ILYA SOMIN: First things first, why should we even care about the problem of political ignorance? Some people say that even if the voters choose out of ignorance, that’s all right — they’re just exercising their individual freedom. John Stuart Mill effectively refuted this sort of argument 150 years ago when he pointed out that voting is not just an individual’s choice. Rather, as he put it, it is “the exercise of power over others.” When we vote for people who will occupy positions of political power, they do not just rule over those who voted for them, they rule over the entire society. And when we exercise power over other people in that way, we have a responsibility to be at least reasonably informed in the way that we do so. So we do have good reason to be concerned about political ignorance.

And the evidence is overwhelming that the amount of political ignorance out there is pretty severe. For example, in our last election in 2014, the main issue at stake was which party would control Congress. Yet in surveys taken not long before the election, only 38 percent of the public even knew which party controlled the House and which party controlled the Senate. Similarly, in most recent elections, one of the big issues is the future of the federal budget, yet surveys consistently show that most of the public has little to no idea how our federal government spends its money. They massively underestimate the percentage of the federal budget that goes to major entitlement programs like Medicare and Social Security, even though these are among the largest items in the budget. On the other hand they massively overestimate the percentage that goes to foreign aid (which is only about 1 percent of the budget). The ignorance that we observe is not just limited to particular issues — it also goes to the very basic structure of government. A recent survey found that only 34 percent of the public even knows the three branches of the federal government: the executive, the legislative, and the judicial.

As I discuss in the book, the problem of political ignorance is not new. Nonetheless the present election does underscore its significance, and nowhere more obviously than in the success of Donald Trump, who says, “I love the poorly educated.” And well he should, because political ignorance is at least a large part of his success. If you look at the major issues that he has stressed in his campaign, virtually all of them involve significant exploitation of political ignorance.

Trump’s campaign first came to prominence when he made that famous, or in many ways notorious, speech about how Mexico is sending us criminals, murderers, and rapists, and thus increasing the crime rate. In reality, social science research consistently shows that immigrants, including Mexican immigrants, have lower crime rates, especially a lower violent crime rate, than native-born Americans do. Far from increasing the crime rate, they are actually lowering it. But polls show that 50 percent of the public believe that immigration is increasing crime, including over 70 percent of Republicans.

The other big issue that Trump has consistently stressed is trade. He talks about how our trade deficits with nations like Japan, Mexico, and others, are a sign of economic failure — of them “beating” us and turning us into “losers.” In reality, as economists across the political spectrum will tell you, a bilateral trade deficit with a foreign nation is no more an indication of economic failure than my trade deficit with my local supermarket is. Sadly they buy many fewer of my products than I buy of theirs, yet that is in no way an economic problem. Surveys reveal that trade is one of the issues with the biggest gap between poorly informed public opinion and relatively well-informed voters and experts.

I should emphasize that Trump is far from the only candidate, either in this election cycle or in others, who has exploited political ignorance. And as I also discuss in the book, today’s levels of political ignorance are actually not unusual. They are roughly similar to what we have had for almost the entire period that we have had modern public opinion polling to measure these things. But in some ways, the problem is more severe than before. Government today is much larger and more complex than it used to be, which makes it even more challenging for relatively ignorant voters to figure out what is going on.

When people hear the data on political ignorance, there is a tendency to think that it must be because the voters are stupid — or alternatively, maybe that the information just isn’t available to them. In reality neither of these two common explanations is likely to be true. The evidence strongly suggests that aggregate IQ scores have actually gone up in the last 40 or 50 years. Similarly, it is not plausible to argue that the information is not available to people. With the internet and other modern technology, it is more easily available than ever before in human history. The problem is not that the voters are stupid or that the information is unavailable. It is that they’re not using their intelligence to learn the information that is out there.

And as it turns out, this is perfectly rational behavior for most voters. If your only reason to become informed about politics is to cast a “better” or more informed vote in an election, that’s not much of an incentive at all, because the chance that your vote will make a difference to the outcome is infinitesimally small: only about one in 60 million in a presidential election, for example.

Some people, of course, do learn about things even though they can’t affect outcomes. Take for example sports fans — they know a lot about sports even though they can’t affect the outcome of games. They just find sports interesting and they love cheering on their favorite teams.

Similarly there are people who in the book I refer to as “political fans”: people who love learning about politics, love cheering on their preferred candidate or ideology or party and so forth. Such people know more about politics than the average voter. But there’s a problem. When you acquire information for the purpose of enhancing your fan experience, you tend to do so in a highly biased way.

Think about how sports fans react to new information about their favorite team: if it reflects well on the team, they love it. If it reflects badly on it, often they tend to downplay or deny it. This is exactly the same way that political fans, those most interested in politics, tend to react to new political information. They overvalue anything that supports their pre-existing views, and undervalue or ignore anything that cuts against them. They also tend to seek out political information only from sources that have the same views that they do.

This is totally illogical behavior if your goal is to get at the truth. On the other hand, it is completely rational if your goal is not primarily to get at the truth but rather to enhance your fan experience. Economist Bryan Caplan calls such behavior “rational irrationality.” It is rational to be highly biased in your evaluation of information when your goal in seeking it out is something other than getting at the truth.

In the book I discuss at some length why it’s unlikely that we can greatly increase levels of political knowledge. But for now I’ll press on to something that I believe actually can reduce the problem of political ignorance — making more of our decisions by voting with our feet, and fewer at the ballot box. We can vote with our feet in a federal system by choosing which state or local government to live under. We can also do so in the private sector by choosing what products to buy or what civil society or religious organizations to join.

Why would foot voting be any better than ballot box voting from the standpoint of political ignorance? After all, we do need to acquire information in both cases. The big difference is one of incentives.

