On Will Wilkinson On Detroit

by hilzoy

My co-blogger publius wrote a post on this piece by Will Wilkinson, on bailing out Detroit, which publius finds callous. But what bothers me about what Wilkinson wrote is not exactly its callousness so much as the sense I have that he is arguing with people who do not exist. Specifically:

“There is nothing that helps people more than high rates of economic growth, compounding, compounding. But everyone is not helped equally. Economic growth requires dynamism, requires “creative destruction,” and some people get trapped in the wreckage, become wreckage. Not everyone is hurt equally. That irks. We should do what we can to limit downside risk consistent with the goal of producing broad prosperity. And we should feel a pang for those whose expectations are disappointed, whose lives turn out harder than they’d hoped. But the impulse to freeze the system, to try to tape all the cracks and staple all the cleavages, to ensure that nobody has to explain to their kid why Christmas this year is going to be a lousy Christmas, that is one of our greatest dangers. Our sympathy, untutored by a grasp of the larger scheme, can perversely make itself ever more necessary. When we feel compelled to act on our uncoached fellow-feeling, next year’s Christmas is likely to turn a bit worse for everybody. And then somebody has to explain to the kids that they can’t find a job at all. Businesses that would get started don’t get started, wealth that would be created isn’t. And in just a few decades, the prevailing standard of living is much, much lower than it could have been had our sympathy been more far-seeing. There is no justice, and great harm, in diminishing the whole array of future opportunity to save a few people now from a regrettable fate.”

This is an argument about why, in general, market economies are better than planned economies, or why, in general, the US economy does better for its people than the Soviet economy did. There was a time when it was worth having this argument. That time was the late 70s and the 80s. During the 1970s, a Republican President instituted wage and price controls to try to stop inflation. Until Jimmy Carter’s administration deregulated the airline industry, it was so tightly regulated that most discounts were illegal. Large chunks of the British economy were nationalized.

Ronald Reagan and Margaret Thatcher were, therefore, arguing against real people who really did favor massive regulation of the economy, regulation of a sort that would now be unthinkable in the US. But who now favors wage and price controls, or the re-regulation of the airlines? No one that I know. Many of us were in favor of market economics, in general, before the fall of the USSR; most of those who were not were convinced by the discovery of just how badly that economy had failed its people. We do not need to be told that a centrally planned economy is a bad idea.

I’m sure that there are some devotees of central planning out there somewhere. But for the most part, those of us who reject free market fundamentalism do so not because we fail to recognize that market economies are generally best, but because we think that this, like most general principles, has exceptions.

There are market failures of various kinds: for instance, externalities that are not captured by normal market mechanisms. There are ineliminable agency problems. There are collective action problems, in which we require government intervention to achieve some outcome that is preferable to everyone, but not individually rational. There are functions that, for various reasons, we do not think should be privatized — for instance, the military.

Moreover, there are cases in which we need to regulate something because while a market mechanism does exist for solving some problem, we find it unacceptable. For instance, there is a market mechanism for dealing with the problem of melamine contamination of milk in China: some kids die, people stop buying milk, eventually, I imagine, someone will come up with a way of letting people distinguish between tainted and untainted milk; in the meantime, creative destruction will sweep up the just and the unjust alike. But we might prefer to prohibit adding melamine to milk, and to inspect dairy facilities, on the grounds that this is better than a market mechanism involving dead kids. (Note: this does not imply that a market mechanism that predictably leads to dead kids is never preferable. The alternatives might be worse, e.g. if they involved more dead kids. The point is just that in this specific case, in which we have a choice between inspecting dairy facilities or testing milk for melamine on the one hand, and dead kids on the other, we might rightly choose the former.)

I’m a pragmatist about these things. Like Matt Yglesias, I think that there are a number of licensing requirements that we might usefully get rid of. I also think it’s always worth asking whether any government solution would be worse than the problem it purports to solve. On the other hand, I think that sometimes, as in the melamine case, the answer to that question is ‘no’.

That said, I find myself arguing in favor of more regulation more often than not. This is not because I favor it in general, though; it’s because, as I said, the playing field has shifted since the ’70s and ’80s. Back then, there were arguments between people who were generally for regulation and people who were generally opposed to it, and I normally found myself in the middle, asking for details of particular cases. Now, however, the pro-regulation camp has more or less vanished, leaving mostly people (like me) who think these questions should be decided on a case-by-case basis, and that there is no general reason for favoring regulation over its absence, arguing against free market fundamentalists who often write as though all regulation were presumptively bad. (I assume they don’t actually believe this — e.g., that most of them do not favor abolishing all food safety requirements — but they sometimes write as though they do. I assume it’s just a habit that conservatism got into back when such arguments needed to be made. But it’s a bad habit nonetheless.)

