« How bad news can be good news | Main | Sherry Glied's new health care paper »
The economics of "bailouts"
Paul Krugman writes:
...(according to Reinhart and Rogoff) the resolution of Sweden’s financial crisis imposed a fiscal burden — that is, required a taxpayer-financed bailout — equal to 6 percent of GDP. That would be $850 billion in America today. Just saying.
It's worth noting that such costs consist mostly of transfers rather than real resource costs. Most of the costs of overinvestment in housing already have been borne in the form of lower living standards, namely we have fewer non-housing goods and services. Making debt obligations whole again does involve higher taxes but most of the money is sloshed around; the government doesn't dynamite any factories or homes. It should bother you if you think taxes are already too high but of course that doesn't describe everyone. Furthermore if the destruction of the debt claim would otherwise have been deflationary, some of that debt can be monetized (thus, taxes don't go up) without raising the risk of inflation. (TC: the Swedish number seems to be wrong, see the first comment.)
Here are a few other points about bailouts, or non-bailouts, as the case may be:
1. Most plans for Fed assistance aren't bailouts at all. It is pretty easy for the Fed or Treasury to virtually wipe out shareholders. The real "bailouts" come when the institutions are allowed to stay open and continue taking risks.
2. The Fed's regulatory powers make crisis deals less than fair. If you, as a bank, don't accept the Fed's terms, you can be prosecuted or thrown in jail or at least ruined by your friendly regulator. Being an advocate of the rule of law, I'm not entirely comfortable with this arrangement, but it does mean that the Fed has a much easier time managing crises. Keep in mind also that the failing banks are indeed the most likely ones to have been criminal, so the unfairness is not usually being applied to the innocent.
3. If you think the managers were in charge, and will remain in charge, the real moral hazard problem is the severance pay for the failed managers, not the so-called bailouts.
4. If you're a critic of bailouts, you can't have it both ways. If the Fed or Treasury is guaranteeing loans, yes that does put taxpayer dollars on the line. But if you think the system can hold up, as do most bailout critics, those guarantees are unlikely to cost very much. The Fed or Treasury may even turn a profit. If you think the system cannot hold up, the bailout is probably necessary even if costly. So you can't claim: "The bailout isn't needed" and also "The bailout will burden taxpayers."
Addendum: By the way, do read David Leonhardt on "what really happened."
Posted by Tyler Cowen on March 19, 2008 at 06:24 AM in Economics | Permalink
Comments
The figure of 6% of GDP for the bailout of the Swedish banks is incorrect. There is no reference to it in the Reinhart & Rogoff paper. It was estimated that the worst case cost at the time could end up at about SEK 100 bln, which was roughly 6% of GDP. In the actual case, the costs were much, much lower. Only Gotabanken failed, and Nordbanken (subsequently Nordea) was nationalised. It recovered quickly and has since been partly privatized. In the end, it is probably that the Swedish government will make more from selling the outstanding shares of Nordea than they paid for the whole mess in the first place.
If so, that would then put the cost at 0% of GDP. (This doesn't mean that governments should do stuff like this at will - the government took on a huge risk. It's just that in this particular instance, it didn't happen to cost the taxpayers anything.)
Posted by: Joe Torben at Mar 19, 2008 9:07:33 AM
2 objections, 1 strenuous:
1. "Keep in mind also that the failing banks are indeed the most likely ones to have been criminal" may be trivially true but does not at all imply that "the unfairness is not usually being applied to the innocent." There is no credible assertion that Bear Stearns was brought down by criminal acts, nor are they present in the vast majority of bank failures. A belief that finance is endemically criminal or that adverse results in leveraged lending are probative of criminal activity is not helpful. Contrast Krugman on the current situation vs Enron.
2. Bailouts can be nominally costly even when the system holds up. To the extent the Fed's given JP a par put on the worst $30B of Bear's assets, it's easy to see how most of that can be burned through even in a reasonably supportive credit environment. Now $30B is terrifically cheap vs. the scenario where Bear pulls Lehman and MS over the cliff, who in turn pull Goldman and Merrill, then Citi and JPM, and so on. But it's hard to assert it will be cheap relative to the $30B commitment, much less profitable. Also, I do think this is the right structure for a bailout - as similar to the structures used by the FDIC in the late 80s - early 90s bailouts, which were much less costly than the RTC liquidation strategy.
