Joseph H. Sommer
This post is the eleventh in a series of thirteen Liberty Street Economics posts on Large and Complex Banks. For more on this topic, see this special issue of the Economic Policy Review.
“[T]he instinct of the trader does somehow anticipate the conclusions of the closet.”
Walter Bagehot is always good for an epigraph. And this epigraph is a good one: going well beyond traders. It also applies to the conjoint instinct of bankers, legislators, and regulators. The “bail-in,” or “single point of entry,” technique of large bank insolvency was conceived by bankers, authorized by Congress, and is being implemented by the FDIC. (Calello and Ervin (2010); 12 USC § 5381 et seq.; Single Point of Entry Strategy). This nascent practice needs a conceptual framework. This post, and the companion article in the Economic Policy Review, suggests one. The framework does more than justify bail-in. It applies more generally to financial firm insolvency. It also provides a new and surprising perspective on bank capital.
Some background may help. For our purposes, bail-in has three steps. First, it relegates parent debt to equity. Now free of debt, the parent can afford to inject capital into its subsidiaries. These two steps are done overnight. As a final step, bail-in distributes new equity to holders of the relegated debt. This process resembles a Chapter 11 reorganization at warp speed. However, the two processes differ in several respects, apart from speed. Most of them are technical, but one is central. Bail-in only affects the parent. The parental capital injection typically “bails-in” all the subsidiaries. Their creditors recover 100 cents on the dollar, with no delay. The parents’ creditors bear all the pain. Ordinary Chapter 11 doesn’t work this way. The reorganized parent can only rescue subsidiaries if such a rescue aids the parent’s creditors. In other words, bail-in gives all the subsidiaries’ liabilities priority over those of the parent; Chapter 11 doesn’t. It’s this priority that this post tries to explain.
Insolvency law should maximize the value of firm assets, and then distribute this value according to prebankruptcy priorities. This notion, popularized by Thomas Jackson (2001), is fine as far as it goes. But what should determine the prebankruptcy priorities? Financial economics (from which insolvency theory derives) has no strong theory of preinsolvency priorities. This should be no surprise. The Modigliani-Miller theorem (M-M) asserts that the capital structure of the firm is irrelevant to the value of the firm. The aggregate value of the assets determines the aggregate value of liabilities, and no slicing or dicing of liabilities, or prioritizing, can change this. This suggests an agnostic view of priorities.
M-M may describe the proverbial widget maker, but it fails for banks. Not all liabilities are created equal. There are two kinds of liabilities. Some liabilities are only worth the net present value of their income stream, just like the textbooks say. Widget makers issue these “capital liabilities,” as do banks. But banks also issue another kind of liability—products, such as deposits, repo, derivatives, or insurance policies. The value of these “financial liabilities” exceeds the net present value of their income stream. These liabilities provide more than future income: they shift risk or provide liquidity. These are valuable services. These may be credit products, but the credit is incidental to the services.
Therefore, M-M fails for financial firms. The aggregate value of their liabilities depends on the priority scheme: who gets paid first from assets. Any dollar paid to financial liabilities discharges both net present value and product value. This dollar is more valuable than the same dollar paid to capital liabilities, which only satisfies net present value. Therefore, a priority scheme that favors financial liabilities is worth more than one that favors capital liabilities, especially if it pays the financial liabilities fast, as well as first. Sometimes, it’s worth much more. It is impaired financial liabilities that create systemic risk, typically through a systemic liquidity shock. Their product value can be enormous when so externalized. Capital liabilities only have a net present value of cash flow, by definition. They have no product value, and thus are far less disruptive when impaired.
Bail-in exploits the Bank Holding Company Act, which for practical purposes restricts the parent’s balance sheet to capital liabilities. Financial liabilities are only in the subsidiaries. Bail-in impairs only the parent’s liabilities: it pays subsidiary liabilities on time and in full. Hence, the subsidiary liabilities—financial liabilities—get priority treatment. This is nothing new in financial insolvency law. Even before the invention of bail-in, financial liabilities got priority treatment—the explicit depositor and policyholder priorities of bank and insurance insolvency law, the bankruptcy safe harbors for derivatives and repo transactions, even the statutory trust of money transmitter law. Even casino chips get into the act—bankruptcy judges give them a priority that has no clear warrant in the statute (in re TCI 2 Holdings LLC, 428 B.R. 117, 180 (Bankr. D.N.J. 2010)). The instinct of judges sometimes anticipates the conclusions of the closet!
Bail-in is novel in only one respect. It deals with international conglomerates, not local entities. Local entity insolvency is impractical here: too slow and too many actors. All the action must happen at the parent level.
I promised a “new and surprising perspective on bank capital.” I can now deliver. All capital liabilities—parent debt and equity—serve the same function in bail-in. They protect the financial liabilities to which they are junior. They are all capital, whether recognized by Basel or not. Admati and Hellwig (2013) believe that bank capital levels must be much higher than those mandated by the Basel process. They may be correct. But parent debt typically exceeds Basel capital by a substantial margin, and bail-in uses this debt to protect the financial liabilities. Bail-in may have already done through parent debt what Admati and Hellwig advocate through equity.
Disclaimer
The views expressed in this post are those of the author and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the author.
Joseph H. Sommer is a counsel and assistant vice president in the Federal Reserve Bank of New York’s Legal Group.
RSS Feed
Follow Liberty Street Economics
Shane, Typically, the subs are funded by their parents, since the parent is the cheapest source of unsecured funding in the organization. This means that the parent can recapitalize their subs by a simple debt-equity swap, with no cash changing hands. The bank subsidiary is an exception, since it, too, is a cheap source of funding and may not have much debt to its parent on the books. Here, the parent may have to inject funds into the bank to recapitalize it. Sometimes, the parent has a big deposit with its bank sub—another source of debt convertible into equity.
Mark, I agree with you analytically on valuation, but pity the poor lawyers who actually have to value the firm in a systemic crisis. There are a number of valuation techniques used by courts in ordinary corporate valuation cases—comparable firms, discounted total cash flow, stock price. The second and third are historical methods, and a systemic crisis tends to be a historical break. Fortunately for bail-in, it is far more important to establish a definitive asset value than a “true” value. To the extent that this value is set by a court, we’re okay. A court’s judgment is “so ordered”—definitive indeed!
mechanically, how are the subs recapilatized back to a solvent state? i get the parent’s balance sheet cleans up when you reclass its risky liabilties as equity, but the sub’s balance sheet has not changed. does the sub recognize an asset on its balance sheet recognizing the value of the parent’s guarantee of the subs debt, or does the parent sell its assets/raise new debt/borrow from fed to buy new equity from its sub?
I do not follow hte author’s point regarding valuation. Financial assets generally are valued as the discount of their income streams. The liabilities of hte bank are the assets of their creditors who will value them in the same general way that the bank values its assets. I agree that bank valuation generally should be performed in abstraction to the cocrete features of the capital structure; who gets what in an insolvency will not alter tha aggreagate payout. The point to the capital structure is to differentate risk and efficiently to bring into conformity the bank’s funding requirements with the risk appetites of its respective funders. This risk differentiation is supported by an insolvency regime that recognizes priority and security. In the US, bail-in uses the holding company structure to establish a structural priority; this can be used to facilitate an efficient resolution of insolvency that preserves the operations of the troubled entity. Such an approach to resolution may have beneficial social consequences by e.g. preserving the money supply and minimizing business disruptions and knock-on defaults.