Results for “statistical discrimination” 24 found
Credit Scores, Criminal Background Checks and Hiding the Bad Apples
A number of people (Slacktivist, Kevin Drum, Matt Yglesias, Megan Mcardle) are debating the use of credit scores in employment. Credit scores are useful at predicting all kinds of things including, for example, car accidents so there is good reason to believe that they are useful in employment. The above commentators tend to focus on the potential for scores to hurt the poor but that is not obvious. Consider a similar issue: Should employers be allowed to use background criminal checks when hiring?
One argument against is that black men are more likely to have criminal backgrounds and thus these criminal background checks discriminate against black men. Let's put aside the normative issues. What’s surprising is that under plausible circumstances criminal background checks can lead to an increase in the employment of black men. The reason is that without the background check employers face a risk that their employees are ex-cons. If employers are very averse to hiring ex-cons then they will seek to reduce this risk and one way of doing so is by not hiring any black men. As a result, a background check allows non ex-cons to distinguish themselves from the pack and to be hired. Furthermore, when background checks exist, non ex-cons know that they will not face statistical discrimination and thus have an increased incentive to invest in skills.
Consistent with this reasoning, although not demonstrative of the net effect, Holzer, Raphael and Stoll find that:
…employers who check criminal backgrounds are more likely to hire
African American workers, especially men. This effect is stronger among
those employers who report an aversion to hiring those with criminal
records than among those who do not.
My view is actually that criminals face too many post-crime impediments to reintegrating themselves within the workforce. Private incentives not to take a risk on an ex-con do not cohere with social incentives to reintegrate workers into society and thus we get too little hiring of ex-cons. As a result, ex-cons face a low opportunity cost of recidivism.
Nevertheless, banning criminal background checks or credit scores is probably not the best way to combat these types of problem. Banning criminal background checks increases the incentive to rely on less accurate statistical discrimination which discriminates against the innocent and reduces the incentive to invest in skills. In short, hiding the bad apples among the good comes at the expense of the good.
Edmund Phelps — Today’s Nobel Prize in economics
Edmund Phelps. Here is the announcement from Sweden.
Here is his autobiography. He was born in Chicago in 1933 and now teaches at Columbia. Here is his CV, and here is another version. Here are recent papers. His Wikipedia entry is a short stub, but watch it grow.
Here is his summary of his research. Here is another good summary of his work. This summary, from Sweden, is the best and most comprehensive, albeit more technical.
His main contribution is a better understanding of the Phillips curve and the dynamics of short-run unemployment and the concept of the natural rate of unemployment. He gave the Phillips curve microfoundations and developed the "expectations-augmented Phillips curve." As the name suggests, the level of inflationary expectations matter for how money will influence output.
Here is his memoir on developing the idea of the natural rate of unemployment. His most influential 1960s work suggested that economies possess a natural rate of unemployment, monetary policy can reduce unemployment only temporarily (NB: in his view this is a conclusion, and should not be an axiom in economic models), monetary policy can reduce unemployment temporarily, and Keynesian economics should not treat the rate of unemployment as arbitrarily at the whim of monetary and fiscal policy. He was also concerned with how the natural rate of employment can change over time; here is his 1997 paper on that topic.
The evolution of Phelps’s thought on how money can matter is complex. His later work stresses monetary non-neutrality, mostly through non-rational expectations and non-synchronized wage and price setting. His work in the 1980s focused on what the concept of rational expectations means in such complex environments.
Do not assume that early Phelps and late Phelps are saying the same things or arguing against the same opponents. Sometimes it is argued that he redefined macroeconomics twice. After criticizing Keynesianism, he later turned against the "rational expectations" point of view. He is a complex thinker, although it can be hard to divine his "bottom line." He fails to fit inside the "macroeconomics boxes" that have developed since the early 1980s, namely real business cycle theory vs. neo-Keynesianism.
Phelps’s work was considered revolutionary in the 1960s, though the subsequent work and influence of Milton Friedman have brought related ideas into the mainstream some time ago.
He also has done work on economic justice and how a Rawlsian maximin analysis might modify the idea of a zero rate of marginal taxation on top earners, as had been suggested by James Mirrlees. Phelps believes that considerations of justice and distribution are important, and neglected, in economic thinking. Once he had a piece in the Journal of Philosophy on ideas of justice in public finance.
