Discussion - comment on John F. Kain and Kraig Singleton, in this issue - Special Issue: Earnings Inequality

I agree with the basic argument of John Bishop's paper, that college will continue to be a good investment in the years to come. I would like to address four questions raised by his paper. The first three might be asked by readers of media reports about hard times for college graduates. Many of these reports, as Bishop points out, are based on projections made by the U.S. Bureau of Labor Statistics. My responses to the questions posed by hypothetical readers are based on a paper that John Tyler, Frank Levy, and I recently published. The questions are as follows:

* Why are so many college graduates anxious about their economic positions, given that they are faring so much better than high school graduates?

* Is there any group of college graduates whose labor market experiences during the 1980s fit the gloomy prediction of the BLS?

* How can John Bishop's optimistic picture of the attractiveness of investing in college be reconciled with everyone's experience of knowing a college graduate from the class of 1994 or 1995 who is working at a coffee bar?

* How can states most effectively help low-income high school graduates pay for college?

Why Are Recent College Graduates (Especially Males) Anxious?

The first line of Table 1 lists the median earnings for 25- to 34-year-old male college graduates in the years 1979, 1989, and 1993, all in 1993 dollars. The second line lists comparable information for 25- to 34-year-old male high school graduates. The third line lists the college graduate/high school graduate earnings ratio, showing clearly the growth in the education-related earnings premium. However, another look at the top line shows that the median earnings of young male college graduates grew only 2 percent between 1979 and 1989 and fell by 4 percent during the subsequent four years.

As the figures in the second line make clear, the education-related earnings differential rose because of the dramatic decline in the earnings of high school graduates. The median income of 25- to 34-year-old high school graduates in 1993 was $20,000; this is less than three-fourths of the median earnings (in 1993 dollars) for the comparable group in 1979. Thus, while young college graduates are certainly better off relative to high school graduates today than was the case in 1979, young male college graduates do not have significantly higher incomes today than their counterparts did 15 years ago. This is one factor that contributes to their anxiety. Another is that most have larger college debt burdens than did their counterparts 15 years ago. This is a consequence of the rapid tuition increases, particularly at public colleges, that Bishop describes in his paper.

Is There Any Group of College Graduates Whose Labor Market Experiences Fit the Stories of the Media Jeremiads?

Table 2 lists two indicators of economic well-being in 1979 and 1989 for four groups of college graduates: 25- to 34-year-old women, 25- to 34-year-old males, 45- to 54-year-old women, and 45- to 54-year-old males. The two indicators are median earnings (in 1993 dollars) and the percentage of the group holding jobs classified as "high school jobs" by economists at the BLS. As the numbers make clear, young women improved their position over the decade of the 1980s; young men held their own; older women improved their position. The one group in a worse average economic position in 1989 than the comparable group in 1979 is the 45- to 54-year-old men. As Bishop explains in his paper, there are reasons to be cautious in interpreting changes over time in the percentage of a group employed in what the BLS refers to as "high school jobs." But the decline over the 1980s in the median real earnings of older male college graduates is unequivocal. The difference between the declining economic position during the 1980s of older men and the stable or improving position for other groups of college graduates is missing from most BLS studies, because they tend to group all college graduates together in their analyses.

Table 1
Median Earnings of 25- to 34-Year-Old Males and Females, in 1993
Dollars

                                    1979        1989        1993

Males       College Graduates     $31,579     $32,336     $31,000

            High School
            Graduates             $27,427     $22,791     $20,000

            Earnings Ratio:
            Coli. Grad/
            H.S. Grad                1.15        1.42        1.55

Females     College Graduates     $19,593     $24,252     $24,340

            High School
            Graduates             $13,719     $13,858     $14,000

            Earnings Ratio:
            Coli. Grad/
            H.S. Grad                1.43        1.75        1.74

Source: The 1979 and 1989 earnings figures were calculated from the
1980 and 1990 Public Use Microdata 1 Percent Samples of the U.S.
Census of Population and Housing. The 1993 earnings figures were
calculated from the March 1994 Current Population Survey.
Table 2
Indicators of Economic Well-Being in 1979 and 1989 for Four Groups
of Four-Year College Graduates

Cohort                            1979          1989       Change

Young Women (25 to 34)
Median Earnings (1993 $)        $19,593       $24,252       23.8
Percent in Jobs Requiring
High School Education              28.2          25.2       -3.0(a)

Young Men (25 to 34)
Median Earnings                 $31,579       $32,336        2.4
Percent in Jobs Requiring
High School Education              25.0          23.2       -1.8(a)

Older Women (45 to 54)
Median Earnings                 $21,552       $26,561       23.2
Percent in Jobs Requiring
High School Education              27.6          23.6       -4.0(a)

Older Men (45 to 54)
Median Earnings                 $52,886       $50,814       -4.1
Percent in Jobs Requiring
School Education                   14.5          17.9        3.4(a)

a In percentage points.

Note: Data are for college graduates who worked at least one week
during 1979 or 1989. All statistics reported in this table were
calculated from the 1980 and 1990 Public Use Microdata 1 Percent
Samples of the U.S. Census of Population and Housing.

What about the College Graduates Working in Coffee Bars?

Figure 1 displays the median earnings of male high school graduates and male college graduates from the ages of 22 to 30, in 1989 and in 1993. The very low median earnings for college graduates at age 22 and 23 are consistent with everyone's anecdotes about college graduates working in coffee bars. But the earnings at age 30 are consistent with the earnings figures reported earlier. So the story is that it does take many college graduates a couple of years to find their way into jobs that have career potential. But it does happen. Moreover, it happened in 1993 as well as in 1989. In fact, the projected earnings profiles for college graduates in the two years are remarkably similar.

In contrast, the charted earnings profiles for male high school graduates are much more shallow, illustrating that not only do they have low earnings at age 22, they also have quite low earnings at age 30. Note also that at each age level, the median earnings of male high school graduates are lower in 1993 than in 1989, illustrating that the deterioration in the earnings of male high school graduates that took place during the 1980s has continued into the 1990s.

How Can States Most Effectively Help Low-Income High School Graduates Pay for College?

In closing, I would like to comment on Bishop's argument that the escalation of tuitions at public colleges and universities should stop because it is keeping many students from college, especially students from low-income families, who could benefit from this investment. I share his concern with the problem. However, I wonder whether a better response than arguing for tuition rollbacks might not be found. Currently, the states contribute about $40 billion to public post-secondary education. Over 90 percent is used to subsidize tuition rates. Given the fiscal situations in most states, this amount is unlikely to increase over the coming years, which means that it will be difficult even to maintain current tuition levels, never mind reduce them.

It seems to me that a better policy would be to allow tuitions to increase and use a significant part of the increased revenue for need-based financial aid, targeted to low-income students, whose enrollment decisions are especially sensitive to costs. This will work only if the availability of such aid is well publicized, and if significant steps are taken to simplify the application process for student aid. But these steps seem manageable, and targeting more of state financial assistance to low-income students seems important to do, given the tight fiscal situation.

Reference

Tyler, John, Richard J. Murnane, and Frank Levy. 1995. "Are more college graduates really taking 'high school' jobs?" Monthly Labor Review, vol. 118, no. 12, December, pp. 18-27.

COPYRIGHT 1996 Federal Reserve Bank of Boston
COPYRIGHT 2004 Gale Group