Slate is having a healthy, but incomplete, debate about the uproar about Brendan Eich’s resignation from Mozilla. Eich donated $1000 to the successful campaign to adopt “Proposition 8” in California in 2008. Prop 8, until it was struck down, banned marriage equality for gays. William Saletan published a satirical article suggesting that everyone be “purge[d]” who contributed to Prop 8.
Other columnists, such as Mark Stern, weighed in to remind readers about the cruelty of the often homophobic TV ad campaign used by Prop 8 supporters. Stern makes the point that much of the campaign was designed to picture gays as recruiting straight children. This column (eventually) discusses why Eich stepped down, but it begins by explaining why neoclassical economists have such a terrible track record in understanding discrimination and its remedies.
Neoclassical Economics’ Inability to Comprehend Discrimination
What does economics have to say about employment discrimination? Neoclassical economics’ inability to comprehend one of the most common and disabling social ills – discrimination – is a classic example of its limitations. Neoclassical economists suffer from a self-inflicted limitation that cripples their ability to discuss discrimination honestly. Under the standard assumption of rationality essential to all their models discrimination cannot exist because it is irrational. Neoclassical economists could respond to the failure of their models by changing their models, but that would cause them to have to admit massive errors and raise the question – why are we listening to these folks? Neoclassical economists have taken two primary approaches to attempt to escape this trap of their own (irrational) design that insists on ignoring so much human behavior.
The First Cheat: A “Taste” for Bigotry
One neoclassical answer to the dilemma is a cheat. There are two sub-variants. In the first, bigotry, which has so often proven murderous, is sanitized, even made positive, by rebranding it as a “taste.” When we say that someone has “taste” it is a compliment. It also (seems) to solve the rationality problem because neoclassical economists are used to (assuming, incorrectly) that tastes are innate (“exogenous”) and stable rather than often being driven by advertising and other social forces. Some people have a taste for vanilla and some have a taste for chocolate. Neoclassical economists assume that all such tastes are equally valid. We don’t see many mass murders in ice cream stores between holders of these diverse tastes. I’m not aware the vanilla ice cream lovers have ever asserted the right to enslave the chocolate lovers.
Gary Becker asserts (contrary to all human experience) that employers who have a “taste” for discrimination can produce segregation at some firms, but only if the bigots are willing to pay a price – a price that will inevitably cause them to be swept away by their unbigoted competitors if markets are sufficiently competitive. The real victim of discrimination becomes the bigot (and the bigot’s shareholders) rather than the bigots’ targets. Becker implicitly assumes that the bigots will go quietly out of business rather than creating extra-legal means of harming firms run by non-bigots. His implicit assumption is often false, but because he makes it implicitly he never needs to defend it and probably does not even recognize he has made the assumption. If he understood history, sociology, or anthropology he would not make the error, but as the leader of the “Imperial Science” who emphasizes that other disciplines’ research isn’t worth reading Becker founders like many other deluded imperialists before him who never understood why their invasion of Afghanistan ended in disaster.
Neoclassical economists who chose to “rebrand” bigotry as “taste” know exactly what they are doing. It’s a cheat – the violation of the rationality assumption has simply been obscured. It is great press for bigots because it not only legitimizes their actions but holds that there is no need for laws against employment discrimination because the market will promptly cure all ills.
The Second Cheat: “Statistical Discrimination” is “Efficient”
The second neo-classical cheat is to give up the standard assumption on perfect, cost-free information. They modify this assumption to argue that “statistical” discrimination is efficient. Traits such as race, gender, sexual orientation, national identity, and age become under this treatment excellent (albeit inherently imperfect) predictors of employee performance. Becker claimed in a book that have proven to be an embarrassments (A Treatise on the Family), women are inherently inferior to men as workers in most fields. There may be a few female “deviants” who it would make sense to train for “male” jobs if we could predict their (deeply unlikely) future success in “male jobs” when they were five years old, but we cannot. Given this inability to identify now the tiny group of potentially successful “deviant” girls, it makes no sense to provide girls the option of undertaking the years of training necessary to qualify the few “deviants” who would actually be as good as males in such jobs. Becker claimed that he used the term “deviant” only in its statistical sense – because, you know, there was no other word for that concept that avoided the obvious and bigoted double insult to both women and lesbians. Come to think of it, he could have used the word “exceptional.”
“Let me emphasize that ‘deviant’ is used in a statistical, and not a pejorative sense.”
But if you are tempted to believe such an obvious lie, consider this sentence.
“Deviant behavior would presumably be more common if deviant biology were more common.”
And what type of “deviant biology” does Becker claim would explain women’s “common” ability to be lawyers, doctors, chemists, CEOs, saleswomen, or hairdressers? It has to be something in the brain, because Becker is not talking about the ability of women to do jobs that require the ability to bench press 400 pounds – he’s talking about all paying jobs.
