Thomas Hoenig Read All of Basel III . . .

By James Kwak

. . . and doesn’t like what he sees. In a post for the Harvard Law School Forum on Corporate Governance and Financial Regulation, the former president of the Kansas City Federal Reserve Bank echoes some of the issues raised by Andrew Haldane, which I discussed earlier. The core problem, for Hoenig, is that Basel III “promises precision far beyond what can be achieved for a system as complex and varied as that of U.S. banking.” Banks were able to arbitrage the risk-weighted capital requirements of Basel II? Well, we’ll close all of those loopholes, one by one. But this cannot be done, given the incentives and power imbalances at work: “Directors and managers . . . will delegate the task of compliance to technical experts, and the most brazen and connected banks with the smartest experts will game the system.”

How do we know this will happen? Just look at history:

“Between 1999 and 2007, for example, the industry’s tangible equity to tangible asset ratio declined from 5.2 percent to 3.8 percent, and for the 10 largest banking firms it was only 2.8 percent in 2007. More incredible still is the fact that these 10 largest firms’ total risk-based capital ratio remained relatively high at around 11 percent, achieved by shrinking assets using ever more favorable risk weights to adjust the regulatory balance sheet.”

In other words, large banks increased their leverage (measured simply) while keeping their risk-weighted capital ratios constant. Conceptually speaking, this can only happen if their asset portfolios were becoming less risky over the period (1999–2007). Does anyone think this was actually happening?

In the long term, Hoenig ascribes the decline of capital levels in U.S. banks to the replacement of market forces by deposit insurance and regulators as guarantors of banks. Even if we shouldn’t go back to the unregulated days of the nineteenth century, he argues that we should go back to the capital levels that applied then. Otherwise, lower capital levels reflect a subsidy from the federal government to bank shareholders, who can take on greater risk because of the government safety net. (Those lower capital levels could also reflect the extent to which government regulation actually makes banks safer, which is something that can be debated.)

That implies a ratio of tangible equity to tangible assets on the order of 10 percent or higher—several times higher than required by Basel III. Some argue this will be bad for banks (an argument that Anat Admati has shredded previously). But another way to put it is that the capital levels prescribed by Basel III are very, very good for banks. Is that what we want?