1
Creating and Distributing Wealth: Whose Responsibility?
Peter J. Hill
The issue of material inequality among people and the appropriate policy response to this reality is one of the most contentious in modern democratic societies. The debate focuses on governmentâs role in reducing that inequality. One of the primary attractions of Marxist socialism was its promise of a more egalitarian society. Now that centrally planned, socialist economies have been found to be both inefficient and exploitative of their citizens, the debate has turned to inequality in market societies. An oft-expressed sentiment is that though markets are efficient, they are unjust, since market processes create significant economic inequality. Therefore, modern egalitarians argue, a legitimate role for government is to equalize the distribution of wealth or income in society.
In this chapter, I shall argue that the opposite is the case: government should not attempt, as an explicit policy, to redistribute income or wealth.1 My argument has three components. First, the concern for material inequality represents a harmful distortion of an appropriate and laudable human motive, a desire to help the infirm, the suffering, and the unfortunate. Second, efforts to achieve justice through redistribution of income foster far greater injustices because of the inequality of political power that such redistribution requires. Third, redistribution as a policy goal involves structuring the rules of the game in such a manner as to inhibit the most important means of poverty amelioration available: wealth creation.
What should we care about?
Most religious and ethical systems, Christianity not least among them, express a level of concern for the poor in society. These belief systems outline a clear responsibility for those with material wealth to alleviate the suffering of the poor and even take as a measure of religious commitment the degree to which one helps the poor.2 For instance, Proverbs 19:17 says, âHe who is kind to the poor lends to the Lordâ (NIV); in contrast, one âwho oppresses the poor shows contempt for their makerâ (Prov. 14:31). However, as legitimate and widespread as is this concern for the poor, addressing poverty by attempting to reduce inequality has serious shortcomings.
Discussions of inequality focus on the relative size of the economic pie held by particular groups. The percentage of income received by each quintile of the population is a common measure of inequality. For instance in the United States in 1997, the lowest 20 percent received 5.2 percent of the national income, the next 20 percent received 10.5 percent, the middle quintile 15.6 percent, the fourth 22.4 percent, and the highest 20 percent received 46.4 percent.3 Income distribution in the United States is clearly not equal. Focusing on the relative size of the shares, however, tells us nothing about the well-being of the poor. All that we know from this data is the relative share each group receives from the total pie. We know nothing about the size of the pie, whether the bottom quintile is malnourished or not, what types of housing they have access to, what their life span is, whether they drink clean water or not, and what level of health care they can purchase.
Another measure that has a strong relationship to the quintile groupings is the Gini index, which varies between zero and one. An economy that has an index of zero is one with complete equality, while an index of one would represent complete inequality. The most recent World Development Indicators from the World Bank list Gini indices that range from .231 (Austria) to .629 (Sierra Leone).4 The Gini index is useful because it summarizes the income distribution data into a single number. As an indicator of economic well-being, however, it faces the same problem that the quintile shares measure does, in that it simply provides no information.
Egalitarians often quote the income distribution data and deplore any rise in inequality because of their inference that greater inequality means greater hardship for those in the bottom portion of the distribution. The fallacy driving this perception is the belief that world wealth is a zero-sum game. If there is a fixed amount of income or wealth in a society, then increases in inequality must indeed represent increases in hardship for some. But total wealth is not fixed. For instance, India has a Gini index score of .378, while the Gini for the United States is .408. The share of national income received by the lowest 10 percent in India is almost twice the share received by the lowest 10 percent in the United States (3.5 percent compared to 1.8 percent).5 If one accepts the egalitarian assumption that greater income equality means a better world, one would assume that the poor in India are better off than the poor in the United States. But the fact that GDP per head in India is $460 compared to $34,260 in the U.S. indicates that the larger share of the pie received by the poor in India tells us little about their economic status relative to the poor in the United States.6
Even when income distribution becomes more unequal (the usual term used is âworsens,â nomenclature that presupposes that an increase in income inequality is bad and that a decrease is good) within a particular economy, this change still provides no information on the well-being of those we ought to be concerned about, namely, the poorest sector of the population. Suppose that the income of the top 20 percent increased by 4 percent over a given period and the bottom 20 percent increased their income by 2 percent. Now contrast this situation with a situation in which the top 20 percent found their income decreasing by 4 percent and the bottom 20 percent experienced no change in their income level. By egalitarian standards, the second situation represents an improvement in the position of the poor relative to the first, for in the first case, income inequality would be increased. And yet in the first situation, unlike the second, the poor would now have access to better health care, higher nutritional levels, and better housing, surely things we ought to care about if we really are concerned about the well-being of the poor.
