Paul Ryan was on CNBC pushing a Zingales and Hart type solution to the financial crisis

The essence of Zingales and Hart is that if CDS spreads on an institution become too wide regulators swoop in to stress test the institution. If an institution fails the stress test it is put into receivership.

The idea has some cute features and I encourage you to read the National Affairs piece in which they lay out there case. The central problem that I see, with market based and transparent triggers, is that they force runs.

Suppose, that the trigger level is set 100 basis points. If on Monday, the CDS contract on XYZ bank closes at 93 basis points then tomorrow the next thing I am going to do is close-out any outstand positions I have with XYZ. However, I know that everyone else is going to do this. So I need to protect myself from being late in line. How do I do this? By buying CDS on XYZ bank and further driving up the spread.

In this way a run on the bank itself will be replaced by a run on the banks CDS which itself may trigger a receivership that otherwise would not have taken place. Allowing these types of market signals is asking for more runs. That’s why the best way is to just show up with the regulators and seize the institution day-of.


The second issue is that Ryan and to a lesser extent Zingales and Hart seem to view “Too Big to Fail” as a policy or attitude. Too Big to Fail is a state of nature. It is a real thing that exists. Even if we lived in an anarchocapitalist state with no government whatsoever some institutions would be Too Big to Fail. That is, it would be in the collective interest of the population to prop up the bank rather than deal with the fallout of its collapse.

Credible threats to be irrational are hard to make. That is if everyone knows its in your best interest to do something it is hard to assure them that you will not do it anyway. Thus unless you set up some arrangement that makes it rational to hang a big bank out to dry the market is going to assume that you will not do so.


Third, Ryan, Zingales and Hart all seem to be under the impression that borrowers have limited downside risk because of implicit government garuntees. Again, this is a fundamental position that borrowers have. You cannot get blood from a turnip. You cannot collect money that people do not have. And, you cannot loose more than everything you have a stake.

Borrowers always face unlimited upside and limited downside. That is what makes borrowing such a good idea and why institution have to reject some borrowers who are willing to pay market rates. In how many markets do you reject paying customers? Borrowing is special, precisely because of the limited downside risk and unlimited upside.

You might argue that without a government guarantee lenders will be more cautious. However, history doesn’t seem to bear that out. Why this happens is complex and seems to have to do with some essential issue with liquidity risk. That is, the riskiness of lending to XYZ bank depends crucially on what other people think the riskiness of lending to XYZ bank is. If everyone thinks XYZ is a good bet then that really, factually, does make XYZ bank a better bet. Its not just mass misperception.

Reality is influenced by the ability to raise money and the ability to raise money depends on perceptions. Thus we have the very weird and hard to deal with situation where reality depends crucially on perception.