Social Democracy and Rational Choice

Henry Milner

• Publisher: Routledge
• Year: 2002
• Print ISBN:
  9780415116992
• eBook ISBN:
  9781134817702
• Topic: Economics
• Subtopic: Comparative Politics
• Index: Economics

Is it possible in this post-socialist world, for equity and efficiency to be reconciled ? Or is a productive welfare state a contradication in terms ? This book addresses these questions in theory and in practice, using the Nordic countries as its case study.
Social Democracy and Rational Choice will appeal to readers interested in comparative institutional and policy analysis, and in particular to those concerned with the future of the welfare state and the latest developments in the Nordic countries.

Part I

TOWARDS A SOCIAL-DEMOCRATIC POLITICAL ECONOMY

What guidelines could a society composed of rational individuals operating within a market economy follow in designing and maintaining institutions that distribute wealth in such a manner as to eliminate poverty and reduce inequality? To ask the question is to answer it: none are readily to be found. Social democrats rarely look to mainstream political economy; and mainstream political economy seldom addresses the egalitarian considerations social democrats bring to bear. Yet, although there is no rational-choice egalitarian school of political economy or anything comparable to be found, important, if often unconnected, developments are taking place. A potentially fruitful, though as yet uncertain, intellectual process is gaining strength.

We noted in the Introduction that rational-choice theory is increasingly breaking out beyond the economic realm to explain the workings of political and social institutions. As rational-choice theory occupies parts of this terrain, it finds itself forced to defend its flanks and, even on its home economic terrain, to take into account certain wider social concerns central to the work of the social theorists whose terrain it is invading. The scene is thus set for important theoretical developments in political economy. Though not necessarily inspired by egalitarian considerations, these developments nevertheless break open new intellectual space for potential advances in social-democratic political economy.

Rational choice's crossing over the indeterminate boundary between economics, the imperial' social science, and the ‘softer' social sciences was precipitated both by the intellectual vacuum left by the collapse of Marxism and the proliferation of electronically transmitted and computed information. Its application has generated significant rethinking of fundamental theory at the centre of the disciplines of political science and sociology and affected work in philosophy and psychology. In their eagerness to jump on the bandwagon driven by mathematically expressed propositions about society testable through computerized data – a territory heretofore occupied almost exclusively by those formally trained in economics – these new political economists are producing a great deal of quantitatively based research. Some is of doubtful quality and without lasting value; yet much may yet prove relevant to development of the theoretical framework required for a social-democratic political economy.

But the process is just in its early stages. Contemporary neoclassical economists increasingly invade the other social sciences' terrain, seeing economic man at work (and play) just about everywhere – ‘the economic approach is a comprehensive one that is applicable to all human behaviour' (Becker, 1986: 112). But economists continue for the most part to ignore the work of social scientists, producing elegant mathematical models based on assumptions which severely limit their application to the actual behaviour of individuals and organizations.

In those areas of research where economics and the other social sciences meet, however, promising developments are taking place (Swedberg, 1990). At the heart of these developments is the widening application of rational-choice theory to the workings of institutions, ‘the humanly devised constraints that shape human interaction' (North, 1990a: 182). Individual preferences and institutional constraints are coming to be understood as linked through a synthesis: ‘institutions are viewed not only as important determinants of preferences over alternative actions, but also … as endogenously determined by individual preference, tradition, and transaction costs' (Ordeshook, 1990: 25). In the course of these next chapters, reference will be made to scholars at the forefront of these efforts, such as Douglass North, Jon Elster, Elinor Ostrom, James Buchanan, Fred Hirsch, Kenneth Shepsle and, especially, Mancur Olson. North, who shared the 1993 Nobel prize, well formulates our theoretical preoccupation.

Institutions are a creation of human beings; hence our theory must begin with the individual. At the same time, the constraints institutions impose on individual choices are pervasive. Integrating individual choices with the constraints institutions impose on choice sets is a major step towards unifying social science research.

