Basic Income on the Agenda: Policy Objectives and Political Chances 9789048505029

Table of contents :
Table of Contents
Preface
Acknowledgements
How Attractive is a Basic Income for European Welfare States?
Part One. Policy Objectives
In Search of the Double-Edged Sword
Basic Income and its Cognates: Partial Basic Income versus Earned Income Tax Credit and Reductions of Social Security Contributions as Alternative Ways of Addressing the ‘New Social Question’
Activation and the Burden of Working: On Instrument Choice by a Responsibility-Sensitive Egalitarian Government
Arguing for a Negative Income Tax in Germany
Hush Money or Emancipation Fee? A Gender Analysis of Basic Income
Prospects for Basic Income in an Age of Inactivity?
Basic Income and Social Europe
Basic Income at the Heart of Social Europe? Reply to Fritz Scharpf
European Basic Income or the Race to the Bottom: Why Politicians Might Come to Think the Unthinkable
Bibliography
Part Two Political Chances
Clues and Leads in the Debate on Basic Income in the Netherlands
The History of an Idea: Why did Basic Income Thrill the Finns, but not the Swedes?
From Concept to Green Paper: Putting Basic Income on the Agenda in Ireland
Short Cuts and Wrong Tracks on the Long March to Basic Income: Debating Social Policy Reform in Germany
Ups and Downs of Basic Income in Denmark
What Reforms are Needed for the Minimum Insertion Income (RMI) in France?
The VIVANT Experiment in Belgium
Notes on the Contributors
Index

Citation preview

Basic Income on the Agenda

Basic Income on the Agenda Policy Objectives and Political Chances

Edited by Robert van der Veen and Loek Groot

Amsterdam University Press

Cover design: Crasborn Grafisch Ontwerpers BNO,Valkenburg aan de Geul Lay-out: Brassica / Wouter Kool, Leiden NUGI: 641/661 ISBN 90 5356 461 6 ã Amsterdam University Press,Amsterdam, 2000 All rights reserved. Without limiting the rights under copyright reserved above, no part of this book may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording, or otherwise), without the written permission of both the copyright owner and the author of this book.

Table of Contents Preface Osmo Soininvaara

7

Acknowledgements

11

How Attractive is a Basic Income for European Welfare States? Loek Groot and Robert van der Veen

13

Part One Policy Objectives In Search of the Double-Edged Sword Paul de Beer Basic Income and its Cognates: Partial Basic Income versus Earned Income Tax Credit and Reductions of Social Security Contributions as Alternative Ways of Addressing the ‘New Social Question’ Philippe Van Parijs, Laurence Jacquet and Claudio Caesar Salinas Activation and the Burden of Working: On Instrument Choice by a Responsibility-Sensitive Egalitarian Government Frank Vandenbroucke and Tom Van Puyenbroeck

41

53

85

Arguing for a Negative Income Tax in Germany Joachim Mitschke

107

Hush Money or Emancipation Fee? A Gender Analysis of Basic Income Ingrid Robeyns

121

Prospects for Basic Income in an Age of Inactivity? Anton Hemerijck

137

Basic Income and Social Europe Fritz W. Scharpf

155

Basic Income at the Heart of Social Europe? Reply to Fritz Scharpf Philippe Van Parijs

161

European Basic Income or the Race to the Bottom:Why Politicians Might Come to Think the Unthinkable Steve Quilley

170

Bibliography

186 5

Part Two Political Chances Clues and Leads in the Debate on Basic Income in the Netherlands Loek Groot and Robert van der Veen

197

The History of an Idea:Why Did Basic Income Thrill the Finns, but not the Swedes? Jan-Otto Andersson

224

From Concept to Green Paper: Putting Basic Income on the Agenda in Ireland Sean Healy and Brigid Reynolds

238

Short Cuts and Wrong Tracks on the Long March to Basic Income: Debating Social Policy Reform in Germany Stephan Lessenich

247

Ups and Downs of Basic Income in Denmark Erik Christensen and Jørn Loftager What Reforms are Needed for the Minimum Insertion Income (RMI) in France? Chantal Euzéby

257

268

The VIVANT Experiment in Belgium Yannick Vanderborght

276

Notes on the Contributors

285

Index

287

6

Preface Osmo Soininvaara, Minister of Health and Social Services, Finland

Throughout Europe, the idea of an unconditional basic income has been long discussed by academics, policymakers, and interest groups such as the unemployed, religious organizations, environmentalists and trade unions. Contemporary issues to which these debates relate today include unemployment, poverty, marginalization, changing patterns of family life, labour market flexibility, and institutional questions concerning the future of social policy under European integration. In Arguing for Basic Income (1992), edited by Philippe Van Parijs, the ethical foundations for granting everyone a fully unconditional subsistence income were explored in depth. The present volume follows this up by examining basic income in relation to concrete objectives of social and labour market policy (Part I), and looking at its political chances across European countries (Part II).This may be taken to reflect the fact that the international debate on basic income has become politically mature in the intervening years. What, then, can reasonably be thought of the title of the book: Basic income on the agenda? An ambitious reading might suggest that basic income, or the closely related negative income tax, is now firmly on the legislative agenda, ready to replace the meanstested and work-related forms of guaranteed income in European welfare states.As the contributions of Part II will show, basic income is not on the agenda in that sense, at least not in any European country, at the beginning of this century. But while the fortunes of the proposal depend on contingent circumstances and fluctuations in political mood, the general outline of the debate is more or less similar everywhere. It is a debate which is not likely to subside. Consequently, basic income is indeed on the agenda of welfare state politics. The challenge posed by basic income rests in its claim that unconditional guarantees have now become indispensable elements of social policy.They are said to cope more efficiently with the uncertainties of increasingly flexible labour markets, and avoid the administrative and incentive pitfalls of means-testing.They are held to provide means of job transition and retooling of labour capacity, as well as supporting unpaid activities in the domestic sphere and the associations of civil society, thereby also contributing to a more equitable redescription of gender roles. However, as proposals to institute unconditional income of one kind or another become more realistic, they also tend to mobilize opposition. This is because the unconditional nature of basic income poses a threat to the primacy of paid work, as the central source of legitimate security. Thus, basic income forces us to rethink the notion of social obligation, in a way that may seem, at times, far removed from the reality of mainstream politics.Yet, the problems which it addresses are undeniably part of that same reality. The contributions collected in Part I can be read as critical examinations of the above claims, following the discussions at the Seventh BIEN Amsterdam Congress on the theme of ‘full employment without poverty’. The main problem here is whether one can envisage alternatives to the social policies and labour market arrangements of existing welfare states that would serve the same goals as basic income does, yet in ways 7

Basic Income on the Agenda that retain the primacy of paid work. In this debate on policy instruments, the editors have resisted the inclination to select contributors according to their degree of bias in favour of the basic income solution.This gives the reader the opportunity to judge how relevant basic income is to the future of social security and labour participation, in the emerging space of ‘Social Europe’. Basic income has become topical again for very practical reasons, which, however, are slightly different from the reasons for which basic income was pursued in the 1980s. Economic development has forced the present European welfare systems to face two difficult challenges.The unskilled labour force does not find its place in the labour market, which leads to a trend of growing structural unemployment. Simultaneously, the obligation to accept a job offer as a condition of receiving unemployment benefits does not function any more. Employers do not hire people just to get their time; they want to have their sincere effort as well. Thus, refusing a job offer does not really lead to losing the benefit. Unemployment benefits have become a twisted picture of basic income that gives bad incentives. It is clearly a handicap to the European countries that the unskilled part of the labour force does not find employment. In the USA, the Earned Income Tax Credit (EITC) has proved to have positive effects. The expansion of the EITC has improved the rather bad position of the working poor and had a major positive impact on the employment situation. In the EITC system the benefit the state pays to those whose wages are low grows as the earned income rises. In the European systems, however, social income transfers diminish – and it is likely that the benefit of earning a part of your income is cut out completely. EITC is not suitable for European circumstances since the European basic rights legislation guarantees everyone the necessary income regardless of participation in the working life. It should, nevertheless, be easy to find the perfect compromise between the European way (the support diminishes as earned income rises) and the American way (the support rises as earned income rises): why not make the amount of support independent of earned income? Each country has its own history of social politics. Acting and thinking have been adjusted to the existing system which makes radical and fundamental reforms unlikely. But the need to rationalize incentives in flexible labour markets, the impossibility of controlling the readiness of the unemployed to accept work, the difficulties in finding employment for the unskilled and the requirement of guaranteed minimum income based on basic rights, lead social politics, step by step, toward a situation much resembling a basic income system that allows, with cause, many means-tested personal benefits. These new systems will vary in names and details but they all converge towards basic income. The biggest and mentally most difficult step to be taken is to combine the income transfer systems with taxation, even though taking this step would make it possible to create rational systems. Basic income thinking was originally inspired by the ideal of freedom. It is contradictory that basic income is now favoured for almost opposite reasons: creating rational incentives at the lower end of the income scale and improving the employment of the unskilled labour force – basic income can be thought of as a ‘marriage of justice and efficiency’. Basic income does, however, still include a dream of a freer society. Making this dream come true is worth aspiring to as systems of social politics inevitably converge towards basic income.When it comes to the ideal of freedom, the smallest details can make a difference. Although it may seem most unlikely that any European country would abandon its 8

Preface old traditions and reform its social politics so radically as to adopt a basic income system, this option is not completely impossible either.The fact that social policies vary much from country to country creates substantial difficulties for the EU. If, sometime in the future, the EU or some of the Member States were to unify their architecture of social security, basic income would indeed be something to build on.

9

Acknowledgements

We wish to thank the sponsors of the Seventh International Congress of the Basis Income European Network (Amsterdam, September 1998) for financial support in producing this volume. Earlier versions of all chapters (save three) have been presented in plenary sessions or workshops of the Congress. The chapter by Frank Vandenbroucke and Tom Van Puyenbroeck (in Part One) as well as the chapters by Chantal Euzéby and Yannick Vanderborght (in Part Two) were included in order to fit the general design and purposes of the book. We warmly thank our co-authors for their co-operation and patience in the different stages of – often extensive – revision.We are also endebted to Osmo Soininvaara for writing the Preface to this collective effort. In addition, we are grateful for the advice and assistance of Philippe Van Parijs in editorial matters. Finally, special thanks are due to the Board of the Vereniging Basisinkomen, in particular Saar Boerlage and Emiel Schaefer, for their help and encouragement.

The Editors

11

How Attractive is a Basic Income for European Welfare States? Loek Groot and Robert van der Veen

A basic income is an income granted unconditionally to all on an individual basis, without a means test or work requirement. It is a type of minimum income guarantee that differs from those that now exist in various European countries. Considered in its pure form, a basic income is paid (1) to individuals rather than households; (2) irrespective of wealth or any income from other sources; and (3) without requiring the performance of any work or the willingness to accept a job if offered. Its relevance to the problems of contemporary welfare states has long been recognized. From the 1970s onwards, forms of basic income have been regularly proposed for adapting the post-war institutions of social transfer to the conditions of a modern economy, in which lifetime security through earned income is no longer the norm. The guiding claim was that an unconditional floor of guaranteed income is justified as the fairest and the most efficient way of reconstructing basic security in the welfare state. Inevitably, the fairness part of this claim led to a principled debate concerning the unconditional features of basic income, in particular the absence of the requirement of willingness to work.The reason for this is obvious, for any moral justification of unconditional basic income has to deal with the objection that it is unfair to hand out transfers to ablebodied persons who are merely unwilling to earn a living. This theme has been explored in depth in Arguing for Basic Income (1992). However, basic income is being increasingly discussed in a more pragmatic context. This discussion is mostly concerned with the efficiency part of the above claim. Here, basic income is proposed as the most viable way of reconciling two of the welfare state’s central objectives – the alleviation of poverty and full employment – while at the same time allowing scope for making the labour market function in a more flexible and dynamic way. One of the reasons why basic income has attracted interest is that it seeks to reinterpret the compatibility of these two central objectives, reflecting the fact that social policy and economic policy can no longer be conceived separately. Thus, in recent years, basic income has appeared on the agenda of institutional reform as a proposal which combines adequate social protection with the demands of flexible labour markets, and may lead to a more equitable distribution of income, paid work, care work and free time among men and women.As such, it has to be pitted against the claims of (partly or fully) competing policy instruments in the welfare state: collective working time reduction, subsidized employment, paid parental care, early retirement, educational and sabbatical leave provisions, the minimum wage, and schemes of ‘workfare’. This book is divided into two distinct but interrelated parts, both of which reflect the main topics of the 1998 Congress of the Basic Income European Network.1 The first part, ‘Policy Objectives’, analyzes the potential of basic income vis-à-vis other policy schemes to achieve the above-mentioned goals of welfare state reform. The second part, entitled ‘Political Chances’, reviews the state of the debate on basic income in var13

Basic Income on the Agenda ious European countries.Although the book can be seen as following up on Arguing for Basic Income, it has a different point of departure. Arguing for Basic Income concentrated on the ethical controversies around unconditional grants, which, as noted above, have been part of the general debate from the start. It attempted ‘... to spell out a plausible conception of the just or good society which could provide firm foundations for the legitimacy of an unconditional income’ (ibid.: 8) and ‘… to derive basic income from an explicit formulation of the ideal of a free, equal or good society’ (ibid.: 29). These controversies, which are concerned with rethinking the regulative principles of solidarity and justice in the welfare state, have not yet come to rest.2 As customary in applied political philosophy, the arguments are exchanged at a high level of abstraction. Most recently, the main question has been whether granting people a substantial, or even the maximally sustainable, basic income is an appropriate instrument of liberal-egalitarian justice. The answers seem to depend on how the egalitarian aim of compensating for unequal endowments can be combined with the liberal aim of holding people responsible for their personal choices. At the level of detailed policy implications in societies with developed welfare state institutions, competing interpretations of liberal-egalitarian justice then generate different positions regarding the desirability of either worksensitive or unconditional compensation to the worst-endowed members of society: the working poor, and those who are more or less chronically excluded from the labour market, and hence depend on transfer income. In this debate, the proposal of introducing a basic income in the welfare state typically serves as a testing ground for exploring the relative attractiveness of abstract normative positions entertained on the political Left.3 While these theory-driven controversies still remain largely unsettled, the fact that they are being fought out in the international arena in the area of welfare-state policies to some extent explains why basic income has become a proposal of some interest in the problem-driven debate on social security reform. It is this debate which provides the perspective of the present book. Rather than trying to link the proposal of basic income to philosophical principles of justice, here, we want to focus on how it fares relative to other policy instruments which claim to promote politically salient objectives of welfare reform.There is widespread consensus among policymakers and other professionals that the welfare system has to be adjusted, but also disagreement about the direction of reform. Is it time to pare down the benefits of the welfare system, by making it ‘meaner and leaner’, allowing it to survive global market competition and more intense intra-European competition for advantageous tax regimes? Or do we instead have to find new ways to maintain the basic protective features of the European welfare state in the face of these challenges? To what extent is the debate on the form that welfare arrangements should take influenced by the problems of the inactivity traps besetting means-tested welfare arrangements, the persistence of (long-term) unemployment among the low-skilled, and the difficulty to get rid of poverty despite the affluence of European countries?4 Last but not least, what requirements of re-design are being imposed by flexible work arrangements, the decline of breadwinner families along with increased female participation, and the greater difficulties of combining work and family responsibilities? In addressing such questions, arguments for and against basic income figure at a far lower level of abstraction.They are more influenced by the concrete social problems in different welfare states than by general normative positions.Yet, as the contribution of 14

How Attractive is a Basic Income for European Welfare States? Frank Vandenbroucke and Tom Van Puyenbroeck in this volume will show, it is important not to lose sight of the link between these two levels of debate.Though most of the contributors of basic income in this book are to some extent moved by egalitarian or solidaristic justice, such normative commitments often appear obliquely, against the background of what is often called the ‘new social question’.This is the real threat of a permanent divide between on the one hand job holders with a strong labour market attachment (secure full-time employment with full entitlements to social insurance) and marginal groups with intermittent labour market careers (insecure access to benefits of social security, and old-age pensions) on the other.5 In Arguing for Basic Income, the importance of basic income for addressing the new social question was described by Van Parijs (1992: 6-7) as follows: … it can no longer be assumed that an overwhelming majority of households can cover their basic needs thanks to the wages they owe to the job one of their members currently holds or to the benefits they owe to the job one of their members used to hold. Under this assumption, central to the conception of the modern welfare state, the safety net of social assistance could be confined to a marginal – and ideally, shrinking – set of cases. For various reasons, this is now very far – and even further – from being the case. Throughout Europe, an increasing number of households have had to rely on social assistance and have become caught in the net it provides.The joint impact of technical change and the internalization of markets is making it increasingly difficult for the economies of advanced capitalist countries to generate a sufficient number of jobs that can be profitable while providing those who hold them with a living wage.The outcome of this process is, increasingly, a ‘dual economy’, a ‘two-thirds society’, in which the most significant divide, as far as material welfare is concerned, is no longer the one that separates capitalists from workers, but the one that separates those who hold proper jobs from the rest of the population.There is no easy way of fighting this tendency. But the replacement of the safety net, in which the weakest and the unlucky get trapped, by a firm unconditional floor, on which they can securely stand – in other words, the replacement of a conditional minimum income scheme by a genuine basic income – is increasingly viewed as an indispensable ingredient in any such strategy.

The new social question thus basically turns around the failure of the existing labour market and social security institutions to ensure access to employment, while at the same time securely protecting people who are in or out of work from falling into poverty. In the same vein, Goodin (2000) made the case that the four pillars – the state, the market, the family and the community – on which social security rests are all crumbling. In the increasingly non-standardized world towards which we are moving, Goodin prefers some form of a basic-income type scheme above means tested and fine-tuned conditional social benefits: ‘Any standard categories or conditions fit any given individual’s actual experiences increasingly poorly, both occupationally (with the flexibilization of the labour market) and personally (with the flexibilization of the marriage market). So, if we want our social policy to cater to people’s real needs and actual circumstances, we will have to abandon categorical and conditional modes of social-policy making’ (ibid.: 147). In capsule form, the main policy objective to be kept in view in assessing the indispensability claim Van Parijs made for basic income in the early 1990s, is ‘full employment without poverty’.6 However, compared to the early 15

Basic Income on the Agenda 1990s, the conditions for counteracting the ‘dual economy’ have become less favourable. On the political Centre-Left at least, this has considerably influenced the diagnosis of the new social question. It is now conceded that flexibility in the organization of labour, footloose factors of production, and processes of tax competition, shifts in the employment volume between exposed and sheltered sectors in the economy, as well as structural increases in the demand for health care and pension provisions of rapidly ageing populations, have worsened the chances of attaining ‘full employment without poverty’.These developments have made it harder for national governments to pursue the goals of employment security and redistributive welfare that were formerly held to be viable, given sufficient political will.Yet at the same time, the Centre-Left is committed to showing that neither the ‘blind forces of globalization’, nor the adverse demographic shifts taking place in advanced welfare states, have robbed national governments of the policy space for pursuing broadly egalitarian strategies. In new and suitably adjusted ways, such strategies are still held to be within reach, depending on the development of the right kind of policies.7 At present there is a great variety of relatively novel policy instruments, amongst them forms of basic income, which may conceivably be put to use in counteracting the emergence of a dual economy, under the more restrictive conditions outlined above. But it should be noted that at present also, the proponents of basic income may have more of a rough ride to stay in the race, for two connected reasons. First, since basic income has to prove itself in an economically and politically more constrained environment, the tax cost of guaranteeing the ‘firm unconditional floor’ – if pitched at the level of subsistence needs – is now even more likely to be perceived as prohibitive than it was a decade ago. As a consequence, only partial and piecemeal movements toward unconditional entitlements of benefit are the ones that are going to figure seriously in the comparative exercise of optimal policy instruments for solving the new social question. Secondly, unlike before the early 1990s, basic income can no longer be presented as the only alternative to the status quo of work- and means-tested conditional benefits, with its large unemployment and poverty traps. For in the policy debate on the new social question, competing instruments have entered the field, such as Earned Income Tax Credit, and various kinds of wage subsidies to employees. These instruments, it is now claimed, are able to remove inactivity traps just as efficiently as a basic income would be able to, and moreover, in ways that avoid the controversial feature of basic income: its loosening of the links between transfer income and the obligation to perform paid work. These changes in the terms of the debate imply that forms of partial basic income and forms of work-oriented benefit and labour-market policies slide into a broad continuum of alternatives, each of which can claim to tackle the new social question in terms of its effects on promoting access to employment at the low end of the labour market, preventing poverty, and fighting social exclusion. The advantages and drawbacks of each of these alternatives thus have to be sorted out by a very careful comparative analysis, and in the course of such analysis, as the contributions of Paul de Beer and of Philippe Van Parijs, Laurence Jacquet and Claudio Salinas clearly show, it is difficult to claim a knock-down superiority for any particular policy alternative.Thus it is perhaps fair to say that basic income, after having won a certain kind of respectability as a normatively unorthodox way of addressing the deficiencies of existing welfare state arrangements, is now taken seriously enough to be put to the test under rather harsher 16

How Attractive is a Basic Income for European Welfare States? conditions than its proponents may have anticipated. One object of this volume is to see if basic income can survive this test.

Four Ideal Types of Basic Social Security Since the rise of unemployment in the mid-1970s, several measures have been taken to curb steadily increasing social expenditures (both absolutely and as a share of GDP) and to counteract the rising trend of unemployment which appeared after each successive recession.The ruling diagnosis was, and still is to a large extent, that the vested security entitlements of European welfare states are too permissive, and that labour markets are ridden with structural rigidities.8 Accordingly, the menu of social-economic policy recommendations has varied from tighter eligibility conditions, shorter durations and lower levels of social benefits, reduction of the level of minimum wages, and the relaxation of firing and dismissal procedures.Also, and in part independently of whether or not the overall unemployment figures in the European Union have improved (as they have in Denmark, Britain, Ireland and the Netherlands), the persistent problem of long-term unemployment at the lower end of the labour market has given rise to the introduction of wage subsidies, and a large variety of job, workfare and learnfare programs. Especially these last types of measures are held to mark the emergence of an ‘active’ welfare state, which tries to reduce ‘passive’ dependency on conditional social assistance benefits. In general, the stringency of the work-test is being raised, while at the same time efforts are undertaken to provide access to the labour market, either directly or indirectly. The idea of the active welfare state attacks the perceived passivity of both recipients and government agencies. On the one hand, active welfare state policies are committed to destroy the ‘legitimate expectation’ of claimants to holding a social benefit, subject only to the ritual of turning up at a job centre and noting that no suitable work is forthcoming. On the other hand, the bureaucracy of welfare administration must be forced into taking an active interest in getting claimants prepared for permanent attachment to the labour market. The administration of social security legislation, as well as the legislation itself, must therefore be changed, by providing job-oriented offers that claimants are no longer in a position to refuse. Such offers involve mandatory participation, in (re)training, public service employment, or subsidized (private or public) employment programmes.The active welfare state approach thus tries to strike a new balance between social security rights and the correlative duties of both claimants and agencies. It claims that this new balance is needed to maintain redistributive solidarity in the hard times of mobile capital and high-quality labour assets. It should be made clear that anyone who is ‘on welfare’ is also being redirected and resocialized to a norm of a self-sufficient working life, and that a growing proportion of the taxes to be paid for implementing solidarity is thus being spent on programs achieving labour participation, rather than on passive income support. The active welfare state approach may culminate in a system of workfare, if the government steps in to provide nonmarket ‘paid jobs of last resort’. These jobs would be expressly designed to permanently employ people whose attachment to the labour market remains weak, despite reasonable efforts.This is the extreme case of ‘activation’. So far, workfare has only been a point of discussion in Europe, rather than a concrete proposal. In 17

Basic Income on the Agenda Figure 1: Four ideal types of basic social security and labour market policies Lax

Enforcement of entitlement conditions

Stringent

Conditional Welfare Narrow

A

BÕ

Wage subsidizing and job-training instruments Conception of work and conditions of entitlement

E Subsidized exemptions from paid work

B

Workfare

C

Participation Income

Broad D

CÕ

Basic Income

so far as it is a genuine response to the new social question, workfare could be motivated by the idea that even artificially created jobs can fulfil social demands which the labour market may not supply under minimum wage constraints; can serve as bridges to regular economic participation while maintaining discipline without loss of dignity; and finally, be a signal on the part of society that active welfare has replaced passive welfare across the board (WRR, 1997).Workfare schemes, provided that they are truly motivated in these solidaristic and participatory ways, constitute one extreme on a two-dimensional continuum of policies, of which a basic income at the ruling subsistence level is the other extreme. Before going further into this, note that the arrangements on which the various possible solutions to the new social question focus are predominantly concerned with universal minimum benefits, rather than with employee-related social insurance, which covers risks of earnings losses from unemployment, sickness, or disability. In what follows, we will concentrate on the treatment of those whose access to these last benefits is limited because they have an insufficiently secure link to the labour market.9 Although there has been a spectacular expansion of social insurance and other earnings-related coverage after the second World War, a growing part of the labour force now has to rely on the transfer payments under a socially guaranteed minimum safety net. Let us call these payments ‘basic security’. In most countries, means-tested social assistance benefits would be the usual form of basic security. Thus understood, the new social question revolves around the different forms which basic security – at a poverty-preventing ruling level of individual or household income – may take, and what types of labour market policy might best accompany these forms. Figure 1 gives a bird’s eye view of four ‘ideal types’ of basic social security.Together, they define a policy space of alternatives for achieving ‘full employment without poverty’. After having explicated the dimensions of this policy space, we will use it to put into a common perspective the matters raised in the contributions which make up Part I of this volume. 18

How Attractive is a Basic Income for European Welfare States? The vantage point of Figure 1 is point A at the top-left corner, which defines basic security of the work- and means-tested type in most existing regimes. We call it Conditional Welfare. Conditional Welfare is the type which motivates our exercise, since its conditionalities set up various unemployment traps.These traps arise because (a) taking a part- or full-time job with a net wage close to the level of the net benefit causes the worker to lose the entire benefit and (b) the rules of implementation are lax enough to allow prospective workers to pass over a job that does not result in a significant improvement of income and non-income related advantages. Moreover, since means-testing is applied to the joint income of adult families, the conditionalities also discourage taking up part-time work by one or both household partners. In many cases also, Conditional Welfare (at least in its pure form) extends to wealth holdings, and does not allow benefit holders to undertake training, follow educational courses, or engage in setting up a business. It thus also discourages activities of human and financial capital formation. From point A one can move along two dimensions, vertical and horizontal. Along the vertical axis, the conditions of entitlement and the conception of what counts as ‘work’, as applied in the work-test, become progressively broader, until one ends up, in point D, with a pure – non-work tested as well as non-means tested – Basic Income, at the ruling social minimum of a single earner. Several varieties of intermediate cases can be located along this dimension, for example a partial basic income at, say, one-half of the social minimum, topped up by conditional welfare, a non-means tested but workconditional income, or a fully means-tested income with relaxed conditionalities of work, which reflect a broader conception of work entitlement. Such intermediate cases might be located at point E in Figure 1. Going back to point A, let us now consider moving along the horizontal axis.This dimension is defined by gradations of increasingly stringent implementation of the entitlement conditions of Conditional Welfare, to some maximum at point B’. The notional move from A to B’ would virtually remove the disincentive aspect of the unemployment trap, since in B’ it would become impossible to refuse any part-time or full-time job which was actually being offered to someone on welfare. This move along the horizontal dimension is only notional, because it does not correspond to any existing or even proposed basic security policy.Yet in one way, the move from A to B’ captures the logic of the active welfare state approach discussed above, in which benefits are to be discontinued as soon as paid work becomes a possibility. Of course, what the move from A to B’ does not capture is the complementary feature of the active welfare state approach.As we have seen above, this consists in making available facilities of job search, training or occupational reorientation, as well as paid work in various kinds of government-sponsored kinds of job (ultimately including the workfare jobs of last resort) which are envisaged as the ultimate consequence of the active welfare state approach.Thus, the policy extreme that we identify as constituting the active welfare state’s ideal type – here called Workfare – must be located further down along the vertical dimension than point B’ in Figure 1 at point B.This is because ‘activating’ policies necessarily involve broadening some of the conditions of entitlement, so as to include all activities which contribute to enhancing the ‘employability’ of the welfare recipient as legitimate grounds for the basic security transfer. In a similar way, one can envisage a notional move along the horizontal dimension, proceeding rightwards from Basic Income in (Point D), to point C’. Obviously, this 19

