The US government agrees to bail out Citibank.

The federal government agreed Sunday night to rescue Citigroup
Inc. by helping to absorb potentially hundreds of billions of dollars
in losses on toxic assets on its balance sheet and injecting fresh
capital into the troubled financial giant.

The agreement marks a new phase in government efforts to stabilize U.S.
banks and securities firms. After injecting nearly $300 billion of
capital into financial institutions, federal officials now appear to be
willing to help shoulder bad assets, on a targeted basis, from specific
institutions.

I don't want to be a self-panicker; these things take time.  But I'm becoming increasingly convinced that the original theory of the bailout--that it would step in and provide a firewall to prevent future failures--has been proven wrong.  I still think it was worth trying, prospectively; there seemed to be at the time, a reasonable prospect that it would save money in the long run by forestalling the need for future bailouts.  But in hindsight, it hasn't worked. 

(I'm well aware that I am about to be slammed for supporting the bailout by a large number of people with shaky command of the concept of "uncertainty".  Hopefully, at least a few of them will pause to consider that confusing getting good results with having a good decision making process is exactly the cognitive error that brought us to this disaster in the first place.)

Where does it leave us, if the firewall has failed?  It's worth looking at what sort of seems to have worked during the Great Depression--I say "sort of" because the more histories of the era I reread, the more it becomes clear to me that economists and wonks drew far too firm conclusions from their study of the period, conclusions like "We know what happened, and therefore how to prevent it from happening again."  If you read the histories with an eye to gaping holes, there's a distressing amount of "Why the economy experienced [the 1930 consumption contraction/the second banking panic/insert your favorite adverse event here] is something of a mystery", after which the author darts, without further comment, to the part of the Great Depression he thinks he can explain.

But I digress.

What does actually seem to have worked in the Great Depression is not the second banking holiday, which was arguably something of a disaster on many levels, but what the Roosevelt administration did with the second banking panic:  sent auditors out to every bank in the country to ascertain whether they were solvent or not.

This process is more difficult now, because it seems to me that the instruments that 1930 banks were dealing with were basically understood.  There was supposed to be a steady stream of interest payments; if those payments weren't coming in, then you new the loan was in trouble.  Harder to know what the value is of a derivative where the underlying payment streams may or may not be impaired.

Nonetheless, I think it would make sense to take some of those laid-off bankers and put them to work under more experienced regulators, and try to develop some metrics for communicating the creditworthyness of the banks.  The ad-hoc nature of the response may have initially meant flexibility, but now it just means panic.

What else are we pretty sure worked?  The FDIC, which eased peoples' fears and stopped bank runs from taking down solvent banks.  Monetary stimulus seems to have worked somewhat, although less effectively than is usual. 

All else is more controversial.  Things like work relief programs may have done a good job at alleviating misery--and I find myself annoyed that palliative measures are being sold virtually exclusively as fiscal stimulus.  The stimulative virtues of extended unemployment benefits are not particularly exciting, but they are an excellent way to ensure that families hit hard by a recession do not suffer from want.  This is important--and so affordable, too!

The things we know, pretty much without a doubt, did not work?  Attempts to keep prices high through government fiat.  FDR tried just about everything you could imagine to halt the Great Depression, and several things that you couldn't, unless you were a very special sort of lunatic.  The attempts at price supports were the worst of the worst.  The National Industrial Recovery Act, which it is not too harsh to call quasi-fascist, probably takes first prize:  it involved creating gigantic cartels of industry and labor, which were intended to collude to keep prices high.  When the Supreme Court said (unanimously!) that the government wasn't allowed to do this, FDR tried to pack it with sympatico cronies like any cheap third-world dictator.  Only the intervention of some brave Democratic senators saved us from both a stunning erosion of American democratic institutions, and a barking mad price-fixing scheme that would have further prolonged the crisis.

The nation's farm subsidy scheme comes in a close second; the object was to keep commodity prices high in an era of general deflation.  Hence milk poured upon the ground and wheat fields burned when people in the cities were starving. This survived the Supreme Court to turn everyone in my mother's home town into permanent Republicans; they weren't economically sophisticated enough to understand why one should destroy food when needy people could have eaten it.

Probably not terribly harmful, but crazy expensive and awesomely lunatic, was FDR's scheme to buy gold.  Some bright boy on his staff had noticed that when the price of gold was high, the price of other commodities was usually also high.  Neatly reversing cause and effect, the government embarked upon a series of gold purchases in an attempt to make the economy grow.  They might more effectively have written to Santa and asked him for a good economy for Christmas.  Thankfully, this was so moronic that eventually even FDR's band of merry tinkerers noticed that it wasn't working, and also, that all the other central bankers were standing behind the privy, pointing and laughing.  It was rapidly abandoned, especially by the standards of the New Deal, which tended to pile program upon program until a sort of old-candlewax effect was achieved.

I don't think we're going to get anything quite that crazy.  But we'll undoubtedly get their mildly manic cousins:  minimum wage laws, stiffer unionization rules, and heavier corporate taxes, on the theory that if demand for something is falling, the best way to fix the problem is to make it more expensive.  There remain a few things that Obama's election has not changed, the tendency of demand curves to slope downward among them.  The effects of these efforts will thus be mixed at best.

We may well also get the automobile (or steel, or airline, or whatever) equivalent of the Farm Bill:  congress may cave into the notion that any industry which was once successful and lucrative, ought to be so forever.  Like the Farm Bill, it will not actually prevent the economy from changing, only make it phenomenally more expensive and less efficient.  But if that sort of thing were enough to stop our legislators . . . well, frankly, I'm not able to imagine such a world, so I can't even provide you a punchy example of what it might be like.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.