Economics Is The Least Influential of All Social Sciences

By Graham Peterson

Economists have a council that reports to the president.  They work at our Federal Reserve Banks.  They’re more often hired by industry.  And they get greater discussion in the New York Times, as Justin Wolfers has just pointed out.

Other social sciences have inferiority complexes about the fact.  Economists have superiority complexes about the fact.  Everyone loves talking about it — economists are most influential.   They only fight about why they are and whether it’s fair and deserved.

But the discussion lacks intuition for sample selection bias.  Economics is only dominant in elite discourse — newspapers, bureau white papers —  and has had little to no influence in the street.  But what if the ideas that really shape people’s lives come from the street?

How does it respond to its lack of street-level influence? Print increasingly algebraic textbooks for kids who can’t remember high-school algebra, and congratulate itself on its sophistication.  For 200 years economists have been publishing polemics about this or that economic fallacy.  At what point do you stop and think, “hey maybe people don’t get it because we’re explaining it poorly”?

Economists don’t do that because economists are rational, because of incentives.  They have convinced a small group of patrons that they possess a very special magic, and those patrons pay them handsomely to talk to themselves.  So why bother making the ideas understandable and useful to anyone else?

On the other hand, everybody in the developed world knows what the subconscious mind is.  Most of them have a grasp of institutional racism.  Couples in trailer parks argue about their lack of communication skills and childhood traumas. Businesspeople furiously consume magazines telling them how to improve their culture.

Ask someone, even an undergraduate economics major, what a comparative advantage is. Or an average cost curve.  Exactly.

If the goal is to influence people who can afford to pay for advice they don’t understand, and give them arguments that they’re helpless to adjudicate or apply to their own lives, then economics is doing a fantastic job.  Keep the barriers to entering the conversation you’re having high, and you can prove you’re important with your salary.

The undergraduates will flood your majors to get the labor market signal whilst having no clue what you’re talking about.  The politicians will rush to hire you to give technical justification to their moral pageantry.  The Sunday morning sophisticate will read your NYT blog and think, “Gee those economists are smart.  Republicans are evil after all.  Thanks, Paul.”

But the rest of the world will be shaped, at the level of its neighborhoods and relationships and workplaces, by the people who give them tools to think and argue and persuade and influence each other.  I’m all for calculus and supply and demand diagrams, but everyone in this debate needs to think a lot harder about whether social science’s first job is to influence suits.

If you believe that society is a top down system in which elites write the ideas that everyone else just dances to (and us intellectuals, for personal reasons, just love that idea), then it matters a great deal that economists are well represented among elites.

But if you believe that society is a bottom up system, or an “emergent order” as economists used to call it, then it matters a great deal that people live and breathe models from other social sciences, that have diffused much more widely.

Advertisements

Why am I ignoring Nigeria?

By Seth Studer

I take a little exception to the smarminess of certain media’s response to the Charlie Hebdo murders. Last week, they inform us, we witnessed two horrific massacres: the murder of 12 satirists in Paris and the murder of roughly 2,000 civilians in Baga (that’s in Borno, Nigeria). But, they continue, judging from CNN, Fox, and your Facebook feed, only one of these terrible crimes got any coverage. To ask the question “which one: the 12 Europeans or the 2,000 Africans?” is to answer it. While the loss of 12 innocent lives and an implied assault on Free Speech (which doesn’t really exist per se in France) rallies millions across the Great White West, virtually no one is speaking for what Teju Cole calls “unmournable bodies” (an eloquent phrase, although the critical theorist’s habit of saying body when you mean person upsets his essay’s thesis). Cole’s essay in the New Yorker (linked above) is intelligent and passionately argued, and he handles his argument’s underlying ethos – the aforementioned smarminess – with more grace than others (the latter article incorrectly states that Nigeria is south of the equator, a reminder that the many truths revealed by postcolonial theory – e.g., global North vs. global South – do not always square with geographical reality). But in general, I felt scolded for paying more attention to France than Nigeria.

