Markkula Center of Applied Ethics

The Income Gap

Recent increases in income and wage inequality in the United States prompt a look at the ethics of income redistribution.

By William A. Sundstrom

As the "American Century" draws to a close, the United States appears to be experiencing a decade of extraordinary economic prosperity. Since the early 1990s, economic growth has been steady and robust, inflation near zero, and unemployment rates at a level that most economists would have dismissed as implausibly low just 10 years ago. This record prompted Alan Greenspan, the chair of the Federal Reserve Board, to declare that the economy's current performance is as impressive as it has been at any time during the past half-century.


Although the benefits of overall economic growth should not be understated, broad indicators of prosperity obscure another, less encouraging economic trend of the past 25 years: a dramatic rise in income inequality. Indeed, by standard measures of real income and wages, the poorest and least educated Americans have experienced a falling standard of living since 1975. Meanwhile, the largest proportionate income gains have come to those in the top income and wealth brackets.

How real are these changes in income distribution, and why have they occurred? To what extent do basic principles of justice and compassion require redistribution? And if they do, how should this be accomplished?

Some evidence

During the 1980s, economists and other observers began to notice striking changes in the income gap between rich and poor. At first, there was considerable dispute over the numbers and how to interpret them. For a time, it seemed possible that the increase in inequality was transitory, a product of the deep recession of the early 1980s, and would be reversed during the subsequent economic recovery.

Eventually, however, the weight of the evidence made the trend unambiguous. It also became clear that economic expansion-even during the current boom decade-was not reducing inequality.

One intuitive measure of inequality is to compare "high-income" and "low-income" families. Suppose we define a high income level as income at the 90th percentile: that is, the income level higher than that of 90 percent of all families but lower than that of the top 10 percent. Similarly, we can define a low income level as the income at the 10th percentile.

During the early 1970s, a family at the 90th percentile had an income about six times that of a family at the 10th percentile. By the late 1980s, this ratio had risen to nearly 8-to-1, and it moved even a little higher during the early 1990s.

Rising inequality in family incomes reflects rising inequality in wages, the most important source of income for most Americans. Wage inequality has increased dramatically for both men and women, especially between 1979 and 1987. In 1996, the hourly wage of a worker at the 10th percentile was $5.17, about 14 percent lower than it had been in 1973 after adjusting for inflation. Meanwhile, the worker at the 90th percentile earned $23.01 per hour, an increase of about 6 percent in real terms since 1973.

Many of the lowest-paid workers are poorly educated. These workers have seen their real earnings fall dramatically over the past 25 years. A man without a high school diploma, for instance, now earns a wage that is worth about a third less than what he would have earned in 1973.

Even if, as some economists have argued, the official statistics overstate the rate of inflation by about 1 percent per year, poorly educated men still earn less now than they would have 25 years ago. Much of the increase in income inequality can be attributed to this growing gap between high-wage and low-wage men.

Although the widening wage gap has been the largest factor accounting for increased income inequality, changes in family structure have exacerbated the trend, especially since the 1960s. As is well-known, single-parent families are much more likely to be poor than are two-parent families, so the growing proportion of single-parent families has helped increase the percentage of families with very low incomes.

While the overall numbers have not been good news from an egalitarian's point of view, some bright spots should not be overlooked. For the first time this century, the pay gap between men and women has been shrinking quite significantly. And the pay gap between African-American and white men has also shown some signs of closing in the last few years. These equalizing trends simply have not been large enough to offset the factors that have increased inequality overall.

Why has inequality increased?

It is always tempting to look for a single cause that can explain a major social or economic event, but in all likelihood the growing gap between rich and poor has many causes. A list of the "usual suspects" is easy to compile; evaluating their relative importance is far more difficult.

Much evidence points toward declining demand for low-skilled workers as a crucial factor. In the view of many economists, the principal source of this shift has been technological change that has allowed firms to economize on low-skilled labor while increasing the demand for highly educated workers. Ironically, these changes are broadly a consequence of the growing importance of computers and automation in the economy, the very technological advances that have helped drive the current economic boom.


