On behalf of Sandler O’Neill + Partners, L.P., we would like to share our thoughts on the Committee’s Consultative Documents Strengthening the Resilience of the Banking Sector and International Framework for Liquidity Risk Measurement, Standards and Monitoring, issued for comment December 17, 2009.
Sandler O’Neill is a full-service investment-banking firm focused on the U.S. financial services sector, particularly insured depository institutions. Our clients include a wide variety of financial firms, among them hundreds of community and regional banks and thrifts and their holding companies. As a firm of financial professionals who work closely with many banking firms, Sandler O’Neill frequently comments on supervisory and other issues important to our clients.
The Committee’s proposals comprehensively address important supervisory concerns arising from the recent international economic and financial crises. These include improving the quantity and quality of capital; adding an international leverage capital requirement to supplement a more comprehensive risk-based framework; adding a complementary but independent liquidity requirement; and attending to macro- as well as micro-prudential considerations, including the procyclicality of bank capital and liquidity regimes.
While we support the Committee’s goals, we believe the proposals would do more harm than good if implemented in their current form. Specifically, they would significantly deleverage the banking system, dramatically reduce industry profitability, exacerbate balance sheet and earnings volatility, and undermine capital formation for the industry. In turn, these unintended consequences would severely damage the international financial system and the national economies that depend upon it. For these reasons, we suggest a number of improvements and alternative approaches to the Committee’s proposals.
Financial Intermediation as Touchstone
We believe that thinking about the role of capital and liquidity requirements in bank supervision is most productive when grounded in the realities of financial intermediation, which remains the core competence of all banks, even large, internationally active ones.
Banks stand between the providers and users of credit. Because providing the service of credit intermediation is a low-margin business, banks must be thinly capitalized to be profitable and vigilantly manage risks that could overwhelm capital and liquidity. Unavoidable concentrations of various sorts, including geographies and industries, compound the difficulty of this balancing act, which is an art rather than a science.
Supervisors must therefore focus first on how well bankers manage risk rather than how much capital or liquidity banks have, because they can never have enough absent watchful risk control. Thus, supervisors must have zero tolerance for incompetent management, an imperative made more important by the moral hazard arising from deposit insurance and the belief that certain financial institutions are too big to fail. As well, they must be ever alert for signs of emerging systemic risk.
Finally, bank supervisors must emphasize the safety and soundness of individual firms over potentially conflicting macro factors such as the availability of credit, for such considerations are capable of undermining good bank supervision, particularly when politically motivated.
Hard Cases Make Bad Law
There is a legal maxim that “hard cases make bad law.” It posits that difficult or unusual facts provide a poor basis for a law or rule of general application that must cover a wider range of less extreme circumstances.
In this regard, it is worth recalling that the abating financial crisis was not a tremor, nor even an earthquake, but rather a cataclysmic realignment of tectonic plates that no bank capital or liquidity regime could have anticipated, avoided, or handled. What is required in such circumstances is what we were lucky enough to get: supervisory improvisation of a high enough order not to be wrong or desperate too often.
To lose sight of this is to run the risk that the primacy of the essential work of roll-up-your-sleeves supervision will be forgotten in the rush to enshrine a complex international capital and liquidity framework promising far more than it can deliver, and at a considerable cost to banks and the economies they serve.
The implications of these realities for reforming the supervisory capital and liquidity framework are threefold: capital and liquidity requirements should be as simple as possible, supervision must be robust, and central banks must stand ready to backstop the banking sector to keep crises from spiraling out of control.
Tangible Common Equity Focus Too Narrow
To enable Tier 1 capital to absorb losses and still allow a bank to remain solvent, the Committee proposes to refocus capital adequacy assessment on a ratio of tangible common equity to risk weighted assets, seeking to improve both the quality of the numerator and the inclusiveness of the denominator.
