Monitoring Corruption: Evidence from a Field Experiment in Indonesia

Abstract

This paper uses a randomized field experiment to examine several approaches to reducing corruption. I measure missing expenditures in over 600 village road projects in Indonesia by having engineers independently estimate the prices and quantities of all inputs used in each road, and then comparing these estimates to villages' official expenditure reports. I find that announcing an increased probability of a government audit, from a baseline of 4 percent to 100 percent, reduced missing expenditures by about 8 percentage points, more than enough to make these audits costeffective. By contrast, I find that increasing grass-roots participation in the monitoring process only reduced missing wages, with no effect on missing materials expenditures. Since materials account for three-quarters of total expenditures, increasing grass-roots participation had little impact overall.The findings suggest that grass-roots monitoring may be subject to free-rider problems. Overall, the results suggest that traditional top-down monitoring can play an important role in reducing corruption, even in a highly corrupt environment.

Corruption in Procurement and the Political Cycle in Tunneling: Evidence from Financial Transactions Data

“…Due to the secretive nature of corruption, most empirical work had been based on subjective perception measures, which is problematic for assessing welfare effects of corruption (Reinikka and Svensson 2006). Recognizing the problems with subjective evidence, researchers have recently turned to studying corruption using policy experiments (e.g., Svensson 2004, Olken 2006), natural experiments (e.g., Caselli and Michaels 2009), and field experiments (e.g., Bertrand et al 2007a;Olken 2007;Ferraz and Finan 2008). However, experiments that allow evaluation of the effects of corruption are rare and often cover a specific area of corrupt economic activities.…”

Can Bottom-Up Institutional Reform Improve Service Delivery?

“…First, monitoring public projects is a public good, so there may be a free-rider problem (Olson, 1971). Second, community monitoring may be prone to capture by local elites (Bardhan, 2002;Bardhan and Mookherjee, 2006;Olken, 2007). Finally, citizens' beliefs regarding the probability of success of the community monitoring exercise may affect their participation rate, which in turn affects the visibility of services.…”

“…The only article assessing the impact of a social audit, although not a traditional one, is Olken (2007), which studies the effect of grassroots participation in Indonesia. This seminal contribution analyzed the effect of additional participation in the community monitoring process on unaccounted expenditure from infrastructure projects.…”

“…First, it provides new evidence of the impact of community monitoring interventions using a unique dataset from the Citizen Visible Audit (CVA) program in Colombia. In contrast with Olken (2007), where the community had the ability to manage the funds to build a given project, in the present study I assess the impact of a program aimed at empowering citizens through increasing their opportunity to make providers and local governments accountable but not as a group of peers. This means that citizens do not have decision power, but only the opportunity to monitor infrastructure projects.…”

What Role for Qualitative Methods in Randomized Experiments?

“…For example, in their study on the impact of reserving village leader roles for women on the provision of public goods in West Bengal and Rajasthan, Duflo and Chattopadhyay (2004) utilized participatory resource mapping (with ten to twenty villagers) and semi-structured interviews to ascertain village-level infrastructure investments and repairs. More recently, Karlan (2009) has rightly stated that "the decision about what to measure and how to measure it, i.e., through qualitative or participatory methods versus quantitative survey or administrative data methods, is independent of the decision about whether to conduct a randomized trial", and outlines further studies that utilize non quantitative methods (including Olken, 2007, andKarlan andZinman, 2009). However, whilst the acknowledgment that qualitative methods can be utilised within a randomized experiment is to be welcomed (see Prowse, 2007, for an early discussion of this issue), Karlan (2009) (2008) and Leeuw and Vaessen (2009) have paid little attention to the possibility of randomized experiments that allow a primary role to qualitative methods (even though this is common within the related field of social policy, reviewed in Molloy et al 2002, see also Gibson and Duncan, 2000;London et al, 2005;Lewis, 2007).…”