India Migration Report 2012: Global Financial Crisis, Migration and Remittances 9780415634052

Table of contents :
Cover
Half Title
Title Page
Copyright Page
Dedication
Table of Contents
List of Tables
List of Figures
List of Abbreviations
Foreword
Preface
Acknowledgements
1. The Global Economic Crisis and Governance of Human Mobility
2. Migration and Development Linkages Re-examined in the Context of the Global Economic Crisis
3. The Global Economic Crisis and Impact on Migration from South-Asian and South-East Asian Countries: What are the Lessons to be Learned?
4. The Effect of the Global Economic Imbalance on Migrant Workers and Economies of the Gulf Cooperation Council
5. The Financial Crisis in the Gulf and its Impact on South Asian Migration and Remittances
6. The Dubai Model and the Impact of the Financial Crisis on South Asian Migrant Workers in the United Arab Emirates
7. Global Financial Crisis and its Consequences on Migrants in Qatar: Macro and Micro Perspectives
8. Global Financial Crisis and the Migrant Labour Market: A Study of Kuwait
9. Low-skilled Indian Construction Workers in the Gulf, Singapore and Malaysia: Return to India, Reintegration and Re-migration
10. Impact of the Global Recession on Migration and Remittances: The Kerala Experience
11. Global Financial Crisis and Return of South Asian Gulf Migrants: Patterns and Determinants of their Integration into Local Labour Markets
12. Inclusive Growth and Economic Crises: Labour Migration and Poverty in India
13. Food Inflation and Financial Crisis in India: Impact on Women and Children
14. Migration, Human Rights and Development
15. Remittances and Financial Participation: A Household-level Analysis in Kerala
16. International Labour Migration: Global Words, National Leads and Local Deeds
17. Broadening Exchanges and Changing Institutions: Multiple Sites of Economic Transnationalism
About the Editor
Notes on Contributors
Index

Citation preview

India Migration Report 2012

This page intentionally left blank

India Migration Report 2012 Global Financial Crisis, Migration and Remittances

Editor

S. IRUDAYA RAJAN

LONDON NEW YORK NEW DELHI

First published 2012 in India by Routledge 912 Tolstoy House, 15–17 Tolstoy Marg, Connaught Place, New Delhi 110 001

Simultaneously published in the UK by Routledge 2 Park Square, Milton Park, Abingdon, OX14 4RN

Routledge is an imprint of the Taylor & Francis Group, an informa business

© 2012 S. Irudaya Rajan

Typeset by Star Compugraphics Private Limited 5, CSC, Near City Apartments Vasundhara Enclave Delhi 110 096

All rights reserved. No part of this book may be reproduced or utilized in any form or by any electronic, mechanical or other means, now known or hereafter invented, including photocopying and recording, or in any information storage and retrieval system without permission in writing from the publishers.

British Library Cataloguing-in-Publication Data A catalogue record of this book is available from the British Library

ISBN 978-0-415-63405-2

To Professor K. N. Raj

This page intentionally left blank

Contents List of Tables List of Figures List of Abbreviations Foreword Preface Acknowledgements 1.

The Global Economic Crisis and Governance of Human Mobility Bimal Ghosh

2.

Migration and Development Linkages Re-examined in the Context of the Global Economic Crisis S. Irudaya Rajan and B. A. Prakash

3.

The Global Economic Crisis and Impact on Migration from South-Asian and South-East Asian Countries: What are the Lessons to be Learned? Vanessa Steinmayer

ix xiii xv xxi xxiii xxvii

1

14

27

4.

The Effect of the Global Economic Imbalance on Migrant Workers and Economies of the Gulf Cooperation Council Olga Marzovilla

47

5.

The Financial Crisis in the Gulf and its Impact on South Asian Migration and Remittances S. Irudaya Rajan and D. Narayana

74

6.

The Dubai Model and the Impact of the Financial Crisis on South Asian Migrant Workers in the United Arab Emirates D. Narayana and Vinoj Abraham

94

7.

Global Financial Crisis and its Consequences on Migrants in Qatar: Macro and Micro Perspectives Hrushikesh Mallick and Udaya Shankar Mishra

122

8.

Global Financial Crisis and the Migrant Labour Market: A Study of Kuwait K. Pushpangadan and M. Parameswaran

131

viii Y India Migration Report 2012

9. Low-skilled Indian Construction Workers in the Gulf, Singapore and Malaysia: Return to India, Reintegration and Re-migration Auke Boere 10. Impact of the Global Recession on Migration and Remittances: The Kerala Experience K. C. Zachariah and S. Irudaya Rajan 11. Global Financial Crisis and Return of South Asian Gulf Migrants: Patterns and Determinants of their Integration into Local Labour Markets Vinoj Abraham and S. Irudaya Rajan 12. Inclusive Growth and Economic Crises: Labour Migration and Poverty in India Arjan de Haan

151

169

197

225

13. Food Inflation and Financial Crisis in India: Impact on Women and Children S. Mahendra Dev

249

14. Migration, Human Rights and Development S. Irudaya Rajan and Charles Nellari

269

15. Remittances and Financial Participation: A Household-level Analysis in Kerala Deepa Iyer

280

16. International Labour Migration: Global Words, National Leads and Local Deeds S. Krishna Kumar

303

17. Broadening Exchanges and Changing Institutions: Multiple Sites of Economic Transnationalism S. Irudaya Rajan and V. J. Varghese

322

About the Editor Notes on Contributors Index

347 348 350

List of Tables 2.1 2.2 3.1 3.2 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 5.10 5.11 5.12 5.13 5.14 6.1

Average Cost of Emigration in Different Migration Corridors, 2008 Policies Aimed at Reducing the Supply and Demand of Foreign Workers is GCC Countries Key Demographic and Economic Indicators for Main Countries of Origin and Destination of Migrants in the West Asian and Asian Region Nominal Growth of Gross Domestic Product (GDP) in Selected Asian Economies 1996–2009 Selected Economies: Real GDP Growth Rates (Annual Per Cent Change) Growth of Population and GDP in GCC Countries Profile of Expatriates in the Gulf, 2009 Share of Employment (in Per Cent) across Economic Activities in GCC Countries, 2007 Projects Affected by the Crisis in the GCC Projects Affected by Subsectors, UAE Government Expenditure in the GCC Countries, 2006–2011 (Percentage of GDP) Number of Return Emigrants to Kerala due to Recession in 2009 Estimates of Return Emigrants to South Asia from the Gulf due to Financial Crisis, 2009 Average Cost of Emigration for Different Migration Corridors from Kerala, 2008 Channels of Migration by Emigrants, 2007 Estimates of Emigrants who Lost their Jobs in the Gulf but have not Returned, 2009 Outflow of Migrant Workers from South Asia to Gulf, 2005–2009 Inward Remittances to South Asian Countries from Migrant Workers, 2000–2009 Distribution of Population by Emirates, 1970–2005

21 23

30 38 78 80 80 82 83 83 83 85 86 86 87 88 88 90 95

x Y India Migration Report 2012

6.2

6.14 6.15 6.16 6.17

Distribution of Population by Age Groups and Nationality Distribution of Population by Level of Education Percentage Distribution of Population (15 Years and Over) by Activity Status, Sex and Nationality, 2008 Labour Force Participation Rate (in Per Cent) by Nationality, Sex and Educational Status, 2008 Industrial Classification of Workers in the UAE Average Annual Earnings Per Worker in the UAE and Simple Growth Rate Occupational Structure of the Workforce in the UAE The UAE’s Oil and Non-oil GDP, 1993–2008 (in Billion Dirham, in Current Prices) Industrial Composition of the Economy Some Macroeconomic Indicators (at Current Prices) Imports, Non-oil Exports and Re-exports of the UAE by Groups of Countries, 2008 (Percentage of Value) Distribution of Population by Percentage of Disposable Income Spent on Food Inflation in GCC Countries, 2002–2007 (in Per Cent) GDP and Current Account Balance of the UAE Losses of Sovereign Wealth Funds of the Gulf Ease of Doing Business

8.1 8.2 8.3 8.4 8.5 8.6 8.7

Growth Rate of the Economy, 1991–2009 Growth Rates by Sectors, 1991–2007 The Share of Oil Revenue in Total Public Revenue Statistical Tests for Lag Length of VAR VAR Descriptive Statistics Structure of the Labour Force in Kuwait (in Thousands) Distribution of Migrant Labour Force by Sector

9.1

Duration of Last Emigration Among the Return Emigrants Current Financial Situation Compared with Financial Situation before Emigration Last Country of Emigration and the comparison of Financial Situation Before Migration Last Country of Emigration and Debt Problems

6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 6.12 6.13

9.2 9.3 9.4

96 97 98 99 100 101 102 103 103 104 105 108 109 110 111 119 133 134 136 138 139 146 146 154 156 156 157

List of Tables Y xi

9.5 9.6 9.7 9.8

Type of Return and Financial Situation after Emigration Type of Return and Debt Problems Last Country of Emigration: Type of Return Debt Problems: Tamil Nadu

159 159 160 162

10.1 EMIs by Employment Status in 2008 and 2009 10.2 Estimate of Unemployed among the 2008 EMIs in 2009 in Kerala 10.3 EMIs by Employment Status, Kerala, 2008 and 2009 10.4 Sector of Economic Activity of the Unemployed, 2008–2009 10.5 Economic Activity of EMIs in 2008 and 2009, KMS 2008 10.6 Country of Residence of EMIs who became Unemployed in 2009 10.7 Unemployment among Males and Females of 2008 EMIs in 2009 10.8 Unemployment Percentage and Rate by Level of Schooling, 2009 10.9 Emigrants of 2008 by Migration Status in 2009 in the RMS 2009 10.10 Average REMs Per Year, 1996–2008 10.11 Reasons for Return of REMs among EMIs of 2008 in 2009 10.12 Percentage and Return Rate by Country of Origin of REMs, 2009 10.13 Sex Composition of REMs among 2008 EMIs in 2009 10.14 Education: Percentage of Total and Percentage of EMIs 10.15 REMs and EMIs: Percentage and Ratio by Employment Sector, 2009 10.16 Number of REMs in Kerala who would need help in Rehabilitation, 2009 10.17 Household Cash Remittances, 2008 and 2009 10.18 Average Remittances per Household, by Religion, 2008 and 2009

172

11.1 Level of Education of REMs in India, 2008 and 2009 11.2 Conditions of the REM Household, 2009

201 203

172 173 176 177 178 179 180 180 181 182 184 185 186 187 189 190 192

xii Y India Migration Report 2012

11.3 Matrix of Distribution of Various Types of Workers in 2009 by their Employment Status in 2008 11.4 Matrix of Change in Earnings of REMs between 2008 and 2009 (in Per Cent) 11.5 Inter-country Variations in Employment Status 11.6 Industrial Distribution of Migrant Workers: Before and After Return 11.7 Logit Model: Determinants of Employment of Return Migrants (Present Status Being Employed = 1; Present Status Being Unemployed = 0) 11A Changes in Employment Status during the Crisis: An Inter-country Comparison 11B Industrial Distribution of Migrant Workers: Before and After Return 11C Variable Definition, Construction and Hypothesis 13.1 Household Responses to the Food Price Increase (in Per Cent): Bangladesh 13.2 Hunger Index and Components in India: States 13.3 Incidence of Malnutrition Among Women (Chronic Energy Deficiency), 2005–2006 15.1 Categorization of Households Based on Migration Experience 15.2 Bank Accounts, Remittances and Migration Categories 15.3 Bank Loans, Remittances and Migration Categories 15.4 Any Bank Interactions, Remittances and Migration Categories 15.5 Remittance Marginal Effects for All Households 15.6 Remittance Effects: Conditioning for Migration Categories 16.1 The Migration Process 16.2 Illustrative Migration–Development Planning Matrix

205 207 210 211 215 219 221 223 255 256 257 286 288 288 288 295 296 308 320

List of Figures 3.1 3.2 3.3 3.4 3.5. 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9

Yearly Labour Migration Outflows from Selected Countries in Asia, 1976–2008 31 Migration Flows Disaggregated by Sex from Selected Asian Countries, 1986–2008 34 Migrant Worker Outflows, Selected Asian Countries of Origin, 2005–2010 42 Number of Migrant Outflows to the United Arab Emirates, Selected Asian Countries of Origin, 2005–2010 43 Remittances Received by Selected Asian Countries, 2005–2009 (in US$) 44 Balance of Payments (in US$ Billion) Official Reserves and Foreign Holding of US Securities of Some Emerging and Developing Economies (in US$ Billion) Population Growth and Net Migration Flows (in Thousands) Percentage Distribution of Population and Labour Force by Nationality, 2008 Distribution of Labour Force by Nationality in Private Sector and Public Sector, 2008 (in Per Cent) Rate of Unemployment, 2008 Main Expatriate Communities in GCC Countries (Values in Thousands) Monetary Base (Values in National Currencies) Interest Rate and Inflation Rate

6.1 6.2 6.3 6.4

Trade Intensity of the Economy, UAE High-end Residential Housing Prices (US$/m²) Office Space under Construction (Million ft²) Top Five Recipients of FDI Inflows in West Asia, 2007–2008 (in US$ Billions)

8.1

Relationship between Diversification Index and Instability Index Ratio of GNP to GDP in Kuwait

8.2

49 51 52 55 57 58 61 62 64 105 106 106 107 135 136

xiv Y India Migration Report 2012

8.3 8.4 8.5 8.6 8.7 8.8 8.9

Impact of GDP Error Term on Remaining Macro Variables Impact of OLX Shocks on the Remaining Four Macro Variables Impact of CPI Error Shock on the Remaining Four Macro Variables GEX Shocks on the Other Four Macro Variables Impact of MOS Error Shocks on the Remaining Four Macro Variables Impact of CPI Error Shocks on the Remaining Four Macro Variables Impact on Labour Market Pre- and Post-financial Crisis

10.1 Unemployment: Before and After Emigration, and After Recession 10.2 Unemployed: Distribution by Age (in Per Cent) 10.3 REMs per 100 EMIs by District of Origin of Return Migrants 10.4 Age Distribution of REMs and Return Emigration Rate by Age 10.5 REMs per 100 EMIs by Education 10.6 Percentage of Households that Received Larger, Equal or Smaller Remittances in 2009 than in 2008 11.1 Age Distribution of REMs 11.2 Comparison of the Age Distribution between REMs and EMIs from India 11.3 Level of Education of REMs 11.4 Changes in Employment Status and Reintegration into the Local Labour Market 11.5 Share of Unemployed in Sample: Inter-temporal and Inter-country Differences

142 142 142 143 143 143 144 174 179 185 186 187 192 199 200 201 208 209

List of Abbreviations ADIA ADIC APL BNP2TKI BOP BPL BRIC BWI CDS CEDAW CIVETS CNRI CPI CV DI DINRICD DIPP EIU EMI EPS EU FAO FCNR FCRA FDI FEMA FIRE GATS GATT GCC GCIM GDP GEX

Abu Dhabi Investment Authority Abu Dhabi Investment Council Above Poverty Line Badan Nasional Penempatan dan Perlindungan Tenaga Kerja Indonesia Balance of Payment Below Poverty Line Brazil, Russia, India and China Building and Wood Workers’ International Centre for Development Studies Committee on the Elimination of Discrimination against Women Colombia, Indonesia, Vietnam, Egypt, Turkey, and South Africa Confederation of NGOs of Rural India Consumer Price Index Coefficient of Variation Diversification Index District NRI Committee for Development Department of Industrial Policy and Promotion Economic Intelligence Unit Emigrant Employment Permit System European Union Food and Agriculture Organization Foreign Currency Non-Resident (Account) Foreign Contribution (Regulation) Act Foreign Direct Investment Foreign Exchange Management Act Financial, Insurance, Real Estate General Agreement on Trade in Services General Agreement on Tariffs and Trade Gulf Cooperation Council Global Commission on International Migration Gross Domestic Product Government Expenditure

xvi Y India Migration Report 2012

GFCF GFMD GNP GoI GoP HLCID HLD HSM HWWI IC ICDS ICVIT IDF IFPRI ILO IMF IOM ISI IT KFC KIA KIP KKNTC KMS LFPR LMP LMRA LNG LPG M2 MD MDG MENA MFLM MICE mmtpa MOIA MOS mtpa

Gross Fixed Capital Formation Global Forum on Migration and Development Gross National Product Government of India Government of Punjab High Level Committee on Indian Diaspora High-level Dialogue Highly-skilled Migration Hamburgisches WeltWirtschafts Institut International Cooperation Integrated Child Development Services Indo-Canadian Village Improvement Trust India Development Foundation International Food Policy Research Institute International Labour Organization International Monetary Fund International Organization for Migration Instability Index Information Technology Kerala Financial Corporation Kuwait Investment Authority Know India Programme Kerala Kettida Nirmana Thozhilali Congress Kerala Migration Survey Labour Force Participation Rate Labour Mobility Partnership Labour Market Regulatory Authority Liquefied Natural Gas Liberalization, Privatization and Globalization Money Supply Managing Director Millennium Development Goals Middle East and North Africa Multilateral Framework on Labour Migration Meetings, Incentives, Conventions and Exhibitions (Industries) million metric tonnes per annum Ministry of Overseas Indian Affairs Money Supply million tons per annum

List of Abbreviations Y xvii

NBER NCAER NCEUS NCRL NDP NFHS-3 NGO NIROMP NORKA NRE NREGA NREGP NRHM NRI NRK NRO NSDP NSS NSSO OBC OCI ODA OECD OLX OMI OPEC OR PBSA PDS PIO PMLU PPP PROGRESA PUNRICD RBI

National Bureau of Economic Research National Council of Applied Economic Research National Commission for Enterprises in the Unorganized Sector National Council for Rural Labour National Development Plan 2005–2006 National Family Health Survey Non-governmental Organization New International Regime for Orderly Movement of People Non-resident Keralites’ Affairs Department Non-resident External (Rupee Account) National Rural Employment Guarantee Act National Rural Employment Guarantee Programme National Rural Heath Mission Non-Resident Indian Non-Resident Keralites Non-Resident Ordinary (account) Net State Domestic Product National Sample Survey National Sample Survey Office Other Backwards Class Overseas Citizenship of India Official Development Assistance Organization of Economic Cooperation and Development Oil Exports Out-migrants Organization of the Petroleum Exporting Countries Oil Revenue Pravasi Bharatiya Samman Award Public Distribution System Person of Indian Origin Palamoori Migrant Labour Union Purchasing Power Parity Programa Nacional de Educacion, Salud y Alimentacion Punjab NRI Committee for Development Reserve Bank of India

xviii Y India Migration Report 2012

REM ROM Q1 Q2 Q4 QCB QIA QNB QP QR RBI RMS SAMA SANEI SC SD SDP SDR SEBI SIEPR SPDC SSA ST SWF TCWF TPR TVEs UAE UK UN UNCTAD UNDP UNHCR UNICEF US US$

Return Emigrant Return Out-migrants First Quarter Second Quarter Fourth Quarter Qatar Central Bank Qatar Investment Authority Qatar National Bank Qatar Petroleum Qatari Riyal Reserve Bank of India Return Migrants Study Saudi Arabian Monetary Agency South Asia Network of Economic Research Institutes Scheduled Caste Standard Deviation State Domestic Product Special Drawing Right Securities and Exchange Board of India Stanford Institute for Economic Policy Research Scholarship Programme for Diaspora Children Sarva Shiksha Abhiyan Scheduled Tribe Sovereign Wealth Fund Tamil Nadu Construction and Unorganized Workers Federation Total Public Revenue Township and Village Enterprises United Arab Emirates United Kingdom United Nations United Nations Conference on Trade and Development United Nations Development Programme United Nations High Commissioner for Refugees United Nations International Children’s Fund United States United States Dollar

List of Abbreviations Y xix

VAR VDC VLIF WEO WPR WTO y-o-y

Vector Auto Regressions Village Development Committee Village Life Improvement Foundation World Economic Outlook Workforce Participation Rate World Trade Organization year on year

This page intentionally left blank

Foreword Migration is a complex global issue, which touches almost every country of the globe today. Virtually all countries are either points of origin, transit or destination, particularly for economic migrants. In the 21st century, international migration is both a cause and consequence of globalization and its impact is both on countries of origin as of destination. These movements are a response to the demands of labour markets which have always been served by different migration modes, from indenture in the 19th century to student mobility in the present times. Migration, and especially labour mobility, is like water — it will flow wherever there is a vacuum. Since over 200 million people currently live outside their country of origin and many more migrate every year, new forms of migration are emerging and new models of migration management are being proposed the world over. Migration impacts across borders and sectors. Its ripples are felt socially, economically and even politically. It is in this context that assessing the impact of the recent global crisis on migration and remittance, especially for India, is both relevant and topical. Edited by Dr S. Irudaya Rajan, Chair Professor, Research Unit on International Migration at the Centre for Development Studies, Kerala, this is the third in the series of India Migration Reports. The first was on ‘Governance and Labour Migration’ (2010); the second on ‘Migration, Identity and Conflict’ (2011); this, India Migration Report (2012), is on ‘Global Financial Crisis, Migration and Remittance’. From ‘governance’ to ‘conflict’ to ‘crisis’ — migration encompasses all of these and more. This volume, organized into 17 chapters, weaves through the intricacies of the 2008–2009 global financial crisis and the impact it had on both migration flows as well as remittances to India. Interestingly, there are detailed studies of different countries as well as of individual states such as Kerala where the findings are not only surprising but reflect the resilience and networking of the simple lowskilled Indian worker. Transnational migration and it’s connect with development are also featured to remind us of the deep economic bearing that such movements of workers has. India is unique in being only the 11th country in the world to have an exclusive Diaspora Ministry — the Ministry of Overseas

xxii Y India Migration Report 2012

Indian Affairs. Not only does this Ministry focus on engagement with the Indians based overseas; it also handles emigration management. Towards ensuring transparent and conducive policies for migration, the Government of India through this Ministry has taken several steps to institutionalize its governance on evidence-based policy research. The Research Unit on International Migration at the Centre for Development Studies is a radiant example of such academic rigour. Surely the findings of this report will be used in fine-tuning migration policies in this important perspective. Dr A. Didar Singh Secretary Ministry of Overseas Indian Affairs Government of India, New Delhi

Preface First of all, I would like to thank the readers of the India Migration Report (IMR) for their overwhelming support to this annual series brought out by the Ministry of Overseas Indian Affairs (MOIA) Research Unit on International Migration (RUIM) at the Centre for Development Studies (CDS), Thiruvananthapuram, Kerala1 and published in collaboration with Routledge, New Delhi. The reprint editions of IMR 2010 and IMR 2011 are already underway. While the IMR 2010 focuses on governance and policy issues in the context of international labour migration, IMR 2011 examines the questions of identity and conflict in the emergent phenomenon of internal migration and the role of protecting the rights of migrants within India. This volume, IMR 2012, third in the series, overviews the impact of the global crisis on migration, return migration, re-integration and remittances, both in the countries of origin and of destination. The first set of four chapters elucidates the global scenario with focus on South Asia, South-East Asia and the Gulf, where several migration corridors exist between Asia and the Gulf. Chapter 1 re-emphasizes the importance of the governance of human mobility at the time of the global economic crisis, followed by Chapter 2, which re-examines migration and development linkages in the same context. Chapter 3 broadly assesses the impact of the crisis on the South and South-East Asia region, where emigration and remittances are major economic issues, and the lessons learnt by the economies of sending countries. Chapter 4 evaluates the effect of the global economic imbalance on workers in the countries of destination, in particular, the Gulf Cooperation Council (GCC). The second set of five chapters is based on both macro- and micro-level analyses on the impact of the global crisis on migration and remittances in several countries of origin — India, Pakistan, Bangladesh, Sri Lanka and Nepal — as well as countries of destination — the United Arab Emirates, Qatar, Kuwait and Malaysia. Chapter 5 assesses the impact of the global crisis in the Gulf on migration,

1

http://www.cds.edu (accessed 20 December 2011).

xxiv Y India Migration Report 2012

return emigration and remittances in the countries of South Asia and concludes that both migration and remittances are resilient. The next four chapters examine the impact of the crisis from the perspective of destination countries in the Gulf and South Asia — the United Arab Emirates, Kuwait, Qatar, Malaysia and Singapore. The third set of four chapters discusses the impact of the economic crisis on return emigration and their coping mechanisms, reintegration, as also poverty and special population groups such as women and children. Chapter 10 reports the findings of the fifth large-scale survey conducted in Kerala in 2009 immediately after the global crisis, which covered both emigrants and return emigrants identified in the third and fourth rounds of Kerala Migration Surveys 2007 and 2008. The subsequent chapter gives the results of the survey of emigrants carried out in five South Asian countries to assess the coping mechanism of migrants at the time of the global crisis. While Chapter 12 discusses inclusive growth and poverty in the context of the global crisis, Chapter 13 assesses the impact of the global crisis on women and children. As with the earlier IMR, the last four chapters talk about the ongoing work of the CDS on the various facets of international migration. The last chapter of the IMR 2012 is an outcome of a large project carried out by the CDS for the European Commission and deserves special mention. Before I conclude, I would like to inform the readers that IMR 2013 will be on the broad theme of social costs of migration — a neglected area of research in the field of migration. As of now, it proposes to focus on Indian diasporas and developments in IMR 2014. We welcome comments and suggestions from readers for future annual themes and also article contributions for the forthcoming volume. The RUIM completes five years in 2011. It was evaluated by the four independent experts and their report is quite impressive. It states: ‘CDS has been engaged in research on migration for over three decades now. Through RUIM, under the sponsorship of MOIA, it has acquired considerable knowledge and expertise and established networks and collaboration. The establishment of RUIM has given regularity and consistency in research on migration at CDS. It has also provided scope for faculty, irrespective of their main research interest, to get engaged in the research undertaken at RUIM. A time

Preface Y xxv

has come for CDS to mainstream and institutionalize RUIM at CDS, as RUIM has the potential to become a national centre for migration research.’ Based on the evaluation, I am pleased to inform readers that the MOIA has decided to extend the unit for the next five years — from 2011 to 2016. Finally, my dream project of launching Migration and Development, an international interdisciplinary journal published by Routledge, UK, is taking final shape and I hope to see the journal in print soon. I look forward to your continued support, guidance and suggestions. S. Irudaya Rajan

This page intentionally left blank

Acknowledgements In working on the third annual series of the India Migration Report 2012, I received a lot of support, comments and suggestions from various quarters. However, I would like to acknowledge the constant support received from the Ministry of Overseas Indian Affairs (MOIA), Government of India, in particular, Dr Didar Singh, Secretary; Mr Atul Kumar Tiwari, Joint Secretary; Mr. Roulkhumlien Buhril, Protector General of Emigrants; and Ms Kale, Economic Adviser. I would like to thank Mr Gurucharan, the Chief executive officer of the Indian Council of Overseas Employment, Government of India. In addition to the regular grants received from the MOIA, I received financial support to conduct research on the global crisis, migration, remittances and return migration from the Department of Non-Resident Keralite Affairs, Government of Kerala and the Asian Development Bank. At the Centre for Development Studies, I always found encouragement, cooperation and guidance from all the faculty members, administrative staff and the library staff — especially, Professor Pulapare Balakrishan, Director; Professor K. N. Nair, former Director; Mr Bimal Jalan, Chairman; Mr Soman Nair, Registrar; Mr V. Sriram, Librarian; Mr V. K. Anil Kumar, Senior Assistant Librarian; Dr T. K. Subramani, former Librarian; and Ms A. Chandra and Mr S. Suresh, Finance Section. I thank them all. I would also like to express my gratitude to all the individuals who have generously contributed to this volume despite their busy work schedules and other commitments. At home, I thank my wife, Hema, and three children — Rahul, Rohit and Catherine — for their emotional support, patience and understanding. My appreciation also goes to the editorial and sales teams of Routledge, New Delhi for all their hard work and dedication that helped bring out this volume on time.

This page intentionally left blank

1 The Global Economic Crisis and Governance of Human Mobility Bimal Ghosh The world migration system has been under increasing strain for

more than four decades. The strain stems from a gathering mismatch between rising emigration pressures and dwindling opportunities for legal entry (especially of low-skilled workers). Instead of trying to bring these two powerful contradictory trends into dynamic harmony through pro-active inter-state cooperation, nations have continued to follow mostly reactive and inward-looking policies and taken short-term or ad hoc measures to manage migration. This has produced a string of perverse results: human and economic costs have risen sharply while the opportunities for enhancing world stability and welfare gains have been largely forgone. The present economic crisis has worsened the situation. If joblessness and poverty continue to rise or remain at a high level in the coming years, derailing the Millennium Development Goals (MDG) timetable, and if, instead of strengthening their cooperation, nations become more inward-looking and reactive in their migration policies, it would be difficult to avoid domestic and inter-state tension; world recovery, too, will be slowed down. Like its predecessors in the past, the present economic crisis will not last forever. Some analysts are foreseeing a ‘business as usual’ scenario for migration in the post-recession future. According to this view, once the recession ends and recovery takes hold, migration will continue to flow as in the past because the structural reasons that drive inter-country movements will remain unchanged. It is important to remember, however, that some of the policy or practical measures taken during a crisis could have long-term effects on the future configuration of migration. It took several decades for international migration to recover from the impact of the restrictive

2 Y Bimal Ghosh

immigration measures taken by countries such as the US (United States)1 and France just before and during the Great Depression of the early 1930s and their adverse effect on the social climate. Equally important is the fact that if migration is not effectively managed during a crisis, and instead becomes a source of serious tension and conflict, host countries — advanced or developing — could be unduly reticent to admit immigrants for many years to come. The corrosive scars can last long. Furthermore, as indicated later in this chapter (third section), it is far from certain that over the medium to long-term all of the structural drivers of migration will remain exactly the same as they are today. Many of the policy and practical measures I have outlined elsewhere (see Ghosh 2011)2 to respond to the recession-driven changes in the configuration of migration and global labour markets would call for close and genuine cooperation between migrant-sending and migrant-receiving countries, without which they will not make much headway. The pre-existence of an agreed international framework for cooperative management of migration would have made it possible to avoid much of the strain being placed on the migration system by the crisis and the resulting widespread concern. It would also have made it easier to launch the necessary remedial measures through closer international cooperation. Nations could have moved forward with greater confidence, avoiding panicky reactions and knee-jerk policy measures as we are witnessing today. The recession has no doubt signalled some fledgling but ominous signs of anti-immigrant and restrictionist trends in both rich and poor countries and regions. This might imply a shift further away from international cooperation. On the other hand, if the gravity and urgency of the migration situation lead nations to work more closely and initiate some joint action to meet the immediate challenge, this could very well open up a new opportunity for closer inter-state cooperation on a lasting basis and serve as a stepping stone towards the establishment of a common international framework for improving governance of human mobility. The US Immigration Act of 1924, which took effect in 1928, imposed serious quota restrictions on the annual flow of new immigrants from most countries. 2 The list of policy and operational measures suggested provide a benchmark against which a country’s current capacity to deal with different aspects of migration policy and management over the short to medium term could be assessed. The present chapter deals only with the governance aspect. 1

Global Economic Crisis and Governance of Human Mobility Y 3

Clearly, this would only happen if, unlike during the Great Depression, nations resist the temptation of becoming isolationist and inward looking. The agreement reached at the recent Group of 20 (G-20)3 summits in London and Pittsburgh (and the follow-up work currently being done by the group’s finance ministers and central bankers) to develop a harmonized global approach to the financial crisis and the acceptance of the need for international policy coordination to ensure orderliness, transparency and confidence in the financial sector, augur well for a similar approach towards improving global governance of human mobility. In times of crises, just as nations can become panicky and short-sighted, they can also become more willing to accept change. Common challenges, as a recent United Nations Development Programme (UNDP) report put it, evoke common responses (UNDP 2009: 3). Sadly, however, unlike the ongoing global efforts to better manage the financial system, no major new initiative has, as yet, emerged to develop a framework for inter-state cooperation to improve governance of human mobility.

Looking Back through History Attempts to build inter-state cooperation on migration within a global framework are not new. In 1927, the League of Nations explored at some length the possible adoption of an international convention to ‘facilitate and regulate’ international exchange of labour. Since then, a long series of initiatives have been taken to develop such a framework. (For details and a chronicle, see Joel P. Trachtman [2009: xv–xvii]). The latest academic initiatives in this area include: a research project on the international law of economic migration (supported by the W. E. Upjohn Institute, and executed by Trachtman at the Fletcher School of Law and Diplomacy), and on global mobility regimes (led by Rey Koslowski at the University of Albany, and supported by the MacArthur Foundation). Also of interest are the observations made by the UNDP in October 2009, just before the Third Session of the Global Forum on Migration and Development (GFMD) in Athens: There are calls to create a new global regime to improve the management of migration: over 150 countries now participate in the Global 3

See http://www.g20.org/about_what_is_g20.aspx (accessed 19 October 2011).

4 Y Bimal Ghosh Forum on Migration and Development. Governments faced with common challenges, develop common responses — a trend we saw emerge while preparing this report (UNDP 2009: 5).

However, the reality is that, despite numerous eloquent pronouncements and a plethora of consultations, governments and intergovernmental organizations have so far remained sluggish in giving a concrete shape to the proposed international framework.

The Constraints: How Real are They? Given the strange paradox, the question could well be asked: how realistic is it to expect nations to agree to adopt and adhere to a common international framework for managing migration? Three main reasons are generally cited as major constraints to the development of such a framework: (a) the lack of a shared goal or a ‘common good’ between origin and destination countries; (b) the absence of reciprocity of interests between the two groups of countries and asymmetry in their bargaining strengths; and (c) the absence of a hegemonic power to sponsor and safeguard such a regime (Ghosh 2007: 97–118, 2009). How valid are these arguments? This is discussed in the following sections.

Lack of a Common Good All civilized states have a collective, as well as an individual, stake in maintaining international stability and peace and promoting economic progress. Orderliness and predictability of human mobility are among the essential conditions for achieving these goals, and they constitute a ‘common good’ that can underpin a global agreement on migration, fully in keeping with the tenets of regime theories. However, individual states also have differing, even conflicting, interests on many individual issues, depending on the countryspecific situation and the type of the migratory flows. But it is precisely this diversity of state interests that offers the potential for bargaining based on trade-off or reciprocity, as practiced in trade negotiations, thereby creating the potential for each state to be a net gainer at the end of the deal.

Absence of Reciprocity There are a number of possible areas of reciprocity that can be explored. These include: trade-offs between opportunities for legal

Global Economic Crisis and Governance of Human Mobility Y 5

entry in destination countries and readmission of irregular migrants by countries of origin; availability of negotiated categories of skilled personnel from origin countries in exchange for destination countries’ acceptance of some of the origin countries’ low-skilled workers; and unmet demand for labour in host countries’ seasonal industries and occupations generally shunned by locals and the supply of such labour from origin countries. Some scope also exists for negotiating reciprocity between the movement of rich countries’ high-level personnel as intra-company transfers under Mode 3 (commercial presence) of the General Agreement on Trade in Services (GATS)4 and the temporary entry into industrial countries of service-providing professionals from developing countries both as employees and as self-employed persons. Additionally, significant trade-offs can be worked out between predictable labour supply from origin countries and alleviation (at least in the short to medium term) of the growing strain on social security funding and slowing down of demographic decline in receiving (industrial) countries. Reciprocity of interests across sectors — for example, access to destination countries’ labour markets in exchange for liberalization of origin countries’ specific product markets, as recently discussed as part of the Doha Round of trade talks5 — also holds some significant promise. The interlocking reciprocity of interests in these and other potential areas plays its part in ensuring a high degree of symmetry in the bargaining powers of both sides. Since more and more countries (at least 25 per cent according to a recent global survey by the International Labour Organization [ILO]) are becoming involved in simultaneously sending and receiving migrants on a relatively large scale (ILO 1999), the protection and treatment of immigrants (and avoidance of ‘tit-for-tat’ practices) also becomes an important area of reciprocity; this too adds to the symmetry in the bargaining powers of the two groups. Although often ignored or forgotten, it is important to note that this makes 4 GATS Mode 3 relates to service delivered within the territory of a country through the commercial presence of the supplier in another country. 5 Migrant-sending countries may agree to liberalize some of their product markets (for agricultural commodities or manufactures) if, in exchange, the destination countries accept some additional labour migrants from origin countries. For example, during the Doha Round, India offered freer access for rich countries’ specific goods to its markets if such countries accepted some additional labour migrants.

6 Y Bimal Ghosh

the protection of migrants’ human rights and good governance of human mobility so closely interwoven (Ghosh 1995, 2003). Although the principle of reciprocity of interests has underpinned some of the existing bilateral agreements between migrant-sending and -receiving countries — as in the case, for example, of readmission agreements — its full potential for building inter-state cooperation still remains to be explored. Outside the European Union (EU), formal reciprocal commitments have generally been confined to issues such as readmission, remittances and planned recruitment, and have taken place in a bilateral or sub-regional/regional setting. Unlike in the case of trade, neither the governments nor the inter-governmental organizations have explored the range of reciprocity that can be negotiated within migration and across sectors, providing an enduring basis for inter-state cooperation. In a recent study on economic migration, Joel P. Trachtman (2009) proposed a structure of a multilateral legal framework within which states could negotiate commitments based on reciprocity of interests, thereby unlocking significant mutual and global welfare gains. In doing so, Trachtman follows a ‘positive list’ approach (under which each state specifies sectors or migration areas in which it would allow entry taking into account its own conditions and interests) to commitments for openness. This thus ensures flexibility, allows for the gradual application of the approach to openness, and makes it politically more realistic (see ibid.: xv–xvii).6 Both migrant-sending and -receiving countries could benefit by giving systematic attention to this avenue of action and by developing their analytical research to identify various areas of reciprocity and negotiating skills to overcome bargaining problems and facilitate efficient co-operation, be it at the bilateral, plurilateral or multilateral level.7

Hegemonic Power vs Collective Inter-state Leadership The role of a hegemonic power, as posited in regime theories, has significantly evolved over time, signalling a decline in the influence This is very much in line with the principle of ‘regulated openness’ as envisioned by the New International Regime for Orderly Movement of People (NIROMP) project, supported by the United Nations (UN) and several European governments. 7 For a discussion of the reasons why a multilateral framework enjoys certain advantages over bilateral and plurilateral settings see Joel P. Trachtman [2009: 271–96], and Bimal Ghosh [2000: 235–39]. 6

Global Economic Crisis and Governance of Human Mobility Y 7

of one or more hegemonic powers in world politics and the potential emergence of a multi-polar world society. Not surprisingly, in recent years, several international agreements or regimes were initiated and/or subsequently sustained by collective initiatives in the UN (despite the absence of any hegemonic support or even resistance from one or more hegemonic powers). In these cases, collective consensus-building rather than hegemonic leadership has been the driving force.8

The Outlook for the Future: Migration in 2025 If mutual confidence and commonality of interests are essential cornerstones for building durable international cooperation in managing migration, the proposed framework, based on a set of transparent and predictable norms and principles, is clearly the answer. Such cooperation should, ideally, take the form of an overall framework agreement as a soft instrument (for instance, a solemn declaration) supported by a set of autonomous, but inter-related sub-regimes in the form of soft and hard instruments, depending on the nature of the issues or areas concerned. Admittedly, as long as there are sharp economic and demographic imbalances between rich and poor countries, there is likely to be some tension in managing migration, despite the existence of an agreed framework for cooperative action. It is conceivable, however, that by 2025 the global migration system would become more stable and the framework of cooperation would work under much reduced tension with the framework agreement itself contributing to a less fraught environment. The expected stability would come from three main sources. The first is concerned with declining pressure for emigration — especially though irregular channels — as a result of an improved Under the conventional regime theory, an international agreement (or regime) is not viable unless it is sponsored and sustained by a hegemonic power. Recent developments in international relations have shown that this need not always be the case, and the regime theory itself is evolving in recognition of this trend. A typical example of such developments is the establishment of the International Criminal Court, which was initially opposed by some of the major powers but has become a reality. The same largely applies to the adoption of the 1990 UN International Convention on the Protection of the Rights of All Migrant Workers and Members of Their Families. The emergence of the G-20 is also indicative of the same trend towards collective inter-state initiative in the economic arena. 8

8 Y Bimal Ghosh

political and economic situation, including in particular high rates of economic growth in a number of migrant-sending and other developing countries such as China, Brazil, India, Indonesia, and South Korea. Several other countries, including Mexico, Turkey, Colombia, South Africa, and Vietnam, too are making rapid strides. The growth prospects of these counties have led banks and financial institutions to introduce new acronyms, wider than BRIC (Brazil, Russia, India and China) that was coined nine years ago, in order to attract investor interest.9 In recent years, many of these countries have witnessed a much faster rate of economic growth than the rich countries and they probably will continue to do so in the coming years. Even if a few suffer slowdowns, the general trend is unlikely to be reversed. Between 2005 and 2010, GDP increased by 25 per cent in Brazil, 47 per cent in India, and 69 per cent in China. This compares with an increase of just 5 per cent in the US, 4 per cent in the Eurozone and 2 per cent in the United Kingdom (UK). Even without China and India, emerging and developing country economies grew by 5.2 per cent in 2010. The shift is also reflected in the relative shares of the two groups in global GDP. In 2000, rich countries accounted for 63 per cent of global GDP at purchasing power parity (PPP); it fell to 53 per cent in 2010. There has been a concomitant rise in the emerging countries’ share in global output, with China and India accounting for 80 per cent of the increase. It is true that, as concerns income per head, these countries remain way behind the rich countries, and it will take them a long time to bridge the gap. However, as I have argued elsewhere (Ghosh 2005a: 181, 2005b), more than the absolute level of income, it is the rate of growth and the perception of the future performance of the economy that influences people’s decision to migrate. As the economic outlook in these and several other middle- and low-income countries continue to improve, many who would have otherwise moved abroad would prefer to stay at home. 9 The US-based bank, HSBC, for example, has suggested CIVETS (Colombia, Indonesia, Vietnam, Egypt, Turkey, and South Africa); while the Standard Chartered Bank has proposed ‘seven percent club’ for countries with regular Gross Domestic Product (GDP) growth of 7 per cent a year or more.

Global Economic Crisis and Governance of Human Mobility Y 9

This does not mean that inter-country income or wage differential will cease to be of relevance to movements across borders. However, it can be expected that even the wage differential between the two groups of countries — especially at the upper ends of the tradeable sectors — will start coming down, as it had happened between rich and poor countries of Europe in the 19th and early 20th centuries (Williamson 2002), and is now happening in countries like India in certain sectors such as information technology (IT) and pharmaceuticals. Between January and July 2010, each of India’s top three groups in the IT sector had to increase pay by 10–20 per cent, and the cost relating to poaching by competitors hit a record US$2 billion according to analysts. In several of these countries, labour productivity is rising rapidly. Labour productivity in India and China, for example, is estimated to be growing at more than five times the global rate. This will help in pushing wages up. Much of the present tension in South–North flows stems from increasing pressures of low-skilled migrants, including those who enter, or try to enter, through irregular channels. But as some of the low- and middle-income countries make rapid strides towards higher levels of growth and income, a change in the characteristics of migrants and composition of flows is likely to take place; this too, as explained further in the chapter, will contribute to lessening the tension. Elsewhere I have made a distinction between migrants who are desperate to move because of joblessness, poverty and hunger (‘survival migrants’) and those who move in search of better opportunities abroad (‘opportunity-seeking migrants’) (Ghosh 2007). The former move under compulsion, have little to lose and hardly care about the risks involved. By contrast, for the latter, migration is a matter of rational choice. They cross borders after a relatively careful assessment of the possible costs and benefits of the move and would generally avoid the risks of illegal entry. They are also normally better equipped to migrate, more informed about conditions in the destination country, and are likely to be more easily accepted by the host society. If higher growth in low- and middle-income countries is accompanied by better distribution of income and job creation (an important condition), there would be a decline in the pressure of low-skilled migration, especially through irregular channels, which is now a main source of tension.

10 Y Bimal Ghosh

True, it is possible that at the same time increased earnings may make it possible for some new people in these countries to move abroad, but most of them are likely to be opportunity-seeking migrants. Thus, the shift in the composition of migration would help in reducing tensions in South–North migration. The second area of anticipated stability relates to rapid and rather unexpected changes in fertility rates in both poor and rich countries, which suggest that demographic imbalances may soon be a less powerful driving factor in inter-country migration. Apart from China, which has had a one-child policy since the early 1970s, in many parts of the developing world (such as Brazil, Indonesia, Iran and even in some areas of India) fertility rates are falling astonishingly fast (Bongaarts 2009; The Economist 2009). According to the latest available figures, the rate of India’s population growth (medium variant) is now hovering around 1.4 per cent a year; by 2010–25, it will fall to 1.10 per cent. These countries are racing through the demographic transition to reach the replacement rate of fertility that is consistent with stable population. If poor people would have greater access to population planning facilities, the fertility rates in many more countries would also drop. A number of countries in Asia and Latin America are having a bulge of working age adult population that has the potential to help them over the coming decades in both capital accumulation per head and rapid economic growth. In 2020, the median age in India will be 28, as against 38 in America, 45 in Western Europe and 49 in Japan. At the same time, significantly enough, in an increasing number of rich countries (such as France, Italy and Sweden) fertility rate is rising again to reach the replacement rate. The combination of these two trends has important implications for future migration: just as the decline in birth rates, coupled with economic growth in less affluent countries, will reduce the demographic push for emigration, the rise in fertility rates in rich countries will lower the demographic and labour market-related demand that encourages immigration. Stability in population growth should contribute to stability in migration flows. This, in turn, will contribute to the diffusion of at least some of the existing tensions, especially those affecting South–North migration. Helping this process, there will be, as discussed in the next section, a concurrent rise in South-South migration, with an increase in the number and diversity of both source and destination countries across the developing regions.

Global Economic Crisis and Governance of Human Mobility Y 11

Finally, stability in South–North migration is also likely to be helped by technological progress and lifestyle changes in rich countries, especially for the younger generation. The increased use of labour-saving and automatic appliances for household work such as cleaning and gardening as well as for low-skilled jobs in farming and industrial processes should reduce the demand for low-skilled workers. These jobs are generally shunned by local workers and the unmet labour demand is met by bringing low-skilled, low-wage immigrant workers from poor countries, often encouraging inflows of irregular immigrants who are cheap and also more docile because of their precarious legal status. The tensions that the situation is now causing are likely to be diminished in the future.

Increased South–South Migration Increased South–South migration will run alongside expansion of trade and economic links among countries in the South. According to recent figures, South–South remittances are now already an important part of global flows of migrant remittances; and they are set to show further increase in the years ahead. In the past, developing countries had focused on efforts to develop collective action in the context of North–South negotiations. Economic and technical cooperation among them, though often a subject of public announcements, had, in practice, remained limited. Things are now changing. A number of developing countries are establishing new economic links through trade, aid and investment, including outsourcing, with their counterparts elsewhere. Some new movements of people are already taking place as part of the process, and prospects are opening up for more to follow. India, like China, has long been noted for its diaspora networks across the globe, including developing regions. What is striking however is that the economic and related migratory links among developing countries are now extending in several new ways and forms. For instance, Indian technology and other companies, including Infosys and Tata, have now 17,000 Indian and non-Indian employees in South America and the Caribbean — a trend which was hardly noticeable before 2000. Increased South–South migration will emerge as part of the ongoing, wider process of economic rebalancing between North and South. The adoption of a framework agreement on migration at the global level, as discussed previously, can facilitate these fast-moving changes and help ensure that the movements will be more orderly

12 Y Bimal Ghosh

and predictable, with much less negative externalities than what we are witnessing now. The forthcoming sessions of the GFMD and the UN High-level Dialogue on Migration should be geared to building the consensus needed to attain this objective. No less important is that the G-20, which so far has kept away from the issue, should take up this challenge and make it part of its agenda. India is among the key players in the G-20 and other international fora: it is also an important migrantsending and -receiving country. It is in a position to play, together with other like-minded countries, an important catalytic role in making this happen. Separately, but linked to this wider process, India can pursue the proposal for a multilateral agreement (or subregime) on labour migration based on inter-country reciprocity of interests in labour markets and across different economic sectors. In doing so, it can revamp and redesign the interesting proposal it had already made during the recent Doha round of trade negotiations — intakes of categories of migrant workers by capital-rich countries in exchange for liberalization of developing countries’ specific product markets.

We are Standing at a Crossroad Is it unrealistic to think of better managing human mobility through an agreed framework of inter-state cooperation? Is the scenario envisioned above too holistic? I don’t think so. The initiative is certainly within our reach. But much depends on whether nations (not just governments, but also the other principal actors — notably civil society and the private sector) are prepared to seriously take up the challenge of international migration, just as they have now started doing to better manage global financial flows in the wake of the recession. Their response to the pressing migration issues, which are linked to the economic crisis, may very well set the course — either by opening up new vistas of change for better governance of human mobility or reaffirming the old ways of trying to muddle through the migration morass.

Acknowledgements An earlier and shorter version of this essay was used as a background paper for the World Migration Report 2010: The Future of Migration: Building Capacities for Change. Geneva: International Organization for Migration, 2010.

Global Economic Crisis and Governance of Human Mobility Y 13

References Bongaarts, John. 2009. ‘Human Population Growth and the Demographic Transition’, in Roger V. Short and Malcolm Potts (eds), The Impact of Population Growth in Tomorrow’s World. London: The Philosophical Transactions of the Royal Society. The Economist. 2009. ‘Fertility and Living Standards: Go Forth and Multiply a Lot Less’, 31 October–6 November. Ghosh, Bimal. 1995. ‘Movements of People: The Search for A New International Regime’, in Commission on Global Governance (ed.), Issues in Global Governance: Papers Written for The Commission on Global Governance, pp. 405–24. London: Kluwer Law International. ——— (ed.). 2000. Managing Migration: Time for a New International Regime?. Oxford: Oxford University Press. ———. 2003. Elusive Protection and Uncertain Lands: Migrants’ Access to Human Rights. Geneva: International Organization for Migration. ———. 2005a. ‘Economic Effects of International Migration: A Synoptic Overview’, in World Migration 2005: Costs and Benefits of International Migration. Geneva: International Organization for Migration. ———. 2005b. ‘Managing Migration: Interstate Cooperation at the Global Level — Is the Emergence of a New Paradigm of Partnership Around the Corner?’, in Interstate Cooperation and Migration. Geneva: Berne Initiative Studies/Swiss Federal Office for Migration/International Organization for Migration. ———. 2007. ‘Managing Migration: Towards the Missing Regime’, in Antoine Pécoud and Paul de Guchteneire (eds), Migration Without Borders: Essays on the Free Movement of People, pp. 97–118. Paris and New York: United Nations Educational, Scientific and Cultural Organization/Berghahn Books. ———. 2011. The Global Economic Crisis and Migration: Where Do We Go From Here?. Geneva: The International Organization for Migration/The Hague Process on Refugees and Migration. ILO. 1999. Migrant Workers. Geneva: International Labour Organization. Trachtman, Joel P. 2009. ‘Negotiating Global Disciplines on Migration’, in Joel P. Trachtman, The International Law of Economic Migration: Toward the Fourth Freedom. Michigan: W. E. Upjohn Institute for Employment Research. http://www.upjohninst.org/publications/titles/ilem.html (accessed 4 October 2010). United Nations Development Programme (UNDP). 2009. ‘Overview’, in UNDP, Human Development Report 2009: Overcoming Barriers: Human Mobility and Development, pp. 1–5. New York: United Nations Development Programme. Williamson, Jeffrey G. 2002. Winners and Losers over Two Centuries of Globalization. Cambridge, MA: National Bureau of Economic Research.

2 Migration and Development Linkages Re-examined in the Context of the Global Economic Crisis S. Irudaya Rajan B. A. Prakash The Global Economic Crisis and Fall in the Demand for Migrant Asian Workers from the GCC Countries The global economic crisis, which is being experienced throughout

the world at present, is considered one of the worst global crises since the Great Depression of the 1930s. The crisis started in the US in 2007 with the collapse of leading investment banks, insurance companies and non-banking financial institutions, and the sub-prime mortgage arising due to excessive credit supply, collapse of the real estate, share and commodity markets, negative saving rates in the US economy, huge external borrowings and increase in the public debt of the US. As the US occupies the position of the financial capital of the world, with large-scale investment from other countries, the US dollar is considered as a world currency, and since the world share market and derivatives market are linked to the American markets, the crisis has spread to almost all countries in the world. It is pointed out that the global economic crisis is likely to roll back the development gains of the last decade and may precipitate a human tragedy in many developing and developed countries. According to World Bank estimates, the global economy is expected to contract by 1.7 per cent in 2009 (Heyzer and Mochida 2009). Due to the slump in global demand, the commodity price boom has been busted. Since peaking in mid 2008, oil prices, for example, have fallen by more than 70 per cent, energy prices by 60 per cent and food and metals by nearly 36 per cent. According to the ILO, 50 million people are likely to lose their jobs during the current crisis and world unemployment has increased from 180 million in 2007 to 230 million in 2008 (ibid.). One of the regions that faced serious consequences

Migration and Development Linkages in Global Economic Crisis Y 15

because of the crisis is that of the Gulf Cooperation Council (GCC) countries. These countries face acute economic problems due to the global crisis because they are heavily dependent on oil exports and migrant workers for their survival. The GCC countries (Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and the United Arab Emirates [UAE]) have the highest share of migrants in the total population among the countries of the world and rely heavily on migrant workers. According to one estimate, the UAE had the highest percentage of migrants, i.e., 75.7 per cent in 2002. The other GCC countries with very a large foreign population are Qatar, Kuwait, Bahrain, Saudi Arabia and Oman. In the GCC countries, migrant workers account for about 70 per cent of the labour force, while in the UAE, the percentage of migrant workers was about 90 per cent in 2000. According to a recent United Nations report (Heyzer and Mochida 2009), the percentage of migrants in the total population was 71.4 per cent in the UAE, 62.1 per cent in Kuwait, 25.9 per cent in Saudi Arabia and 25.9 per cent in Oman in 2005. According to a World Bank estimate, the GCC countries that are facing a massive decline in construction activities, trade and commerce, other services, manufacturing, etc., will face a fall in growth rate from 5.8 per cent in 2008 to 1.1 per cent in 2009 (ibid.). This indicates a situation of acute recession. Further, a sizeable proportion of migrant workers in construction, trade, manufacturing, agriculture and domestic service sectors will lose their jobs. The ILO has estimated that 10 per cent of unskilled workers from the GCC countries will return home in 2009 (ibid.). The return of large numbers of migrant workers from the GCC countries will create serious economic problems in the migrant households of the labour exporting countries. The jobless unskilled and semi-skilled migrants who cannot find employment in the labour market after return will be pushed into poverty. The major labour exporting countries which will be adversely affected are India, Pakistan, Egypt, Yemen, Bangladesh, Sri Lanka and Philippines. These countries account for 85.76 per cent of the total migrant population in GCC countries. A recent estimate suggests that there has been large scale return of Indian migrants from the UAE due to the current crisis (Rajamony 2009). In 2008, the UAE remained the single most important destination for Indian workers and the total stock of Indians was estimated at around 1.5 million. Indian workers account for 42.5 per cent of

16 Y S. Irudaya Rajan and B. A. Prakash

the labour force in the UAE. The number of Indians who availed amnesty and left the UAE in 2007 was 95,000. During the current crisis, it is estimated by the Ministry of Overseas Indian Affairs (MOIA), Government of India (GoI) that the number of Indians who returned is estimated is between 50,000 and 150,000. But according to another estimate, the total number of Indians in the UAE who returned due to the global crisis during the period September 2008– June 2009 was 200,000 (ICOE 2009). Among the total stock of Indian migrant workers in the GCC countries, the majority are from one Indian state — Kerala. As Kerala relied heavily on migration to the Gulf and remittances from the migrant workers, any return of migrants in large numbers will create serious economic consequences. Reduction or stoppage of the flow of remittances will create very acute economic problems for the thousands of households depending largely on the income of migrants. Lack of absorption of returned migrants into labour market will increase the level of unemployment. Serious adverse effects will occur in the labour market, consumption, savings and investment of Kerala due to the return of migrant labour. The relationship between migration and development is more closely linked in the GCC countries than in any other part of the world. Prior to 1973, all the GCC countries were poor, backward economies with weak infrastructure using traditional technology and population largely engaged in conventional occupations. The rise in oil prices in 1973, which led to rapid increase in the revenues of the countries, sparked off a process of industrialization and social change characterized by massive investment in social infrastructure. As these countries had a small population, they majorly relied on migrant workers for building up their economies. Almost all social and economic infrastructure, industrial units and modernisation of production and service activities were achieved mainly with the remittances of migrant labour. A good part of the investment in industrial production, trade, commerce, real estate, educational and medical services were made either by foreigners or joint ventures of the natives and foreigners. The newly rich natives began to employ migrant workers even for household work due to large availability of cheap labour. Thus, the contribution of migrant workers and foreigners were substantial in the transformation of the traditionally backward economies into modern economies. As a major part of all production and service activities of the GCC economies are handled

Migration and Development Linkages in Global Economic Crisis Y 17

by the migrants, any large-scale reduction in the migrant population will have very serious adverse consequences on all economic fronts in these countries. In the case of migrant labour, the GCC countries followed a migration policy to minimize the cost of labour. They prohibited settlement type of migration to avoid the social expenditure required to maintain the dependants of the migrants. On the other hand, the policy that has been followed is the promotion of contract migration. These countries impose a number of restrictions on the migrants regarding bringing along their dependants, stay, and changing jobs and employers. In the case of the majority of low-paid categories of semi-skilled and unskilled workers, a major portion of the cost of migration such as visa, travel, medical check-up, labour card, incidental expenditure, etc. was shifted to the migrants. The workers usually got a low wage rate due to lack of minimum wage legislation, the practice of stagnant wage rate and lack of payment of nonwage benefits as per the labour laws, and ineffective labour dispute settlement machinery. Thus, these countries got the services of the migrant workers for the development of their economies without payment of a fair wage or a wage including non-wage benefits as per international standards.

Global Economic Recession makes the Vulnerable Migrant Workers more Vulnerable The migrant workers in GCC countries are basically vulnerable because they belong to the contract migration category. The ILO has classified international migration for employment into two major categories: settlement and contract migration (ILO 1989). Settlement migration is a traditional type of migration where people migrate from one country to another to secure jobs and settle there. On the other hand, contract migration occurs when a worker is officially granted permission to enter another country and take up employment in a given job, and a contract is conducted on his behalf or between him and the employer or enterprise for which he will work. Contract migration itself can take different forms such as ‘individual contract’ and ‘collective contract’.

Conditions of South Asian Workers in the Gulf Studies on the conditions of Indian migrants in the Gulf countries reveal that most of the migrant workers are vulnerable as they are

18 Y S. Irudaya Rajan and B. A. Prakash

temporary contract workers, who are at the mercy of the employers and not protected by any civil or labour laws relating to wage, working conditions and treatment by employers. A study conducted by a High Level Committee on Indian Diaspora appointed by the Government of India to look into the condition of Indian migrant workers in the Gulf (GoI 2001) revealed the following: South Asian migrants face several problems related to labour contracts in the Gulf countries: alteration of the provisions of the original labour contract after arrival, non-compliance with the provisions of labour contract by the employer, denial of job, wages and non-wage benefits as per the original labour contract, etc. In several cases, the employers force skilled workers to do unskilled jobs with low wages. There is no effective administration mechanism to penalize the employer who violates the provisions of the contract. The civil, criminal and labour laws in the Gulf are totally biased in favour of the native employer in all aspects of labour disputes. Arbitrary reduction and non-payment of salaries are common practices in the Gulf. Salaries are often not paid when due, and sometimes not paid at all for several months towards the end of the contractual period, resulting in non-receipt of salaries for the work done. Besides this, the Gulf countries have not fixed any minimum wages for different categories of workers. Non-payment for legitimate overwork, denial of medical benefits, deduction of work permit fees from the amount due to them and denial of return airfares to their countries of origin at the end of the contract period are other common issues plaguing the migrant worker. As per the labour contract, a person has to work for eight hours per day. But in reality, employers force migrant workers to work for 10–12 hours per day. In the case of housemaids and other domestic workers, employers force them to work continuously for several hours without proper rest. The living and working conditions of the unskilled and semiskilled South Asian workers in the Gulf are not at all satisfactory. They are accommodated in crowded labour camps in small rooms with four to eight bunker beds. The facility provided to them such as toilets and kitchens are inadequate and unsatisfactory. In addition, the transportation arrangements for labourers from their camps to the work sites are poor and unsatisfactory. In the Gulf countries, labour laws do not cover domestic workers such as housemaids, gardeners, cooks, bearers and such other

Migration and Development Linkages in Global Economic Crisis Y 19

personnel who work within a household. The general complaint is that these workers are not treated properly. In many cases, the treatment is harsh and unfair. Female workers working as housemaids and domestic helps are extremely vulnerable in most of the Gulf countries. While professional migrants are allowed to bring along their families, other categories of workers such as the unskilled and semiskilled are not given family visas. Income limits are fixed for the emigrants; they are given proper visas when they attain the stipulated limit. For instance, in the UAE the monthly income limit is 4000 Dirhams and in Kuwait it is 400 Kuwaiti Dinars. According to the practice currently prevalent in the Gulf countries, migrant workers are required to submit travel documents such as passports to the employer on arrival. This enables the employer to exercise undue control over the employee and to ignore or alter the terms of employment, which often results in some of the migrants becoming undocumented workers. No effective labour dispute settlement machinery is available to settle labour disputes arising due to contract violation, non-payment of salaries and other benefits, and harassment by employers. The Labour Courts are inaccessible to ordinary migrants. In the case of female domestic workers, they cannot approach labour courts as they are not covered by any labour laws. Concern for their families left behind in the countries of origin, difficult living and working conditions, lack of job security, etc., has often led to depression and melancholia among the workers, even resulting in suicide in extreme cases. Shepherds and agricultural workers have to work in remote areas with minimal or no contact with the outside world. A study conducted by us about the employment, wages and working conditions of Indian migrant workers in the UAE in 2002 based on 361 sample emigrants also arrived at similar conclusions (Zachariah et al. 2004). It revealed that the Indian migrant workers in the UAE are those in the unskilled, semi-skilled and skilled categories engaged in construction, production, transport and service activities. Nearly one-fifth were engaged in professional and technical-related work. Those with an educational level of primary and below secondary worked as unskilled and semi-skilled labourers. Majority of the migrants worked more than eight hours per day and one-tenth of them worked between 10–14 hours per day. The main problems faced by

20 Y S. Irudaya Rajan and B. A. Prakash

several Indian emigrants in the UAE were non-payment of salaries, denial of wages and non-wage benefits as per the contract, refusal by employers to release confiscated passports and non-payment of charges for the air ticket to return to their place of origin. Though the UAE labour laws stipulate that the employer should pay for the emigrant worker’s air ticket to return home, most of the emigrants are forced to spend for this from their own savings. Nearly one-third of them live in worker camps and in a majority of cases, the number of persons per room ranges from four to six.

Impact of the Global Crisis on Migrant Workers 1. The construction sector is the worst hit by the crisis and a large number of construction workers from India, Pakistan, Egypt, Yemen, Bangladesh, Sri Lanka and Philippines have been forced to return due to loss of employment. It is reported that half of the UAE’s construction projects, totaling US$ 582 billion have either been put on hold or cancelled, resulting in substantial loss of jobs. 2. In the Gulf countries, most private sector jobs, especially in agriculture, construction and household/domestic services, employ foreign workers. A prolonged economic crisis will therefore affect the livelihood of vulnerable workers in the region, including the large population of migrant workers, as well as remittance flows, tourism, etc. 3. Due to the non-completion of projects and premature termination of contracts, employers incur heavy losses which result in non-payment of salaries, reduction in salaries, and lengthening of the hours of work without wages for those who continue to work. As a large number of unskilled and semi-skilled workers have incurred heavy debts for meeting the cost of visa, travel documents and work permits etc, they cannot return to the home country immediately (Table 2.1). 4. Another category of migrant workers who will face more acute problems due to the global crisis is women working as housemaids. This category of workers has been facing severe problems such as non-payment of wages, over-working, lack of holidays, physical abuse, sexual exploitation and harassment. It is reported that Sri Lankan housemaids with higher salaries have been sent back and those with minimum pay are allowed to continue with increased workload.

Migration and Development Linkages in Global Economic Crisis Y 21 Table 2.1: Average cost of emigration in different migration corridors, 2008 Countries India to Bahrain India to Kuwait India to Oman India to Qatar India to Saudi Arabia India to UAE India to UK India to USA

Average cost 57,172 53,951 56,840 66,316 74,606 61,308 56,589 42,080

Note: Estimated by the authors based on the Kerala Migration Survey 2008 conducted among 15,000 households in Kerala, financed jointly by the Department of Non-Resident Keralite Affairs, Government of Kerala and Ministry of Overseas Indian Affairs, Government of India.

