Do Voters Know What They’re Doing? An Interview with Christopher Achen.

 _{A voter being interview by a pollster (Smithsonian)}

Christopher Achen is co-author (with Larry Bartels) of a stunning new account of the forces that drive American politics: Democracy for Realists. The subtitle gives the theme away: “Why Elections Do Not Produce Responsive Government.”

Achen is a professor of politics at Princeton who’s known for his legendary skill at imaginative quantitative research. Once, on a hunch that irrelevant bad news can have an impact on voters, he spent a month in the basement of the Princeton library studying what happened in the aftermath of the worst series of shark attacks in United States history. It was in 1916. It took place in New Jersey. Four people died and the hotels in the affected beach communities emptied out, killing the local economy. What Achen discovered by studying local voting records closely was that voters punished the president of the United States (Woodrow Wilson) when he came up for re-election in November, voting heavily against him. This was proof positive that voters aren’t terribly rational. Wilson, after all, couldn’t have done anything to stop those shark attacks. Achen tells the story in the book. (Disclosure: I learned about the story from a paper he wrote years ago and included it in my own book, Political Animals, after interviewing him.)

His coauthor Larry Bartels, formerly at Princeton and now at Vanderbilt, is the well known author of Unequal Democracy: The Political Economy of the New Gilded Age.

The interview was conducted by email.

Rick Shenkman: Before we get to the Great Depression and other historical events, let’s start with a question on a lot of people’s minds these days. Do voters know what they’re doing

Christopher Achen: Sometimes the voters decide wisely, as they are expected to do in the conventional view of democracy. That's what we call "the folk theory."  But unfortunately, it is not at all uncommon for the voters to make very foolish choices, especially in primaries and in referendums, when they get no guidance from their partisanship. 

One example is California voters' enactment of Proposition 13 in 1978, which mandated dramatic property tax cuts and sharp reductions in local budgets. The result was poor public services, including understaffed and undersupplied fire stations.  When the Oakland Hills Firestorm struck in 1991, it could not be controlled.  More than 3,000 homes burned down and 25 lives were lost in just 24 hours.  The Federal Emergency Management Agency (FEMA) reported afterward that Proposition 13 was a major cause of the fire's getting out of control.  In effect, the voters had chosen to burn their homes down.
There are many such examples, some of which we set out in our book.  Even in presidential elections, most voters are remarkably poorly informed about the issues in the campaign.  The folk theory is great for Fourth of July speeches, but it is a very misleading way to think about democracy.

Most political scientists believe that voters can be trusted to make somewhat rational decisions. You and Larry Bartels argue this is pretty much nonsense. What first convinced you of this?

In large part, our teaching.  Every year, professors go over the same materials, both classic readings and more recent research, helping students understand the current state ofknowledge. Wehave to understand the research thoroughly ourselves in order to explain it and to assist students in making sense of contradictory findings and opinions.

Inevitably, over the years one comes to see that some lines of research are stronger than others.  In our own case, we saw that the arguments of Angus Campbell, Philip Converse, Warren Miller, and Donald Stokes in their seminal book on voting, The American Voter, were more powerful than their competitors.  And so we have adopted their views and the views of those who have come after them and have extended and strengthened their findings, as we try to do ourselves.

But don’t the views of most voters line up pretty well with the candidates they vote for?

Again, sometimes. But to a remarkable degree, as decades of survey research prove, the voters begin election seasons with no opinions on the major issues of most campaigns.  As time passes, they adopt the views of the candidates they favor. Donald Trump's voters this year favor building a wall across Mexico, for example--an idea they got from him.  But his extremist foreign policy views—his skepticism about NATO, for example—leave little trace in most voters' thinking.  Similarly, one recent poll suggested that fully 25 percent of those who favored Bernie Sanders in the primaries now support either the libertarians or Trump. That makes it all too clear that a great many candidate choices do not reflect settled, thoughtful policy judgments.

If voters aren’t deciding which party to support on the basis of rational choice, how are they making this decision?

