Community monitoring interventions to curb corruption and increase access and quality in service delivery: a systematic review

ABSTRACT

There is a belief that allowing communities monitoring power over providers could be beneficial for improving service delivery and reducing corruption in service delivery. In community monitoring interventions (CMIs), the community is given the opportunity to observe and assess providers’ performance and provide feedback to providers and politicians. This systematic review and meta-analysis appraises and synthesises evidence on the effects of CMIs on access and quality of service delivery and corruption outcomes in low and middle-income countries. The results indicate evidence of beneficial effects of CMIs on service delivery quality and on helping to curb corruption. The potential benefits of CMIs on access to and quality of services are likely to be higher when interventions are designed so that contact between both actors are promoted, and tools for citizens to monitor agents’ performance are provided. However, more rigorous research is needed to address this hypothesis.

KEYWORDS:

• Community monitoring
• community-based
• corruption
• service delivery
• systematic review
• education

Acknowledgements

Many academics and practitioners read the review protocol and gave us relevant feedback. In particular, we want to thank Bénédicte de la Brière, Seollee Park and Vincenzo Di Maro for thoughtful comments. Additionally, we would like to thank reviewers from 3ie and the International Development Coordinating Group (IDCG) of the Campbell Collaboration. The usual disclaimer applies.

Disclosure statement

There are no known conflicts of interest that the team is currently aware of. The team has not been part of any organisation that has implemented projects in this area nor has any interests in promoting particular findings due to personal relationships with individuals or organisations who will benefit from these. Ezequiel Molina has conducted research on community monitoring in Colombia (Community Visible Audits) as part of his dissertation work. He studied the effects of the programme on corruption as well as political influence of the community over policy making.

Notes

1. There is also a new Global Partnership for Social Accountability (GPSA), which is a coalition of donors, governments and civil society organisations aiming to improve development results by supporting capacity building for enhanced citizen feedback and participation to monitor service delivery. GPSA aims to reach overall funding of 75–125 million USD over the next 7 years. To date, 15 countries have joined the GPSA: Bangladesh, Belarus, Colombia, Dominican Republic, Honduras, Indonesia, Kyrgyzstan, Malawi, Moldova, Mongolia, Mozambique, Philippines, Senegal, Tajikistan and Tunisia.

2. For the purpose of this review, service delivery involves not only services but also construction of necessary infrastructure to carry out those services. As a result, we will talk indistinctly between service delivery and project performance.

3. The term bottlenecks has been used in the literature (Lieberman, Posner, and Tsai 2013) to refer to constraints that limit the effectiveness of community monitoring programmes.

4. In order to give salience to information practitioners use an array of instruments to attract the citizens’ attention. We are not aware of any CMIs where these incentives were embedded in the theory of change and properly assessed. This appears to be a knowledge gap for CMIs.

5. In some parts of the world, the state fails completely to provide services and to monitor illegal private service provision. Even under these environments, when citizens can choose providers overall providers’ performance may increase.

6. We also aimed to answer a third review question – ‘to what extent do geographic region, income level or length of exposure to interventions explains the variation in findings?’ – using moderator analysis. Due to the limited number of studies, we were unable to perform planned moderator analyses by study design and length of exposure to intervention. Moderator analysis by region did not reveal big differences.

7. We critically appraised the consistency among ratings by our coders through interrater assessment (McGraw and Wong 1996). The absolute agreement intraclass correlation is 0.70 with a 95 per cent CI [0.21, 0.95]. In case of disagreement, a third researcher determined the rating.

8. ASE (Kling et al. 2004) is interpreted in the same way as SMD. It is calculated by combining several measures for the same outcome into a unique average standardised treatment effect (ASE), by estimating a seemingly unrelated regression system for K-related outcomes:

where is a K by K-identity matrix. The average standardised treatment effect is estimated as

where is the point estimate on the treatment indicator in the kth outcome regression and is the standard deviation of the control group for outcome k (Björkman, de Walque, and Svensson 2013).

9. In some cases, these correlations were available in the studies’ databases, or where easily obtainable from tables reported in the papers. When not available, we assumed a value of 0.5 and checked whether the results changed substantially for extreme correlation values.

10. All randomised field experiments report no statistical difference between treatment and control groups.

11. In both cases, we changed the sign of the effect size so it can be interpreted properly (that is a positive effect size means that corruption has been reduced).

12. The study also evaluates the effect of external audits, which did reduced corruption, but we did not include it in the meta-analysis because it does not fall into any of our four intervention categories.

13. This finding is consistent with those reported by the author, who argues that ‘increasing grassroots participation in monitoring had little average impact’ (Olken 2007).

14. Finally, we did not include Banerjee et al. (2010) as it is not a measure of corruption they use, but rather they look at whether the treatments to increase community monitoring generated additional nonteaching resources for the schools. They found that none of the interventions have any effect.

15. Actually, the medium-term impact of the first intervention is assed in Björkman, de Walque and Svensson (2013). However, to avoid confusion, we designate the latter as the main reference for the second intervention and Björkman and Svensson (2009) for the first intervention.

16. We were not able to compute neither SMD nor RR for these outcomes due to lack of information.

17. The short-term impact of this intervention is also not statistically significant, but it is not reported in the table since we were not able to compute RR.

