Are Demographics Destiny?

Self-interest, sex, snakes, and the making of our political preferences

The Hidden Agenda of the Political Mind: How Self-Interest Shapes Our Opinions and Why We Won't Admit It, by Jason Weeden and Robert Kurzban, Princeton University Press, 387 pages, $29.95

Rich people hate taxes, right? Or at least that's what a rudimentary understanding of self-interest would suggest. But the story immediately gets more complicated when Warren Buffett and Bill Gates start arguing that their tax rates should be higher, even as the pro-business U.S. Chamber of Commerce cries out in opposition. And according to data from the U.S. General Social Survey, the owners of manufacturing plants and retail stores are more likely than lawyers and stockbrokers to oppose a tax increase. What's going on here?

One standard explanation is that the majority of lawyers and financial experts tend to be concentrated in major cities, among the more cultured, progressive elite, and thus tend to vote for Democrats. Meanwhile, most factory owners and retail-store CEOs are scattered across the conservative heartland and thus are more likely to vote Republican. There is some evidence for this, particularly in the donation patterns for the legal and finance industries compared to the manufacturing and retail industries.

Yet such an explanation is profoundly unsatisfying. It's amusing to believe we live in a country where there exists an "eastern elite" and a "left coast" with Jesusland in between. But the standard red state/blue state tropes for why people develop specific political preferences are fraught with caveats, from "weird" liberal Austin, Texas, to the Republican mayors of New York City. Plus, while plenty of social science evidence suggests that political parties can shape their members' views, there still is the question of why a person self-selects into a political ideology in the first place.

The Pennsylvania Laboratory for Experimental Evolutionary Psychology psychologists Jason Weeden and Robert Kurzban believe they have the answer: a more robust understanding of what constitutes a person's self-interest.

In their new pop-political-psychology book, The Hidden Agenda of the Political Mind, the two researchers define self-interested behavior as "advancing any of a range of people's typical goals, whether directly involving material gain or something more subtle that advances someone's progress over the longer-term." To the degree that individuals identify with a particular group, whether it is based on gender, race, or socioeconomic status, they will perceive what is good for the group as in their own "inclusive interests" as well. Thus, they argue, people will favor political policies that are "in the interest of themselves, their families, their friends, and their social networks." Find what are in the interests of a person or their group, and you'll find their political preferences.

How does this explain the difference between factory owners and stockbrokers? Weeden and Kurzban note that the owners of physical businesses rely on depreciating assets that they have to replace over time. They therefore have an interest in limiting the extraction of assets from their operations. Individuals who make a living by amassing information to improve their specialization, by contrast, can expect to command higher salaries over time, and they don't have as much of an interest in ensuring tax rates are kept low. For Weeden and Kurzban, the reason lawyers are statistically less likely to be concerned about tax increases than manufacturing plant owners is that, all else being equal, the lawyers have a greater confidence that they will probably just make more money next year anyway.

Of course, in the real world, all else is never equal. Recognizing this, Weeden and Kurzban perform a statistical version of fracking, shooting highly pressurized regression analysis into General Social Survey data to cross-compare dozens of demographic factors to find good predictors of political preferences.

The General Social Survey includes characteristics like gender, race, education level, faith affiliation, and income, as well as behavioral patterns like how often one attends a religious service, consumes alcohol and/or recreational drugs, or has sex. Using a range of statistical tools, the researchers looked for correlations between these demographics and stated political preferences.

The result is over 150 pages of appendices that document a wide range of groups who trade off among their competing interests in different ways. The book naturally focuses on the policy areas where correlations are strongest, including abortion, gay rights, taxation, redistribution, and affirmative action. Areas where the correlations are weaker, specifically foreign policy and the environment, get relegated to a few paragraphs of discussion at the end of the book.

On the whole, the data are fascinating and presented in an easy-to-consume format, even for people whose idea of a fun Friday night generally doesn't include pivot tables or linear regressions.

For instance: If you're interested in finding communitarians, look for highly educated whites who frequently attend church services and have lower socioeconomic statuses. If you want to find libertarian women, look for white, heterosexual atheists or Christians with infrequent church attendance, high education levels, and incomes above the median.

