In its July session, the European Parliament approved some of the strictest rules in the world on the bonuses paid to bankers. The aim is to curb risk-taking by financial institutions.

The new rules require that no more than 30 per cent of bankers’ bonuses be paid in cash, that between 40 per cent and 60 per cent be deferred for at least three years, and that at least 50 per cent be invested in “contingent capital,” a new form of debt that converts to equity when a financial company is in distress. The most innovative aspect of these new rules is that the limits do not apply only to financial institutions’ chief executive officers, but to all the top managers (though the definition of top managers is delegated to national parliaments).

The alleged justification for this major interference in private contracting is the systemic effect that these bonuses can have. High pay in the banking sector, so the argument goes, rewards success but does not penalise failure. Managers can easily move from firm to firm when things go badly, avoiding any punishment. This system rewards managers for taking risks, even when the risk is excessive. This distortion is perceived to be one of the main causes of the 2008 financial crisis.

The problem with this argument is that there is no evidence supporting the first crucial link in its logic. Much research has tried to establish a connection between bankers’ compensation schemes and risk-taking, but has failed to find one. At most, such research has found that more highly paid executives took greater risks, but it is unclear whether this is a cause or an effect. Executives in highly leveraged institutions should be paid more, because they bear more risk.

To be sure, these investigations are limited to the top five executives, for which data is publicly available. Unfortunately, there is no publicly available data to establish a causal relationship between bonuses’ pay-for-performance sensitivity and risk-taking for lower-level managers.

In this respect, the Financial Crisis Inquiry Commission (FCIC), established by the United States government, has a unique opportunity. Thanks to its subpoena powers, the FCIC can collect and analyse such data. It is to be hoped that when its report is published in December, we will be able to answer this question.

If we assume that a causal relationship exists, the European directive seems to be fairly well designed, with one main shortcoming. It is well designed because it interferes not with the level of compensation (as many have demanded), but with the form that this compensation takes. It requires that most of an annual bonus not only be deferred for three years, but also that it be put at risk.

If the company performs poorly in the three years, the manager will lose part or all of his or her accumulated bonus. This reduces incentives to take risk, though it does not eliminate them.

The main shortcoming is that these restrictions can be circumvented easily, since they apply only to bonuses, whereas banks maintain discretion over the mix between salary and bonus. Currently, bank managers receive their bonuses at the beginning of each year, with the level based on their individual performance during the previous year. It would be very easy to transform last year’s bonus, based on last year’s performance, into this year’s salary. The salary, which can be paid entirely in cash, will be renegotiated every year, thereby skirting all the regulatory restrictions. Without direct government intervention, it would be difficult to fix the problem.

In large financial institutions, however, the incentive to gamble at taxpayers’ expense does not apply only to managers; it extends to bondholders, who are de facto protected by the government. Having access to insured credit, banks’ shareholders find it irresistible to borrow excessively. Restricting managers’ incentive pay without changing shareholders’ incentives will only force shareholders to be more actively involved in the company and choose other ways to increase the level of risk-taking.

If the problem is the moral hazard implied by being too big to fail, the solution is not to restrict pay, but to eliminate the hazard by forcing shareholders to issue more equity or lose their stock when banks’ debt starts to become risky. As Oliver Hart and I explained in a recent paper, this can be done easily, with a regulator intervening every time the credit-default swaps on the financial institutions’ debt becomes too high.

If we want to intervene on pay in addition to (not instead of) reforming capital requirements, the most effective way is a variation of the tax imposed by former British Prime Minister Gordon Brown: a special tax on all compensation above a certain threshold that is not paid in stock. This tax would have two positive effects: it would induce banks to recapitalise, thereby reducing their excessive leverage, while forcing managers to have more skin in the game.

If the solution is so simple, why has no elected body implemented it? My fear is that politicians want to be perceived as tough on bankers, but have no interest in really fixing the problem.

© Project Syndicate, 2010, www.project-syndicate.org.

The author is professor of entrepreneurship and finance at University of Chicago Graduate School of Business and co-author, with Raghuram G. Rajan, of Saving Capitalism from the Capitalists.