Weakening derivatives’ super-priority

By Felix Salmon
December 18, 2009
Brad Miller for pointing me to Mark Roe's excellent column on super-priority for derivatives in bankruptcy:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Thanks to Brad Miller for pointing me to Mark Roe’s excellent column on super-priority for derivatives in bankruptcy:

True, somebody has to win and somebody has to lose when there is not enough money to go around. But these super-priorities warp the financial industry’s incentives to avoid problems… If financial players set their deals up perspicaciously with enough collateral, they need not worry if a counterparty such as AIG or Lehman collapses, as they will be paid before regular creditors. Some players – Goldman Sachs and JPMorgan are those talked about – set up parts of their AIG and Lehman deals to take advantage of these super-priority provisions. Knowing now what can happen in a financial meltdown, other financial players will do the same in the future, making the financial system more fragile…

If the derivatives and repo transactions were not so favourably treated in bankruptcy, financial players would seek other ways to protect themselves. That effort would channel them into stabilising the financial system or at least into making sure they were dealing with stable counterparties…

Priority facilitates liquidity and risk-spreading, but others pay for part of those benefits because they lose in bankruptcy and the financial system is potentially rendered less stable…

Priority proponents say we should fear financial contagion: one institution defaults and then its counterparty cannot meet its own obligations. Priority, they say, helps to contain the contagion. This could be so, but the opposite can also be true. We saw last year that priority also spreads contagion: as AIG and Lehman weakened, financial counterparties made legitimate collateral calls that, without their super-priority, would not stick in bankruptcy. This meant more of the weakened institutions’ counterparties knew that they would have to make similar collateral calls to stay even. This run-like process further weakened AIG and Lehman, and their weaknesses spread outward.

I’m particularly impressed by the way in which Roe makes a clear distinction between liquidity and stability — two things which are too often conflated. Indeed, in extremis, liquidity can sometimes be a bad thing, insofar as excess liquidity in one area of the markets (most likely Treasuries) can act as a super-magnet, unhelpfully pulling risk capital from everywhere else.

And there’s another problem with derivatives super-priority, which Roe doesn’t delve into: it encourages financial institutions to recast ordinary trades as derivatives contracts, in a multi-trillion-dollar game of beggar-my-neighbor where everybody needs super-priority because everybody else has it. The losers, of course, are less sophisticated investors who are naive enough to play in cash markets and buy real securities.

Goldman Sachs has sworn up and down that it was always fully hedged in its dealings with AIG. But it can only credibly say that because of super-priority provisions. Without super-priority, Goldman would never have allowed itself to amass such an enormous exposure to AIG, and AIG as a result would have been less vulnerable to a run from derivatives counterparties with collateral demands.

So let’s hope that derivatives super-priority is ended as part of an eventual financial-reform bill. Along with giving regulators the ability to treat late collateral grabs as a voidable preference in bankruptcy.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
6 comments so far

Adding the word “super” to “super-priority” makes it sound somehow different than the concept of priority which is already long established in bankruptcy law. Why is the priority of collateralized counterparties in a derivative transaction somehow worse than the priority of senior secured creditors over unsecured creditors in a regular bankruptcy case? Why is one kind of priority good and another bad? I think that priority is a useful concept as long as the rules are applied fairly and consistently. It helps to improve the functioning of credit markets to have different seniorities of creditors. It’s basically the same concept that allows a consumer borrower to take out a mortgage on their house in which the lender gets first claim on that house in the event of bankruptcy. Is that unfair to the credit card issuer who doesn’t get a claim on the house? Should the mortage lender not try to collect on the mortgage payments because that might force the borrower to fall behind on their credit card bills? The argument is fallacious.

Posted by o_nate | Report as abusive

Maybe You, Felix and Brad Miller should read some of FDIC material (and the FDI Act) on the matter before drawing wild conclusions.
“In order to reduce systemic risks, however, both the Bankruptcy Code and the FDI Act have long provided exceptions from many of these restrictions for certain defined financial contracts—securities contracts, commodity contracts, forward contracts, repurchase agreements and swap agreements. In the FDI Act, these contracts are defined as qualified financial contracts, or QFCs. The statutory safe harbors allow parties to these contracts, with some exceptions, to terminate and net their exposures and liquidate any pledged collateral to protect them from losses that could result from market fluctuations if the counterparties were subject to lengthy insolvency proceedings.

Under the FDI Act and other U.S. insolvency laws, a party to a derivative contract has four key rights after commencement of insolvency proceedings……”
http://www.fdic.gov/bank/analytical/fyi/ 2005/101105fyi.html

Posted by alea | Report as abusive

Alea:

Somebody needs to explain the ASSUMPTION that the exemptions that grant super priority to derivatives and repos “reduce systemic risk”. Yes, I know, the FDIC and the President’s Working Group and many others have stated that this is the case, but I’ve never seen a coherent explanation of WHY this would be the case.

(i) You don’t need super-priority to grant derivatives traders the right to net their contracts. This can be done under the supervision of a bankruptcy court.

(ii) When legal scholars look at the systemic risk argument they conclude that it is at best a red herring (http://papers.ssrn.com/sol3/papers.cfm? abstract_id=589261) and at worst a facade to facilitate a direct transfer to financial institutions (http://papers.ssrn.com/sol3/papers.cfm? abstract_id=1265070)

(iii) We’ve just seen how super priority makes it possible for bankers to run legally on other banks.

I’ve been making a similar argument to Professor Roe’s at my blog and just posted a powerpoint version (http://www.slideshare.net/csissoko/the- legal-foundations-of-financial-collapse) .

Posted by csissoko | Report as abusive

“In order to reduce systemic risks…..The statutory safe harbors allow parties to these contracts, with some exceptions, to terminate and net their exposures and liquidate any pledged collateral to protect them from losses that could result from market fluctuations if the counterparties were subject to lengthy insolvency proceedings.”
That’s clear enough, carrying positions for a lengthy and uncertain period time is riskier than netting and closing them out immediately.

Posted by alea | Report as abusive

I understand how the safe harbor provisions protect each individual bank from losses. I just don’t understand how to relate the fact that individual banks are protected from losses to systemic risk.

It seems to me that managing the risk of losses is precisely what banks are supposed to do. How can protecting banks from losses help them play the role of credit risk managers in the economy?

My question really is: what is the model of systemic risk that explains the safe harbors? Is the FDIC claiming that there is some linear relationship between the losses incurred by individual bank and systemic risk? If so, why does this relationship exist? I really just don’t understand.

Posted by csissoko | Report as abusive

In a forthcoming article in the Harvard Law Review, I like Roe argue that the bankruptcy priority for derivatives discourages monitoring and thus probably contributed to the 2008 financial crisis. I also show how these exemptions encouraged AIG to take on correlated risks, selling guarantees on mortgage-backed securities while at the same time buying up subprime MBS for its own investment portfolio. Contrary to standard accounts, this conduct probably was rational from the perspective of AIG’s shareholders. Here’s a link:

http://papers.ssrn.com/sol3/papers.cfm?a bstract_id=1394995

Posted by RS3 | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/