Regulate bank behavior, not capital, to prevent next crisis
Since the financial crisis, bank capital has doubled at U.S. banks — and yet it’s unclear that the financial system is any safer as a result.
The Federal Reserve and other regulators are now deciding whether to slightly lower those capital levels, spurring debate among Washington policymakers and academics.
Yet first and foremost, it needs to be accepted that higher capital levels for banks have been embraced by regulators as a fix for a far more complex problem, namely financial fraud. None of the banks that failed in 2008 stumbled due to inadequate capital. Instead, banks such as Citigroup, Lehman Brothers, Bear Stearns, Washington Mutual, Countrywide and Wachovia all failed or were acquired because of a lack of liquidity tied to an equally large deficit of confidence.
When uncertainty regarding the liabilities of these banks caused investors to run, the banks died. Investors could no longer ascertain the noncapital liabilities of these large, market-facing financial institutions — as a result, liquidity disappeared and the banks collapsed. The fundamental truth of the 2008 crisis is that confidence is more important than capital, but you will never hear a regulator or member of Congress say that in public.
Another one of the fallacies that drives the fixation on bank capital among regulators and policy analysts is the idea that large capital buffers absorb losses in times of economic weakness. In fact, income is the chief measure of whether a bank can absorb credit losses. Contrary to statements from former Federal Deposit Insurance Corp. officials Sheila Bair and Tom Hoenig that U.S. banks consumed “double their equity and required trillions of dollars in liquidity and other assistance backed by taxpayers to bail them out and stabilize the economy,” the banking industry in fact consumed virtually no additional capital except among the banks that actually failed.
Following the 2008 financial crisis, the banking industry diverted tens of billions of dollars in operating income to provisions for future credit losses. The industry prepaid four years of deposit insurance assessments to fund the FDIC’s resolution of failed banks, but virtually the entire cost of the cleanup was paid for out of operating income — not capital, as so many regulators and policy analysts falsely suggest. Indeed, a large part of these provisions for credit losses were not needed for the industry cleanup and were eventually recaptured back into bank earnings.
While many banks suspended common equity dividends during the post-2008 cleanup, the only banks that actually consumed capital to offset credit losses were banks that were actually taken over by the FDIC. Seen from this perspective, the chief use of bank capital is to reduce losses to the FDIC’s bank insurance fund when a depository runs out of cash. From the perspective of the credit markets, the chief indicators of the probability of default of a bank are operating income and reputation, not static levels of capital.
The actual results in 2008 reveal the reality. The largest banking institutions were too systemically important to be taken into FDIC receivership (or were otherwise impractical to shutter). Four of the 10 largest banks at year-end 2017 — 44 % of the industry by asset size — had to be rescued by stronger institutions, or, in the case of Citigroup, required extraordinary governmental support to remain afloat.
Seen from the perspective of risk and liquidity, the Dodd-Frank Act provisions such as the Volcker Rule are far more important to moderating the risk-taking of large banks than are the mechanistic capital rules maintained by prudential regulators. If you understand that the 2008 crisis sprang from the ancient well of financial fraud, including the making of bad loans and the sale of bad securities by Wall Street, then the issue of capital falls into its appropriate, but secondary, place behind limits on the activities of banks, including their behavior in a professional and ethical sense.
Advocates of higher capital levels commit another error when they suppose that constraining the available resources of a bank via higher capital rules and other restrictions somehow makes a bank safer. In fact, the current regime rewards banks for taking greater risks with their remaining available net liquidity because of the need to achieve equity and asset returns demanded by investors. Thanks to the 2017 tax bill, bank equity returns to investors have risen by a third, but pretax returns on assets and equity are still 30% below 2008 levels.
The Federal Reserve Board and Office of the Comptroller of the Currency are correct in their desire to understand capital as a function of the risk taken, but ultimately, modeling such a measure is extremely challenging. The best insurance against bank failure, especially during periods of high stress, is to moderate bank risk-taking by enforcing the laws against financial fraud.
The failure of more than half of the large U.S. banks by assets in 2008 had nothing to do with capital, but everything to do with confidence. When we accept that confidence in banks is more a function of the behavior of bankers than the accumulation of capital, then we will have a system of prudential regulation that is effective and supported by investors and the public.