Twitter: Joshua Gans- @rurounimaikeru that's the way. 98 or so more to go
- Why I’m Hoping My Son’s Kindergarten Teacher is Good http://j.mp/cwOgoU
- Sharing ... Who Pays For Credit Card Rewards? http://n.pr/9N1utH
- Scott Adams replicates Paul Romer's idea. http://bit.ly/9rdwoa
Readership
Oct
19
Moral hazard misunderstanding watch
October 19, 2008 | 14 Comments | Joshua Gans
“It will exacerbate moral hazard” has become the standard criticism of most government action going on to deal with the financial crisis. I wince almost every time as commentators rarely identify actual moral hazard. Let’s take an example from an opinion piece by Milind Saythe (from the University of Canberra):
“A blanket financial guarantee is totally uncalled for and will only exacerbate moral hazard. That is, people will take increased risks as they know that the Government will bail them out if the outcome is adverse. With the blanket guarantee in place, financial institutions would be encouraged to do exactly what led to the crisis: lend to sub-prime borrowers. The appropriation of profit and apportionment of loss must stop.”
This quote appeared again in The Australian on Saturday. This statement is just plain wrong.
Let me remind everyone when moral hazard occurs: it occurs when a decision-maker taking a risky action can appropriate a return on the upside but foist costs on others on the downside. So who are the decision-makers in Saythe’s statement. Maybe they are the banks, their shareholders and their managers. Without a guarantee, if there is a run on a bank that is unfounded, the bank collapses and those people lose. With a guarantee, if there is a run on a bank that is unfounded, the bank collapses, deposits are paid back by the government and those people still lose. So there is no moral hazard there. Replace unfounded with founded due to “lending to subprime borrowers” and you have the same analysis. The decision-makers bear the cost to them on the downside in either case. Indeed, with the guarantee, there is less a chance of an unfounded bank run and so they get to keep more of the upside. That is a moral benefit not a hazard.
So maybe the decision-makers are someone else. Perhaps the borrowers or depositors. But do these people have any special information about the risk profile of the bank. I can imagine that they might for some hedge fund but for mass market financial institutions, that strains credibility. But it is true, the guarantee does reduce incentives to act on information (say rumours of insolvency). But that is hardly a bad outcome.
The final actor that might be subject to moral hazard is the government. The problem is that the guarantee has them bear more of the costs of poor monetary and fiscal management not less. Once again a moral benefit.
Take claims of “moral hazard” with a grain of salt. It is not at all clear people are using the term appropriately. I’ll post incidents here so at least they can bear more of the costs of loose terminology.
Comments
14 Responses to “Moral hazard misunderstanding watch”
Joshua, it’s pretty clear to anyone who’s looked at the US savings and loan debacle what the author meant.
You can accept deposits and pay, say, 10%, which is well over the going rate, but you can offer a guarantee of repayment, via the govt guarantee, so that depositors are confident that you’re “safe”. Meanwhile, you can lend at 15%, 20%, or whatever to the riskiest of borrowers, and for a while, you’re creaming off 5 or 10 percent. Then the borrowers fail to repay, and the govt picks up the tab.
The risk taking is on the part of the managers of the bank.
You’re failing to recognise agency problems when you lump bank shareholders in with the managers. If the bank manager can collect $10m, $50m or whatever for risky bets made with shareholder capital, he’s going to do it.
The point is the manager of the institution will take the same whether guaranteed or not. Lehman Brothers was not guaranteed while Bank of America is.
To be clear, the bank shareholders may be subject to the manager’s moral hazard but that doesn’t have anything to do with whether depositors are guaranteed or not.
WaMu and Wachovia were guaranteed.
Joshua, let’s instead say the moral hazard is that the bank no longer has to fear a run. People’s deposits are guaranteed by the government, their fears are assuaged, therefore the bank no longer has to plan for that contingency.
There are no costs they can pass on to anyone, but they can act in such a manner that might be held to greater scrutiny should the guarantee not exist.
For mine, moral hazard isn’t so stringently defined as you have done above. It is simply where the risks of a certain action you take are a assumed by another party, and whether you would act differently because of this.
The government have said all deposits are guaranteed. No need for the public to be edgy and threaten the banks funding by withdrawing. Does this mean the banks will be more careful with their creditors money, or less?
I’m very confused by the point you’re trying to make. Is the definition of moral hazard wrong? or the use of the term? or the application?
Via the Wikipedia I found this by Larry Summers.
His argument I understand, but don’t entirely agree. Are you saying something similar or very different?
