This section is very interesting, though it might be less exciting than the others in this chapter. It focuses on the question of state provision of public goods and addressing negative externalities. The last section takes up forms of “practical paretianism”, some very influential today. One of the things distinctive about Jerry’s liberalism is his attitude to a state function that is widely accepted, namely, provision of public goods.
In 25.1 Jerry comments on the quotation from Green that opens the chapter. At the end of the rich passage, Green says “That, however, is the beginning, not the end, of the state. When once it has come into being, new rights arise in it and further purposes are served by it.” Green was not as adverse about the further purposes as we might be. Jerry wishes to examine what further purposes states may serve after it has assumed the role “as interpreter and protector of social morality”. Might there be a function of states as “providers of services and goods that are not morally required”? In our time, ever since the rise in the influence of economics or what used to be called political economy, the first task that comes to mind is the provision of public goods or, more colloquially, the remedying of “market failures” (i.e., externalities). With what are known technically as public goods, the benefits accrue to people independently of their contribution to their production. They thus create what is called a “collective action problem”, which often approximates an n-person PD. The two defining features of a public good are indivisibility (once produced it is available at no additional cost to anyone) and nonexcludability (it is not feasible or efficient to exclude people from enjoying the good). The thought is that states may or should step in and facilitate the production of the good, taxing people for the costs. If all goes well, the result will be mutually beneficial for all or, more colorfully, progress towards the Pareto frontier.
Jerry rightly points out that there are alternative institutions and communities capable of addressing many public goods problems. Given how dangerous state action can be — and I would add, clumsy — Jerry argues that we should seek assistance first in non-statist approaches. But he concedes that there are “times when the state and its coercive power appear to be the only viable way to cope with some problems. In these cases, the provision of public goods is, at least in principle, capable of publicly justifying state coercion” (533). He rightly points out that the costs of providing the good in question must considered when raising the question.
25.2 considers a problem with this abstract case for the state provision of public goods, namely, “that few goods are purely public”. If some Members of the Public prefer no good to particular packages of goods and costs, then no proposal for producing the good will pass. The result is constraining, “a severe restriction on the range of justifiable public policy.”
The key to understanding the restrictive implications of Jerry’s framework lies in seeing that it is not so much disagreement about the good — e.g., protection against lunatic adversaries, clean water – but in the different valuations people ascribe to the value of this good (and its costs) relative to other goods. “Again, though, to publicly justify an optional policy it is not sufficient to show that well accept these values or goods, but that there is a trade-off rate or ranking that Members of the Public endorse in their unrestricted set of values” (537).
25.3 takes up “practical Paretianism”, contemporary ways of evaluating the benefits and costs of policy proposals. The familiar Paretian principle, favoring one policy over another, is the move from one to the other makes everyone better off (strong) or makes at lease one person better and no one worse off (weak). The difficulty of using this principle — virtually all policies make someone worse — has attracted many to a principle named after Kaldor and Hicks. It counsels favoring a policy over another if the move from one to the other create enough gains so that the losers could be compensated. (If the losers are compensated, then we have a Pareto principle after all.) There are some technical worries about the Kaldor-Hicks principle — which don’t deter its devotees — but Jerry focuses on a peculiarity: “The Pareto criterion, which was based on the denial that expense of others” (540). Kaldor-Hicks is attractive to moral utilitarians or consequentialists who are not friendly to Paretian principles.
Jerry mentions a number of other kinds of Practical Paretianisms, the next being Buchanan and Tullock’s well-known supermajority rules. After a brief discussion he considers exempting some from particular policies (e.g., Sikhs from wearing motorcycle helmets). He criticizes the proposal of some, including Tom Christiano, who suppose that democratic practice is mutually beneficial over the long run. Jerry describes and analyzes a situation where over the long run all voters are worse off than they were at the outset, a case owed to Buchanan and Tullock, but really one of the first of now a very long list of situations where democratic institutions produce perverse effects. It is now customary to talk about the failures of political institutions compared to market failures.
Some reactions. Jerry’s classical liberalism is interesting as it is different from close siblings in contemporary political thought. Much of contemporary libertarianism — one thinks especially of Nozick here — is one step away from anarchism. (John Simmons’ phrase with regard to Locke was “On the Edge of Anarchy”.) Jerry thinks states of some kinds can be justified. In this section of the chapter he is sceptical about the extent to which states have the room that many now think they have in matters of public policy. Many will be unhappy with his scepticism — though I should emphasize he does not argue that there are no justified public policies. But his scepticism is different, in motivation and structure, from that of many familiar libertarians. I have no criticisms now; I need to spend more time thinking about the core of his account of public justification from which these implications flow.
Let me mention one observation about Jerry’s taxonomy of state functions. One can think of law as a public good, subject to the usual problems posed by free-riders and the difficulty of excluding beneficiaries. And state provision of law can be justified by appeal to these features. In my remarks in the last post I noted that, in principle, some forms of redistribution can be mutually beneficial. (I have called these “cooperative” forms of redistribution elsewhere.) There is a way of reading Hobbes’ and similar accounts (e.g., Jim Buchanan’s in Limits of Liberty, which Jerry cites). It’s not easy individuating the functions or tasks of states. I don’t think this will undermine anything that Jerry defends. But it does make it harder to see how to constrain states, to structure them so that they do what we decide they should do and not other things, a task that should interest everyone who believed our states should be limited.