What Does 9.5% Mean?

By James Kwak

This week, the Federal Reserve approved its final rule setting capital requirements for banks. The rule effectively requires common equity Tier 1 capital of 7 percent of assets (including the “capital conservation buffer”), with a surcharge for systemically important financial institutions that can be as high as 2.5 percent, for a total of 9.5 percent. That sounds like a lot, right?

If it sounds like a lot to you, it’s probably because (a) it’s higher than capital requirements before the financial crisis and (b) the banking lobby has been saying it’s a lot to anyone who will listen. But apart from some people thinking that higher is better and others thinking that lower is better, you rarely get any basis for understanding what the numbers mean.

In school, you learn that a 9.5 percent capital ratio means that a bank can sustain a 9.5 percent fall in the value of its assets before it becomes insolvent. But that’s clearly not true, for multiple reasons. First, if a bank were to report that its capital fell from 9.5 percent of assets to 2 percent of assets, it would fail the next day as all of its short-term creditors pulled out their money in (justifiable) panic. In other words, it’s not clear how much of that capital buffer is really a buffer.

Second, that’s 9.5 percent of risk-weighted assets. We all know what risk-weighting means in theory, but few people have a firm grasp on what it means for the actual numbers. As of September 2012, for example, Goldman Sachs had $949 billion in balance sheet assets, but only about $436 billion in risk-weighted assets—although that would increase to $728 billion under new risk-weighting rules. As an illustration, if risk-weighted assets are only half of actual assets, then a 5 percent drop in asset values would be enough to wipe out a 9.5 percent capital buffer.

Third, and most important, there’s measurement error. As many have noted before, Lehman Brothers had something like 11 percent Tier 1 capital two weeks before it went bankrupt. What this means is that, as Steve Randy Waldman said and as I discussed in an earlier post, “Bank capital cannot be measured.”

Given systemically important financial institutions and imperfect regulations, capital requirements have an important role to play. But we should be setting them with the understanding that banks fail before they run out of capital, capital is difficult to measure, and the errors all come out the same way—in the banks’ favor. In practice, of course, it’s all politics: even if Daniel Tarullo wants higher capital requirements, there’s a limit to what he can get through the Board of Governors, and there’s a limit to what Ben Bernanke thinks he can get while remaining an independent agency. That’s the bottom line to remember—not that our new capital requirements are the outcome of some reasoned discussion about how much capital banks really need to protect the rest of us from their misadventures.