But, an egalitarian may respond, the poor measure their well-being with respect to other people, and if a societyâs income distribution becomes more equal, the poor feel they have improved their lot, even if their economic status in absolute terms has declined. It is true that individuals often measure their economic status with regard to the status of others, but that does not mean we should allow such comparisons to drive public policy. To do so capitulates to the base human tendency to envy what others possess. Policies that focus upon income equality legitimate envy. Rather than encourage people to rejoice when their fellow citizens do well, such policies encourage people to focus on resenting the achievements of others instead of celebrating their own achievements.7 If some in a society find ways to create wealth, and if that wealth creation does not harm the well-being of others, why should those who have not shared fully in that wealth creation be resentful?8 Envy and resentment will never be eliminated, but building public policy upon these most base of human motivations is a bad idea.
Focusing on material inequality creates several other problems. Most discussions of inequality have implicit assumptions about legitimate and illegitimate reasons for differences in income. Differences in effort and ability are often deemed to be valid reasons for the generation of differences in wealth, whereas luck, discrimination, and inheritance are seen as unfair or unjust reasons for the generation of such differences. However, two experiments indicate that substantial differences in income can be generated simply through differences in peopleâs willingness to work, in conjunction with minimal ability differences.9 In Canada, an experiment gave men and women the opportunity to participate in something called the Cannabis Economy. (This experiment was designed to determine the effects of marijuana smoking on productivity.)10 All of the participants were twenty to twenty-eight years old and all had twelve years of education or more. Subjects were not allowed to leave the experimental facilities, and they were provided with housing for the ninety-eight days of the experiment. They had the opportunity to weave woolen belts on small portable handlooms if they so desired, for which they received $2.50 per belt. The material was supplied free of charge, and they could work as many or as few hours as they wished. The subjects used their earnings to purchase food from the hospital cafeteria and the canteen. They could also take home any savings at the end of the experiment.
A second experiment occurred in the Central Islip State Hospital in New York. Chronic psychotics were given tokens for their performance of janitorial chores. The jobs were optional: the patientsâ basic needs of food, housing, and medical care were provided, so the tokens were only used to purchase additional consumption items. All participants had the requisite skills to perform the tasks.
In both economies the earnings of the highest group were over ten times those at the bottom. The Gini indices generated in these experimental âsocietiesâ were very similar to the indices of most Western economies. The conclusion of the researchers was that differences in abilities and willingness to work are sufficient to generate the income differences seen in the United States and other Western countries. When the effects of luck, original wealth position, and natural disasters were removed, there were still large variations in earnings. Although this does not answer the question of whether existing income distributions are legitimate or not, it does raise the issue of whether or not intervention to rectify existing distributions is actually an attempt to alter what could be called a ânatural distribution,â a distribution reflecting peopleâs innate abilities and their desire for leisure or work.
What type of inequality?
