عنوان مقاله [English]
نویسندگان [English]چکیده [English]
Empirical studies often employ Solow method to calculate productivity. However, potential correlation between inputs and the unobserved firm-specific productivity shocks in the estimation of production function parameters leads to a biased estimation. This implies that firms that have a large positive productivity shock may respond by using more inputs. In recent years, several techniques offering improvements over Ordinary Least Squares have been proposed for estimating production functions. These methods offer simultaneity bias and selection bias corrections not accounted for in an OLS framework. In this paper, we review the econometrics methodology associated with estimation of productivity, and use an econometrics method to control simultaneity bias and selection bias problems. We estimate the productivity of firms using Levinsohn and Petrin (2003) method and panel data for 10 selected 2-digit (ISIC2) manufacturing industries for the period 2000-2007. The results show that the average productivity growth rate is 1.79 per cent with the highest rate of 9.62 percent in 2001 and the lowest of rate -3.98 percent in 2002. Among the manufacturing industries, other non-metallic mineral products and textiles have the highest productivity growth and rubber and plastic products and electrical machinery and apparatus have the lowest productivity growth.