5. Conditions are expected to become harsher for many of the unskilled and semi-skilled categories of workers as they try to hang on to their jobs and stay in the countries. They will be affected by reduction in pay, longer working hours and deterioration in working conditions. 6. The abundant supply of Asian workers in the pre-global crisis period resulted in the stagnation of wages in the case of construction workers, housemaids and domestic workers and other unskilled and semi-skilled categories of workers. The global crisis will result in further reduction in their wages. For instance, the UAE has already witnessed 10 strikes organised by the Asian labourers when the oil prices were above US$ 100 per barrel. 7. Currently, a large number of workers who are in the Gulf are undocumented workers. Many of them became so due to the malpractices of the employers such as refusal to release the passport or denial of consent to switch jobs, etc. These categories of workers will face further exploitation at the hands of their employers in the new context of reduced demand for jobs. 8. The Gulf countries are using this opportunity for implementing more measures to reduce the dependence on unskilled and semi-skilled categories of workers. Some countries may follow deliberate policies of exclusion or expulsion of migrant workers. All the GCC countries have been following immigration

22 Y S. Irudaya Rajan and B. A. Prakash

policies to reduce the size of the unskilled and semi-skilled migrant workforce. 9. In the labour market, the vulnerability of the migrant workers created due to the global crisis makes them an easy target for unscrupulous employers as they can be underpaid, hired or dismissed at a moment’s notice and denied non-wage benefits due to them as per labour law or practice. 10. Decline in the flow of remittances to labour sending countries may affect the living standards of those households depending mainly on migrant remittances. In general, countries depending heavily on migrant remittances will face very serious adverse economic consequences.

Need for Change in Immigration Policies Immigration Policies of GCC Countries The GCC countries have been following a policy of temporary contract migration to circumvent the social expenditure needed to maintain migrants and to prevent the integration of foreigners with the local population. The migrant workers are discriminated against and paid lower wages and provided with lesser non-wage benefits as compared to the native workers. There is practically no effective labour law or labour dispute settlement mechanism in the Gulf countries. Besides this, the GCC countries have been following a deliberate policy of curtailing the size of the foreign migrant worker population through a number of immigration policy measures designed to reduce labour demand and discourage the inflow of labour from abroad (Shah 2006). Among the policies that aim at reducing the supply of foreign workers are: increased cost of living for migrant workers, nabbing and deportation of overstayers and undocumented workers, stricter visa regulations, and curbs on visa trading. The major policies that aim to reduce the demand for foreign workers are creation of job opportunities for nationals through training and market mechanisms, and indigenization of the labour force through administrative mechanisms (Table 2.2). Among the GCC countries, Saudi Arabia has been following migration policies which give unquestioned authority to the employer and offer no protection to the interest of the migrant workers. The basic elements of the migration policy regulations of Saudi Arabia are follows:

Migration and Development Linkages in Global Economic Crisis Y 23 Table 2.2:

Policies aimed at reducing the supply and demand of foreign workers in GCC countries Policies aimed at affecting

Supply of foreign workers Direct and indirect taxes on migrant workers (e.g. health fees)

Demand for indigenous workers Creating job opportunities for Nationals Vocational training for nationals Enhancing private sector benefits

Periodic Amnesties Stricter regulation of visa issuance in sending countries

Encouraging nationalisation through market-based measures Fees for employing expatriates Cash benefits for employing nationals

Restriction of visa trading; stricter implementation of laws

Forcing nationalization of labour force through administrative measures Nationalization of the public sector workforce Quotas on expatriates Quotas on employment of nationals Ban on hiring expatriates in certain sectors Tightening of immigration legislation

Source: Shah (2006).

1. Foreign or non-Saudi workers cannot enter the country unless they have a sponsorship of an eligible employer, or a permitted Saudi household in the case of domestic workers. 2. When a foreign worker enters the country, he/she has to get a residency card or iqama1 and a work permit. 3. Once the worker gets the iqama card, he/she has to give his/ her passport to the kafeel (the sponsor);2 the passport will be returned only when he/she leaves the country, as the iqama card stands in place of the passport. 4. A non-Saudi worker cannot change his/her sponsor or job unless the sponsor issues a release. 5. The sponsor has the right to refuse to give a release, deport the worker or ask the authority to ban the worker from coming back to the country for two years. The passports of residents who work in Saudi Arabia generally have to be surrendered to the sponsor while in the country. The iqama is the Saudi residence permit that residents use for identification instead of their passports. 2 The kafeel is the sponsor in whose name the visa is issued. 1

24 Y S. Irudaya Rajan and B. A. Prakash

6. Foreign workers in professional, technical and managerial occupations with post-secondary education have the right to bring their families with them to Saudi Arabia, while relatives of workers with lower education have the right to make only a short visit that might be extended to three months. 7. In addition, a foreign worker is prohibited from checking directly with the Passport Department for any official arrangements concerning his residency or visa. He/she has to do this though the sponsor who hands over the worker’s passport to an authorized person to make the arrangements. Thus, a large fall in demand for the migrant workers due to the global crisis on the one hand and the policy of GCC countries to reduce the size of the migrant workers on the other, have resulted in the proliferation of unhealthy practices in the labour market, making migrant workers all the more vulnerable.

Changes in Migration Policies to address Vulnerable Workers in the context of the Crisis Follow ILO’s Guidelines to Protect the Rights of Migrants In the context of the global crisis making the vulnerable workers more vulnerable, the labour receiving countries in the Gulf may be prepared to accept international law to provide protection to migrant workers and their families. The ILO has put forward an international law having three fundamental aspects: Equality of treatment between regular migrant workers and nationals. Core universal human rights apply to all human beings, including all migrants. A broad array of international labour standards providing for protection in treatment and conditions at work (including occupational safety and health, hours of work, remuneration, non-discrimination, freedom of association, and maternity leave) apply to all workers.

Protect the Migrant Workers in the Context of Global Crisis The labour receiving countries should avoid forced expulsion of migrant workers, extend the labour inspection to sectors and

Migration and Development Linkages in Global Economic Crisis Y 25

workplaces where migrant workers are concentrated to ensure decent treatment, and to discourage the tendencies to cut pay and resort to exploitative and discriminatory practices. Bad labour practices such as forcing workers to work long hours, non-payment of wages and other benefits, confiscation of passports and other documents, and harassment of workers by employers should be prevented. Adequate compensation should be paid to the workers who are forced to return due to the economic crisis. The labour receiving countries should implement economic revival packages to generate more employment opportunities and social protection measures for migrants.

Labour Contracts, Protection of Rights and Labour Courts The labour receiving countries should introduce model labour contracts for various categories of workers, specifying the details of wages and non-wage benefits to be paid and should take steps to ensure their strict implementation. All labour receiving countries in the Gulf should fix the minimum wages and other working conditions like hours of work per day, paid holidays, accommodation facilities, provision for medical treatment, and meeting travel cost for return journey. Harassment, ill-treatment and sexual exploitation of female domestic workers by employers should be treated as criminal acts and the civil courts should treat these cases at par with those involving the other citizens of the labour-receiving country. Special labour laws should be formulated in the labour redressal machinery for all categories of female workers such as domestic workers in the households, and skilled, semi-skilled and unskilled female workers employed in establishments.

Reduce Cost of Migration and Ensure Migrant Rights The labour sending as well as receiving countries should make changes in the practices of recruitment, role of recruiting agents, hiring practices, etc., to ensure the reduction in the cost of migration to the minimum level or nil. Pre-departure training should be given to fresh workers who are going to foreign countries, especially to the Gulf.

26 Y S. Irudaya Rajan and B. A. Prakash

The labour sending countries of South Asia and South-East Asia should form a union like Organization of the Petroleum Exporting Countries (OPEC) to ensure the implementation of minimum wages, zero cost of migration, decent working conditions, labour laws to protect the workers and special laws for female workers in the labour receiving countries. The labour sending countries should collectively take a decision not to send workers to those countries which are not prepared to fix a minimum wage and offer decent working conditions.

References Government of India (GoI). 2001. Report of the High Level Committee on the Indian Diaspora. New Delhi: Ministry of External Affairs. Heyzer, Noeleen, and Shigeru Mochida. 2009. The Global Economic and Financial Crisis: Regional Impacts, Responses and Solutions. New York: United Nations. Indian Council of Overseas Employment (ICOE). 2009. ‘Impact Assessment of Global Recession on Indian Migrant Workers’. Paper presented at the international conference on Financial Crisis in the Gulf and its Impact on South and South-East Asian Migrants, 21–22 July, Centre for Development Studies, Thiruvananthapuram. International Labour Organization (ILO). 1989. World Labour Report, 1989: Employment and Labour Incomes — Government and its Employees Statistical Appendix, Vol. 4. Geneva: Inter-national Labour Organization. Shah, Nasra M. 2006. ‘Restrictive Labour Immigration Policies in the Oil-Rich Gulf: Effectiveness and Implications for Sending Asian Countries’. http:// www.un.org/esa/population/migration/turin/Symposium_Turin_files/P03_ Shah.pdf (accessed 7 November 2011). Rajamony, Venu. 2009. ‘Impact of the Financial Crisis on Indian Migrant workers in the UAE’. Paper presented at the international conference on Financial Crisis in the Gulf and its Impact on South-East Asian Migrants, 21–22 July, Centre for Development Studies, Thiruvananthapuram. Zachariah, K. C., B. A. Prakash, and S. Irudaya Rajan. 2004. ‘Indian Workers in UAE: Employment, Wages and Working Conditions’, Economic and Political Weekly, 39 (22): 2227–34.

3 The Global Economic Crisis and Impact on Migration from South-Asian and South-East Asian Countries What are the Lessons to be Learned? Vanessa Steinmayer1 The Global Economic Crisis, which burst out in 2008, has been a

severe setback to the world economy with only few countries being spared from the crisis. It began as a financial crisis, hitting first the financial sector and richer income groups, but soon trickled down to those groups who usually belong to the most vulnerable of society — among them migrant workers. Asia, which had previously seen a decade of unprecedented economic growth, had hoped to be spared from the crisis. But it soon became clear that through the slowdown in global demand, Asia’s economies were severely affected by the crisis, many of them falling into a recession or at least experiencing a major slowdown in economic growth. International migration has been increasing in the past decade and is one of the most visible outcomes of globalization. Migrant workers have largely contributed to the bustling economies of the GCC. Migration within the Asian continent is increasing and some countries have opened themselves to migrant workers due to demographic developments. Other countries have turned from countries of net out-migration to countries of net in-migration. The GCC countries, the major destination for most labour migrants from Asia sector, have been affected by the crisis through the decline in oil-price but have also experienced high financial losses from the financial crisis. Many of the ambitious construction projects were reported to be stopped or postponed.

1 The views expressed here are strictly the views of the author and do not necessarily represent those of the UN.

28 Y Vanessa Steinmayer

The question is whether the global economic crisis of 2008 has changed these trends. As a reaction to the global slowdown, several countries announced more protectionist policies which included providing jobs for the locals first. Media drew a bleak picture, reporting about large numbers of Asian migrants returning from the GCC countries. Large returns of migrants would put development in their countries of origin at risk since remittances from migrant workers have in the past contributed to reducing poverty, ensuring education and have helped their countries to keep current account deficits under control. A reduction of remittances would have severely affected these economies. The main message of this chapter is that although the impact of the crisis on migrants from Asia was less severe than often feared, and in some countries even almost visible, countries of origin and destination of migrants should learn from the situation to improve the quality of migration by providing better protection of migrant workers. The chapter will give an overview of key migration trends in Asia that have evolved in the past and will assess the impact of the crisis on migration and the migrant workers themselves. It will then conclude with recommendations to improve the management of migration for the benefit of all stakeholders.

Key Migration Trends in Past Decades Migration has a long history in Asia. Historically, the movement of people was associated with the rise and the decline of empires. The arrival of colonial powers in the 16th century has also led to migration flows — be it voluntary or forced — such as the movement of labour forces within Asia or from Asia to Africa by colonial powers, or the arrival of merchants from Asia — mainly India and China — to Africa. But the large labour migration flows that can be observed now are a relatively recent phenomenon, starting with the oil-boom in the countries of the Arabian Gulf in the 1970s, and later enforced by globalization. Migration has reshaped the economies and societies of the countries of destination and origin and has become a structural element of the economies and societies of several countries. Within this general trend towards more migration there are other trends which will be discussed subsequently. The Asia-Pacific region hosts four large migration hubs. The bulk of labour migrants from the Asia-Pacific, especially migrants from

Global Economic Crisis and Impact on Migration Y 29

South Asia, migrate to the oil-rich countries of the GCC.2 The other two important hubs for labour migrants are the more affluent economies of North-East Asia — namely Japan, Hong Kong, China, Republic of Korea, and Taiwan Province of China — and SouthEast Asia, especially Brunei Darussalam, Malaysia and Singapore. In 2005–2006 an estimated 65–70 per cent of Asia’s labour migrants migrated to work in West Asia, while around 10–15 per cent migrated to South-East Asia, and another 10–15 per cent to North-East Asia.3 The fourth migration hub is migration from Central Asian countries, especially Armenia and Tajikistan to the Russian Federation. Many countries are in need of in-migration for demographic and economic reasons. Overall, migration tends to flow from low- or lower middle-income countries, typically with medium to high fertility, to countries with high income and low fertility. Table 3.1 shows that most countries of net in-migration are countries with middle or high income and fertility below the replacement level of 2.1, except for Malaysia which is a country of in-migration with relatively high fertility. The GCC countries form a special case as high-income countries with high fertility, but small populations with low labour force participation of the local population, especially women. Countries with small populations, high income, high economic growth and low fertility such as Hong Kong (China), Macau (China), and Singapore are typical countries in need of migrant workers. Due to these demographic trends with a shrinking labour force and rapidly ageing population that raise the need for caretakers, the trend towards increased migration is irreversible. Although external economic shocks may temporarily reduce migrant inflows, it cannot be expected that this trend will be fundamentally reversed (Regional Thematic Working Group 2008). In South-East Asia, high- and middle-income countries have poorer neighbouring countries which lead to push and pull factors for international migration. Malaysia and Thailand draw the bulk of their migrant workers from neighbouring countries — Indonesia in the case of Malaysia, and Myanmar in the case of Thailand. Income gaps combined with the possibility to migrate through long and porous borders are main drivers of the migration flows. The GCC countries are Bahrain, Kuwait, Qatar, Oman, Saudi Arabia, and United Arab Emirates. 3 Economic and Social Commission for Asia and the Pacific (ESCAP) Labour Migration Database. 2

30 Y Vanessa Steinmayer Table 3.1: Key demographic and economic indicators for main countries of origin and destination of migrants in the West Asian and Asian region Population, total (thousands) (2010) Bangladesh Hong Kong, China India Indonesia Japan Korea, Rep. Kuwait Macao, China Malaysia Nepal Oman Pakistan Philippines Qatar Saudi Arabia Singapore Sri Lanka Thailand United Arab Emirates

GNI per Net Fertility rate, Population capita, Atlas migration total (births growth rate method (2010)/ per woman) (per cent), (current population (2010) 2010 US$), 2009 (%)

148,692

2.16

1.25

580

–1.0

7,053 1,224,614 239,871 126,536 48,184 2,737 544 28,401 29,959 2,782 173,593 93,261 1,759 27,448 5,086 20,860 69,122

1.14 2.54 2.06 1.42 1.39 2.25 1.16 2.57 2.59 2.15 3.20 3.05 2.20 2.64 1.37 2.24 1.53

1.04 1.32 0.98 –0.07 0.39 2.41 2.01 1.57 1.68 1.89 1.77 1.68 2.90 2.13 1.11 0.80 1.10

31,570 1,220 2,050 38,080 19,830 43,390a 39,550 7,350 440 17,890b 1,000 1,760 … 17,210 37,220 1,990 3,760

+7.6 –0.2 –0.8 +0.4 –0.1 +9.7 +14.3 0.6 –0.6 +6.0 –1.4 –2.1 +18.8 +3.6 +6.6 –2.3 +1.1

7,512

1.71

10.90

26270c

+10.9

Source: World Bank, World Development Indicators 2010, Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat, World Population Prospects: The 2010 Revision, http://esa.un.org/ unpd/wpp/index.htm (accessed 13 December 2011). Note: a Data from 2007 b Data from 2008 c Data from 2004

Overall Increase in Migration Flows Migration outflows as a whole have increased over time, especially from Bangladesh, India, Indonesia and the Philippines, which are also the countries with the largest migration outflows in recent years (Figure 3.1). In the recent decades, more countries started to systematically encourage out-migration, such as Nepal where

Global Economic Crisis and Impact on Migration Y 31 Figure 3.1: Yearly labour migration outflows from selected countries in Asia, 1976–2008

Source: ESCAP Labour Migration Database.4 4 1. Bangladesh: Bureau of Manpower, Employment and Training; see http://www.bmet.org.bd/Reports/Flow_Migration.htm (accessed 1 July 2009). 2. China: 1979–1991: Wang Shengjin, ‘China’s export of labor and its management’, Asian Pacific and Migration Journal, 4 (2–3), 1995; 1992: China’s Foreign Economic Relation and Trade, People’s Republic of China, Almanac of China’s Foreign Economic Relations and Trade 1993; 1993: China’s Foreign Economic Relation and Trade, People’s Republic of China, Almanac of China’s Foreign Economic Relations and Trade 1994; 1994: China’s Foreign Economic Relation and Trade, People’s Republic of China, Almanac of China’s Foreign Economic Relations and Trade 1995; 2002–2007: Ministry of Commerce, China, ‘Statistics of China’s Overseas Contractual Projects, Labor Service, Design and Consultant Service in 2007’, http://hzs.mofcom.gov.cn/aarticle/date/200801/200801053543 77.html (accessed 15 December 2008). 3. India: 1976–1987: ILO. 1990. Statistical Report 1990: Asian Regional Programme on International Labour Migration. Bangkok: UNDP–ILO; 1988–1994: Ministry of Labour (Government of India [GoI]), 1994–95. Annual Report 1994–95, p. 120. New Delhi: Ministry of Labour (Government of India); 1997–2002: Ministry of Labour (GoI), 2002–2003. Annual Report 2002–2003, p. 212. New Delhi: Ministry of Labour (GoI), http://lisd.delhi.nic.in/AnnualReport2002-03/lab13.pdf (accessed 1 November 2007); 2003–2007: Ministry of Overseas Indian Affairs (MOIA). 2007–2008. Annual Report 2007–2008, p. 62. New Delhi: MOIA (GoI). http://moia.gov.in/ shared/linkimages/200.pdf (accessed 18 June 2009); 2008: Centre for Development Studies, Trivandrum, from Ministry of Overseas Indian Affairs, India. 4. Indonesia: BNP2TKI (Badan Nasional Penempatan dan Perlindungan Tenaga

32 Y Vanessa Steinmayer

migration flows have been increasing in the past decade.5 Afghanistan’s Ministry of Labour is also starting a programme to encourage official labour migration.6 The increasing number of countries officially deploying workers, which also includes least developed countries, raised concern about a raise at the bottom for migrant workers’ wages. Some countries of origin are reluctant to ask for better protection of migrant workers from their countries fearing that employers would seek workers from other countries instead, especially countries with even lower income standards and that have a lower bargaining power on the global job market. The countries in the region of the GCC, which had seen unprecedented migration inflows since 1974, remain the main destination countries for migrants from Asian countries of origin, especially from South Asia. The most important destinations are Saudi Arabia and the United Arab Emirates, with Qatar rising as another prominent destination in recent years.

Intra-Asian Migration on the Rise Not only has the number of migrants increased, but the destinations have also become more diverse over time. Some South-East Kerja Indonesia), http://www.bnp2tki.go.id/content/view/180/87/ (accessed 20 September 2009). 5. Pakistan: Pakistan Bureau of Emigration and Overseas Employment, http://www.beoe.gov.pk/Downloads.asp (accessed 20 December 2008). 6. Philippines: 1998–2007: Department of Labor and Employment, Republic of the Philippines, http://www.poea.gov.ph/html/statistics.html (accessed 20 October 2011). 7. Sri Lanka: Sri Lanka Bureau of Foreign Employment, http:// www.slbfe.lk/feb/statis_main.asp (accessed 7 September 2009), 8. Thailand: 1975– 1988: ILO. 1990. Statistical Report 1990: Asian Regional Programme on International Labour Migration. Bangkok: UNDP–ILO; 1989–1991: ‘Trends in Asian Labor Migration, 1992’, Asian Migrant, 6 (1): 23, 1993; 1992–1993: Gullaprawit, C., and J. W. Huguet. 1995. ‘International Migration Data for Thailand’, Asian and Pacific Migration Journal, 4 (4): 621–26; 1993–1996: Asia Migrant Database, http:// www.smc.org.ph/atlas/data/Thai1.htm (accessed 5 November 2007); 1997– 2007: Sciortino, Rosalia, and Sureeporn Punpuing. 2009. International Migration in Thailand 2009. Bangkok: International Organization for Migration, and Thailand Overseas Employment Administration, Ministry of Labour, 2008. 5 http://www.smc.org.ph/misa/misa_pdf_download/NP.pdf (accessed 10 December 2011). 6 United Nations Economic and Social Commission for Asia and the Pacific (UNESCAP) 2010.

Global Economic Crisis and Impact on Migration Y 33

Asian countries, namely Malaysia and Thailand, have turned from countries of net out-migration to countries of net in-migration. Due to ageing populations with a rapidly shrinking labour force, East Asia has increasingly become a region of destination for labour migrants, especially from the Philippines and South Asia. The main destination countries in East Asia are Hong Kong (China), Macau (China) and Taiwan Province of China (ESCAP Labour Migration database). New destination countries also include Japan and the Republic of Korea. Japan has long sought to deal with the labour shortages through relocation of enterprises to labour-abundant countries and technology upgrade. This approach has its limitations, however, since not all jobs can be relocated to labour-abundant countries, especially jobs in the health sector. Japan is increasingly becoming a destination for health care personnel especially from Indonesia and the Philippines (Regional Thematic Working Group 2008). As another rapidly ageing country, the Republic of Korea responded to labour shortages with the design of a temporary labour migration scheme called the Employment Permit System (EPS), which establishes quotas for migrant workers, to be accepted according to the needs in each sector. Quotas can be re-adjusted every year (Park 2008).

Increasing Labour Migration of Women Although the majority of temporary labour migrants are generally men, the number of women migrants has been increasing in the past decade. From Indonesia, the Philippines and Sri Lanka, the number of migrating women exceeded the number of migrating men in some years. Although only about 10 per cent of the migrants from Bangladesh and Nepal are women, the number of migrating women has been increasing (Figure 3.2). Moreover, a large number of women migrants remain statistically invisible as some countries do not publish sex-disaggregated data of migrant outflows. The reasons why an increasing number of people felt the need to migrate can be found both on demand and supply sides. On the one hand, there was increasing demand for female-dominated jobs such as domestic work, caregivers and health care work from the GCC countries, and in East and South-East Asia. On the other hand, increased unemployment of men in many countries of origin, combined with job opportunities for women abroad, motivated many women to

34 Y Vanessa Steinmayer Figure 3.2: Migration flows disaggregated by sex from selected Asian countries, 1986–2008

Source: ESCAP Labour migration database, from official sources.

migrate. Research has also shown that women — more than men — often opt for migration to escape from unwanted marriages, abusive relationships or rigid social systems. Breakup of families is sometimes not the consequence, but rather, the cause for migration (Regional Thematic Working Group 2008). Yet, migration of women raises a social concern that is different from that for the migration of men. Female migrants are more vulnerable than male migrants because of the nature of jobs they typically undertake. The issues of children being left behind and changing role patterns in the case of female migration have not been fully explored and would deserve further research. Most female migrants work in low-skilled categories and especially as domestic helpers. This is especially the case with Bangladesh, Indonesia and Sri Lanka; more than 80 per cent of migrant women migrate as domestic helpers. The nature of the work of the majority of migrant women makes them particularly vulnerable, since domestic work is not covered by the labour law of most destination countries in the GCC and in Asia. However, the jobs women migrants are typically taking have proven to be more crisis resistant than those most migrant men engage in, such as the construction industry, which typically responds very

Global Economic Crisis and Impact on Migration Y 35

quickly to external shocks like the 2008 global economic crisis. This poses a dilemma for policymakers: while jobs such as domestic work are relatively crisis-safe and can ensure remittance flows even during crises, these are at the same time the most vulnerable ones.

Remittances: A Curse or Panacea? Remittances are considered the most visible outcome of labour migration. In the case of many South and East Asian countries, remittances have increased in line with the increase of labour migration outflows. For several Asian countries, remittances have become the most importance source of foreign income. In the current discussion, remittances are often praised for being the most stable source of foreign income for developing countries, with less volatility than Official Development Assistance (ODA). For many countries in South Asia, remittances have helped to keep current account deficits under control in spite of a chronically negative trade balance. Moreover, research has shown that remittances can reduce the probability of current account reversals (Bugamelli and Paternò 2005). Remittances can also increase debt sustainability by reducing the country risk. Since remittances not only increase household but also government income, they can increase the government’s revenue base and reduce the marginal cost of raising revenues (Chami et al. 2008). Indeed, in times of low economic growth and increasing trade deficits, remittances have sustained the current account deficits of the largest remittance-receiving economies of Asia, especially Nepal, but also Bangladesh and the Philippines. Overall, remittances have been more stable than other capital inflows such as ODA or Foreign Direct Investment (FDI). Studies have also shown that remittances have led to poverty reduction and remittance-receiving households are on an average better-off than non-remittance-receiving households (see Pernia 2008). However, remittances can also have negative macroeconomic impacts. Ralf Chami et al. (2008) have shown evidence for the following: Remittances may reduce the governments’ incentive to maintain fiscal policy discipline. Remittances can decrease competitiveness of tradeable goods due to exchange rate appreciation. Remittances can reduce political will to enact policy reform. As remittances insure the public against adverse economic shocks,

36 Y Vanessa Steinmayer

they reduce pressure on households and the need to implement reforms that would enhance pro-poor growth. Higher remittance-receipts are also associated with lower control of corruption, quality of government and rule of law indicators (Abdih et al. 2008). While it is well known that remittances have a positive income effect at household level, it still has not been proven whether they have a positive impact on economic growth. It is even argued that remittances may delay growth-enhancing policies, reduce the labour force participation and lead to riskier investments (Chami et al. 2008). Possible negative macroeconomic effects show that remittances are not a panacea; they may reduce poverty but do not necessarily enhance economic development. In fact, if not adequately used, they may lead to development traps similar to other incomes such as natural resources (Pernia 2008).

The Global Economic Crisis 2008–2009: The Situation The 2008 Global Financial Crisis has often been regarded as the greatest global crisis after the Great Depression in 1929, which was characterized by skyrocketing unemployment and hyper-inflation. Concerns were raised that this crisis would similarly lead to higher unemployment which would also affect migrant workers. With many countries depending on remittances for financing developments, this raised concerns about economic, human and social development in migrant-sending countries and the situation of the migrants themselves.

Transmission Mechanisms of the Crisis to Destination Countries of Asian Migrant Workers Having started as a housing crisis in the US in 2007, the crisis developed into a major liquidity crisis in September 2008. It quickly turned into a global crisis and several Asian economies were severely affected, especially those with a large financial sector, such as Singapore and Hong Kong. Other Asian countries, such as China, were affected by the crisis through the slowdown in global trade.

Global Economic Crisis and Impact on Migration Y 37

The region of the GCC, one mainly still living from oil exports, was affected by the crisis through the decline in oil-price. Most of these countries have learned to deal with oil-price fluctuations and built up large reserves during the hike in prices. Dubai is the only exception, where only a small part of GDP is generated by oil. With most of Dubai’s ambitious construction projects being bank-financed and having been built on high-end demand, the Emirate has been seriously affected by the crisis.

Economic Effects of the Crisis on Economies of Asia Different from the 1997 crisis — which was a financial crisis with the epicentre in South-East Asia, especially Thailand — the 2008 crisis affected Asia through decline in real demand. Consequently, the sectors most affected were export-oriented ones such as electronics and automobiles, as well as sectors depending on availability of credit, such as construction. The automobile sector had already previously suffered a decline in demand due to high fuel prices and was further hit through the decline in global demand and by global liquidity shortages. After October 2008, industrial production fell significantly in Japan, Republic of Korea, Taiwan Province of China and Singapore and this fall continued in the first half of 2009 (EIU 2009). Most Asian industrialized economies had negative economic growth in 2009, or at least experienced a slowdown in economic growth, for instance, China (see Table 3.2). The economic setback was however less severe than in 1998, where the effects of the 1997 crisis became most visible in Asia.

Impact of the Global Economic Crisis on Migrant Workers Impact and Policy Reactions Related to Migrant Workers in the GCC Countries As stated in previous sections, temporary labour migration dominates labour migration flows from and within the Asian region. In theory, it would be expected that large numbers of temporary labour migrants would be forced to return in times of economic crises. Experience to date from the crisis has shown that job losses as well as return of migrant workers in East Asia were higher than in the GCC region. In the GCC region, migrants as well as employers

8.98

4.20 1.33 4.82

8.04

6.12 2.32 7.04

10.05

1997

9.83

1996

–7.25

–8.09 –2.12 –5.71

8.33

1998

7.82

2.54 –0.05 8.87

7.60

1999

9.83

9.47 2.87 7.71

7.91

2000

Source: World Bank, World Development Indicators 2011.

China Hong Kong SAR, China Japan Korea, Rep. Macao SAR, China Malaysia –0.41

–1.15 –0.22 3.67

8.54

2001

4.73

3.77 0.27 6.83

8.62

2002

0.96 1.35 2.78 16.06 4.79

9.41

2003

9.12 2.55 4.40 31.77 6.68

9.79

2004

8.98 1.51 4.37 7.78 4.90

10.59

2005

5.17 1.57 4.92 20.56 4.20

12.61

2006

5.39 1.87 5.07 14.64 5.55

14.02

2007

Table 3.2: Nominal growth of gross domestic product (GDP) in selected Asian economies 1996–2009

–0.16 –1.12 1.76 18.94 6.00

9.66

2008

–0.78 –5.13 0.40 –0.31 –2.91

8.89

2009

Global Economic Crisis and Impact on Migration Y 39

developed a number of coping mechanisms to avoid preliminary returns. This may be largely due to the high recruitment costs for the migrants themselves as for the employers. Although migrants have been affected by retrenchments, some countries have introduced some measures to protect migrants. Although experiencing growth losses, most GCC countries had built up large reserves which allowed fiscal space to react with counter-cyclical spending. For example, Saudi Arabia announced the continuation of its ambitious construction projects that had already started (EIU 2009). Qatar also reacted to the crisis with countercyclical spending in construction. Although the protection of the rights of migrant workers in the GCC countries in general remains an area of concern, the crisis has also shown that several GCC countries have adopted measures addressing the situation of the migrants in the crisis. Migrant workers who were the most affected by the crisis were construction workers in the Emirate of Dubai, where the construction sector had contracted sharply, with media reporting of 30–40 per cent of Dubai’s construction projects having been suspended or postponed during the crisis.7 Many companies did not massively lay off workers, but rather sent their workers on extended — and unpaid — leave, to save separation cost on the one hand, and on the other hand being aware of the fact that once migrant workers are sent home, hiring new migrant workers would be costly. The Government of the UAE has responded to this situation by permitting any worker who has not been paid for more than two months to change jobs without requiring the No-Objection Certificate, which had previously been required to be allowed to change jobs.8 According to anecdotal reports, many migrant workers actually changed jobs within the UAE, moving from Dubai to oil- and gas-rich Abu Dhabi, which was less affected by the crisis. A remaining area of concern is the sponsorship system, through which most migrants working in the GCC countries are recruited. The migrants’ work permit is only valid while working for their sponsor, Bladd, Joanne. 2009. ‘52% of UAE’s construction projects suspended’. ArabianBusiness.com, 5 February, http://www.arabianbusiness.com/545802-52of-uaes-construction-projects-suspended (accessed 20 October 2011). 8 Issa, Wafa. 2009. ‘Ministry allows unpaid workers to switch jobs’. Gulf News, 23 July, http://archive.gulfnews.com/articles/09/07/24/10334226.html (accessed 20 October 2011). 7

40 Y Vanessa Steinmayer

with changes being allowed only during the first three months. Recruitment through this system is costly for the employer as well as for the workers themselves. Moreover, the kafala system has often been criticized for creating room for irregularities and exposing migrants, especially female domestic workers, to higher vulnerability, since they totally depend on their employers. Employers are also made accountable in case workers change jobs illegally, which often drives employers to keep the passports of migrant workers. It appears that the crisis has increased awareness about the problems related to the sponsorship system as it raises issues for employers and migrant workers. In August 2009, Bahrain was the first country to remove the kafala system.9 This move not only reduces the vulnerability of migrant workers to some extent, but also allows the Labour Authority to control migrant flows better. In September 2010, Kuwait announced a similar move.10 Although most GCC countries are continuously seeking to increase the participation of locals in the labour markets, it is not likely that they would be able change their dependence on foreign workers, given the demographic situation of migrant workers being more than local ones in many of the smaller GCC countries such as Kuwait, Qatar and the UAE.

Policy Reactions in Asian Countries of Destination for Migrant Workers In many countries on the globe, migrant workers have been disproportionately affected by retrenchments. However, this was not due to their status as migrant workers but because of the sectors they typically work in. In countries of destination where migrants come to settle for long-term, it is unlikely that migrants will return due to a temporary economic shock. However, in countries where migrant workers are considered as temporary, they typically lose their work permits with their jobs. In Singapore, foreign workers were more affected by retrenchments during the crisis than resident Singaporeans. In the first half Harmassi, Mohammed. 2009. ‘Bahrain to end “slavery” system’. BBC News, 6 May, http://news.bbc.co.uk/2/hi/middle_east/8035972.stm (accessed 10 December 2009). 10 Toumi, Habib. 2010. ‘Kuwait to scrap sponsorship system’. Gulf News, 26 September 2010, http://gulfnews.com/news/gulf/kuwait/kuwait-to-scrapsponsorship-system-1.687703 (accessed 3 June 2011). 9

Global Economic Crisis and Impact on Migration Y 41

of 2009, more than 21,000 foreign workers left Singapore, while employment among residents increased by 7,000. With unemployment of resident Singaporeans rising to levels too high for Singaporean standards, there was an increasing anxiety that skilled foreigners were taking jobs that could be filled by locals. In addition, concerns were raised that the growing number of low-skilled foreign workers would have negative impacts on Singaporean ‘culture and language’ (EIU 2009). At the same time, Singapore sought to better integrate recently arrived foreign workers by offering them English courses and announced better protection of foreign workers from 1 January 2010. This would include: higher basic medical insurance coverage for foreign workers, and more punitive actions against employers who fail to pay their workers’ salaries promptly.11 The Republic of Korea responded quickly to the crisis; already in December 2008, it froze recruitment under its Employment Permit System (EPS), which involves temporary employment for migrant workers under bilateral schemes. In March 2009, the quota for migrant workers was reduced significantly from 72,000 to 17,000 yearly and the Korean Ministry of Labour announced plans to introduce a quota of foreign workers on construction sites in order to protect Korean workers from competition with foreign workers. At the same time, the Government of the Republic of Korea addressed the situation of migrant workers being laid off by extending the period allowed to find a new job from two to three months (Martin 2009). Malaysia, with the manufacturing sector severely affected by the crisis, announced that it would continue the official policy further in order to reduce the intake of foreign workers. In Malaysia, the intake of foreign workers had been periodically on the policy agenda, including during the 1997 crisis. As a result of the 2008 crisis, the Government discouraged employers from hiring foreigners and cancelled the visas of a number of foreign workers. Towards the end of 2009, this policy led to a shortage of about half a million workers in the manufacturing, service and agricultural sectors. The Government then declared its decision to accept foreign workers in sectors where Malaysian workers were not available.12 SMC (Scalabrini Migration Center), ‘Asian Migration News: Singapore’, 1–30 September 2009, http://www.smc.org.ph/amnews/amn0909/amn0909. htm#Singapore (accessed 26 November 2009). 12 SMC (Scalabrini Migration Center), ‘Asian Migration News: Malayasia’, 1–30 September 2009, http://www.smc.org.ph/amnews/amn0909/amn0909. htm#Malaysia (accessed 26 November 2009). 11

42 Y Vanessa Steinmayer

In Taiwan Province of China, the Council of Labor Affairs (CLA) announced that foreign workers would be granted an extension period if they were unable to change employers within the stipulated period because of valid reasons. But it also instructed to enforce the ceiling for the employment of foreign workers, which is at 20 per cent of the total workforce, more strongly. Employers failing to comply with this rule would have their quota revoked completely.13

Impact on Migrant Workers’ Outflows While early returns due to the crisis appeared to be rather limited, many companies reacted to the crisis by slowing down new recruitment. Most countries of origin that had already made available recent data on migrant workers outflows reported reduced migrant worker outflows in 2009 compared to 2008 (see Figure 3.3). Figure 3.3: Migrant worker outflows, selected Asian countries of origin, 2005–2010

Source: ESCAP Labour Migration Database.

The strongest reductions of migrant worker outflows were experienced by Bangladesh (–51 per cent in 2009) and India (–28 per cent). While India reported an increase in 2010, migrant outflows 13 SMC (Scalabrini Migration Center), ‘Asian Migration News: Taiwan’, 1–30 September 2009, http://www.smc.org.ph/amnews/amn0909/amn0909. htm#Taiwan (accessed 26 November 2009).

Global Economic Crisis and Impact on Migration Y 43

from Bangladesh continued to decrease. However, reduced outflows from Bangladesh are largely due to a sharp decline in outflows to Saudi Arabia due to deteriorated diplomatic relations. Deployments to the UAE especially dropped sharply for Bangladesh, India and Pakistan. The Philippines slightly increased deployments to the UAE (see Figure 3.4). This can be explained by the fact that migrant workers from Bangladesh, India and Pakistan were mainly employed in the construction sector. However, it also has to be noted that 2008 seems to be an outlier year with regard to deployments to the UAE, with exceptionally high deployments due to the start of several ambitious construction projects. As the start of new construction projects was slowed down, the number of additional workers required dropped. Filipino workers are mainly employed in the services sector, which had reacted to a lesser extent to the crisis. Figure 3.4: Number of migrant outflows to the United Arab Emirates, selected Asian countries of origin, 2005–2010

Source: ESCAP Labour Migration Database

Impact on Remittances During the crisis, it was often projected that remittance flows may reduce as a result. However, the bulk of remittances in most Asian countries originate from migrant workers in the GCC countries, rather than from the large diasporas in the US and the UK. In countries where migrant outflows have reduced after the crisis, remittances may drop in recent years when the contracts of those workers who had migrated in boom years will end.

44 Y Vanessa Steinmayer

Different from expectations during the global financial crisis in 2008, the impact of the crisis on remittances was limited. In Bangladesh, Nepal, Pakistan, and Sri Lanka, remittance receipts increased in 2009 as compared to 2008. Even the growth in remittances remained remarkable in most countries. Only in Bangladesh, remittance growth slowed down slightly, but still grew at 18 per cent as compared to a growth of 36 per cent in 2008. India experienced a slight drop of 1 per cent in remittance receipts in US dollars, but remittances in Indian rupees increased.14 In Indonesia, where migrants were more affected by the 2008 crisis, remittances remained stable as most migrants migrate to Malaysia (see Figure 3.5). Figure 3.5: Remittances received by selected Asian countries, 2005– 2009 (in US$)

Source: World Bank World Development Indicators

Conclusion and Recommendations A number of conclusions can be drawn for countries of origin of migrants and for host countries. For host countries, it has been shown 14 MOIA, GoI. 2011. Annual Report. New Delhi: MOIA (GoI). http://moia. gov.in/writereaddata/pdf/Annual_Report_2010-2011.pdf (accessed 11 May 2011).

Global Economic Crisis and Impact on Migration Y 45

that once migrants are in the country, they may not easily return to their countries of origin. They may still remain in the country, even as irregular migrants, and accept lower paid jobs under even more difficult conditions. Policies limiting the intake of foreign workers to protect the local ones have often not protected local workers but rather led to higher employment of irregular workers. The crisis has also shown that more possibilities have to be created to promote safe migration. Migrants are among the most vulnerable segments of society and the migration has exposed them to even more vulnerabilities. This has to be addressed by countries of origin and destination. In this crisis, several countries seem to have already started introducing some measures to improve the quality of life of those migrants who are already in the country, by providing them better protection, facilitating possibilities to stay, reforming temporary migration schemes while at the same time trying to reduce the quantity of migrants coming to the country. More would need to be done in this respect. For countries of origin of migrants, it would be important to cooperate rather than compete in the employment of migrants abroad. Formulating a common position on the protection of migrants and their families and implementing common policies could be for the benefit of all migrants. Moreover, the crisis shows again that migration is not a replacement for development nor is it a development strategy per se. Reliance on remittances can have risks similar to relying on commodity exports. This shows that providing jobs in the country itself and economic and social development should still be a priority. At the same time, governments can seek to benefit from existing opportunities of international migration while reducing the social cost of migration.

References Bugamelli, Matteo, and Franceso Paternò. 2005. Do Workers’ Remittances Reduce the Probability of Current Account Reversals?. Washington, DC: World Bank, Development Research Group, Trade Team. Chami, Ralf, Adolfo Barajas, Thomas Cosimano, Connel Fullenkamp, Michael Gapen, and Peter Montiel. 2008. Macroeconomic Consequences of Remittances. Washington, DC: International Monetary Fund. Economic Intelligence Unit (EIU). 2009. Country Reports: China, Hong Kong, India, Japan; Malaysia, Pakistan, Saudi Arabia, Singapore, South Korea,

46 Y Vanessa Steinmayer Taiwan, Thailand, United Arab Emirates. October, Economist Intelligence Unit. Martin, Philip. 2009. ‘Recession and Migration: A New Era for Labour Migration?’, India Migration Report, 43 (3): 671–91. Park, Young-bum. 2008. Admission of Foreign Workers as Trainees in Korea. Bangkok: International Labour Organization. Pernia, Ernesto. 2008. ‘Is Labor Migration Good Development Policy?’. [unpublished]. http://www.unescap.org/esid/Meetings/Migration09/6.pdf (accessed 20 October 2011). Regional Thematic Working Group on International Migration including Human Trafficking. 2008. Situation Report on International Migration in East and South-East Asia. Bangkok: International Organization for Migration, Regional Office for Southeast Asia. UNESCAP. 2010. ‘Proceedings and Recommendations of the Workshop on Strengthening National Capacities to Deal with International Migration’. 22–23 April 2010. Bangkok: United Nations Economic and Social Commission for Asia and the Pacific. http://www.unescap.org/esid/Meetings/ Migration10/Proceedings.pdf (accessed 20 October 2011).

Databases United Nations COMTRADE Database (2009). United Nations Population Division (2008): International Migrant Stock: The 2008 Revision, http://esa.un.org/migration/index.asp?panel=1, accessed 2 December 2009. World Bank (2009): World Development Indicators.

4 The Effect of the Global Economic Imbalance on Migrant Workers and Economies of the Gulf Cooperation Council Olga Marzovilla In May 1981, Saudi Arabia, Bahrain, the UAE, Kuwait, Qatar and Oman set up the Gulf Cooperation Council (GCC) in order to achieve closer cooperation and integration in all sectors. Despite the small size of most GCC countries and the prevalence of unusable land in Saudi Arabia, the Gulf economies have become major actors in the international global arena during this new millennium. This can be explained by two main reasons:

(a) The GCC area includes 40 per cent of world oil reserves as well as 23 per cent of natural gas reserves, which makes it one of the leading producers and exporters of hydrocarbons. (b) Due to foreign exchange reserves piled up over the new millennium, following the growth in the world’s oil demand and the standing increase in oil prices from 2001 to the first half of 2008, the GCC countries can have a considerable impact on international financial markets. Additionally, three of the leading sovereign wealth funds currently operating on international markets belong to this area — namely, the Qatar Investment Authority (QIA), the Abu Dhabi Investment Authority (ADIA), and the Kuwait Investment Authority (KIA). The above considerations help explain why the GCC area continued to grow at a mean annual rate of 6.4 per cent even in 2008, while null or negative rates featured in the rest of the world following the explosion of the international economic and financial crisis (IMF 2010b).

48 Y Olga Marzovilla

In fact, the Gulf countries would appear to have been affected only indirectly, and to a limited extent, by the financial crisis that started spreading worldwide from the second half of 2007. Conversely, they were markedly affected by the global economic imbalance that kept worsening throughout the new millennium and was ultimately the cause of the financial crisis. The importance attained by these countries as leading actors in the world economic arena, along with the possible impact that their domestic developments may produce on the international milieu, point to the appropriateness of reconsidering the effects of the said imbalance in the light of some characteristics that feature in the GCC members — so as to outline some possible guidance for the future, especially with a view to the monetary union they have planned.

The Global Economic Imbalance in the Dollar Area The global economic imbalance was the subject of a lively debate in economic literature, including contributions from, amongst others, Ronald McKinnon (2005, 2006 [with Gunther Schnabl], 2007); Ben S. Bernanke (2005); Michael P. Dooley, David Folkerts-Landau and Peter Garber (2003, 2004); Amar Bhide and Edmund Phelps (2005); and Richard Cooper (2005). Although there is no consensus on its causes, it is generally agreed that it affects the structure of the balance of payments of the main world economic actors — with particular regard to those of the dollar area. Over the last decade, the growing deficit of the US current account balance has been accompanied by the growing surplus of various emerging economies. Since 1983, the US balance of payments has been characterized by the dual presence of a current account deficit and a systematic and growing surplus in the financial account. However, the opposite imbalances of the two sections of the external accounts have expanded further during the new millennium and found their counterpart in the rising current account surpluses of the GCC countries, China and developing Asian countries (Figure 4.1). On different grounds, these countries have shown major surpluses in the balance on goods and services. While the successful export-led policies, as fostered by the major inflow of direct foreign investments, may account for the growing surplus of developing

Sources: IMF (2009b, 2010d); US Department of Commerce, Bureau of Economic Analysis.

Figure 4.1: Balance of payments (in US$ billion)

50 Y Olga Marzovilla

Asian economies, the increase in oil prices and exports underlies the surpluses shown by oil producers as well as by GCC countries. In both cases the surplus has facilitated the creation of official reserves, most of which have been used to acquire foreign financial assets — in particular, US assets. Accordingly, these economies have become major holders of US financial assets (Figure 4.2). More specifically, China overtook Japan in 2009 as the leading holder of US assets, while GCC countries quadrupled their assets in the 2002–2009 period and now rank seventh on the list of foreign holdings of US securities — after China, Japan, the UK, Cayman Islands, Luxembourg, and Belgium (US Department of the Treasury 2010).1 These processes contributed to reinforcing the status of the US as the world’s leading debtor, with net capital imports amounting to 38.2 per cent of the world’s imports in 2009 (IMF 2010a); while China and GCC countries have become the main creditors. Although Gulf countries have surpluses in their balances of payments, the US has produced considerable effects due to their interaction with three major features of their economies, namely: (a) Population dynamics; (b) Labour market structure; (c) Exchange rate regimes.

Population Dynamics and Migration Flows As shown in Figure 4.3, from 1950 to the end of the first decade in the new century the population of GCC countries rose from 4 million to about 40 million people — with a growth rate that was among the highest in the world and was favoured not only by the natural population growth rate, but also by significant net migration flows. The high dependence of the Gulf area on hydrocarbons production and exports played a key role in its population dynamics as well

1 In fact, these data underestimate the actual financial holdings of GCC countries. While in the 1970s petrodollars were deposited with the international banking system and invested directly by the latter, most proceeds from oil surpluses are currently recycled via other financial markets, which hide the ultimate identity of the individual investors and also make it more difficult to monitor and quantify the actual capital flows targeted abroad.

Sources: IMF (2010c, 2010d); US Department of the Treasury (2010). Note: a Excludes Hong Kong, Macau, and Taiwan. b Includes GCC Countries, Iran and Iraq.

(in US$ billion)

Figure 4.2: Official reserves and foreign holding of US securities of some emerging and developing economies

Population growth and net migration flows (in thousands)

Source: UN (2009) Database. a Note: The statistics for the year 2010 is based on estimates.

Figure 4.3:

Global Economic Imbalance Y 53

as in the labour market structure. Indeed, the small population of the GCC countries, the lack of specialized labour, the high inactivity rate and widespread illiteracy resulted in the imbalance between the offer of domestic labour and the increasing demand related to the development of the oil sector. This gap has been bridged by importing foreign workers since the 1930s, when major oil fields were discovered and the first extractions began. Since then, migration flows have become a particular feature of the GCC area. They can be broken down into three sub-periods as a function of the foreign workers’ countries of origin as well as of their sector-related utilization.

1930s to the First Half of the 1970s This period featured the influx of foreign workers mainly from the Arab region. The conflicts and tensions that were rife in those years, including the many Arab–Israeli conflicts, fuelled major migration flows of Palestinians, Yemenites, Egyptians and Lebanese towards the Gulf economies. They were mostly unskilled workers that met the increasing demand of the oil sector for labour and were accepted in the various destination countries easily because of their language, cultural and religious similarities. These flows grew consistently throughout this period and peaked during 1970–1975 — which includes the first oil shock — when about one million workers moved to this area.

1975–2000 The flows increased further in the second sub-period. The experience gathered in the 1970s and 1980s — when the major oil revenues piled up after the 1973 and 1979 oil shocks were depleted in a few years’ time and the external surplus, along with the budget surplus, turned into a deficit — raised awareness about how transitional the positive effects produced by the increased oil revenues were. This led the GCC countries to implement policies that were targeted at diversifying production activities so as to reduce the dependency of their economic growth on the ebb and flow of oil prices. These policies also resulted in fostering the creation and growth of the private sector, since they were supported by the introduction of privatization and liberalization processes. These developments fuelled the demand for manpower and can account not only for the major flows observed during this period,

54 Y Olga Marzovilla

but also for the more diversified sector-related distribution of foreign labour along with the increased demand for skilled, highlyspecialized workers. However, the most striking changes have to do with the origin of migration flows. Indeed, the remarkable growth in the incidence of foreign workers of Arab origin led to some concern both in political and in social and economic terms. As for the former, there were fears about the possible spread of extremist and radical positions within the framework of a pan-Arabist ideology. As for the latter, the concern was mostly focused on the expectations of Arab migrants to remain indefinitely in the Gulf countries, where they usually also took their families. Conversely, Asian workers proved less demanding: they accepted lower wages and temporary work, their families stayed in their home countries, and they were easier to employ and dismiss given the brokerage function discharged by Asian agencies in their recruitment. The de-Arabization of labour was accelerated following the first Gulf War in 1991. About two million workers from countries that had supported Iraqi claims (Yemenites, Palestinians, Jordanians, Sudanese) were expelled from Kuwait and Saudi Arabia and replaced by many Asian workers from India, Pakistan, Philippines, Indonesia, Bangladesh, and Sri Lanka. Overall, the incidence of the Arab component over the total migrants in the GCC region fell between 1975 and 2004 from 72 per cent to 32 per cent, to the benefit of the Asian component (Kapiszewski 2006).

2000–2010 The third sub-period confirmed and emphasized the trends that had already begun during the second sub-period. The diversification of economic structures became the leading feature of this phase, along with the steady increase of oil prices that fuelled such diversification. However, the fast accumulation of foreign exchange reserves in small economies, where wealth is markedly concentrated and the public sector is under the control of a handful of powerful families, led entrepreneurs to invest mainly in areas where they could be executed more quickly and easily — which caused the unbridled growth of real estate, building, and services sectors, i.e., of the socalled FIRE (Financial, Insurance, Real Estate) economy. Due to

Global Economic Imbalance Y 55

the limited domestic population, the feverish diversification process required additional inflows of foreign workers and led to new peak net migration flows — which were estimated to amount to about 2.5 million in the last decade (Figure 4.3). Although it is impossible to gauge the real dimension of this phenomenon,2 over 10 million migrant workers are estimated to have entered the GCC region between 1950 and 2008 — whereby foreigners nowadays make up a considerable percentage of the total Figure 4.4: Percentage distribution of population and labour force by nationality a, 2008b

Sources: Labour Market Regulatory Authority (LMRA), Kingdom of Bahrain, Bahrain Labour Market, 2009; Central Informatics Organization of Bahrain (CIO), Statistical Abstract, 2007; Institute of Banking Studies of Kuwait (Kibs), Economic and Financial Data base for 2009; Statistics Authority of Qatar, Labour Force Sample Survey, Results, December 2008; Saudi Arabian Monetary Agency, 45th Annual Report; UAE Ministry of Economy, UAE in Numbers 2007; UAE Ministry of Economy, Central Department of Statistics, Labour Force Survey, 2008. a Note: The countries under examination are as follows: BA (Bahrain); KU (Kuwait); OM (Oman); QA (Qatar); UAE (United Arab Emirates); SA (Saudi Arabia). b Data for Bahrain and Oman on population refer to 2007; those for UAE refer to 2005 for population and to 2006 for labour force.

The difficulties here have to do with the poor quality of the statistics available from the countries in this area — which tend to downsize the numbers related to foreigners and emphasize those for the domestic population — as well as with the impact produced by illegal migration (Cadène and Dumortier 2008). 2

56 Y Olga Marzovilla

population and the foreign labour force outnumbers the domestic one in almost all the countries of this area (see Figure 4.4). As shown in Figure 4.4, the incidence of migrants in the total population ranges from 27 per cent in Saudi Arabia to about 80 per cent and 90 per cent in the UAE and Qatar, respectively — while the foreign labour force rates are higher, being in excess of 80 per cent in Kuwait (84.4 per cent), the UAE (92 per cent) and Qatar (94 per cent).

The Impact of Migration on the Labour Market It is understandable that the remarkable proportion of foreigners in the overall population and labour force has resulted in several problems for the Gulf economies — first and foremost the need to ensure citizens’ rights by preserving their cultural identity. This requirement has led to the adoption of measures since the 1970s in order to ensure that foreigners could only stay on a temporary basis. To that end, they have been excluded from nationality rights, political and ownership rights, and welfare rights. Additionally, highly restrictive conditions have been imposed on family reunion, the duration of employment contracts has been rarely in excess of two years, and a recruitment mechanism has been introduced that relies on a sponsorship system (kafala), whereby any foreigner intending to work in a Gulf country should find a national of that country (kafil) who stands surety for him/her. In this way, the sponsorship system has increased the dependence of foreign workers on private nationals of the GCC countries, allowing the latter major flexibility in setting salaries and labour conditions — which often borders on exploitation and black market situations. While several constraints limit decision-making power of foreigners, a number of benefits have been granted to nationals — including the right to study and receive medical care, free transportation systems, precedence in public employment and certain private sector activities, high wages, generous retirement benefits and almost nonexistent taxation. Thanks to the bountiful welfare system and the granting of specific rights, these States are enabling their nationals to benefit from oil export revenues. Thus, it can be argued that on the whole, the high incidence of migrants in the overall population and labour force in the GCC

Global Economic Imbalance Y 57

countries has resulted in the unequal allocation of rights and duties, which has influenced the labour market and ultimately resulted in its segmentation. One major gap that has opened up is the one separating the private from the public sector. The first is where most foreign workers can be found. In all GCC countries the proportion of foreign labour in the total labour force is definitely higher than that of nationals, with rates in excess of 90 per cent in Qatar (99.5 per cent) and Kuwait (96.6 per cent) (Figure 4.5). This can be accounted for by considerations relating both to the demand for and the offer of labour. On the demand side, entrepreneurs prefer foreigners because the recruitment mechanisms can ensure lower wages, longer work hours and higher possibilities for dismissal. On the offer side, the higher competition rate in the private sector makes the latter less appealing to nationals as it affects wage levels and labour conditions. Figure 4.5: Distribution of labour force by nationality in private sector and public sector, 2008a (in per cent)

Sources: Labour Market Regulatory Authority (LMRA), Bahrain Labour Market, 2009; Institute of Banking Studies of Kuwait (Kibs), Economic and Financial Data base for 2009; Statistics Authority of Qatar, Labour Force Sample Survey, Results, December 2008; Saudi Arabian Monetary Agency, 45th Annual Report; UAE, Ministry of Economy, UAE in Numbers 2007, UAE Ministry of Economy, Central Department of Statistics, Labour Force Survey, 2008. a Note: Data of Oman refer to 2007.

58 Y Olga Marzovilla

Furthermore, the employment positions made available by the private sector are regarded as either excessively menial or inadequate by the individuals. The domestic workforce can mostly be found in the public sector, which it finds appealing because of the high wages, generous retirement benefits, favourable work conditions, and the social status that public employment gives. In all the GCC countries there is a high proportion of nationals in the public sector (Figure 4.5), the highest rates being those found in Saudi Arabia (92 per cent), Oman (84.7 per cent) and Bahrain (80.3 per cent). Thus, migration flows have created two segments in the labour market that are totally different, not only due to the nationalities involved, but also in terms of their respective flexibility; indeed, the private sector is highly flexible,3 while the public sector is completely not. Figure 4.6:

Rate of unemployment, 2008a

Sources: SAMA, 45th Annual Report; UAE Ministry of Economy, Central Department of Statistics, Labour Force Survey, 2008; Qatar Statistics Authority, Labour Force Sample Results, 2008; Institute of Banking Studies of Kuwait, Economic and Financial Database for Bankers, 2009; Ministry of Labour and Bahrain Centre for Studies and Research, Labour Force Survey, November 2004. a Note: Data of Bahrain refer to 2004; data of the UAE refer to 2005.

The wage flexibility index as estimated by the World Economic Forum — ranging from one to seven — shows very high values ranging from 5.4 per cent in Saudi Arabia to 6.2 per cent in Qatar; conversely, the rigidity of employment index — ranging from one to 100 — shows very low values ranging from 27 per cent in Saudi Arabia to 13 per cent in Kuwait (Schwab and Martin 2009). 3

Global Economic Imbalance Y 59

The rigidity of the public segment in the labour market can account, furthermore, for the paradox whereby high unemployment rates can be found among nationals of several countries in this area (Figure 4.6) in spite of the high demand for foreign labour. The rigidity in question actually hampers the migration of national and foreign workers between the public and the private sector and thus leads the former towards the public sector, which has by now reached saturation and can no longer absorb the labour force adequately. The cases of Saudi Arabia, the UAE and Bahrain are especially significant in this aspect: the unemployment rate is close to 10 per cent, 14 per cent and 18 per cent respectively as regards nationals, while the rates concerning foreigners are very low (0.4 per cent, 2.6 per cent and 2 per cent respectively), and the foreign labour force makes up almost 80 per cent of the total labour force in the private sector. The same applies to Oman, where unofficial sources estimate a total unemployment rate of about 15 per cent (CIA 2010). This problem is compounded if one considers the data concerning the youth unemployment rate, which is on average in excess of 30 per cent throughout the area. Faced with the urgent need to cope with this issue, national policies have followed a three-fold approach: (a) the production structure has been diversified further to stimulate demand for labour in a way that can meet the requirements of a young, rapidly growing population, which the oil sector is unable to ensure on its own given its capital intensive nature; (b) educational and training policies have been implemented to promote skills that are appropriate to the needs of the labour market; and (c) the nationalization of the labour force has been fostered (Saudization, Omanization, Kuwaitization, etc.). As for the latter, governments in the GCC countries have tried to increase the proportion of nationals in the workforce by way of various measures, including the ban on recruiting foreigners or the obligation to reserve labour quotas for nationals in certain activities; the granting of facilitations to businesses if they employ nationals, coupled with the imposition of taxes in case they recruit foreigners; and the extension of the benefits featuring in the public sector to some private sector occupations so as to make them more appealing to nationals. However, nationalization policies have only proved successful in the public sector. In the private sector, market logic has continued to predominate. Indeed, the higher competitiveness of foreign labour

60 Y Olga Marzovilla

in terms of wages, productivity and work conditions have led companies to ignore the prohibition on recruitment and disregard the facilitations envisaged for the employment of nationals. On the other hand, the fast pace of growth in these economies and their high population growth rates have determined new needs and demands that the national component of the population is unable to meet because they lack the required skills and/or are not interested. Additionally, nationalization policies have reinforced social segmentation, increasing the unequal allocation of rights and duties and fuelled mutual tensions and grievances between nationals and foreigners. Segmentation was not only between foreigners and nationals: it could also be found within the former, where different segments could be defined depending on nationality, work executed, education levels and skills, and underlying cultural models. This has translated into a sort of social scale — ranging from Westerners to non-Gulf Arabians up to Asians — as well as into major differences in terms of wage levels and living conditions. At the bottom of the scale there are contract labourers. These labourers are to be found mostly among Asian workers and they are usually recruited by brokerage agencies in the respective countries of origin. They are unskilled, illiterate workers and usually live in labour camps, from which they are taken daily to the places where they have to work. Given their humble conditions and the recruitment mechanisms, they are especially prone to the risk of exploitation and abuse, including lower wages, forfeiture of their passports, deduction of visa-related costs from their wages, black market trafficking in visas, extended working hours, delayed payment of their wages, dangerous working conditions, and segregation in labour camps as they are prohibited from living in urban areas and/or going to urban areas at night. Accordingly, the unequal distribution of wealth, rights and duties in a society featuring a large number of foreigners fuels a climate of conflict not only among nationals, but also among expatriates. Against this backdrop, the greatest fears felt by GCC nationals over the past few years have resulted from the fast growth rate of Asians — in particular Indians and Bengalis, who make up the majority of foreigners as well as include most contract labourers (Figure 4.7). In the context of a highly unequal, segmented, increasingly conflicting society, the effects of the global economic imbalance were produced. Those effects were increased by the characteristics of the exchange rate regimes of the Gulf countries.

Global Economic Imbalance Y 61 Figure 4.7: Main expatriate communities in GCC countries (values in thousands)a

Source: Kapiszewski (2006: 10). a Note: Estimates for various years — Bahrain, Oman, Saudi Arabia: 2004; Kuwait: 2003; Qatar, UAE: 2002.

Exchange Rate Regimes in the GCC Countries For over 30 years the GCC countries have been formally or informally pegging their currencies to the dollar.4 However, since 1 January 2003, the Gulf economies have officially adopted a dollar peg regime as the first step towards their full monetary integration.5 The reasons that led the GCC members to link their currencies to the dollar are basically two-fold: 1.

2.

Since international oil prices are in dollars, pegging national currencies to the US currency can ensure the stability of export earnings and government revenues, reducing foreign exchange risk. The GCC countries anchored their currencies to that of a country with strong institutions and traditions of stability in

4 Oman has officially pegged the riyal to the US currency since 1973; while Saudi Arabia, Bahrain, Qatar and the UAE, despite having de jure tied their currencies to the Special Drawing Right (SDR) until 2001, de facto have pegged the dollar at a fixed rate since the 1980s. Even the Kuwaiti dinar, which was formally tied to a basket peg until 2002, has always shown a pronounced stability against the dollar. 5 However, Kuwait withdrew from the agreement in May 2007 and reinstated the previous basket peg regime.

62 Y Olga Marzovilla

order to import that stability along with the credibility and confidence into their respective economies. Indeed, the dollar peg allowed the GCC countries to keep price dynamics basically stable for over 20 years. Still, the worsening of the global economic imbalance that started in 2002 along with the widening of the gap between the growing deficit of the US balance of payments and the balance surplus of Gulf countries ultimately resulted in turning the dollar peg into a vehicle of instability from the anchor country to the GCC economies — via both a liquidity effect and a cost effect.

Inflationary Effects Produced by the Dollar Peg: The Liquidity Effect The dollar peg translated the standing surpluses of the balances of payments in the GCC countries — resulting from the increase in the world oil demand as well as in oil prices — into monetary base increases (Figure 4.8). This led to the well-known dilemma of the so-called impossible trinity, i.e., the impossibility of simultaneously pursuing the three objectives consisting of internal balance, external balance and exchange stability in the presence of fixed exchange rates. Figure 4.8:

Monetary base (values in national currencies)

Source: IMF (2009a).

Global Economic Imbalance Y 63

Indeed, the need to defend the exchange rate and prevent capital inflows made it difficult to keep monetary circulation under control. This led to the alignment of the GCC member interest rates to the lower US rates, at a time when the rapidly growing economies of the area would have required more stringent monetary policies. This can be seen quite clearly 2007 onwards, when the short-term interest rates of the Gulf countries fell in parallel with those of the US, despite the pressure of inflation existing in their economies (Figure 4.9). The consequences were negative real interest rates — which encouraged borrowing — and the expansion of monetary offer in its broadest sense. This fuelled inflation and gave rise to speculative bubbles in those areas where bottlenecks were especially rife — for example, the real estate sector. The mean inflation rate as measured from the consumer price index (CPI) rose in the GCC countries as a whole from 0.2 per cent in 1998–2002 to 10.2 per cent in 2008 — the peak values were recorded in Qatar (15 per cent), Oman (12.6 per cent) and the UAE (11.5 per cent). In Saudi Arabia, where inflation was always lower than 1 per cent, consumer prices increased from 2006 onwards, so that the 2008 inflation rate was as high as 10 per cent (Figure 4.9). The peak increase concerned food and housing prices, which affect the living standards of the poorest part of the population to a greater extent. More specifically, in the 2004–2008 period, food prices rose by 24–30 per cent in Gulf countries with peak increases of 29.7 per cent and 30.1 per cent in Qatar and the UAE respectively. As regards the housing prices, the increase was in excess of 46 per cent in the Emirates (UAE Ministry of Economy 2008) and peaked at 60 per cent in Qatar (QCB 2008). Thus, it can be argued that — in the presence of diverging economic cycles in GCC countries as opposed to the US — the dollar peg translated an inflation process initiated by the increase in the international oil demand as well as in oil prices into the expansion of the monetary base, which ultimately fuelled inflation.6 The inflation effect was also due to a cost effect that was fostered by the dollar peg and amplified by the features of the local labour market. From an econometric standpoint, this conclusion is supported by a study performed by Nasser Saidi, Fabio Scacciavillani, Aathira Prasad and Fahad Ali (Saidi et al. 2009) to identify the factors that influence inflation pressure in GCC countries. 6

Interest rate and inflation rate

Sources: Kuwait Central Bank; Qatar Central Bank; Saudi Arabian Monetary Agency; UAE Central Bank; http://www.global-rates.com (accessed 8 September 2010); IMF (2010b). a Note: UAE data refers to the central bank overnight interest rate.

