Results for “statistical discrimination” 28 found
A few of you have raised objections to my recent statistical discrimination hypothesis on the grounds that, if it were true, minority group members who “made it to the top” should be the super-achievers, since they had to pass through so many screens and implicit taxes. I wrote back the following (edited) in an email:
Maybe, but I think you are assuming fixed quality of talent within each sector.
Let’s say there are two sectors. In the first, the CEO sector, women face statistical discrimination and there are multiple levels. In the second there is no statistical discrimination, let us call it women’s tennis but of course there are other examples too.
It could be that most of the talented women, those who can judge where they really will succeed best and most easily, flock to the latter sectors. In which case the winners in the CEO sector need not be so special, including in the presence of discrimination.
That also means that employers and intermediaries have no special incentives to hunt for that talent: it has run away to other, less discriminatory sectors (and lowered wages in those sectors, I might add).
By the way, here was one good comment by Willitts on that post:
…if the signal of skill is “years of experience,” then the person filtered at the lower level will always look objectively worse at the higher level filters.
You’re assuming that the higher level decision makers have the opportunity (and desire) to consider walk-on candidates. You’re also assuming that those walk-ons will have adequate means to signal their superiority to those who passed through the filters.
I might be able to be the best CEO who ever lived, but my lack of management experience would never get me in the door for an interview. If (mild) discrimination at several rungs of the ladder kept me from rising to the penultimate rung, I’d have ZERO chance of attaining the top rung, not merely a small chance
Most elite performers are impressive throughout their lives. But they can stay constantly motivated, by rising through the ranks quickly. A stat-discriminated person might not have that advantage either.
Well, with the first gatekeeper there’s really no motivation to gamble on the marginalized, because a) there’s not yet a big gap and b) doing so would give up the profit of stereotyping (see the Bayesian analysis referenced in #3 of Dan’s post).
And as you get to the subsequent gatekeepers, a larger actual ability gap starts to form because the discrimination of prior gatekeepers prevented the marginalized group from gaining valuable experience.
I will continue to ponder this problem.
Let’s say there is only a mild amount of statistical discrimination in a system. Not prejudice, just a social judgment that some groups are more likely to succeed at some tasks than others. Most people, for instance, do not expect women to reach the NBA, but I would not from that conclude they are prejudiced.
But now introduce a further assumption. There are multiple layers of evaluation, and at each layer people, and institutions, wish to be seen as successful talent spotters, mentors, and coaches. High schools wish to promote students who will get into good colleges. Colleges wish to invest in students who will get into best grad schools, or get the best jobs. Firms wish to hire workers who will rise to CEO, even if elsewhere. And so on. Let’s say there are ten levels to this “game.”
Each level will apply its own “statistical discrimination” tax, whether intentionally or not. Say for instance there is (mild) statistical discrimination against women at the CEO level. Firms that wish to hire and promote future CEOs will be less likely to seek out women to hire, including at lower levels. This may or may not be conscious bias; for instance the firms may decide to look for certain personality traits that, for whatever reason, are harder to find in women. They’ll simply be making decisions that give them plaudits as good talent spotters.
Colleges will then consider similar factors in their decisions. And so will high schools. And so on. In equilibrium, all ten levels of the game will levy a partial “statistical discrimination tax,” with or without conscious bias in thee discriminatory direction.
Does this sound familiar? It is a bit like the double/multiple marginalization dilemma in microeconomics. The number of discrimination taxes multiplies, at each level. Just like the medieval barons put too many tolls on the river. All of a sudden the initially mild statistical discrimination isn’t so mild any more, due to it being applied at so many veto-relevant levels. (As you will recall from the double marginalization problem, each supplier does not take into account the effect of his/her mark-up or tax on the gains from trade elsewhere in the system.)
So say the “Bayesian rational” level of statistical discrimination is a five percent discount. You can get far more than that as the actual effective tax on the disadvantaged group, with everyone in the system behaving in a self-interested manner.
And of course these taxes will discourage effort from the disadvantaged groups, to the detriment of efficiency and also justice.
