Non-Governmental Monitoring of Local Governments Increases Compliance with Central Mandates: A National-Scale Field Experiment in China (With Sarah Anderson, Mengdi Liu, and Bing Zhang) [request] Revise and resubmit, AJPS.
Abstract — Central governments face compliance problems when they rely on local governments to implement policy. In authoritarian political systems, these challenges are pronounced because local governments do not face citizens at the polls. We designed and implemented a national-scale, randomized field experiment in China to test whether the public dissemination of performance ratings of municipal governments by a non-state actor caused municipal governments to release more information about the management of pollution to the public as mandated. We find significant and positive treatment effects on compliance with the release of information to the public after only one year that persists with reinforcement into a second post-treatment year. These results reveal important roles that non-state actors can play in enhancing accountability of local governments in authoritarian political systems.
Escaping the Valley of Disengagement: Two Field Experiments on Motivating Citizens to Monitor Public Goods (With Jacob Skaggs and Daniel Nielson). [PDF] Revise and resubmit, BJPS.
Abstract — Beyond signaling their preferences through voting, protest, and lobbying, citizens can inform governments directly and routinely about the implementation of policies and the delivery of public services. Yet citizens lack incentives to provide information when they do not expect governments to be responsive or when they do not value public services, which often prevents governments from providing public goods effectively. In two field experiments, we studied potential remedies to this dilemma related to solid waste services in Uganda. We randomly assigned reporters to be recruited by community nomination and to be recognized by community leaders in an attempt to select for and motivate information sharing. We also randomly assigned reporters to hear from government about how their reports are used to make real improvements to waste services. Community nominations and public announcements did not increase reporting. However, responsiveness boosted participation over several months and the effect grows larger over time, highlighting the critical role of timely responsiveness by governments for initiating and sustaining information flows from citizens.
How Exposure to Election Violence and Fraud Neutralizes the Effects of Corruption on Turnout: A Field Experiment in Uganda (With Ryan Jablonski, Paula Pickering, and Daniel Nielson) [PDF]
Abstract — Information about corruption ought to motivate voters to show up at the polls to “throw the bums out.” However, in elections characterized by fraud and violence, voters may not expect their vote to matter, and may have to worry about repercussions for political participation. We theorize that fraud and violence should condition the effect of corruption information on turnout. We evaluate this theory using a large randomized-controlled trial conducted during Uganda’s 2016 district elections. We treated eligible voters with factual information about irregularities in the management of local government budgets using SMS messaging. We find that this information had hypothesized turnout effects, but only for respondents with less exposure to reported election violence and for those who had high expectations of fair voting. These results imply that electoral violence and fraud impede citizens from seeking accountability from their elected officials at the polls and may help indemnify public corruption.
Massive Citizen Reporting is Too Inconsistent and Costly to Improve Public Services: A Framework and Field Experiment (With Patrick Hunnicutt and Polycarp Komakech) [PDF]
Abstract — Governments around the world are investing in technologies that allow massive, frequent, and localized contact with citizens, though there is little evidence about the impacts of the streams of data these technologies create on the delivery of public services. We report a large-scale randomized controlled trial that involved recruiting 50 citizens in each of 100 neighborhoods across Kampala, Uganda to provide weekly reports on the delivery of solid waste service via an SMS-messaging platform to a municipal government, resulting in 17,520 verified and usable reports during the study period. Citizen reporting did not reduce waste accumulation. More positively, reporting reduced the amount of burning and unmanaged waste piles for a time, but this positive result did not persist after an unexpected staff restructuring in the unit responsible for waste management. Waste collection did not improve in zones with more reports or more dissatisfied reporters. Using our observations as participants in development and deployment of the platform and interviews with key staff at the government agency receiving citizen reports, we show how the adoption of new technologies to collect data from citizens requires both new capacities and data consistent enough to reduce uncertainty about the allocation of effort. We provide a formal framework for analyzing the challenge of utilizing citizen-sourced data for the management of public services. Citizen-sourced data must be both low-cost relative to alternatives and consistent enough to reduce uncertainty about decisions related to public effort.
Community Monitoring Does Not Activate Oversight of Revenue Sharing at Bwindi National Park, Uganda (With Brigham Daniels) [PDF]
Abstract — Citizen monitoring of the performance of government officials often fails because the information provided by citizens does not reach authorities who have the ability and responsibility to oversee and sanction wayward officials. Working with the Uganda Wildlife Authority, we attempted to overcome this common problem for a national park revenue-sharing program that is plagued by corruption and the misdirection of funds. In a field experiment, citizen monitoring was channeled to nationally-appointed officials who have specific responsibilities for oversight. In field audits, we find no evidence that monitoring improved the delivery of revenue-sharing projects or increased satisfaction with revenue sharing among residents. Follow-up interviews with officials indicate that the information from citizen monitoring was not used for oversight. It is not possible to use bottom-up monitoring to solve governance problems where institutions are well-suited for obfuscation, corruption, and blame shifting.