Figure 4.9:

Global Economic Imbalance Y 65

Inflationary Effects Produced by the Dollar Peg: The Pass-Through Effect The dollar peg converted the depreciation of the US currency, which featured in the 2002–2008 period, compared to the currencies of the main trade partners of the GCC. Indeed, it also entailed the depreciation of the GCC currencies as compared to the euro, the yen, and the UK pound, raising prices in the national currency for imported goods. Regarding this, special importance should be attached to the depreciation of the US dollar compared to the euro, since Europe is the main import area for the GCC members.7 This effect plays a major role in the Gulf countries, as their economies are: small-sized; open to international trade; highly focused on oil production; have limited agricultural production and manufacturing, for which they need to import a large share of their consumer goods, raw materials, intermediate inputs and capital goods; and have a segmented labour market and a considerable proportion of foreign workers. Special importance should be attached in this context to the fact that these countries are highly dependent on imports of farming produce and food, whose prices markedly affect the living conditions of working classes and, above all, migrant workers. Food, in particular, is the main component of their imports — with peak values recorded in 2008 in Kuwait (16 per cent), Saudi Arabia (13.2 per cent), Oman (10.9 per cent) and the UAE (9.2 per cent) (WTO 2009). Against this backdrop, it can be easily understood that — given the dollar peg — any increase in import prices resulting from the depreciation of the dollar can jeopardize the standard of living of the working classes and fuel demands for wage rises along with social tensions.

The Influence of Labour Market Features on Inflationary Dynamics In the context of a highly segmented and conflictual labour market, where migrants predominate and the allocation of rights and duties is imbalanced, the effects of inflation, initiated by the oil price 7 The pass-through effect was recently estimated to range between 25 per cent and 35 per cent (Al-Qudsi et al. 2008).

66 Y Olga Marzovilla

increase and amplified by the dollar peg in the presence of the global economic imbalance, should be assessed. The price increases, and in particular those of food and housing, translated into a reduction in real incomes for a large section of the population — especially the poorer migrants — and resulted in wage increase claims amongst growing tensions. The highest wage increases were granted in the public sector. From 2005 to 2007, public salaries were repeatedly increased in all GCC member countries — the peak growth being the one found in the UAE (125 per cent) (GulfTalent.com 2008). This made public sector employment more appealing, and therefore made it more difficult for the private sector to employ national workers. Accordingly, the private sector’s dependency on foreign manpower increased further at a time of strong wage demand due to the need to defend the purchasing power of wages. This resulted in a steady increase of basic salaries: the highest increases were granted by the construction and banking sectors. More specifically, in the 2005–2010 period, wages rose in the GCC countries by about 47.6 per cent, with the highest growth rate recorded in 2008 (11.4 per cent) (GulfTalent.com 2011). Wage increases were also favoured by the effects produced by the dollar peg on the value of migrants’ remittances. The dollar depreciation during 2002–2008 could be observed not only with regard to the main currencies, but also in respect of some of the countries whence the main migration flows originate. In particular, from December 2003 to December 2007 the dollar depreciated by 17 per cent and 33.5 per cent compared to the Indian rupee and the Filipino pesos respectively. Due to the dollar peg, the Gulf currencies also showed the same level of depreciation, which resulted in reducing the value of remittances as calculated in the currencies of the migrants’ countries. This data should not be overlooked, considering the high proportion of migrants in the total population and labour force in the GCC countries. The temporary work contracts and the many obstacles hampering family reunions are leading migrants to send a large part of their wages to their families8 and assess their possible savings in the currencies of their countries (Razgallah 2007; Nauful and Termos Saudi Arabia ranks second worldwide in terms of remittances sent abroad, while the remaining GCC countries are among the first 30 on this list (World Bank 2008). 8

Global Economic Imbalance Y 67

2009). In this context, the dollar depreciation contributed towards fuelling the claims for wage increases. The greatest claims came from Indian workers, who form the majority of foreign workers and, in particular, contract workers. Being increasingly aware of their numbers and faced with the exploitation and vexatious recruitment mechanisms they are subjected to, they have staged major strikes and demonstrations that have led to an increase in their wages at a pace faster than those of workers from other countries in recent years (GulfTalent.com 2008). In conclusion, the particular structure of the labour market in the GCC countries interacted with the exchange rate regime giving rise to a mixed inflation in which the pressures on the cost side overlapped with those on the demand side, fostering tensions and protest among foreign workers.

Conclusion: The Need for the GCC Countries to Amend their Exchange Rate Regimes The GCC countries are characterized by a small indigenous population, the need to import labour from abroad to meet the demand coming from the oil sector and the diversification processes of the production structure, a high proportion of immigrants in the population and in the labour force, and wide-ranging inequalities in social and economic terms. Given these features, it is important for them to reconcile growth with social stability. The latter requires a more equitable allocation of rights and duties along with not only more effective actions in terms of human rights, but also more stable economic dynamics, such as preventing the redistribution effects of inflation from increasing inequalities in income allocation. The experience of the new millennium has shown that the dollar peg has been a major vehicle of inflation for GCC members and suggests that it should be amended. An alternative might consist of a basket peg system, whereby national currencies could be anchored to a basket of strong currencies that mirror the direction and intensity of commercial and financial flows in the international market.9 For a more in-depth overview of the benefits resulting from a basket peg to the GCC countries, see Abed et al. (2003); Aleisa et al. (2008); Habib and Stràsky (2008); Khan (2008); Marzovilla (2010); Marzovilla and Mele (2010). Some studies attempted to determine the appropriate composition of this basket: see Aleisa et al. (2008); Jen and Bindelli (2008); Saidi et al. (2008). 9

68 Y Olga Marzovilla

Given this context, it would be appropriate to include the euro in the said basket. Indeed, including the euro can reduce the risks related to possible exchange losses, pass-through effect and unwanted liquidity changes due to the exclusive anchorage to the US dollar.

Exchange Risks Oil exports are the main item of export among the GCC countries, with total exports ranging from 40 per cent in the Arab Emirates to 83 per cent in Kuwait. Such sales are targeted mainly at Asian countries. Conversely, Europe is the main partner of the Gulf countries in terms of their imports, which consist basically of food and manufactured products. Taking this framework into account, special importance should be attached to the features of the exchange rate regime. The currency revenues related to exports are essentially in dollars, both because oil is quoted in US dollars and because the latter is widely used as transaction currency in Asian countries. Conversely, their imports are paid mostly in euro, given the current practice whereby European countries quote their exports in the respective national currencies. Accordingly, the euro/dollar exchange rate plays a key role in the GCC region and in the past decade it has caused significant currency losses because of the trend towards depreciation of the US currency compared to the euro. This would appear to suggest that greater consideration should be given to the euro in the GCC countries’ exchange rate system.

The Pass-through Effect The importance of imports to the Gulf area from Europe increases the risks related to the pass-through effect. Europe is actually the source of 33.4 per cent of the imports for Qatar, 31.9 per cent of the imports for Saudi Arabia, 25.9 per cent for Kuwait, 24.7 per cent for the UAE, 15.1 per cent for Oman, and 11.4 per cent for Bahrain (WTO 2010).10 Given this framework, the dollar depreciation against the euro — as was the case in the 2002–2008 period — entailing the The per cent rates related to the US are comparatively smaller: 16 per cent for Saudi Arabia, 10.8 per cent for Kuwait, 9 per cent for Qatar, 9.3 per cent for the UAE, and 5.4 per cent for Oman. 10

Global Economic Imbalance Y 69

depreciation of Gulf currencies, might translate into a major increase in the price of the goods imported from the EU, such as resulting in significant inflationary pressure on the cost side. The latter pressure might be reduced by introducing a basket peg, in which the weight allocated to the euro should mirror its use in trade transactions of GCC members.

The Liquidity Effect The unquestionable predominance of import flows to the Gulf countries from the EU over export flows can account for the trade surplus that the European region taken as a whole generally shows with respect to GCC region — unlike the deficits shown by other major trade partners of the Middle Eastern countries (WTO 2009). This is imperative to devising an exchange rate system that can appropriately meet the requirements coming from the GCC members. Indeed, creating links to the currency of an area that — given the structure of its trade relations with Gulf countries — mostly shows trade surpluses can allow such countries to partly recover their monetary sovereignty by limiting the expansion effects on the monetary base that result from their being anchored to the currency of a country that shows systematic trade deficits with respect to them.

Stabilization of Remittances The dollar depreciation that has featured throughout the last decade, along with the oscillations observed in the last part of the decade (i.e., 2000–2009), has produced destabilizing effects on the value of the remittances from foreign workers who make up the majority of the overall labour force. These continuous oscillations have given rise to uncertainty among expatriates, influencing their families’ living conditions and fuelling tensions in the labour market. The basket peg may reduce those uncertainties by enabling greater stability of nominal effective exchange rates.

 One may therefore conclude that the advantage of a basket peg including the euro would consist of stabilizing nominal effective exchange rates along with the value of remittances in the currencies of the migrants’ countries, while reducing the risks due to the

70 Y Olga Marzovilla

pass-through effect and restoring some measure of flexibility in monetary policies. Although the dollar has appreciated in recent times and there have been uncertainties regarding the euro, the experience of the past decade shows that the weakness of the US currency — which has prevailed over most of the past decade — may no longer be construed as an episodic event. Indeed, it mirrors a world that is no longer as asymmetrical as the one that emerged from World War II and where globalization is advancing and speculation is on the rise, while the global economic imbalance is worsening. Given this context, one may not rule out a new trend of dollar depreciation in the coming years and, in the presence of the dollar peg, a new inflationary spiral in GCC economies. The major costs due to inflation in Gulf countries — given their economic, demographic, and social structures — would point to the advisability of amending their exchange rate regimes by relying on a basket peg.

Acknowledgement The paper was first presented at the conference ‘Un mondo in movimento: approccio multidisciplinare ai fenomeni migratori’, Società Italiana di Economia, Demografia e Statistica, 27–29 May 2010, Milan.

References Abed, George T., S. Nuri Erbas and Behrouz Guerami. 2003. The GCC Monetary Union: Some Consideration for the Exchange Rate Regime. Washington, DC: IMF. Aleisa, Eisa A., Shawkat Hammoudeh and Yuan Yuan. 2008. ‘External and Regional Shocks in the GCC Region: Implications for a Common Exchange Rate Regime’. Working Paper 426, Economic Research Forum. http://www. erf.org.eg/CMS/uploads/pdf/1218952201_426.pdf (accessed 3 November 2011). Al-Qudsi, Sulayman, Ala’ Kaloti, Faris Numan, Shifa Obeid and Hanan Marar. 2008. ‘Trade Volatility. The Dollar Peg and Inflation in the GCC Countries: Some Preliminary Research Findings’, The Arab Bank Review, 10 (1): 13–39. Bernanke, Ben S. 2005. ‘The Global Saving Glut and the U. S. Current Account Deficit’. Sandridge Lecture, Virginia Association of Economics, Richmond, Virginia, 10 March. Bhide, Amar and Edmund Phelps. 2005. ‘A Dynamic Theory of the China– U.S. Trade: Making Sense of the Imbalance’. Working Paper 4, Center of

Global Economic Imbalance Y 71 Capitalism and Society. http://www.columbia.edu/~esp2/Bhide_Phelps_ China_000.pdf (accessed 3 November 2011). Cadène, Philippe and Brigitte Dumortier. 2008. ‘L’impact politique des flux migratoires dans les États du Conseil de Coopération du Golfe’, L’Espace Politique, 4: 69–82. Central Intelligence Agency (CIA). 2010. The World Factbook. Washington, DC: Potomac. Cooper, Richard N. 2005. ‘Living with Global Imbalances: A Contrarian View’. Policy Briefs in International Economics no. PB05-3, Institute for International Economics. http://www.iie.com/publications/pb/pb05-3.pdf (accessed 3 November 2011). Dooley, Michael P., David Folkerts-Landau and Peter Garber. 2003. ‘An Essay on the Revived Bretton Woods System’. Working Paper no. 9971, National Bureau of Economic Research (NBER). ———. 2004. ‘The Revived Bretton Woods System’, International Journal of Finance and Economics, 9 (4): 307–13. GulfTalent.com. 2008. ‘Gulf Compensation Trends 2008’. http://www. gulftalent.com/home/Gulf-Compensation-Trends-2008-Report-20.html (accessed 3 November 2011). ———. 2011. ‘Employment and Salary Trends in the Gulf 2010–2011’. http:// www.gulftalent.com/home/Employment-and-Salary-Trends-in-the-Gulf2010-2011-Report-26.html (accessed 3 November 2011). Habib, Maurizio Michael and Jan Stráský. 2008. Oil Exporters in Search of an External Anchor. Frankfurt: European Central Bank. International Monetary Fund (IMF). 2010a. Global Financial Stability Report: Sovereigns, Funding, and Systemic Liquidity. Washington, DC: International Monetary Fund. ———. 2010b. Regional Economic Outlook: Middle East and Central Asia. Washington, DC: International Monetary Fund. ———. 2010c. Regional Economic Outlook: Western Hemisphere: Taking Advantage of Tailwinds. Washington, DC: International Monetary Fund. ———. 2010d. World Economic Outlook: Rebalancing Growth. Washington, DC: International Monetary Fund. ———. 2009a. International Financial Statistics Yearbook 2009. Washington, DC: International Monetary Fund. ———. 2009b. Regional Economic Outlook: Middle East and Central Asia. Washington, DC: International Monetary Fund. Jen, Stephen and Luca Bindelli. 2008. ‘A 70:15:15 Currency Basket Numeraire for the GCC’, The Tree of Liberty, 25 January. Kapiszewski, Andrzej. 2006. ‘Arab versus Asian Migrant Workers in the GCC Countries’. UN Expert Group Meeting on International Migration and Development in the Arab region, 15–17 May, Beirut. Khan, Mohsin S. 2008. The GCC Monetary Union — Choice of Exchange Rate Regime. Washington, DC: Peterson Institute for International Economics. Marzovilla, Olga. 2010. ‘Un regime di cambi per i Paesi del Gulf Cooperation Council: Dollar Peg o Basket Peg?’, Economia, impresa e mercati finanziari,

72 Y Olga Marzovilla 1 (Gennaio–Aprile): 1–18. Marzovilla, Olga and Marco Mele. 2010. ‘From Dollar Peg to Basket Peg: the Experience of Kuwait in View of the GCC Monetary Unification’, Global and Local Economic Review, 14 (1): 19–48. http://www.gler.it/Gler_XIV_ No1/GLER_XIV_No1_2010_cap_2.pdf (accessed 4 November 2011). McKinnon, Ronald. 2005. ‘Trapped by the Dollar Standard’, Journal of Policy Modeling, 27 (4): 477–85. ———. 2007. ‘The Worth of the Dollar’. Stanford Institute for Economic Policy Research (SIEPR) Policy Brief 123, Stanford University. http:// www.stanford.edu/group/siepr/cgi-bin/siepr/?q=system/files/shared/pubs/ papers/briefs/policybrief_feb07.pdf (accessed 3 November 2011). McKinnon, Ronald and Gunther Schnabl. 2006. ‘Current Account Surplus and Conflicted Virtue in East Asia: China and Japan under the Dollar Standard’. Hamburgisches WeltWirtschafts Institut (HWWI) Research Paper, 4 August. http://www.stanford.edu/~mckinnon/papers/McKinnonSchnabl%20HWWI%20_AP_UV_Aug%203%202006.pdf (accessed 3 November 2011). Nauful, George and Ali Termos. 2009. ‘The Responsiveness of Remittances to the Oil Price: The Case of the GCC’, Discussion Paper no. 4277, Institute for the Study of Labor (IZA). http://ftp.iza.org/dp4277.pdf (accessed 4 November 2011). Qatar Central Bank (QCB). 2008. The Thirty Second Annual Report. Financial Stability & Statistics Department, Qatar Central Bank. http://www.qcb.gov. qa/English/Publications/ReportsAndStatements/AnnualReports/2008.pdf (accessed 6 November 2011). Razgallah, Brahim. 2008. ‘The Macroeconomics of Workers’ Remittances in GCC Countries’. Working Paper no. 410, Economic Research Forum. http://www.erf.org.eg/CMS/uploads/pdf/1213518033_410.pdf (accessed 6 November 2011). Saidi, Nasser, Fabio Scacciavillani, Aathira Prasad and Fahad Ali. 2008. ‘The Exchange Rate Regime of the GCC Monetary Union’. Economic Note 3, Dubai International Financial Centre. ———. 2009. ‘Inflation in the GCC: An Analysis of the Causes and Implications for Monetary Policy’. Economic Note 5, Dubai International Financial Centre. http://www.difc.ae/sites/default/files/DIFC_ Economic20Note55Bv06Ref5D_0.pdf (accessed 7 November 2011). Schwab, Klaus and Xavier Sala-i-Martin. 2009. The Global Competitiveness Report 2009–2010. Geneva: World Economic Forum. https://members.weforum. org/pdf/GCR09/GCR20092010fullreport.pdf (accessed 7 November 2011). UAE Ministry of Economy. 2008. Population and Vital Statistics — Statistic Abstract 2008. Abu Dhabi: Government of the UAE. United Nations (UN). 2009. World Population Prospects: The 2008 Revision. New York: United Nations. US Department of the Treasury. 2010. ‘Preliminary Report on Foreign Holdings of U.S. Securities at the End – June 2009’. 26 February. http://www.treasury.

Global Economic Imbalance Y 73 gov/press-center/press-releases/pages/tg568.ASPX (accessed 7 November 2011). World Bank. 2008. Migration and Remittances Factbook 2008. Washington: World Bank. World Trade Organization (WTO). 2009. International Trade Statistics 2009. Geneva: WTO. http://www.wto.org/english/res_e/statis_e/its2009_e/ its2009_e.pdf (accessed 7 November 2011). ———. 2010. ‘Statistics Database: Trade Profiles’, October. http://stat.wto.org/ Home/WSDBHome.aspx?Language=E (accessed 7 November 2011).

5 The Financial Crisis in the Gulf and its Impact on South Asian Migration and Remittances S. Irudaya Rajan D. Narayana The financial crisis had its origins in the United States (US) in 2008, spread to Europe, and then to Japan. The effect of the crisis has been slow to manifest in the six Gulf Cooperation Council (GCC) countries.1 Their basic strengths — a public-funded banking sector and huge trade surplus due to the export of oil, the price of which saw unprecedented increase in a span of six months in 2008 — shielded the GCC economies from the adverse impact during the initial days of the crisis. This, coupled with significant inward foreign direct investments to all GCC countries, except Kuwait, also had a beneficial impact (ESCWA 2008). The GCC economies, however, began to feel the impact of the global crisis from the last quarter of 2008. The most significant indicator was the slowdown in the gross domestic product (GDP) growth rate in 2008 and the negative growth rate in 2009 in some of these economies. In the financial sector, the stock markets in all the GCC countries recorded a decline owing to the withdrawal of the foreign institutional investors. A number of privately-funded domestic and international projects in the Gulf region had reportedly been cancelled or abandoned, leading to a large number of layoffs or retrenchment of the workforce. Countries that are more exposed to global capital, investment and consumption demand face a greater risk of being affected by the crisis than others. For instance, Dubai in the United Arab Emirates (UAE), which depends heavily GCC countries are: United Arab Emirates, Saudi Arabia, Qatar, Bahrain, Kuwait and Oman. 1

Financial Crisis in the Gulf Y 75

on international capital, tourism and real estate, seems to be more adversely affected than other countries. On the other hand, Saudi Arabia, which has only 25 per cent foreign workers compared to much higher proportions in other GCC economies might be much less affected than others (Zachariah and Rajan 2009a). The slowdown in the growth rates of GCC economies has particular significance for the South Asian expatriates who are the main migrant labour in the GCC countries. This would, it was expected, affect the flow of migration and cause unforeseen large scale return emigration and falling remittances (Kapiszewski 2006). In this context, the following research issues become important: (a) How has the crisis affected the demand for South Asian migrant workers in the Gulf countries? (b) What strategies did the emigrants adopt to cope with the situation at their place of work (countries of destination), and what is the likely impact of the crisis on the home country in terms of decline in remittances, if any? (c) Did countries in South Asia see large-scale return emigration? Did they find a decline in the outflow of emigrant labour to Gulf countries and inward remittances from them? The chapter is a modest attempt at analyzing these issues in detail.

Approach and Methodology Following an assessment of the trends in expatriate workers and the employment structure in the GCC countries based on published data, mapping the trends and patterns of international migration, preferred countries of destination and trends in remittances over a long period of time is attempted. In addition to the macro assessment of the situation, the study is based on two surveys: (a) return emigrants in the countries of origin who lost their jobs in the countries of destination due to the financial crisis; and (b) return migrants who come back as per the terms of contract migration.

Return Emigrant Survey This survey was conducted among the emigrants who lost their jobs and were forced to return home because of the financial crisis

76 Y S. Irudaya Rajan and D. Narayana

in the Gulf. It was also aimed at examining their coping mechanism after their return to their home country. The survey was canvassed among 50 return emigrants each in the four countries of South Asia — Pakistan, Bangladesh, Nepal and Sri Lanka. In India, the survey was canvassed among 250 return emigrants in five states of India, selecting 50 each in Andhra Pradesh, Tamil Nadu, Kerala, Maharashtra and Punjab. Thus, the total number of return emigrants surveyed was 450 among the five countries of origin in South Asia. However, we confess that it was difficult to locate the emigrants who lost their jobs in the countries of destination and returned to the countries of origin. The return emigrant survey collected information on household details, the profile of return emigrants, household economic assets, employment, remittances and their utilisation, household expenditure pattern, reasons for return, adaptation and coping mechanisms.

Return Migrant Resurvey, 2009 Return migration from the Gulf is the normal process of contract migration. Migrants from South Asia go on contract work to the destination countries; and once the contract ends, they, in the normal course, return to the countries of their origin. As of now, we have no estimate of return emigration from the Gulf to South Asia. However, the Centre for Development Studies (CDS) has completed four large-scale migration surveys (1998, 2003, 2007, and 2008) over the last decade. One of the research objectives of this project is to assess the flow of forced return emigration, or return emigrants before the expiry of contract from the Gulf region to South Asia. To assess both regular return migrants and the crisis-led return emigrants from the Gulf, we revisited emigrant households from the 15,000 households of the 2008 Kerala Migration Survey (KMS). We estimated the extent of crisis-led return emigrants to Kerala after the revisit. Further in this chapter, we are applying the same methodology and projecting the figures to estimate the number of return emigrants from the Gulf to South Asia. In addition, the return migration resurvey 2009 also estimated the number of emigrants who lost their jobs in the Gulf, but had chosen to remain there without returning to their countries of origin. This is new information (‘lost job, but have not returned’) which will also be generated for South Asia.

Financial Crisis in the Gulf Y 77

Financial Crisis and Growth in the Gulf 2 The global crisis originating in the US, spreading to Europe and to Japan has affected the Middle East through a large fall in price of oil, reversal of capital inflows, depression of property and equity markets, and losses in sovereign wealth funds. The effect of the crisis varied across the countries depending upon country characteristics, such as high share of oil in total exports, large quantum of re-exports, sizeable share of services in GDP, especially transportation, trade, hotel, and restaurants. In the region as a whole, growth declined from 5.1 per cent in 2008 to 2.4 per cent in 2009. Among the oil producing countries, the sharpest slowdown was in the UAE, where the exit of external funds contributed to a large contraction in liquidity, a sizeable fall in property and equity prices, and substantial pressure in the banking system. At the other end of the spectrum is Qatar, which grew by about 9 per cent in 2009 (Table 5.1). Interestingly, the comparison of growth forecast for 2009 and the realized growth for the countries shown in Table 5.1 presents some important patterns. For the developed countries, the contraction forecast and the realized hardly reveal much of a difference but the recovery is expected to be quicker. For the South Asian countries as a whole, the realized growth is much better than the forecasts and the recovery is also rapid. The GCC countries show a mixed pattern: both the UAE and Kuwait witnessed contractions greater than the forecasts, the rest of the countries except Qatar reported growth rates higher than the forecasts; the growth recovery in 2010 and 2011 is on the expected lines (Table 5.1).

Gulf Economies: Population and GDP Growth The population of the GCC countries has increased by 8.39 million between 2000 and 2008, an increase which is almost the size of GCC excluding Saudi Arabia in 2000. The large increase in populations of almost all the Gulf countries is spawned by the high average growth 2 This section is based on the six country reports prepared by the research teams at the CDS. Most of the members of the research team have already visited the Gulf to assess the reality by talking to several stakeholders, and additional research is done by the authors of this report.

4.62

0.22

7.51

6.32

0.55

1.70

4.83

3.89

Bahrain

Kuwait

Oman

Qatar

Saudi Arabia

United Arab Emirates

Bangladesh

India

2001

4.56

4.85

2.65

0.13

3.20

2.57

3.01

5.19

2002

6.85

5.78

11.89

7.66

6.32

2.01

17.33

7.25

2003

2005

2006

8.19

5.55

9.24

6.02

10.62

9.39

3.03

15.03

6.79

5.14

7.90

9.21

9.82

Country of Origin 6.11 6.30 6.53

9.69

5.27

17.72

5.33

10.24

Country of Destination 5.64 7.85 6.65

2004

6.32 (6.31) 9.30 (9.89)

8.07 (8.38) 2.51 (4.46) 6.38 (6.81) 15.35 (26.76) 3.52 (2.02) 6.34 (6.06)

2007

Table 5.1: Selected economies: Real GDP growth rates (annual per cent change)

5.59 (5.96) 7.29 (6.40)

6.12 (6.31) 6.33 (5.53) 6.18 (12.84) 16.40 (25.42) 4.63 (4.23) 7.41 (5.14)

2008

5.00 (5.64) 4.52 (5.68)

2.64 (3.11) –1.14 (–4.82) 3.02 (3.59) 17.99 (8.65) –0.91 (0.60) –0.60 (–2.47)

2009

5.38 (5.78) 5.61 (9.67)

3.47 (3.96) 2.39 (2.33) 3.80 (4.72) 16.37 (15.96) 2.90 (3.42) 1.55 (2.43)

2010

6.01 (6.26) 6.89 (8.37)

3.94 (4.50) 4.34 (4.44) 6.00 (4.68) 8.90 (18.58) 4.40 (4.51) 3.29 (3.18)

2011

1.98

–1.55

0.184

0.751

2.462

Pakistan

Sri Lanka

Japan

United States

United Kingdom

2.097

1.599

0.262

3.96

3.22

0.12

2.818

2.51

1.414

5.94

4.85

3.95

6.24

7.67

3.12

7.67

6.18

3.72

2.758

3.637 2.058

2.939 2.838

2.779

Major Developed Countries 2.744 1.934 2.039

5.45

7.37

4.68

2.392 (2.36) 2.028 (1.95) 3.022 (2.69)

3.19 (3.41) 6.02 (5.64) 6.80 (6.80) –0.641 (–1.20) 1.111 (0) 0.707 (–0.07)

4.70 (6.10) 5.95 (1.64) 5.95 (5.95) –6.197 (–5.22) –2.751 (–2.63) –4.086 (–4.89)

3.60 (4.86) 2.50 (3.37) 2.20 (3.54) 0.515 (2.82) –0.049 (2.64) –0.396 (1.70)

3.25 (2.98) 3.50 (4.79) 3.59 (7.0)

2.168 (1.50) 3.53 (2.31) 2.121 (2.02)

4.81 (4.01) 4.50 (2.75) 4.98 (7.0)

Sources: International Monetary Fund (IMF), World Economic Outlook (WEO) Database, April 2009; IMF, WEO Database, October 2010. Note: Figures within parentheses are from WEO 2010; Data for 2009–2011 and 2010–2011 are forecasts in WEO 2009 and WEO 2010 respectively.

5.63

Nepal

80 Y S. Irudaya Rajan and D. Narayana

of GDP between 2000 and 2009. Only Kuwait showed a large increase in population with a relatively low increase in GDP (Table 5.2). Along with population growth, the proportion of expatriates in the population has shown an increase. In Qatar and the UAE, expatriates constitute over 80 per cent of the total population, and in Kuwait they account for close to 70 per cent of the total population. While in Saudi Arabia and Oman expatriates constitute slightly over a quarter of the population, Bahrain has over 40 per cent non-nationals in the total population. The proportion of expatriates in the labour force moves with the proportion in the population. Over 60 per cent of the expatriates are from South Asia and in some countries, such as Oman and Bahrain, they account for over 90 per cent of the expatriates (Table 5.3). Overall, the flow of South Asian migrants to the GCC countries is a function of GDP growth in the latter. Table 5.2: Growth of population and GDP in GCC countries Population (million) Country Kuwait Qatar Saudi Arabia UAE Bahrain Oman Total

2000

2009

Increase

Increase (%)

GDP growth 2000–08 (%)

2.217 0.606 20.474 2.995 0.670 2.402 29.364

3.443 1.098 24.897 4.764 0.779 2.769 37.750

1.226 0.492 4.423 1.769 0.109 0.367 8.386

55.30 81.19 21.60 59.07 16.27 15.28 28.56

6.68 11.17 3.91 7.74 6.29 5.36 –

Source: WEO Database, April 2009.

Table 5.3: Profile of expatriates in the Gulf, 2009

Country Kuwait Qatar Saudi Arabia UAE Bahrain Oman GCC

Population (in thousands)

Expatriates to total population (%)

% of South Asian expatriates to total expatriates

3,443 1,098 24,897 4,764 779 2,769 37,750

68.8 86.5 27.8 81.0 43.0 28.4 40.3

52.8 68.4 54.9 68.7 95.1 89.0 61.8

% of expatriates in the labour force 83.9 92.5 55.8 89.8 58.3 64.3 –

Sources: Compiled from various sources such as ‘Country Reports’ prepared by the CDS team, reported figures by the respective Embassies in the Gulf, and the following four publications: (a) UN (2009), (b) Kapiszewski (2006), (c) Taattolo (2006), and (d) Shah (2009).

Financial Crisis in the Gulf Y 81

Employment Structure in the GCC Countries In the GCC countries, over 50 per cent of the workforce is employed in manufacturing, trade and construction. Kuwait and Saudi Arabia are the exceptions, where the share of public administration and defence is rather high (Table 5.4). The share of construction in total employment increased rapidly during 2001–2008 in some of the GCC countries. For instance, in the UAE the share of construction sector employment increased by five per cent points during the period. In Saudi Arabia, the increase in employment in the construction sector during the period was of the order of 300,000. As construction is one of the major sectors attracting expatriate labour, it is important to analyze the effect of the crisis on that sector. Housing finance and private utilities have taken a severe beating along with finance institutions in the current crisis. A survey of projects (worth at least US$10 million) in mid-2009 reported 10–30 per cent cancellations or orders to put on hold in the GCC countries (Table 5.5). Dubai, which has about 60 per cent of all projects in the GCC, has taken the largest hit, which in turn has affected the GCC as a whole. Interestingly, the crisis has affected all subsectors — from commercial projects to residential properties — as the illustration from the UAE would show (Table 5.6). While new project starts have come down in the UAE, there is persistent high-level activity in ongoing projects that would be ‘the envy of many’ elsewhere in the world. There is evidence of increased construction activity in Abu Dhabi, Sharjah and Ajman. Thus, while new starts have come down and those about to be initiated have been put on hold, a lot of work continues to be carried out for current projects.

Gulf Crisis and South Asian Labour: The Links The link between economic growth and labour flow is through the growth in manufacturing, trade and construction. Construction, in particular, attracts large numbers of expatriate labour from South Asia. Any of the factors adversely affecting construction would affect the labour. The quick rebound of oil prices by mid-2009 and the not too depressing current account and budget balances have made the governments of the GCC countries bolder and induced them to continue major infrastructure investments. The increase in government expenditures (as per cent of GDP) was close to 10 points in most of the countries (Table 5.7), except in Bahrain and Qatar. Fiscal policy

0.47 0.01 0.49 17.54 0.13 29.86 24.62 6.55 4.20 3.46 7.54 0.01 1.24 0.24 2.11 0.06 0.21 0.30 100.00

Agriculture, Hunting and Forestry Fishing Mining and Quarrying Manufacturing Electricity, Gas and Water Supply Construction Wholesale, Retail Trade and Car Repairs Hotels and Restaurants Transport, Storage and Communication Financial Intermediaries Real Estate & Renting Services Public Administration and Defence Education Health and Social Work Community and Personal Services Domestic Services Extra-territorial Organizations and Bodies Not Classified by Economic Activity Total

9.09 0.44 1.96 10.77 0.33 34.68 16.18 5.97 1.30 0.29 1.77 0.02 0.76 1.91 1.08 9.96 2.59 0.90 100.00

Oman 2.60 0.08 1.90 4.43 0.01 14.23 14.03 2.89 3.85 1.21 5.59 14.75 5.23 2.40 4.18 21.86 0.11 0.66 100.00

Kuwait 1.92 0.43 5.27 8.69 0.66 37.14 12.28 1.96 4.33 1.09 3.43 6.35 3.16 2.55 1.54 8.79 0.21 0.18 100.00

Qatar a

1.3 13.0 1.2 20.6 20.0 4.2 6.2 1.4 3.3 10.8

1.32 7.28 0.96 10.22 16.10 3.20 4.42 1.08 3.22 18.03 11.96 4.33 2.26 10.79 0.13 0.01 100.00

4.5 8.4 – 0.01 100.00

b

b

5.0

a

UAE

4.69

Saudi Arabia

Sources: ILO Labour Statistics, ILO; Statistical Year Book 2008, Ministry of National Economy, Oman; For Saudi Arabia and the UAE the figures are taken from the country reports. Note: For Bahrain and the UAE figures show the paid employment by economic activity. For Oman figures show expatriate workers in private sector. The Kuwait figures are for the year 2005. a Fishing is included in Agriculture, Hunting and Forestry. b Education and Health are included in Public Administration and Defence.

Bahrain

Activity

Table 5.4: Share of employment (in per cent) across economic activities in GCC countries, 2007

Financial Crisis in the Gulf Y 83 Table 5.5: Projects affected by the crisis in the GCC Projects and finances involved

Country Kuwait Qatar Saudi Arabia UAE Bahrain Oman

Projects under construction

Projects cancelled/ on hold

Total project worth (US$ billion)a

% Cancellation

90 124 442 1,372 148 95

18 7 106 566 54 8

114 42 387 900 36 38

17 – 19 29 27 8

Source: Proleads. a All projects including cancelled/on hold. Note:

Table 5.6: Projects affected by subsectors, UAE Subsector

Projects under construction

Total number of projects

% Cancellation

340 288 495 249 1,372

487 406 712 333 1,938

147 (30) 118 (29) 217 (30) 84 (25) 566

Commercial projects Hospitality business Residential properties Retail projects Total Source: GulfBase.com (2009).

Table 5.7: Government expenditure in the GCC countries, 2006–2011 (percentage of GDP) Country

2006

2007

2008

2009

2010

Kuwait

31.83

29.94

40.15

47.36

43.22

2011 44.28

Qatar Saudi Arabia United Arab Emirates Bahrain Oman

26.42 31.96 18.39 28.48 34.44

25.37 34.36 18.98 28.70 35.33

24.52 30.81 21.22 28.00 29.42

26.66 44.54 32.11 31.42 38.73

23.22 42.80 28.34 30.83 37.38

22.47 40.75 23.07 29.53 37.02

Source: WEO Database, October 2010.

has played a crucial role in cushioning the impact of global crisis in the GCC countries. The interventions in the banking sectors have also been decisive. A further boost has been the, albeit lower, but healthy GDP growth

84 Y S. Irudaya Rajan and D. Narayana

in the whole of South Asia in 2009, 2010 and 2011, and the forecast of higher growth rates in 2012. South Asia and China have emerged as the major trading partners of GCC economies and the trade outlook does not look very depressing. However, the continuing adverse factors have been the depressed real estate and equity prices in the GCC countries, particularly in Dubai. The recovery would remain fragile as long as private investment does not stimulate growth. FDI, which had played a major role in the high growth of the pre-crisis days, fell drastically in 2009 in almost all GCC countries, except Oman, Qatar and Saudi Arabia. The fall in the UAE was from US$13.7 billion in 2008 to US$4 billion in 2009 (UNCTAD 2010). It is unlikely that the situation will improve till the Dubai World3 crisis is resolved.

Impact of the Crisis on South Asian Migrant Workers This section is devoted to an assessment of the impact of the crisis — on the South Asian migrant workers in terms of return emigration; on flows of labour emigration from Asia to the Gulf; and on inward remittances to South Asia. The assessment is based on the summary results of the emigrant household surveys and survey of return emigrants carried out to understand the coping mechanisms of individuals and families in times of crisis.

Return Migration to South Asia from the Gulf, 2009 All agencies working on migration and remittances in the South Asian countries and the Gulf region predicted an exodus of return emigrants from the Gulf to their countries of origin following the crisis. The CDS, which has undertaken four large-scale migration surveys in Kerala during the last 10 years to estimate the number of emigrants, return emigrants and remittances, revisited the households of the 2008 survey in 2009 (Zachariah et al. 2006)4 to arrive at 3 Dubai World, a holding company owned by the Government of Dubai that manages about 90 entities, asked to delay payment on US$26 billion of debt for six months, shaking the confidence of investors holding Government’s debt. 4 Return Emigrant Survey 2009 was conducted at the CDS and sponsored by the Department of Non-Resident Keralite Affairs, Government of Kerala (for more details of the survey report, see Zachariah and Rajan [2009a]). The field work was carried out during 16 June–September 2009.

Financial Crisis in the Gulf Y 85

reliable estimates of return emigrants. All those in the original sample who had returned were asked to cite the reasons for returning to Kerala. The questionnaire provided 10 possible reasons for return, among which the following three could be attributed to recession: job loss and return due to financial crisis, expiry of contract (renewal of contract did not take place as expected due to recession), and compulsory expatriation. The estimates of return migrants due to the crisis are provided in Table 5.8. Table 5.8: Number of return emigrants to Kerala due to recession in 2009 Sample Total Emigrants in 2008 based on 2008 Kerala Migration Survey Return emigrants among emigrants of 2008 in Return Migration Survey in 2009 Return Emigrants to Kerala due to financial crisis and recession

Kerala

3,953

2,193,412

304

168,681

110

61,036

Source: Zachariah and Rajan (2009c).

If we deduce that out of the stock of 2.19 million emigrants from Kerala, about 61,036 crisis-led migrants returned, then what could be the number of return emigrants from the Gulf to South Asia? According to the database available from various sources (both formal and informal), we arrived at a figure of 9.5 million South Asian emigrants in the Gulf; and the projected return emigrants from the Gulf region to South Asia at about 263,660.5 The country-wise estimates of return emigrants are provided in Table 5.9. One can also estimate the number of return emigrants from the countries of destination in the Gulf to countries in South Asia. For instance, India had a stock of 1.7 million migrants settled in the UAE; and in the projected number of return emigrants from the UAE, there were 47,000 Indians. Why is the number so small as compared to the large prediction? We postulate two important features of Gulf migration from South Asia as responsible for this: (i) the cost of migration to the Gulf; and (ii) the peculiarities of the channel of migration. South Asians incur 5 These could be underestimates as the composition of migrants from Kerala would have a lower proportion of unskilled workers.

86 Y S. Irudaya Rajan and D. Narayana Table 5.9: Estimates of return emigrants to South Asia from the Gulf due to financial crisis, 2009 Country Kerala India Pakistan Bangladesh Nepal Sri Lanka South Asia

Stock of emigrants 2,193,412 5,050,000 2,300,000 900,000 250,000 975,000 9,475,000

Return emigrants due to crisis 61,036 140,526 64,002 25,044 6,957 27,131 263,660

Source: As estimated by the authors.

huge costs to migrate to the Gulf. According to the Kerala Migration Survey 2008, the cost of migration to the Gulf varied between `53,951 to Kuwait and `74,606 to Saudi Arabia — between US$1,200–1,660 at an exchange rate of `45 per US$ (see Table 5.10). This applies to all the South Asian countries (also see UNDP 2009; Zachariah and Rajan 2009b; Rajan and Prakash 2009). The high cost of migration to the Gulf caused many emigrants to borrow from various financial sources. Under such conditions, even if the expatriates lost their jobs in the Gulf, they would prefer not to return home fearing inability to repay the debt already contracted there. They would rather accept any job at a lower wage and send home remittances to repay their loans even during a crisis in the destination country. Table 5.10: Average cost of emigration for different migration corridors from Kerala, 2008 Countries Kerala–Bahrain Kerala–Kuwait Kerala–Oman Kerala–Qatar Kerala–Saudi Arabia Kerala–UAE Kerala–UK Kerala–US

Average cost (in `) 57,172 53,951 56,840 66,316 74,606 61,308 56,589 42,080

Source: Zachariah and Rajan (2009b).

Another characteristic of South Asian migration to the Gulf is the part played by the social network, which consists of friends and relatives who perform a major role in the channel of migration flows by

Financial Crisis in the Gulf Y 87

arranging visas and other requirements for the emigration process. For instance, an all-India survey conducted by the CDS for the ILO and the MOIA revealed that close to 80 per cent of Indian emigrants utilized their friends and relatives as an important channel for migration (Table 5.11). This also ensured that in the event of a job loss, they could rely on someone to provide them temporary support. Table 5.11: Channels of migration by emigrants, 2007 Channel

Male

Female

Total

Female

Total

330 3 41

185 0 7

515 3 48

74.20 0.70 9.20

88.52 0.00 3.35

78.7 0.5 7.3

71 445

17 209

88 654

16.01 100.00

8.10 100.00

13.5 100.0

Friends and relatives Government agency Foreign employer Private recruitment agencies Total

Male

Source: Rajan et al. (2011).

Lost Jobs in the Gulf and have not Returned to the Country of Origin So there is a category of migrants, termed ‘lost jobs in the gulf and have not returned to the country of origin’ (Zachariah and Rajan 2009c: 21), who remain unemployed in the destination country and continue to look for jobs in sectors less/not affected by the crisis, at lower wages and poorer working conditions (see Zachariah et al. 2004). The Return Emigrant Survey 2009 conducted in Kerala offered a unique opportunity to estimate the number of those who lost their jobs in the Gulf countries due to the crisis there. According to the estimate made by the authors, of the 2.2 million emigrants from Kerala, about 39,396 persons lost their jobs between 2008 and 2009 but have not returned to their country of origin (Zachariah and Rajan 2009c). Applying the same methodology to estimate the number of South Asian migrants who lost their jobs in the Gulf yields the number 170,181 (Table 5.12).

Outflow of Workers from South Asia to the Gulf, 2009 As there was no official data to show the extent of outflow from South Asia to the Gulf, we estimated the possible trends using reasonably well-managed databases from these countries. All the countries in South Asia, except Sri Lanka, have reported decline in the flow of workers to the Gulf. The projected decline for India is

88 Y S. Irudaya Rajan and D. Narayana Table 5.12: Estimates of emigrants who lost their jobs in the Gulf but have not returned, 2009 Place Kerala India Pakistan Bangladesh Nepal Sri Lanka South Asia

Stock of emigrants

Number of those who lost their jobs but have not returned

2,193,412 5,050,000 2,300,000 900,000 250,000 975,000 9,475,000

39,396 90,703 41,310 16,165 4,490 17,512 170,181

Source: As estimated by the authors.

huge — about 280,000 — followed by Pakistan with just 12,000 (See Table 5.13). As regards the destination, the decline was large for the UAE, which has been more severely affected by the crisis than the other countries in the Gulf. But this was more than compensated by the increase in the number who left for Saudi Arabia. This pattern holds for India, Pakistan, Sri Lanka and Nepal. Thus, the crisis has changed the migration and demographic dynamics of South Asian workers in the Gulf region.

Inward Remittances to South Asia, 2009 The money that migrants send home is important not only to their families, but also to their country’s balance of payments. In many developing countries, remittances represent a significant proportion of the GDP as well as foreign exchange receipts. In a 2009 World Bank Table 5.13: Outflow of migrant workers from South Asia to Gulf, 2005–2009

2005 2006 2007 2008 2009

India

Pakistan

Bangladesh

Nepal

Sri Lanka

454,628 618,286 770,510 818,315 538,090

127,810 172,837 278,631 419,842 407,077

207,089 307,620 483,757 643,424 N.A.

88,230 128,306 182,870 169,510 152,272

192,004 170,049 188,365 215,793 226,299

Source: This table is based on the country papers prepared by the respective country teams at the countries of origin for this project.

Financial Crisis in the Gulf Y 89

Migration and Development Brief (Ratha and Mohapatra 2009), India was ranked number one in terms of the volume of remittances with US$52 billion in 2008 (4.2 per cent of GDP) among the South Asian countries. Bangladesh was ranked eighth and Pakistan ranked 11th in terms of remittances. On the other hand, Nepal is listed as one of the top 10 countries with the highest share of remittances to the GDP at 22 per cent. When the crisis hit in 2008, the World Bank (Ratha et al. 2008) stated, ‘The outlook for remittances for the rest of 2008 and 2009–10 remains as uncertain as the outlook for global growth, oil and non-oil commodity prices, and currency exchange rates.’ After several years of strong growth, remittance flows to developing countries began to slow down significantly in the third quarter of 2008 in response to a deepening global financial crisis. In late 2008, on request from the Government of Kerala, the CDS prepared a report called the ‘Global Financial Crisis and Kerala Economy: Impact and Mitigation Measures’ (CDS 2008). The report predicted that the remittances to Kerala were expected to increase from `30.122 crore in 2007 to `42.917 crore in 2008. Both the World Bank and the CDS reports predicted that the inflows of remittances to South Asia and Kerala were likely to continue, when the general expectation was that it would drastically fall. Our estimates based on the simple average of remittances for the available months from the country reports prepared by the teams suggests that all the countries of South Asia are resilient to the crisis in terms of remittances (Table 5.14). Our estimates put the growth in remittances to India at 3 per cent, from US$52 billion in 2008 to US$53 billion in 2009. The World Bank report ‘Migration and Remittances Trends 2009’ (Ratha et al. 2009) confirmed our estimates and said that the outcomes were better than expected, but that there were significant risks ahead. Why did remittances not decline in South Asia? From our study, the following six observations can be made: (a) the debts contracted to meet the high cost incurred during their migration kept the emigrants from returning to the countries of origin in spite of the layoffs; (b) the predictions of a large exodus of return emigrants from Gulf did not come true; (c) though the outflow declined in the first half of 2009, it has still not significantly affected the stock of South Asian migrants in the Gulf; (d) the appreciation of the US dollar vis-à-vis South Asian currencies, for instance, the current exchange

90 Y S. Irudaya Rajan and D. Narayana Table 5.14: Inward remittances to South Asian countries from migrant workers, 2000–2009 India

Pakistan

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009a 2009b

12,890 14,273 15,736 20,999 18,750 22,125 28,334 37,217 51,581 47,000 53,227

1,075 1,461 3,554 3,964 3,945 4,280 5,121 5,998 7,039 8,619 8,856

2000–01 2001–02 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08 2008–09a 2008–09b

10.73 10.25 33.45 –10.71 18.00 28.06 31.35 38.60 –8.88 3.19

35.91 143.26 11.54 –0.48 8.49 19.65 17.13 17.36 22.45 25.81

Bangladesh

Nepal

Sri Lanka

US$ million 1,968 2,105 2,858 3,192 3,584 4,314 5,428 6,562 8,995 10,431 10,525

111 147 678 771 823 1,212 1,453 1,734 2,727 3,010 2,812

1,166 1,185 1,309 1,438 1,590 1,991 2,185 2,527 2,947 2,892 3,308

32.43 361.22 13.72 6.74 47.27 19.88 19.34 57.27 10.38 3.12

1.63 10.46 9.85 10.57 25.22 9.74 15.65 16.62 –1.87 12.25

Percentage change 6.96 35.77 11.69 12.28 20.37 25.82 20.89 37.08 15.96 17.01

Source: World Bank and the country reports prepared by the research team for this project from five countries of South Asia. a World Bank estimates. Note: b As estimated by the authors.

rate of the Indian rupee to the US dollar is same as in 2001–2002; (e) the continuous rise in oil prices generating more income in the Gulf; and ( f ) the ‘reverse migration’ of the crisis-led return emigrants back to the Gulf.

Conclusions The unravelling of the impact of the crisis on output and employment in the US had generated an anticipation of large-scale retrenchment of expatriate labourers in the Gulf region. The anticipated misery and the need for rehabilitation had got some governments

Financial Crisis in the Gulf Y 91

in South Asia thinking about plans for the returning migrants. We find that the dimensions of the impact were not as large as was feared earlier. For a stock of approximately 9.5 million South Asian emigrants in the Gulf, the number who returned due to the crisis was about 264,000 (just around 2.78 per cent); and the number of those who lost jobs but continued to stay in the Gulf was about 170,000 (1.80 per cent). Overall, less than 5 per cent of the South Asian emigrants had lost jobs owing to the crisis. But such an impact had not adversely affected the yearly flow of out-migrants from South Asia. Compared to the outflow of about 2.2 million workers in 2008, the number in 2009 was close to 2 million, which was higher than the corresponding figure in 2007. As regards remittances (in US$), the annual percentage increase since 2004–2005 has been over 20 per cent for the South Asian countries, except for Sri Lanka. The magnitude of the increase has taken a hit all over South Asia, except Pakistan from where the outflow of migrants had been increasing at around 50 per cent every year since 2005. But remittances have not fallen following the crisis; either it has remained stagnant or has shown a mild rise depending upon the estimates. As most of the South Asian currencies have depreciated against the US$ (to which the GCC currencies are pegged) from late 2007, remittances in terms of domestic currencies would have shown an increase. Survey of migrant households in South Asia confirmed these estimates as 94 per cent of the households reported regular remittances during the crisis period as in previous years and recorded no significant change in the use of remittances. Overall, less than 5 per cent of the stock of South Asian migrants in the Gulf had lost jobs and had either returned, or were struggling to continue in the Gulf. The flow of workers from South Asia to the Gulf had also not been affected to any significant extent, but there were changes in the origin (in favour of Pakistan) and the destination (in favour of Saudi Arabia) of the flow. The volume of remittances into South Asia too did not show any remarkable decline. In sum, the crisis of 2008–2009 is a blip on the radar of labour migration to the GCC countries as far as South Asia is concerned. A few thousands lost jobs and had to come back but the number migrating to the Gulf has hardly fallen. With the quick recovery experienced in 2010 and 2011, the migration flow is going to increase and the resultant outcome — remittances will follow.

92 Y S. Irudaya Rajan and D. Narayana

Acknowledgements This project was funded by the Asian Development Bank; the Ministry of Overseas Indian Affairs, Government of India; and the Department of Non-Resident Keralite Affairs, Government of Kerala.

References Centre for Development Studies (CDS). 2008. ‘Global Financial Crisis and Kerala Economy: Impact and Mitigation Measures’. Report submitted to the Government of Kerala. Economic and Social Commission for Western Asia (ESCWA). 2008. ‘Foreign Direct Investment Report’. E/ESCWA/EDGD/2008/Technical Paper. 13 September, New York: UN. GulfBase.com. 2009. ‘UAE Construction Worst-hit in GCC’, 17 September. http://www.gulfbase.com/site/interface/NewsArchiveDetails.aspx?n= 110724 (accessed 14 November 2009). Kapiszewski, Andrzej. 2006. ‘Arab versus Asian Migrant Workers in the GCC Countries’. UN Expert Group Meeting on International Migration and Development in the Arab region, 15–17 May, Beirut. Rajan, S. Irudaya and B. A. Prakash. 2009, ‘Migration and Development Linkages Re-Examined in the Context of the Global Economic Crisis’, Invited Paper for the Civil Society Days of the 3rd Global Forum on Migration and Development, 2–3 November, Athens. Rajan, S. Irudaya, V. J. Varghese and M. S. Jayakumar. 2011. Dreaming Mobility and Buying Vulnerability: Overseas Recruitment Practices and its Discontents in India. New Delhi: Routledge. Ratha, Dilip and Saket Mohapatra. 2009. ‘Revised Outlook for Remittance Flows 2009–2011: Remittances Expected to Fall by 5 to 8 Percent in 2009’. Migration and Development Brief 9, Migration and Remittances Team, Development Prospects Group, World Bank. http://siteresources.world bank.org/INTPROSPECTS/Resources/MD_Brief9_Mar2009.pdf (accessed 27 October 2011). Ratha, Dilip, Sanket Mohapatra and Ani Silwal. 2009. ‘Migration and Remittance Trends 2009: A Better-Than-Expected Outcome So Far, but Significant Risks Ahead’. Migration and Development Brief 11, Migration and Remittances Team, Development Prospects Group, World Bank. http://siteresources. worldbank.org/INTPROSPECTS/Resources/334934-1110315015165/ MigrationAndDevelopmentBrief11.pdf (accessed 27 October 2011). Ratha, Dilip, Sanket Mohapatra and Zhimei Xu. 2008. ‘Outlook for Remittance Flows 2008–2010: Growth Expected to Moderate Significantly, but Flows to Remain Resilient’. Migration and Development Brief 8, Migration and

Financial Crisis in the Gulf Y 93 Remittances Team, Development Prospects Group, World Bank. http:// siteresources.worldbank.org/INTPROSPECTS/Resources/334934-1110 315015165/MigrationAndDevelopmentBrief12.pdf (accessed 25 October 2011). Shah, Nasra M. 2009. ‘Trends and Policies for Contract Worker Mobility of Asians to the Gulf Cooperation Council Countries’. Paper presented at the 2009 International Union for the Scientific Study of Population (IUSSP) ‘International Population Conference’, 27 September–2 October, Marrakech. Taattolo, Giovanna. 2006. ‘Arab Labor Migration to the GCC States’. Working paper, Jean Monnet Observatory on trans-Mediterranean relations, Project on Arab Labour Migration. United Nations (UN). 2009. International Migration 2009. New York: United Nations Department of Economic and Social Affairs, Population Division. United Nations Conference on Trade and Development (UNCTAD). 2010. World Investment Report: Transnational Corporations, Agricultural Production and Development. New York: United Nations. United Nations Development Programme (UNDP). 2009. Human Development Report 2009: Overcoming Barriers: Human Mobility and Development. New York: United Nations. Zachariah, K. C., B. A. Prakash and S. Irudaya Rajan. 2004. ‘Indian Workers in UAE: Employment, Wages and Working Conditions’, Economic and Political Weekly, 39 (22): 2227–34. Zachariah, K. C., P. R. Gopinathan Nair and S. Irudaya Rajan. 2006. Return Emigrants in Kerala: Welfare, Rehabilitation and Development. New Delhi: Manohar Publishers and Distributors. Zachariah, K. C. and S. Irudaya Rajan. 2009a. Migration and Development: The Kerala Experience. New Delhi: Daanish Publishers. ———. 2009b. ‘Migration Monitoring Study, 2008: Emigration and Remittances in the Context of Surge in Oil Prices’. Research Report, Centre for Development Studies, Thiruvananthapuram. ———. 2009c. ‘Impact of the Global Recession on Migration and Remittances in Kerala: New Evidences from the Return Migration Survey 2009’. Research Report, Centre for Development Studies, Thiruvananthapuram. http://cds. edu/download_files/wp432.pdf (accessed 31 October 2011).

6 The Dubai Model and the Impact of the Financial Crisis on South Asian Migrant Workers in the United Arab Emirates D. Narayana Vinoj Abraham Despite their basic strengths of a public-funded banking sector and huge trade surplus, the global financial crisis started affecting the Gulf Cooperation Council (GCC) region since late 2008 with a decline in Gross Domestic Product (GDP) growth and plunging of the stock markets, eventually leading to a large number of layoffs or retrenchment of the workforce. It influenced the prospects of South Asian migrant labour as workers in the region. Possibly, the flow of migration would have slowed down with other effects such as a surge in return migration, declining wages and remittances. This chapter seeks to assess the impact of the recession on key industries in the United Arab Emirates (UAE), which employs a majority of migrant workers from South Asia, and to examine the migrants’ strategies to cope with the crisis. To provide a background for the assessment, an analysis of the structure, composition and growth of the UAE economy and employment have been undertaken, based on published sources of data. The assessment of the impacts of the financial crisis on labour is based on field interviews of workers and employers in Dubai and Abu Dhabi during September 2009. This chapter is divided into seven sections, including the introduction. The second section presents the population numbers, and composition with respect to age, sex and nationality, followed by an analysis of the trends in labour and employment in the UAE over the long stretch. The fourth section presents the growth and structure of the economy, with the fifth examining the impact of the global financial crisis on growth of the UAE economy. The sixth section

Dubai Model and Impact of Financial Crisis Y 95

takes up the issue of the impact of the crisis on employment and work conditions and the last one provides insights into the immediate future.

Demographic Characteristics The UAE has approximately 4.1 million people (2006 estimate). The 2005 census showed that nearly 80 per cent of the population living in the UAE consisted of non-citizens, or expatriates. However, this concentration was mostly within the urban areas, where around 85 per cent of the population comprised non-citizens, while in the rural areas this was only 57 per cent. The UAE is a highly urbanized economy with more than 82 per cent of the population living within the urban areas. The most populous emirate in the UAE is Abu Dhabi, accounting for more than 34 per cent of the country’s population. It is closely followed by Dubai at 32 per cent. Sharjah accounts for another 20 per cent. The rest of the population is distributed among the other emirates. Over the last 30 years there has been a shift of the share of population from Abu Dhabi to all other regions, especially Dubai and Sharjah (Table 6.1). A regional classification of the UAE population shows that while in all the emirates the non-citizen population constitutes the major share of population, Dubai, Sharjah, Abu Dhabi and Ajman had more than 75 per cent of the population as expatriates. The highest was in Dubai, accounting for 90 per cent of the population. The age distribution of the population is such that the largest proportion of citizens is in the age group of less than 19 years, accounting for more than 49 per cent, while that in the working age Table 6.1: Distribution of population by Emirates, 1970–2005 Year

Abu Dhabi

1970 1975 1985 1995 2005

38.0 43.4 41.0 39.1 34.1

Dubai Sharjah Ajman 32.8 26.5 26.9 28.6 32.2

14.1 15.3 16.6 16.7 19.3

3.0 3.5 4.0 5.0 5.0

Umm Al- Ras AlQuwain Khaimah Fujairah 1.2 1.2 1.4 1.5 1.2

7.9 7.1 7.0 5.9 5.1

3.0 3.1 3.2 3.2 3.1

Total UAE 100.0 100.0 100.0 100.0 100.0

Sources: Calculated from Statistical Abstract, 1999, 2000, 2001, 2002; Ministry of Economy 2008.

96 Y D. Narayana and Vinoj Abraham

group 20–60 years is 47 per cent. As regards the non-citizens, the distribution is very different. Compared to the citizens, among the non-citizens only 17.7 per cent are below the age of 19 years, and the working age group accounts for 81 per cent of the population. Thus, the overwhelming presence of non-citizens in the working age groups is a prominent characteristic of the demography of the UAE (Table 6.2). Table 6.2: Distribution of population by age groups and nationality Distribution of population by age

Age group (years)

Citizens

Non-citizens

Total

less than 19 20 to 35 50 to 35 50 to 60 above 60 Total

49.1 27.5 15.5 4.0 3.9 100.0

17.7 43.4 33.0 5.0 0.9 100.0

23.9 40.2 29.5 4.8 1.5 100.0

% share of population by age Citizens Non-citizens 40.9 13.6 10.5 16.7 52.5 19.9

59.1 86.4 89.5 83.3 47.5 80.1

Total 100.0 100.0 100.0 100.0 100.0 100.0

Source: Census, 2005, Central Department of Statistics, Ministry of Economy, UAE.

The national identities of the expatriate population in the UAE are not statistically verified. However, anecdotal evidence suggests that about 50 per cent of the population is from South Asia and about 23 per cent is of other Arab and Iranian origins. Westerners and East Asians account for about 8 per cent of the population. The UAE has undergone a silent revolution in terms of educational achievements. From nearly half the population being illiterate in 1975, the level of illiteracy declined to 9.3 per cent in 2005. The number of people with poor levels of education has also come down substantially. For instance, the share of population that had at least primary level of schooling was only 29 per cent in 1975, which increased by leaps and bounds to 77 per cent in 2005. The proportion of people who had undergone tertiary education was less than 6 per cent in 1975, which came up to 18.4 per cent by 2005 (Table 6.3). Since more than 80 per cent of the population consists of noncitizens, much of these changes in educational achievements are however, carried over from their countries of origin. This also means that there is substantial difference in the skill profile of migrant workers in the country.

Dubai Model and Impact of Financial Crisis Y 97 Table 6.3: Distribution of population by level of education Level of education Illiterate Read and Write Primary Preparatory Secondary and Equivalent Above Secondary and Below University First University Degree and Equivalent Postgraduate Degree Not Stated Total

1975

1980

1985

1995

2005

43.7 27.1 7.9 6.6 9.2

31.8 22.5 12.2 10.2 13.6

26.4 25.1 11.6 11.9 13.8

19.8 17.1 17.4 16.0 15.9

9.3 13.9 14.6 17.8 25.9

1.1

2.6

3.2

3.3

4.0

4.0 0.3 0.2 100.0

6.5 0.6 0.0 100.0

7.2 0.7 0.0 100.0

9.5 0.7 0.2 100.0

12.8 1.5 0.1 100.0

Source: Census, 1975, 1980, 1985, 1995, 2005, Central Department of Statistics, Ministry of Economy, UAE.

Labour and Employment: Patterns and Trends The labour force participation rate (LFPR)1 in the UAE was 72.6 per cent in 2008. Of the total population, 69.7 per cent was employed and 2.9 per cent unemployed. Among the population not in the labour force, the largest share was that of housewives, accounting for 14.3 per cent of the population, followed by students at 10 per cent. However, it is to be noted that 58.2 per cent of the women in the UAE were not in the labour force (Table 6.4). In the UAE, the WPR and the LFPR among the citizens was lower than that among the non-citizens. Among the latter, the WPR was as high as 77 per cent while for the citizens it was only 39.3 per cent. Labour force participation among the non-citizens was 79 per cent while for the natives it was 46 per cent. A majority of the native population was thus not in the labour force. Nearly 24 per cent of the native population consisted of students (Table 6.4). There is considerable difference in the LFPR between the citizens and non-citizens. While among the citizens the LFPR is low for the illiterates and less educated, among the non-citizens the LFPR in 1 LFPR is defined as those in the labour force (working or employed plus unemployed) divided by population aged 15 years above. Workforce participation rate (WPR) is the ratio of working or employed to the total population.

98 Y D. Narayana and Vinoj Abraham Table 6.4: Percentage distribution of population (15 years and over) by activity status, sex and nationality, 2008 Sex Activity status Employed Unemployed — Worked at Some Point Unemployed — Never Worked Total Labour Force Student Housewife Unable to Work 65+ Years (Not Working) Others Total Not in Labour Force Total

Male

Nationality

Female

Citizen

Non-citizen

Total

87.6

36.8

39.3

77.1

69.7

0.5 1.2 89.4 7.7 0.0 0.3 0.9 1.7 10.6 100.0

0.5 4.5 41.8 14.3 40.5 0.3 1.2 1.8 58.2 100.0

1.0 5.3 45.6 23.8 19.7 1.2 3.8 5.9 54.4 100.0

0.4 1.7 79.2 6.6 13.0 0.1 0.3 0.7 20.8 100.0

0.5 2.4 72.6 10.0 14.3 0.3 1.0 1.7 27.4 100.0

Source: Labour Force Survey, 2008, Central Department of Statistics, Ministry of Economy.

these groups is very high. However, at higher levels of education there is a convergence in the LFPR between the two groups (Table 6.5). It may also be noted that the LFPR varies a lot between male and female within and between citizens and non-citizens. Among the citizens, the LFPR rises steadily with education both for male and female; but for female citizens with below secondary level of education the LFPR is less than 10 per cent, and with university education or above it rises to over 75 per cent. Among non-citizen women, the pattern is just the opposite — being high for those with very poor education and university education and dropping sharply for those with secondary level education. Despite around 80 per cent of the population being composed of non-citizens, close to 40 per cent of the total population reports of primary level of education or lower. Such a situation has arisen owing to the evolving structure of the UAE economy. The oil price hike and the massive investments led to a construction boom which attracted a large inflow of unskilled labourers from South Asia who are poorly educated. Although the largest share of GDP in the UAE is contributed to by the petroleum sector its share in total employment is only about 1 per cent. On the other hand, the largest share of employment in the UAE is in the construction sector, accounting for 21.5 per cent

18.7 39.3 65.0 45.3 73.5 92.9 63.0

Illiterate Read and Write Primary Preparatory Secondary University and above Total

Citizen

1.9 4.4 6.3 8.7 30.7 76.4 28.2

Female 7.6 20.7 40.2 29.4 52.9 83.5 45.6

Total 98.1 98.7 97.2 85.1 90.9 98.9 94.1

Male 53.6 74.7 56.1 37.8 37.1 52.8 47.1

Female

Non-citizen

Source: Labour Force Survey, 2008, Central Department of Statistics, Ministry of Economy.

