Is the Growing Gender Gap in College Enrollments a Cause for Concern?

A report released last Tuesday by the American Council on Education, and discussed in various media articles this week, indicates that over 55% of college students are women. This reflects a continuing upward trend in women's share of enrollments for the past 30 years. It is ironic that an earlier 1992 study claimed that colleges were biased against women because women were intimidated to speak up, the type of course work emphasized favored men, etc. In a political response to at most a minor problem, Congress unwisely passed "gender equity" legislation during the 1990's.

The gender gap in enrollments is especially large for lower income African-Americans and Latinos, and is negligible for children from middle and upper income white families. I do not believe there is reason to be concerned about the overall growth in the relative number of women college students-good for them- but I continue to worry about the performance of African American and Latino young men.

On pretty much all objective measures, women deserve to have greater college representation than men because they study harder, get better grades, are more likely to graduate from high school, complete their school work in a more timely fashion, write better, and in other ways outperform young men. Schools competing in trying to get the best students naturally respond to this, and end up selecting larger numbers of young women than young men. Women still remain a minority, however, in the sciences, engineering, business, and economics.

The trend toward increased college enrollment of women will continue the growth in the education of women in the labor force compared to that of men. This should further narrow the gender gap in earnings, a gap that has already narrowed greatly since the mid-1970's. Since the education of younger women is exceeding that of men, will the gender earnings gap begin to reverse signs, so that women will earn more than men?

In answering this question, first note the study by Francine Blau and Lawrence Kahn, which shows that the gender pay convergence slowed during the 1990's even though the education of women in the labor force continued to grow relative to that of men. This slowdown in convergence is consistent with my belief that earnings of the average women in the labor force will not rise above that of the average man, although an increasing fraction of women in the labor force will have higher hourly earnings than men. While the gap between the education of women and men in the labor force will continue to grow, the commitment of women to careers will remain below that of men, despite the claims about their career ambitions from the selected college women in the media stories on the enrollment gap.

Women will continue to have much greater responsibilities for child care than men do. That means sometimes long periods of being out of the labor force, more reluctance to work overtime, less willingness to take jobs that require much out of town travel, greater likelihood of taking absences to care for sick children, and other behavior that lowers both hourly earnings and hours worked. All these differences continue to be found in Sweden, perhaps the country with the greatest degree of gender equality, and they would apply in even greater force to the United States and other countries.

Although young women do considerable better on average then men in school, the variance in performance among men is much greater than the variance among women, and admisssion policies should depend on variance as well as mean performance. Due to their greater variance, many more men drop out of school, have failing grades, study little, have disciplinary problems, and the like. Greater variance also implies, however, that many more of the outstanding students are men. Certainly after many decades of teaching economics, I would confirm that my female students have done better on the average, while the men were more likely to be at both tails of the distribution; that is, men were more often both very bad and very good. I would hasten to add, however, that I have had a considerable number of very exceptional female students too.

Larry Summers got into trouble by suggesting that the variance in gender difference in achievement- a gender differ in performance is found on many dimensions of behavior- might have a genetic basis. It surely might for reasons put forward by many biologists, but I believe (and I am confident that Summers would agree) that the difference in variance is mainly explained by interactions between genetic and environmental forces. Young girls may be discouraged from high achievement, or young women may recognize that they will have and want childcare responsibilities, and realize that this will cut down on their career commitments.

Of great social concern is the very poor performance by African American and Hispanic young men compared to young women of the same race or ethnicity. African American young men not only tend to drop out of high school more and are less likely to go to college, but the men are also far more likely to end up as delinquents, in jail, murdered, unemployed, and in other bad circumstances. This to me is the most serious racial issue in the United States, and is only partly reflected in the much higher college enrollment rates of African American women than men.

Perhaps African American boys are more affected than girls by the absence of fathers in their households, or negative peer pressure is more harmful to boys, or drug selling and other crimes is more appealing to them compared to school, and so on. I am not going to try to solve such a major problem in this post, except to indicate that legalizing drugs would help African American young men, and so too would any steps that can be taken to stabilize the family structure of African Americans.

The final issue I address is whether it is proper for colleges to use an affirmative action plan for men; that is, to have easier admission standards for male applicants to bring enrollments closer to 50-50 for men and women. I believe it is a perfectly legitimate strategy. Since the US higher education system is highly competitive, different schools should be allowed to choose their policies on these and most other issues. Then they compete for students and for funds from donors by offering different programs, including the ratio of female to male students.

An additional factor in this case is that usually does not apply to affirmative action programs to help racial or ethnic minorities is that the group facing higher standards, female applicants, often want schools to try to get more male students so that their social life would be better. After all, most schools that formerly had students of only one sex-such as Princeton and Vassar- have become coed to provide a better social and perhaps also intellectual life. Since affirmative action toward men would be supported not only by men, but also by many women, easier standards for male applicants seems to be a desirable policy for many colleges.