Public Choice Theory: Why Government Often Fails

Speakers
Antony Davies,

Release Date
October 9, 2017

Topic

Democracy and Voting Economics Government
Description

Governments don’t work the way most people think they do. Public choice theory explores how voters, politicians, and bureaucrats actually make decisions. Prof. Antony Davies explains.

  1. Behavioral Economics Ep. 5: What You Need to Know About Public Choice (video): Erika Davies and Prof. Antony Davies give an introduction to public choice economics and describe how insights on human behavior in the private sector can be applied to predict human behavior in the public sector.
  2. Public Choice: Why Politicians Don’t Cut Spending (video): Prof. Ben Powell uses insights from public choice theory to explain why politicians, despite what they may promise to voters, rarely cut government spending.
  3. Schools of Thought in Classical Liberalism, Part 3: Public Choice (video): Dr. Nigel Ashford gives a brief overview of the intellectual figures and ideas associated with public choice theory; part of a larger series on schools of thought in the classical liberal tradition.

ANTONY DAVIES: Public choice economics is a field in economics in which we take what we understand about how humans behave, which is normally applied to humans in the private sector, and we apply this knowledge to humans in the public sector. What emerges is an interesting dichotomy between how we imagine government works and how it actually works.
For an example, suppose we take some problem that we face. Here, as we think about government, let’s imagine government in the broadest terms. Three groups of people: voters, politicians and bureaucrats. We, as a society, encounter this problem that we want to solve, and the problem, let’s suppose, is that some workers don’t earn a living wage. The way we imagine government works when we say to the government, “Fix this problem.”, is the following. We have the voters, whose goal it is to help all people earn a living wage. The way they go about doing that is by voting for politicians who support living wage laws. Then we have the politicians. We imagine that their goal is to do what’s in the best interest of society. How do they go about doing that? By enacting and voting for living wage laws. Then finally we have the bureaucrats, who we call the public servants. Their goal, we imagine, is to serve the public. The way they go about serving the public in this instance is by executing these living wage laws for the good of the poor. What we imagine happens is unicorns and rainbows. All these wonderful things that will occur if we simply say to the government, “Fix this problem.”
Now let’s look at the government through the eyes of a public choice economist. A public choice economist looks at these three groups, voters, politicians and bureaucrats, and understands that these are human beings who face limitations and they have desires. Let’s start with the voter. In the view of a public choice economist, the public choice economist says, “Here’s a voter. He’s a human being. What’s his goal?” The goal of all human beings is to maximize their happiness. Some people’s happiness comes from consuming things, collecting lots of money, lots of stuff. Other people’s happiness comes from helping others, from doing good. Economists don’t make any judgment as to what it is that makes you happy. We simply say that people pursue the things that make them happy. Here are the voters. Their goal is to maximize their happiness. How do they go about doing that? They will become informed and they will vote if the benefit of doing so exceeds the cost.
Now we have the politicians. The politicians also are human beings, which means their goal is to maximize their happiness. How do they go about maximizing their happiness? By attracting at least 50% of the voters. If the politician does not attract 50% of the voters to vote for him, he’s no longer a politician. He’s now something else. The politicians that persist will be the ones that attract, whose goal it is to attract 50% of the voters.
Finally we have the bureaucrats. The bureaucrats are human beings. Just like the rest, their goal is to maximize their happiness. How do they go about doing that? I will argue that the way the bureaucrats go about maximizing their happiness is by crafting their jobs to satisfy their needs. What actually emerges is not the unicorns that we imagine when we say the government should do “this”.
You might say, “Yes, but we live in a democracy, and the whole point of a democracy is to guard against this selfish behavior. All we have to do is vote in better people, and if we vote in better people, we can have our unicorns and rainbows back.” Let’s think this through for a moment. Think about voter behavior in a democracy.
Suppose we have two groups of people, a large group of people over here, we’ll call this Group A, and a smaller group of people, Group B. I’m going to propose a law for this group of people. The law goes like this, “We will take $10 from every person in Group A, burn half of it, and whatever’s left, we’ll give to the people in Group B.” The question is, how are people going to vote on this proposed law?