18. It is important to note why we think this is a quality measure and not an access measure. Access is related to getting the service. However, you can get the service and the fact that you had to wait makes it less valuable and of lesser quality.

19. When different test scores where reported (for example language and math test scores), we previously pooled them following the procedure explained before.

20. It should be noted that we are excluding two studies for which we were not able to compute standardised effects (Table 12) but which found significant effects of CMIs on our outcomes of interest. Keefer and Khemani (2011) examine the proportion of children tested in the village public school who could read sentence and paragraphs (ASER literacy test), finding that the information campaign resulting from communities’ access to radios enhanced literacy tests. Pandey, Goyal and Sundararaman (2009) examined the percentage of children who could pass different learning tests, including reading and writing competences and mathematics abilities.

21. Additional qualitative and quantitative evidence support this. For example, Gaventa and Barrett (2012) perform a metacase study of 100 interventions aim to increase citizen engagement in service delivery. For the 828 outcomes from the 100 reviewed case studies, only 153 came from interventions where the final goal was to strengthen the responsiveness and accountability of the state to provide services. Results indicate that 55 per cent of those 153 outcomes were positive and 45 per cent were negative. Negative results were associated with failure of citizens to participate, due in part to fear of backlash against those who speak out and a sense of tokenism in the participation mechanism.

Additional information

Funding

3ie Systematic Review Grant provided the funding for this systematic review.

Notes on contributors

Ezequiel Molina

Ezequiel Molina is an economist in the Education Global Practice of the World Bank. He works on issues of institutional reform and service delivery with research spanning the areas of governance, education, gender and poverty and inequality. Recent publications include a systematic review on the effectiveness of social accountability interventions and a paper assessing the quality of service delivery in Sub-Saharan Africa. He also has co-authored the book Economic Polarization, Institutions and Conflict in Latin America. Ezequiel holds a PhD in Political Economy from Princeton University, and a B.A. and M.A. in Economics from La Plata National University (Argentina).

Laura Carella

Laura Carella is Junior researcher of the Center for Distributive, Labor and Social Studies (CEDLAS) at Universidad Nacional de La Plata (UNLP) in Argentina. She holds a Master in Economics from the UNLP. She is a Proffesor of the Economics Department at UNLP and teaches undergraduate courses on Econometrics and Microeconomics. She has also taught graduate courses on Microeconomics. Hermajor fields of research are Economics of Education, Health Economics and Social Policy.

Ana Pacheco

Ana Pacheco is a Senior Researcher and Executive Coordinator for the Center for Distributive, Labor and Social Studies (CEDLAS) at Universidad Nacional de La Plata (UNLP) in Argentina. Pacheco has a Ph.D in Economics from Universidad Nacional de La Plata. She teaches undergraduate Macroeconomics, and her research is focused on development economics, with a special emphasis on applied econometrics and computed general equilibrium models.

Guillermo Cruces

Guillermo Cruces (PhD in Economics, LSE) is Under-Secretary of Development at the Treasury Ministry in Argentina. He is on leave as the deputy director of the Center for Distributive, Labor and Social Studies (CEDLAS) at the Universidad Nacional de La Plata, Argentina (UNLP). He is also a researcher at Argentina’s National Scientific and Technical Research Council (CONICET), and a research fellow at IZA. His research is focused on labor economics and distributional analysis in Latin America and the Caribbean, and on the economics of perceptions and reference groups in general. He teaches at the graduate and undergraduate level at the Economics Department of the UNLP, and he is invited professor of labor economics at the Universidad de San Andrés (UdeSA), Argentina. He has published in journals such as the Journal of Political Economy, Journal of Public Economics, American Economic Journal – Macroeconomics, Labour Economics, Journal of Population Economics, Brookings Papers on Economic Activity, Journal of Development Studies and Economia. He has edited books and contributed to collective volumes and reports, and recently published the book Growth, Employment and Poverty in Latin America(Oxford University Press, 2017, with G. Fields, D. Jaume and M. Viollaz).He has worked previously for the UK’s Department for Work and Pensions and for the Development Studies Division of the UN’s Economic Commission for Latin America and the Caribbean. He has also been a researcher at STICERD, London School of Economics and Political Science, where he obtained an MSc and a PhD in Economics, and a visiting scholar at Harvard’s DRCLAS and at University of California at Berkeley.

Leonardo Gasparini

Leonardo Gasparini is the director of the Center for Distributive, Labor and Social Studies (CEDLAS) at Universidad Nacional de La Plata (UNLP) in Argentina. Gasparini has a Ph.D in Economics from Princeton University. He teaches undergraduate and graduate courses on Income Distribution and Labor Economics at UNLP. He has published articles on income distribution, social policy and labor issues in several international journals such as the Journal of Public Economics, Review of Income and Wealth, Social Choice and Welfare, Economic Development and Cultural Change, Journal of Economic Inequality, Journal of Income Distribution, Journal of International Development, and others. He has written several books on development issues. Gasparini has taught courses and given seminars on distributive issues in almost all Latin American countries. Gasparini has been awarded with the Guggenheim fellowship 2009.