Panel Paper: Independent Professional Bureaucracies and Street-Level Corruption: Evidence from Latin America

Saturday, November 8, 2014 : 8:30 AM
Santo Domingo (Convention Center)

*Names in bold indicate Presenter

Laura Langbein, American University and Pablo Sanabria, University de los Andes
We examine whether making bureaucracies more professional reduces street-level corruption in Latin America. Previous scholarship, based on cross-sectional data from many nations, finds a negative relationship. We examine a difference-in-difference relationship between changes in corruption and changes in the law and practice of civil service quality in countries comparable in level of democracy and economic growth (Argentina, Brazil, Colombia, Ecuador, Guatemala, Mexico, Nicaragua, Peru and Venezuela) during the period 2004-2012. So that changes in the law and practice of bureaucratic independence precede changes in bribe requests, we lag the changes in law and practice relative to changes in bribe requests. We examine bribe requests from police, educators, hospitals, national and local government employees, courts, and government offices that relate to the workplace. Using a panel (i.e, repeated measures) of each bribe-type, we find that changes in laws making bureaus more independent within a country correspond with increased bribe requests in the same country, but that “in practice” improvement of bureau independence has no effect. However, the effect sizes, even when they are statistically significant, are not substantively significant. We find similar results using other estimators in robustness tests. We conclude that regulatory advances in countries with less developed civil services do not necessarily mean that those regulations are effective in reducing street-level corruption.  If street-level corruption is affected by legal changes at the national level, the lag between changes in laws and street-level corruption is likely to be so delayed that any claim of causality is likely to be questionable.  Further, it may well be that the link is just what we found.  Contrary to the expectation that “good” laws have “good” outcomes, many scholars expect that bureau independence will increase opportunities for corrupt exchanges, even though that is contrary to reformers’ intent.  It is also possible that the causal connections between laws and street-level corruption, if there are any, exist at the local rather than the national level.

Full Paper: