The case of the $2.3 billion trading loss at UBS holds many lessons for any company that trades derivatives. Remember, UBS wants to claim its trader was a rogue that victimized the company. There was a time that a bank could shout ‘rogue’ as an effective excuse of senior management. But that time is now long past. There have been plenty of penetrating questions asked about the self-evident shoddiness of UBS’ control systems. Another area that deserves scrutiny is the compensation system. How is a trader’s pay determined?
Pay should be for performance. But what counts as performance?
The metrics for performance on a proprietary trading portfolio should be different from the metrics for performance by a market maker. A market maker ought to be compensated, in part, for how successfully s/he is hedging trades. For a market maker, outsized gains on the unhedged component should not count towards a bonus, whereas for a trader running a proprietary portfolio, they should.
How was performance measured at UBS’ Delta One desk? If the traders were being rewarded based on the total profitability of the desk, then UBS was incentivizing them to speculate, and the Delta One desk shouldn’t be described as customer facing or market making.
This same lesson applies to non-financial companies that hedge through their own trading desk. The metrics for performance on hedging should incentivize minimizing risk. The metrics should measure risk reduction. When the desk reports big profits — after netting out the matched positions — that’s a bad sign, not a good one.
If you pay out bonuses when bets payoff, be prepared to see some bets that lose big, too. It’s not a rogue trader if the risky bets are rewarded by the compensation system.