Estimating Production Functions Using Inputs to Control for Unobservables

Abstract

We introduce a new method for conditioning out serially correlated unobserved shocks to the production technology by building ideas first developed in Olley and Pakes (1996). Olley and Pakes show how to use investment to control for correlation between input levels and the unobserved firmspecific productivity process. We prove that like investment, intermediate inputs (those inputs which are typically subtracted out in a value-added production function) can also solve this simultaneity problem. We highlight three potential advantages to using an intermediate inputs approach relative to investment. Our results indicate that these advantages are empirically important.

Innovation and Firms’ Productivity Growth in Slovenia: Sensitivity of Results to Sectoral Heterogeneity and to Estimation Method

“…Additionally, our productivity proxy may fail to control for contemporaneous growth in inputs which may conceal the actual productivity dynamics. In order to control for this we use a TFP measure of productivity estimated by the Levinsohn-Petrin (2003) method. For obvious reasons this is done for manufacturing firms only.…”

Capital Misallocation During the Great Recession

“…Since traditional estimators used to overcome endogeneity issues (…xed e¤ects, instrumental variables) have not proven satisfactory for the case of production function, a number of semiparametric alternatives have been proposed. Both Olley and Pakes (1996) and Levinsohn and Petrin (2003) have developed a semiparametric estimator that addresses the simultaneity bias. The key di¤erence between the two methods is that Olley and Pakes (1996) use investment whereas Levinsohn and Petrin (2003) use materials used in production as a proxy for TFP.…”

“…Both Olley and Pakes (1996) and Levinsohn and Petrin (2003) have developed a semiparametric estimator that addresses the simultaneity bias. The key di¤erence between the two methods is that Olley and Pakes (1996) use investment whereas Levinsohn and Petrin (2003) use materials used in production as a proxy for TFP. Since data on investment is readily available and often non-zero at the …rm level but data on materials is not, I follow Olley and Pakes (1996) to estimate the production function.…”

Determinants Of Productivity Of Ukrainian Firms

“…In the KVED-2010, active from 1 January 2012, the number of service industries has been increased (a higher level of disaggregation) in comparison to KVED-2005 The definitions of variables used in our empirical study and their summary statistics are reported in Table 1. The level of firm productivity was measured by TFP calculated on the basis of the Levinsohn-Petrin (2003) methodology. Some industries were omitted from the analysis due to problems with calculating factor input shares.…”