Public Choice Theory reveals hopelessly screwed up incentives

There are innumerable arguments about the legitimacy and morality of government and its proper scope, all of which are worth exploring.

Public Choice Theory examines an entirely different question than what government ought to do. It asks what government actually does.

Public Choice is the study of the operations of government using the analytical tools of economics. In short, the same assumptions about human interests and actions are applied to the political market as the market for economic goods.

The real power in this theory is its ability to reveal how poorly government works in practice. It does not address lofty notions of the ideal state, or the right and wrong of state action, but rather examines the actual operations of states to see if they are effective at achieving their stated goals.

Public Choice theory by that name is relatively new among schools of economic thought, but the application of economic insights to political institutions is not new.

This is a basic overview of the core insights of Public Choice Theory.

Why government?

The need for government is typically justified by the claim that there are certain “public goods” which cannot be supplied by the voluntary forces of the market, but are nonetheless beneficial to all members of society.

The standard analysis describes goods like roads, for example, and spends a great deal of time analyzing the incentives in the market to explain why roads will not be sufficiently provided. It is supposed that everyone would benefit from a road and it would be “non-excludable”. That is, once constructed, it would be too costly to prevent members of society from using the road whether they paid for it or not. Because of this non-excludability, rationally self-interested individuals would be unwilling to contribute to a voluntary fund for the maintenance of the road. Each individual knows that their contribution is small relative to the entire road fund, and furthermore that without their contribution, they will still enjoy use of the road. With these incentives facing everyone, no one will contribute and the road will decay. It is in no one’s individual interest to pay for road maintenance, but everyone would be better off if each person contributed.

You may substitute any number of “public goods” for roads, but the standard analysis is the same. It looks closely at the incentives in the market, deems them unfit to provide the good in question, and concludes that the good must be provided by the state.

Let’s grant, for the sake of argument, that the analysis of the market is correct. (There is ample evidence to the contrary of course: Nearly any “public good” you can imagine has at some point been, or is even today, provided by the private sector, even though theorists sometimes fail to imagine how.) If we accept the inability of the market to sufficiently provide the good, there’s still something missing in the standard story.

All the time is spent analyzing what would happen in a market of rationally self-interested individuals and what the incentive structure would produce. When deemed insufficient, no time is spent analyzing what happens when government attempts to provide the good. (This is what Art Carden has colorfully dubbed the political economy of the Underpants Gnomes.) What happens if we use the same assumptions and analytical rigor when examining government provision of public goods as we do for the market?

Political self-interest

In order for government to solve public goods problems it would require selfless politicians. The political actors who use tax dollars to maintain the road would have to be counted on to discharge this duty rather than, say, spend the money on something else or give it to a subpar contractor who happens to be their friend. But if we are consistent in our analysis, we must treat politicians as rationally self-interested people too. They have every incentive to act to their own benefit at the expense of the taxpayer.

How can the self-interest of the political class be kept in check? The textbook answer is democracy.

Democracy as a restraint

Does democracy ensure that the political class will pursue the interest of the public, rather than their own? The answer is a resounding ‘no’. The reason is because voters are also self-interested.

It is well known that, statistically, an individual vote in a state or national election is meaningless. The odds of one vote changing the outcome of a national election are worse than the odds of winning the lottery. The odds of getting in a car accident on the way to the polls are greater than the odds of an individual vote making a difference. In other words, the possibility of an individual vote resulting in measurable benefits to that individual is almost nonexistent.

In order for democracy to keep the self-interest of politicians in check, voters need to have an understanding of what they’re voting on and what policies are good for the whole of society. This would take a tremendous amount of time and effort. A single bill may be several hundred pages of technical legalese, and most elected officials vote on hundreds of bills in each term. For a citizen to be informed enough to know what policies are good for society vs. good only for the politicians is incredibly costly. Yet the individual vote of a citizen has almost no chance of changing the outcome or conferring any benefit. The rational response is to be ignorant of policies, because the cost of being informed is so much greater than the chance of benefiting from being informed. A dedicated, informed voter has one vote that is cancelled out by just one ignorant voter.