Under Becker’s view stereotypes are useful heuristics that developed because society and parents discovered that the stereotypes efficiently sort children in order to optimize scarce training resources. Sexist stereotypes based on spurious, unspecified “biology” should rule our approach to the (non) education of females – that’s just good “science.”
Becker decried the entry of women into the paid workforce as reducing “specialization” and resultant “efficiency.” He married (pun intended) the (also deeply flawed) concept of “comparative advantages” from international trade with Smith’s parable of the pins (Smith’s ode to the efficiency of specialization) to claim that the classic “generalist” “job” (the “traditional stay-at-home mom”) represented the epitome of “specialization.” (Yes, he was blind to the irony.) I need to emphasize that I am not making any of this up. Becker also expressly called “children” a “commodity.” Becker claimed that “women” (who he implicitly assumed were essentially a homogenous “entity”) had a comparative advantage in the household over men because they can lactate. He emphasized that even the most minor comparative advantage should lead to strict gender “specialization” in which women would be stay-at-home moms while all men should work in paid sector.
Of course, the reality is that some women (and some men) specialize in a host of “household” specialties, e.g., childcare and house cleaning), but they are paid for doing so. Other women specialize in being chemists, lawyers, and everything else in the paid sector. The result is far greater specialization than under Becker’s sham specialization. This was already apparent at the time Becker began researching his infamous Treatise. Becker’s Nobel address stresses his exhaustive research on the Treatise (published in 1981). By 1982, 26.8% of medical school graduates and one-third of law school graduates were women. That’s a lot of “deviants” in “traditionally male” jobs. Today, roughly half of med and law school grads are women.
“Writing A Treatise on the Family is the most difficult sustained intellectual effort I have undertaken.
Trying to cover this broad subject required a degree of mental commitment over more than six years, during many nighttime as well as daytime hours, that left me intellectually and emotionally exhausted.”
In addition to his obvious sexism, it is plain that Becker remained clueless about women as individuals with different desires, skills, and dreams. Instead, he saw them as a homogenous input in a production function. His “six years” of research and writing was not merely “emotionally exhaust[ing]” – it was also bizarre, for he must never have seriously sought out a female perspective. He was such a prisoner of his own prejudices that he could never understand the imprisoning aspects of prejudice on the lives of women.
Becker Receives the Nobel for Proving Embarrassingly Wrong
Becker’s “scientific” support for the “efficiency” of “statistical discrimination” proved to be another theoclassical embarrassment, but the Nobel committee loved his “Imperial” forays and gave him the Prize in 1992 as the economist who failed in more fields than any other economist. The committee’s short statement of the basis for the award was all about rewarding imperialism.
“The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 1992 was awarded to Gary S. Becker ‘for having extended the domain of microeconomic analysis to a wide range of human behaviour and interaction, including nonmarket behaviour’.”
By 1992, there was no conceivable defense of Becker’s claims about the efficiency of gender specialization. Census data show that male and female part-time labor participation rates were equivalent and full-time (paid) labor participation rate for women was about 60% of the male rate (see Figure A, below). Becker’s award is another illustration of the Swedish Central Bank’s uncanny ability to award a Prize at the very moment that the falsity of the winner’s theories becomes obvious to anyone that follows the field. 1992 was the first year in which the percentage of women high school graduates aged 25-29 who had obtained at least a college degree exceeded the rate for men of that same cohort (28% v. 27%).
“In 1992, women earned more associate’s, bachelor’s, and master’s degrees than men, whereas in 1977 the reverse was true. Though fewer doctoral and first-professional degrees were awarded to women than to men in 1992, the gap has narrowed considerably over time. For example, the percentage of first-professional degrees earned by women rose dramatically between 1960 and 1993: from 2 percent to 42 percent of all law degrees; from 6 percent to 38 percent of all medical degrees; and from 1percent to 34 percent of all dentistry degrees.”
Paid labor force participation rates for women in the prime adult working years were above 70% by 1992 – and had risen steadily since Becker’s first published his (falsified) critique of the loss of “specialization” and “efficiency” that purportedly arose from having women work in the paid sector.
The situation today demonstrates that had we followed Becker’s recommendations about not bothering to educate women for “deviant” careers we would not only have denied half our population the educational and job opportunities they cherish, but done irreparable harm to our economy.
“ACADEMIA & EDUCATION
Key Facts from Academia & Education pulled from sources below.
Women represent 51% of the nation’s PhDs, 51% of business school applicants, 67% of college graduates, and more than 70% of 2012 Valedictorians in the US.
Nationally, about 58 percent of US college undergraduates are women, with some campuses at 70 percent.
Nationally, women are 57 percent of all US college students but only 26 percent of full professors, 23 percent of university presidents and 14 percent of presidents at the doctoral degree-granting institutions.
In 2011, 37 percent of young women had at least a bachelor’s degree compared to 29 percent of young men in the US.
In 2009, 57.2% of Bachelor Degrees, 60.4% of Masters Degrees, and 52.3% of Doctoral Degrees awarded in the US were awarded to women.”
