"What gets measured gets managed", according to the common business management slogan. If governments are to manage for results, then clearly it is critical to measure the performance of their public employees. But what, specifically, should be measured? How should those indicators be measured? And what should the consequences be for good (or bad) performance? Unfortunately, the answers to these questions are not at all straightforward.This paper examines some of the key troubles that plague personnel evaluation systems when the when the rhetoric of performance meets the hard realities of implementation. Selected insights for performance evaluation are presented from the field of social psychology, as well as informative lessons from the experience of OECD countries. Then, preliminary data are presented from an examination of the Colombian system for evaluating judges and their staff, drawing tentative conclusions for how that system either evades or falls victim to the troubles that are commonplace in personnel evaluation systems.The broad lessons of this paper for designing personnel evaluation systems are: i) keep the rewards (and punishments) modest for all but the most extreme cases, ii) introduce the evaluation system cautiously, as unintended consequences can create as many performance problems as the system was meant to solve, and iii) the personnel evaluation system itself must be subject to rigorous, periodic evaluation.

Inglés