And I probably deserve a scolding. Did mainstream news outlets focus on France over Nigeria as the consequence of a bias toward white Europeans? Absolutely! Was the attack on Charlie Hebdo more frightening and noteworthy to Western audiences than the massacre in Baga because the former represents an attack on the imagined “center” of Western civilization rather than its “periphery”? You bet!

So should 2,000 murder victims be more “newsworthy” than 12 murder victims? I think it depends on the circumstances. 

Anyone who hasn’t been following Boko Haram over the past many months is an irresponsible consumer of world news. The mass violence last week represents the terrifying apex of an ongoing story. We spent much of 2014 preoccupied with the horrors inflicted upon the Nigerian people by this radical group (even Michelle Obama got involved, which got American conservative media involved, etc., etc.). The Charlie Hedbo massacre, meanwhile, fell out of a clear blue sky. Both discrimination against Muslims and Muslim unrest in France are ongoing, but nothing concrete or obvious precipitated this attack. These murders arrived on our screens demanding a context. Hence, the intense coverage.

And for me, intense coverage of the Charlie Hebdo massacre is essential not merely because it reinforces Western commitments to free speech (commitments that tend to get waylaid when they’re needed most). Coverage is essential because France is an important European nation in the grips of a major rightward political and cultural shift, one that could potentially turn more strident, more xenophobic, and more violent. After a half century on the fringes (and apparent defeat in the face of European unification), Europe’s right-wing parties (as opposed to its right-of-center parties) are, ahem, on the march. In the United States, extreme right-wing rhetoric has benefited from decades in the mainstream: a speaker’s racism or xenophobia can be carefully coded and embedded in speeches about tax policy. In Europe, the far right has been far wilder and wilier. They’ve retained their ugliness and wear it explicitly on the surface. (Whenever one of my liberal friends unfavorably compares America’s conservative politics with Europe’s socialist policies, I remind them, “Yes, you like their left wing, but you don’t want their right wing.”) Meanwhile, since the 2007/08 global banking crisis, nationalism in Europe – both right-wing and left-wing – has resurged to levels not seen in decades. Because of their knotted political and economic ties to Germany (or Russia), the peoples of Europe are seeking social and cultural distinction. Secession movements have gained renewed traction in the geographical and political expanse between Scotland and Crimea. Consequently, Germans and Russians are also asserting their national character in ways that, twenty years ago, would have seemed taboo.

This, for me, is the context of the Charlie Hebdo attack, far removed from the bloodshed in Nigeria (admittedly, all things connect in our post-post-colonial world, as African expats like Cole convincingly demonstrate). Note that the above paragraph doesn’t include the word “Islam.” I don’t think you need to dwell much on radical Islam to understand the socio-cultural dynamic that drives millions of French residents into the streets. From a French perspective, however, immigration from the Muslim world underscores every aspect of the current national identity crisis. Thus, when an event like the attack on Charlie Hebdo occurs, you get 3.7 million people in the streets and attacks on Muslims.

This, to me, is a very big story indeed.

Two thousand people died in Nigeria last week, it’s true, but 3.7 million people marched throughout France yesterday – roughly one million in Paris alone. What do those one million want? What do they represent? Many of them are doubtless sympathetic with France’s Muslim minorities. Few among them are likely to be extreme French nationalists (though more of them are sympathetic with French nationalism than Western liberals would like to imagine). Whatever their motives, this represents a good moment to take France’s cultural temperature. The context demands it. Your first response to Charlie Hebdo should be an unequivocal condemnation of the murders and support for free speech. But your second response, given the atmosphere in Europe, should be concern for liberalism in France. Because, contrary to what the news coverage is telling you, continental Europe is not historically an easy or natural home to liberal values. And because a march can be a mob by another name.

Slamming Into People On the Sidewalk Is Not Evidence of Patriarchy

By Graham Peterson

Apparently the internet has made a big hullabaloo about “manspreading” (a spinoff of the already ridiculous word “mansplaining”) that is supposed to identify pernicious misogyny — men sitting with their legs spread on the subway.  Sitting with legs crossed, forward, or spread certainly is a gendered and classed phenomenon.