Basic Income on the Agenda move is of only analytical significance, since the conditions of basic income – citizenship status or long-time residence – are relatively simple to monitor (which is not to say that they would always be simple to enforce in a given country). For the present purpose, however, the real interest of moving along the horizontal dimension to point C’, is to locate our fourth ideal type of basic security, Participation Income, which is a special form of work-conditional citizen’s income, originally proposed by Atkinson (1993b). We arrive at this type by moving upwards along the vertical dimension, to point C. The move from C’ to C signals the fact that a Participation Income, even though it is not means-tested, like basic income, is forthcoming only in return for the performance of ‘socially useful activities’, such as care work for the young, disabled or elderly, or community work of officially approved kinds, in contrast to basic income. These unpaid activities figure as grounds of entitlement to Participation Income, in addition to the entitlements generated by paid work, or generated by the activities of enhancing labour-market employability that the active welfare state defines. It should be noted that Participation Income was envisaged by Atkinson as an independent and non-means tested alternative to Basic Income, with work-related conditions that would render it invulnerable to the powerful charge – traditionally directed at Basic Income – that it violates the notion of reciprocity. So understood, Participation Income is a form of basic security that must take the enforcement of its work conditionalities just as seriously as Workfare does, the difference with workfare being that the conditions for receiving a Participation Income are more inclusive. Note however, that the arrangements of Workfare, as described in Figure 1, are also subject to means-testing, while Participation Income is not meant to be means-tested. Thus one might envisage intermediate forms between Workfare and Participation Income on the line BC, and of course one could also imagine laxer forms of Participation Income, which would start to resemble Basic Income, as one moved along the line CD in the policy space of basic security. So far, our description of the policy space has concentrated on the possible legislative and implementation forms of basic security. In practice, these forms need to be complemented by policies of labour market regulation appropriate to each. Roughly speaking, one can locate policies aiming at enhancing the demand and supply of paid work at the top end of Figure 1, while policies that facilitate the transition of individuals between paid and unpaid work would be located at the bottom end. Policies of this first kind include wage subsidies and tax credits for workers, as well as various forms of training- and job-programmes. Policies of the second kind, by contrast, include provisions for parental leave, sabbaticals, subsidies on going from full-time to part-time work, or subsidized early retirement schemes. It should be noted, however, that these last policies have not yet been developed in the trajectory of minimum income guarantees, on which Figure 1 focuses. In so far as such policies exist at present, they are typically attached to a secure employee status offering social insurance benefits, a point to which we return later on. Moving from left to right in Figure 1 will ordinarily require the introduction of supporting labour market policies of one kind or the other. For instance, if one wants to transform Conditional Welfare into something close to Workfare (along line AB), then the corresponding increase in the stringency of benefit conditions will only be seen as reasonable, if the recipients of basic security transfers also get better chances to actually fulfil the tighter conditions. This, in turn, requires labour market policies, such as a 20

How Attractive is a Basic Income for European Welfare States? wage subsidy for long-term unemployed, combined with a retraining programme.The more these supporting labour market policies succeed in expanding employment opportunities of their target groups, the more scope there will be to impose strict duties on welfare recipients to accept work, and penalize non-co-operation. At the Workfare extreme, finally, these labour market programs would become an integral part of basic security legislation, since they would have to involve permanent jobs of last resort. In a similar manner, one can argue that moving from Basic Income in the direction of Participation Income (along line DC) will require supporting policies that actually enable the recipients to meet the broader conditions of work that this last form of basic security wishes to impose. In part, such policies are of the kind that are traditionally regarded as subsidized exemptions from paid work. As a final note, basic social security in advanced capitalist countries is mainly situated along the line AB, with the U.S. more workfare-oriented, and Europe more welfare-oriented, but perhaps increasingly shifting rightwards. What happens on the line DC is strictly virtual, since neither a basic income nor a participation income actually exist anywhere in welfare states. One can at most detect elements of these pure types here and there. Admittedly, our characterization of existing arrangements does no justice to the great, mainly group-specific, variety in entitlement conditions (nor to the great variation in basic social provisions across European countries, e.g., in Italy conditional guaranteed minimum income is not universally granted). For instance, care law exemptions for single parent families might be taken to constitute a diluted form of a Participation Income. Mandatory job programs for youthful unemployed definitely belong to Workfare. Universal and non-means tested state pensions can be seen as a group-targeted form of Basic Income, if they are paid out regardless of whether persons have worked before age 65, as in the Netherlands. A similar remark applies to the rather lumpy classification of labour market instruments in Figure 1. Although there is a vast array of different kinds of wage subsidies – varying in type, level, duration, target groups and administrative and regulatory details – the primary policy objective of wage subsidies is to raise the level of employment by impinging on either the labour demand function for low-productivity work, through employer subsidies, or on the labour supply function, through employee subsidies. Under the former, the wage subsidy reduces wage costs for the employer, while under the latter it acts as a stimulus to accept low-paid jobs. Although both types of wage subsidies are equivalent in a neo-classical framework with invariably clearing labour markets, in real life there may be a definitive preference of one type of instrument above the other. As Anton Hemerijck remarks in his contribution below, the shortfall of low-paid jobs in Europe makes it most obvious to adopt employer wage subsidies (such as a reduction of social security contributions), whereas in the U.S., the great number of working poor calls for employee wage subsidies (like an Earned Income Tax Credit). Equally, although there are numerous public employment programs and (often unco-ordinated) constituency-based training programs in most European countries, these are predominantly temporary expedients, and they are targeted to the longterm employed at the far end of the labour queue. In other words, they fall short of a permanent and comprehensive workfare scheme, in which the government truly acts as an employer of last resort. Genuine workfare would call for a much larger task than any European government (except Sweden in the past, as paradigmatic case of European workfare) has started yet. 21

Basic Income on the Agenda Despite the variety in the actual configuration of different welfare arrangements and instruments, then, the four ideal types of basic security, as linked to the most appropriate labour market instruments, may be helpful to discuss specified objectives of social policy.

Fighting Poverty, Unemployment and Gender Inequality With this overview in mind, we now ask which types of basic social security and labour market policies are recommended by the contributors of Part I, for reducing unemployment without increasing poverty or income inequality, and for enhancing a more equal gender division of paid and unpaid work. As we will see, some authors are in favour of ‘packaging’ schemes, that is, they choose positions in between the ideal types represented in Figure 1.The central question posed by Paul de Beer is which of the following instruments – an Earned Income Tax Credit, a general wage cost subsidy, direct job creation,10 or a basic income – can be seen as a ‘double edged sword’, in terms of success in simultaneously reducing poverty and unemployment. In the short run, there is a definite winner, direct job creation. If the additional jobs are targeted on the long-term unemployed, then each job reduces unemployment, and ends poverty for the job-holders, provided the remuneration is above the poverty line. If the longterm unemployed are forced to take such jobs under threat of their social benefit being withheld, then this policy comes very close to the Workfare scheme of Figure 1. Crucial for the long-term impact is the extent to which targeted job creation helps the long-term unemployed to move on to regular (non-subsidized) jobs. If it does not help, then the policy amounts to an expansion of the public sector proportionally to the number of jobs created, with higher tax burdens than would have been the case without the job program. Now consider basic income.This policy is inferior to the job creation policy in the short run, but may become superior in the long run, provided two conditions are met. Firstly, a basic income should actually work as an incentive for the unemployed to take up part-time jobs (which will also be facilitated if many who would otherwise work full-time choose to work part-time under a basic income scheme, and so leave more jobs for the unemployed to fill). Secondly, the higher tax rates required to finance the basic income should not raise gross wage costs (and hence reduce labour demand), nor impair investment in human capital. Even if the reader agrees with De Beer that these two conditions are likely to be met in a liberalized market economy, then still his conclusion is rather sobering for those who think that the introduction of a substantial basic income is around the corner. For even if basic income would prove to be the sharpest double-edged sword in the long run, the greater effectiveness of job creation, or the nearly equal performance of an Earned Income Tax Credit in the short run, might be more attractive to politicians under the pressure of re-election. What distinguishes Conditional Welfare and Workfare on the one hand from Participation Income and Basic Income on the other is not merely the conception of work, but also the presence or absence of means-testing. Both a Participation Income and a Basic Income are non-means-tested, whereas social assistance under welfare and workfare schemes are usually means-tested (including not only social assistance, but also other benefits such as rent subsidies, or subsidized day-care centres). Under the 22

How Attractive is a Basic Income for European Welfare States? present arrangements, those who draw from basic security face a very high marginal tax rate on the income they could earn, up to the level of the means-tested benefits.11 There are various ways to remedy this ‘inactivity trap’: through an Earned Income Tax Credit, or a reduction of social security contributions for low-paid work, or by means of a partial basic income. To compare these three policy alternatives in the unified framework of a common minimum guaranteed level of basic security, and a common minimum wage, is the aim of Van Parijs, Jacquet and Salinas.12 Their starting point is that with the rise of the new social question, more and more people find themselves in the range of a 100 percent marginal rate of taxation.This makes it imperative to study available options for cutting back this prohibitively high rate. With commendable rigour, the authors demonstrate the following ‘equivalence theorem’. Assuming that all income is formal wage income, the three policies to alleviate the inactivity trap are equivalent.That is to say, given a suitable choice of parameters, Earned Income Tax Credit, reduction of social security contributions and partial basic income will result in exactly the same trajectories of post-tax-and-transfer income, given that the level of basic security and the minimum wage is the same in all three cases. Since Earned Income Tax Credit policies are used in Anglo-Saxon countries, while most continental European countries are trying out some form of reduction of social security contributions for low-paid work, this insight seriously challenges the strong preferences held by politicians in favour of either of these two policies, and their almost unanimous rejection of the (in principle, equivalent) partial basic income. If it became widely appreciated, this equivalence theorem might perhaps induce the more thoughtful policymakers to re-adjust their stance towards a partial basic income. But this is only the first part of the story.Van Parijs, Jacquet and Salinas devote the remainder of their contribution to sorting out where the equivalence claim between the three different policies breaks down.Without giving a full overview of their argument, it turns out that a partial basic income is better than the other two policies for poor households with some unearned income (e.g. pensioners and divorcees with alimonies), for low-earning self-employed and for low-paid voluntary working-time reducers. Moreover, a partial basic income might have a strongly favourable impact on the bargaining power of low-paid workers towards employers.There is, however, one important drawback. Partial basic income requires higher marginal tax rates on intermediate-range and high-wage workers. Due to its unconditionality, partial basic income has greater tax costs, which are avoided by Earned Income Tax Credit and social security contribution exemptions, because these last policies are targeted at the low-wage employed. This disadvantage may or may not be compensated by two types of behavioural effects of a partial basic income on welfare claimants and low-wage workers. First, fewer applications for supplementary social benefits, and secondly, the ‘reshuffle effect’ of increased work-sharing. Regarding the first type of behavioural effect, the higher the partial basic income is, the higher the number of welfare recipients will be who can earn a small supplementary wage income which brings them above subsistence level, and hence takes them off welfare. Additionally, reduced applications for supplementary benefits can be expected because of ‘the inconvenience to the claimant of claiming the full means-tested, willingness-to-work-tested guaranteed minimum income (intrusiveness, dole queues, job applications, risk of being caught ‘cheating’, etc.)’.These effects will work more strongly to the extent that welfare recipients under 23

Basic Income on the Agenda present arrangements hesitate to take a job because of ‘the fear of leaving the relative safety of a regular flow of benefits for the uncertainty of a job that may pay late or prove too demanding’. The second behavioural effect of a partial basic income that may compensate its higher tax cost is the ‘reshuffle effect’. Under a partial basic income, more full-time workers will choose to work part-time, freeing up jobs to the unemployed, while at the same time these unemployed have greater incentives to accept part-time jobs. As a result, both savings in social expenditures from fewer applications for supplementary benefits, and a more equal distribution of jobs among the workforce through work sharing (and hence, a smaller caseload of supplementary benefits) may be large enough to compensate for the higher fiscal costs of a partial basic income. To the extent that this holds good, a partial basic income would then compare favourably to the two other policy alternatives of Earned Income Tax Credit and reduction of social security contributions. So, by a different route,Van Parijs, Jacquet, and Salinas arrive at a similar overall conclusion to De Beer. Taking into account its likely behavioural effects, a partial basic income may be superior to other alternatives on the table, once there is enough time to let those effects play their part in the complicated institutional environment of the labour market and social policy constraints. However, partial basic income is bound to be the politically least feasible policy option, not least because giving unconditional income to the rich, in the interests of the poor, is a message difficult to understand for the ordinary citizen. Frank Vandenbroucke and Tom Van Puyenbroeck address the policy choice between proportional working hour employment subsidies and basic income from an entirely different perspective. This is the perspective of ‘responsibility-sensitive egalitarianism’, that is to say, a criterion of justice which seeks to compensate for low productivity, but holds people responsible for their own choices of income and leisure.They first explain why even a well-defined egalitarian yardstick of this type does not in itself dictate either of these two types of policy. The choice of employment subsidies or basic income (or some combination of the two) is shown to depend on the government’s stance towards the non-pecuniary (dis)advantages of (un)paid work, leisure and labour market participation.This implies that under responsibility-sensitive egalitarian principles of distribution, the government’s view on the work ethos is decisive for the choice of policy instruments. For example, if participation in paid work is highly valued by the government, because of its beneficial consequences for sustaining solidarity, the integration of immigrants, or the development of skills and social contacts, while it does not believe that these effects would result from participating in activities outside the domain of paid work, then the government may definitely favour employment subsidies and supplementary activation programs above the basic income solution. Conversely, if the general wisdom is that paid work imposes heavy burdens on workers (revealed in stress, burnouts, incidence of disabilities, etc.) whereas non-market activities (care and volunteer work) and leisure are experienced as highly valuable, then this would give a basic income a much better chance.13 Vandenbroucke, who is currently the Minister of Social Affairs in the Belgian Labour-Liberal government, and one of the most progressive proponents of the active welfare state approach, thus makes the point that the choice of the most appropriate instruments of basic security policy depend both on principles of egalitarian justice and on democratic judgements concerning the good life of citizens. 24

How Attractive is a Basic Income for European Welfare States? Joachim Mitschke takes issue with the conditional welfare arrangements in Germany for their manifest inability to provide enough low-wage jobs. As he points out, this is partly caused by the wage bargaining policies of the unions, which set the lowest wages above the productivity of low-skilled workers. The other cause of the German failure to induce employment at the low end of the labour market is that minimum social benefits are contribution- rather than tax-financed. This raises the non-wage labour costs of employers. Job creation and training programs are only of limited use, Mitschke argues, because ‘qualifications can only be improved within narrow talent-, milieu-, and age-specific boundaries’. Mitschke is also pessimistic about the adoption of temporary wage subsidies (e.g. equal to half the pre-tax wage up to one year):‘To fetch an unsubsidized job at the same wage rate, after the subsidy ceases, the former social assistance recipient would have to double his productivity within (less than) one year’.These perversities of employment-unfriendly social security can be set right by introducing a negative income tax, or ‘Bürgergeld’, of which Mitschke has designed several variants that have been widely discussed in Germany from the late 1980s onwards.These proposals show the potential of a negative income tax for harmonizing a wide range of existing instruments of social policy directed at the low end of the labour market. A negative income tax would act as a uniform and permanent graded wage subsidy directed to the workers, without the behavioural drawbacks of temporary wage subsidies, or the administrative complexities of local wage subsidizing policies. Having looked at several contributions that address the dual objectives of poverty alleviation and employment growth, we now turn to the issue of gender in the labour market. With reason, Ingrid Robeyns can claim that with minor exceptions, serious analyses of how basic income impacts on different policy objectives are gender blind. The same applies to other policy instruments. In this respect, it would be of interest to find out what a workfare scheme, with its narrow focus on the virtues of paid work, would mean for the position of women. Robeyns, however, is concerned to investigate the effects of basic income in her systematic overview of the literature. Among the female population, a precarious attachment to the labour market is much more common than among the male population. This fact justifies taking a critical look at the contention, dear to proponents of basic income, that basic income will act as an ‘emancipation fee’. It will, so proponents say, improve the bargaining position of (especially poor) women against bosses and husbands, and also encourage men to start working part-time, in order to take a larger share in care work. Robeyns compares the plausibility of this contention with the opposite view, which maintains that basic income, far from acting as a stimulus of emancipatory practices, is essentially a form of ‘hush money’. In this view, basic income will act to serve as a housewives’ wage, despite – or, for some, because of – the fact that a basic income recognizes the social value of unpaid work for both men and women, in a ‘gender-neutral’ way.As a result, so the hush money argument goes, men and women would sort themselves spontaneously into the traditional categories of breadwinners and housewives, and there would be hardly any way to complain about that, since it would be the consequence of ‘free choices’. In between these two extremes is the view of Hermione Parker. She argues that a basic income is to the advantage of women, not because it is especially women-friendly, but because present social security arrangements are far more biased towards male breadwinners, given that these arrangements are still strongly based upon the presumptions of the postwar 25

Basic Income on the Agenda welfare state: the stable, long-term employment relationship of the male family provider. This gives men a large stake in preserving the structure of earnings- and work-historyrelated social insurance benefits of unemployment, disability and retirement.Against this background, a basic income, as well as a participation income, would be reforms that revalue the importance of non-market work for society at large, and reduce the stakes involved in building up social insurance rights through paid labour. As Robeyns shows, considerable clarity in this controversy can be attained by distinguishing three dimensions: (i) the short-term and long-term effects of basic income, according to whether (ii) women are highly skilled or low skilled, and whether (iii) women have a weak or a strong labour market attachment. In the short run, a substantial basic income will probably decrease women’s overall labour supply.This might be taken to support the hush money thesis. In the long run, however, when new generations enter the labour market, among which it is more commonplace for men to work part-time, and take a greater share in domestic work, it is far less obvious that hush money effects will dominate the emancipation effects of a basic income. Robeyns’ distinction by groups then indicates that the real winners from a substantial basic income will be low-skilled women with a weak labour market attachment: greater economic independence, more bargaining power, and no loss of non-pecuniary advantages of paid work, given their already low participation rate. On the other hand, it is unlikely that career women, being high-skilled, and having paid-work as a central life interest, will give up their jobs to live from a basic income only.The evidence is mixed for the other two groups of women in Robeyns’ classification. Perhaps it is fair to say that the final evaluation of a basic income, from the standpoint of women’s emancipation, depends as much on the expected effects as on one’s view of what emancipation is all about. Does gender equality require no educational and occupational segregation, as well as no gender division of labour, or can some of this be tolerated, if equality of opportunities still results in different choices reflecting different preferences between men and women? These questions clearly emerge from Robeyns’ thorough treatment of the empirical issues. As a final note, we would like to point to an idea only casually mentioned by Robeyns.This is a basic income targeted at children, which would cover the material costs of upbringing, as well as some of the opportunity costs of child care. Of course, even though such a basic income would replace all present child allowances as well as existing subsidies for child care and family tax allowances, it will still be a very expensive measure. But it would offer each parent – especially mothers, as long as they are the ones predominantly taking care of the children – much more of a real opportunity to choose freely between full-time work, full-time care work, or any mix of these. The essays discussed above suggest that participation income or basic income are promising ways to address the problems posed by the new social question, or to empowering low-skilled women. However, as Anton Hemerijck’s essay makes clear, experience in the Netherlands has shown that less radical measures in between basic income and workfare can serve to address the new social question as well. In particular, the Dutch policy of sustained wage moderation has proved to be successful in terms of employment growth.This policy has recently been flanked by targeted wage cost subsidies for long-term unemployed and low-wage workers, the introduction of modest schemes for educational, parental and sabbatical leaves, and a central agreement between unions, employers and government which traded a relaxation of firing and 26

How Attractive is a Basic Income for European Welfare States? dismissal procedures for full-time core workers against greater job and social security rights for flexible and part-time workers. Despite the apparent ability to accommodate the trend towards more flexible labour markets – the Netherlands has an unparalleled high share of part-time workers – the Dutch miracle has not yet succeeded in overcoming structural inactivity, nor the precarious position of specific groups on the labour market.As Hemerijck points out, structural inactivity has remained more or less constant at about 20 percent of the labour force, and it is concentrated among the low skilled, older workers, immigrants and women. The majority of the older, predominantly low-skilled, workers are either on early retirement or on disability insurance. Finally, the problem of long-term unemployment, still about half of total unemployment, persists as long as the ‘baby boom’ cohorts of younger and better skilled workers, along with increased participation of women, absorb employment growth. For the near future Hemerijck is in favour of embracing wage subsidies following a ‘negative income tax logic’, as the most effective and feasible route to increase participation of the most vulnerable groups. Hemerijck is right in stressing that in a corporatist climate, with the commitment of all parties (unions, employers’ associations, and government) much can be achieved by piecemeal engineering. However, it may be doubted that the Dutch case can serve as the benchmark of a ‘best practice policy mix’ for other European countries. For one, wage moderation can be expected to have less favourable results on labour demand when adopted Europe-wide than when it is practised in one country (the ‘beggingthy-neighbour’ argument). Secondly, shared commitments and collective agreements on wage moderation may be more difficult to achieve in countries like Germany, France and Britain, with their more militant trade unions, than in the consensus-oriented society of the Netherlands.This raises the question of the next section, whether European countries would at all be able to find new ways of co-operating in the search for full employment without poverty.

A Participation-Based Eurodividend: The Key to Social Europe? To introduce the European dimension, consider the following statement, in which Atkinson (1998: 148-9) reiterated his earlier plea (1993b) for a European Participation Income: It is noteworthy that, despite the attention which basic income has been given, and despite finding supporters in all political parties, the scheme has not got close to being introduced. If one asks why, then, in my judgement a major reason for opposition to basic income lies in its lack of conditionality. There are concerns that it will lead to dependency, or stateinduced social exclusion. I believe therefore that, in order to secure political support, it may be necessary for the proponents of basic income to compromise – not on the principle of no test of means, nor on the principle of independence, but on the unconditional payment… I believe that such a Participation Income offers a realistic way in which European governments may be persuaded that a basic income offers a better route forward than the dead end of means-tested assistance. New ideas in this field are needed, and this one combines the twin concerns with poverty and with social exclusion.

As this passage makes clear,Atkinson is in favour of reforming the present means-test27

Basic Income on the Agenda ed basic social security arrangements (in terms of Figure 1, Conditional Welfare) into non-means tested basic provisions, preferably conditioned on a broadly applied worktest. Atkinson’s preference for non-means tested social benefits is motivated by the problems of the new social question, in particular the fact that a two-tier structure of (non-means tested) social insurance and (means tested) social assistance strongly discriminates against the most vulnerable in the labour market. This focus on the detrimental effects of means-testing explains why Atkinson prefers a participation income to a basic income, on the assumption that the work test of the former would be capable of securing the required political support for doing away with means tests in basic security.This view is attractive, even though a more relaxed work test would have to be enforced seriously, as we have pointed out in our discussion of Participation Income as an ideal type of basic security (see Figure 1).This may well prove to be no easy task, in view of the reasonable disagreement as to what counts as ‘participation’. In any case, the proposal for implementing some kind of participation income at the European level, here stated in guarded terms by Atkinson, is essentially favoured by Steve Quilley, as we will discuss below.14 But before turning to Quilley’s argument, we want to introduce the exchange of views between Fritz Scharpf and Philippe Van Parijs on the importance of European co-ordination. As Scharpf argues powerfully, the hope that any form of ‘Social Europe’ will come about without concerted decision-making at the European level is entirely unrealistic. Competitive pressures resulting from the single economic market (shortly to include several Eastern European countries), and from a single currency, would continue to have an adverse impact on the national social and economic policies of the Member States.According to Scharpf, economic integration constrains national welfare arrangements in two important ways. First, in a relentlessly competitive market without borders, governments are forced to shape social policies with an eye to the establishment of a competitive welfare state. Secondly, even if one country could manage to provide generous welfare arrangements, while preserving its international competitiveness, sooner or later it would become vulnerable to welfare migration, particularly of lowskilled workers from other countries in or outside the European union. In the longer run, this would undermine the viability of the generous policies. Tax competition between European countries, and unco-ordinated reshaping of welfare arrangements is likely to lead everywhere to a separation between ‘individualistic’ social insurance and ‘solidaristic’ redistributive programs. If so, the new social question would be reinforced.That is, the gap in lifetime prospects between a core of medium- and high-skilled workers with stable job and income security assets, and a marginalized workforce which has to rely on basic security, would widen. Why? According to Scharpf, workers would tend to regard social insurance entitlement as a form of delayed earnings (because of the actuarial fairness between contributions paid and coverage provided). Therefore, social insurance would be less vulnerable to tax resistance everywhere, and it thus would not create competitive disadvantages. By contrast, generous tax-financed redistributive programs would be likely to run into far more resistance, on the part of the net contributors.As a result, keeping such programs at a high level would be electorally unpopular, hence far more vulnerable to being eroded by tax competition. Countries that decide to maintain generosity in basic security would have to pay the price of reduced international competitiveness, while at the same time attracting welfare migration.As long as social policy remains a national pre28

How Attractive is a Basic Income for European Welfare States? rogative, the institutional separation of social insurance and basic security will continue to induce new rounds of competitive pressure on basic security. In the absence of European policy co-ordination, this may lead to a ‘race to the bottom’. Under such conditions, it seems obvious that national states would want to avoid introducing a taxfinanced and highly redistributive basic income scheme, or a participation income. So in Scharpf ’s view, the only hope for the disadvantaged is that competing redistributive alternatives of basic income, like the Earned Income Tax Credit and reductions in wage costs, can still command sufficient political support. But is this all one can say? Not entirely. For Scharpf ’s realistic assessment is logically compatible with some kind of a European basic income.This, however, would require a prior effort at co-ordination of social policy. For example, if Member States can somehow first agree to guarantee a system of Europe-wide minimum income standards (for example, related to the per capita income of Member States), and if, moreover, the rich Member States were willing to fund the required additional expenditures, then this would check the tendency of levelling down basic security provisions. Against this more optimistic background, Scharpf says:‘if such support were forthcoming – perhaps through a redirection of the cohesion funds – it might be necessary and possible to avoid the work disincentives inherent in the present social assistance programs of the richer Member States. In that case, basic income programs, most likely in the form of the negative income tax, might indeed provide the most effective and efficient solution to the twin problems of poverty and unemployment, and European policymakers might well be persuaded to promote this strategy’. Van Parijs’s reply to Scharpf seizes on this possibility of prior co-ordination in a ‘Social Europe’. His assessment is that a European participation income along the lines suggested by Atkinson, would then be the way forward in an attempt to reduce the danger of fiscal and social competition between Member States still further. As we explained in the previous section, while a participation income avoids means-testing and the ensuing inactivity traps, it is politically more viable because its broad participation conditions signal a concern for tying the receipt of transfer income to the reciprocal performance of ‘socially useful activity’. It is thus a good candidate for counteracting the divisive effects of the separation between social insurance and basic security. For these reasons,Van Parijs proposes to recast the term ‘participation income’ as a ‘Euro-dividend for all active European citizens’. He sees two further advantages. First, if the Euro-dividend were to be set at a uniform level – which would probably be close to the level of the minimum income guarantee of the poorest countries in the EU – this would provide a universal floor, on top of which each more affluent Member State could then maintain its own favourite mix of welfare arrangements. Secondly, a uniform Euro-dividend, even if it were low in comparison to the average European level of basic security, would considerably increase inter-European solidarity. For, irrespective of its tax base, a uniform Euro-dividend would raise redistribution from rich to poor countries, or from booming to depressed areas, far beyond the present redistributive effect of the structural funds.15 Steve Quilley’s political reflections chime in with Scharpf ’s assessment as well. Since there is hardly a chance that any full-fledged basic income scheme will get off the ground at the national level under increased intra-European competition, a framework of co-ordination is imperative for getting politicians ‘to think the unthinkable’. In a way, taking this as a starting point is to admit that the project of basic income depends 29

Basic Income on the Agenda on a degree of institutional co-operation that may seem rather utopian, given that social harmonization is at the tail end of the agenda of European integration at present. Quilley’s main concern is to address this problem. He argues that while the debate on basic income has become more mature in clarifying the ethical desirability of unconditional grants, as well as the feasibility of (revenue-neutral) schemes of basic income, ‘…such intellectual clarity does not flow naturally into a plausible political project’. The debate, so far,‘has offered few clues as to possible points of departure and the politics of transition’. It is necessary, therefore, to reflect on why it makes good political sense, in the longer run, to regard a small Euro-dividend of the type envisaged by Van Parijs (see above) as the nub of a ‘basic income trajectory’, that is to say, as a strategic measure that could create the conditions for its own ‘metamorphosis into a more substantive redistributive mechanism’.The merit of Quilley’s contribution to Part I is that he seeks to locate the political opportunities that exist, at present, for discussing the new social question at the European level, in ways that would move likely solutions away from Conditional Welfare and Workfare, and towards Basic Income or Participation Income – that is to say, downwards along the vertical dimension in the policy space of Figure 1. According to Quilley, there are several reasons for expecting a more favourable political constellation over time. First (and especially looking at the problem from the British point of view), the operation of the European single market under conditions of increasing flexibilization is creating pressures to introduce an element of fairness and even-handedness in the process of corporate restructuring and relocation. Secondly, the inter-regional and national inequalities resulting from the single market can be sustained only if they are partially legitimated by a sense of European citizenship. But while both of these features show that the logic of competition that has driven European integration so far is creating a political need for redistributive social cohesion, this in itself does not imply a policy preference in favour of basic income. As Quilley admits, the redistribution involved in basic security seems to be tied more closely everywhere to the work ethic, targeting, and means testing. If anything, then, European basic security is moving from Conditional Welfare to Workfare, along the horizontal dimension in the policy space of Figure 1.There is no straightforward way to achieve a ‘ninety-degree change of direction’.What one can envisage, however, is a possibility to attract political forces toward Participation Income, by appealing to the ideology of a broadened and civic work ethic, which would be compatible both with the Anglo-Saxon rhetoric of communitarianism and self-help that underlies at least some of British New Labour’s work-to-welfare policies, and some of the more enlightened forms of the ‘active welfare state’ on the continent. Thirdly, Quilley strongly endorses the funding of the Euro-dividend through a combination of ecotaxes, because taxing ‘environmental bads’ (rather than productive activities) may seem a more acceptable ethical basis for the distribution of universal benefits, and also because linking basic income to the Green movement is politically opportune: ‘The emergence of a space for a green and socially responsible capitalism, is perhaps the nearest thing to the ‘big economic idea’ for the left since Keynesianism’. As Quilley emphasizes, these features of the political landscape create no more than windows of opportunity, which can be utilized in the development of a ‘basic income trajectory’. But it should be understood that the effectiveness of such a trajectory will require dropping the idea that basic income, in whatever form, is a self30

How Attractive is a Basic Income for European Welfare States? contained blueprint to be put up on the political agenda as a straightforward choice. Thus, any realistic attempt to move the politics of European redistribution in the direction of more universality and less conditionality will have to be a ‘war of position’, of institutional and political jockeying, in which proponents must seek to link these objectives with other political agendas, emphasizing similarities of purpose, rather than focusing on programmatic contrasts. While there is certainly a lot to be said in favour of Quilley’s realist utopianism, it does of course presuppose that the idea of basic income has gained sufficient foothold in the European Member States. We now want to examine to what extent this can be said to be the case.

