I haven’t had much to say about the stimulus bill because (a) I’m not trained as an economist, and don’t have a strong personal view on whether stimulus will “work” or not; and (b) the debate is happening in something of a vacuum because the Republicans have not put anything remotely serious on the table as an alternative. I rather liked Greg Mankiw’s alternative proposal but, then again, I wanted to cut the payroll tax and raise consumption taxes before we went into a recession, so that just tells you more about my general policy preferences rather than anything in particular about the merits of the proposal as stimulus.
I probably shouldn’t opine much about the banking situation either because, you know, I’m still not trained as an economist. But I do have personal views here on what will and won’t work, and I’m, you know, a blogger, so you’re going to hear it whether you like it or not.
Some commenters on my last post on this topic seem to think that I was criticizing Geithner solely for lousy presentation (I was criticizing that, but not solely that), and that I was distressed by the negative market reaction (not the case). My primary concern about the Geithner “plan” is that there is no plan – it’s deliberately vague and confusing. That’s possibly because they don’t want to tell people what they are really planning to do – in which case it will weaken, not strengthen, business confidence. Or it’s possibly because they don’t know what they are going to do and are still making things up as they go along. I suspect the latter.
Why keep things vague and confusing? I think Martin Wolf focuses on the key reason: because the government does not want to conclude that there is a real solvency problem out there in the banking system; it wants to leave open the door to the possibility that this is still primarily a liquidity problem, and therefore will not require such drastic intervention.
But how could you know which is which? Well, if you knew the precise future trajectory of housing prices, GDP, unemployment, etc., you could make some good estimates about projected losses on mortgage-backed securities, corporate loans, credit cards, etc., which would tell you how solvent the banks actually are. That’s presumably what Geithner is talking about doing when he talks about performing a “stress test” – but the problem is that any such test is going to be premised on some estimate on the efficacy of government policy. And government policy, to date, has been largely reactive, and unclear about its ultimate objectives.
So here are some thoughts.
The collapse of housing prices is the original cause of both the banking solvency problems (as the first assets to go bad were mortgage-backed securities) and the consumer recession (which, in turn, is what is driving the larger economic collapse that is creating worse solvency problems for banks). But the government emphatically should not be trying to reflate the housing market as a policy response. Housing prices got seriously out of line with fundamentals – in some parts of the county (particularly southern California, the Las Vegas area, and southern Florida) insanely out of line. Overpriced housing is bad public policy across multiple dimensions – it’s bad for family formation (young families can’t afford space for kids), bad for the environment (encouraging sprawl in the search for cheaper housing), bad for overall economic productivity (because too much capital is tied up in real estate); it’s just bad.
Rather, the government needs to figure out ways to reduce the debt burden on borrowers, and thereby stabilize household finances. There are a variety of ways to do this – mandated cram-downs, transfer of title to the lender in exchange for a multi-year rental agreement, etc. The key impact of any of these kinds of large-scale interventions would be to crystallize the losses in the mortgage securitizations because a lot of loans would be restructured or turned into titles to houses that cannot be sold out from under the renter (except perhaps to the government, if the government decided to become a bottom-fishing buyer of last resort of residential housing).
That crystallization of losses is precisely what bankers have feared, because it would do a lot to answer the question of valuation. But that’s a feature, not a bug. If the value is set “too low” that just means there’s been a transfer of wealth from lenders to homeowners. Given that there’s another ongoing transfer of wealth from taxpayers to lenders as the government recapitalizes the banks, and that the homeowners are the taxpayers, this doesn’t bother me much.
At this point, you need to decide what to do about the banking sector. The big problem with nationalization is, as I’ve said before, you can’t eat just one. Nationalizing one weak firm guarantees an attack on the second-weakest firm, and a chain reaction ensues. Basically, it doesn’t make any sense for anybody to invest equity in a bank that might be nationalized. And, as the President noted, the U.S. banking sector is just way bigger than Sweden’s, and way more important to the global economy, which should give anyone favoring a wholesale nationalization of the banking sector pause.
But it also doesn’t make sense to invest public money to prop up failed institutions. If a bank is insolvent, it should fail. Bailing out equity investors sets a terrible and dangerous precedent. Quite apart from the fact that it creates moral hazard among such investors, it removes any incentive for the bankers themselves to run their banks well. Rather, the incentives run on the one hand towards looting for personal gain and deployment of capital in ways that placate political masters. (There’s also a powerful incentive to skew numbers in whatever way will get the best deal from the government.)
The government needs a triage mechanism to identify which institutions deserve an infusion of public funds on terms that will allow the banks to continue to function as private entities (i.e., not quasi-nationalization a la AIG), and which deserve to be put out of their misery. The best such mechanism would be a market mechanism: an agreement to invest alongside private investors that put new capital into the banks. But for that mechanism to work properly, the government would have to commit not to fully bail-out bond-holders of banks that fail. The perceived commitment that bondholders will never lose money has made pricing credit a game of political prognostication. The government wants insolvent banks to fail faster and solvent banks to get stronger faster – and gobble up the bits of the failed banks.
Why does the government not want banks – even insolvent banks – to fail, and bondholders to lose money? Partly because there are a lot of people who think the Lehman failure was what got us into this mess. But that’s only the case if you believe that we’re fundamentally in a liquidity crisis, and I don’t think we are. We’re in a solvency crisis, and the root of that is in the collapse of the housing market; to the extent that we also have an independent problem of market confidence, it’s not because Lehman failed (the only element of true panic that flowed from the Lehman failure was in the money markets, and that was resolved by extending government guarantees to money market funds – another area that will have to be returned to with new regulation once the market returns to something resembling normal) but because the government’s actions since Lehman have been so ad hoc; nobody with private capital knows the rules, so nobody wants to play.
But another big reason to fear, though, is that insurers own such a huge percentage of bank debt (and bank preferreds) that letting bank bondholders lose money would mean devastating the investment portfolios of the American insurance industry. That probably means there needs to be some kind of government entity created to provide additional support to the insurance industry – probably in the form of reinsurance rather than additional capital infusions, though I don’t want to suggest too strong a preference on that score. The point, though, is that we don’t want to misallocate capital in the banking sector because of a fear that properly allocating capital will hurt other sectors of the economy. We want to allocate capital properly, and intervene where necessary to preserve the viability of critical institutions.
Everything I’ve outlined above is consistent with what Geithner proposed. But lots of other policy choices would also be consistent. We need to know where we’re going – we need to make some real choices. I’m waiting for evidence that this Administration sees that.
If we can stabilize the housing market (not by propping up asset values but by crystallizing the losses and stabilizing householder finances), and if we can stabilize the banking sector (coming up with a triage mechanism to direct public and private capital to the most-viable banks and seizing and breaking up those that are insolvent without large capital infusions and cannot attract private capital), and if we can avoid a further meltdown in the insurance sector, then I think we’ll have done a great deal to provide a much needed floor under the economy, and end the free-fall that we’re in. In the medium term, the savings rate has to go up and productivity has to go up in order to pay off the excess debt we’ve incurred, and public policy should be organized around trying to facilitate those things and mitigate the social costs thereof. But in the short term, I really still believe fixing the banking sector matters more than fiscal stimulus because the banking sector is where credit comes from. The nightmare of the Japanese lost decade is a very real risk if we rely on spending money to keep demand up and don’t bite the restructuring bullet.