EDITOR’S NOTE: Goizueta Newsroom regularly shares news on the school, its faculty, staff, students and alumni. But we also take pride in sharing vital pieces of research from the world-class business school faculty. Look for regular updates on published pieces of research like this one…

Title: In Search of Informed Discretion: An Experimental Investigation of Fairness and Trust Reciprocity

Authors: Kristy L. Towry, Associate Professor of Accounting, Goizueta Business School; Victor S. Maas (Erasmus); and Marcel van Rinsum (Erasmus)

Journal: The Accounting Review (2012)


Kristy Towry, Associate Professor of Accounting, Goizueta Business School

When it comes to incentive compensation, only so much can be governed by the terms of a performance-based contract. Rewarding good performance within a group of workers sometimes requires managerial discretion that incorporates information from a variety of sources such as interviews with team members and examination of work documents. But are managers willing to obtain this costly information? If so, does the degree of willingness vary under certain conditions?

Kristy Towry, Associate Professor of Accounting at Goizueta Business School, explores these and other questions with colleagues from Erasmus University in new research on the use of informed discretion in compensation decisions. Towry uses theory from behavioral economics to predict that managers’ willingness to obtain costly information increases as it becomes more difficult to gauge an individual’s contributions from aggregate performance measures. The researchers also predict that managers’ willingness is higher when the team’s aggregate performance is better. Experimental results back-up the predictions as well as a more fundamental notion that managers don’t simply act according to the classic “Homo Economicus” model, but instead are willing to incur a personal cost to provide fair and trustworthy performance assessments.

To study the issue, the researchers conducted an experiment involving 126 students at a business school in the Netherlands. Participants repeatedly played a game at computer terminals in which they were randomly assigned into triads – one student played the role of manager while two others acted as employees. Monetary amounts in the game were expressed in terms of an experimental currency called a Lira and participants assigned the employee role started the game with an endowment of 10 Lira; those in the manager role were given 15 Lira. Employee and manager effort in the game was expressed in terms of monetary cost. Employees, for example, selected an effort level of 0 to 10, which corresponded to a payment of 0 to 10 Lira.

Once employees selected their effort level, managers were given the task of splitting between the two workers a bonus pool that grew in size as the combined effort of the employees grew. Managers had access to an aggregate performance measure for the collective work of the two employees, but also could spend anywhere from 0 to 5 Lira for the chance at accessing better information about individual worker performance. While managers in the game had the key decision to make in terms of testing the researchers’ hypotheses, employees had to think strategically about how they could amass the most Lira. If one employee in a triad selected an effort level of 10 while the other employee selected an effort level of 0, for example, the aggregate performance of the two workers would have generated a bonus pool of 15 Lira. In such a scenario, the employee who opted to make a big effort would be betting that the manager would elect to buy the information needed to recognize individual performance and award the diligent worker accordingly. The employee who opted to make no effort, meanwhile, would be betting that the manager wouldn’t bother to spend money for the extra information and would simply split the 15 Lira of bonus funds in half – leaving the lazy employee with more total compensation than the diligent worker. Managers weren’t rewarded for making bonus pool decisions that recognized the relative contributions of employees.

Study participants played the game eight times; during one randomly selected match, players actually were paid a half Euro for every Lira they amassed in the game. Results showed that participants in the employee role selected effort levels that were equal to slightly more than half their endowments, signaling at least some degree of trust that managers would try to be fair in handling the bonus pool. Study participants in the manager role typically bid about 2 Lira, suggesting that most managers are willing to obtain additional, costly information. The likelihood of managers bidding for the information increased as aggregate performance increased. But when scores for aggregate performance were either very high or very low, managers were less likely to bid for the information. This result fit with the researchers’ predictions since aggregate performance at the extremes is more informative about the contribution of all employees.

“Managers’ willingness to pay is not solely driven by inequality aversion,” the researchers concluded. “Instead, it seems to originate in an experienced need to enforce a social norm of fair cooperation and to repay trust.”

– Chris Snowbeck