Merit, Tenure, and Bureaucratic Behavior: Evidence From a Conjoint Experiment in the Dominican Republic

Abstract

Bureaucratic behavior in developing countries remains poorly understood. Why do some public servants -yet not others -work hard to deliver public services, misuse state resources, and/or participate in electoral mobilization? A classic answer comes from Weber: bureaucratic structures shift behavior towards integrity, neutrality, and commitment to public service. Our paper conducts the first survey experimental test of the effects of bureaucratic structures. It does so through a conjoint experiment with public servants in the Dominican Republic. Looking at merit examinations and job stability, we find that Weber was right -but only partially. Recruitment by examination curbs corruption and political services by bureaucrats, while enhancing work motivation. Job stability, by contrast, only decreases political services: tenured bureaucrats are less likely to participate in electoral mobilization. Examinations thus enhance the quality of bureaucracy (motivation and lower corruption) and democracy (electoral competition); job stability only enhances the quality of democracy. Forthcoming in Comparative Political Studies

Civil service management and corruption: What we know and what we don't

“…Rauch and Evans (), Dahlström et al (), and Rubin and Whitford () find no significant associations between job stability and corruption in cross‐sectional country‐level data. Similarly, in a conjoint experiment in the Dominican Republic, Oliveros and Schuster () do not identify a significant effect.…”

“…Recanatini et al () provide additional evidence from African and Latin American agencies, although their measurement strategy renders distinction between recruitment and promotion impossible. More recently, Meyer‐Sahling et al () provide evidence based on individual‐level data from a survey of public servants in five Eastern European countries; and Oliveros and Schuster () conduct a conjoint experiment in the Dominican Republic, finding the, to our knowledge, first (survey) experimental support for the argument that merit recruitment limits corruption. Finally, Bersch et al () use administrative data from Brazilian agencies to show a positive correlation between agency politicization and corruption.…”

“…Since corruption is a sensitive subject, this approach, of course, has its own pitfall: survey responses to corruption questions may suffer from bias. Recent studies point to two solutions: survey experimental methods, such as list experiments, randomized response techniques and conjoint experiments (e.g., Gingerich ; Oliveros and Schuster ); and the use of ‘big data’ proxy measures for corruption, such as measures of competition in public procurement (Charron et al ).…”

“…Scholars seeking statistical causal identification will, of course, wish to go beyond associations towards field experiments, regression discontinuity designs (RDDs), or (other) instrumental variables. Strong causal identification has been a blind spot in the literature to date (with few exceptions, e.g., Barr et al ; Armantier and Boly ; Oliveros and Schuster ). As Table shows, experiments are few and far between and are mostly concentrated in economics journals.…”

Legal reform need not come first: Merit‐based civil service management in law and practice

“…That examinations in the DR remained marginal in number and were subject to manipulation is not to say that they had no effect when they took place. In fact, Oliveros and Schuster () show that merit recruits are more motivated to work hard, are less corrupt and are less likely to engage in clientelism in the DR.…”