In Wilkinson’s piece, this bad habit shows up as an assumption that people who support the bailout need a lecture on the general virtues of the free market, and on why the fact that some business goes under is not, in general, an argument that the state should have intervened. It shows up more specifically here:

“If employees of the Big Three deserve to have taxpayers pay part of their relatively lavish salaries, then employees at thousands of failing businesses deserved the same. They had no chance of getting it, though, simply because they don’t have the right history with Washington. There is no other reason.

Really? No other reason? Offhand, I can think of several other reasons, most of which are included in the Jon Cohn article publius cited. (Ezra Klein posted a shorter version here.) I have no idea whether Cohn’s arguments are right, though he’s generally very trustworthy. However, he claims that the Big Three have reformed a lot of the management practices that got them into trouble, that a lot of those reforms involve serious cash up front, and that this leaves them without the deep reserves they’d need to weather this crisis. If so, the case for intervention is a lot stronger, and a bailout looks less like throwing money into a bottomless pit and more like solving a temporary, albeit severe, liquidity problem.

He also argues that for reasons specific to this case, the Big Three would be forced into Chapter Seven bankruptcy rather than Chapter Eleven; that is, into forced liquidation rather than reorganization. As Cohn notes, in this case “the company’s institutional knowledge would end up on the proverbial shop floor.” Moreover, the economic dislocation would be huge:

“Letting GM fail would have potentially catastrophic effects on the economy. It probably couldn’t reorganize under Chapter 11, so it’d have to liquidate. That would take down most of the suppliers, potentially sinking the rest of Detroit with it. Realistic estimates of new unemployment range from 1 to 3 million, once you include the ripple effects in communities with shuttered suppliers, dealers, etc.”

The suffering this would produce is not the suffering of one unemployed person multiplied by 1-3 million. It’s an absolutely enormous dislocation, the kind that sinks entire regions. Moreover, while I have precisely no ability to assess this claim, some people argue that letting GM fail would cost the government more than bailing it out. This would alter the cost-benefit calculus considerably.

In general, if we do bail out the automakers, we plainly ought to do so in a way that penalizes its shareholders and management. (Avoiding moral hazard matters.) Moreover, if there are any tough choices that they have been resisting, we should insist that they make them. We should not bail them out if that just means keeping zombie companies alive. But if, in this case, it makes sense to do so, given the enormity of the dislocation, the losses (economic and the loss of institutional knowledge) that would follow from the destruction of these companies, the costs of not bailing them out, etc., then we ought to bail them out in some way that causes the people responsible for their predicament sufficient pain to avoid moral hazard.


To me, what makes Wilkinson’s piece callous is not that he tells us that there might be benefits that outweigh the costs of letting the automakers fail. Obviously there could. It’s that he did not bother to argue the specifics: to explain why, in this case, the costs of a bailout really do outweigh the benefits. Instead, he took the invocation of a piece of economic dogma as sufficient to make his case.

It would be callous were I to react to the Chinese melamine contamination by saying: we can’t have government inspections of milk producers; that’s government interference, and as the experience of Stalinist Russia shows us, government interference is always counterproductive. If I take those dead kids seriously, I ought to ask myself: is it really obvious that just having people test milk for melamine will lead to the institution of gulags, the killing of kulaks, and so forth? If not, don’t I need some more fine-grained argument to deal with this case?

Likewise here: it is wrong to suppose that just invoking the general virtues of free markets is sufficient to deal with this case. The dislocations caused by any of the Big Three going under would be immense. It is different from the bankruptcy of individuals or small businesses, precisely because of its scale (and because there might be specific reasons, in this case, to think that the remedies available in the normal cases are not available in this one.) It’s not at all obvious that there is no way to prevent this dislocation that’s consistent with avoiding moral hazard. One might therefore think that some more specific argument might be required; that using a one-size-fits-all argument, in this case, is enough.

At least, one might think this if one took the costs of letting the automakers go under seriously, or even if one had noticed that while virtually no one questions the general point Wilkinson makes, a lot of people question his conclusion. That Wilkinson does not see the need for anything more than an invocation of the general virtues of free markets, and an account of why we should not in general intervene when firms go under, shows that he is not thinking seriously about this case. And that, to me, is the real problem with what he wrote.