Economic Facts and Fallacies

by Thomas Sowell

Those whose sensibilities are offended by what they see out of airplane windows can of course close the shades. But some prefer instead to disrupt the lives of millions of people on the ground.

When and where there are significant differences between women and men in their employment, pay, or promotion, discrimination can be inferred and, where there has been a lessening of such disparities over time, it has been due to a lessening of discrimination under the pressures of government, the feminist movement or a general increase in enlightenment.

It is one of the central fallacies of our time.

Even in the twenty-first century, “two-thirds of the world’s illiterate adults are women,” according to The Economist magazine. However, at the other end of the educational spectrum, women in the most industrially advanced countries are going on to higher education in numbers comparable to men—and, in some countries, more often than men. In Japan there are 90 women enrolled in higher education for every 100 men, in the United States 140 women for every 100 men and, in Sweden, 150 women for every 100 men.1 Nor is such predominance of women purely quantitative. In 2006, the New York Times reported, “at elite institutions like Harvard, small liberal arts colleges like Dickinson, huge public universities like the University of Wisconsin and U.C.L.A. and smaller ones like Florida Atlantic University, women are walking off with a disproportionate share of the honors degrees.”

The replacement of human muscle by machine power in our own times has so reduced the importance of physical strength that it may be difficult today to imagine how important that factor was in centuries past. For example, at one time desperately poor people in China, living on the edge of starvation, often killed newborn baby girls because only boys were likely to grow strong enough, soon enough, to produce enough food to sustain themselves, and the poorest families had little or no surplus food with which to supplement what a girl could produce with primitive implements on small farms. For such desperately poor people, baby girls were often seen as a threat to the physical survival of the family.

The replacement of human muscle by machine power, and the growing importance of industries and occupations not dependent on either, have made sex differences and age differences no longer as significant as they had once been.

The economic consequences could be seen in the rising age at which people reached their peak earnings, now that experience and skill were more important than physical strength.

Another physical difference between women and men, child-bearing, has continued to have major economic consequences. Mothers as a group tend to fall furthest behind men in income, as competing domestic responsibilities reduce the ability of women with babies and small children to be able to maintain continuous, full-time employment in the workforce. This factor is especially important when it comes to high levels of achievement in the most demanding professions:

In the arts and sciences, forty is the mean age at which peak accomplishment occurs, preceded by years of intense effort mastering the discipline in question. These are precisely the years during which most women must bear children if they are to bear them at all.

In reality, the proportion of women in the professions and other high-level positions was greater during the first decades of the twentieth century than in the middle of the twentieth century—and

Declines in the representation of women among academic faculty during this era occurred even at women’s colleges, run by women, such as Smith, Wellesley, Vassar, and Bryn Mawr,8 so this trend could hardly be attributed to increased male employer discrimination against women.

A closer scrutiny of facts suggests that what changed over these decades was not employer discrimination but women’s marriage and child-bearing patterns.

During the early decades of the twentieth century, when women’s representation in higher level occupations and in the postgraduate education required for such occupations was higher than in the 1950s, the median age at which women first married was also higher than at mid-century.9 Most of the women who staffed women’s colleges during this earlier era were not married at all.10 Neither were most women who taught in elementary and secondary schools, until the late 1940s.11 As the median age of marriage began to decline, the representation of women in high-level occupations and among recipients of postgraduate degrees also declined.

The decline in women’s median age of first marriage ended in 1956 and began to rise thereafter. The birth rate also began to decline, from 1957 on, and by 1966 the birth rate was again as low as it had been back in 1933.12 Women’s share of postgraduate degrees closely followed these reversals of trends in age of marriage and birthrate.