He also wrote some well-known papers on what intergenerational justice means, the optimal accumulation of capital, and whether those allocations will prove sustainable and consistent over time. He asks what kind of principles should govern how much capital we should leave for the next generation. His 1961 work on capital theory formulated the notion of a "golden rule" of capital accumulation. It asked what savings rate would maximize per capita income on an ongoing basis. The concepts behind this work remain important for work on capital accumulation and also the sustainability of natural resource use and environmental policy. Phelps also generated the counterintuitive result that the savings rate can be too high, and that all generations could be better off with a lower savings rate. He does not, however, seem to think that this latter idea is policy-relevant. The best summary of this work on capital theory is here, scroll through a bit.
Lately he has been working on the possibility of subsidies for hiring low-wage labor and Eastern European transitions. Here is his book on wage subsidies. Here is a more popular Phelps piece on wage subsidies. He has also done work on the structural dynamics of economies and the underlying factors behind economic innovation. Here is an early Phelps paper on technological diffusion; surprisingly it is his most frequently cited work according to scholar.google.com. He looked to education and population size as key factors driving the rate of economic growth; this piece is a precursor of later work on endogenous growth theory.
Phelps also wrote a 1972 paper on statistical discrimination, one of the earliest formal economic treatments of that topic.
Here is Phelps on Project Syndicate, the link offers numerous essays on current events. The European malaise stems from lack of dynamism. He opposed the Bush tax cuts. Here is Phelps on the rise of the West and the need for humane capitalism. He has a broadly classical liberal slant but has adopted the modern liberal idea that distribution requires government intervention into labor markets and other parts of a modern economy. He has a strong concern with the moral foundations of a free society.
Here are his cites on scholar.google.com. 4600 is a relatively low number for a Nobel Laureate. Vernon Smith for instance has over 40,000. In part this relatively small number reflects the older nature of Phelps’s major contributions, and that often his ideas have been absorbed but without citation. Furthermore Phelps does not always write within the context of the most contemporary debates.
Over the last twenty years Phelps has spent a great deal of time in Europe. In general his European influence and reputation is stronger than in the United States.
My take: It is hard to argue with this pick. It is a good selection. His 1960s macro work was true, important, and extremely influential. The capital theory work endures and provides a foundation for subsequent theory. The overall scope is impressive, and Phelps’s concerns never strayed far from the real world.
But Phelps is not an economist who has influenced my own thinking much if at all. His major contributions were absorbed, and were standard fare, by the time I was a young’un. For instance I drunk the same macro milk through the writings of Milton Friedman. I find him to be a murky writer, and often he is frustrating to read and hard to pin down. His advocates would characterize him as a "rich" thinker.
What this Prize means: The big questions still matter. Unemployment, economic growth, labor markets, capital accumulation, fairness, discrimination, and justice across the generations are indeed worthy of economic attention. Phelps contributed to all of those areas. Normative questions matter. Relevance and breadth triumph over narrow technical skill.
Addendum: The U.S. has now won six Nobel Prizes in a row, but I bet we don’t get the Peace Prize this year.