(North, 1990a: 5)

In belatedly recognizing the work of Ronald Coase with the 1992 prize, economists were moving away from the narrow neoclassical paradigm that took the institutional context as fixed, given and exogenous (Shepsle, 1989: 132).

1

THE POLITICAL ECONOMY OF DISTRIBUTIVE JUSTICE

ECONOMICS AND INSTITUTIONS

Political economy is built on the individual choices and actions of human beings, that is, the assumption that the individual is the ultimate arbiter of his or her welfare. This utilitarian democratic postulate states simply that individuals, acting according to the logical prescriptions of rational action (see Chapter 2), seek to efficiently achieve their aims. Institutions facilitating such actions are said to be efficient.¹

There is a renewed interest in institutions among social scientists consistent with the approach of political economy. The form and structure of institutions are increasingly being analysed for their effect on policy outcomes, no longer dismissed as mere by-products of the conflict of interests or classes.

Individual action is assumed to be optimal adaptation to an institutional environment, and the interaction between individuals is assumed to be an optimal response to each other. Therefore the prevailing institutions (the rules of the game) determine the behavior of the actors, which in turn produces political or social outcomes.

(Tsebelis, 1990: 40-1)

For example,

if the institutional constraints result in the highest payoffs in the economy being criminal activity, or the payoff to the firm is highest from sabotaging or burning down a competitor, or to a union from engaging in slowdowns and make-work, then we can expect that the organization will be shaped to maximize at those margins. On the other hand, if the payoffs come from productivity-enhancing activities, then economic growth will result.

(North, 1990b: 396)

This ‘new institutionalism' is thus gradually removing the artificial barrier set at the end of the last century between political science and economics. Economists no longer assume that once their methods identify rational policy choices, political institutions ‘rationally' implement these. Political science can no longer consign economic choices to a lower order of concern distinct from the processes resolving the great issues that politics addresses.

It is increasingly accepted that the analysis of policy choices within the ‘Pareto-optimal' framework is constraining when institutions are taken into account. Pareto-optimal transactions are defined as those where at least one individual is made better off, but no individual is made worse off; thus Pareto-optimal assessments serve to identify rational, that is, welfare-enhancing, transactions, but in a narrow arena. Once the choice concerns the rules governing transactions, that is, institutions, Pareto-optimality leads to few interesting outcomes. The challenge facing political economy is to widen the parameters for assessing institutional arrangements without degenerating into subjectivity and ideology.

For example, Pareto-optimal considerations would leave the choice to employers as to whether to train their workers – a choice resulting in an increase in aggregate welfare. But institutional arrangements get in the way. In a free labour market, entrepreneurs can ‘poach' workers trained at great expense by their competitors. Such poaching is by definition Pareto-positive since both the worker and the new employer voluntarily enter into the ‘poaching' transaction (except where the knowledge that she² might have to pay the external costs of this transaction discourages the original employer from providing the training in the first place). The overall effect of this state of affairs on aggregate welfare is thus hardly positive. In the language to be developed here, the theoretical challenge is to incorporate such institutional or ‘second-order' considerations into political economy. Applications of game theory, as we shall see, provide a fruitful means of taking up this challenge.

EQUALITY AND EFFICIENCY IN ECONOMICS

Once we are in a position to incorporate second-order considerations, we can be concerned not only with economic growth in a democratic community but also its capacity to distribute resources fairly, and thus to take rational-choice theory into areas of political economy traditionally dominated by the left. In order to incorporate redistributive justice into political economy we need to add a second postulate. Rational-choice political economy is built on an economic expression of the first postulate, that of democratic utilitarianism, or efficiency. The straight definition of efficiency usually invokes the notion of Pareto-optimality. An outcome is efficient if it is impossible to increase the welfare of one individual without reducing that of another. According to this definition, a measure, the effect of which is to take a dollar from the richest individual and provide two dollars to the poorest cannot be presumed to be efficient.

Such a measure, however, is sometimes defined as ‘potentially Pareto-optimal' since it places the community in a position to compensate the rich man made worse off. The notion of potential Pareto-optimality thus opens space for redistribution, but only limited space. For if the amount available to the poorest individual were only one dollar, the transaction could not be efficient. In order to add a second postulate alongside that of efficiency, one of equality, we must transcend even potential Pareto-optimality. The argument has already been broached in the Introduction through alternate distributions A, B and A' presented in Figures 0.2 and 0.3. Because of diminishing marginal utility, welfare is increased when a greater proportion of a fixed aggregate income is in the hands of the poor. The reasoning behind this is disputed by few professional economists (Olson, 1990): the marginal utility of an additional unit of income decreases with the amount the individual receives, and vice versa. From such reasoning we can make interpersonal comparisons of utility since, while we cannot presume that poor individual X will gain greater utility from the dollar than rich individual X, we can presume that the average utility for all the poor from this additional dollar will be greater than for all the rich.³

Such considerations have generally not been incorporated into mainstream political economy. It is not, I suggest, a matter of excluding normative factors. Adding the second postulate does not make political economy normative. The second postulate is indeed normative, but so is the first. Any theory based on utilitarianism is normative. If I as a believer in the natural superiority of the landed aristocracy, or the rich, or better educated, or more fair skinned (and these were hardly extreme views at many points in world history) reject the first postulate, there is no non-normative means of convincing me of the contrary. The same is true if I am convinced that the pursuit of economic growth – the natural outcome of the efficiency postulate – is wrongheaded since it comes at the expense of man's divine mission on earth. It is ultimately no less normative to take the maximization of individual utility as axiomatic than it is to start from the desirability of redistribution based on the postulate of equality.

Why then has practice placed the first postulate into the realm of science and the second into that of morality? Leaving aside the explanation that economists are part of some international bourgeois conspiracy, we come upon the simple answer. Political economy is a hard business, and it is easier to limit comparisons to Pareto-optimal distributions and build a body of work upon only the single postulate of utility maximization or efficiency than to attempt to synthesize two potentially contradictory postulates. Adding a second postulate, namely that of equality, to the single postulate of efficiency complicates the model. If efficiency is the only consideration, then the mathematical process of translating individual choices into social choices allows for more elegant theoretical models; hypotheses can be more readily formulated mathematically. Indeed, the elegant geometrical figures that grace the published work of economists, in which welfare gains and losses are captured by the different geometrical shapes formed where lines and curves meet, are meaningful in terms of a single postulate of utility. Incorporating any consideration of equality in distribution confuses the picture.

The above is not meant to deprecate work in the mainstream which can and does sufficiently approximate real-world behaviour and attitudes to generate useful explanations and even applications. A great deal of knowledge has been accumulated about the relationship between institutional arrangements and economic growth under given conditions. Work in political economy which has spurred on the richest explanations, investigations and applications has sought to identify and analyse those institutional arrangements compatible with economic growth pure and simple. And there is some truth in the idea that considerations of distributive justice take a higher billing as the level of overall wealth rises. Thus in many places in the world, including the ruins of the state-run economies in Eastern Europe and the former USSR, applications based on such considerations are justifiably being sought out. But, at least among the mature industrial democracies (the OECD countries), analyses based on economic growth to the exclusion of any concern with distribution are inadequate.

It is thus considerations of practicality and not ideology that mainly account for the fact that no systematic social-democratic political economy has emerged. Apart from the difficulty of the task, there is, I suggest, no intrinsic obstacle to the development of a political economy applicable to the institutional arrangements of industrial democracies based on the combined postulates of democratic efficiency and distributive justice. As noted in the Introduction, the two postulates are clearly in contradiction at the extremes, where both equality and inequality inhibit efficiency. But within the limits set by the most and least redistributive existing industrial democracies, there are clearly a number of sustainable outcomes, or ‘structurally induced equilibriums' (Shepsle, 1989: 137) based on different systems of institutional arrangements.

