This View of Life Anything and everything from an evolutionary perspective.
FIND tvol:
We Like To Think People Vote Against Their Self-Interest. Research Shows It’s Not True.

Sure, politicians are self-interested – the greedy jerks. But what about the rest of us? Well, we’ve got some good news and some bad news.

The good news is that lots of social scientists let the general public off the hook when it comes to self-interest. In the 2008 book, The American Voter Revisited, a team of political scientists reported: “The current scholarly consensus holds that self-interest is not a major determinant of issue attitudes or voting choices.” Built on decades of research, such claims have been profoundly influential within political science and beyond.

But now the bad news. As is obvious from the title of our new book – The Hidden Agenda of the Political Mind: How Self-Interest Shapes Our Opinions and Why We Won’t Admit It – we disagree sharply with the consensus. Our claim isn’t that self-interest is the only factor in public opinion and party preferences. Instead, we’re saying that it’s, well, a “major determinant.”

Sign up for our newsletters

I wish to receive updates from:

Challenging self-interest deniers on their own terms

Self-interest denials are often presented with a list of specific claims. We open our investigations in the book by checking in simple ways whether those claims are true.

One example is: “The unemployed do not line up behind policies designed to alleviate economic distress.” To check, we looked at a large, representative American dataset and compared unemployed people with people working full time. Who thinks it should be the government’s responsibility to provide a decent standard of living for the unemployed? Roughly three quarters of the unemployed but less than half of those working full time. Who thinks there should be more spending on unemployment benefits? Over half of the unemployed but only around a quarter of those working full time.

Despite claims to the contrary, views on unemployment benefits are in fact a clear-cut example of self-interest mattering in public opinion. Indeed, we examine a number of such claims about the absence of self-interest effects – involving affirmative action, healthcare, school funding, and so on – and find that most of the claims are false.

What do we mean by “self-interest”?

Usually “self-interest” is limited to short-term economic benefits to one’s self and family members. As we indicated above, however, even on this limited definition, the self-interest denials don’t hold up. Self-interest in fact does matter.

This sense of “self-interest” is often contrasted with “group interests” (involving racial groups, religious groups, etc.), which are seen as different from “self-interest.” The typical claim is that “self-interest” doesn’t matter much but “group interests” do.

We begin from a different viewpoint, one grounded in evolutionary psychology. An evolutionary perspective can help highlight various strategic conflicts within societies that go beyond income redistribution, including fights over in-group preferences and conflicting reproductive lifestyles. We don’t think there’s any clear reason to call it “self-interest” when one of these domains is at issue but not self-interest for the others.

We view the distinction between “self-interest” and “group interests” as something of an illusion. Consider these examples. If I’m unemployed, and I favor benefits for the unemployed, that’s not “group interests” but “self-interest” on the usual definitions, even though a bunch of unemployed people I’m not related to are also helped. But if I’m a religious minority, and I oppose discrimination against religious minorities, that’s not “self-interest” but “group interests,” even though my own self is probably helped by reduced discrimination.

The problem is that “group interests” are conceived as something fundamentally non-self-interested. Yet most of the examples of “self-interest” and “group interests” are things that both help one’s self and – at the same time – help unrelated people similarly situated to oneself. It doesn’t make sense to say that favoring unemployment benefits is self-interested if you’re unemployed but that favoring equality (or privileges) for your particular minority (or majority) group is not.

Not only that, but the self/group distinction has other problems. Typically, “self-interest” explicitly includes a group – family members. Further, people are themselves often better off when their friends and allies are better off; people can have an individual stake in the positive (and negative) effects of policy outcomes on their allies.

Still further, there are important matters left uncategorized by the self/group definitions. Suppose, for instance, that I’m employed right now; it’s not in my short-term economic interests to have an unemployment safety net in place. But, suppose that I am likely to become unemployed in the future, and, for this reason, I favor the continued existence of government unemployment insurance. Many of our social science colleagues wouldn’t call this either “self-interest” or “group interests,” but to us, it should be. Finally, our view is that not everything is about money. Suppose I’m a young person who engages in casual sex but wants to delay having children. It seems fairly clear that it’s in my “interest” to oppose policies that make it hard for me to achieve these goals. For instance, although I might not plan to use abortion in the service of delaying having children, it’s likely in my interests to keep the option legal.