Male

Educational status 88.0 91.8 87.9 70.6 71.3 83.3 79.2

Total 87.9 94.4 93.2 76.3 87.3 98.4 89.4

Male

28.3 62.0 43.2 28.1 35.2 57.4 41.8

Female

Total

Table 6.5: Labour force participation rate (in per cent) by nationality, sex and educational status, 2008

67.8 84.1 80.3 59.9 66.9 83.3 72.6

Total

100 Y D. Narayana and Vinoj Abraham

in 2008, followed by wholesale, retail trade and repairing services, accounting for 19.5 per cent, and manufacturing industries, accounting for 13 per cent. Other important sectors of employment include government services and domestic services (Table 6.6). Employment in the UAE economy grew by about 8 per cent every year during 2001–2008. While most sectors of the economy reported employment growth of between 7–9 per cent, crude oil, electricity and gas, and mining and quarrying reported fairly low rates of growth of employment. At the other end, employment was growing at over 10 per cent per annum in construction, real estate, and financial corporations sector. Table 6.6: Industrial classification of workers in the UAE Percentage share of employment Industrial sector Agriculture, Livestock and Fishing Crude Oil and Natural Gas Quarrying Manufacturing Industries Electricity, Gas and Water Construction Wholesale Retail Trade and Repairing Services Restaurants and Hotels Transport, Storage and Communication Real Estate and Business Services Social and Personal Services Financial Corporations Sector Government Services Sector Domestic Services of Households Total

Employment growth (%), 2001–2008

2001

2004

2007

2008

8.0 1.3 0.2 12.8 1.6 15.8

6.9 1.2 0.2 13.0 1.2 20.2

5.0 1.1 0.2 13.0 1.2 20.6

4.7 1.1 0.2 13.0 1.2 21.5

0.09 4.83 7.94 8.27 3.09 12.83

19.2 4.4

19.5 4.5

20.0 4.2

19.5 4.3

8.19 7.78

6.6

6.0

6.2

6.3

7.27

2.9 4.7 1.3 11.1

3.0 4.3 1.1 10.8

3.3 4.5 1.4 10.8

3.3 4.6 1.5 10.8

10.07 7.64 10.11 7.56

10.0 100.0

8.1 100.0

8.4 100.0

8.1 100.0

4.77 8.01

Source: Calculated from Ministry of Economy 2008.

Annual earnings in nominal terms grew by about 7 per cent during 2001–08. Most of the sectors recorded 6–8 per cent growth annually; the exceptions were agriculture, construction, and restaurants and hotels, with 1–5 per cent growth in earnings. But there were

Dubai Model and Impact of Financial Crisis Y 101

four sectors seeing over 12 per cent increase in earnings per annum, namely crude oil, mining and quarrying, electricity and gas, and the financial corporations sector (Table 6.7). Table 6.7: Average annual earnings per worker in the UAE and simple growth rate In 1000 UAE Dirhams Industrial sector

2001

Non-financial Corporations Sector 28.6 Agriculture, Livestock and Fishing 16.4 Mining and Quarrying 73.4 Crude Oil and Natural Gas 81.1 Quarrying 27.0 Manufacturing Industries 22.7 Electricity, Gas and Water 42.5 Construction 31.6 Wholesale, Retail Trade and Repairing Services 24.9 Restaurants and Hotels 22.7 Transport, Storage and Communication 52.9 Real Estate and Business Services 23.1 Social and Personal Services 25.9 Financial Corporations Sector 116.1 Government Services Sector 108.5 Domestic Services of Households 10.1 Total 36.8

Average annual growth rate (%) 2001–2008

2005

2007

2008

29.3

39.1

43.9

6.65

17.4 78.3

18.2 128.1

18.4 143.5

1.56 11.95

88.8 27.1 22.2 63.8 30.2

146.3 37.4 29.8 81.7 38.1

164.2 42.2 34.5 88.2 42.3

12.81 7.07 6.53 13.43 4.26

25.9 23.1

34.5 28.9

40.5 32.7

7.80 5.48

53.3

81.1

85.0

7.58

24.1

31.4

35.5

6.67

26.5

34.6

38.2

5.91

131.4

216.6

242.4

13.61

123.6

156.5

161.1

6.05

11.4 38.8

14.1 52.3

15.0 57.1

6.12 6.91

Source: Calculated from Ministry of Economy 2008.

Turning to the relationship between employment growth and growth in earnings, most of the sectors showed comparable rates of growth in both. But the sector which showed the highest rate of growth in employment, namely construction, showed the lowest increase in nominal earnings. The UAE reported of over 7 per cent rate of inflation on average since 2002 and the wages of construction

102 Y D. Narayana and Vinoj Abraham

workers in real terms would have declined by about 3 per cent every year during this period. The occupational structure of the UAE shows that 51.6 per cent of the workers were engaged as production, transport, equipment operators and labourers. Sales and service workers were the second highest in proportion, accounting for 18.3 per cent. While clerks and related workers who accounted for 10.5 per cent of the total workforce in 1975 declined to 3.9 per cent in 2005, the share of highend occupations such as professional, administrative and managerial workers increased from 9.5 per cent in 1975 to 21.6 per cent in 2005 (Table 6.8). Table 6.8: Occupational structure of the workforce in the UAE Percentage share of occupation Occupation

1975

1985

1995

2005

Professional, Technical and Related Administrative and Managerial Clerks and Related Sales and Service Agriculture and Animal Husbandry Production, Transport, Equipment Operators Occupations not Adequately Defined Not Stated Unemployed, Never Worked Before Total

7.5 2.0 10.5 21.8 4.6 52.2 0.0 0.1 1.2 100.0

11.2 2.1 11.9 28.4 7.2 38.7 0.0 0.1 0.5 100.0

5.7 12.4 5.3 24.9 21.0 43.1 0.3 0.0 0.9 100.0

8.1 13.5 3.9 18.3 16.0 51.6 0.8 0.0 2.0 100.0

Source: Ministry of Economy 2008.

Growth and Structure of the Economy The Dubai model demonstrates the success of harnessing locational and comparative advantage through state-led capitalism. Dubai has been developed with remarkable success as a hub and critical channel to facilitate commercial interaction between the East and the West. The main drivers of the model are service-oriented industrialization and state-of-the-art infrastructure. The model has helped the UAE to diversify from a largely oil economy to a predominantly non-oil economy. The share of the non-oil GDP in the total GDP has been above 60 per cent and continues to remain at that level despite the high oil prices in recent years (Table 6.9). Led by such diversification the economy has been growing at an average 8 per cent per annum during 2000–2008.

Dubai Model and Impact of Financial Crisis Y 103 Table 6.9: The UAE’s oil and non-oil GDP, 1993–2008 (in billion Dirham, in current prices) Year

Oil GDP

Non-oil GDP

GDP

Share of non-oil GDP (%)

1993 1994 2000 2003 2005 2006 2007

47.3 44.6 86.7 109.9 185.1 224.1 261.8

85.9 97.4 171.3 229.6 328.0 440.5 467.9

133.2 141.9 258.0 321.8 513.1 624.6 729.7

64.49 68.64 66.40 71.34 63.93 64.12 64.12

Source: Pradhan (2009).

The largest share of GDP in the economy is generated from petroleum and its various products. More than 44 per cent of the GDP came from crude oil and natural gas extraction in 2008. Also, the share of this component in GDP has increased from 32 per cent in 2001 to 44.5 per cent in 2008. Manufacturing is the second most important sector of the UAE economy, accounting for around 13 per cent of the GDP. Other prominent sectors include construction, services such as trading, real estate, financial corporations and government services (Table 6.10). Table 6.10:

Industrial composition of the economy % Distribution of GDP (at constant prices)

Industrial sector

2001

Agriculture, Livestock and Fishing 3.31 Crude Oil and Natural Gas 32.41 Quarrying 0.26 Manufacturing Industries 13.11 Electricity, Gas and Water 1.81 Construction 6.43 Wholesale Retail Trade and Repairing 8.37 Restaurants and Hotels 1.95 Transport, Storage and Communication 6.84 Real Estate and business Services 7.24 Social and Personal Services 1.48 Financial Corporations Sector 6.23 Government Services Sector 9.85 Domestic Services of Households 0.71 Total 100.00 Source: Calculated from Ministry of Economy 2008.

2005

2008

2.10 38.88 0.20 13.80 1.60 6.36 8.53 1.63 5.77 6.81 1.55 4.97 7.32 0.49 100.00

1.52 44.56 0.19 12.29 1.21 7.40 7.48 1.46 5.21 5.95 1.30 4.61 6.33 0.48 100.00

104 Y D. Narayana and Vinoj Abraham

The macroeconomic indicators show that the savings rate in the UAE is high, between 31 and 41 per cent of the GDP since 2001 (Table 6.11). There also seems to be some increase in the savings– GDP ratio from a low 27.6 per cent in 2002 to 38.4 per cent in 2008. However, the investment rate is much lower than the savings rate. The Gross Fixed Capital Formation (GFCF) as a ratio of GDP remained somewhere around 20 per cent. There was in fact some decline in the GFCF–GDP ratio from 23.7 per cent in 2001 to 21.5 per cent in 2008. This widening gap between savings and investment may be a worrying factor for the economy, especially since this is the trend at the eve of the financial crisis. The declining GFCF–GDP ratio in the face of a rising savings–GDP ratio may be due to an active policy of investment in the rest of the world. Table 6.11:

Some macroeconomic indicators (at current prices)

Savings–GDP Ratio GFCF–GDP Ratio Exports–GDP Ratio Imports–GDP Ratio Current Account

2001

2002

2003

2004

2005

2006

2007

2008

38.8 23.7 73.4 60.7 15.1

27.6 22.9 73.2 64.4 4.7

31.4 22.7 79.4 65.8 8.7

32.9 21.0 89.0 74.3 11.9

32.9 19.1 88.5 68.0 13.8

37.2 19.2 87.0 64.4 18.0

40.9 19.6 90.4 80.5 21.3

38.4 21.5 97.8 86.4 16.9

Source: Calculated from Ministry of Economy 2008.

The UAE is a highly open economy with almost the entire income being generated through trade, either in oil or in non-oil products, or in services. The Exports–GDP ratio was as high as 97.8 per cent in 2008 while the Imports–GDP ratio was 86.4 per cent (Figure 6.1). Both the ratios have been increasing since 2001 showing that the trade dependence of the economy had been increasing in this decade. The trading ties of the UAE were strongest with non-Arab Asian economies, especially India and China. Nearly 48 per cent of the international trade volume of the UAE was with non-Arab Asian economies (Table 6.12). The next important partners were the European countries accounting for 25.35 per cent of the trade. In the UAE, re-export is the most important form of trade activity, especially from the Dubai ports. It was in fact these re-export markets that made the non-Arab and European countries important trading partners for the UAE. As already indicated, the growth of the economy of the UAE in the recent period is led by large investments in construction, real

Dubai Model and Impact of Financial Crisis Y 105 Figure 6.1:

Trade intensity of the economy, UAE

Source: Calculated from Ministry of Economy 2008.

Table 6.12: Imports, non-oil exports and re-exports of the UAE by groups of countries, 2008 (percentage of value) % Share of groups of countries Groups of countries

Imports

GCC Other Arab China India Rest of Non-Arab Asian Non-Arab African European American Oceanic Others Total

4.54 1.44 11.26 10.95 24.80 3.35 30.33 10.07 2.30 0.98 100.00

Non-oil exports 27.15 11.51 1.62 29.68 11.50 3.20 9.85 1.49 0.75 3.30 100.00

Re-exports 9.63 12.50 1.98 23.37 28.50 5.70 13.79 2.76 0.18 1.62 100.00

Total trade 7.32 4.49 8.61 14.95 24.50 3.82 25.35 7.91 1.75 1.29 100.00

Source: Calculated from Statistic Abstracts, Ministry of Economy, UAE.

estate, and financial corporations sectors. The pace of growth witnessed in the real estate sector over 2007–2008 was remarkable. The cost of residential properties in Abu Dhabi went up by 53 per cent,

106 Y D. Narayana and Vinoj Abraham

and the value of land more than doubled between 2005 and 2007. Growth in Dubai has been greater, a sense of which may be obtained from Figures 6.2 and 6.3. Residential rents have risen sharply making it unaffordable to even service professionals, leading many of them to relocate to Sharjah, Ajman and Ras al Khaimah. The spread effect has led to all round real estate booms. Figure 6.2 High-end residential housing prices — US$/m²

Source: Colliers International.

Figure 6.3: Office space under construction (million ft²)

Source: Colliers International.

Dubai Model and Impact of Financial Crisis Y 107

The rental price of property has been driven by a shortage of supply. The imperfections of the realty market, however, have led to a situation where commercial developers focus on high-end properties. The high-end property segment has attracted foreign investment following Dubai’s relaxation of restrictions on foreign ownership in 2002. Restrictions on foreign ownership are more stringent in Abu Dhabi and other emirates. The rising house prices in the US and Europe also made Dubai an attractive destination for foreign investors. A large proportion of demand for property in Dubai comes from Russia, India, Pakistan and Iran. Further, a large proportion of the purchases were speculative, off-plan funded by mortgage lending. New regulations introduced in Dubai on mortgage lending and re-sale of property is expected to moderate the trend. Foreign Direct Investment (FDI) inflows into West Asia increased since 2003 for the sixth consecutive year in 2008. The increase was largely due to a significant rise of inflows to Saudi Arabia, Turkey and the UAE (Figure 6.4) and was driven largely by real estate, petrochemicals, refining, construction and trade. Among the other countries, Qatar saw a sizeable increase in inflows into Liquefied natural gas (LNG), power and water. It rose slightly in Bahrain, and Figure 6.4: Top five recipients of FDI inflows in West Asia,a 2007– 2008 (in US$ billions)

Source: UNCTAD (2009). Note: a Ranked by the magnitude of 2008 FDI inflows.

108 Y D. Narayana and Vinoj Abraham

fell in Kuwait and Oman. FDI in services, such as telecom, banking, power, and water has become more prominent in recent years. The large investments in construction, real estate and manufacturing were attracting 200,000 workers a year to the UAE during 2000–2008. About 75 per cent of the UAE labour force was from Asia and Indian workers constituted 42.5 per cent of the population. The Indian embassy of the UAE estimates that ‘50 percent of the 1.8 million Indian expatriates in the UAE are unskilled workers, 25 per cent semi-skilled and 25 percent skilled professionals’ (Woertz et al. 2008: 20). As the unskilled and a large proportion of the semi-skilled workers are housed in labour camps and their rent, water and electricity charges are borne by their employers, the major outgo from their wages is the expenses on food. The food expense as a percentage of disposable income does not show much variation across the GCC economies. Only less than a quarter of the population spends less than 10 per cent of the income on food, but for over a quarter of the population food expenses eat up over one-fifth of the income (Table 6.13). In the UAE, the rent component is also a fairly large outgo — 21–50 per cent — for some segments of workers. Food inflation, which was non-existent in the GCC countries until 2002, developed into a serious problem in the Gulf after 2005. Table 6.13: Distribution of population by percentage of disposable income spent on food Percentage of population

Percentage spent on food

Kuwait

Qatar

Saudi Arabia

UAE

Bahrain

Oman

Less than 10 10–20 Above 20 Don’t know

14 46 29 11

24 43 18 15

22 38 26 13

24 44 20 13

22 39 23 16

13 40 35 10

Source: Woertz et al. (2008).

The UAE and Qatar were particularly affected (Table 6.14). A major part of the inflation could be attributed to the global rise in food prices. The Food Price Index of the FAO increased by 57 per cent between March 2007 and March 2008. This was on top of an increase of 23 per cent from 2006. The GCC countries responded to the unprecedented inflation by rising salaries. In the UAE, the governments of Abu Dhabi, Dubai and Sharjah announced a pay hike of

Dubai Model and Impact of Financial Crisis Y 109 Table 6.14:

Inflation in GCC countries, 2002–2007 (in per cent)

Country

2002

2003

2004

2005

2006

2007

Kuwait Qatar Saudi Arabia UAE Bahrain Oman World

0.5 0.2 0.2 2.9 –0.5 –0.7 3.7

1.2 2.2 0.6 3.1 1.7 –0.3 3.2

1.3 6.8 0.3 5.1 2.3 0.9 3.9

4.3 8.8 0.7 7.1 2.6 1.9 3.6

3.3 11.8 2.2 11.3 2.1 3.1 3.5

6.6 12.8 6.5 11.6 8.4 5.3 4.2

Source: Woertz et al. (2008).

70 per cent to all public sector workers. It is not clear to what extent these measures protected the expatriate workers.

The Impact of the Crisis on Growth The projections made by the International Monetary Fund (IMF) suggest that after the fairly high growth rates of GDP during 2003– 2008, ranging between 6 and 12 per cent, growth decreased to the lowest rate recorded after 1993 to –3.2 per cent in 2009. But unlike the earlier fall in growth in 1993, this time around there is not going to be an early rebound. The growth rate is predicted to remain around 3.5 per cent between 2011 and 2014 (Table 6.15). Even by 2016 the growth rate is projected to reach only 4.2 per cent. Thus, though there was only some marginal shrinking of the UAE economy in 2009, a complete recovery from the crisis may take more than five years. During this period it is also expected that the per capita income may decline. The current account balance of payments of the UAE had been in surplus since the 1990s. The current account balance, which was projected to be negative for the years 2009 and 2010, turned out to be 3.0 and 7.7 per cent of the GDP as per the World Economic Outlook (IMF 2009). This has been one of the strengths of the country. The surplus oil revenues of the GCC countries that flew into US Treasury Bills since the 1970s helped fund expansions in the 1980s and 1990s when oil prices fell. In the late 1990s, Sovereign Wealth Funds (SWFs) started investing in riskier assets abroad, such as stocks and real estate. The rising oil prices in the 2000s saw this trend gain in strength. Recent years saw them investing in industries such as energy, automotives, aerospace, real estate and technology that benefit them. The UAE had taken the lead in such investments.

110 Y D. Narayana and Vinoj Abraham Table 6.15:

GDP and current account balance of the UAE

Year

GDP constant prices (Annual per cent change)

GDP per capita in Dirhams (constant prices)

1990 1993 1996 1999 2002 2005 2008 2009 2010 2011 2012 2013

23.56 –2.21 5.38 3.14 2.65 8.19 (8.6) 7.41 (5.3) –0.60 (–3.2) 1.55 (3.2) 3.29 (3.3) 4.65 (3.8) 5.09 (4.2a)

91,331.80 82,761.63 84,307.02 75,685.93 80,413.26 87,089.14 93,784.87 90,503.91 89,232.87 89,487.71 90,919.00 92,765.83

Current account balance (per cent of GDP) 22.07 8.30 9.05 1.63 4.95 17.97 15.80 (7.4) –5.59 (3.0) –1.05 (7.7) 3.06 (10.4) 5.23 (10.5) 6.89 (11.2a)

Source: IMF World Economic Outlook (WEO) Database, April 2009. Note: Figures within parentheses are from IMF WEO Database, April 2011. After 2010 the data is estimated. a For 2016.

The recent collapse of the real estate and equity markets generated large losses for SWFs (Table 6.16). Turmoil in the financial markets affected global FDI flow in 2008 and 2009, which declined by 14 per cent in 2008, and the rate of decline accelerated in the first half of 2009. While the bulk of the decline has been in the developed countries, the developing and transition economies saw an increase in 2008. However, FDI inflows have begun declining in developing and transition economies as well since late 2008. Mergers and Acquisitions fell in developed and developing countries in the first half of 2009. Greenfield investments began to feel the impact in the fourth quarter of 2008; there have been announcements of cancellation or postponement of many projects. The steadily worsening outlook for the world economy since the third quarter of 2008 has hit the initiation of development projects across the region. The number of international banks willing to lend to projects in GCC countries shrunk sharply from 45 in 2006 to 12 at the end of 2008. Outward FDI activities have become part of the diversification policy of GCC countries away from oil- and gasbased economies. The financial crisis and the significant loss have made them averse to risk and they are reducing the purchase of foreign assets.

Source: UNCTAD (2009).

533 262 65 385 16 1282

Value (as on December 2007) –183 –94 –27 –46 0 –350

59 57 28 162 –33 273

Net inflows

Changes in value Capital gain/loss

Losses of sovereign wealth funds of the Gulf

Abu Dhabi Investment Authority (ADIA), Abu Dhabi Investment Council (ADIC) Qatar Investment Authority (QIA) Kuwait Investment Authority (KIA) Saudi Arabian Monetary Agency (SAMA) Other GCC GCC Total

Agency

Table 6.16:

328 228 58 501 84 1200

Value as on December 2008

–40 –36 –41 –12 0 –27

Gain/loss on December 2007 portfolio (%)

112 Y D. Narayana and Vinoj Abraham

While telecom, banking, power, water and real estate attracted foreign investment, the ongoing crisis dried up credit in a number of infrastructure, residential, commercial and tourism related real estate projects. The lack of liquidity forced developers to either cancel or suspend projects. The private sector played an important role in the emergence of West Asia as the world’s largest market in project finance, surpassing Western Europe and North America. The deepening global financial crisis dried up this source of project finance affecting project prospects forcing governments to increase direct public funding of projects. The drying up of international debt flow has led to relaxing of the tight monetary policy to fund infrastructure projects. The UAE, along with other countries of the GCC, has pegged its exchange rate (which implies money supply changes with the change in Net Foreign Assets of the Central Bank) to the US dollar (US$). As oil prices began rising money supply too began surging. All this has changed following the crisis and the fall in the oil price. Liquidity became the number one concern of the UAE Central Bank as foreign finance dried up. It was feared that the lack of available finance would put a brake on economic growth. The Central Bank announced infusion of AED 50 billion into local financial markets in September 2008. As the eligibility criteria and cost of the Central Bank credit was not favourable to the commercial banks there were no takers for this facility. However, on 8 October, the UAE slashed its repo rate to 1.5 per cent and the premium to 150 bps2 (at a lending rate of 3 per cent).

The Effect of the Crisis on the Workers For the purpose of eliciting information on the impact of the crisis on the expatriate workers, a field survey was conducted in the UAE. The team members were D. Narayana, Vinoj Abraham, S. Irudaya Rajan, and K. C. Zachariah. The dates of visit were from 3–9 October 2009. The purpose of the visit was to conduct interviews with stakeholders such as employers, migrant employees, government agencies, civil society organizations, researchers, journalists, etc. and to conduct a questionnaire-based sample survey among migrant workers in the UAE. Based on the background study, the following sectors 2

Base points.

Dubai Model and Impact of Financial Crisis Y 113

were chosen for in-depth analysis: manufacturing, construction, trade and repair services, transport, storage and communication, and finance and investment. The regions covered were Dubai and Abu Dhabi. The team visited two labour camps and conducted interviews. Interviews were held with middle-level workers in marketing and sales sectors and with workers at high-end executive positions as well. We also spoke to two labour recruitment agents, and conducted telephonic interviews with journalists. Discussions were held with the labour attaché at the Indian Consulate at Dubai and that at the Indian Embassy in Abu Dhabi. Interviews of investment bankers, construction company managers and exporters, and interaction with two researchers in Dubai elicited a lot of information on the issue.

Profile of Workers Expatriate workers in Dubai belong mainly to the unskilled and skilled category. Construction workers form the major share of skilled and semi-skilled workers. A visit to the construction worker camps revealed that most workers were from Kerala, but the workers opined that the profile was changing with more workers coming from Gujarat, Rajasthan, Andhra Pradesh, and Punjab. There were also many coming from Bangladesh and Pakistan. Most of the workers resided in labour camps constructed at industrial regions of Jabel Ali, Al Quoz and Sonapur. The labour camps are a way of developing commercial activity in a region as well. First, labour camps are built in remote regions. New commercial centres would come up to cater to the workers. This would attract other services in the region. This in turn would attract greater domestic activity and convert that region into the urban continuum of Dubai. The typical construction worker is provided with accommodation in the labour camps with free electricity and water. In some well-maintained labour camps, food and medical facility is provided as well. The contracts are usually for a period of three years. At the time of joining air travel and visa expenses are paid for by the sponsoring company. Some companies provide health insurance, gratuity and tickets to travel to their home country once every two years. The impression based on the fieldwork is that salary package for construction workers would be in the range of AED 650–1,000 per month. The actual pay would be inversely related to the other

114 Y D. Narayana and Vinoj Abraham

benefits and perks, so that the actual unit labour cost across companies may be more or less the same. The work time is around nine hours, with the possibility of getting over time for another hour at the rate of one and a quarter hours’ pay. The work week consists of six days. Earlier the government regulations on labour were lax which meant that the workers used to work for more than nine hours and at times on all seven days. But now the government is taking a more stringent view of labour conditions at the work sites and in labour camps. Workers lamented that the salary was not very attractive as compared to that in Kerala, their home state, but it was the job security for longer periods that attracted workers to Dubai. Moreover, the savings rate would be higher in Dubai since there are not many avenues to spend.

Pre-Crisis Economic Boom and Workers Dubai was traditionally a trading centre. The whole of the UAE, and especially Dubai, was experiencing an economic boom during the period 2003–2008. This came about because of the following reasons: (a) the oil price hike, (b) the strict investment regulations in US and Europe following the 9/11 terror attack which restricted investment from the Gulf in the West, and (c) the politically neutral position of Dubai in a region where there was lack of trust in their own ruling regimes. Dubai was being projected as the global capital of tourism and entertainment. The investors pumped speculative investment into real estate. The focus shifted from trading to real estate. This in turn propped up the growth of the derivative industries, both in the manufacturing and services sector. A housing bubble developed and real estate-based speculative investment increased. In 2005, nearly 150,000 new residences were taken up. In 2008, it had increased to 3 million residences. Super normal profits were the norm for many industries during the period. This boom led to a massive rise in salaries for some sections of workers, and along with it the rentals in Dubai. Oil companies were giving incentives to their staff to get more workers. Salaries were unrealistically high. For instance, one respondent said that his salary as a manager in a Dubai firm in 2003 was AED 10,000 per month, which increased to AED 20,000 in 2007, and by 2008 it was a whopping AED 30,000. Credit was easily available, though the cost of credit was high. But availing it was not an important issue as

Dubai Model and Impact of Financial Crisis Y 115

workers has a lot of disposable income. Many workers pushed up their spending levels, with a high share of loan component in their expenditure. The lifestyles among workers changed. However, another problem was in the offing in some sectors. While the sectors that were not derivates of the construction and real estate ones did not grow to the extent of the latter, the workers in these slow-growing sectors were already in trouble as the rental cost reached up to 50 per cent of their total earnings. Many workers in fact relocated their families back to their home country during the boom.

The Crisis-affected Sectors and Workers The crisis had affected a large number of workers in some way or the other. A research consultant named Dr Janardhan, based in the UAE, estimates that nearly 40 per cent of the workers have been affected. The effect varied across different industries, occupations, and skill levels, depending upon the number of years of experience. The strategies adopted by the workers to cope with the crisis varied. The impression from the field visit was that, belying the general expectation, expatriate workers did not leave the country in large numbers. The effect of the crisis can be seen mainly in the construction sector. The main consequence was that construction was stopped or put on hold wherever work has not reached an advanced stage. But wherever it had progressed to an advanced level, the work continued. The major example is that of Nakheel, which has put on hold its dream project of a one-mile high building, taller than the currently tallest one, Burj Khalifa in Dubai. They had also conceived of plans to do construction work similar to Palm Jumeira, to be called Palm Deira and Palm Jabel Ali. But all of these have been put on hold. In this firm, not only construction workers but even managers were asked to leave. The firm issued certificates of job assurance in other companies to the retrenched workers. Interviews with construction contractors gave the impression that construction activities declined to about half of what it was in 2008. Workers in some construction companies such as the Arabtec Construction Company had to leave the country because the company was severely affected by the crisis. In most cases the construction companies would provide the construction workers with basic minimum facilities including food and accommodation. Hence, for low-skilled workers in the construction

116 Y D. Narayana and Vinoj Abraham

sector, the effect was not as severe. However, when the company faced sickness or closure these workers had no option of staying back in the country. The decline in the construction sector has affected the workers’ salary, and their job security. As per the industry view, nearly 40 per cent of the population was affected by the crisis. About 10 per cent of the workers have suffered job losses. Almost 10 per cent have gone on leave and another 5–10 per cent are still sticking on to jobs at wages much lower than they were receiving earlier. Many have also moved to Abu Dhabi, Qatar and Saudi Arabia in search of employment. During the interviews, it was discovered that while many workers were not sent away by the company, all incentives and perks such as increments, travel allowances, insurances, etc. had been stopped. This has led to a situation wherein many workers have left of their own accord due to low salary. Though the brunt of the crisis is falling on the migrant workers it is not that the local workers are completely excluded from its effects. According to sources, at least 2 per cent of the nationals have been affected by the crisis. For instance, Al Futtaim, a private construction company in the UAE, sacked some nationals earning high salaries, which created a controversy. However, a government rule has come about since the incident stating that unless there is a violation of law nationals cannot be sacked. The foreign banks have been severely affected by the crisis, whereas local and national banks have not been influenced to that extent. One of the prominent foreign banks in the UAE, Citibank, which used to employ about 3,500–4,000 Indian workers in Dubai has retrenched almost all of them since 2009. Export and re-export too have been hit badly. A prominent electronic goods exporter from Dubai says that exports to Central and Latin America have come down by at least 80 per cent, while exports to Africa have reduced by 30–40 per cent in the last one year. With Dubai being traditionally a re-exporting port, the crisis and the decline in exports have affected cargo movement and hence workers in the sector as well. Meetings with the cargo industry workers revealed that people are losing jobs. There are no replacements for workers who quit and the workload increases. Another sector that felt a severe impact included the MICE (Meetings, Incentives, Conventions and Exhibitions) industries.

Dubai Model and Impact of Financial Crisis Y 117

The demand in this sector being highly elastic, the crisis led to a fast decline. As per informed guesses, most hotels have sent at least 40 per cent of their expatriate workers back. The insurance sector is also suffering. As for the workers, since they worked on commission basis rather than drawing fixed salaries, the decline in the industry affected the workers directly. Real estate agents were also affected badly with loss of jobs, salary cuts, and shifted to commission-based work. The biggest source of revenue for the media was property advertisements. This reduced drastically in 2008 and 2009. Even within government services, such as the Dubai municipality, the livelihood security of the workers is being jeopardized. The Dubai municipality used to provide accommodation to its workers earlier. But since the crisis they have issued notice to the workers that accommodation would be provided only till December 2009, after which the workers are liable to pay rents or else will have to vacate the accommodation.

The Effect of the Crisis on Workers The most rampant effect of the crisis was cuts in salary, stoppage of increments and discontinuance of benefits and perks. Almost all the workers interviewed reported one of the above events in their work life. This phenomenon was worse in case of middle-income group workers who were engaged in semi-skilled and skilled jobs. Most employees depend on their own arrangements for accommodation and food. Salary cuts ranged from 10–30 per cent of the levels in 2007 just before the crisis. However, it is to be kept in mind that prior to the crisis there had been a large rise in the salaries and wages due to the boom in the sector. Hence, the decline in wages is a relative one when compared to the boom in 2007. Yet the effect of this cut cannot be seen as trivial. This is so because the boom had also lifted the expectations of most workers leading to a high-spending lifestyle. Many workers stated that the rise in wages during the boom period led to their taking large loans for various purposes. The crisis however led to a situation wherein most workers were in debt traps. And precisely, these traps acted as a lock-in for workers, who could not leave the country even if they wanted to. At the level of high-skilled workers and managerial class the effect had been curiously positive. Most positions at the managerial levels were headed by European expatriates who claimed large salary packages. Many companies decided to lay off their European workers

118 Y D. Narayana and Vinoj Abraham

and placed south Asian workers who needed to be paid much less. However, results from a survey among workers showed that almost all workers perceived the crisis as a threat to their job in the UAE.

Workers’ Coping Strategies One of the options given commonly to the workers was to go on an extended leave, often christened as open leave. Workers were relieved from work and asked to take a long leave until the company was ready to take them back after the crisis. Many workers stated that often this strategy was used by companies to retrench workers with higher wages and recruit workers with lower wages. Another option was to work on half salary for a period of time, usually three to four months, till the company moved out of the red. Industry opinion tells us that salary cut has been to the tune of 20–30 per cent. This led to deterioration of conditions of living for many workers. Their families had to be repatriated and many workers reported of moving from houses with double rooms to single rooms or dormitories. The Emirate foundation, with support from the UAE government, had been trying to mitigate the effect of the crisis through various interventions such as providing free accommodation to workers who lost jobs during the crisis. Also, the UAE government has extended the grace period to be in the UAE without employment up to six months.

Into the Future The population of the UAE is expected to increase by about 80 per cent between 2007 and 2018. It is expected to grow to 5.62 million by 2011, 6.62 million by 2014 and 8.05 million by 2018. Both Abu Dhabi and Dubai foresee a doubling of their populations during this period. Abu Dhabi is planning extensive public works, housing and public transport, and Dubai plans to introduce a state-wide health care system for all residents by 2012. The growth of population and the developments in the oil sector (oil prices between US$70 and US$100 per barrel over 2009–2015) would lead to larger import of food and consumer goods and decline in current account balance. However, the economy is expected to diversify into manufacturing, tourism, trade, financial services and logistics. Abu Dhabi’s plans to build a new port and expand airport facilities are all pointers to it.

√ √

√ √ √ √ √

Kyrgyz Republic √ √ √ √ √ √ √

Macedonia, FYR

√ √

√ √ √ √

Belarus

√

√ √

UAE

√

√ √ √ √ √

√ √

Columbia

√

√ √ √

√ √

Tajikistan

√

√

√ √

Egypt, Arab Republic

Source: World Bank and IFC (2009). Note: Economies are ranked on the number and impact of reforms. First, Doing Business (ibid.) selects the economies that implemented reforms making it easier to do business in three or more of the Doing Business topics. Second, it ranks these economies on the increase in rank on the ease of doing business from the previous year. The larger the improvement, the higher the ranking as a reformer.

√

√

√ √ √ √

√

Rwanda

Ease of doing business

Starting a business Dealing with construction permits Employing workers Registering property Getting credit Protecting investors Paying taxes Trading across borders Enforcing contracts Closing a business

Economy

Table 6.17:

120 Y D. Narayana and Vinoj Abraham

Crisis is a period of opportunity and governments implemented regulatory reforms, making it easier to do business. Starting and operating a business, strengthening property rights, improving the efficiency of commercial dispute resolution and bankruptcy procedures has been the focus of many governments. The UAE has also been at the forefront of reforms. As regards the rankings of the economies in World Bank’s Doing Business (World Bank and IFC 2009), the GCC countries are all on top of the list and improving. The 2009 and 2010 rankings of the GCC economies are: Saudi Arabia (15, 13), Bahrain (18, 20), United Arab Emirates (47, 33), Qatar (37, 39), Kuwait (52, 61), and Oman (60, 65). In the list of 183 economies, Singapore and Hong Kong continue to be in the first and third positions respectively by continually pushing forward and staying proactive. The UAE, which was almost 50 notches below Singapore, has improved by 15 ranks. The strength of the UAE lies in the quick and hassle-free export and import procedures and lower shipping costs. The UAE takes on average less than half the time than other countries to complete importing and exporting — nine and eight days respectively. Cost of shipping a container — US$593 per export container and US$579 per import container — is half of the rates prevalent in Middle East and North Africa (MENA) and the OECD. The massive budget and current-account surpluses, owing to the oil price spike, provided unprecedented liquidity in financial markets, in turn boosting investment and growth. The oil boom has corresponded to an overall expansion of employment opportunities in the region, and a decline in regional unemployment rates from 15 per cent in 2000 to 12.7 per cent in 2005. Suzanne Maloney (2009) writes: Many of the larger states appear to have placed a premium on projects with real potential to accommodate the region’s fast-growing labour market. Saudi Arabia has invested $200bn in new ‘economic cities’ intended not just to compete with traditional hubs like Dubai and Bahrain but to generate large-scale employment, as the 2,000 planned factories and 800,000 planned jobs of King Abdullah Economic City and the 1.3 m projected jobs for an agribusiness city in Hail suggest. The GCC states have approximately $1tr in infrastructure investments already in the pipeline, or as much as double that amount if all the announced projects are actually launched (ibid.: 134).

Dubai Model and Impact of Financial Crisis Y 121

Much of this investment is in power generation, water desalination, education and housing (for details of projects and projects cancellation, see Chapter 5). The recent surge in oil prices has further boosted these trends.

References International Monetary Fund (IMF). 2009. World Economic Outlook: Crisis and Recovery. Washington, DC: International Monetary Fund. Maloney, Suzanne. 2009. ‘The Gulf’s Renewed Oil Wealth: Getting it Right This Time?’, Survival, 50 (6): 129–50. Ministry of Economy. 2008. National Abstract UAE 2008. Abu Dhabi: Government of the United Arab Emirates. Pradhan, Samir. 2009. ‘Economic Recession and the Middle East’ World Trade: Recent Policy Trends and Implications’, in Simon J. Evenett (ed.), Broken Promises: a G-20 summit report by Global Trade Alert. London: Centre for Economic Policy Research. United Nations Conference on Trade and Development (UNCTAD). 2009. World Investment Report: Transnational Corporations, Agricultural Production and Development. New York: United Nations. Woertz, Eckart, Samir Ranjan Pradhan, Nermina Biberovic and Christian Koch. 2008. ‘Food Inflation in the GCC Countries’. GRC Report, Gulf Research Center. http://www.grc.ae/data/contents/uploads/food_inflation_3609.pdf (accessed 7 November 2011). World Bank, and International Finance Corporation (IFC). 2009. Doing Business 2010: Reforming through Difficult Times, pp. 1–9. Basingstoke: Palgrave Macmillan. http://www.doingbusiness.org/~/media/FPDKM/Doing%20 Business/Documents/Annual-Reports/English/DB10-FullReport.pdf (accessed 27 October 2011).

7 Global Financial Crisis and its Consequences on Migrants in Qatar Macro and Micro Perspectives Hrushikesh Mallick Udaya Shankar Mishra The impact of the global financial crisis that originated in the United States (US) housing sector is widespread with its varying impact on economies, subject to their levels of integration with the developed and other emerging economies. The Middle East economies are typical of such integration in terms of trade and investment along with the absorption of migrant workers from developing economies. Since Qatar is an oil-led economy, any sluggish demand for oil in the world economy on account of the financial crisis may have its bearing on the Qatar economy and its migrant workers. In this context, we examine the impact of the global financial crisis in Qatar. Qatar is an independent emirate with an estimated population 1,638,829 in August 2009 and a geographic area of 4,400 sq. mi (11,400 sq. km). The ethnic composition in Qatar in 2008 indicates that around 40 per cent of the population consists of Arabs, the rest are Indians (18 per cent), Pakistanis (18 per cent), Iranian (10 per cent), and others (14 per cent) (CIA 2009). This indicates that more than one-third of Qatar’s population (36 per cent) is of Indian and Pakistani origin, reaffirming the substantial South Asian presence in Qatar. The economy is one of the fastest growing ones, consistently registering the highest per capita income among the six gulf countries since 2000 to 2008. Its per capita income stood at nearly $57,935 in 2007, ranking as the fifth-highest in the world and top among the GCC, and it further indicated a record level of $70,629 in 2008. This is backed up by a higher gross domestic fixed capita formation constituting around 43 per cent of Gross Domestic Product (GDP) in 2007 that resulted in higher real growth.

Global Financial Crisis and its Consequences on Migrants in Qatar Y 123

Economy of Qatar The economy of Qatar is mainly dominated by oil and natural gas, accounting for almost 82.42 per cent in 2010 (QCB 2010). The Government’s strategy is towards diversifying the economic base beyond hydrocarbons. Such diversification includes the development of chemicals, steel, cement, and fertilizer industries, and banking. For centuries, the economic activity had been centred around camel breeding, fishing and pearl diving. In the 20th century, the economy was transformed by the discovery of oil and, later on, vast reserves of gas off the north-east coast. Qatar has the largest nonassociated gas field yet discovered, and is now a major exporter of liquefied natural gas and condensates. The State of Qatar conducts its principal oil operations through the State-owned Qatar Petroleum (QP), which manages Qatar’s oil, gas, fertilizer, petrochemicals and refining enterprises in Qatar and abroad. At the current production pace, oil reserves are expected to last more than 40 years. Qatar’s proven reserves of gas are the thirdlargest in the world, exceeding 900 trillion cubic feet (14 per cent of the world’s total proven gas reserves).1 Within Middle East, Qatar has the second highest proven gas reserves after Iran. Qatar is now the world’s largest producer of liquefied natural gas (LNG), with a capacity of more than 31 million metric tons per annum (mmtpa), and expected to reach 78.1 mmtpa LNG exports by the end of 2011 (QNB 2011). By then, Qatar will also account for one-third of the world’s LNG supply. It began exporting LNG to Japan in 1996. Further, phases of North Field gas development costing billions of dollars are in various stages of planning and development, and it has concluded agreements to export gas via pipelines to the UAE, and to Spain, Turkey, Italy, the US, France, South Korea, India, China, Taiwan, and the United Kingdom (UK) via ship. Given the ensuing global meltdown and its effect on the economies, the scaling down

1 The QP Company has initiated and developed two major Liquefied Natural Gas (LNG) projects with foreign shareholders for the purpose of utilizing the North Field gas for exports in the form of LNG. These projects are Qatargas and RasGas which currently operate LNG facilities with a combined production capacity of around 38.6 million tons per annum (mtpa). These facilities are undergoing major expansions to meet additional LNG export opportunities.

124 Y Hrushikesh Mallick and Udaya Shankar Mishra

of expansion activities would have a definite impact on the employment and migrants in Qatar.

An Assessment of the Impact of the Crisis on Various Sectors The year on year (y-o-y) variation in quarterly output in different sectors and aggregate activities would provide a clear picture regarding the impact of the financial crisis on different sectors of the economy. In oil and non-oil sectors, although the annual variation in output was positive during 2007–08, on considering the y-o-y percentage quarterly variation of output, it registered a minimal overall growth rate of 0.3 per cent output in the fourth quarter (Q4) of 2008, on comparing the growth rates in the first (Q1) and second (Q2) quarters of 2008 and then it showed negative growth rate in Q1 2009. The sectors and sub sectors such as mining and quarrying (involving the oil and gas producing sector) recorded negative growth rate in Q4 2008 and in Q1 2009. While other non-oil sectors registered a lower growth rate in Q4 2008 and in Q1 2009, the manufacturing sector saw a negative growth rate in Q1 2009 as compared to its positive growth rate in the preceding quarter. Although the construction sector started growing at a slower rate since Q2 2008, its growth has, however, unexpectedly risen in Q1 2009 by almost 1.0 per cent from the preceding quarter. Similarly, while transport and communications, finance, insurance, real estate and business services, electricity, gas and water show some sign of recovery from previous declines, trade, restaurants and hotels revealed a declining growth rate in Q1 2009. Overall, the oil sector was severely hit in Q4 2008 and Q1 2009, while the non-oil sectors showed some sign of recovery in Q1 2009. This negative growth in the oil sector could be due to the drastic fall in international oil prices, cutting down the scale of production and output in the economy. The QCB is now following a contractionary monetary policy as revealed by the negative growth rate of money supply, with the government following a contractionary fiscal policy stance as reflected in the higher budgetary surpluses in Q1 2009 and low spending. While the governments across many countries have enhanced their expenditure levels through various stimulus packages for averting the recession in view of the financial crisis, the government in Qatar has restrained its expenditures. This restraint measure may be on account of a predicted crunch in the economic activities of the domestic economy resulting in falling tax revenues and sluggish

Global Financial Crisis and its Consequences on Migrants in Qatar Y 125

world demand. The sluggish world demand has its implication for Qatar’s oil revenues in terms of keeping down the revenue receipts of the government at a lower level. The capital market indicators such as Securities Market Index and Market Capitalization also show downward trends. This disinflationary spell of the crisis might have had its impact on the migrant workers in Qatar, especially in the construction sector which is the largest sector in terms of employment, and related sectors such as oil and gas, which is the largest sector in terms of output, as oil and gas sector to a greater extent is interlinked with construction activities for extraction, storage and sale.

Liquidity Scenario There was substantial overall expansion of domestic credit and liquidity in Qatar between 2005 and 2008. The liquidity expansion — as a result of the rise in demand deposits, savings and time deposit and foreign currency deposits — led to the expansion of credit and increase in energy prices in the domestic market in 2008. Nevertheless, the growth of domestic credit and liquidity reflected in broad money supply declined by 4 per cent and 20 per cent respectively in 2008 from their previous levels during the crisis period. The fall in overall growth rate of domestic credit in 2008 could be due to the lesser growth in the allocation of credit in the merchandise sector. Further, the overall domestic credit and liquidity registered negative growth rates of 5.9 per cent 2 and 2.50 per cent respectively during Q1 2009, compared to their growth rates at Q4 2008. This could be the consequence of liquidity restraining measures adopted by the banking and financial institutions, including the QCB. However, although output growth of the construction sector in 2008 got adversely affected due to the financial crisis, the share of the land, housing and construction sector in total domestic credit has been rising. The higher spending by the State on infrastructure projects, education projects, health and social welfare projects, and various other hotels, commercial and residential construction projects have been the main reasons for the substantial growth in this sector. Bank credit allocation to this sector shows no signs of slowing down even during Q1 2009 with a share 2 This is on account of negative growth of credit in the service sector in the first quarter of 2009 (QNB 2009).

126 Y Hrushikesh Mallick and Udaya Shankar Mishra

of 23.8 per cent constituting the second highest portion of domestic credit facilities as of March 2009, next to the personal sector.

External Sector The external account of Qatar reflects that the economy always maintains a surplus in the current account as well as in the overall balance of payment (BOP). The surplus in the overall BOP is mainly due to the surpluses in the current account, as exports earnings of Qatar always exceed more than twice than that of what it pays for its imports. However, the rate of growth of exports has already stagnated on comparing the rate of growth of imports; as a result, the current account balance has come down in recent years. Further, in total export earnings, oil and gas (mineral fuels and products) constitutes the major source of total export earnings, followed by chemicals and related products. Its share was as much as 83 per cent in 2002 and went up to 90 per cent in 2006. Since the rate of growth and per capita income of Qatar attracts repatriate workers from other countries (mainly Asia) into its economy, Qatar is found to be a net remitter. It remits around 7 per cent of its GDP as worker remittances, and around 6.6 per cent of GDP as transfers. The overall BOP position shows that it had a surplus of 17.8 per cent of GDP in 2007, which indicates a healthy and cushioning external economic position of Qatar. But unfortunately the data for the entire crisis period is not available in order to have a comparative perspective. Asia is the major exporting destination for Qatar’s crude oil, with the region accounting for 90 per cent of the total oil exports in 2008. Japan, Spain, South Korea and India are the main importers of LNG from Qatargas and Rasgas. Qatar emerged as the world’s biggest LNG exporter in 2008. However, a steep decline in crude oil prices ($44 p/b) in 2009 has resulted in a downward movement in the export of crude oil. The IMF World Economic Outlook (IMF 2009) states that the steep decline in the price of oil is hitting the region hard. As external financing conditions have deteriorated and capital inflows reversed, many equity and property markets have suffered substantial losses. Despite supportive policies, growth is projected to slow down and inflation pressures to subside considerably 2011–2012. At the same time, the external and fiscal balances are set to worsen sharply, as oil-exporting countries utilize the buffers accumulated during the boom years to cushion the impact of the crisis (IMF 2009).

Global Financial Crisis and its Consequences on Migrants in Qatar Y 127

Employment Scenario The government and public sectors are the major employers in Qatar followed by mixed and private financial intermediaries. The proportion of non-Qataris engaged is very less in the government sector, followed by even lesser presence in the public sector (government company/establishments), mixed and private financial intermediaries and hotels. However, the proportion of non-Qataris engaged in the public sector showed some dramatic improvement in 2006. The proportion was 40.3 per cent in 2002 and has improved to 75.9 per cent in 2006. The mixed sector, which is a relatively less employmentabsorbing sector, shows that it consistently engages above 80 per cent non-Qataris in the total engaged population from 2002–2006. Although the highest proportion of non-Qataris is engaged in the private sector, the size of their total employment is quite negligible. The proportion of Asians in total has been increasing over the years. It has risen from 32 per cent in 2004 to 38.3 per cent in 2006. The proportion is found to be higher in the public and mixed sectors and has been rising in both as compared to the private sector, and has declined in the government sector. Since during the crisis, the private sector is likely to be more vulnerable, the Asians engaged in the mixed and private sectors are likely to be affected severely by the close-down of the companies or due to employment retrenchment strategy of firms in the private sectors during the crisis. As the migrant population constitutes a substantial proportion in the total engaged population in Qatar and Asians alone account for more than 60 per cent of the total employment among the nonnationals, it is of interest to know their presence by economic activities, and occupations across different sectors such as private, government, public and mixed. The highest percentage of nonQataris is engaged in the financial intermediation in the private sector, followed by those employed in hotels. Further, in the private sector, a sizeable percentage of non-Qataris take up clerical jobs. A good percentage of non-Qataris also work as service workers, shop and market sales workers, and a lesser proportion as legislators, senior officials, managers, technicians, and associate professionals and professionals. The economic activities with higher employment capacity have a lower average monthly salary. The average monthly wage is found to be less in the construction sector (QR 3985). The next highest

128 Y Hrushikesh Mallick and Udaya Shankar Mishra

employment-oriented activity, which is a mix of wholesale and retail trade, repair of motor vehicles, motorcycles, personal and household goods, also have a low average monthly salary of QR 3740 compared to that in construction activity and the average of all average monthly salaries in all economic activities.

Evaluation from a Stakeholder’s Perspective: A Field Experience Towards obtaining such a perspective, interviews were held with small and medium construction contracting companies, commercial bankers, migrant workers from India and Nepal, and the embassy officials (India and Nepal) in Qatar at the end of 2009. The overall perception was denial of any impact of the crisis. The small companies got severely affected and in some cases they closed their businesses, temporarily asking the workers to go home and come back after two to six months when the company started functioning again. Thus, there was reduction in investment and delay in sanctioning of new projects. However, the employment scenario remained unaffected in medium and large companies as there was no retrenchment of workforce as they were looking forward to the projects. The large companies could sustain the crisis because of economies of scale effect. Crisis, if any, was seen in the short-term, in terms of shrinking private investment and government projects but the activities were peaking again. Company staff opined that there was no significant effect of the crisis in Qatar. Only the construction sector seemed to have seen some visible impact. A few workers might have gone home after working for two to four years, to return after six months, as there was no work available. This is partly cyclic and not purely a result of the financial crisis. The movement of Indians to Qatar remains the same with returnees as well as new migrants. The net migration is positive although it is increasing at a slower pace than before. But, it will perhaps be an oversimplification if we attribute any observed change to the crisis. There is no restraint by the government on visa regime. Anyone who wishes to work here can work. Further, some of the Indian migrants staying for long have small business and occupations to fall back upon in case of loss of employment, However, the construction workers from Nepal and Philippines have borne the brunt of the crisis in multiple ways, like delayed payment

Global Financial Crisis and its Consequences on Migrants in Qatar Y 129

of salaries, and reduction in non-monetary incentives like duration of leave, overtime payments, etc. This is primarily because workers from these countries are working on low-paid contracts and also in low-skilled jobs. The minimum wage varies between QR 600 and QR 800 depending upon the negotiation between the contractors and embassies concerned. The Qatari government has taken measures of consolidation of banking sector, such as buying equities through acquisition of bad debts to prevent the liquidity crunch and to show that banks are performing well. There is no liquidity crisis and thus, the crisis has no great impact. However, the banks have lots of funds but are unable to lend to private sector because of their liquidity crunch. Thus, the government has injected funds into the banking sector resulting in liquidity but no credits are available to the private sector as banks are scared of lending to the real estate sector and small companies dealing with the residential and commercial estates. The government has been spending on education, health and infrastructure. The activities have been expanding and the economy is diversifying. The government is working towards achieving equalproportionate share of 50:50 in the composition of GDP share of oil and gas, and non-oil and gas by the year 2015 as against the current share of oil and gas, i.e., 62:38. The Qatar government has planned for major gas projects as there is recovery in the global oil and gas prices. The government may initiate major gas projects as the budget turns this year. It has put up a good growth in 2009 and is predicted to put up a strong growth in 2010. The overall situation is stable but the question is whether the confidence can be restored soon. The Qatar economy has already laid out mega projects replacing the old infrastructure, and the economy’s outlook is good.

Concluding Observation The analysis shows that the Qatar economy is predominantly an oil and gas economy as more than 60 per cent of contribution to the GDP comes from this sector. The government has recently been making efforts to diversify the economic activities from oil to service and industrial sectors. It is observed that there are some indications of the impact of the crisis at the end of 2008 as well as in the first quarter of 2009 in the Qatar economy, as reflected by the falling output in major sectors, falling exports, declining inflation, projected downward growth rate of output, and downward

130 Y Hrushikesh Mallick and Udaya Shankar Mishra

movement of financial indicators such as falling growth rate of domestic credit and money supply (M2) and stock market index. Given the falling trends of output, the construction sector and wholesale and retail trade; repair of motor vehicles, motorcycles, personal and households goods; domestic services; and manufacturing are the major employment absorbing sectors in Qatar, where there could be more concentration of repatriate workers as statistics show that government sector employs a very negligible percentage of repatriates and a slightly higher proportion of non-nationals in the lower-paid jobs such as elementary occupations, with a few in the higher paid jobs as they work as professionals. Once there was a falling trend in output, it must have affected the employment status of the non-Qataris in those sectors. In order to have a better understanding on the impact of the financial crisis, especially on migrant workers from South Asia, a small survey was conducted to cover the economic activities and occupations in different sectors where there is higher presence of South Asian migrants. The survey showed that the impact is largely restricted to the construction sector and this has produced an adverse impact on the lives of migrants in terms of long period leave as there is no work in the current employment, delays in salary release to the workers and unavailability of extra hours of work to the workers for earning additional income. It is largely the small companies and corporate involved in the construction sector business who have suffered. The large businesses could sustain the crisis because of their economies of scale. Unless the recovery process starts sooner, the big corporate is also likely to witness a collapse, which would severely affect the life of the migrants.

References Central Intelligence Agency (CIA). 2009. The World Factbook. Washington, DC: Potomac. International Monetary Fund (IMF). 2009. World Economic Outlook: Crisis and Recovery. Washington, DC: International Monetary Fund. Qatar Central Bank (QCB). 2010. The Thirty Fourth Annual Report. Doha: Financial Stability and Statistics Department, Qatar Central Bank. http://www.qcb.gov.qa/English/Publications/ReportsAndStatements/ Documents/Annual_Report_2010.pdf (accessed 7 November 2011). Qatar National Bank (QNB). 2009. Qatar Economic Review, May. Doha: Qatar National Bank. ———. 2011. Qatar Economic Review. Doha: Qatar National Bank.

8 Global Financial Crisis and the Migrant Labour Market A Study of Kuwait K. Pushpangadan M. Parameswaran The world economy is more interdependent now than ever before

due to the process of liberalization and globalization in the world economy. As a result, any economic disturbance in one country/ a group of countries is transmitted very quickly to the rest of the world economy. The current financial crisis that originated mainly from developed economies is no exception. The only caveat is that the impact is not uniform across the countries and varies according to the structure of their foreign trade, financial markets, and institutions to cope up with such crisis. In this chapter, we examine the impact of the crisis on the migrant labour market in the oil exporting economy of Kuwait. Before undertaking the analysis of the impact on migrant labour, we have provided a brief account of the performance of the economy at the aggregate and the disaggregate level. In order to place issues in the right perspective, we have examined a five-sector macroeconomic model of the economy with emphasis on the effect of oil exports on national income (Gross Domestic Product [GDP] in constant prices) and cost of living index. The macroeconomic impact is then linked to the migrant labour market for the recession in the world economy. The major difficulty in executing such an evaluation is to locate the required information for the period after the financial crisis began in 2007. Therefore our analysis is limited by the availability of data. The database consists of secondary sources from international agencies (World Bank, IMF, etc.), various agencies of the Kuwait government (Ministry of Planning and Central Bank of Kuwait, etc.). The severity of data is most felt in the empirical analysis of the

132 Y K. Pushpangadan and M. Parameswaran

labour market. Primary data collection is the only way out, which requires more time than was available with the team’s study tour that lasted a week. Therefore, the empirical part of this report is based on the field visits and our personal interaction with the stakeholders (firms, migrants and government officials) who are affected by and concerned with the crisis. We have structured the chapter in the following way. We have summarised the economic performance of the Kuwait economy in the next section. The third section undertakes a five-variable vector auto regressions (VAR) analysis for assessing the macroeconomic reality of the Kuwait economy in the wake of the present crisis. In the fourth section, we assess the impact of the crisis on the migrant labour market using the supply–demand model. The implications of the model are then subjected to empirical verification from field level observations in the fifth section. Finally, the last section contains the summary and conclusions of the study.

Performance of the Kuwait Economy The country of Kuwait, with a geographic area of 17,818 km2, is surrounded by large powerful neighbours: Saudi Arabia to the south, Iraq to the north and Iran to the east. It had a population of 2.4 million in 2004 and the per capita GDP was around US$38,000 in 2007. The International Monetary Fund (IMF) World Economic Outlook database (as given in Table 8.1) showed a negative growth rate of –22.9 per cent in 1991 reaching the lowest –0.66 per cent in 2009. But the initial period (1991–1993) refers to war years and therefore rates are outliers. The rates after 2007 are based on projections. Therefore, the actual (normal) performance of the economy is best captured by analysing the period, 1994–2007. The growth rate for the period, 1994–2007, varied from –0.77 per cent in 1999 to 6.9 per cent in 2003. The economy was subject to wide variation in growth rates and an understanding of the sectoral variations is essential for analyzing the impact of the financial crisis on migrant workers. Keeping this in view, we have examined the sectoral growth rates of the economy in Table 8.2. The differential behaviour of the growth rate also indicates that the ‘fallacy of composition’ exists in the sense that the behaviour of sectoral growth rates do not add up to the aggregate growth performance of the economy. The instability as indicated by the fluctuation in the sectoral growth rates may increase or decrease or remain the same. In order to minimize the instability in income, one strategy is

Global Financial Crisis and the Migrant Labour Market Y 133 Table 8.1: Growth rate of the economy, 1991–2009 Year

GDP growth

Year

GDP growth

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000

–22.92 17.81 12.63 3.59 0.59 0.25 1.06 1.56 –0.77 1.98

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

0.09 1.28 6.94 4.23 4.38 2.17 1.07 2.66a –0.66a –

Source: IMF (2009). a Projected growth rate. Note:

to diversify the production structure of the economy. We examine the validity of such a strategy for the Kuwait economy during the period. This requires the measurement of both instability and diversification of the economy. The economy wide instability index (ISI) is defined as the same as the average of the absolute deviations of the GDP from a linear trend for each of the seven sectors. The diversification index (DI) is defined as DI = 1 – Herfindahl index, where Herfindahl index is the sum of the squares of share of GDP in total GDP in constant prices. If the indices move in the opposite direction, then diversification reduces instability, otherwise not. This hypothesis is examined in Figure 8.1. Figure 8.1 clearly shows an inverse relationship only in the second half of the period under consideration — 1991–2007. But correlation analysis does not show any significant relationship. Therefore, it can be concluded that no policy initiatives have been undertaken on diversification during the period to reduce the instability in the economy. In such an uncertain world, it is pertinent to ask what has happened to the net factor income earned abroad during prefinancial crisis. This is examined in Figure 8.2 by looking at the ratio of Gross National Product (GNP) to GDP. The ratio declined from a peak of 150 in 1991 to 111 in 2004. But we do not have any up-to-date information on the ratio. It is interesting to observe that the ratio shows a secular decline in the postwar period suggesting that the factor income (mainly from its return from investment abroad) is slowing down and cannot be better at time when the investment climate is not very good in the developed

–23.44 –26.17 26.17 20.12 6.89 0.48 5.46 –1.74 3.63 .00 8.69 6.38 2.93 .00 3.25 4.75 0.78 2.24 12.47 5.56

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Mean SDa CVb

–21.80 23.62 24.02 3.96 –0.21 0.48 –0.07 0.78 –4.34 3.64 –0.88 –6.26 8.20 3.79 5.44 1.94 1.96 2.60 10.31 3.96

Mining, utilities

Source: IMF (2009). a SD = Standard Deviation. Note: b CV = Coefficient of Variation.

Agriculture, hunting, forestry, fishing

Year

–23.22 11.66 10.11 6.91 3.37 –1.61 0.30 1.80 0.22 –5.47 –2.68 10.51 6.97 4.60 2.55 3.60 1.83 1.85 8.00 4.32

Manufacturing

Table 8.2: Growth rates by sectors, 1991–2007

–29.08 32.55 1.40 2.68 .00 –1.99 –1.26 –0.82 0.35 2.52 20.72 10.24 2.83 3.66 2.59 3.91 1.96 1.82 11.24 6.17

Construction –23.68 19.66 –1.13 –2.58 1.60 –1.41 .00 2.50 0.82 –0.86 0.72 8.30 2.95 2.51 –0.20 6.20 1.05 0.96 8.19 8.53

Wholesale, retail trade, restaurants and hotels –24.98 17.83 9.14 4.63 –0.55 5.08 2.59 4.88 5.06 4.63 4.48 –0.09 14.04 10.78 9.77 –2.99 3.90 4.01 9.11 2.27

Transport, storage and communication –23.50 37.49 –1.78 2.50 1.75 –5.81 1.92 1.85 –0.07 3.55 2.00 4.34 3.1 3.13 2.49 4.14 1.49 2.27 11.19 4.92

Other activities

Total –22.97 27.52 10.61 3.49 0.86 –1.71 0.69 1.39 –1.82 2.63 0.37 0.35 6.10 3.96 4.15 2.74 1.86 2.36 9.39 3.97

Global Financial Crisis and the Migrant Labour Market Y 135 Figure 8.1: Relationship between diversification index and instability index

Source: As estimated by the authors.

countries reeling under financial crisis. But we have information on the oil revenue during the beginning of the crisis, i.e., the year 2007. This source of income is examined in Table 8.3. The oil revenue as a percentage of total public revenue peaked in 2006–2007 and declined by 50 per cent in 2007–2008. This will reduce the revenue expenditure of the government, affecting the growth of the economy. The analysis indicates that the economy is subject to increasing instability, and declining oil revenue and net factor income earned form abroad. The implication on the labour market can be assessed only if we understand the macroeconomic performance of the oil economy.

Oil and the Macro Economy: Sims–Hamilton Model The first task is to identify a framework for capturing the macroeconomic reality of the oil economy of Kuwait. A review of litera-

136 Y K. Pushpangadan and M. Parameswaran Figure 8.2: Ratio of GNP to GDP in Kuwait

Source: As estimated by the authors.

Table 8.3: The share of oil revenue in total public revenue Year 2002–03 2003–04 2004–05 2005–06 2006–07 2007–08

Oil revenue (OR)

Total public revenue (TPR)

OR/TPR (%)

5,498.5 6,149.9 8,170.5 12,955.4 14,511.5 7,449.9

6,219.0 6,937.3 8,962.4 13,728.1 15,509.3 8,320.3

88.65 91.16 94.37 93.57 89.54

88.42

Source: Annual Report, 2005, 2008, Ministry of Finance and Ministry of Planning, Government of Kuwait.

ture on the field suggests two pioneering works, Christopher A. Sims (1980) and James D. Hamilton (1983), for the purpose. Sims (1980) had suggested a six-variable system of VAR for understanding macroeconomic reality. This has become the standard model in literature. Hamilton modified the six-variable VAR to make it suitable for oil-importing economies. This Sims–Hamilton version of the six-variable macroeconomic system included: two variables on output (real GNP, Unemployment), three price variables

Global Financial Crisis and the Migrant Labour Market Y 137

(implicit price deflator for non-farm business income, hourly compensation for worker, import prices), and a single series on money supply (MOS) for financial sector. A similar analysis of an OPEC economy is the study on Bahrain conducted by Rizwan Tahir and Ahmed Abdul Ghani (2007). They used a five-variable VAR which included GDP, consumer price index (CPI), government expenditure (GEX), value of oil exports and money supply. In our model of Kuwait, the following variables are included: (a) GDP, (b) oil exports (OLX), GEX, MOS and CPI.

The Data The data for the study is obtained from the various publications of the Government of Kuwait (Planning, Central Bank, etc.) and international agencies such as the IMF and the World Bank. The analysis is restricted to the period, 1991–2007, largely dictated by the availability of data. All the variables are deviations from a linear trend so that they are stationary.

VAR Model Specification In order to undertake a formal analysis of VAR, we have to sort out two issues: (1) lag structure of the variables; and (2) stability of the lag coefficients. The choice of the lag structure is based on a multiple statistical criteria suggested in Lutkepohl (2006). The results are reported in Table 8.4, which indicates a lag length of two by all criteria. However, the final choice of the lag structure is based on the stability of the matrix of lag coefficients. The VAR (2) is again subjected to a stability test. The test depends on the eigenvalues of the matrix of lagged coefficients. The VAR (2) process is stable if all the eigenvalues of the matrix of lagged coefficients have modulus value less than one (ibid.: 15). By this test only VAR (1) satisfies the stability condition and hence we use VAR (1) for our analysis. Having decided on the lag structure of unity, we undertake VAR (1) analysis. It comes in three varieties: reduced form, recursive, and structural (Stock and Watson 2001). Our analysis concentrates only on two versions — reduced and recursive — of VAR (1). The two versions are used here to summarize the co-movements of the five variables. The summary results are: (a) Granger causality tests; (b) forecast error variance decompositions; and (c) the impulses response analysis.