Let’s start with a simpler question. Who wants to vote at all? If I ask the people in Group A and Group B, “Would you like to vote for or against this law?” all of them would raise their hands and say, “Yes, we would like to vote.” What’s the outcome of the vote? The outcome of the vote is the proposed law will be defeated. It will be defeated why? Because although it’s great for the people in Group B, they’ll all vote for it, they’re a small number of people. The people in Group A, this law is bad for them. They outnumber the people in Group B. They’re going to vote against it. Consequently, the law will be defeated. That raises another question, which is, “Is this good for society?” Clearly it is. This is a stupid law. It should be defeated. In defeating it, society is better off.
This thought experiment is normally how people imagine voter behavior in democracy working. In fact, it’s missing an important point. The important point missing is what economists call the information and voting cost. In information and voting cost is the cost, maybe in terms of money, maybe in terms of time and energy or effort, but it’s the cost of first being aware that there’s a vote pending. Two, reading up or investigating or finding out what it is that’s under consideration. Three, reading this proposed law and deciding whether, in the end, you think this is good for you or bad for you. Deciding how you’re going to vote. Finally, getting up out of your chair, going to the voting place, voting, coming back home. All of that is costly, sometimes in terms of dollars, sometimes in terms of money, but it’s costly.
In our thought experiment, what we’ll do is simulate this information and voting cost with a $20 fee. I’m going to propose the same law for society. We’re going to take $10 from everyone in Group A. We’re going to burn half of it. We’re going to give what’s left to Group B. The people can vote, for or against. I don’t care how they vote, but to vote at all, you must pay $20. That $20 represents the information and voting cost. Now, ask yourself, “Who is going to want to vote in this instance?” The people in Group B clearly are going to want to vote. They stand to gain a lot from this law. Although they have to pay $20 to vote, what they stand to gain exceeds the $20 they have to pay. Therefore, they will be interested in voting. Look at the people in Group A. The people in Group A each have to pay $10 if this law is passed. But the cost of fighting the law actually exceeds the cost of living with it. The people in Group A have a strong incentive not to vote at all.
What happens? What happens is in our thought experiment, this law passes unanimously. The only people who turn up to vote are the people in Group B, and the people in Group B unanimously vote for it, and the law passes unanimously. Is society better off? No. Society’s actually worse off. This is a bad law, yet this is something that has emerged from the democratic process once you understand that an information and voting cost is a real thing.
This is just a thought experiment, but you see this all over the place. Take, for example, sugar. There is a sugar tariff. All sugar that comes into the United States has a tariff attached to it. Therefore when you buy sugar, you pay more than you would otherwise. Who benefits from this? The people who benefit are American sugar producers. It turns out sugar requires an equatorial environment to grow, so to grow sugar in the United States, you have to build these large buildings and simulate an equatorial environment. This is all very costly. When you’re done, the cost of growing sugar in the United States ends up being much higher than the cost of growing sugar in more equatorial environments. Consequently, when American sugar producers try to compete with foreign sugar producers, they can’t. Foreign sugar producers can produce sugar at a lower cost.
The American sugar producers have an incentive to go to Congress and say, “Would you please impose a tariff on imported sugar so we can compete?” Congress, made up of politicians who are looking to get elected, turn to the electorate and say, “Hey. What do you think about this?” What the politicians see are basically two groups of people. A large group of people who are not sugar producers, who have other concerns in their lives, can’t be bothered with reading the minutiae involved in a proposed law, find the idea of tariffs boring to begin with. When the politician turns to them and says, “What do you think about this law involving tariffs?”, a lot of them just tune out. The information and voting cost for these people tends to be higher than the benefit that they perceive they would get by paying attention. Meanwhile, the sugar producers of the United States are very interested in this law because they stand to gain a lot. When the politician says, “What do you think about this?”, all the sugar producers and the people who work for them and the people they buy things from, all stand up and say, “Yes. This is a good idea. Let’s pass this law.”
I, as the politician, see then this sea of voters, one group of whom isn’t even paying attention because they don’t think it’s worth their time, and another group who is intensely paying attention, is interested in my passing this thing. I have a strong incentive now to pass this law. Therefore we end up with a sugar tariff. That’s just in the case of tariffs on sugar. Multiply this by all sorts of industries and all sorts of special interest groups that are looking for benefits from legislation, and all of a sudden you find that this information and voting cost is more than simply just a trick in a thought experiment. It’s something that actually pervades all of the democratic process.