I am indebted to Anecdotal for a useful query related to this discussion.
Asaf Zussman has a 2013 Economic Journal paper on this topic (pdf, gated), here is the abstract:
Using a combination of randomised field experiments, follow-up telephone surveys and other data collection efforts, this article studies the extent and the sources of ethnic discrimination in the Israeli online market for used cars. We find robust evidence of discrimination against Arab buyers and sellers which, the analysis suggests, is motivated by “statistical” rather than “taste” considerations. We additionall find the Arab sellers manipulate their identity in the market by leaving the name field in their advertisements blank.
That abstract could be more informative, here are some concrete results from the paper, noting that market participants do not wish to start a transaction which will then later end up cancelled:
1. Both questionnaire answers and market behavior show discrimination towards Arabs.
2. “The overall [seller] response rate to emails is 22% higher for the Jewish than for the Arab buyer.”
3. An Arab buyer offering a car’s posted price receives the same amount of response as a Jewish buyer requesting a 5-10% discount off the posted price.
4. Based on questionnaires, unfavorable attitudes towards Arabs are positively correlated with Jewish religiosity and negatively correlated with education.
5. Jewish questionnaire responses are correlated with actual marketplace discrimination against Arab transactors. This concordance of words and action is by no means always the case in other studies of discrimination.
6. Arab buyers discriminate against sellers from their own ethnic group, although not as much as Jews discriminate against Arab sellers.
7. The share of used car advertisements without seller name is 10.8% for Jewish sellers, 29.5% for Arab sellers, and 16.7% for sellers with “shared” names.
…[Tim] heavily emphasizes a few experiments showing that statistical
discrimination could be a "self-fulfilling prophesy." For example, he
describes a resume experiment where otherwise identical fake resumes
with "black names" were less likely to get a response. "High-quality
applicants were more likely to be invited for an interview, but only if
they were white. Employers didn’t seem to notice whether black
applicants had extra skills or experience." If that is how employers
treat black applicants, what’s the point of trying? As Tim asks, "Why
bother to get a degree or work experience if you are young, gifted, and
But is it really true that the market fails to reward blacks for
getting more education? Is it even true that the market rewards them
less? I tested these claims using one of the world’s best labor data
sets, the NLSY. The results directly contradict Tim’s self-fulfilling prophesy story. Blacks actually get a substantially larger
return to education than non-blacks! The same goes for experience,
though the result is not statistically significant. The real lesson of
the data is that if you are young, gifted, and black, you should get a
ton of education, because it has an exceptionally large pay-off.
Why would this be so? I’m not sure, but one simple story is that counter-stereotypical
behavior stands out. When my sons were young, my wife was working a
lot, so I often took my kids places on my own. Funny thing: Time and
again, strangers came up and said, "Wow, you’re such a great dad!" But
there were moms of young kids doing the same thing in plain sight, and
the strangers rarely praised them. Why not? Because a dad taking care of two babies is counter-stereotypical, which grabs people’s attention.
Purely anecdotal, yes. But it is consistent with the small academic
literature on counter-stereotypical behavior. If you clearly violate
expectations, people not only notice; they often over-react.
The upshot is that stereotypes may actually be self-reversing
rather than self-fulfilling. The marginal payoff of distinguishing
yourself from the pack is high if people think poorly of the typical
member of the pack.
Bryan has much more on the unpleasant truths about discrimination. Read the whole thing.
We use data from Airbnb to identify the mechanisms underlying discrimination against ethnic minority hosts. Within the same neighborhood, hosts from minority groups charge 3.2 percent less for comparable listings. Since ratings provide guests with increasingly rich information about a listing’s quality, we can measure the contribution of statistical discrimination, building upon Altonji and Pierret (2001). We find that statistical discrimination can account for the whole ethnic price gap: ethnic gaps would disappear if all unobservables were revealed. Also, three-quarters (2.5 points) of the initial ethnic gap can be attributed to inaccurate beliefs of potential guests about hosts’ average group quality.
That is from a newly published paper (AEA) by Morgane Laouénan and Roland Rathelot.
