One way the public sector can allocate resources in the presence of market imperfections is by a dictator. However, famous resource allocators like Joseph Stalin, Adolph Hitler, Pol Pot, and Idi Amin have generally given this alternative a bad name. As a result, most people seem to prefer the alternative of MAJORITY VOTING. In the next few chapters, we want to investigate whether majority voting is likely to improve upon the resource allocations produced by the private sector in the presence of market imperfections. To do that, we first need to understand how majority voting works.

City streets are widely acknowledged as being public goods.¹ Let's consider an example from the make-believe city of Dinky, North Dakota. The little township of Dinky consists of five homes at the end of a cul-de-sac. The cul-de-sac is not paved. In fact, in its current form it's a miserable excuse for a road, consisting solely of dirt and gravel. Every time the snow melts or there is a significant rain, the area turns into a sea of mud and pot holes. As a result, each of the five households has given a lot of thought to the possibility of hiring a contractor to pave the cul-de-sac. After checking around, they locate a contractor who is willing to pave their cul-de-sac for $25,000.

The table below represents how much happiness each of these families would receive if the cul-de-sac was paved.

As the table shows, each household has a different "willingness to pay" value. This is because some of the families live at the end of the cul-de-sac and have to drive over more potholes than others. In addition, some of the households own cheap compact cars which have a tough time making it through the mud and the sludge; while others own Range Rovers and Jeeps that can go through almost anything.

And now for the really important question: Should Dinky get paved? From the perspective of maximizing society's happiness, the answer is a resounding yes. The total dollar value of happiness from paving the cul-de-sac is $10,000+$8,000+$6,000+$3,000+$3,000 = $30,000. In contrast, since the cost of paving the road is $25,000, we can infer that society will lose $25,000 of happiness from having the resources used to pave the road--the laborers of the road crew, the tar and asphalt, etc.--withdrawn from other activities. The net effect will be an increase in society's happiness of $5,000.

But will the households of Dinky work together to arrange this resource transfer? Let's first consider how resources would be allocated under a PRIVATE SECTOR arrangement. This would be problematic, to say the least. While we know the willingness to pay values of the five households, this is private information. Each household only knows how much happiness it would receive. Because of this, and because the residents of Dinky are no stranger to the phenomenon of FREE RIDING, there is no telling what is going to happen. Maybe all five will agree to contribute enough money so that they road is paved. That would be great.

On the other hand, maybe some of the families will try to free ride. That is, some of the families might not really be willing to pay their "willingness to pay" values. For example, suppose the Ewings decide to free ride off of the other families. They tell their neighbors that, personally, they enjoy the mud and the potholes. On top of that, it keeps those pesky door-to-door salesmen away. They lie about the fact that they really want a paved road. And it just might work. After all, the Andersons, Bakers, Custers, and Demeters together sufficiently value paving the road. These four families might voluntarily contribute enough money to pull it off. Then again, they might not. Another family might also try to free ride. If that happened, then the whole project would fall apart, and the road would not be paved. No one household gets sufficient happiness out of the paved road to pay for it themselves. It requires coordination with other households. This points out the problem with private sector allocations of public goods. If people free ride, or if it's too costly to collect revenues from those who benefit, there will be a tendency to devote too few resources to the production of the public good. In this case, the road might not be built or a lesser road, such as a gravel road, will be chosen over the optimal solution of a paved road.

Now consider what happens when the resources are allocated under a majority voting arrangement. Suppose the town of Dinky puts a proposition to pave the road on the ballot in the next general election. In the table below, we reproduce the "willingness to pay" values from above on the "Benefits" row, but now we include two additional rows. Since the township of Dinky will arrange with a contractor to have the road paved, it will need to raise tax revenues. Let's assume that Dinky raises $5,000 in taxes from each family if the proposition to pave the cul-de-sac passes. We represent this personal tax cost to each family on the "Costs" row. The "Vote" row reports how each family will vote. If the household's personal "Benefits" are greater than their "Costs," they vote YES on the proposition. Otherwise they vote NO. Here are the results. (Note: we let each household have one vote.)

The people have spoken! The proposition to pave the cul-de-sac is passed by a 3 to 2 vote. The road shall be paved, and society's happiness shall be increased. It may seem a little unfair that the Demeters and the Ewings had to pay $5,000 in taxes when they only valued the road at $3,000. But once again, we are not concerned with fairness. We are concerned with maximizing society's happiness. In the private sector, the road might not have been paved because the community had no way of forcing everybody who received a benefit from the road to pay for it. When the government intervenes--precisely because it's the government--it can force people to pay taxes so that enough money will be collected to pay for the road paving. It is the ability of the government to coerce individuals to pay for things that they otherwise wouldn't pay for that enables the government to increase society's happiness in this case.

Will majority voting always result in public goods being provided if they increase society's happiness? The answer is no. This is easily seen by manipulating a few numbers to obtain the table below. If it happened that the Bakers valued the road paving at $10,000 (instead of $8,000) and the Custers valued it at $4,000 (instead of $6,000) we would obtain the opposite vote outcome.

In this case, the vote on the road paving proposition would be 3 to 2, against. The road would not be paved under majority voting. And that's a shame. After all, paving the cul-de-sac still would generate a net increase in society's happiness of $5,000. The residents of Dinky would still receive $30,000 of happiness ($10,000+$10,000+$4,000+$3,000+$3,000) from paving the road; while withdrawing resources which would have produced $25,000 of happiness elsewhere in the economy.