What you’re observing here is something economists call concentrated benefit and dispersed cost. In this case, the benefit of the law, taking the $10 from everybody in Group A and giving it to Group B, the benefit from this law is concentrated in the hands of small number of people. That’s the people in Group B. The cost is dispersed over many many people. Those are the people in Group A. Consequently, the way we imagine lawmaking works, is that we look at the law and if the benefits of the law exceed the costs of the law, the law gets passed. In fact, that’s not the way it works. We look at the law and if the benefit per capita, benefit per person who’s benefited is larger than the cost per person, the cost per person on whom the costs are imposed. If that benefit per person exceeds the cost per person, then the law gets passed even if the total benefit is small and the total cost is large.
The way voters end up behaving is that they exhibit what we call rational ignorance. Rational ignorance means simply that the voters are ignorant of the law and it is perfectly rational for them to be so, because it doesn’t make sense that they would spend their effort figuring out what this law says and fighting it. To do so is actually more costly than living with the bad law in the first place.
You might say, “Yes, but we live in a republic, not a democracy.” The whole point of a republic is to get around this information and voting cost. The way we get around it is by electing representatives. We understand that the voters don’t have the time or the energy or the knowledge to study legislation. The information and voting cost is high. Therefore the voters will hire somebody to do that for them. We call this person a representative.
You could imagine the people in Group A hiring representatives to represent them, and the people in Group B hiring representatives to represent them. These representatives, then, will hash out whether this law is a good idea or not and vote on behalf of the people in the way the people would vote if they had the time and the energy to study the law themselves. That’s an interesting idea, but it raises a problem of its own, which is how do we divide our populace into districts and assign representatives to them? What we would like is that we assign representatives in an electorally fair way. Here’s what I mean by that. Let’s suppose we have 60% of the population who likes green stuff and 40% of the population who likes red stuff. If we were to establish representatives to represent this population, we would want the representative body, the Congress, to be comprised of 60% green representatives and 40% red representatives. That is, we want the Congress to look like the population.
One way to do that is something like the following. We put these 10 households in a district and we say, “You 10 households. You get to elect a representative. What kind of representative would you like?” These 10 households are all green, so if you ask them they will say, “We would like a green representative.” Then we put these 10 households in a district and we ask them. They’re green. They would also like a green representative. As will these 10 households. Now these 10 households are all red. If we ask them what kind of representative they would like, they’ll say, “Give us a red representative.” Similarly with these 10 households. When you’re done allocating representatives to the people, you end up with fair representation. 60% of the population is green, 40% red. 60% of the representatives are green, 40% red. That works. Maybe.
What happens when these people you have elected have the ability to re-draw the Congressional districts? For example, suppose instead of dividing the Congressional districts like I showed you, we do it differently. We put these 10 households in a district. We ask them, “What sort of representative would you like?” These people would like to have a green representative, these people would like to have a red representative. This district is mostly green, so if we ask this district, they will say, “Give us a green representative.” If I put these 10 household in a district, they will also elect a green representative, as will those 10 household, those 10 and those 10. Notice what’s happened now. By altering how we define the Congressional districts, we have completely silenced the minority. The red households now have no representation in Congress at all.
If you think that’s perverse, look at this. Suppose we were to put these 10 households in a district. These 10 households are mostly red, so if we ask them, “What kind of representative would you like?”, they will say, “Give us a red representative.” If we put these 10 households in a district, they’ll similarly elect a red representative. If we put those 10 households in a district, they’re almost entirely green. They’ll elect a green representative, as will those 10 households. Finally, here’s our last district comprised of these 10 households. They’ll elect a red representative. Look what’s happened. We have taken the same population we had before, redistricted it, and now we have a majority representation in Congress of a minority of the population.
The moral of the story is the information and voting cost means that we will have a tendency to pass laws that are bad for society because they benefit a small group of people and the cost is spread out over so many people that it’s not worth the large number of people’s time to figure out what this law is and to fight it. We think maybe we should have instead a representative government. The problem with the representative government is, depending on how we draw our district lines, we can have fair representation, which is fine, or the problem is we could completely silence the minority, or give the minority majority representation.
This is an example of what many consider to be the most gerrymandered state in the country. Gerrymandering is, of course, altering Congressional district lines to benefit the party in power. This is North Carolina. You can see the Congressional districts. Clearly someone has taken the time to think very carefully how to draw these Congressional district lines to give more representation to whatever party it is that was in power. In a democracy, voting results in worse outcomes for society. In a republic, gerrymandering can silence the minority or give majority representation to a minority of the population.
