Bad Bank Bafflement

A number of comments on Kevin’s link to Buiter’s discussion of bank nationalisation have brought up the idea of a “bad bank” that can be used to take over non-performing assets.   I think this issue is important enough to hoist out of the comments and onto the front page!   It should be noted that Buiter is discussing this idea in the context of a fully nationalised British banking system.  The “bad bank” idea has an “economies of scale” advantage in that case. so that all bad loans can be grouped together and dealt with by a team specialising in getting the best long-run return for the government from working out bad loans.

Outside the context of full nationalisation, I’m not sure I understand the “bad bank” idea or why it has caught on in the Irish media over the last day or two.   I’ve been puzzling over this the last few days and then found a post by Paul Krugman that expressed my puzzlement far better than I could.  In a post entitled bad bank bafflement (good post title!), Krugman says: “The idea of setting up a “bad bank” or “aggregator bank” to take over the financial system’s troubled assets seems to be gaining steam.  So let me go on record as saying that I don’t understand the proposal.  It comes back to the original questions about the TARP. Financial institutions that want to “get bad assets off their balance sheets” can do that any time they like, by writing those assets down to zero — or by selling them at whatever price they can. If we create a new institution to take over those assets, the $700 billion question is, at what price? And I still haven’t seen anything that explains how the price will be determined.”

In the Irish case, perhaps someone could explain to me how the bad bank proposal gets at this question.  What price would the Irish government pay to struggling banks for their underperforming loan portfolios?  Why should the government pay a price above current market value rather than, for instance, provide additional capital to cover the implicit losses and thus increase the government’s equity share?

As Krugman notes, in the US case, the answer to these questions appears to be that Bernanke believes the market is systematically underpricing a wide range of mortgage-backed securities and that the US government may break even (or perhaps profit) from buying them at above market rates and selling them later or holding to maturity.  Whether Bernanke is right or not is open to question.  But is there any reason to think a similar logic applies to Irish commercial property loans?

8 replies on “Bad Bank Bafflement”

There are rumours that the British government is considering a complicated scheme whereby the government will insure against losses. It’s another idea for Dublin to consider, alongside whatever happens in the US.

This would be a floor, a pretty high floor, on the price of these dodgy loans. There would certainly have to be massive payouts by the state and the state would lose money in my humble opinion, but some free money is going to have to be handed out at some stage.

The difficulty is in setting the premium that will be paid to the state by the bank. Is there a market based mechanism that could set the premiums or guide the state as to which loan books are riskiest?

Valuation of the assets transferred into an AMC or Bad Bank has been the stumbling block in the way of implementing the first version of the US TARP. But it’s not impossible.

There’s one obvious solution, canvassed by myself last September in the comment section of (see my posting at

The essence of my idea is that, in return for the toxic assets, the AMC pitches in enough government money to make the bank whole, but the government gets warrants for the bank shares whose exercise price rises with the recovery performance of the toxic assets.

The schemes subsequently used for the recapitalization of the Swiss bank UBS and also more recently for Citi contain similar features.

It is crucial to be very clear in advance as to who is bearing upside and downside risk in any financial restructuring arrangements if the taxpayer’s interests are to be protected — as I spelled out for the Anglo case in my comment to Kevin’s posting on Buiter’s idea yesterday.

It is presumably easier to put a value on distressed real estate loans than toxic MBSs. And even if the AMC gets a bargain, the deal could be structured to allow the Exchequer to share in any subsequent gains, as PH suggests. The RTC used PPPs and other such schemes.

Karl asks, Why not provide additional capital to cover the implicit losses and thus increase the government’s equity share? Presumably the banks will need more capital to plug the hole resulting from selling the bad loans to the AMC. Transferring them to a specialist distressed asset company has two advantages: (1) bankers have little expertise and experience in managing distressed assets, and (2) bankers who made the bad loans face perverse incentives in dealing with those loans. Folks at the AMC face better incentives.

FYI: Here is a follow up posting from Paul Krugman’s blog on the “bad bank” idea. It argues that a bad bank is useful as part of a temporary nationalisation strategy (Anglo?), but problematic as a means of shoring up other institutions.

More on the bad bank

OK, I’ve been doing more homework on the “bad” or “aggregator” bank idea that seems to be gaining ground. And here’s what I think: it’s mainly based on a false analogy.

What people are thinking about, it’s pretty clear, is the Resolution Trust Corporation, which cleaned up the savings and loan mess. That’s a good role model, as far as it goes. But the creation of the RTC did not rescue the S&Ls. The S&Ls were rescued by (1) having FSLIC seize them, cleaning out the stockholders (2) having FSLIC pay down enough debt to make them viable (3) reselling them to new investors. The RTC’s takeover of the bad assets was just a way for taxpayers to reclaim some of the cost of recapitalizing the banks.

What’s being contemplated now, if Sheila Bair’s interview is any indication, is the creation of an RTC-like entity without the rest of the process. The “bad bank” will pay “fair value”, whatever that is, for the assets. But how does that help the situation?

It looks as if we’re back to the idea that toxic waste is really, truly worth much more than anyone is willing to pay for it — and that if only we get the price “right”, the banks will turn out to be solvent after all. In other words, we’re still in Super-SIV territory, the belief that fancy financial engineering can create value out of nothing.

Color me skeptical. I hope the buzz is wrong, and that something more substantive is being planned. Otherwise, we’re looking at Hankie Pankie II: Paulson may be gone, but officials are still determined to believe in financial magic.

The alternative to the ‘bad bank’ idea is the ‘bad insurance’ idea. Keeps the toxic stuff of the Government’s balance sheet (hey, if it’s good enough for the banks …)

Here’s the details on the FDIC deal with Bank of America. Wonder would this work in Ireland?

On the other hand, a similar suggestion for the UK has been met with a less than positive response by some:

Krugman is arguing very persuasively today for nationalisation, again based on the impossibility of correctly pricing the junk assets. Which brings me back to the question I posed the other day: is the argument valid in the Irish case? If it isn’t, because of the scale of the bad debts and so on, that is one thing. (But then the counter is that the guarantee is already in place.) If on the other hand the only argument against it — admittedly a powerful one — is political, as some were suggesting, then you have a pretty powerful case for an immediate election, surely?

Comments are closed.