Policy as a Contract

In my previous post, I made the argument that economists need to think carefully about the political process. Yes, exchange occurs in markets, but exchange also occurs in other, non-market contexts. In this post, I would like to dig in a little more and explore this idea of politics as a form of exchange.

In order to think about this, I would like to organize the discussion by starting with the Coase Theorem. What the Coase Theorem basically says is that in a world with transaction costs, the assignment of property rights matters for the outcome. (As an aside, people often incorrectly assert that the Coase Theorem says that the assignment of property rights doesn’t matter when there are no transaction costs. This is not the Coase Theorem. This is a corollary of the Coase Theorem.)

Suppose a railroad is built and runs parallel to farm land. When a train passes by a farm, it might create sparks and catch the neighboring crops on fire. This imposes a cost on the farmer(s). What happens in this sort of scenario? The Coase Theorem says that what happens will depend on the definition of property rights. Does the railroad have the right to throw sparks from the railroad tracks even it harms the adjacent land? Or does the throwing of sparks violate the farmers’ property rights to the land?

In a world of zero transaction costs, it doesn’t matter how property rights are assigned. If the value of running the trains exceeds the costs of the damages, the parties can bargain away the problem.

However, in a world of positive transaction costs, the assignment of property rights matters. If the transaction costs of negotiating are sufficiently high and the railroad’s bundle of property rights includes the ability to throw sparks, the farmers will be stuck with damaged crops or possibly be forced to abandon the adjacent land for farming. On the other hand, if the property rights belong to the farmers and the cost of operating the train plus the cost of the damages plus the transaction costs exceed the value of running the trains, then no trains will run.

Thus, when transaction costs are high, this naturally raises questions about what should be done.

One way of addressing what should be done is to focus on the legal system. How should property rights be assigned? This approach to the problem amounts to considering whether we would be better off without the train or with damaged crops. Property rights should be assigned based on this judgment.

Another way of addressing the problem is by addressing transaction costs directly. Could public policy reduce transaction costs sufficiently to allow private bargains to occur? Or, alternatively, can public policy be designed to produce the outcomes that would occur in the absence of transaction costs? For example, if the railroad’s bundle of property rights includes the right to throw sparks, policymakers could require railroads to purchase enough land adjacent to the tracks to prevent any sort of fire or could require all trains to install devices to prevent sparks.

The railroad/farm example is an externality. I’ve written about externalities before and the various roles for policy under different circumstances, focusing specifically on how many people are harmed and how many are doing the harming. However, the lessons about transaction costs and the corresponding role of public policy need not apply only to externalities.

In my last post, I argued that public policies that relate to the family look a lot like an inter-generational contract. The basic idea was that public policy can facilitate something that resembles a desirable inter-generational contract even when the family itself cannot produce the contract on its own.

When I think about public policy, I like to examine it through this lens. Why do particular policies exist? When I offer the contractual argument for family-related policies, I do not get a lot of pushback. This is probably because things like public schools, Medicare, Social Security, and laws against child abuse are quite popular. Nonetheless, I am especially interested in taking this approach with unpopular policies. If an unpopular policy exists, it might be due to flaws in the political process. However, it also could be that the policy is not well understood and a “policy as contract” approach might be helpful.

One set of unpopular policies that I have written about is maritime policy. Programs like the Maritime Security Program and the Jones Act (explicit and implicit subsidies to the merchant marine) are quite unpopular, at least in my professional circle. In fact, I once heard an economist state confidently that no economist alive could explain why these policies exist. Well, challenge accepted.

Like the family example, I see these maritime policies as a contract for mutually beneficial exchange that would not be possible without public policy. For example, consider that a navy does not need to be the same size during a war as it does during times of peace. Thus, maintaining a peacetime fleet the size of a wartime fleet would be costly, not only in terms of money, but in terms of the opportunity cost. As such, what a government would like is to maintain a peacetime navy that would be capable of expanding rapidly during times of war. In addition, the army might want to have ships available to help transport supplies long distances. So how can they accomplish this?

One potential solution would be to turn to the merchant marine. The navy and the army could use these ships and their crews to immediately increase naval capacity. It certainly seems like there are gains from trade to be had here. When a war emerges, the navy and the merchant marine could negotiate a deal with the merchant marine in which the navy gets the ships and their crews and the companies that operate these ships receive compensation and maybe the crew receives hazard pay.

Of course, the transaction costs of negotiating are likely high. For example, suppose that the country goes to war. The government goes to the merchant marine to negotiate for its equipment and services. The government might be subject to a hold-up problem. Furthermore, even if the government is able to negotiate the terms, there is a significant opportunity cost. The ships that were previously carrying goods all over the world are now being used as a naval auxiliary. Who will carry these goods in their stead? (If your answer is foreign ships, you might want to read my paper for examples of when this wasn’t the case.)

Alternatively, the government could simply seize the ships. They could announce that they are using their governmental authority to take the ships and using them as a naval auxiliary. The problem is that this would work one time, at best. If there’s an anticipation that ships will be taken by the government during times of war, private operators will under-invest in ships during times of peace.

Despite these difficulties, there are potential gains from trade. One solution would be for policymakers to create a policy in which the merchant marine received subsidies during peacetime in exchange for service during wartime. This not only resembles a contractual solution, but the subsidy would actually lead to excess capacity during peacetime such that the opportunity cost of switching these ships from commercial to military use is reduced, if not eliminated.

Like the railroad example, we can think of this in the context of externalities and the Coase Theorem. The merchant marine produces a private benefit, but also a social benefit in that its ships and crew could be used by the navy in an emergency. If transaction costs were zero, the two parties would simply bargain until they agreed to a deal. However, given that the ships are privately-owned and the transaction costs associated with negotiating are likely to be high, no such deal will occur. Thus, to facilitate the gains from trade, a policy emerges that replicates a contractual solution between the parties.

As I said in my previous post, I think we should spend more time thinking about public policy using this contractual approach.