Basic Income in European Countries The second part of this volume is entitled ‘Political Chances’. It offers seven papers, each of which illustrates major aspects of the political fortunes and failures of the basic income proposal in European countries, without necessarily attempting to cover every relevant development. In order of population size: Germany (Stephan Lessenich), France (Chantal Euzéby), Finland and Sweden (Jan-Otto Andersson), the Netherlands (Loek Groot and Robert van der Veen), Belgium (Yannick Vanderborght), Ireland (Sean Healy and Brigid Reynolds) and Denmark (Erik Christensen and Jørn Loftager).16 These papers enable one to assess the claim with which we started: that basic income has evolved to the stage where it has entered the political agenda, that is, has become a recognized political possibility. But it is important at the outset to see what it means to say that basic income is ‘on the political agenda’. Does it just mean that it is regularly discussed in the broad forum of public opinion (the popular media, position papers of policymakers, academic conferences) or does it mean that properly costed reform proposals involving a full or partial basic income, participation income, or negative income tax, have been, or are about to be, put up for decision on the legislative agenda of a ruling government? Obviously, the answer must be varied: everything that goes on in between these two definitions should be counted. Then there is the separate question of judging the results of these discussions and decisions. Since at present the policy space of basic security is really only inhabited at the top half of Figure 1, if basic income has at all been on the agenda of a particular country for a certain length of time, this has not yet had much of a definitive policy impact. So far, then, the ‘agenda status’ of the proposal must signify that basic income has been considered and then rejected, only to resurface at a later date, perhaps in a different form, and with different arguments, for a new round of consideration and rejection, or hopeful postponement, as the case may be.At the end of the period considered – roughly from the late 1970s up into the first months of the new millennium – the commentator may be in the downswing of facing up to the latest rejection (‘basic income is now nearly dead’), or be able to give a more upbeat report to the effect that ‘finally’, the idea has been taken up by a social movement, is being ardently discussed in the media, is being seriously debated in parliament, or is at least being singled out for further study, by some prestigious commission, or governmental advisory board.As the papers of Part II show, moreover, the agenda status of basic income between different countries must be considered against the background of the widely varying social security systems that have been in existence in each of them in the last two decades, as 31

Basic Income on the Agenda well as the country-specific major events that induce changes in the climate of policy, such as the re-unification process in Germany. In some cases, the institutional status quo has effectively prevented basic income from being admitted at all into the arena of discussion. This is the case most notably in Sweden.As Andersson shows, the reason is the highly developed Swedish welfare state, with its wide and effective coverage of social insurance, its long history of active labour market policy, and the firm hegemony of a social-democratic work ethic. In this setting, basic income does not seem much of an issue, primarily because the area of basic security itself does not yet stand out as being particularly problematic. Even where basic security certainly is seen to be problematic, as in Germany, the predominance of social insurance, the sheer complexity of the patchwork of social assistance, and the tradition of corporatist wage bargaining practices present conditions under which a basic income solution, such as Mitschke’s Bürgergeld, may in the end be rather far removed from what are perceived to be politically viable deviations from the status quo, as Lessenich notes in his instructive discussion of the policy stance of the social-democratic Friedrich Ebert Stiftung. However, in France, basic income is largely confined to the margins of academic discussion, precisely because it seems to be more of a pressing problem to get a universal system of basic security – one which covers all age groups – off the ground in the first place, following up from the 1998 Revenue Minimum d’Insertion (RMI). In such a political context, as Euzéby puts it politely, basic income is an ‘interesting but uncertain idea’.The reason is that policymakers with solidaristic goals find it more rewarding to start working on a basic pension system, and to take steps in the direction of including the young, the intermittent workers in the service sector, and the self-employed, in one integrated national safety net. From that point of view, rightly or wrongly, further questions about the conditionalities of social assistance are ones that have to be decided along the way, in order to advance at all. In addition, policymakers in France are strongly concerned to improve the employability conditions of the RMI, which has up to now largely failed to connect its clients to the labour market. In view of this goal, as Euzéby notes, it again stands to reason that a move to basic income should be preceded by intermediate and far less radical measures. Given what we have said up to now, the natural home of basic income policies would seem to be countries with highly developed (and cost-of-living or wage indexed) universal basic security provisions with high rates of take-up. Roughly, these are represented by the smaller Nordic welfare states of Belgium, Denmark, and the Netherlands. Here, it would at least seem that insofar as the ideological constraints (the recent activism of the work ethic) and financial restrictions (budget neutrality, the EMU-norms) allow, there is some scope for a gradual erosion in the conditionalities of a social minimum framework which is already securely in place. However, judging from the contributions from these countries, the current political prospects of promoting a basic income with open vizier are not favourable. This message is conveyed in different ways. In the Danish case, as Christensen and Loftager show, basic income was effectively removed from the political agenda in the 1990s by a decisive ‘top down’ ministerial intervention, silencing the Radical Liberal, Green, and generally the ‘alternative’ voices in favour of the idea. In Belgium, the sudden advent of a political party (VIVANT), which was committed to a clear and unambiguous form of basic income head on, has not seriously affected even the state of the 32

How Attractive is a Basic Income for European Welfare States? public debate, according to Yannick Vanderborght. Significantly also, the efforts of this party have hardly been supported by the two Green parties, both of which have been consistently in favour of basic income since the late 1970s. And while in the Netherlands, after twenty years of dramatic ups and downs, a small – though analytically significant – element of negative income tax has been built into the recent tax reform, the political understanding shared by policymakers and politicians alike is that this, surely, has nothing at all to do with the concept of basic income. If one believes that these countries can resist the grim process of convergence to the ‘competitive welfare state’ (as described by Scharpf) long enough, however, it would be natural to think that astute ‘institutional jockeying’ (as advocated by Quilley) may go some way to achieve slow shifts towards forms of negative income tax or participation income Finally, there are two countries that might be regarded as the ‘jokers in the pack’ – Ireland and Finland. Both of these seem to have in common a climate and history of social policy which is more conducive to sudden changes, and is more receptive to the central intuitive argument of fairness, efficiency and simplicity, which is associated with basic income. Sustained economic growth from a relatively backward starting position (Ireland), and the combined impact of high unemployment and exposure to the electronic age in the 1990s (Finland) may have something to do with this apparent volatility in the policy space of basic security. Cultural variables may also matter, in the case of Finland. If it is true, as Andersson argues in his survey, that as compared to Swedes, Finnish workers and citizens are more disposed to go for innovative and radical solutions, then some of the many concrete proposals that have been circulating in Finland since the 1980s involving basic income or negative income tax may be actually implemented.The more so, since Osmo Soininvaara, one of the main players in this field, has taken office as Minister of Social Affairs in a coalition including the Finnish Greens (see his Preface to this volume). We resist the temptation to speculate at length on these cases.There is not enough information to go on, as yet, to decide whether the impression that basic income is on the fast track in either Ireland, Finland or both, is justified. One small remark about Ireland, though.The sudden popularity of basic income can partly be explained by the fact that although the Irish economy is booming, the incidence of poverty remains high. The former gives some leeway to increase social expenditures (Healy and Reynolds use the wonderfully optimistic expression of ‘revenue buoyancy’), while the latter forces politicians who want the poor to share in the fruits of growth to seriously consider radical solutions.The case also illustrates how a relatively small but excellently well-organized community group (the Conference of the Religious of Ireland CORI) can have a major impact on the political agenda.Yet the outcome of CORI’s costed proposal for a full basic income is extremely uncertain, until the Irish government has actually taken an official stand on the substance of the proposal – which it has not yet done, despite its promises.The Dutch experience, meanwhile, has shown that as soon as the political establishment takes basic income reform proposals seriously, such proposals are subjected to extremely harsh tests, and moreover, that actual decisions on them tend to be shifted forwards in time, when an outright rejection is inopportune. It is to be hoped that the Irish élan will be kept alive if the window of opportunity were to narrow to the point of piecemeal implementation, and the long march through the institutions starts to take its toll. 33

Basic Income on the Agenda

Some Common Themes The surveys of Part II speak eloquently for themselves. By way of concluding this introduction, however, we want to look at some common themes. First, the initial attempts to launch basic income on the agenda in the 1970s (at least in countries with long histories of the idea) came from relative outsiders, whose interest was to promote a critical politics of free time, as opposed to the mainstream productivist politics of full employment. Secondly, what helped these ideas to gain wider currency in politics has invariably been sudden rises of unemployment, together with the idea of jobless growth, or – later in time – the fear that economic growth is permanently linked to insufficient demand for low-skilled labour. Thirdly, the radical unconditionality with respect to the willingness to work has invariably proved to be a stumbling block for basic income to become part of the mainstream alternatives, despite its heightened salience under conditions of high unemployment. Apart from the popular spectre of ‘able-bodied scroungers’, the objection to the no-work feature of basic income translates naturally into the recurrent (and highly damaging) images of ‘generalized state dependency’ and of ‘legitimizing the ongoing process of social marginalization’ (see Christensen and Loftager, below). In part, the intellectual state of the debate on basic income may be measured by the frequency with which convincing answers to these standard points are brought into play. As the account of Healy and Reynolds suggests, at least some of the success of moving basic income closer to the Irish legislative agenda must have depended on the sheer didactic effort of the proponents (notably, but not exclusively, CORI) in patiently countering a host of popular objections. Many commentators to Part II note the unmistakable shift of European social policy towards the active welfare approach, since the early 1990s. As Lessenich observes for the German case, this has put basic income advocates in the difficult position of having to argue that their proposals are the most cost-effective method of how to make them all work, which he rightly considers to be ‘utterly sobering for any apologist of basic income in its purist sense’. But, as the histories of the Dutch, Danish and Finnish cases show quite clearly, the no-work objection predates the shift to active welfare policies. It has always been quite powerful, and there is reason to think it will remain powerful – not merely as a policy conception, but mainly as an electoral constraint on feasible policies – as long as the no-work objection seems to be widely shared by net contributors to basic security and net beneficiaries alike. Note, however, that the extent of this presumed sharing is far from being a well-researched fact. It is not known exactly what ‘the electorate’, or ‘the ordinary people’ actually think about issues of income conditionality, since nowhere are citizens being asked to give their opinion on those issues in any systematic way, apart from the occasional survey question. Fourthly, even though it is difficult to generalize across the European ideological landscape, the social-democratic parties have always been highly critical of basic income, whereas the radical Left (early on, in the 1970s) and the Greens (in the 1980s and 1990s) were basic income’s most loyal allies.This is so, at least, if they do not come to power all of a sudden. See below, where Lessenich acidly remarks on how the German Greens silently dropped their advocacy of a more or less basic-income-friendly proposal, on entering the Red-Green government coalition. 34

How Attractive is a Basic Income for European Welfare States? Social democracy has its roots in the trade union movement, and to some extent it remains stamped by the old social question of capital against labour. Unsurprisingly, therefore, the social policy of much of modernized Labour is strongly oriented to the activity of paid work, though predominantly in the form of employability and of disciplining transfer recipients in acquiring work skills, as the developments in the United Kingdom reported by Quilley also show. Outside of Sweden, where an enlightened attitude to paid work has always been a marked feature of the welfare state, there are recent signs of an increased willingness to engage in the discourse of integrating unpaid work and leisure. A good example is the Dutch Labour Party’s 1994 election slogan of a ‘relaxed organization of work’, which refers to a kind of lifetime ‘employability’ in both paid and unpaid activities, with flexibility of choice to move in between the two. But just as in Sweden, this more relaxed attitude is at present largely confined to the sphere of work-related social insurance. With respect to basic security, Dutch socialdemocrats generally still tend to insist on the centrality of paid work, perhaps driven by a vaguely Victorian logic of the need to discipline the poor. This explains why the enlightened stance of ‘unpaid work counts as well’ can nicely go together with branding basic income as a defeatist policy, which is bent on trading away the ‘right to work’ for those ‘who need it the most’, against a politically insecure ‘right to income’. The same may be said for Denmark, on Loftager and Christensen’s account.There, as in the Netherlands, basic security is moving towards the activation of welfare clients, in tailor-made instatement contracts with administrative case-workers.17 In the Danish case, this goes together with the further expansion, and the great popularity among workers, of comprehensive schemes of paid leave. In terms of Figure 1, one might be tempted to see this last development as a vertical movement, in the direction of the ideal type of Participation Income. But as we stressed, that would be a mistake. For Figure 1 exclusively depicts the policy space of basic security, not that of work-related social insurance. Paid leave, obviously, is the privilege of those performing paid work in collective (or in semi-private) arrangements, from which an increasing number of people are excluded.The distinctive hallmark of a basic income, and similarly of a participation income, is to extend such privileges universally, at the level of basic security. The sympathy and occasional support for basic income of Green parties across Europe can be traced to the progressive middle-class intuition that the productivist stress on employment sits uneasily with promoting environmentally-friendly ways of life. Nevertheless, as the Greens have moved to the centre of politics in Europe, they tend to be more in favour of a moderate partial basic income strategy (Germany and the Netherlands), or for shortening the standard working week, as in France. As Quilley notes, this might be explained by the fact that social policy is only one item, and not necessarily the main one, on Green platforms of ecological modernization. Taxing ‘environmental bads’ instead of labour may be a way of creating a new, and more generous, foundation for basic security. But it may equally be used to simply shift the tax base (as in the recent tax reform in the Netherlands), or to create the fiscal means for assisting investment in renewable energy sources. In any case, up to now, Green parties have not yet actively engaged in bringing forward the national or European eco-dividend, as envisaged by Van Parijs and Quilley. Fifthly and finally, moral intuitions and efficiency concerns are connected in politics.This can be seen from another recurrent theme. Understandably, those who regard 35

Basic Income on the Agenda basic income reforms as morally suspect attempts to provide ‘handouts to everyone’ will also tend to regard such reforms as fiscally and economically unsound.There is no need to illustrate this standard point by reference to the surveys of Part II. But as the debate on basic income has matured over the years, a host of more sophisticated efficiency arguments have emerged from the heart of political exchanges. One of these is the impossibility theorem, as we have called it in our report on the Dutch debate.The impossibility theorem asserts an inescapable dilemma: in its economically or politically feasible range, a basic income will be too low to be socially acceptable, while in its socially acceptable range – which must be close to the ruling level of conditional basic security – basic income is too expensive for it to be either economically sustainable at all, or too expensive, given political constraints on marginal taxation levels. This line of reasoning has been used both in Denmark and the Netherlands to great effect. It obviates the need to argue the moral issue directly, granting for the sake of argument that unconditional incomes would be ‘socially acceptable’, if they could feasibly be pitched at the level of the ruling social minimum. Now, if a sensible proponent of basic income admits that this, indeed, is not economically feasible at present, then she is also forced to admit that a partial basic income is the only possibility, and that in order to be socially acceptable, the partial basic income will need to be supplemented with conditional benefits. The next move of the opponent is now to grant that a full basic income could combine adequate social protection with the demands of flexible labour markets; that it could lead to a more equitable distribution of income, of paid work, of care work and of free time between men and women; to less involuntary unemployment; to better working conditions for low wage workers; to a removal or shortening of poverty traps; and to a reduction in the administrative costs of providing social security. However, the opponent then goes on to say that while all of this might be true, the beneficial effects of a partial basic income would surely not be significant enough to make the reform worth the trouble. The opponent finishes by concluding that the modified basic income proposal will not make good on its initial promises, and would still be more expensive than a streamlined and more employment-oriented scheme of basic security would be. Even though this argument has often been put forward by opponents who also resist basic income on moral grounds, it is by no means an easy one to answer. In fact, as we show below, the argument has contributed to the development of costed basic income models in the Netherlands, which attempt to track the dynamic behavioural effects of specified changes in tax rates and benefit levels under various conditionalities of the benefits. In view of the uncertain validity of such models, however, arguments of the type of the impossibility theorem also point to the need for generating more reliable knowledge by promoting local experiments with basic income schemes.18 With this last point, we are back where we started in this Introduction: the undisputable success of the basic income discussion of the last two decades has been to set higher standards of expectation. We now invite our readers to judge to what extent these standards are being met in the contributions that follow.

36

How Attractive is a Basic Income for European Welfare States?

Notes 1

2 3

4

5 6 7 8

9

Founded in 1986, the Basic Income European Network (BIEN) aims is to serve as a link between individuals and groups committed to, or interested in, basic income, and to foster informed discussion on this topic throughout Europe.The name has turned out to be somewhat mistaken by now, as not only European, but also politicians and scholars from North and South America have joined the network. BIEN publishes a Newsletter which provides an up-to-date and comprehensive international overview on current ideas and publications, organizes bi-annual BIEN-congresses where people from more than twenty countries meet to report and discuss on all aspects of basic income and related proposals, including its moral, economic, political, social, and fiscal justifications, feasibilities, alternatives, objections, premises, experiments and outcomes. More information, as well as the previous BIEN Newsletter issues, can be found at BIEN’s Web-site: http://www.etes.ucl.ac.be/BIEN/bien.html. See White, 1997;Van Parijs, 1995;Van Donselaar, 1997;Vandenbroucke, 1999a; and Van der Veen, 1998. In the normative discussion, moreover, the debate around basic income is also concerned with exploring the consequences that work-sensitive or unconditional forms of redistribution may have for the self-respect of claimants (Wolff, 1998) and for the extent to which the labour market will work so as to compensate for the disutility of lowly productive jobs (Groot, 1999 Chapter 4). The contributions of Van Parijs et al. and De Beer in this volume revolve centrally around this hard exercise. It may be said, perhaps, that the proponents of basic income in the traditional sense of full unconditional grants at social minimum level will not receive maximum reassurance from what is said. A response which is invited by the exercises of examinations of ‘unheeded differences’ by Van Parijs et al., and De Beer’s cautious assessment of short- and long-term advantages, is to revert to the well-tested strategy of pointing out that if basic income cannot win out on any one or two policy objectives, then surely it may still prove superior on a more inclusive set of objectives. We shall not be offering examples of this seductive reasoning here, but point out the connection between the themes of Robeyns (gender equality) and Hemerijck (citizenship) in Part I and to the ‘middle range effect’, described by Fitzpatrick (1999: 44-45) as follows:‘... in terms of each policy objective taken individually basic income would be trumped by the other proposals..., but scores rather well when the entire range of objectives is taken into account... the reason why basic income has been ignored is because of the tendency of politicians and policy-makers to examine social objectives in isolation’. See Van Parijs et al. in this volume; Offe and De Deken, 1999; and Standing, 1999. This was one of the main themes of the Basic Income European Network Congress in Amsterdam, 1998. For similar assessments of the present situation, see e.g.Van Parijs, 1993: 134-39; Scharpf, 1999; Standing, 1999; and Vandenbroucke, 1998. Poverty and unemployment traps (giving rise to hysterisis among long term unemployed), binding minimum wages restricting the number of low paid, labour intensive service jobs at the bottom end of the labour market, restrictive firing and dismissal procedures, and extensive collective labour agreements (see e.g. De Neubourg, 1995). We do not want to suggest, of course, that present arrangements of earnings-related social insurance or tax-financed benefits are immune to the call for reform.

37

Basic Income on the Agenda 10 Direct job creation occurs through generous marginal wage subsidies for additional jobs open to long-term unemployed workers only. 11 This is due to the high withdrawal rate on all net labour income up to the social security benefit, and because working may engender additional costs (e.g. travelling expenses, costs of day-care centres, etc.). It is no exception that welfare recipients face a marginal tax rate of 100 percent or more.Those caught in these poverty and employment traps therefore experience a strong disincentive to work: they can only escape the poverty trap when they succeed in finding a job with pre-tax earnings considerably higher than the level of their benefits.All (part-time) work with pre-tax earnings below a certain threshold will not yield any financial reward for those on welfare. This may partly explain why part-time work is mainly undertaken by women.Women with a working partner are not entitled to any benefit and therefore they do not face the poverty trap. For them, any part-time work will raise family income. 12 Note that the authors do not include stringent workfare policies among the alternatives to be considered. Their comparison does not cover the whole of the horizontal dimension in Figure 1. 13 Even in that case, it still has to be shown that a basic income policy is better than, say, a comprehensive scheme of paid – sabbatical, parental, educational – leaves. 14 In the concluding chapter, Atkinson (ibid., 152) writes in less guarded terms: ‘As we have seen, the popular recommendation of greater targeting is less promising than appears at first sight.We have to search for new approaches, particularly at the European level.The way forward, in my view, is in (reformed) social insurance at the national level, coupled with a Participation Income to provide a European minimum’. 15 See Genet and Van Parijs (1992) for a clear exposition. 16 We regret not having been able to include a contribution from the United Kingdom, in which basic income has been promoted actively for many years by the broadly based Citizens’ Income Trust. However, for a comprehensive survey of the debate in the United Kingdom, see Fitzpatrick (1999). 17 See Van der Veen (1999). 18 Comparing basic income and workfare (or the shift towards activating labour policies replacing passive welfare), it is interesting to note that a basic income experiment may serve as the right counter-experiment for all kinds of workfare-oriented experiments.As argued by Peck and Theodore (2000: 124-5), the recent popularity of workfarism in the US and the UK can to a large extent be attributed to the positive results of local workfare experiments in the early 1990s.Workfare experiments show the effects of mandatory welfare-to-work programs compared to the normal treatment (e.g. the duty to apply for jobs, the duty to resume work as soon as possible) of a control group of welfare beneficiaries. Running a workfare and basic income experiment simultaneously might show what a difference it makes when recipients must participate, as a condition of income support, in programs designed to improve their insertion in paid work as under workfare, or when they can freely choose themselves what to do as under basic income. Because there are no basic income experiments going on, we can only guess what the differences would be.

38

Part One Policy Objectives

In Search of the Double-Edged Sword Paul de Beer

As is well-known, the United States and continental Europe have shown a remarkable divergence in socio-economic development the last 15 to 20 years (e.g. OECD, 1994c; Nickell, 1997). On the one hand, unemployment rates in the United States have remained quite low, at about the same level as in the 1960s and 1970s, while income disparities have risen sharply and poverty has increased. On the other hand, in most continental European countries unemployment levels mounted in the first half of the 1980s during a deep economic recession, fell only slightly in the second half of the 1980s and rose again in the first half of the 1990s, only to decrease recently. But income inequality either remained stable or increased only moderately while the poverty rate also showed no marked increase.This diverging development seems to be a perfect illustration of the famous (or should one say, notorious?) ‘trade-off between equality and efficiency’ (cf. Okun, 1975). Or, to put it in plain English, you can’t have your cake and eat it. Each country seems to be faced with a choice between either a low unemployment rate and a high poverty rate, or a high unemployment rate and a low poverty rate.1 The former outcome can be attained by a (neo-)liberal free market policy, combining low tax rates with a residual and harsh welfare state, the latter by a Keynesian policy of extensive market regulations, a generous but lax welfare state and high tax rates. If this trade-off between efficiency and equality is some kind of ‘natural’ (or economic) law, as many seem to believe, the only option is to move along a downward sloping transformation curve of efficiency and equality.This means that one has to pay for every small reduction of the unemployment rate by an increase in the poverty rate, and vice versa (see Figure 1). The position of different countries on this curve would then simply be a matter of social preferences instead of inferior or superior economic performance. However, it seems that both Americans, who worry about their high poverty rate, and Europeans, who are concerned by their massive unemployment, believe it is possible to improve their performance by either reducing poverty without simultaneously increasing unemployment, or reducing unemployment without increasing poverty.That is, they attempt to shift the trade-off curve inwards instead of moving along the curve. The American attempt to shift the trade-off curve inwards amounts to abstaining from interference with the unfettered labour market while alleviating poverty by some sort of income relief for the working poor. This is accomplished by the so-called Earned Income Tax Credit (EITC), a tax subsidy which equals a fixed percentage of selfearned income, up to a (quite moderate) income ceiling.The European approach, on the other hand, tries to change the lax and passive welfare system into more active labour market policies, without reducing the level or duration of their social benefits substantially. A range of measures are being adopted to achieve this, most of which amount to either a wage cost subsidy to employers, in order to boost low wage 41

Basic Income on the Agenda

Figure 1: The trade-off between poverty and unemployment

poverty

?

unemployment

employment, or an income incentive to the unemployed to encourage their search efforts and willingness to accept a low paid job. In this paper I will try to assess the impact of both kinds of policies and compare them with a third possible way to try to shift the trade-off curve, viz. the introduction of a basic income scheme. In their paper in this volume,Van Parijs et al. show that it is possible to shape these different measures in such a way that they are in effect equivalent as far as their direct impact on net incomes is concerned. However, it is unlikely that these measures will really be equivalent in practice. First, they will be implemented in different labour market contexts and with the different purposes of the policymakers in mind. Secondly, they will generally not be equivalent as far as the (dis)incentives for the subjects of these measures are concerned. So, despite the theoretical similarities, it seems justified to consider them as really different policies to tackle both unemployment and poverty in practice.

Short-term versus Long-Term Effects and Relative versus Absolute Poverty Measures The test for the success of these different policies is of course the expected outcome in terms of unemployment level and poverty rate. In assessing the impact of a particular policy it is important to differentiate between short-term and long-term effects. Of course, it would be possible to reduce both unemployment and poverty considerably in the short term by raising benefit levels and simultaneously creating a lot of additional jobs for the unemployed in the public sector. But it is generally believed that the huge increase in tax rates that would be necessary to finance such a policy would have detrimental effects in the long term, so the favourable short-term effects would not last. 42

In Search of the Double-Edged Sword In the following analysis, I will distinguish between three time periods. First, I will assess the initial redistributive impact of a particular measure, before economic agents have had time to react to it. Secondly, the short-term effects which manifest themselves after the agents have adjusted fully to the particular measure, will be examined. And thirdly, the focus will shift to the long-term effect which results from the interaction of the behaviour of all economic agents. A full and consistent assessment of these longterm effects is very difficult and calls for an analysis with an advanced econometric general equilibrium model. I will not attempt such an analysis and will restrict myself to a few remarks on the possible long-term effects of different policies, especially their impact on economic growth. The economic growth rate is of crucial importance with respect to a specific interpretation of the poverty level. As is well-known from the extensive literature on poverty measures, in the long run it matters greatly whether one uses a relative poverty measure or an absolute poverty measure (see, e.g., Foster, 1998).A relative poverty rate is calculated as the share of the population that has an income of less than half or two thirds, or some other percentage, of the average or median income of the population. Whether one is poor relatively speaking depends therefore on the incomes of other, non-poor people in the same population. An absolute poverty measure, however, is independent of the average or median income of the population. It counts the number of people (or the share of the population) with a real income below a certain absolute poverty threshold or poverty line.A rise or fall in the average income of the population does not affect the absolute poverty line. Since the level of both the relative and the absolute poverty line can be fixed arbitrarily, it makes no difference in the short term which poverty measure is applied (although the level of this threshold matters greatly, of course). But in the long term, using a relative measure might result in a completely different conclusion than using an absolute measure. For example, if there is a secular rise in national income (or GDP) per capita while income inequality remains constant, the relative poverty line will keep pace with the per capita income, while the absolute poverty line, by definition, will remain the same. Hence, the relative poverty rate will not change whereas the absolute poverty rate will drop sharply. In this respect, it is important to note beforehand that in the 1980s and 1990s the United States and Europe did not diverge as far as economic growth is concerned: from 1980-1996 real GDP per capita grew on average by 1.7 percent a year in the fifteen countries of the European Union and by 1.6 percent in the United States.2 This implies that, as long as income disparities remain constant, the absolute poverty rate in the US and in Europe would fall in about the same rate. For one country to perform better than the others it would have to succeed in either reducing relative income inequality or boosting economic growth.

Income Support for the Working Poor To make things as simple as possible, I start with an ‘ideal-type’ liberal market economy, with a very flexible labour market, without a statutory minimum wage (or a very low one) and an austere welfare system. I will further suppose that not only all unem43

Basic Income on the Agenda ployed people who are dependent on a social benefit are poor, both by a relative and an absolute poverty standard, but also a considerable number of the working population, whose wages are below the poverty line. This ‘neoclassical’ labour market can be regarded as a stylized version of the present-day American labour market. I will also suppose that an Earned Income Tax Credit (EITC) is introduced, which raises the net disposable income of all employed people whose wage is below or a little above the poverty line. Of course, the immediate impact of the EITC will be to lift the income of some, but probably not all, of the working poor above the poverty line. Scholz (1994) estimated that about one in every four households in the US that are eligible would be taken out of poverty by the EITC.The income of the non-working population would however not be affected by this measure.The poverty rate would therefore be slightly reduced, but only for people already employed. How would people react to the EITC? For people in a low paid job the EITC is effectively an incentive to earn more, since every self-earned dollar is, up to some limit, supplemented with 40 dollar cents.3 The main impact of this incentive would probably be to increase the number of hours that people in low-paid jobs are willing to work. This would lift an additional number of people above the poverty threshold. People in the so-called phase-out range, earning more than the income which offers the maximum EITC, would however experience a disincentive, because 21 cents of every extra dollar are withheld. This might induce them to reduce their number of working hours, although it seems unlikely that their net income would fall below the poverty line. Although the EITC widens the net income gap between non-working people and people with a low paid job, it is questionable whether this would much encourage the willingness to accept a low paid job. After all, the benefit level of the unemployed would already be so far beneath the poverty line that an extra incentive to look for a job can hardly be expected to exert much influence on their job search behaviour. The long-term effects of the EITC are hard to predict, but would probably differ not much from the short-term effects.The crucial point, as observed above, is whether the EITC would speed up or slow down economic growth. To judge the impact on economic growth one has to weigh the probable positive effects of the incentive for the low wage earners against the disincentive for the (much larger) group in the phaseout range of the EITC and the overall increase in the tax rate which would be necessary to finance the EITC. If anything, it seems likely that this would slacken economic growth a little. Since it is unlikely that the EITC would result in a substantial increase in the number of working people, there is not much compensation to be expected from that side. As far as our two poverty measures are concerned, the relative poverty rate would probably fall slightly as a consequence of the EITC, especially in the long run when the working poor with a part-time job would succeed in expanding their working hours. But, since the EITC would not alleviate the poverty of the non-working poor, it is unlikely that the relative poverty rate would fall to the level of continental European countries.The absolute poverty rate would probably also fall, but to what extent seems to depend far more on the general economic development – which is hard to predict – than on the specific impact of the EITC.

