Geographic ring-fencing is a regulatory concept that practically everyone considers a bad idea: an impediment to efficient capital and liquidity allocation in good times, and a potential systemic risk in bad times. But in an August 1914 kind of way, we are watching a resigned march in that direction across the world. The Unites States began this march—playing the role of Gavrilo Princip, for those of you who do not think of Franz Ferdinand as a 2000s rock band—but now has a clear and present opportunity to stop it. At the same time, it can mitigate a potentially counterproductive and likely unintended consequence of yesterday’s tax reform bill as it affects foreign banks.
Ring-fencing and Its Costs
Ring-fencing in international regulation is the idea that a global bank must pre-position prescribed amounts of capital and liquidity in each jurisdiction where it operates—that is, trap capital and assets in those countries, as opposed to maintaining them in its home country headquarters, where they can be deployed anywhere across the organization as needed. The problems with this approach are numerable:
- Because the cost of international expansion is necessarily higher, banks are incentivized to operate in fewer countries, depriving countries of needed investment and credit.
- For those countries where banks do elect to continue to operate, costs go up.
- Because banks are less able to promptly dispense resources to growth areas, economic opportunities in those areas are lost.
- To the extent that an overseas subsidiary experiences trouble, the bank has less free capital or liquidity to support that subsidiary. (Think of ring-fencing as disbanding the town fire department in favor of giving each house in the town a fire extinguisher.)
- To the extent that regulators want to order banks to fill capital or liquidity shortfalls, the same constraints apply.
- In the event of failure (more below), each host country—that is, a country supervising an overseas subsidiary—has the means to seize local capital and assets, with less incentive to participate or cooperate in an orderly, centralized resolution led by the home country resolution authority.
The costs of ring-fencing are felt not only by foreign countries facing disinvestment by U.S. banks but also by the United States facing disinvestment by foreign banks. And this could hurt prospects for future American economic growth, given that foreign banks provide about one-third of all U.S. business loans and hold approximately 20% of all U.S. banking assets, including loans to American consumers and businesses. Foreign banks also provide significant capital markets services to U.S. businesses.
The Rush to the Fence
So, why would regulators engage in such counter-productive ring-fencing? Originally, the answer was clearly the experience of Lehman Brothers’ failure, but now it appears to be more mutual distrust and lack of understanding.
The dissolution of Lehman Brothers taught a harsh lesson, as some overseas subsidiaries of Lehman lost out as the U.S. bankruptcy and various overseas insolvency proceedings played out. Some U.S. creditors fared better in the U.S. proceedings than overseas creditors fared in their local proceedings. In the wake of that experience, no host country wanted to be left out in the cold, and therefore wanted assurance that the depositors or creditors of their local subsidiary would not suffer relative to home country depositors and creditors. One obvious way to accomplish that goal was to require subsidiarization and/or other measures to trap the necessary capital and liquidity in local hands.
At that point, however, a wise group of lawyers, academics and policymakers devised a better way to resolve banks: the single-point-of-entry resolution strategy that is now the approach adopted by seven of the eight U.S. global systemically important banks (GSIBs) in their recently approved living will plans, and is expected to be pursued by all eight U.S. GSIBs in their next plan submissions. The concept involves a resolution (either under the bankruptcy code or a regulatory receivership) whereby all the overseas (and domestic) subsidiaries and branches of a bank remain open and operating, while losses are borne by the equity and debt holders at the parent holding company, through what has come to be called a bail-in (as opposed to a government bail-out). Lest anyone consider this approach hypothetical, note that a recent Federal Reserve regulation requires the eight U.S. GSIBs to hold more than $1.5 trillion in loss absorbing debt and equity to facilitate this outcome—known as total loss absorbing capacity or TLAC. Worldwide, GSIBs are expected to hold about $4 trillion of TLAC. Coupled with industry-wide adoption of a contractual provision prohibiting close-outs of derivatives in the course of such a bail-in, the Lehman problem appeared to have been solved, and the need for ring-fencing radically reduced.
And yet the push continued, reflecting the fact that regulators simply did not trust each other post-crisis. What if a home country regulator refused to use those abundant resources to keep an overseas subsidiary open, and instead just let it fail? Of course, it is hard to imagine a scenario where that would not be a deeply counterproductive action, as it likely would destabilize the overall firm and deprive the home country receiver of all the intangible value in the subsidiary. But once bitten (Lehman), twice shy.
As a result, the Financial Stability Board decided that it would be prudent to require the bank to make what amounts to a down payment of loss absorbency in each overseas jurisdiction. Thus, the notion of “internal TLAC” was born. Like your mortgage down payment, internal TLAC makes it expensive for a bank holding company or its regulator to walk away from a foreign subsidiary. Most in the industry thought a 50% down payment of total loss absorbency was more than sufficient; it would mean that a bank subject to a general TLAC requirement of 20% would have to place 10% loss absorbency (a combination of equity and long-term debt) at each material subsidiary—a lot of money to abandon. What bank or home country regulator would cut off a subsidiary that already had 10% loss absorbency, forfeiting that much capital and triggering a destabilizing liquidation of a valuable asset?
In its wisdom, the FSB decided in 2015 to set internal TLAC equal to a higher range of 75-90% of external TLAC, with the exact amount for each bank to be determined through a consultation among its Crisis Management Group (CMG)—a committee of its home country regulator and the host country regulators of its material subsidiaries.