I once watched a little girl’s father tell her “close your legs; you ain’t a boy,” while she sat.  I feel more comfortable full-crossing my legs when I’m in a higher status environment like a nice restaurant (note that this class reversal – where upper class men are more effeminate – seems to give the lie to hegemonic masculinity) .  Women close and cross their legs because of the tradition of skirt wearing, but skirt wearing is gendered.  There’s no obvious reason why men don’t wear skirts.

So leg crossing is gendered; this much is obvious.

But leg-crossing norms are old and evolved largely independent of train riding.  Train riding norms also are also old, and evolved mostly independent of leg crossing norms.  The CTA in Chicago has a routine announcement on trains about keeping your bags off the seat next to you.  The CTA is 100 years old.  After 100 years people still don’t pick up their bags for other passengers.  And the announcement about bags is not aimed at a gender.

It seems unlikely that after paying focused attention to the problem of rider comfort and issuing repeated warnings about courtesy for years, and after paying a not-small amount of attention to leg crossing norms for years, we’ve just now uncovered a ubiquitous example of patriarchy at their intersection.

The uncritical discussion of manspreading hasn’t stopped internet feminists from taking it even further, though.  The logic here is, “we have very little evidence for our beliefs, so the time is ripe to project them even further.”  Now the complaint is that men will not defer to women’s paths on the sidewalk.  The evidence is a “social experiment” in which an activist walked down the sidewalk and refused to change her course when men were coming her way.

Screen Shot 2015-01-10 at 12.43.56 PM

It would be an interesting experiment if it were actually an experiment.

Did she just wait for men to end up in her path at random, or did she walk into men’s paths?  What kind of men did she walk into?  What kind ended up in her path randomly?  White, black, suits, sneakers, taller, shorter?  Were they on a cellphone or distracted by a car wreck?  Did she make or avoid eye contact with people coming toward her?

If she did walk into their paths, how many paces away did she merge in?  Did that give them ample time to see her?  Did she walk into women at the same rates as men, and then compare proportions?  When she compared across groups, did she hold constant the context like the block they were on or racial composition of the neighborhood?

Did she observe other people walking toward one another in order to ascertain which patterns of sidewalk walking are general and which particular?

None of these factors got discussed or controlled for.  The experiment gets an F in an undergraduate course.  A woman who was angry about women’s (supposed) lack of body sovereignty just started walking into men on the street, called it an experiment, and five thousand people subsequently broadcast it, nodding along.

Like I said, these are interesting questions.  Leg crossing is certainly gendered, and playing chicken on the sidewalk certainly can be about dominance.  I’ve played chicken with many a young thug on the sidewalk in order to assert myself in the hood.   Our high-school principle used to walk straight down the hallway without swerving.

It might not be the case that not being able to get a seat on the subway, or able to get to work without colliding into people, is evidence of patriarchy.  But we’ll never know as long as “look at this picture of this dooshbag wearing a suit on the subway with his legs uncrossed lol retweet” and “I’m paranoid about patriarchy so I created an experiment with no controls and wow I’m surprised I confirmed my hypothesis lol retweet” is the standard of evidence.

A Response to the Best Arguments That Academic Discrimination Doesn’t Exist

By Graham Peterson

Fabio Rojas is generous and clever, and his review of the Duarte et al. paper on academic discrimination lays out some of the best arguments that academic discrimination doesn’t exist.  I’m going to take them down here, and explore my alternative in a different post.

Sociologists might be overwhelmingly liberal because we deal with lots of research that have social implications.

The causal arrow goes in the opposite direction.  Interest in disaffected populations began because of, e.g., Durkheim’s thesis that opposition to outsiders reveals the mechanics of the insiders, or e.g., Merton’s thesis that black crime revealed a problem with modernity’s rapid normative shifts.

Interest in disaffected populations was originally aimed at a generalized interest in the construction of the social.  Now we study them because we’re trying to undo oppression, or as the second to last banner of the ASA stated, “Interrogating Inequality.”

Early twentieth century sociologists were cautiously optimistic about modernity and capitalism — and politically, were justifying and trying to understand social democracy, bureaucracy, capitalism, urbanity — not radical politics of oppression starting with Rousseau, detouring at Marx, and ending with Bourdieu.

