Technology gap, imported capital goods and productivity of manufacturing plants in Sub-Saharan Africa

ABSTRACT

This paper uses firm-level data from Ghana, Tanzania and Kenya to examine the effect of capital goods imports on domestic firms' productivity, and the role firms' technology gap plays in aiding the transmission of knowledge embodied in capital goods to domestic firms. The results show that increasing imports of capital goods and closing technology gaps have positive effects on productivity. Furthermore, domestic firms with technology standards farther from international best practices benefit more from capital goods imports. The results also imply that trade liberalization policy aimed at eliminating tariffs on capital goods will significantly improve the performance of technically incompetent firms in the African manufacturing sector.

KEYWORDS:

• Capital goods
• technology gap
• productivity
• Sub-Saharan Africa

JEL Classification:

• O55
• O33
• F14

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Other firm-level studies also find similar results of positive relationship between trade and firm-level productivity. Using firm-level data from Indonesia, Blalock and Veloso (2007) show that importing is a source of international technology transfer. They also find that firms in industries supplying increasingly import-intensive sectors have higher productivity growth than other firms. Fernandes (2007) also finds similar evidence of positive effect of imports on the productivity of firms in Columbia.

2. To access this data, visit: http://www.csae.ox.ac.uk/datasets/cd/c d-main.html. We exclude data for firms in Nigeria and South Africa due to insufficient observations.

3. See for instance the works of Keller and Yeaple 2009; Keller 2010; Blalock and Gertler 2004; Sjoderbom and Teal (2002) also argue that substantial differences in capital intensity across firms in SSA may make labor productivity a very poor measure of firm performance. Hence rather than using labor productivity to measure firm performance, it is desirable to use TFP, which relates output to all inputs in the production process.

4. Blalock and Gertler (2009), Kasahara and Rodrigue (2008), Pavcnik (2002) and Topalova and Khandelwal (2011) employ a similar procedure for firm-level production function estimation.

5. For instance, Hu, Jefferson, and Jinchang (2005) argued that domestic firms seem to be insignificant for technology transfer in the case of China.

6. For this reason, we later examine the differential effect of capital goods imports on TFP for firms in different technology gap categories. See Section 5.4.

7. We have also carried out robustness checks to see if replacing log of capital and log of labor with firm's age, export status and ownership structure changes the results. The key results still hold. We do not report these extra results in the paper for brevity, but they can be requested from the authors.

8. A Hausman test on the relevant time varying variables also confirms our choice of FE estimation. We present both RE and FE results but we discuss FE results only.

9. We experimented by including a quadratic term for materials per worker and an interaction term between human capital and imported capital goods in an attempt to capture diminishing returns to materials per worker and the potential that knowledge from imported capital goods can be transferred through improved human capital. While the coefficients had the expected signs, they were not significant and the coefficient on the interaction of technology gap and capital goods was not qualitatively different from those reported here.

10. A good instrument should be uncorrelated with the error term and correlated with the endogenous explanatory variable (see, for example, Wooldridge 2006). In the data, tariff peaks are correlated with capital goods imports, but uncorrelated with the error term, and are thus a suitable instrument to use in the estimation.