BLS, Claudia Golden, DOL, FiveThirtyEight, Gender Discrimination, Gender Gap, High-Risk Occupations, Job Flexibility, Mark Perry, Millennial Pay Gap, Mutually Beneficial Trade, Non-Wage Compensation, Obama Discrimination Data Mandate, Occupational Choice, Pay Differentials, Risk Aversion, STEM, The Economist, Tyler Cowen
Debates on social issues are often plagued by facile comparisons that distort the underlying facts. The alleged gender pay gap involves such comparisons. The Obama Administration proposed new rules last month intended to address a difference in median earnings between men and women, demanding data reports on various demographics from firms with 100+ employees. Mark Perry points out that the pay gap in the Obama White House is about the same as the national difference. Can there be any reasonable explanation for these disparities?
One key to understanding the debate is that the difference in aggregate pay between men and women (17% in 2014, according to the Bureau of Labor Statistics) is not a divergence in pay for equal work! However, that is the gist of the fraudulent narrative so often heard from the White House and elsewhere. The truth: 17% is the difference in the medians of two large distributions of working adults, one for men, one for women, covering all occupational categories. The discrepancy, which has declined sharply over the past 35 years, is explained today by fewer hours worked among women and “differences in educational attainment, work experience, and occupational choice.” These differences are well known, but gender-gap warriors conveniently overlook the following facts, as established by the Department of Labor:
- There is more part-time work among women;
- Women lose more experience to childbirth, child care and elder care;
- Women demand more job flexibility and non-wage benefits (and that costs);
- Women are disproportionately under-represented in dangerous occupations;
- Women are disproportionately under-represented in STEM fields (Science, Technology, Engineering, Math);
Interestingly, the last point may have more to do with a broader range of talents possessed by females who are skilled at math, relative to men, which leads to a greater variety of career options. An implication: non-STEM occupational choices by women are often voluntary and not the result of discrimination. And those choices are often driven by considerations other than cash remuneration.
As to the risk of physical danger, in 2010, men were almost 12 times as likely as women to suffer fatal injuries on the job. There is no question that high-risk occupations have higher wages. Apparently, women choose not to pursue opportunities in these occupations. An earlier study found that single parents, male and female, were the most risk averse in their choice of occupation, and that married women with children are more risk averse than married men with children. Of course, it is possible that some employers have requirements in terms of physical strength that favor men. Either way, the job-risk gap almost certainly contributes to the measured-wage gender gap, but it has little to do with gender discrimination per se.
Earnings are sensitive to factors such as full-time / part-time status, continuous job tenure, and the likelihood of extended leaves of absence. This is supported by a research finding cited in The Economist, that partners in lesbian relationships tend to out-earn married straight females. The division of responsibilities in the home is surely part of the story: lesbian couples tend to split chores more equally than straight couples. Millennial couples (ages 25-34) are also more likely to split household chores equally; the gender pay gap for millennials is much narrower than for older age cohorts, and it is nonexistent for childless millennials. Millennial women have more than closed the education gap as well.
When gender differences in hours, tenure, absences, education, and job hazards are considered, as well as the full menu of compensating non-wage benefits available, the wage gap is essentially nonexistent. Yet President Obama’s proposed data mandate would carry high compliance costs and likely cost jobs as well. The purpose of the regulation is to make it easier for various groups to sue employers on the basis of wage discrimination. But observation of such a gap, wherever it might exist, is not prime facie evidence of discrimination; it is more than likely to be the result of private, voluntary agreement.
Is it possible that certain attitudes or behavioral characteristics of women generalize to poorer outcomes, relative to men, in negotiations? Tyler Cowan reports on research that suggests as much, based on “laboratory” experiments in which participants played repeated games involving actual rewards. In one experiment, the rewards depended on the acceptance of an offer to share a pot, and both men and women made lower offers to female partners than to males. However, when the partner was a woman, females were markedly stingier in their offers than males. Those women are tough! But seldom are real-world “deals” so one-dimensional, and controlling for all considerations of value is often impossible. In any case, trades rarely take place when the parties don’t find them to be mutually beneficial.
Fortunately, in labor markets, when differentials in skills and experience matter, discrimination is practiced only under a self-inflicted penalty on the discriminator. In the case of wage-based gender discrimination, the employer will tend to overpay for equivalently-skilled male help. Discrimination of this sort impairs a firm’s ability to attract the best employees and harms its competitive position. Nevertheless, the extent to which the market’s self-regulation confers benefits on individual participants depends upon their vigilance: buyer beware (caveat emptor) and seller beware (caveat venditor) are keys to real economic freedom. Most importantly, in all things, beware government edicts. Markets are the best regulator.
Sidebar: I was referred to an article on FiveThirtyEight by my friend John Crawford. The main subject matter of the article is off-topic and its conclusions are incorrect (I might post on it soon), but many of the charts are interesting; the third chart is really fascinating! It shows that women, by age 30, tend to belong to households that are higher in the income distribution than men who come from the same point in the distribution of household-income in childhood. This is true at every point in the childhood household-income distribution! Are there advantage(s) for women that can account for this? A few guesses: a lower rate of incarceration of women by age 30; women have higher marriage rates by age 30; women “marry up” more than men, both in terms of the ages and incomes of their spouses; women who don’t marry live with their parents more than men do (?). There could be other explanations, and the relationship may not hold at later ages. Still, it’s noteworthy that such a reverse “gender gap” exists in the data.
I close with a quote from Harvard’s Claudia Golden, from “A Grand Gender Convergence: Its Last Chapter” (HT: Marginal Revolution):
“The gap is much lower than it had once been and the decline has been largely due to an increase in the productive human capital of women relative to men. Education at all levels increased for women relative to men and the fields that women pursue in college and beyond shifted to the more remunerative and career-oriented ones. Job experience of women also expanded with increased labor force participation. The portion of the difference in earnings by gender that was once due to differences in productive characteristics has largely been eliminated.
What, then, is the cause of the remaining pay gap? Quite simply the gap exists because hours of work in many occupations are worth more when given at particular moments and when the hours are more continuous. That is, in many occupations earnings have a nonlinear relationship with respect to hours. A flexible schedule comes at a high price, particularly in the corporate, finance and legal worlds.“