The result is what economists call “rational ignorance”. Voters are ignorant of policies and positions because to be otherwise is a burden with no reward. But there are some people for whom knowledge of policies is beneficial; namely, the small groups that are directly affected by those policies.

A bill to give a $100 million subsidy to Acme co. is worth a great deal to that company. They would not be foolish to spend $99 million lobbying for its passage, as they would still come out $1 million ahead. Voters, on the other hand, have no incentive to lobby against the bill because spread out across taxpayers it might cost each just a few dollars, while active opposition — even just a letter to a Congressperson — might take hours of time that could be spent doing something worth more than a few dollars. This is why democracy results in concentrated benefits and dispersed or diffused costs. The rationale of politics is to provide benefits to concentrated interests and spread the costs as far and wide as possible — including into the future by way of borrowing or inflating to pay for it.

The obvious result is a myriad of special interests seeking benefits at the expense of the broader public. In the end, everyone is worse off, but every group has the incentive to continue to seek privileges, if for no other reason than to offset the costs they are bearing for the privileges lavished on every other group. This is why Frederic Bastiat described the state as, “That great fiction by which everyone tries to live at the expense of everyone else.” It has been described elsewhere as a game where people stand in a circle and the state takes a penny from each person, then awards five pennies to one of them at the end of the round (the other five being kept by the state). The game is repeated until each person has been the “winner” of five pennies at least once. They all begrudge the loss of a single penny each round, but all eagerly expect to be the winner in another round, not realizing that at the end of the game every single person in the circle has less than they started with.

The incentives in a democratic system lead to special interests lobbying for and receiving privileges at the expense of society. Far from keeping the self-interest of politicians in check, instead democracy promotes and rewards it, so long as those politicians also provide benefits to every imaginable minority and hand the majority the bill.

But let’s ignore all that…

Let’s assume away rational ignorance on the part of voters. Let’s pretend that voters will expend every effort to become knowledgeable and constrain the self-interest of politicians. Let’s say the “will of the voters” can keep officials in check.

But what is “the will of the voters”? It is not an easy question to answer. Let’s walk through the selection of a preferred policy through the democratic process. The policy in question is what to do with troops in a particular country where the US is involved in a conflict.

Option A: Keep troop levels the same

Option B: Increase troop levels

Option C: Remove all troops

Now let’s look at the preferences of three different voters.

Voter 1: A>B>C — Prefer to keep the same level of troops, but if any change is going to occur would rather increase troops and “get the job done” than to pull out.

Voter 2: B>C>A — Prefer to increase troops to “get the job done”, but if that’s not going to happen better to pull out entirely than keep the same level of troops.

Voter 3: C>A>B — Prefer to remove all troops, but short of that, better to leave the same number of troops there than to add more.

Whether or not you agree with the preferences of these voters, it is clear that each of them has a rational sequence of preferences among the given policy options. You probably have met people who hold each of these views. To determine the “will of the voters”, let’s put these options to a vote and see what policy the elected officials should follow…

In a vote between policy A and B, policy A would win. Two of the voters prefer A to B. If we put policy B and C to a vote, policy B would win. Finally, if we put policy A and C to a vote, policy C would win.

So what is the will of the voters? According to the votes, they prefer A>B, B>C, and C>A. In practical terms, it means the “will of the voters” is to have the same level of troops instead of more, more troops instead of none, and no troops instead of the same number. That would be like a person saying that, between Snickers, Baby Ruth and Heath bars, they prefer Snickers above all, followed by Baby Ruth, followed by Heath, which they prefer to Snickers. This is a non-transitive set of preferences, and is one of the definitions of a mentally impaired person. The will of the voters is a logical impossibility. This is called vote cycling, or Arrow’s Impossibility Theorem.

You can see how, based on the structure of the voting process, entirely different preferences can emerge. This means that even if the voters were well informed, democracy would fail to provide a clear “will of the voters” for politicians to be accountable to. Take this simple example of three clear policies and substitute a number of politicians each with positions on dozens of different policies and it is utterly impossible to know what the “will of the voters” is based on the results of elections.