 144. Killick, Tony. (1993). The Adaptive Economy: Adjustment Policies in Small, Low-Income Countries, World Bank, Washington D.C. 145. Koop, G. (1992), ‘“Objective” Bayesian Unit Root Tests’, Journal of Applied Econometrics, 7: 65-82. 146. Kreuger, A.O. (1974), ‘The Political Economy of the Rent-Seeking Society’, American Economic Review, 64 (3) June 291-303. 147. Laffont, J.J. (1988), ‘Hidden Gaming in Hierarchies: Facts and Models’, Economic Record, 64 (187) December: 295-306. 148. Laffont, J.J. (1990), ‘Analysis of Hidden Gaming in a Three-Level Hierarchy’, Journal of Law, Economics and Organisation 6 (2), Fall 301-324. 149. Laffont, Jean-Jacques (1996). ‘The French Electricity Industry’, in Gilbert and Kahn (1996, b), pp. 406 – 456. 150. Lange, O. and Taylor, F. M. (1938), On the Economic Theory of Socialism, University of Minnesota Press, Minneapolis. 151. Lee, Henry and Darani, Negeen (1995), ‘Restructuring and the Environment’, CSIA Discussion Paper 95-13, Kennedy School of Government, Harvard University, December. 152. Leibenstein, H., (1976). Beyond Economic Man – A New Foundation for Microeconomics, Harvard University Press, Massachusetts. 153. Lencz, Imrich (1977), ‘Prognoses of the Consumption of Energy, The Czechoslovak Experience’, p. 85, in William D. Nordhaus (ed.) International Studies of the Demand for Energy, 1977, North-Holland, Amsterdam, pp. 83 – 94. 154. Levinthal, D. (1988), ‘A Survey of Agency Models of Organisation’, Journal of Economic Behaviour and Organisation, 9 (2), March: 153-185.

11

155. Littlechild, Stephen C. (1983), Regulation of Telecommunications’ Profitability, HMSO, London. 156. Littlechild, Stephen C. (2000), ‘A Review of UK Electricity Regulation 19992000’, DAE Working Paper 0026, University of Cambridge, November. 157. Ljung, G.M. and Box, G.E.P. (1978) ‘On a Measure of Lack of Fit in Time Series Models’, Biometrika, 65: 297-303. 158. Mackay, R. and Weaver, C. (1978), ‘Monopoly Bureaus and Fiscal Outcomes: Deductive Models and Implications for Reform’, in Tullock, G. and Wagner, R. (eds.) Policy Analysis and Deductive Reasoning, Spring, D.C. Heath. 159. MacKinnon, J.G. (1990), ‘Critical Values for Cointegration Tests’, UC San Diego Discussion Paper 90-4; reprinted in R.F. Engle and C.W.J. Granger (eds.) (1991) Long-Run Economic Relationships: Readings in Cointegration, chap. 13, Oxford University Press, New York. 160. Macrakis (ed.) (1974), Energy: Demand, Conservation and Institutional Problems, The MIT Press, Cambridge, MA. 161. Mahalingam, Sudha (2000). “Power Reforms in Trouble”. Frontline, March 17, pp. 94-96. 162. Malia, M. (1995), ‘The Nomenklatura Capitalists: Who’s Running Russia Now?’, The New Republic, 212 (21): 17-22. 163. Manzetti, L. (1994), Privatisation South American Style, Oxford University Press, Oxford. 164. Massie, R. K. (1980), Peter the Great: His Life and World, Alfred A. Knopf, New York. 165. McChesney, F.S. (1987), ‘Rent Extraction and Interest Group Formation in a Coasean Model of Regulation’, Journal of Legal Studies 20: 73-90. 166. McChesney, F.S. (1997), Money for Nothing, Harvard University Press, Cambridge. 167. Mehta, A. (2000), Power Play: A Study of the Enron Project, Orient Longman, New Delhi. 168. Menon, R.V.G. (1999), 'Energy Planning: The Need for a Departure from the 'Kerala Model'' , in Oommen M.A. (ed.) Rethinking Development: Kerala's Development Experience, Vol. 2: 459-469, Institute of Social Sciences and Concept publishing Co., New Delhi.

12

169. Mills, T.C., (1990), Time Series Techniques for Economists, Cambridge University Press. 170. Mirrlees, J.A. (1976), ‘The Optimal Structure of Incentives and Authority Within an Organisation’, Bell Journal of Economics, 7(1) Spring 105-131. 171.