Econ Journal Watch — new issue
In this issue:Hospitals, communication, and dispute resolution: Florence R. LeCraw, Daniel Montanera, and Thomas A. Mroz criticize the statistical methods of a 2018 article in Health Affairs, and tell of their effort to get their criticisms into Health Affairs.Health Insurance Mandates and the Marriage of Young Adults: Aaron Gamino comments on the statistical modeling in a 2022 Journal of Human Resources article, whose authors Scott Barkowski and Joanne Song McLaughlin reply.Origins of the Opioid Crisis Reexamined: A 2022 article in the Quarterly Journal of Economics on the origins of the opioid crisis assigns considerable explanatory weight to the introduction and promotion of OxyContin. Robert Kaestner looks at the empirics behind the conclusion and suggests that it is without much foundation.Temperature and Economic Growth: As he did in the previous issue of this journal, David Barker investigates a piece of Federal Reserve research purporting to show that high temperatures decrease the rate of economic growth. Barker looks under the hood, replicates, and reports.Classical Liberalism in Romania, Past and Present: Radu Nechita and Vlad Tarko narrate the classical liberal movements in Romania, from the beginning of the 19th century, through the awful times of the 20th century, and down to today. The article extends the series on Classical Liberalism in Econ, by Country.Edward Westermarck’s Lectures on Adam Smith, delivered in 1914 at the University of Helsinki. Westermarck, of Finland, was an influential sociologist, anthropologist, and philosopher. His lectures are remarkably attentive toward Smith’s Theory of Moral Sentiments. The lectures are translated and introduced by Otto Pipatti.French economic liberalism versus occupational privilege: In 1753, Vincent Gournay wrote a memorial blasting the exclusionary privileges conferred upon guilds. The Chamber of Commerce of Lyon replied, and Gournay then responded with another memorial. The three-part exchange is translated here for the first time, and introduced by Benoît Malbranque.Professor McCloskey’s 1988 Letter Responding to a Letter from the President of Penn State: In 1988, Donald (now Deirdre) McCloskey received a letter about a passage in The Applied Theory of Price in an exercise on discrimination in labor markets. The letter and McCloskey’s response are reproduced here.EJW Audio:
- Sheilagh Ogilvie on 900 Years of European Guilds
- Art Carden on William H. Hutt
- David Barker on Temperature and Economic Growth
EJW books from CL Press:
Gender Differences in Peer Recognition by Economists
Card et al. study the selection of fellows to the prestigious Econometrics Society showing essentially that prior to about 1980 there was modest discrimination against women. Between 1980 and 2005 about equal access but since 2005 a large bias towards women. Not surprising but citation metrics give us a way of comparing selection with achievement.
The key result can be seen in the raw data–compare the green line of at least 3 top-5s with the red line of selection as an ES fellow.
Here is the abstract to the paper with more details.
We study the selection of Fellows of the Econometric Society, using a new data set of publications and citations for over 40,000 actively publishing economists since the early 1900s. Conditional on achievement, we document a large negative gap in the probability that women were selected as Fellows in the 1933-1979 period. This gap became positive (though not statistically significant) from 1980 to 2010, and in the past decade has become large and highly significant, with over a 100% increase in the probability of selection for female authors relative to males with similar publications and citations. The positive boost affects highly qualified female candidates (in the top 10% of authors) with no effect for the bottom 90%. Using nomination data for the past 30 years, we find a key proximate role for the Society’s Nominating Committee in this shift. Since 2012 the Committee has had an explicit mandate to nominate highly qualified women, and its nominees enjoy above-average election success (controlling for achievement). Looking beyond gender, we document similar shifts in the premium for geographic diversity: in the mid-2000s, both the Fellows and the Nominating Committee became significantly more likely to nominate and elect candidates from outside the US. Finally, we examine gender gaps in several other major awards for US economists. We show that the gaps in the probability of selection of new fellows of the American Academy of Arts and Sciences and the National Academy of Sciences closely parallel those of the Econometric Society, with historically negative penalties for women turning to positive premiums in recent years.
Was the New Deal Racist?
In recent decades, “racial disparity” has become the central framework for discussing inequities affecting African Americans in the United States. In this usage, disparity refers to the disproportionate statistical representation of some categorically defined populations on average in the distribution of undesirable things—unemployment, low wages, infant mortality, poor education, incarceration, etc. And by corollary logic, such social groupings are also found to be statistically underrepresented in desirable things—wealth, income, educational attainment, etc.
…Identifying disparate treatment or outcomes that correlate with racial difference can be a critical step in validating a complaint. However, the inclination to fixate on such disparities as the only objectionable form of inequality can create perverse political incentives. We devote a great deal of rhetorical and analytic energy to the project of determining just which groups, or population categories, suffer or have suffered the worst. Cynics have sometimes referred to this brand of what we might term political one-downsmanship as the “oppression Olympics”—a contest in which groups that have attained or are vying for legal protection effectively compete for the moral or cultural authority that comes with the designation of most victimized.
Even short of that cynical view, a central focus on group-level disparities can lead to mistaken diagnoses of the sources and character of the manifest inequalities it identifies.
Good art by women is cheaper
A rose painted by another name would cost more. In a new paper*, four academics show that art made by women sells for lower prices at auction than men’s, and suggest that this discount has nothing to do with talent or thematic choices. It is solely because the artists are female.
