Two Cheers for Basel III…or Waiting for Basel 3.5

Citibank chief Vikram Pandit warns in the Financial Times against excessive increase in capital and liquidity requirements.  This is a tough argument to make in the aftermath of a serious crisis in which banks and other financial institutions were caught short on both counts.

The op-ed below was co-authored with Masami Imai,  who is is Associate Professor of Economics and Chair of the East Asian Studies Program at Wesleyan University.  It is based on a longer working paper we presented at the Economic History Association meetings in September and policy paper we published at Voxeu.

******************************************************

The recent endorsement by regulators from across the industrialized world of the more stringent bank capital requirements proposed by the Basel Committee on Banking Supervision was good news.

But it was only a first step in shoring up the world’s banks.

Rolled out almost exactly two years after the spectacular collapse of Lehman Brothers, the new, tougher capital standards, dubbed Basel III, further tighten the standards agreed upon in the Basel and Basel II accords.

Tougher capital requirements promote bank safety and soundness in two important ways.

First, capital provides an important security for creditors.  Capital represents an important reserve that creditors can claim if the bank stumbles.  Further, if the bank suffers a shortfall in cash flow, it can easily cut the dividend on equity; suspending interest payments, however, poses an existential threat.

Second, because capital represents the equity of bank owners—sometimes referred to as “skin in the game”—greater capital should discourage excessive risk-taking.  Additionally, it increases the confidence of bank creditors, reducing the likelihood of panic withdrawals.

Bankers ought to have more skin in the game though.  With the current limited liability system, bankers themselves have far less at risk than in the days of the unlimited liability bank or investment bank.

Although going back to unlimited liability is not practical, there is a middle ground between the unlimited liability system that was common in the nineteenth century (and even, among a few venerable institutions) and into the late twentieth and our current limited liability system.

Unlimited liability is impractical for most large financial institutions today.  Those with even a modest nest egg would be reluctant to invest even a small part of it into such an institution, since failure—even for reasons that had nothing to do with excessive risk-taking—could mean personal financial ruin.   Shares would instead end up in the hands of “widows, orphans, and other impecunious individuals” who would not have the resources to settle the debts of a failed bank.

A second possible solution is to allow firms to issue shares with some additional liability.  This was popular with British firms before World War I.  A company might issue shares with a “nominal” value of £20, of which, say, only £12 was “paid-in.”  This meant that if the firm needed more capital—for expansion, or for liquidation–it could call upon shareholders to pay in an additional £8 of “uncalled liability” per share.  Many US banks in the 19th and early 20th century operated under a system of “double” or “triple” liability, in which shareholders were liable for an additional defined amount over and above what they had paid for their shares

Our research on the American and British historical antecedents indicates that banks that operated with some form of extended liability—that is, with more skin in the game—undertook less risk than those that operated under conventional limited liability.  Although such arrangements did not eliminate financial crises in either country, they clearly mitigated some of the worst tendencies of the financial system.

An element of extended liability should be part of Basel 3.5.

This entry was posted in Financial Crisis, financial regulation and tagged , , , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *