Motivating bureaucrats: Group incentives and organisational performance of local governments

Project State

One key challenge in the decentralisation of the public sector is to be able to manage a large number of bureaucrats, who are the ones who are ultimately in charge of executing public works and delivering public goods. Not only is there a relative scarcity of qualified potential employees, but also employment in the public sector has been commonly used as a patronage tool, filling posts in local governments with political allies who might not be qualified to execute their tasks.

The proposed project aims to provide empirical answers to core questions in the decentralisation of the public sector literature, investigating ways in which incentives can be used to improve productivity for public sector workers.  Implementing high power incentives in the public sector (unlike in the private sector) is difficult for a number of reasons: (i) high unionisation usually restricts the implementation of pay for performance schemes, (ii) contracts are not flexible, restricting the principal's power to terminate them, using termination as an incentive, (iii) as in other organisations, outcomes depend on the effort of multiple agents, making the moral hazard problem more salient, and harder to link individual effort to outcomes, and (iv) political appointments are prevalent, enhancing coordination problems in teams and reducing the power of internal monitoring to reduce shirking.

In this pilot project, we will carry on preliminary field exploration and piloting different group incentives to increase public sector productivity and reduce the incidence of political appointees in Beninese communes  (the lowest administrative level of government). In the subsequent project, we will randomly assign the universe of Beninese communes into a control or a treatment group, in which communes will enter a tournament in which the best performing ones will be rewarded with a bundle of non-monetary and monetary incentives (to be distributed among all the workers in the municipality).  The impact of this treatment will be tested on a wide variety of outcomes related to the performance of bureaucrats and the commune government. In particular, we will focus our attention in outcomes related to team performance and productivity, as well as intermediate outcomes related to the processes and “best practices”, such as (i) quantity and quality of public goods provided, (ii) time allocation, team composition, and collaboration within the organisation, (iii) worker satisfaction with their jobs and the work environment, (iv) citizen satisfaction with the performance of the local government, (v) checklist of “best practices” for local governments (e.g. existence of a development plan, budgeting processes, do they follow standard public sector procurement procedures, etc.), (vi) participation of political appointees in teams and hiring practices.