The proposed capital measure in the numerator would exclude mortgage servicing rights, core deposit intangibles, deferred tax assets dependent on future earnings, and minority interest. The Committee also proposes a mirror-image deduction from capital for a bank’s investment in other banks as well as for any shortfall in provisioning for estimated future loan losses. Some analysts estimate that these deductions could total as much as $200 billion for U.S. banks, which, assuming 10% capital levels, could cause a deleveraging of the U.S. banking system by as much as $2 trillion, which would be an unacceptable restraint on the economy.
The Committee also proposes to eliminate trust preferred securities as a form of Tier 1 capital. Because about $129 billion of trust preferred securities are outstanding in the United States, their elimination from Tier 1 capital could prompt a sharp deleveraging of bank balance sheets by as much as $1.3 trillion. To avoid this additional shock to banks and the economy, trust preferred securities should remain an acceptable form of Tier 1 capital or, worst case, should be grandfathered and phased out of capital over some intermediate time period.
Given the variations in tax laws and accounting conventions across jurisdictions, banks need the flexibility to issue a variety of capital instruments that count as Tier 1 capital, including common stock, preferred stock, and trust preferred securities. For example, in Europe some forms of Tier 1 eligible preferred stock can be structured to be tax deductible. By contrast, preferred stock is not tax deductible in the United States, although trust preferred securities are. Another problem is the strict ownership limitations imposed on common equity in the United States, which hamper the ability of small and mid-sized banks to raise substantial amounts of common equity, thereby placing them at a competitive disadvantage compared to larger capitalization banks. Trust preferred and preferred stock have no such voting ownership limitations. For reasons such as these, having a variety of Tier 1 capital instruments is important for all banks, but particularly for thousands of small and mid-sized U.S. banks, which generally have access to fewer sources of capital.
The proposed exposure measure in the denominator would limit netting of counterparty exposures, including off balance sheet assets, which would be added to risk weighted assets applying a 100% credit conversion factor. The reduced capital measure and enlarged exposure measure would create the need for substantial additional capital if the calibration of such ratios were not carefully managed and phased in over an extended period.
In short, neither the capital measure nor the exposure measure can be discussed in a vacuum. Rather, they must be considered within a framework that correlates them each with the other as well as with a baseline minimum required ratio. The more restrictive the component parts of the ratio are, the lower the required minimum ratio will have to be. Adjustments will have to be made to require a baseline minimum ratio that provides meaningful capital discipline without so constricting banks that they cannot profitably perform their critical function of credit intermediation within the financial system.
Better Alternatives to Capital Deductions
Some of the proposed deductions from common equity or Tier 1 capital may be better handled through adjustment to risk weighted assets. For example, mortgage servicing rights represent an important asset category for U.S. banks active in mortgage lending. The deduction of such assets from common equity would likely reduce availability of credit and exacerbate the current pressure on U.S. home prices.
A higher risk weighting of mortgage servicing rights could better recognize the higher risk of their income stream but not unduly punish the providers of mortgage credit so important to the U.S. economy. Similarly, applying higher risk weights to investments in other banks, core deposit intangibles, and deferred tax assets would address the more volatile asset valuations without forcing disruptive deleveraging.
Recognizing that the greatest threat to a bank's balance sheet is deterioration in asset quality, the Committee has proposed that shortfalls in provisioning for expected losses be fully deducted from the common equity component of Tier 1 capital. The difficulty of quantifying such shortfalls is but one of many implementation problems.
An alternative approach to aligning capital with embedded credit risk is now being employed by U.S. supervisors, who are requiring many banks to limit the ratio of adversely classified assets to Tier 1 capital plus reserves to less than 50%. The loan classification system is well understood, and such a capital measure would provide a workable, comprehensive metric.
Liquidity Measures “Not Ready for Prime Time”
The liquidity coverage ratio and net stable funding ratio attempt to provide objective, measurable standards of minimum liquidity. However, they have numerous problems that must be addressed and re-calibrated to assess the advisability of adoption.
These problems include the following: the sources of liquidity are not defined broadly enough; the required liquidity for unfunded commitments and backup credit facilities are not consistent with actual experience in the recent crisis and would constrain lending if adopted; runoff assumptions for core deposits are not consistent with actual experience in the recent economic downturn; and disallowance for agency mortgage backed securities and held-for-sale mortgages would essentially eliminate liquidity for mortgages.