From Cui, Li and Zhang:
We conduct four randomized field experiments among 1,801 hosts on Airbnb by creating fictitious guest accounts and sending accommodation requests to them. We find that requests from guests with African American-sounding names are 19.2 percentage points less likely to be accepted than those with white-sounding names. However, a positive review posted on a guest’s page significantly reduces discrimination: When guest accounts receive a positive review, the acceptance rates of guest accounts with white-sounding and African American-sounding names are statistically indistinguishable.
In other words, taste based discrimination is weak but statistical discrimination is common. Statistical discrimination happens when legitimate demands for trust are frustrated by too little information. Statistical discrimination is a second-best solution to a problem of trust that both owners/sellers/employers and renters/buyers/workers want to solve. Unfortunately, many people try to solve statistical discrimination problems as if they were problems of invidious prejudice.
If you think the problem is invidious prejudice, it’s natural to try to punish and prevent with penalties and bans. Information bans and penalties, however, often have negative and unintended consequences. Airbnb, for example, chose to hide guest photos until after the booking. But this doesn’t address the real demands of owners for trust. As a result, owners may start to discriminate based on other cues such as names. Instead market designers and regulators should approach issues of discrimination by looking for ways to increase mutually profitable exchanges. From this perspective, providing more information is often the better approach. As Cui, Li, and Zhang write in a HBR op-ed:
Our recommendation is for the platform companies to build a credible, easy-to-use online reputation and communication system. Bringing information to light, rather than trying to hide it from users, is more likely to be a successful approach to tackling discrimination in the sharing economy.
Addendum: See also Tyler and I in The End of Asymmetric Information. We need to work with information abundance rather than try to push against the tide.
Asians in America faced heavy discrimination and animus in the early twentieth century. Yet, after institutional restrictions were lifted in the late 1940s, Asian incomes quickly converged to white incomes. Why? In the politically incorrect paper of the year (ungated) Nathaniel Hilger argues that convergence was due to market forces subverting discrimination. First, a reminder about the history and strength of discrimination against Asians:
Foreign-born Asians were barred from naturalization by the Naturalization Act of 1790. This Act excluded Asians from citizenship and voting except by birth, and created the important new legal category of “aliens ineligible for citizenship”…Asians experienced mob violence including lynchings and over 200 “roundups” from 1849-1906 (Pfaelzer, 2008), and hostility from anti-Asian clubs much like the Ku Klux Klan (e.g., the Asiatic Exclusion League, Chinese Exclusion League, Workingmen’s Party of CA), to an extent that does not appear to have any counterpart for blacks in CA history. Both Asians and blacks in CA could not testify against a white witness in court from 1853-73 (People v. Hall, 1853, see McClain, 1984), limiting Asians’ legal defense against white aggression. The Chinese Exclusion Act of 1882 and the “Gentlemen’s Agreement” in 1907 barred further immigration of all “laborers” from China and Japan.
…Asians have also faced intense economic discrimination. Many cities and states levied discriminatory taxes and fees on Asians (1852 Foreign Miner’s Tax, 1852 Commutation Tax, 1860 Fishing License, 1862 Police Tax, 1870 “queue” ordinance, 1870 sidewalk ordinance, and many others). Many professional schools and associations in CA excluded Asians (e.g., State Bar of CA), as did most labor unions (e.g., Knights of Labor, American Federation of Labor), and many employers declined to hire Asians well into the 20th century (e.g., Mears, 1928, p. 194-204). From 1913-23, virtually all western states passed increasingly strict Alien Land Acts that prohibited foreign-born Asians from owning land or leasing land for extended periods. Asians also faced laws against marriage to whites (1905 amendment to Section 60 of the CA Civil Code) and U.S. citizens (Expatriation Act 1907, Cable Act 1922). From 1942-46, the US forcibly relocated over 100,000 mainland Japanese Americans (unlike other Axis nationalities, e.g. German or Italian Americans) to military detention camps, in practice destroying a large share of Japanese American wealth. In contrast, blacks in CA were eligible for citizenship and suffrage, were officially (though often not de facto) included in CA professional associations and labor unions that excluded Asians, were not covered by the Alien Land Acts, and were not confined or expropriated during WWII.
