Yves here. One of my pet peeves has been the way legislators and regulators have come up with various remedies to the Too Big to Fail problem, like living wills and Dodd Frank Article II resolutions, and have gone into Mission Accomplished mode. We’ve written lots of nitty-gritty posts debunking ideas like these when they were fresh.
As we stressed then, dealing with a problem like that of TBTF banks is like disarming a very large, live bomb. It has to be done carefully and in a certain order so as not to create lots of devastation. And as top risk manager Richard Bookstaber stressed in his book, A Demon of Our Design, the first step in reducing risk in a tightly-coupled system is to reduce the tight coupling. Implementing any other type of risk reduction effort will make matters worse. We’ve identified derivatives as the biggest source of tight coupling, which means it needs to be given top priority in “end TBTF” efforts, and needs to be addressed in a informed manner.
This very readable post by Perry Mehring illustrates that the officialdom still has a very long way to go. While they’ve at least included some aspects of derivatives risk in their efforts (like moving many to central clearinghouses), Mehrling exposes the glaring problems in one element of the plan to deal with derivatives that banks intermediate.
And Merhling makes a key point: that too much of the thinking about TBTF focuses on solvency of individual institutions. In a system of market-based credit, the authorities feel they must keep financial markets operating. Bear was not bailed out because it was a systemically important institution by virtue of size. It was seen as too important to let fail because it was a critically important player in the credit default swaps market by virtue of being the number three prime broker. Money market funds were rescues not to protect the funds themselves, but because they were a major source of funding in the repo market, which was and remains an critical source of liquidity to the major capital markets trading firms. Similarly, the Administration’s considerable efforts to paper over chain of title issues in the mortgage-backed securities market weren’t to save Bank of America, which was heavily exposed due to its acquisition of Countrywide. It was because it perceived the MBS market to be too big to fail.
BY Perry Mehrling, a professor of economics at Barnard College. Originally published at his website
The word has come down, “Never again!”
On October 14, 2008, the US Treasury announced a plan to recapitalize the US banking system, to the tune of $250 billion, starting with the nine biggest banks who were forced to take the money, whether they wanted to or not. The government got its (our) money back, but that’s not what matters. Private risk-taking for private profit on the upside seems to require private risk-taking for private loss on the downside. And implementation of that principle has been taken to require very substantially increased total loss absorbing capacity (TLAC), starting with larger regulatory capital buffers but including also a second tranche of supplementary capital in the form of “bailin-able” bonds that can be converted to equity in times of stress.
That’s the state of play at the moment. Sounds good, right?
The problem is that derivative contracts, written under the standard ISDA master agreement, give non-defaulting counterparties the right to immediate termination, which in effect puts them at the front of the line relative to other creditors. Just so, when Lehman filed for bankruptcy, their derivative counterparties terminated their exposure, often on terms disadvantageous to Lehman (and hence to Lehman’s other creditors). Just so, a marginally solvent firm became deeply insolvent. Having this experience in mind, regulators want to make sure that the enhanced TLAC is not simply absorbed by this kind of front-of-the-line contract termination, leaving nothing for the other creditors and hence requiring government backstop to ensure ongoing operation.
To prevent this, regulators have come up with something called the ISDA Stay Protocol which they have pressured the major dealer banks to sign. Under this protocol, termination rights are delayed by 48 hours. The idea is that this should be enough time to arrange for the (private) recapitalization of a failing bank, which will then present the non-defaulting counterparties with a new healthy counterparty, and so eliminate the need for termination. The idea is to keep the entire book of derivative exposures alive for two days while a bridge is built to transfer that book from the old counterparty to the new one.
Sounds good, right?
The problem is that hedge funds, and others who see themselves losing a valuable termination right, are reluctant to sign the protocol, and they have good reasons to be reluctant.
From their point of view, the whole point of the termination right is to enable the non-defaulting counterparty to find their own new counterparty, immediately. If your hedge has just disappeared, you need to find a new hedge, ASAP. Two days of limbo, during which you are supposed to trust a new untested resolution process, and after which you will be presented with a new counterparty chosen by someone else, is a very different matter. Most obviously, two days in a stressed market situation is forever. Prices can move, and you can be required to make very large payments, all the while your hedge is frozen in the resolution process; now it is your capital on the line. In effect, what the regulators seem to have in mind is recruiting derivative counterparty capital as a third line of defense. No thanks.
From a money view perspective, what is most troubling about this entire debate, on both sides, is the unrelenting emphasis on solvency, not liquidity, and the consequent (inadvertent?) implicit assumption of efficient markets.
Price is not necessarily always equal to value, and market pressure on either sell or buy side can push price pretty far from value. (Don’t trust me on this, listen to Fischer Black.) This is the central message of modern models of the economics of the dealer function. Dealers make money by absorbing imbalances in market demand on their own balance sheet, buying below value when everyone is selling, and selling above value when everyone is buying. Dealers do these trades because they expect them to be profitable when, in course of time, they are able to reverse them.
But dealers have in mind always that, in order to reap that profit, they need to be able to survive the course of time. Forty eight hours can be too long. In times of stress, even stress that involves only an individual market or individual counterparty, profit-motivated dealers will be less willing to supply liquidity to the market. As a result, price can be expected to deviate much farther from value than in normal times. Traders know that, and can be depended on to protect themselves ex ante by stepping away from counterparties at the very first sign of trouble, or by finding other trading and clearing venues.