(Originally published here.)
How much should banks be allowed to borrow? According to the Office of the Comptroller of the Currency and the Federal Deposit Insurance Corporation, the answer is “less.” Yesterday, the regulators proposed stricter equity capital standards for banks with at least $700 billion in assets.
Combined with similar new rules set by the Federal Reserve last week, this would be an improvement. But none of the regulators has gone nearly far enough to ensure that banks won’t bring us to a financial meltdown like the one that threatened the global economy in 2007 and 2008.
Understandably, bankers like to exploit this situation by taking as much as risk as possible. When their bets pay off, they get to keep the winnings. When they misjudge their risks, it creates losses that have to be distributed somewhere.
For a variety of reasons, governments don’t like dumping these losses directly on the banks. The result is an enormous subsidy paid by the rest of society. (Bloomberg View’s editors have written about this subject on occasion.) The Dodd-Frank legislation promised to eliminate the subsidy by banning bailouts, although few believe that this promise is credible in a hugely complex, globally interconnected financial system.
Given this reality, financial reformers have rightly focused on finding ways to remove the large subsidies banks extract. The best way to do this is to ensure that bank shareholders and bank employees bear the risk of bank losses. I’ve previously argued that all banks should fund 100 percent of their assets with equity. You can’t lose more than 100 percent of what you start with in the absence of borrowed money. (Derivatives can make this slightly more complicated but aren’t an insurmountable issue).
So far, U.S. regulators have been more modest: Large bank holding companies should fund at least 5 percent of their total assets with Tier 1 capital, while insured depository subsidiaries must fund at least 6 percent of their total assets with Tier 1 capital. (Note that these “leverage ratios” are not strictly comparable to the 3 percent rule in the latest Basel Accords, contrary to some news accounts.)
Crucially, if the new “generally applicable leverage ratio” falls below 5 percent, regulators are empowered to restrict bonus payments, dividends and share buybacks. That’s good for two reasons. First, it’s much easier to boost equity capitalization by retaining earnings rather than going to the market and issue new shares. Second, this rule has teeth: The real risk-takers in most banks aren’t usually in senior management; rather, they often operate in the middle of the org chart. They may or may not care about the share price of their employers, but they are acutely sensitive to the prospect of a lower bonus.
My hope is that these new rules are only the beginning of a much stricter regime. In particular, the renewed focus on leverage ratios should offset the weaknesses of the existing rules governing the ratio of capital to “risk-weighted” assets. While intellectually defensible, the concept of risk-weighting for regulatory purposes is deeply flawed.
For example, in a new paper, the Bank for International Settlements studied the variations in risk-modeling techniques used by big global banks and found enormous deviations. The exact same set of hypothetical assets and liabilities generated regulatory capital ratios as low as 5.9 percent and as high 15.7 percent, depending on the risk model used. That kind of variation might be tolerable if equity capital requirements were higher, but the current low level creates too much room for error.
The higher the minimum leverage ratio, the less likely that banks will extract subsidies and impose losses on the rest of the economy. So let’s hope the new rules are just the first of many steps in the right direction.
(Matthew C. Klein is a contributor to the Ticker. Follow him on Twitter.)