Janet Yellen is keeping an eye on repo.

Photographer: Chip Somodevilla/Getty Images.

Regulators Want to Slow Runs on Derivatives

Matt Levine is a Bloomberg View columnist. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz and a clerk for the U.S. Court of Appeals for the Third Circuit.
Read More.
a | A

Nobody quite knows what it means for a bank to be "too big to fail," so the regulators in charge of solving the problem have an understandable focus on tidiness. A bank that fails tidily, sensibly, in neat little compartments, probably won't do much damage to anyone else. A bank whose failure is sprawling and incomprehensible might well turn out to be catastrophic.

So the preferred mechanism for winding up a possibly too-big-to-fail bank these days is largely about compartmentalization. You put all of the important, messy stuff into subsidiaries -- put the deposits in a bank subsidiary, the repurchase agreements and derivatives in a broker-dealer subsidiary, etc. -- and put those subsidiaries under a "clean" bank holding company with a fairly large amount of capital and long-term debt. Then if things go horribly wrong, the holding company's shareholders and bondholders are the ones who lose money, shielding the people who have messier and more systemic claims on the subsidiaries. The regulators swoop in and recapitalize the holding company, or just sell the subsidiaries to other, healthier banks, in any case without ever interrupting service at the systemic subsidiaries. All the bad stuff happens at the holding company, all the important stuff happens at the subsidiaries, and you try to avoid mixing the two. Then all you have to do is make sure that the holding company has enough equity and long-term debt to shield the subsidiaries against any plausible bad outcome.

But to make this work you really need to keep things in their boxes. Derivatives have a tendency to want to jump out of their boxes. In particular, if bad things are happening at a large and systemically important bank holding company, there isn't a lot of reason for the bank's derivatives counterparties and repo creditors to stick around. Repo is meant to be a super-safe place to park your money overnight; if it looks like a repo counterparty might default, then you look for a different counterparty. And derivatives are just supposed to work: If Bank A owes you money under an interest-rate swap, and you owe Bank B money under an offsetting swap, and Bank A defaults, then all of a sudden you have an unanticipated unhedged risk. So if your derivatives or repo counterparty gets in trouble, you bail immediately to protect yourself. (Also there is always the possibility of making a lot of money on the unwind.)

QuickTake Defusing Derivatives

But while this is individually rational, it is systemically bad. As Janet Yellen put it yesterday:

The crisis underscored that when a large financial institution gets into trouble, its failure can destabilize other firms. This is because large banking organizations are connected with each other by the business they do together and through the contracts that result from that business. Indeed, in the 21st century, a run on a failing banking organization may begin with the mass cancellation of the derivatives and repo contracts that govern the everyday course of financial transactions. When these contracts, known collectively as Qualified Financial Contracts or QFCs, unravel all at once at a failed large banking organization, an orderly resolution of the bank may become far more difficult, sparking asset firesales that may consume many firms.

So yesterday U.S. banking regulators proposed new rules to prevent that from happening. The rules basically say that a bank subsidiary's derivatives and repo contracts can't be cancelled for 48 hours after the bank's holding company files for bankruptcy or otherwise enters resolution proceedings.  This gives the regulators two days to swoop in and conduct the neat resolution of the bank before its derivatives spill out everywhere and create a mess. 

One impulse behind the new rules is obvious enough: They really should make bank resolution tidier, which should make the derivatives and repo trades safer. Generally speaking, U.S. law has figured out how to deal with a company that doesn't have enough money to pay its creditors. There's a bankruptcy system with an order of priority, in which secured creditors get paid first, then unsecured creditors, then shareholders if anything is left over. And the bankruptcy system includes an "automatic stay" that prevents creditors from demanding their money back immediately, because that would destroy value and be bad for all the creditors. They have to wait until the bankruptcy court figures out how to divide the money in a way that maximizes value for everyone.

But everyone long ago concluded that regular bankruptcy is far too slow and unpredictable for financial contracts, so QFCs are just exempt from the automatic stay: If a bank goes bankrupt, derivatives and repo counterparties can just demand their money back as though the Bankruptcy Code doesn't exist. Speed and certainty are so important to the functioning of these contracts that the ordinary rules don't apply. But remember that the automatic stay is meant to preserve value for creditors, and that demanding their money back can destroy value. Lehman taught that, sort of.