Think about this: if you’re like most people, you probably spent more time acquiring information the last time you bought a car or a TV set than the last time you decided who to support for president or any other political office.

Is this because your TV is more important than who runs the government, or deals with more complicated issues? Probably not. It’s because you knew that the decision on the TV set would actually make a difference. On the other hand, with the presidency, there is only a tiny chance that your vote will actually change the result, so you take that decision less seriously. The same goes for decisions about what jurisdiction to live in, in a federal system. This applies not just to your incentive to seek out information, but also your incentive to analyze it in an unbiased way.

What are the implications of this? One is that we should favor greater political decentralization. When more powers devolve to the state and local levels, more decisions can be made by voting with your feet and therefore in a framework where people have more incentive to be well-informed. Similarly in many cases, we may want to devolve more issues out of the government entirely into the private sector, where you can vote with your feet even more easily than in a federal system. I do not contend that any of this proves that we should have the maximum possible decentralization or the maximum possible limitation of government power. Political ignorance is not the only issue that we should take into account in thinking about the role of government in society. If you read the book, as I hope you will, and agree with all my arguments, as I also hope you will, you may not favor the decentralization and limitation of government power as much as I do. However, you will favor both to a greater degree than you yourself would if you thought you lived in a world where political ignorance is not a serious problem.

JOHN SAMPLES: This book raises the question: why do we have elections? Is it just because we’ve always had them and people would want them? Are they desirable given the degree of voter ignorance? And after all, foot voting itself does not require elections.

SOMIN: That’s a good question and it raises an issue people often bring up when I speak on this issue: “Doesn’t your argument imply that we should do away with democracy entirely?” If you go back to the ancient world, this was exactly Plato’s answer.

Plato argued that voters don’t know what they’re doing, so we should replace democracy with rule by some sort of informed elite. You see some scholars today arguing for a less radical version of the same approach, for instance, giving more power to well-informed bureaucrats and the like.

I think that democracy still does have one important virtue, which is that, while the voters are generally ignorant, there are some forms of government failure that are so blatant and so obvious that even ignorant voters will tend to notice them and therefore punish the incumbents at the polls. In the book I discuss the striking fact that in no modern democracy has there ever been a mass famine — even though mass famines, including ones deliberately created by the government, are actually quite common in authoritarian regimes.

The reasons for this are fairly obvious: when a mass famine is going on even ignorant voters will tend to notice it and punish the incumbents. So democracy still has important advantages over dictatorship. But democracy with an extremely large government that is hard for the voters to keep track of, and that leaves a lot of room for political ignorance to cause harm, is not nearly as good as more constrained, more limited and decentralized democracy, where there is a greater role for foot-voting. It is not my view that democracy will always and everywhere be the best possible system. But I do believe that, for the foreseeable future, our best option is probably to limit and decentralize democracy rather than to try to do away with it entirely.

SAMPLES: I want to pose to you a certain kind of libertarian argument that I think you will recognize: over a long period of time we’ve had what we call “big government,” and it’s been pretty stable. And so often libertarians will say well, the people don’t want that — the elites are corrupt. The problem, really, is to move aside the elites, and then more libertarian, freedom-loving policies will happen. What do you say to libertarians like that, of a more populist bent?

SOMIN: Proponents of every ideology, not just libertarians but also conservatives and many on the left, have this notion that “the people are really on our side.” If it doesn’t look like it, it’s only because some evil elites have manipulated things, whether it be government bureaucrats and technocrats, or in Bernie Sanders’s version the 1 percent and the corporatists, or in the right-wing populist version, the liberal elites in the universities and the media and so forth.

Actually looking at public opinion data, as I have done, is a good cure for those kinds of notions. It reveals that the public does not fully support libertarianism or anything even close to it. It doesn’t fully support traditional conservatism or leftism either, and it also reveals that the government, while not completely responsive to public opinion by any means, is at least significantly influenced by it.

It is in fact the case that most voters favor having much more government than most libertarians would want, and this is not simply the product of manipulation by a small elite. That raises the question of what prospect there is for limiting government or decentralizing it as I would like to do? After all, if most voters are ignorant and don’t pay attention to facts and counterarguments, why would they pay attention to arguments that I make in this book, or to any arguments for limited government?

In the last chapter of the book, I do discuss some reasons for hope. One is that although voters do favor much more government than I think is desirable, and they have contradictory desires, they do have a deep suspicion of government and politicians. And a lot of survey data shows growing distrust of government in both the U.S. and many other countries.

Sadly, often this is channeled into thinking, “Well, if only we elect the right person and give them power they will fix things; they will bring change we can believe in,” as Obama said, or they will appoint the best people and solve all of our problems, as Donald Trump says. But it may be possible to channel some of this into minimizing the power of the institution of government itself.

Historically there are some things that for decades and centuries most people thought government should be doing, which today there is a broad consensus they should not be. State promotion of religion used to be thought of as one of the main functions of government. What could be more important than that? Over time, at least in the Western world, we’ve realized not that religion isn’t important but that governments tend to make bad decisions about it. The same thing might happen on some other issues.

Secondly some advanced modern democracies have actually succeeded in greatly reducing or decentralizing their government: recent examples include places like Canada, Ireland, New Zealand, and a couple of others mentioned in the last chapter of the book. I don’t think there can be a libertarian revolution any time soon. Indeed most revolutions of any kind tend to cause more harm than good, in part because of political ignorance. But it might be possible to have an evolution, if not toward libertarianism as such, then toward greater decentralization and limitation of government power. I would point to recent survey data, including some collected in a Cato paper you co-authored, which shows that, on average, Americans have more confidence in state and local government than in the federal government. This suggests that there might be at least some prospect of a constituency for decentralizing government power on some important issues.