44

In Search of the Double-Edged Sword

Activating the Passive Welfare State Continental European welfare states have been quite successful in fighting poverty, but since the early 1980s they have shown a deteriorating performance regarding their unemployment level. This is often imputed to the passive character of these welfare states, which have been far more concerned with compensating income losses as a result of unemployment and illness than with the re-entry of the beneficiaries of the welfare system into employment. Recently, most European countries have switched to more activating labour market policies without seriously impairing the protection offered by their welfare state. One of the means to achieve this is to supply wage cost subsidies to employers who create (extra) low paid jobs for the unemployed.This is intended to increase the number of jobs at the lower end of the labour market but above the poverty line, and to promote the outflow of non-working poor from welfare to work. It is important to distinguish between two kinds of wage subsidies, viz. general and marginal wage subsidies.The former apply to all low-wage jobs, the latter only to extra jobs or to vacancies that are filled by (long-term) unemployed job seekers.The impact of both kinds of wage subsidies will be discussed separately. A general wage cost subsidy is in effect equivalent to a shift in payroll taxes from low wages to higher wages.This is sometimes called a ‘Robin Hood policy’ (Bovenberg, De Mooij and Van der Ploeg, 1994).The immediate impact on unemployment and poverty is, of course, nil, since this kind of wage subsidy is only effective if it induces employers to create extra jobs. The positive impact of lower payroll taxes on the demand for low skilled workers in the short run will be at least partly offset by the displacement of high skilled workers. The net effect on total employment depends on the wage cost elasticity of the demand for low-skilled labour compared to the demand elasticity for high-skilled labour. If the demand for low-skilled workers is more elastic than the demand for high-skilled workers, as seems likely from empirical research,4 the net effect on employment of an upward shift in payroll taxes will be positive. Nevertheless, this net effect is probably quite small. Besides, it does not necessarily imply a falling unemployment level, since the extra jobs might also be taken by new entrants to the labour market, e.g. school leavers and housewives.This is quite likely to happen, since the wage of the new jobs will in most cases be not much higher than the social benefit of the unemployed, who will therefore experience hardly any financial incentive to apply for these new jobs.This problem might be aggravated if many of the new jobs are part-time jobs and if meanstested subsidies and income supplements are reduced or withdrawn if one accepts a job (which causes the so-called poverty trap). One may object that this still reduces the hidden unemployment of people who are not officially registered as unemployed but are nevertheless looking for a job. However, this would not contribute much to alleviate poverty, since most of these newcomers probably belong to non-poor families. A wage cost subsidy might also have an unintended negative effect if employers lower the wage of existing jobs that are just above the income ceiling of the subsidy in order to become eligible for the wage subsidy. Although the workers in these jobs would not drop below the poverty line, the position of this low income group would nevertheless deteriorate. 45

Basic Income on the Agenda So, although a general wage cost subsidy will probably increase employment in the short run, the expected decline in unemployment is small and the impact on the poverty rate negligible. The long-term effects of a general wage subsidy might differ from the short-term effects, firstly, if the wage subsidy hampers the upward mobility of low wage earners, and secondly, if part of the wage subsidy is appropriated by the workers. The former effect is likely to occur since the cut-off point of the wage subsidy puts a very high marginal tax rate on workers whose wage rate surpasses this point.This might be a disincentive to invest in skills, something which, in the long run, also hinders productivity growth and hence economic growth. If the labour market is in equilibrium, the part of the wage subsidy that ultimately results in a rise in the net wage (instead of a reduction of wage costs) is determined by the elasticity of labour supply relative to the elasticity of labour demand. As explained above, in the short term labour supply might be quite elastic because of the influx of school leavers and housewives, so not much of the wage subsidy would be shifted to the employees. However, in the long term, this additional labour supply might dry up and the labour market might become more tight, which would increase the opportunity for workers to appropriate a (large) part of the wage subsidy. Of course this would improve their income (but they were already above the poverty line), although it would impair the positive impact of the wage subsidy on employment. To summarize, a general wage subsidy or, equivalently, an upward shift of the payroll tax, probably increases employment in low-paid jobs, although this positive effect might be slightly weakened in the long term. However, this may result in only a slight decrease in the number of unemployed and hence the poverty rate would hardly change at all. Since investment in skills is somewhat discouraged, a general wage subsidy might have a small detrimental effect on economic growth, which would increase absolute poverty a little in the long term. A marginal wage cost subsidy is only paid out for extra jobs or for hiring unemployed job-seekers.As with a general wage subsidy there is no immediate redistributive impact on poverty or unemployment, since a marginal subsidy is, by definition, only paid out if something happens first. In the short term, a marginal wage cost subsidy is of course much cheaper than an equivalent general wage cost subsidy, since only additional jobs or formerly unemployed job-finders are subsidized. So, given a particular budget, a much larger financial incentive can be given. In this respect the most efficient application of a wage subsidy seems to be the creation of additional jobs for the (long-term) unemployed in the public sector. Every Euro spent on these jobs would directly benefit the unemployed who are employed in these jobs. If the wage of these jobs is above the poverty line, this would also immediately reduce the poverty rate. Of course, creating an extra job is more expensive than paying an unemployment benefit or social assistance, so a rise of the tax burden would be necessary to balance the government budget.This tax rise might have some deleterious effects on employment, weakening the positive impact of the job creation scheme somewhat. Creating extra jobs in the public sector might also be more expensive than is at first apparent, because part of the unemployed who obtain these jobs would also have succeeded in finding a job otherwise. This so-called deadweight loss can be quite large, since in many active 46

In Search of the Double-Edged Sword labour market programs a process of ‘creaming’ takes place: the unemployed with the best chances on the regular labour market are selected to take part in the program (cf. Erhel et al., 1996). Nevertheless, in the short term it seems likely that creating additional jobs in the public sector reduces both unemployment and poverty substantially. The long-term effects are, however, far more insecure. The more successful a job creation scheme is, the more likely it is to reduce effective labour supply on the regular labour market, causing an upward pressure on wages. This hurts employment growth in the private sector in the long term.An important issue is, therefore, whether people in subsidized additional jobs move on to regular, unsubsidized jobs. If they do not, the job creation scheme amounts to a structural expansion of employment in the public sector, which has to be financed on a permanent base.The unemployment benefits or social assistance that can be ‘used’ to finance the subsidized jobs in the shortterm do, in principle, have a temporary character. So, in the long term, subsidized jobs cannot be considered ‘additional’ and the full weight of them will be added to the tax burden.This might worsen the deleterious effects on the overall employment level and on economic growth. In the long term, it is also hard to persist in discriminating between the formerly unemployed and other job seekers, so it will be tempting to open these jobs to all applicants. In effect, the question of the long-term impact of subsidized jobs boils down to the question of the economic sustainability of a structural expansion of the public sector. Since the mounting unemployment level in continental Europe in the 1980s is often – at least partly – blamed on the size of the public sector, it is rather doubtful whether in the future an expansion of the public sector could be the solution to Europe’s unemployment problem. If the expansion of the public sector and the rise of the tax burden hurts economic growth and employment in the private sector, one ends up worse than where one started from. In the end, this might result in a substantial increase in the absolute poverty rate, and perhaps even in the relative poverty rate.The warning of the poor Swedish performance in the 1990s should be taken to heart in this respect. To conclude, creating additional jobs in the public sector seems to be quite an effective means to reduce both unemployment and poverty in the short term, but the durability of these positive effects in the long term is questionable. Instead of subsidizing employers in order to promote job growth, some European countries try to induce the beneficiaries of social benefits to actively search for and/or accept available jobs. This is accomplished by some sort of income support to lowwage earners in the form of a subsidy or a tax credit.Although the exact form may differ, such income support is similar to the EITC.The impact on (un)employment and poverty might, however, be different because it is applied in a different labour market context.Whereas the EITC is meant to lift the working-poor out of poverty, in Europe income support is intended to stimulate the transition from social benefit to work by alleviating the poverty trap. With respect to this purpose it is important whether the level of income support depends on the hourly wage rate or on monthly or annual income. In the latter case, most people who receive the income support will probably have a part-time job. But many of them would not be entitled to social benefit in case they were out of work, since their (male) spouse has an income above the poverty line. So, by mainly inducing housewives to accept a part-time job, this measure will reduce neither unemployment nor poverty. If the income support is restricted to working 47

Basic Income on the Agenda people with a low hourly wage (and perhaps a full-time job), the problem might be that there are simply not enough jobs for which the unemployed are qualified. At the same time the unemployed who receive a social benefit will not have any incentive to accept a part-time job, of which a larger number might be available. In the long run – as neoclassical theory predicts – there is probably not much difference between the impact of a subsidy to employers and a subsidy to employees, so it is likely that the aforementioned effects of a general wage subsidy also apply to an income-support measure. To summarize the preceding sections, softening the harsh effects of a liberal free market economy by way of a tax credit for the working poor will probably reduce the poverty rate somewhat, without worsening the unemployment rate, both in the short term and long term. It might therefore result in an inward shift of the trade-off curve, but this effect will probably be quite small.The non-working poor will find no relief, since income support is targeted to the working-poor and the chances of finding a job are not improved. Adding general wage cost subsidies for employers to the generous and passive European welfare state will probably increase employment, but hardly reduce (formal) unemployment and the poverty rate. In the long term these effects will be even smaller than in the short term. A job creation program in the public sector seems to be far more effective in the short term and may reduce both unemployment and poverty substantially.The long-term effects however are dubious and might well result in a serious deterioration of both the unemployment rate and the poverty rate (the absolute rate in particular). Income support to people with a low-paid job might stimulate labour market participation of housewives, but will probably not result in a lower unemployment or poverty rate.

Basic Income: The Double-Edged Sword? Although each of the policies mentioned above will very likely ameliorate the unemployment-poverty trade-off slightly, either in the short term or in the long term, none of them can really be considered a double-edged sword, that is, an instrument that cuts both unemployment and poverty at the same time and is sustainable in the long term. Hence, one might wonder whether perhaps an entirely different instrument, namely a basic income scheme, turns out to be such a double-edged sword after all. To assess the impact of a basic income it is, of course, very important to specify what kind of basic income one has in mind. Because I am looking for some kind of ‘golden mean’ between the American and the European model, it seems reasonable to start with a basic income which is comparable to the level of social assistance in the United States.5 In fact, I will simply assume that this kind of basic income is introduced in a liberal free market economy.This means that the basic income will replace all existing social benefits (except wage-related unemployment and disability insurance), but will be added to the gross income (wage, profit) of all other people. To finance this basicincome scheme the income tax rate will have to be raised by about 10 to 15 percentpoints.6 48

In Search of the Double-Edged Sword In contrast with the aforementioned policies, a basic income is not targeted at the groups that are most in need of assistance, such as the poor and the long-term unemployed. Accordingly, a basic income is often charged for being inefficient: the bulk of basic incomes is paid out to people who are neither poor nor unemployed, so there is a very large deadweight loss. However, if the basic income is implemented as a negative income tax, most people will not actually receive a basic income since it will be cancelled out by the rise in the marginal tax rate. So, in effect, a basic income is also targeted at the low-income groups.A basic income is not, however, tagged to employment status: people with and without a job are treated alike.This has the advantage that a basic income can act as a wage cost subsidy and an income supplement at the same time. Besides, there is no danger of displacement effects, since both employed and unemployed, low-skilled and high-skilled people receive the basic income. The immediate redistributive effect of the introduction of a basic income will be a substantial increase in the net income of people with a low wage and people with (at present) no income, mainly non-working spouses whose husbands earn an income. The higher-income groups will of course lose some money, since for them the rise in the tax rate exceeds the basic income.The immediate impact on the poverty rate will be similar to that of the EITC, for at first, only the working poor profit from the basic income. But since the basic income level is higher than the EITC, the magnitude of the initial effect on poverty will be larger. Nevertheless, due to the fact that the basic income is, by assumption, much lower than the present social-benefit level in Europe, the poverty rate among the non-working population will be much higher than in present-day Europe. In the short term a basic income will evoke different reactions as compared to the EITC. A basic income does not incite employed people to supply more working hours, as the EITC does. On the contrary, both the income effect (of the basic income itself) and the substitution effect (of the higher tax rate) will encourage workers with a low-paid job to reduce their working hours, although it is unlikely that this will cause them to fall below the poverty line. However, a basic income does work as an incentive for the unemployed to try to find a part-time job.After all, the extra income is not deducted from the basic income, so every self-earned dollar or Euro, net of income tax, will be added to disposable income. It is quite probable therefore, that many social-benefit recipients will try to supplement their basic income with a part-time job. On an unfettered and flexible labour market many of them may indeed succeed in getting a small job (especially if many employed people want to reduce their working hours simultaneously), thereby raising their income just above the poverty line. The poverty rate might then drop appreciably. Nevertheless, non-working people who do not succeed in getting a job and who have to live on only their basic income, will stay below the poverty line. The financing of the basic-income scheme might hurt employment if the tax rise is shifted onto employers, and wage costs rise. However, both theoretical and empirical evidence suggests that a rise in the marginal tax rate might compensate the upward pressure on wages of a higher average tax rate (e.g. Koskela and Vilmunen, 1996; Gelauff and Graafland, 1994). This is explained by the fact that a higher marginal tax rate reduces the net worth for employees of a given rise of the wage cost for the employer. So, relatively speaking, it is cheaper to shift priorities from a wage raise to, e.g., a reduc49

Basic Income on the Agenda tion in working hours or an increase in employment. Hence, it is quite likely that a simultaneous rise in both the average and the marginal tax-rate, will not exert a large negative influence on the demand for labour. If a basic income turns out to be a stimulus for shorter working hours, there might even be a substantial increase in the number of jobs as a result of worksharing. In the long term, the crucial point, as before, is whether a basic income will exert a substantial influence on the structural rate of economic growth. Again, this is hard to predict. If the rise in the tax rate, needed to finance a basic income, does not put an upward pressure on wages, there is probably no negative impact on capital investment and hence on productive capacity. As far as investment in human capital is concerned, the negative effect of a higher (marginal) tax rate on the return on human capital may be counterbalanced by the fact that investment costs are lower. The net effect of a basic-income scheme on human capital might, therefore, be quite small, perhaps even negligible. So there does not seem to be a decisive argument that a basic-income scheme will hurt economic growth. The long-term impact of a basic income on unemployment and poverty may not differ much from the short-term effects.The most important question is which share of the non-working population will be able to find a (part-time) job in the long term. Since there is no impediment to a completely flexible labour market under a basic-income scheme, it might turn out that, ultimately, the greater part of the non-working population will succeed in finding a (small) job and hence climb above the poverty line.

A Trade-Off between Short-Term and Long-Term Prospects The foregoing discussion of three different strategies to shift the trade-off curve inwardly by simultaneously reducing poverty and unemployment, does not end with a definite winner.The reason is that there is not only a trade-off between unemployment and poverty, but also a trade-off between the short-term and the long-term impact of labour market policies. If one restricts attention to the short term, there certainly is a clear winner: creating a large number of extra jobs in the public sector for the (long-term) unemployed reduces both the unemployment rate and the poverty rate considerably in the short and medium term. But there is a serious risk that this kind of policy will turn out to be the worst alternative in the long run. Compared to the extension of the public sector, the introduction of a tax credit for low paid workers seems quite innocuous. However, both in the short term and the long term the positive effects are probably quite modest, because the EITC only affects the income of the working poor, who in Europe make up only a small portion of the poor. A basic income-scheme will initially fall short compared to an extension of public sector employment, since its level will inevitably be below the minimum social-benefit level in Europe. Nevertheless, its short-term impact on the poverty rate will be favourable compared to the EITC, since many non-working people who succeed in finding a part-time job will profit from a basic income. In the long term, the crucial issue is the impact of a basic income on the economic growth rate.The economic sustainability of a basic income is subject to much debate. Definitely, many economists are convinced that a substantial basic income scheme is not feasible in the long term. 50

In Search of the Double-Edged Sword Nevertheless, it might also be argued that the abolition of the poverty trap and the even spread of the marginal tax-rate over all income groups would be quite favourable for employment growth and for upward mobility of low income groups. This might offset the negative effects of the high average tax-rate under a basic-income scheme. However, even if my optimism is justified that a basic-income scheme comes closest to a double-edged sword in fighting both unemployment and poverty in the long term, the trouble is that it is certainly not the most effective means in the short term.This fact might explain why it is so hard to find popular and political support for introducing a basic-income scheme. Most people – and certainly most politicians – are not very interested in the possible favourable effects in five or ten years’ time. And it is, indeed, quite harsh to ask the present poor to wait another ten years until they can be drawn out of poverty. So, even if one is convinced of the favourable effects of a basic income in the long term, it might still be wise to propose a basic income disguised as a program that favours the poor and reduces unemployment in the short term. Fortunately, as Van Parijs et al. show in their contribution to this volume, different policy measures can be shaped in such a way that they are, in fact, equivalent. Hence, it seems perfectly possible to start, e.g., with an EITC and gradually transform it into a real basic income. In other words, perhaps one should first forge some ‘ordinary’ single-edged swords and later on try to fuse them into the double-edged Excalibur of a basic-income scheme.

Notes 1

2

3

4 5

The United Kingdom seemed, at least until quite recently, to perform particularly badly by combining the worst of both worlds: sharply increasing poverty and a high unemployment rate at the same time. The Scandinavian countries, however, seemed to have escaped the trade-off in the 1980s by keeping both unemployment levels and poverty rates quite low. The demise of the Swedish model in the 1990s, however, has cast doubt on the sustainability of this kind of policy in the long run. Although economic growth (change in real GDP per year) was higher in the US (2.5 percent) than in the EU (2.0 percent), this was fully compensated for by the higher population growth in the US (1.0 percent) as compared to Europe (0.3 percent, excluding the effect of the unification of Germany). Source: OECD (1995a and 1997). This applies from 1996 on only for households with at least two qualifying children. For households with one child and with no qualifying child the credit rate of the EITC is 34 percent and 7.65 percent respectively. Source: Scholz (1994). Hammermesh (1993: 117), considering a number of empirical studies, states that,‘... it seems fairly safe to conclude that additonal education (....) reduces the labor-demand elasticity.’ That is, the Temporary Assistance for Needy Families (TANF), which has replaced the Aid to Families with Dependent Children (AFDC). If one should choose a basic-income level comparable to the level of social benefits in Europe, it is clear that in the short run this would reduce the poverty rate substantially, since the social benefit recipients would receive at least as much as at present, while all people without income and most workers with a low income would get a higher income. The economic sustainability of such a generous and therefore very expensive basic-income scheme is, however, very shaky.The discussion would therefore be reduced to the question whether such a high basic income can be sustained in the long run.

51

Basic Income on the Agenda 6

52

Suppose that both the social benefit level and the basic income level amount to a fraction f of the average net labour income. Denote the share of the adult population that is employed by e and the share that is dependent on social benefit by s.Then, under the present system, the tax burden t on the employed population that is needed to finance the social benefits, can be calculated from: te=fs(1-t). Hence t=fs/(e+fs). Now, suppose a basic-income scheme is introduced in the form of a negative income tax. For employed people the basic income will then be deducted from their income tax and the basic income for people on social benefit will replace their social benefit.Then only the basic income for non-working people without social benefit is added to the government outlays. So the new tax rate t’ can be calculated from: t’e= f (1-e)(1-t’). Hence t’= f (1-e)/(e+ f (1-e)). Suppose both the social benefit and the basic income are equal to 40 percent of the average net labour income (f =0.4), half of the population is employed (e=0.5) and a quarter is receiving social benefit (s=0.25).Then, under the present system, the tax burden needed to finance the social benefits is 17 percent, while a tax burden of 29 percent is needed to finance a basic-income scheme, hence an increase by 12 percent-point. Similar changes in the tax rate result if other realistic figures are used.

Basic Income and its Cognates Partial Basic Income versus Earned Income Tax Credit and Reductions of Social Security Contributions as Alternative Ways of Addressing the ‘New Social Question’ Philippe Van Parijs, Laurence Jacquet and Claudio Caesar Salinas*

At least three types of transfer policies have been advocated as ways of fighting unemployment without worsening poverty: reductions of social security contributions, the earned income tax credit, and an unconditional basic income (possibly in the form of a negative income tax). Many proponents of some of these are passionate opponents of some of the others.Yet, in the institutional context of much of today’s Western Europe and within some broad limits, each of them can, in principle, be so calibrated as to generate exactly the same profile of tax rates and the same pattern of disposable incomes as the corresponding variant of each of the other two. So, despite all the heat, is there nothing to choose between? Not quite. For beyond the prima facie equivalence, a closer look reveals important, sometimes unexpected differences between the three policies. These differences concern, for example, the impact of the policies on the unemployment trap, on the development of self-employment and informal work, on the accumulation of human capital, on workers’ bargaining power, on Trade Union attitude and on political feasibility. Using a battery of simple graphs, the article spells out these differences and explores their significance.

Introduction: Full Employment without Poverty Europe’s new social question can be said to consist in an emerging dilemma between high unemployment and worsening poverty.This new social question arguably stems from the very solution that had been gradually developed, over the last century, for the old social question, the unsustainable inequality between capitalists and proletarians generated by the industrialization process.Along with public involvement in the accumulation of physical and human capital, the furthering of workers’ rights – most prominently in the form of collective bargaining, labour legislation and social insurance schemes – gradually turned full jobs from unavoidable toil into valuable assets. The resulting pattern of distribution seemed to be on the right track as long as * This paper was written within the framework of project PAI P4/32 ‘The new social question’ (Belgian Federal Government, Prime Minister’s Office, Federal Office for Scientific, Technical and Cultural Affairs). Earlier versions were presented at the Harvard Centre for European Studies (29 April 1998) the Oxford Centre for the Environment, Ethics and Society (14 May 1998) and the VIIth Congress of the Basic Income European Network (Amsterdam, September 1999). Useful comments by Anne Alstott, the editors of this volume and all members of the PAI project are gratefully acknowledged.

53

Basic Income on the Agenda access to a job and the entitlements associated with it – including, for example, extra child benefits in periods of involuntary unemployment or the right to a decent pension for a worker’s widow without any employment record – were secured to the overwhelming majority of households. For a variety of interacting reasons – including technical change, European economic integration, so-called globalization, privatizations, marital instability, educational homogamy, etc. – such a broad coverage can no longer be taken for granted.1 This is why a new social question arises. It consists in a growing proportion of households proving unable to secure access to adequate job assets, with cumulative consequences for both the monetary and non-monetary aspects of the welfare of all their members.This new social question may therefore be said to generate a new class divide, based on job endowments, characteristic of welfare-state capitalism and crucially different from the old class divide, itself based on the ownership of material means of production and rooted in industrial capitalism’s old social question.2 To investigate how the new social question might be solved, one first needs to further specify the underlying diagnosis. Under developed welfare-state capitalism, income redistribution within the population of working age operates mainly through the taxation of labour income (personal income tax, employers’ and workers’ social security contributions) and the distribution of the proceeds to people who are involuntarily unemployed (unemployment compensation, means-tested minimum income guarantee, disability allowances). However effective at reducing income poverty, this pattern of redistribution, if sufficiently developed, displays a strong tendency to generate persistently high levels of unemployment, as a growing number of households becomes unable, for the reasons listed above, to durably achieve through their labour a net income that exceeds the level of social protection.To reduce unemployment without worsening poverty – and to thereby tackle the dilemma in which the new social question expresses itself – one therefore needs to challenge the (near-) exclusive focus of redistribution on the involuntarily unemployed. In other words, one needs to channel more of the explicit or implicit transfers (1) towards low-paid workers (in-work benefits), and/or (2) towards people who choose to stop working or to work less (chosen-time subsidies). This distinction can help us map the landscape of policy proposals currently under discussion, and thereby structure the key socio-economic debate among people with both a sense of justice and a sense of reality.3 The introduction or expansion of in-work benefits aims to increase the total volume of employment by boosting, sometimes indistinguishably, either the demand for labour (the number of low-paid jobs that are profitable for potential employers) or the supply of labour (the number of low-paid jobs that are acceptable for potential employees). They can take the form of (a) reductions of employers’ social security contributions; (b) reductions of workers’ social security contributions; (c) direct employment subsidies or tax credits (or job voucher reimbursements) to firms on account of the number of workers they employ; (d) direct earnings subsidies or refundable tax credits restricted to workers; or (e) employment-motivated subsidies to public sector jobs. These reductions or transfers may be allocated either (1) only to low-paid jobs, or (2) indiscriminately to all jobs (but with a net benefit to low-paid jobs only, owing to the way the scheme is funded), or (3) only to jobs with a characteristic correlated with low pay (such as having an incumbent with a low level of education or a long period of unemployment). 54

Basic Income and its Cognates Unlike in-work benefits, chosen-time subsidies do not purport to boost the total volume of (profitable and acceptable) employment, but to distribute it differently. They include (a) compensation for voluntary early retirement (whether part-time or fulltime), (b) compensation for voluntary career interruption (whether part-time or fulltime, short-term or long-term, restricted to specific reasons, such as parental or educational leave, or unrestricted) and (c) reductions of income tax or (employers’ or workers’) social security contributions for those who choose, individually or collectively, to reduce their working time. These two types of policies seem to work in opposite directions. In-work benefits can be viewed as addressing the unemployment trap: they try to make it easier, more attractive, less costly, at least for those with a low earning power, to increase their paid working time by accepting low-productivity jobs. By making them acceptable, they aim to bring them into existence: the unemployment trap is also relevant to the demand side. Instead, chosen-time subsidies can be viewed as addressing the employment trap: they try to make it easier, more attractive, less costly, at least for those with a low earning power, to decrease their paid working time. Nonetheless, some policy proposals belong to both types. This is the case for any form of general lump-sum transfer or basic income scheme (henceforth BI), which provides transfers to all adults, whether or not in paid employment, either unconditionally (citizen’s income) or subject to their making some contribution in a sense that extends significantly beyond full-time waged employment (participation income): the narrower the interpretation of the contribution condition, the weaker the chosen-time-subsidy character of such proposals and the greater their in-work-benefit nature. Basic income schemes in this sense may or may not be integrated with a universal child benefit system. They may be sufficient to cover basic needs (full basic income) or they may not (partial basic income).They may take the form of ex ante payments to all (in the social dividend version) or rather of refundable income tax credits (in the negative income tax version, or NIT, whether of the linear or non-linear variety). Finally, the per capita level of the payment may be affected or not by the composition of the household.4 In-work-benefit policies tell the unemployed that they can keep some transfer if they work. Chosen-time-subsidy policies tell the employed that they can have some transfer if they work less. BI policies tell both the unemployed and the employed that they can have an (explicit or implicit) transfer whatever they do.

