They're too timid, I argue in my new FT column.
The proposed new ratios strike a compromise between countries (including the US and Britain) that wanted tougher rules and others (notably Germany) that pressed for less stringency. The current minimum ratio of common equity to risk-adjusted bank assets is just 2 per cent - a figure that, with hindsight, should be viewed as laughable. This is to rise to 7 per cent under the new rules, including a proposed buffer of 2-3 per cent. When capital falls below the buffer zone, banks would have to curb pay and/or dividends.
A recent paper by Samuel Hanson, Anil Kashyap and Jeremy Stein underlines a crucial point: to be any use, the regulatory minimum capital ratio in good times must substantially exceed the market-imposed standard in bad times: "Thus if the market-based standard for equity-to-assets in bad times is 8 per cent, and we want banks to be able to absorb losses on the order of, say, 4 per cent without pressure to shrink, then the regulatory minimum for equity-to-assets in good times would have to be at least 12 per cent." The authors add that 4 per cent is a conservative estimate - cumulative credit losses at US banks between 2007 and 2010 were roughly 7 per cent of assets.
Banks have been emphasising the costs of even modestly higher capital ratios. They have a point but they exaggerate. Estimates by the Bank for International Settlements and others suggest that the penalty in higher lending costs and lower economic growth would be small, and outweighed by the benefit of fewer crashes.
The real danger in capital minimums that bind when they need to is not lower growth but regulatory arbitrage. The better the system works for regulated financial firms, the greater the danger that business will move to the unregulated. An appropriately strict regime for banks and other covered institutions would require new rules for the rest of the shadow banking system - including minimum collateral requirements for asset-backed securities regardless of the buyer.
We want to hear what you think about this article. Submit a letter to the editor or write to firstname.lastname@example.org.