Getting Serious about Ending Bank Bailouts
This article originally appeared in Forbes.
On July 23 – roughly five years from the date the Dodd-Frank bill was passed – I had the opportunity to testify before the U.S. House Committee on Financial Services about whether we have fully eliminated the risks of bank bailouts, and particularly the “too-big-to-fail” problem. My full written testimony is available here. My oral testimony follows.
Chairman Hensarling, Ranking Member Waters, other Members, it’s a pleasure and honor to share my thoughts with you about addressing the too-big-to-fail (TBTF) problem, and more generally, the problems of bank instability, credit collapses, and financial burdens on taxpayers that result from private risk taking at public expense.
Title II of Dodd-Frank is supposed to ensure orderly liquidation of TBTF banks (now called systemically important financial institutions, or SIFIs), but is more likely to institutionalize bailouts by establishing specific procedures through which they will occur.
Rather than pretending that we will have the legal mechanisms and political will to liquidate SIFIs, we should focus on preventing them from becoming insolvent. That means focusing on the adequacy of bank capital and cash.
Book equity is a poor measure of the true value of equity. When banks suffer losses on tangible assets (such as loans) they typically delay loss recognition; overstating equity capital allows them to avoid curtailing risky activities.
Furthermore, the book value of equity does not capture losses of intangible assets: lost servicing income, other fee income, and reduced values of relationships with depositors and borrowers have been the primary drivers of loss in bank value since 2006.
We should raise equity capital ratio requirements, but we cannot rely only on book equity ratios to measure bank health. We need to measure the economic value of equity, and put in place reliable regulatory requirements that ensure banks will maintain adequate and meaningfully measured equity capital.
I propose requiring, alongside a book equity requirement, that large banks maintain a substantial proportion of funding in contingent convertible debt (CoCos) that converts into equity on a dilutive basis when the market value of equity persistently falls below 10% of assets. Dilution ensures that bank managers face strong incentives to replace lost equity in a timely manner, to avoid dilutive conversion of CoCos.
Bank CEOs would have an incentive to maintain a significant buffer of equity value above the 10% trigger. They would increase that buffer voluntarily if the riskiness of banks’ assets rose, resulting in a self-enforcing risk-based equity requirement based on credible self measurement of risk, in contrast to the current system of risk measurement gaming by banks.This CoCos requirement would virtually preclude SIFI bailouts – bailouts cannot occur if banks remain distant from the insolvency point.
Additionally, stress tests could be a promising means of encouraging bankers to think ahead. As they are structured, however, stress tests are a Kafkaesque Kabuki drama in which SIFIs are punished for failing to meet unstated standards. That not only violates the rule of law, the protection of property rights, and adherence to due process, it makes stress tests a source of uncertainty rather than a helpful guide to identifying unanticipated risks.
And the penalties for failing a stress test are wrong. Limiting dividends makes sense for a capital impaired bank, but not for a healthy bank in compliance with all its regulatory requirements. In that case, it is an inappropriate incursion into the decision making of the board of directors.
Finally, the stress testing standards currently being applied by the Fed are not very meaningful.
We can do better.The Fed should be required to provide clear guidance about how the risk of loss will be estimated. Stress tests should be an input into capital requirements, not used to control dividend decisions of healthy banks. Finally, stress tests should focus on the loss of economic value, by analyzing consequences for bank cash flows, divided by line of business, using data from bank managerial accounts.
Liquidity requirements are another good idea being implemented poorly. A better, simpler approach would require SIFIs to maintain reserves at the Fed of 25% of their debt. To avoid turning that into a tax, reserves should bear market interest. This would require banks to hold a significant proportion of their assets in riskless debt. This would not bind on SIFIs today, given their huge excess reserve holdings. Nor would it have been binding in the early 1990s. But it would have been very helpfully binding on SIFIs leading up to the recent crisis. Large banks held 25.8% of their assets in cash plus treasuries plus agency securities in January 1994. That percentage fell to 17.2% by January 2001, and to 13.5% by January 2008.
Applying to SIFIs the right combination of regulations governing book equity, CoCos, stress tests and reserves would virtually eliminate the risk of too-big-to-fail bailouts. But that is not the only bank bailout risk we face. The most important source of systemic risk for small banks – one that was visible both in the 1980s and in the 2000s – is their excessive exposure to real estate lending. Real estate risk is highly correlated and hard to shed in a downturn. As of January 2008, roughly three-quarters of small bank lending was in real estate loans. Large banks had lower exposures, but they held real estate loans on balance sheet equal to 32.6% of total assets, and that excludes MBS exposures, on and off balance sheet.
The obvious answer is to limit bank real estate lending, forcing real estate financing to migrate to REITs, insurance companies, and other more natural providers of real estate finance. Banks would become more focused on lending to small and medium-sized enterprises.
In summary, for SIFIs, I suggest a 10% book equity-to-assets minimum requirement (alongside a 15% book equity-to-risk-weighted assets requirement), a 10% CoCos issuance requirement with a market-informed conversion trigger (as described above), a 25% remunerative cash reserves-to-debt requirement, and a stress testing regime that is transparent, disciplined and focused on bank cash flows. In addition, for all banks, I propose limiting real estate lending.
These reforms not only would virtually eliminate the too-big-to-fail problem; they would stabilize the entire banking system, protect taxpayers, reduce regulatory uncertainty, and improve the performance of banks.
Charles W. Calomiris is a member of the Shadow Open Market Committee and the Henry Kaufman Professor of Financial Institutions at Columbia University.
Interested in real economic insights? Want to stay ahead of the competition? Each weekday morning, e21 delivers a short email that includes e21 exclusive commentaries and the latest market news and updates from Washington. Sign up for the e21 Morning eBrief.