Business Columns & Blogs

Links between race, education and earnings are complex

Ed Lotterman: Real World Economics
Ed Lotterman: Real World Economics

What people earn on average differs among gender and race. Men earn more than women and whites earn more than members of other races. But beyond such general facts, things get complicated: What jobs are people seeking? What education and experience are they bringing? Are differences between full- and part-time work factored in? It takes work to make true apples-to-apples comparison.

The information in an ongoing study of postgraduate outcomes in my home state of Minnesota’s gives myriad examples of this. While the data is from just one state and one mix of races and ethnic origins, the patterns would be similar elsewhere.

The headline is that two years out of college, whites outearn other groups, with Asians and Hispanics nearly tied for second and third, blacks further behind in fourth, and those considered in two or more races and American Indians well at the bottom. In percentage terms, recent Asian and Hispanic graduates earn about 4 percent less than whites, blacks about 6 percent less and American Indians about 15 percent less.

That is only for people with full-time, year-round jobs. Many recent grads don’t have such jobs. Factor in hourly earnings of those with part-time work and results differ. Asians’ wages are only 5 percent below that of whites, blacks 9 percent, Hispanics 10 percent and American Indians 21 percent.

The fraction of recent grads in part-time work is in inverse proportion to wages, with American Indian highest at near 47 percent and Asians and whites down near 42 percent. Data limitations mean 20 percent of each group is “status unknown,” so one should not draw too-fine conclusions in this category. But the higher the proportion of a group in part-time or seasonal work, the lower its average annual earnings.

The sectors in which people find jobs also is important. Ten percent of Asians in the study work in manufacturing, with a median hourly wage of $22, but only 5 percent of African-Americans, with a median 70 cents less. For American Indians, manufacturing does not even appear in the top 10.

The “Temp Help” rubric is the fourth-most common job for blacks and sixth for Asians. For both groups, it accounts for 6 percent of all jobs,but does not show up in the top 10 for any other group.

“Nursing and residential care” is in the top 10 for all except Asians but varies from more than 14 percent for blacks to less than 5 percent for Hispanics and whites. This is one sector in which the median wage for blacks, at $17.50 an hour, is greater than for any other group, including whites, who have a median 50 cents lower. America Indians in the sector earn 17 percent less than blacks.

Note these are broad “industries” and not specific professions or majors. “Nursing and residential care” includes registered nurses and people collecting linens. “Manufacturing” includes skilled toolmakers and common laborers in shipping and receiving.

There are useful detailed studies from the same data set. One, “Racial Disparities in Wage and Employment After Graduation,” by Alessia Leibert, gives tantalizing hints. For example, black and white men in health care earn equal amounts, but Asians nearly $5 per hour less. For women in the same sector, there is a $1 drop from white to black women and another 50 cents for Asians. In liberal arts jobs, there is only a 70-cent spread from the lowest paid, black men, to the highest paid, white women. In auto mechanics, black and Asian men have identical hourly earnings, but in construction, blacks earn nearly $2 an hour less. In engineering, white men earn some $2.50 an hour more than women, but for Asians, the relationship is almost exactly reversed.

The proportion of a racial or gender group in a particular occupation or sector is also key. For engineering, construction and mechanics, the proportion of all white workers who are in these well-paying careers is more than twice as high as for blacks. Tellingly, out of a population of over 100,000 graduates, only one African-American woman was in engineering. None were in construction and only two Asian women were. There were no female African-American mechanics or repair technicians and only one female Asian in this career.

Nursing, however, is a relatively good field for black women. The proportion of white and Asian registered nurse graduates who work in hospitals, the highest-paying category, is roughly twice that of blacks. But the median wages of the blacks who do work in hospitals is $2 to $3 higher than for Asians and whites. The proportion of blacks working in nursing and residential care facilities is more than twice that of whites or Asians, and again, black wages are $1.50 to $2 higher per hour than for Asians or whites. Nursing is one sector where, overall, the average wages for black women are at least as high as for whites.

This is a lot of detail. The point is to illustrate that many different factors enter into earnings outcomes for racial and gender groups. Looking through the data on occupational earnings and reading the analyses teaches much.

These studies all deal with postsecondary graduates. Within this group, there are different educational levels — from certificate programs of less than a year to doctorates. In general, minorities are over-represented in shorter-duration alternatives and under-represented in baccalaureate and postgraduate programs.

These recent studies represent just the beginning of a multiyear initiative. These tabulations don’t give any clear answers or solutions. But they are valuable in helping us understand the key issues better.

St. Paul economist and writer Edward Lotterman can be reached at