If an enemy had deliberately sought to cripple our economy they could not have come up with a better plan than Becker’s blunder. Fortunately, women ignored Becker. Sadly, economists failed to show similar wisdom. Economists do not simply reinforce failure – they declare victory in their “imperial” ambitions and give Nobel prizes to generals who led them to their worst failures at the moment that the catastrophic nature of their failures becomes most glaring.
There is No Discrimination: The “Victims” are Inferior
Alternatively, neoclassical economists assume that there is no discrimination. Employers do not hire particular people from disfavored groups because such groups are in fact inferior. I will dignify this with a refutation beyond that presented above in which the markets have refuted Becker’s claim that women were inferior paid workers due to “biolog[ical]” defects that Becker could not identify.
Markets, Despite their Manifold Defects, Can be Hostile to Discrimination
If we back off neoclassical economists’ extreme claims about markets invariably destroying bigotry, we are left with an important result. Markets do have some anti-discriminatory tendencies. CEOs are special. They are the face and leader of firms. Consider two hypotheticals that will help illustrate the point. What if Eich (remember Eich, this is an article about him) had given $1000 to a group favoring a proposition banning interracial marriage? Eich has not explained his opposition to marriage equality. The most common basis for such opposition is religious. As the Virginia trial court that convicted the Lovings of the “crime” of interracial marriage declared, the basis for opposing such marriages was also the Christian religion. Does anyone think that Eich would have been considered for the job as Mozilla’s CEO if he was on record as opposing interracial marriage? Opposition to interracial marriage was once the overwhelming norm in the U.S. – it is now something that polite society does not tolerate in any CEO.
The second hypothetical asks whether Eich would have been considered for the position of Mozilla’s CEO if he had supported a proposition that would have barred a much larger group from being able to marry, such as those in which one or more of the parties was either infertile plus all same sex couples. I’ll assume for purposes of discussion that this would bar 15% of California couples from marrying. Does anyone think that Eich would have considered for the job as Mozilla’s CEO if he was on record as opposing marriage for this 15% of the population?
These hypotheticals demonstrate why the issue for the firm’s choice of CEO has nothing do with “purging” or “first amendment” rights (sic). It is simply terrible business to pick a CEO who outrages a material number of the firm’s customers. It’s particularly bad business in high tech firms based in Northern California. I have written several columns explaining why states that discriminate against gays make will become a nightmare for large firms that operate in multiple states. Think what would have happened if those state bills allowing firms to refuse service to gays had gone into effect in Kansas and Arizona. The firm invites the CEO of a prospective client to a fancy restaurant to pitch her for business. The restaurant owner decides that she must be a lesbian and asks her to leave the restaurant. This would have to happen exactly once before firms went ballistic and demanded an immediate repeal of the law immunizing firms from liability for such denials of service to those they assumed to be gay.
Restaurant owners who wanted to refuse service to gays would, as a practical matter, have to post signs on their doors and web sites announcing their intent to discriminate. Imagine the scene of a restaurant that lacks such a sign telling a patron who has sat down for a meal to leave because they think the patron is gay? Do you think he will leave quietly? How many patrons will cheer on a restaurant that causes a screaming match when they explain, or refuse to explain, how they decided that (1) the patron was gay and (2) they would refuse to serve him? Then think of the special joys – the weeks of media coverage – when a patron they denied service because they thought he was gay turns out be straight. How many restaurants do you think would find that a good business practice? Making clear that you despise (sorry, “love” – but refuse to serve) potential customers is a poor business practice.
How many non-gays will refuse to enter a restaurant that has a sign announcing that it discriminates against gays? How will customers and employees react to firms who choose to hold business meetings in restaurants that post discriminatory signs? Now add weapons and alcohol to this joyful mix, because the states most likely to pass laws encouraging discrimination against gays are also the most likely to allow weapons in bars and restaurants.
Here’s a third thought exercise (a variant on the second). What if gays represented 30% of the population? Does anyone think that Eich would have been considered for the job as Mozilla’s CEO? The only reason that there is an issue at all about Eich is that gays are such a small percentage of the population. What the criticism of Eich demonstrates is that one need not be gay to object to discrimination against gays. Mozilla just discovered that a very large number of Americans are appalled by those who discriminate against gays. Similarly, the percentage of the population that is in families that are interracial is still fairly small, but the number of Americans who are appalled by those who would like to discriminate against interracial marriage is vast.
The interesting question is why a majority of Mozilla’s board of directors did not understand this reality – even when some of their colleagues warned against appointing as CEO a person who was an affront to such a large number of Mozilla’s customers. To understand market reactions to discrimination it is essential to understand the social forces that often drive markets. Neoclassical economists, of course, often ignore social forces, which is one of the reasons they are such bad economists when it comes to understanding discrimination. Mozilla’s board members have no such excuse for ignoring a fundamental rule of business – don’t gratuitously piss off your potential customers.