Most advocates of government as an agency for the redistribution of income assume that this task can be carried out with few or no costs. Such is not the case. The achievement of greater economic equality can only come with the creation of great political inequality. Only when the coercive power of government is extended into numerous private spheres of action can substantial gains in material equality be realized.11 In addition, the use of the term âredistribution of incomeâ implies an original distributor. But such is not the case, either. In a free society, incomes result from the uncoerced interactions of individuals pursuing their own goals. They buy and sell products, transfer resources by gift, work for wages and salaries, and control most of their own lives. Since there is no original distributor of income, it is not possible to have redistribution unless some social agency becomes that distributor. Therefore, setting out to âredistributeâ income requires a substantial alteration of institutional structures, an alteration that concentrates enough power in the hands of the coercive agency of government that choices are restricted. Robert Nozick has pointed out that to suggest income be redistributed is akin to arguing for the redistribution of marriage mates.12 Since there is no one distributing marriage mates, it is difficult to see how one could redistribute them unless one replaced voluntary interaction with coercive, top-down direction. The same applies to income.
It is clear that all societies at all times generate substantial differences in income. The 2001 World Bank data quoted earlier indicate that some societies are more equal than others, but all societies have a degree of inequality. Even in Austriaâthe most egalitarian among the large number of countries for which data are reportedâthe bottom 20 percent of the income distribution receives only 10.4 percent of total national income, while the top 20 percent receives 33.3 percent. The experimental economies reported earlier show that in societies where fortune, inheritance, and access to means of production are equalized, the simple willingness to work will still generate quite unequal incomes. In other words, the diversity of humankind naturally reflects itself in material inequality, and thus efforts to reduce inequality require the use of substantial coercive power.
If individual rights are recognized and protected by government under the rule of law, economic inequality will result. Therefore, reducing inequality means moving away from the rule of law and stable property rights. Equality before the law, defined as the equal treatment of people in equal circumstances, is one of the great attainments of the modern liberal order. Reducing that equality to reduce material inequality is a costly project.
In the twentieth century, between 80 and 100 million people lost their lives at the hands of communist governments.13 These deaths came not from defending borders against external aggression, but from attempts at social engineering in which the primary motivating force was radical egalitarianism. Stalinâs elimination of the kulaks, Maoâs Great Leap Forward and Cultural Revolution, and the killing fields of Cambodia represent the most egregious cases of attempts to achieve a socially sanctified objective through the coercive power of government. Equality before the law and recognition of individual rights were sacrificed to a higher goal: remaking society to meet a stated objective. While modern democracies do not use coercion in the brutal manner of these communist societies, it is important to remember that the desire to create greater equality has been a motive force behind some of the greatest tragedies of the twentieth century.
Some of the same problems that plagued the attempts of communist regimes to remake their societies according to a social ideal plague modern democracies in their attempts to redistribute income. A sufficient concentration of power to perform the âgoodâ of redistribution can be captured and used for evil ends. Most egalitarians do not want complete equality, believing that some inequality occurs because of meritorious behavior or meritorious differences in individuals. But the information problems in determining merit are enormous. Much of what we call merit depends upon motive, and motive is difficult to ascertain. Is the physician earning $200,000 a year motivated by the desire to relieve suffering or is she concerned with her own well-being? What about food production? How much of it is carried out for laudable reasons and how much for selfish? It is beyond the capacity of government to obtain adequate information about meritorious behavior, and even if such information could be obtained, reaching any sort of consensus about the trade-offs between different forms of behavior and individual rights would be even more difficult.14 Therefore, attempts to rectify material inequality will be arbitrary in nature and will seem quite unprincipled. Once the redistribution project is embarked upon, it is unlikely that the citizens of the redistributing society will find that the results meet even minimal standards of justice.
This is not to say that societal judgments about merit are inappropriate. The only question is, who should make them and how should rewards and punishments be meted out? Numerous institutions, ranging from families to churches to communities, need to make judgments about moral worthiness. Moral considerations are of crucial importance for a well-functioning society and moral judgments need to be enforced by societal pressure. But these numerous decentralized groups find it far easier to generate appropriate information about merit than does centralized government. Due to their different functions, they can measure merit along many different dimensions. ...