So the new rule just re-introduces longstanding normal bankruptcy ideas to bank failure, but in a faster and cleaner and more bank-regulatory way. You don't get an automatic stay that lasts until everyone gets together with a judge and works out a plan of reorganization; you get a two-day stay while bankers and regulators work frantically over the weekend to sell the bank subsidiaries to more solvent acquirers who can seamlessly take over the derivatives and repo trades. The sense of urgency that regulators felt in 2008 is embedded in the new rules; the idea is that 48 hours should be plenty of time to work out a resolution so clean and compelling that, by the time derivatives counterparties are allowed to bail, they won't want to.

This is a real and good and sensible goal. There are objections, and they are classic objections; the main objection is that, if you ban derivatives flight after some cutoff point (the bankruptcy filing), you'll get flight before that cutoff, which will make the crisis worse:

The Managed Funds Association, a hedge fund trade group, published a paper last fall suggesting the rules would actually harm financial stability by encouraging investors to exit trades at the first sign of trouble, lest they be hemmed in by the termination restrictions after a bankruptcy filing.

I suppose that is true but, I don't know, you have to start somewhere.  It is at least plausible to think that, on balance, the new rules will make things safer not only for banks, but also for the repo and derivatives creditors who are losing the right to cancel trades under the new rules. Creditor flight is a collective action problem, and the new rules might solve it for those creditors and leave them better off than they'd be in an every-counterparty-for-itself world.

But I think there is another, opposite impulse behind the rules too. They are not just about making derivatives and repo contracts safer; they are also about making them feel less safe. They seem to have had that effect, too; one writer says that the rules have "thrown into question a core tenet of securities finance transactions." These things were supposed to be safe, protected by instant cancellation and collateral and buy-in rights. Now they are more uncertain. 

The deep fragile magic at the heart of banking is that a bank funds its risky assets by issuing risk-free money-like liabilities. At its most basic level -- Bailey Brothers' Building & Loan takes deposits and makes mortgage loans -- this is good and simple and well-understood and socially useful. But even there it is a fragile magic: If all the depositors want their money back at once, it's a disaster. We know that's a problem, we have known that for decades, and for traditional banking we have solved the problem. We have the Federal Reserve as lender of last resort, and the Federal Deposit Insurance Corp. as a deposit insurer, to prevent traditional runs on traditional banks from being a problem. We have in essence made banking a public-private partnership : Risk-free money claims are socially useful, and productive loans are socially useful, and private bankers have the right skills and incentives to make the loans, while the government and the central bank have the credibility to guarantee that the claims are risk-free.

But there is a lot of unease and uncertainty about expanding that public-private partnership, both to other sorts of assets (not good small-business loans but evil trading-desk positions), and to other sorts of money-like claims (not good retail deposits but evil repo agreements). Obviously banks like to push the limits: If counterparties think that repo agreements and derivatives trades with big banks are risk-free and, as it were, part of the general public-private partnership between banks and the state, then they won't demand to be compensated for credit risk when they fund banks via repo and derivatives. (This is called the "too-big-to-fail subsidy.") And it is arguably socially useful too: Investors want safe money-like assets, and they want bond-market liquidity from bond dealers, and it's certainly possible that banks are the best places to provide those safe assets and that liquidity.

But the partnership -- in which banks issue safe claims and use the money on risky productive activities, and in which the state backstops those safe claims -- has never been explicitly negotiated, outside of the classic deposits-and-loans business of banking. On the asset side, lots of people think that insured deposits and bond trading shouldn't mix (this is the Volcker Rule, sort of ). And on the liability side, you might worry that repo and derivative claims look too safe -- that if investors think those claims are as safe and money-like as deposits, then they will give banks too much short-term information-insensitive funding, leading to banks that are unstable and overleveraged and run-prone. And then, when a run happens, banks and investors will look to the Fed and the FDIC for a backstop, because at some level they thought that was the deal.

The new rules are a way of saying: That's not the deal. Your repo claims and derivatives assets are not safe.  We hope they'll be safe -- and the new rules are meant to allow for efficient bank resolution that will protect the value of those claims -- but there is no backstop, no certainty, no magic. If your bank counterparty fails, you'll probably get your money back, but not by way of a seamless instantaneous closeout. First you'll have to spend 48 hours pondering the dangers of counterparty credit risk. The new rules demystify some of the magic of banking. They force counterparties to examine the trick more closely, and to confront the fact that it covers up but doesn't quite eliminate risk.

  1. We talked about this mechanism a bit back when the banks' "living wills" -- kind of the second-best resolution mechanism -- were in the news. It is the "Single Point of Entry (SPOE)" approach to using the regulators' "Orderly Liquidation Authority" (OLA) under Title II of Dodd-Frank, along with the long-term debt and "total loss-absorbing capacity" (TLAC) rules for global systemically important banks (GSIBs). There is no shortage of initialisms.

