Economics: Understanding Public Choice

This lesson has not been reviewed.

Please purchase the lesson to review.

We've seen how the free market fails to provide the efficient amount of public goods, and how government - collective action, cooperative action - can give an improved outcome. So now you might wonder: How does the government work to deliver the efficient amount of public goods? Do governments work according to some different set of rules? Well, some people believe so. But economists are going to be slow to give up their belief that people are always behaving rationally, in self-interest. And if you change the rules of the game, they're just going to learn to play by a different set of rules.
Now the government may work differently than the market, but we can still imagine that with this different set of rules, the people involved - the voters, the politicians - they're all going to be pursuing self-interest in whatever environment they find themselves in. And this helps us to understand what economists call "public choice theory" - the tools of economics applied to understanding how governments work, and how public goods get provided, and whether or not the outcomes are in some sense efficient, or good.
Let's look at some instances of public choice theory applied to understanding how voting works. Now voting is a peculiar thing. Voting is different than the market. In the market, if you want something and you want it really bad, you can pay more money for it. That is, if you want the cheeseburger and I want the cheeseburger, the person who bids the highest price will get it. However, in politics everyone has one vote; therefore, you're limited in your ability to express the intensity of your preferences, and that leads sometimes to some weird outcomes. Let's look at some instances of voting rules and how they play out in public choice situations.
First, let's look at how majority voting can lead to an inefficient outcome. Suppose we are in a city where people are voting on whether or not to build a new park by the highway, and let's suppose there are three people who live in the city - Person A, Person B, and Person C - and they're going to have an election to decide whether or not they are going to pay the taxes that it would take to build this park. And let's suppose the total cost of this park is $300.00, and that if this election is successful - that is, if the park is approved - each person will have to pay an equal share of the cost of the park. So the taxes that are going to be imposed in this case will be one-third of the cost of the park on each of the three voters. So the cost to each voter is going to be $100.00. The total cost of the park is $300.00, and the cost per voter is $100.00.
Now let's suppose that we know something about each of these three voters. What we know is the satisfaction that they would get from being able to play in this park, and let's suppose it's different for each person. Let's suppose the guy on top really likes parks a lot, and if he goes to the park, he's going to get a satisfaction of $110.00 worth of pleasure, being able to visit this park daily. Let's suppose that he would pay $110.00 rather than do without the park. The next guy is also a big park fan, but he doesn't enjoy it quite as much. Maybe he lives further from the site of the park, and his benefit is only $105.00. And let's suppose our last citizen is not so crazy about the park; it's worth $75.00 to her - not as much as it is to the other two citizens.
Now, first question: Is this park going to be approved in an election? Well, look, our Voter A finds that the benefit is greater than the cost - that is, the satisfaction is more than enough to compensate for the extra taxes. He is going to vote in favor of the park - "yes." The next guy gets a benefit of $105.00 - greater than the taxes he's going to incur; he likes this idea, he votes "yes" for the park. And our third voter finds the park is not going to be worth enough to her to compensate for the extra taxes; she votes "no." Well, the park is going to be approved by majority vote. But here's the rub: The park should not be approved according to economic analysis because the benefits are not enough to compensate for the costs. The total cost of providing this park for this city is $100, $100, $100 - $300.00; the total amount that we had to raise in taxes covered the cost of the park. But how much benefit is provided in total by the existence of this park? - $110.00 plus $105.00 is $215.00, plus $75.00 is only $290.00 worth of benefit. As you can see, the benefit is less than the cost, and by the economic way of thinking, this park would be an inefficient use of society's resources. But majority vote approves it anyway simply because enough voters find it individually profitable to have the park, while socially the park is unprofitable.
So there you have it, an instance of an inefficient election outcome; majority voting can lead to an inefficient result. Now we're going to see how majority voting may fail to yield any meaningful result at all - what we call the "impossibility theorem of voting." The implications of the impossibility theorem are best seen in a story. Suppose we have a society where we have a million dollars to spend, and we're considering three possible public goods projects -- a streetlight, we'll call that Project A; a public television station, we'll call that Project B; and a park, we'll call that Project C. Now, which of these three projects does our society really want?
Now hold on, does that statement really even make sense - "projects that our society wants"? I mean, does society "want" anything? Individuals want things; individuals have clearly-defined preferences, but can we clearly define the preferences of society? This is what the impossibility theorem is all about. It says that there is no consistent way of taking individual preferences and adding them up into the preferences of society that could then be used to guide public choice.
Let me show you how it works. Now we've got three voters in our society. Here's our first guy, and let's suppose that he expresses his preferences thusly: He thinks that the park is the best use of the money, he likes the streetlight next, and he thinks the public television state is third. Our second voter has got a different set of preferences. He likes the public television station the best, the park second, and the streetlight last. And our third voter thinks that the streetlight is the most important thing, the public television station is second, and the park is third. Now these preferences are clear, right? The individual preferences are clearly stated, but can we add them up now to get the will of the people? Well, it turns out that we can't; there's no consistent rule for adding these up, and it's easy to see the problem that we get by majority vote in trying to choose which of these projects is the most desired by society.
