Our review starts where an earlier exercise ended. In 2015, two of us reviewed empirical studies on the same topic published between 1991 and 2015. The new paper covers the years since then. From an initial 200 publications or papers suggested in response to a survey among 55 experts or that we identified ourselves, we reviewed 32 studies covering effects of participation (14), transparency (16), or both at the same time (two). Our focus was on studies with a research design that could credibly claim some causal impact of government-led interventions.
Overall, we find that recent empirical studies have started not only to focus on a wider range of contexts and topics, but have also become more nuanced, thereby moving towards a more systematic understanding of interactions between policy instruments and contexts.
The most credible causal evidence comes from studies of audits. This literature has not only grown in geographical scope, but also advanced methodologically with new studies exploiting natural experiments that capture real-world effects of audits – something that governments might ultimately be most interested in. Several studies highlight the importance of timely audits to facilitate electoral accountability, while others stress the importance of the independence of auditors. A re-analysis of seminal work in Indonesia focuses on both audits and participation, hinting at a potential trade-off: top-down audits, whilst being effective, might crowd-out community monitoring efforts.
New studies on participation have moved beyond the initial focus on Brazil where such approaches were pioneered. Several investigate how the introduction of participatory budgeting relates to development outcomes and public goods provision, highlighting the crucial role of context, history, and organisational structures. While these studies add nuance and provide insights on mechanisms, they often stop short of proving causality.
Other studies have started to focus on revenue generation, providing evidence that participation might increase tax revenues or tax morale. The possibility of such a virtuous cycle of good governance is encouraging for governments attempting to boost revenue collection. Finally, studies focused on procurement indicate that more open practices reduce corruption and thereby ensure more effective spending of public funds.
This all indicates that the field of fiscal openness research has not only grown quantitatively but also matured qualitatively. Recent studies provide credible causal evidence, capture long-run and real-world effects, and provide more nuanced insights about when, where, and how fiscal openness can contribute to transforming governance and ultimately improve lives.
We end with some reflections that might help in advancing this field of studies even further:
From local to national. While rigorous causal claims have been made in decentralized settings (such as municipalities), much advocacy and funding focus on national-level reforms. By disaggregating outcomes or exploiting staggered rollouts, more causal evidence might be generated at national levels. National-level events – such as the release of budget information – can provide empirical opportunities to capture the impact of fiscal transparency, for instance with public opinion surveys.
Working with governments. We encourage more collaboration between researchers, practitioners and governments to better measure the impact of fiscal openness initiatives, and to improve the policy relevance of academic studies. This might include staggered or randomized rollouts of these initiatives to generate information about their impact.
From programs to systems. Budget systems consist of complex networks of institutions, including finance ministries, line ministries, legislatures, and auditors, and across different levels of government. Beyond studies of one-off interventions or specific programs, more work is needed to understand the interlinkages among different actors over longer periods.
Clarifying the transparency-participation nexus. There is a striking lack of evidence on the relationship between transparency and participation: when are they complements or substitutes? If the latter, what mechanisms are more effective, and under what conditions? More widely, are there limits to participation in budgetary decisions? Future research should tackle these questions.
Understanding trade-offs, especially in low capability contexts. More research is needed to consider opportunity costs and whether focusing on transparency or participation might come with negative externalities for other government actions one might value. Progress in this area, and others, would profit from deeper engagement between scholars using qualitative and quantitative methods, and between researchers and practitioners.
[1] Martin Haus is a doctoral candidate at the London School of Economics and Political Science (LSE), Joachim Wehner is Associate Professor in Public Policy at the LSE, and Paolo de Renzio is 2021/22 Policy Fellow at the Center for Advanced Study in the Behavioral Sciences at Stanford University.
Note: The posts on the IMF PFM Blog should not be reported as representing the views of the IMF. The views expressed are those of the authors and do not necessarily represent those of the IMF or IMF policy.