If you want to find lawyers who are opposed to tax hikes, look for ones who got lower SAT scores and attended bottom-tier schools; they will tend to have more pessimistic expectations about their future earnings capacity. The higher an individual's human capital, the less likely they are to be concerned about future tax hikes. That goes for accountants and hot dog cart owners alike.

Weeden and Kurzban essentially argue that if we know an individual's demographic traits, we will know her interests, and we can predict her political preferences with a high degree of confidence on most (not all, they stress) matters of importance in the American zeitgeist. Just look for what is good for people with those demographic traits. If, say, we know that an individual has had five or more sexual partners, is sexually active outside of a marriage, does not have kids, and goes to bars or drinks once a week, that person has an interest in the accessibility of contraceptives. And if an individual has sex only in the context of a committed relationship, does not cohabit outside of marriage, has children, and attends a religious service at least once a month, that person has an interest in reducing the accessibility of contraceptives, to raise the cost to partners or potential partners of living a competing lifestyle.

Using this information, Weeden and Kurzban find strong relationships between these demographic traits and views on abortion, marijuana legalization, and whether pornography should be legal. The results are impressive, as the demographic data on lifestyle preferences do a pretty good job of predicting what an individual's political preferences might be. However, predicting preferences is different than explaining why the preferences were developed in the first place.

Weeden and Kurzban's book lands in the midst of a contentious debate in the academic community about how to understand the political mind. The leading story in social psychology today says that political and lifestyle preferences originate in our values. Psychologists such as Steven Pinker and Michael Gazzaniga have popularized the academic understanding that our brains come wired at birth to engage the world around us as social animals. We are born with evolved instincts to avoid potentially lethal creatures (e.g., an instinctive fear of snakes) or foods (e.g., a built-in disgust mechanism when facing rotting flesh or off-colored fruits).

We have also evolved moral instincts for engaging other people, gut feelings about human suffering, personal freedom, fairness, loyalty, purity, and the value of social norms. Psychologist Jonathan Haidt argues these universal moral intuitions in the brain are then molded and shaped by the culture we grow up in and the environments we spend time in. Our life experiences and group identities inform our worldviews (or moral intuitions), and the variance in culture and environment leads to vast differences in ideas about how important individual liberty is relative to purity, how fairness should be defined, and other matters.

As our values inform our preferences, we seek to advance not just those political ideologies but the value system that underlies them. Not abstract "higher-level" values, but structural instincts about what is right and wrong that have been and are continuously shaped by our life experiences and group identities.

In practice, this means that competing ideas about what constitutes a fair tax code are based not simply on individuals' self-interest but on the relative weight they give to different values. If someone values avoiding human suffering at a relatively higher level than preserving individual liberty, he is more likely to favor higher tax regimes that can pay for more social services. If someone defines fairness in terms of the value of equality, she is more likely to be concerned about a growing income gap than an individual who thinks about fairness in terms of people getting what they deserve. If someone values purity more than individual liberty, he will develop lifestyle preferences that lead to an interest in reducing contraceptives and subsequently opposing legalized abortion.

Weeden and Kurzban dutifully acknowledge that values can predict and explain our politics. But ultimately they emphasize their position as standing in contrast to Haidt, Pinker, and others who've developed theories of what causes people to wind up with their political preferences.

Weeden and Kurzban's book is well worth a read. But in embracing a tautological definition of self-interest, Weeden and Kurzban have simultaneously used the most accurate definition of human behavior and the least helpful definition for explaining ideologies.

Yes, it is reasonable to argue that all behavior is self-interested-given the limited knowledge a person has when making any particular decision, and assuming that "maximizing one's preferences" can include advancing the interests of whatever groups the individual identifies with. But that still leaves open the question of what defines the "interests" in the first place.

The more robust explanation for people's political preferences is found in their underlying values, which guide their self-interests and are shaped by life experience and group identities. Suss out the primary values of a person or group and you'll discover what defines their interests and their political preferences.

So how do we explain Warren Buffett's desire to write a bigger check to the IRS? Rather than rooting around in the demographic characteristics of lawyers versus factory owners, consider whether a person views harm reduction or promotion of individual liberty the more important value. Look for whether the person emphasizes or de-emphasizes equality. Those who value individual liberty will be the ones looking at their 1040s with disgust, while the billionaires who care about equality will be more cheerful when the tax man comes knocking.

Anthony Randazzo ([e-mail]) is director of economic research at Reason Foundation.