138 Y K. Pushpangadan and M. Parameswaran Table 8.4: Statistical tests for lag length of VAR Lag

Log L

LR

FPE

AIC

SC

HQ

0 1 2

–486.58 –440.98 –382.18

NA 59.01 41.50∗

9.01E+18 9.35E+17 5.13E+16a

57.83 55.40 51.43a

58.07 56.88 54.12a

57.85 55.55 51.70a

Source: Central Bank of Kuwait; World Development Indicator; IMF 2009. a Note: Indicates lag order selected by the criterion LR: sequential modified LR test statistic (each test at 5 per cent level). FPE: Final prediction error; AIC: Akaike information criterion; SC: Schwarz information criterion. HQ: Hannan–Quinn information criterion.

Granger Causality Test The first summary results are from reduced VAR (1). In the reduced form VAR, there are five equations, one for each of the five variables. Each variable is specified as a function of its own lagged value, the lagged values of the remaining four variables, and a serially uncorrelated error term. The specification implies that the only current variable that affects the current dependent variable is the stochastic error term. Granger causality statistics examine whether lagged values of one variable help to predict another dependent variable. Table 8.5 summarizes the Granger causality results for the reduced form five-variable VAR (1). The tests show that OLX causes GDP. This is not a surprising result since the entire economic activity depends on the oil exports of the country. The other findings are: OLX causes CPI and GEX causes CPI. This suggests that oil exports and government expenditure generate inflationary pressure in the economy. These findings are important for our later analysis when the impacts are measured in terms of real consumption. However, GDP and GEX, GDP and CPI have feedback system in the sense that causality runs both ways (Lutkepohl 2006: 42). The other standard summary statistics is based on the decomposition of forecasting error variance known as Innovation accounting in VAR literature. Such an analysis requires the specification of the recursive VAR model. The recursive model ‘constructs the error terms in each regression equation to be uncorrelated with the error in preceding equations’ (Stock and Watson 2001: 103). Here the order

Global Financial Crisis and the Migrant Labour Market Y 139 Table 8.5: VAR descriptive statistics A. Granger-Causality Test for Reduced Form VAR (1) Dependent variable in regression Regressor

GDP

OLX

GEX

MOS

CPI

GDP OLX GEX MOS CPI

0.00a 0.07a 0.00a 0.98 0.00a

0.50 1.00 0.44 0.81 0.89

0.00a 0.50 1.00 0.91 0.34

1.00 1.00 0.95 0.00a 0.37

0.07a 0.00a 0.00a 0.34 0.30

B. Forecast Error Variance Decomposition from the recursive VAR (1) ordered as GDP OLX GEX MOS CPI B (i) Variance Decomposition of GDP Period

SE

GDP

OLX

GEX

MOS

CPI

1 2 4 8 12

458.23 604.01 730.51 1003.63 1072.59

100.00 79.95 56.43 30.70 28.21

0.00 7.09 5.13 6.20 5.75

0.00 8.86 27.63 18.11 20.17

0.00 0.18 6.90 40.29 38.90

0.00 3.90 3.88 4.67 6.94

B (ii) Variance Decomposition of OLX Period

SE

GDP

OLX

GEX

MOS

CPI

1 2 4 8 12

204.42 228.02 258.42 327.56 345.67

53.35 48.92 39.58 25.25 23.78

46.64 45.50 35.79 25.05 22.76

0.00 5.53 18.34 14.16 16.27

0.00 0.00 5.66 33.01 32.60

0.00 0.03 0.61 2.49 4.58

B (iii) Variance Decomposition of GEX Period

SE

GDP

OLX

GEX

MOS

CPI

1 2 4 8 12

7.85 10.00 12.00 13.49 14.17

32.23 36.44 31.02 26.33 24.06

27.96 22.59 18.14 14.96 14.24

39.79 31.49 22.87 25.34 23.58

0.00 4.02 21.66 23.83 28.80

0.00 5.43 6.28 9.53 9.30

140 Y K. Pushpangadan and M. Parameswaran B (iv) Variance Decomposition of MOS Period

SE

GDP

OLX

GEX

MOS

CPI

1 2 4 8 12

267.67 389.33 523.57 587.50 616.65

0.54 0.26 0.89 3.03 2.94

8.41 8.98 7.85 6.74 6.77

10.41 5.28 5.27 11.50 11.06

80.61 83.51 79.28 68.68 69.31

0.00 1.95 6.69 10.02 9.90

B (v) Variance Decomposition of CPI Period

SE

GDP

OLX

GEX

MOS

CPI

1 2 4 8 12

12.94 16.48 22.20 27.47 28.67

15.32 13.00 7.65 6.47 6.50

1.90 5.39 6.49 6.04 6.04

41.42 40.04 24.85 21.57 21.35

0.07 12.38 44.33 49.21 50.17

41.27 29.16 16.65 16.68 15.91

Source: As estimated by the authors.

of the five variables (GDP, OLX, GEX, MOS and CPI) becomes very important. In our version, the order is based on the economic theory of interdependence among the variables. The dependent variable in the first equation is GDP and the regressors are lagged values of all the five variables included in the VAR and an error term. In the third equation, for example, the dependent variable is GEX and the regressors are lagged values of all five variables plus the current value of the first two variables in the following order, GDP and OLX. The final equation in the recursive VAR contains CPI as a dependent variable and the independent variables are the lagged values of all five variables plus the current values of the first four variables in the order-GDP, OLX, GEX, MOS. Obviously, the error term in the second equation is uncorrelated with the first equation in the system, third with the second, fourth with the third and fifth with the fourth. It may be noted that the error forecast depends on the order of the specification. The results are reported in Table 8.4. The major highlights are as follows. We make a distinction of short run (two-step forecast) and long run (12-step forecast) for interpreting the results in Table 8.4. For example, about 80 per cent of the two-step forecast error variance of the GDP (Table 8.4) is accounted for by its own innovations and only 20 per cent from innovations in other equations (Lutkepohl 2006). The major share of the 20 per cent (9/20 = 45 per cent) goes to GEX. In the case of MOS it is 84

Global Financial Crisis and the Migrant Labour Market Y 141

per cent from its own innovations (Table 8.4, B [iv]). But the short run behaviour is not the same for the remaining three equations in the system. For example, in the case of OLX (Table 8.4, B [ii]), about 46 per cent is contributed by own innovations and 49 per cent is accounted for by GDP innovations. In the case of GEX (Table 8.4, B [iii]), only 31 per cent is contributed by its own innovations, 59 per cent by innovations in GDP and that of OLX is 23 per cent). In the case of CPI, which determines the real consumption of wage earners the impact is different; the own innovation is only 29 per cent, and 40 per cent of the remaining 71 per cent of variance is contributed by innovations in GEX. The contribution of GEX is 11 per cent higher than own innovations in OLX. Let us examine the long run impact by considering 12-step forecast error variance. Own contribution of innovation in GDP, OLX, GEX and CPI is lower than the contribution of MOS. The impact of MOS is highest in CPI. In other words, MOS is the most important macro variable in the system that affects the innovation accounting of forecast error variance. This recursive model can also be used for assessing the impact of shocks in one equation to another variable by assuming all other errors to remain zero. The impact responses are given in Figures 8.3 to 8.8. It ‘traces out the response of current and future values of each of the variables in the current value of one of the VAR errors, assuming this error returns to zero in subsequent periods and that all other errors are equal to zero’ (Stock and Watson 2001: 106).The impulse responses in Figure 8.3 give the shock in GDP, Figure 8.4 in OLX, Figures 8.5 and 8.6 in GEX, Figure 8.7 in MOS, and Figure 8.8 in CPI. All the variables are plotted with confidence interval of two standard errors. The major highlights are as follows. In the case of GDP shocks, an unexpected one-percentage point increase in GDP reduces inflation (CPI) and GEX in the short run (Figure 8.3). This would mean that slowing down of the economy leads to higher inflation indicating stagflation in the economy. In the case of shocks in OLX, the impact is mostly negative, in the short run, in all the four variables (Figure 8.4). In Figures 8.5 and 8.6, the CPI has a negative impact on OLX in the long run and on MOS in the short run. MOS has positive effects on all the four variables in the short run. The impulse analysis and innovation accounting suggest that MOS is inflationary and economy experiences stagflation. In order to examine the effect on labour market, we examine the coping mechanisms between the firms and labourers analytically as shown below.

142 Y K. Pushpangadan and M. Parameswaran Figure 8.3: Impact of GDP error term on remaining macro variables

Source: As estimated by the authors.

Figure 8.4: Impact of OLX shocks on the remaining four macro variables

Source:

As estimated by the authors.

Figure 8.5: Impact of CPI error shock on the remaining four macro variables

Source:

As estimated by the authors.

Global Financial Crisis and the Migrant Labour Market Y 143 Figure 8.6: GEX shocks on the other four macro variables

Source: As estimated by the authors.

Figure 8.7: Impact of MOS error shocks on the remaining four macro variables

Source: As estimated by the authors.

Figure 8.8: Impact of CPI error shocks on the remaining four macro variables

Source: As estimated by the authors.

144 Y K. Pushpangadan and M. Parameswaran

Impact on Migrant Labour Force The impact of the global financial crisis on migrant labour in an oil exporting economy is examined here in the conventional supply– demand framework. The global crisis has slowed down and in some cases there has been negative growth in the industrialized world since the crisis began in 2007. As a result, the economic activities of the oil exporting countries have also slowed down, affecting the demand for labour in general and migrant labour in particular. To understand the coping mechanism of migrant labour, which is the main purpose of this investigation, let us assume the migrant labour market is in equilibrium before the crisis as shown in Figure 8.9. The equilibrium emoluments rate is E∗ and the corresponding employment L∗ before the crisis. After the crisis, the labour demand shifts leftward (from D1 to D2) as a result of the stagnating world income and the new equilibrium is lower emoluments rate — E1∗ — and lower employment, L1∗. The migrant unemployment is given by L∗–L1∗. Obviously, there is a welfare loss due to reduction in both producer’s and labourer’s surpluses after the crisis. However, there exist other adjustment possibilities in the labour market. For example, the Figure 8.9: Impact on labour market pre- and post-financial crisis

Source: As estimated by the authors.

Global Financial Crisis and the Migrant Labour Market Y 145

same level of employment before the crisis can be maintained provided the market can adjust to the lower emoluments rate, E2∗. In this case, there is no unemployment but substantial reduction in the emoluments rate. What are strategies available to the stakeholders in this situation? For the migrant labour, the option is either to quit the present job and return home. In this case, the return migrants is of the order of L∗–L2. The other coping mechanism is to adjust living standards to the new level of income. One possibility of cutting down the cost of living is to send the family to their country of origin until the emoluments rate reaches the pre-crisis period. This is only a short-run response. In the short run, there is no guarantee that the same effort level will be coming from the labourers for a reduced emoluments rate. One coping mechanism for firms in this situation is to recruit labour from abroad for the low emoluments rate and replace them with the experienced one so that the supply curve shifts to the right (S2) and the market adjusts to the new equilibrium, E2 and L∗. The net benefit of strategy is to be assessed after accounting for search costs and productivity changes of the fresh recruits. This is an empirical question. The analysis indicates that the economy is subject to increasing instability, and declining oil revenue and net factor income earned from abroad. The implication on the labour market can be assessed only if we understand the macroeconomic performance of the oil economy.

Emoluments Rate For empirical verification, we need to know the structure of the migrant labour force in Kuwait. This is examined in the Table 8.6. The percentage of non-citizens in the total employed labour force was 13.9 per cent for males and 23 per cent for females in 1996. This increased to 14.6 per cent for males and 6.7 per cent for females in 2004. The distribution of migrant labour force among sectors is given in Table 8.7. The sectoral distribution of the migrant labour force, as indicated in Table 8.7, is concentrated in public administration, wholesale and retail trade, construction, and manufacturing. The trend remains more or less the same in construction and manufacturing but has declined in public administration and trade (wholesale and retail). It is interesting to note that the unclassified category has almost doubled during the period and it is most likely that it reflected the

146 Y K. Pushpangadan and M. Parameswaran Table 8.6: Structure of the labour force in Kuwait (in thousands) Citizen

Non-citizen

Year

Male

Female

Total

Male

Female

Total

Total

1996 1999 2002 2004

123.87 142.50 162.00 179.60

60.75 75.10 96.50 119.60

184.62 217.60 258.50 299.20

751.72 815.00 820.20 1,052.00

203.74 215.50 241.40 283.00

955.46 1,030.50 1,061.60 1,335.00

1,140.08 1,248.10 1,320.10 1,634.20

Source: Ministry of Finance and Ministry of Planning, Government of Kuwait.

increase in domestic services such as house maids and servants at the household level. The impact can be assessed only through a detailed study of sectors with migrant concentration.

Coping Mechanisms: Evidence Impact analysis on labour market and coping mechanisms devised by stakeholders can be assessed only if we conduct structured primary surveys among the main stakeholders directly or indirectly affected by the financial crisis. Ideally, questionnaire-based information should be collected and analyzed separately for firms, migrant workers and the public policy makers. Such a survey could not be executed for two reasons. First was the team’s lack of time availability. The visit was only for a week and no survey could be conducted during such a short period of time. The second reason, more Table 8.7: Distribution of migrant labour force by sector Sectors

2003

2005

2006

Agriculture and Fisheries Mining and Quarrying Manufacturing Construction Electricity, Water and Gas Wholesale and Retail Trades Transportation and Communication Finance and Business Services Public Administration Unclassified Total

2.1 0.2 6.8 9.1 0.2 19.1 3.2 4.6 45.7 8.9 100.0

2.2 0.1 6.1 8.8 0.2 16.4 3.1 4.7 43.9 14.4 100.0

2.1 0.1 6.0 9.6 0.2 16.0 3.3 4.8 41.5 16.5 100.0

Source: Ministry of Planning, Statistics and Census Sector, Government of Kuwait.

Global Financial Crisis and the Migrant Labour Market Y 147

scientific, was the absence of any record of migrant population with the embassy. As a result, there is no way of selecting a representative sample of migrants from the migrant community for detailed analysis. Therefore we are left with field observations and in-depth interviews with selected stakeholders for evidence. We interviewed the following: (a) managers of two money exchange companies, (b) two construction companies, (c) and one logistic company; (d) representatives of the Kuwait Chamber of Commerce; (e) headmaster of an Indian School; and ( f ) a number of migrant workers through their associations. The general impression we got from these discussions is as follows. The crisis happened at a time when the country was having a higher rate of inflation. The sources of the inflation were (a) higher price of imported items, particularly of food items, because of global food price inflation; and (b) investment of a part of the huge current account surplus in the domestic economy, particularly in the real estate sector that increased overall demand in the economy.1 Our discussion with migrant workers revealed that cost of living increased mainly because of increase in rent, cost of education and price of food items. But the wages remained the same for most of the unskilled workers of the labour market, thus reducing real consumption. Although the financial crisis has not led to widespread retrenchment of workers, it has led to a reduction or stagnation of wages as well as of delayed payment of wages as predicted by our supply–demand model and later confirmed by the migrant interviews. The extent of job loss, family repatriation and shifting of children’s education cannot be assessed since we do not have any direct estimates of the same. The interview also revealed that companies requested their employees to go for extended vacation without pay, and reported of unilateral reduction in benefits and delayed payments of wages which may/may not be based on the actual impact of crisis. This again cannot be verified with statistical validity. Our interview with Human Resource Managers of the construction companies confirm that they got lesser number of projects during the year 2008 due to the financial crisis. They were also informed that a number of construction projects in the commercial and oil 1 This information is obtained from the discussion with the representatives of the Kuwait Chamber of Commerce.

148 Y K. Pushpangadan and M. Parameswaran

sectors have either been cancelled or shelved in 2008. One construction company, which employs around 28,000 workers, revealed that they had a surplus of around 1,000 workers because of the reduction in the number of projects. They had retrenched around 500 of them and the remaining workers were relocated to other Gulf countries. However, workers of the same company, whom we interviewed later, said that the company retrenched around 6,000 workers. The managers of the logistic company revealed that the crisis did not affect them simply because a major chunk of their activities is in maintaining the supply chain of the American army located in Kuwait and Iraq, which is immune to any financial crisis. Once the army moves back to America, a substantial number of migrant workers would be retrenched. The fixed wage earners, especially at the lower levels of the migration ladder, may be the worst affected as indicated by the close-down of most of the small shops in the region where a large number of unskilled workers live. This may be attributed to the stagflation in the economy. Unfortunately, very little is known about their coping mechanisms, particularly of the petty traders and workers in household chores. In order to assess the impact we need a complete listing of migrants by occupation structure. Such details are not available anywhere and therefore one has to wait until such information is provided by the respective embassies.

Summary and Conclusions The economy of Kuwait showed stagnation in the year 2007, at the beginning of the global financial crisis. But the projected growth rate by the IMF indicates a meltdown in 2009. The sectoral growth rates show mixed results: slowdown as well as meltdown. There is no uniformity in the meltdown process among the sectors. The analysis indicates that the economy is subject to increasing instability, and declining oil revenue and net factor income earned from abroad. The impulse analysis and innovation accounting using five-variable VAR suggest that money supply has an impact on all other four macro-variables (GDP, oil exports, government expenditure and consumer price index). The stagnation of the economy has a tendency towards inflationary pressure in the economy. In other words, the crisis has a stagflation effect on the economy. In order to examine the impact on the labour market, we looked at the coping mechanisms between the firms and labourers analytically. The analytical result

Global Financial Crisis and the Migrant Labour Market Y 149

predicted generalized decline in total emoluments. In the long run, even fresh recruits from the country of origin may be resorted to for replacing the experienced workers in the countries of destnation. The field-level observations and interviews with stakeholders confirmed that there is reduction in the emoluments rate of firms. The strategies pursued include, among other things, cutting down fringe benefits, delayed payments of wages and salaries, and forced vacation for an extended period. The study recommends the urgent need for the assessment of impacts on low-skilled migrant labour in general and domestic helps in particular. This requires a complete list of migrants with the embassies. Preliminary investigation shows that there is no such record of with the Kuwait Embassy. This has to be urgently sorted out by Ministry of Overseas Indian Affairs in consultation with Department of Immigration and the Ministry of External Affairs. To begin with, the information available with the immigration section of all international airports in India on migrants and return migrants should be shared with the respective Indian embassies so that systematic management of migrants can be initiated with human face.

Acknowledgements This study is part of a larger project to understand the impact of Global Financial Crisis on migrant labour in Gulf countries sponsored jointly by the Asian Development Bank and the Ministry of Overseas Indian Affairs, Government of India. We are thankful to the Indian Embassy for arranging meetings with migrant associations, financial institutions and firms. We are also thankful to Pratheesh N. P. for being an excellent research assistant. Of course, the authors are responsible for the contents and limitations, if any.

References International Monetary Fund (IMF). 2009. World Economic Outlook: Crisis and Recovery. Washington, DC: International Monetary Fund. Hamilton, James D. 1983. ‘Oil and the Macroeconomy since World War II’, Journal of Political Economy, 91 (2): 2828–48. Lutkepohl, Helmut. 2006. ‘Forecasting Temporally Aggregated Vector ARMA Processes’, Journal of Forecasting, 5 (2): 85–95. Sims, Christopher A. 1980. ‘Macroeconomics and Reality’, Econometrica, 48 (1):1–48.

150 Y K. Pushpangadan and M. Parameswaran Stock, James H. and Mark W. Watson. 2001. ‘Vector Autoregressions’, Journal of Economic perspectives, 15 (4): 101–15. Tahir, Rizwan and Ahmed Abdul Ghani. 2007. ‘Modelling of Bahrain’s Economy: A Vector Autoregression (VAR) Approach’, Middle East Business and Economic Review, 19 (1): 23–38.

9 Low-skilled Indian Construction Workers in the Gulf, Singapore and Malaysia Return to India, Reintegration and Re-migration Auke Boere An International Labour Organization (ILO) presentation by Patrick Taran in 2009 stated that migrants are often ‘the last hired and the first fired’ (Taran 2009: 1). The global economic crisis has in many regions of the world created a setting where this has turned out to be true for an increasing number of migrant workers. In the Gulf, hundreds of thousands of Indian migrants who came to work there in the construction sector on temporary labour contracts have felt the consequences of the economic downturn. Dubai has especially been severely hit, since it is the region’s key financial centre. Furthermore, unlike other emirates and Gulf states, Dubai does not possess any oil reserves. In South-East Asia, Singapore and Malaysia are important destinations for low- and unskilled Indian labour migrants and here retrenchments are also followed by return migration. Coming back to a country of origin is often not a sinecure. Old patterns of living can be hard to pick up again and new ones might be tough to establish. Family relations can partly have been eroded due to newly acquired identities or ideas by both the migrant and the family in the home community. Additionally, many Indian migrants have faced extremely difficult and harsh conditions while abroad, which inevitably will leave an imprint on the life after emigration. But not all returnees make an effort to reintegrate. Low- and unskilled construction workers often do not stop migrating after their first project and a second, third or fourth migration is not uncommon. Hence it is not unlikely that many of the recently laid-off construction workers will, despite the crisis, try to emigrate again.

152 Y Auke Boere

Although reintegration is thus not a self-evident phenomenon, not every returnee will have the wish to re-emigrate1 and for them reintegration issues are very relevant. To what extent do these reintegration issues play a role for the Indian migrants and their families and what kind of issues should we think of? Also, are there any policies aimed at return migrants and their reintegration? This chapter goes into reintegration issues and their importance, depending on the scale of re-emigration efforts. Furthermore it will elaborate on the (very few) policies that exist for Indian return migrants. It is the momentum caused by the crisis in combination with the lack of research on reintegration, circular migration and re-emigration of south Indian migrants which makes this research important. Hugo Graeme (2003) states that temporary and circular migrations have not been granted the attention they deserve in scientific literature, considering that this form of migration is increasingly on the rise. Hein De Haas (2005) states that ‘the clear-cut dichotomies of “origin” or “destination” and categories such as “permanent”, “temporary” and “return” migration are increasingly difficult to sustain in a world in which the lives of migrants are increasingly characterized by circulation and simultaneous commitment to two or more societies’ (ibid.: 1247). This chapter deals with the impact on reintegration of these forms of migration in south India, backed by fieldwork carried out with return migrants who have been working in the construction sector in the Gulf, Singapore and Malaysia. After shortly going into the methodology used during the fieldwork, a general picture will be drawn of the reasons of return and reintegration issues of the returnees in this survey. We will examine the communities of Tamil Nadu (taken together) and Narsingi in Andhra Pradesh in detail, because figures were dramatically different between the states and localities in which the surveys were held. In the last section, we will review the existing policies for return migrants with some concluding remarks and recommendations.

Methodology The main methodological base of this research is a survey held among return low- and unskilled Indian construction workers in

I use the word ‘re-emigration’ in this chapter because ‘re-migration’ is in some scientific literature synonymous for ‘return migration’. 1

Low-skilled Indian Construction Workers Y 153

several localities in south India. For this survey a questionnaire was used to get insight into the characteristics of the migrants, as well as their status and issues before, during and after emigration. Although the focus of the research is more on the situation after return, questions were also asked about the phases of pre-departure and emigration itself, because all migration phases are inextricably connected to each other and earlier phases in the migration project determine for a large part the severity and kind of reintegration issues returnees face. For the accomplishment of the fieldwork I was largely dependent on the local trade unions that were aligned to the Building and Wood Workers’ International (BWI), the global union federation for labourers working in construction. They had key informants who could bring me into contact with the return migrants. Because of this, the fieldwork was carried out at locations where they had their contacts. The fieldwork locations were Isrampally (18 respondents) and Narsingi (26 respondents) in Andhra Pradesh; around the city of Perumbavoor in Kerala (18 respondents); around Tirupattur and Pudakottai in Tamil Nadu (61 respondents); and in and around Kallakkurichi, also in Tamil Nadu (20 respondents). In addition to the face-to-face interviews, phone calls were made to the interviewed migrants to gather more qualitative data about their return stories and the specific problems they faced since being back in India. Lastly, to get a clearer picture of the general context of Indian return migration, several semi-structured interviews were conducted with stakeholders in the field of Indian migration and construction.

Reintegration Patterns and Propensity to Re-emigrate The crisis and its concomitant retrenchments have urged many south Indians to return home. What has been the exact reason heading back home? What are the issues that appear for the returnees in their home communities and with which factors does this correlate? In this chapter, generalizations will be done based on the sample survey conducted by the author.

Reasons for Return For low- and unskilled migrant workers, emigration to the Gulf, Singapore or Malaysia will inevitably include return, since strict

154 Y Auke Boere

immigration rules prohibit them from settling permanently in one of these destination countries. Sometimes the migrants manage to extend their working period abroad for several years or even decades and others manage to re-emigrate soon after return. But permanent settlement is effectively banned by the destination authorities and most migrants return within five years after the end of the labour contract (in this sample 78.3 per cent; Table 9.1). In this survey, 76 (53.1 per cent) respondents said they chose to return themselves, which means that not less than 67 (46.9 per cent) were forced to do so. When returnees had themselves chosen to return, three reasons proved to be important. ‘Low salaries abroad’ was for 50.7 per cent of the voluntary returnees a very important reason to go back to India. This is striking, since the main reason for leaving India in the first place is the pursuit of higher salaries. It is likely that the economic crisis has caused some salaries to become lower than what had been promised and indicated by fellow migrants. Other important reasons for respondents who had chosen to come back were ‘family and friends’ (for 25.3 per cent a very important reason; 42.7 per cent found it at least important) and ‘difficult labour conditions’ (important reason for 32 per cent). Reasons like ‘difficult living conditions’, ‘ill health’, ‘suitable employment in India’ and ‘homesickness’ were not often mentioned by the returnees. Forced returnees frequently cited closed companies, expired visas and contracts that had come to an end and could not be renewed as reasons. But as already mentioned, there was great versatility in the exact nature of the forced returns and often the story was much more complex than a simple ending of the labour contract.

Table 9.1: Duration of last emigration among the return emigrants Duration Less than 1 year 1–2 years 2–3 years 3–4 years 4–5 years More than 5 years Total respondents Source: Survey conducted by the author.

Number

Per cent

16 29 38 20 9 31 143

11.2 20.3 26.6 14.0 6.3 21.7 100.0

Low-skilled Indian Construction Workers Y 155

Jianaselan says:2 I was working for a company which did not want to improve my grade and salary, although I managed to get a license as a piling rig operator. Then I got an opportunity at another company, so I cancelled my job at the first one. But due to the economic crisis eventually I could not enter the other company, and without a contract I had to return to India.3

Rajendran has quite a different story to tell: ‘My employer in Bahrain asked me to pay money to renew my visa and labour contract, but the sum was too high in comparison with my salary. That is why I had to return to India.’4 Sometimes the stories are more poignant. Thandambani reveals: ‘I worked for a bad company in Saudi Arabia, so I switched to another one. This was illegal, however, and I was caught by the police. After five months in jail I was sent back to India.’5 Ramesh, from Narsingi, too had tough experiences: I went on a visit visa to Dubai. At a certain point I lost my job and then I spent eight months there illegally without employment. Then I decided to go to the police myself, so they put me in jail. Luckily it lasted only two days, because after that I had arranged money for my return ticket with another family in Narsingi.6

After return, no matter what the reason, life continues in south India, at least for a short period. Income and access to money are obviously of crucial importance for setting up or revamping a life back home. Money also increases the chances in the labour market, for instance, by investing in one’s own business, which consequently helps in both economic and social integration. When comparing the financial situation before the first emigration with the situation after the last emigration, it becomes clear that the migration project for most respondents is not a financial success.

2 All interviews were conducted by the author and all of them were return migrants. Names have been changed to protect the identity of the respondents. 3 Interview with Jianaselan, Isrampally, Andhra Pradesh, 10 March 2010. 4 Interview with Rajendran, Pudakottai, Tamil Nadu, 14 April 2010. 5 Interview with Thandambani, Kallakkurichi, Tamil Nadu, 21 March 2010. 6 Interview with Ramesh, Narsingi, Andhra Pradesh, 14 March 2010.

156 Y Auke Boere

Table 9.2 shows that only 37.8 per cent of the respondents said that they were financially better-off after emigration. Another 29.4 per cent, however, felt their situation had remained the same, while 32.9 per cent indicated that they were in a worse financial situation. These figures indicate that what by many migrants is seen as the key to a more well-off existence often turns out to be a financial disappointment or even a nightmare.

Analysis on the Basis of Destination Country It is interesting to investigate whether differences occur when comparing respondents according to the destination country. The statistics for Dubai show a striking deviation with the other destinations (Table 9.3). Only 20 per cent of the respondents who went to Table 9.2: Current financial situation compared with financial situation before emigration Opinion

Number

Per cent

54 42 47 143

37.8 29.4 32.9 100.0

Better Unchanged Worsened Total Source: Survey conducted by the author.

Table 9.3: Last country of emigration and the comparison of financial situation before migration Current financial situation compared to before emigration

Last country of emigration Dubai Other Gulf States Malaysia Singapore Total

Number Per cent Number Per cent Number Per cent Number Per cent Number Per cent

Better

Unchanged

Worsened

Total

11.0 20.0 15.0 34.9 6.0 50.0 22.0 66.7 54.0 37.8

13.0 23.6 18.0 41.9 4.0 33.3 7.0 21.2 42.0 29.4

31.0 56.4 10.0 23.3 2.0 16.7 4.0 12.1 47.0 32.9

55.0 100.0 43.0 100.0 12.0 100.0 33.0 100.0 143.0 100.0

Source: Survey conducted by the author.

Low-skilled Indian Construction Workers Y 157

Dubai said that they have improved their financial situation, while 56.4 per cent of the migrants indicated that their situation has worsened. In contrast, migrants who have been to Singapore revealed that they have improved their financial situation in 63.6 per cent of the cases, while only 12.1 per cent reported that they are worse-off. Debts were the most commonly mentioned and striking concern; 40.8 per cent of the respondents called debts a very problematic issue, while 58.5 per cent found it problematic at least to some extent (Table 9.4). Debts are often incurred when investing for migrating abroad again. Passports, visas, tickets, and recruitment charges cost the migrants often more than what they can pay for out of their own savings and therefore loans have to be taken from private moneylenders or friends and relatives. Certain private moneylenders charge high interest rates which further increase the debt burden in the course of time. Ramamoorthy reveals: For going abroad, I got a loan of one lakh from a private lender in the village. Because I lost my job abroad, I could not pay back the money, and due to the interest rate the debt increased. So now I have to go abroad again to repay the loan. However, in India I do not see my debt as a problem, because I made an agreement with the private moneylender that I cannot pay back the money as long as I am in India.7 Table 9.4: Last country of emigration and debt problems Debt problems Last country of emigration Dubai Other Gulf States Malaysia Singapore Total

Number Per cent Number Per cent Number Per cent Number Per cent Number Per cent

Very problematic

Problematic

Some problems

No problems

Total

34.0 61.8 12.0 27.9 4.0 36.4 8.0 24.2 58.0 40.8

6.0 10.9 1.0 2.3 1.0 9.1 7.0 21.2 18.0 10.6

4.0 7.3 5.0 11.6 0.0 0.0 1.0 3.0 10.0 7.0

11.0 20.0 25.0 58.1 6.0 54.5 17.0 51.5 59.0 41.5

55.0 100.0 43.0 100.0 11.0 100.0 33.0 100.0 142.0 100.0

Source: Survey conducted by the author. 7

Interview with Ramamoorthy, Tirupattur, Tamil Nadu, 25 March 2010.

158 Y Auke Boere

Karyppaiah has a similar story to tell: I borrowed ‘175,000 from a private lender to pay the recruitment agent. But I earned only 16 Ringgit a day in Malaysia (approximately ` 225), so I could not save enough. The interest rate of the loan is 5 per cent, so the debt has now increased to ` 200,000.8

Table 9.4 shows that migrants who had been to Dubai considered debt problems as very problematic or problematic in 72.7 per cent of the concerned cases, while for returnees from Singapore and the other Gulf countries taken together this was only 45.4 per cent and 30.2 per cent respectively. These figures show that Dubai seems to be a much riskier destination than the other receiving countries of Indian low- and unskilled labour migrants.

Type of Return Jean-Pierre Cassarion (2004) states that ‘return refers to a preparation process that can be optimally invested in development if it takes place autonomously and if the migration experience is long enough to foster resource mobilization” (ibid.: 275–76). In other words: the type of return is incredibly important for the success rate of a migration project. When it suddenly comes to an end it can have severe consequences for the situation back in the home community. When exploring the relation between the type of return (forced or chosen) and the financial situation after emigration, a clear indication emerges that the migrants who decide to come back themselves often fare better than migrants who are forced to do so. Voluntary returnees said that their financial situation had improved in 46.1 per cent of the cases, against 19.7 per cent of respondents who revealed that they were financially worse-off after returning to the home country (unchanged for 34.2 per cent). For returnees who were in some way compelled to return, only 28.4 per cent reported that they were financially better-off after leaving; 23.9 per cent said the situation had remained the same, but for 47.8 per cent it had worsened (Table 9.5). The figures are more clear-cut even when examining the relation between type of return and debt problems. Among voluntary returnees, debt problems were very problematic for 24 per cent, 8

Interview with Karyppaiah, Pudakottai, Tamil Nadu, 27 March 2010.

Low-skilled Indian Construction Workers Y 159 Table 9.5: Type of return and financial situation after emigration Current financial situation compared to before emigration Type of return Chose to return Forced to return Total

Better

Unchanged

Worsened

Total

35.0 46.1 19.0 28.4 54.0 37.8

26.0 34.2 16.0 23.9 42.0 29.4

15.0 19.7 32.0 47.8 47.0 32.9

76.0 100.0 67.0 100.0 143.0 100.0

Number Per cent Number Per cent Number Per cent

Source: Survey conducted by the author.

problematic for another 8 per cent and somewhat problematic for 9.3 per cent (41.3 per cent in total), while 58.7 per cent of them said that they have no debt problems at all. Forced returnees, however, have very problematic debts in 59.7 per cent of the cases; 13.4 per cent had problematic debt problems and 4.5 per cent had some problems (a total of 77.6 per cent). Only 22.4 per cent of the forced returnees stated that they had no debt problems at all (Table 9.6). Murugesan states: ‘I got a loan from the brokers. But because I already had to come back to India after six months, I could never make enough money to repay these debts’.9 Veerupandiyan has had a similar experience: There was not enough work in the company and according to Singapore labour laws the company had to reduce the amount of Table 9.6: Type of return and debt problems Debt problems Very Some problematic problematic problems

Type of return Chose to Return Forced return Total

Number Per cent Number Per cent Number Per cent

18.0 24.0 40.0 59.7 58.0 40.8

6.0 8.0 9.0 13.4 18.0 10.6

7.0 9.3 3.0 4.5 10.0 7.0

No problems

Total

44.0 58.7 15.0 22.4 59.0 41.5

75.0 100.0 67.0 100.0 142.0 100.0

Source: Survey conducted by the author. 9

Interview with Murugesan, Pudakottai, Tamil Nadu, 28 March 2010.

160 Y Auke Boere manpower. Luckily Singapore labour laws also oblige the company to provide the return tickets. But still I cannot repay the loan I got from a private lender to pay the recruitment agent.10

There are also big differences between the type of return and the country from where this return has to take place. When comparing the last country of emigration with the type of return, it shows that 65.5 per cent of the respondents who went to Dubai were forced to come back. Singapore also scores relatively high here with 48.5 per cent, while from the other Gulf States only 27.9 per cent were forced to return (Table 9.7).

Re-emigration Return is often not the final stage in the migration process of Indian construction workers abroad. Migration is a continuum and the process can endure a whole working life for some migrants. In this sample, for 35 per cent of the respondents the last emigration was not the first. The percentage of returnees with the intention to re-emigrate is higher, however. In the total sample, 39.9 per cent of the respondents indicated that they were absolutely sure about re-emigrating when they would get the chance, while another 25.2 per cent were at least considering it (65.1 per cent in total). Of this group, 35.2 per cent wanted to go to the same country as the last time and 30.8 per cent wished to go to another country. Another 33 per cent was not sure, Table 9.7: Last country of emigration: Type of return Last country of emigration Type of return Chose to Return

Dubai Other Gulf States Malaysia Singapore Number

Total

19.0

31.0

9.0

17.0

76.0

Per cent 34.5 Forced return Number 36.0 Per cent 65.5 Total Number 55.0 Per cent 100.0

72.1 12.0 27.9 43.0 100.0

75.0 3.0 25.0 12.0 100.0

51.5 16.0 48.5 33.0 100.0

100.0 67.0 100.0 143.0 100.0

Source: Survey conducted by the author.

10

Interview with Veerupandiyan, Pudakottai, Tamil Nadu, 28 March 2010.

Low-skilled Indian Construction Workers Y 161

what often means that the migrant is willing to go anywhere, as long as he will get a good salary for the work he provides. Comparing the actual re-emigration rate and the intention to re-emigrate thus indicates that the former figure is not so much lower than the figure for people who are absolutely certain about re-emigration, but almost twice as low when the hesitating migrants are also involved in the comparison. For re-emigration, money too continues to be an important push to leave India. ‘Higher salaries’ were the most prominent reason to go abroad (very important for 78.5 per cent, important for 94.6 per cent). Not being able to find a job in India was a very important reason for 18.3 per cent, an important reason for 21.5 per cent and of some importance for 9.7 per cent (49.5 per cent in total). For 50.5 per cent, repayment of debts was a very important reason to go abroad again, important for another 16.1 per cent and of some importance to 10.8 per cent (77.4 per cent in total). Status improvement was very important for 21.5 per cent and important for another 26.9 per cent. Not being able to re-adapt in India did not play any significant role in the motivations for potential re-emigrants. Also, unemployment seems to be an important factor for leaving India again. Of the unemployed respondents, most have an absolute wish to re-emigrate (23 out of 34, 67.6 per cent). They constitute 40.3 per cent of the group of returnees who are certain about their wish to re-emigrate. Of these potential re-emigrants who are jobless, 16 respondents said that they are also not searching for employment in India. For these returnees reintegration issues seem completely irrelevant, because they are not trying to reintegrate.

Analysis for Separate Research Localities Varied data was found for the different settings that the fieldwork was carried out in. Therefore, it is considered essential to also describe these localities separately and elaborate on their results individually.

Tamil Nadu In Tamil Nadu, questionnaires were canvassed at several locations in two different regions: in villages around Tirupattur, Pudukottai district, and in and around the city of Kallakkurichi, which all showed similar results. In total, 81 valid interviews were conducted in this state.

162 Y Auke Boere

Most respondents in Tamil Nadu were younger than 35 years old (61.2 per cent) and had at least finished middle school (83.8 per cent). Before emigration, most people in the sample were either working in agriculture (40.7 per cent) or the construction sector (30.9 per cent). For many respondents in Tamil Nadu migration seems to have been a rewarding choice: 55.6 per cent stated that they were betteroff than before emigration. Only 16 per cent said that they were worse-off after emigration. At the same time, 46.9 per cent of the respondents were revealed to be problematically indebted at least to some extent (Table 9.8) and for many of the returnees this was mainly caused by taking a loan to proceed overseas. Re-emigration is being considered by 77.8 per cent of the respondents in Tamil Nadu, with 50.6 per cent of them being absolutely sure about it. This is a very important observation, because it indicates that for many migrants in Tamil Nadu, despite the economic crisis, reintegration is not really considered or convincingly taking place. Many still visualize their immediate future abroad and do not fully focus on setting up a living and working life in Tamil Nadu. Of the potential re-emigrants, 42.6 per cent wants to go to the same country as last time, 36.1 per cent to a different country and 21.3 per cent are not sure.

Andhra Pradesh Narsingi is a small town located about 50 km from the metropolis of Hyderabad in the state of Andhra Pradesh. Although Narsingi is well-connected with Hyderabad, economic activity in this community appeared to be very low. Out of the 26 returned construction workers interviewed in Narsingi, 15 respondents did not have any education at all (57.7 Table 9.8: Debt problems: Tamil Nadu Debt problems Very Problematic Problematic Some Problems No Problems Total Source: Survey conducted by the author.

Number

Per cent

24 11 3 43 81

29.6 13.6 3.7 53.1 100.0

Low-skilled Indian Construction Workers Y 163

per cent), another three had only finished primary school and three had finished middle school. Most respondents were active parttime in some informal occupation like manual work (12), agriculture (seven) or construction (three), but nobody was employed on a formal basis. The employment opportunities in Narsingi are clearly very limited. The vast majority of respondents in Narsingi had gone to Dubai (23 respondents; 88.5 per cent), while the others had been to Saudi Arabia. For everybody, except one, this had been the only emigration project so far. In terms of profession, 22 had simply been helpers while abroad, being unskilled to perform any other job. Higher salaries and repayment of debts were the main reasons to take up the adventure. While abroad, contact with relatives at home was scarce, with 22 being in touch with their family only once a month or less, while temporary leaves during the contract did not occur. The majority of labour migrants from Narsingi had gone to the Gulf on a tourist visa, making it illegal for them to actually work there. However, these men were unaware of their impending illegal status, or at least of the full consequences of this situation. Nagaraju says: ‘I went on a visit visa to Dubai, so after a while I became undocumented. I was caught by the police and spent two weeks in jail. After that I managed to borrow money from friends to pay the return ticket, so I could come back to India’.11 Ganesh went to Dubai on a visit visa as well. When he got exploited by the second company he worked for, he handed himself over to the police. ‘The private agency which had recruited me showed understanding and borrowed money for the return ticket. However, now they are threatening me to return the money’.12 Rajkumar went on a work permit to Saudi Arabia, but the company he was supposed to work for paid very low wages. ‘So I left that employer and started working outside, but I was legally not allowed to do that. I was arrested and sent to jail. Another family from my home village lent me money for the return ticket, which my family later paid back’.13 Many of the respondents in Narsingi have been exploited and were paid much lower salary than promised. Some of the respondents had been caught by the police and spent time in jail. Others Interview with Nagaraju, Narsingi, Andhra Pradesh, 15 March 2010. Interview with Ganesh, Narsingi, Andhra Pradesh, 15 March 2010. 13 Interview with Rajkumar, Narsingi, Andhra Pradesh, 15 March 2010. 11 12

164 Y Auke Boere

had handed themselves over to the police or called in the help of the Indian embassy when they discovered their illegal status. They only managed to go back to Narsingi by taking another loan from relatives or acquaintances in Narsingi in addition to the loan they had already taken from private lenders to go abroad and loans that many had already taken before leaving. After return, debts for these migrants had only multiplied. In Narsingi, everybody’s financial situation had become worse, most often much worse, and everybody was heavily indebted, mainly due to their emigration. After coming back they found themselves in the same occasional (under)employment (73.1 per cent) for low salaries again, making the situation even more unfavourable. The hard experiences in the Gulf have demotivated the majority from considering re-emigration, although eight of them still say they would give it another try (30.7 per cent). The case of Narsingi makes it clear that there is dire need for stronger institutional support of (potential) migrants, not only before or during emigration, but also after return, when problems continue at home. Lack of employment opportunities in the home community make it impossible to repay the towering debts. Therefore the last section of this article will go into policies for return migrants that (scarcely) exist in India.

Policies for Indian Return Migrants Although there are multiple institutions which try to address the needs of Indian labour migrants before they emigrate for instance, by providing pre-departure trainings — or during emigration, hardly any support policies exist for return migrants. In fact, none of the respondents in the survey indicated having received support from any institution whatsoever when back in India, except for bank loans (with concomitant interest rates) in a few cases. In Kerala, especially, it seems that return migrants need their awareness raised about the possibilities that actually exist for them to receive support.

Non-resident Keralites’ Affairs Department The Non-Resident Keralites’ Affairs Department (NORKA) is a special government department for international and internal migrants from Kerala, which also provides some schemes for return migrants. Returnees are still recognized by NORKA as NonResident Keralites (NRKs) for a certain amount of time, depending

Low-skilled Indian Construction Workers Y 165

on how long the person has been working abroad. For example, when somebody has been working in the Gulf for 25 years, he is still considered an NRK and has the right to certain welfare schemes. Norka-Roots is the field and implementation agency of the NORKA. According to S. M. Najeeb of Norka-Roots, an important scheme that NORKA has set up for return migrants is the NRK Welfare Fund (also Pravasis Welfare Fund), set up in January 2009. Keralites in the age group of 18–55 working abroad are the main contributors and can donate a minimum of ` 300 a month. The government of Kerala also contributes to the fund, and the returnees can also participate in the fund for a minimum contribution of ` 100. With this fund only pension schemes are created currently. This scheme is meant for the payment of pension to the members and deemed members who have completed 60 years of age and remitted contribution for not less than five years. Also, family pensions are paid from this scheme on the death of a member or a deemed member who has remitted contribution for not less than five years. About 35,000–40,000 members have already enrolled, but considering the total number of Indian expatriates this is only a fraction. In the future the fund should also be used for purposes like insurance benefits, scholarships benefits, etc.14

Palamoori Migrant Labour Union The Palamoori Migrant Labour Union (PMLU), based in Mahabubnagar (formerly named Palamooru) in Andhra Pradesh, claims to be the only trade union in India that solely works with migrant workers. They have been very active with repatriating Indian construction workers (mainly from the Gulf) who went on a tourist visa and ended up in jail. The PMLU does not have specific policies for return migrants, but they do form labour cooperative societies for this group. This means that if there is any government construction work available, the PMLU will write to these labour cooperative societies to make the return migrants aware of the availability of this opening. Apart from this, the PMLU pushes the government to create employment opportunities for return migrants.15 14 Source: http://www.norka.gov.in/ (accessed 10 August 2010); Interview with S. M. Najeeb, Thiruvananthapuram, Kerala, 17 March 2010. 15 Interview with P. Narayanaswamy, Chairman of PMLU, Mahabubnagar, Andhra Pradesh, 6 May 2010.

166 Y Auke Boere

Other trade unions like the Tamil Nadu Construction and Unorganized Workers Federation (TCWF) and the Kerala Kettida Nirmana Thozhilali Congress (KKNTC), counterparts of the PMLU in Tamil Nadu and Kerala respectively, perform similar tasks, but all of them are for now mainly focused on assisting the migrants before (pre-departure and para-legal trainings) and during emigration. Apparently, the number of policies in India specifically aimed at return migrants is very limited, since the above was all that was found. Professor Bernard d’Sami from Loyola College in Chennai underlines this lack of attention, especially by the Indian government, for migrants in general and return migrants in particular. According to him, attention to migrant issues is still in an infancy state.16

Concluding Remarks and Recommendations The story of the low- and unskilled return migrant who worked in the construction sector is versatile. On the one hand, there is the successful returnee who worked for years in a construction company for a decent wage and has managed to build a concrete house for himself and his family; on the other hand, there is the wrecked returnee who has been exploited throughout the whole process of migration and has to find a way to deal with his enormous burden of debt. Depending on many factors, some migrants succeed and others fail. It is as a consequence of this that reintegration patterns and tendencies to re-emigrate are shaped. A wide range of factors can be decisive in the success rate of a migration project and these precluding phases are considered essential here for explaining the period after emigration. They elucidate to a large extent how the situation of the individual returnee and his family back in south India was created. The place of origin seems to be a major structural factor that influences a returnee’s destiny to a great extent. Before emigration, it makes a huge difference whether a migrant lived in a relatively affluent community in Kerala or one of the communities in Andhra Pradesh described in this chapter. Migrants and potential migrants influence each other and to a large degree help determine each others’ paths. At the same time, recruitment agencies that choose certain communities for their practices make it more likely that migrants in 16

Interview with Bernard d’Sami, Chennai, Tamil Nadu, 18 February 2010.

Low-skilled Indian Construction Workers Y 167

these communities make similar choices, most notably going to the same destination country and working for the same company. Hence, migrants from the same origin communities will often have similar experiences. Consequently, there is a strong correlation between place of origin, destination country and situation after return. Most unlucky respondents from Narsingi for example had been to Dubai, while more successful migration experiences were found in Tamil Nadu from where a large number of returnees had been to Singapore. After their rough experiences in Dubai, the returnees in Narsingi found themselves in dreadful financial situations with unbearable debts and no prospects of a decent future. Many respondents in Tamil Nadu, however, stated that they were better-off than before going abroad and their future seemed to look much brighter. The major reintegration issues that returnees had to deal with were debt problems. These debts are often caused through the investments made for the migration project, e.g., for tickets, visas, passports and fees for the recruitment agents and become direr due to interest rates. The migration project has often led to so many debts that the precarious financial situation has become unsolvable, especially in Narsingi. Debt problems add to other economic problems like unemployment, underemployment and low salaries. Employment is found again, often within one year, but frequently in a low-paid job or one with insufficient working hours. What should be kept in mind always is that re-emigration is an important phenomenon in Indian low- and unskilled labour migration. More than half of the respondents in this research indicated that they considered going abroad again and this makes it clear that reintegration is for many returnees not self-evident, or at least not granted priority. Policies specifically designed for return migrants are scarce. Clear policies have been established only in Kerala. The government department for NRKs — NORKA — provides an important welfare fund, which will be extended and diversified in the nearby future. Given the high number of returnees who keep on coming back to India with debts and the incessant reports about terrible labour and living conditions and foul treatment in many destination countries, the need for putting a stop to these malpractices during the migration project is evident. Unavoidably, these harsh experiences leave a mental imprint on the returnees and have a bearing on their

168 Y Auke Boere

life in India after return. The government of India and state governments, especially those of Tamil Nadu, Kerala and Andhra Pradesh should make more efforts to stop the exploitation and extortion of migrant workers that persistently takes place during the entire migration process. As long as malpractices keep on taking place in Gulf and South-East Asia migration, a better solution for underemployment in peripheral communities in south India could be migration to urban areas within India, where employment opportunities for construction workers are increasing and wages are rising. India’s Central Statistical Organization reported a 7.5 per cent increase in construction employment within India in late August 2010 (Al Jazeera 2010). In contrast to especially the nearly bankrupt emirate of Dubai, this indicates that India’s urban areas provide great opportunities for people from less affluent rural regions in India. An internal migration project also brings fewer risks and will prohibit most debt problems from evolving in the first place. Therefore, it is recommended here to government institutions, NGOs and trade unions to promote internal migration to urban areas within India among potential migrants instead of a risky adventure abroad.

References Cassarion, Jean-Pierre. 2004. ‘Theorising Return Migration: The Conceptual Approach to Return Migrants Revisited’, International Journal on Multicultural Societies, 6 (2): 253–79. Graeme, Hugo. 2003. ‘Circular Migration: Keeping Development Rolling?’. University of Adelaide and Migration Policy Institute. http://www. migrationinformation.org/feature/display.cfm?ID=129 (accessed 19 March 2010). De Haas, Hein. 2005. ‘International Migration, Remittances and Development: Myths and Facts’, Third World Quarterly, 26 (8): 1243–58. Taran, Patrick. 2009. ‘The Impact of the Financial Crisis on Migrant Workers’. Paper presented at the conference ‘17th Organization for Security and Cooperation in Europe (OSCE) Economic and Environmental Forum, Second Preparatory Conference’, 16–17 March, Tirana. http://www.osce.org/ eea/36454 (accessed 7 November 2011). Al Jazeera. 2010. ‘India’s Growth Picks up Speed’, 31 August. http://english. aljazeera.net/business/2010/08/20108319435395930.html (accessed 18 October 2010).

10 Impact of the Global Recession on Migration and Remittances The Kerala Experience K. C. Zachariah S. Irudaya Rajan This chapter on the impact of the global recession on Kerala emigra-

tion was undertaken by the Research Unit on International Migration at the Centre for Development Studies (CDS), Thiruvananthapuram, at the request of the Department of Non-resident Keralite Affairs (NORKA), Government of Kerala. The objectives of the study were clearly spelt out by NORKA in terms of a framework of questions on emigration from and return emigration to Kerala since the beginning of the global recession. The broad objective of this chapter is to assess the impact of the global economic recession on the employment and economic conditions of emigrants from Kerala. How many of Kerala’s emigrants lost their jobs abroad since the beginning of the global recession and how many have been forced to return home? What are the social and economic implications of such job losses and the subsequent return of the emigrants for their families and for the Kerala economy? What measures can be adopted to minimize the adverse impact of return emigration on the socio-economic scenario of the state? What measures can be adopted to rehabilitate the emigrants who have returned as a direct consequence of the global recession, particularly those who were employed abroad in low-level jobs? In the last two years, the CDS had conducted two large-scale migration surveys in Kerala and both were financed by the NORKA. The first survey was carried out during March–July 2007. It covered a sample of 10,000 households selected at random from all the 63 taluks1 in the state. Among these, 1,768 had an emigrant (EMI), 1

A taluk is an administrative division under a district.

170 Y K. C. Zachariah and S. Irudaya Rajan

1,065 had a return emigrant (REM) and 2,556 households had either an EMI or an REM (for more details on the Kerala Migration Survey [KMS] 2007, see Zachariah and Rajan 2007). The second study was carried out during August–December 2008, just before the global recession. It was based on a larger sample of 15,000 households selected at random from all the 63 taluks in the state. About 2,702 of the 15,000 households had an EMI from that household. Similarly 1,765 of the sample households had an REM and 3,981 households had either an EMI or an REM (for more details on the KMS 2008, see Zachariah and Rajan 2009). The new study (Return Migrants Study 2009 or RMS 2009), is designed to cover 6,537 households (2,556 from the 2007 study and 3,981 from the 2008 study). The non-response was 113. Data was collected from 6,424 households. The field work was carried out during 16 June–7 September 2009. The following data was collected through surveys and used for this report: (a) Current information on members of selected households in the 2007 and 2008 studies. (b) Current information from the (old) EMIs and the REMs identified as such in the 2007 and 2008 studies. (c) Current information on new EMIs and REMs (those not included in [b] above). The data enables estimation of the number of EMIs and proportion of those who underwent changes in migration and employment status since the previous survey (2007 or 2008). It is also possible to study the socio-economic conditions of those who were adversely affected by the global recession with respect to employment. Although all the 6,537 households which were identified to have an EMI or REM in the 2007 and 2008 surveys were expected to be visited by an investigator, only 6,424 households could be contacted as others had moved out of the area without leaving any contact address. The necessary information was collected from these households with the help of a structured questionnaire. Data from these 6,307 households form the basis for this chapter. This chapter uses data related to EMIs and REMs identified as such in the KMS 2008 (not in the KMS 2007), and new EMIs (those who moved after December 2008) REMs (those who came back to

Impact of the Global Recession on Migration and Remittances Y 171

Kerala after December 2008) in households with an EMI or REM according to the KMS 2008.

General Information about the Sample Households The sample households in this study are not a random sample from the 8 million households in Kerala in 2009. The sample includes only those households with an EMI or a REM. Therefore, it is not possible to make generalizations from the sample about the total number of migrants from the state, total remittances to the state, etc. However, the sample would provide an unbiased estimate of parameters directly related to EMIs. An example is the proportion of the 2008 EMIs who returned to Kerala during the time period between the 2008 survey and the 2009 survey. If this proportion is applied to the total number of EMIs from Kerala in 2008 (2,193,412), it is possible to get an unbiased estimate of the number of EMIs who had returned to Kerala since the 2008 survey. As the sampling was done independently in the various districts and as the sampling fraction was different in different districts, estimates were obtained independently for the 14 districts and these estimates have been added up to get the state-level estimate.

Unemployment among Emigrants Due to the Recession Important questions to consider are: how many of the EMIs from Kerala who continue to stay abroad have lost their jobs in recent months as a result of the global recession? How many have become unemployed?

Employment Before and After Recession We start this analysis by comparing the employment situation of EMIs before the recession period (October–December 2008) with that of the same set of EMIs after the recession (June–August 2009) (see Table 10.1). A total of 3,953 EMIs from the KMS 2008 survey were included in the RMS 2009 survey. Of them, 3,456 were employed in 2008, 39 were unemployed, and 458 were not in the labour force. The employed constituted 87.4 per cent of the total number of EMIs and the unemployed were 1.0 per cent. About 11.6 per cent were not in the labour force.

172 Y K. C. Zachariah and S. Irudaya Rajan Table 10.1:

EMIs by employment status in 2008 and 2009 Status in 2008

Status in 2009

Employed

Unemployed

Not in LFa

Total

Employed Unemployed Not in LFa Total Employed Unemployed Not in LFa Total Employed Unemployed Not in LFa Total

3,301.0 98.0 57.0 3,456.0 95.5 2.8 1.6 100.0 98.3 89.1 11.8 87.4

27.0 2.0 10.0 39.0 69.2 5.1 25.6 100.0 0.8 1.8 2.1 1.0

30.0 10.0 418.0 458.0 6.6 2.2 91.3 100.0 0.9 9.1 86.2 11.6

3,358.0 110.0 485.0 3,953.0 84.9 2.8 12.3 100.0 100.0 100.0 100.0 100.0

Source: Zachariah and Rajan (2010). a LF = Labour Force. Note:

By the time of the 2009 survey (June–August 2009), the number of employed EMIs had decreased to 3,358, and the number of unemployed EMIs had increased to 110. Thus, the number of unemployed EMIs increased by 71 between the years 2008 and 2009. The increase could be attributed to the recession.

Unemployment during the Recession An estimate of the number of those unemployed among the EMIs in 2008 was 22,610 (Table 10.2). The corresponding number among the same number of EMIs in 2009 was 60,099. Thus, the increase in Table 10.2: Estimate of unemployed among the 2008 EMIs in 2009 in Kerala Unemployed 2008 Kerala

39

Number of Unemployed in 2008 Number of Unemployed in 2009 Increase Source: Zachariah and Rajan (2010).

EMIs in 2008

2009 Sample 110

3,953

Kerala

Unemployed 2008

2,193,412 22,610 22,610 60,099 37,489

2009 60,099

Impact of the Global Recession on Migration and Remittances Y 173

the number of unemployed EMIs is 37,489. This is an estimate of the number of EMIs who became unemployed during 2008–2009 (the recession period). Thus, about 37,000 additional EMIs became unemployed during the eight-month period between November 2008 and June 2009 (recession period). This, however, is not the same as the number of EMIs who lost their job.

Loss of Jobs during the Recession In order for a person to lose his/her job, he/she must first be employed. Table 10.3 gives the distribution of EMIs by employment status in 2008 and 2009 (state-level estimate). Table 10.3: EMIs by employment status, Kerala, 2008 and 2009 Status in 2008 Status in 2009

Employed

Unemployed

Not in LFa

Total EMIs

Employed Unemployed Not in LFa Total

1,831,635 54,378 31,628 1,917,640

14,982 1,110 5,549 21,640

16,646 5,549 231,937 254,132

1,863,263 61,036 2,69,113 2,193,412

Source: Zachariah and Rajan (2010). a LF = Labour Force. Note:

Among the 1,917,640 EMIs who were employed in 2008, 54,378 became unemployed by 2009, i.e., during the recession months. This is an estimate of the job losses abroad among Kerala EMIs during the months of recession. Job loss is much higher than the increase in unemployment. However the net job loss is very much lower. At the time when 54,378 employed EMIs lost their job, 14,982 unemployed EMIs and 16,646 EMIs who were not looking for a job in 2008 got employment during the recession months. Thus, net employment loss is only 22,750 (54378–14982–16646). How significant is this estimate of the number of EMIs who became unemployed or the number of employed EMIs who lost their job due to the recession? 54,000 EMIs who lost their job or 37,000 EMIs who became unemployed are by no means very huge numbers considering that there were more than 2 million EMIs from Kerala in 2008. The unemployed or those who lost their job are less than 3 per cent of the total number of EMIs. It is also instructive to

174 Y K. C. Zachariah and S. Irudaya Rajan

compare the unemployment due to the recession with unemployment among these same EMIs in Kerala before emigration. The number of unemployed EMIs following the recession is 110 (in the sample). The number of those who were unemployed among them in Kerala before their emigration was 805, more than seven times the number of those unemployed after the recession (Figure 10.1). Thus, although unemployment has increased considerably (relatively speaking) the unemployment among Kerala EMIs after the recession is very much lower than the unemployment among them in Kerala before emigration. In spite of the impact of recession, from the point of view of employment, the EMIs are in a much better position than they were before they left Kerala. Figure 10.1: Unemployment: Before and after emigration, and after recession

Source: Zachariah and Rajan (2010).

Are all these job losses due to the global recession? It is certainly not unusual for employed persons, migrants or non-migrants, to become unemployed even in ‘normal’ times (when there is no recession). Some EMIs lose their jobs every year. We can conclude that all the job losses among the EMIs in 2008 was due to the global recession during the first half of 2009 only if such job losses do not occur in normal years. Unfortunately, comparable data for a non-recession year are not available. Some approximations are, however, possible. The unemployment rates (as a percentage of the labour force) among EMIs in some non-recession years are available from KMS 1998, 2003 and 2008.

Impact of the Global Recession on Migration and Remittances Y 175

They are given as follows: 1998 2003 2008 2009

0.73 1.89 1.12 3.17 (recession period)

Source: KMS, 1998, 2003 and 2008.

The rates for the non-recession years are much lower than the rate for 2009 (the recession months), indicating that the global recession could indeed have played a major role in raising the unemployment rate in 2009. A second source of data indicating the role of recession is the panel data from KMS 2003, which gives the sector of employment of about 4,000 persons in 1998 and 2003:

EMIs in 1998 EMIs in 2003 Increase

Employed

Unemployed

958 878 –80

6 15 +9

Increase in number unemployed as per cent of employed in 1998

= 0.94

Source: KMS, 2003.

Thus, the percentage increase in unemployment among the EMIs during a five-year non-recession period is relatively very small when compared to the corresponding increase during the eight-month period of the global recession. A logical conclusion from these analyses is that global recession had an effect on the employment situation among Kerala EMIs abroad. About 54,000 employed EMIs lost their job and 37,000 additional EMIs became unemployed during the first half of 2009. It would not be very far from the truth to attribute almost all of this additional unemployment to global recession. Our answer to the first question raised by NORKA is that there was a net increase in unemployment of 37,000 among Kerala EMIs and 54,000 employed EMIs lost their job during the recession.

Transition to the Unemployment It was mentioned that the number of unemployed EMIs had increased by 71 persons (in the sample). Which sector of economic activity did they come from? Table 10.4 (sample data) shows that

176 Y K. C. Zachariah and S. Irudaya Rajan Table 10.4: Sector of economic activity of the unemployed, 2008– 2009 Activity status

2008a

2009b

2 – 23 1 – 1 – 2 – 2 6 2 39

4 2 56 4 – 1 31 2 – 1 9 – 110

Employed in Central/State Government Employed in Semi-Government/Aided Organizations Employed in Private Sector Self-Employment Unpaid Family Work Labour in Agricultural Sector Labour in Non-Agricultural Sector Job Seekers Job not Required Students Household Work Pensioners Total Source: Zachariah and Rajan (2010). a Economic activity in 2009 of those unemployed in 2008. Note: b Economic activity in 2008 of those unemployed in 2009.

two persons came from the government service category, two from semi-government services, 33 from the private sector, three from the self-employment sector and 31 from the non-agricultural labour sector. These numbers add up to 71. The changes in the sectors outside the labour force cancel each other out. Thus, most of the newly unemployed came from the private sector and from among the labourers in the non-agricultural sector.

Transition of the Employed While increase in unemployment during the recession is the principal focus of this chapter, transitions in the employment sector (movement from one sector of the economy to another) throw additional light on the impact of the recession on the migrant’s employment status. How have the EMIs manoeuvred themselves through the various employment avenues to avoid potential job loss? This is an important aspect of the impact of recession on employment of the EMIs. A person faced with imminent job loss moves to another job (if possible), sometimes with a lower salary, to avoid becoming unemployed. Information about such transitions is equally instructive in

Impact of the Global Recession on Migration and Remittances Y 177

measuring the adverse impact of recession and devising policies to deal with them. In this case, it seems that information on such movements is more significant than that on job losses. Between 2008 and 2009, the number of employed persons decreased by 98 (Table 10.5), the number of unemployed persons increased by 71 and the number of those not in the labour force increased by 27. As a result, some employment sectors gained and some others lost. Which were the sectors that gained and which were the ones that lost? Table 10.5 indicates that the number of EMIs working in the private sector increased by about 38 per cent. At the same time, the number of EMIs working as labourers in the nonagricultural sector has decreased by 100 per cent. There were also significant losses (relatively) in employment in government and semi-government sectors. Table 10.5: Economic activity of EMIs in 2008 and 2009, KMS 2008 Activity status

2008

2009

Employed in Government Sector Semi-Government Sector Private Sector Self-Employment Unpaid Family Worker Agriculture Sector Labourer Non-Agriculture Sector Labourer Job Seekers Job not Needed Students Household Work Pensioners Others Total Employed Unemployed Not in Labour Force

79 65 2,135 129 8 7 1,033 39 4 218 122 4 110 3,953 3,456 39 458

74 12 2,938 90 15 228 1 110 4 218 135 7 121 3,953 3,358 110 485

Difference –5 –53 803 –39 7 221 –1,032 71 0 0 13 3 11 0 –98 71 27

Source: Zachariah and Rajan (2010).

