The Existence of Factor Substitution in the Primary Aluminium Industry. A Multivariate Error Correction Approach on Norwegian Panel Data

Abstract

This paper presents an econometric analysis of factor demands in the Norwegian primary aluminium industry using annual panel data for individual plants. Focus is on testing theoretical and technical restrictions. The translog cost function approach is applied, and a multivariate error correction model with the cost shares of labour, raw materials and electricity with individual fixed effects is estimated. Capital is assumed to be quasi-fixed. The hypothesis of fixed input coefficients in this industry is rejected, but the estimated own and cross price elasticities suggest that relative price variation has limited effect on conditional factor demands.