Mishra, R. (1984), The Welfare State in Crisis, Wheat-sheaf Books, Brighton.

172. Mohammad, S. and Whalley, J. (1984), ‘Rent Seeking in India: Its Costs and Policy Significance;, Kyklos, 37 (3): 387-413. 173. Moore, T. G. (1970), 'The Effectiveness of Regulation of Electric Utility Prices', Southern Economic Journal, 36 (April): 365-375. 174. Morris, Sebastian (1990), ‘Cost and Time Overruns in Public Sector Projects’, Economic and Political Weekly, November 24: M-154 – M-168. 175. Morris, Sebastian (1996). ‘Political Economy of Electric Power in India’, Economic and Political Weekly, Vol. 31, Nos. 20 & 21, May 18 & 25. 176. Morris, Sebastian, (2000). “Regulatory Strategy and Restructuring Model for Gujarat Power Sector”, Economic and Political Weekly, June 3, pp. 1915-1928. 177. Mukherjee, S. (1998). ‘Fis Seek Quality Escrow Accounts for Loan to IPPs’ The Economic Times, November 14: 1. 178. Munasinghe, Mohan and Warford, Jeremy J. (1982), Electricity Pricing: Theory and Case Studies, The Johns Hopkins University Press, Baltimore. 179. Myrdal, G. (1968), Asian Drama: An Enquiry into the Poverty of Nations, Penguin Books, Middlesex. 180. National Working Group on Power Sector (1994) Current Power Policies: A Critique, September/October. 181. Nelson, C.R., and Kang, H. (1981), ‘Spurious Periodicity in Inappropriately Detrended Time Series’, Econometrica, 49:741-751. 182. Nelson, C.R., and Kang, H. (1984), ‘Pitfalls in the Use of Time as an Explanatory Variable in Regression’, Journal of Business and Economics, 2: 73-82. 183. Nelson, C. R. and Plosser, C. I. (1982), ‘Trends and Random Walks in Macroeconomic Time Series: Some Evidence and Implications’, Journal of Monetary Economics, 10:139-162.

13

184. Newberry, David M., and Green, Richard (1996). ‘Regulation, Public Ownership and Privatisation of English Electricity Industry’, in Gilbert and Kahn (1996 b). 185. Newberry, David M., and Pollitt, M. G. (1997), The Restructuring and Privatisation of Britain’s CEGB: Was It Worth It?’, Journal of Industrial Economics, 45(3): 269-303. 186. Newbery, David M. (1998). ‘Rate of Return Regulation Versus Price Regulation for Public Utilities’, in Newman (1998), pp. 205 – 210. 187. Newman, Peter (ed., 1998). The New Palgrave Dictionary of Economics and the Law, Vol. 3, Macmillan Reference Ltd., London. 188. Niskanen, W. A. (1971), Bureaucracy and Representative Government, Aldine, Chicago. 189. Niskanen, W. A. (1973), Bureaucracy, Servant or Master ? Institute of Economic affairs, London. 190. Olson, M. (1965), The Logic of Collective Action, Harvard University Press, Cambridge, Mass. 191. Pachauri, R. K., (1975) Dynamics of Electrical Energy Supply and Demand: An Economic Analysis, Praeger Publishers, New York. 192. Pagan, A.R. (1989), ‘20 Years After: Econometrics 1966 – 1986’, in B. Cornet and H. Tulkens (eds.), Contributions to operations Research and Econometrics, The MIT Press, Cambridge, MA. 193. Parikh, Jyoti K. (1981) Modelling Energy Demand for Policy Analysis, Planning Commission, Government of India, New Delhi. 194. Parikh, J., et al., (1994), Planning for Demand Side Management Options in the Electricity Sector, Tata Mcgraw-Hill, New Delhi. 195. Parikh, K. S. (2001), ‘Thinking Through the Enron Issue’, Economic and Political Weekly, 36(17), April 28: 1463-1472. 196. Peltzman, S. (1971), 'Pricing in Public and Private Enterprises: Electric Utilities in the United States', Journal of Law and Economics, April. 197. Peltzman, S. (1976), ‘Toward a More General Theory of Regulation’, Journal of Law and Economics, 19 August: 211-240.

