Shifts in labor demand and supply - Special Issue: Earnings Inequality - Panel Discussion

The presentations at this excellent conference have shed further light on rising inequality, one of the truly big stories in American economic life over the last 20 years. The enormous disparities in the fortunes of American families in recent years have largely been associated with labor market changes that have increased overall wage inequality and shifted wage and employment opportunities in favor of the more-educated and more-skilled. Less-educated young men have suffered unprecedented losses in real earnings and are at greater risk of nonemployment than in years past, both in absolute terms and relative to more-skilled workers. In short, the U.S. labor market has experienced a massive twist against "disadvantaged" workers - those with limited education or skills or from impoverished families and neighborhoods - that has diminished their earnings prospects and made it more difficult for them to keep their families out of poverty and intact.

Many analysts believe a key driving force behind these changes has been a strong shift in relative labor demand against the less-educated and those doing more routinized tasks and toward more-educated workers and those with problem-solving skills. Changes over time in wage inequality can be thought of as being the outcome of a footrace between technology (the demand for skills) on the one side and the supply of educated labor on the other side. It is clear that the technology and demand side has been winning the footrace, outstripping supply and stretching out the wage structure during most of the past two decades. These demand shifts favoring the more-skilled have been reinforced by changes in pay-setting norms, increased competition in many product markets, increased immigration of less-educated workers, and the weakening of institutions that have protected non-college workers (for example, the decline of unions and the erosion of the real value of the minimum wage). While much debate exists concerning the relative importance of these different underlying causes of rising inequality and increased returns to skill, none of the suspected factors show any apparent signs of abatement.

The Role of Macro Policy

Strong macroeconomic performance traditionally has been a crucial factor in improving the labor market prospects for disadvantaged workers. But the experiences of the long boom of the mid and late 1980s and the current U.S. expansion suggest that sustained economic growth by itself, unassisted by specific initiatives to deal with increased structural labor market barriers facing the less-skilled, is unlikely to be sufficient to reverse recent trends in inequality or to overcome increased labor market barriers facing the disadvantaged in America's inner cities.

Market incentives for increased individual educational investments and skills upgrading can play some role in alleviating growing inequality in the United States. The large increase in the college wage premium in the 1980s has been associated with an increase in college enrollment rates from 49 percent of high school graduates in 1980 to more than 60 percent in the early 1990s. Evidence from U.S. time series and cross-country studies strongly suggests that rapid expansion of the supply of more-educated workers narrows earnings differentials and improves the labor market position of the less-skilled. But the process of supply adjustment can take many years, and many disadvantaged individuals face financial and informational barriers to pursuing further education and training. Furthermore, the overall supply of college graduates has not grown very rapidly in recent years, as John Bishop showed, because the current baby bust cohort is quite small. Not many 40-year-olds return to college when the college premium expands.

These facts suggest a number of different strategies. First, we could try to improve the supply side of the labor market, as Frank Levy discussed. Obviously, primary and secondary education is key to that, although access to higher education is important as well. Second, we could try to affect the demand side of the labor market. We are not going to shut down the borders to trade; that would be foolhardy. But we could undertake some form of targeted demand policies, such as employer-side wage subsidies for economically disadvantaged workers, based either on people or on place. Third, government could play a better role in trying to make work pay, through an expanded earned income tax credit, possibly a higher minimum wage, or even doing more with the tax system. Fourth, we could do more to match up jobs and people who have very little connection to the labor market, such as welfare recipients and disadvantaged youth. Given that a lot of state and local governments will be making these decisions, we should draw lessons from the past on which approaches work best.

Choosing Policies That Work

Our 30 years of experimenting since the Great Society with training and wage subsidies and location-based assistance policies have given us a menu of options from which government can make its current decisions. We have had a number of negative messages, but this is probably the one area in the government budget where we have the most random-assignment evidence on which programs actually might work. So from this menu of options, policymakers such as state governors could make better-informed decisions than those made in the past.

