Stereotypes Are Poisoning American Politics
I was born in West Virginia and spent all of 10 days there as an infant before my family moved to Ohio. Perhaps that's a license for me to say why Appalachians are poor, drink too much, and voted for Donald Trump. The best-selling and widely praised "Hillbilly Elegy: A Memoir of a Family and Culture in Crisis" by J.D. Vance, proceeds along those lines. But I shouldn't single out that book: Sloppy analysis of collections of people -- coastal elites, flyover America, Muslims, immigrants, people without college degrees, you name it -- has become routine. And it's killing our politics.
Three laws guide this bogus analysis of groups. First, define the group by the outcome you are trying to explain. Second, invoke a stereotype and exaggerate it. Third, endow the group with innate permanent properties, akin to racial characteristics. Together, these errors establish a kind of collective guilt, blaming an entire ill-defined group for the failings of its individuals, even if the offenders are a tiny minority. This is both divisive and false -- and all the more toxic because of its flavor of intellectual propriety.
Let's take each of these laws in turn.
Vance (who's actually a third-generation Appalachian immigrant to Middletown, in western Ohio) tells how his grandparents smashed up a pharmacy and threatened a clerk who'd told his son not to play with a toy on display: "If you say another word to my son, I will break your fucking neck." Hence, the definition: "Destroying store merchandise and threatening a sales clerk were normal to Mamaw and Papaw. That's what Scots-Irish Appalachians do when people mess with your kid."
When Appalachians move to western Ohio, Vance notes, "hillbilly values spread widely along with hillbilly people." How do you know they're still hillbillies? Because they wreck pharmacies. Defining Appalachians as those who are poor, uneducated, and violent, we find that Appalachian culture causes poverty, lack of education, and violence.
Ayaan Hirsi Ali wrote a memoir in 2006 -- "Infidel" -- that was roughly the Muslim equivalent of "Hillbilly Elegy". I respect Ali for her intelligence and courage, but question her group analysis. She defines "genuine" or "devout" Muslims by whether they support terrorism: "Every devout Muslim who aspired to practice genuine Islam -- the Muslim Brotherhood Islam, the Islam of the Medina Quran schools -- even if they didn’t actively support the [9/11] attacks, they must at least have approved of them.… It was about belief."
Sadly, training in economics offers little protection against this kind of circular reasoning. Sir Paul Collier, a distinguished scholar, devoted his widely praised book on "The Bottom Billion" to another group defined by outcome and not much else. In a later work, "Exodus," he extended his inquiry to groups of migrants from poor to rich countries, influencing debates about immigration in both the U.S. and U.K. One insight was that "the countries now at the bottom are distinctive not just in being the poorest but also in having failed to grow." That's true, but it yields no more information than the initial misleading aggregation: It's like saying, "Those at the bottom are distinctive not just in being at the bottom but also in being at the bottom." When you order the world’s seven billion people from richest to poorest, there will always be exactly a billion people in the bottom billion.
The second law of pseudo-analysis of groups says, reinforce stereotypes. In the case of Appalachia, these have a long history -- "Deliverance" and all that. Recent fires in the Smoky Mountains aroused suspicions about the "moonshine stills of the poor, ignorant hillbillies." This takes the badly defined group fully into the realm of caricature. (Note that when Appalachians complain about such accounts, Vance and other analysts see a flawed culture that "makes it hard for Appalachians to look at themselves honestly." )
Group stereotypes typically have a kernel of truth, reflecting some trait which is over-represented -- but the likelihood of the trait's occurring is then greatly exaggerated. Consider the elderly Floridian. Florida's proportion of elderly people is indeed above the national average, but not as much as you think -- 17 percent are 65 or older, compared to a national average of 13 percent. Another stereotype is the heartless conservative. In surveys, conservatives are more likely than liberals to say they don't care about the weak and vulnerable -- but most conservatives say they do care. Poverty is higher among Appalachians than the national average, but not by much -- Kentucky has a poverty rate of 18.5% and West Virginia 17.9%, compared to a national average of 14.7%. The typical Appalachian isn't poor (and doesn't trash pharmacies).