14

198. Perron, P. (1988) ‘Trends and Random Walks in Macroeconomic Time Series: Further Evidence from a New approach’, Journal of Economic dynamics and Control, 12: 297-332. 199. Perron, P. (1989), ‘The Great Crash. The Oil Price Shock and the Unit Root Hypothesis’, Econometrica, 57: 1361-1401. 200. Perron, P. and Ng, S. (1996), ‘Useful Modifications to Some Unit Root Tests with Dependent errors and Their Local Asymptotic Properties’, Review of Economic Studies, 63: 435-465. 201. Pescatrice, D. R. and Trapani III, J. M. (1980), 'The Performance and Objectives of Public and Private Utilities Operating in the United States', Journal of Public Economics, 13 (April): 259-276. 202. Phillips, P.C.B., (1986),’Understanding Spurious Econometrics’, Journal of Econometrics, 33: 311-340.

Regression

in

203. Phillips. P.C.B. (1991 a), ‘to Criticise the Critics: An Objective Bayesian Analysis of Stochastic Trends’, Journal of Applied econometrics, 6: 333-364. 204. Phillips, P.C.B. (1991 b), ‘Bayesian Routes and Unit Roots: De Rebus Prioribus Semper est Disputandum’, Journal of Applied Econometrics, 6: 435474. 205. Phillips, P.C.B., and Durlauf, S.N., (9186), ‘Multiple Time Series Regression with Integrated Processes’, Review of Economic Studies, 53: 473495. 206. Phillips, P.C.B. and Ouliaris, S. (1990), ‘Asymptotic Properties of Residual Based Tests for Cointegration’, Econometrica, 58: 165-193. 207. Phillips, P.C.B. and Perron, P. (1988), ‘Testing for a Unit Root in Time series regression’, Biometrica, 75: 335-346. 208. Pillai, P. P. (1981) Dynamics of Electricity Supply and Demand in Keralam – A Macro-Econometric Analysis, Agricole publishing Academy, New Delhi. 209. Pillai, N. Vijayamohanan (1991). Seasonal Time-of-Day Pricing of Electricity Under Uncertainty – A Marginalist Approach to Kerala Power System. Unpublished Ph. D. Thesis. University of Madras, Chennai. 210. Pillai, N. Vijayamohanan (1995) “Liberalisation of Poverty” Paper presented at a seminar on ‘Economic Issues Relating to Social Welfare in Post-Liberalisation India’ on February 10 in the Department of Econometrics, Madras University, Chennai .

15

211. Pillai, N. Vijayamohanan (1999 a) ‘Reliability Analysis of a Power Generation System: A Case Study’, Productivity, July – September, pp. 310-318. 212. Pillai, N. Vijayamohanan (1999 b), ‘Adoption of Energy Efficient Lamps in Keralam’, Productivity, 40(3), October-December: 440-450. 213. Plosser, C.I. and Schert, W.G., (1978), ‘Money, Income and Sunspots: Measuring Economic Relationships and the Effects of Differencing’, Journal of Monetary Economics, 4: 637-660. 214. Posner, R. A. (1969), 'Natural Monopoly and Its Regulation', Stanford Law Review, 21, February: 548-643. 215. Posner, R. A. (1975), ‘The Social Costs of Monopoly and Regulation’, Journal of Political Economy, 83 August: 807-827. 216. Quinglian, He (2000), China’s Descent into a Quagmire, reprinted in The Chinese Economy, 33 (3) May-June. 217. Raj, K. N. (1973), 'The Politics and Economics of Intermediate Regimes'', Economic and Political Weekly, July 7: 1189-1198. 218. Ramsey, J.B. (1969), ‘Tests for Specification Errors in Classical Linear Least Squares Regression Analysis’, Journal of the Royal Statistical Society B, 31: 350-371. 219. Rao, M. Govinda, Shand, R. T., and Kalirajan, K. P. (1998-99). ‘State Electricity Boards: A Performance Evaluation’, The Indian Economic Journal, Vol. 46 (2), October-December. 220. Rao, S.L. (2000). “Electricity Reform and Regulation: Some Issues”, Economic and Political Weekly, June 24, pp. 2231-2234. 221. Reddy, A.K.N. (2001), ‘California Energy Crisis and Its Lessons for Power Sector Reform in India’, Economic and Political Weekly, 36(18), May 5: 1533-1540. 222. Reddy, A.K.N., D’Sa, Antonette, and Murthy, K.V.N. (2000). “The Curate’s Egg”, Deccan Herald, March 31. 223. Reddy, S.B. (1995), Energy Efficient Options: Techno-Economic Potentials for Mitigating Greenhouse Gas Emissions, Report prepared for Energy Management Centre, Ministry of Power, Government of India, New Delhi. 224. Reimers, H.E. (1992), ‘Comparisons of Cointegration’, Statistical Papers, 33: 335-359.

Tests

for

Multivariate

16

225. Rhodes, G.F. and Fomby, T. B. (eds.) (1990), Advances in Econometrics, Vol. 8: 3-69, JAI Press, Greenwich, CT. 226. Rose, N.L. (1987), ‘Labour Rent-Sharing and Regulation: Evidence from the Trucking Industry’, Journal of Political Economy, 95: 1146-1178. 227. Rose-Ackerman, S. (1999), Corruption and Government: Causes, Consequences and Reform, Cambridge University Press, Cambridge. 228. Ross, S. (1973), ‘The Economic Theory of Agency: the Principal’s Problem’, American Economic Review, 63 (2), May: 134-139. 229.

Samuelson, P. A. (1970), Economics, McGraw-Hill, 8th edn., New York.

230. Sankar, T.L., and Ramachandra, Usha (2000). Electricity Tariffs Regulators: The Orissa Experience” Economic and Political Weekly, May 27, pp. 1825-1834. 231. Sant, Girish and Dixit, Shantanu (1996) ‘Beneficiaries of IPs Subsidy and Impact of Tariff Hike’ Economic and Political Weekly, Vol. 31, No. 15, December 21: 33153321. 232. Sargan, J.D. (1964),’Wages and Prices in the United Kingdom: A Study in Econometric Methodology’, in P.E.Hart, G. Mills and J.K. Whitaker (eds.), Econometric Analysis for National Economic Planning, Butterworth, London; reprinted in D.F. Hendry and K.F. Wallis (eds.), Econometrics and Quantitative Economics, 1984 Basil Blackwell, Oxford. 233. Sargan, J.D. and Bhargava, A.S. (1983), ‘Testing Residuals from Least Squares regression for Being Generated by the Gaussian Random Walk’, Econometrica, 51:153-174. 234. Schmidt, P. and Phillips, P.C.B. ((1992), ‘LM Test for a Unit Root in the Presence of Deterministic Trends’, Oxford Bulletin of Economics and Statistics, 54: 257-287. 235. Schwert, G.W. (1989), ‘Tests for Unit Roots: A Monte Carlo Investigation’, Journal of Business and Economic Statistics, 7: 147-159. 236. Shah, C. M., Dalal, C. M., and Patel, K. H. (1985). ‘Minimization of Losses in Power System – A Case Study of Reduction of Distribution Losses from 16 % to 10 %’ Irrigation and Power Journal, Vol. 42 (3), pp. 275 – 284. 237. Sims, C.A. (1988), ‘Bayesian Scepticism of Unit Root Econometrics’, Journal of Economic Dynamics and Control, 12: 463-475.

17

238. Sims, C.A. and Uhlig, H. (1991), ‘Understanding Unit Rooters: A Helicopter Tour’, Econometrica, 59: 1591-1599. 239. Smiley, R. H. and Greene, W. H. (1983), 'Determinants of the Effectiveness of Electric Utility Regulation', Resources and Energy, 5: 65-81. 240. Srinivasan, K. (1996), ‘Indian Power Policy, Enron and the BoP’, Economic and Political Weekly, 17 August: 2207-2214. 241. Stigler, G.J. (1971),’The Economic Theory of Regulation’, Bell Journal of Economics, 2 Spring: 3-21. 242. Stigler G.J. and Friedland, C. (1962), 'What Can Regulators Regulate? The Case of Electricity', Journal of Law and Economics, 5 (October): 1-16. 243. Stiglitz, J.E. (1975), ‘Incentives, Risk and Information: Notes Towards a Theory of Hierarchy’, Bell Journal of Economics 6(2) Autumn 552-579. 244. Tansil, J. and Moyers, J.C. (1974) ‘Residential Demand for Electricity’, in Macrakis (ed.) (1974), 375-385. 245. Taylor, L. D. (1977) 'The Demand for Energy: A Survey of Price and Income Elasticities', in William D. Nordhaus (ed.) (1977). 246. Thakur, U. (1979), Corruption in Ancient India, Abhinav Publications, New Delhi. 247. Tirole, J. (1986), ‘Hierarchies and Bureaucracies: On the Role of Collusion in Organisations’, Journal of Law, Economics and Organisation, 2 (2), Fall: 181-214. 248.

