David R. Henderson  

Mark Thoma Doesn't Get It

Where is the Future?... This Mimic is Different...

Government as Deux Ex Machina

Almost all economists recognize that there are some market failures that must be corrected by government intervention, the disagreement is over their prevalence. Some economists see widespread and costly market failures, and that government can intervene effectively to overcome them.

This is from Mark Thoma.

Notice a category missing? The category includes me. It's economists who do "see widespread and costly market failures" but don't see how "government can intervene effectively to overcome them." The debate between advocates of government intervention like Mark and critics like me is still unjoined--by Mark. What he has not shown us and, more upsetting, what so few advocates of government intervention even try to show us, is how a government regulator will have the right incentive to do the right thing. Will the government regulator be fired if he screws up? Not typically. Will he get a huge bonus if he does something right? Not typically. And how, with a centralized information system, will he get the information needed to make a good decision, something I wrote about in the context of dealing with terrorism? I don't read Thoma as much as Arnold does, but I read him a fair bit and I've never seen him deal with these issues. Have you?

Comments and Sharing

CATEGORIES: Regulation

COMMENTS (15 to date)
Joel writes:

I see this false dichotomy everywhere. I find it most upsetting when I read it in economics textbooks.

Lord writes:

Krugman has,

david (not henderson) writes:

I believe a large part of the problem is that most of us schooled in mainstream economics have blindly accepted the notion of "widespread and costly market failures".

"Costly" or "failure" by comparison with what outcomes achieved by what means? If the market outcome cannot be realistically improved upon, then it is by definition optimal and there are no costs or failure.

William writes:

I hate to impose overly exclusive rules, but in my opinion, someone who talks about market failure without mentioning Coase or Demsetz is not worth serious consideration.

As david (not henderson) points out, you have to be talking about and comparing actually-existing alternatives. It's useless to compare the world as it exists with some world that you make up.

To quote Coase's "The Problem of Social Cost":

"Actually very little analysis is required to show that an ideal world is better than a state of laissez faire, unless the definitions of a state of laissez faire and an ideal world happen to be the same. But the whole discussion is largely irrelevant for questions of economic policy since whatever we may have in mind as our ideal world, it is clear that we have not yet discovered how to get to it from where we are."

Bob writes:


Those are just unsupported claims. Sure, you can cherry pick past instances where regulation seems to have worked well, just like there are examples of deregulation working out well.

Because regulation creates rigid, durable rules - which has both benefits and costs - it seems like the burden of proof should fall on those who wish to regulate. I conceed the market failure on pollution, and that regulation can provide a net benefit. But even pollution is messy (pun intended) as those who wish to regulate expand the meaning (e.g., the EPA moving to regulate carbon).

mobile writes:

To be fair to Mark Thoma, I found the missing category after reading another, oh, two sentences past the excerpt.

Almost all economists recognize that there are some market failures that must be corrected by government intervention, the disagreement is over their prevalence. Some economists see widespread and costly market failures, and that government can intervene effectively to overcome them. Thus, an active, interventionist government is required to ensure that markets are functioning correctly. More libertarian types tend to both see fewer market failures and, more importantly, believe that government is not very effective in intervening to correct problems. It's only very large, very obvious cases where government can help, and those are far and few between. Some never see them at all.
Justin writes:


Krugman argues that libertarianism fails because politicians bestow favors on private business under the status quo. Isn’t this basically the same as saying libertarianism fails because our government isn’t libertarian?

He could have done this on any issue. “Libertarians oppose out of control military spending and reckless foreign interventions. However, in the real world politicians fund wasteful military spending to keep jobs in their districts and then vote to attack Vietnam, Iraq, Afghanistan, etc. And don’t say that we just need better politicians. If libertarianism requires incorruptible politicians to work, it’s not serious.”

David N. Welton writes:

Mark Thoma responds, rather sharply, demonstrating very much that he does "get it", although he may not always agree with "it":


I think he has something of a point about the ideology on display. As someone who isn't that knowledgeable about economics, one of the more interesting ways of "cross-checking" various authors is observing where they go against the political grain. For instance, some of Krugman's older writing, or Republican Greg Mankiw favoring a carbon tax.

If someone always favors no government, or (at the other extreme) government ownership/intervention in everything, it seems that one's reading time might be better dedicated to more nuanced commentators who can be counted on to vary their responses with the situation.

Hyena writes:

The dichotomy is proposed because a market failure means that there is an incentive issue within the scope of enterprises seeking financial profits.

The obvious recourse against this is deference to another forum where actors seek a different set of profits (prestige of various sorts) and so face a different set of incentives.

Ted writes:

I think you have a slight literacy problem. He said "almost all economists." Almost all is a factually correct statement, you simply aren't one of those "almost all." Almost all economists see market failures where the government can improve outcomes. More often than not they think some version of taxes and subsidies can do the trick, but many see a role for regulation as well.