A Puzzling Equivalence In this light, the introduction of an unconditional BI can be viewed as the simplest and most radical among a whole range of proposals that attempt to tackle unemployment without worsening poverty, through changing the pattern of income redistribution under welfare-state capitalism.5 The purpose of this chapter is to contribute to the identification of the main differences between the consequences one can expect from the introduction of an unconditional BI6 and from its two most serious contenders, qua strategies for tackling the core of the new social question, namely the earned income tax credit, currently most popular in the US and the UK, and reductions in the social security contributions on low-paid work, currently most popular in continental countries. Identifying these differences with the help of the most appropriate analyti55

Basic Income on the Agenda Graph G1 Current situation K

Net Income

t’

45º

F D

M

E t = Rate social security contributions t’ = Rate of income taxation

t

O

A

B

C

Gross Income

cal tools and on the basis of the most relevant available evidence is essential to feed a well-informed and dispassionate debate on these issues.7 But, as we shall see, it is no straightforward matter, for much of the standard modelling of a BI and its alternatives is simply blind to the crucial differences between them. A stylized picture In order to bring these differences into focus, we shall start by comparing, on the background of a highly simplified picture of the present situation in continental Europe, the impacts of the three policies on disposable income and material work incentives. To represent in the simplest possible way the (purely arithmetic) relationship between net and gross income under the present tax and benefit system, we initially assume that all gross income is waged income earned in the formal sector and that each household consists of a single adult. Graph G1 offers a stylized picture of the lower range of the income distribution under this assumption.The 45° line represents what the situation would be in case there were no taxation and households would therefore receive the full amount of their gross earnings. Social security contributions are assumed to be paid at source at a proportional rate (t), starting from the very first Euro earned (O). Income taxation proper is also proportional (at rate t’), but only kicks in at a higher level (C): any gross income lower than OC is tax-exempt. Moreover, there exists a means-tested guaranteed minimum income scheme (henceforth GMI), in the form of transfers making up the difference between the net earnings of any household and some chosen minimum level (OM). In graph G1, the lightly shaded area represents the social security contributions. The darker area represents the linear income tax.Triangle MEO corresponds to the means56

Basic Income and its Cognates

Graph G2 Net average tax rates of taxation %

t+t” t+ t’ t

O

A

B

C

Gross income

tested GMI payments to households with pre-transfer incomes comprised between 0 and AB. Finally, the thicker line shows the resulting relation between net (post-taxand-transfer) and gross (pre-tax-and-transfer) income. In graph G2, the solid line shows the corresponding profile of the net average tax rate, defined as the ratio of net taxation (tax plus social security contributions minus benefits) to gross income. In graph G3, the solid line shows the corresponding profile of the effective marginal tax rate, defined as the ratio of the increase in net taxation (increase in tax and/or social securi57

Basic Income on the Agenda Graph G3 Marginal effective rates of taxation %

100

t+t” t+t’ t+t * t O

A

B

C

Gross income

ty payments plus decrease in benefits) to the increase in gross income which prompts it. Graphs G1 and G3 highlight the crucial fact that households with (potential) gross incomes lower than OB are stuck in the unemployment trap.The profile of net income is horizontal in this range, and the effective rate of taxation is 100 percent. If they are unable to earn a gross income higher than OB, households have no financial incentive whatever to earn anything through declared work and are in this sense trapped in a nowork situation. Nor do firms have any incentive to create any such job, even when they are not legally prevented from doing so, as they would find it hard attracting and retaining suitably motivated workers. As characterized above, the new social question can be very simply portrayed in our stylized graph: it consists in a steady increase in the proportion of households whose earning power falls in the OB range, as a result of the interaction of technological, economic and social factors. It is not obvious that such a 100 percent effective rate of taxation on the lowest earnings should be regarded as unfair and urgently abolished.To start with, this highly regressive profile of marginal rates (G3) is fully consistent with an even more highly progressive profile of average rates (G2): those who lose most marginally are also those who gain most overall.8 Moreover, under assumptions commonly made in the optimal taxation literature, the sustainable maximization of the lowest incomes requires the highest marginal rates to affect the lower layers of earned income.9 However, the observation of a steady upward trend in the numbers of the long-term unemployed or of the beneficiaries of GMI schemes has fed a strong suspicion that the lower (OB) range of potential earnings is getting more crowded. And this is a source of concern – precisely the very concern which defines the new social question –, in part because of the direct and indirect impact of this phenomenon on the sustainable maximization of the lowest incomes, but also because exclusion from paid work arguably matters as such, not only as a cause of low income.11 To deal with the problem, our stylized graphs provide 58

Basic Income and its Cognates unambiguous guidance: one must find a way of reducing the unemployment trap, i.e. of lengthening left-ward the lower part of the U-shaped curve of marginal rates (G3), so that it encompasses a larger share of the active population. Reduction of social security contributions (RSSC) One striking feature of the situation depicted in G1 is that some households with gross earnings higher than the income guarantee (OM) nonetheless fall into the trap because of social security contributions pushing their net earnings below OM.11 One obvious suggestion, depicted in G4, is to scrap social security contributions on all earnings below the minimum income (OA = OM), while lowering them to a decreasing degree in the next range (say, between A and C, the point from which income tax starts being paid) and collecting whatever is needed to achieve budget neutrality – in a purely arithmetic sense, i.e. abstracting for the moment from any behavioural effect – by raising the rate of taxation (from t’ to t’’) in the upper range (beyond C).The size of this adjustment will (arithmetically) depend on the number (and distribution) of households in range OC. Because of the unemployment trap mechanism, one has every reason to expect that the range OB will be practically empty.The required net funding should therefore reduce to what is needed to finance the relatively small discount on the social security contributions of households in the BC range. In graph G4, the lightly shaded areas represent the amount by which social security contributions are being reduced and the matching increase in income taxation. The triangle ODF represents the part of the social security contributions that has been scrapped. The darkly shaded area shows the amounts still payable as contributions. People with a gross income higher than C will pay exactly the same amount as before in social security contributions, even though the lower part of their earnings (OA) has also been exonerated.The reason is that the gradual phasing out of the reduction in the AC range amounts to imposing, compared to the initial situation, a far higher rate of social security contribution on each Euro earned in this range.This is reflected in the flattening of the slope of DF, relative to that of EF (45°–t), which has been calibrated to make the reform exactly neutral for someone earning gross income OC. The thick line in G4 shows the resulting relation between net and gross income. The dotted line in G2 shows the corresponding profile of the net average rate of taxation (inclusive of social security contributions and benefits). The dotted line in G3 shows the corresponding profile of the effective marginal tax rate.When compared to the thick lines in G1 and G3, which represent the current situation, this highlights the crucial fact that the unemployment trap, though far from abolished, has been reduced from OB to OA. Barring a reduction in the GMI level M, this reduction could only be achieved at the expense of a significantly increased marginal rate in the BC range and a milder increase in the marginal rate on higher incomes (see G3). In other words, if reductions of social security contributions are to significantly alleviate the unemployment trap without worsening poverty, it is (nearly) an arithmetic necessity that each Euro earned by some of the less well paid among the current workers will be effectively taxed (inclusive of social security contributions) at a significantly higher rate than before.12 Yet, by no means does it follow that they can be regarded as the victims of the operation. Quite to the contrary, as graph G2 shows, they are the main net beneficiaries of the reform in terms of net average tax rates, and hence disposable incomes. 59

Basic Income on the Agenda Graph G4 Reductions of social security contributions Net Income

K

t” 45º F

G D

M

E t = Rate social security contributions t’’ = Rate of income taxation

t

O

A

B

C

Gross Income

Earned income tax credit (EITC) Let us now turn to something (apparently) altogether different. First introduced under the Ford administration, massively expanded under the first Clinton administration, the so-called Earned Income Tax Credit (EITC) has now displaced Aid to Families with Dependent Children (AFDC) as the USA’s main federal transfer programme.13 The central idea of the programme is quite simple. In order to alleviate the poverty of poor workers without damaging their work incentives, let us supplement their earnings with a refundable tax credit the level of which grows proportionally with the level of earnings up to a point at which it stabilizes, before being gradually phased out. EITC itself does not give anything to households who do not work at all. In Europe at any rate, it could not therefore be considered as a complete alternative to the existing income support system. But EITC makes sense as a way of improving work incentives and reducing the unemployment trap even if the existing GMI is maintained.14 To understand this, let us take again as our point of departure a stylized picture of the current situation (graph G1). On this background (with social security contributions collected at source), we assume EITC to come into operation by giving all working households with gross incomes not exceeding OA (if there are any) a refundable tax credit that rises proportionally to their earnings. We want the chosen variant of EITC to be as comparable as possible to the RSSC scheme explored above, and therefore calibrate it in such a way that this rise occurs at a (constant) rate exactly equal to the rate (t) at which social security contributions are levied (graph G5). Beyond level OA of gross earnings, the tax credit is gradually phased out until it disappears altogether for gross earnings in excess of OC. Beyond C, income tax is paid at a somewhat higher rate in order to finance the cost of the reform.This cost is the sum of all the tax credits conceded to earners in the OC range (triangle DOF). But it essentially reduces 60

Basic Income and its Cognates

Graph G5 The Earned Income Tax Credit K

Net Income

t” 45º Q F

G S

M

D E t = Rate of social security contributions

t”=Rate of income taxation

t

O

R

A

W B

C

Gross Income

(arithmetically speaking) to the set of (shrinking) tax credits in the BC range (triangle DEF). For prior to the reform, there was no financial incentive to earn anything in the OB range, and the latter can therefore be expected to be practically empty. One may wonder how EITC’s positive effect on work incentives can survive its being combined with a GMI scheme. For in the range in which the tax credit grows with each Euro of earning (OA), the effective marginal rate remains 100 percent. The whole point, it would seem, is lost. But this inference is incorrect. For the tax credit on the lower layers of income does not only affect the pre-GMI incomes of those whose gross incomes entirely lie in range OA. Its key effect is that it makes it easier to reach a level of gross income (now lowered from OB to OA) above which working starts making financial sense. This can again be clearly seen by examining, on graph G5, the profile of the thick line that depicts the relationship between gross and net income after the reform.This profile, it turns out, is exactly the same as in the case of RSSC (graph G4), and so are the profiles of both average and marginal tax rates (G2 and G3). Because of the specific rates deliberately chosen for the tax credit and its phasing out, the effects of EITC and RSSC coincide exactly in the range OC. And the identity also holds in the range above C, since the net cost of modifying the profile of disposable income in the OC range is obviously the same in both cases and must therefore be reflected in an identical increase in the linear tax rate on the highest incomes. In both cases, therefore, the unemployment trap is reduced by the distance AB, the net (arithmetic) beneficiaries are in the AC range, and marginal rates increase on all gross incomes in excess of OB.

61

Basic Income on the Agenda Partial basic income (PBI) In order to get a large number of households out of the unemployment trap, another, again apparently quite different, idea is to give each of them a basic income (BI), or taxfree income floor, which they can keep whatever they earn.We might think of introducing this BI at level OM, which would enable us to get rid of the current GMI and the associated trap altogether. But let us suppose, for the time being, that we want to proceed cautiously and decide to start with a partial basic income (or PBI) at some lower level OP. In order to make comparison with the previous schemes as convenient as possible, let us simply choose P as the intersection of the DF line and the vertical axis.15 Handing out an income to all, instead of only to the poor, is of course a hugely expensive operation, which we assume to be funded by a massive linear increase (by t*) in the income tax as from the first Euro earned. Existing social security contributions are untouched, and the GMI is kept at the same level OM, with means-tested payments reduced by OP as a result of every household’s means being increased by the amount of the PBI. This reform is depicted in graph G6.A PBI is given to each household at level OP.This has the immediate consequence of shifting the 45° line upwards, as every household’s pre-tax income rises by amount OP. The lightly shaded area below the new 45° line represents the proportional social security contributions, the rate of which remains unchanged (t).Taxation proper, by contrast, needs to be massively expanded to fund the huge new outlays on the PBI.To secure this funding, a proportional tax is now raised from the first Euro earned, at a rate t* chosen in such a way that someone with a pretax-pre-transfer-income of OC ends up, as in the initial situation, with a net income of CF. In G6, the effect of this additional tax on net income is depicted by a straight line PF passing exactly through D because of the specific choice made for the level of OP. Whatever outlay is not paid for in this way has to be funded by an adjustment in the tax rate that applies beyond C. By how much taxation needs to rise in this range is given by the total cost of the PBI minus the taxes now raised in the OC range and the means-tested payments now no longer made in the AB range.16 The total cost of the new tax required to finance the PBI (minus the savings from the reduced cost of the means-tested GMI) is then represented by the massive expansion of the darkly shaded area, relative to what it was in the initial situation (graph G1).17 Once again, we can bring out the consequences of the reform by focusing on the resulting profiles of disposable incomes, average and marginal tax rates. Despite the very different intellectual gymnastics we have just gone through, we end up with the very same three after-reform profiles as in graphs G2 to G4. First, the level of OP and the implied linear tax rate were chosen in such a way as to achieve the same profile of disposable income in the OC range as with the previous two policies. Second, precisely because of these identical profiles in the OC range, the net cost to be picked up by increased taxation beyond C – the complex net result of the introduction of PBI payments, the increase of taxation and the decrease of means-tested payments in the OC range – is also necessarily identical to what it is under the other two policies. Hence, despite the massively larger tax revenues required under this third policy, the economically relevant marginal rates and the politically sensitive disposable incomes are exactly equal, at every level of gross income, under all three schemes. 62

Basic Income and its Cognates Graph G6 Partial Basic Income

45º

Net Income

45º

45º *

t

K

Y

t”

t*

t G M

X

D

t P

O

F

T E H

I

Z

V

U

A

B

C

Gross Income

The challenge This simple exercise in graph-assisted elementary arithmetic yields the challenge that constitutes our point of departure. We have explored three tools for alleviating the unemployment trap which are commonly associated with very different approaches and certainly sound extremely different: scrapping social security contributions on low wages by exonerating the lower layer of everyone’s earnings, introducing a refundable tax credit reserved for the working poor, and paying all households, rich and poor, a uniform grant.Yet, in each case, we ended up with an identical net impact: all three tools reduced the unemployment trap to exactly the same extent; all three tools modified the distribution of disposable incomes in exactly the same way; all three tools generated an identical pattern of marginal taxation. Of course, this arithmetic exercise was rough and simplistic: it started from an oversimplified picture of the existing complex system of social insurance and assistance; it did not use any precise estimate of the distribution of households across the various ranges of earnings; and it did not incorporate any conjecture about the way in which economic agents would react to the new incentive structure. Bringing in each of these important considerations would of course affect the outcomes. But for the purpose of formulating our challenge, this does not matter in the least. For the changes brought about by using the three tools, as depicted in our graphs, are strictly equivalent, and can therefore be expected to be affected symmetrically by the introduction of each of the considerations just mentioned. In particular, it can no doubt be pointed out that it is only when the behavioural effects of the various reforms are taken into account – as 63

Basic Income on the Agenda households and firms adjust to the new pay-off structure – that the differences between the various policies can come to light. But how could any such difference emerge if the new payoff structures are identical in all three cases? Indeed, when scrutinized closely, most, if not all, existing attempts to formally model the behavioural impact of any of the three policies (in a context in which a GMI is and remains in place) would apply just as much to the corresponding variant of either of the other two policies. Such a formal equivalence constitutes at least prima facie a disturbing challenge for anyone who embraces (sometimes passionately) any one of the three policy proposals, while rejecting (sometimes no less passionately) some of the others. It might be tempting to object that the whole exercise from which the challenge emerged was rigged, since in each case levels and rates were selected so as to generate the equivalence. Some actual proposals of each type are quite remote to the particular one we have considered, and they may therefore generate very different effects.Actual RSSC proposals, for example, tend to recommend that one should scale down contributions on the lower layers of earnings without scrapping them altogether. EITC, as implemented in the United States, differs from the variant selected above because the level of the tax credit, after growing proportionally to earnings as these increase from zero, and before being gradually phased out, remains constant over some range. Some PBI proposals rely at least in part on the taxation of land, energy or consumption, rather than exclusively on an adjustment of the income tax. EITC as it exists and most NIT proposals make the amount of the benefit crucially dependent on the composition of the household. Countless more variations of this kind could be mentioned, sometimes implying massive departures from the profiles of marginal and average tax rates depicted in graphs G2 and G3. Such variations matter a great deal if the effects of specific proposals are to be assessed. But they are irrelevant to the challenge itself, at least within the broad limits within which the parameters of each of the three proposals can be adjusted to preserve the equivalence. What about a full basic income? It must, however, be acknowledged that there are such limits. In particular, the equivalence claim becomes rather far-fetched as one moves from the sort of substantial but partial BI depicted in G6 to a full BI (at level OM), one that could entirely replace the existing GMI without increasing poverty and would enable households to increase their net income as from the first Euro of gross earnings. True, the same net relation between gross and net income could still be achieved through a RSSC that would allow for negative social security contributions, i.e. net employment subsidies, in the OA range. It could also be achieved through an EITC with refundable tax credits making net income jumping all the way to OM as from the first Euro of earning. But these would hardly be natural ways of expanding either RSSC or EITC.18 Advocates of a full BI may therefore feel immune from the threat of the equivalence result. Except in a strained, administratively cumbersome way, what they propose and want cannot be achieved through some version of RSSC or EITC. This is true, but does not make the equivalence irrelevant, even to them. For as a measure designed for immediate implementation, there are two strong reasons for focusing on a partial rather than a full BI. One is that the sudden introduction of an individual BI at the current level of GMI for single people (inclusive of housing allowances, if such exist) would unavoidably create a massive upheaval in the distribu64

Basic Income and its Cognates tion of disposable income between various types of households, in particular at the expense of one-adult households and in favour of two- or more-adult households.The infliction of serious and sudden losses on a broad set of households is not desirable, not only for reasons of political feasibility, but also for reasons of intrinsic fairness which this is not the place to spell out.19 While decisive for short-term proposals of a strictly individual full BI, this objection is much weaker in the case of household-based variants of BI, which are able to mimic the current dependence of GMI levels on household composition and thereby to avoid these massive shifts.20 Compared with this strategy, however, a strictly individual but partial BI, combined with a residual household-based and means-tested GMI, has the symbolic advantage of asserting more clearly that this is not merely improved social assistance but the expression of equal citizenship, and it has above all the crucial practical advantage of making entitlements independent of living arrangements for the bulk of the population, which makes it both administratively cheaper and less intrusive of people’s privacy. Secondly, both an individual and a household-based full BI would naturally be designed to enable households to improve their net incomes as from the first Euro of gross earnings. However, compared to a PBI strategy, this would have the unavoidable consequence of increasing marginal rates of taxation higher up, in the much more crowded middle range of the distribution of potential gross earnings, and therefore of having an impact on economic activity (via weaker incentives to train, seek promotion, etc.) that is likely to be more negative than maintaining an unemployment trap for the least productive, at the very bottom of the distribution of earnings.21 This is no decisive argument at all against moving towards a full BI, especially as both the average gross earnings and their dispersion of gross earnings can be expected to keep growing. But it is definitely an argument for adopting a cautious approach, one which reduces significantly the depth of the trap without getting rid of it in one swoop. Combined with a residual GMI, an individual and universal PBI would do precisely that.22 In this light, the equivalence issue gains a different significance. For supporters of an (eventually) full BI, it offers a potential for converting to a PBI – the first stage of what they want – policymakers who have already travelled so far as to advocate RSSC or EITC.This potential is strengthened if a closer look at the equivalence comes up with some hidden advantages for PBI. It is weakened if closer scrutiny reveals instead some significant benefits of RSSC or EITC, to which the equivalence claim was blind. Of course, a supporter of a full BI is unlikely to regard these benefits as decisive. It is nonetheless in her interest to understand their nature.The most powerful plea is one that grasps the full strength of the opponents’ case. Even for those whom the equivalence claim leaves undisturbed, a close look at it is most likely to repay the trouble.

Unheeded Differences Perhaps the most obvious challenge to the validity of the equivalence claim is that its formulation completely overlooks the crucial distinction between social insurance contributions and redistributive taxation. Once this distinction is made, policies which scrap or reduce the contributions of the low-paid can no longer be viewed as equivalent to policies which provide tax credits or cash benefits without reducing these contributions. For in the latter case (EITC and PBI), the entitlements coupled to the 65

Basic Income on the Agenda insurance contributions remain intact, whereas they are curtailed in the former (RSSC).This asymmetry would stick if the insurance principle were seriously upheld by the advocates of RSSC. But it seldom is, if ever, and rightly so. If the aim is to fight unemployment without increasing poverty, it does not make sense to reduce the oldage pensions, unemployment benefits, health care entitlements or sick pay of people with low earnings, as a quid pro quo for an increase in their take home pay.The versions of RSSC which are relevant in the present context – and which are actually put forward – are therefore versions which do not affect social insurance entitlements any more than EITC or PBI.The putative equivalence still sticks.Yet, there are important differences that derive from the nature of the three reform schemes, rather than from the particular variants adopted.

Unearned Income, Self-Employment and the Informal Sector Actual proposals do not only vary in terms of the profiles they imply.They can also be more or less selective. For example, RSSC proposals are sometimes restricted to specific industrial sectors, or to newly created jobs, or to workers with a low level of education or a long record of unemployment. EITC was initially confined to families with children. In the form of a non-contributory and non-means-tested pension, a BI restricted to the over 65s already exists in some countries. Many of these features can be readily disqualified as potential responses to the equivalence claim, as they can easily be mimicked within the framework of the other two policies. But some of them – for example, a restriction to waged workers of the formal sector – cannot be so easily dismissed because of their link with intrinsic differences between the three schemes. This difference can only appear as one relaxes the simplifying assumption made so far that household incomes consist entirely of formal sector wages. Indeed, because of their constitutive features, RSSC, EITC and PBI do not relate in symmetric fashion to the distinctions between earned (or labour) and unearned (or non-labour) income, between waged employment and self-employment, and between formal and informal income.This can be brought to light, in the simplest possible way, as follows. PBI better for poor rentiers, pensioners and divorcees? Let us define unearned (or non-labour) income as any form of income that does not consist in direct compensation for the performance of labour, for example, interest on savings, capital gains, contributory pensions, invalidity allowances, or alimonies received from divorced partners. As a simplifying, but not too unrealistic, assumption, suppose that unearned income is not subjected to anything analogous to social security contributions, and hence is only taxed, in the initial situation, from level C, with net unearned income rising along the 45° line OK. Obviously, for someone with nothing but unearned income, RSSC and EITC leave the profile of disposable income unchanged (along MDK), except for the usual tax adjustment for gross incomes exceeding C. But PBI stands out. The new levy introduced to fund the PBI applies at the same rate (t*) and from the first Euro to both unearned and earned income. Hence, GMI aside, the net income of a household with nothing but unearned income will move along PK in graph G6.23 There will therefore be a net gain for households with an unearned income in the UA 66

Basic Income and its Cognates range, who can now combine the latter with the PBI in such a way that they reach a net income in excess of the GMI and hence escape the range in which they would qualify for means-tested payments.There will also be a net gain – represented by the distance between the old OK line and the new PK line – for households whose unearned income falls between OA and OC. Somewhat paradoxically, the PBI scheme is the only one that imposes a tax on low unearned income – at a sizeable rate and from the first Euro – and yet it makes some of the more modest recipients of this unearned income better off that they would be under the other two proposals and under the status quo. If unearned income were, for some reason, exempted from the new levy or subjected to a much lower rate, the profile of net unearned income (inclusive of PBI) would not be lifted from OK to PK, but from OK to PY, with the necessary consequence that even recipients of unearned incomes far in excess of OC would benefit. Since all are entitled to the PBI, there is of course no principled reason to exempt any income recipient from contributing to it. But there may be pragmatic reasons, for example when the cost of detection of major categories of unearned income is prohibitive or their responsiveness to taxation very high. Under such conditions – which quite plausibly apply to capital income – a PBI would seem to necessarily involve, relative to both the status quo and the other two policies, a large transfer towards the recipients of unearned income, including very affluent ones.This would count as a serious, indeed conceivably as a fatal objection to PBI (and, even more, to a full BI) if one could not rely on the existence of a strong positive correlation between capital income and other income. Most people with a sizeable capital income can be expected to also have either a more detectable form of unearned income (typically, retirement pensions) or labour income on which a levy can reliably be raised. In the terms of graph G6, few households with a total income above C, if any, will therefore be net beneficiaries of the proposal, as the additional levy on their labour or pension income will completely recapture their PBI, or nearly so, and increased taxation beyond C is likely to erase any remaining gain. Nonetheless, it is true that, compared to the status quo, EITC and RSSC, PBI is more favourable to modest rentiers, pensioners, recipients of alimonies and other forms of unearned income, even if the new levy applies equally to unearned and earned income. Not only does PBI shrink (from OA to OU) the savers’ poverty trap (i.e. the range in which there is no incentive to accumulate entitlements to such income at the lower end of the income distribution), but it also improves the net incomes of unearned and earned income recipients alike in the AC range, whereas EITC and RSSC restrict this improvement to earned income. Because of this difference, the net cost to be made up by higher taxation (on all incomes) above C is unavoidably higher under the PBI. How much higher depends on how much unearned income is being received by comparatively poor households. EITC and PBI better for small shopkeepers and freelance artists? Let us now leave unearned income aside and consider the asymmetry between waged and self-employed labour. If these are subjected to the same level of social security contributions, RSSC, EITC and PBI have strictly symmetric effects, assuming of course (as is reasonable under the stated condition) that RSSC is not restricted to waged labour. But take the other extreme case, in which self-employed labour is con67

Basic Income on the Agenda tribution-free.24 RSSC then applies exclusively to waged labour and achieves two things. It brings the marginal rewards of self-employed and waged labour in line with one another up to level A (in graph G4), thus bringing down the unemployment trap for waged employment (OB) to the same level as for self-employment (OA). At the same time, it steeply increases the gap between the marginal rewards for the two types of work in the AC range, with the self-employed moving from D to K, and the waged workers from D to F, as their gross earnings increase (see graph G4). EITC, under these conditions, has a very different impact, even if its level is determined (as assumed throughout) by a person’s post-contribution (as opposed to gross) income (see graph G5).While treating waged employment in the same way as RSSC, it lifts the net rewards for self-employment above the 45° line (to OSQ) over a range of gross (and hence net) income from self-employment OW, itself equal to CF, the net wage at which the tax credit for waged employment is completely phased out. Hence, while EITC reduces the unemployment trap from OB to OA for waged employment (like RSSC), it reduces this trap from OA to OR for self-employment (whereas RSSC leaves it unchanged). Under PBI, all earnings are subjected to the same increase in taxation from the first Euro, but being exempted from social security contributions, earnings from selfemployment can exceed the GMI level (PM) more quickly, with the unemployment trap for self-employment lowered from OA to OU, while the unemployment trap for waged employment drops, as usual, from OB to OA (graph G6). Consequently, in the extreme case in which self-employment escapes all social security contributions (and to a reduced extent in all cases in which it is subjected to a lower rate than waged labour), EITC and PBI can be viewed, relative to RSSC, as favouring self-employment by making it profitable from a lower level of gross earnings (OR in graph G5, OU in graph G6) than was previously the case. As usual, the net gain for the low-earning selfemployed (RW in graph G5, UC in G6) needs to be made up by higher taxation above C. How much higher depends on how high a share of total income is earned by selfemployed people in the relevant ranges. PBI better for LETS traders, volunteers and drug dealers, but worse for the tax base and moral guidance? Finally, consider a third possibility of asymmetry in the taxation of two types of income: labour income can be either formal – and hence subjected to discounting in the means-testing of benefits, to social security contributions and to taxation – or informal and escaping all of this. Informal work in this sense can consist of drug dealing and moonlighting at odd jobs, but also of producing one’s own food, looking after the household in exchange for a share in one’s partner’s purchasing power or performing volunteer work in exchange for a number of in-kind advantages (free meals and phone calls, computer use, and accommodation) or taking part in a network of (untaxed) exchange of services (the so-called LETS schemes). If a person’s labour income is entirely informal, obviously she could not, by definition, benefit from either RSSC or EITC, whereas she might, in principle, benefit from PBI. But given that there is a means-tested GMI in the background, to the full amount of which she is entitled (under the above assumptions) since she has no formal means, this does not make any difference.Whether or not any of the three schemes is in place, the person could be said to use her guaranteed income OM as a full BI. Her informal 68

Basic Income and its Cognates income can be fully combined with it, and her total income therefore rises from M along the top 45° line of graph G6. Suppose now, more realistically, that what income is taken into account for the sake of means-testing is considerably more comprehensive than what is taken into account for the sake of taxation (and social security contributions): it includes, for example, an estimate of the market value of one’s home and other belongings, or the claims one may be able to make on the help of close relatives, whether in cash or in kind, or some degree of self-production, or various black-market earnings far more likely to be detected by the social worker than by the tax controller. To make the contrast in the sharpest form, let us suppose that informal income is completely untaxed, but fully part of the means by reference to which means-tested benefits are being assessed.The profile for an informal worker then becomes MDK in the initial situation as well as under RSSC and EITC (see graphs G1, G4 and G5). But here again, and even more sharply than with unearned formal income, PBI stands out. For the informal worker receives the PBI at level OP and can start earning at a zero rate of tax as from that level (along the PY line in graph G6). Hence her disposable income profile is MXY rather than the less favourable MDK. Relative to the other two formulas, therefore, PBI can be seen as a subsidy to informal labour at all levels of earnings: someone with informal earnings at level CK, for example, will remain at that level after taxes and transfers under RSSC and EITC, while being promoted to CK+CI under PBI (see graph G6). How much of an advantage or a disadvantage this is heavily depends on the nature of the informal work thereby encouraged and on the way it is distributed among various levels of earning power. If poor people have little access to informal income sources and all the PBI does (in this respect) is top up the underground income of a handful of wealthy mafiosi, it constitutes, at best, wasted money. But suppose, perhaps more realistically, that the potential for informal earnings is not insignificant in the lower reaches of earning power.Whereas PBI has exactly the same impact on the formal unemployment trap as the other two schemes (down from OB to OA), it has a dramatically stronger impact, under the stated assumptions, on the informal unemployment trap, which it brings down from OA to OV (in graph G6), while the other two schemes leave it unaltered at OA. Owing to the PBI, people need to possess an informal earning power of only OV for them to have access to a total net income in excess of OM and hence find themselves beyond the maximum limits for means-tested benefits.25 Exiting the range in which the means test applies not only has the advantage of avoiding what may be regarded as unwelcome intrusions into the claimants’ privacy (on a scale that far exceeds what a tax controller could get away with). More important for our present concerns, it enables poor people in the VA range (graph G6) to get out of the trap and start earning in the informal sphere without being taxed at an effective rate of 100 percent. Moreover, the administrative costs of the means test can be expected to shrink significantly, at least if only a small proportion of those currently on means-tested benefits have earning powers in the very lowest range OV. On the other hand, this very reduction in the reliance on the means test may be considered a serious disadvantage by those who view transfer schemes as an opportunity for enforcing proper conduct.The very fact that a PBI would, relative to the other two schemes, reduce the scope for meddlesome casework from the OA range to the OV range of earning power then counts as an argument against it.26 So does, less controversially, the fact that the dramatic lowering of effective taxation on informal 69

Basic Income on the Agenda income in the VA range under PBI, may lead some people who would be earning a formal income beyond level OA under any of the other two schemes to substitute some informal for formal earnings, thereby reducing the tax base. This effect may or may not be offset by the fact that those in the VA range of earning power will have, relative to the other two schemes, a somewhat greater incentive to do some formal work too and, above all, to maintain their skills and sanity.