The first thing we have learned on the negative side is that it is extremely hard to turn around the lives of people who have become disconnected from the mainstream educational system and dropped out of high school. Countless programs have attempted to help disadvantaged youth who have dropped out of high school and, aside from the Job Corps, a very expensive residential program, almost all have shown very little return. On the other hand, a number of recent demonstration projects suggest we can be more successful by starting earlier to work to keep kids in high school and prevent dropouts. The Quantum Opportunities program is a good private sector example, and the Department of Education has run a number of very successful demonstration projects: not traditional programs that help a 16-year-old get a summer job and do not last very long, but rather programs that start at age 14 or earlier and set up an inexpensive infrastructure with extra tutoring, together with a group at school responsible for helping kids make connections to the labor market. Some of the best examples, like the "I Have a Dream" programs, also guarantee some financial assistance for college. A number of these programs have had substantial effects on high school dropout rates and college attendance rates, and certainly they seem like potentially good uses of the funds that states will have available.

The second thing we have learned is that the returns to getting more education, such as attending college, are particularly high for those from disadvantaged backgrounds. Thus, the limited response of this group is not because they themselves do not generally experience high returns. When we have seen interventions such as increasing access to college or cutting tuition levels and studied them as natural experiments for estimating the rates of return to schooling, people from lower-income households have been the most affected. These are people on the margin who decide whether or not to go to school when you change access or tuition levels. When you estimate their rates of return, as David Card did in a recent survey, they look higher than the average difference in earnings between college- and high-school-educated workers, which suggests that capital market constraints are important. That does not mean that we know exactly the right ways to reduce the cost of education. But access to education combined with information seems to have a very high return for low-income people with high abilities. Policies to prevent dropouts and increase access to college do not work complete miracles, but they are also not that expensive when targeted to those at the margin, for example, in inner cities.

In another area, we have learned from the Gautreaux program and from a number of other quasi-experimental programs that neighborhoods, and the spatial concentration of the poor, do seem to matter. There is no chance in the world that the public will agree to huge residential dispersion policies, as the Baltimore experience with the Moving to Opportunity (MTO) program and the Mt. Laurel decision indicate. Small-scale attempts have a role, however, as shown in the current MTO program that, despite Baltimore talk radio disparagement, is in operation in the Baltimore metropolitan area as well as in Boston, Chicago, New York, and Los Angeles.

A striking characteristic of this program is that the majority of those who agree to participate in it say that the primary reason they want to move out of their neighborhood is because of problems with crime and worry for their children, but they lack the resources to leave public housing. Most claim to have been victimized by crime within the previous six months. In terms of transportation, 87 percent of them do not have cars, and the vast majority do not have driver's licenses. It is, therefore, plausible that these people are not choosing a place to live after evaluating neighborhood and transportation possibilities, but rather that public housing is the one place where they can get a subsidized living situation. Dispersion policies could accomplish a bit here, and what I call place-based people policies could do a lot more. This would not be subsidizing employers with tax breaks for setting up warehouses in enterprise zones, but rather targeting training and human resources funds towards areas with greater needs. Such programs may be less stigmatizing than those based on individuals' characteristics, such as the targeted jobs tax credit.

Finally, good returns may come from greater investments in improving information for kids. A number of mentoring programs provide such connections. Project Strive in Harlem is a good example: It provides training and two years of follow-up services for youth, where they try to make connections with and help resolve problems with employers. States and localities can do a lot to break down the barriers between the offices of central-city Job Training Partnership Act agencies and suburban employers, providing connections beyond just the transportation link.

In conclusion, massive increases in human capital investments would be required to overcome the changes of the past 15 years, increases in the $100-billion-a-year range for a decade, based on some estimates by Jim Heckman. We are certainly not going to embark on such an investment. But in a limited "cut and invest" budget situation, we could probably target our money better. States and localities should be looking at the research on what has worked and what has not, to determine how to use possible future block grants and their current resources. Also, these policies will be more effective in an environment of tighter labor markets.

Panelist Lawrence F. Katz, Professor of Economics, Harvard University.

COPYRIGHT 1996 Federal Reserve Bank of Boston
COPYRIGHT 2004 Gale Group