Does performance pay increase productivity? Evidence from a medical typing unit

Using unique data from a medical typing unit (2011–2014), this paper empirically explores the influence of performance pay on employee productivity in a risk-less low incentive intensity environment. We analyze a routine task performed by lower-level employees. We analyze a setting with strong sectoral agreements and thus a clear (and comparatively high) basic wage level set by the general collective bargaining agreement. We find that despite the risk-less design and low incentive intensity of the setting, switching employees from fixed wages to partially performance-based compensation (PPBC) significantly increases the number of typed pages by 9.53%. Over a one-year period, this rate of increase translates into approximately one additional month of output at a fraction of the usual cost. We also find that these results do not disappear over time. In contrast, the observed gains split into an initial increase in productivity following the switch to PPBC (+3.25%) and additional gains while an employee remains on PPBC. Therefore, we assume that a type of learning begins once an employee switches to PPBC and that this learning is the most pronounced during the first months following the change in compensation. Furthermore, we find no evidence of potentially misplaced incentives in the selection of tasks by the employees (i.e., cherry picking) and no indications of reduced quality.