Suppose we have an election, and all three of these projects are on the ballot at once, and we ask each voter to tell us how we should spend the million dollars. Well, this guy's going to say that the park is the priority, and he's going to vote for it; this guy will vote for the TV station; and this woman will vote for the streetlight. We're deadlocked; society has told us nothing - there is no clear will of the people. Suppose instead of putting all three of these projects on the ballot at once, we do what people usually do in an election; that is, we let people make a choice between two alternatives. That is, we have pairwise comparisons. So we take one thing off the ballot, and let people vote on the remaining two items.
Well, let's see what happens in that case. Suppose we take the streetlight off the ballot. Which of the remaining two projects would win the election? Well, this guy prefers the park, this guy prefers the TV station, and she prefers the TV station; so the TV station would be beat the park in an election. Suppose now that we take the TV station out of the picture and make a pairwise comparison between the streetlight and the park. He likes the park better, he likes the park better, and she likes the streetlight better; so the park would beat the streetlight in a pairwise election. Suppose now we take the park out of the picture and make a comparison between the streetlight and the TV station. Well, he prefers the streetlight, he prefers the TV station, and she prefers the streetlight; so the streetlight beats the TV station in a pairwise comparison.
Now hold on, this is beginning to sound like that game of scissors and paper and rock, where the rock can beat the scissors but the scissors can beat the paper and the paper beats the rock; things seem to be going around in circles. Well, that in fact is the case, and if you're having a set of pairwise comparisons - that is, you're having one election and then the winner goes up against the remaining candidates, you can wind up with any outcome simply by choosing the order of the elections carefully. Let me show you what I mean.
Suppose what you really want to wind up with is the TV station - Option B - and you're a clever politician and you're going to stage a set of elections so that people will actually vote for this TV station. The way to get this TV station in is to make sure that it's running against something that it can beat. Now you know that the TV station can beat the park; however, you know that the streetlight will also beat the TV station if it goes up against it directly. So you've got to arrange this election so that the streetlight is out of contention before the TV station enters the picture. So the first election that you want to see held is between the streetlight and the park. If the streetlight and the park are running against each other, he votes for the park, he votes for the park, she votes for the streetlight, and the streetlight is now out of the picture - people have voted for this park. Then in the second election, you run the park against the TV station, and see what happens. With the streetlight out of contention now, you've got two voters - right here and right here - who prefer the TV station, and the park loses, and the TV station is how we're going to spend public money.
By now it's clear that any of these three projects could wind up winning a series of elections depending on the order in which the elections are held, and that is a kind of weird outcome. I mean, does the society really prefer the TV station or the park or the streetlight? The answer is, we don't know, because there's no consistent clear way of adding up individual preferences to get to society's preference. There is no way for us to come up with a society preference ordering that guides us in how to spend public money, because we can use any set of rules we want to, to create this social preference function; and, depending on the rules we use, we'll wind up with a different recommendation. That's the impossibility theorem at work, and its implication is that we should be cautious in the faith that we place in majority voting systems to express the will of the people when it comes to how to provide public goods.
Now we can also use public choice theory to explain the behavior of politicians, and this leads to one more model - this is the model of the "median voter." Suppose we are in a society where people hold a variety of political views that range from those on the far left to those on the far right. And suppose, just for the sake of our example, that people's political views are distributed evenly along the political spectrum; that means any little chunk that I take out of the political spectrum here has the same number of people in it - people are uniformly distributed along this line. Now that's probably not the way things are in society; in fact, people's politics are probably distributed according to a bell curve, with the majority of people's opinions clustered toward the center. I could do this example with a bell curve, but it would be a bit more complicated visually, and give us no really useful additional information. So let's assume for the sake of simplicity that people's views are distributed evenly along the political spectrum.
Now let's see what's going to happen. People are, of course, going to vote for whichever politician is closest to their point on the political spectrum - whichever politician is closest to their particular views. So let's suppose that politician red locates himself down here at the far left end of the spectrum, and politician green locates herself down here at the far right end. Well, all of these voters down here are going to vote green, and all of these voters over here are going to vote red, and the voters in the middle are going to look both ways to see which politician is closest to their particular position.
So if we take this line and find the point that's halfway between green and red, we're going to divide the voters between the two parties. Now red is a strong left politician, and green has strong right views. But if green is pragmatic, green will realize that she can get more votes by moving closer to red, because as green moves this direction towards the left, all of these voters that are down here in the far right into the spectrum still find green the most attractive alternative. And yet now the dividing line is moving further and further to the left; that is, more voters now find that green is closer to their preferred position. So green is going to keep moving, keep moving, keep moving to the left until green is right up against red. See, now green has a big share of the market over here; lots of voters now find green closer to their preferred position, and red is stuck down here with just a little niche on the far left.