Finally, excess liquidity requirements could prompt the re-emergence of an unregulated shadow banking market because such requirements would prevent banks from achieving attractive returns. One industry observer has estimated that the proposed ratios could require banks to raise up to $3.3 trillion globally in medium-term funding, which the capital markets would likely have difficulty absorbing. To the extent that the proceeds from such a liquidity raise could not be invested in higher yielding loans, bank earnings would be reduced, thereby making it more difficult for banks to raise equity capital.
Procyclical Perils Exacerbated
It is well known that financial markets are procyclical in that they tend to amplify the business cycle, and that regulatory capital requirements currently exacerbate procyclicality.
However, the proposed move to forward-looking loss provisioning is problematic for several reasons. The use of expected loan losses versus historical experience to set loan loss reserves creates the potential for timing differences in measurement of risk, creating additional deferred tax assets that would reduce Tier 1 capital a second time after provisioning has already reduced it once. The problem of inconsistent approaches among tax, accounting, and banking authorities in the United States is compounded by securities regulators, who have criticized banks for “earnings management” as a result of reserving against potential loan losses. Finally, forward-looking provisioning would eliminate the ability to use earnings from a loan portfolio to support future losses, creating a mismatch between revenues and expenses.
We think the concern over procyclical provisioning is better addressed through the use of the previously mentioned ratio of adversely classified assets to Tier 1 capital plus loan loss allowance, which is intuitive and workable, and already actively used by U.S. supervisors in their regular exams and reviews.
The proposal that unrealized gains and losses recognized on the balance sheet be included in the common equity component of Tier 1 capital would amplify, not moderate, the procyclical effects of regulatory capital. The suggestion is useful in raising pointedly the question of what role general accounting principles such as GAAP or IFRS should play in regulatory accounting principles crafted for the safety-and-soundness supervision of banks.
Alignment of Business Models, Accounting, and Supervision Required
Banks are neither mutual funds nor hedge funds. In the United States the regime of capital requirements and supervision, including principles of regulatory accounting and reporting, has evolved to assure that they remain what they are: financial intermediaries that pursue lower-risk lines of business in exchange for federal deposit insurance and relatively modest, reliable returns. Recent history is replete with cautionary tales of institutional investors that mixed – to their great grief and sudden demise – the combustible ingredients of highly leveraged balance sheets, risky business strategies, and the pursuit of total return.
Bank supervisors must reject the indiscriminate application of fair value accounting to bank operations. The more extensively banks are forced to apply fair value accounting to their core competence of credit intermediation, the greater will become the pressure to manage to total return rather than the spread relationships between earning assets and funding liabilities that traditionally have characterized the management of their balance sheets. Such a shift in thinking from spread management to a trading account mentality would be foreign to most bankers. Therefore, the debate over accounting principles appropriate to bank supervision is neither academic nor inconsequential, for upon its outcome hinges nothing less than the safety and soundness of banks.
In conclusion, while we applaud the Committee's efforts to constructively address the financial crisis by prescribing comprehensive reform of the capital and liquidity requirements for banks, we think the proposals as structured would do more harm than good. They would significantly deleverage the banking system, dramatically reduce industry profitability, exacerbate balance sheet and earnings volatility, undermine capital formation in the industry, and drive traditional credit intermediation into the unregulated shadow banking market. We have proposed a number of adjustments to the architecture of the Committee’s proposals which, if implemented, would capture the spirit of change desired while dampening the unintended negative consequences.
Joseph Longino Thomas W. Killian
cc: The Honorable Timothy F. Geithner
Secretary of the Treasury
U.S. Department of the Treasury
The Honorable Ben S. Bernanke, Chairman
Board of Governors of the Federal Reserve System
The Honorable Sheila C. Bair, Chairman
Federal Deposit Insurance Corporation
The Honorable John C. Dugan
U.S. Comptroller of the Currency
The Honorable John E. Bowman, Acting Director
Office of Thrift Supervision