« "Fixing the Broken Government Policy Process" | Main | "Underwater, But Will They Leave the Pool?" »

Saturday, January 23, 2010

"A Modest Proposal" to Reform How Bank Executives are Paid

Axel Leijonhufvud has a proposal that attempts to overcome the "I win, you lose" nature of banking by imposing a double liability scheme that forces managers to pay for their own mistakes.

He does not believe that attempts to limit leverage, something I've been advocating, will be effective since the rules can always be "circumvented by highly motivated and very smart people." However, this objection applies to his proposal as well, something he acknowledges parenthetically toward the end of his argument. So do both, restructure banking pay along the lines suggested below or along the lines suggested elsewhere, and also impose rules that attempt to bound leverage. Then hope that these measures, in combination with other policies, will have the intended effects:

A modest proposal, by Axel Leijonhufvud, Vox EU: The crisis has shown that the commercial banks, the investment banks, the shadow banks, and sundry other financial institutions have assumed too much risk. It has also been demonstrated that when risks do materialise, the costs are borne by the taxpayers and the unemployed – not by bankers. The bankers have been playing “I win, you lose” with the general public. What happened to the capitalist system where everyone was supposed to pay for their own mistakes, not for those of others?

Proposed reforms: Less than reassuring

The reforms proposed to deal with this problem have, by and large, been limited to reformulated capital requirements for banks and macroprudential supervision of the system as a whole. To date, the ideas advanced are less than reassuring.

  • The Basel capital requirements were actually procyclical. That problem will presumably be fixed somehow. But it is quite another matter to formulate capital requirements that cannot be circumvented by highly motivated and very smart people, and that will effectively constrain leverage in the financial system as a whole.
  • As for macroprudence, a reliable methodology does not yet exist. Even if it did, it is far from obvious that macroprudential forewarnings would produce timely political action.

Fixing the distortion at its source: Restructuring bankers’ liability

What has been received little attention so far is the structuring of liability in banking (Sin 2008). In the early days of fractional reserve banking, bank owners were subject to unlimited liability. In the US, double liability for bank shareholders was common up to the Great Depression. All American investment banks were partnerships into the 1970’s and the last one, Goldman Sachs, only converted itself into a limited liability corporation about a decade ago. Back in the 70s, US financial sector liabilities were less than 20% of GDP. Today the figure is close to 120%. Leverage has increased enormously. Back then, stringent liability rules inhibited risk taking. In the days when such liability provisions applied to banks, “conservative” was the adjective habitually attached to “banker.” It does not fit the high-stakes gambling “quants” of recent years.

If bank stockholders were subject to double liability today, it would hardly be possible to attract enough capital into banking for the needs of a modern economy. Most stockholders have little or no ability to control the risks that bank managements assume. Liability provisions, however, could be applied directly to managers rather than to stockholders so as to change the incentives to assume risk that bank executives face.

Equity for shareholders and equity for employees: The E-shares proposal

To do this, one would first have to create a special type of equity for bank employees. These E-shares would be subject to double liability. If the bank were to fail, the holders would be liable for a sum equal to the value of their shares on the date that they were originally received. E-shares would be non-marketable but exchangeable one-for-one for ordinary shares at market value five years after the owner has left the employ of the financial institution in question.

The second element of such a liability scheme would regulate the extent to which executive compensation would have to be in the form of E-shares. Lower-level employees who are not among the decision makers of the bank should naturally not be part of the program. So, the first $150,000 – or wherever the poverty line is supposed to go in banking nowadays – would be exempt. Compensation would be entirely in ordinary salary. Starting at some such level part of compensation would have to be in E-shares with the proportion rising to, say, 80% at the CEO and CFO level.

Obviously, the specific rules of a liability scheme of this sort can be tweaked in a number of ways. Indeed, only recently Neil Record suggested a similar proposal in the Financial Times (Record 2010).

Benefits of linking decision makers’ liability to their risk taking

Within wide limits such a scheme should have a socially beneficial effect on risk attitudes within the institutions affected – it might even bring back the conservative banker species from the brink of extinction.

The policy can be expected to have some interesting subsidiary effects as well. In the internal politics of banks, double liability should strengthen the influence of risk managers and make it less countercyclical than it has been in recent years. Some of the admirable capacity for innovation that the financial sector has exhibited in recent years would be diverted to risk management (although no doubt much of it would go towards circumventing the accountability that the double liability system would impose).

Within the financial institutions, double liability would cause executives to take a lively interest in the activities of divisions for which they have no direct responsibility. On occasion, one must imagine, dissension might severely strain internal relations. For the individual bank, all this would raise the cost of operations, but from the standpoint of systemic risk the effects would be desirable. If all the insurance executives of AIG had had the personal incentive to learn what was going on in that London unit, internal opposition to its operations might well have brought it under control.

These predictable consequences of double liability would reduce the perceived advantages of conglomerate banking. It will not solve the too-big-to-fail problem by itself, but it just might induce the big banks to spin-off some lines of business.


  • The banks have found high leverage a simple way to gain rates of return several times higher than the rates of profit earned in the rest of the economy – as long as the going is good.
  • The great escalation of executive compensation in finance derives mainly from this business strategy.
  • Double liability would change the risk-return calculation that executives would apply to the strategy.

We would expect to see lower leverage ratios – and a moderation in executive compensation packages. This way of doing it just might work better than Basel rules of leverage and government imposed limits on compensation.

A variant of the scheme would make E-shares marketable. It would be somewhat more cumbersome to administer. Investors who are not bank employees would have to post collateral adequate to back the double liability and records tracking all transfers of shares would have to be maintained. It might however make the market more efficient. The discount on E-shares relative to ordinary shares would reveal to the market what insider’s think of the bank’s prospects. And when E-shares start to trade at negative prices, it is time for the regulatory agencies to step in and take over the bank.

Double liability – or some variant thereof – makes eminent sense. Too much so, perhaps, to be politically feasible.


Record, Neill (2010), “How to make the bankers share the losses”, Financial Times, 6 January.

Sin, Hans-Werner (2008), “What can be learned from the banking crisis”, VoxEU.org, 17 December.

    Posted by on Saturday, January 23, 2010 at 12:42 PM Permalink  Comments (6)


    Feed You can follow this conversation by subscribing to the comment feed for this post.