Underneath the GMI A second set of differences emerges – even if all income consisted of formal waged income – as soon as one allows for the possibility that some households’ incomes may fall below the GMI by virtue of their choosing to earn less than this amount. That some households may have an income below the GMI is an obvious possibility in those countries in which the GMI is restricted to certain categories of households (e.g. to families with children or to single parents). But even in those countries in which there is an effective and general GMI applying to all types of households, it is possible for people to find themselves under level OM because the right to GMI payments is usually subjected, not only to a means test, but also to some willingness-to-work test. Consequently, people who willingly reduce their working time in such a way that their gross income falls in the OA range are typically not entitled to having their disposable income lifted to level OM. If an income equal to OM were strictly necessary for survival, this would be an insignificant fact, as no one would make use of such an option. But in reality, even if the GMI does not exceed sheer survival requirements, account must be taken of the scope for intertemporal transfers (living off one’s savings or one’s loans) and interpersonal transfers (above all within households). Hence, finding oneself in the OA range as a result of choosing to give up, be it temporarily, part or all of one’s job, is far from inconceivable. PBI better for low-paid working-time reducers? For this reason, it is not irrelevant to look at the profile of disposable incomes, under the three policies, in the gross income range OA. Under RSSC and EITC, disposable income falls steeply along the 45° line (DO) as people voluntarily reduce their working time (see graphs G4 and G5). Under PBI, in sharp contrast, disposable income goes down far more mildly in the same circumstances (along DP in graph G6).27 This basic fact can easily be used to clarify and qualify our earlier suggestion that, unlike employment subsidies of any kind (such as RSSC and EITC), BI schemes can be advocated, not only as unemployment-trap-reducing in-work benefits, but also as employment-trap-reducing chosen-time subsidies. When displaying the apparently identical net outcomes of all three policies, the thick line that shows the profile of disposable incomes in graphs G4 to G6 is simply blind to the fact that if people choose to earn less than OA, RSSC and EITC send them down the DO track, whereas PBI keeps them on the DP line. Relative to RSSC and EITC, therefore, PBI cheapens, and hence encourages, working time reduction in the OA range. How much of an encouragement this proves to be depends on how many people find it feasible and sufficiently attractive to live below the GMI with voluntarily reduced working hours.28 70

Basic Income and its Cognates For those who believe that the expansion of individually chosen part-time jobs is desirable, this is an important argument in favour of PBI against RSSC and EITC. But others believe that such an expansion is undesirable – for example, because of the training or organizational costs it generates, because of the gender bias in its distribution, or because of the lesser involvement in the work atmosphere that may be associated with it.They can therefore use the very same difference to substantiate their preference for RSSC and EITC over PBI.They must, however, be careful not to overstate their case. For whereas the policies have sharply diverging relationships to working time reduction at sub-minimum levels of earnings (the OA range), all three of them favour it in a strictly symmetrical way in the probably more crucial BC range. For people whose earnings are in this range, the three policies decrease identically the cost of reducing gross income, and hence working time, since any voluntary move from C to A is now matched by a smaller decrease in disposable income than was the case in the initial situation (graph G1).29 PBI necessarily more expensive in an economically relevant sense? Like the asymmetries pointed out when different types of income are considered, the asymmetry just emphasized underneath the GMI level unavoidably upsets the net cost equivalence between the three schemes, and hence their impact on the tax adjustment required in the upper reaches (above C) to secure budget neutrality. But which way the resulting net cost difference will go is a complex and contingent empirical matter. Voluntary working time reducers in the infra-minimum range OA are entitled to higher net transfer payments under PBI than under RSSC and EITC.This justifies a strong prima facie expectation that the latter two schemes must be cheaper than the former two, not only because the people who happen to find themselves voluntarily in that range in the initial situation (graph G1) will cost more under PBI, but above all because PBI creates a greater incentive to give up some tax-paying working time and move into this range. But there are two countervailing effects which may more than offset this difference. Firstly, the financial sanction if one were to openly admit that one deliberately keeps one’s working hours down in the OA range is notably tougher under RSSC and EITC (it corresponds in both cases to the gap between MD and OD in graph G6) than under PBI (it corresponds in this to the gap between MD and PD). Given the choice between being fully and (allegedly) involuntarily unemployed and being partly and (openly) voluntarily unemployed, some people may choose the former option under RSSC and EITC (and hence cost the government the full amount OM), whereas they would choose the latter option under PBI (and hence cost the government the smaller amount OP, minus the tax and contributions, which rise from zero to DY=OP as gross earnings rise from zero the break-even level OA).This assumes (plausibly, it seems to us) that the inconvenience to the claimant of claiming the full means-tested, willingness-to-work-tested GMI (intrusiveness, dole queues, job applications, risk of being caught cheating, etc.) are neither so great that everyone would choose instead to work part-time under RSSC and EITC, nor so mild that no one would choose the parttime option even under PBI. PBI would then offer a potential for savings by dissuading people from relying on the full amount of the GMI and encouraging them instead to earn a (low) taxable income. Secondly, there is the work sharing aspect of the encouragement of voluntary working time reduction by PBI. As mentioned before, this encouragement – espe71

Basic Income on the Agenda cially the one which is specific to PBI (in the OA range), as opposed to the one common to all three schemes (in the BC range) – is only sizeable for those with a low earning power. But this is also the category of workers which is most significantly plagued by involuntary unemployment. The very fact that some voluntarily reduce their working time, be it temporarily – for example, in order to look after their children or acquire further education – frees slots for some of the others. This is of course at the core of the chosen-time-subsidy aspect of the PBI strategy. But it is also a potential source of savings, as the net transfers to these newly employed people shrinks from OM (or whatever higher level of unemployment benefit they are entitled to) to OP minus the taxes and social security contributions on their earnings. Suppose for example that two workers with gross earnings OB decide to halve their working time under PBI whereas they would not have done so under RSSC or EITC (because their disposable income would then have fallen too steeply along GDO, rather than GDP, in graph G6). Suppose further that both halves of their jobs are taken up by an unemployed previously claiming benefits at level PM (in addition to her PBI at level OP). PBI payments are obviously unchanged. So are taxes and social security contributions, with the third person now paying what the first two no longer pay. But a net saving of PM (or more) is booked by virtue of the unemployment benefit no longer needing to be paid. This is the dividend from work sharing, from the mutually beneficial redistribution of leisure. Added to the dissuasive effect on GMI claimants mentioned before, it is bound to offset at least in part the direct net cost that stems from PBI, giving a better deal to the low-paid voluntarily underemployed.What the net balance is – and hence whether the tax rate beyond C needs to be readjusted upward or downward relative to RSSC and EITC – depends on such factors as the availability, among the involuntarily unemployed, of suitable substitutes (qua location and skills) for the voluntary timereducers, and can therefore not be told a priori. PBI better for low-paid quitters and strikers? The same difference in what happens below A is also most relevant to assess the difference between the three policies in terms of the redistribution of bargaining power they bring about. If a worker chooses to quit because (s)he is fed up with her working conditions or pay level, the immediate cost (s)he has to bear is smaller by amount OP under PBI than under the other two policies. Similarly, workers on strike would keep receiving their PBI (which could therefore be regarded, among many other things, as a systematic collective subsidy to strike funds), while they would obviously lose all benefits from EITC or RSSC. 30 For highly paid workers, this is of little significance, partly because the PBI corresponds to only a small fraction of their earnings, and partly because, for most of them, their bargaining power derives mainly from the alternative job opportunities they have. But for poorly-paid, low-skilled workers, even a low PBI can make a significant difference to their bargaining power, with noticeable consequences for both the working conditions and the pay of the jobs in which employers will want to keep them. It does not follow from this difference that the quit rate can be expected to be higher or strikes to be more frequent under PBI. Simply, the fact that the threat of quitting or striking needs to be taken more or less seriously, depending on the scheme adopted, will systematically lead to different deals. This argument, again, is two-edged, as some may 72

Basic Income and its Cognates regard this equalization of bargaining power (however modest) as a major danger: for those concerned that there should be enough people to do cheaply the dirty jobs, EITC and RSSC offer, for this reason, a better promise. PBI better for human capital accumulation? In the discussion on BI or NIT, it is sometimes pointed out that the incentive to improve one’s skills and thereby one’s earning power will be reduced. This can be attributed in part to the rise in marginal tax rates upward of gross income OB (in graph G6): the return to skill acquisition, and productive effort generally, is significantly reduced in the BC range, in which many people start their working careers, and to a lesser extent also beyond C.The negative impact on incentives can also be attributed in part to the fact that some young people will be satisfied to supplement their PBI with the income from a low-skilled more or less regular job in the AB range, whereas in the absence of a PBI they would have bothered to acquire further skills so as to fit into a more productive job in the BC range.31 The overall result would be a less educated labour force and a less productive economy than could be, and would otherwise be, the case. This is a potentially important argument, which applies in exactly the same extent to RSSC and EITC as it does to PBI, since in the AC range all three policies amount to straightening DEF into DGF (see graphs G4 to G6). How weighty this argument is against the family formed by all three policies needs to be assessed bearing in mind, among other considerations, the following three. First, intra-organizational promotion provides non-monetary incentives (power, titles, non-taxable privileges) no less than monetary ones, and the former are left unaffected by the change. Second, rises in marginal tax rates may go hand in hand with rises in marginal returns, if it operates on the background of a growing inequality in gross earning power – which is precisely one central aspect of the trends which are creating the problem in which the new social question consists.Third, the human capital of the people stuck in the unemployment trap is getting quickly eroded, owing to a powerful positive feedback process of skill obsolescence and motivational adjustment, which the three policies, by reducing the unemployment trap, are helping to keep in check. All these considerations apply symmetrically to all three policies. If there is a significant difference between them, as regards human capital accumulation, it must again be due to differences arising below the GMI level. One possibility is that some people may be happy to combine their PBI not only with a low-paying job in the AB range, but also with an even lower-paying job in the OA range, one which would yield them a far lower disposable income under RSSC and EITC (see the gap between PD and OD in graph G6).Would some people really choose to durably stick to such a modest standard of living by voluntarily refraining from upgrading their skills? Perhaps. But this difference is quite likely to be more than offset by the following three countervailing differences. First, one standard reason for voluntarily and substantially reducing one’s working time in such a way that one’s disposable income temporarily falls below M, is precisely that one may wish to invest in further education and training so as to boost one’s future earning power. In this fashion, a PBI can work as a permanently available study grant, whereas RSSC and EITC cannot. Second, more speculatively, another major reason for choosing to temporarily reduce one’s earnings below the (permanent) subsistence level is one’s desire to spend 73

Basic Income on the Agenda more time with one’s children at the time they need it most. Given some striking standard empirical facts about the determinants of educational achievement – such as the strong predictive power of one’s mother’s educational level – freeing poor households from a compelling pressure to work full time at the expense of bringing up their children as they would like to can be a major contribution to the productively usable human capital of the next generation. Thirdly, the impact on bargaining power associated, as pointed out above, with the differential cost of quitting and striking is likely to show up in the training content of the low-productivity jobs (in the AB range) which the reform aims to make viable: workers with a higher bargaining power will be less willing to put up with low pay unless the job provides them with a useful experience or training which could help them move up.The heavier the cost to them of quitting or striking, on the other hand, the less discriminating they can afford to be and the more likely the low-skilled are to be stuck in dead-end jobs.32 Whether directly or indirectly, therefore, the improvement of the lot of those who choose to find themselves in the OA range would seem to enable PBI to foster a better educated work force than RSSC and EITC.

Universality PBI more effective as an instrument against poverty? Finally, the most striking difference between PBI and the other two instruments is that it is given ex ante to all, rich and poor.33 In terms of the fight against income poverty, this has the obvious advantage of making sure that at least some purchasing power reaches the poorest. The rate of take up for universal benefits is systematically higher than for means-tested ones, owing both to the difficulty of providing the information to those entitled to claim them and to the stigmatization that tends to accompany any benefit scheme restricted to the poor. Of course, a PBI does not lift people out of poverty and therefore does not make the GMI dispensable. But for those people in the OA range for whom the GMI scheme does not do what it is supposed to do, whether permanently (because of the inhibition generated by intimidating or humiliating procedures) or temporarily (because of information failures or file-processing delays), it makes a big difference whether they can move along the PD curve, as the PBI regime enables them to, rather than along the much lower OD curve, as the other two policies restrict them to in the most favourable case, or even along the even lower OZ curve if tax credits take a while to reach their beneficiaries under EITC (see graph G6). How much of a difference this makes clearly depends on the relative effectiveness of GMI and PBI at honouring the entitlements of people in the OA range.The greater the gap, the greater the additional net cost of PBI (to be offset again by tax adjustments beyond C), but also the stronger the prima facie case in its favour as an incentive-friendly but nonetheless effective tool against income poverty. On the other hand, it may be argued that it is precisely this administratively straightforward nature of the transfer, the fact that it does not rely on means-tested transfers, that makes PBI less effective as a strategy for tackling those aspects of poverty which do not reduce to income poverty. Desperate isolation, the inability to manage one’s income or a shaky mental health may often be no less crippling than the lack of adequate purchasing power. Against this background, the social workers’ implementation 74

Basic Income and its Cognates of the means test, while sometimes perhaps unpleasantly intrusive, provides precious opportunities for them to effectively identify and tackle these further problems. This argument would definitely be worth a close look if we were here talking about a full BI. But a PBI is coupled with a residual GMI which gives social workers the opportunity and power to help as aptly as they can in those cases in which non-income aspects of poverty require it. Indeed, by automatically softening the blow of sudden adversity (by virtue of its dispensing with the need for any claiming procedure) and by lifting more people out of the means test than the alternative schemes (at least if there is something to the informal work argument and the working-time-reduction argument) a PBI might enable social workers to do a better job by concentrating, in a less emergency-dominated way, on the people who really need their professional help. PBI better at handling the uncertainty side of the unemployment trap? Field researchers have pointed out that what keeps many of the excluded from accepting a job or actively looking for one is often less the lack of an income differential between no-work and work than the fear of leaving the relative safety of a regular flow of benefits for the uncertainty of a job that may pay late or prove too demanding. For the risk of a period without any income is something households at the margin of solvency cannot responsibly afford.34 If this is the case, the trap may not be suppressed for someone with a gross earning power of OB for example, by the sole virtue of the corresponding net income being lifted by the reform from BE to BG (for example in graph G5). People may therefore stick to joblessness with the lower income level OM, because of the feeling that the higher income BG is less safe: the job can be lost, and in case it is lost, going back to OM is often far from automatic. Hence, it may prove of far more than marginal importance to point out that under PBI, not under the other two policies, part of BG is made totally safe, because people keep it no matter what.With a PBI, the net income BG of someone with gross income OB is neither less nor more than under the other two policies, and the income-differential aspect of the alleviation of the poverty trap is therefore identical (as reflected in graph G3, or in the thick lines of graphs G4 to G6). But under PBI, jumping from earning nothing to earning OB is less of a risk than under EITC or RSSC because only the difference HG (in graph G6) is affected by uncertainty, while the firm flow BH is unaffected. The importance of this difference can vary from the trivial to the crucial, depending on (1) how large a share of net income the PBI represents, (2) how insecure the relevant jobs and earnings are (private sector precarious jobs, self-employment, workers’ co-operatives, etc.)35 and (3) how complicated, lengthy, humiliating and in other ways difficult it is to get the flow of means-tested benefits running again once the job has vanished, or has been taken away by the employer or been given up by the worker. PBI less appealing to Trade Unions ? The fact that PBI is paid ex ante to all may be better, for the reason just explained, for the sake of fighting unemployment. It might also provide a reason for expecting PBI to be less popular with the Trade Union movement than the other two policies.36 That this should be the case is not obvious from the considerations explored so far.True, PBI is comparatively more favourable to informal labour than the other two schemes, which makes it unlikely to be a vote catcher among organized formal workers. But 75

Basic Income on the Agenda compared to RSSC, it has the advantage of leaving untouched the level of social security contributions – which, unlike tax revenues, are often to a notable extent under Union control. Moreover, PBI can be seen as a massive contribution to strike funds. Of course, unlike strike funds of the standard form, PBI is not Union-controlled. It gives a power to strike without the Union’s consent and may therefore prove a mixed blessing from the standpoint of the Union’s leadership and its ability to effectively mobilize working people. Nonetheless it should, all in all, strengthen the bargaining power of organized labour. So, why are the Unions not more enthusiastic? If a Union’s objective can be reduced to the interest of its median member and if the latter can be depicted as earning a gross wage in excess of OC – and hence having nothing to gain from the reform – then the Union’s being at best lukewarm should come as no surprise. But this would apply symmetrically to all three policies and would therefore do nothing to predict a specific reluctance to embrace PBI. The PBI’s ex ante payment may be the key. For even if a worker’s net income, her effective marginal tax rate, her comprehensive average tax rate and the income difference between her income at work and out of work (whether social assistance or unemployment benefit) are exactly the same in all three cases, her take-home pay is significantly lower under PBI than under the other two schemes (in graph G6, IF instead of CF for a worker with level CK of gross earnings).The difference is exactly the amount of the PBI, which the worker’s household receives independently of the worker’s pay package.The household should not care, but the Union will, as the fraction of the worker’s material welfare which it is perceived to control, or at least affect, will be significantly curtailed.This perception, no doubt, rests on an illusion, since in all three cases the public tax-and-benefit system modifies the distribution of gross incomes in exactly the same way. But this is an illusion of a particularly credible and possibly powerful kind. If take-home pay is a far smaller proportion of a household’s income under one scheme than under another, how can this fail to affect the degree of unionization or the ease with which the Union leadership can mobilize the rank and file?37 To the extent that Union support is important to secure the adoption of one of the policies, this effect, if significant, provides a tactical case for EITC, RSSC and the NIT version of PBI against the latter’s universal, ex ante version. But there may be more at stake than political feasibility. The Unions’ concern with the non-decrease of takehome pay may generate a different dynamics of labour costs depending on which of the policies is adopted. For if such a concern is of decisive importance, the introduction of a universal PBI and any increase in its level would affect collective bargaining very differently from corresponding advances with EITC, RSSC or NIT. As organized workers insist on protecting their net wages, not only their net incomes, they would trigger off a wage-inflationary process, with detrimental effects on the achievable trade-off between unemployment and non-accelerating inflation. PBI politically less attainable but more stable? A final asymmetry that stems from the same fundamental difference between the ex ante PBI and the other schemes concerns their political chances in the democratic process. There is bound to be a tremendous difference in perception between an income that suddenly starts being paid separately to every citizen (PBI), and such more localized and less visible changes as a modification in the schedule of social security 76

Basic Income and its Cognates contributions which will remain unmentioned in any document routinely received by workers and citizens (RSSC), or a tax reimbursement to the working poor (EITC). In all likelihood, workers in the BC range will soon notice, under each of the three policies, that they have become better off, though possibly only with some delay, owing to the length of the tax procedure (especially with EITC). Gradually, people will hopefully also start discovering the increased potential in the AB range or indeed, as far as voluntary underemployment is concerned, in the OA range. Moreover, the tax adjustment above C, if significant, is also going to be felt by some. But all this is relatively minor and local, compared to the highly visible introduction of an income henceforth paid from the same source to every citizen, coupled with a dramatic increase in the apparent rate of taxation. There are at least three reasons why this feature may make the introduction of PBI more difficult than equivalent variants of the other two alternative policies – and indeed the much closer NIT – even assuming that there is no difference whatever in their economic consequences and in their impact on the material fates of all categories of households. One reason is simply that, being less conspicuous, the other policies can be phased in more swiftly by politicians clever enough to alert the likely beneficiaries without attracting too much attention from those likely to lose out (those with incomes above C). A second reason is that opponents can easily exhibit the (alleged) absurdity of handing out benefits to scores of people who do not need them in the least, namely with a gross income in excess of OC. 38 It would not be very hard for them to be persuasive, as their audience may well fail to see the equivalence between PBI and the corresponding NIT (with all transfers on graph G6 netted out) and be eluded by the subtle superiority of giving to the rich in terms of the interests of the poor (whether because of higher take-up rates or the trap aspects discussed earlier in the present section). A third reason is the scare of high apparent tax rates. Whereas RSSC and EITC (and NIT) can be sold as tax cuts (admittedly compensated by a somewhat higher taxation in the range above C), PBI cannot but be perceived as a massive increase in government expenditure and taxation. One can no doubt devote some energy to explaining that what matters is the profile of effective tax rates and disposable incomes, which happens to be exactly the same as in the tax-cut alternatives, and that one should think about the cost of redistributive programmes quite differently from the way one thinks about the cost of substantive expenditure programmes.39 But, in some countries at any rate, it is nonetheless likely that opponents to the proposal (whether in good or bad faith) will manage to focus public attention so intensely on this huge cost to the tax payer that, for a mainstream politician, defending it will look tantamount to political death. PBI, therefore, does not fare too well in terms of political attainability. On the other hand, it may fare better in terms of political stability.The experience with the Alaskan Dividend Scheme provides some evidence in support of this conjecture. Despite occasional observations that the money would be better spent if it were more targeted or if it were used for public investment, the scheme is proving a political sacred cow.40 By contrast, being less visible, tinkering with contribution rates and tax rebates to the lowpaid may for this reason be easier, not only to introduce, but also to roll back. Further indirect support for this claim can be gathered from the discussion on family allowances.True, there has been a wave of reforms subjecting previously universal child benefits to a means-test of varying severity on the grounds that a universal payment to 77

Basic Income on the Agenda all mothers regardless of a family’s means was wasteful (as Thatcher’s government put it). But the first two countries to travel this road have since returned to a universal system (Denmark in 1981 and Japan in 1985) and similarly motivated government meanstesting plans were successfully resisted in several other countries, including the UK in 1987, Ireland in 1991, Sweden in 1995 and France in 1998. Moreover, comparative evidence seems to support the view that the level of the transfer is more secure when everyone receives it than when only the poor do. For example, the level of child benefits as a proportion of average wages is systematically higher in countries which have stuck to a universal system.41 Greater political irreversibility is a precious advantage if one is certain that reducing the unemployment trap, as done by each of the three policies, is the way to go. Otherwise, it may be regarded as a serious defect.What if, for example, the disincentives to work or train in the BC range – which are, as we have seen, a necessary corollary of the stronger incentives to work in the AB range – are so strong that a dramatic net loss in economic efficiency results, once the full effects of having a more laid back, less skilled labour-forced (drifting leftward on our graphs as worker cohorts succeed each other) make themselves felt? In this case the very superior political robustness of the PBI scheme (if established) would prove a handicap, as it would prevent swift readjustment once the negative effects of the policy were detected and their full size assessed.

Conclusion: Partial Basic Income on the Agenda? Underneath the prima facie equivalence between matching variants of the three policies, our very stylized graphical analysis has thus enabled us to highlight a number of potentially important differences. Once we laid aside the distinction between taxes as pure levies and social contributions as insurance premiums, we were left with three basic sources of differences: how the three policies relate to different types of income; how they treat those who voluntarily find themselves under the GMI level; and whether they rely on ex ante payments.This systematic inventory of what we believe to be the main relevant differences is obviously inconclusive on many counts. For most of the differences we explored, it was possible to state a priori, if there was an advantage, which way it would go. But our assessment of the significance of this advantage has always been rather vague and tentative, based on casual or indirect evidence – if any. In many cases, a far more serious empirical evaluation of the differential impact would be useful, as would, in some cases, more complex theoretical modelling. However tentative, our exploration should force supporters of each of the three policies to qualify a number of arguments they routinely make in favour of their favourite.This holds, in particular, for those who support a universal PBI. If they do so out of a basic concern for improving the options open to the worst off – a plausible normative foundation for the objective of fighting unemployment without worsening poverty – they will no doubt have been pleased to see their case upheld in connection with the impact of universal payments on take-up and traps and the importance of unconditional support under the poverty line for bargaining power and human capital formation. But there are at least two important differences between PBI and its alternatives which present supporters of PBI with serious challenges. 78

Basic Income and its Cognates One concerns PBI’s political feasibility. Owing to its ex ante nature and the implied height of the apparent rates of taxation it requires, we have seen that there are reasons to expect a universal PBI to be comparatively less popular with organized workers and public opinion.True, much of this unpopularity can be traced to variants of an illusion – basically the failure to perceive the sense in which a tax expenditure and a benefit are equivalent. But this illusion may well be so resilient as to block any direct introduction of a PBI. One can no doubt endeavour to undermine it through effective pedagogical efforts. But this is likely to prove a Sisyphus task. Alternatively, one can try to use opportunities to introduce a universal PBI as a uniform dividend on a commonly owned asset (royalties on oil extraction, for example), or as uniform compensation for a commonly suffered nuisance (energy consumption, for example), or even out of non-inflationary money creation.42 But this is unlikely to provide more than a very low floor. Beyond this, the most promising way forward for shrewd politicians who believe in PBI’s substantive preferability is probably to first introduce an appropriately calibrated uniform refundable tax credit (or NIT), soon to be inconspicuously turned into a tax-creditable benefit (or PBI).43 The second challenge may be viewed as more serious still, as it does not concern PBI’s achievability, but its very desirability in a context in which informal (i.e. legally or illegally untaxed) income represents a major option. Providing one can (plausibly) assume that the assessment of GMI entitlements is more sensitive to informal income than taxation is, we have seen that PBI is unquestionably more favourable to informalincome earners than the other two policies. It is not altogether absurd to regard some boosting of the informal sphere as a meaningful objective in itself. In one interpretation, for example, political ecology can be defined by this objective and contrasted on this basis with liberalism and socialism.44 However, this assumes an unrealistically benign perception of what is bound to be a very heterogeneous bundle. It might be good, not just for the quality of human relations but also for the long-term performance of the economy, if people are less hooked on monetary rewards and spend more time with their families, their vegetable gardens, their neighbourhoods and their voluntary associations. But some informal work is ruled by no less strong a cash nexus than formal work, or is performed within the framework of a barter relationship with no intrinsic virtue compared to monetary exchange. Indeed, much informal work, precisely because it escapes the standard controls of formal activities, may be more degrading, unsafe and in other ways unpleasant than had it been performed formally. If this is the typical case, there is not much of a case left for promoting informal activities. Moreover, the PBI will be wasted on some high-income people from whom little or nothing will be clawed back in taxes. And by increasing the relative attraction of untaxed activities, it may jeopardize the sustainability of the GMI, and hence of the level of poverty alleviation already reached. How much of a threat this really is depends, for example, on how much a country’s economy relies on non-monetary exchange, on how weak or corrupt its tax authorities are, and on how much dishonesty its citizens feel comfortable with. Under such circumstances, PBI is unlikely to be better than EITC, for example, as way of tackling unemployment without worsening poverty, even if political feasibility were no problem. Or at least it will not become superior until the background institutions have changed in such a way that the bulk of sizeable market incomes has become sufficiently visible to a sufficiently effective tax administration. 79

Basic Income on the Agenda These two challenges do not really overturn the fundamental presumption in favour of a PBI – and, beyond it, in favour of a full individual BI – for those whose ultimate objective is neither maximal output nor maximal employment, but is captured in a conception of justice that gives priority to the real freedom of the worst off.45 But they do invite them to shed any dogmatism in terms of immediate political agendas. If the cosmetic-didactic dimension of politics is taken into account, if the administrative background conditions are given the attention they deserve, the most direct route is not always likely to be the best one. Under certain circumstances, some of the other strategies for fighting unemployment without increasing poverty may be either more feasible or more appropriate. Under certain circumstances, BI may need to step aside and give way to one of its cognates.This is a concession, but no sacrifice of principles to expediency.To best serve one’s principles, one must not stubbornly advocate, under all circumstances, the immediate implementation of their most straightforward expression.To wish that one day, everywhere, an individual and universal BI should belong to the acta, does not prevent one from believing that sometimes, in some places, it is better that one of its cognates should be on the agenda.

Notes 1 2

3

4

5

80

On the shape and causes of the growing dispersion of (potential) earnings in OECD countries, see e.g.Atkinson (1993a), Gottschalk and Joyce (1993),Wood (1994),Atkinson (1998), etc. See Van Parijs (1987). One can conceive of other characterizations of the new social question and the corresponding new class divide, for example by emphasizing the unprecedented importance of one’s generational location, as a result again of a set of interacting factors such as rising life expectancy, rising potential cost of health care, development of the old age pension system and increasing ability to knowingly cause (and avoid) long-term environmental damage. This belief is consistent with our awareness that several relevant policy proposals are being left out of the picture, in particular (1) policies which are ethically unacceptable (for example, lowering the general level of income protection or sending working women back home); (2) policies which, however well intended, would be counterproductive (for example, protectionism or across-the-board reduction in the maximum working time); (3) policies which do not constitute substitutes but (arguably) complements to the sort of socio-economic reform we want to focus on (for example, macroeconomic co-ordination or continued education). The notion of basic income, as used here, is therefore broader than the definition standardly used by BIEN in three important respects: it is not necessarily paid to all on an individual basis (even if the entitlement is individual, its level may depend on household composition), nor quite without means test (in the sense in which even a uniform refundable tax credit is means-tested), nor quite without work requirement (in the sense in which even a broadly characterized participation condition can be construed as a work requirement). See e.g.Atkinson (1995),Van Parijs (1996), Schokkaert,Van der Linden and Van Parijs (1997), Van der Linden (1996; 1999) and Standing (1999) for a set of arguments in favour of basic income as an attempt to tackle (what is here described as) the new social question. See also Fitzpatrick (1999) for a recent overview of arguments for and against.And see the web site of the Basic Income European Network (BIEN) for a comprehensive, regularly updated, annotated bibliography (http://www.econ.ucl.ac.be/etes/bien/bien.html).

