April 20, 2014

NG | When Debunking the Gender Pay Gap Gets Awkward

Print More


Mark J. Perry and Andrew G. Biggs, scholars at the American Enterprise Institute, wrote in a Wall Street Journal column that the Obama administration’s push to close the gender wage gap is deceitful. “The numbers bandied about to make the claim of widespread discrimination are fundamentally misleading and economically illogical,” Perry and Biggs claim, because once you control for educational background, job risk and industry, the wage gap between men and women disappear. The title proclaims that the “77 cents on the dollar” battle cry is a myth.

Part of Perry and Biggs’ argument focuses on the Obama administration’s use of aggregate wage measures to come up with 77 cents. Comparing median weekly earnings of all men to all women fails to account that men work more, that men choose riskier jobs and that working mothers choose jobs with more flexibility. Aggregate wage measures are too blunt, and they are right. But repudiating aggregate wage comparisons for other factors like occupational choices and marital status muddles the issue of the gender wage gap, contrary to the word Perry and Biggs chose — “myth” — that implies they succinctly debunked the existence of the issue.

Attributing the “wage gap” to different occupational choices requires explaining why women “choose” lower paying jobs or “choose” jobs that happen to pay lower. “Women often choose fields of study, such as sociology, liberal arts or psychology, that pay less in the labor market,” they affirm. Okay, sure: but why? What is Perry and Biggs’s explanation that women happen to choose lower-paying, less risky or flexible jobs? If women clearly know that they could earn more, why don’t they make their husbands stay at home so they could become financiers and engineers? To this, they remain oddly silent, playing the act of messengers simply describing the world as it is. This deafening silence is revealing, for such questions are on the cusp of the rabbit hole of “choice” and social expectations that the debunkers seem keen to avoid. In trying to show that gender pay inequality is not so bad, the debunkers realize they must point fingers at a much larger, fractious and nebulous cause beyond “economic choice” that they simply do not want to touch.

It is easy to see why the debunkers, often in the neoclassical economics vein, tip-toe around discussing the “real cause,” for the cause is antithetical to the “rational expectations” metaphors they use. The intellectual historian Daniel T. Rodgers, in his elegant and eloquent book Age of Fracture, charts the rise of this rational expectations movement is closely tied with Chicago School economist Robert Lucas. “All the ‘irrationalities’ that psychoanalysts had relegated to the unconscious, that anthropologists had relegated to ‘culture’ … could be more compellingly explained by shifts in costs played out against stable preferences. Economic actors did not simply react to economic information: Rather, Lucas maintained, they learned to anticipate economic actions, decoding the rules of thumb of other economic actors and foreseeing their line of action.”  Markets were aggregates of anonymous, fluid actors anticipating policy outcomes with lightning-quick accuracy. Socially detached, with no gender, race and historical structural constraints, these rational actors made decisions based purely on their own preferences and resources. Situations that disadvantaged women in the work field like those that spurred the Fair Credit Reporting Act, among others legislative attempts to equalize the economic field for women, are delegated as mere footnotes in history.

The conceptual mismatch — using a model that assumes no structural constraints to test whether a structural constraint exists — is not only circularly reasoned, but awkward because it attributes the effects of real structural constraints on an individual’s “preference.” “The choice of college major is quite free. … Likewise, many women prefer to stay home with their children,” Perry and Biggs generalize. Ramesh Ponnuru, writing in Bloomberg View, responds similarly: “There’s no reason to think that women will ever, on average, have the same preferences as men about combining employment and parenthood.” These explanations are unsatisfying. Even within the high-skill, high-paying jobs that the debunkers weirdly confine their analyses to, outright gender discrimination is rampant. It is possible that women choose to not enter these industries because they have ‘maternal instincts’ or ‘lack confidence’ — but it is just as likely that they are discriminated against. And regardless of biological imperatives or confidence, it is unclear why we should embrace an economic system that values ‘feminine’ labor less than plain old labor.

Outside of the high-paying jobs that the debunkers focus on are minimum wage industries that are, more often than not, predominantly female than predominantly male. The trade-off the debunkers make between a smaller salary and more family time falls apart in these industries: Because of a lower wage, timekeeping clerks, maids and house cleaners must work full-time only to earn less. “If women were paid 77 cents on the dollar, a profit-oriented firm could dramatically cut labor costs by replacing male employees with females,” Perry and Briggs theorize. There is no need to speculate; the numbers show that firms already do this.

And for the econometrics nerds: Perry and Biggs fail to control that women could opt to work less because of smaller salaries. By conflating the relationship between wages on hours and hours on wages, their arguments are biased towards seeing the relationship as one-sided.

The 77 cents figure might not be entirely accurate, and the gap might not be entirely discriminatory. Explaining the gender wage gap is awfully complicated, with factors jumbled on factors. But don’t describe it as a myth and justify that with a stick diagram of supply and demand.