The 1970s saw women’s share of doctoral degrees rise. By 1972 that share was again as high as it had been back in 1932. It was much the same story with Master’s degrees, where it was 1972 before women’s share of these degrees reached the level of 1930, except for the World War II years when millions of young men were away in the military. With both Master’s degrees and doctorates, women’s share declined precipitously after the war to levels below those of the 1930s.13 These were of course the years of the “baby boom,” indicating again the role of child-bearing in limiting women’s educational and career prospects.

Women’s rise in higher-level occupations in the second half of the twentieth century continued to follow the rise in their age of marriage, which rose sharply and finished the century significantly higher than it was at the beginning,14 while the birth rate fell sharply and was much lower at the end of the century than it was at the beginning.15 As the age of first marriage climbed...

However, even if we were to find zero economic differences between those women and men who were truly comparable, that would not mean that women and men as a whole had the same income or the same likelihood of being hired or promoted, if the sexes as a whole were distributed differently between full-time and part-time employment or in different fields or levels of education or in other ways that affect people’s economic prospects. In short, even an absence of discrimination would not mean an absence of male-female economic differences.

The distribution of women and men in various occupations has long differed, partly due to restrictions placed on women and partly due to choices made by women themselves.

Put differently, how much of male-female differences in income has been due to employer discrimination and how much to other differences arising from social restrictions or other factors is a question rather than a foregone conclusion. Many social restrictions, especially in the past, have been based on attempts to forestall problems growing out of the attraction of the sexes for one another.

While families had incentives to curtail women’s work outside the home, employers had countervailing incentives to try to tap this large potential source of workers. Early New England mill owners, for example, tried to reassure parents of the safety and propriety of letting their daughters work in their businesses by having all-female workforces, often overseen by older women who in effect were chaperons, especially when the young women lived away from home.

Even before the industrial era, highly respected and affluent families were likewise able to attract live-in maids, either because the supervision or reputation of these particular families were considered to be some assurance of lower risks of sexual misconduct or because the poorest families needed the daughter’s earnings so much as to have little choice but to take chances that other families would not. The sexual molestation of servant girls by their employers, their employers’ sons, or male servants, was among these dangers, as was the succumbing to temptation by the girls themselves.

Thus whole occupations could be off-limits to women simply because it was mostly men who worked in those occupations and the few women who might want to work in such occupations were not considered by employers to be worth the adverse effect which their presence could have on the productivity of the men.

Although physical strength is no longer as major a factor as it once was in the days when agriculture was the largest economic endeavor in most countries, or during the times when heavy industry or mining dominated various countries’ economies, there are still particular industries today where considerable physical strength remains a requirement.

Women are obviously not as likely to work in such fields as men are—and some of these are fields with jobs that pay more than the national average. While women have been 74 percent of what the U.S. Census Bureau classifies as “clerical and kindred workers,” they have been less than 5 percent of “transport equipment operatives.” In other words, women are far more likely to be sitting behind a desk than to be sitting behind the steering wheel of an eighteen-wheel truck. Women are also less than 4 percent of the workers in “construction, extraction, and maintenance.” They are less than 3 percent of construction workers or loggers, less than 2 percent of roofers or masons, and less than one percent of the mechanics and technicians who service heavy vehicles and mobile equipment.

Such occupational distributions have obvious economic implications, since miners earn nearly double the income of office clerks when both work full-time and year-round.21 There is still a premium paid for workers doing heavy physical work, as well as for hazardous work, which often overlaps work requiring physical strength. While men are 54 percent of the labor force, they are 92 percent of the job-related deaths.

Within the external limitations placed on the range of occupations open to women have been further limitations due to occupational choices made by women themselves. In addition to avoiding occupations requiring more physical strength than most women possess, women have tended to make career choices influenced by the likelihood that they would at some point or other become mothers. Since motherhood has usually entailed a period of withdrawal from full-time work outside the home, the cost of such withdrawal becomes a factor in occupational choices.

Quite aside from formal seniority rules, interruptions of labor force participation, in order to take care of small children until they are old enough to be placed in day care facilities while the mother returns to work outside the home, mean that a woman may have fewer years of job experience than a man of the same age, since such interruptions are less common among men.