Despite this intense discrimination, Asian (primarily Japanese and Chinese) incomes converged to white incomes as early as 1960 and certainly by 1980. One argument is that Asians invested so heavily in education that convergence has been overstated but Hilger shows that convergence occurred conditional on education. Similarly, convergence was not a matter of immigration or changing demographics. Instead, Hilger argues that once institutional discrimination was eased in the 1940s, market forces enforced convergence. As I wrote earlier, profit maximization subverts discrimination by employers:
If the wages of X-type workers are 25% lower than those of Y-type workers, for example, then a greedy capitalist can increase profits by hiring more X workers. If Y workers cost $15 per hour and X workers cost $11.25 per hour then a firm with 100 workers could make an extra $750,000 a year. In fact, a greedy capitalist could earn more than this by pricing just below the discriminating firms, taking over the market, and driving the discriminating firms under.
If that theory is true, however, then why haven’t black incomes converged? And here is where the paper gets into the politically incorrect:
Modern empirical work has indicated that cognitive test scores—interpreted as measures of productivity not captured by educational attainment—can account for a large share of black-white wage and earnings gaps (Neal and Johnson, 1996; Johnson and Neal, 1998; Fryer, 2010; Carruthers and Wanamaker, 2016). This literature documents large black-white test score gaps that emerge early in childhood (Fryer and Levitt, 2013), persist into adulthood, and appear to reflect genuine skills related to labor market productivity rather than racial bias in the testing instrument (Neal and Johnson, 1996). While these modern score gaps have not been fully accounted for by measured background characteristics (Neal, 2006; Fryer and Levitt, 2006; Fryer, 2010), they likely relate to suppressed black skill acquisition during slavery and subsequent educational discrimination against blacks spanning multiple generations (Margo, 2016).
…A basic requirement of this hypothesis is that Asians in 1940 possessed greater skills than blacks, conditional on education. In fact, previous research on Japanese Americans in CA support this theory. Evidence from a variety of cognitive tests given to students in CA in the early 20th century suggest test score parity of Japanese Americans with local whites after accounting for linguistic and cultural discrepancies, and superiority of Japanese Americans in academic performance in grades 7-12 (Ichihashi, 1932; Bell, 1935).
Hilger supplements these earlier findings with a small dataset from the Army General Classification Test:
…these groups’ cognitive test performance can be studied using AGCT scores in WWII enlistment records from 1943. Remarkably, these data are large enough to compare Chinese, blacks and whites living in CA for these earlier cohorts. In addition, this sample contains enough young men past their early 20s to compare test scores conditional on final educational attainment, which can help to shed light on mechanisms underlying the conditional earnings gap documented above.
Figure XII plots the distribution of normalized test score residuals by race from an OLS regression of test z-scores on dummies for education and age. Chinese Americans and whites have strikingly similar conditional skill distributions, while the black skill distribution lags behind by nearly a full standard deviation. Table VIII shows that this pattern holds separately within broad educational categories. These high test scores of Chinese Americans provide strong evidence that the AGCT was not hopelessly biased against non-whites, as Neal and Johnson (1996) also find for the AFQT (the successor to the AGCT) in more recent cohorts.
From Hilger’s conclusion:
Using a large and broadly representative sample of WWII enlistee test scores from 1943 both on their own and matched to the 1940 census, I document the striking fact that these test scores can account for a large share of the black, but not Asian, conditional earnings gap in 1940. This result suggests that Asians earnings gaps in 1940 stemmed primarily from taste-based or some other non-statistical discrimination, in sharp contrast with the black earnings gap which largely reflected statistical discrimination based on skill gaps inherited from centuries of slavery and educational exclusion. The rapid divergence of conditional earnings between CA-born Asians and blacks after 1940—once CA abandoned its most severe discriminatory laws and practices—provides the first direct empirical evidence in support of the hypothesis of Arrow (1972) and others that competitive labor markets tend to eliminate earnings gaps based purely on taste-based but not statistical discrimination.