In accord with a great many other social scientists in recent decades, we believe that group loyalties are fundamental. People think of themselves as members of various groups—their political party, their race and ethnicity, their occupation, their religion or its absence, their gender, their place of birth, and many other groups. Sometimes those group memberships are lightly held.  But in other instances, they are fundamental to people's sense of self, in which case we call them "identities."  Many identities have important roles in politics.

Larry Bartels and I believe that the best way to understand people's attitudes and vote choices is to understand which identities are important to them. Typically, those identities determine not only how people vote, but also how they think about campaign issues. That is, the identities are fundamental:  They drive the voters' ideas and choices, not the reverse.  And we summarize a great deal of other scholars' evidence and add some of our own to demonstrate that those effects are very powerful. 

When people think that your party does not really welcome them, they will vote against you even if they agree with your policy views right down the line.

We, the voters, are pretty good at adopting views that make our choices seem rational and sensible, choices that we made on completely different grounds. That is, we rationalize.  And so we sound sensible and coherent.  But no one should be fooled.


Of course, it's obvious to all of us that OTHER people often use their identities rather than their policy views to think about politics, and that they are expressing opinions that are just rationalizations of their underlying group interests. The trick is to learn to see it in yourself, where it turns out to be just as powerful as in those benighted individuals with whom one disagrees.

Most historians and political scientists see great rationality in the behavior of voters during the Great Depression. Voters saw that Herbert Hoover’s policies weren’t working, so they elected FDR in a landslide. Pretty rational. But that’s not your take, is it?

No, it's not.  We show that the groups that should have rationally moved toward Roosevelt on policy grounds mostly did not do so. The people who did were those who lived in states where the economy got better after he took office. In those states like Oklahoma, where the economy got worse, they voted for the Republicans. Then as now, ideology mattered little. Then as now, as James Carville famously put it, "It's the economy, stupid."

You suggest in the book that if people were rational then worldwide they would have tossed out only conservatives whose policies contributed to the Depression. But they didn’t.

That's right. People just voted against the incumbents, no matter who they were—leftists, rightists, or centrists. People vote for a change in bad economic times.  They want to throw the bums out, even if they are replacing them with bums who will be worse for them.  They don't care.  And again, the accumulated evidence from many scholars for that view now extends over many decades and many countries, not just the Depression.

You note in the book that when an issue is salient to a voter they may act more rationally. How was this illustrated by what happened in Watergate?

Our argument about Watergate is based on the survey research from that period.  We show that Republicans who were repelled by Watergate often moved away from their party. That gave them to some degree a new identity. And then they began adopting more Democratic attitudes on issues completely unrelated to Watergate.  All that is quite odd from the conventional view of democracy, but it is just what one would expect if identities are fundamental and if identities drive policy views.

We also examine cases of conflicting identities, notably those involving abortion after the parties began to take different views on that topic in the Eighties.  Many people found themselves in the "wrong" party where abortion was concerned.  Of course, the issue was more important to women than to men, whichever side of the issue they adopted.  We show that disproportionately, when abortion attitudes and party conflicted, women changed parties.  By contrast, men cared less, and they disproportionately resolved the conflict by changing their abortion attitudes.

Irrationality can clearly lead to disaster. Is there some way to make voters more rational?

Fifty years ago, education was widely thought to be the solution to voter irrationality.  But education has risen dramatically, and nothing has changed.  The percent of people who know the name of their Member of Congress is just the same as it was half a century ago, for example.

Similarly, the Internet makes political information easy to obtain, but very few use it for that purpose, and those that do are often confused by misinformation. None of the usual reform proposals have worked, and none seems likely to work.  No doubt some marginal improvements can be made, but people are always going to be people. They are busy, and tedious policy debates cannot compete for their limited leisure time with family, sports, and other accessible and engaging topics.

We feel that a better solution is to follow the wisdom of the Founders.  That means making a place for parties and political professionals in our thinking about democracy.  Of course, the people need a voice.  But it needs to be balanced by the voice of leadership.  Unfortunately, people often think that they are above political parties, and that parties and professional politicians get in the way of real democracy. That's how we got frequent referendums in the Progressive states on all sorts of complex topics. And that's how you burn your house down, both physically and metaphorically.

Thank you.