Globalization of the economy may also have played a role. In the past 25 years, low-skilled American workers have experienced increasing competition from both low-paid immigrants and low-paid workers living in other parts of the world who produce goods that compete with American products. Although the effects of globalization are not negligible, most economists think they are less significant than the effects of technological change on wage inequality.

Finally, institutional changes in the labor market have had an impact. These include the declining membership in labor unions in the private sector and the declining real value of the legal minimum wage, which until recently had been severely eroded by inflation.

International comparisons support this analysis. The pressures on low-skilled wages from technological change and globalization have affected all developed economies, but the growth of inequality has been far greater in the United States. International differences in wage-setting institutions and income-transfer programs may explain the different outcomes. They may also help account for the higher unemployment rates in many other developed economies, a perspective that casts the U.S. experience in a more favorable light.

Why should we care?

Compared with the residents of many other economically advanced countries, Americans seem to have a high tolerance for economic inequality. The growing gap between rich and poor has prompted intermittent public discussion about fairness but little sustained dialogue regarding what, if anything, is wrong with income inequality and what ought to be done about it.

Part of the problem is that low-paid workers lack the political voice to raise these issues in the political sphere. But another aspect of the problem is that Americans seem to have lost sight of the compelling moral arguments for reducing economic inequality, beyond merely serving the self-interest of the poor. Nothing is more symptomatic of this ideological vacuum than the fear of mainstream politicians that, by speaking out on inequality, they will be perceived as fomenting "class warfare."

But income inequality raises basic ethical issues that should be the subject of public debate and inform government policy. Moral arguments for reducing income inequality follow from a number of basic principles, including the following:

Public Compassion or Sympathy: As members of a larger community, we simply should and do care how others are doing. Income growth at the top of the distribution provides the means to act on our compassion by improving the material well-being of those at the bottom.

Fairness: Morally arbitrary differences among us arising from the circumstances of our birth, upbringing, or current position in the market economy should not be permitted to create drastic differences in basic well-being or human dignity. In a just society, those with a better draw in life's lottery should contribute toward the welfare of those with a poorer draw. Furthermore, to the extent that family income helps determine a child's life chances, fair equality of opportunity depends on providing families with adequate resources to give their children a good start.

Desert: The income a person receives should have some relationship to what she or he deserves. Arguments based on desert have often been used to defend market-derived inequality: In the labor market, pay is a function of productivity, and lower pay reflects a smaller contribution to production. But in our everyday language, we also link desert to effort: The low-skilled worker who puts in a long, hard day's work may in this sense be as deserving as the high-powered lawyer or CEO.

Democratic Solidarity: A demo-cratic community cannot tolerate too much social stratification before it loses its demo-cratic character. This is perhaps most evident in the relationship between income and political influence in modern American politics. Although the wealthiest candidates and causes do not always win, they are always sure of being heard.

Positive Liberty: Americans place great value on freedom as the absence of constraints on conscience or expression (negative liberty). But the exercise of freedom in practice depends on access to the basic resources and capabilities necessary to act. The absence of positive liberty is nowhere more evident than in the limited options of the poor.

Self-interest: It is at least plausible that, after a point, increased inequality might lead to social or political instability and impede economic performance. Self-interest, even among the rich, might be served by moderating inequality.

These principles point toward a special concern for inequality that arises from low incomes, or poverty. The exercise of basic freedoms and participation in the political and social life of one's community typically require some minimum standard of living. This minimum is not fixed and absolute, but depends in part on the social standards and expectations of the community. Concern about inequality thus properly focuses on the social safety net and the problems of the working poor, though it should not ignore the concentration of economic power at the upper end of the income distribution.

What should be done?

Individual and communal acts of charity will always play a role in reducing the adverse effects of income inequality, but significant reductions of inequality will depend upon the government's power to tax, transfer, and regulate. Inevitably, however, redistributive policies involve real costs, in terms of both their economic impact and the infringements of property rights that accompany them. Thus there is a trade-off between equity and other values. To minimize the damage to these other values, two principles of policy design-efficiency and efficacy-should guide us as we evaluate proposals for reducing inequality.