  2. I realize that in practice lots of derivatives are in the bank subsidiaries, which is controversial (the controversy is referred to as "swaps push-out"), but I tend to agree with John Crawford and Tim Karpoff that in a single-point-of-entry world this doesn't matter that much. But the compartmentalization in the text is stylized, and lots of stuff ends up getting put into the bank subsidiary that might more neatly live elsewhere.

  3. I mean, that's not exactly what they say; I am interpreting. But the gist is that a "qualified financial contract" of a systemically important bank affiliate "may not permit the exercise of any default right with respect to the covered QFC that is related, directly or indirectly, to an affiliate of the direct party becoming subject to a receivership, insolvency, liquidation, resolution, or similar proceeding." So if the "direct party" is a bank or broker-dealer subsidiary, the QFC can't have a default provision relating to the bankruptcy or resolution of the holding company. (Or any other subsidiary, but in the tidy world of single-point-of-entry resolution, why would another subsidiary need to enter resolution? I'm sure there are reasons. I am again stylizing the mechanics.)

    The necessary complement here is that "Under the clean holding company component of the Board’s recent TLAC proposal, the top-tier holding companies of U.S. GSIBs would be prohibited from entering into direct QFCs with third parties": That is, in the tidy new regime, only subsidiaries can do derivatives and repo contracts, and (one would hope) only holding companies should ever need to be resolved, so the failure and the derivatives contracts should always be separated.

    The 48 hours thing is also a bit loose; the derivatives stay is "the period of time beginning on the commencement of the proceeding and ending at the later of 5:00 p.m. (eastern time) on the business day following the date of the commencement of the proceeding and 48 hours after the commencement of the proceeding."

  4. More or less. From yesterday's Notice of Proposed Rulemaking:

    The Bankruptcy Code largely exempts QFC counterparties from the automatic stay through special “safe harbor” provisions. Under these provisions, any rights that a QFC counterparty has to terminate the contract, set off obligations, and liquidate collateral in response to a direct default are not subject to the stay and may be exercised against the debtor immediately upon default. (The Bankruptcy Code does not itself confer default rights upon QFC counterparties; it merely permits QFC counterparties to exercise certain rights created by other sources, such as contractual rights created by the terms of the QFC.)

    The Bankruptcy Code’s automatic stay also does not prevent the exercise of cross-default rights against an affiliate of the party entering resolution. The stay generally applies only to actions taken against the party entering resolution or the bankruptcy estate, whereas a QFC counterparty exercising a cross-default right is instead acting against a distinct legal entity that is not itself in resolution: the debtor’s affiliate.

    The safe harbors cover many derivatives (section 362(b)(6), (17) and (27)) and repos (362(b)(7)).

  5. In some ways it is too soon to tell how much value the Lehman bankruptcy destroyed, though JPMorgan made a ton of money canceling derivatives trades with Lehman.

  6. The Bankruptcy Code addresses the analogous problem with rules about "preferential transfers" made before bankruptcy, but that is surely not practicable with derivatives and repo contracts. Speed and certainty, remember.

  7. In this and what follows I am fairly obviously drawing on various things I've cited before, including especially Morgan Ricks's "The Money Problem" and Mervyn King's "The End of Alchemy," as well as Gary Gorton's writings about safe assets and crises (of which the most recent is "The History and Economics of Safe Assets") and this famous Steve Randy Waldman post. What I call "deep fragile magic" King calls "alchemy"; he wants to end it. 

  8. I owe this use of the phrase to Ricks.

  9. I mean, strictly speaking, the Volcker Rule allows market-making and in some sense wasn't supposed to reduce liquidity. But in practice separating market-making from proprietary trading isn't that easy.

  10. Yesterday the bank regulators also proposed the Net Stable Funding Ratio rules, which require banks to "maintain a minimum level of stable funding relative to the liquidity of their assets, derivatives, and commitments, over a one-year period" -- that is, not to rely too much on flighty wholesale money-claim liabilities to fund long-term assets.

  11. I am exaggerating here in one respect because no one really thinks that derivatives claims are "safe"; everyone knows, after Lehman, that they have not only market risk but also counterparty risk. Still, you collateralize and have closeout mechanisms and can hope that they're safe at least up to the amount of collateral. The new rules put even that in doubt.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Matt Levine at mlevine51@bloomberg.net

To contact the editor responsible for this story:
James Greiff at jgreiff@bloomberg.net