It is not the case that sociology’s theories are inherently radical and, whoops, therefore we attract radicals.

How would people possibly know the party preference of mathematicians?

Superficial tastes that otherwise seem to have no semiotic connection to political values end up strongly correlated to those values nevertheless.  Fashions, senses of humor, type of car, are easily observable in a hiring process (which drives discrimination) and in longer-run training (which drives selective homophily).

They cover a number of studies showing that liberals admit they would discriminate, while others do not, [but] my hypothesis that overly zealous academic liberals are simply more motivated to admit personal fault, which means they deviate from the socially desirable answer at much, much higher rates.

This is wishful.  The idea here is that lefties are disproportionately interested in personal growth through psychological self evaluation, and that they are always checking prejudicial biases.  But no.  Lefties are interested in checking their privilege when it comes to disaffected groups, not conservatives.

Lefties admit to discriminating against conservatives because there’s nothing embarrassing to “admit.”  Protecting themselves and their friends from dirty-pool conservatism is a proud moral duty.  This is why when we survey people on whether or not they’ve raped a woman or would discriminate against conservatives, we use the definitions of rape and discrimination in the prompt instead of calling it rape and discrimination.

Rapists do not believe rape is rape; it is a deserved reprisal and/or sex they are entitled to.  Discriminators do not believe discrimination is discrimination; it is deserved selection based on presumed population characteristics, like “black people don’t work hard,” or “conservatives are dogmatic.”

One point that is never brought up is that liberal disciplines become noticeably more conservative if they try hard enough. For example, it is my impression that economics used to be dominated by Keynesians up until about 1960 or so.

This is correct, and incorrect.

Correct, Milton Friedman was openly maligned way before he had a national stage.  Rose Friedman told Deirdre at a cocktail party, after Milton had gotten his Nobel, “I would trade all this for some respect for Milton early in his career.”  The Chicago School of economics is named after just one university, and the Virginia School of political economy the same way, because the people who got them together formed defensive coalitions in otherwise hostile climates.

But the success of conservatives and libertarians in economics is not an example of a discipline “trying hard enough.”  Economists routinely and abusively malign one another for their politics.  Right leaning economists have carved out the space they have because they got on board with a vector of ideological homogeneity that is much more important to economists — mathematical theory and statistical methods.

Liberals tend to dominate in areas that are low paying [like academics] and focus on issues like education, learning, care giving, and culture. These include the arts, entertainment, academia, K-12 teaching, nursing, social work, and science. Once you add some high income, conservatives start appearing in large numbers (e.g., the Dem/GOP is way different for doctors vs. nurses; artists vs. art managers; lawyering vs. other humanities oriented work; physical science vs. engineering).

This is a compelling argument that I have never heard.  But it lacks economics.  What Fabio often calls “helping” professions are often low skill professions for which there is a lot of competition, or which are valued really highly at the margin, but consumed in enormous amounts.  This is the water/diamond paradox – we value water more than diamonds but because we consume so much water and marginal utility diminishes, water costs less than diamonds.

The fact that helping professions pay less is not a result of the fact that we don’t value helping.  We value helping a lot, and that’s why we consume so much helping, which is why its price is low.  The idea that helping provides people a compensating differential in which they are willing to give up wages lacks empirical demonstration (but is a nice story sociologists might tell themselves to validate themselves) and is, once again, normatively biased against “non-helping” professions.

Is there any evidence other than Enron documentaries about evil capitalists that people in “non-helping” professions do not seek the same personal transcendence, and meaning that “helpers” do?  No.  Are paper towels and petroleum helpful, and do people who work in those industries experience the same camaraderie and productive influence other people do?  Yes.

Anyway, academics is only low paying in certain disciplines, and pays extremely well in others.  Chemistry graduate students at U of C make something like $40,000 a year.  We make $25,000.  UIC Economics, a department ranked about 60th, hired two economists before I left at $100,000 a piece.  And at the top of any discipline, academics pays some of the highest wages in the world.