Hilger’s other research is here.
That is a new paper by Camelia Simoiu, Sam Corbett-Davies, and Sharad Goel, the abstract is familiar but depressing:
In the course of conducting traffic stops, officers have discretion to search motorists for drugs, weapons, and other contraband. There is concern that these search decisions are prone to racial bias, but it has proven difficult to rigorously assess claims of discrimination. Here we develop a new statistical method—the threshold test—to test for racial discrimination in motor vehicle searches. We use geographic variation in stop outcomes to infer the effective race-specific standards of evidence that officers apply when deciding whom to search, an approach we formalize with a hierarchical Bayesian latent variable model. This technique mitigates the problems of omitted variables and infra-marginality associated with benchmark and outcome tests for discrimination. On a dataset of 4.5 million police stops in North Carolina, we find that the standard for searching black and Hispanic drivers is considerably lower than the standard for searching white and Asian drivers, a pattern that holds consistently across the 100 largest police departments in the state.
For the pointer I thank the excellent Samir Varma.
Devah Pager’s article in the latest American Journal of Sociology demonstrates an important relationship between race, criminal record and employment. She sent out pairs of black and white young men to apply for entry level jobs, gave them similar records except that one was randomly selected to have a criminal background. She then analyzed who was called back for an interview and got some interesting results:
1. Unsurprisingly, for both blacks and whites, reporting a criminal record drastically reduced the chances of a call back.
2. Black men *without* the criminal history were less likely to be called back than white men *with* criminal records.
3. Having a criminal record is more damaging for black applicants than for white applicants.
This, I think, is a nice challenge to the whole statistical discrimination thesis, where employers use race as a proxy for other unmeasured variables. The Pager study shows that even when employers have full information on their applicants, they often prefer a white ex-convict than a similar black man without a criminal record.
Update: Dmitri Masterov writes to tell me about point #2 – Pager showed that the difference between the two groups was not statistically significant.
In an NBER paper, Blair and Chung find that occupational licensing reduces labor supply significantly. I had expected that occupational licensing would be worse for blacks than for whites because it imposes an additional locus of discrimination but that effect seems to be opposed by a certification effect (the license helps black workers to overcome statistical discrimination) so the net effect is not as bad for blacks as for whites:
We exploit state variation in licensing laws to study the effect of licensing on occupational choice using a boundary discontinuity design. We find that licensing reduces equilibrium labor supply by an average of 17%-27%. The negative labor supply effects of licensing appear to be strongest for white workers and comparatively weaker for black workers.
An Institute for Justice report by Morris M. Kleiner, the dean of occupational licensing studies, and Evgeny S. Vorotnikov attemps to calculate the net loss to the US economy from occupational licensing and concludes that when all costs are considered it is on the order of $200 billion annually.
In preventing people from working in the occupations for which they are best suited, licensing misallocates people’s human capital. In forcing people to fulfill burdensome licensing requirements that do not raise quality, licensing misallocates people’s human capital, money and time. And with its promise of economic returns over and above what can be had absent licensing, licensing encourages occupational practitioners and their occupational associations to invest resources in rent-seeking instead of more productive activity. Taking these misallocated resources into account, we find potential costs to the economy that far exceed those from deadweight losses and that likely provide a more complete picture of the extent to which licensing reduces economic activity.
…we find licensing costs the American economy $197.3 billion in misallocated resources.
The original Sears mail-order catalogue changed how African Americans in the South shopped:
…the catalogue format allowed for anonymity, ensuring that black and white customers would be treated the same way.
“This gives African Americans in the Southeast some degree of autonomy, some degree of secrecy,” unofficial Sears historian Jerry Hancock told the Stuff You Missed in History Class podcast in December 2016. “Now they can buy the same thing that anybody else can buy. And all they have to do is order it from this catalogue. They don’t have to deal with racist merchants in town and those types of things.”
In a heartfelt essay Ashlee Clark Thompson explains how the “grab and go” technologies now being tested at Amazon Go made her confront lessons learned from decades of shopping while black:
The idea of walking into a store, taking an item or several off the shelves and strolling right back out again boggled my mind. It ran counter to everything I had learned about being black and shopping.