Muslims are over-represented among the tiny number of terrorists worldwide, which leads many to wildly overestimate how likely Muslims are to be terrorists or jihadist fighters. (More than 99.99 percent aren't.) We also stereotype terrorists as mainly attacking Americans or Western Europeans, whereas Western Europe and the U.S. actually accounted for less than 3 percent of deaths from terrorism worldwide between 2001 and 2015. Muslim terrorists mainly attack Muslims.
"The Bottom Billion" said of its subject that the group's "reality is the fourteenth century: civil war, plague, ignorance." Over the previous four decades, one hundredth of 1 percent of Africans had died annually in wars. "Exodus" said, "Nigerians radically, deeply, do not trust each other" and "Nigerian immigrants to other societies tend to be untrusting and opportunistic." Survey data show Nigeria to be close to the world median on measures of trust -- or about as trusting as France. (Neither the original study nor the survey data say anything about "opportunism.")
In the U.S., we "coastal elites" are likely to condemn stereotypes when they involve immigrants, nonwhites, or religious minorities. But we're more accepting of stereotypes that portray Southern, Midwestern, uneducated, working class whites as stupid, racist, and homophobic. (You wonder why so many rejected our advice on how to vote?) On the other hand, if you're thinking, "Speak for yourself" -- you're right! "Coastal elite" is another stereotype. We don’t all disdain Appalachia or flyover America.
The crucial point is that all these stereotypes purport to be findings. In fact, they're the opposite: a refusal to see vast individual variation within groups.
The third and most dangerous law of pseudo-analysis of groups is to treat each stereotyped collection as a race or its near equivalent, a permanently embedded culture. The idea is that all group members have a biological or innate propensity to behave a certain way. Studies suggest instead upsurges of bad or extreme behavior by some members of such groups for historical, changeable, circumstantial reasons. There's no evidence for innate, permanent traits.
We understand this for some groups, but not for others. We understand Floridians don’t have a biological tendency to be elderly, and that Florida wound up with more old people for reasons of history and climate. But many anti-racist liberals see Appalachians as akin to an inferior race innately prone to racism. When it comes to Muslims, there's a receptive audience for Ayaan Hirsi Ali’s message about the 9/11 attacks: "This is based in belief. This is Islam."
In the late 19th and early 20th centuries, terrorists were likely to be Eastern European or Russian. In the 1970s they were Irish Catholics and Protestants. In the 1980s and 1990s, Peru's Shining Path and Sri Lanka's Tamil Liberation Tigers weren't Muslim.
Collier sees migrants as carriers of the persistently dysfunctional cultures they absorbed in their countries of origin. Yet U.S. immigration laws once discriminated against Catholics and Jews from Eastern and Southern Europe. Asian migrants used to inspire vicious xenophobia in the U.S.; now they're seen as model immigrants. More recently, some have started to notice that African immigrants in the U.S. are highly educated (41 percent have college degrees, higher than any other immigrant or native racial group except for Asians).
The three laws come together to sustain the loathsome idea of collective guilt. This idea is a gross affront to justice. Whatever group we belong to, each of us should be held responsible for our own misbehavior, violence, and racism. Wrong in itself, collective guilt also presumes a nonexistent capacity to alter group behavior. Collective blame of whites or sub-groups of whites hasn't defeated racism. How exactly are heterogeneous, leaderless, decentralized groups supposed to control the behavior of their members? Vance calls on Appalachians to alter a "culture that increasingly encourages social decay." To what Central Committee on Appalachian Culture is he appealing?
One of the greatest insights of economics is that individual incentives work while group rewards and punishments don’t. Collective guilt doesn't work to change anyone's behavior. In the end, collective guilt, fashioned from bogus analysis and delight in stereotypes, is mere slander. It's a formula for constant antagonism and it's poisoning American politics.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the editor responsible for this story:
Clive Crook at firstname.lastname@example.org