Inequality

Abstract

In the museum of the Louvre, in Paris, sits “The Raft of the Medusa”, one of the most celebrated paintings of the nineteenth century, the work of Theodore Gericault. Here it is, in reproduction (Fig. 18.1):

Appendix: Theories of Distributive Justice

This chapter is about the positive analysis of inequality; this appendix addresses the normative side. In the end, most debates about inequality—whether it is a problem and what to do about it—center not on disagreements over the facts or how they are explained, but on the principles that ought to govern a “fair” distribution. In philosophy, this is the realm of distributive justice. Here our purpose is not to develop complete theories of fair distribution, weighing the arguments pro and con, but simply to present the core principles and the motivations behind them. The point is to be explicit about the criteria we are using when we pass judgment on economic inequality and to avoid fruitless disputes based on unexamined normative assumptions.

Before continuing, we should be clear that the theories we are about to look at all have one element in common, an attempt to apply logical analysis to the problem of inequality. This is not how many people think about it, however. For most of human history, and into the present as well, it has been more common to see the world as made up of “natural” hierarchies: some groups are viewed as above others by virtue of their parentage, gender, physical appearance, cultural aura or social or religious standing. This is the view that has given us kings, castes, the veneration of elders, racial and religious classifications, rigid gender roles, and tight networks of loyalty and patronage. From this angle, an economic distribution would be regarded as just if it corresponds to whichever hierarchies are seen as natural and proper. Even though this is a common perspective—especially if you acknowledge its subconscious force—here we will put aside all appeals to traditional authority and consider only the rationalist arguments associated with philosophy.

Keeping this commitment in mind, we will take up six general approaches to justice: that any outcome that results from neutral rules is just, that rewards should be in proportion to contribution or to effort, that fairness should depend only on initial equality of opportunity, that outcomes should be as equal as possible, and that the primary restriction on inequality should be that an acceptable minimum be guaranteed to all.

1. Neutrality of rules. For many centuries, the Chinese empire was administered by a corps of mandarins, highly educated scholars and officials. These were the highest positions one could aspire to if one was not born to a noble family. To become a mandarin you had to pass a difficult test; only those who had spent many years in study would be able to do this. Admission to the schools was based on achieving an initial level of skill at reading and writing. Of course, for most of Chinese history, only a small portion of the population ever attained literacy, and this meant that most households—mainly peasants and artisans—had no real chance to place a son (this was an all-male institution) in the mandarin corps. Would people of today accept this system as just?

It was extraordinarily unequal; mandarins lived a life that most of the population could only dream of. There wasn’t anything approaching equal access to the schools that funneled children into the mandarin class. If you were poor you were mostly out of luck, although, on occasion, a poor child did manage to ascend through the system. Nevertheless, one argument in favor of this custom is that the rules were the same for everyone, at least everyone who was male. Nothing in the procedures prohibited poor families from trying to get a child admitted to a school, nor was there any penalty to children from poor families if they made it as far as the final exam. The rules were perfectly neutral.

Even today there are some who argue that, as long as the rules that govern who gets ahead are neutral, making no distinctions between anyone, whatever outcomes result should be accepted as fair. Admission to universities and hiring decisions should be “blind”, taking no account of any information about the candidates other than requirements that apply equally to all. What would be unfair, from this point of view, would be any rule that did make distinctions, that applied one set of criteria to this person and another set to someone else.

The great advantage of neutrality as a principle of justice is that it minimizes the opportunity for elites to rig the game in their own favor. As soon as you allow procedures that deliberately give an advantage to some over others, you create incentives for insiders to use them to make sure that the advantage is theirs. In the Chinese example, suppose, instead of having a single exam and scoring system for everyone who wanted to become a mandarin, officials were able to choose which applicant would get which exam or could give extra credit to some, but only some, of the exam-takers. In all probability, the children of those who already had the most wealth and power would be given special advantages, and those who came from poor backgrounds would find their way blocked by barriers they might never see.

The disadvantage of neutrality is that the societies we live in are not neutral. As in China during the empire, today we have wealth and poverty, children with every advantage money can buy and others who are up against enormous odds. If the rules are neutral, nothing will intervene to reduce the inequality we have inherited from the past. In the words of Anatole France, “The law, in its majestic equality, forbids the rich as well as the poor to sleep under bridges, to beg in the streets, and to steal bread.” Do we want the laws to make distinctions—to have one set of laws for the rich and another for the poor? Maybe not, but then we might want something else, other programs, policies or institutions that do more than simply determine what is legal or illegal, but intervene in other ways to limit economic inequalities. What these measures should be is the question the other criteria below try to address.

