Speeches & Testimony
"The Relative Role of Debt in Bank Resiliency and Resolvability"– Remarks by FDIC Vice Chairman Thomas M. Hoenig. Presented to the Peterson Institute for International Economics, Washington, DC.
January 20, 2016
In the bankruptcy reorganization of a failed commercial business, the conversion of debt into equity or some revised debt instrument is often the protocol. This has long proven useful in transitioning a failed business back into a productive enterprise. However, mandating increased levels of debt as part of a broad, prescribed resolution strategy has potential effects that, paradoxically, may undermine the very financial stability being sought.
This may be the case with TLAC. Total Loss–Absorbing Capacity, as set out by the Financial Stability Board, requires large, interconnected banking firms to hold certain levels of long-term debt.1 Its intention is laudable: to improve the resolvability of banking firms to assure that, should they fail, they can be resolved without turning to the taxpayer for help. The proposed U.S. version of TLAC would require global systemically important banks (GSIBs) to maintain TLAC-related debt at the holding company with restrictions on ownership.2 However, in many instances, to meet the proposal's goals, firms would have to not only maintain current levels of debt, but also would have to add debt to their balance sheets.
In my remarks today, I will highlight TLAC's effects and suggest that it be used in a more limited and discretionary manner. I will suggest that its use depend on a bank's business model and Title I resolution plan. Such an approach would acknowledge that not all business models are the same and that different resolution strategies might work better for different financial firms –– and for financial stability.
Resolution Planning and SPOE
TLAC is a necessary component of the single point of entry (SPOE) resolution strategy that has been proposed but not yet adopted in the United States.3 It therefore is difficult to discuss TLAC without first describing SPOE.
The FDIC initially proposed SPOE to facilitate an orderly resolution of a failed financial institution under Title II of the Dodd-Frank Act. Under SPOE, only the top–tier bank holding company of the failed firm would be put into receivership. In theory, the parent firm would have enough long-term debt to both absorb any remaining losses after equity is depleted and to recapitalize subsidiaries so that they could remain well–capitalized and continue operations under a new holding company. The presumption is that this would all be accomplished without causing overwhelming panic and necessitating additional government intervention.
Since SPOE was first discussed, the concept has migrated and become the principal bankruptcy plan under Title I of the Dodd–Frank Act for all but two of eight U.S. GSIBs.4 The basic theory of SPOE in a bankruptcy reorganization is the same as for SPOE in a FDIC Title II receivership. In this instance, the financial institution would go through bankruptcy without the critical operating subsidiaries themselves becoming insolvent or requiring government support. These operating units could include subsidiary insured banks, broker-dealers, and asset managers.
Resolution Planning and TLAC
A successful SPOE resolution, under Title I or Title II, assumes sufficient internal long-term debt that can convert to equity in order to effectively recapitalize operating subsidiaries of the holding company following its failure. The Federal Reserve's TLAC proposal further specifies that a clean holding company, with this long–term debt, is required in order to remove obstacles "to the orderly SPOE resolution" of a financial institution. In short, the TLAC proposal is designed to require a firm to have sufficient long-term debt to facilitate a SPOE solution both through conversion of current debt and, if need be, added debt.
According to the estimates in the proposal, the GSIBs would have a combined long-term debt requirement of approximately $680 billion. Existing debt issuance would cover much of this amount, but the proposal estimates that there is a shortfall of approximately $100 billion.
Roughly half of this projected debt shortfall would be met through issuing new debt to replace "near–eligible" debt. The remaining half of the shortfall could, according to the proposal, "then be filled through the issuance of eligible external [long–term debt] in the place of existing deposits or other lower—cost liabilities." The impact section of the proposal concludes by noting that the aggregate increase in cost of funding, which would come directly out of bank earnings, would be in the range of $680 million to $1.5 billion annually.