…I grabbed one of the orange Amazon Go bags and began to make my way around the perimeter of the store. I was studying the various bottled waters and debating whether to get fizzy or still, or a bottle of kombucha, when I realized what I was really doing: I was stalling. The fear I had carried with me for decades reared its head as I stood in front of the refrigerated display. I was afraid to make a choice, remove it from a shelf and put it in my bag. I was afraid someone would pop out from behind a display of Amazon-branded merch and scream, “Get your hands off that!” And I was mad that this fear couldn’t even let me fully enjoy an experience that’s designed for everyone to grab and go, no questions asked.
Eff this, I thought. I’m getting some Vitamin Water.
Once the plastic bottle hit the bottom of my reusable bag, I glanced around to see if anyone noticed. The Amazon employees shuffled around the small store and restocked shelves. Tourists chatted in small groups as they pointed and looked for the sensors that were keeping track of our every move. One guy with his phone on a selfie stick recorded himself as he selected snacks. And then there were the folks for whom the novelty had worn off and just wanted a vegetarian banh mi sandwich.
No one cared what I was doing. Is this what it feels like to shop when you’re not black?
…Amazon Go isn’t going to fix implicit bias or remove the years of conditioning under which I’ve operated. But in the Amazon Go store, everyone is just a shopper, an opportunity for the retail giant to test technology, learn about our habits and make some money. Amazon sees green, and in its own capitalist way, this cashierless concept eased my burden a little bit.
The similarities in these cases are interesting but so are the differences. In the Sears case most of the effect of diminished discrimination was driven by greater competition in one-shop towns. In the one-shop town the owners sometimes took a share of their monopoly profits in invidious racism–this appears to explain why shop owners would prevent blacks from buying more expensive products (or perhaps the one-stop shop had to cater to racist customers who demanded invidious discrimination.)
In the Uber case my bet is that a large share of the reduction in discrimination was due to the fact that Uber drivers don’t carry cash and so are less worried about robbery and the app increases safety because it records in detail rider, driver and trip data. In other words, the Uber system reduced the value of statistical discrimination. It’s difficult to know for sure, however, because there was probably also some decline in invidious discrimination brought about by Uber hiding some rider information from drivers until trips are accepted.
The last case, the Amazon Go case, is in part a decline in the value of statistical discrimination since shoplifting is no longer a problem (in theory, assuming the technology works) but in this case the decline in statistical discrimination is driven by much finer discrimination. The moment a shopper enters the Amazon Go store, Amazon knows their name, address, entire shopping history, credit history and potentially much more. Moreover, a shopper’s every movement within the store is tracked to a level of detail that no store detective could ever hope to match. To the customer, especially the black customer, it may feel like they are no longer being watched but in fact they are watched more than ever before–the costs of technological monitoring, however, are mostly fixed which means that everyone is monitored equally. No need for statistical discrimination in the panopticon.
Addendum: A good dissertation might be to incorporates the cost of information, the value of statistical discrimination and the demand for invidious discrimination in a general theory that explains the various cases mentioned here and the effects of information bans such as ban the box.
That is a 2011 AFPS paper by Sarah F Anzia and Christopher R Berry, here is the abstract:
If voters are biased against female candidates, only the most talented, hardest working female candidates will succeed in the electoral process. Furthermore, if women perceive there to be sex discrimination in the electoral process, or if they underestimate their qualifications for office, then only the most qualified, politically ambitious females will emerge as candidates. We argue that when either or both forms of sex‐based selection are present, the women who are elected to office will perform better, on average, than their male counterparts. We test this central implication of our theory by studying the relative success of men and women in delivering federal spending to their districts and in sponsoring legislation. Analyzing changes within districts over time, we find that congresswomen secure roughly 9% more spending from federal discretionary programs than congressmen. Women also sponsor and cosponsor significantly more bills than their male colleagues.
I also would consider the alternative hypothesis that the women legislators are simply more conscientious and less wrapped up in themselves. Nonetheless this result is one possible equilibrium relevant to the recent MR discussions on statistical discrimination.