However, the Federal Reserve decided in December 2016—unilaterally and with no such consultation—to set the minimum internal TLAC requirement in the United States at 90% of external TLAC in the case of a U.S. intermediate holding company (IHC) of a foreign GSIB. Furthermore, the Federal Reserve required these U.S. IHCs to meet a stand-alone minimum internal debt requirement, in addition to the overall internal TLAC requirement (which can be met with any combination of debt and equity). The concept of the stand-alone debt requirement for internal TLAC was considered but specifically rejected by the FSB. Bang!
The result of the Federal Reserve’s unilateral action has been profound. Most immediately, the cost of doing business in the United States for foreign banks rose. The European Commission retaliated with a proposal to require U.S. banks to establish an intermediate holding company for their EU operations, with an internal TLAC requirement also set at 90% of external TLAC. Other jurisdictions around the world are preparing to follow suit—even though the likely result will be banks that are less global, serving fewer customers in fewer countries at higher cost, yet more brittle under stress. Foreign regulators are not shy about their motivation: they consistently express frustration with the Federal Reserve for firing the first ring-fencing shot, as well as a belief that the United States would, in the event of a U.S. bank failure, block any funds from leaving the country. (That last bit is almost certainly wrong, but shows the level of mistrust that the Federal Reserve’s 90% requirement has engendered.)
Required TLAC Debt Treated as Tax Evasion
As though this were not bad enough, one component of this week’s federal tax reform legislation is a so-called base erosion alternative tax designed to prevent foreign firms from avoiding U.S. tax. Its definition sweeps in interest payments on internal TLAC debt made from a U.S. subsidiary of a foreign bank to its home country parent. So, remarkably, a U.S. regulation now requires U.S. subsidiaries of foreign banks to borrow from their parent via the Federal Reserve’s stand-alone debt requirement, and U.S. tax law treats that borrowing as an inappropriate evasion of law and taxes it accordingly.
This treatment is all the more destructive to foreign banks operating in the United States because they are required to maintain a significant portion of their internal TLAC in debt. Without a stand-alone debt requirement, a foreign bank could choose to meet its internal TLAC requirement with more equity and less debt, thereby reducing its tax burden. Doing so would have the added benefit of actually making the U.S. IHCs more resilient to stress, as equity is a superior loss-absorbing instrument anyhow. As we noted in a comment letter to the Federal Reserve at the time of its proposed rule:
It would be counterintuitive to prohibit a covered BHC from substituting equity for long-term debt securities since equity can absorb losses both inside and outside of a bankruptcy or Title II [resolution] proceeding, and therefore function as both going-concern and gone-concern capital. In contrast, absent a consensual debt restructuring outside of a bankruptcy or Title II proceeding, long-term debt securities can only absorb losses in a bankruptcy or Title II proceeding, and therefore generally function only as gone-concern capital. Thus, we believe that long-term debt securities should be permitted but not required to satisfy any minimum TLAC requirements in excess of regulatory capital requirements.
Meanwhile, the Federal Reserve’s policy of imposing a stand-alone debt requirement as part of internal TLAC has thus far been rejected by every other G-20 country that has implemented internal TLAC requirements. It stands alone as an endorsement of debt over equity as a loss absorber.
The 75% Solution
Fortunately, incoming leadership at the Federal Reserve has an opportunity to correct all these problems, quickly and easily, by simply setting the U.S. internal TLAC requirement at the low end of the permissible range established by the FSB in 2015. Specifically, the Federal Reserve could propose to reset the internal TLAC requirement for foreign banks operating in the United States at 75% of external TLAC, and ideally do so in conjunction with reciprocal action by the Bank of England, European Central Bank, Bank of Japan, and Swiss Banking Authority. A memorandum of understanding among that small group would cover nearly all the world’s GSIBs and almost all of their material subsidiaries.
(Notably, the Bank of England has already taken a similar step, setting its internal TLAC requirement at 75% of external TLAC, with the chance of going higher if it has reason to distrust the home country. As is its wont, the Bank of England vests that discretion solely in itself, whereas some process would be required under U.S. law. But precedent exists.)
At the same time, the Federal Reserve could rescind its stand-alone debt requirement, thereby removing the prohibition on substituting equity for debt—yes, that is as backwards as it sounds—in meeting an internal TLAC requirement. This would make the U.S. treatment consistent with that of all other parties to the FSB agreement.
A 75% solution would put the brakes on global ring-fencing; free up capital for foreign banks looking to expand credit in the United States; reinforce trust among global regulators; and remove some of the sting of U.S. tax reform for foreign banks (as well as lessen any further incentive for foreign countries to retaliate). And it does not require Congressional action or multi-agency rulemaking, or renegotiation of the FSB agreement.
Rarely do the benefits of a potential regulatory reform so clearly and greatly exceed its costs.
 The Treasury Department has already recommended in its first report on the U.S. financial system that the Federal Reserve revisit this requirement. U.S. Department of the Treasury, A Financial System that Creates Economic Opportunities: Banks and Credit Unions (June 2017) at 18.
Disclaimer: The views expressed in this post are those of the author(s) and do not necessarily reflect the position of The Clearing House or its membership, and are not intended to be, and should not be construed as, legal advice of any kind.