The underlying premise was stated in the Introduction: people in a democratic industrial society, given the opportunity to choose between living in one of two roughly equally rich societies, could be expected to choose the one where wealth is distributed more equally and poverty is less extensive. When we buy insurance, we give up consumption when resources are greater in order to have them available when resources may be lesser. Similarly, since we cannot know whether we will be in a stronger or weaker position in the future, we rationally protect ourselves (and our loved ones) through supporting redistributive policies. Rational individuals favour efficient public social insurance programmes that redistribute income within lifetimes (such as pensions or unemployment insurance), since in this manner lifetime utility is maximized.

Social insurance is a good example to illustrate the complexity of goals being pursued by an efficient social-democratic innovation. A well designed social insurance programme seeks to correct market failures (e.g. myopia, the tendency to discount the future too heavily) and hence improve on the efficiency of the private market. At the same time, it makes possible a measure of redistribution consistent with most people's preferences. But altruism will not sustain inefficiencies in redistribution. Well-designed social security systems achieve internal efficiency through co-insurance or some other means of discouraging free riding, similar to that in private insurance. The seller of insurance reduces the price of the premium by adding a deductibility provision requiring those insured to pay the first part out of their own pockets. This is a disincentive to reduce what is known as ‘moral hazard', the increase in the likelihood the contingency insured against will occur because of the insurance. Moral hazard is faced by insurance companies insuring against acts (like car accidents) that can be voluntarily self-inflicted (Olson, 1987: 216).

There is a second-order logic here: without recourse to co-insurance, insurance premiums would increase significantly to pay for self-induced claims (or for the investigators needed to ferret them out), and hence fewer people could afford to protect themselves against adversity. This result is anti-efficient, but also anti-egalitarian as those left out are usually the most vulnerable. It also sets in motion a logic that erodes the cultural base for socially responsible and potentially altruistic behaviour, since it tells those who are not ‘ripping off the system that they are the ones being ripped off in high premiums.

Rational choice teaches us that we cannot ignore the need to reward wealth-promoting risk and to punish unproductive rent seeking. One ignores incentives and disincentives at the cost of diminishing the pie out of which pieces are to be cut to help the needy. Americans ended up paying hundreds of billions for bad debts incurred by savings and loan companies after the Reagan Congress allowed the savings and loans to invest just as they wished and still guaranteed the depositors' money. Similarly, the Scandinavian Social Democrats opened up financial markets in the 1980s which enabled the banks to speculate on property under a taxation system that rewarded borrowing money to invest in real estate. The result was a bubble the recent bursting of which forced the governments, at great cost, to bail out several banks.

Americans pay exorbitant medical fees to cover insurance premiums for medical practitioners at the mercy of a system of litigation which (through allowing lawyers to be paid on a contingency basis) places no disincentives on taking court action. In Sweden, sick-leave and long-term disability policies failed to take into account that workers might call in sick when healthy in the absence of any financial disincentive, especially when their employers had no financial incentive to monitor their actions. Yet sick leave was the exception that proves the rule for, as a rule, people pay deterrent fees when receiving medical services, prescription drugs, etc. in the Nordic countries.

Undoubtedly the intellectual challenge of finding complementary applications of the two postulates is immense, for we seek not just to explain, but to generate guidelines for action of real human beings, what Hargreaves-Heap (1989: 210) calls ‘useful theory'. Generating such guidelines makes exacting demands on the theory. In order to show that an institutional goal is feasible, we must show that at every practical stage in the formation of the desired institution, institutional requirements are compatible with concurrent individual behaviour. If at any stage the institutional structure is not compatible with the behaviour of individual agents, the former ...

Table of contents

• Front Cover
• Social Democracy and Rational Choice
• Title Page
• Copyright
• Contents
• List of figures
• Preface
• INTRODUCTION
• Part I: Towards a social-democratic political economy
• Part II: From theory to practice
• Part III: The prospects for Nordic social democracy
• Conclusion
• Notes
• Bibliography
• Index