The authors used a sample of 1.9m transactions in art auctions across 49 countries in the period from 1970 to 2016. They found that art made by women sold at an average discount of 42% compared with works by men. However, auction prices can be distorted by a few famous artists whose output is perceived as extremely valuable. If transactions above $1m are excluded, then the discount falls to 19%.
…the researchers used a computer programme to generate paintings and randomly assign the results to artists with male or female names. They then asked participants to rate the paintings and ascribe a value. The experiment found that affluent individuals (those most likely to bid at auctions) attributed a lower value to works which the programme assigned to a woman. Clearly, this gap was unrelated to the artistic merit of the picture.
I don’t quite think that shows (non-statistical) discrimination, perhaps more convincing is this:
The average discount applied to the work of a given female artist was lowest in countries where women were more equal. (There are some exceptions to the rule, such as Brazil, where women’s art was highly rated.)
The good news is that the female discount has fallen over time. For transactions under $1m, the study calculated, the discount has dropped from 33% in the 1970s to 8% after 2010.
If you not wealthy and wish to collect art, buy textiles, they are much cheaper and very often the creators are women rather than men. They are not as a whole less aesthetically valuable than paintings, except perhaps for paintings at the very very highest levels. But within painting, prices for Gwen John vs. Augustus John have been in parity for some while now, same with Frida Kahlo and her male contemporaries, or Natalia Goncharova vs. her peers, the latter I check on a regular basis in fact I was just perusing them today.
Here is The Economist source piece. Here is the original research., “Is gender in the eye of the beholder? Identifying cultural attitudes with art auction prices”, by Renée Adams, Roman Kräussl, Marco Navone and Patrick Verwijmeren.
Google Bans Bail Bond Ads, Invites Regulation
Google: Today, we’re announcing a new policy to prohibit ads that promote bail bond services from our platforms. Studies show that for-profit bail bond providers make most of their revenue from communities of color and low income neighborhoods when they are at their most vulnerable, including through opaque financing offers that can keep people in debt for months or years.
Google’s decision to ban ads from bail bond providers is deeply disturbing and wrongheaded. Bail bonds are a legal service. Indeed, they are a necessary service for the legal system to function. It’s not surprising that bail bonds are used in communities of color and low income neighborhoods because it is in those neighborhoods that people most need to raise bail. We need not debate whether that is due to greater rates of crime or greater discrimination or both. Whatever the cause, preventing advertising doesn’t reduce the need to pay bail it simply makes it harder to find a lender. Restrictions on advertising in the bail industry, as elsewhere, are also likely to reduce competition and raise prices. Both of these effects mean that more people will find themselves in jail for longer.
As with any industry, there are bad players in the bail bond industry but in my experience the large majority of providers go well beyond lending money to providing much needed services to help people navigate the complex, confusing and intimidating legal system. Sociologist Joshua Page worked as a bail agent:
In the course of my research, I learned that agents routinely offer various forms of assistance for low-income customers, primarily poor people of color. It’s very difficult for those with limited resources to get information, much less support, from overburdened jails, courts, or related institutions. Lacking attentive private attorneys, therefore, desperate defendants and their friends and families turn to bail companies to help them understand and navigate the opaque, confusing legal processes.
…In fact, even when people have gone through it before, the pretrial process can be murky and intimidating….[A]long with walking clients through the legal process, agents explain the differences between public and private attorneys and the relative merits of each. Discussions regularly turn to the defendant’s case: Is the alleged victim pressing charges? Will the case move forward if he or she does not? When is the next court date? If convicted, what’s the likely punishment? Any chance the charges will get dropped?
…In a classic 1975 study, sociologist Forrest Dill argued:
One of the key functions performed by attorneys in the criminal process is to direct the passage of cases through the procedural and bureaucratic mazes of the court system. For unrepresented defendants, however, the bondsman may perform the crucial institutional task of helping to negotiate court routines.
Dill’s observation still rings true: bail agents and administrative staff (at least in Rocksville) act as legal guides for defendants who do not have private attorneys—and at times they provide this help to defendants with inattentive hired counsel. They provide information about court dates and locations, check the status of warrants, contact court staff on defendants’ behalf (especially when the accused have missed court or are at risk of doing so), and, at times, drive defendants to their court dates. These activities help clients show up for court, thereby protecting the company’s investments.