Basic Income and its Cognates 6

7

8

9

10 11

12

13

In this chapter, we shall focus on the social-dividend version. But much of what will be said about it also applies to the NIT version, with the level of the lump-sum refundable tax credit pitched at the same level as the social dividend. However, there are important differences (see Van Parijs, 1995: 35-37), some of which will be pointed out as we go along but which are not the focus of this essay. With a number of specific features to be presented shortly, our inquiry fits therefore into a long sequence of attempts to assess the merits of alternative transfer schemes on the basis of a careful comparison of their differential consequences, from Zeckhauser (1971) and Garfinkel (1973) to Phelps (1994) and Besley and Coate (1995), for example. In the OA range, people lose everything they earn, since each Euro earned is cancelled by an equal reduction in the benefit. But it is precisely people whose income is in the OA range who are the net beneficiaries of the scheme. The intuitive ideas behind these results are lucidly explained, for example, in Piketty (1997). If the aim is to maximize the tax yield, raising the rate of tax on a particular slice of income has both a positive effect because it increases what is collected from anyone with an income at least as high as the income levels concerned, and a negative effect because it reduces the incentive to generate income (through work, saving, effort, etc.) for anyone whose income level falls within the range concerned. Obviously, the positive (or distribution) effect is the higher, the lower the income slice concerned: when taxing a low slice, one taxes the income of nearly everyone, when taxing a high slice, one collects income from only a few.The negative (or incentive) effect, on the other hand, will depend both on the number of people whose income level happens to fall within the slice considered and on how responsive they are to a lowering of what they can gain from an increase of their working time, effort level, etc. If there are relatively few people with very low (potential) earnings, one would therefore need them to have an extremely high level of responsiveness (or tax elasticity), compared to people with a higher income, to justify a lower, or even an equal rate of taxation on the lowest slice of income. See Van Parijs (1995: chapters 2 and 4) and, more briefly, the discussion between Scharpf and Van Parijs in this volume. To make the comparison more straightforward, we shall generally write as if we were talking exclusively about workers’ social security contributions. Under the received wisdom among economists, however, it does not make much of a difference whether the social security contributions are levied on the workers’ or on the firm’s side. Whether under perfect competition or under reasonable non-competitive assumptions, whatever balance of forces leads the workers’ net compensation (and hence their share in the burden of the levy) to settle at a certain level under one administrative arrangement will lead it to settle at the same level under the other. See, however, Rasmussen (1994) and Muysken and Van Veen (1996) for attempts to come up with a significant discrepancy. This necessity holds under the assumption that the income of the better-paid workers is not to be taxed at a rate approaching 100 percent either. Only nearly, because it is imaginable, if there are sufficiently few workers, not only in AB, but in the AC range as a whole (in graph G4), that the lifting of DE to DG and the upward shift of EF (henceforth starting from G, while keeping the slope unchanged) will cost so little that the marginal tax rate on gross income above C, while unavoidably much higher than before, will still fall short of 100 percent. In 1998, the federal share in the cost of TANF (Temporary Assistance to Needy Families), the programme that replaced AFDC in 1996, was a shrinking US$ 20 billion, compared to a growing US$ 22 billion for EITC (Figures from the Budget of the US Government kindly pro-

81

Basic Income on the Agenda

14

15

16

17

18

19

20

21

22

23

82

vided by Anne Alstott). For illuminating critical discussions of EITC, see Alstott (1995) and Shaviro (1997). Keeping this GMI in the background is essential to the arguments below. These, therefore, hardly apply (even in highly stylized fashion) to the US, since (1) the main assistance programme for able-bodied working-age people (earlier AFDC, now TANF) provides aid only to poor single mothers with children and a very small number of poor married couples with children, while all others have to rely on local and state general assistance programmes which provide very little cash aid, if any; (2) even for those covered by TANF there is now a time limit of five years. At this stage, the PBI can be interpreted either as a citizen’s income (if it is granted without any activity condition) or a participation income (if it is subjected to a broad condition of participation). It can also be interpreted as either a social dividend or a negative income tax. Because of the PBI’s effect on lower incomes prior to the means test, less households qualify for means-tested payments (only those in the OA range, no longer all those in the OB range), and those who qualify are entitled to lower payments (a maximum of PM, instead of a maximum of OM). In some proposals, for example Pelzer’s (1996; 1999) Ulm Modell, the BI is supposed to be funded by a specific proportional Bürgergeldabgabe (citizen’s income levy) with a broader income base than both social security contributions and the ordinary personal income tax (no exemptions allowed). The equivalence of PBI and EITC with RSSC becomes far-fetched as soon as they enable (irrespective of the GMI) households with gross earnings of less than OA (=OM) to reach a net income of OM or more. Even if regressive tax schedules are ruled out (if explicit marginal taxation in the OA range is not higher than beyond A), these limits should encompass, in the Western European context, quite substantial PBI levels. See Van Parijs (1995: section 4.6) on the importance of predictable taxation to avoid negative employment rents. Even the introduction of a partial BI could not avoid significant shifts, however astutely the various available tax instruments are used to minimize sudden downward shocks on some categories of (not very well off) households. (See Gilain and Van Parijs (1995) and Terraz and Joyeux (1998) for micro-simulations of various scenarios in the case of Belgium.) See, for example, the versions of NIT advocated by Mitschke (1985; 1995), Haveman (1988), Brittan and Webb (1991), Bourguignon and Chiappori (1997), Godino (1999), and endorsed by Castel (1999), etc. In Graph G6, net earnings would no longer fall on the PF line, but, for example, on a straight line MF, with a bigger tax adjustment beyond C. Just as a flattening of the net gains in the BC range was a price to pay for getting the AB range out of the trap, a flattening of the gains in the BC range is a price to pay for getting everyone out of the trap. In principle, a full BI could do it too, simply by imposing, somewhat oddly, a 100 percent tax rate on the lower layer of income (as in a variant studied e.g. by Salverda, 1984). But given that a full BI is not needed to achieve such a profile – whereas it would be to abolish the trap –, the first argument in favour of a PBI becomes decisive. The measure is therefore neutral for someone with an unearned income of OC.The difference FK between the net income of a household with unearned income OC and that of a household with earned income OC corresponds to the amount of social security contributions to be paid by the latter.

Basic Income and its Cognates 24 While not yielding any less entitlements to benefits than waged labour.As mentioned earlier, we are treating taxes and contributions as equivalent in this respect. 25 The difference also holds, in a gradually attenuated form, when the earning power accessible to the poor is a combination of formal and informal earning power.As the formal component increases, the extent of the reduction of the trap gradually shifts and shrinks from VA to AB (in graph G6), whereas for the other two schemes it expands from nothing to AB. Needless to say, the difference is also reduced as informal income is treated less differently from formal income by the tax system (at the limit, they are taxed at the same rate and with the same probability) and more differently by the means-tested GMI (at the limit, any level and kind of informal income can be combined with the GMI, just as with the PBI). 26 On this disadvantage of any variety of BI relative to NIT (as interpreted), see e.g. Petersen (1997: 58): whereas the means which determine the level of the NIT transfer take the right to support payments by spouses and relatives into account, the introduction of a nontested transfer program would destroy the important role of self-responsibility in the family. 27 For NIT, two interpretations are possible. Either the payment of the negative tax is subjected to exactly the same (willingness-to-)full-time-work requirements as the payment of the GMI, in which case the disposable income of a willing working-time reducer follows the sharp DO slope. Or it is kept distinct as an unconditionally refundable tax credit, in which case the relevant slope is the milder one DP, the same as under PBI. Under the latter interpretation, people with a gross income below A could claim some amount even if they are not job seekers as an advance on their NIT, and a further amount only if they are job seekers.This can easily become confusing for both the public officials in charge and, even more, for the ordinary claimant. As long as a residual conditional GMI remains in place, therefore, the strict, PBI-equivalent NIT is quite tricky to implement. I shall nonetheless assume that such implementation is possible. 28 To judge by the experience of Belgium’s voluntary career interruption benefit scheme, this number may well be considerable.The scheme was started in 1986 by the then Employment Minister (and later Director General of the ILO) Michel Hansenne.The number of beneficiaries at any single time has risen from less than 10,000 in 1987 to around 50,000 in 1996 (about 2 percent of total employment), most of them women (86.5 percent) and most of them retaining a part-time job (69 percent). See Szabo (1997: 16-18) for further data. 29 This is one of the central reasons why Phelps (1997) rejects EITC as badly designed, in favour of his own hourly wage subsidy restricted to full-time workers. 30 Under a sufficiently broad specification of the participation condition (part-time work, education, child care, etc.) most of what applies to PBI in this section also applies to Atkinson’s (1993b; 1996) participation income. One exception concerns the power to strike, as it would be hard to construe strikers as fulfilling the participation condition. 31 These two facets of the human capital argument against BI are being presented as the central objection against it, respectively, by Bovenberg and Van der Ploeg (1995) and by KrauseJunk (1996). 32 The first argument (educational grant) is closely related to Standing’s (1986; 1992; 1999) case for BI as a strategy for facilitating the back-and-forth between work and education which our flexible economies require. The second argument is related to Mayhew’s (1991) cautious claim that the introduction of a BI should lower the relative attraction of low-skill poor-training full-time jobs from both the employers’ and the employees’ standpoint, and thereby help a country such as Britain out of a low quality/low skills equilibrium.

83

Basic Income on the Agenda 33 Contrary to the features of PBI discussed in the previous two sections, universality is not shared by NIT, which will therefore side, in the discussion to follow, with EITC and RSSC. 34 See Delvaux and Cappi (1990), Jordan et al. (1992). 35 This ties up with the old theme, developed by Nooteboom (1986) and Meade (1989) among others, that a BI would be a way of spreading entrepreneurship in the population, and also with the idea that there is a natural complementarity between a BI and both a co-operative economy and a share economy (see Van Parijs, 1995: section 6.4). 36 By no means do we want to imply that there are no voices in the Trade Union movement that support the BI approach (see, for example,Van Berkel and Hindriks, 1991; De Henau, 1996). But with the remarkable and partly enigmatic exception of the Voedingsbond FNV in the Netherlands (Van Berkel et al., 1993), there is to our knowledge no single Union leadership which has openly and systematically advocated it (which cannot be said for RSSC, employment subsidies or even forms of NIT). 37 In some countries, this effect may be reinforced by another, which derives from the differential impact of the various schemes on unemployment compensation. Under both EITC and RSSC, the level of unemployment benefits should be hardly affected, if at all. But under PBI, their (post-tax) level needs to be reduced by an amount about equal to the PBI. In some countries (such as Belgium), the distribution of a large proportion of the unemployment benefits is managed by the Unions and a significant fraction of the Unions’ revenues takes the form of a share in the amount distributed by way of compensation for the expenses incurred in the provision of this service. If the latter is dramatically reduced, the Unions may fear that a significant part of their funding will become less secure. Even if this is not the case, the material importance of the Union’s role in the (unemployed and potentially unemployed) workers’ eyes may be notably reduced. 38 James Tobin remembers, for example, that, in the 1972 democratic primaries, George Mc Govern’s demogrant proposal (which Tobin, as his chief economic adviser, had managed to squeeze into his platform) was ridiculed by his rival Hubert Humphrey on precisely this ground (see Van Parijs, 1998: 8). 39 This point is neatly made by Shaviro (1997). 40 See Olson and O’Brian (1990), Brown and Thomas (1994). 41 See Gauthier (1996: 164-166) on the back and forth between universality and means-testing since 1976, and on the putative effect of universality on benefit levels. See also Bezat (1998) on the recent U-turn in France. 42 An old and suspicious idea recently rehabilitated by Huber (1998; 1999). 43 If the strength of illusions makes a NIT version the best strategy, part of the case for an individual PBI (versus a household-sensitive full basic income) delineated at the end of the first section becomes shaky. 44 Each of the three main ideologies of industrial and post-industrial society can then be represented in a triangle whose angles correspond to a fully market-ruled, a fully state-governed and a fully autonomous society: liberals, socialists and greens are interpreted as claiming that the optimal mix of activities can be reached by moving somewhat closer to one of these angles from where our societies currently are. See Van Parijs (1991). 45 Very little has been said here to establish such a presumption, which is vindicated at length in chapters 2-4 of Van Parijs (1995).

84

Activation and the Burden of Working On Instrument Choice by a Responsibility-Sensitive Egalitarian Government Frank Vandenbroucke and Tom Van Puyenbroeck*

Introduction With regard to employment policy and welfare reform, there is a large degree of consensus among policymakers and scholars that taxes and benefits must not lead to a situation in which poor individuals (or their families) face very high marginal tax rates when they take up a job or when their hours of work increase. Benefit systems that are too selective are beset by inactivity traps: they discourage the labour market participation of low-skilled workers. In academic research, various proposals, related to basic income or negative income taxation, are put forward to remedy such inactivity traps. Obviously, other approaches to the incentive problem for low-wage earners are possible, such as (a) topping up lowskilled workers’ purchasing power by selective tax credits, (b) increasing their net pay by lowering personal social security contributions on low earnings, or (c) supporting sufficiently high minimum wages for low-skilled workers by selectively subsidizing employers.These alternative instruments reflect not only technical differences, but also more fundamental differences in approach.Therefore, it is useful first to assess alternative instruments from a normative vantage point, that is, by examining the conceptions of distributive justice underpinning their use, before assessing them with more explicit reference to the particular problems created by tax and benefit systems in economies plagued with involuntary unemployment. In this paper we focus on this normative issue.Yet we start with a broader question, namely, why the problem of inactivity traps has gained heightened attention in debates on social policy over the last years.We present the idea of an active welfare state, which adopts increased participation as a central goal of social policy. In our view, this idea should also be based upon a responsibility-sensitive egalitarian conception of social justice.The active welfare state and responsibility-sensitive egalitarianism will act as the general background for our discussion of the relationship between a government’s normative conceptions and its eventual choice of instruments. Subsequently, we explain why defining a government as responsibility-sensitive egalitarian does not suffice to determine which distributive instruments it is supposed to choose. Making specific choices depends on the collective decision rule which is used and on normative positions taken by a government that do not reduce to responsibility-sensitive egalitarianism.We look at each of these in turn. Next we present and discuss a collective decision rule that is based on the work of John Roemer.We pro*

We are grateful to Alan Chipp, Loek Groot, Jeroen Knijff, Erik Schokkaert, Dirk Van de Gaer, Robert J. van der Veen, and Philippe Van Parijs for comments on earlier versions of this paper.

85

Basic Income on the Agenda ceed by discussing the impact of a government’s normative stance with regard to its conception of personal responsibility. In the remainder of the paper we emphasize the impact of a government’s conception of individual well-being on the eventual outcome of using this decision rule.While we keep the level of formal analysis to a strict minimum, we point out that a formal model can be constructed allowing for a more general analysis of the issue (Vandenbroucke, 1999a: chapter 3; 2000).We will refer to some of the results of this model insofar as they indicate how optimal policy design, with regard to the use of instruments such as labour subsidies or a basic income, is influenced by a government’s conception of individual well-being. The primary aim of such a model is to make clear how differences in a government’s normative conception lead to differences in policy choices. Such a highly simplified representation of reality is valuable, if only because it shows that, other things being equal (e.g. the economic environment), the way a government publicly values an individual’s well-being has a profound impact on the choice of instruments.We briefly discuss whether one can go further and identify criteria for choosing between the different optimal policies established by this model. Yet we also point out that such a discussion may overlook some essential elements in current debates on social policy. This point is developed more fully in the last but final section, where we return to the idea of the active welfare state and indicate in which respect the model falls short of capturing some of the active welfare state’s essential constitutive elements.We provide a short summary and conclusion at the end.

Remedying Economic Inactivity in an Active Welfare State A fruitful discussion of inactivity traps requires a broader perspective relating to social policy at large. In various countries, the very functioning of the traditional welfare state has been increasingly criticized. Briefly, doubts have been raised about the feasibility of sustaining the welfare state, not only in view of demographic changes, but also because new kinds of social risks have emerged (such as long-term unemployment for the lowskilled) and social needs (such as the need to combine work with family life), which tend to increase even further the demand for social policy intervention. Hence, current political thinking about the future of the welfare state is also characterized, not surprisingly in view of the sustainability problem, by increased emphasis on ‘getting people back to work’.This explains why the fight against inactivity traps is high on the social policy agenda. Combating inactivity traps is not the only – indeed, not even the most important – hallmark of a renewed social policy. To safeguard the welfare state, one must alter its working methods, for example, by gearing social policy intervention more towards preventive actions, by complementing social spending with social investment in human and social capital, etc.This new kind of welfare state can be characterized as an active welfare state (Vandenbroucke, 1999b), to express both the change of methods (‘an intelligently active state’) and the change of goals (‘a state of active people’, i.e. a state combining social protection with citizens’ active participation in society). Arguably, the potential sustainability problem of the current welfare state has served as an impetus to the debate on the re-orientation of social policy. However, this problem in itself cannot be taken as an ethical underpinning of the plea for increased partic86

Activation and the Burden of Working ipation.The fundamental argument should start from the recognition that participating in society can be regarded as an individual ‘functioning’, in the terminology of Amartya Sen. If one advocates genuine equality of opportunity, active participation in society should be included in any individual’s option set. Cohen states that pursuing equality of opportunity is tantamount to removing ‘obstacles to opportunity from which some people suffer and others don’t’ (1999: 354). He distinguishes socialist equality of opportunity from right-liberal and left-liberal alternatives in that ‘it seeks to correct for all unchosen disadvantages, disadvantages, that is, for which the agent cannot herself reasonably be held responsible, whether they be disadvantages that reflect social misfortune or disadvantages that reflect natural misfortune’. We believe this maxim should apply to a social-democratic, that is, an egalitarian active welfare state. Hence, the egalitarian active welfare state, so conceived, is bound to address questions of individual responsibility.Thinking about distributive justice on egalitarian lines compels us to distinguish personal ‘luck’ from personal ‘choice’. It is not surprising that the notion of individual responsibility emerged on the social agenda at the same time as the call for a new kind of activating social policy. Some of the new social risks are more predictable than their traditional counterparts, and more easily linked with personal behaviour. For example, being long-term unemployed is closely associated with having a lower level of educational achievement. Achievement in education obviously depends on circumstances, such as family background and innate talents, but also on personal effort and choice. Hence, socio-economic developments in themselves already explain why questions regarding one’s personal responsibility are showing up more than they used to do in discussions about welfare. But, as we have just indicated, these questions also arise if one limits the discussion – as we do in this paper – to more abstract thinking about normative issues. In this paper we will confine our discussion to the case where government is committed to the idea of responsibility-sensitive egalitarianism.

The Idea of Responsibility-Sensitive Egalitarianism and the Practice of Policy Making On the most general conceptual level, egalitarians sensitive to the idea of responsibility consider it unjust for individuals to be disadvantaged relative to others because of personal characteristics for which they are not held responsible.1 Justice requires ‘equal access to advantage’: if everybody has such equal access, differential advantage must reflect genuine choices for which people are responsible.2 When phrased in such a general manner, this responsibility-sensitive conception of equality provides little guidance to a government in the actual design of its distributive policies. More specific substantive questions have to be answered if one wants to understand why and to what extent a responsibility-sensitive egalitarian government would actually prefer particular instruments to achieve particular goals.What is the government’s actual conception of personal responsibility, that is, which individual characteristics does the government take the citizens to be personally responsible for? What is the government’s conception of the individual well-being of citizens, or, what constitutes an advantage according to this government? In addition, there is an operational question to be answered: how will an egalitarian government actually choose between 87

Basic Income on the Agenda alternative policies? These questions are examined in the following sections. We shall handle the operational question first.

Responsibility-Sensitive Egalitarian Policy Choices How will a responsibility-sensitive government actually choose between different policies? Let us start from the preliminary observation that the idea that distributive justice should distinguish between luck and choice fits in naturally with the idea that true equality is equality of option sets. Le Grand (1991: 87), for instance, writes: ‘(O)ur judgements concerning the degree of inequity inherent in a given distribution depend on the extent to which we see that distribution as the outcome of individual choice. If one individual receives less than another owing to her own choice, then the disparity is not considered inequitable; if it arises for reasons beyond her control, then it is inequitable’. Le Grand concludes: ‘(A) distribution is equitable if it is the outcome of informed individuals choosing over equal choice sets’.We propose taking as our point of departure the view that ‘equality of option sets’ provides a definition of equality tout court. Admittedly, our broad phrasing of what responsibility-sensitive egalitarianism amounts to, does not refer to equal option sets, but rather to equal access to advantage, where ‘advantage’ refers to a government’s evaluation of an individual’s well-being (whereby we take it that well-being can be measured in a cardinal and interpersonally comparable way).To explain the relationship between the general idea of ‘equal option sets’ and ‘equal access to advantage’, we propose to focus immediately on the issue of participation in the labour market. A group of people has identical option sets in the labour market when everybody in that group has access to everybody else’s job, for the same salary, in similar circumstances, etc. Calling this situation ‘equal access to all jobs’, one thus assumes that it holds if the only reason why individual A has a different job to individual B is because she genuinely chooses to have a different job.There are a variety of reasons why this situation is not the same as ‘equal access to advantage’. For example, people’s well-being in distinct jobs may be evaluated by a government in the same way, so that two individuals may have equal access to advantage even when their option sets are not identical. Conversely, different people may not be able to convert identical jobs into the same level of personal advantage. Nevertheless, if a government values the advantage of a job independently of the person who holds the job, then ‘equal access to all jobs’ is a sufficient condition for ‘equal access to advantage’. Throughout we will assume that a government does indeed value advantage in such a way.We will use in our numerical examples the (highly unreal) situation of equal access to all jobs (hence, equality) as a useful counterfactual, serving as a benchmark for judgements on justice. Consider then a situation in which only two jobs are on offer, fully described by the following data: job F: full-time, salary 10; job P: part-time, salary 5; zero option Z: not working, income 0. 88

Activation and the Burden of Working Call this option set I. Consider then option set II: job F: full-time, salary 9; job P: part-time, salary 6; zero option Z: not working, income 3. Let us further assume that alternative policies can give all people access to all the options in option set I (by using a policy I), or to all the options in option set II (by choosing a policy II). Hence, both policies admit equality as a result. Note that, while both admit equality, they are clearly different. For example, policy II is naturally associated with the establishment of an unconditional basic income, whereas policy I rules it out. We now want to indicate how a responsibility-sensitive egalitarian government would actually decide which option set is best for the collectivity. There is a natural way to proceed here. We could, first, determine for option set I the choice each individual would effectively make when confronted with that option set.3 Second, the government would proceed to value the benefits and burdens of this choice for each individual using an objective, synthetic measure of advantage. Third, one could average over all individuals, counting each individual for one in the weighting procedure. This last step clearly has a utilitarian flavour. More precisely, it can be seen as expressing a utilitarian conception of impartiality.The same exercise could be performed with regard to option set II, and the government would choose that policy which yields the highest average advantage. It should be noted that this ‘natural way to proceed’ builds into the conceptual framework of responsibility-sensitive egalitarianism an additional idea. That is, when making choices between policies, one should be impartial between the individuals with whom one is concerned, and simply aggregate the advantage those individuals choose to obtain.This feature captures an essential aspect of a methodology developed by John Roemer (1994; 1996a: 279; 1996b) to implement responsibility-sensitive egalitarian justice. Let us provide a more detailed account of this implementation procedure. Consider a world in which individuals differ from each other with regard to their economic productivity and with regard to their preferences over income and time for non-market activity (i.e. their propensity to work). Accordingly, one can subdivide the population into ‘productivity types’ (people having a certain level of productivity in common) and ‘preference tranches’ (people having a certain preference ordering in common). Let us represent citizens by indexed capital letters, the letter indicating a preference tranche and the subscript indicating a productivity level. So, A1 is a person belonging to preference tranche A and productivity type 1. Productivity type 1 comprises a range of citizens {A1, B1, C1,...}, preference tranche A consists of citizens {A1,A2,A3,...}. Next, we state explicitly which individual characteristics are assumed to be the subject of personal responsibility and which are not. Except for the following section, we will assume in this paper that implementing responsibility-sensitive egalitarian justice requires choosing an allocation rule that satisfies the principle ‘no individual should be put at a disadvantage relative to his fellows because of his lower productivity level; if such a disadvantage is the result of his preferences, we do not consider that unjust’. (Any model simplifies reality, of course. One may also consider productivity as a metaphor for productive talents and preferences for working time as representing economic choice.) 89

Basic Income on the Agenda Roemer then proposes to reason as follows. Ideally, what the responsibility-sensitive egalitarian would like to do is choose the policy which equalizes advantage across productivity types, but allows it to differ within the productivity types according to the choices people make concerning their working time. Or, put another way, two people with different productivity levels that have the same propensity to work should end up with the same advantage, whereas two people with the same productivity level but different propensities to work should not have the same advantage. Of course, full equality is not always feasible. In such cases, the responsibility-sensitive egalitarian has to satisfy herself with an allocation rule that maximins advantage across productivity types for each preference tranche. In other words, she must look for the policy that yields, for each tranche of preferences, maximal advantage for the worst-off productivity type. In terms of our notation, what the egalitarian government would like to do is to maximize the advantage of the worst-off productivity type within tranche {A1, A2, A3,...}, do the same within preference tranche B={B1, B2, B3,...}, etc. Now, as such a series of maximizations cannot be performed simultaneously, Roemer proposes to choose the policy that maximizes a weighted average of the minimum advantages across productivity types, where the weight assigned to a given preference tranche is its frequency in the population. Hence, if, for example, the ‘productivity 1 type’-individuals are the worst-off in terms of advantage in each of the preference tranches A, B, C, ..., N, Roemer proposes a method which maximizes a weighted average of the advantage of A1, B1, C1, ..., N1. In his own words, this position can be described as one where, when looking at a preference tranche, it is Rawlsian; among tranches, it is utilitarian, in giving equal consideration to each tranche.4 Roemer justifies this method by appealing to the fact that egalitarians who hold people responsible for their preferences must, for that reason, be impartial to variations in individual preferences in the population. Responsibility-sensitive egalitarians have to count ‘each preference for one’. We now state two basic assumptions that serve to simplify the setting. First, we assume that it is indeed the case that the worst-off citizens belong to a single productivity type (e.g. type 1, as in our examples above). This is not, of course, a trivial assumption; it depends not only on empirical facts, but also on the way one measures individual advantage. Second, it is assumed that productivity levels and preferences are distributed independently from each other.Within every productivity type one finds the same distribution of preferences; within every preference tranche one finds the same distribution of productivity types. Under these assumptions, Roemer’s procedure is straightforward and proceeds as we indicated in our examples above. First, identify the (unique) worst-off productivity type.5 Then, select the option set that maximizes average advantage for the worst-off group. Since by assumption the distribution of preferences within these worst-off groups corresponds to the population-wide distribution of preferences, impartiality between the preferences of the worst-off is equivalent to impartiality between the preferences of all individuals.6 At this point, we stress two things. First, a collective choice rule such as the one just presented is needed to guide choices between states of affairs, i.e. to turn from a situation in which a responsibility-sensitive egalitarian could only distinguish equal states from unequal states, into one where actual policy choices (between option sets) can be 90

Activation and the Burden of Working made. This need for guidance appears both in situations where different allocation rules can achieve equality (as in option sets I and II), as well as in situations where only maximin can be satisfied (where some individuals would enjoy better option sets, but where the policy that maximizes the position of the worst-off is chosen). Second, when calling this collective choice rule ‘impartial’, we actually refer to the impartial way of aggregating interindividual differences in ‘responsible traits’. Clearly, the very fact of using an objective evaluation of personal well-being,‘independently of the person who holds the job’, will introduce a bias in favour of those individual preferences most in fitting with the publicly affirmed normative valuation.

The Government’s Conception of Personal Responsibility Given that individuals have multiple characteristics, policy decisions in a framework such as the one just presented also depend on the question of which of these characteristics people are held responsible for, and which they are not. Recall that in the very simple world we consider, people have two characteristics: their level of productivity and their preferences for non-market activity. The responsibility-sensitive egalitarian way of choosing between policies can be contrasted with a case in which an egalitarian government does not consider individuals to be responsible for their preferences (next to considering them not responsible for their productivity).7 The profound difference between these two conceptions of responsibility can be illustrated starting from our option set example. Whereas both option sets depict an equal state if one holds people responsible for their propensity to work, neither of them does so if the government thinks people are not responsible for their preferences. Indeed, since the government then thinks people do not genuinely choose the number of hours they work, equality would only be obtained in our simple setting if the reward is the same for everybody. More generally, if we still take the maximin rule as expressing egalitarianism, such a government would not look at averages across preference tranches. It would choose a policy that maximizes advantage for the worst-off individual across both preference tranches and productivity types. Clearly, such a government would opt for a policy that differs from the one chosen by its responsibility-sensitive counterpart. Since such a government holds people responsible for fewer characteristics, it will in general redistribute more (referring to the example, it would prefer an option set in which one receives the same ‘salary’, whether one chooses job F, job P, or the zero option).This does not imply that it is ‘more egalitarian’ than a responsibility-sensitive government.They simply have a different conception of what equality (and compensation) implies. Before we proceed, we should emphasize that the conception of individual responsibility and the conception of advantage are two distinct normative issues that each bear on the ultimate choice of instruments. The first bears on how the government aggregates ‘something’, whereas the second determines what it aggregates. In fact, the conception of advantage would also be important if the government is responsibilityinsensitive or even non-egalitarian if its aggregation method is merely to add up advantages.Yet, it is clear that both questions are pertinent when linking responsibilitysensitive egalitarianism with government decision-making. 91

Basic Income on the Agenda To keep our exposition brief, we will focus in the remainder of the paper on the impact of the conception of advantage on the choices of a responsibility-sensitive egalitarian government. (We refer the reader to an appendix for an additional discussion about the influence of the conception of personal responsibility.) Hence, we now return to the setting in which people are held responsible for their preferences but not for their productivity.

