“Combating the Crisis in Government Accountability: A Review of Recent International Experience”

Dilip Mookherjee

IED Discussion Paper 117, December 2001

Many developing countries and transition economies are currently occupied with (or contemplating) reforms in the organization of public service delivery. Most of these reforms have been targeted at problems of resource diversion and inappropriate project selection that seem to characterize traditional development programs administered by centralized bureaucracies. These problems are commonly thought to arise from the absence of accountability pressures on government employees appointed by a distant central government, and accentuated by poor information available to policy makers concerning citizens’ needs and priorities. Dilip Mookherjee’s paper provides a perspective on the types of reforms recently initiated and their effects in a number of developing countries.

Mookherjee distinguishes between different categories of reforms aimed at enhancing accountability. The most traditional of these consists of reforms within the centralized bureaucracy, whereas the more recent ones substitute bureaucratic implementation with decentralized mechanisms. The former category of reforms were designed to implement superior incentive schemes for bureaucrats, involving salary reforms combined with increased effort at monitoring bureaucrat performance. Examples are tax administration reforms in some Latin American and African countries following the fiscal crises of the 1980s. In many cases, however, these types of reforms typically ran into

design and implementation problems. Most important among these were lack of political will to reform bureaucracies from within, combined with a tendency of high level superiors to stall and prevent high level crackdowns.

Mookherjee argues that the frustrating experiences with attempting to reform bureaucracies directly have led an increasing number of countries to experiment with the alternative approaches involving decentralization of delivery systems. The common element has been a reallocation of authority over the delivery system away from centralized bureaucracies to citizens or to their locally elected representatives. Two forms of local empowerment are distinguished in the paper. The first involves political decentralization, or delegation of authority over service procurement and delivery to locally elected governments. The second alternative involves civic decentralization, direct empowerment of citizens or their non-governmental representatives. Detailed examples of political decentralization from various countries such as Brazil, Bolivia, Uganda and Philippines are discussed, followed by schooling reforms in Mexico, Uganda, El Salvador, Nicaragua, Pakistan and Colombia which directly empowered local communities and parent associations.

While data and endogeneity problems make an exhaustive evaluation of these reforms difficult at this point, it seems that decentralization is capable of significant impact if properly implemented. This requires clear demarcation of roles of local institutions via constitutional amendments, complementary financing arrangements, and expansion in citizens’ access to information and legal resources. Mookherjee ends by discussing lessons for policy makers in India, and directions for future research.