At the same time, some persons who were not in the labour force in 2008 (those engaged in household duties) got employment during the recession period.

178 Y K. C. Zachariah and S. Irudaya Rajan

Characteristics of the Unemployed Data on the characteristics of the unemployed provides important additional information needed to formulate policies to redress their problems. From which district of Kerala did they hail? In which countries did they live? How many are males and how many females?

Country of Residence of the Unemployed More than half the number of unemployed (in 2009) among the 2008 emigrants is from the UAE (53 per cent). About 14 per cent are from Saudi Arabia. Kuwait is the country of residence with the third largest number of the unemployed EMIs. Relatively fewer EMIs became unemployed in other destination countries (see Table 10.6). Table 10.6: Country of residence of EMIs who became unemployed in 2009 Country UAE Saudi Arabia Kuwait Oman Qatar Bahrain Maldives US Others Total

Unemployed

Per cent unemployed

EMIs

Unemployed per 100 EMIs

58 15 7 4 3 4 1 1 17 110

52.7 13.6 6.4 3.6 2.7 3.6 0.9 0.9 15.5 100.0

1,469 867 214 279 202 186 7 186 543 3953

3.9 1.7 3.3 1.4 1.5 2.2 14.3 0.5 3.1 2.8

Source: Zachariah and Rajan (2010).

Sex More than 85 per cent of the unemployed among the EMIs in 2008 were males. Females constituted 14.5 per cent of the total. However, the unemployment rate is higher among females, 5.93 per cent compared to 2.94 per cent among males. The overall unemployment rate is 3.2 per cent (Table 10.7).

Age The largest number of unemployed EMIs is in the age group of 25–29 among all the age groups. This group also has the highest

Impact of the Global Recession on Migration and Remittances Y 179 Table 10.7: Unemployment among males and females of 2008 EMIs in 2009 Sex Males Females Total

Unemployed

Labour force

Percentage

Rate (%)

94 16 110

3198 270 3468

85.5 14.5 100.0

2.9 5.9 3.2

Source: Zachariah and Rajan (2010).

unemployment rate — 4.72 per cent compared to the overall average of 3.2 per cent (Figure 10.2). Figure 10.2: Unemployed: Distribution by age (in per cent)

Source: Zachariah and Rajan (2010).

Levels of Schooling Nearly one-third (32.7 per cent) of the unemployed EMIs are from among those with 10th standard education. A fairly large number of the unemployed EMIs (18.2 per cent) are those with 12th standard education. However, there is no systematic upward or downward relation between the unemployment ratio and years of schooling. This relation between the years of schooling and unemployment rate among EMIs is given in Table 10.8.

Return Emigration Due to the Recession How many Kerala EMIs returned to Kerala in recent months as a consequence of the global recession?

180 Y K. C. Zachariah and S. Irudaya Rajan Table 10.8: Unemployment percentage and rate by level of schooling, 2009 Unemployment Education Less than Class 6 Classes 6–9 Classes 10–12 Classes 12+ Others Total

Unemployed

Labour force

Rate

8 14 57 29 2 110

190 686 1,574 1,006 12 3,468

4.2 2.0 3.6 2.9 16.7 3.2

% 7.3 12.7 51.8 26.4 1.8 100.0

Source: Zachariah and Rajan (2010).

EMIs of 2008 by Migration Status in 2009 The 2009 survey collected information on the current migration status of all the 3,953 EMIs identified in the 2008 survey. How many of them remained EMIs and how many are in Kerala as REMs? The data is given in Table 10.9. Table 10.9: Emigrants of 2008 by migration status in 2009 in the RMS 2009

EMI REM Total

Number

Per cent

3,649 304 3,953

92.3 7.7 100.0

Source: Zachariah and Rajan (2010).

Of the 3,953 EMIs (sample) in the KMS 2008, 304 or 7.7 per cent returned to Kerala by the time of the 2009 survey (during the recession period).

Return Emigration: State-level Estimate At the state level, about 7.7 per cent of the EMIs of 2008 became REMs by 2009. If district-wise ratios are applied to the total EMI in 2008, the number of EMIs who returned to Kerala in 2009 would be 173,339. This is for the period between 2008 KMS and 2009 RMS, roughly November 2008 to June 2009. This period coincides with the months of recession. Thus, an estimate of recession-related return

Impact of the Global Recession on Migration and Remittances Y 181

emigration among Kerala EMIs is 173,000 (roughly for a period of eight months). Thus, the average return emigration for a year varies from 102,876 during 1996–1998, to 70,005 per year during 2001–2003, and to 139,363 per year during 2006–2008 (Table 10.10). Taking the estimate for the latest year, for an eight-month period, return emigration during 2006–2008 would be 93,000. We have thus the following results. Return emigration during: Recession Period November–June 2009 (eight months) = 173,000 Non-Recession Period average 2006–2008 (average for = 93,900 eight months) Return emigration has indeed accelerated since the beginning of the global recession. An estimate of the number of EMIs who returned to Kerala during November 2008 to June 2009 is 173,000. But can we conclude from this analysis that each of the 173,000 EMIs that returned to Kerala did so as a result of the global recession? Table 10.10: Average REMs per year, 1996–2008 Three year average (per year) 1996–1998 102,876a Three year average (per year) 2001–2003 70,005b Three year average (per year) 2006–2008 139,363c Source: Zachariah and Rajan (2010). a Note: Based on 1998 Survey. b Based on 2003 Survey. c Based on 2008 Survey.

Return Emigration due to the Recession The present survey can provide an answer to this question to a large extent. This survey included a question on the reasons for the migrant’s return. The answers are given in Table 10.11. Table 10.11 indicates that EMIs generally return to Kerala due to a number of reasons. Expiry of contract is one cause for return. This was a major reason cited by the REMs in 1998 and 2008 (the nonrecession years). Similarly, many EMIs return because conditions back home are such that their presence is required there. This is so particularly in the case of female REMs. Their return has nothing to

Source: Zachariah and Rajan (2010).

Lost Job due to Financial Crisis Expiry of Contract Compulsory Repatriation Low Wages or Not Getting Promised Wages Poor Working Conditions Harsh Behaviour of Employer Ill Health Problems at Home Others Total

Reasons for the return 0 5 0 1 0 0 3 9 30 48

49 14 3 22 24 39 256

Female

65 30 10

Male

50 14 3 25 33 69 304

65 35 10

Total

19.1 5.5 1.2 8.6 9.4 15.2 100.0

25.4 11.7 3.9

Male

Table 10.11: Reasons for return of REMs among EMIs of 2008 in 2009

2.1 0.0 0.0 6.3 18.8 62.5 100.0

0.0 10.4 0.0

Female

16.4 4.6 1.0 8.2 10.9 22.7 100.0

21.4 11.5 3.3

Total

52.7 57.3 58.2 66.5 77.3 100.0 100.0

21.4 32.9 36.2

Cumulative

91,085 99,052 100,759 114,986 133,765 173,032 173,032

37,022 56,940 62,631

Number

Impact of the Global Recession on Migration and Remittances Y 183

do with the recession. Another case is when EMIs return because they cannot withstand the hostile climatic conditions in some Gulf countries. This too is not connected in any way to the recession. Others come back because they fail to get the salary or wages they were promised. Thus, most of the EMIs could have returned home in 2009 for reasons not related to the global recession. In 2009, the largest number of EMIs returned to Kerala because they lost their jobs due to the financial crisis — 21 per cent (or 37,000) quoted this as the reason in the sample. Another 3.3 per cent underwent compulsory retirement. If they are also added, about 24.7 per cent or about 43,000 EMIs returned due to the recession. About 11.5 per cent returned because their contract had expired and was not renewed. This is a common experience among EMIs in the Gulf region and not substantially related to the recession as such. EMIs cited this as the reason for their return in 2003 and 1998 too. However, for argument’s sake, if such returnees are also added, the number would increase to 36.2 per cent or 63,000 persons. The number of return EMIs attributable to the global recession would then be at the most 63,000. Therefore, the number of EMIs who returned home due to recession-related reasons could be a minimum of 37,000 and a maximum of 63,000.

From Return Emigrants to Emigrants Some of the REMs identified in the KMS 2008 re-migrated out of the country to become an EMI once again. They were about 9.55 per cent (or about 110,494) of the total REMs. Thus, in the first half of 2009 (the recession period), 173,000 EMIs returned to Kerala and 110,000 former REMs became EMIs once again. The net change is the return of 63,000 former EMIs back to Kerala.

Characteristics of the Return Emigrants Country of Residence of Emigrants The largest number of REMs came from the UAE: 46.4 per cent. But the return emigration rate was highest among EMIs in Kuwait, at 15.0 per cent followed by the UAE with 9.6 per cent. Saudi Arabia contributed 23 per cent and Kuwait 11 per cent to the flow of REMs

184 Y K. C. Zachariah and S. Irudaya Rajan

to Kerala. Together with the UAE, these countries accounted for about 80 per cent of the REMs from among the 2008 emigrants (Table 10.12). Table 10.12: Percentage and return rate by country of origin of REMs, 2009 Countries UAE Saudi Arabia Kuwait Oman Qatar Bahrain US Maldives Others Total

REMs

EMIs

Per cent

141 69 32 21 16 8 3 3 11 304

1,469 867 214 279 202 186 186 7 543 3,953

46.4 22.7 10.5 6.9 5.3 2.6 1.0 1.0 3.6 100.0

Rate per 100 EMIs 9.6 8.0 15.0 7.5 7.9 4.3 1.6 42.9 2.0 7.7

Source: Zachariah and Rajan (2010).

District of Origin of Return Emigrants The majority of REMs hailed from Malappuram district, i.e., 14.8 per cent. But return emigration as percentage of total EMIs is highest among those from Palakkad district at 10.3 per cent. Malappuram, Thiruvananthapuram, Thrissur and Kozhikode each had more than a tenth of the total REMs in the state. Palakkad, Thiruvananthapuram, Kannur also have a relatively high return emigration ratio (Figure 10.3).

Sex Composition As the EMIs are mostly males, the vast majority of those who returned are also males, i.e., 84 per cent of the total. However, the ratio of REMs is higher among females. While 7.6 per cent of male EMIs returned, the number is as much as 8.5 per cent among female EMIs (Table 10.13).

Age Composition The largest number of REMs is in the 25–35 age group, which accounts for 42.5 per cent of the total. However, the ratio of REMs is highest in the 15–19 age group and among those over 55 years of

Impact of the Global Recession on Migration and Remittances Y 185 Figure 10.3: REMs per 100 EMIs by district of origin of return migrants

Source: Zachariah and Rajan (2010).

Table 10.13: Sex composition of REMs among 2008 EMIs in 2009 Sex

EMI

REM

Per cent

Males Females Total

3390 563 3953

256 48 304

84.2 15.8 100.0

Rate (%) 7.6 8.5 7.7

Source: Zachariah and Rajan (2010).

age. More than 11 per cent of the youngsters in the 15–19 age category have returned. Similarly, more that 10 per cent of the elderly also returned during the recession. Children and the elderly were more prone to return during the recession than persons in the prime working ages (Figure 10.4).

Educational Attainment EMIs with educational level up to the 10th standard contributed the largest number of REMs (29.3 per cent). Those with 12 years of schooling constituted 17.4 per cent. However, the rate per EMI is not the highest in these groups (Table 10.14). The rate of return emigration from among the 2008 EMIs is relatively high among EMIs with less than six years of schooling. On the whole, the rate decreases with years of schooling, but seems to reach the highest point at the very fag end (Figure 10.5).

186 Y K. C. Zachariah and S. Irudaya Rajan Figure 10.4: Age distribution of REMs and return emigration rate by age

Source: Zachariah and Rajan (2010).

Table 10.14: Education: Percentage of total and percentage of EMIs

Less than Class 6 Classes 6–9 Classes 10–12 Classes 12+ Others Total

REM

EMI

REM per 100 EMI

Per cent

45 48 144 62 5 304

417 745 1,654 1,119 17 3,952

10.8 6.4 8.7 5.5 29.4 7.7

14.8 15.8 47.4 20.4 1.6 100.0

Source: Zachariah and Rajan (2010).

Economic Activity Among the EMIs who returned during the recession months, the largest proportion, i.e., 32 per cent, were unemployed. About 20.4 per cent of the returnees were those working in the private sector, 15.5 per cent were non-agricultural workers and 9 per cent were engaged in household work. Thus, about 46.1 per cent were working, 21.6 per cent were not in the labour force and 32.2 per cent were unemployed (Table 10.15).

Impact of the Global Recession on Migration and Remittances Y 187 Figure 10.5:

REMs per 100 EMIs by education

Source: Zachariah and Rajan (2010).

Table 10.15: REMs and EMIs: Percentage and ratio by employment sector, 2009

Government Semi-government Private Sector Self-employed Agricultural Labour Non-agricultural Labour Unpaid Family Worker Job Seekers Job not Required Students Household Work Pensioners Too Old to Work Too Young for School Disabled Others Total

REM

EMI

Percentage

Ratio

0 0 62 23 8 47 0 98 2 11 28 1 2 9 2 11 304

74 12 2,938 90 15 228 1 110 4 218 135 7 3 95 3 20 3,953

0.0 0.0 20.4 7.6 2.6 15.5 0.0 32.2 0.7 3.6 9.2 0.3 0.7 3.0 0.7 3.6 100.0

0.0 0.0 2.1 25.6 53.3 20.6 – 89.1 50.0 5.0 20.7 14.3 66.7 9.5 66.7 55.0 7.7

Source: Zachariah and Rajan (2010).

188 Y K. C. Zachariah and S. Irudaya Rajan

Impact of Unemployment and Return Emigration The second set of questions raised by NORKA relates to the impact of recession on EMIs, their families, and the state, and to policies to minimize the adverse consequences of their return. What is the social and economic impact of job loss and subsequent return on Kerala EMIs, their families and the Kerala economy? What measures can be adopted to minimise the adverse impact of their return on the socio-economic situation in the state? What measures can be adopted to rehabilitate the EMIs who returned as a direct consequence of the global recession, particularly those in low-level jobs? The impact of the global recession on EMIs is not only through job losses and return emigration, but also through reduction in salaries or wages.

Return Emigration due to Recession The number of EMIs who lost their jobs in the destination country and yet remained there is estimated to be about 37,000. Among them, those who returned to Kerala as a result of recession-related reasons are estimated to be a maximum of 63,000. Of this, only 28,000 were unemployed at the time of the 2009 survey and it is they who actually needed to be rehabilitated more than any other group (Table 10.16). However, to these REMs, we have to add the number of EMIs who have undergone salary cuts. This is a largely unknown number. But a good estimate of their magnitude could be gauged on the basis of the impact on the remittances that they send back home (see Table 10.16).

Impact of the Recession at the State Level While recession could affect the household of the REM in many ways — social, psychological and economic — this chapter is confined to the economic impact only. We assume that most of the other consequences come about through economic factors. Much of the economic impact occurs through the remittances which the households receive from their members working abroad. In the KMS, a distinction was made between ‘Household Remittances’ and ‘Total Remittances’. Studies like the KMS or the RMS can at best measure household remittances. Fortunately, it is household

Impact of the Global Recession on Migration and Remittances Y 189 Table 10.16: Number of REMs in Kerala who would need help in rehabilitation, 2009 Sample

Kerala

3,953 304

2,193,412 168,681

Among the EMI of 2008, those who returned Due to Recession as reasons for return

110

61,036

Among those who returned during the crisis (110), their status as EMI in 2008 Employed in 2008 Unemployed in 2008 Not in labour force in 2008 Total

106 1 3 110

58,817 555 1,665 61,036

Of the 107 employed EMI (2008) who returned Employed in 2009 Number of unemployed Not in labour force Total

54 50 2 106

29,963 27,744 1,110 58,817

Total EMI, 2008 REM among the EMI of 2008

Source: Zachariah and Rajan (2010). Note: Number of those who would require help because of unemployment = 27,744.

remittances that matter the most as far as the impact of the recession on households is concerned.2 Before proceeding further a word of clarification is necessary. As the sample of households included in this study has not been selected at random, it cannot provide an absolute measure of household remittances or any other kind of remittances in Kerala in 2009. What this study can do at best is to measure the relative change (per cent increase or decrease) in the remittances received by the EMI households between 2008 and 2009. Such a measure of relative change, however, is fairly adequate to measure the impact of the global recession on household remittances. Table 10.17 shows the total cash remittances and the districtwise cash remittances per household in 2008 and 2009. The table indicates that for the state as a whole, cash remittances increased by 5.9 per cent between the two 12-month periods, November 2007 KMS and RMS included questions about remittances received by any member of the household during the 12-month period prior to the survey. 2

337 283 255 155 129 18 142 329 199 477 274 61 165 221 3,045

Thiruvananthapuram Kollam Pathanamthitta Alappuzha Kottayam Idukki Ernakulam Thrissur Palakkad Malappuram Kozhikode Wayanad Kannur Kasaragode Kerala

Source: Zachariah and Rajan (2010).

2008

District

337 283 255 155 129 18 142 329 199 477 274 61 165 221 3,045

2009

Number of households 15,570,000 17,582,000 17,471,600 8,601,500 11,429,995 1,300,000 10,718,000 22,866,997 13,029,000 28,652,500 16,504,996 6,250,999 11,277,500 9,826,000 191,081,087

2008

Remittances

Table 10.17: Household cash remittances, 2008 and 2009

15,824,800 18,214,999 16,830,999 12,350,000 16,957,000 705,000 11,081,000 22,905,000 12,902,000 28,458,000 16,548,000 7,700,000 13,492,999 8,438,999 202,408,796

2009 46,202 62,127 68,516 55,494 88,605 72,222 75,479 69,505 65,472 60,068 60,237 102,475 68,348 44,462 62,752

2008 46,958 64,364 66,004 79,677 131,450 39,167 78,035 69,620 64,834 59,660 60,394 126,230 81,776 38,186 66,473

2009

Per household 1.6 3.6 –3.7 43.6 48.4 –45.8 3.4 0.2 –1.0 –0.7 0.3 23.2 19.6 –14.1 5.9

Per cent increase 2008–2009

Impact of the Global Recession on Migration and Remittances Y 191

to October 2008 (the non-recession period) and October 2008 to September 2009 (the recession period). This is quite an unexpected result. The recession is supposed to have decreased remittances, not increased them. The increase could, however, be partly explained by the fact that the remittances in 2009 are for a 12-month period which includes some non-recession months also. Secondly, the recession could, in some cases, actually increase remittances. The EMIs who chose to return home permanently as a result of the recession brought all their accumulated wealth with them, leading to some households showing very large increases in remittances and others showing decreases. The subsequent analysis shows that this is indeed the case. That there has been an increase in remittances during the recession period is supported by the statistics of South Asian countries such as Pakistan, Nepal, Bangladesh, Sri Lanka, etc., which indicate significant growth in remittances in 2009 (Rajan and Narayana 2010). The all-India statistics on remittances also support an increase in remittances to India.3 Therefore, the observed rise in remittances in Kerala in 2009, although somewhat unexpected, is highly plausible. Thus, the economic impact of recession at the state level is not a matter that merits much concern. Remittances to the state during the recession months seem to have increased by about 6 per cent.

Impact of the Recession by Religious Affiliation All the three religious groups show an increase in remittances per household, the highest increase being among the Christians. Remittances per household increased by 17 per cent among Christian households compared to a mere 1.9 per cent increase among the Muslim households and 5.4 per cent among the Hindus (Table 10.18).

Impact of the Recession at the Household Level Heterogeneity in the relative change in remittances becomes more evident when the analysis is done at the household level.

3 ‘Indian Banks have mobilized $2.7 billion during April–September 2009 from non-resident Indians (NRIs) against inflows worth $1.1 billion in the year ago period’ (Economic Times 2009: 4).

192 Y K. C. Zachariah and S. Irudaya Rajan Table 10.18: Average remittances per household, by religion, 2008 and 2009 Religion Hindus Christians Muslims Total

2008

2009

59,622 59,738 67,038 62,752

62,812 69,742 68,331 66,473

% Increase 2008–2009 5.4 16.8 1.9 5.9

Source: Zachariah and Rajan (2010).

Figure 10.6: Percentage of households that received larger, equal or smaller remittances in 2009 than in 2008

Source: Zachariah and Rajan (2010).

Earlier analysis indicated that all districts did not experience increase in remittances. Even in districts with positive growth, not all households reported the same. Some households received smaller remittances in 2009 as compared to 2008. Some households got larger remittances in 2009 than in 2008 (Figure 10.6). About 1.383 million (18.3 per cent) out of a total of 7.566 million households in Kerala in 2008 had received some amount of remittances in 2008. Among them, 605,000 households (43.7 per cent) received larger remittances in 2009 as compared to what they received in 2008. About 347,000 (25.1 per cent) households received the same amount in 2009 as they did in 2008. The remaining 431,000 households (31.1 per cent) received smaller amounts as remittances in 2009 vis-à-vis 2008.

Impact of the Global Recession on Migration and Remittances Y 193

It was mentioned above that 31 per cent of the households experienced decline in remittances, with the decrease being minimal in some cases, but quite significant in others.

Policies to Rehabilitate Return Emigrants Lastly, we look at policies and programmes to overcome the recession-related problems. What measures could be adopted to minimize the adverse impact on the socio-economic situation in the state? What measures can be adopted to rehabilitate the EMIs who have returned as a direct consequence of the global recession, particularly those who were employed abroad in low-level jobs? This study has brought to light some of the broad dimensions of the problems created by the global recession on Kerala EMIs — the number of EMIs who became unemployed, lost their jobs abroad, were forced to return to Kerala, the countries in which the returnees had been working, the districts they hailed from, the sector of economic activity they were engaged in before the recession, and their demographic and socio-economic characteristics. The study also gives a rough estimate of the number of households that received smaller amounts as household cash remittances in 2009 compared to what they received in 2008 and the number of households that received remittances in 2008, but not in 2009.

Groups that Deserve Special Consideration in Combating the Adverse Impact of the Recession (a) Families in Kerala of the 56,000 EMIs abroad who lost their jobs (b) Families in Kerala of an unknown number of EMIs abroad who suffered loss of income due to salary cuts and increase in the cost of living (c) 173,000 EMIs who returned home during the recession (d) 61,000 EMIs who returned in 2009 as a direct result of the recession (e) 28,000 REMs who are still unemployed ( f ) 431,000 Kerala households that received smaller amounts as remittances in 2009 compared with what they received in 2008

194 Y K. C. Zachariah and S. Irudaya Rajan

(g) 90,000 Kerala households that received no remittances in 2009 although they received remittances in 2008 This is useful background material for arriving at policies and programmes to minimize the adverse impact of recession on the households of the EMIs or REMs in the state. However, they are just background information and not policies or programmes. This chapter recommends that such policies and programmes be put together in consultation with NORKA officials (who have a wider grasp of the migration situation in the state), taking into consideration the results of this study. While putting together these policies and programmes, this chapter suggests that the following points should also be taken into consideration. First, the number of Kerala EMIs who returned due to the recession is not as large as is often reported to be. The number is unlikely to be more than 63,000. Among these returnees, only about 28,000 are unemployed. Second, even as EMIs returned to Kerala, new EMIs and former EMIs (who had returned earlier) are going back to the Gulf and other destination countries, thus regaining their EMI status. The recession has not been much of a deterrent and has not stood in their way for re-emigration. Third, household remittances at the state level have increased and not decreased during the recession months, although there are several households where remittances have decreased. About 6 per cent of the households that received remittances in 2008 did not receive any remittances in 2009. Fourth, NORKA has recently announced a number of new measures to help EMIs and REMs. These include (a) training programme for the prospective EMIs, and (b) financial assistance to returnees by way of loans from Kerala Financial Corporation (KFC). These measures would go some way in easing the problems of EMIs and REMs. The Government could undertake more such measures including one to provide future EMIs with multiple skills that would enable them to move from one employment sector to another according to the need of the day. Fifth, the present study seems to indicate that the efforts to utilize the expertise which the REMs have gained abroad for the state’s development are more likely to succeed now than at any time in the

Impact of the Global Recession on Migration and Remittances Y 195

past. A concerted effort to utilize these for skill up-gradation and industrial development in the state is worth undertaking in the present context. This would be the best rehabilitation package for REMs that the government can offer at present. Sixth, the global recession has affected not only EMIs but also non-migrants, persons who have never stepped out of the state. The global recession has affected most of the export-dependent industries in Kerala: coir, fishing and cashew, to name a few. Not only were the owners of the enterprises affected by the recession, the workers too felt the impact, some of whom have lost their jobs. Although these workers themselves do not bring in any foreign exchange directly, the industries which utilize their labour contribute significantly to the foreign exchange earnings of the country. Should not the package that is being developed to help the REMs equally benefit nonmigrant workers and others affected by the global recession? Lastly, the study recommends that the government should give as much or more importance to ‘problem-preventing measures’ aimed at future EMIs as to ‘problem-solving measures’ aimed at returning emigrants.

Acknowledgements The RMS 2009 was financed by the Department of Non-Resident Keralite Affairs (NORKA), Government of Kerala and executed by the Ministry of Overseas Indian Affairs (Government of India) Research Unit on International Migration at the Centre for Development Studies (CDS), Kerala. We are grateful to Mrs Sheela Thomas, Principal Secretary to the Chief Minister, and Secretary, NORKA for her continued support. An earlier version of this chapter was presented at an open seminar on 1 December 2009, chaired by Mrs Sheela Thomas, Secretary, NORKA; Professor D. N. Narayana, CDS; and Dr K. N. Harilal, Member, State Planning Board, Government of Kerala, as discussants. Comments received from the chairperson, discussants and participants of the seminar are gratefully acknowledged.

References Economic Times, 2009. ‘NRIs Double Funds Parked at Indian Banks to $2.7 bn’, 13 November. Rajan, S. Irudaya and D. Narayana. 2010. ‘The Financial Crisis in the Gulf and its Impact on South Asian Migrant Workers’. Working Paper no. 436, Centre

196 Y K. C. Zachariah and S. Irudaya Rajan for Development Studies, Thiruvananthapuram. Zachariah K. C. and S. Irudaya Rajan. 2007. ‘Migration, Remittances and Employment: Short-term Trends and Long-term Implications’. Working Paper no. 395, Centre for Development Studies, Thiruvananthapuram. ———. 2009. Migration and Development: The Kerala Experience. Daanish Publishers, New Delhi. ———. 2010. ‘Migration Monitoring Study 2008: Emigration and Remittances in the Context of Surge in Oil Prices’. Working Paper no. 424, Centre for Development Studies, Thiruvananthapuram.

11 Global Financial Crisis and Return of South Asian Gulf Migrants Patterns and Determinants of their Integration into Local Labour Markets Vinoj Abraham S. Irudaya Rajan The global financial crisis had many casualties including the financial sector debacle, export stagnation and falling oil prices. However, these effects were largely visible in the developed world and the oil-rich regions. Most of the developing world remained relatively insulated from the global crisis mainly due to its poor integration with the global market and also due to the strong regulatory role the governments still imparted in many of these countries.1 Yet, the crisis reached these countries mainly transmitted through its effects on international migration. No other part of the world depends on migration and remittances as a source of economic development than the South Asian region, where a large number of migrant workers have built their hope for a different and brighter future than their past. The crisis — in this case, the decline in the Gross Domestic Product (GDP) growth in general in the Middle East region — led to the loss of jobs and means of livelihood for many workers from South Asia (Parameswaran 2009; Rajan and Narayana 2010). However, the loss of jobs in the gulf is compounded by the fact that in their own home countries the rehabilitation and reintegration of these workers is tedious and often the returnees are thrust with forced choices.

The output of advanced economies grew at 3 per cent and –0.6 per cent during 2008 and 2009 respectively, while the corresponding growth rate for emerging and developing economies was much better at 6.1 and 2.4 per cent (IMF 2009). 1

198 Y Vinoj Abraham and S. Irudaya Rajan

Migrants returning to their home country would imply more workers in these developing economies which are already overburdened with shrinking jobs due to the effect of the crisis.2 Though the overall effect of the global crisis on South Asian developing economies has been less as compared to developed ones, given the fact that the inherent capacity of such economies is weak any decline in the growth rates would greatly diminish their ability to cope with an influx of return migrants. The moot question then, for the return migrants, boils down to their employment prospects in their home country and its determinants. This chapter tracks a set of return migrants in five South Asian countries and looks into the process of reintegration of these migrants into the labour markets of their home countries. This chapter is based on a primary survey conducted in five South Asian countries — Nepal, Sri Lanka, Bangladesh, Pakistan and India. The survey was conducted by the Centre for Development Studies (CDS) in 2009 as part of a larger study on the impact of the financial crisis on migration and funded jointly by the Asian Development Bank through their South Asia Network of Economic Research Institutes (SANEI) research funding initiative, and the Ministry of Overseas Indian Affairs, Government of India, hereafter referred to as the CDS survey of international migration (2009). The survey was administered by local academic institutions in their respective countries. A small sample of 50 return migrants (REMs) and their households were covered in each country, except in India, where the sample size was 250 REM households — in total the sample size being 450 households and REMs. The next section provides an analysis of the patterns in return migration and the third section deals with the patterns and process of reintegration of the migrants into their local labour markets. The question of their positioning in the labour market and its determinants is analyzed in the fourth section. The last section provides the concluding remarks.

2 Estimates show that growth in South Asia decelerated in 2008, falling from 8.6 per cent in 2007 to below 7 per cent. It is projected to decline further to around 6 per cent or below in 2009, before recovering to around 7 per cent in 2010 (Asian Economic Monitor, Asian Development Bank, July 2008; December 2008; July 2009).

Global Financial Crisis and Return of South Asian Gulf Migrants Y 199

Profile of the Return Migrants and their Household The Return Migrants Before we look at the patterns in employment of the REMs, we provide a profile of them. As shown in Figure 11.1, the median age of REMs ranged between 29 for REMs in Bangladesh to 39 for those in Sri Lanka. As can be seen from the box plots, the two whiskers of most countries seem to be evenly distributed, except in the case of India wherein there is the presence of a substantial share of REMs above the age of 40. For males in general, the age range 30–39 is at the centre of their working age range. It is at this peak working age that most REMs had to return back to their home countries. During periods of normalcy, the average age of REMs would be in the higher range of 40–45. However, this lower range of age of REMs after the crisis shows that while the more experienced workers continued to remain in the host country, it was the younger workers with lesser experience that had to leave. Figure 11.1:

Age distribution of REMs

Source: CDS survey on international migration, 2009.

A comparison with the EMIs in 2008,3 the year that immediately preceded the crisis period shows that in India, the REMs in 2008 — a normal year — followed a regular expectation curve (see Figure 11.2). 3 The source for this comparison is the Migration Monitoring survey conducted by the Migration Research Unit of the CDS with financial assistance from the Department of Non-Resident Keralite Affairs, Government of Kerala.

200 Y Vinoj Abraham and S. Irudaya Rajan

The REMs among the younger age groups were very less compared to older age groups. The peak age group of REMs was 35–54 years of age. But in the crisis year, the peak age group of REMs was 25–29, followed by 30–34 years of age. Moreover, the pattern shows that the REM pattern in 2009 followed exactly the same trajectory as that of the EMIs in 2008. This suggests that REMs during the crisis year were the ones who had migrated in the immediate preceding years. In other words, the crisis had affected the recent young migrants the worst as compared to older and more experienced migrants in India. This pattern may be more or less the same in other South Asian countries as well. Figure 11.2: Comparison of the age distribution between REMs and EMIs from India

Source: CDS survey on international migration, 2009; Migration Monitoring survey of CDS, 2008.

In general it can be seen that the level of education among the REMs was low. More than 60 per cent of the REMs who came back due to the crisis had only completed their primary-level schooling or less, while another 34 per cent had only completed their secondaryor higher secondary-level of education (see Figure 11.3). However, this is not uniform across all countries. For India and Bangladesh, more than 65 per cent of the REMs had less than upper primary education, while for Pakistan and Nepal it was close to 40 per cent only. Pakistan and Nepal had the largest number of REMs who were educated more than the upper primary level.

Global Financial Crisis and Return of South Asian Gulf Migrants Y 201 Figure 11.3:

Level of education of REMs

Source: CDS survey on international migration, 2009.

A comparison between the patterns in the education level of REMs in 2008 and that of REMs in 2009 shows a kind of selfselection in crisis-related migration. Firstly, the self-selection process can be seen in the case of normal year REMs as well (Table 11.1). For instance, when the EMIs with a below primary level of education in 2008 were 14.7 per cent, in the same year the REMs with the same level of education were 23.5 per cent. Thus, a larger share of less-educated workers seemed to be returning while the proportion of those going to the Gulf was lesser. But as at higher levels of skill the share of EMIs increased much beyond that of REMs. The widest gap between EMIs and REMs was at graduation level or above — 27 per cent EMIs as compared to 10 per cent REMs. However, in the year of the crisis this self selection seems to have accentuated, Table 11.1:

Level of education of REMs in India, 2008 and 2009 EMIs/REMs in 2008 REM 2009

Below Primary Upper Primary Completed Secondary Completed Graduation Completed

34.1 32.5 27.0 6.4

EMI 2008

REM 2008

14.7 38.6 26.7 20.0

23.5 44.1 22.0 10.3

Source: CDS survey on international migration, 2009; Zachariah and Rajan (2010).

202 Y Vinoj Abraham and S. Irudaya Rajan

wherein only 6.4 per cent of REMs were graduates while 66.6 per cent of REMs had merely completed upper primary or lower. Thus, the crisis seems to have forced more of the less educated to return to their home country rather than stay on in their host country, while the share of those educated beyond school levels among REMs was relatively low. If schooling can be taken as a proxy for skills then it can be stated that the potential for resilience and stability increased with education in times of crisis.

The Features of the Return Migrant Households We shall now examine the characteristic features of the typical migrant household (Table 11.2). This is to highlight the conditions of life, as well as the pressure that is exerted on the migrant worker to meet the demands of his household. Typical migrant households are fairly large with nearly five members each. But there exist substantial inter-country variations. While the household size in Pakistan was more than six members, in India it was less than five members, for Sri Lanka and Nepal it was approximately four members, and for Bangladesh less than four members. Now, if we look at the average dependency ratio, Pakistan — which has the highest household size — also has the highest dependency ratio4 at 3.2. This means that approximately 3.2 non-workers depended on every worker in the sample household. For Sri Lanka the corresponding ratio was 2, while for India and Bangladesh it was approximately 1 and less than 1 for Nepal. It is also worth noting that on an average about 80 per cent of the total income of a migrant household in 2008 was derived from migrant remittances. Across countries it ranged between 60 per cent and 80 per cent. This shows the importance of remittances in the income of these households. Given this very high level of dependency on remittances, the compulsions on the migrant workers to continue working even during a crisis situation need to be recognized. Even when they return to their home countries, the stress on these workers to find alternative sources of income would be substantially high.

Dependency ratio is defined as the ratio of non-workers to workers in the household. 4

34.00 49.00

60.00 70.00

49.00

80.00

3.70 0.90 0.60 8.80 0.37 61.80

Bangladesh

62.00

58.00

6.10 3.20 0.70 12.40 0.52 332.90

Pakistan

70.00

56.00

4.70 1.00 0.80 12.80 0.53 246.90

India

65.00

54.00

4.70 1.20 0.80 12.20 0.51 257.90

Weighted average

household.

Note: a Household’s income dependence on remittances is defined as the share of migrant remittances in the total income of the

Source: CDS survey of international migration, 2009.

4.40 2.00 0.70 13.00 0.54 364.00

Sri Lanka

4.10 0.60 0.80 12.00 0.50 327.70

Nepal

Conditions of the REM household, 2009

Household Size Dependency Ratio Household’s income dependence on remittancesa Asset Score Asset Index Migrant’s average monthly income in 2008 (US$) Migrant’s average monthly income in 2009 as proportion of the same in 2008 (%) Per capita monthly average income in 2009 as proportion of the same in 2008 (%)

Table 11.2:

204 Y Vinoj Abraham and S. Irudaya Rajan

The asset levels of the migrant households show that they belong to the median levels of the asset scores with their household asset score being around 12 on a maximum achievable scale of 24.5 The poorest asset-level position belonged to Bangladesh, while the best was Sri Lanka’s at 13. From the point of view of crisis resilience, the ownership of these assets plays a very important role in determining the ability to diversify into other areas of employment. For a glimpse at the effect of the crisis on the household, we can look at per capita income levels of the migrant and their households in these countries. The decline in the earnings of the migrants is visible in their households as well — the per capita income declined such that in 2009 it was only 65 per cent of the corresponding figure for 2008. This is visible across all countries. However, it may be noted that while the share of per capita income declined along with that of migrant average income, the decline in per capita income across all the countries was lower than that of the migrant’s income. This implies that with the crisis the households also diversified into income sources other than remittances which helped them tide over the situation partially.

Trends and Patterns in Employment of Return Migrants To understand the change in the employment status of the migrant workers after their return due to the crisis, we compare the employment status in the sample between the years 2008 and 2009 — the first year is a normal year and the second a crisis one. As can be seen in Table 11.3, 51.3 per cent of the regular workers in 2008 had shifted their status to casual workers in 2009, while another 17 per cent of the unemployed in 2008 also became casual workers. Only 29 per cent of the casual workers in 2009 belonged to the category of casual workers in 2008. Among the regular workers in 2009, 94 per cent had been regular workers in 2008 as well. Among the selfemployed in 2009, almost 50 per cent had had regular employment in 2008. Among the unemployed in 2009, 73 per cent had been regular workers in 2008. Among those who did not report their employment status in 2009, 56 per cent had been regular workers in 2008. For the assets considered and construction of the asset index, see Appendix Table 11C. 5

51.3 93.9 48.7 73.1 55.6 69.3

Regular workers 2.6 1.4 40.8 3.9 2.8 8.9

Self-employed 17.1 1.4 1.3 18.0 9.7 8.2

Unemployed

0.0 2.7 4.0 2.6 29.2 6.7

Not reported

Total 100.0 100.0 100.0 100.0 100.0 100.0

Source: CDS survey of international migration, 2009. Note: Each row represents the distribution of workers in 2009 as per their activity in 2008. For instance, the first row shows that among all the casual workers in 2009, only 29 per cent belonged to the category of casual workers in 2008.

29.0 0.7 5.3 2.6 2.8 6.9

Casual workers

Employment status in 2008

Matrix of distribution of various types of workers in 2009 by their employment status in 2008

Casual Workers Regular Workers Self-Employed Unemployed Not Reported Total

Table 11.3:

Employment status in 2009

206 Y Vinoj Abraham and S. Irudaya Rajan

The fact that they had reported their employment status in 2008 and did not in 2009 also points to possibility of social stigma attached to reporting a lower employment status in 2009 as compared to 2008. One important conclusion that can be drawn from this is that the employment status of the REMs was in general worse off than it had been in the host country. There was a clear trend towards casualization, self-employment and unemployment in the crisis year, while formal regular employment share declined drastically during the crisis. Along with the change in status, the workers also experienced a substantial dip in earnings after returning to their home countries. The REMs in the sample saw a 46 per cent decline in their average monthly income in 2009 as compared to that in 2008 (see Table 11.4). This decline was felt among workers of all employment statuses. The largest reduction was for self-employed and casual workers who became unemployed following the crisis. Casual workers who became regular workers on return also experienced a near 100 per cent drop in their earnings. This is true across countries of origin as well. The average monthly income of the migrants had fallen in all the countries in 2009. For the Sri Lankan migrants it declined so drastically that the income in 2009 was only 34 per cent of their income in 2008. It may be noted that the Sri Lankan migrants were on an average earning the most among all the countries, and it was they who experienced most severe cut in their incomes. Bangladeshis, who earned the least among these countries, still earned 80 per cent of their income in 2008. The survey had covered the immediate employment history of the REM during the crisis as a cross-sectional picture of five time frames. The employment status was enquired for the periods; (a) one year to six months prior to return, (b) six months prior to return, (c) two weeks before return, (d) one month after return and (e) the present status.6 It can be seen that while regular employment was enjoyed by nearly 90 per cent of the workers during the first and second periods — i.e., one year preceding the return and six months preceding the return — two weeks before the return, the share of regular employment declined to 55 per cent, and the share of unemployed increased from almost zero in the previous periods to about The average time for REM workers between their return to the home country and the survey was approximately 10 months. 6

–64.4 –32.0 –47.7 –61.0 –97.3 –50.8

Regular workers –95.1 26.9 3.4 –100.0 –66.7 –14.5

Self-employed 638.0 0.0 0.0 0.0 0.0 160.9

Unemployed

0.0 194.3 –72.7 0.0 –100.0 84.7

Not reported

Total –45.7 –29.1 –38.2 –61.9 –96.7 –46.1

Source: CDS survey of international migration, 2009. Note: Each row represents the change in earnings of workers of each category in 2009 as compared to 2008. For instance, the first cell shows that casual workers in 2008 who remained the same in 2009 also experienced a decline in earnings of 15 per cent in 2009 as compared to 2008.

–15.0 –100.0 –59.0 –100.0 –68.0 –27.1

Casual workers

Employment status in 2008

Matrix of change in earnings of REMs between 2008 and 2009 (in per cent)

Casual Workers Regular Workers Self-Employed Unemployed Not Reported Total

Table 11.4:

Employment status in 2009

208 Y Vinoj Abraham and S. Irudaya Rajan

33 per cent (Figure 11.4). One month after their return, however, unemployment among the group increased to more than 75 per cent, while there was some increase in self-employment to 10 per cent. In the host country, the present status of these migrants was such that about 42 per cent of them were still unemployed, while 20 per cent of them found self-employment, and 18 per cent of the workers took up casual employment. Regular employment had the smallest share at 15 per cent. This is a clear account of how the crisis led to loss of jobs in the Gulf, and their integration into the labour market at lower employment status which included lower wages on their return. Figure 11.4: Changes in employment status and reintegration into the local labour market

Source: CDS survey of international migration, 2009. Note: For the detailed table, see Appendix Table 11A.

Across the five countries there are some differences in the patterns of re-entry to the labour market. More than 75 per cent of the workers remained unemployed immediately after returning across all countries. Later, the share of unemployed among REMs declined in general in all the countries. But the share of unemployed in the present status seems to be high for Nepal and India where the share was 58 per cent and 45 per cent respectively, while for the rest of the countries it reduced to 35 per cent or less. Thus, there seem to be

Global Financial Crisis and Return of South Asian Gulf Migrants Y 209

country-specific variations in the way the workers have entered the home labour market and found jobs. Figure 11.5: Share of unemployed in sample: Inter-temporal and Inter-country differences

Source: CDS survey of international migration, 2009.

There were also considerable inter-country variations following the crisis in terms of the employment pattern (Table 11.5). Though there was a general increase in casual employment, it was the highest in Bangladesh at 27 per cent. Here, as we had seen earlier, the unemployment rate was the lowest. When we consider both these trends together, it seems to suggest that Bangladeshi migrants could not afford to wait unemployed for opportunities on returning to their home country, probably due to their household conditions. Moreover, these migrants who had the lowest level of education could not expect to get highly-skilled regular employment. While in Nepal and Sri Lanka the largest share of workers joined as regular salaried workers in their countries, in India and Pakistan the most attractive propositions were for self-employees, accounting for a share of 25 and 31 per cent respectively. Looking at industrial distribution of workers, during their tenure in the Gulf the largest proportion of workers was engaged in

210 Y Vinoj Abraham and S. Irudaya Rajan

Present status

Table 11.5:

Nepal Sri Lanka Bangladesh Pakistan India Total

Inter-country variations in employment status Selfemployed

Regular salary

Casual wage labour

Unemployed

Others

Total

6.00 8.00 19.23 31.37 25.30 21.36

26.00 36.00 3.85 21.57 9.64 15.73

10.00 20.00 26.92 11.76 19.28 17.84

58.00 36.00 30.77 35.29 44.98 43.43

0.00 0.00 19.23 0.00 0.80 1.64

100.00 100.00 100.00 100.00 100.00 100.00

Source: CDS survey of international migration, 2009.

the secondary sector, accounting for 51 per cent of the workers (Table 11.6). Within the secondary sector itself, construction accounted for 41 per cent and another 36 per cent was engaged in the tertiary sector, of which hotel, transport, storage, and communication industries accounted for the largest share at 10.8 per cent. The primary sector accounted for only 2 per cent and 11 per cent did not report their industry of employment. This broad pattern was true for migrants from both India and Pakistan while for Nepal and Sri Lanka the patterns showed some difference. Migrants from both these countries had a higher percentage working in the tertiary sector than in the secondary sector — 44 and 54 per cent respectively. Also, their presence in the construction industry was much lower compared to the Indians or Pakistanis — only 24 and 20 per cent share respectively. We had earlier seen that Sri Lankan and Nepalese migrants had a larger proportion of regular workers and their average monthly earnings were above the mean. These aspects indicate that their conditions of work, industry, and earnings were better than workers from India, Pakistan and Bangladesh before the crisis. After the crisis and their return there have been substantial differences in their choice of industries. Overall, the share of the secondary sector declined from 51 to 26 per cent, while that of the tertiary sector increased from 36 to 44 per cent. There was a significant rise from a mere 2 per cent to 16 per cent for the primary sector. This shift in the industrial structure of these workers is in consonance with the industrial structure of the host region, wherein most South Asian countries have a dominant tertiary sector while the secondary sector is under-developed. Moreover, this shift, especially to the

10.0 0.0 59.6 4.1 5.4 11.0

Nepal Sri Lanka Bangladesh Pakistan India Total

4.0 0.0 0.0 6.1 1.7 2.1

0.0 0.0 26.2 12.1 20.4 16.2

Primary

24.0 20.0 14.9 46.9 55.2 42.3

0.0 12.5 0.0 27.3 26.3 18.5

Construction

42.0 46.0 29.8 51.0 58.5 51.3

4.8 34.4 4.8 36.4 31.4 26.0

Secondary

14.0 20.0 6.4 12.2 8.7 10.8

9.5 31.3 0.0 18.2 10.2 12.1

H & R; TS & C

Source: CDS survey of international migration, 2009. Note: H & R = Hotels and Restaurants. TS & C = Transport, Storage and Communication.

23.8 0.0 61.9 3.0 3.7 14.0

Nepal Sri Lanka Bangladesh Pakistan India Total

Unreported

Industrial distribution of migrant workers: Before and after return

Country

Table 11.6:

Present

Six Months before Return

8.0 16.0 2.1 2.0 7.1 7.1

23.8 21.9 0.0 3.0 7.3 8.7

Others

44.0 54.0 10.6 38.8 34.4 35.7

71.4 65.6 7.1 48.5 44.5 43.8

Tertiary

100.0 100.0 100.0 100.0 100.0 100.0

100.0 100.0 100.0 100.0 100.0 100.0

Total

212 Y Vinoj Abraham and S. Irudaya Rajan

primary and tertiary sectors, implies a lack of employment opportunities in the host countries and hence getting accommodated in labour surplus sectors such as agriculture and low-end services mainly through self-employment and casual work. However, it may be noted that in case of Nepal and Sri Lanka there were no entrants to the primary sector. In case of Nepal the REMs got maximally absorbed in the Tertiary sector (71.4 per cent) while in Sri Lanka they got absorbed both in the services (65.5 per cent) and secondary sector (34.4 per cent). As we had seen earlier, the share of REMs in Nepal and Sri Lanka had who had regular work was higher than other countries. Also, they had higher rates of unemployment than other countries. Viewing from their previous conditions of work it can be argued that the REMs in Nepal and Sri Lanka had a greater buffer than their counterparts in India and Pakistan and they chose to remain unemployed rather force themselves into lower quality of employment. While for Pakistan and India the REM workers got involved more in tertiary sector and primary sector as self employed and casual workers.

Determinants of Finding Employment in the Home Country We now turn to examining what determines the status of REMs being employed or unemployed in their home countries on return from their host countries. For this purpose we estimate a logit model with the binary choices of being employed or unemployed.

Analytical Framework Following the crisis the process of integration of the REMs needs to be viewed from the context of developing economy labour markets. The typical labour markets in developing economies are characterized by surplus labour supply conditions, weak labour demand conditions, and hence, low wage rates. These economies, owing to their surplus labour conditions, absorb a large share of their workers at low value to industries such as traditional agriculture or lowend personal services, and in poor working conditions such as being casual labour or being self-employed. In other words, the typical labour markets in developing economies are characterized by the presence of a large informal sector. Moreover, these labour markets work under conditions of institutional rigidities such as caste,

Global Financial Crisis and Return of South Asian Gulf Migrants Y 213

gender and other labour market institutions, information asymmetries and segmentations. It is into these labour markets that the migrants have to enter on return from the Gulf. The possibility of finding employment of their choice may be limited. This may be so because of many reasons. Firstly, the industries that they have worked under in their host country may not exist or may be under-developed in the home country. Secondly, these workers may be over-skilled for industries in the developing economies, and hence may not find the desired job. Thirdly, though these workers may have worked in relatively poor work conditions in their host countries, they may not be ready to take up employment under similar conditions in the home country due to the social stigma attached to such kind of employment in the home country. Due to these reasons, the REM may not find an employment of his/her choice immediately on arrival. Because of the weaknesses in the market, the time taken to look for a job and the cost involved may be high. It then follows that REMs who have the ability to undertake this job search, both in terms of cost and time, would remain unemployed rather than take up an immediately available employment. However, if the opportunity cost of searching for employment is very high then the worker may not wait for an employment of his choice, but take up any that is available. Now, positions that are available with ease in developing economies would be low-end work in all sectors and any type of work in low-wage paying industries — that demand low levels of skill and experience. Based on this analytical context it can be expected that REMs who found work on return were the ones who could not afford the cost of searching for a job as well as those who expected that even if they were to suffer this cost, the probability of finding employment of choice would be very less due to low levels of education or rigidities caused by institutional barriers such as caste. The cost of job search would be highest for the ones who do not have any past savings to depend on, or have poor asset positions within the household, or if the household has very high dependence on the migrant as the prime income earner of the household. The ones that remained unemployed are the ones that could afford the cost of job search, owing to their past earnings or the relatively better asset position and lower household dependency on remittances.

214 Y Vinoj Abraham and S. Irudaya Rajan

Using this analytical framework we fit a logit model to analyze the determinants of the probability of being employed for the REM. The following model is set for analysis: Empi = a + bXi + ui Wherein the dependent variable Emp = 1 if the current status of the ith REM is employed, and Emp = 0 if the current status of the REM. The independent variables x are defined in the next section.

Empirical Analysis and Results From the analytical framework it can be seen that the probability of the REM being employed depends on two set of factors. One set of factors are positive factors that affect the ‘employability’ of an REM and hence his/her ability to find suitable employment. The other set of factors are negative factors that force the REM to take up employment due to unfavourable economic circumstances. For the variable definitions, see Appendix Table 11C. We use country dummies to control for country-based heterogeneity. Robust standard errors were estimated to correct for heteroscedasticity. We report both the logit coefficients and the odds ratios for ease of interpretation. The model has overall high significance with the Wald chi2 at 34.5. It may be noted that in the analysis the observations from Bangladesh were dropped because sufficient information on the variables was not available. As expected the variable ‘Age’ and ‘Age squared’ have yielded positive and negative signs respectively and both the coefficients are significant. This implies that the REMs of the older age group had a greater probability of finding employment on return than younger ones. Yet, beyond a threshold age, the older REMs may not be able to find employment compared to REMs younger than them. However, the odds ratios show that while older REMs had odds of 1.19 of being employed as compared to younger REMs, beyond the threshold age the odds decline to just one. Therefore, the odds are not different for the older and younger REMs beyond a threshold age. Contrary to expectations the coefficient for the variable ‘Education’ had negative coefficients and was significant at least at 10 per cent level. We had hypothesized that workers with higher education would have more demand in a labour market that typically has a dearth of skill supply. However, it can be observed that REMs

Global Financial Crisis and Return of South Asian Gulf Migrants Y 215 Table 11.7:

Logit model: Determinants of employment of return migrants (present status being employed = 1; present status being unemployed = 0)

Variables Age Age2 Education Secondary Education above Secondary Asset Index Industry (Services) Household Income dependency Dependency Rate Backward Caste General Caste Country (Sri Lanka) Country (Pakistan) Country (India) Constant Number of Observations Wald chi2(13) Prob>chi2 Pseudo R2

Logit coefficients (z values)

Odds ratio

0.1714* (1.65) –0.0022* (–1.63) –0.5650* (–1.67) –1.1505** (–2.06) –0.0602* (–1.66) –0.6657*** (–2.6) 1.2618*** (3.04) –0.2205 (–1.1) 0.9635** (2.34) 0.8668** (2.2) 1.2714** (2.29) 0.7717 (1.42) 0.3446 (0.79) –3.3604 (–1.63) 330 34.55 0.0010 0.1001

1.19 1.00 0.57 0.32 0.94 0.51 3.53 0.80 2.62 2.38 3.57 2.16 1.41

Source: Based on estimation by the authors. Note: Z values in parentheses; Z values estimated on the basis of robust standard errors. ∗, ∗∗,∗∗∗ are significant at 10, 5 and 1 per cent level respectively.

with secondary level of education or education beyond school had a lesser probability of being employed than not. Moreover, as the level of education increased the probability of finding employment declined. An REM educated up to secondary level had only half the

216 Y Vinoj Abraham and S. Irudaya Rajan

chance of that of an REM who was educated just up to primary level. On the other hand, the REM with an education level that is higher than schooling had the odds of only one-third that of a primary educated person of getting a job. Education, it seems, acts as a barrier to finding employment in the context of developing economy labour markets. It may also be argued that REMs with a higher level of education are probably ready to wait longer to find skill-matched employment rather than take up skill-mismatched jobs due to social stigma and higher savings from their previous employment. The ‘asset index’ too has a negative coefficient, as expected. A higher level of asset index lowers the probability of the REM getting employed. This is based on the argument that households of REMs with a greater asset index have greater well-being in general and probably have greater past savings that would allow the REM to remain unemployed and get involved in job search rather than be forced to take up employment that is not of his/her choice. When the REM’s household asset position is low the worker does not have the luxury of job search, but takes up the first such available opportunity. The coefficient is significant at a level of 10 per cent. The REM’s previous ‘industry’ in the host country is an important determinant of his employment probability in the home country. The industry dummy coefficient is negative and significant at 1 per cent. This indicates that if the REM’s work experience had been in the services sector than the secondary sector then his/her probability of finding employment would be lesser as compared to workers with secondary sector experience. The service sector workers probably do not get such employment opportunities. However, since their returns were higher than the secondary sector workers in general, it can be assumed that they are also ready to incur greater costs on job search. ‘Household income dependency’ came out be positive and highly significant. The odds ratio shows that the probability of an REM with high household dependency on him/her finding employment is 3.5 times higher than that of an REM with low household dependency. It can be argued that when the household dependency on the REM is very high during his work in the host country, it is not easy to substantially reduce this dependency. Hence, the REM would be forced to accept employment of any type in any industry, which in

Global Financial Crisis and Return of South Asian Gulf Migrants Y 217

turn implies that his probably of being employed would be higher than one that has low household income dependency. But dependency rate was not significant. ‘Caste’, as expected, is highly significant and positive. It implies that REMs high up in the caste order had a greater chance of getting employment as opposed to REMs belonging to the low end of the caste order. Both backward and general castes had odds of 2.5 times that of the scheduled castes and tribes for finding a job on their return from the Gulf. It appears that traditional social segregations such as caste would continue to impose discriminatory entry barriers even if the migrant had returned from regions with nondiscriminatory labour practices. Among the ‘country dummies’, only the Sri Lanka dummy came out to be significant and positive, indicating that the probability of REMs in Sri Lanka finding employment was three times higher than that of REMs in Nepal, while in other countries it was significant.

Conclusion The global financial crisis had affected the stock and flow of international migration from the developing world to the developed world in myriad ways, of which one of the most conspicuous aspects was the loss of their source of livelihood and return of thousands of migrant workers to their home country. Often the rehabilitation and reintegration of these workers is marked by forced choices of employment. This study was an attempt to understand their process of reintegration in five South Asian countries. Analysis of the patterns shows that there was a process of selfselection of REMs based on age and education, wherein the younger and less educated were over-represented in the sample. The households of these REMs were large and depended very heavily on remittances as their main source of livelihood. There was a clear decline in the households’ per capita income following the crisis. On return, the employment statuses of REMs were in general worse-off than in their host country with high share of casualization, self-employment and unemployment in the crisis year, while the proportion of formal regular employment declined drastically; their average monthly earnings declined by an average of 46 per cent

218 Y Vinoj Abraham and S. Irudaya Rajan

and they got employed in industries with poorer employment conditions. REMs from Sri Lanka and Nepal, who had better employment conditions in their host economies, seemed to be getting integrated into their home economies either at better conditions of work and industries, or chose to remain unemployed. Indian and Pakistani REMs, who had worked at relatively inferior conditions of work in the Gulf, had to find employment relatively faster than REMs in other countries. They in turn got absorbed in industries which had inferior conditions of work. The analysis of the determinants of being employed suggests that those who found employment on return were in fact driven by economic compulsions to reduce their job search period and cost. Hence, we find that REMs with lower education, greater household dependency, and poorer assets were the ones who took up employment, while REMs with higher levels of education, larger assets in the household, lower level of household dependency and work experience in specific industries tended to remain unemployed. The direction and magnitude of the determinants of employment tend to suggest a process of forced or constrained choice of employment for REMs rather than getting involved in job search to maximize their potential gains. And the constraints mainly included their low level of skills, weakly diversified income sources for their households and a low level of household income. For the policy makers it is important to note that, in effect, it is the REMs with employment at the time of the survey that are worse-off than those unemployed. Therefore, it may be important to target remedial measures on improving the earnings capacity of the REMs along with a stronger social security net that would enhance their ability to withhold their labour when needed to.

Nepal Sri Lanka Bangladesh Pakistan India Total

Nepal Sri Lanka Bangladesh Pakistan India Total

One month after return

2.00 0.00 10.00 13.73 15.42 11.44

6.00 8.00 19.23 31.37 25.30 21.36

Self-employed

2.00 4.00 0.00 0.00 5.00 3.65

26.00 36.00 3.85 21.57 9.64 15.73

Regular salary/wage employed

0.00 2.00 10.00 0.00 8.75 5.84

10.00 20.00 26.92 11.76 19.28 17.84

Casual wage labour

94.00 94.00 75.00 86.27 70.00 78.10

58.00 36.00 30.77 35.29 44.98 43.43

Unemployed

Changes in employment status during the crisis: An inter-country comparison

Present status

Table 11A:

Appendix

100.00 100.00 100.00 100.00 100.00 100.00

100.00 100.00 100.00 100.00 100.00 100.00

Total

(Table 11a continued)

2.00 0.00 5.00 0.00 0.83 0.97

0.00 0.00 19.23 0.00 0.80 1.64

Others

Nepal Sri Lanka Bangladesh Pakistan India Total

Nepal Sri Lanka Bangladesh Pakistan India Total

Six months before return migration

One year to six months before return migration

0.00 6.00 0.00 9.80 1.64 2.88

0.00 6.00 0.00 7.84 2.07 2.91

0.00 2.00 0.00 6.00 2.89 2.66

Self-employed

93.48 90.00 80.00 60.78 94.26 88.70

93.62 90.00 87.50 60.78 93.36 88.62

68.00 36.00 28.57 24.00 67.77 56.66

Regular salary/wage employed

Source: CDS survey of international migration, 2009.

Nepal Sri Lanka Bangladesh Pakistan India Total

Two weeks before return migration

(Table 11a continued)

4.35 4.00 8.00 29.41 0.82 5.53

6.38 4.00 0.00 27.45 0.83 5.08

10.00 2.00 4.76 6.00 2.48 3.87

Casual wage labour

2.17 0.00 0.00 0.00 2.87 1.92

0.00 0.00 12.50 3.92 3.32 3.15

22.00 60.00 66.67 64.00 26.45 36.56

Unemployed

0.00 0.00 12.00 0.00 0.41 0.96

0.00 0.00 0.00 0.00 0.41 0.24

0.00 0.00 0.00 0.00 0.41 0.24

Others

100.00 100.00 100.00 100.00 100.00 100.00

100.00 100.00 100.00 100.00 100.00

100.00 100.00 100.00 100.00 100.00 100.00 100.00

Total

Nepal Sri Lanka Bangladesh Pakistan India Total

Nepal Sri Lanka Bangladesh Pakistan India Total

Nepal Sri Lanka Bangladesh Pakistan India Total

Present

One month after return

Two weeks before return

0.0 0.0 80.6 5.3 6.5 14.1

0.0 0.0 88.6 0.0 18.5 35.7

23.8 0.0 61.9 3.0 3.7 14.0

Unreported

2.6 5.0 0.0 10.5 1.1 2.0

0.0 0.0 11.4 0.0 25.9 19.4

0.0 0.0 26.2 12.1 20.4 16.2

Primary

30.8 10.0 5.6 63.2 52.4 41.8

0.0 0.0 0.0 57.1 23.5 17.8

0.0 12.5 0.0 27.3 26.3 18.5

Construction

53.9 45.0 13.9 63.2 55.7 50.2

33.3 0.0 0.0 57.1 29.6 22.5

4.8 34.4 4.8 36.4 31.4 26.0

Secondary

15.4 10.0 2.8 10.5 9.2 9.4

33.3 0.0 0.0 28.6 8.6 7.8

9.5 31.3 0.0 18.2 10.2 12.1

H & R; TS & Ca

Industrial distribution of migrant workers: Before and after return

Country

Table 11B:

7.7 20.0 2.8 0.0 8.7 8.0

0.0 100.0 0.0 14.3 4.9 6.2

23.8 21.9 0.0 3.0 7.3 8.7

Others

100.0 100.0 100.0 100.0 100.0 100.0

100.0 100.0 100.0 100.0 100.0 100.0

100.0 100.0 100.0 100.0 100.0 100.0

Total

(Table 11b continued)

43.6 50.0 5.6 21.1 36.8 33.8

66.7 100.0 0.0 42.9 25.9 22.5

71.4 65.6 7.1 48.5 44.5 43.8

Tertiary

Nepal Sri Lanka Bangladesh Pakistan India Total

One year to six months before return

12.2 0.0 52.0 0.0 3.3 9.1

10.0 0.0 59.6 4.1 5.4 11.0

Unreported

4.1 0.0 8.0 5.9 2.1 3.2

4.0 0.0 0.0 6.1 1.7 2.1

Primary

24.5 20.0 12.0 47.1 55.0 41.9

24.0 20.0 14.9 46.9 55.2 42.3

Construction

42.9 46.0 30.0 51.0 58.7 51.4

42.0 46.0 29.8 51.0 58.5 51.3

Secondary

14.3 20.0 6.0 11.8 8.7 10.6

14.0 20.0 6.4 12.2 8.7 10.8

H & R; TS & Ca

Source: CDS survey of international migration, 2009. a H & R; TS & C = Hotels and Restaurants; Transport, Storage and Communication. Note:

Nepal Sri Lanka Bangladesh Pakistan India Total

Six months before return

Country

(Table 11b continued)

8.2 16.0 2.0 2.0 7.9 7.5

8.0 16.0 2.1 2.0 7.1 7.1

Others

40.8 54.0 10.0 43.1 36.0 36.4

44.0 54.0 10.6 38.8 34.4 35.7

Tertiary

100.0 100.0 100.0 100.0 100.0 100.0

100.0 100.0 100.0 100.0 100.0 100.0

Total

Variable definition

Age of the worker

Square of the age of the worker

Level of education: up to upper primary is base category; education up to higher secondary level = 1, education beyond schooling = 2

Broad Industrial Category: Secondary is base Category, Services = 1

The following assets were considered: Land, House/Building, Motor Car, Bike/Scooter for own use, Taxi/Truck/Lorry/Auto Rickshaw, Refrigerator, Television, VCR/DVD/MP3 player/Music systems, Land phone, Personal Computer/Laptop, Cable connection, Mobile phone. Ownership of Land and House was given a weightage of 5, Motor car and Taxi/Truck/Lorry/Auto Rickshaw were given 2.5 and all the rest were given 1. In total, the asset score could vary between 0 and 24. Further, the asset index was calculated as each household’s asset score divided by 24.

Share of the REMs earnings prior to return in the total household’s income

Ratio of household members that are either in the age group of less than 15 or above 65 or number of unemployed in the household to employed

Caste the household belongs to: Scheduled Caste/Tribes is base category; Backward Castes = 1, General castes = 2

Countries

Variable name

Age

Age Squared

Education Dummy

Industry Dummy

Asset Index

Household Income Dependence

Dependency Rate

Caste Dummy

Country Dummy

Source: Developed by the authors.

Variable definition, construction and hypothesis

Table 11C:

224 Y Vinoj Abraham and S. Irudaya Rajan

References International Monetary Fund (IMF). 2009. World Economic Outlook: Crisis and Recovery. Washington, DC: International Monetary Fund. Zachariah, K. C. and S. Irudaya Rajan. 2010. ‘Impact of the Global Recession on Migration and Remittances in Kerala: New Evidences from the Return Migration Survey (RMS) 2009’. Working paper no. 432, Centre for Development Studies, Thiruvananthapuram. http://cds.edu/download_files/ wp432.pdf (accessed 7 November 2011). Rajan, S. Irudaya and D. Narayana. 2010. ‘The Financial Crisis in the Gulf and its Impact on South Asian Migrant Workers’. Working paper no. 436, Centre for Development Studies, Thiruvananthapuram. http://cds.edu/download_ files/wp436.pdf (accessed 7 November 2011). Parameswaran, M. 2009. ‘Financial Crisis and the Kerala Economy’. Working paper no. 441, Centre for Development Studies, Thiruvananthapuram. http:// cds.edu/download_files/wp441.pdf (accessed 7 November 2011).

