If You’re So Smart …

16 June 2009 at 2:34 pm Nicolai Foss 1 comment

| Nicolai Foss |

… that it makes sense to delegate a lot of decision-making authority to you when you perform as an agent for a principal, you may also be so smart that you can game the incentive plan. In “Ability and Agency Costs: Evidence From Polish Banking,” Douglas Frank and Tomasz Obloj, both INSEAD, argue (rightly, IMO) that the link between cognitive ability and agency costs has not yet been studied in agency theory.

It is not a priori obvious which effect will dominate, the benefits from delegating decision-making authority to smart agents or the additional agency costs that smart agents may give rise to. Empirical work is called for. Based on data from a large Polish retail bank, where managers are placed on a complicated incentive plan, Frank and Obloj find evidence that smarter bank managers (those with superior insight in the dynamic aspects of their own — complicated — optimization problem) use their decision-making autonomy to pursue private benefits. A very interesting read and highly recommended! Here is the abstract:

Theory and evidence suggest that performance-based pay can increase productive effort but may also give rise to distorted effort choices (multitasking or “gaming”). Performance-based pay is often coupled with delegation of broad decision-making authority. Much existing work assumes that the gains to delegation are increasing in the agent’s (cognitive) ability. However, this work does not address the possibility that such ability may be correlated with agents’ ability to game their incentive plans. In certain settings, agency losses from ability could outweigh the productivity gains. We investigate this possibility using data from a large Polish retail bank. We find that branch managers select the interest rate and size of consumer loans in a manner consistent with gaming of their incentive plan. Furthermore, evidence of this gaming behavior is stronger for managers with a better understanding of their incentive plan. Finally, we estimate that the bank loses between three and twelve percent of its profits from managers’ pricing decisions and that the losses are greater for the more knowledgeable managers. We estimate these agency costs through a novel empirical strategy, using managers’ position in their incentive plan as a supply shifter to identify the bank’s demand for loans.