Fabio’s points are well taken and extremely thoughtful.  Voices like his are needed in this debate.  But he is wrong.

Is Prejudice Like A Mouse?

By Graham Peterson

They say that if you see a mouse in your house, they’re everywhere.  Mice have every reason to hide from you, so if one of those little bastards is so hungry that he’s willing to scurry into your kitchen while the lights are on, it’s a good guess that the walls are crowded and food is getting scarcer.

The social justice movement uses the same logic when reasoning about prejudice and discrimination.  If you can pick up a trace of discrimination in a resume audit study, for instance, it is safe to assume that such racism or sexism exists just about everywhere else — because like the mice — people want to hide their racism and sexism.

That might be true, but I doubt it.

I don’t see a lot of people hiding their racism and sexism.  I see people keeping their racism and sexism to sympathetic company, sure.  And I see a lot of unabashed racism and sexism among, say, the police in my Chicago neighborhood, Blacks at 55th and Garfield, Irish city workers, and among some of the hicks in my home town.

Social justice advocates assume that, like themselves, one who’s had one’s prejudice exposed will be shocked and disgusted at oneself, and change. It follows that there must be a gigantic subconscious conspiracy in which people lie to themselves, and one another, hiding their prejudice.

But it could be the case that discrimination is actually hard to detect because it’s actually uncommon.

Maybe it’s the case that in some boundary-keeping settings like policing, hiring, and dating, where people have little information about one another, they use crappy observables to determine whether they’ll get along.  And maybe outside these particular situations, people really don’t think much about, or have to invoke, consciously or unconsciously, their prejudice.

Discrimination is in this view extremely difficult to detect because it is simply rare, not because people are actively (or inactively, in an embodied or subconscious or habitus-y way) hiding their prejudice.

Note that it still wouldn’t really take much discrimination to maintain a world that looks pretty segregated.  People mostly choose to stick with their own, and people who do attempt to transgress boundaries only rarely stir up prejudicial sanctions.  Like murder, most people just prefer not to murder one another, so murder is rare and so are murder trials.

If this is the way discrimination actually works, one way to make the world less segregated would be to caution people to be nicer and get more information about one another at social boundaries.  I think we do a lot of that and should keep it up.

But a particularly bad way to do it would be to claim that everyone is constantly fomenting nasty attitudes toward one another, constantly putting each other into homogenized and prejudicial categories, and constantly needing to be called out for it.

Demanding apologetic prestations from people, or screaming at them to shock them out of their “denial,” is not how we’re going to make the world a more tolerant place.

Rare but influential boundary discrimination means that we need to focus our energies on particular rituals in which people evaluate one another and have poor information, and persuade them to get better information than skin color, gender, religious affiliation, etc.

That’s tolerance.  Prejudice is not like a mouse.

Interdisciplinarity Requires Patience and Tolerance

By Graham Peterson

It should really be an obvious point that working with people from outside your own tribe takes patience and tolerance.  You have to translate constantly; you might not crack coconuts the same way; and your hypothalamus will kick up and say “gross run!” or “fuck that, kill!” when it senses something foreign.

Alas, in the academy there is enough pretension about quality, dedication, honesty, and rigor to power a weather balloon, and we tend to reserve those moralistic judgments to our friends and presume The Other Team lacks them.

So it happens that when people read outside their field, they get really prickly when someone makes mistakes that a specialist from inside their field wouldn’t have made.  “This is low quality [relative to what my peers who have worked on it for decades have done]!  This person is dishonest [relative to people who have better information]!  This person is lazy!”

Here’s how this type of thing goes.  Brad DeLong illustrates nicely for us, because he completely lost his mind about David Graeber’s Debt: The First 5,000 Years.

DeLong found mistake after mistake that Graeber made concerning recent economic history — a topic that economic and monetary historians have covered exhaustively (because of its richness of quantitative historical records).  We are then supposed to infer that the rest of the book is probably garbage too.  A lot of people (maybe not DeLong) also conclude that Graeber is stupid, dishonest, or incompetent.

These are exactly opposite the conclusions one should draw in such a situation.