So what's red's best move going to be? Red's best move is going to be jump over green to get between his rival and the largest share of the voters. Well, two can play at that game, and green is going to jump over red, and the two are going to jockey back and forth until they find themselves practically on top of each other at the center of the market. The competition for voters leads these two politicians to press themselves towards the center.
Now this is something that we talk about all the time - politicians clustering towards the center; there's really no difference between the candidates, they wind up saying the same things. But that's not always going to be the case. What if it's true that voters in the end, if they don't find you close enough to their position, simply drop out of the market, because they're disaffected, and they won't vote for anyone at all? If it's the case that moving too far away from the extreme positions causes you to lose votes, then the voters will start to - the politicians will start to spread out, so as to keep the extreme voters in the market. If there's a chance that you could lose people on a particular issue, then you'll find yourself spreading yourselves out, moving back towards the ends, in order to bring voters back in. And this happens frequently when a third-party candidate enters the market and threatens to take away the extremes. You'll see that politicians tend to differ more on issues that people hold extreme and firm opinions about. So "hot button" issues tend to divide politicians, whereas issues that are less controversial and not make-or-break issues for voters tend to lead towards more clustering towards the center.
So the median voter model is really asking us where is the center? Where are most of the voters clustered? If the voters are clustered towards the left, we'll find both politicians clustered towards the left. If the voters are clustered towards the center, that's where the politicians will be. Studying where the politicians are clustering together, that is telling us where the median voter is. Where's the person in the middle, where's the person that the politicians are trying to get on top of. When the government provides public goods, it collects tax money from constituents and uses that money then to pay for roads, and, bridges and streetlights, and public television, and so forth.
But public choice theory alerts us to the fact that once a big pot of money has been gathered through taxes, or through any method, for that matter, people are going to begin to swarm around it, wanting a piece. That is, people who are fans of public television are going to lobby the government for more public television, and people who are fans of streetlights are going to lobby the government for more streetlights. Special interests form around public goods, as well as transfer payments. Other people will show up just simply wanting a chunk of tax money for their own particular interests. And before you know it, there are a lot of people swarming around the central government's tax collections, spending their time and driving around organizing to try to influence the machinery of government to divert more of this tax money in the direction of their special interests.
Now, this seems like good, clean fun, doesn't it? It's just another form of competition - competition for the government's largess. However, to an economist, this kind of competition is not benign; that is, it is wasteful - wasteful competition, in the form of what we call "rent seeking." Rent seeking refers to people's willingness to burn up their own or someone else's resources in an activity that is purely redistributed. That is, by showing up in Washington and lobbying congress and organizing voters and trying to get a bigger share of the pie for yourself, you're not actually creating anything - you're just kind of mud-wrestling over a limited pot of money that's been taken from other people. Nothing is being created; it's just a fight over the stuff that's already here.
So the time that people are spending in this activity is time that they could be spending doing something else, creating something like pizza, that's valuable to society. And the resources that people are burning up in this activity are resources that would have other uses, other places - he could be delivering Meals on Wheels, or helping a friend move his furniture. So rent seeking refers to a form of wasteful competition, when people burn up resources or impose opportunity costs on society, or otherwise shrink the overall pie just to try to get a bigger chunk of the smaller pie for themselves. The cost of rent seeking is that things that society would otherwise have are no longer available to us; it's just as if people were burning up resources - that is, making these things no longer available. It's as if they're simply vanishing so that people can get a bigger chunk of the tax money for themselves. And this is the cost to society of all of that activity - lost resources.
So, rent seeking is socially costly; it creates no net benefit, but it does impose resource costs on society. And public choice theory alerts us to different aspects of government involvement in the economy that are likely to provoke or promote rent seeking.
Next time you're engaged in some kind of activity, ask yourself, "Am I actually creating value here? Am I writing a song? Am I accumulating human capital in the form of additional knowledge? Am I baking bread? Am I making someone happy?" Or "Am I engaged in a rent seeking activity? Am I simply trying to get a bigger share of an already created pie by burning up my own time, my own energy, my own resources. Creative activities are ones that add to the overall value of society - productive employment of resources. And managers are always looking for ways to promote productive competition - competition that makes people work harder, or faster, or create more beautiful things. Rent seeking competition, or wasteful competition, on the other hand, actually shrinks the pie. And one of the complaints about government is that the tax pots that are collected by the government encourage rent seeking - lobbyists, litigation, and other wasteful efforts to redistribute income in the direction of special interests.
Market Failures
Public Goods
Understanding Public Choice Page [4 of 4]