Abstract
The impact of public infrastructure on output or productivity is a subject of continuing debate. At one extreme are studies that suggest that there are high rates of return to infrastructure investment and those that suggest that the impact is essentially zero or negative. In this paper, we argue that it is inconceivable that efficient investments in large stocks of public capital would provide an output/productivity benefits beyond the direct provision of amenities, especially for developing countries like those in Sub-Saharan Africa where public infrastructure stocks are almost certainly below optimal levels. Using firm level data and key physical public infrastructure assets in Uganda, we empirically test the hypothesis that the impact of of infrastructure development on output/productivity is more significant in an economy where there are bottlenecks causes by and underdeveloped infrastructure.
To take into account the serious deficiencies in models used in previous studies, we extend the basic production function approach and apply different formulations and functional specifications. Our approach is comprehensive as individual measures and a composite index are used in the analysis. Final conclusions are based on the results from the model with the preferred functional form, which is decided on the basis of statistical performance and consistency with theory. The findings from our chosen model (translog production function) are that the estimated elasticity between public infrastructure and private sector production is positive, big in magnitude, and significantly different from zero (at 1% level). We also find complementarity between public infrastructure and private capital and substitutability between public infrastructures and private labour employment. The results provide rationale for increased efficient public infrastructure investments in Uganda.
To take into account the serious deficiencies in models used in previous studies, we extend the basic production function approach and apply different formulations and functional specifications. Our approach is comprehensive as individual measures and a composite index are used in the analysis. Final conclusions are based on the results from the model with the preferred functional form, which is decided on the basis of statistical performance and consistency with theory. The findings from our chosen model (translog production function) are that the estimated elasticity between public infrastructure and private sector production is positive, big in magnitude, and significantly different from zero (at 1% level). We also find complementarity between public infrastructure and private capital and substitutability between public infrastructures and private labour employment. The results provide rationale for increased efficient public infrastructure investments in Uganda.
Original language | English |
---|---|
Place of Publication | Den Haag |
Publisher | International Institute of Social Studies (ISS) |
Number of pages | 70 |
Publication status | Published - Apr 2006 |
Externally published | Yes |
Publication series
Series | ISS working papers. General series |
---|---|
Number | 424 |
ISSN | 0921-0210 |
Series
- ISS Working Paper-General Series