The Search for Symmetry: How bankers' bonuses can set-up the wrong incentives

In my last post I described how the Scalability of transactions allows banks to pay large sums to attract the best people. But there is another feature of the bonus system that is equally fascinating, and goes some way to explain why banks suffered such massive losses in the 2008-09 financial crisis.

Bonuses are, by their nature, asymmetric. Employees get paid a base salary. Then those who perform particularly well may get a bonus on top. The ones who perform particularly poorly do not, as a rule, get a fine, or an ‘anti-bonus.’ The worst that can happen to them is they lose their jobs.

In non-financial jobs this asymmetry is not such a problem, as the contribution the employee makes to their company is also asymmetric. A great employee can create a lot of value for their firm, but a bad employee doesn’t destroy equivalent amounts of value. Take a law firm for example: highly productive associates may work lots of billable hours doing quality work. Their firm benefits from the work and pays them a large bonus to try and retain them. A poor associate, who works few billable hours and does poor quality work, will get no bonus, and may get the boot. But the poor performer is unlikely to lose the law firm a lot of money, as the work is quality-checked before it goes to a client – the deficiencies are noticed and rectified by the good performers.

However the financial sector, or more particularly the trading sector, doesn’t work this way. On investment banks’ trading desks, the banks’ own funds are put at risk by traders taking positions on the financial markets. If the desk makes a profit, the traders get a cut, which can lead to multi-million pound bonuses for ‘top performers’. However, if the desk makes a loss, the traders are not exposed to the downside. They may lose their jobs, but it is the bank that has to foot the millions or billions of pounds of losses. This asymmetry gives traders an incentive to make large and risky bets, whether or not they are the best thing for the investment bank’s bottom line.

This kind of behaviour is a key cause of trading scandals such as Jerome Kerviel’s €4.9bn loss at Société Générale. But moreover, it led to an accumulation of risky assets (such as securitised loans) on banks’ balance sheets, and helped sustain the over-valuation of these assets in the run up to the recent financial crash.

In the wake of the crash, banks and regulators are looking hard at how performance ought to be rewarded. Multi-year guaranteed signing-on bonuses for star performers are being phased out. More bonuses are being issued as restricted stock, which becomes worthless if a trader leaves the firm; this gives them a financial penalty if they lose their job, a kind of ‘anti-bonus’ if they perform particularly badly. Likewise with ‘clawback clauses’ that can retrospectively reduce a trader’s bonus if the positions they take turn out to make a loss in the long-term. These kinds of a reforms are a promising step in avoiding a repeat of the crisis; though they also add an element of the unknown and untried: I would not be surprised if the new set of incentives for bankers leads to some unintended consequences further down the line.