Both Brink Lindsey and Steve Teles have devoted a large part of their response essays to addressing my criticisms of aspects of their project. I am grateful for their valuable contributions to the discussion. But I remain unpersuaded on the central point in contention between us: whether limiting government power is an essential element of a strategy for combating “capture.”
Lindsey and Teles contend that the existence of nations that combine large governments with relatively low levels of capture suggests that the limitation of government power is not necessary to greatly reduce capture. Unfortunately, the evidence they cite for this proposition is not very impressive.
International comparisons on capture are extremely difficult. But looking at the nations most comparable to the United States – long-established Western democracies – it does not appear that they do significantly better than the United States at preventing capture. Many European nations with relatively large governments – most notably France, Germany, Italy, and Spain – are also notorious for harmful policies that favor special interests. For example, severe employment regulations benefit unions, current employees, and some firms, but also lead to extensive long-term unemployment, especially for the young and poor. European agriculture policy is notorious for massive special-interest subsidies that artificially inflate food prices, also disproportionately harming the poor. Some countries, including Britain, have zoning restrictions that impede mobility and raise housing prices even more than those in the United States do. And this list could easily be extended.
Lindsey cites evidence indicating that there is an inverse correlation between “capture” and government spending. But the data he is referring to actually show that there is an inverse correlation between government spending and regulation. This does not prove that having a larger government doesn’t increase the risk of capture. At most, it shows that there may be tradeoffs between two different types of government intervention: spending and regulation. Some argue that the latter is more susceptible to capture than the former. But even if this is true, it does not follow that the former is not also problematic. Moreover, it is likely that most of the correlation Lindsey points to is driven by the fact that many poor nations with dysfunctional economies have high levels of regulatory intervention (that is one of the reasons why they are poor), but also they have too impoverished a population to be able to extract more than a small fraction of GDP in tax revenue from it.
Lindsey’s argument does point to the following interesting scenario: it is theoretically possible that the government could spend a lot of money but confine the spending to a narrow range of functions. We would then have a government that is large in size but small in scope. Such a government might be less dangerous and less prone to capture than what we actually have: a government that is large in both size and scope. It might also be preferable to a government that spends less money but regulates and controls a wide range of activities. The possibility of a big-spending government with a narrow range of functions is worth greater consideration than we can give it here. But, in the meantime, we should not forget that actually existing government both spends and regulates on an incredibly wide range of issues.
Lindsey argues that his position is supported by the experience of the late 19th century Gilded Age, when there was extensive political corruption even though government was small by modern standards. It is not clear that Gilded Age corruption and capture really was so extensive by modern standards, or even by comparison with other governments at the time. The study Lindsey cites on corruption actually shows that it declined substantially during the Gilded Age, particularly during the early part of it, in the 1870s. But to the extent that capture did occur on a substantial scale, the explanation is actually consistent with my theory. During the late 19th century, state and federal governments began to take on a range of new functions, such as engaging in extensive regulation of railroads, utilities, and other large enterprises. As Yale political scientist Samuel DeCanio shows in an important recent book on the period (which I reviewed here), many of these interventions were either deliberately conceived by interest groups for the purpose of regulatory capture or at least exploited by them after the fact; he also explains how – much like the modern instances of capture documented in Lindsey and Teles’ book – the business interests and allied political elites were able to exploit voter ignorance and the complexity of regulatory policy to get away with it. This finding cuts against Edward Glaeser and Andrei Shleifer’s conjecture (relied on by Lindsey) that the shift towards regulation by bureaucratic agencies occurred because they were less vulnerable to capture than courts. Actual Gilded Age interest groups seeking to capture the political process made the exact opposite calculation; bureaucracies would be easier to capture.
It was the new, more complicated functions of government that were the ones most often captured in the Gilded Age, not the more traditional functions of preventing violence, protecting property rights, and the like. Of course government was still relatively small by modern standards. As a result, no instance of Gilded Age capture resulted in harm on anything like the scale of that caused by more extensive modern regulations, such as zoning restrictions that cut millions of people off from valuable job and housing opportunities.
