Abstract
In recent years, a number of studies have shown that institutions are crucial to economic development. This literature has highlighted a positive relationship between the quality of the macroinstitutional environment and the performance of reform policies conducted in some key sectors of the developing countries' economies. This paper explores this relationship for the case of telecommunications, the sector among the infrastructure industries that worldwide has probably experienced the most important structural changes. More specifically, we investigate the extent to which this relationship is robust to the level of development of these countries. We specify an econometric model for a data base of observations on a panel of 32 developing countries covering the period 1985-1999. The sample is decomposed into two panels according to GDP per capita, thus allowing us to test for the existence of a level of revenue below which, once the effect of reforms variables, in particular, market liberalization and privatization of the incumbent, has been controlled for, the quality of institutions plays only a minor role. We find that the impact of the institutional quality on the performance of the industry is more perceptible in the sample of countries with the lower GDP per capita. Thus, within the developing countries, the marginal effect of an investment in improving the institutional quality is higher in the countries with lower revenues. This result is in line with the recent trend of international donors to reallocate resources to long term policies for improving institutional mechanisms in less developed countries.
| Translated title of the contribution | Institutional quality and economic performance: The case of telecommunications in developing countries |
|---|---|
| Original language | French |
| Pages (from-to) | 51-81 |
| Number of pages | 31 |
| Journal | Revue d'Economie du Developpement |
| Volume | 23 |
| Issue number | 3 |
| DOIs | |
| Publication status | Published - 1 Dec 2009 |
| Externally published | Yes |