An Economist’s View of Compensation and Honesty
Posted by Alex Drexel on November 13, 2008
I always liked the field of economics because it allows us to model complex situations (given simplifying assumptions) which then give us insight into the relationships between variables and how they might react given some kind of change. The analysis becomes even more intriguing when we try and model human behavior. So let’s take a walk down memory lane for those of you took a microeconomics class in college as we apply some of what we learned to ‘ourselves’ and the failure of a rewards system.
Economists often refer to the concept of utility – it’s basically a measurement of value or happiness someone derives from a transaction. We can apply this concept to the amount of value a worker derives from their employment. After a bit of research into a worker’s preferences, we can start to build a utility function where overall value of employment is dependent on a number of factors; e.g. U = (1/2 Compensation + 1/2 Flexible hours). As you can see, the function expresses how utility is affected as each factor is adjusted, and it also shows the tradeoffs someone must make if they want to keep a constant level of utility when factors are adjusted up or down. The tradeoffs a person is willing to make at a constant level of utility is expressed in graphical form as an Indifference Curve; a person is willing to be on any point of the indifference curve since the amount of utility they derive will be the same no matter where they are on the curve.
Now that we have a function that captures preferences, we also need one that captures the constraints on those preferences – we all would like as much compensation and flexible hours as possible, but there are practical limits to these amounts and they must be recognized in tandem with individual preferences as we try and use this model to predict behavior. In order to maximize utility, an individual will make a choice where their indifference curve is tangent to the constraints they face (see graph below).
The graph explains a situation where an employee’s utility is a function of money and honesty. Based on the slope of their indifference curve at various levels, they are willing to compromise honesty for more money.
An example of this situation could be where Jim, who works for a bond rating agency, is paid an annual bonus based on the revenue generated by his company. If Jim is honest and gives a poor rating for a mortgage backed security, the investment bank who paid the rating agency to rate the security may take their business somewhere else and the resulting revenue loss for the agency would impact the Jim’s bonus. This constraint placed on Jim by the incentive plan is represented by the line titled “Incentive Plan 1”. If we match this constraint with Jim’s preferences for money and honesty, Jim would choose HA amount of honesty in the report he produces and $A amount of bonus. If Jim isn’t too worried about his personal reputation (and more about his income), such an incentive plan would consistently produce inaccurate ratings and could damage the long term revenues for the firm. A company can adjust the constraints they place on employees by adjusting their incentive systems – this will impact the choices people make. If the agency changes its bonus formula for Jim so that less of it is dependent on short term revenue, it will cause the constraint curve to shift downward – this yields a higher level of honesty for Jim (HA->HB) and greater long term profits for the rating agency via stronger brand/reputation for the analysis they produce.
It’s hard to measure the preferences of individuals. We will all have different indifference curves; Al Capone’s indifference curve in the model above will look quite different to Mother Teresa’s. But measuring preferences at an aggregate level is a more realistic exercise. As compensation professionals design incentive plans, it’s important for them to recognize that, on average, employees will reach a point where they will be willing to ‘game the system’, manipulate performance measures or make some other compromise in their interests of maximizing utility. Therefore, it’s not enough to hire ‘honest people’ – you need to look at your rewards structure.
– Alex Capone