2. Reward according to contribution. A plausible approach would be to say that what individuals get out of a social or economic system should depend on what they put into it. The main virtue is pragmatic: it provides a strong incentive for each person to make as great a contribution as possible. The result will be more to share for everyone, at least if this theory holds true. Indeed, one of the attractive features of the Market Welfare Model is that, if its conditions are met, individuals and organizations (like firms) will earn from the market exactly what each puts in. Thus, each worker would receive the economic value of his or her marginal product, and if that value is correctly measured by prices—as it would be in a Market Welfare Model world—then the worker’s earnings equal the consumer’s benefit from this work. If the assumptions of the Market Welfare Model are not fulfilled but we continue to accept reward according to contribution as our guiding principle, we should try to fix the market failures or other impediments responsible for supply curves not representing social costs or demand curves not representing social benefits. And failing that, we should use the benchmarks of an idealized perfect market to help us set prices and regulations that will approximate those of a Market Welfare Model world as closely as possible. This, as you should know by now, is the program laid out by conventional welfare economics; see, for instance, the discussion in Chap. 6. The normative underpinning of this project is the belief that rewards in the economy should be governed by contribution.

The ethical basis for this approach is reciprocity, that each should give and get in equal measure. Nevertheless, there are ethical dilemmas that have weakened the appeal of contribution as the main criterion. One is that the contribution one ends up making to society is strongly influenced by pure chance. A worker who works for a particularly innovative or well-run company contributes more than one who works for an average or less effective outfit. According to the principle of reward for contribution, it is appropriate that the first worker should earn more than the second. Nevertheless, it is often a matter of luck which type of job one ends up getting. Perhaps Worker 1, who is employed by the high-powered company and gets a fatter paycheck, has connections that helped her get through the door, or perhaps she was just at the right place at the right time. Is it fair that she should get the greater reward?

Another issue has to do with native human talent. Some simply have greater physical or mental resources to work with. One worker might have to work twice as hard to have the same effect as another; shouldn’t this extra dedication be worth something? Are results all that should matter?

3. Reward for effort. Such concerns have led many to adopt a modified version of the contribution principle in which it is the input of the individual (or group), and not the output, that determines rewards. From an ethical standpoint, we are rewarding what people give of themselves for others, even if chance or the uneven distribution of talent does not always translate this effort into tangible benefits. As a practical matter, this approach has the virtue of providing strong incentives to all members of society to work hard, whether or not their skills are in the most demand. Of course, this virtue is also a vice, since rewarding people for simply trying may not give them enough incentive to apply their energies to the things society places its highest value on. In general, market economies do not generate rewards for effort apart from its role in the ultimate product being produced, so some interference in labor or other markets is required if effort per se is to be recognized.

Here is a conundrum you may want to think about in connection with reward for effort: Suppose there are ten workers working on an assembly line. Each operates a different machine, each operation is equally necessary and each is equally difficult. The line moves at a common pace, so workers can neither increase nor decrease the speed of their work. The output of the line is a flow of products that can be sold for a given amount of money. If reward for contribution is the criterion, each worker should get the same pay. If, on the other hand, the workers are paid in accordance with their effort, the least capable workers will earn the most, since they have to put in more effort to keep up with the pace of the line. Their extra pay will in turn mean lower pay for the most capable, who are, in effect, being punished for having more strength or agility. Is this fair?

4. Equality of opportunity. Perhaps the world is too complex to measure either contribution or effort with any accuracy. There are so many different kinds of effort to compare, and the ways contributions are combined in real-world production systems makes it difficult to tell just who contributed what. Moreover, perhaps both contribution and effort should be recognized, although not in any precise combination. In that case we might be drawn to an approach that says, let all start with an equal chance to succeed, and let effort, contribution and luck determine outcomes however they will. Specifically, the criterion of equal opportunity embodies two elements, that there should be a moment in each individual’s life (the “starting point”) at which equality should reign, and that the rules that govern success should not be biased toward any particular individuals or groups. These are extremely demanding requirements, and it is doubtful that any existing society meets them completely; yet they could serve as goals to be pursued. In practical terms, the first will usually require substantial intervention in market outcomes, since the advantages that children of rich parents would otherwise have need to be offset, but the second is usually thought of as compatible with the way markets should work if they are regulated to be transparent and fair.