Posted by: misplaced trust co. at Mar 19, 2008 9:16:42 AM
"If you're a critic of bailouts, you can't have it both ways. If the Fed or Treasury is guaranteeing loans, yes that does put taxpayer dollars on the line. But if you think the system can hold up, as do most bailout critics, those guarantees are unlikely to cost very much. The Fed or Treasury may even turn a profit. If you think the system cannot hold up, the bailout is probably necessary even if costly."
There are a certain set of facts which allows you to have it both ways. If Bear Stearns turns out to be insolvent, and "the system" could withstand a standard court-approved liquidation of its assets, then this is a bailout, of Bear Stearns' creditors, and it is unnecessary.
The first half of that hypothetical is something we will likely find out eventually. The second half is something I personally don't have much of a handle on, so any arguments you can provide or point to in that regard would be much more useful than just asserting it as a fact. But even if "the system" would have failed in the absence of *any* government intervention, could the Fed have targeted its intervention to those victims more innocent than Bear's creditors?
Posted by: Anthony at Mar 19, 2008 9:58:38 AM
The Fed gave JP Morgan a $30 billion non-recourse loan against Bear Stearns' MBS assets. If the Fed supposed those assets were temporarily illiquid (that is, worth more in the long run than at present), why should the loan be non-recourse? And if the Fed believed those assets really were garbage, why should the Fed effectively buy them?
The whole thing stinks. JP Morgan will take any profits, and the Fed (that is, middle-class taxpayers) will take all losses. I don't see prudent regulatory action, I see rent-seeking and back-scratching.
As for the moral hazard question, it's true that Bear Stearns' shareholders got wiped, but remember, the directors and managers are the ones susceptible to moral hazards-- the public shareholders have no control over the business and in most cases don't know a thing about how it is run. And what about the MBS originators?
Posted by: Vercingetorix at Mar 19, 2008 10:37:42 AM
The first comment is wrong: the Reinhart Rogoff paper does reference the number, see page 4, footnote 1.
Posted by: Isaac at Mar 19, 2008 11:29:18 AM
It's worth noting that such costs consist mostly of transfers rather than real resource costs.
Sure - no problem. You have some money in the bank prepared for a nice future, retirement etc. Now either the bank goes bankrupt, or you are taxed on capital so that the government can bail it out or the inflation just eats your money. That's the cost of 3 possible scenarios. It shouldn't bother you, right...?
Furthermore if the destruction of the debt claim would otherwise have been deflationary, some of that debt can be monetized (thus, taxes don't go up) without raising the risk of inflation.
No, you don't get it. Either the banks go bankrupt which IS deflationary, only that half of the people don't get their money, so they couldn't care less if there is deflation or not. Or, the debt will be monetized, in which case you will get inflation. It's that there is too much money for too little goods - either you get less money (in which case you will see stable prices of food and falling prices of houses) or you will monetize it (which will mean stable prices for houses and skyrocketing prices of food, energy etc.).
4. If you're a critic of bailouts, you can't have it both ways. If the Fed or Treasury is guaranteeing loans, yes that does put taxpayer dollars on the line. But if you think the system can hold up, as do most bailout critics, those guarantees are unlikely to cost very much.
And if I think that the system cannot hold up, that there will be a domino effect and total crash - and that the government still shouldn't bail them out?
Posted by: andy at Mar 19, 2008 12:25:37 PM
"The first comment is wrong: the Reinhart Rogoff paper does reference the number, see page 4, footnote 1."
That is incorrect. There is no reference in that footnote, even though discuss the number they state in the main text.
Posted by: Erik at Mar 19, 2008 1:38:09 PM
I realize I'm always criticizing Cowen, but hopefully I at least come off as respectful...
If you're a critic of bailouts, you can't have it both ways...[Y]ou can't claim: "The bailout isn't needed" and also "The bailout will burden taxpayers."
Sure I can. (Of course, there might indeed be many people who are contradicting themselves as you suggest, but I don't think I'm one of them.)
For years many cynics of financial engineering, outsourcing, trade deficits, etc. have been saying, "These big investment banks are making stupid decisions." At the time, as a free marketeer, I responded, "So they are taking bigger risks with their own (clients') money than you on the sidelines think is appropriate? Give me a break."
In a truly free market economy, if you make a bad investment, you eat the losses. The bigger the losses, the bigger the lesson for everyone during the next bubble. Those losses are real; people really have been overvaluing mortgage-backed securities. Cowen is acting as if government bailouts will render the mistakes irrelevant.