Interests in action

Our view, then, is that if individuals are benefitting from policies based on their own group memberships, without sacrificing anything, then it’s arbitrary to say that’s not “self-interest.” We also think it’s reasonable to call it self-interest when someone favors a policy that is likely to help them beyond the very immediate future, or when someone favors a policy that advances their own non-economic lifestyle choices. To convey our broader sense of the notion of interests, we use the expression “inclusive interests.”

To us, the big question is whether we can predict people’s policy opinions from demographic features that might indicate where their interests lie: race, religion, sexual orientation, gender, education, sexual history, income, age, and so on. Are people with high incomes especially likely to favor lower income taxes? Are religious minorities especially likely to oppose religious discrimination? Are people who sleep around while delaying having children especially likely to favor reproductive rights?

And, indeed, in our analyses we found a wide range of evidence indicating that self-interest is a “major determinant” of issue opinions. The key is in recognizing that different kinds of political issues relate to different interest-relevant demographic features.

If you want to see some simple examples of what we’re talking about, go check out our recent essay on, which also includes an interactive political calculator allowing you to enter a short selection of demographic features and see how typical political opinions shift as a result.

But no self-interested person would vote

Reactions to these arguments have popped up involving a supposed counter-argument we wish to address: If people were really motivated by self-interest, then they wouldn’t vote because each individual’s vote is essentially always irrelevant to election outcomes. In fact, if it’s just self-interest, given that their votes don’t matter, they wouldn’t bother developing political opinions at all.

To be clear, we don’t have strong opinions on the question of why people sometimes vote and sometimes don’t. Maybe there are self-interested motives for voting beyond single-handedly determining elections outcomes, or maybe there aren’t. (It’s complicated, and we find it odd when people claim to know that there aren’t.)

The data we have analyzed relates almost entirely to the demographic correlates of particular policy opinions in surveys of the public. The claim has existed for decades that those relationships don’t show strong evidence of people preferring the alternatives that would advance their self-interest. We argue that a reasonable look at the data demonstrates that there is strong evidence.

Maybe there are self-interested motives for expressing opinions in favor of positions advancing one’s self-interest, motives that go beyond single-handedly determining election outcomes, or maybe there aren’t. Our argument is that, for whatever reason, people often do express such opinions – they really do seem on average to express preferences for the policies that help themselves.

Talk about the passion.

We don’t deny that there are political issues where self-interest-based explanations don’t seem to work (we discuss some specific examples in our book). We don’t deny that any given regularity has plenty of individual exceptions (something we repeatedly demonstrate by providing simple statistics where every reader can understand that the findings explain many cases but not all). We don’t deny that various other factors beyond demographic self-interest matter (we acknowledge some of these in the book and leave the door open for a range of others).

We’re responding to claims that self-interest rarely matters in public opinion. Our analyses indicate that it often matters. It’s beside the point to counter this with claims that it sometimes doesn’t matter.

We understand that our views are causing, in the words of one reviewer, “dismay and considerable anger.” When we poked the public with a New York Times essay on election day, these reactions were, well, vividly displayed in the online comments: “fatuous,” “absurd,” “a joke,” “ridiculous,” “hogwash,” “insulting and arrogant,” “bunk,” “nonsense,” “utter nonsense,” “a swamp of nonsense.”

We’re the bearers of bad news. We get it. In fact, it’s a major theme in our book – it’s the “and why we won’t admit it” part of our title. People strongly prefer to think that their own political views are motivated by principles and a well-reasoned concern for others, and will often passionately resist evidence that challenges those pleasing (and, we might add, self-interested) narratives.

But what are we, as researchers, to do? Are we really supposed to stick to crowd-pleasing conclusions? Maybe that would be in our own self-interest, or maybe not. But, as we’ve said, it’s not always obviously about self-interest, and even when it is it can be complicated.

Jason Weeden, a lawyer and psychology researcher, and Robert Kurzban, a professor of psychology at the University of Pennsylvania, are the authors of The Hidden Agenda of the Political Mind: How Self-Interest Shapes Our Opinions and Why We Won’t Admit It.

1 Comment

Join the discussion

One Comment

  1. Lesley Newson says:

    Your first couple of lines turned me right off. I have known a number of politicians and I would say that they are hard-working idealistic people. Have you read research that supports your dim view of all politicians or did you just think that is is cool to begin begin a piece of scientific writing with a cliché?