There are two large difficulties with equal opportunity as a principle of justice. First, what exactly should be this hypothetical moment of perfect equality—the equal starting line, to use the metaphor of a footrace? Should it be birth? This means that the unequal distribution of luck before birth must be counterbalanced, so that those who are congenitally stronger or more clever should be disadvantaged in equal measure. If that doesn’t appeal to you, then you perhaps imagine a moment even before birth and before genetic qualities are doled out. But related to this is the problem that, as soon as we are born, we begin to do things or have things done to us that, if not offset, will lead to unequal life chances down the road. The further back we push the moment of equality, the more subsequent inequality we must accept. If opportunities are to be equal at birth, then the advantages that some get in childhood will not count against “equal opportunity”. Perhaps you would set a much later age for the “starting point”—say 18. This commits you to much greater intervention to offset all the many pluses and minuses that can accrue by that age, including many that are due to the choices that children make for themselves. At the same time, it can be seen as a bit heartless, since it doesn’t allow for second chances. If someone discovers what they truly want at the age of 25 or so, too bad: they missed the moment of equality and they will have to make do with whatever opportunities they are lucky enough to still have. This sort of criticism can be addressed by requiring a multiplicity of “somewhat equal opportunities” that can reappear as one grows older, but then the criterion loses its sharpness: how equal must these second chances be and how many must be offered?

The second large difficulty is that equality of opportunity is compatible with almost any level of general inequality, as we defined it in this chapter. Suppose, for example, that you have a society that works according to this rule: every year a lottery is held with just one winning number. The individual who wins that year gets everything—every last penny of income, all the wealth, the land, everything of value. Everyone else must beg for enough to survive on. The principle of equal opportunity demands only one thing, that the lottery be perfectly fair, so that each person has the same chance to be tycoon-for-a-year, but surely this demand does not go far enough. Can extremely unequal divisions of life’s good things be viewed as just simply because the system is fair at the moment just before division?

5. Equality of outcomes. This principle, which we can call egalitarianism for short, does not concern itself with who gets more or how they acquired it; more versus less is itself the problem. The ideal world would be one in which the rewards of our common social and economic life are apportioned perfectly equally. This would presumably be true not at a given moment, such as in equal opportunity, but at all times.

Of course, perfect equality is not really possible. It would take too large a bureaucracy to measure and distribute everything in perfectly equal measure, and we would have the problem of figuring out how much of one sort of good, like security, should be worth another, like income. In practical terms, egalitarianism is an orientation toward more equality of outcomes in preference to less. Thus, to take the case of income distribution, an egalitarian would prefer a more equal distribution (a lower Gini coefficient), all else being equal. It is a preference for equality as a value in itself.

One of the most vexing practical difficulties faced by egalitarians is determining just what it is that should be made more equal. Money income is a logical candidate, but questions arise: Should large families have more income than smaller ones? Should workers with more dangerous or uncomfortable jobs receive extra pay to make up for their hardship? Should people with special needs, for instance due to disabilities, receive the extra income they need to meet them? These are not abstract puzzles in logic; they arise whenever we try to apply egalitarian principles to real human beings.

Another set of problems arises from what egalitarianism is not—not a system that assigns value to contribution or effort. This is an ethical concern, since contribution and effort both make claims on our sense of justice. It is also an issue of great practical importance, since it is very likely that, without the incentives that unequal rewards depend on, people would neither contribute enough work nor be guided to work in the most beneficial ways. To the extent this is true we face what has been called a tradeoff between equity and efficiency, where equity means egalitarianism and efficiency the production of the greatest quantity of economic value. It should be remembered, however, that, while economic incentives are indispensable, they are not the only forces that motivate us, and sometimes they get in the way of more desirable, intrinsic motivators. (See the discussion in Chap. 10.) Thus, the tradeoff between equity and efficiency is real, but not everywhere and always. In any case, it is difficult to claim that egalitarianism should be the only ethical criterion, and that no weight should be given to contribution or effort, particularly since some degree of inequality will persist in any real-world situation.

6. Guaranteed minimum rewards. In the twentieth century there was renewed interest in approaches to justice that emphasize the needs of those at the bottom of the distribution. Perhaps what we should look for, it was reasoned, is not a principle that governs every issue of distribution for everyone, but which focuses on the crucial needs of the worst-off. There are two main variants at the present time, the brainchildren of John Rawls and Amartya Sen.