If a policy is efficient, it will achieve its redistributive goals at minimal cost to the economy as a whole. Perhaps surprisingly, the criterion of efficiency is usually better served by policies that treat the symptoms, rather than the causes, of inequality.

For instance, to the extent that import competition is a source of downward pressure on low-skilled wages, protectionist trade policy could counteract the trend. Yet protectionism comes at a high cost to consumers and trade-dependent sectors of the economy. More efficient would be a policy that directly enhanced low-skilled workers' incomes, whether through training, subsidies, or minimum wages.

The second guideline, efficacy, requires that policies target the individuals and families most in need of income support. By this criterion, the minimum wage is a rather poor tool for redistribution: A surprisingly large percentage of minimum wage workers reside in fairly prosperous households.

A policy that does a better job targeting the working poor is the earned income tax credit (EITC), which has become one of the most important components of federal antipoverty policy. The EITC is a federal tax credit on every dollar earned, up to a maximum (currently about $3,700 per year for families with two children). The credit phases out as a worker's earnings rise beyond a certain level. By design, the EITC can only augment the incomes of families that contain workers.

Political rhetoric favors training and education programs as the best means of upgrading the skills and thus the pay of low-wage workers, but these also fail the efficacy test. Unfortunately, studies of job search and training programs have typically found that while they can help low-skilled workers find jobs, they seldom lead to significant increases in earnings potential, and thus seldom lift poor workers out of poverty. There is little reason to believe that job training, even on a massive scale, would do much to reverse the trend in wage inequality.

While broader educational reform is central to any long-run strategy for improving the living standards of the poor, its effects are also uncertain and will only be observed over the course of a generation or more.

If narrowing the gap between rich and poor is to be accomplished in the here and now or the near future, old-fashioned income redistribution policies still have much to recommend them. Income transfers to the poor, financed by a system of progressive taxation, have the virtue of targeting those in need without undue disruption of the economy.

The form such transfers might take is as much a function of political feasibility as it is a matter of fairness or economic efficiency. Progressive economists, especially in Europe, have often favored a guaranteed basic income, or negative income tax, which would be paid regardless of employment status. But in the current political climate of the United States, offering an income to the poor with no strings attached is not a realistic policy option.

Increasing the EITC and expanding programs that support poor workers (such as job search assistance, health insurance, and housing and child care subsidies) are less radical but potentially quite effective ways to mitigate the worst effects of the inequality trends of the past 25 years. Such policies must still be augmented by an adequate system of social insurance-a safety net- for those Americans who are simply unable to earn enough to escape poverty.

Finally, government can take measures to reduce the advantages of the very rich in the political sphere, where income inequality threatens basic democratic values. Serious campaign finance reform is the logical approach here.

Although egalitarianism is less politically popular than it was earlier this century, the current political climate is in some ways conducive to reopening the debate about distribution. Indeed, the recent welfare reform and increases in the minimum wage and EITC have moved poverty onto the political front burner. And the remarkable overall performance of the economy may lead many to ask why we cannot now afford to reduce the wide disparities in living standards and life chances that still exist among Americans.

William A. Sundstrom is associate professor of economics at Santa Clara University and a member of the Markkula Center for Applied Ethics Steering Committee.

Further Reading

Bernstein, Jared, and Mishel, Lawrence. "Has Wage Inequality Stopped Growing?" Monthly Labor Review 120 (December 1997).

Blank, Rebecca M. It Takes a Nation: A New Agenda for Fighting Poverty. New York: Princeton University Press, 1997.

Danziger, Sheldon, and Gottschalk, Peter. America Unequal. New York: Russell Sage Foundation, 1995.

Gottschalk, Peter, and Smeeding, Timothy M. "Cross-National Comparisons of Earnings and Income Inequality," Journal of Economic Literature 35 (June 1997).

Symposium on Wage Inequality, Journal of Economic Perspectives 11 (Spring 1997).