May 13, 2010
The early 20th century psychologist B.F. Skinner invented a crude, and occasionally cruel, box for measuring the behavior of rats. He discovered and statistically proved that a rat’s behavior was effected by what FOLLOWED a positive or negative reinforcement. Not what preceded the reinforcement. In other words, even rats think about the consequences of their actions.
So why aren’t we applying this fundamental knowledge to compensation practices in America?
Before Skinner, Russian psychologist, Ivan Pavlov discovered the phenomenon of conditional responses by making a dogs salivate when presented with the mere IDEA of a tasty snack. Ergo, dog’s actions can be manipulated by dangling the right bacon-flavored carrot in front of their hyper-sensitive noses.
Again, why can’t managers across America understand that HOW they incentivize (i.e., compensate) their employees will lead to HOW they act?
I believe the recent, and many of the past, troubles that have caused major disruptions in our financial markets are a result of poorly designed compensation packages. Alter the way our decision makers are compensated and you’ll alter the way they act. It’s a practice I not only preach, but practice as my clients know.
Take the most recent Great Recession. Sure the conditions that led to a collapse in the U.S. housing market are complicated, but I postulate that it was the manner many investment bankers and mortgage brokers are compensated that led to the disaster.
First the investment bankers. The “carrot” of huge cash bonuses at year end based on a 12 month period of performance incentivizes decision makers to enhance that 12 month’s performance as much as possible. Anyone who can earn enough in one year to skate through life comfortably isn’t going to think about the long-term. Especially if they’re 28 years old and surrounded by similarly-minded bloated egos. The counter argument that if things collapse after 12 months the decision maker could lose his or her job doesn’t fly when you’re talking about 5, 6, and 7 figure bonuses. If you had the opportunity to make $5 million in one year WITHOUT regard for the future, wouldn’t you seriously consider it? Regardless of your ethics, it would be a constant thought in your mind which would undoubtedly effect your behavior. No ifs, ands, or buts about it. It’s human nature.
Next, the mortgage brokers. This rare breed had one goal in mind prior to the crisis. Sell a loan to anyone with a pulse, collect a commission, and move on. This behavior was enabled because a mortgage broker is paid to close a deal. What happens to the client and the loan has no impact whatsoever on their compensation. It’s a transaction. Get it closed, get paid, and move on. This must change if we don’t want to see another disaster in the mortgage market. If the mortgage broker was paid over five years based on the timeliness and completeness of his or her client’s payments, I guarantee we’d never have seen the liar loans, no doc loans, and the invention of negative amortization, interest-only, adjustable rate loans that fueled the housing crisis.
Rating agencies were another major contributor to the recent financial collapse. Rating agencies are paid to rate securities. Fine. But when you’re paid by the party that created the security, conflicts abound. With the three major competitors—Standard & Poors, Moodys, and Fitch—all vying for the lucrative contracts, Wall Street’s securitization machine simply shopped the three for the best rating. If Moody offered a BBB and Fitch was willing to go to AA, guess who got the contract? Once again, it’s simply human nature.
Similar to the ratings agency conflict of interest is the investment banker-security analyst conflict of
interest that resulted in the global research analyst settlement in 2003. After decades of dangling favorable or threatening unfavorable “buy, sell, or hold” ratings on equities, Wall Street’s troubled practices were finally exposed to the public and they were slapped on the wrist with a $1.4 billion fine. Prior to this ruling, equity analysts were paid exorbitant sums to opine on the value of a company and to provide a so-called price target. Their opinion was heavily influenced because their compensation was heavily influenced by the even more exorbitant investment banking fees that go along with raising capital for the very same companies that they were opining about. Call my company and “strong buy” and you’ll get your piece of the 7% commission on our next $300 million secondary. Call my company a “hold” and you’ll be left off the capital raise team. The global settlement tried to segregate to two businesses by making it illegal for the two sides of one business to share information with each other. IMHO, that’s like legislating that your brain can’t tell your lungs how much oxygen your heart needs. What’s the point of having a central decision maker (i.e., a brain) if they can’t cooperate. Good luck running a marathon.
Rather than ramble on about more inequities fueled by poorly designed compensation practices, I’d like to make a simple plea to the well paid directors that sit on corporate boards…
You control the purse strings for the top decision makers in America. Stop using “relative compensation” as an excuse for dolling out preposterous pay packages and design a plan that makes a manager rich if they significantly outperform the company’s business plan, properly paid if they reach those goals, and worried about their job AND livlihood if they underperform. Set reasonable time frames like three to five years and link their long-term compensation to difficult-to-manipulate measurements like economic value added (EVA) rather than easy-to-manipulate measures like earnings per share (EPS). Treat their pay packages like you would your children’s upbringing. Think of it in decades, not quarters or years. Quit incentivizing them to think like a paid employee and more like a fiduciary or founder or parent. Then let human nature work FOR the business and watch the lessons of Pavlov and Skinner reap rewards for years to come.
It’s the compensation stupid.