12 Inclusive Growth and Economic Crises Labour Migration and Poverty in India Arjan de Haan Introduction: Migration as Safety Valve Economic crises can have enormous impact on migration, both

short-term and long-term, and direct and indirect. The economic crisis of 1929 led to a decrease in international migration, and the return of large numbers of internal migrants as employment declined. For example, in February 1931, almost 60,000 jute mill workers — about one-fifth of the total — were dismissed as a result of the trade depression, and according to the labour commissioner the employees ‘melted like snow in summer’ (quoted in de Haan 1994: 27), returning to their villages of origin.1 The 1973 oil crisis also led to a fall in labour migration, which in Europe was succeeded by a period of migration for family reunification, and in Asia by booming migration to the Gulf (Rajan and Kumar 2010). To some extent the 1997–1998 crisis in Asia had a similar impact, but this was mostly short-term. In most cases, declines in job opportunities during crises were accompanied by protectionist measures; however, after 1997 employers in Asia soon asked for ensuring sufficient labour supply (Castles 2009; Migration DRC 2009). The effect of the 2008 crisis is still unfolding, with continued financial instability, budget cuts in Organization of Economic Cooperation and Development (OECD) countries, high unemployment, and a distinct political shift and resurging anti-immigrant sentiments.

1 R. N. Gilchrist, Labour Commissioner for the Government of Bengal, 1932 (de Haan 1994: 27).

226 Y Arjan de Haan

Since 2008, international migration has been affected with new migration declining. However there appears to have been little return migration, partly because migrants may not be going home if they fear they may not be able to return, and because conditions at home are equally bad (Castles and Miller 2010). International remittance declined by 6 per cent in 2009 as compared to the year before, but was expected to be picking up again in later years (Ratha et al. 2010). Internal migration too was affected, particularly in the areas and industries that were most closely linked to the global financial and economic crisis, such as the diamond industry in Surat, which is discussed later in the chapter. In coastal China an estimated 20 million jobs were lost, and tens of thousands of factories closed down throughout 2008. While many workers returned to their areas of origin during the Spring Festival of 2009, closely monitored by the government, many also stayed back, or in fact returned early to make sure that they were able to get a job when the economy stabilized. Labour migration has often been considered a safety valve: for workers to diversify livelihood sources in case of needs, and for economies to adjust labour supply in up- and down-turns. This chapter argues that this view is too limited and that migration is an integral part of economic development and processes of production and accompanying insecurity, through patterns of mobility that are structured historically and socially. In the case of India, migration patterns have not changed much as a result of the crisis, which is partly a result of the fact that the impact of the global crisis was cushioned in India, but also because patterns of migration are closely inter-twined with both organization of production and social structures, which are changing but only gradually. This chapter therefore takes a longer-term view, and enquires whether it is likely that patterns of migration and their links with poverty have changed since the early 1990s, and whether current social policies under the Congress-led government have become inclusive of poor labour migrants.2 Within this, it focuses on the migration patterns of deprived social groups, to analyze whether migration does form a route out of poverty, and whether specific policies for these groups exist or should be recommended. This discussion is placed against the on-going debates on changes in patterns

2

This builds on de Haan (2010).

Inclusive Growth and Economic Crises Y 227

of employment and job creation in India during the period of economic liberalization (Chandrasekhar 2010; Ghosh 2010), under the ‘inclusive growth’ policies since 2004, in the light of around 93 per cent of India’s workforce being under ‘informal employment’ (Sengupta 2009), and growing inequalities (Sarkar and Mehta 2010). The rest of this chapter is structured as follows. The next section briefly discusses general and global findings on the links between poverty and labour migration, focusing on population movements that remain within national borders, which is most relevant from a poverty perspective. This is used in the sections after that to describe changes in migration patterns in India. The concluding section discusses the implications of these insights for the notion of inclusive growth, and reflects on the analysis of migration and possible policies to enhance migrants’ well-being and ability to participate in India’s ‘disequalising growth’.

Facts and Questions about Migration and Poverty The subject of migration remains contested. While there is generally a preference for, or at least acceptance of, people moving when there is a demand for labour, there are equally strong voices arguing for reducing the number of immigrants, as currently in Europe (Harris 2010) — voices that become stronger at times of crises. Academically, disciplinary differences regarding approaches to understanding migration continue to exist. As described in more detail in ‘Inclusive Growth? Labour Migration and Poverty in India in the 21st Century’ (de Haan 2010), the literature on international migration can be summed up as follows, highlighting that while there are general lessons regarding migration, links between migration and poverty are deeply context-specific. Movement of people is much more common than is usually assumed, and has existed for much longer than often acknowledged. Moreover, recent international migration literature tends to neglect the fact that most migration remains within the Global South (UNDP 2009), and national borders — much of that therefore remains unrecorded and barely discussed after the 2008 crisis, for example. Labour migration also takes many forms, of which the classic rural–urban transition is only one: movement of entire households is also merely one of them, and a more common form is the movement of one member of the household who retains links with his/her origins, migrating for varying periods of time.

228 Y Arjan de Haan

The question of who migrates does not have a simple answer. In different contexts, socio-economic groups migrate, prompted by various capabilities, opportunities, and differentiated access. Most regions and countries have varying propensities for migration to specific kinds of destinations and opportunities: simple push–pull models tend to be inadequate to explain the complexity of segmented migration streams. The poorest people may not be able to migrate, and the most poverty-stricken areas do not necessarily have the largest numbers of migrants. The phenomenon of chain-migration leads to a great deal of path dependency in terms of migration patterns.3 With technological change, migration has tended to become more selective, with fewer opportunities for unskilled workers. Gender and age are crucial to understanding migration processes. There are no generalities about whether men or women migrate, despite a great deal of path dependency and gender stereotyping. In most places, young men might be over-represented amongst migrants, thus constituting particular gendered and household impacts, but women have always been mobile as well, moving for ‘traditional’ female occupations and newer ones — including unskilled and semiskilled manufacturing and service sector jobs (Gaetano and Yeoh 2010). Female migrants tend to be particularly vulnerable and suffer from labour market discrimination and violence. Reasons for migrating are diverse. Statistical data tend to record only one reason for migration, thus underestimating the complexity of migration. Migration often arises from desperation, lack of work, or indebtedness. For many households, seasonal migration is part of regular household strategies. But much migration is also driven by the hope or idea of better opportunities, and broadening horizons, or is part of the rites of passage for young men and women. Consequences of migration are equally diverse. At the household level, migration usually improves income and well-being, but often tends to maintain levels of living. Areas of origin usually show clear signs of migrant remittances, but there appear to be relatively few cases of transformation of areas, thus also suggesting that migration Hein de Haas (2008) reviews theories of chain migration and Jyoti Parimal Sarkar (2010) describes the process with respect to Bangladeshi migrants in India. See Iversen et al. (2011) for a theory of recruitment using networks and incentive mechanisms, with empirical testing using 1999–2000 National Sample Survey (NSS) data. 3

Inclusive Growth and Economic Crises Y 229

tends to be part of regions’ economies rather than initiating economic development, or causing severe brain drain (de Haas 2007; Papademetriou and Martin 1991). The question of impact on areas of destination has received relatively much attention, but assessments tend to vary — from migrants enhancing growth and well-being, to being a drain on local economies and upsetting local cultures.

Indian Migration Trends: Are they Changing? Movement of workers across the Indian subcontinent is an ageold phenomenon. This depends on changing patterns of economic development, and is partly related to levels of poverty, but with little evidence that migration contributes to reducing regional disparities. The poorest states like Bihar and Uttar Pradesh provide the largest numbers of migrants, who tend to move to richer areas like Delhi, Maharashtra and Gujarat. The patterns of movement are also prompted through chain-migration — with Kerala and Punjab being prime examples with respect to international migration — but similar segmentation of migration streams exists within India. The speed of urbanization in Asia — including India — has not been as fast as often assumed. The rural–urban transition is happening at moderate speed, and evidence suggests that this is associated with increased unaffordability of India’s cities in terms of land and basic services, and slum clearances, with observers fearing that this ‘exclusionary urban growth’ is likely to intensify in the future (Kumar 2010; Kundu 2009). According to the NSS data, the percentage of urban households that had moved during the previous year went slightly up — from 2.2 to 3.3 per cent — between 1993 and 2007–2008 (NSSO 2010: 16–17). Definitions of migrants and sources of data matter a great deal. According to the Census of India 2001, 309 million people were migrants based on the place of last residence — with more than twothirds being female — and there was an increase of 27 per cent since 1991 (see Bhagat 2011 for an extensive discussion on the Census of India 2001 data). Using NSS data — excluding marriage migration to get a better perspective on labour migration — about 10 per cent of the total population was classified as migrants, and this percentage has hardly changed (see de Haan and Dubey 2006 for a comparison between NSS data for 1987–1988 and 1999–2000). According to the data, the out-migration rate for males from rural areas was 9 per cent,

230 Y Arjan de Haan

and for females 17 per cent in 2007–2008; this includes a large proportion of marriage-related migration. NSSO (2010) presents most the recent changes in migration — including all categories of migrants — between 1983 and 2007–2008, and these are discussed in Chapter 1 of the India Migration Report 2011 (Rajan 2011: 1–6) as well. The proportion of migrants in both urban and rural areas has gone up by 3–5 percentage points, but this is entirely due to increase in female migration. The proportion of male migrants declined in rural areas and stayed the same in urban areas. In both years, intra-rural migration was still the most important form of migration (for further discussion, see Rajan and Mishra 2011). Much of the movement of labour in India has remained circular.4 This is of course the case with migration for seasonal occupations, particularly in rural areas (Deshingkar and Akter 2009; Deshingkar and Start 2003; also compare with Jha 2008), especially those with a high concentration of out-migrants from India’s poorest areas, for instance, western Orissa.5 But workers involved in urban occupations also tend to maintain links with areas of origins.6 This is confirmed by casual observation regarding domestic and other workers in cities, where very few have severed links with their villages of origin even if working in the city for many years. My own research among industrial labourers in Calcutta showed the same (de Haan 1994). The return of workers during the 1931 crisis was indeed part of a pattern of circulation, which was only very gradually changing, as my fieldwork in the early 1990s showed, with many workers remaining one foot in both urban and rural areas, and many moving back as the industry was in structural decline. In her research in the summer of 2009 for her MA dissertation, Astha Kapoor (2011) found similar patterns of responses: 4 The National Council for Rural Labour (NCRL) estimated the number of seasonal and circular migrants in rural areas to be 10 million; while other estimates suggest at least 30 million (Smita 2008: 5). 5 ActionAid estimates that approximately 2 million people migrate from the predominantly tribal districts of western Orissa to work in brick kilns of Andhra Pradesh (ibid.: 11). 6 NSSO (2010) shows 12 per cent of the population to consist of ‘return migrants’; the relatively smaller number of migrant households as compared to household members migrating suggests that migration remains predominantly circular.

Inclusive Growth and Economic Crises Y 231 When the factories didn’t reopen in November [2008] after Diwali, many of the workers left for their villages . . . Return migration . . . was a logical option in this case because the cost of living in the city is significantly higher than that in the villages. Most of the people who moved back had land to go back to, which means that the ties to the city are often superseded by the ties to the village because of the existence of agricultural land . . . However . . . even though the realisation of the crisis had set in by December, the masses only moved back in April . . . after the school exams of the children finished . . . a number of workers chose to struggle for a few months for the sake of their children’s education (ibid.: 45).

There is thus no contradiction between relatively low rates of urbanization, and high levels of population mobility. Apart from the fact that much migration remains within rural areas, the rural–urban migration too in many cases is a continued circular process, rather than constituting the classic rural–urban transition.

Migrants’ Characteristics: A Macro–Micro Paradox Most of the existing research describes individual motives for migration. The NSS data highlights that migration for marriage is quoted as the reason for 90 per cent of rural female migrants, with ‘search for employment’ as the factor for 18 per cent of migrants according to the 1987–1988 NSS data and 14 per cent in 1999–2000 (de Haan and Dubey 2006).7 In 2007–2008, 61 per cent of households had migrated for employment-related reasons, while only 10 per cent individuals had migrated for employment-related reasons — 46 per cent among men, and 1 per cent among women. While of course economic opportunities are central to labour migrants, field studies tend to reveal a variety of motivations for migration, shaped by conditions at origin and destination, and by the patterns of recruitment and migration networks. Family structure shapes both the urgency and limitations to migrate,8 as larger families tend to have a greater need to diversify resources, and the ability to maintain labour inputs when part of the family migrates. Age is key of course in terms of employment possibilities (young Over 50 per cent of urban male migrants quoted ‘employment related reasons’. 8 The pattern of circular migration described above is more adequately captured in the context of a family-oriented migration model (Parida 2010). 7

232 Y Arjan de Haan

men having most opportunities) but also in terms of a rite of passage, and the drive within youngsters to explore the world outside their immediate vicinity (and sometimes, away from parental control). Similarly, social–cultural factors structure migration patterns, including in terms of female migration and labour force participation, as discussed further in this chapter. Reasons for migration are very complex, and simple categories may lead to serious misreading of broader patterns, such as in the case of female migration ‘for marriage’. Studies on migration of the poorest workers have argued that it is inappropriate to conceptualize this along the model of individual choices postulated by neoclassical and dualist models. According to Jan Breman (1985), with capitalist production and old exploitative socio-economic relations in western India, labour migration leads to, and is part of, extreme forms of deprivation; while David Mosse et al. (2002) show the conditions of bondage in migration processes amongst adivasis. These forms of migration are usually organized through labour contractors, who often are moneylenders in areas of origin. A variety of studies show the diversity amongst labour migrants. The recent NSSO report on 2007–2008 data (NSSO 2010) lists the incidence of migration among different income groups, showing a higher propensity of migration of households in the top income deciles than in the lower ones.9 In line with this, we found for 1987– 1988 and 1999–2000 that poverty rates amongst migrants were much lower than amongst non-migrants, and the average years of schooling of migrants was higher, in both years, and in both rural and urban areas (de Haan and Dubey 2006). The Indian Village Studies project also showed that migrants were educationally better placed than non-migrants (Connell et al. 1976). However, we also suspected that the group of migrants was diverse, and found that inequality — measured by a Gini coefficient of household consumption — was higher amongst migrants than among non-migrants (de Haan and Dubey 2006). Patterns of landownership amongst migrants also illustrate diversity. John Connell et al. (1976) showed that the landless were least likely to migrate. K. N. S. Yadava et al. (1997) found a positive

9 For short-term migrants, this was the reverse (NSSO 2010: 93) but the numbers of recorded short-term migrants is very low.

Inclusive Growth and Economic Crises Y 233

relationship between landholding and migration in India (and that migrant households are socio-economically and educationally relatively better-off). In a comparative study in the 1980s, Amarjit S. Oberai et al. (1989) showed that in Bihar the poor and landless were (slightly) more likely to migrate; in Kerala the middle peasantry migrated more often, and in Uttar Pradesh all landed groups except the largest cultivators migrated frequently. While there is some degree of path dependency, these patterns change over time, as Arvind Das’s narrative (1985) about a village in Bihar shows, where sons of landowners were amongst the first to migrate, followed by the less well-off. An identical pattern has been detected with respect to the international migration from Punjab, where better-off earlier migrants sponsored the less well-off (Pettigrew 1977). Similarly, such diversity is demonstrated by data on the caste background of migrants. Indentured labour migrants formed an average broad-middle sample of India’s rural population (Tinker 1974), a pattern that seemed to be continued in the migration to the old colonial industries (de Haan 1994). Ben Rogaly et al. (2002) showed how caste is responsible for segmented migration streams in rural West Bengal. Jan Breman’s seminal work (1985) in western India stressed on the overrepresentation of lower castes and Harijans in circular migration. In the study of Mahabubnagar village in Andhra Pradesh, Vijay Korra (2010) observed that all Reddy (powerful caste) households had a migrant in urban areas, whereas migration to rural areas was much more common among Madiga (Dalit) and landless marginal farmer households. NSS data provides information about migration among different castes, though under-recording may be particularly serious here, as the types of migration of the most deprived groups may remain unrecorded more often than that of those better-off. Analysis of NSS data for 1987–1988 and 1999–2000 highlighted striking differences in the mobility of social groups: the proportion of recorded migrants among Dalits and Adivasis is on an average lower than among other groups (de Haan and Dubey 2006).10 NSS data for 1999–2000 and 2007–2008 referring to migrant households suggests a slightly different picture, with a somewhat lower proportion of migrants amongst

10

For an analysis of the corresponding 2001 Census data, see Bhagat (2011).

234 Y Arjan de Haan

Dalits, but higher share of migrants amongst Adivasis in urban areas — a fact that may be related to reservation. Patterns of migration change over time. Priya Deshingkar et al.’s re-survey (2008) in Madhya Pradesh showed that circular migration had become more beneficial and empowering for the poor, while also bringing greater returns to those with skills or strong social networks. It also showed increased participation by higher castes (and women) in migration as opportunities became more rewarding. In Palanpur in western Uttar Pradesh, higher castes were more prominently represented among migrants in 1983–1984, but in earlier years lower castes had secured outside jobs (Lanjouw and Stern 1989). Thus, a micro–macro paradox seems to emerge from this brief overview. National-level data highlights that migration is selective, with opportunities biased against the poor, which may be reinforced with technological change. But micro studies show very high rates of migration amongst poorest and socially marginalized groups, including in bonded and child labour. To ensure a better insight into the dynamics of migration and its relations to poverty, an interdisciplinary approach to the study of migration is needed, as demonstrated with respect to the gendered nature of migration.

Gender and Age Structure Migration Migration in India, as elsewhere, is strongly gendered, with regional and class differences, thus structuring the potential of migration to be an inclusive force. Census and NSSO data shows very low rates of female labour migration: between 70 per cent (in Kerala) and 94 per cent (in Bihar) of women in migrant households quote marriage as the primary reason for their move, though their labour force participation increases after marriage (Shanthi 2006: 22–25). Many studies show that women do migrate, but often face more barriers to mobility and access to opportunities.11 They receive lower wages, particularly in rural and casual urban occupations, and if they have low levels of education.12 They are highly represented in extremely exploitative occupations like sugarcane cutting (Teerink

11 For example, see Mukherjee (2001); Mehra and Gammage (1999); Neetha N. (2011) with respect to female domestic workers. 12 S. Madeshwaran (2010: 467) finds particularly high female/male differentials among scheduled caste regular rural workers.

Inclusive Growth and Economic Crises Y 235

1995), which are likely to go unrecorded in national statistics. The under-recording of this female migration in the Census of India and the NSS can be related to the fact that respondents are required to give only one reason for migration and that working women that move for marriage are not recorded as labour migrants; cultural inappropriateness to emphasise economic role, particularly vis-à-vis male interviewer; and the emphasis on primary and full-time work (Shanthi 2006: 5). Gender patterns of migration vary spatially and temporarily. For example, across India patterns of female rural-to-urban migration have differed in northern and southern India, with the south having higher rates of female migration (Singh 1984).13 This has been related to cultural norms, in particular northern Indian practices relating to the seclusion of women that affected rates of female out-migration and employment. Gendered patterns of labour migration also change over time. In the early 20th century, manufacturing industries had a substantial proportion of female workers, which declined gradually, thus contributing to a very low percentage of female labour in India’s formal employment. Mirroring a global trend, since the 1980s there has been a gradual trend of increasing female employment, in new manufacturing and service-sector jobs, and into selfemployment (Rustagi 2010). Apart from signalling gradual changes in migration patterns of migration, this may also be reflecting the pattern of increasing disparities within the labour market, with the better-educated benefiting more, while many remain trapped in poverty, and particularly women and children confined to the unregulated informal sector.

Impact of Migration The question of the impact of migration remains difficult to answer because of (a) the diversity of migrants and selectivity of migration, (b) the difficulty of calculating costs of migration, and (c) the complexity of isolating the impact of migration and remittances from broader household or livelihood strategies. Analyzing the impact

K. Shanthi’s analysis (2006) of the 55th round of the NSS shows slightly higher incidences of female migrants who were ‘never married’ in southern states and West Bengal. 13

236 Y Arjan de Haan

of migration using household survey data tends to be particularly limited,14 and needs to calculate the investment migrants and their families have to make before moving as well. Remittances have been among the most important impacts of migration in India. There is increasing evidence about the amounts that migrants send home and their effect clearly visible in states like Kerala. International migrants on an average sent back ` 52,000. Migrants within India remitted ` 13,000 on an average — hence, as 19 per cent of households reported an out-migrant, this is a not an insignificant contribution.15 Over 90 per cent of the income was spent on household consumption goods, confirming micro-study findings that show that remittances lead to limited investment. A small share of remittances in rural areas was used for debt repayment, whereas in urban areas savings and investment were a more common way of utilizing remittances. However, there are many negative sides to migration. Migrants may have to return without savings and remitting — a fact that may go unrecorded in the statistics — and even be unable to earn back the investment. Many migrant labourers have no option but to take children along (Smita 2008), thus potentially contributing to increased child labour and gaps in education, and thus possibly further transmitting poverty across generations. Health of migrants is adversely affected by poor living and working conditions, increased exposure to infectious diseases, lack of access to health care, and emotional stress related to the movement. Ravi Srivastava and S. K. Sasikumar’s conclusion (2003) that despite positive impacts on incomes and investment internal migration from poorer areas functions as a ‘safety valve’ is still valid.

14 For example, William Joe et al. (2009) analyze the net gain of rural–urban migration based on the probability of migrant and non-migrant population in different income quintiles (1999–2000 NSSO data), and show that migrants have much lower probability to be in bottom quintiles than non-migrant population in areas of origin. As they acknowledge, this does not provide information about individual households’ income. 15 Source: NSS data 2007–2008. This estimate is far below that of private transfers by the Reserve Bank of India. Remittances are particularly high in Kerala, Goa and Punjab, with recently a rapid increase in Uttar Pradesh. For a detailed analysis of national and international remittances, see Tumbey (2011).

Inclusive Growth and Economic Crises Y 237

At the bottom of the income hierarchy, the benefits from migration are likely to be least, and in cases of bonded labour, migration may actually reinforce conditions of bondage (Mosse et al. 2002). The NSSO data also appears to confirm the hypothesis postulated by Michael Lipton (1980) that migration can increase inequalities, as the better-off can afford to invest in migration opportunities (including international) that are likely to have higher returns. Remittances for rural households in the lowest income group were on average one-quarter of that for those in the highest income group; while this may still be progressive because other sources of income are even more unequally distributed, migration on an average reinforces initial inequalities. The impact of migration thus continues to constitute major questions, conceptually, empirically, and politically. The questions are increasingly important, as labour markets and migration patterns appear to become increasingly unequal. This poses a major challenge for the debate on inclusive growth.

Migration and the Paradigm of Inclusive Growth [M]igrant workers are the most vulnerable and exploited among the informal sector workers, and have not received any attention in the labour policy . . . effective and large-scale effort for vocational training in the labour intensive occupations is required . . . amenable to the special needs of the entrants to informal labour markets. In the destination States, the focus of public policy (including Labour Policy) should be to improve the conditions under which the bulk of these in-migrants live and work (Planning Commission 2007: 74).

The Government of India’s 11th Five Year Plan expressed the expectation that accelerated growth would lead to increased migration (ibid.: 21). The recent World Bank report Accelerating Growth and Job Creation in South Asia (Ghani and Ahmed 2009b), which attributes great importance to labour legislation as a cause of underutilization of labour, assumes that labour mobility is ‘the natural mechanism for promoting faster and inclusive growth regionally and globally’ (Ghani and Ahmed 2009a: 26). They argue that mobility within South Asia’s countries is low, with 96 per cent of the Indian population living in the state in which they were born, and attribute this to large distances, poor infrastructure, cultural factors, language and poor education, as well as policies preventing migrant

238 Y Arjan de Haan

workers from using safety nets in states to which they have migrated (Ghani and Ahmed 2009a: 26).16 However, the accumulated knowledge does not bear out the conclusion that mobility is low. Neither is there clear evidence that patterns of migration in India are changing substantially. The NSS and the Census of India confirm that proportions of migrants have not been increasing. The data even suggests a decline in rural male migration, and increasing selectivity — which would be in line with evidence on patterns of job-less growth and growing regional inequalities in India. At the same time, micro-studies show that while some migration does contribute to socio-economic improvements, other forms of migration can be hugely exploitative. This section therefore argues that public policy is critical for beneficial outcomes of migration, and thus for making economic growth more inclusive. First, policy objectives tend to have a strong bias towards reducing numbers of (internal) migrants. For example, rural development programmes and the social protection schemes like the National Rural Employment Guarantee Act (NREGA) often include an objective to reduce numbers of out-migrants, often specifically referring to (but not defining) ‘distress migration’.17 Emerging evidence does show that the NREGA led to some decrease in distress migration from villages, with workers stating preference to work in and around their villages, rather than bearing the social and other costs of migrating elsewhere in search of work (Sengupta 2009: 220, 223). Moreover, the National Commission for Enterprises in the Unorganized Sector (NCEUS) is hopeful that ‘programmes have made a significant contribution in not only enhancing income levels of the poor but have

16 The idea that migration in India has been low historically has a long genealogy in colonial and early post-colonial periods (Bhagat 2011). Various studies use net migration as evidence of labour mobility (Cashin and Sahay 1996; Purfield 2006), thus largely missing complicated patterns of migration. Kaivan Munshi and Mark Rosenzweig’s emphasis (2009) on permanent migration also biases their perspective on why mobility in India is ‘so low’. 17 According to the NREGA guidelines, its basic objective is to ‘enhance livelihood security in rural areas by providing at least 100 days of guaranteed wage employment in a financial year to every household whose adult members volunteer to do unskilled manual work.’ Other objectives include generating productive assets, protecting the environment, empowering rural women, and reducing rural–urban migration.

Inclusive Growth and Economic Crises Y 239

been helpful in stemming the rural and urban migration of the poor also’ (Sengupta 2009: 251). This emphasis on reducing migration per se is problematic. Migration tends to be part of households’ established livelihood strategies, which they are unlikely to give up easily, and public policy should focus on supporting these strategies rather than trying to minimize this. In fact, evidence shows that it is as likely that migration will continue or even increase with development (as enhanced resources become available and access improved) as decrease. Patterns of migration usually change, but mobility is part of societies’ development paths (many Indian states with high rates of out-migration also have substantial in-migration). This does not imply a call for a laissez-faire approach to policy on the presumption that migration leads to optimal outcomes; rather, policies that try to reduce exploitation of migrants should be based on an understanding of the complexity of migration patterns, motives and outcomes, and take into account migrants’ perspectives. Second, migrants often remain invisible in policies, law and statistics. This can enhance the vulnerability of migrants, excluding them from social services and rights. While there is a wealth of empirical studies on migration in India, the insights from the complexity of migration does not sufficiently inform the large-scale data collection. The NSS data may be under-recording migration, but also and perhaps more importantly insufficiently recognizes the complexity of migration processes, including with reference to female migration. We need a much better understanding of the changes in migration and employment that are now reflected in the 2007–2008 data. And when migrants are visible in empirical studies, they tend to be portrayed as victims — of economic exploitation, sexual oppression — thus denying migrants’ agency (Kapur 2010). Much of the analysis of migration has departed from a dualistic model, with its emphasis on the gradual absorption of a labour surplus into a modern sector. This also appears to be the perspective in the 11th Five Year Plan (Planning Commission 2007: 63).18 For decades this modernization and transition from rural and informal into an urban modern sector has happened at a moderate pace,

18 See Harris and Todaro (1970); Hatton and Williamson (1992); Lewis (1954); Ranis (2003); Todaro (1969).

240 Y Arjan de Haan

and there is recent evidence of continued stagnation and even some reversal of growth in the modern or organized sector. A focus on this slow transition tends to ignore the multiple forms of labour mobility that happens with those two sectors of the dual model. This theoretical focus may feed into public and policy debates, which tend to have — or at least obtain at points of time — an antimigrant bias. Policy makers tend to regard migrants as vagrants, or a threat to social order, and/or to national or regional identity. There is also a risk that the assumption of half-way existence on the road to modernization may hinder creative thinking about ways in which migrants can be supported. As the quote at the start of this section shows, the 11th Five Year Plan is explicit in the recognition of a severe gap in policies vis-àvis migrants.19 However, the dominant policy paradigm has been to improve the investment climate through which jobs would be generated, and support provided to the poorest through the flagship schemes (with the NREGA seen as supporting employment, whereas it is primarily a cash transfer scheme). There is little in terms of legislation and policies to support labour. The NCEUS recommendations that focused on support to the informal sector seem to have been shelved. Improvement of employability of and conditions for migrants, thus, does not appear to be high on the agenda, as even support to the ‘informal sector’ appears unpopular amongst policy makers. While more policy analysis is required in this area, the conclusion of Ravi Srivastava and S. K. Sasikumar (2003) largely seems to hold: ‘legislation (regarding migrants) fails because regulatory authorities are over-stretched; the state sees migrants as a low priority’ (ibid.: i), and acts like the Inter-State Migrant Workman Act remain without teeth. Moreover, many social protection schemes, tend to have a ‘sedentary bias’, focusing on the ‘resident’ population, and tending to exclude migrants. Working mostly in the informal sector, migrants 19 The Plan argues that if basic minimum conditions to the new migrants are not available, economic growth should be restrained. It argues for better implementation of legislations for migrants, through the Building and Other Construction Workers (Regulation of Employment and Conditions of Service) Act, 1976; the Building and Other Construction Workers (Cess) Act, 1976; the Workmen’s Compensation Act, 1923; the Minimum Wages Act, 1948; and the Unorganized Workers’ Social Security Bill, 2007 (Planning Commission 2007).

Inclusive Growth and Economic Crises Y 241

tend to lack work-related social security, and are seldom unionized. Migrants may not have access to the PDS and housing schemes. Schooling may not be available to migrants, particularly if they move seasonally. Health care and even immunization may not be accessible, either because of an absence of or large distance to health care centres and anganwadis, or because of discrimination against groups of migrants.20 Finally, the policy gaps vis-à-vis migrants are larger for India’s socially deprived women and men — including Dalits and adivasis — than for other migrants. While micro-studies highlight the large numbers of people moving for casual work, often over short-distance, the macro data shows continued under-representation of the most deprived in labour market and migration statistics. Over and above invisibility and lack of supportive policies or anti-migrant hostility, they suffer from stereotypes and denial of basic rights, as it is unlikely that their social identity becomes irrelevant after they move, even though some of the worst forms of caste discrimination are less prominent in urban areas. Marginalized migrant groups suffer additional barriers to accessing health and education provisions. Similarly, they are discriminated against in the labour market, as migrants as well as Dalit and adivasi women and men. Important programmes to address the discrimination against deprived groups have been in existence for decades. There have been significant achievements notably through affirmative action. However, disparities between groups have not disappeared, and the political emancipation has not been accompanied by a similar decisive process of social transformation. Programmes for the excluded have by and large conformed to the welfarist model of India’s social policy. As stressed by Shenggen Fan et al. (2000), the policies have not addressed some core aspects of discrimination, particularly in the sphere of labour markets, and the challenges to address the multiple disadvantages of being migrants and socially stigmatized thus remain. Thus, while national policy does recognize a policy gap in addressing the exclusion of migrants, much remains to be done to make

20 The National Rural Heath Mission (NRHM) recognizes the specific health needs of migrants as a group, but it is not clear whether separate initiatives for migrants are included.

242 Y Arjan de Haan

growth more inclusive at the national level.21 These challenges are particularly large for women and most deprived groups, for whom ‘regular’ forms of deprivation are augmented by the sedentary bias of public policies.

Conclusion: Supporting Migrants While many micro-studies describe the often extreme vulnerability and exploitation of migrants in India, macro-studies have continued to struggle with the role of migration in economic transformation, including during periods of crises. This chapter argues that an interdisciplinary approach is needed to chart the complexity of migration dynamics and how they relate to poverty and well-being. These questions are increasingly pertinent, as the policy objective of inclusive growth is an explicit response to the knowledge that India’s economic model has produced high growth figures, but with increasing inequalities and without a transformation in terms of absorption of labour within a ‘modern’ or ‘formal’ sector. A key insight into migration patterns globally and within India is that while migration is critical in many households’ livelihoods, it does not by itself produce structural change. People’s responses to opportunities are structured by initial economic, political and even social–cultural conditions. Migration can and does enhance wellbeing of individuals and the broader economy and society. But often migration reinforces inequalities within areas of origin, substantial benefits go to the better-off and well-connected regions, and extreme exploitation including bonded and child labour may be intensified through the grip of labour contractors and moneylenders. The changes in migration patterns over the last two decades appear to reflect the uneven growth pattern India has been experiencing, with increasing inequalities, and limited creation of ‘decent’ jobs. Patterns have not been stagnant, with, for example, women playing an increasing role in the labour force and migration. But macro data indicates that those better-off are benefiting more from opportunities provided through migration than lower-income groups, while micro-level studies continue to show severe discrimination against Jayati Ghosh (2010: 157–60) building on the Committee on the Elimination of Discrimination against Women (CEDAW) recommendations lists a number of essential proactive policies related to migration and gender empowerment. 21

Inclusive Growth and Economic Crises Y 243

poor migrants. Not much macro-data is yet available to assess the impact of the financial and economic crisis of 2008 and beyond on migration, but while the impact on international migration and remittances has been less dramatic than expected, internally the crisis is likely to have reinforced patterns of circular migration. This chapter has argued that inclusive growth policy can do more that at present to address the exploitation and exclusion of migrants. New social protection schemes like the NREGA unquestionably enhance the well-being of the poor during periods of downturn in the economy, but also as a more general safety net. But more can be done to enhance the participation of migrants in the growth process. This ought to start from accepting the reality of labour outside the modern sector, among which migrants occupy a specific position, as economic growth has not and will not in the near future substantially change the structure of employment, with over 90 per cent in the so-called informal sector. A range of initiatives shows that social policy does not need to exclude migrants and their families (see, for example, Smita 2008). Rural livelihood programmes and non-governmental organizations (NGOs) have experimented with support programmes for migrants, to ensure that they can avail of their basic rights and have the necessary identification and documentation. Impressive examples include Janarth’s schools for children of sugarcane migrant workers in western Maharashtra, SETU’s education interventions to prevent migration of children to unsuitable work sites in Gujarat, and Vikalpa’s support to children of migrants in villages in Bolangir and Nuapada districts in Orissa. National-level Sarva Shiksha Abhiyan (SSA) has introduced special schemes for out-of-school children, and encourages states to work with NGOs to reach children of vulnerable groups. Ravi Srivastava and S. K. Sasikumar (2003) argue that labour legislation can be enforced more strictly, with modification of laws where necessary. Advocacy campaigns can be mounted to address stereotypes about migrants, and migrants’ awareness of rights can be strengthened. The inclusive growth policy would make a great leap forward if these initiatives are forcefully promoted across the Indian economy and labour market.

References Bhagat, Ram B. 2011. ‘Internal Migration in India: Are the Underclass More Mobile?’, in S. Irudaya Rajan (ed.), Migration, Identity and Conflict: India Migration Report 2011, pp. 7–24. New Delhi: Routledge.

244 Y Arjan de Haan Breman, Jan. 1985. Of Peasants, Migrants and Paupers: Rural Labour Circulation and Capitalist Production in West India. Oxford: Oxford University Press. Cashin, Paul and Ratna Sahay. 1996. Internal Migration, Centre–State Grants and Economic Growth in the States in India. Washington, DC: International Monetary Fund. Castles, Stephen. 2009. ‘Migration and the Global Financial Crisis: A Virtual Symposium, Update 1.A: An Overview’, http://www.age-of-migration.com/ uk/financialcrisis/updates/1a.pdf (accessed 30 April 2009). Castles, Stephen and Mark J. Miller. 2010. ‘Migration and the Global Economic Crisis: One Year On, Update 1L: April 2010’. http://www.age-of-migration. com/uk/financialcrisis/updates/migration_crisis_april2010.pdf (accessed 8 September 2010). Chandrasekhar, C. P. 2010. ‘India’s New, High Growth Trajectory: Implications for Demand, Technology and Employment’. V. B. Singh Memorial Lecture, 52nd Annual Conference of the Indian Society of Labour Economics, Dharwad, 17–19 December. Connell, John, Biplab Dasgupta, Roy Laishley, and Michael Lipton. 1976. Migration from Rural Areas: The Evidence from Village Studies. Delhi: Oxford University Press. Das, Arvind. 1985. Changel: The Biography of a Village. New Delhi: Penguin. de Haan, Arjan. 1994. Unsettled Settlers: Migrant Workers and Industrial Capitalism in Calcutta. Hilversum: Verloren. ———. 2010. ‘Inclusive Growth? Labour Migration and Poverty in India in the 21st Century’. Keynote Paper, Annual Conference of the Indian Society of Labour Economics, Dharwad, December 2010. de Haan, Arjan and A. Dubey. 2006. ‘Are Migrants Worse Off or Better Off? Asking the Right Questions’, Margin — Journal of Applied Economic Research, 38 (3): 9–26. de Haas, Hein. 2007. ‘Turning the Tide? Why Development Will not Stop Migration’, Development and Change, 38 (5): 819–41. ———. 2008. ‘The Internal Dynamics of Migration Processes’, Journal of Ethnic and Migration Studies, 36 (10): 1587–1617. Deshingkar, Priya and Daniel Start. 2003. Seasonal Migration for Livelihoods in India: Coping, Accumulation and Exclusion. London: Overseas Development Institute. Deshingkar, Priya, Pramod Sharma, Sushil Kumar, Shaheen Akter and John Farrington. 2008. ‘Circular Migration in Madhya Pradesh: Changing Patterns and Social Protection Needs’, The European Journal of Development Research, 20 (4): 612–28. Deshingkar, Priya and Shaheen Akter. 2009. ‘Migration and Human Development in India’. Human Development Research Paper 2009/13, United Nations Development Programme. http://hdr.undp.org/en/reports/global/hdr2009/ papers/HDRP_2009_13.pdf (accessed 11 July 2009). Fan, Shenggen, Peter Hazell and Sukhadeo Thorat. 2000. ‘Government Spending, Growth and Poverty in Rural India’, American Journal of Agricultural Economics, 82 (4): 1038–51.

Inclusive Growth and Economic Crises Y 245 Gaetano, Arianne M. and Brenda S. A. Yeoh. 2010. ‘Introduction to the Special Issue on Women and Migration in Globalizing Asia: Gendered Experiences, Agency, and Activism’, International Migration, 48 (6): 1–12. Ghani, Ejaz and Sadiq Ahmed. 2009a. ‘Sustaining Rapid Growth in South Asia’, in Ejaz Ghani and Sadiq Ahmed (eds), Accelerating Growth and Job Creation in South Asia, pp. 3–41. New Delhi: Oxford University Press. ——— (eds). 2009b. Accelerating Growth and Job Creation in South Asia. New Delhi: Oxford University Press. Ghosh, Ajit. 2010. ‘Economic Development, Employment and Social Protection in India’. Presentation for Panel Discussion on Labour Markets and Social Protection in Emerging Economies, Dharwad, 16 December. Ghosh, Jayati. 2010. ‘Migration and Gender Empowerment’, in S. Irudaya Rajan (ed.), Migration, Identity and Conflict: India Migration Report 2010, pp. 134–62. New Delhi: Routledge. Harris, Nigel. 2010. ‘Immigration and State Power’, Economic and Political Weekly, 45 (5): 8–11. Harris, John R., and Michael P. Todaro. 1970. ‘Migration, Unemployment and Development: A Two Sector Analysis’, American Economic Review, 60 (1): 126–42. Hatton, Timothy J. and Jeffrey G. Willamson. 1992. ‘What Explains Wage Gaps between Farm and City? Exploring the Todaro Model with American Evidence, 1890–1941’, Economic Development and Cultural Change, 40 (2): 267–94. Iversen, Vegard, Kunal Sen, Arjan Verschoor and Amaresh Dubey. 2011. ‘Job Recruitment Networks and Migration to Cities in India’, in S. Irudaya Rajan (ed.), Migration, Identity and Conflict: India Migration Report 2011, pp. 144–72. New Delhi: Routledge. Jha, Veena. 2008. ‘Trickle down effects of Inter State Migration in a Period of High Growth in the Indian Economy’. Working Paper 253/08, Centre for the Study of Globalisation and Regionalisation, University of Warwick. Joe, William, Priyajit Samaiyar and Udaya S. Mishra. 2009. ‘Migration and Urban Poverty in India: Some Preliminary Observations’. Working Paper 414, Centre for Development Studies, Thiruvananthapuram. Kapoor, Astha. 2011. ‘Diamonds are for Never: The Economic Crisis and the Diamond Polishing Industry in India’, in Peter A. G. van Bergeijk, Arjan de Haan and Rolph van der Hoeven (eds), The Financial Crisis and Developing Countries: A Global Multidisciplinary Perspective, pp. 200–14. Cheltenham: Edward Elgar. Kapur, Ratna. 2010. Makeshift Migrants and Law: Gender, Belonging, and Postcolonial Anxieties. New Delhi: Routledge. Korra, Vijay. 2010. ‘Nature and Characteristics of Seasonal Labour Migration: A Case Study in Mahabubnagar District of Andhra Pradesh’. Working Paper 433, Centre for Development Studies, Thiruvananthapuram. http://cds.edu/ download_files/wp433.pdf (accessed 27 August 2010). Kumar, Prem. 2010. ‘Declining Number of Slums: Nature of Urban Growth’, Economic and Political Weekly, 45 (41): 75–77.

246 Y Arjan de Haan Kundu, Amitabh. 2009. ‘Exclusionary Urbanisation in Asia: A Macro Overview’, Economic and Political Weekly, 44 (48): 48–58. Lanjouw, Peter, and Nicholas Stern. 1989. Agricultural Changes and Inequality in Palanpur 1957–1984. Development Economics Research Programme Paper 24. London: London School of Economics. Lewis, W. Arthur. 1954. ‘Economic Development with Unlimited Supplies of Labour’, The Manchester School, 22 (2): 139–91. Madeshwaran, S. 2010. ‘Labour Market Discrimination in India: Issues in Theory and Methods’, The Indian Journal of Labour Economics, 53 (3): 457–80. Mehra, Rekha and Sarah Gammage. 1999. ‘Trends, Countertrends, and Gaps in Women’s Employment’, World Development, 27 (3): 533–50. Migration Development Research Centre (DRC). 2009. ‘Migration and the Financial Crisis: How will the Economic Downturn Affect Migrants?’. Briefing Paper 17, Sussex Centre for Migration Research, Brighton. http:// www.migrationdrc.org/publications/briefing_papers/BP17.pdf (accessed 11 July 2009). Mosse, David, Sanjeev Gupta, Mona Mehta, Vidya Shah, Julia Rees and the KRIBP Project Team. 2002. ‘Brokered Livelihoods: Debt, Labour Migration and Development in Tribal Western India’, in Arjan de Haan and Ben Rogaly (eds), Labour Mobility and Rural Society, pp. 59–88. London: Frank Cass. Mukherjee, Shekhar. 2001. ‘Low Quality Migration in India: The Phenomena of Distressed Migration and Acute Urban Decay’. Paper presented at the 24th International Union for the Scientific Study of Population Conference, 18–24 August, Salvador. Munshi, Kaivan and Mark Rosenzweig. 2009. ‘Why is Mobility in India so Low? Social Insurance, Inequality, and Growth’, Working Paper 14850, National Bureau of Economic Research, Cambridge. http://www.nber.org/papers/ w14850 (accessed 9 December 2011). N., Neetha. 2011. ‘Closely Woven: Domestic Work and Internal Migration of Women in India’, in S. Irudaya Rajan (ed.), Migration, Identity and Conflict: India Migration Report 2011, pp. 219–35. New Delhi: Routledge. National Sample Survey Office (NSSO). 2010. Migration in India 2007-08: NSS 64th Round (July 2007–June 2008). New Delhi: National Sample Survey Office, Ministry of Statistics and Programme Implementation. http://mospi. nic.in/Mospi_New/upload/533_final.pdf (accessed 9 December 2011). Oberai, Amarjit S., Pradhan H. Prasad and M. G. Sardana. 1989. Determinants and Consequences of Internal Migration in India: Studies in Bihar, Kerala, and Uttar Pradesh. New Delhi: Oxford University Press. Papademetriou, Demetrios G., and Philip L. Martin (eds). 1991. The Unsettled Relationship: Labour Migration and Economic Development. New York: Greenwood. Parida, Jajati Keshari. 2010. ‘Determinants of Migration and Remittance in India: An Empirical Evidence’. Paper presented at the 52nd Annual Conference of the Indian Society of Labour Economics, 17–19 December, Dharwad. Pettigrew, Joyce. 1977. ‘Socio-economic Background to the Emigration of Sikhs from Doaba’, Punjab Journal of Politics, 1 (1): 48–81.

Inclusive Growth and Economic Crises Y 247 Planning Commission. 2007. Eleventh Five Year Plan: 2007–12 — Volume 1: Inclusive Growth. New Delhi: Oxford University Press. http://planning commission.nic.in/plans/planrel/fiveyr/welcome.html (accessed 9 November 2011). Purfield, Catriona. 2006. Mind the Gap — Is Economic Growth in India Leaving Some States Behind?. Washington, DC: International Monetary Fund. http://www.imf.org/external/pubs/ft/wp/2006/wp06103.pdf (accessed 8 September 2009). Rajan, S. Irudaya (ed.). 2011. Migration, Identity and Conflict: India Migration Report 2011. New Delhi: Routledge. Rajan, S. Irudaya and U. S. Mishra. 2011. ‘Facets of Indian Mobility: An Update’, in S. Irudaya Rajan (ed.), pp. 1–6. New Delhi: Routledge. Rajan, S. Irudaya and Prabhat Kumar. 2010. ‘Historical Overview of International Migration’, in S. Irudaya Rajan (ed.), Governance and Labour Migration: India Migration Report 2010, pp. 1–29. New Delhi: Routledge. Ranis, Gustav. 2003. ‘Is Dualism Worth Revisiting?’. Center Discussion Paper No. 870, Economic Growth Center, Yale University, New Haven. http:// www.econ.yale.edu/growth_pdf/cdp870.pdf (accessed 9 December 2011). Ratha, Dilip, Sanket Mohapatra and Ani Silwal. 2010. ‘Outlook for Remittance Flows 2010–11: Remittance Flows to Developing Countries Remained Resilient in 2009, Expected to Recover During 2010–11’. Migration and Development Brief 12, Migration and Remittances Team, Development Prospects Group, World Bank. http://siteresources.worldbank.org/INTPROSPECTS/ Resources/334934-1110315015165/MigrationAndDevelopmentBrief12.pdf (accessed 9 December 2011). Rogaly, Ben, Daniel Coppard, Abdur Rafique, Kumar Rana, Amrita Sengupta and Jhuma Biswas. 2002. ‘Seasonal Migration and Welfare/Illfare in Eastern India: A Social Analysis’, in Arjan de Haan and Ben Rogaly (eds), Labour Mobility and Rural Society, pp. 89–114. London: Frank Cass. Rustagi, Preet. 2010. ‘Continuing Gender Stereotypes or Signs of Change: Occupational Pattern of Women Workers’, The Indian Journal of Labour Economics, 53 (3): 481–500. Sarkar, Jyoti Parimal. 2010. ‘Cross-Border Migration in Developing Countries’, in S. Irudaya Rajan (ed.), Governance and Labour Migration: India Migration Report 2010. pp. 113–33. New Delhi: Routledge. Sarkar, Sandip and Balwant Singh Mehta. 2010. ‘Income Inequality in India: Preand Post-Reform Periods’, Economic and Political Weekly, 45 (37): 45–55. Sengupta, Arjun K. 2009. The Challenge of Employment in India: An Informal Sector Perspective, Volume 1, New Delhi: Academic Foundation. http:// nceuis.nic.in/The_Challenge_of_Employment_in_India.pdf (15 August 2009). Shanthi, K. 2006. ‘Female Labour Migration in India: Insights From NSSO Data’. Working Paper 4/2006, Madras School of Economics, Chennai. http:// www.mse.ac.in/pub/santhi_wp.pdf (accessed 9 December 2011). Singh, Andrea Menefee. 1984. ‘Rural-to-Urban Migration of Women in India: Patterns and Implications’, in James T. Fawcett, Siew-Ean Khoo and Peter Colin Smith (eds), Women in the Cities of Asia. Migration and Urban Adap-

248 Y Arjan de Haan tion, pp. 81–107. Boulder: Westview Press. Smita. 2008. Distress Seasonal Migration and its Impact on Children’s Education. Brighton: Consortium for Research on Educational Access, Transitions and Equity (CREATE) and University of Sussex. http://www.create-rpc.org/ pdf_documents/PTA28.pdf (accessed 10 November 2008). Srivastava, Ravi and S. K. Sasikumar. (2003. ‘An Overview of Migration in India, its Impacts and Key Issues’. Paper presented at the Regional Conference on Migration, Development and Pro-Poor Policy Choices in Asia, 22–24 June, Dhaka. http://www.eldis.org/vfile/upload/1/document/0903/Dhaka_CP_ 2.pdf (accessed 9 December 2011). Teerink, Rensje. 1995. Women and Seasonal Labour Migration in India. Research Report, Indo–Dutch Project Alternatives in Development, Amsterdam. Tinker, Hugh. 1974. A New System of Slavery: The Export of Indian Labour Overseas, 1830–1920. London: Oxford University Press. Todaro, Michael P. 1969. ‘A Model of Labor Migration and Urban Unemployment in Less Developed Countries’, The American Economic Review, 59 (1): 138–49. Tumbey, Chinmay. 2011. ‘Remittances in India: Facts and Issues’. Working Paper No. 331, Indian Institute of Management, Bangalore. United Nations Development Programme (UNDP). 2009. Human Development Report 2009: Overcoming Barriers. Human Mobility and Development. Basingstoke: Palgrave Macmillan. Yadava, K. N. S., S. S. Yadava and R. K. Sinha. 1997. ‘Rural Out-migration and its Economic Implications on Migrant Households in India — A Review’, The Indian Economic Journal, 44 (2): 21–38.

13 Food Inflation and Financial Crisis in India: Impact on Women and Children S. Mahendra Dev During the last few years, there has been a significant increase in global food prices due to several structural and cyclical factors. India has experienced high food inflation in the last two years. People from lower income groups spend 60–70 per cent of their income on food and have little capacity to adapt as prices rise and wages may not adjust accordingly. Apart from the problem of increasing food prices, India is also facing the adverse impact of the global financial crisis since the third quarter of 2008. India has largely avoided the impact on the banking system. However, the crisis has adverse impact on the liquidity situation and economic growth in India. This in turn can have an adverse effect on the poor and food security of the country. Although the underlying causes for the rise in food prices and financial crisis are different, they are interconnected through their implications on financial stability, food security and political security (von Braun 2008a). At the global level, the capital was diverted from the collapsing housing market to speculation in agricultural futures. Speculative activities were partly responsible for the rising global food prices. The food crisis not only increased general inflation but also affected macroeconomic policies. Similarly, the financial crisis can affect employment, poverty, agriculture investment and social sector expenditures. Therefore, both food and financial crisis may have adverse impact on food and nutritional security of India and undermine the poverty reduction efforts and the gains over the last several years if large sections of the population do not cope with rising prices and the financial crisis. These two crises can potentially further exacerbate and deepen existing vulnerabilities in India for the poor and disadvantaged groups including women and children.

250 Y S. Mahendra Dev

The coping strategies due to these crises could have an adverse effect on the human development of women and children. The objective of this chapter is to examine analytical issues that would identify the pathways of the effects of rising food prices and the financial crisis on households, particularly women and children. It also outlines desirable macro and sectoral — in particular social protection — policies and measures that would mitigate the negative effects and have the strongest effect on safeguarding levels of living of the households, in particular in terms of nutrition, health, education, and enhance child protection.

Pathways that Lead to Poor Outcomes for Women and Children Both food and financial crises would affect the conditions of women and children. This section discusses the pathways for the poor outcomes among these vulnerable groups.

Pathways of Poor Outcomes Due to Rise in Food Prices There are several pathways that lead to poor outcomes in nutrition, health and education of children due to increase in food prices. One can group them into four categories: (a) impact on poverty, (b) macroeconomic impact and its affect on employment and social sector, (c) impact on nutrition and social protection programmes, and (d) women’s well being and intra-household decisions.

Impact on Poverty A recent study in eight countries estimates that the increase in food prices between 2005 and 2007 increased poverty by 3 percentage points on an average. Extrapolating these results globally suggests that, as a result of the rise in food prices, total world poverty may have increased by 73–105 million people (World Bank 2008). High food prices would have different effects on net sellers and net buyers. However, net buyers are large in number including all urban poor and majority of rural poor. It is true that net sellers are likely to benefit from rising food prices. However, the constraints on agriculture may prevent the farmers from responding in the short run. Some of the small producers who have a marketable surplus could become worse-off with higher prices. This is because typically

Financial Crisis and Food Inflation Y 251

a small producer sells the surplus immediately in the post-harvest season, when prices are low, and buys food when prices are high. In India, the rising food prices would have an adverse affect on the poor as (a) a large share of the expenditure of the poor is on food and (b) there are many more poor households who are net buyers of staple food than net sellers. The increase in food prices erodes the purchasing power of the poor. The incomes of the poor will fall and those who are just above the poverty line are likely to become poor. The progress towards achieving poverty and hunger would affect the Millennium Development Goals (MDGs) for some time to come. The impact on poverty would have four effects on the poor. These are: (a) nutrition status of pregnant and lactating women and of pre-school children, (b) the health status of women and children, (c) increase in child labour and withdrawal of children from school, and (d) the distress sale of productive assets.

Macroeconomic Impact and Effect on Employment and Social Sector Rising food, commodity and oil prices have increased the general inflation. Apart from the food price rise, increase in general inflation would hurt the poor and vulnerable sections. The reduced economic activity due to rise in food prices and general inflation would affect social sectors like education and health. At the macro level, reduction in economic growth would lead to reduction in tax/GDP ratio. This would have an adverse impact on employment and social sector expenditures. Even if growth does not decline, increase in food prices and the resultant increase in subsidies, tax concessions, etc. can reduce freedom of governments to raise social sector expenditures. This may adversely affect women and children in terms of education and health outcomes.

Impact on Nutrition and Social Protection Programmes Increase in food prices would worsen nutrition, especially among infants (0–24 months), pregnant and lactating mothers. Children incur long-term consequences from short-term shocks. The countries hardest hit by the food crisis already have the highest malnutrition rates. The malnutrition situation would deteriorate further with increase in food prices.

252 Y S. Mahendra Dev

The food price rise would affect the food consumption of households which in turn reduces consumption among women and children. Also, the households may spend more on cheaper, high calorie staples and less on foods rich in protein and vitamins, such as meat, fish, dairy, fruit and vegetables, thereby reducing the quality of their diet. This will have significant negative repercussions for morbidity, mortality, cognitive abilities, and growth. According to Howarth Bouis (2008), there are four basic factors for the above conclusion: ‘(I) Expenditures on non-staple foods by poor consumers comprise 40–60% of total expenditures on food; (II) Demand for food staples (rice, wheat, maize etc. depending on the geographical region and culture) is highly inelastic. Income and price elasticities for food staples in the aggregate are low; (III) In diets, minerals and vitamins are concentrated in non-staple foods; energy is concentrated in staple foods and (IV) Current intakes of vitamins and minerals are already too low, resulting in high prevalence rates of micronutrient deficiencies. Modest decrease in current intakes of minerals and vitamins will drive these prevalence rates significantly higher, with severe consequences for the nutritional status of the poor and public health’ (ibid.: 1).

The micronutrient malnutrition will be a widespread health consequence of high food prices. According to Christine Klotz et al. (2008), increased staple food prices around the world are forcing households to reduce their consumption of micronutrient rich foods, which will have a range of health consequences depending on the pre-existing nutritional status of a population. Rising food prices would affect the social protection programmes, particularly the food-based schemes. For example, India has food supplementation programmes such as Integrated Child Development Services (ICDS) and mid-day meal schemes for primary school children. The Government of India provides subsidized food grains and bears the cost of converting food grains into freshly cooked meals by including pulses, oil and vegetables. These social protection programmes benefit a large segment of the vulnerable sections.

Women’s Well-Being and Intra-household Decisions The high food prices have different effects on women consumers and producers. The rise in food prices could increase incentives for producers. But due to constraints on input supplies and other

Financial Crisis and Food Inflation Y 253

factors, past evidence suggests that higher prices may not necessarily stimulate production by female farmers (Quisumbing et al. 2008). At the household level, the burden on the daily lives of women time may increase with increase in food prices as they try to manage the household budget with increasing expenditure on food and/or paid employment, in turn affecting children’s nutrition and reducing the time available for their care. Intra-household dynamics are important to know the impact of rise in food prices on children. Intrahousehold allocation of resources to children depends on several factors such as women’s empowerment and education, the household wealth and asset base, social protection measures reaching the household, etc (Holmes et al. 2008). One of the coping mechanisms of the households relates to decisions regarding children’s activities and time. In some cases, children may be withdrawn from school to take up paid work activities. In some other cases, children will have less time for studying and leisure because of involvement in paid or unpaid activities.

Pathways that Lead to Poor Outcomes due to Financial Crisis The financial turmoil that started in the US financial system as a result of defaults of sub-prime mortgage loans has blown into an unprecedented financial crisis. It has affected the international money, credit, equity and foreign exchange markets. The global situation deteriorated massively after mid-September 2008 following the collapse of Lehman Brothers Holdings Inc., one of the top five investment banks in the US. There has been a massive choking of credit since then and a global crash in stock markets. The slowdown that was expected in the global economy became much worse with the US, Europe and Japan moving into recession. A crisis of this magnitude in industrialized countries is bound to have an impact on India. The indirect impact of the crisis was transmitted through capital flows, financial markets and trade (real sector). The foreign exchange reserves have declined in India and created a liquidity crisis due to global crisis as well as tightening of monetary policy earlier. It affected the exports from India and thus led to decline in growth and employment. On the other side, the lower growth adversely affected social sector expenditures. More importantly, the financial crisis is likely to have negative impact on the agricultural sector and food security. The pathways

254 Y S. Mahendra Dev

through which the agricultural markets will be affected are on both demand and supply sides (FAO 2008). The slowing down of economic growth would affect demand for agricultural commodities. Prevailing uncertainty and consequent negative market expectations could further dampen demand. On the supply side, reduction of agricultural prices may result in cutback in agricultural production. Although oil and fertilizer prices declined, in general, input prices could be sticky and this would affect the supply of agricultural commodities. The financial crisis is likely to have a negative impact on agricultural credit which is widely regarded as a major constraint to agriculture in developing countries. Farmers who took advantage of rising agricultural prices to invest in raising production may find themselves unable to pay off their debts because of falling output prices. The combination of falling agricultural prices, reduction in agricultural investment and access to credit may have a negative effect on agricultural production. It would have serious implications for food security. Poverty reduction efforts may be stalled because of low employment and volatility in food prices. The negative impact on household income would affect the well-being of women and children.

Vulnerabilities and Impact on Children The vulnerabilities have both short-term and long-term impact on children. Decline in food consumption and expenditure on health would affect the nutrition of children. As mentioned earlier, other effects include reduction of expenditure on non-food items like health and education, selling of productive assets by the family, children dropping out of school, increased use of child labour, debt, etc. A study of Andhra Pradesh shows that the impact of risk responses was increased indebtedness. The extent of borrowing that took place from moneylenders and the rich farmers is quite high (nearly 90 per cent) (SERP 2001). Increased indebtedness is not only a short-term burden but can also limit the ability of poor households to invest in future income-generating activities, including that of the children. Looking at the responses for the different groups, some people from the Scheduled Caste (SC) and Scheduled Tribe (ST) sections, and backward classes had to revert to seasonal migration. Migration affected their children’s education and health. Interestingly,

Financial Crisis and Food Inflation Y 255

dependency on family and friends was most popular among the backward classes, limited with SCs and not reported by STs, reflecting the poverty status of these groups in general who are little able to assist each other financially (SERP 2001).

Rising Food Prices and Household Responses A Rapid Price Impact Survey was conducted in Bangladesh in 2008 by the World Bank to observe the household responses to price rise (Viswanath 2008). Around 1,200 rural and 800 urban households were interviewed in this survey. The results of the survey are given in Table 13.1. It shows that around 43 per cent households in rural areas reduced their educational expenses while 9 per cent of rural households have taken children out of school.

Hunger Index in India The vulnerabilities across states in India can be examined by looking at the hunger index of the International Food Policy Research Institute (IFPRI) (Menon et al. 2008). The hunger index shows that relatively developed states like Gujarat, Maharashtra and Karnataka have a rank of hunger index that is higher as compared to states like Uttar Pradesh and Rajasthan, and closer to states like Orissa and Chhattisgarh (Table 13.2). It shows that poverty and hunger are not strictly correlated and vulnerability of children is high even in developed states like Gujarat, Maharashtra and Karnataka. Table 13.1: Household responses to the food price increase (in per cent): Bangladesh

Reduce Quantity of Food Intake Switch to Lower Quality Food Reduce Non-food Expenditures Spend Savings/Sell Belongings Take Out Loans Gift/Help from Community Members Take Children Out of School Decrease Education Expenses Work More/Increase Production Stop Loan Payment Source: Viswanath (2008).

Urban

Rural

Total

72.3 86.9 86.0 43.5 46.3 0.9 6.5 32.5 24.6 3.1

77.9 87.8 86.5 46.5 59.9 9.3 8.8 43.2 39.8 6.1

75.7 87.5 86.3 45.3 54.5 6.0 7.9 39.0 33.8 4.9

256 Y S. Mahendra Dev Table 13.2: Hunger index and components in India: States

State India Andhra Pradesh Assam Bihar Chhattisgarh Gujarat Haryana Jharkhand Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal

Proportion of Prevalence underweight of calorie among children under less than five nourishment years old 20.0 19.6 14.6 17.3 23.3 23.3 15.1 19.6 28.1 28.6 23.4 27.0 21.4 11.1 14.0 29.1 14.5 18.5

42.5 32.7 36.4 56.1 47.6 44.7 39.7 57.1 37.6 22.7 59.8 36.7 40.9 24.6 40.4 30.0 42.3 38.5

Under-five India mortality State India rate, reported hunger hunger as deaths per index index hundred score ranking 7.4 6.3 8.5 8.5 9.0 6.1 5.2 9.3 5.5 1.6 9.4 4.7 9.1 5.2 8.5 3.5 9.6 5.9

23.31 19.54 19.85 27.30 26.65 24.69 20.01 28.67 23.74 17.66 30.90 22.81 23.79 13.64 20.99 20.88 22.17 21.00

3 4 15 14 13 5 16 11 2 17 10 12 1 7 6 9 8

Source: Menon et al. (2008).

It may be observed that social disparities overwhelm regional disparities regarding poverty and vulnerability and incidence of malnutrition. For example, the numbers in Table 13.3 show that SCs/STs and Muslims suffer more from poverty and vulnerability as compared to Other Backward Classes (OBCs) and other castes. The incidence of malnutrition (chronic energy deficiency) among women is the highest for SCs/STs, followed by OBCs and Muslims. The malnutrition among women for OBCs is much lower than other castes. This is true of all the states in India.

Possible Ways of Tackling the Problem of Adverse Impact on Children In this section, we discuss the policies at macro and sectoral level and social protection measures needed in order to protect children from the two crises and improve their well-being. It is clear that the food

Financial Crisis and Food Inflation Y 257 Table 13.3: Incidence of malnutrition among women (chronic energy deficiency), 2005–2006 State India Andhra Pradesh Assam Bihar Chhattisgarh Gujarat Haryana Jharkhand Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal

Total population

SC/ST population

OBC population

Muslim population

Others

35.6 33.5 36.5 45.1 43.4 36.3 31.3 43.0 35.5 18.0 41.7 36.2 41.4 18.9 36.7 28.4 36.4 39.1

42.7 38.4 34.5 58.4 46.6 48.0 36.4 44.6 41.8 24.1 48.7 43.6 50.6 26.7 44.1 36.1 43.6 45.2

36.0 37.0 30.4 43.3 44.8 40.5 33.1 45.3 34.9 18.6 42.2 36.1 39.6 18.0 33.3 26.8 35.4 39.2

35.1 27.6 46.0 49.6 28.9 37.0 49.0 47.3 26.9 15.6 37.4 23.8 63.5 22.5 36.1 20.7 36.4 37.4

27.5 22.2 31.4 31.4 26.2 22.9 26.6 26.5 31.7 18.0 27.7 34.5 30.7 14.4 32.3 8.3 28.3 31.5

Source: Compiled from Kannan (2008). The data is from the 2005–2006 National Family Health Survey (NFHS-3).

price increase and financial crisis have a number of potential implications for children’s welfare and that policy and programming should seek to minimize the negative impact. In this context, it is important to ensure that we have measures that combine policies aimed at alleviating household poverty along with specific child-focused policies.