When someone from outside your field screws up details from inside your field, it is probably the case that they have a much better grasp on the details from inside their field.  Concluding otherwise is like showing up in Tahiti and concluding that the Tahitians have nothing to teach you about Tahiti because they speak English poorly.

If you stop listening to the foreigner, or trash them repeatedly on your blog, you lose the opportunity to educate them about your field, and to be educated about theirs.  I think there are a lot of things wrong even with Graeber’s interpretation of prehistoric economy, but I had to suppose that a world famous anthropologist was not completely incompetent or lying to me, in order to consider the issue.

The Consequences of Ideological Homogeneity in Gender Research

By Graham Peterson

Paula England is a fantastic researcher. Her work in gender is some of the strongest there is, especially her criticisms of Gary Becker’s sexual division of labor, the idea that gender roles are as natural for household efficiency as are kitchen roles at T.G.I. Friday’s.

So I was disappointed to see a tweet from her showcasing her new data on sex. After they sex, men and women report differently whether the woman climaxed. As the chart below shows (credit England), it is a pretty big difference. Men report around twice the rate of female orgasm that women do. England put it on Twitter and tagged it: “Hint: delusion.”

orgasmgender.xlsx

Omar Lizardo, also one of the smartest people in sociology, posted the same graph to Facebook and tagged it: “#banmen.” He was joking of course, and so was England, probably, but not without a point. Jokes that land, do, because they’re meaningful.

This issue is pretty important to a lot of people. Guys want to believe they’re doing a good job, of course, either because they’re egotists, or because they’re givers. And women, presumably, want to come. England and Lizardo seem to believe men are egotists. And a lot of women probably agree! But a lot of men, for their own reasons, probably do not.

Either way, the question sociologists and women want answered here is: are men egotistical sex havers? These data can’t adjudicate that question.

What they can tell, maybe, is the rate at which men are overestimating themselves, remaining agnostic on why men overestimate themselves. To do that we need to recognize that some percentage of the discrepancy comes from women pretending to orgasm.  About that, England and her co-author say in Contexts,

[The discrepancy] may be because women fake orgasms to make men feel better, and men are misled by this; we learned in qualitative interviews that some women do this, but don’t know how prevalent it is.

Now that’s a strange thing to say, because there are a lot of data on the prevalence of faking.

The first hit on a Google search for “prevalence of woman faking orgasm” is a PubMed article stating that 50% of women report that they have faked an orgasm at least once (interestingly, 25% of men report faking too). A little more digging pulls a writeup of another study on orgasms that estimates that women fake “routinely” 25% of the time. Across many other studies, estimates land between 50% and 66% of women faking, says the second study’s literature review.

Now most of this research asks whether women have faked an orgasm, not how often they have, so it’s not possible to know how much of the gap in the chart comes from faking. But the percentage of women who fake is enormous, suggesting that unless women fake only once, and then have a come to Jesus, faking is prevalent. If the chart above controlled for faking, the men’s bar would take a haircut that might not inspire humor. It’s a relatively well studied issue.  It is just not the case that, “[we] don’t know how prevalent [faking] is.”

Now I love polemics, and I think they’re good for science. I think people should be transparent about their priors, and argue stuff out.  So, in the spirit of transparency, I can’t fault Lizardo and England for their feminist prejudices, and I’ll allow myself some bias in the other direction.

Why did two of the nations top sociologists feel it was nbd lol to conclude from one set of descriptive statistics that men are deluded? Because sociology is ideologically homogenous. No one on any side of this discussion bothered to double check. When scientists share the same beliefs, nobody checks, and the quality of science suffers.

With better data on the frequency of faking we can know how much of that gap shrinks. With better data on how often men ask if their partners orgasm, we can know how many care but have bad information (many men cannot tell when a woman climaxes). Whatever is left in the discrepancy is the rate at which men overestimate themselves, and that’s an issue worth addressing. That is when we ask whether masculine overestimation comes from wishful altruism, or careless egotism.

But it’s difficult to imagine that anyone will go get those data, when researchers are satisfied to high five each other about delusional men, before doing perfunctory Google searches.