(a) Rawls: Maximize the minimum well-being. Rawls, perhaps the most influential philosopher in the field of justice in modern times, takes as his starting point two principles, objectivity and risk aversion. Objectivity means that our evaluation of an economic or social order should not depend on our own place in it; it should be the same no matter which role we come to occupy. His device for achieving this was an imaginary “thought experiment”: suppose we were about enter a society (through birth, for example), but we first had a chance to put an evaluation on it, prior to knowing who we would come to be, including who our parents would be, what our physical inheritance would be, etc. This evaluation would be ideal, according to Rawls, since it would be based on a perfectly objective analysis.

The second principle is that such an evaluation should be governed by risk minimization. What would weigh most heavily in our judgment would be the potential to be someone who is unhappy or oppressed in a society, which we could identify with having a very low income. If this is the case, we would rank societies on the basis of how well-off the worst off person is. If the total income of the economy were a given, Rawls’ theory would dovetail with perfect egalitarianism, since under that rule the worst-off person would be as well-off as possible. Rawls assumes, however, the necessity of economic incentives to motivate production, so that too much equality might well reduce the rewards attainable by the worst-off. The ideal, in his estimation, would be that balance of equality and inequality that maximizes the position of the bottom person in the economic distribution. As a global proposition, this would be difficult to apply, but we might be able to employ it when looking at a particular distributional issue, such as whether the wages for a particular group of workers ought to be raised. Even so, it would be a difficult empirical task to estimate just how much particular distributional adjustments are likely to change economic growth, the extent and direction of innovation, and the like. Toward the end of his life, Rawls retreated from more rigid formulations of his principle.

(b) Sen: Maximize the fulfillment of human capabilities. Amartya Sen follows in the footsteps of Aristotle, who argued that human beings have a common nature and by realizing our potential we can achieve “flourishing”. Aristotle in effect advocated the use of social science to observe a variety of societies to see, empirically, under what conditions their members flourished, so that we could replicate the best of these features in our own ideal. In the more than two millennia that have transpired since Aristotle, however, we have learned much more about how to do such observation, and we have a much broader base of human possibility to draw from.

Sen believes that we are in a position today to make an informed judgment along Aristotle’s lines, provisional (in the spirit of all science) but with real practical implications. We will look at the details of his approach in the next chapter when we focus on poverty, but for now it is enough to mention the broad outlines. Human beings are said to have capabilities, modes of functioning in the physical, psychological and social universe. Exercising strength, solving problems that interest us, engaging with others—these are the sorts of things we all do, but in different ways in different cultures. Sen, who is a Nobel laureate economist as well as a philosopher, suggests that all societies grant “entitlements” to people to enable them to access the resources that make the exercise of capabilities possible. These entitlements may take the form of income, or they may be social obligations to provide particular goods or opportunities, or they may be distributed politically. However we come by them, we need enough of them and in the right combination to fully exercise our capabilities—to flourish.

Although he has not presented his theory in quite this way, it would be a reasonable deduction from his approach to regard the universal attainment of full human functioning as the primary ethical norm. If this goal were achieved, his system of justice would be indifferent regarding the distribution of the “extra” goods not needed for realizing our capabilities. In this way it could be considered a cousin to Rawls, since it would have the primary effect of raising the well-being of the worst off.

This presentation of Sen’s theory of capabilities may appear highly abstract, too abstract to be useful in practical situations. This criticism will be addressed in the following chapter, when we see how it has actually been applied, but it should be conceded that this difficulty has never quite been dispelled. There is disagreement among Sen’s followers regarding which capabilities are fundamental and how their fulfillment can be measured. There is also a potential risk in any approach which claims to tell us what we “really” need, since economists and philosophers have their own biases and blindspots. On the other (third?) hand, one of the main things we pay philosophers to do is advise us on what we need and how we should regard ourselves.

To summarize, here we have five principles of just distribution. The first three of them try to do too much, and it would probably be a mistake to apply them in every situation. The last (combining Rawls and Sen) at best does too little, since they tell us only about what is fair for those at the bottom, and not for those in the middle or at the top. The fourth (equal opportunity) is both too demanding in its requirements and also not demanding enough. In short, they all have their flaws.

This brief survey should inspire a measure of humility, since whatever yardstick you adopt you will be vulnerable to counterarguments. Also, as in the other truly difficult problems we have explored in this book, you should be encouraged to be flexible, to be willing to use more than one framework when circumstances seem to call for it.