The bail agents are not purely altruistic, they are in a competitive, service business and it pays to help their clients with kindness and care. When I asked one bail agent why he was so polite to his clients and their relations–even when they had jumped bail–he told me, “we rely a lot on repeat business.”
Ian Ayres and Joel Waldfogel also found that the bail bond system can (modestly) ameliorate judicial racial bias. Ayres and Waldfogel found that in New Haven in the 1990s black and Hispanic males were assigned bail amounts that were systematically higher than equally-risky whites. The bail bondpersons, however, offered lower prices to minorities–meaning equal net prices for people of equal risk–exactly what one would expect from a competitive industry.
My own research found that defendants released on commercial bail were much more likely to show up for trial than statistical doppelgangers released by other methods. Bounty hunters were also much more likely than the police to capture and bring to justice people who did jump bail. The bail bond system thus provides an important public service at no cost to the public.
In addition to being wrongheaded, Google’s decision is disturbing because it is so obviously a political decision. Google has banned legal services like bail bonding and payday lending from advertising on Google in order to curry favor with groups who have an ideological aversion to payday lending and the bail system. Google is a private company so this is their right. But every time Google acts as a lawgiver instead of an open platform it invites regulation and political control. Politicians on both sides will see that Google’s code is either a quick-step to political power without the necessity of a vote or a threat to such power. Personally, I don’t want to see greater regulation but if, for example, conservatives decide that Google doesn’t represent their values and threatens their interests, they will regulate.
Google’s decision to use its code as law is an invitation to politicization. Moreover, Google is throwing away its best defense against politicization–the promise of neutrality and openness.
Firms that Discriminate are More Likely to Go Bust
Discrimination is costly, especially in a competitive market. 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. The basic logic of employer wage discrimination was laid out by Becker in 1957. The logic implies that discrimination is costly, especially in the long-run, not that it doesn’t happen.
A nice test of the theory can be found in a paper just published in Sociological Science, Are Business Firms that Discriminate More Likely to Go Out of Business? The author, Devah Pager, is a pioneer in using field experiments to study discrimination. In 2004, she and co-authors, Bruce Western and Bart Bonikowski, ran an audit study on discrimination in New York using job applicants with similar resumes but different races and they found significant discrimination in callbacks. Now Pager has gone back to that data and asks what happened to those firms by 2010? She finds that 36% of the firms that discriminated failed but only 17% of the non-discriminatory firms failed.
The sample is small but the results are statistically significant and they continue to hold controlling for size, sales, and industry.
As Pager notes, the cause of the business failure might not be the discrimination per se but rather that firms that discriminate are hiring using non-rational, gut feelings while firms that don’t discriminate are using more systematic and rational methods of hiring.
As she concludes:
…whether because of discrimination or other associated decision making, we can more confidently conclude that the kinds of employers who discriminate are those more likely to go out of business. Discrimination may or may not be a direct cause of business failure, but it seems to be a reliable indicator of failure to come.
Racist tips?
We collected data on over 1000 taxicab rides in New Haven, CT in 2001. After controlling for a host of other variables, we find two potential racial disparities in tipping: (1) African-American cab drivers were tipped approximately one-third less than white cab drivers; and (2) African-American passengers tipped approximately one-half the amount of white passengers (African-American passengers are 3.7 times more likely than white passengers to leave no tip).
Many studies have documented seller discrimination against consumers, but this study tests and finds that consumers discriminate based on the seller’s race. African-American passengers also participated in the racial discrimination. While African-American passengers generally tipped less, they also tipped black drivers approximately one-third less than they tipped white drivers.
The finding that African-American passengers tend to tip less may not be robust to including better controls for passenger social class. But it is still possible to test for the racialized inference that cab drivers (who also could not directly observe passenger income) might make. Regressions suggest that a "rational" statistical discriminator would expect African Americans to tip 56.5% less than white passengers.
I’ve read the abstract but not yet the paper. Note the authors wish to ban tipping [NB: they call it mandatory tipping] to limit racism. Thanks to Mitch Berkson for the pointer.