That’s voters, only one of these groups that we mentioned when we talk about government. Let’s talk about politicians’ behaviors. Remember, politicians are human beings just like everyone else. Picture the voters. Suppose the voters are thinking about a level of government service. This might be something like the military or public education or public transportation. Something like that. You can imagine, amongst all the voters, there is a lot of people who would like a low level of this government service. You could imagine as well there’s another large body of voters who would like a high level of this government service. You could imagine perhaps there is a smaller number who would like a low to medium level of this government service, and a smaller number who’d like a medium to high, and then finally you might imagine that the fewest number of voters of all would like a medium level of this government service.
Now, not all preferences for government service look this way, but a lot do. Here’s an example. Think about the military. There are a large number of voters who believe that the United States has no business intervening militarily in foreign countries. We should have a small military that’s devoted strictly to defending our borders. Then we have a large number of people who believe that we should bomb democracy into everybody on the planet. What we tend not to have are many people here, who believe we should have a military that’s large enough to be expensive, but not large enough to be effective. What you typically get with government service is this binary split of the voters between those who would like a small amount of the government service and those who would like a large amount.
Superimpose on top of these voters a couple of politicians. There’s this guy. He’s running for office. Remember, his goal is to get elected. He says to the voters, “If you elect me, I will provide you with this level of government service.” His opponent comes along and the opponent says, “Elect me and I’ll provide you with this level of government service.” What do the voters think as they look at these two politicians? These people like the guy on the left because the guy on the left has promised to deliver exactly the level of government service they want. These people like the guy on the right. He’s promised to deliver exactly the level of government service they want. These people over here, they don’t like either one of them, but of the two, they consider the guy on the right to be the lesser of the evils. He’s going to provide a level of government service that they don’t like but is better than the level of government service the other guy’s offering. When I ask these voters, “who are you going to vote for?”, these people vote for the guy on the left and all of those people will vote for the guy on the right.
What happens? What happens is the guy on the left is aware that this is how the voters are perceiving the two candidates. He will alter his position. He’ll come out and say, “I have been consistently misrepresented in the media. I have never been for a low level of government service. I have always been for this level of government service.” Now what happens? The voters rethink their positions. These people think that this guy is going to offer the level of government service they like the best. These people think this guy’s going to offer the level of government service they like the best. What does the guy on the left do? The guy on the left holds a press conference and says, “If you elect me, I promise you I will provide this level of government service.” Now what happens? All these voters like this guy. These voters like this guy. You end up with each of the candidates getting about 50% of the vote. What level of government service do we get? We get the level of government service that the least number of people wanted.
This effect occurs so frequently, economists have given it a name. It’s called the median voter theorem. The median voter theorem says majority voting will yield the median voter’s preference even if the median preference is the thing that’s liked by the least number of people. You see here, for example, why if you pick any American at random and ask them, “What do you think about public education in the United States?” Invariably they will say, “We have a serious problem with public education.” Almost all Americans will say that. They’ll say it for two different reasons. Half of them will say it because they believe that we’re spending too much on public education. We should be spending a lot less. The other half of them are going to say it because they believe we’re spending too little on public education and should spend a lot more. What we’re actually doing is spending the amount that satisfies the median voter, which tends to be the amount that the least number of people are interested in.
The conclusion with politicians is, “They don’t seek the common good.” They seek to get elected. That’s not to say that there are no politicians who care about the common good. It’s to say the following. If you give me two politicians who are equal in every way except one, this guy’s primary motivation is to seek the common good, where this guy’s primary motivation is to get elected, on average, this is the guy who’s going to win. Consequently, on average, the politicians that you will get are politicians whose primary duty is to get elected. Seeking the common good is something further down the list. How do politicians go about getting elected? They seek to satisfy the median voter, even if the median voter is a small portion of the population.
We’ve seen how voters behave through the eyes of a public choice economist. We’ve seen how politicians behave through the eyes of a public choice economist. Let’s now think about how bureaucrats behave. This is the county courthouse in the town I live in. If you have any business with the county, you go here. If you want a dog license you go here. If you want a variance to put up a fence you go here. You deal with these people. This is in western Pennsylvania, and in western Pennsylvania it rains all the time except when it’s snowing.
