I spent last week engaging with researchers on issues of transparency and accountability, and it generated a few thoughts I’d like to share. I had the opportunity to meet with the great team of researchers and research managers on the T4D research team, involving Harvard’s Kennedy School of Government and R4D, and from there I went to the 2014 American Political Science Association’s (APSA) annual conference (on the theme ‘Politics after the Digital Revolution’), where I facilitated a working group on government transparency and accountability.
Questions and More Questions
The T4D team is looking at a very specific issue of accountability: community action to improve health outcomes. It will reveal a wealth of insights on this specific topic. At the APSA conference, the questions being asked (and partially answered) were much more varied. Yet in the working group, which was composed of an extremely diverse group of researchers, we came to the conclusion that the broad question that we need to wrestle with is:
How does transparency contribute to accountability?
Or phrased more specifically, what kinds of information allow which actors with what capacities and incentives to leverage which accountability mechanisms under what conditions?
Plug in credible answers to the components of that question, and you have the skeleton of a Theory of Change for a transparency and accountability campaign.
However, this formulation of the question about how transparency contributes to accountability is quite linear and may present such interventions in isolation from the broader context. Throughout the week, I was reminded of the importance of thinking about the systems in which accountability efforts are embedded. Accountability ecosystems are composed of dynamic interactions between actors, institutions, mechanisms and processes related to government responsiveness and accountability in a defined context. Thus, any specific pro-accountability strategy or action must consider (and potentially address) the other moving parts around it, which will have an impact on the intended outcomes.
At the APSA conference, many researchers were seeking to understand one particular aspect of the accountability ecosystem, and one often not explicitly addressed by many ‘transparency and accountability’ interventions: elections. For democratic governments, elections are the principal (and in some sense, only) mechanism by which citizens can hold their elected officials to account for their decisions and actions. However, in most formally democratic countries (including Western democracies), this accountability mechanism does not function as intended. The reasons behind this vary, but often include some combination of the following:
- Clientelism and vote buying
- Electoral fraud
- Voter intimidation
- One party state and/or unrepresentative political parties
- Ethnic voting
Several researchers at APSA were also thinking about information and voting patterns. What kinds of information do citizens need to be able to hold specific elected officials accountable for acts of corruption or failure to deliver public goods? The answer to this question varies significantly across contexts, but is important for understanding the roll that elections could play in accountability ecosystems, given the limitation mentioned above. (for one piece of evidence on this issue, see here)
The case of elections highlights a key element of accountability ecosystems: the potential gap between formal and legal rules, institutions and processes, and what actually happens on the ground. An additional example could be the existence of a Freedom of Information law on the books, which does not ensure that citizens have effective access to information. Officials may lack the capacity and/or incentives to comply with access to information requests, and when they do provide information, its quality and usability may vary significantly.
Finally, and briefly, accountability ecosystems allow us to ‘think outside the box’ about what factors and processes influence accountability. Stephen Kossak has written on the role of the political organization of the poor on government prioritization of primary education. At the APSA conference, one researcher presented evidence from South Korea, Taiwan and the Philippines tracing the mechanisms by which the former two countries were able to make much more progress on governance and control of corruption than the latter. An important initial enabling condition in Korea and Taiwan was not one most transparency and accountability efforts would take into consideration: land reform. In those two countries, successful land reform significantly reduced inequality, which in turn weakened clientelsim and elite capture, setting the stage for improvements in control of corruption and governance (even under weakly or non-democratic regimes). Neither political organization nor land reform features prominently in many approaches to government accountability and anti-corruption.
Measurement and Research Approaches
My engagement with researchers throughout the week reinforced the need for diverse and mixed methods approaches to answering questions about transparency and accountability, including evaluating campaigns and other efforts. The T4D research project is based on a very integrated mixed-methods design. Research presented at APSA was based on a diversity of research approaches. Historical process tracing is an important complement to surveys, experiments, ethnography and statistical approaches. Better still is to combine many of these approaches. Issues of governance and accountability are very complex, and don’t tend to boil down to a couple key factors. Thinking about accountability ecosystems reveals even more factors and influences related to a given outcome. In addition, approaches need to be flexible and adaptive, taking advantage of new windows of opportunity and shifting tactics when they aren’t showing progress. What kinds of research and evaluation tools will help practitioners and funders understand what is working and how to strategically leverage their scarce energy and resources? Process tracing can be very intensive, and isn’t perfect, but can help us drill down to better understand how causes and effects are linked, potentially enabling activists to understand and reflect on which strategies and tactics are best suited to their goals.