Someone has to eat those losses, and I think it should be the institutions that made the bad decisions, not the taxpayers. Not only is this fair, but it also will provide the best incentives for the future. How is this a contradiction?
For an analogy, I also believe: "The government doesn't need to fund higher education, and if it does so it will cost the taxpayers billions." Is that statement somehow internally inconsistent?
Posted by: Bob Murphy at Mar 19, 2008 2:58:50 PM
My question is, why can't we design a system that doesn't need to be bailed out so often? I'm thinking of a more decentralized system, like the internet, which is designed to withstand the failure of large sections of its network.
Posted by: Ryan at Mar 19, 2008 3:24:51 PM
I was going to post my own comments, but Bob Murphy already covered everything I would have said.
Posted by: Dr. T at Mar 19, 2008 8:29:39 PM
"Furthermore if the destruction of the debt claim would otherwise have been deflationary, some of that debt can be monetized (thus, taxes don't go up) without raising the risk of inflation."
So - if we assume that debt deflation is unnacceptable and that losses should not be realised we can monetize debt to a large degree with no inflationary effects. Ok, let's assume I accept that argument. I remain fundamentally confused as to why the banks / wealthy investors in hedge funds should be the principle beneficiaries of the Fed's pseudo-monetization (lax repos which get rolled over, TAF, etc). I would think it would be more 'democratic' to credit, say, the poorest 20% of Americans with tens of thousands of dollars directly into their bank accounts. Whether you agree or not with this thought experiement is not really the issue - I guess my point is there are philosophical reasons to be opposed to monetization. Fairness is the issue here with bailouts.
Unfairness could eventually lead to a loss of confidence in the currency itself on a psychological basis or a backlash manifesting itself as ridiculously populist demands placed upon the government, amongst other possible consequences. Maybe its unlikely monetization would unravel in this way because the common man doesn't understand it. But let's be clear about the philosophical problems and the risks.
And this is coming from someone who works at a major i-bank and benefits greatly from the Cantillon effects of money creation happening at my doorstep.
Posted by: N-z at Mar 20, 2008 4:48:44 AM
Anthony's point is right on: the Bear takeover was a bailout of Bear's creditors, counterparties and clients with assets on deposit. The Fed has announced that there is no need for due diligence if doing business with a large investment house. That's like the firemen dropping oily rags from his firetruck.
Here's a way to have it both ways: Bear's bankruptcy would have cost many creditors and counterparties, but would not have triggered a systemic meltdown. If that's the case, then it was not needed, but will cost taxpayers.
Posted by: Bill Conerly at Mar 23, 2008 4:44:32 PM
To bailout or not to bailout. That is the question. In both cases those with mortgages they can't afford for homes that are not worth what they paid are in real trouble. Let's bail them out and let the perpetrators go under. Bear Stearns and their ilk need to take responsibilty for poor fiscal management. We keep bailing them out and they keep coming back like the living dead. Let them die a natrual death, diminsh the gene pool so to speak, and let them all know in the future there is no bailout. By the way, why is it that none of them is complaining aobut regulation during this "regulatory bailout?"
Posted by: richard lilliston at Mar 31, 2008 11:26:29 AM
Let's give credit where credit is due! What happened, obviously, was excess asset valuation, pure and simple. Sure, everything that has been mentioned as a "cause" of the problem, bundled securities, greed, sub-prime mortgages, terrible lending practice, greed, etc., all these things "aggravated" the symptoms, they didn't cause the problem. The root of the problem existed, as usual, at the federal level. Excess valuation of anything in our economy is almost exclusively caused by oversupply on the monetary side. That is the "definition" of excess asset valuation, too little product and too much money. Too much deficit, too much money supply and too much trade deficit to fund the credit that caused the too much money. Another case of wolves gaurding the chicken coup, Bernanky, Paulson and his boys, have nothing to lose[unless of course they actually sink the ship] and everything to gain from excess asset valuation. Those in power own the bulk of the assets and inflation does nothing but make them richer. The rest of us work to pay off what small share of assets we are trying to own, and inflation curbs our ability to maintain our payments. Zero inflation would be the goal for a "fair" free market, as at that point the market could work on level ground and be most effective. of course, how plausible is that when inflation is measured by the cost government stabilized food products rather than things that actual "inflate" during expansionary time, such as energy, commodities and housing? The system on paper is good and fair, but obvious in recent events, the actual game is highly rigged and the outcome is already determined.
Posted by: gene at Oct 5, 2008 3:49:11 PM
