In policy debates, some see state action as the obvious solution; others say the same about civil society and/or markets. But listen closely, and you’ll often find that everyone’s gone negative: They have lots of bad things to say about the other side, and not much in favor of their own. That’s the “pretty pig” problem: We can all see the downsides—many of them quite real—with one system, and so we conclude, a bit too quickly, that the other one must be better.
As I’ve noted before, that approach leads to disagreement without engagement, as the advocates on both sides ignore the problems of their own preferred system: The state I can imagine is clearly a good solution to the real world commercial system I’m immersed in, but that state doesn’t exist, and its powers are not stable or reliable.
A sensible view of the problem is Jason Kuznicki’s 2017 book, Technology and the End of Authority. The book centers civil society, displacing what Kuznicki argues has been a misplaced focus on the state. A review of the concept of “political authority” gives very little justification for treating the state as the permanent “moral center” of society. At best, government is a historically contingent tool for solving certain problems: security, coordinating solutions to collective action problems, adjudication of disputes, and maintenance of public order.
This is not a contract theory, or a natural rights theory, though it might be reconciled with those views. The government is a technological answer to the question of how to solve the problems of living together. We’ve come to think of the state, with its conventional bundle of coercive practices, as if it were merely a natural development. But there is no reason to accept that claim empirically, and there is no good cause to give it any moral force. The reason that “government” and “state” have the meanings we associate with those words is that their functions, and those coercive practices, were chosen by human agents making compromises based on limited information. They had to do their best to create institutions that elicited information and gave workable incentives to people in the future, while working within the technological constraints of their own eras. There is no particular reason to defer to the “authority” of the government, unless government as a tool continues to serve our needs.
Technology is a key moving part in the justification for state action; it’s also a limit on the use of state power. The role of the state as a contingent tool must always be evaluated against the capacities of decentralized private action, and the parameters that govern this comparison are fluid and constantly evolving.
It’s therefore unsurprising that those who sought to control and direct the state’s coercive powers created a powerful mythology of the state’s inevitability and necessity. But the state is a bundle of coercive powers handed down from the past, and the contents of the bundle are not inevitable. It’s possible the whole thing may not even be necessary. The western philosophical tradition exaggerated the moral centrality of the state, as a means of reducing the costs of compliance and enforcement; fair enough. But modern technology increasingly forces a re-evaluation, and the once-useful mythos of necessity is now an obstacle that must be dealt with.
Which brings up the main point of this essay: the usual problem of private rights involves a distinction, and an enforcement, between what is yours, and what is mine. The problem modern society now faces is a different distinction: what is mine, and what is ours? That is, what kinds of rules, activities, and direction of resources will be done by decentralized groups, and what functions will be carried out by the state?
Who Will Build the Roads?
You might have heard the question in market-liberal circles: In a stateless utopia, who will build the roads? That question is a great example for my purposes, because roads are important and tangible. And yet the question is so often discussed on a high plane of abstraction—and that makes it harder to see how physical and social technologies interact in the real world.
One important motivation for the founding and spread of public choice economics is to correct for the excesses of the market failure justification for state action. The logic of that justification is that there is literally no way for decentralized private action to provide public goods, or to correct for externalities. Public choice economics challenges that hasty conclusion. And when it considers a question like “Who will build the roads?,” it’s informed by empirical evidence.
“Public goods,” in particular, are defined as those useful goods and services that markets cannot provide, because the item is non-rival—it doesn’t get used up when others use it—and non-excludable—it’s difficult or expensive to charge the average cost price, and the marginal cost of a new consumer is zero.
The term “public goods” doesn’t refer to the public good—that is, all benefits to the general population—it’s about which things meet the technical conditions of being non-rival in production and non-excludable in consumption. The public good is often served by using private commerce, by saying “this is mine,” not “this is ours.” Think of grocery stores, retailers of electronics, restaurants, movie theaters, coffee shops, and many other spaces. All of the thousands of private sources of public good need to be insulated from state direction and control, because no single actor could know enough to act correctly, and they would have no incentive to act correctly if they did. Private action, both in commerce and in voluntary private associations, serves the public good.
But there are still public goods that the state should provide. Are roads among them?
Having a road network is certainly good for the public. But to be a public good in the economic sense, roads would have to be immune from congestion, or the effects of increased use, and they’d have to be very expensive to provide on a fee-contingent basis.