Rapidly changing computer technology, for example, means that computer engineers and programmers must be constantly upgrading their skills to keep up with advances in their field. Similarly, tax accountants must keep up with changing tax laws, and attorneys must keep up with changes in laws in general, in order to effectively serve their clientele—and therefore continue to have a clientele to serve. To drop out of such fields and then return in a few years after children have gotten old enough to be put into day care facilities can mean having fallen significantly behind developments in these occupations, and therefore having lower earning capacity, whether as an independent practitioner or as an employee of a firm.

From the standpoint of a young woman looking ahead when making career choices, the relative rates of obsolescence of given knowledge and skills in a given field become a serious consideration in choosing a field in which to specialize.

It has been estimated that a physicist loses half the value of his or her knowledge in four years, while a professor of English would take more than a quarter of a century to lose half the value of the knowledge with which he or she began that career.23

Given the asymmetrical effects of career obsolescence on woman and men, it is hardly surprising that women tend to work in fields with lower rates of obsolescence—as teachers and librarians, for example...

When a multimillion-dollar lawsuit is in progress or a death penalty case is being appealed, the attorneys involved cannot simply quit work at five o’clock and go home. If the case requires working nights and weekends, then the attorneys have to work nights and weekends, in order to build the strongest case they can before they are scheduled to appear in court.

In principle, it does not matter whether the attorney is male or female but, in practice, with women more often than men carrying the burden of domestic responsibilities for children and the care of the home, careers that involve much unpredictable night and weekend work are less attractive to women. Having it all—a career and a family and an upscale lifestyle—is fine but doing it all is often harder for a woman, given the usual division of domestic responsibilities between the sexes and the inevitable differences in childbearing. Those young women who think ahead may take this into account in choosing a career, while older professional women have often decided that “having it all” is not worth the toll it exacts.

In principle, this is the same problem for men and women. In practice, however, a mother is more likely to stay home with the children while the father is tied up at the office, or has to fly off someplace to deal with legal emergencies, than a father is to stay home while the mother does the same.

A Harvard Business Review survey among people whose earnings were in the top 6 percent showed that 62 percent worked more than 50 hours a week and 35 percent worked more than 60 hours a week. Among those who held “extreme” jobs—extreme in both hours and stress—less than one-fifth were women. Moreover, even among those people who held such high-pressure jobs, women were only half as likely as men to say that they wanted to still be working like this five years afterwards.

I have been presumptuous enough to counsel new Ph.D.’s in biology as follows: If you choose an academic career you will need forty hours a week to perform teaching and administrative duties, another twenty hours on top of that to conduct respectable research, and still another twenty hours to accomplish really important research.

In general, men and women alike tend to prefer regular hours and less stressful work, so that jobs with these characteristics can attract both sexes more readily and, because of supply and demand, pay less than similar jobs in more taxing situations.

The main reason why women still get paid less on average than men is not that they are paid less for the same jobs but that they tend not to climb so far up the career ladder, or they choose lower-paid occupations, such as nursing and teaching.

The most important reason why women earn less than men is not that they are paid less for doing the very same work but that they are distributed differently among jobs and have fewer hours and less continuity in the labor force.

Among college-educated, never-married individuals with no children who worked full-time and were from 40 to 64 years old—that is, beyond the child-bearing years—men averaged $40,000 a year in income, while women averaged $47,000.

Even women who have graduated from top-level universities like Harvard and Yale have not worked full-time, or worked at all, to the same extent that male graduates of these same institutions have. Among Yale alumni in their forties, “only 56 percent of the women still worked, compared with 90 percent of the men,” according to the New York Times.32 It was much the same story at Harvard:

While those women are the best judges of what suits their own individual circumstances, priorities, and sense of well-being, third parties looking at statistical data see only the artifacts of disparities based on paychecks.