One of the critical functions of any sales incentive system is pay for performance. But performance is often hard to measure depending on the industry.

Your sales may be up 50% but are you selling 50% more units or has the cost just inflated 50%? Even if you are paid on margin or profit, is your profit inflated by selling older cheap inventory at prices dictated by higher replacement costs?

Is activity in your market place influenced by monetary stimulation? Has housing inflation help your commissions as a real estate sales person? Has a new tax law stimulated life insurance sales?

Are your commissions as a trader influenced by the new tax stimulus package? Has your success rate been due more to probability than you would care to admit?

While you want to pay for performance, what you really want to do is pay for performance you can control. That is a more challenging proposition.

Likewise we often credit presidents with success and failures without consideration as to what degree they actually controlled the outcome.

Was Clinton’s economic boom due to his policies such as his initial tax increase on income, or was the surplus due to his tax cuts on capital gains and dividends? How was his financial performance influenced by the decline in the Soviet Union and our huge military cut back? How was it influenced by the growth of junk bond financing, the high tech boom, or the resolution of the S&L crisis under George H. Bush?

How has the War in Iraq affected the value of the dollar? How effective were the Bush Tax cuts in the post 911 economic recovery? How much has the price of food and oil been affected by economic policy in China and Russia?

Presidential decisions clearly have a major impact, but it is often challenging to separate the impact from their decisions from the randomness and impacts from less controllable factors.

We often are quick to bestow credit and blame more based on our partisan biases than on any intelligent analysis. We are reluctant to distinguish between the results the president controlled and the consequence of randomness.