Tollison, R.D. (1982), ‘Rent Seeking: A Survey’, Kyklos 35 (4): 575-602.

249. Taylor, L. D. (1975) 'The demand for electricity: a survey', Bell Journal of Economics and Management Science, Vol. 6, No. 1, (Spring). 250. Tullock, G. (1967), ‘The Welfare Costs of Tariffs, Monopolies and Theft’, Western Economic Journal 5 June 224-232. 251. Tullock, G. (1980), ‘Rent Seeking as a Negative-Sum Game’, in Buchanan, J. M., Tollison, R.D. and Tullock, G. (eds.) Toward a Theory of the Rent-Seeking Society, Texas A & M University Press, College Station: 16-36. 252. Turvey, Ralph (1968), Optimal pricing and Investment in Electricity Supply, MIT Press, Cambridge, Mass.

18

253. Turvey, Ralph and Anderson, Dennis (1977), Electricity Economics, The John Hopkins University Press, Baltimore. 254. Tyner, W.E. (1978), Energy Resources and Economic Development in India, Martinus Nijhoff Social Sciences Division, Boston. 255. Tyrrell, T.J. (1974), ‘Projections of Electricity Demand’, in Macrakis (ed.) (1974), 342-359. 256. Unnikrishnan, P.V., et al. (1997), 'A Demand Side management Model for Low Tension Electricity Consumers in Kerala', Proceedings of the Ninth Kerala Science Congress, January, Thiruvananthaapuram: 422-436. 257. Upadhyay, A.K. (1996), ‘Power at Any Price’, The Economic Times, January. 258. Upadhyay, A.K. (2000), ‘Power Sector Reforms: Indian Experiences and Global Trends’, Economic and Political Weekly, March 18: 1023-1028. 259. Vickers, J. and Yarrow, G. (1988). Privatisation: An Economic Analysis, MIT Press, Cambridge, MA. 260. Wiener, N. (1956),’The Theory of Prediction’, in E.F. Beckenback (ed.), Modern Mathematics for Engineers, McGraw-Hill, New York ,pp. 165-190. 261. Weisskopf, T. (1992), Russia in Transition: Perils of the Fast Track to Capitalism’, Challenge, no.11: 28-37. 262. White, H. (1980), ‘A Heteroscedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroscedasticity’, Econometrica, 48: 817-838. 263. Williamson, O.E. (1968),’Economies as an Antitrust Defence: the Welfare Tradeoffs’, American Economic Review, 58 March: 18-36. 264. World Bank (1979), India: Economic Issues in the Power Sector, A World Bank Country Study, November. 265. World Bank (1992), World Development Report: Development and the Environment, Oxford University Press. 266. World Bank (1995). Economic Developments in India: Achievements and Challenges, World Bank country study, Washington D.C. 267. World Bank (1996), India: Five Years of Stabilisation and Reform and the Challenges Ahead, Washington D. C.

19

268. World Bank (1997), World Development Report 1997: The State in a Changing World, Oxford University Press, New York. 269. World Bank (1998 a). Adjustment Lending: An Evaluation of Ten Years of Experience, Washington DC. 270.

World Bank (1998 b). India 1998 Macroeconomic Update. Washington D.C.

271. World Bank (2000 a), 2000 World Development Indicators, Development Data Centre, Washington DC. 272. World Bank (2000 b), World Development Report: Entering the 21st Century, Oxford University Press. 273. World Bank (2000 c), The Private Sector and Power Generation in China, World Bank Discussion Paper No. 406, February, Washington DC. 274. Yule, G.V. (1926), ‘Why Do We Sometimes Get Nonsense Correlations Between Time Series ? A Study in Sampling and the Nature of Time Series’, Journal of the Royal Statistical Society, 89: 1-64. 275. Zachariah, K. C., et al. (1999). Migration in Kerala State, India: Dimensions, Determinants and Consequences, (mimeo) Centre for Development Studies, Thiruvananthapuram. 276. Zaidi, A.M. and Zaidi, S.D. (1981) (eds.) The Encyclopaedia of the Indian National Congress, Vol. 15: 1955-1957; S. Chand and Company Ltd., New Delhi. 277. Zellner, A. (1979), ‘Causality and Econometrics’, Carnegie-Rochester Conference Series 10, K. Brunner and A.H. Meltzer, (eds.), North Holland Publishing Company, Amsterdam, pp. 9-50.

