We all “know” that markets can’t produce public goods. The reason is that a public good is non-rival (my use doesn’t reduce the amount available to you) and non-excludable (you can’t prevent me from enjoying the public good, even if I don’t pay). The canonical (cannon-ical?) public good is national defense: What’s defended is an area; the cost doesn’t go up with population. And the air force can’t say, “Okay, you can bomb that one guy’s house because he didn’t pay. But that’s it; no other bombing, because everyone else paid up.”
There are two problems with the basic claim about the necessity of state provision. The first is that sometimes people do contribute to public goods, voluntarily or because institutions manage to arrange incentives in a way that solves the problem. The second is that the state has problems of its own and we can't compare real markets (which are imperfect) with imaginary, perfect government. I have made the argument about government failure at some length elsewhere (“The Anatomy of Government Failure”). In this essay, I’m going to consider the possibility that markets can produce public goods more often than traditional economics may recognize.
The problem is not bad design. Our systems of government are the product of long thought and efforts at improved structure over time. The idea that constant reforms and tweaking of rules is the answer is one of the central mistaken conceits that public choice theory was created to combat.
The Dynamics of Institutional Discovery
Supporters of state action argue that governments learn because government action is the consequence of intentional choices and correct motivations. Markets, to the extent that they are simply emergent processes, are not capable of learning. So the difference does not come down to perfection; pro-intervention welfare economists never believed in perfection (as I noted in an earlier piece about A.C. Pigou).
Instead, the difference comes down to the possibilities for improvement. The view that government can learn better than markets, which many would say originated at Cambridge and metastasized across the Atlantic to Harvard, M.I.T., Yale, and other universities on the East Coast, is an argument about progress.
As Roger Backhouse and Steven Medema put it (“Economists and the Analysis of Government Failure: Fallacies in the Chicago and Virginia Interpretations of Cambridge Welfare Economics”), the difference in learning capacity is important:
What emerges, then, is that the difference between the Cambridge welfare economists and their modern counterparts at Chicago and Virginia was not that the former were guilty of committing the ‘nirvana fallacy’ or that they were naive about political processes. Political processes were as central to the policy conclusions of the Cambridge welfare economists as they are to modern public choice theory and the literature on law and economics — indeed, because they did not see government as a homogenous entity, it was even more important for them to examine such processes than it is for modern economists who work with a simplified conception of government. The main difference is that because they were willing to work within the confines of rational choice theory — because, in a sense, they were willing to be more ‘neoclassical’ — Chicago and Virginia developed techniques for analysing political processes that would probably have been rejected by Sidgwick, Marshall and Pigou, even if they had been available to them.
The rational choice approach, with its assumption of stable preferences, is central here, for it effectively rules out the evolutionary view of human improvement that was central to the Cambridge vision.
To put the question more clearly: what are the source and effect of evolutionary pressures in the public sector and in the private (market) sector? To the extent that market processes must take institutions as (more or less) given while the state can change institutions rapidly, this seems to imply that states are our best hope for solving vexatious public goods problems.
Consider the economists Backhouse and Medema mention. Henry Sidgwick, following J.S. Mill, had simply asserted that markets could not handle the problem of public goods. This was taken as a prima facie case for government action: markets can’t adjust, so governments must. But there is an interesting story, told in part by R.H. Coase in his 1974 paper “The Lighthouse in Economics.” Coase argued that markets are in fact quite flexible and innovative, even nimble, when it comes to emergent notions of institutions that can solve collective problems, or what look like prisoners’ dilemmas.
We all recognize that the use of the concept of externality is much older than Pigou, but the debate over lighthouses in the 19th century brings the problem into sharp focus. John Stuart Mill, in his 1848 Principles of Political Economy, notes that “no one would build lighthouses from motives of personal interest,” because they could not collect the fees necessary to cover costs. Mill’s conclusion was therefore that “it is a proper office of government to build and maintain lighthouses … since it is impossible that the ships at sea which are benefitted … should be made to pay a toll.”
Later, Henry Sidgwick generalizes from the lighthouse to the existence of a larger set of problems in the first (1883) edition of his Principles of Political Economy. He notes that “there is a large and varied class of cases” in which voluntary private exchange would underprovide goods that produce positive externalities. In particular, Sidgwick claims that “it may easily happen that the benefits of a well-placed lighthouse must be largely enjoyed by ships on which no toll could be conveniently placed.”
One is reminded of Pigou’s description, which I quoted in the earlier AIER post, of what economists “in their studies can imagine.” In this case, Mill and Sidgwick, sitting in their studies, could not imagine a solution, and so they simply assumed that markets couldn’t solve the problem. As I noted above, an obvious problem is the failure to imagine the problems that a real government will have in coming up with a good solution. But in this case, the failure of imagination is not the inability to foresee problems of government, but rather an incapacity to imagine innovations in market institutions.
If, as Mill and Sidgwick claim, a positive externality persists without being captured by those who would benefit, then that means that a group of people are leaving a lot of money on the table. Why would they fail to organize and find a way to provide the useful service?
They didn’t fail. In fact, they succeeded. As early as 1820, in England (home of Mill and Sidgwick, who could have gotten out of their studies and gone for a walk), most — more than three-quarters — of all lighthouses had been built, and were being operated, by private individuals.
The costs were paid by user fees levied in nearby ports. The whole system operated in the near-total absence of government activity. It is true that the state enforced the collection of the private fees, but that is true of any fee-for-service arrangement, like a clothing store or chocolate shop. That is, it is true that government officials backed the enforcement of contract, and would use force if payment were not made. But Mill and Sidgwick assumed that since they could not imagine a private solution, government provision of the service itself must be necessary.
And that’s not true. Markets are at least as adaptable as government, and in many cases the capacity for innovation, especially regarding local externalities, is far greater. In fact, as Candela and Geloso point out in a Public Choice paper, the range of creative responses can be remarkable. The solution is not a pure market outcome, of course, because government action and enforcement of property rights are an indispensable part of efficient market processes. The lighthouses of coastal England in 1840 were not pure market entities, but rather a kind of hybrid.
It is quite true that a standard bilateral fee-for-service arrangement would fail to provide enough lighthouses — at first. But that failure would have consequences.
Ships need lighthouses, and entrepreneurs want profits. The “externalities mean markets fail” perspective rests on an assumption that entrepreneurs are passive and not very bright. By the time an economist has identified an externality or public good as a market failure, a group of people who stand to benefit may already have devised a private institution that solved the problem. It just never occurs to the economist to check to see what actually happened, because economists are more interested in theory, and in policy, than in actual markets and how entrepreneurs correct errors in allocations.
Further, if we assume static, inert market actors, we should not assume that government agents are active, perfectly informed, and public-spirited. People are more or less the same, according to what public choice scholars call the behavioral-symmetry assumption. If there is a difference based on sorting, you have to ask yourself which way it goes. Do impatient, innovative people go to work for large bureaucracies or private companies? It’s not plausible to think that positive innovations occur exclusively, or even primarily, at the level of the state.