You’ll notice here this dark area is an underground garage, which is quite nice because if it’s raining or snowing, you can go in there to park and your car doesn’t get wet or snowed on. There’s an elevator you can go up to where you need to go and do whatever business you need to do, and come back down. If you’re ever in this town and you have business with the county government and you go here and you go down the street and you turn to go into this parking garage, you’ll see inside there a sign. The sign says, “Parking for courthouse employees only.” If you’re there as a citizen to do business with the county, you have to park somewhere else. Where? I don’t know. There are a couple of places here. You can go down the street there. This parking, the nice stuff, is reserved for the people who work in the building.
Think about this. Have you ever gone to a store and seen that all of the prime parking out front close to the door is reserved for the people who work at the store? Typically not. Typically these spaces are reserved either for the handicapped or a good manager will tell his workers, “Don’t park near the building. Park as far away as you can to reserve as many spots as we can close to the entrance for the customers, the people who are going to come here.” Why do they do that? This is very different behavior from what you see at the county courthouse. Businesses do that not because the manager of the business is somehow more altruistic than the people who work at the courthouse. It’s not because the manager of the business is smarter than the people who work for the courthouse. They’re the same brand of homo sapien in both instances, the public sector or the private sector.
The difference is one of motivation. In the case of the business, if I don’t make it as easy as possible for you to come to my store and to transact business with me, you will go elsewhere and I’ll lose your business. That’s not the case for the county courthouse. With the county courthouse, if I want a dog license or I want a variance to put up a fence or whatever, I have to go here. I can’t go to the county courthouse in the next county over. They’ll send me away. If I have parking tickets I have to pay I have to go here. I can’t go to some other county courthouse. In fact, I don’t even have the option to not go at all. They will send a nice gentleman with guns and handcuffs and force me to come and do business with them. The reason you get things like this, the prime parking reserved for the people who work here, is because of the difference in incentives. The people in the county courthouse do not have to make me the center of their lives because I have no choice but to deal with them. The people in the private sector have to make me the center of their lives because I have the ability to go elsewhere.
I had to renew my license recently. In Pennsylvania you get this thing in the mail and it says, “Pennsylvania’s driver’s license services are now online.” They give me this web address. The web address they couldn’t even be bothered to buy something neat like renewyourlicense.com. It’s dmv.state.pa.us. Fine. Whatever. This is a nice web address. I can go here. I can renew my driver’s license.
I go to a browser. I type this thing in. What do I see? I see this. There are 36 links on this page. I spend about five minutes clicking and backing out and clicking and backing out and clicking and backing out before I discover that this the link I want, “Driver’s license, photo ID”. Fine. I’m going to click on this thing. I’m going to renew my driver’s license. You click on that link and you go here. Here you have another 25 or 30 links and again another five minutes of clicking and backing out and clicking and backing out and clicking and backing out before I discover that this is the link is want, “renewing your driver’s license”. Okay, good. I’m going to click on this. My driver’s license will be renewed. You click on that link and you go here. This says, “Renewing your non-commercial driver’s license” … I don’t know what all that means, but we have step number one. This is good news. This means something’s happening. Step number one. “Receive an invitation to renew DL60A/DL60R, or obtain form DL143.” I don’t know what that means.
Imagine that you have to buy a screwdriver and you go to Home Depot. What normally happens? What normally happens is you walk into Home Depot. You walk up to the person. You say, “Hello. I’d like to buy a screwdriver.” If they’re not busy, they’ll often come out from behind the counter and say, “Come with me this way. The screwdrivers are over here.” If you need screwdrivers, maybe you need screws. “We have screws. We have 50% off on hammers this month.” Maybe you want to buy a hammer. They will do everything they can to make it as easy as possible for me to conduct business with them because if they don’t, I will go elsewhere. Imagine what would happen if you need to buy a screwdriver and you go to Home Depot and you say, “Hello. I’d like to buy a screwdriver.” They say to you, “You’ll need to fill out form DL60A/DL60R or obtain form DL143.” What do you do? You leave. You go somewhere else. Pennsylvania can do this because if I want a Pennsylvania driver’s license, I have no option except to deal with these people.