.

20

Author Index Arrow, K. 183, 383 Averch, H. 220, 273, 274 Banerjee, Nirmala 109 Baumol, W.J. 379 Bhagwati, J. N. 301, 425 Bhargava, A. 115 Box, G. E. P. 106, 108 Brennan, T. J. 292, 297 Buchanan, J. M. 381 Cheng, B. S.

Nelson, C. R. 115, 152 Newberry, D. M. 290 Newbold, P. 118, 150 Niskanen, W. A. 381, 382

120

Dickey, D. A. 115, 153 D’Sa, Antonette 350, 354 Doornik, J. A. 106, 110, 150, 153, Engle, R. F.

Mehta, A. 410, 415, 429 Morris, Sebastian 54, 171, 343, 350, 370, 429, 430 Murthy, K. V. N. 350, 354 Myrdal, G. 400, 406, 432, 433

117

Fuller, W. A. 115 Galbraith, J. K. 403 Granger, C. W. J. 117, 118, 150, 153 Green, Richard 290, 292, 403 Halvorsen, R. 154 Hansen, B. E. 106, 107 Henderson, P. D. 154 Hendry, D. F. 110, 150, 153 Jenkins, G. M. 108 Jha, prem sankar 345 Johansen, S. 117 Johnson, L. 220, 273, 274 Joskow, Paul L. 288, 298 Juselius, K. 117 Kalecki, M. 376, 377, 386, 388 Kannan, K. P. 389, 404 Kreuger, A. O. 401, 425 Lange, O. 385, 401 Leibenstein, H. 42 Littlechild, S. C. 221, 275

Pachauri, R. K. 109 Pagan, A. R. 150 Parikh, Jyoti K. 23, 109 Parikh, K. S. 23 Peltzman, S. 287, 381, 425 Perron, P. 116 Phillips, P. C. B. 118 Pillai, P. P. 109, 154 Pillai, N. Vijayamohanan 23, 30, 109, 369, 389, 404 Plosser, C. I. 115, 152 Raj, K.N. 387 Rao, K. P. 230, 235, 236 Rao, M. Govinda 33, 47 Rao, S. L. 54 Reddy, A. K. N. 350, 354 Reimers, H. E. 117 Rose-Ackerman, S. 402, 431 Samuelson, P. A. 426 Srinivasan, T. N. 425 Stigler, G. J. 287, 380, 381, 425 Taylor. F. M. 401 Taylor, L. D. 102, 103, 123 Thakur, U. 375, 432 Tullock, G. 382, 425 Williamson, O. E. Zachariah, K. C.

379 40

1

Subject Index availability 29–31 Averch_Johnson effect 220, 273, 274– 275, 422–423 bilateral contracting 265 California’s power crisis 269, 289, 293, 294–296 competition in power sector 261 conditions for 262–264 expanded wholesale competition 264–265 retail competition 265 consumption commercial energy India 4–6, 23 other countries/world electricity India 38–40 other countries/world 38–39

1–2

2,

corruption (also see political economy) at the high-up 393–397 in energy theft 391–393 bursting 397–398 inflated capital costs 395–397 in private sectorisation 399–404 international penetration 408– 409 in time & cost overruns 183– 185, 394–395

cointegration 108, 117– 118 model adequacy 104–107, 110–114 unit roots 107–108, 115– 117 of the less correlatables 119– 126 electricity demand forecasting 98–104, 126–129 Annual Power Surveys (APS) 100–101 for Kerala 103–104, 126–129 electricity pricing; (also see regulation) general 217–218 cost plus/full cost/mark up 218–222 for Kerala power system 232–233 in regulated utilities 220–222 price capping 221 rate of return regulatory 220 marginal cost pricing 222–226 for Kerala power system 233–236 in France, etc. 237 Electricity regulatory Commission (ERC) 317 Central (CERC) 318–320 State (SERC) 319, 320 Enron entry of 410–415 effects of 416–424

electric power system 2–4 forced outage 30 electricity demand analysis econometric models 99–104 causality 98, 109, 118– 119, 150