The Impact of the Metric of Advantage on Policy Choices The two different option sets in our example instantiate a difference with regard to the underlying ‘principle of reward’. In option set I, a full-time job yields twice the income of a part-time job, whereas in option set II a full-time worker holds an income that is only one third higher than that of a part-time worker. Moreover, not working implies no income in option set I, whereas a work-independent basic income of 3 is available in option set II.We have nonetheless indicated that, if people are held responsible for their preferences, the two option sets result in equality (i.e. equal access to all jobs is assumed to be fulfilled in both cases). The question arises of whether responsibilitysensitive egalitarianism would force a choice between the two underlying principles of reward. Some have claimed that responsibility-sensitive egalitarianism entails a principle of natural reward.Thus, according to Fleurbaey (1995: 685), an ethic of responsibility conveys the ideal that a society ‘should let the agents exercise their responsibility and bear the consequences of such exercise, without trying to distort their outcomes in a particular way and with particular incentives. If there is some ‘natural reward scheme’, it should, according to this view, operate as freely as possible’.The differentials in individuals’ responsible characteristics should operate fully. One can formulate various axioms specifying a principle of natural reward. Such an axiom could be constituted by the following requirement: if all individuals are identical with regard to the traits for which the government does not hold them responsible, there must be no difference between the pre- and post-transfer distribution of resources in society. Hence, in the setting we discuss, this axiom demands that there be no redistribution when all individuals have the same level of productivity. In Vandenbroucke (1999a) it is argued that responsibility-sensitive egalitarianism by itself does not lead to the imposition of natural reward, but we will not develop this issue here.We do want to show a related point, however. On the basis of what has been said previously, we want to highlight the interconnection between principles of rewards and metrics of advantage.This interconnection is most visible when, starting from given specific assumptions, one establishes a highly simplified optimization model. Given specific assumptions, it can be shown that with the use of a Roemerian ‘impartial’ collective choice rule, principles of reward and metrics of advantage completely determine each other. Or, stated differently, if decision making by a responsibility-sensitive (egalitarian) government is based on this collective choice rule, the actual government’s conception of individual well-being will drive the choice between various ways of framing the option sets (and hence between their associated principles of reward).8 This can be illustrated by some simple calculations on the basis of our earlier example. Recall that the two option sets allowed for equality, and recall the three-step pro92

Activation and the Burden of Working cedure we indicated for choosing between them. First, we determine the choices people would effectively make in each of the option sets. Then, we assess the level of advantage associated with these choices. Finally, we average advantage over all individuals, counting each individual for one in the weighting procedure, after which the two option sets can be compared. It can already be seen that ‘valuing advantage’, which comes in at the second step, will thus be quite crucial in determining the eventual outcome. To illustrate this, we need assumptions about people’s effective choices. Hence, we assume the following responses: Given option set I, 50 percent of the population choose job F; 50 percent of the population choose job P; no one chooses the zero-option. Given option set II, 40 percent of the population choose job F; 55 percent of the population choose job P; 5 percent of the population choose the zero-option. Consider, next, some alternative public systems of metric to calculate the individual advantage associated with particular choices. In each alternative system we count the advantage of income as being equal to 1 per unit of income.The difference between the metrics is based on different conceptions of the burden of paid labour. More precisely, the government thinks that, when one does not take into account the monetary reward, paid work is on balance a burden for people.Work may be a mixed blessing: paid work brings some benefits (developing human capital, structuring one’s life,…) and some burdens (having less time for the family, less leisure, etc.).We will use ‘the burden of working’ as shorthand for the non-monetary advantages and disadvantages of labour market participation, where we assume that, on balance, the former are outweighed by the latter. We measure the burden of working in a particular job as a disadvantage expressed in numbers:9 Metric M1: disadvantage job F=0; disadvantage job P=0; Metric M2: disadvantage job F=5.5; disadvantage job P=2.5; Metric M3: disadvantage job F=7; disadvantage job P=3. Metric M1 implies that the government measures advantage only in terms of the income people have. Metric M2 implies that government conceives individual well-being as being negatively influenced by working and metric M3 associates an even higher disadvantage with the burden of working. A lower level of disadvantage can be connected with a government that attaches relatively more weight to the (positive) value of nonmonetary rewards associated with participation in economic activity.Admittedly, this is a very rough and even somewhat ambiguous way of incorporating the value of participation in society.We return to the complexity of the issue under review later on. 93

Basic Income on the Agenda Given our assumptions, calculations of average advantage yield the following result:10 Option set I

Metric M1 Metric M2 Metric M3

7.5 3.5 2.5

Option set II

7.05 3.475 2.6

Clearly, if government measures people’s well-being only in terms of money, option set I is preferable as it yields a higher average advantage in comparison to option set II. But if we increasingly value time for non-market activity (i.e. if we attach growing disadvantage to working time), there comes a point at which option set II is preferable (e.g. given metric M3), because policy I’s strong work incentives lead people to overwork on advantage-maximizing standards. Broadening the perspective to a case where more option sets are available, the example suggests that to each metric corresponds a best option set. Since each metric describes a definition of advantage, we may summarize this by saying that, given the Roemerian collective choice rule, to each governmental conception of individual well-being, i.e. to each public metric of individual advantage, corresponds a collective choice of the ‘best’ option set, and hence, a ‘best’ policy, (and a ‘best’ principle of reward). This implies that no further philosophical arguments are required within the framework of this model to explain differences in policy choices. For example, choosing option set I, characterized by the absence of a basic income, is not justified by appealing to a principle of ‘reciprocity’ (no productive contribution, no income). Conversely, if in the above setting the government chooses option set II, this is not because it appeals to deeper philosophical foundations justifying an unconditional basic income. Of course, this last remark should be put into the right perspective. It should by no means be considered as a proof of the superfluous character of such foundations per se. For instance, White (1997) appeals to a principle of reciprocity to reject basic income; following White, we may supplement the principle of equal access to advantage by a principle of reciprocity.The approach presented here does not rely on such a principle; in a sense, it boils down to ‘advantage-maximizing’ among equal-access-toadvantage structures.

A more General Assessment of the Influence of Different Normative Vantage Points on Policy Choice The conceptual apparatus presented in the previous sections can be generalized into a formal model in which government has different policy instruments at its disposal. Unless explicitly indicated, we assume in this and the following section that government uses earned-income taxation and a labour subsidy as instruments. More specifically, we consider a simple case in which government taxes labour income at a constant rate t (where we assume t £ 1).To this income tax scheme is added the possibility that government taxes or subsidizes individuals in a uniform way (independently of earned income).That is, we allow for the possibility that individuals must, next to their earned 94

Activation and the Burden of Working

Figure 1: An optimal policy track

s a1

1

a0 0

1

t

income taxes, all pay the same tax or receive the same benefit. Denoting this fixed amount as a transfer B, it follows that the complete income tax scheme studied here can take the form of a negative income tax, thus creating an unconditional universal basic income (in casu when the transfer B>0). The government can also use a labour subsidy, with the subsidy rate s proportional to the time spent in paid work.The choice between different policies therefore boils down to a choice between different instrument vectors (B, t, s). Such a model allows for a more general systematic assessment of alternative instruments from a normative vantage point than in the previous sections. It can be used to check how optimal policies – i.e. optimal choices of instruments (B, t, s) – are influenced by differences in the metric of advantage. It is not our intention to present and discuss such a formal model here (see Vandenbroucke, 1999a: chapter 3; 2000). Nevertheless, it has a clear conceptual similarity with the examples hitherto provided, and we can therefore repeat and discuss here some of the findings that can be obtained from such a model. As in the previous section, a key variable in this model is the government’s conception of individual well-being. Income enters positively into the public advantage metric; working time – or more precisely the burden of working time – enters negatively. If one scales the impact of working time on an individual’s overall advantage by a (variable) policy parameter, thereby effectively allowing for different metrics of advantage, one can construct a figure such as Figure 1.This figure displays the ‘optimal policy track’, i.e. the locus combining optimal values of the instruments s and t for different metrics of advantage. 95

Basic Income on the Agenda The horizontal axis of Figure 1 measures the earned-income tax rate t, the vertical axis measures the subsidy rate s.The bold line is the optimal policy track for these instruments: each point on the track between a0 and a1 is an optimal combination of income tax rates and labour subsidy rates, maximizing the position of the worst-off productivity type among all preference tranches. In other words, each point on this track represents the ‘best choice’ a government would make. Now each point on the track also corresponds to a different metric of advantage.At the starting point a0 the government considers paid work a heavy burden, giving it a heavy weight in its measurement of people’s advantage. At the other end of the track, at a1, the government thinks paid work is not a burden at all.At that point it only considers people’s income when measuring their advantage. In other words, at a1, maximizing a citizen’s advantage amounts to maximising that citizen’s income. Hence, on the basis of such a model, one can draw the following conclusion regarding the relationship between the normative stance of a government (in casu the way it values time for non-market activity) and its eventual instrument choices: a. The optimal earned-income tax and labour subsidy rate increase as labour counts less as a burden in the government’s conception of people’s individual advantage. It follows that if a government’s conception of individual well-being narrows down to a ‘money’s all that matters’ stance, then it will induce people to spend as much time as possible on the shop floor. Let us next consider by way of example those cases where the tax rate is at its (assumed) maximal value of 1. The optimal policy track still prescribes higher labour subsidy rates for increasing values of a.11 In view of the observation that a higher a can be interpreted as capturing a government with a more outspoken emphasis on the positive value of participation in economic activity, the fact that such a government continues to subsidize a working individual, even if t is fixed, does not come as a surprise. To understand conclusion a intuitively, it is important to note that it is indeed plausible that a government with a higher a would choose to subsidize labour at a higher rate, since this ceteris paribus leads to higher advantage (the underlying model takes it that people will work more if they earn more). Both an individual’s disposable income and her hours worked increase with a higher subsidy, and the amount of work performed yields less disadvantage when a increases. Of course, granting more subsidies requires the government to increase its revenues, in order to keep its budget balanced. Increasing revenues implies raising t, or lowering B, or both. Suppose that government would then finance the increased subsidy entirely by increasing the uniform tax (or, if this transfer has the opposite sign, by lowering the unconditional basic income) B.This would run counter to the idea that the earned income tax instrument allows us to compensate for differences in individuals’ productivity, which, to recall, is a characteristic that is not considered as being within the realm of individual responsibility. For that reason, the rise in s that follows from a higher a will be accompanied by a rise in t. The fact that earned-income taxation rises as the burden of working becomes less important may seem a bit odd by itself. But, as just indicated, this feature follows from the fact that the government has more instruments at its disposal.To see this, note that 96

Activation and the Burden of Working Figure 2: 2:An an optimal optimal policy policy track track when when only only taxes taxes are are available available Figure

s 1

0

a0

a1

t

1

Figure 2: 3: An an optimal policy track with constrained subsidy level

s 1

a

a1 s*

a0 0

1

t

97

Basic Income on the Agenda there would not be a rise in the earned-income tax rate if the government could not observe an individual’s working time and could therefore not use the subsidy instrument. Indeed, in such a setting, it can be shown that the optimal tax rate decreases as a increases.That is, when no subsidies can be used it is optimal to let the net reward for working increase – hence, let the tax rate decrease –, as the burden of working becomes less important in the public metric of advantage. This case is illustrated in Figure 2. (Intuitively, this can be compared with our earlier numerical example: metric M3 considers the burden of work relatively more important than metrics M1 and M2. On the basis of metric M3 the government selects an option set with a lower net reward for working). Finally, Figure 3 displays the result for an intermediate case in which there is a constraint of the type s £ s* (covering for example a setting in which a government pays a limited wage subsidy – which benefits low-paid workers relatively more, but which is universal qua technique – to avoid training disincentives).The outcome is a mixture of the unconstrained case considered in Figure 1 and the regime without labour subsidies as captured by Figure 2.

Basic Income And what about B, the constant term in the labour income tax scheme, which represents a universal unconditional basic income in those cases where B > 0? A balanced budget requirement allows one to consider B as a ‘residual’ instrument, the value of which could be retrieved by solving B(t,s) once the other optimal values for the instruments have been found. It must be noted that, in principle, a policy can be devised that would keep a given basic income level unaffected. That is, a policy can be designed which fulfils the requirement that any rise in government outlays for labour subsidies is exactly offset by a rise in earned-income taxes, so that B neither decreases nor increases. The crucial question is whether such a B-preserving policy would be compatible with the optimal policy track as displayed in Figure 1, which already links the level and movement of the two instruments s and t. Stated otherwise, if s is required to increase with t along the optimal policy track at a more rapid rate than necessary to keep B at a given level, then basic income will fall, given an upward move along the optimal policy track. Or to put it in yet another way, basic income will then decrease as the normative policy parameter a rises. It can be proven that this will indeed always be the case. In the neighbourhood of the optimal policy track – which depicts an optimal relationship between s and t – an increase in the subsidy leads to a lower level of B (which still is, given the government’s objective, an optimal level).Two conclusions can thus be drawn as regards the relationship between basic income and the responsibility-sensitive egalitarian government depicted in this paper: b. The government will, for a given set of instruments and constraints, propose a lower basic income level as the burden of work in the government’s conception of individual well-being becomes less important (in other words, the more exclusive the importance the government attaches to income, the lower the basic income will be). 98

Activation and the Burden of Working Figures 4a,b: The conception of well-being and optimal instrument values (a: no constraint on s, b: constraint s £ 0.25)

0,9 0,7 0,5

t s B

0,3 0,1

a

-0,1 -0,3

M -0,5

0,9 0,7 0,5 0,3 0,1

s t B

-0,1 -0,3 -0,5

c. Given a set of instruments and constraints (particularly given the balanced budgetrelationship between B, t, and s), there is an irreducible conflict between the level of the labour subsidy and the basic income level. An alternative statement of conclusion b is that a lower basic income level is associated with a government valuing more highly the non-monetary rewards associated with 99

Basic Income on the Agenda participation in economic activity. Referring to our discussion in the previous section, conclusion b implies that, in general, a rise in s will not be exclusively financed by a rise in t.What underlies this result is the fact that a change in the metric of advantage entails a change in the overall reward structure for paid labour, which can be achieved by shifting the balance between income taxation and the unconditional transfer B. Finally, we point out that conclusion b also holds when the model operates with an additional constraint on the level of s.12 A fortiori, it also holds for the restricted model without labour subsidies (i.e. where s is restricted to the value zero, as in Figure 2). Again, the underlying reason for this result is that one seeks to increase the net reward for working when the burden of work is regarded as less important. To conclude the discussion of this more general model of optimal instrument choice, we offer an alternative graphical representation of the conclusions identified in this and the previous section. In Figures 4a and 4b, the value of a Î [0, 1] is measured on the horizontal axis: from left to right, the weight of the burden of working declines in the normative conception of people’s individual well-being. The optimal values of the three distributive instruments as identified by the model (conditional on the parameter values we used for this specific simulation) are given for the case that was previously considered in Figure 1, as well as for a mixed scenario as depicted in Figure 3. Comparing the optimal policy track of Figure 1 with its counterpart in Figure 4a, note that the labour subsidy rate s, indeed, keeps rising (initially) even when the earned-income tax rate t is at its upper bound.The fact that the optimal policy track in the unconstrained regions of Figures 1 and 3 is a straight line is mirrored in Figures 4a and 4b, since in such regions the t- and s-curves are related through a linear transformation. Our conclusion a can be inferred directly from the alternative representation in Figure 4a, whereas Figure 4b shows the reversal in the movement of t as soon as s is constrained. Focusing on the work-independent transfer B, one sees that, initially, we have a basic income, but as a increases the unconditional transfer decreases.This provides an illustration of conclusion b.As we have stated earlier, and as is shown in Figure 4a, B can even become negative, that is, a poll-tax. Overall, the two figures also reveal the ‘residual’, budget-balancing character of B, thus providing a graphical illustration of conclusion c.

An Optimum Optimorum? A model such as the one outlined produces a systematic ‘catalogue’ linking different distributive policies with different public conceptions of advantage and/or with different conceptions of personal responsibility. It thus adds flesh and bone to the generic concept of a responsibility-sensitive egalitarian government.To repeat, for the simple setting outlined above, a government that gives little weight to time for non-market activity will, other things being equal, opt for a policy with higher labour subsidies and a lower basic income. Or, as stated earlier, it is the government’s stance towards the nonmonetary advantages and disadvantages of labour market participation that can ceteris paribus be considered as the driving force between different policy choices. Some may ask whether there exists something like an optimum optimorum on the optimal policy track.That is, whether the different optimal policies are not merely different per se, but differ in offering greater or lesser justice. In other words, one could 100

Activation and the Burden of Working take the findings of the model as a point of departure and then ask whether a policy favouring labour subsidies over basic income offers greater or lesser justice than its opposite. Within the framework of the model, this requires one to focus on choices between different metrics of advantage. In this respect, we first point at a negative result that follows from conclusion c identified above. Conclusion c is important with respect to the operational value of the real freedom inclusion rule discussed by Van der Veen (1997; 1998) in the context of the broader debate on basic income.Van Parijs (1995) defines real freedom as the ability to do what one might want to do (irrespective from what one actually does).Van der Veen (1997; 1998) suggests an ordering of option sets on the basis of this concept:‘a person’s real freedom is said to improve from one regime to another if and only if his choice set unambiguously expands…If someone’s choice sets in two regimes contain non-overlapping income-leisure combinations, it follows that the extent of his real freedom in this regimes cannot be compared’. (1997: 276-277) The irreducible conflict between s and B when the metric of advantage changes results in any pair of alternative option sets (defined by t,s, and B) not being comparable on the basis of the real freedom metric.The real freedom inclusion rule has consequently no operational value as a means to help someone choose between alternative definitions of advantage.13 An alternative approach starts from the observation that metrics of advantage and principles of reward determine each other, given the adoption of a Roemerian ‘impartial’ collective choice procedure and specific assumptions of the model under review. If one accepts the argument stated earlier that a principle of natural reward is intimately linked to responsibility-sensitive egalitarianism, then one can use such an (axiomatic) principle to identify its associated advantage metric, which we may then call the ‘neutral conception of advantage’.This would evidently amount to the identification of a specific point on the optimal policy track, given the parameter values in the complete model. While we have indicated above that one may raise fundamental doubts about the fact that responsibility-sensitive egalitarianism necessarily entails such a principle of natural reward, it is still true that someone could wish to rely on it to select a specific policy. When this is the case, the following result becomes highly relevant (see Vandenbroucke, 1999a). If government opts for a neutral conception of advantage – by adhering to the axiom that there should be no redistribution when all individuals have the same productivity level –, and if it can achieve equality by using earned-income taxes and labour subsidies, then it will not cash out a basic income. In other words, it follows from a model such as the one outlined above that a neutral government that can achieve equality (as its optimal policy), will use earned-income taxation exclusively to fund labour subsidies.14,15 Of course, it is clear that if one introduces an external argument for choosing a metric of advantage, a particular point on the policy track may be ‘justified’. But, to engage in such an exercise against the background of real-life debates on social policy carries the danger of missing another, more important point.To wit, that the underlying model is extremely simple. We limited our analysis to redistribution via a (Walrasian) labour market in this paper.The only individual endowment that was considered is ‘internal’, namely, the individual’s productivity level. Obviously, individuals 101

Basic Income on the Agenda also have external endowments, such as houses or capital, yielding an income that can also be redistributed in a welfare state. Moreover, active participation in society is not limited to participation in the labour market.With specific reference to the idea of the active welfare state mentioned in the introduction, it is therefore essential to consider what this model does not capture.

Beyond the Model: Policy Choices in the Active Welfare State Let us return to the idea of the active welfare state and its ethical foundations as they were identified in the first section, and confront them with the set-up of the model. It is important to point out that the model reduces the discussion on participation to a simple dichotomous choice. ‘Increasing participation’ in the model only refers to participation in the labour market. Of course, such an interpretation is appropriate if one wants to study inactivity traps.Yet the perspective should be broader in current discussions on social policy. Our goal should refer to participation in society, rather than to a narrower interpretation confined to labour market participation. Indeed, one should recall that the ethical justification of the active welfare state considers participation in society as one of the social bases of respect and self-respect, thereby making clear that one should not merely look at participation in the labour market. For, evidently, the social bases of respect and self-respect can also be strengthened through other, non-market activities such as taking care of a child, a parent, a friend, voluntary social or cultural work, etc. One could claim that such activities are nevertheless subsumed in the model, more specifically through the individual advantage associated with time for non-market activity. Still, this is a rough and indirect manner to treat them. It is a rough manner because it does not allow us to distinguish between various forms of non-market activities – e.g. ‘pure leisure’ versus voluntary social work, versus taking refresher courses, etc. – which could be valued differently by the government. Moreover, such valuations would then again be dependent on normative choices (relating to the question of to what extent such activities are considered as bases for respect and self-respect), which in this instance could be captured in principle by some multidimensional metric of advantage. It is an indirect manner because the advantage of time for non-market activity appears in the model only as the mirror image of the disadvantage associated with the burden of paid labour (which turns the model into a simple dichotomous model). More generally, the model lacks an explicit valuation of participation in social activity, be it in or out of the formal labour market. Indeed, also with regard to participation on the labour market, there is no direct reference to its capacity as a social basis of respect and self-respect. In short, it would be erroneous to consider the model as one that describes how different conceptions of participation would influence policy design. Put into the right perspective, it is a model which is suited to capture the influence of public valuations of some specific qualitative aspects of paid labour (such as stress, or its opportunity cost in terms of time for other activities), but it does not address the importance of participation per se. Finally, the model is very much a simplification because it has no intertemporal dimension and, therefore, can only incorporate a limited set of distributive instruments. The broader conception of participation in society requires one to think about the 102

Activation and the Burden of Working most appropriate combination of work and non-market activities over the entire life cycle (e.g. should we not design policies such that, as compared to the current situation, people work less between ages 25 and 40, and more between ages 55 en 65?). A ‘relaxed labour market’ would allow people to leave and re-enter the labour market easily (e.g., in order to care for someone, to update their qualifications, or simply to replenish their reserves; see Vandenbroucke, 1999b). The model we have presented cannot capture this conception of a relaxed labour market. Assuming that it could capture it, and neglecting the other problems we have just discussed, it is not at all clear whether a government granting a high value to an individual’s time for non-market activity would still opt for a high basic income, or whether it would instead favour other policies such as attaching to every job a right to full-time or part-time sabbatical leave, which acts as a more direct (qualitative) supplement to the monetary reward of paid labour.

Summary and Conclusion ‘Getting (poor) people back to work’ is an important concern in many of today’s welfare states, although opinions differ on why and how one should achieve this goal. It is precisely this matter of differing opinions which we wanted to focus on in this paper. We have shown how the publicly affirmed relative valuation of income and time for non-market activity drives the choice of a responsibility-sensitive egalitarian government between earned income taxes, labour subsidies, and an unconditional basic income. The model, we believe, can illuminate the choice between the instruments it considers. For example, it highlights the irreducible conflict between the level of the labour subsidy and the basic income level, or points out the absence of basic income if the government is committed to a neutral conception of advantage and can achieve equality.Yet, it is still a limited model. It is, in particular, too limited to offer an adequate description of an active welfare state, which is also founded upon the notion of responsibility-sensitive egalitarianism, but rests on a broader interpretation of participation as a goal for society and allows for a larger set of policy instruments. Hence, one should avoid extensive optimum optimorum-debates within the confines of a limited model.We nevertheless think it is worthwhile to convey the basic message that a discussion about how to get rid of inactivity traps is not merely a matter of choosing the most appropriate instrument given economic circumstances.We have to make a complex political choice on at least two issues: a conception of responsibility, and, as we stressed in this paper, a conception of individual well-being. As we have mentioned in the previous section, if a richer model were to be constructed, it would ultimately need to address the same kind of issues in a multidimensional way. Therefore, even if one constructs a model that directly addresses the broader idea of participation and/or allows for more instruments, one still faces the issue that the way in which one measures the quality of life or the level of advantage – in particular that of the worst-off – co-determines policy choices with regard to distributive instruments. ‘Being a responsibility-sensitive egalitarian’ is no sufficient basis for guiding policy choices. Implementing policies presupposes that one makes choices concerning the metric used to evaluate the quality of life; a choice that can only emerge from democratic debate in general and from the political process in particular. 103

Basic Income on the Agenda Finally, it serves to note that such choices are implicitly made in daily practice, such as through the use of poverty and inequality measures that serve as benchmarks for social policy.To the extent that these benchmarks are exclusively based on the distribution of income or wealth, they presuppose an extremely biased conception of individual well-being. Indeed, just as it has become generally acknowledged that GDP is not a very adequate indicator of economic development, this model carries the message that monetary inequality measures are at best only partial indicators of distributive justice.

Appendix The Government’s Conception of Individual Responsibility The model in Vandenbroucke (1999a; 2000) also systematically compares different normative conceptions of individual responsibility. With reference to the figures in the main text, the difference between a government that holds people responsible for their preferences and a government that considers individual preferences as non-responsible traits can be equally captured by a difference in the parameter a. Specifically, except for some limiting cases, a government that does not hold people responsible for their preferences will have a lower a. Hence, the results that were identified above mutatis mutandis carry over to a case in which one analyses differences in normative conceptions regarding personal responsibility.We can consequently state the following conclusions: a’. All things being equal, a government that considers people responsible for their preferences concerning time for non-market activity will propose a higher earnedincome tax rate and labour subsidy rate than a government that does not hold people responsible for their preferences.16 b’. For a given conception of advantage, and given a set of instruments and constraints, a government that considers people responsible for their preferences will propose a lower basic income level than a government that neither holds people responsible for their productivity nor for their preferences for working. Obviously, the choice of instruments is not exclusively driven by normative conceptions, but by the economic environment as well.To illustrate this, we reproduce here a simple formal expression for the optimal earned-income tax rate in a regime without labour subsidies.The optimal t in this case depends on two variables, the policy parameter a, and a factor J that captures pre-tax inequality: t=

0.5 - a J 1-aJ

where

J=

(wL / w) 2 s w2 / w 2+1

with w denoting the citizens’ individual productivity. Hence, J is a factor that decreases with two measures of inequality: (i) the productivity level of the worst-off, wL, divided by average productivity, and (ii) the coefficient of variation of the productivity levels. In line with our previous discussion, this expression shows that t decreases as the policy parameter a rises. But it also shows that, in a regime without labour subsidies, earned income taxes will decrease when inequality (as measured by J) decreases. 104

Activation and the Burden of Working

Notes 1

2 3

4

5

6

7

8

9 10

11

12 13 14

We are well aware that behind this phrase lies the difficult debate as to what ‘responsibility’ means, i.e. whether one is responsible for a certain characteristic because one is actually ‘in control’ of it, or whether one is held responsible for that characteristic on the basis of a mutual agreement. This succinct statement is inspired by Cohen (1993: 28; 1989: 916). This first step is the ‘natural way’ for economists to proceed: they traditionally assume that the real value of an option set lies in the best use that can be made of it, which is under their usual assumptions the use that is actually made. This can be criticized from a freedom perspective, i.e. it can be said to neglect the importance of flexibility for individuals (the importance of having options which are not actually taken up, but which might be taken up given some change in preferences). The expression ‘utilitarian’ may be somewhat misleading in this context since we are not comparing utilities, but a publicly affirmed objective notion of advantage. Nevertheless, it clearly reveals the similarity with the utilitarian conception of impartiality. Note that the simple example deals with a case in which two equal states are compared. Therefore, in this example everybody belongs to the worst-off (or best-off) type, implying that we will have take into account the effective choices of the entire population to calculate average advantage. This is notably the way we proceed when choosing between the two option sets of the example (as can be seen in more detail in note 10). Moreover, if these basic assumptions hold true, Roemer’s approach yields the same results as those obtained through a slightly different implementation procedure proposed by Van de Gaer (1993) and Van de Gaer, Martinez, and Schokkaert (1998). On this, see Vandenbroucke (1999a). At the risk of introducing some confusion in terminology, such a government could be labelled ‘Rawlsian’, since Rawls in his mainstream exposition rules out individual responsibility for ‘work effort’ (Rawls, 1971: 312). Of course, as we mentioned at the end of the previous section, the conception of advantage will influence the choice between policies as soon as ‘advantage’ is an argument in the social objective function of a government. This implies that we allow for the inclusion of ‘time for non-market activity’ in the Rawlsian bundle of social primary goods. For instance, for the entry combining option set I with metric M2, the average advantage is found using the formula 0.5 x (10-5.5) + 0.5 x (5-2.5) = 3.5. Or, given option set II and metric M1, using (0.4 x 9) + (0.55 x 6) + (0.05 x 3) = 7.05, etc. It can be seen on the figure that the track eventually stops at the point where s=t=1. This feature follows from the assumption that t can be at most 1, and from a labour supply constraint in the complete model ensuring that an individual cannot work more than a certain maximum number of hours.The latter requirement is fulfilled if the subsidy is not set higher than the income tax rate. Cf.Vandenbroucke (1999a). Referring to the appendix, it also cannot help to choose between different conceptions of personal responsibility. As the model in Vandenbroucke (1999a) allows for another revenue source, viz capital income taxation, and for spending on education, the result as identified there states that, under the conditions identified in the main text, there will be no positive basic income

105

Basic Income on the Agenda unless there is a positive residual left after having funded educational expenditures generated by the capital income tax. In other words, if this residual were negative, B