Peer-reviewed article · 2026
How Street-Level Bureaucrats' Traits Affect Community Leaders' Assessments of the Government
Nico Ravanilla, Dotan Haim, and Matthew Nanes · Journal of Politics 88(1)
In brief
Evidence from community policing in the Philippines on how a police officer's personal traits change how village leaders judge that officer, and whether the effect extends to trust in government as a whole.
Abstract
Research shows that bureaucrats' performance is shaped by their personal traits, such as public service motivation, cognitive ability, grit, and “big five” personality. What impact do these traits have on citizen assessments of service delivery and governance? To test this question, we randomly assigned police officers to villages during a community-oriented policing program in the Philippines. This program involved meetings between police officers and community leaders, citizens who serve as key conduits between the population and the state. We find that community leaders' assessments of individual officers were meaningfully affected by officers' personal traits, but those traits matter less for assessments of broader institutional efficacy. These findings suggest that bureaucrats' traits can improve public service provision by shaping how citizens approach encounters with specific government representatives. However, they place limits on the suggestion that exceptionally positive (or exceptionally negative) bureaucrats can shape attitudes towards state institutions more generally.
Research annotationQuestion, variables, design, findings, and mechanism at a glance▾
- Research question
- Do the personal traits of frontline bureaucrats change how community leaders judge those officials and the state institution they represent?
- Independent variable (X)
- Police officers' personal traits, measured before the intervention using established survey instruments and behavioral games: public service motivation, cognitive ability defined as working memory capacity and attention, grit, and the Big Five personality traits. Traits reach communities through two randomizations, of which officers participate and of which village each officer is assigned to.
- Dependent variable (Y)
- Community leaders' assessments of the lead officer assigned to their barangay across effort, reliability, responsiveness, creativity, trustworthiness, approachability, confidence, empathy, and overall quality, and their assessments of institutional police effectiveness across capacity and intentions, procedural justice, and overall quality.
- Identification strategy
- The analysis uses two randomizations embedded in a community-oriented policing experiment in Sorsogon Province, Philippines: which officers, from 705 eligible, joined the problem-oriented policing intervention, and which of the 99 treated barangays each officer was assigned to. Because village assignment is random, officer traits are orthogonal to pre-existing local views of the police. The estimator is OLS regressing each assessment by a community leader of an officer on z-scored trait indices for public service motivation, cognitive ability, grit, and the Big Five, controlling for village characteristics including population, demographics, education, religion, rebel presence, crime rate, and distance to the police station, with standard errors clustered at the officer level. The sample is 284 community leaders in 95 barangays evaluating 95 senior officers.
- Main findings
- Officer traits jointly predict assessments of the individual officer, with the test of no joint effect rejected at the 5 percent level and shifts on the order of 0.2 to 0.4 standard deviations, but the direction is not uniform. Public service motivation predicts consistently better assessments, with a one standard deviation increase raising the overall quality rating by 0.18 points, and extraversion predicts better assessments. Cognitive ability runs the other way, predicting predominantly more negative assessments, as does agreeableness. Grit predicts higher perceived effort and empathy and nothing else. At the institutional level the traits do essentially nothing: the largest effect on trust in the police consistent with the estimates is 0.07 standard deviations for public service motivation and 0.15 for the Big Five index.
- Mechanism
- None of the measured traits predict attendance at the meetings, the conventional performance measure, so traits work through intangible qualities of the encounter itself rather than through effort supplied. The effect stops at the individual officer because institutional reputations are persistent: a lifetime of prior interactions and observed institutional behavior is not moved by one well-matched officer.
- Why it matters
- Bounds an influential claim in public administration: selecting for better bureaucrats improves specific encounters, but will not by itself rebuild attitudes toward state institutions. It is also, to the authors' knowledge, the first study to randomly assign real-world encounters between bureaucrats and citizens, which isolates trait effects from organizational and environmental factors.
- Speaks to
- street-level bureaucracybureaucrat quality, personality, and public sector performancecitizen-state contact and government legitimacycommunity policing in developing democracies
Cite
@article{ravanilla2026bureaucrats,
title = {How Street-Level Bureaucrats' Traits Affect Community Leaders' Assessments of the Government},
author = {Nico Ravanilla and Dotan Haim and Matthew Nanes},
journal = {Journal of Politics},
year = {2026},
doi = {10.1086/734550},
}