Independent System Operator (ISO) 265, 269

2

‘intermediate regimes’ 376, 377, 386– 388 K. P. Rao Expert Committee 230–231, 235, 236, 371 loss of load probability (LOLP) 30 nationalisation 379, 380, 385–386, 389 natural monopoly problem 378–380 over-manning

41

plant load factor (PLF) 29–32, 58–59 political economy a generic analysis 378–388 of corruption 391–404 ‘principal-agent’ problem 382–385 private interest theories 380–382 Poolco/Power Pool/Exchange

265

Power Purchase Agreements (PPA) 91, 61, 349, power sector India general 6–9, 25–26 ilfare (also see corruption) 304–305, 307, 347, 389–390 organisation 299–301 power deficit 26, 28, 305–306 reforms (see power sector reforms) regulation 301–304 Kerala 11 reforms (see power sector reforms) power sector reforms the background 257–259 the agenda 261–284

experiences in Argentina 271–273 Chile 271 China 260 England 267–268 India a critique 339–358 Accelerated Power Development Programme 324–325 Electricity Bill, 2000 325–328, 390 liquid-fuel-based projects 315–316 mega power plants 314– 315 policy of 1991 312 private sector participation 311–314 problems 330–334 responses 328–329 reform background 305–310 transmission sector 316–317 Kerala 320, 334–339, 363 other States Andhra Pradesh 320, 322, 360 Haryana 320, 323, 361–362 Karnataka 320, 322, 362–363, 365 Orissa 320, 321, 353, 354–358, 408–409 Uttar Pradesh 323, 366 – 367 Scandinavia 269–270 South-East Asia 259–260 USA 268–269 lessons from the experiences 284–296 power tariff (also see SEBs; electricity pricing) in India 226–227

3

in Kerala 228–232 privatisation 398–399 corruption in (see corruption) Rajadhyaksha Committee Report 91, 226, 304–305, 307, 452 regulation, electricity supply industry 273–284 of rates fixed price 273, 276 rate of return 273–274, 278 price cap 274–276, 278 revenue 276 in Argentina 282 Canada 279–280 Chile 281–282 China 283–284 England 277 India (see power sector: India) Scandinavia 280–281 South-East Asia 282–283 USA 278–279 Santhanan Committee on Prevention of Corruption 400, 406, 427, 428, 432, 433 State Electricity boards (SEBs) capacity utilisation 26, 29–32 commercial losses 69–72 due to inefficiency 75–76 due to subsidised power sale 72–74 cost of depreciation and interest 64 - 66 fuel 56–57 O&M and E&A 63–64 power purchase 57–63 cost of inefficiency in accumulated interest charges 65–66 in establishment &

administration 64, 390 (also see over-manning) in technical operation 58–59 in project implementation (see time & cost overruns) in revenue collection 74–75 cost of the ‘over up’ of the power losses 76–78 financial performance 69–80 internal resources 79–80 organisational inefficiency 41–43 physical performance 27–43 power purchase 38 power theft 36, 202, 209, 210, 427 reform/restructuring (see power sector reforms: India) subsidised power sale 68–69 tariff and revenue realisation 66–69 technical inefficiency 28–36 T & D loss (see transmission (and distribution loss) time and cost over-runs 59–60, 157–195 causes 173–183 cost of delays 169–172 projects affected Azhutha Diversion 165 Chimony 163–164, 168 Idukki 158, 167, 168 Idamalayar 59, 60, 159, 167, 430 Kakkad 59, 60, 158, 159–160, 168 Kallada 59, 60, 161, 168 Kuttiady Tail Race 165 Kuttiady Extension 167, 168 – 169, 431 Kuttiar Diversion 165–166 Lower Periyar 59, 60, 161 Madupetty 60, 162–163, 167 Malampuzha 162, 167 Malankara 163 Peppara 60, 164, 168 Pooyankutty 162

4

Poringalkuthu LB Extension 164 Sabarigiri Augmentation 159 Vadakkeppuzha Diversion 166 Vazhikkadavu Diversion 166–167 ‘There is no alternative’ (TINA) 391, 398, 404–405, 439 third party access (TPA) 262–265, 270 transmission and distribution loss (T & D loss) causes 34–38, 200–210 cost savings 35–37, 210-213 in India 26, 33–38 in Kerala 33, 34, 36, 200 in other countries 33 in other states 33, 34, 36 non-technical loss 198–200, 208 technical loss 198–200 unbundling of power utility 263 Venkataraman Committee Report 226, 302–303, 306, 438, 439

5