Hensarling leverage ratio plan unworkable without stress tests
Bank regulatory capital requirements have become way too complex, and Rep. Jeb Hensarling, chairman of the House Financial Services Committee, wants to simplify them. His idea (as part of the proposed Financial Choice Act) is to permit a big bank to maintain a 10% leverage ratio to qualify for exemption from many of the other prudential banking rules. A bank that meets this 10% ratio would be relieved of the various capital surcharges and many other regulatory complexities to which big banks now are subject.
Hensarling has the right basic idea but the wrong benchmark.
The bill appears to establish the 10% leverage ratio as the mark of a bank whose strength is beyond dispute. Ten percent is indeed a higher leverage ratio than large banks have maintained historically. But it tells us nothing about the riskiness of the bank’s business or the size of its exposure to economic downturns. Indeed, standing alone, a leverage ratio based on historical financial statements tends to induce banks to take extra risks of the sort that cause losses in economic downturns, especially downturns that severely affect real estate values.
The Federal Deposit Insurance Corp. has used leverage ratios as its key capital data for many years, but its experience demonstrates that a capital regime based on historical data does not work, in part because it does not look at what happens to the bank’s assets in a severe downturn.
By contrast, a regime based on forward-looking, stress-tested capital has been successful in the short time that it has been in effect. And it works logically to take account of the particular risks and capital level of each individual bank without preconceived risk weights or assumed business plans.
Before 2008, the capital ratios banks maintained — based on historical financial statements — may have appeared to be adequate. But history becomes irrelevant in a bad recession when capital is most needed. Regardless of what amount of capital a bank may have had in good times, it must maintain a ratio in bad times that is high enough for the market not to lose confidence in the bank and create a run. And it is exactly in those bad times when a bank cannot raise more capital at a reasonable price, and when assets formerly thought to be liquid become difficult to sell without significant loss.
The effective way to simplify banking regulation while maintaining bank soundness would be to accept that stress testing — admittedly a complex and expensive process — is the best way (that we know of) to measure the amount of bank capital needed when it really matters. A stress-tested capital benchmark (not a historically-based capital benchmark) should qualify banks for significantly reduced regulatory scrutiny.
The stress-tested benchmarks should be set high enough that the bank still will be perceived as strong in the bad times. But since stress tests evaluate the bank’s actual risks, which a flat leverage ratio requirement cannot do, a stress-tested capital standard could be significantly lower while still being more effective than Hensarling’s 10% standard.
The stress-tested ratios could be, in my opinion, a 5% to 6.5% leverage ratio and an 8% to 10% risk-based ratio. These ratios would be higher than the post-stress-test minimums the Federal Reserve Board now applies, but that increase would be justified by the vast amount of regulations and intrusiveness that could be eliminated for banks that met the test consistently.
Smaller banks could qualify for the same streamlined regulatory treatment if they achieved the same results under a simplified stress test using the same severe economic scenario presented to larger banks.
Since the 1980s, U.S. commercial banks have been required to maintain minimum levels of regulatory capital, both in terms of a simple leverage ratio — that is, net worth to total assets — and in terms of risk-based assets. But these requirements have not been effective in enabling the regulators to predict which banks are likely to fail very far in advance or to assist the regulators in changing a bank’s conduct when the regulators have believed the bank to be taking excessive risks. Traditional regulatory capital requirements have failed for two reasons: One, they are based on historical financial statements, and two, examination findings are secret and are not self-executing.
In 2009, after the financial crisis, the U.S. government was able to draw a line that ended the crisis by conducting stress tests of 18 large American banks. The banks that passed (most of them) got a clean bill of health. The banks that failed were told that they must raise new capital — and they did so promptly.
The Federal Reserve Board built on the success of the 2009 stress tests by creating the Comprehensive Capital Analysis and Review for the nation’s largest banks. The most important feature of the CCAR stress test is that a bank must demonstrate that it has adequate capital after nine calendar quarters of performance under a severely adverse economic scenario. If the bank’s capital is projected not to be adequate at any time during the nine quarters, then the bank must submit a revised capital plan that shows adequate capital throughout the period.
The CCAR stress testing regime has three great benefits.
First, it does not depend on historical financial statements. Only the stress-tested balance sheet counts. Secondly, enforcement is self-executing. A bank that fails the test must submit a new, compliant capital plan. Finally, the results are made public, so stockholders, customers and creditors can see how the bank is likely to perform in a deep recession.
This type of forward-looking regulatory capital regime works, even though we have only a few years of experience with it. It has flaws — as a complex process, it always will have flaws — but those flaws are being dealt with, including by increasing transparency to enable the public to comment on the Fed’s procedures.
In international markets, big American banks have outperformed big banks from anywhere else in the world since passing the initial stress tests in 2009. That is not a coincidence. Strong banks attract deposits, are desirable counterparties and make more loans. Rigorous, believable stress tests have become the surest indication of a strong banking system that can compete globally and serve national markets that need a safe place for people’s money and credit for individuals and businesses.
The crisis may have resulted in regulatory provisions that are unnecessary or excessive. Stress testing is not one of them. Indeed, stress testing is the best path to reducing regulation while achieving effective prudential supervision of banks and their holding companies.