A nice cup of CoCo December 7, 2010 at 9:08 am

The idea of paying bankers’ bonuses in stock or stock options always struck me as odd. After all, equity is a call on the value of the firm after debt has been paid, and you maximise the value of the call by increasing volatility. Stockholders, then, naturally prefer risky high leverage structures – hardly the behaviour you want in a banker. It would make a lot more sense to force the banker to write the bank a put, so that they are incentivised to prevent disaster.

Now, according to Reuters, Barclays might be about to do just that. Specifically they are apparently looking into paying bonuses partly in contingent convertibles:

These securities are bank bonds that turn into equity when things go awry, for instance when losses mount and the bank’s equity capital ratios fall, thereby boosting capital. So, in essence, adding CoCos to the bonus mix would be a novel way for BarCap to force bankers to contribute some of their loot to their employer’s capital cushion, thereby helping to minimize political opprobrium over pay.

OK, this is not quite as good a paying them in reverse convertibles, but it is a good start. Throw in a five year vesting period, clawback provisions for individual malfeasence, and you have the start of something interesting and worthwhile.

4 Responses to “A nice cup of CoCo”

  1. How about paying bonuses in 30 year unsecured bonds of the employer? This would focus bonus recipients on the long term health and stability of their firm.

  2. But… you’re just magnifying a another bad disincentive that already exists. Namely the encouragement of trading desks to take risks that are correlated with the rest of the firm. A traders bonus is the result of two factors: the total bonus pool and his individual performance. Already you can see that traders that take risks that are negatively correlated with the firms get bigger bonuses than those with positive correlation. In years such a trader does well the bonus pool is smaller (if not basically zero), and in years when the big bonus pool is available he gets small or no bonuses.

    Paying in long term securities tied to the firm’s performance makes this even worse. If you’re paying cash every year you might still take uncorrelated risks, you have no incentive to, but you have little disincentive to. In any given year most likely the bonus pool size and your trading strategy’s performance aren’t too likely to cancel each other out. With 0 correlation half the time, your performance and the bonus pool will be in the same direction.

    However over a 10+ year period you’ll see traders actively seek out correlated trading strategies. This is because you basically only have a single “bonus pool period” rather than a new one every single year. Even if their strategies do well, they will never get paid out in the state of the world where the firm does poorly. Thus they have no reason to care about their profits in those states of the world.

  3. “Already you can see that traders that take risks that are negatively correlated with the firms get bigger bonuses than those with positive correlation.”

    Typo and awkwardly phrased on my part should read:

    “Traders that take risks positively correlated with the firms’ get bigger bonuses than those that take negatively correlated risks.”

  4. Doug

    Thank you. I agree with your thoughts, indeed (a long time ago) I wrote something along somewhat similar lines. One solution is risk adjusted compensation where you look at the marginal contribution to the firm’s risk. This is a different dimension though to the ‘what do you pay bonuses in’ one, and even if you figure out a better way of figuring out how much to pay, what form that pay takes needs to be discussed. It is for the latter question that I like the answer ‘a reverse convert’.