The congestion problem for roads has recently been much commented on; the implementation of “congestion pricing” in New York City and other areas has resulted in substantial improvements in drive times. On the other hand, for many rural roads there is little or no congestion, unless there is an accident, and the “non-rival” condition is usually met.
For many years, the real difficulty was that charging for use of the road was expensive. It took human toll booth operators, and the long lines at toll booths were a deadweight loss—a simple inefficiency, with no upside anywhere. But digital tools and electronic sensors have completely changed the cost calculus, to the point where roads are obviously excludable.
I have a North Carolina “Quick Pass” transponder attached to the inside of my windshield, behind the mirror. It lets me drive on (almost) any toll road in twenty northeast and southern states, and tolls are collected instantly and seamlessly from my pre-arranged account. It’s easier to pay tolls than to pay for a movie ticket or groceries.
That seems like a pedestrian (sorry) point, but it is exactly the sort of thing that Kuznicki was talking about. Technological changes are moving the boundaries of what the state must, or ought, to do. Voluntary associations, what public choice founder James Buchanan called “clubs,” are now able to accomplish collective goals that once required state action.
Perhaps we shouldn’t be surprised. Private, collective, voluntary action has long supplied public goods in northern Europe. Far from being socialist, Scandinavian countries have for centuries used voluntary, decentralized, private organizations to provide “public” goods, especially roads. A defined group of users pays dues, sets rules, and often accepts state subsidies in exchange for granting broader access.
Roughly two-thirds of Swedish roads (by length, not traffic) are managed by private road associations. There are about 60,000 associations responsible for more than 130,000 kilometers of roads. Of course, that means that the average association manages about 2 kilometers of road, and the median is less than that: some are simply roads that branch off a main road and serve 50-80 property owners. (That number can vary widely and is larger in more urban areas).
About 24,000 of those associations receive public subsidies, but the subsidy comes with conditions: if they take the money, they must keep the road open to general vehicle traffic. Some of these associations even handle ferry services in the Stockholm archipelago. That’s not socialism, but neither is it standard capitalism; it is decentralized, voluntary, user-based provision inside a legal framework overseen by the state.
In Finland also, most of the roads, as measured by length, were built on the same model. Finland has about 370,000 kilometers of private roads linking roughly 30,000 businesses, 40,000 farms, 250,000 homes, and 190,000 holiday homes to the public network. These roads are financed and managed by owners through road maintenance associations.
In Norway, roads are financed primarily through direct user charges via AutoPASS; tolling now spans hundreds of toll stations across dozens of transport projects. That’s not a membership club in the Swedish-Finnish sense, but it follows the same basic logic of excludable, user-financed infrastructure, rather than fully socialized provision. Norway is perhaps the clearest example why roads are no longer “public goods”: technology has changed the conditions under which collective activities can be financed and provided.
Wanted: a Theory of the State
Infrastructure finance shows how the conditions that justify full state provision are actually quite strict, and changes in technology—in many activities, not just roads—mean that the question of what the state must do, and for that matter what it should do, is always pressing. We can’t have a useful answer to those questions, however, unless we have some comprehensive vision of what a state is for in the liberal society.
As noted at the outset, the usual answer is that the state should serve the public good. The popular conception of the public good would make this charge too vague, and overly broad. The standard economics answer is that the state should produce public goods, that is, things that have collective benefits, but which cannot be provided properly by private action. But that’s not really a theory of the state, by itself.
Law is organized force; the state exists to enforce the law and adjudicate disputes over its application. That means that the state cannot legitimately extend its direction and control beyond the proper scope of the use of force. The justification for the use of force must be, and can only be, the incapacity of voluntary private action to supply the lack or need.
In short, the state as an instrument, not a moral end in itself. But that means that the proper role of the state is historically and technologically contingent, and it should constantly be reexamined and questioned. Problems that once required state action may now be solvable in other ways. Of course, the reverse is also true: problems that once did not require state action may deteriorate or devolve into concerns that private action cannot correct. Naïve claims that the state should always get bigger (or, for that matter, smaller) are misleading. Liberals need a theory of the state to understand how the state should be reformed and directed.
Claims that “the state should do that” must be evaluated fairly, not simply dismissed on ideological grounds. But if someone wants to argue that a function is truly essential to the state, the burden is on them to show that the problem really is best or only solvable by state coercion. Applying that standard, if we can do it seriously and consistently, will give us a greater intellectual credibility and a much smaller state.
