chapter  4
45 Pages

4 Principles and practice

WithROY BAHL AND RICHARD M . BIRD

Introduction The theory of expenditure assignment tells us that, under the right circumstances, certain services, including the infrastructure needed to provide them, can be more efficiently provided by subnational governments. At least in industrial countries, infrastructure investment is decentralized to a significant extent (Estache and Sinha 1995). In OECD countries, even such big projects as ports and airports are often locally managed and funded (Bel and Fageda 2009). In the European Union, for example, the share of subnational investment in economic infrastructure is 60-70 percent in the older member states and 40 percent in the new member states; subnational shares in social investments (e.g., in schools and hospitals) are even higher (Kappeler et al. 2012). In low and middle income countries, economic theory provides less guidance to best practice. This is mostly because the assumptions beneath the decentralization model do not fit the developing country setting as well. The result is a great deal of variation in the extent to which responsibility for infrastructure service delivery is assigned to local governments (Alm 2010). In the countries included in Table 4.1, for instance, the equivalent of a large share of subnational own revenue is devoted to infrastructure. In fact, the amount spent on infrastructure by subnational governments in Peru is greater than the amount they raise from own sources. Even in countries in which subnational governments are important investors in infrastructure, however, it is by no means obvious whether expenditures on infrastructure have been over-assigned or under-assigned to such governments, or whether the right expenditures have been devolved.2 The open question is whether a change in governance arrangements (or in the incentives to policy-makers that lie beneath these governance arrangements) would lead to an improvement in the delivery of infrastructure services.3 The institutional arrangements in question include not only the structure of subnational governments but the nature of political representation, spending and revenue mobilization powers, borrowing restrictions and practices, the degree of autonomy in management, the design and implementation of intergovernmental transfer systems, and the accountability of governments at all levels to those for whom they presumably speak.