Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
In this laboratory experiment we study the use of strategic ignorance to delegate real authority within a firm. A worker can gather information on investment projects, while a manager makes the implementation decision. The manager can monitor the worker. This allows her to exploit any information gathered by the worker, but also reduces the worker's incentives to gather information. Both effects are influenced by the interest alignment between manager and worker. Our data confirm the prediction that optimal monitoring depends non-monotonically on the interest alignment between managers and workers. Managers also show some preferences for control that seem to be driven by loss aversion. We also find mild evidence for hidden benefits and costs of control. However, behavioral biases have only limited effects on organizational outcomes.