Strengthening state capacity through reforming the civil service and traditional institutions

Where I do research and analytical work:

Research Overview

The inability of the state to implement policies, collect taxes, and govern the national territory is a fundamental and pervasive problem in developing countries, particularly those affected by issues of fragility, conflict and violence (FCV). While these states rely on civil servants—both in the formal public sector and in the informal traditional sector—for their capacity to effectively implement policies and to provide essential services, there is ample evidence that the lack of qualification and accountability of civil servants, corruption and patronage may compromise public service delivery and state effectiveness. Even when state agents are physically present, they often do not behave in accordance with their obligations and actions are often constrained as a result of customary practices, poor infrastructure and monitoring, low mobility, as well as the lack of financial resources to pay wages and provide incentives.

My research in this area is driven by the need to address these constraints (institutional or otherwise) in order to improve basic service delivery in weak states, both in the formal public sector, but also in informal traditional and/or non-state governance spaces. It covers a wide range of issues, including the reform of traditional institutions, accountability strengthening of service providers, civil servants’ recruitment and deployment strategies, patronage networks, misalignment of incentives and capacity issues that hinder the effective delivery of services. Bellow I discuss the two complementary workstreams in this research agenda.

• Research reforming traditional governing institutions in FCV

In another project in Zimbabwe, my co-author and I use an experiment covering 270 villages to investigate the effectiveness of an intervention designed to reform governance, accountability and social inclusion in the country’s customary institutions. Though existing studies suggest low accountability is a key source of mal governance of traditional institutions, we argue that this does not have to be the equilibrium. We show that, at least in the case of Zimbabwe, that strong norms shape deliberations in traditional institutions. Specifically, the experiment shows that broadening the group of potential advisers to village chiefs leads to more inclusive village governance by traditional leaders. We also find that incentivizing the inclusion of new civil society leaders in governing circles results in more open decision-making and better outcomes for the chief’s political opponents (e.g., more favorable court decisions and food aid).

Click here for information and publications on specific projects by country

Click here to access relevant publications, working papers and policy reports

• Research reforming traditional governing institutions in FCV

The second strand of my research in this area also focuses on constraints and civil service reforms in the public sector. Inadequate performance of civil servants is not only the result of a lack of monitoring by well-intended principals, but can be the result of the lack of adequate accountability mechanisms (Laudati et al., 2020) and complex networks of redistribution and patronage inside the state administration (Sanchez de la Sierra and Titeca 2017), which interact with poor selection of civil servants (often intentional) on key aspects such as their ability, the intrinsic motivation for public service, and proneness to corruption (Finan et al. 2015). A better understanding of how social networks operate to influence the behavior of civil servants, and how to optimally screen and allocate them across positions to improve their contribution to state performance, therefore, has the potential to significantly increase the performance of civil servants, state effectiveness and public welfare (Ashraf et al. 2017).

This strand of my research focuses on a couple of objectives. The first objective is to identify key constraints to human resources processes, including in effectively recruiting and deploying talents throughout public administration units (e.g., lack of skills, patronage networks, etc.), career management and advancement (e.g., salaries, promotions, misalignment of incentives, etc.), which often hinder effective public service delivery. For instance, I am working with the WBG and government of the DRC to conduct a nationally representative survey of civil servants working at all levels of government (n= 8,000). This survey is first of its kind. It will provide a detailed picture of how public institutions work and point to reforms needed to strengthen state capacity and to improve service delivery in FCV countries.

A second objective is to test targeted interventions designed to address some of the identified constraints to effective public service delivery. In one study, joint with Guo Xu (UC Berkley) and Raul Sanchez (Chicago), I am leading a field experiment, based on a joint WBG-DRC government project, that evaluates the recruitment and deployment of freshly minted young professionals (YPs) into the public administration. The program will train 1,000 YPs selected competitively from a pool of 45,000 applicants. Selected YPs are randomly assigned to administrative units across the DRC, accounting for their home base. The experiment’s factorial design will allow us to test the effectiveness of allocating YPs to administrative units as well as the marginal effects of insider allocation (i.e. being deployed to a unit in one’s home district) or outsider’s allocation (i.e., being deployed outside one’s home district) on existing patronage networks and on the performance the productivity of YPs.

Finally, in another experimental study joint with Peter van der Windt (NYU-AD) and Ann Laudati (UC Berkeley), we examine how to best strengthen the accountability of service providers in fragile states in order to improve service delivery, particularly in the education and health sectors. The study is a field experiment that uses factorial design across 339 in Eastern DRC to isolate the effects of community participation, government ministries’ participation or both in the decision-making related to donor-funded projects in these sectors. The data suggest no differences across treatments in the education sector. In contrast, we find that community participation improves the accountability of actors in the health sector, improves health services and increases villagers’ satisfaction with these services. To further explain these findings, we 1) explore the channels previous studies have suggested via which such change may occur (information, collective action and oversight) and 2) build on in-depth qualitative data to understand these findings.

Click here for information and publications on specific projects by country

Click here to access relevant publications, working papers and policy reports