Government actions against African-Americans (subject to the beginning of Jim Crow segregation) and Native Americans (many of whom were forcibly confined to reservations) did of course, inflict far greater harm than any regulatory capture. With respect to these groups, state and federal governments exercised far more comprehensive control than over almost any part of the population today. These atrocities were less the result of capture by narrow interest groups than a consequence of government doing the bidding of ignorant and prejudiced majority public opinion. The latter is often an even greater menace than the former, and provides additional reasons for keeping the powers of the state within tight bounds.
Teles contends that the dangers of capture can be reduced if the government relies on relatively simple “shove” forms of regulation rather than more complicated “nudges.” I share much of Teles’ skepticism of nudging. But it doesn’t follow that shoves are better. Whether the government relies on nudges or shoves, any effort to regulate a wide range of complicated markets and social interactions inevitably requires making difficult, nuanced tradeoffs, and addressing a variety of complicated issues. To take two of the examples in Lindsey and Teles’ book, there is no clear, simple, and transparent way to micromanage land use across a wide area (as current zoning policies do in many cities), or to set up licensing rules for hundreds of different professions. Of course, government could potentially take a crude, blunderbuss approach to these matters that ignores variations in local conditions and other subtleties. Even if that somehow reduces capture by increasing transparency, it is likely to cause even greater harm in other ways.
Both Teles and Lindsey emphasize that significant reductions in the size and scope of government, even if desirable, are simply politically infeasible, and therefore should be rejected as a “counsel of despair” (as Lindsey puts it). They are too quick to dismiss the possibility of shrinking the state. It is indeed unlikely that we can constrain government as much as libertarians would ideally prefer, certainly not in the near future. But it is possible to reduce it to much lower levels than is the case today.
Such countries as Ireland, Canada, and New Zealand have managed to achieve major reductions in spending and regulation during the last thirty years, and all three, in recent years, outscore the United States in the rankings of economic liberty published by the Cato Institute and the Heritage Foundation (the United States now ranks only 11th in the former ranking, and 17th on the latter). International comparisons suggest that the United States has plenty of room for improvement, and that we are far from having the smallest government that can possibly be achieved under modern conditions. And, of course, what has been achieved elsewhere may not represent an absolute limit either.
The claim that reducing the size and scope of government is infeasible is also at odds with much of the agenda Lindsey and Teles advocate in their excellent book, in which they recommend major reductions in zoning, licensing, some forms of financial regulation, and government protection for intellectual property. If this program were fully implemented, it would by itself qualify as a substantial reduction in the size and scope of government. If it is feasible to do all these things, we should not be so quick to dismiss the possibility that some other forms of government intervention can also be reduced, or at least decentralized to lower levels of government in order to empower foot voters. Decentralization may have considerable political potential in an era where the public increasingly trusts state and local governments more than Washington on a variety of issues.
None of this denies the obvious fact that the short-term political environment is extremely daunting for efforts to scale back government. Both major political parties are moving in the wrong direction on that score. But, of course, the same can be said for most of the procedural reforms that Lindsey and Teles advocate. There is little immediate political enthusiasm for them either, and a good deal of potential opposition from influential interest groups is likely to arise when and if these ideas actually gain traction.
Any substantial reform of the current regulatory system faces serious political obstacles. But there is no reason to think that significant limitation and decentralization of government power is less feasible than the procedural reform elements of the Lindsey-Teles program – especially when it comes to the many important points on which the two actually overlap. Moreover, the limited-government agenda has at least one notable advantage over one focused on technocratic procedural reform: it is more likely to energize and inspire activists, intellectuals, and at least some ordinary voters. It is easier to get people enthusiastic about liberty, “getting the government off our backs,” and (to put a more left-liberal slant on it), empowering the disadvantaged to control their own lives, than to rally them to the cause of improving regulatory procedure.