Other estimates find similar results, and firm-specific estimates are quite revealing. For example, Credit Sights findings "estimate that Wells Fargo has the biggest shortfall at $46 billion, with JPMorgan a distant second at $19 billion, followed by Citigroup at $14 billion, Bank of America at $6 billion and State Street at $2 billion."5
The proposal affects firms in other ways as well. For example, Wells Fargo and Bank of New York–Mellon do not have large broker-dealers and have not adopted a SPOE resolution strategy in their Title I plans.6 In the second round of reviews of the resolution plans, regulators determined that Wells Fargo had the makings of a credible plan, but under the proposed TLAC rule, the company would have to issue additional long–term debt because its current levels are insufficient to meet the TLAC standard.7 The effect, of course, is to encourage them to adopt a SPOE resolution strategy. Also noteworthy, and ironic, is that the former investment banks in general have TLAC surpluses, as they are more reliant on debt funding than deposit funding.
TLAC, as it is applied broadly to the industry, would necessarily affect leverage and in some instances place pressure on firms to change their business model. In doing so, it also would encourage the industry and firms to add risks that may be inconsistent with conventional safety and soundness principles or what they would otherwise pursue. As firms issue large amounts of additional, and more expensive, long-term debt and increase their leverage position, such funds must earn commensurately higher returns to meet debt service and avoid default. Such requirements may encourage banks to acquire higher-yielding and higher-risk assets, or expand into non-bank activities, such as broker-dealer operations.8,9
Also, while authorities might prefer that firms swap out debt instruments, reduce deposits, or issue additional capital to maintain existing levels of leverage, the companies are likely to attempt to increase leverage to meet added earning demands and maintain returns.10
Additionally, in a future recession, if earnings become insufficient to make holding company debt payments (a likely outcome), then regulators would face the prospect of having to allow the transfer of resources out of the bank to avoid default on holding company debt. Remember also that unlike dividends, there can be no suspension of debt interest payments. These outcomes would likely weaken the bank and would seem to undermine the principle that the holding company should be a source of strength for the insured bank.
To appreciate the potential effects that TLAC might have, it is useful to look at the history of trust–preferred securities (TRuPS) at banks prior to the last crisis. At that time, regulators permitted firms to issue these instruments at the holding company and to include them as part of their capital. Although TRuPS are essentially long–term debt instruments, it was argued that they would be available to help absorb losses in a crisis, relieving the firms from having to hold equity. As the crisis unfolded, the weaknesses of TRuPS revealed themselves as firms often had to remove resources from the operating banks in an effort to move cash to the holding company and avoid default. TRuPS placed significant pressures on insured banks to continue making large debt–service payments which most certainly exacerbated losses to the banking system. As it turns out, Congress wisely eliminated the use of these instruments as part of industry capital when it passed Dodd–Frank.
Finally, the TLAC proposal, while increasing leverage, offers no assurance that the amount of debt required would prove sufficient to avoid a failure or panic. There is currently a debate about how much TLAC is enough to both absorb losses after equity is extinguished and also provide enough additional equity to assure the operating subsidiaries are well capitalized. The greater the possible losses, the greater amount of TLAC required, and of course firms want to hold less TLAC while some regulators judge the need for more.
Thus, a blanket rule mandating TLAC –– which requires financial firms to issue substantial additional debt and which increases the leverage in an already highly-leveraged industry –– should only be embraced with great reluctance. The long–term debt requirement would place added earnings demands on the banking system and could be counter–productive, especially during a period of financial stress.
Both regulatory and banking professionals should be extremely cautious in this regard as it is paradoxical to suggest that the best way to manage the effects of excess leverage and financial vulnerability is to require more leverage, potentially raising financial vulnerability. At a minimum, a guiding principle for TLAC when addressing the question of resolution should be that it have a neutral effect on the leverage position of the firm and the industry.
TLAC is closely tied to a SPOE resolution strategy to manage the failure of a GSIB. But there are other strategies that might also work and be less reliant on debt For example, in the U.S. an alternative strategy for certain banking firms could be a bridge bank in which the FDIC is appointed receiver of the insured depository institution while the holding company, and possibly some of its other subsidiaries, declare bankruptcy. The FDIC would place some assets and liabilities of the old bank in the receivership (bad bank) and the rest in the new bridge bank (good bank). This strategy would allow the critical operations of the bank to continue following failure, as the FDIC or its designee operates the bank until it is wound down, sold, or spun-off in accordance with the least cost to the taxpayer requirements. This solution could require less debt or may encourage more equity, depending on a host of factors. The point is that a one-size-fits-all solution is not necessarily the best outcome for the industry or the public.