All right. I still don’t know what DL60A/DL60R is. I’m going to hope it’s that piece of paper they sent me in the mail and just skip step number one and go to step number two. Step number two. “Return the complete application with a check or money order made payable to PennDOT”, that’s Pennsylvania Department of Transportation, “in the amount indicated on the form.” Here we have in red, “Cash or credit/debit cards cannot be accepted.” They want a check. I have a checkbook. I write a check maybe once every five years to these people. I get out my check and I’m writing the check. My son, who’s in high school, walks by. He says, “What are you doing?” I say, “I’m writing a check.” He says, “What’s a check?” Anywhere else on the planet, when you want to buy something online, what do you do? You go there and they’ll take your credit card or a debit card or you could pay with PayPal. You can even find some places that will take things like bitcoin. These people want a check. Fine. I’ll write them a check.
Step number three. A camera card will be mailed. You should receive it in seven to ten working days after the form is processed. Once you receive your camera card, you take it, with appropriate identification, to the driver’s license center where they give you a new driver’s license. Step back and imagine what just happened here. They sent me in the mail a piece of paper. I go to a browser and I type in this web address and I get a whole bunch of links. I click and I back out and I click and I back out and I click and I back out. Ten minutes later I figure out what it is I need to do, and I write a check. I write a check. I put it in the envelope. I put a stamp on the envelope. I use stamps once every five years to mail the check that I write once every five years to these people. I then take this envelope to my mailbox, put it in, go back to the house, wait seven to ten working days, come back to the mailbox. In there is another piece of paper. I then take it to my car. I drive to the photo license center. I sit in the chair. They take my picture. They give me the license. Total time to transaction, seven to ten working days.
Imagine if these people faced competition. How would this transaction work? I’d get a text. Ring. “It’s time to renew your driver’s license. Click here.” I would click. It would say, “We have your credit card on file. Do you want to use the same credit card or do you want to change it?” I’ll use the same credit card, so I click. “Put your face in the camera.” Click. “Do you like the picture?” Yes. “Hit print.” It would print on my printer. Total time to transaction, 30 seconds if I don’t know what I’m doing. How do I know that this is how it would work if a private business were handling licenses instead of the government? This is the way it works everywhere else. If you would like to buy, God knows why, “The Best of Bowie”, you can go to amazon and type in “The Best of Bowie” and you will get this. It says, “Buy now with one click.” All you do is click. You click here and they’ll send you a CD. Even better than that, you click, you could download the thing. Thirty seconds later you’ve got this music that you wanted.
This isn’t even the latest technology. The latest technology is this. That’s a magnet. It goes on your washing machine. When you are running low on laundry detergent, you push the button. That’s all you do. You just push the button. It’s pre-registered with your credit card and your address and all of that stuff. You push the button. Three days later a box of Tide shows up on your doorstep.
Why the difference in behavior between the bureaucratic organization and the private sector organization? The difference is not because the people over here dumb, the people over here are smart, or the people over here are selfish and the people over here are altruistic. The difference has nothing to do with the people. They’re the same brand of homo sapien. The difference is in the incentives. The people over here in the private sector have an incentive to put me at the center of the lives because if they don’t, I will walk away. The people over here do not have the incentive to put me at the center of their lives. Therefore they do what any human being would do, what those human beings would love to do but can’t. They put themselves at the center of their lives. So you get things like arcane descriptions of forms that have to be filled out and parking spaces that are reserved for the people who work here.
The problem here is fundamentally, bureaucrats behave the way they do for two reasons. One, there’s no profit motive. If they behave well, they don’t get increased profits in which they get to share. If they behave poorly, they don’t incur losses that now cause their stockholders to come down on top of them. There’s no competition. There’s nowhere else for people to go. Therefore you get this natural behavior that the economist would predict.
Does this mean, ultimately, the government can’t make good decisions? That, I believe, is the wrong question. The right question is because humans in both the public and the private sector are fallible, they will make errors. The right question is not, “Who’s going to avoid making errors, markets or government?” The right question is, “When they do make errors,” because they will, “who has the greater incentive to identify and correct those errors?” An economist will argue that often, perhaps all of the time, the answer is people in the private sector, because they must respond to profit and loss incentives, because their customers can walk away.
The moral of the story is be careful what you wish for. When you see a problem and you say, “If only the government would,” because in reality we face all kinds of problems like rational ignorance, gerrymandering, the median voter theorem, and a bureaucratic monopoly. Because of that, these wonderful things you imagine that will happen when you say, “If only the government would,” often don’t happen at all. What you get is something very different.