Motivation crowding and the federal civil servant: Evidence from the U.S. internal revenue service

Anthony M. Bertelli

Research output: Contribution to journalReview article

Abstract

Pay-for-performance reforms create "high-powered" incentives for civil servants to meet or exceed specified performance objectives as measured by such things as customer satisfaction. Economists and social psychologists have advanced the claim that high-powered incentives for performance may empirically lessen the effect of civil servants' intrinsic motivation toward achieving agency goals (motivation can be "crowded out"). Nonetheless, well-designed pay-for-performance incentives may "crowd in" intrinsic motivation. A number of federal agencies and subagencies have undergone personnel management reforms that raise the specter of this pattern of "motivation crowding." Does it happen? Is intrinsic motivation crowded in or crowded out? This paper employs item response theory to create measurement models for the estimation a latent trait of intrinsic motivation for employees of the Internal Revenue Service (IRS) and Office of the Comptroller of the Currency (OCC) using data from the 2002 Federal Human Capital Survey. The IRS, but not the OCC, implemented a paybanding system that imposed high-powered performance incentives on supervisors, but not on non-supervisory personnel. Results suggest that the IRS reward structure crowded in intrinsic motivation at the lowest levels, but that at the highest levels of motivation intrinsic motivation is crowded out, a pattern not seen in the OCC data.

Original languageEnglish (US)
Pages (from-to)3-23
Number of pages21
JournalInternational Public Management Journal
Volume9
Issue number1
DOIs
StatePublished - 2006

ASJC Scopus subject areas

  • Business and International Management
  • Public Administration

Fingerprint Dive into the research topics of 'Motivation crowding and the federal civil servant: Evidence from the U.S. internal revenue service'. Together they form a unique fingerprint.

  • Cite this