For the pointer I thank Michelangelo L.
Several loyal MR readers requested I cover this topic. My views are pretty simple, namely that I am a fan of the movement. Police in this country kill, beat, arrest, fine, and confiscate the property of black people at unfair and disproportionate rates. The movement directs people’s attention to this fact, and the now-common use of cell phone video and recordings have driven the point home.
I don’t doubt that many policemen perceive they are at higher risk when dealing with young black males, and that is part of why they may act more brutally or be quicker to shoot or otherwise misbehave. I would respond that statistical discrimination, even if it is rational, does not excuse what are often crimes against innocent people. For instance, a man is far more likely to kill you than is a woman, but that fact does not excuse the shooting of an innocent man.
I also don’t see that citing “Black Lives Matter” has to denigrate the value of the life of anyone else. Rather, the use of the slogan reflects the fact that many white people have been unaware of the extra burdens that many innocent black people must carry due to their treatment at the hands of the police. The slogan is a way of informing others of this reality.
“Black Lives Matter” is a large movement, if that is the proper word for it, and you can find many objectionable statements, alliances, and political views within it. I don’t mean to endorse those, but at its essence I see this as a libertarian idea to be admired and promoted.
2. The demography of Appalachia, recommended. And does higher education propel regional population growth?
3. “The empirical findings are consistent with a model of statistical discrimination where female executives are better equipped at interpreting signals of productivity from female workers. The evidence suggests substantial costs of under-representation of women at the top of the corporate hierarchy.” Link here.
4. Lunch with the FT is now on Medium, here is Marc Andreessen.
I can think of a few candidate theories:
1. His views are the right views, more or less, and American voters recognized this.
2. A quite significant percentage of America is very directly racist. I don’t mean statistical discrimination here, I mean “downright racist.”
3. Give Ray Fair (NYT) his Nobel Prize right here and now, economic conditions truly predict election results at the national level.
4. The “third term Party fatigue” effect is stronger in national elections than we had thought.
5. Hillary Clinton is a weaker candidate than many people had thought. Maybe so, but that has to be unpacked a bit more. I would try “the Democratic national establishment doesn’t understand why much of America trusts it so little, so it keeps on doing and saying unpopular things. Those things include elevating some candidates and also encouraging them to take particular stances.”
6. As Robert D. Putnam suggested, ethnic diversity can lower the quality of governance, and this is one step along that path toward greater fractiousness. This may blend into racism, but much of it is simply “fear of being in the losing coalition.” The common claim that the electorate is more polarized than before fits into this. You might try Ezra Klein’s podcast with Arlie Hochschild.
7. America is not ready for a woman president. Or maybe it has to be a different kind of woman president, noting that Hillary, while she has passed through many filters, has not passed through the “truly popular with normal voters filter” in the same way that say Thatcher and Merkel did. And no, New York isn’t normal, sorry people.
8. The Democrats have plenty of policy proposals, but only the Republicans are running on ideas. And very often an idea beats no idea, even if the idea on the table is a bad one.
I don’t agree with #1, and while #4 sounds like a plausible part of the story to me, as a truly major explanation I find it hard to square with Obama’s continuing popularity. #3 kicks in but as a dominant force, it seems hard to elevate when median household income just grew at 5.2%, inflation is low, there is no major war, gas prices are low, and asset prices are high.
On #2, I see #5 as a more convincing statement of related ideas, while admitting #2 is a factor. How well the Democrats do in the Senate might give us some bead on the relative import of #5.
Overall I am seeing a lot of room for #5 and #6 and #7 and #8. Presumably 5, 6, and 8 are hard for many Democrats to admit, and I genuinely wonder how their thoughts run in the quiet of their homes. Some are plugging hard for an extreme version of #2, but, as long as we are considering matters of prejudice, I find the gender bias of #7 easier to swallow. We did after all just elect Obama for two terms in a row, and we have never ever had a woman president or even a serious contender before.
If, I wish to stress that word if. But that he is still in the running, and making it close, is reason enough to ponder these questions.