Here is a personal anecdote that sheds some interesting light on the subject. The long-time (and a decade since retired) CEO of a well-known bank in Australia said to me in a private conversation about 10 months ago that he thought that all the noise around moral hazard was largely an academic nonsense. And this is coming from a character well known to be exceedingly conservative in regard to all matters related to risk management. He commented that when he was running his bank he and his colleagues assumed that there was an implicit guarantee protecting the bank’s livelihood insofar as it was generally a given that the RBA would not allow it to fail. Yet he also claimed that in the event that this guarantee was ever exercised all the senior management and decision-makers at the bank in question also assumed that their careers and reputations would be irreversibly destroyed (ie, having presided over a bank failure that required them to rely on the government’s support). This is certain true if you look at the damage to the reputational capital of the executives running Bear Stearns, who have also had much of their personal wealth eviscerated (nb: in contrast to traditional workers most of these investment bankers had their compensation locked into long-term equity plans, which were all but blown-away in the bail-out). The individual’s message was simply that the presence or absence of the bank guarantee had no risk-inducing impact on his behaviour as CEO whatsoever: if it ever came into play he was cactus anyway.
Whoever says ‘moral hazard’ first totally wins!
I remember watching Lou Dobbs on CNN once many months ago – don’t ask me why I did this – try to say that all his opponents were wrong because they didn’t consider moral hazard, except he then went on to give a completely and utterly wrong explanation of it in a very serious and outraged tone. Here’s one of his lines, though not the one I saw: “Moral hazard should prevail for the likes of Citibank and for the other irresponsible, irresponsible financial institutions and the consequences.”
OK, I know what he means, it’s just that putting ‘moral hazard’ in there makes no sense.
Another reason economists should try not to use in-house jargon so much. ‘Moral’ is a very bad word to include here. Not that I have a better idea in this case. The ‘hidden action’/'hidden information’ terminology is slightly better jargon (rather than ‘moral hazard’/'adverse selection’), but still pretty much impenetrable jargon.
Josh, I’m puzzled by this post.
Surely in a world without deposit insurance, institutions that took larger risks would have to pay a premium for doing so? It is not that there would be no variation in risk taking, but that the cost of funding would be higher for riskier banks. Basically, deposit insurance severs the link between probability of default and funding costs.
Yes, shareholders lose regardless of whether the deposit insurance is in place or not. Yes, there is an agency problem between managers and shareholders. Yes, managers lose their jobs if the bank fails. However, what you are missing is that there is a positive probability that the bank does not fail, and even if it does, managers will have drawn an income related to the institutions profitability in the interim. Deposit insurance can increase that profitability by increasing the wedge between funding costs and risk.
Are you arguing that this is not the case? If you are, that would run contrary to most of the international literature on the subject.
Perhaps you agree with this in theory but doubt it in practice because you doubt that depositors can monitor institutions effectively? That might be true for small depositors but I don’t know whether it is the case for large depositors. After all, we expect people to understand the tradeoff between risk and return in their non-deposit investments.
Sinclair, that definition of moral hazard is right. But it doesn’t necessarily accompany all guarantees unless particular conditions are met. I argue they are not met.
Labor Outsider, for actions that bring down institutions, I do not believe that deposit insurance exacerbates the manager-shareholder moral hazard. Put simply I think that managers have ample incentive not to cause their institutions to go bankrupt. Their lifetime earnings are impacted immensely. Actually, let me qualify, unless they were close to retirement or something but even then there is a legacy issue.
On depositor monitoring, where moral hazard can theoretically be changed by the government guarantee, I do not think there is an asymmetric information issue. Depositors can’t really assess risks. So what they do is employ rating agencies to do so. Are you going to claim that those rating agencies are able to assess risk?
Interesting post, every one is assesing these events as they go and to some extent thinking out a loud.
Harry Clarke mentions moral hazard in his most recent interesting account of the financial crisis.
Not sure the way he sees it is correct, whether moral hazard has and is playing a part is certainly worthy of further investigation and debate.
Enjoyed reading your interesting post Joshua. You may have seen the references to moral hazards in the Senate Economics Committee yesterday and Dr Ken Henry’s responses on the issue…
Josh’s point here deserves to be made often and loudly. Moral hazard does not apply to Dick Fuld; his name will forever be a byword for management failure. The moral hazard point relies instead on something quite different: the ability of non-equity funders (including small depositors) to accurately assess the risk to which they are exposing themselves. Information asymmetry makes this unlikely.
The fact that so few people understand this point is illuminating. I have been genuinely surprised that so few people have teased out the issues the way Josh has done here.
I used to think that all this implied ratings agencies should somehow be funded to distribute their assessments to bank depositors. But it turns out, of course, that this would have made no difference at all.
Chris Joye’s understanding of the RBA’s expected coursee of action in the case of bank failure is exactly right. My understanding is that in the case of the Bank of Adelaide collapse, the RBA marched in and explained to the board what was going to happen, whether they liked it or not. And as I recall, the RBA told that story for years afterwards. They explicitly wanted it known that bank depositors would be broadly protected, but that shareholders and management of failed banks faced the equivalent of summary execution.
[...] all those armchair economists out there, Joshua Gans has as post that clarifies what is meant by ‘moral hazard’. It’s a term that gets bandied around quite a lot these days, often [...]