For me, I think there are obvious cases where taxes and some regulation can improve social welfare. For example, a carbon tax could improve the environment. Or, in the area of regulation, there are situations where it's impossible to compensate for damages ex-post and tort law is unable to resolve the situation effectively. Those situations I would deem is appropriate for some regulation, although I'm skeptical if its execution would be done well.

More often than not though, I find the government will probably screw up any attempt to make an improvement so more often than not the government should probably leave the market failure unless it's so significant it can't be ignored.

By the way, there are actually good theoretical grounds to believe the government will screw up regulation - and it's buried in the same papers people use to argue for regulation and intervention. The seminal paper on externalities is Greenwald and Stiglitz's 1986 paper "Externalities in Economies with Imperfect Information and Incomplete Markets." Basically they show that there almost always exists some tax instrument that can make a Pareto improvement in almost any case of market failure - be it moral hazard, adverse selection, incomplete markets etc. Often economists are implicitly using this paper as their framework when they recommend tax instruments (or any intervention, like certain regulations, that would mimic the outcome of a tax policy). There is a slight problem though. The model relies on a unique assumption that people just seem to assume is true without ever questioning it. It assumes that the government faces the same information constraints as the private sector. I don't buy that for most cases the government has identical information. More often that not, I think they have significantly worse information and so the Greenwald-Stiglitz theorem breaks down immediately. It's shocking to me economists don't realize how key that assumption is and how unrealistic it is most of the time.

The_Orlonater writes:

"For example, a carbon tax could improve the environment. "

No it won't.

Back to the topic of the blog posts, I also think Gordon Tullock made a really good argument in the later chapters of his great book The Politics of Bureaucracy against how bureaucracies run.

The_Orlonater writes:

I also haven't noticed many of these advocates of government action "correcting" market failures try to address as to how they plan to do it and how to do it effectively. The only possible manner in trying to employ such "correction" through their logic is to use bureaucratic methods. The debate should be boiled down to this, it is grossly overlooked and ignored.

Alan Crowe writes:

Mark Thoma says: "More libertarian types ... believe that government is not very effective in intervening to correct problems." This strikes me as a clear enough example of not getting it.

Markets fail. Governments intervene. Government intervention may ameliorate market failure or exacerbate it. For example one, agriculatural subsidies may increase production of unhealthy foods that market mechanism already over supply. For example two, government intervention in the market for recreational intoxicants deters the substitution of vodka by cannabis.

The rhetoric of government intervention is invariably that of fixing market failure. Since this rhetoric is used equally to justify amelioration and exacerbation it does not itself indicate the likely outcome. Government has incentives that may, or may not, include living up to its rhetoric, and these incentives determine the outcome.

The goal of economics is a naturalistics and mechanistic account of society. Business has its rhetoric of serving customers and we can understand the mechanism behind this: a poorly served customer can go elsewhere. Or can he? We see the incentives for buying up rival businesses or for enlisting government to create burdens that fall more heavily on rivals. We also see that these tactics are not part of the rhetoric of business. Government has its rhetoric of correcting market failure. Politicians have other means of retaining power, such a gerrymandering boundaries and favouring organised voting blocks.

Economics must apply the same kind of mechanistic analysis to both business and government. It cannot a pious science which assumes that the government are the good guys. It is supposed to be an empirical science; it could, perhaps, discover that the government are the good guys.

Mark Thoma's erroneous piety makes him say "government is not very effective in intervening to correct problems". Government isn't intervening to correct market failure. Government is intervening to pursue its own agenda. Empirically we notice that pursuit of its own agenda involves using the rhetoric of "correcting market failure". Does it involve, as Thoma assumes, trying to correct market failure? If it does, then there is a story to tell, much like our story of businesses serving their customers, and it will be an illuminating story, that points out the things to watch out for, that government might be getting up to instead of doing what it says.

fundamentalist writes:

The whole "market failure" debate is dishonestly slanted against markets by definition. Look up what people mean by market failure. They generally mean that the market won't do well at preventing pollution, crime, and foreign invasions. This is a straw man. No free market economist in history has ever argued for no government at all until Rothbard, and he still argued for the rule of law with courts and private police. No one has ever expected the markets to fulfill the role of the state. So claiming that markets would fail at doing the state's work is just stupid. Yet that is the definition of market failure most people work with. No, markets don't do well at preventing pollution, crime or foreign invasions. Even Rothbardian libertarians rely on the courts to handle pollution and crime. But those are not market failures.

No one in history has ever suggested that crime, pollution and foreign invasion should be the work of markets. Pollution is a private property issue and the state has always had the job of protecting private property even among the most anarchic capitalists. But by falsely defining those as market failures, socialists think they open the door for greater state intervention in the economy.

Pandaemoni writes:

@ fundamentalist:

Actually, when I hear "market failure" I simply think of non-Pareto optimal market equilibria. It's not limited to the failure to provide secific goods to society (though there would be a market failure in those cases too).

Comments for this entry have been closed
Return to top