Asymetric Incentives

Yesterday’s testimony before the Financial Crisis Inquiry Commission gave Sheila Bair another chance to air her opinion that the incentive plans in the banks’ capital markets (institutional trading and sales) divisions are a root cause of the crisis.

Years ago I saw this first-hand with a well-known trader that headed a group I was part of.

To set the stage, there is one day each year that brings out every bit of Wall Street’s neuroses — the day we get our “numbers.”  Some call it Bonus Day.

For people with more than five years in the business, that day is typically three quarters (or more) of the total compensation for the year.  Since working on Wall Street is more or less about the money, that makes Bonus Day three times as important as the other 364 days of the year, combined.

In a strange way, the short talk with your boss on the day you get your Number is also the only time all year when you can be a little introspective.  My chief lieutenant in the analytic team used his 15 minutes to ask our trader boss a theoretical question about business strategy:

If you had a choice between two strategies for the year, both with equal expected return of 10%, which would you take?

A) Lower end of range 8%, upper end of range 12%


B) Lower end of range 0%, upper end of range 20%

Without hesitation, our trader boss chose option B, the choice neither of us analytic types would take.

When asked why, the trader explained that even with a 0% return, the firm would pay something as a bonus to keep him after the bad year, but it would pay him half the winnings in the good year.  Besides, he was confident that he would most likely choose the winning trades in either scenario.

It was a classic case of Heads, I Win, or Tails, the House Loses.

When the industry as a whole sets up its incentives this way, it guarantees that a few of its people will push the risk envelope far enough that some of them will blow up their employers.  That’s how we end up having a hundred year flood every five or ten years.

At the very top of the big capital markets firms, they do not actually want to have any of their employees making bets that could blow up the firm.  The problem is that they can’t prevent it.

The reason you pay a lot to hire really smart people to work 80 hour weeks understanding really complex businesses is so your firm can do that business, hopefully with a sustainable competitive advantage caused by barriers such as lack of expertise or lack of capital.

The other side of the coin defined by really smart, really driven people trying to maximize their paychecks is that no one can understand everything every one of those smart people is doing.

A Government bond trader can discover an anomaly in the Kidder accounting system that makes Treasury Strip trades look profitable when the discounts are really the substitute for normal interest payments.

A CDO banker can take Merrill’s desire to be a big force in structured finance bond underwriting, and use the company balance sheet to support “riskless” investment in super-senior AAA CDO’s, so that the firm has a perpetual motion machine once it buys (First Franklin) a steady source of subprime mortgage raw material to make into bonds.

A successful commodities trader 12 time zones away from the Head Office can use the nearly infinite leverage of the futures market to bet more than the net worth of Barings.

All the bank counterparties to Long Term Capital can see a small piece of the giant “relative value” correlation trade they participate in, and conclude that the geniuses there aren’t taking huge risks.

It comes down to what I call the Controller’s Conundrum.

Every incentive system that is based on profitability rewards its most profitable team members with the highest pay.  That’s a given.  The highest pay will draw the smartest, most intensely driven people.  Another given.

I heard in the testimony by the Gang of Four being interviewed in Tuesday’s hearing that they were going to pay their top risk managers just as much as they pay their top producers.


Here’s why that is impossible:

Top producers get extraordinary pay when they have a great year.  A great year includes outperforming the top producers in similar departments at all the competitors.

Since no management team can know at the beginning of the year which of their departments will be most profitable, and even then whether their producers will be top among the competitors, the pay scale for the risk managers would have to be set as if that kind of success were a sure thing, else the producers would still be paid far more than the controllers.

If you think they could wait until the year is finished to determine risk controllers’ pay like they do with the producers, you will necessarily co-opt the controllers to become enablers for outsized (accounted) profits by the producers.

This is what happened when Wall Street paid its stock analysts based on investment banking fees during the dot-com bubble.

A stock analyst puts out a glowing recommendation on a stock with price targets that draw in buyers.  The company looks at the market and decides to get some of that easy money, and hires the investment bank that they think can sell more of their stock at higher prices than the other investment banks.  The dot-commers come to the big investment bank with the most optimistic view of the value of the stock and pays them 5% to 7% of the capital raised as a fee.  The stock analyst gets paid millions, more or less directly driven by the heights he can rationalize for the stock prices.  Rinse and repeat.

It’s a positive feedback loop if profitability drives the “independent” agents’ pay along with the producers’ pay.

It’s only after a major disaster that the controllers get a raise, and that’s probably after the former controllers are fired.

As a friend pointed out one day, Wall Street is really good at nailing barn doors shut, making sure the horses don’t get back into the barn after they escape.

We won’t even go into the second order Controllers’ Conundrum — who watches the watchers?



2 Responses to Asymetric Incentives

  1. Honkytonker says:

    Maybe the answer is to turn all these firms back into partnerships. If external regulation of performance fueled by OPM without recourse creates paradox, maybe self-regulation will work. BWTFDIK.

    • hhill51 says:

      Unfortunately, the logic of employees making the choice my old boss made is always true, even in partnerships (take the biggest risk you can)…. until the person making that choice is a partner, and even then, once there are more than a few dozen partners, the same game theory logic applies with the one vs. the many…
      I’ve got a secret for you, just don’t tell the public shareholders — the profits are already split among the employees (upwards of 50% in many years at many firms), and with equity bonuses now paid, the public shareholders get diluted on the future earnings even while the compensation pool from current earnings sticks in that 40% to 70% band, year after year,

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: