Sam Seitz

Americans tend to have great pride in their democratic form of government. While Congress, Trump, and most other federal agencies are deeply unpopular, the ideas of democracy and freedom of speech are still viewed as sacrosanct by the American body politic. Unfortunately, there is little evidence to suggest that these values and institutions lead to the kind of noble and engaged citizen-voters so frequently venerated in the American political mythos. Most voters are myopic, hypocritical, and clueless. They support contradictory policy positions, fail to understand basic facts, and are overly eager to extrapolate from personal experience when trying to make decisions for the country as a whole. This doesn’t mean that democratic governments are bad (indeed, vast bodies of social science research suggest that democracies are less prone to fight amongst themselves and are more resistant to repatrimonialization than more authoritarian forms of government). However, it does undercut the popular, but completely unsubstantiated, myth that the voters always know best and deserve deference in all instances.

Christopher Achen and Larry Bartels masterfully demonstrate the ineptitude of voters in their book Democracy for RealistsThey start out by thoroughly demolishing what they term the “folk theory” of democracy. Specifically, they argue that governments largely fail to represent the interests of voters because a) voters frequently don’t know what they want and b) partisanship and group herding ensure that voters nearly always conform to their party’s orthodoxy. In their words, elections are nothing more than a “coin toss” or game of “musical chairs.” Instead of representing the aggregate will of the electorate, most electoral outcomes are nothing more than the result of random exogenous variables completely outside the control of the government. The book is replete with fascinating anecdotes and statistical studies that hammer this point home, but a few, in particular, stand out.

In one chapter examining the relationship between economic growth and voting tendencies, Achen and Bartels find that economic decline tends to strongly correspond with the fall of incumbent parties. This suggests that voters lack a unified economic worldview, because instead of standing by theories of free market liberalism or Keynesian government interventionism, most voters instead rely on short-term economic conditions to determine their vote. Some might argue that this behavior is rational. After all, few voters are trained economists, and thus it’s too much to expect them to have nuanced economic worldviews. Using economic indicators to determine the veracity of a party’s economic platform would thus appear rational. But this story of rationality is undercut by Achen and Bartels’ finding that voters only care about economic conditions in the months preceding an election. Instead of looking at a party’s economic record over its entire tenure in office, voters concentrate solely on the period of time directly before an election. This is obviously irrational, as short-term economic swings usually have more to do with market fluctuations than any kind of government policy. If a government oversaw 15 consecutive quarters of economic growth followed by one quarter of slight economic decline, most reasonable observers would characterize its handling of the economy as exceptional. But according to Bartels and Achen’s statistical findings, we would expect support for this government to fall (all things being equal) due to the incredibly myopic nature of voters.

Voters’ inability to accurately gauge economic conditions and make effective cost-benefit calculations was further illustrated in Achen and Bartels’ examination of an Illinois referendum on fire station funding. Voters in one Illinois county were asked to vote on taxes used to fund local fire houses. They ultimately chose to lower taxes, depriving the fire department of much-needed funds and thus significantly increasing its response time to fires. While the tax cuts led to a small increase in voters’ disposable income, this was offset by the higher insurance premiums needed to cover the increased risk of fire damage. Indeed, voters ended up paying more in insurance price hikes than they saved in tax reductions. But, because most voters lack the ability to think beyond first-order considerations, they ended up selecting the inferior economic policy of tax cuts and insurance hikes.

Voters also suffer from an inability to accurately determine what the government can reasonably control. An examination of New Jersey voters in 1916 showed that they severely punished the incumbent Democrats for a number of high-profile shark attacks off their coast. As Achen and Bartels point out, the rapidity with which the attacks occurred meant that there was effectively nothing that the government could do beyond warning swimmers and killing as many sharks as possible. Nonetheless, voters failed to show any understanding for the impossible situation the government was in, and New Jersey witnessed a ten point swing against Woodrow Wilson during the 1916 presidential election. To borrow a line from Bartels and Achen, voters have an annoying tendency of “kicking the dog to get back at a difficult boss at work.” Instead of rewarding and punishing parties for their record, voters use elections as a form of catharsis, punishing incumbent parties for anything and everything that has gone wrong, even if the government had nothing to do with creating the problem.

Least surprisingly, but most troublingly, Achen and Bartels convincingly argue that partisan identities bias voters’ understandings of even basic facts. For example, voters tend to look to their party’s platform when forming opinions on issue areas for which they lack strong convictions. And, when voters already have an entrenched opinion, they simply project their beliefs onto the party and assume without any evidence that the party leadership shares their views on controversial issues. Unfortunately, as Achen and Bartels show, this means that people view objective facts through partisan lenses. One example of this occurred during the Clinton presidency. Over Clinton’s tenure, the U.S. federal budget deficit fell precipitously. However, many Republicans reported that they believed the deficit had grown. Interestingly, Republicans with a moderate level of political knowledge were the most likely to hold erroneous views about the budget. They knew enough to know that the president was a Democrat, and they knew deficits are generally considered to be bad. Therefore, they projected their partisan biases onto the data and came to the wrong conclusion. Very politically knowledgeable Republicans knew the deficit had fallen, but they attributed this more to the GOP-controlled Congress than to Clinton. And although low-information Republicans didn’t know anything about the deficit, they were actually more accurate than moderately knowledgeable Republicans because their ignorance shielded them from partisan biases.

Another example of this kind of partisan interpretation of objective data occurred immediately after Donald Trump won the U.S. presidential election. Gallup asked Republicans and Democrats about their confidence in the U.S. economy, and the results were astounding. Despite nothing having changed (Trump had won the election, but he wouldn’t be inaugurated for another two months), partisan perceptions of the economic health of the U.S. shifted drastically: the number of Republicans who believed the economy was getting better rose 33 percentage points, while the number of Democrats who thought the economy was improving fell 15 percentage points.


In short, Democracy for Realists paints a bleak picture of the average voter. Instead of using objective facts to determine political allegiance, voters use their political allegiance to interpret objective facts. And, on the rare occasion that a voter does try to reward or punish a political party for its governing record, he often uses comically absurd indicators to determine a party’s effectiveness. Instead of looking at long-term economic data, he will care solely about the monthly or quarterly indicators for the period of time immediately before an election. Instead of looking at government domains such as defense and infrastructure, he will punish the government because of the random movements of shark populations. To be blunt, the average voter is an idiot. And even politically savvy voters are easily manipulated by partisan signaling and group social pressures. Again, this does not mean that democracy is a bad form of government: Allowing voters to (imperfectly) hold a government accountable is surely preferable to a system of completely unconstrained authoritarianism. What it does mean, however, is that voters can be easily manipulated by charlatans and demagogues. Therefore, it is important that we, as voters, resist these nefarious actors. Otherwise, we’ll just continue to be played for fools.