Performance Measures and Intra-Firm Spillovers : Theory and Evidence

We revisit the question of how performance measures are used to evaluate business unit managers in response to intra-firm spillovers because prior studies have documented conflicting empirical evidence. Specifically, we are interested in variation in the relative incentive weightings of aggregated “above-level” measures (e.g., firm-wide net income), “own-level” business unit measures (e.g., business unit profit), and specific “below-level” measures (e.g., R&D expenses) in response to spillover arising from either the focal manager's effect on other business units' performance or the other units' effect on the focal manager's performance. Our theory highlights existence of an interaction between the two directions of spillovers. In our empirical work, we account for the interaction effect. Based on a purpose-developed survey of 122 business unit managers, we report that the incentive weighting on above-level measures increases by approximately 50 percentage points when managers face both types of spillovers (as opposed to one type of spillover)