Production function for measuring returns to scale in the garment sector: a case study of Bangladesh

Abstract

This paper empirically examines whether the production function of the garment sector of Bangladesh is operating under increasing returns to scale. The Cobb-Douglas production function is estimated using the advanced level of econometric techniques based on primary data. GLS estimates indicate that the productivity of labor force is 22.17% and the productivity of capital investment is 61.02%, statistically significant at any significance level. Thus, the production function of the garment sector of Bangladesh is operating under decreasing return to scale. From the GLS estimate of R2 , it is clear that about 99.97% of the total variation of the dependent variable output is explained by the fitted regression equation. Thus, the model fits the data very well. The diagnostic test results show that there are no problems of autocorrelation, heteroscedasticity, autoregressive conditional heteroscedasticity and normality of the random error terms. The test results also confirm that the model is correctly specified. It is also found that the average marginal productivity of labor force is 0.0236 and the average marginal productivity of capital investment is 0.5222 in the garment sector of Bangladesh. Since the marginal productivity of labor force is smaller than the capital investment, it can be said that these factories cannot reduce production costs by shifting resources from capital intensive techniques to labor intensive techniques. As a result, this sector cannot generate additional employment opportunities, which is not a good sign. The CUSUM and CUSUMSQ tests results confirm that the preferred production function of the garment sector of Bangladesh can be used for policy decision-making purposes.