The political economy of efficient public good provision: evidence from Flemish libraries using a generalised conditional efficiency framework
Provision of most public goods (e.g., health care, library services, education, utilities) can be characterised by a two-stage ‘production’ process. The first stage translates basic inputs (e.g., labour and capital) into service potential (e.g., opening hours), while the second stage describes how these programmatic inputs are transformed into observed outputs (e.g., school outcomes, library circulation). While the latter stage is best analysed in a supply-demand framework, particularly in the former stage one would like to have efficient public production. Hence, unlike previous work on public sector efficiency (which often conflates both ‘production’ stages), this paper analyses how political economy factors shape efficient public good provision in stage one (using local public libraries as our centre of attention). To do so, we use a specially tailored, fully non-parametric efficiency model. The model is rooted in popular Data Envelopment Analysis models, but allows for both outlying observations and heterogeneity (i.e., a conditional efficiency model). Using an exceptionally rich dataset comprising all 290 Flemish public libraries, our findings suggest that the ideological stance of the local government, the wealth and density of the local population and the source of library funding (i.e., local funding versus intergovernmental transfers) are crucial determinants of library efficiency.
|Date of creation:||Jun 2009|
|Date of revision:|
|Contact details of provider:|| Web page: http://feb.kuleuven.be/Economics/|
When requesting a correction, please mention this item's handle: RePEc:ete:ceswps:ces09.10. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (library EBIB)
If references are entirely missing, you can add them using this form.