Macro Policies The macro policies should follow pro-poor and people-centric policies. Growth and equity objectives should be pursued simultaneously rather than following the ‘growth first and equity next’ approach. Here we are concerned with policies that mitigate household poverty in order to reduce the negative impact on women and children and improve their well-being. Several governments have responded in varying degrees to contain the rise in prices as well as to mitigate the adverse effects on the poor. The Indian government has used several policies to

258 Y S. Mahendra Dev

contain prices. So far the policies taken by the government in the first round were aimed at stabilizing food prices. It has used instruments like trade bans, administrative controls on private trade, import tariffs, subsidies, etc. to contain food prices. However, medium- to longer-term agenda should be to improve agricultural productivity and improve working of social protection programmes. In order to reduce the impact of the financial crisis, India has taken several monetary and fiscal measures. The monetary and fiscal measures announced by the RBI and government are in the right direction — to improve liquidity and investor confidence in the economy. But, there are different views on the policies that have to be taken to come out of the financial crisis. One view is that demand should be increased through direct fiscal action by the government. The objective of the larger government spending must be the reversal of the squeeze on the living standards of the people. ‘In short, the new paradigm must entail inter alia a foodgrain-led growth strategy (on the basis of peasant, not corporate, agriculture), sustained through larger government spending, which simultaneously rids the world of both depression and financial and food crises’ (Patnaik 2008). In other words, the stimulus package has to be towards the agricultural sector and primary food producers. There is another view that is given by the National Commission for Enterprises in the Unorganized Sector (NCEUS 2008). According to the NCEUS, immediate attention has to be given to protect the livelihood security, employment and income of the vast majority of people who are either poor or vulnerable and in doing so overall economic growth has to be stimulated as well. The higher economic growth of the Indian economy in the last 20 years bypassed 77 per cent of the people. Most of them are illiterate or have only primary level education, suffer from malnutrition and ill-health, and are unemployed or belong to the working poor, or are self-employed unorganized workers. Among the farmers, 84 per cent are marginal and small farmers working on not more than 2 hectares of land. Agricultural labourers, who are the bottom layer of the occupational structure, include a majority of SC and ST workers, and are around 89 million in number. The smallest segment of the unorganized sector (with investment below ` 500,000 in Plant and Machinery) gets less than 2 per cent of net bank credit while investment below ` 2,500,000 gets 4.3 per cent of formal sector credit.

Financial Crisis and Food Inflation Y 259

The NCEUS is of the view that given the effect of the global financial meltdown on the economy and particularly the poor who are the unorganized workers, a major response is urgently required. The Commission advocates a major fiscal stimulus comprising of: (a) programmes to boost pro-poor public investment in physical and social infrastructure; (b) programmes/schemes which protect and promote incomes of the poor; (c) expansion in scope and coverage of social security schemes of the unorganized workers. The macro pro-poor policies are expected to help increase the incomes for the poor households and reduce the negative effects of higher food prices and the financial crisis. Improvements in household income can have a positive effect on the well-being of women and children.

Sectoral Policies Growth in agriculture, rural non-farm sector, and sectoral level policies on health and nutrition and women’s empowerment are important in order to tackle the problem of rise in food prices and the financial crisis.

Raising Agriculture Growth Agriculture growth is one of the important components in resolving the food price and financial crises. With the onset of the financial crisis, economic growth has slowed down. Low economic growth is likely to have second round effects on investment and productivity with direct impact on food prices and food security (von Braun 2008b). The International Food Policy Research Institute (IFPRI) examines the following two recession scenarios during the period 2005–2020 at a global level (ibid.). Scenario 1: Economic growth is reduced by 2–3 percentage points depending on world regions, and agricultural productivity and investments are maintained by wise policy. Scenario 2: Economic growth is reduced as in scenario 1, and agricultural investment and productivity also decline, in line with the reduced economic growth. Food prices would increase significantly under Scenario 2. Poor people are likely to consume fewer calories due to recession.

260 Y S. Mahendra Dev

Globally, 16 million more children will be malnourished in 2020 when compared with the baseline scenario. However, if, developing countries can maintain agricultural productivity and investments under a recession (Scenario 1), the negative effects of the financial crisis can be avoided. Under this scenario, increase in cereal prices and the number of malnourished will be lower and per capita calorie consumption much higher by 2020. The analysis by the IFPRI indicates the importance of agriculture for food security of households, women and children. Therefore, India has to focus more on the agricultural sector. There are several changes required for India to achieve 4 per cent growth and equity in agriculture. The supply and demand side constraints have to be removed to increase growth. The support systems have to be tuned in order to improve productivity and incomes of farmers with emphasis on small and marginal farmers and dry land areas. One of the differences between the ‘green revolution’ in the 1960s and 1970s and the present ‘second green revolution’ is that risk is higher in the latter approach as it has to concentrate more on dry land areas. Trade liberalization has also raised the uncertainty. Thus, policies have to keep in mind increasing risk in agriculture. Agricultural policies have to be gender sensitive as the share of women is increasing. Cost reduction in agriculture is important for competing in a globalized world. The crop sector may not be able to grow at 4 per cent per annum. But horticulture and allied activities like dairy, poultry and fisheries have to grow at the rate of 6–7 per cent to achieve 4 per cent growth in agriculture. There are basically six factors which need to be focused on in the short and medium terms. These are: (a) infrastructure, (b) land and water management, (c) research and extension, (d) inputs including credit, (e) marketing including price policy, and (f) diversification and development of the rural non-farm sector. Institutions have to be developed in all these aspects.

Rural Non-farm Sector Poverty cannot be removed with 55 per cent of workers in the agricultural sector. There is a need to promote the rural non-farm sector. India currently produces about 50 million tonnes of fruits and 90 million tonnes of vegetables. Only 2 per cent of these fruits and vegetables are processed as against 23 per cent in China, 78 per cent in the Philippines and 83 per cent in Malaysia. Half of those engaged

Financial Crisis and Food Inflation Y 261

in agriculture are still illiterate and just 5 per cent have completed higher secondary education. Even in 2004–2005, around 60 per cent of rural male workers and 85 per cent of rural female workers were either illiterate or had been educated up to primary level. In other words, education and skills are constraints. India can observe China for a lesson on rural transformation. China has experienced a structural transformation in the last three decades. The state’s role has been decisive in building up the physical and social infrastructure (including land reforms). India should learn reforms in agricultural growth, rural non-farm employment, public investment and human development from China’s example. The impact of growth on poverty reduction is quite significant (Rao 2005). China started with agricultural reforms. Agricultural growth was quite high. The economic and institutional reforms in the whole economy created space for rural non-farm sector (Township and Village Enterprises [TVEs] and others). Diversification towards the rural non-farm sector in China is one of the primary factors responsible for rural poverty reduction (poverty is only 3 per cent). This was partly due to agricultural productivity.

Sectoral Intervention in Nutrition and Health Economic growth alone cannot address the problems of nutrition and health of women and children. Sector-specific interventions are required in order to improve nutrition and health outcomes. For example, a package consisting of expanded child and material immunization, antenatal care coverage, nutritional supplementation (including promotion of exclusive breast feeding) and home based neo-natal services (including treatment of pneumonia) is likely to bring about significant reduction in both infant mortality and child malnutrition (Deolalikar 2004). There is also a need for cooperation across sectors. There is a need for synergy between the processes directed at improving drinking water facilities, sanitation and public hygiene, access to elementary education, nutrition and poverty alleviation and those that increase access to public health and medical services.

Women’s Empowerment and Impact on Child Well-being Advancement in women’s status and well-being would have a positive spill over impact on child welfare. A comprehensive review

262 Y S. Mahendra Dev

of the linkages between women’s empowerment and child wellbeing brought out clearly that there are four mechanisms for transmitting the effects of the former onto the latter (Jones et al. 2007). They include maternal education, economic empowerment, intrahousehold decision-making power and community-level empowerment. Mother’s education is positively associated with better child education, health and nutrition outcomes. Women’s economic empowerment — i.e., greater access for women to financial resources — not only improves the status of women within the household but also leads to more investment in their children. As a result of greater economic resources, women’s empowerment may translate into better outcomes for children because mothers are likely to advocate the interest of their children in intra-household bargaining and be taken seriously by their male partners.

Social Protection Measures Comprehensive social protection programmes are required to address the negative effects due to food price crisis and financial crisis. If we leave everything to market and growth, we cannot take care of risk and inequality. The presence of social protection measures can maintain acceptable minimum floor for social cohesion and can prevent irreversible losses of human capital in hard times. The recent theory and evidence ‘offers a new perspective on social protection policies in poor countries, suggesting that there is scope for using these policies to compensate for the market failures that perpetuate poverty, particularly in high-inequality settings’ (Ravallion 2003: 27). India has many social protection programmes. The major schemes that exist currently for the poor in India fall into four broad categories: (a) food transfer like public distribution system (PDS) and supplementary nutrition, (b) self employment, (c) wage employment, and (d) social security programmes for unorganized workers. There are also many direct programmes for women.1 Some of the social protection schemes have to be scaled up and effectiveness has to be improved in order to reduce the negative effects of food price rise and the financial crisis.

1

See GoI (2005).

Financial Crisis and Food Inflation Y 263

Social Security Programmes for Unorganized Workers It is well known that only organized workers, constituting less than 10 per cent of total workers in India, get all the social security benefits. In other words, more than 90 per cent of Indian workers who belong to the unorganized sector do not have contingent social security benefits. The social security problems of the unorganized workers can be divided into two sets. The first one is the capability deprivation in terms of inadequate employment, low earnings, and low health and educational status, which are related to general deprivation of poorer sections of the population. The second one is adversity in the sense of absence of adequate fallback mechanisms to meet contingencies such as ill health, accident, death, and old age (NCEUS 2006). At present the central government, the state governments and NGOs provide social security benefits to some unorganized workers. But, these initiatives constitute less than 10 per cent of the unorganized workers. The NCEUS proposed a universal but minimum level of social security for the unorganized workers. It consisted of: (a) sickness and maternity, (b) disability and death and (c) old age security in the form of a national pension for those belonging to below poverty line (BPL) households and a contributory provident fund for those belonging to above poverty line (APL) households. The NCEUS proposal is estimated to cost only half a per cent of the GDP to the government for all the unorganized workers.

Measures of Fiscal Stimulus and Social Protection to Reduce Negative Effects of Financial Crisis The NCEUS proposed several urgent measures and fiscal stimulus to protect the informal economy. These measures are: (a) Enhance Pro-poor Public Investment: This includes expansion of rural infrastructure consisting of roads providing connectivity, housing, drinking water, sanitation and rural production infrastructure. (b) Strengthen and Expand the National Rural Employment Guarantee Programme (NREGP): With few exceptions, the employment generated under the NREGA is less than 100 days. The Commission provides suggestions for strengthening and expanding the NREGA.

264 Y S. Mahendra Dev

(c) Introduce an Urban Employment Guarantee Programme: In order to complement the NREGA, an urban employment guarantee programme has to be introduced. The activities can include low-income housing (especially for slum dwellers), electrification, water supply, slum improvement, low cost waste management, etc. (d) Strengthen and Expand Self-Employment Programmes: A number of measures have to be taken to graduate micro finance to livelihood finance. The measures can assist both farm and non-farm workers who are likely to face reduced prospects of employment in the wage market. (e) Special Programme for Marginal and Small farmers: Institutional credit for small and marginal farmers has to be improved. Special programmes like land improvement, minor irrigation, capacity building, training, etc. can be given to these farmers. ( f ) Enhancing Access to Credit to Micro Enterprises: In times of financial crisis, the small producers will be rationed out of the credit market. Steps have to be taken to increase credit to unorganized sector micro enterprises. (g) Create a National Fund for the Unorganized Sector: A national fund for the unorganized sector has to be established in order to step up credit and developmental effort for this sector. (h) A Programme for Skill Development in the Informal Economy: The government has provided measures to expand skill training. But they remain restricted mostly to organized workers. There is a need to have a programme for receiving on-job training in unorganized enterprises. This scheme aims at workers who have primary, but less than secondary level education. (i) A National Minimum Social Security for Informal Workers: As mentioned above, the Commission proposed a universal but minimum level of social security for all the unorganized workers in the country. ( j) Strengthen and Expand Investment in Human Development of the Poor and Vulnerable: The families and children in the informal economy suffer from capability deprivation. Given the conditions of poverty and vulnerability and low human capabilities, there is a need to strengthen and expand

Financial Crisis and Food Inflation Y 265

programmes that improve human development of these households. Here the programmes such as the Sarva Shiksha Abhiyan (SSA), the National Rural Health Mission (NRHM), the PDS, and the ICDS should be expanded to the unorganized workers and strengthened. In order to save the informal sector from collapse, the NCEUS has sought a special package of ` 58,000 crore to implement all the above programmes. If implemented, these programmes would create about 57 million additional jobs over the next five years.

Investing in Social protection Programmes The social protection programmes discussed in this section can help families to cope with the double crisis of rise in food prices and financial meltdown. These programmes also include child-specific policies such as ensuring adequate nutrition, school feeding programmes, health, and education for children. At the international level, there are many examples of good social protection programmes. Cash transfers are preferred to food or other in-kind transfers as cash increases the purchasing power of the households (see Heady and Fan 2008). Conditional cash transfers that have worked well include the food-for-education program in Bangladesh, Mexico’s Programa Nacional de Educacion, Salud y Alimentacion (PROGRESA) programme and the Bolsa Escola in Brazil.

Governance The focus of reforms can now be shifted to more efficient delivery systems of public services. It has been recognized that better governance is very important for equitable development. This is important for better implementation of sectoral policies and social protection programmes. Social mobilization, community participation and a decentralized approach are needed. It may, however, be noted that governance has to be contextualized in relation to socioeconomic environment. Appropriate institutions are needed for better implementation of policies and programmes. For example, rural institutions in areas like land, water, marketing of agricultural and non-agricultural products, credit, technology and infrastructure are needed for better governance. Similarly, people-centric programmes

266 Y S. Mahendra Dev

and institutions are needed for better implementation of social protection schemes. Finally, the rights approach plays an important role in improving implementation of development programmes. Right to food, health, education, employment, information, etc. puts pressure on governments to deliver the services to citizens.

Concluding Observations The objective of this chapter was to examine analytical issues that would identify the pathways of the effects of the rising food prices and the financial crisis on households, particularly on women and children. It also outlined desirable macro, sectoral and social protection policies and measures that would mitigate the negative effects and protect/improve levels of living of the households, in particular with regard to nutrition, health, education, and enhance child protection. The poor and vulnerable were significantly left behind even before the two crises. Rising food prices and the financial crisis would further undermine food security and livelihoods of the most vulnerable by eroding their already limited purchasing power. India has many social protection programmes. The present major schemes for the poor in India fall into four broad categories: (a) food transfer like the PDS and supplementary nutrition, (b) self-employment, (c) wage employment, and (d) social security programmes for unorganized workers. There are also many direct programmes for women. Some of the social protection schemes have to be scaled up and effectiveness has to be improved in order to reduce the negative effects of the food price rise and financial crisis. Child specific programmes are also needed.

References Bouis, Howarth. 2008. ‘Rising Food Prices will Result in Severe Declines in Mineral and Vitamin Intakes of the Poor’. HarvestPlus, International Food Policy Research Institute, Washington, DC. http://alternativeenergy.procon. org/sourcefiles/2008HBouisfoodprices.pdf (accessed 9 December 2011). Deolalikar, Anil B. 2004. Attaining Millennium Development Goals in India: Roles of Public Policy and Service Delivery. Washington, DC: World Bank. Food and Agriculture Organization (FAO). 2008. ‘Agricultural Commodity Markets and the Financial Crisis’, Food Outlook, November. Government of India (GoI). 2005. ‘Economic Survey 2004–2005’. Ministry of Finance, Government of India.

Financial Crisis and Food Inflation Y 267 Heady, Derek and Shenggen Fan. 2008. Anatomy of a Crisis: The Causes and Consequences of Surging Food Prices. Washington, DC: International Food Policy Research Institute. Holmes, Rebecca, Nicola Jones and Steve Wiggins. 2008. ‘Understanding the Impact of Food Prices on Children’. Plan UK and Overseas Development Institute, London. http://www.odi.org.uk/resources/docs/3345.pdf (accessed 12 December 2011). Jones, Nikola, Madhuri Mukherjee and S. Galab (2007), ‘Ripple Effects or Deliberate Intentions? Assessing Linkages between Women’s Empowerment and Childhood Poverty’. UNICEF/Young Lives Social Policy Paper 002, Overseas Development Institute. Kannan, K. P. 2008. ‘Dualism, Informality and Social Inequality: An Informal Economy Perspective of the Challenge of Inclusive Development in India’. Presidential Address, 50th Annual Conference of the Indian Society of Labour Economics, 13–15 December, Giri Institute of Development Studies, Lucknow. Klotz, Christine, Saskia de Pee, Andrew Thorne-Lyman, Klaus Kraemer and Martin Bloem. 2008. ‘Nutrition in the Perfect Storm: Why Micronutrient Malnutrition will be a Widespread Health Consequence of High Food Prices’, Sight and Life, (2/2008), pp. 6–13. http://www.sightandlife.org/ images/stories/pageimages/content/magazine/s%26l_magazine_0208.pdf (accessed 12 December 2011). Menon, Purnima, Anil Deolalikar and Anjor Bhaskar. 2008. The India State Hunger Index: Comparisons of Hunger across States. Washington, DC: International Food Policy Research Institute. NCEUS (National Commission for Enterprises in the Unorganized Sector). 2006. ‘Social Security for Unorganized Sector’. National Commission for Enterprises in the Unorganized Sector Report, ILO Subregional Office for South Asia, New Delhi. http://www.ilo.org/gimi/gess/RessShowRessource. do?ressourceId=2990 (accessed 12 November 2011). ———. 2008. ‘The Global Economic Crisis and the Informal Economy in India: Need for Urgent Measures and Fiscal Stimulus to Protect the Informal Economy’. Note submitted to the Prime Minister of India, National Commission for Enterprises in the Unorganized Sector, New Delhi. Patnaik, Prabhat. 2008., ‘The Present Crisis and the Way Forward’. The Hindu, 14 November. Quisumbing, Agnes, Ruth Meinzen-Dick and Lucy Bassett. 2008. ‘Helping Women Respond to the Global Food Price Crisis’. Policy brief 7, International Food Policy Research Institute, Washington DC. Rao, C. H. Hanumantha. 2005. Agriculture, Food Security, Poverty and Environment. New Delhi: Oxford University Press. Ravallion, Martin. 2003. The Debate on Globalization, Poverty and Inequality: Why Measurement Matters. Washington, DC: World Bank, Development Research Group, Poverty Team.

268 Y S. Mahendra Dev Society for Elimination of Rural Poverty (SERP). 2001. ‘Livelihood Assessment Report’. Society for Elimination of Rural Poverty, Hyderabad. Viswanath, Tara. 2008. ‘Impact of Rising Food Prices on Poverty: South Asia Region’. Presentation at the South Asia Regional Conference on Managing Food Price Inflation, World Bank, 15–16 November, Dhaka. von Braun, Joachim. 2008a. Food and Financial Crisis: Implications for Agriculture and the Poor. Washington, DC: International Food Policy Research Institute. http://www.ifpri.org/sites/default/files/publications/pr20.pdf (accessed 9 December 2011). ———. 2008b. Rising Food Prices: What Should be done?. Washington, DC: International Food Policy Research Institute. World Bank. 2008. ‘Double Jeopardy: Responding to High food and Fuel prices’. G8 Hokkaido-Toyako Summit, World Bank, 2 July, Toyako.

14 Migration, Human Rights and Development S. Irudaya Rajan Charles Nellari Migration has become a major force in accelerating the progress

of human society. The normal flow of migration can help people broaden their horizons, promote economic and social development and cultural exchange among countries, optimize allocation of human, technical and financial resources, and enhance international cooperation and common development.1 It is an undeniable fact that migration brings forth prosperity to countries of origin as well as countries of destination. Migrants and their families benefit from greater opportunities and better living conditions in countries of destination. If managed and administered in a systematic and sustainable manner, migration provides a ‘comparative advantage’ for stakeholders such as migrants, and countries of origin and destination. Unfortunately, migration does not bring out the expected level of development in all situations. Issues such as trafficking in women, smuggling of children and irregular migration continue to overshadow the advantages of migration. Many experts attribute the reason for not achieving the full potential of migration to the lack of respect for migrant rights. Millions of migrants feel victimized, humiliated, discriminated against and exploited at many stages of migration. It is incomprehensible to compare migrants to ‘commodities’ (UNHCR 2009), leaving the destiny of migrant workers solely to the ‘market forces’ of demand and supply theories. Migrants deserve to be treated with respect, dignity and human rights. There is a strong link between migrant rights and achieving development.

1 See Delegate Statement of the Republic of China at the High Level Dialogue on International Migration and Development, United Nations General Assembly, 2006. http://www.un.org/webcast/migration/pdfs/china-e.pdf (accessed 1 December 2011).

270 Y S. Irudaya Rajan and Charles Nellari

The purpose of this research is to discuss and analyze whether promotion and protection of migrant rights contributes to the development of countries of origin and destination. Since migration is a process with many phases, responsibilities of origin country and destination country shift periodically. The process can be categorized as pre-departure, post-departure/work and return to origin country. Countries of origin and destination have varying and interchanging responsibilities at different levels of migration. This chapter attributes shared responsibilities to both origin countries and destination countries. After all, migration is a universal phenomenon across the world and almost all aspects of activities are influenced by the effects of migration. Individual nations can no longer handle or resolve very complex ramifications emanating from the international aspects of migration. Hence, it is important for all nations to address the issues relating to migrant rights and their relationship with development.

Recognition of Migrant Rights ILO Conventions and International Instruments helped establishing standards for rights and protection to be afforded to migrants during all stages of the labour migration process (UNICEF and UNDP 2007). According to this, countries shall adhere to the standards set by the international documents upon ratifying such Conventions. The rights of migrant workers are derived not only from Conventions specifically addressing migrant worker rights,2 but also from basic rights which apply to all human beings.3 It is also important to reiterate that basic human rights exist whether or not a country has ratified a Convention. The fact of the matter is that the rights do exist even before Conventions recognize them. Also, ratifying a Convention alone may not promote and protect migrant rights unless that country takes required measures to implement the standards. This may lead to a large ‘gap’ between promise and delivery in actual practice (Wickramasekara 2008). While international legal framework provides for a ‘standard’ for the protection of migrant

2 The Migration of Employment Convention, the Migrant Worker Convention, the International Convention on the Protection of the Rights of All Migrant Workers and Members of their Families. 3 The Universal Declaration of Human Rights.

Migration, Human Rights and Development Y 271

workers, many countries around the world lack the political will and action beyond ‘lip service’ to make migration work for development (Wickramasekara 2008). Conventions protecting the rights of children and prohibiting racial discrimination, discrimination against women and torture apply to citizens and to all migrants (UNHCR 2009). Nevertheless, many countries fail to extend equal rights and protection to migrants even in the areas where those rights apply universally. This has to be evaluated under the specific guidelines that migrants in regular status shall receive treatment no less favourable than those of nationals in certain matters pertaining to employment.

Nexus between Migration and Development Emergence of migration and the development of human society are closely linked.4 There is a direct link between Migration and Development. Developments have to be viewed from the perspective of origin country, destination country and migrant. The human element of migration is the most notable factor in the process. The individual choice of migration is ‘. . . a courageous expression of the individual’s will to overcome adversity and to live a better life’ (UN General Assembly 2006: 5). Migration opens the doors of opportunity to millions of people who seek employment and a better standard of life. Countries of origin benefit from remittances from their citizens working abroad. Those remittances help in infrastructure developments, health care, education and basic necessities. Those who migrate to other countries receive training, education, new skills and effectively transmit the acquired knowledge to the countries of origin when they return. Migration brings a range of benefits to destination countries. Most importantly, migrant workers fill the need for workers. Labour shortage was a serious problem in Japan in mid 1980s (Kuwahara 2005). Japanese nationals were not interested in certain categories of jobs (ibid.). Employers had to desperately seek foreign workers in order to meet the demand (ibid.). It is estimated that 94 million migrants are economically active worldwide as migration involves work to support oneself and one’s family (UN 2006). Many nations lack sufficient personnel in specific areas like

4 See http://www.un.org/webcast/migration/pdfs/china-e.pdf (accessed 1 December 2011).

272 Y S. Irudaya Rajan and Charles Nellari

health care and technology. Health care professionals migrating to such countries provide sustainable health care support to communities in the destination countries.5

Development through Better Protection of Migrant Rights Promotion and protection of migrant rights not only provides fair treatment to millions of migrants, but also enhances migrant contributions to origin countries as well as the destination countries. It can also provide for a more sustainable and less troublesome migration process. There are many ways to evaluate and connect development to migrant rights. The migration benefits of the US, Canada, Europe and Australia are achieved by extending better treatment to their migrant population. Most member countries of the OECD have policies aimed at attracting highly-qualified migrants and international students. These countries benefit greatly from migration through attracting the most talented personnel by affording great opportunities in the destination countries. These countries guarantee rule of law and equal protection of law to migrants as well as their own citizens. Migrants get integrated into the destination countries much faster due to the rights and protection afforded to them. There is clear linkage between development and migrant rights as these countries attract a highly talented category of professionals. The economic value of such talents is manifested in their contributions to the communities they get integrated into. The benefits of origin countries proportionately increase in conjunction with the rights afforded to migrants. Remittance to origin countries is an important measuring aspect towards assessing how much an origin country would benefit from increased protection of migrant rights. The countries of origin get sustainable support during economic downward trends through an uninterrupted flow of foreign remittance. Right to affordable health care is an important factor directly connected with the development of migrant workers. Philippines deployed over 119,000 nurses between 1992 and 2006 with an average of 8,000 nurses deployed per year. Health professionals from India and Philippines provide a notable number of health care professionals in the Organization of Economic Cooperation and Development (OECD). India is a significant source of doctors and nurses to Canada, the UK and the US. 5

Migration, Human Rights and Development Y 273

Many countries provide health insurance to all workers, including migrant workers. It protects migrant workers’ health issues and saves millions of productive work hours. Health care protection of migrant workers benefits both countries of origin and destination as saved work hours lead to better utilization of human resources and increase production and remittance.

‘Gaps’ in Migration and Development The current state of affairs of migrant rights is portrayed by the way millions of migrants are treated in different parts of the world. A combination of international instruments projects the standards by which migrants are to be treated and the rights that are to be made available to them. That gives an idea of how the scenario ‘ought to be’ compared to what the scenario ‘is’. There are huge ‘gaps’ between what these two paradigms portray. Policy makers on migration around the world have to analyze the reasons why the phenomenon of migration does not bring positive results as it ‘ought to’. Virtually, all the covenants and international documents addressing migration are ‘non-binding’ in nature. International documents are based on the principle that every country has sovereign authority to regulate and administer its migration policy. Nations hold on to this unrestricted power very closely. Any attempt to dilute this protective cover is treated with utmost resistance. However, advancements in bilateral and multilateral agreements on trade and commerce have changed the rigidity of ‘protectionism’. World Trade Organization (WTO) and General Agreement on Tariffs and Trade (GATT) negotiations encouraged member nations to compromise on their power in controlling the tariffs and flow of goods across their borders. Cross-border movements of ‘natural persons’ are addressed in the GATS. Taking this analogy, sovereign nations can come to a ‘general agreement on the movement of natural persons’ once the ‘comparative advantages’ surpass constraints. As of now, nations are hesitant to compromise on their sovereign authority on migration issues. The reasons may be attributed to the lack of an international mechanism to deal with grievances as provided by the WTO and the GAAT. The current scenario on international migration lacks a global mechanism and vision to implement international standards for the protection of migrants. Lack of a ‘global regime’ to deal with international migration is one of the reasons why the ‘gap’ between two standards of migrant rights — (a) what the right ‘is’, and

274 Y S. Irudaya Rajan and Charles Nellari

(b) what the right ‘ought to be’— exists. As we have seen the correlation between the migrant rights and development, the existence of ‘gaps’ in recognizing and enforcing migrant rights shall influence the developments associated with migration.

State Responsibilities in Protecting Migrant Rights Many nations lack the political will required for protecting and promoting the rights of migrants despite their being signatories of covenants and instruments promoting basic human rights. There shall be no distinction between a citizen and a migrant as these rights are fundamental and protect all human beings regardless of their nationality and legal status (OSCE et al. 2007). There are other conventions6 stipulating States to outlaw all types of discrimination on the grounds of race, colour, descent, national or ethnic origin. All migrant workers and their family members shall be protected from arbitrary arrest and detention. One of the common violations of human rights against migrants is the restriction of ‘freedom of movement’.7 All human beings, including migrants, shall be protected from slavery, forced labour, degrading or inhuman treatment. Human trafficking, discrimination at workplaces and sexual harassments do raise specific challenges to migrants in particular. Unfortunately, most of these rights remain only on ‘paper’ as the ‘gaps’ exist in States’ implementation of recognized standards of treatments of migrants.

Lack of Migrant Rights Impacts on Developments Migration is not a ‘stand alone’ phenomenon. It is closely linked with a wide range of economic and social issues such as employment, education, medical care and public health.8 There are many factors of migrant rights affecting the development of individuals, countries of origin and destination. We shall assess the factors affecting the mobility of women migrants on development. The policies of both countries of origin and destination shape the prospects of women in the migration process (UN General Assembly 2006). See the International Convention on the Protection of the Rights of All Migrant Workers and Members and members of their Families (ICRMW) (1990). 7 See ICRMW, Articles 12 and 13. 8 See ICRMW, Article 14. 6

Migration, Human Rights and Development Y 275

The provision of independent legal status and permission to work to migrant women is crucial for their personal development.9 Migrant women face more restrictions than men in many countries. Some countries restrict women within a certain age group from emigrating due to their vulnerability to sex trafficking.10 Some destination countries restrict women from working in certain fields.11 These exclusive restrictions on migrant women in both countries of origin and destination hamper their chances of economic independence, autonomy and social status. This may also cause social and psychological problems for them. In addition to these adverse impacts on the personal development of these migrant women, the restrictions have also impact the countries of origin in terms of loss of potential remittances. Destination countries miss the opportunity to integrate the migrant women into their workforce. The reality is that the imposition of these restrictions only expose migrant women to more vulnerabilities of exploitation. It is recommended that international migration should be gender-sensitive in order to ensure empowerment of migrant women.12 Visa restrictions imposed by some destination countries do not allow women to accompany their husbands. These restrictions are based primarily on the income earning capacity of migrant workers. Hundreds of thousands of migrant workers live apart from their families for a number of years while they work as migrant workers. This negatively affects the very stability of their marriage and family. This also raises a number of concerns including social and psychological issues for both women and children. Thousands of households in countries of origin switch to single parenting due to severe restrictions imposed in the countries of destination on lowincome migrant workers for sponsoring their spouse and children. Unreasonable restrictions on the issuance of visas to dependants by destination countries have affected the normal and orderly flow of migration, and women and children in particular. Furthermore, such restrictions hamper the personal development of migrants and Ibid. See, for instance, GulfNews.com 2008. 11 See http://www.saudiinfocus.com/en/show_news.asp?nid=842 (accessed 30 November 2010). 12 See ICRMW, Article 17. 9

10

276 Y S. Irudaya Rajan and Charles Nellari

members of their family. Many migrants are forced to reduce their stay in the countries of destination due to being separated from their families, thereby affecting the countries of origin and destination in remittance and manpower. The phenomenon of ‘sponsorship’ and restrictions associated with it affect migrant rights. As it existed in some countries, a migrant employee was bound to work under a specific sponsor during the entire duration of the sponsorship. Departments of naturalization and labour coordinate to implement the ‘sponsorship’ rule. Migrant workers wishing to change the employer, who was also the sponsor, would invariably get a ‘ban’ from entering the country for a specific period. Certain categories of professionals were exempted from the ‘ban’ provided they got approval from the department of labour to change their employment. Nevertheless, the majority of migrant workers do not fall under the exempted category. Hence, their freedom of changing jobs is subject to severe restrictions. Unfortunately, these restrictions violate freedom from forced labour and freedom of movement within the country and the right to leave.13 Employers and recruitment agents ‘confiscate’ the passports of the migrant workers to ensure that they do not leave without their permission. These three restrictions on migrant rights — ‘sponsorship’, ‘ban on re-entry’, and ‘confiscation of passports’ — adversely affect ‘development’ in three dimensions. First, the affected migrant worker is stripped off his/her rights of choice in employment and freedom of movement. This not only hinders the individual growth potential of the migrant worker, but also demeans the individual dignity of the person. Second, the countries of origin find these restrictions on their citizens demeaning and objectionable. This may eventually reduce the natural flow of migrants and thereby reduce the advantages of migration — such as foreign remittance — to the country of origin. Third, these destinations countries may be discredited by the international community for their unreasonable restrictions on migrant workers and become a less preferred destination for future migrants. The restrictions of ‘sponsorship’, ‘ban’ and ‘confiscation’ also lead to the growth of an undocumented and unauthorized workforce due to the severity of restrictions. There are reports that some countries started liberalizing the movement of workers after 13

See ICRMW, Article 16.

Migration, Human Rights and Development Y 277

working with experts from ILO for changing the system (Aziz 2008). Removing the restrictions on changing sponsors and surrendering passports to employers will enhance the rights of migrant workers. This will not only help ‘development’ of migrant workers and countries of origin and destination, but also advance positive aspects of sustainable migration. Legal systems of each country play a very important role in protecting the rights of migrant workers. Once a migrant crosses borders to the county of destination, s/he is subject to the domestic laws of that country. International covenants ratified by member nations may not become effective until and unless the nations incorporate those policies in their domestic laws. Migrant rights promulgated in international covenants such as freedom of association and the right to collective bargaining, the right to equality and non-discrimination, freedom from forced labour and elimination of child labour need to be incorporated and implemented in domestic legal systems to get the benefits of protection to migrant workers. Procedural and systemic constraints such as inherent prejudices against migrants may adversely affect the actual implementation of migrant rights despite the incorporation of laws and policies in the domestic legal systems. Therefore, States have to weed out such systemic restrictions for the effective protection of migrant rights. Procedural fairness and transparency are equally important for the guarantee of migrant rights. Quality and transparency of legal systems can be considered as a measuring tool for evaluating the level of actual migrant rights in a country. Nations with fair and transparent legal systems achieve more development as testimony to the links between protection of migrant rights and development.

Conclusion A correlation between migrant rights and development is evident from many perspectives. One way of measuring development is to assess the growth and advancement of individual development of migrants and their family members. Migrants optimize their potential where their rights are respected and they are treated with decency. Migrant workers get integrated into such a community quickly and act responsibly. Another approach to measure the linkage between migrant rights and development is to compare various situations wherein migrants are afforded with rights and denied

278 Y S. Irudaya Rajan and Charles Nellari

such rights. This may be observed from the laws and policies of countries that affect the rights of migrant workers, and their correlation to development. It is clear that nations affording better migrant rights manage their migration in a more sustainable manner and achieve comparative advantage over those countries that do not uphold migrant rights to the level of expectation. Natural principles of justice of equal protection of law and fairness apply to both migrants as well as the citizens of the countries of destination. At the least, countries shall be obliged to provide ‘no less favourable’ treatment to migrants compared to that of their own nationals. Migrants shall not be deprived of life, liberty and property except according to the procedures established by law. Nations must fully honour human dignity and human rights of migrants. It is imperative to recognize that the potential for development is higher where migrants are provided with the standard of rights and protection established by international covenants and instruments. There exist ‘gaps’ between the ‘current state’ of migrant rights and ‘international standards’ set by international organizations such as the ILO. These ‘gaps’ neutralize or negate the developments associated with migration and migrant rights. Nations have to actually incorporate and implement the international standards of migrant rights in their domestic legal system. It is equally important to maintain fair and transparent procedures while adjudicating disputes affecting migrant workers. There is no single way to resolve issues affecting migrant rights and development. There has to be increased partnerships among nations and agencies to achieve the optimum goal. Humane and orderly migration will maximize social and human development. No nation can afford to ignore its own development by ignoring migrant rights. They are closely connected and interdependent.

References Aziz, Ahmed Abdul. 2008. ‘GCC Countries Liberalizing Movement of Workers: ILO’. Khaleej Times, 23 January. http://www.khaleejtimes. com/DisplayArticleNew.asp?xfile=data/theuae/2008/January/theuae_ January678.xml§ion=theuae&col= (accessed 1 December 2011). GulfNews.com. 2008. ‘Malaysia Rejects Women Travel Restrictions’, 5 May. http://gulfnews.com/news/world/other-world/malaysia-rejects-womentravel-restrictions-1.103697 (accessed 1 November 2011).

Migration, Human Rights and Development Y 279 Kuwahara, Yasuo. 2005. ‘Migrant Workers in the Post-War History of Japan’, Japan Labour Review, 2 (4): 25–47. http://www.jil.go.jp/english/JLR/ documents/2005/JLR08_kuwahara.pdf (accessed 1 December 2011). Organization for Security and Co-operation in Europe (OSCE), International Organization for Migration (IOM) and International Labour Office (ILO). 2007. ‘International Legal Framework for the Protection of Migrant Workers’, in Handbook on Establishing Effective Labour Migration Policies in Countries of Origin and Destination, pp. 27–36. Vienna: Organization for Security and Co-operation in Europe, International Organization for Migration and International Labour Office. UN General Assembly. 2006. International Migration and Development: Report of the Secretary-General. New York: United Nations. http://www.unhcr. org/refworld/pdfid/44ca2d934.pdf (accessed 1 December 2011). United Nations (UN). 2006. Trends in Total Migrant Stock. New York: United Nations, Department of Economic and Social Affairs, Population Division. United Nations High Commissioner for Human Rights (UNHCR). 2009. ‘Migration and Development: A Human Rights Approach’. http://www2. ohchr.org/english/bodies/cmw/docs/HLMigration/MigrationDevelopment HC’spaper.pdf (accessed 1 December 2011). United Nations International Children’s Emergency Fund (UNICEF) and United Nations Development Programme (UNDP). 2007. ‘Migration, Human Rights and Sustainable Human Development’. Global Forum on Migration and Development 2007. http://www.globalmigrationgroup.org/ uploads/gmg-topics/children/3.A_Migration_human_rights_sustainable_ human_development_UNICEF_UNDP.pdf (accessed 1 November 2011). Wickramasekara, Piyasiri. 2008. ‘Globalisation, International Labour Migration and the Rights of Migrant Workers’, Third World Quarterly, 29 (7): 1247–64.

15 Remittances and Financial Participation A Household-level Analysis in Kerala Deepa Iyer The dramatic growth of remittances in recent years has exposed their

ability to reshape migrants’ source countries at the macroeconomic and microeconomic levels. A form of self-help, their sending and usage can be harnessed by households to promote change. Studies have demonstrated the positive correlation of remittances with a variety of outcomes, like poverty reduction, economic growth, educational results, entrepreneurship, access to health care, lower income inequality, and lower infant mortality. While policymakers and financial institutions seem to perceive a positive relationship between remittances and financial development, these claims remain empirically untested at the microeconomic level. Theoretically, remittances might have the potential to improve financial depth (and vice versa) via various channels. To illustrate one such channel, the remittance transaction and subsequent usage of remittance money could serve as an access point for a recipient household, enhancing demand for other formal financial services, which in turn could influence financial service supply. To date, the scant literature on remittances and financial development has mainly addressed the issue at the cross-country level, relating aggregate measures of remittances from balance of payments statistics to country-level measures of financial depth and breadth. These studies are all concerned with exhibiting causality between remittances and financial development at the macroeconomic level. In all, these studies indicate that remittances demonstrate the ability to engender financial development at the cross-country level.1 Yet, studies that address the issue at the household level are rare.

1 Specifically, see Aggarwal et al. (2006); Gupta et al. (2007); Peria et al. (2008); Toxopeus and Lensink (2007).

Remittances and Financial Participation Y 281

Cross-country and aggregate studies cannot accurately examine this relationship as a demonstrated macroeconomic relationship may not be valid at the microeconomic level. It is plausible that countries that receive larger volumes of remittances exhibit higher banking breadth and depth (measured by indicators like deposits or credit to GDP per capita) even if, at the household level, there is little or no association between remittances and banking. For this reason, in order to really assess whether remittances ‘cause’ financial development, microeconomic data must be analyzed. The existing empirical literature regarding the determinants of remittances and financial intermediation has not sufficiently considered this phenomenon at more localized levels within countries. The few studies that address the relationship between remittances and financial development at the household level are all regionally concentrated in Latin America; only one study ventures out of this region, instead considering Moldova. Only three studies to date have addressed the relationship between remittances and banking within a more microeconomic framework. Asli Demirgüç-Kunt et al. (2006) studied the relation between the fraction of households that receive remittances and banking breadth (branches or accounts per capita) or depth (deposits or credit to GDP per capita) at the county level in Mexico (ibid.). They exhibit a strong positive association between remittances and financial depth and breadth. Maria Peria et al. (2008) analyzed household surveys from 11 Latin American countries, finding that the proportion of remittance recipient households with bank accounts is significantly greater than non-recipient households in five countries. To establish causality, they used household data in El Salvador and imitated Demirgüç-Kunt et al.’s analysis (2006) in Mexico, aggregating remittance and banking data at the county level. Here, they observe that remittances to ‘cause’ financial depth. Finally, Fernando Avila and Eva Schlarb (2008) analyzed household data from Moldova, using propensity score matching and instrumental variable analysis to indicate that remittances have a strong positive effect on households’ propensity to have a bank account and save money. These microeconomic studies’ findings are not adequate because except Avila and Schlarb (ibid.) they all aggregate remittance or banking data within a specific locality (for example, at the county level or by observing proportions of remittance recipients and individuals with bank accounts in a locality). Naturally, such aggregation is problematic as it ultimately says little about household behaviour, making for an unconvincing policy paradigm.

282 Y Deepa Iyer

Household-level data can shed light on the relationship between economic behaviour, development outcomes and policy at the level of the individual household (Deaton 2007). Household surveys contain variables related to both potential instruments of policy (in this case, remittance flows) and outcomes (banking), facilitating further assessment of the associations between the two. In all, there exists a clear need for more household-level case studies from a multitude of countries to complement cross-country comparisons. This chapter addresses these gaps by analyzing the relationship between remittances and banking in the Indian context. India is the world’s largest recipient of remittances: it is projected to have received a striking US$27 billion in 2007 (Ratha and Xu 2008: 125). As one of the world’s top emigration countries in absolute terms, the inflow of remittances to India demonstrates a meteoric rise in recent years. Workers’ remittances make up the majority of these flows. Remittance flows to India have far outstripped both FDI and other sources of official development finance (Maimbo et al. 2005). In terms of financial development, banking in India is considered to be somewhat mature with respect to supply and highly unequal in terms of usage, especially between rural and urban areas. Much of the population exhibits low access to and usage of formal financial services. Thus, India is a fascinating venue for the study of migration, remittance, and banking behaviour at the household level. The chapter focuses on the state of Kerala, which accounts for a majority of India’s migrants and a large slice of its remittances (Kurien 2002). Kerala’s main migration flow is to the Gulf region, and the majority of migrants involved in this remittance corridor are unskilled and semi-skilled workers pulled from rural households (ibid.). Using household level data from the third Kerala Migration Survey (KMS 2007), put out by the Centre for Development Studies (CDS) in Thiruvananthapuram, Kerala, this chapter examines the relationship between remittances and banking outcomes when conditioning for a household’s migration experience. More specifically, it studies the differences between remittance recipient and non-recipient households in relation to banking indicators of financial depth (use of financial services), conditioning for whether a household has experienced emigration, out-migration,2 return emigration, return

2

Out-migration refers to internal (inter-state) migration within India.

Remittances and Financial Participation Y 283

out-migration, or a combination of these categories. Engagement with the banking sector is assessed by observing whether households have a bank account or taken on a bank loan or both. The analysis comprises of two parts — (a) a preliminary investigation that simply associates bank-related dependent variables to remittances, conditioning on migration experience, and (b) a probit analysis that regresses banking indicators against remittance recipient status with and without controls for observable household, migrant, and return migrant characteristics. Results refute the findings of the macroeconomically-oriented literature on remittances and financial development, which exhibit remittances to ‘cause’ financial development at aggregate levels when this is not necessarily the case at the microeconomic level. It additionally adds to the small set of microeconomic studies concerning remittances and banking. These studies were imperfect because they tended to aggregate data within a locality, without sufficiently considering household-level differences. In addition, they could not adequately control for observable differences between households that do and do not receive remittances. Finally, none of these studies differentiated between remittances emanating from internal and international migration, which is of importance because households and migrants who engage in each type of migration may have extremely different characteristics. This chapter addressed these gaps by conditioning for households’ migration experience, by controlling for observable characteristics of households, migrants, and return migrants, and by considering internal migration and remittances as a potentially valid influence on financial development. Findings indicate that remittance recipient households are much more likely to have bank accounts. Yet, this is because households that send emigrants (and, to a lesser extent, out-migrants) are much more likely to have bank accounts, whether they receive remittances or not. Conditional on having a migrant (which is necessary to receive remittances in the first place), remittance recipient households are no more likely to have bank accounts than non-recipient households. In addition, controlling for observable household characteristics other than migration status does not eliminate the apparent relationship between receiving remittances and having a bank account. Thus, unobservable differences between households that send migrants and households that do not are what drive whether a household has a bank account. In other words, remittances do not

284 Y Deepa Iyer

‘cause’ a household to have a bank account. Aggregate studies that analyze the relationship between the fraction of people in a locality receiving remittances and the fraction having a bank account, or those that use national measures of remittances and financial depth, will make a mistake. Even if these studies control for average demographics, they will erroneously conclude that remittances cause people to use banks when, in fact, they do not.

Empirical Methodology Marked by high rates of international and internal migration for decades, Kerala is a fitting ‘laboratory’ because of its history of migration and the enormity of its migration and remittance statistics. This chapter analyzes data from a large-scale household survey conducted by the CDS. The KMS is a unique survey — the only one of its kind in India mainly dedicated to the study of migration. It collects data from a representative sample of households across Kerala. Since 1998, it has been conducted every five years as part of a larger series of surveys put out by the CDS. This study uses cross-sectional data from the most recent KMS, officially released in 2008. Conducted between April and September 2007, the KMS was administered to 10,000 households randomly selected via multistage random sampling from all 14 districts, 63 taluks (subdivision of a district), and 999 panchayats (village councils) within Kerala. For more extensive details on the sampling design of this survey and the survey questionnaire, see K. C. Zachariah and Irudaya Rajan (2007: 81–82, 102–14). The survey classifies individuals in each household as emigrants (EMI), return emigrants (REM), out-migrants (OMI), return outmigrants (ROM), or non-migrants. Emigrants are defined as individuals who once belonged to the household then left Kerala and were living outside India at the time of the survey (in other words, the household’s international migrants). Return emigrants are individuals within the household who have returned to the household at the time of the survey after living abroad. Out-migrants are individuals who once belonged to the household then left Kerala and were elsewhere within India at the time of the survey (in other words, the household’s internal migrants). Return out-migrants are those household members who have returned to the household at the time of the survey after living elsewhere within India. For the

Remittances and Financial Participation Y 285

purpose of this analysis, ‘non-migrants’ are individuals within the household who have never migrated, internationally or internally. Essentially, the survey questioned each household’s head (which could be either a non-migrant or a return migrant) on the characteristics of the household — its migrants, and return migrants. I use simple cross-tabulations as well as multivariate regressions to empirically analyze the relationship between remittances and banking at the household level. I assess the relationship, both across all households and then within the group of households that have migrants (which of course is necessary to receive remittances in the first place) or return migrants. Analyzing the relationship conditional on households’ migration experience is a way to control for unobservable differences between households that send migrants and households that do not. It helps determine whether controlling for observables is sufficient to establish a causal relationship between remittances and banking or whether unobservable differences must be considered as well. Three variables were created to measure households’ interaction with the financial system. The first variable — bank account — represents whether or not the household has a bank account. The second variable — bank loan — represents whether or not the household has taken on a loan within the year preceding the survey, and the third variable — any bank — represents whether the household has had any interaction with a bank, either by having a bank account or by taking on a loan in the last year. Remittance-specific independent variables include remittance, a dummy for whether or not a household receives monetary remittances; and monetary amount, a variable for the monetary value of the remittances received by the household. To simplify the types of households of interest, an informal hierarchy of migration categories or specifications of interest were created. Households were grouped into categories based on their migration experience. Table 15.1 indicates the different migration categories and the percentage of the sample that they comprise. These values indicate that the majority of households have never experienced migration, while, unsurprisingly, emigration represents a larger phenomenon than out-migration in Kerala. There is some overlap between migration categories due to the style of classification. These specific groupings of migration experience were selected from a pool of other possibilities for the sake of simplicity. Households were essentially grouped based on how far

286 Y Deepa Iyer Table 15.1:

Categorization of households based on migration experiencea

Dummy variable

Percentage of sample

EMI OMInoEMI

17.68 (1768/10000) 6.46 (646/10000)

EMIorREMnoOMInoROM

22.89 (2289/10000)

OMIorROMnoEMInoREM

13.73 (1373/10000)

REMnoEMInoOMInoROM

7.43 (743/10000)

ROMnoEMInoOMInoREM

7.69 (769/10000)

noEMInoOMInoREMnoROM

60.53 (6053/10000)

Proxy for Emigrant households Out-migrant households without emigrants Households that have a migrant abroad, currently or in the past (households with emigrants or return emigrants, but no out-migrants or return out-migrants) Households which have a migrant in another Indian state, currently or in the past (households with out-migrants or return outmigrants, but no emigrants) Households which once had emigration but no longer do (households with only return emigrants) Households which once had out-migration but no longer do (households with only return out-migrants) Households which have never experienced emigration or out-migration

Source: Special tabulation made from the third Kerala Migration Survey by the author. a The migration categories are disparate; some variables possess overlap while Note: other potential categories are not considered for the sake of simplicity, and thus the percentages in the second row will not add to 100 per cent.

away they had sent their migrants. This was done because the literature indicates that key differences exist between international and internal migration that, in turn, can influence remittance volumes. Because the costs of international migration are higher3 and incomes 3 According to the third KMS, average emigration costs include payments to intermediaries and agents, passport, visa, airfare, emigration clearance, and loss

Remittances and Financial Participation Y 287

abroad are more, it is likely that international remittance volumes are higher (in accordance with ‘family loan’ theory4 and as exhibited clearly by the KMS data5) and are possibly remitted more through formal channels than internal remittances, which can always be carried back home by the migrant (Kapur and Witsoe 2008). Thus, it was decided that emigration ‘matters most’ as it involves going ‘farthest away’. In order to study out-migration, households with at least one out-migrant but no emigrants were utilized. To differentiate between international and internal migration further, households with only emigrants or return emigrants and households with only out-migrants or return out-migrants were considered. To study return migration, households with only return emigrants and households with only return out-migrants were assessed. Lastly, households which had never experienced migration of any sort were considered. By categorizing the population into different migration categories based on households’ migration experiences, it is possible to observe how the association between receiving remittances and financial participation differs across different types of households.

Preliminary Analysis Tables 15.2, 15.3 and 15.4 indicate the relationship between each of the financial participation variables (bank account, bank loan and any bank) with remittance, for all households and conditioning on migration categories. In all three tables, households that do not have

due to fraud; out-migration does not cost nearly as much because out-migrants do not have to front the costs of passport, visa, airfare or emigration clearance, and thus there is less incentive to go through agents who charge high fees. 4 The family loan repayment theory, which is part of the New Economics of Labour Migration Model on migrants’ motivation to remit, suggests that emigrants remit more than out-migrants because they have more of an obligation to ‘repay’ their families for fronting high emigration costs. Ethnographic evidence suggests that family loan theory is the most salient motivation to remit among Keralite Gulf migrants (Kurien 2002; Osella and Osella 2000). 5 Calculated from the KMS 2007, the mean monetary amount of remittance received by emigrant households is ` 45,561.51, while the mean received by outmigrant households is ` 2293.344, indicating that quite a large disparity exists between the amount of money remitted home by emigrants and out-migrants.

288 Y Deepa Iyer Table 15.2: Bank accounts, remittances and migration categories

Migration category (specification variable) All Households EMI= 1 OMInoEMI= 1 EMIorREMnoOMInoROM= 1 OMIorROMnoEMInoREM= 1 REMnoEMInoOMInoROM= 1 ROMnoEMInoOMInoREM= 1 noEMInoOMInoREMnoROM = 1

No remittances

Receives remittances

% with bank account (#/#)

% with bank account (#/#)

73.72 (6188/8394) 88.27 (301/341) 82.90 (514/620) 86.61 (828/956) 81.21 (1089/1341) 86.45 (587/679) 80.40 (607/755) 69.58 (416/5979)

90.60 (1455/1606) 90.68 (1294/1427) 84.62 (22/26) 90.47 (1206/1333) 87.50 (28/32) 93.75 (60/64) 100.00 (14/14) 86.49 (64/74)

Source: Special tabulation made from the third Kerala Migration Survey by the author.

Table 15.3 Bank loans, remittances and migration categories Migration category (specification variable) All Households EMI= 1 OMInoEMI= 1 EMIorREMnoOMInoROM= 1 OMIorROMnoEMInoREM= 1 REMnoEMInoOMInoROM= 1 ROMnoEMInoOMInoREM= 1 noEMInoOMInoREMnoROM= 1

No remittances

Receives remittances

% with bank loan (#/#) % with bank loan (#/#) 19.50 (1637/8394) 19.06 (65/341) 22.26 (138/620) 20.71 (198/956) 21.92 (294/1341) 20.91 (142/679) 21.06 (159/755) 18.90 (1130/5979)

19.36 (311/1606) 18.92 (270/1427) 23.08 (6/26) 18.60 (248/1333) 25.00 (8/24) 21.88 (14/64) 28.57 (4/14) 22.97 (17/74)

Source: Special tabulation made from the third Kerala Migration Survey by the author.

Table 15.4: Any bank interactions, remittances and migration categories No remittances

Receives remittances

Migration category (specification variable)

% with any bank interactions (#/#)

% with any bank interactions (#/#)

All Households EMI= 1 OMInoEMI= 1 EMIorREMnoOMInoROM= 1 OMIorROMnoEMInoREM= 1 REMnoEMInoOMInoROM= 1 ROMnoEMInoOMInoREM= 1 noEMInoOMInoREMnoROM= 1

75.02 (6297/8394) 88.86 (303/341) 83.71 (519/620) 87.66 (838/956) 82.40 (1105/1341) 87.63 (595/679) 81.85 (618/755) 70.97 (4243/5979)

91.03 (1606/1462) 91.03 (1299/1427) 84.62 (22/26) 90.92 (1212/1333) 87.50 (28/32) 95.31 (61/64) 100.00 (14/14) 87.84 (65/74)

Source: Special tabulation made from the third Kerala Migration Survey by the author.

Remittances and Financial Participation Y 289

any current migrants and received remittances (for example, households that only have a return emigrant or a return out-migrant) were not assessed. An asterisk delineates these values in the three tables. This is because without any current migrants their remittances come from people who are not members of their household. Table 15.2 shows the relationship between receiving remittances and having a bank account. When considering all households without conditioning on migration experience, a large disparity between remittance recipient and non-recipient households is observed when it comes to having a bank account. Households that receive remittances exhibit a far higher percentage of having a bank account than non-recipient households. However, this difference almost completely disappears when conditioning for various migration categories. When comparing households that have experienced migration in the past (households with only return emigrants, or households with only return out-migrants) to those that have never experienced migration, it appears that the former exhibit much higher percentages of having a bank account (86.45 per cent for households with only return emigrants and 80.40 per cent for households with only out-migrants) than the latter (69.58 per cent). Table 15.3 shows the relationship between bank loans and remittances. When considering all surveyed households, there appear to be few differences among the percentages of households that have taken on bank loans between remittance recipient and non-recipient households. This effect (of an apparent lack of difference) is consistent when conditioning with the range of migration categories. Table 15.4 describes the relationship between bank interactions and remittance. When observing all households, there is quite a large difference in the percentage of households having any bank interaction between non-recipient and remittance recipient households. Like bank accounts, however, this disparity diminishes considerably in size when conditioning for migration categories. Once again, households that have experienced migration in the past seem to exhibit much larger percentages of having any bank interaction (87.63 per cent for households with only return emigrants and 81.85 per cent for households with only return out-migrants) than those that have never experienced any migration (70.97 per cent). Three key findings emerge from the preliminary analysis. First, when considering the relationship between having a bank account or having any bank interaction and receiving remittances for all

290 Y Deepa Iyer

households, the large positive difference that is observed between non-recipient and remittance recipient households in the percentage of households having a bank account all but disappears when conditioning for various migration categories. Second, there appears to be little connection between receiving remittances and receiving a bank loan when considering all households; this effect is consistent even when conditioning for migration categories. Third, when comparing households that have experienced migration in the past to those that have never experienced migration before, the former exhibit much higher percentages of having a bank account than the latter. The relative disappearance of the large disparity observed between recipient and non-recipient households for bank account and any bank when conditioning on various migration categories suggests that the relationship between financial participation and receiving remittances is more complex than accounted for by many of the cross-country and aggregate analyses to date. While remittance recipient households are certainly far more likely to have a bank account or any interaction with a bank, conditioning on migration experience indicates that non-recipient households are just as likely to have a bank account or any bank interaction as remittance recipient households. This is because households that send migrants are more likely to have bank accounts, whether they receive remittances or not. Studies to date have not generally considered potential differences between different types of households based on migration status. It may be that the actual sending of a migrant (something which is unobservable) — in other words, the indirect experiencing of migration by a household — somehow inclines households towards engaging more with the banking sector. Cross-country analyses like Reena Aggarwal et al. (2006) and Sanjeev Gupta et al. (2007) use aggregate remittance and banking data at the national level, which is naturally problematic because remittances do not necessarily ‘cause’ financial development at the microeconomic level. This suggests that studies that consider the relationship between banking and remittances within a macroeconomic or aggregate framework need to make allowances for differences in this relationship at a microeconomic level. The second key finding is uniform across all types of households: there appears to be little connection between remittances and credit (i.e., taking on a bank loan). A possible reason for the virtually non-existent relationship between remittances and bank loans

Remittances and Financial Participation Y 291

is that receiving remittances adds to a household’s income, having a dampening effect on demand for credit in general. Interestingly, Thorsten Beck et al.’s (2008) household-level analysis also observed that receiving remittances appears to share an ambiguous relationship with bank credit. The third finding indicates that households with return emigrants or return out-migrants that do not have anybody currently abroad or in another state in India appear to be more likely to have bank accounts or any bank interaction than those that have never experienced migration before. This is interesting because it points to a potential relationship between return migration and the propensity of having a bank account or any bank interaction. One possibility is that return migrants might ‘bring home’ certain observable and unobservable characteristics that make their households more likely to have a bank account or any bank interaction.

Probit Analysis In addition to the preliminary analysis, regression analysis is necessary to determine whether controlling for households’ observable characteristics eliminates the apparent relationship between remittances and having a bank account or having any bank interaction. All equations were estimated using a probit model, as the ‘bank’ dependent variables used can each take on only two values, and the probit model is traditionally used to fit data relating to a dichotomous outcome. As previously stated, the key dependent variables of interest are bank account, bank loan and any bank, dummy variables that measure whether a household has a bank account and taken on a loan from a bank, or done either of these. The primary independent variable of interest is remittance, a dummy for whether or not a household receives monetary remittances; monetary amount is also of interest, as it measures the monetary value of remittances received by a household. Controls in the regressions included household controls, emigrant controls, return emigrant controls, out-migrant controls, and return out-migrant controls. Household controls attempted to reflect the wealth and socio-economic background of the household in its natural state, before emigration or remittances could potentially influence the household. Household controls can broadly be categorized into:

292 Y Deepa Iyer

Demographic characteristics of the household: household size, the male count and female count of the household, the proportion of dependants in the household, age of the household head, a dummy variable for the household head’s sex, a dummy variable for whether the household was located in an urban or rural area. Education characteristics of the household: education of the head of household, education of the most educated person in the household. Religious characteristics of the household: dummy variables for whether the household is Hindu, Muslim, or Christian. Employment characteristics of the household: the proportion of household adults6 employed in government or semi-government jobs, the proportion of household adults employed in the private sector, the proportion of household adults who are self employed, the proportion of household adults employed in manual labour, and the proportion of household adults who are unemployed. The income proxies of household quality and type of cooking fuel were not considered as household controls because they do not accurately reflect the household’s socio-economic background before migration or remittances; remittances can and often are used to refurbish houses or improve living conditions (Zachariah and Rajan 2007). Emigrant, out-migrant, return emigrant, and return out-migrant controls were intended to reflect the characteristics of a household’s migrants or return migrants. Because households could have multiple migrants or return migrants, controls were specified for the most educated migrant (or return migrant) and the migrant (or return migrant) who migrated or returned the earliest (i.e., who had been away from the household for the longest period of time or had since returned to the household). This provided a meaningful method of accounting for differences between households’ multiple migrants or return migrants. If overlaps existed — for example, if two migrants were both, concurrently, the most educated migrant — then various other specifications were used to remove the overlap (in this case, first time since migration, then sex, then age, and lastly, the 6

An adult is defined as an individual who is 18 years of age or older.

Remittances and Financial Participation Y 293

relationship with the head of household were used as ‘tiebreakers’). If the case was that two or more migrants could be the migrant who migrated earliest, then education, sex, age, and relationship with the head of the household were, in that order, used as needed to diminish the overlap. The controls can be broadly categorized as follows: Characteristics of the most educated (return) emigrant (or outmigrant): time since migration, dummy variables for destination country, age at the time of the survey, a dummy variable for sex, education level, dummy variables for the migrant’s relation to the head of the household. Characteristics of the (return) emigrant (or out-migrant) who migrated or returned first: time since migration, dummy variables for destination country, current age, a dummy variable for sex, education level, the migrant’s relation to the head of the household. (Return) emigrant (or out-migrant) employment characteristics: the proportion of (return) emigrant (or out-migrant) adults employed in government or semi-government jobs, the proportion of (return) emigrant (or out-migrant) adults employed in the private sector, the proportion of (return) emigrant (or out-migrant) adults who are self-employed, the proportion of (return) emigrant (or out-migrant) adults employed in manual labour, and the proportion of (return) emigrant (or out-migrant) adults who are unemployed. The relationship between financial participation dependent variables and receiving remittances, with or without conditioning on specific migration categories, was assessed by using variants of Equation (1): yi = B0 + B1 remittancei + B2 monetaryamounti + B3 Xi + B4 Zi + ei (1) Within this equation, yi signifies one of the three financial participation variables (bank account, bank loan, or any bank). Xi constitutes a vector of household controls, while Zi represents a vector containing controls for emigrants, out-migrants, return outmigrants, or return emigrants, depending on the regression being run. All regressions used robust standard errors and clustered standard errors at the panchayat level.

294 Y Deepa Iyer

First, equation (1) was run for all three dependent variables with all households in the survey without any controls. Then, equation (1) was run for all dependent variables controlling for observable household characteristics using Xi. This was done in order to observe whether potential differences in banking and remittance behaviour could be accounted for by a household’s observable characteristics. Next, equation (1) was run by conditioning on various migration categories. First, equation (1) was run, conditioning on EMI= 1 (in other words, that a household had at least one emigrant) without any controls, with only household controls (Xi), and with both household (Xi) and emigrant controls (Zi). Equation (1) was then run specifying that OMInoEMI=1 (that a household had out-migrants but no emigrants), without any controls, with only household controls (Xi), and with both household and out-migrant controls (Zi). Next, equation (1) was run specifying that EMIorREMnoOMInoROM=1 (that a household only had emigrants or return emigrants), without any controls, with only household controls, with household and emigrant controls, and with household, emigrant and return emigrant controls. Finally, equation (1) was run specifying that OMIorROMnoEMInoREM=1 (that a household had only outmigrants or return out-migrants), controlling for only household characteristics, household and out-migrant characteristics, and household, out-migrant and return out-migrant characteristics. The migration categories of REMnoEMInoOMInoROM, ROM noEMInoOMInoREM and noEMInoOMInoREMnoROM were not used to condition any variants of equations (1). This is mainly because households with only return emigrants, only return outmigrants, or no migrants or return migrants cannot, technically, be receiving remittances from a prior member of their household. Table 15.5 summarizes the key marginal effects of the probit regressions in relation to the key independent variable of interest — remittance. Firstly, it is apparent that the marginal effects for remittance in relation to bank account and any bank are positive and highly significant. Additionally, controlling for a household’s observable characteristics does not make a large difference as the effect continues to be highly significant, although the magnitude of the effect diminishes slightly. Controlling for observable household characteristics has little impact, failing to eliminate the observed relationship between receiving remittances and having a bank account (bank account) or

Remittances and Financial Participation Y 295 Table 15.5:

Remittance marginal effects for all households

All households

No controls

With controls

Marginal effect

Bank account (df/dx)

Bank account (df/dx)

Remittance

0.148∗∗∗ (0.0199) Bank loan –0.00504 (0.0150) Any bank 0.141∗∗∗ (0.0188)

0.130∗∗∗ (0.0177) Bank loan 0.0168 (0.0161) Any bank 0.125∗∗∗ (0.0168)

Remittance

Remittance

Source: As estimated by the author. Note: (Heteroskedasticity-robust standard errors clustered by panchayat in parentheses.) ∗∗∗ p