Insights on Changes in Lifetime Earnings Inequality and Educational Attainment

Data about income inequality are usually snapshots of the income distribution at a particular period of time—the equivalent of a photo that shows how much certain individuals made in one year. But sometimes we want to know how the distribution changed over longer periods of time by following the same individuals as they move along in their careers. In other words, we’d like an extended video clip of the income distribution.

A recently released Urban Institute report by economist Joshua Mitchell provides one such metaphorical video. Two of the report’s main findings are one well-known fact and an under-appreciated one. First, inequality between men of different educational levels is increasing. Men with college degrees are increasingly earning more than men who only went to high school. That’s a story most readers will be familiar with.

But Mitchell also finds that inequality within educational groups is rising as well. The difference between the earnings of a highly-paid member of an educational group, say high school graduates, and the earnings of a low-paid paid member of the same group has been increasing. This second fact has important implications for how we understand the rise in income inequality and the relative importance of boosting college attendance.

Mitchell used Social Security earnings data to produce cumulative earnings for participants in the Census Bureau’ Survey of Income and Program Participation, or SIPP. This allows him to link income information to education attainment for men in birth cohorts from 1940 to 1974. And one important note: Mitchell looks at only men because including women in these long-term trends can be challenging due to the large increase in female labor force participation over these years. That trend would complicate the analysis and make drawing lessons from the data more difficult.

The first major result from the report is summarized by the paper’s Figure 5. The chart shows the cumulative earnings of men at ages 43 to 47 by educational group and the years they were born. The first group of bars shows the cumulative earnings for men born in 1940 through 1944 with the level of educational attainment increasing left to right.

As you can see, the median men with college and advanced degrees always earned more than the median man in the other groups, but the relative difference has been increasing with each cohort. Thus, the value of a college degree and further education has been increasing with time.

The second major result can be found in the paper’s Figure 7. The chart is showing the distance between the earnings of a man at the 25th percentile of an educational group and the earnings at the 75th percentile. The larger the number on the y-axis, the higher the inequality within the educational group. And Mitchell’s analysis clearly shows the rising inequality within groups. The increase has been most dramatic for high school graduates.

How does this new view of the data help us better understand our economy? First of all, Mitchell’s findings complicate the traditional story of why income inequality has risen so much. The widely accepted narrative, called skill-based technological change, or SBTC, explains rising inequality as the result of rising demand for highly skilled labor due to technology interacting with a supply of skilled labor that isn’t rising as quickly. The result is higher pay and incomes for college educated workers and rising inequality. Mitchell does find that more highly educated workers are paid more, which supports the skill-based technological change theory, but the rising within-group inequality he documents means SBTC can’t be the sole reason for rising income inequality.

One explanation for this increasing within-group income inequality could be an increase in employers’ ability to discern the performance of different workers. This could result in high-performing workers of any education tier becoming better compensated over time. Certainly improvements in performance measurement made possible by advances in information technology make this plausible, but the decline of unions could also be behind this trend.

Mitchell’s analysis also sheds light on why college attendance and completion rates for men haven’t been increasing as much as we’d expect given rising average pay for college graduates. The key is that rising inequality between educational groups means that there are high school graduates at the high end of the distribution who earned more than those at the bottom of the college distribution. So for some men, not going to college is a perfectly rational economic decision.

This point has been made before by Equitable Growth’s Executive Director, Heather Boushey, Center for Economic and Policy Research economist John Schmitt, and Thomas Piketty in his new book. Mitchell’s work further strengthens this claim. For anyone wondering why women are now outstripping men when it comes to college degrees, Mitchell’s analysis also provides a possible answer: college just isn’t worth it for some people.