The TLAC proposal would apply to GSIBs with little differentiation among them. Long-term debt dominates the proposal and very little allowance is given for equity held on the balance sheet. This seems short-sighted, since greater amounts of equity increase liquidity and reduce the likelihood of failure. Therefore, there might be the opportunity to reduce the long-term debt requirement based on the firm's resolution strategy and reliance on equity and debt for funding.
What this suggests ultimately is that the Title I resolution planning process should be the sole mechanism to define the resources and mechanism necessary to assure resolution without public support. Each firm and its regulators should identify debt and equity requirements that have a neutral effect on the firm and the industry's leverage positions, while also allowing for bankruptcy without public support should a firm fail. The process also could be used as the planning tool should, for whatever reason, Title II be the ultimate resolution outcome.
I would argue that such an approach is more judicious than requiring a sweeping TLAC rule. Under Dodd–Frank, each Title I plan is prepared in detail by management and reviewed and approved by supervisors. Each living will is intended to be based on the specific business model of each firm and the bankruptcy strategy that would work best for that firm. Thus, instead of applying a blanket long-term debt requirement to facilitate one resolution strategy, regulators could use the existing Title I planning process to tailor long–term debt levels to a firm's business model, capital structure, and resolution strategy, better assuring a private sector resolution should it be needed. And finally, as I have long urged, each firm's Title I plan, particularly its reliance on debt versus equity within the plan, should be made public, as this would enhance the market's role in assuring the right balance within the plans.
In bankruptcy, the conversion of standard debt is often part of workable reorganization strategy. But banking authorities should be cautious before demanding increased levels of debt for individual banking firms or encouraging an increase in the effective leverage of an industry.
It is difficult to accept that more leverage solves the destabilizing problem of leverage. If additional debt is useful to a specific firm, the amount should be determined on a firm–by–firm basis as part of the living will program, with the balance between debt and equity carefully judged. During this process, banking companies would have to show that debt could be successfully used in bankruptcy without weakening balance sheets or ultimately costing the taxpayer money to resolve a failure.
Adding leverage to the banking system in the hope that this will prove to be stabilizing is a gamble. There is considerable evidence that strategies that encourage increasing leverage within the banking industry have in past crises only exacerbated losses. The goal is resilience and, if necessary, resolution without government or taxpayer assistance. That is best achieved not by increasing leverage, but by requiring the right balance of debt and added equity for each firm and the industry as a whole.
The views expressed are those of the author and not necessarily those of the FDIC.
Thomas M. Hoenig is the Vice Chairman of the FDIC and the former President of the Federal Reserve Bank of Kansas City. His research and other material can be found at http://www.fdic.gov/about/learn/board/hoenig/
3 http://www.gpo.gov/fdsys/pkg/FR-2013-12-18/pdf/2013-30057.pdf. As of this date, the FDIC has not adopted this strategy.
4 Public versions of the resolution plans are available at http://www.federalreserve.gov/bankinforeg/resolution-plans.htm.
6 Public versions of the resolution plans of Wells Fargo and Bank of New York–Mellon: http://www.federalreserve.gov/bankinforeg/resolution-plans/wells-fargo-2g-20150701.pdf and http://www.federalreserve.gov/bankinforeg/resolution-plans/bk-ny-mellon-3g-20150701.pdf
8 This argument is often offered as a case against requiring more equity. However, equity is far less demanding on earnings, dividends and cash flow than fixed debt payments, especially under stress.
9 There is already some evidence that Wells Fargo is beginning to expand its investment banking activities: http://www.wsj.com/articles/wells-fargo-warms-up-to-risk-1428877356.
10 See FT, January 14, 2016, "Basel Committee softens new rules on bank capital," Laura Noonan, Mark Arnold and Caroline Binham.