Modelling technical change in Italian agriculture: a latent variable approach

This paper presents an alternative approach to the measurement of technical change. It is based on the latent variable level of technology that enters explicitly the input demand system and on a hypothesis about the innovation generating process. By adding measurement error equations, the behavioral system can be viewed as a Multiple Indicators/Multiple Causes (MIMIC) model. The parameter estimates are obtained with a maximum likelihood estimator which involves the implicit covariance matrix. The analysis refers to Italian agriculture and the results provide some evidence on the nature and level of technical change during the years 1961-1991. © 2000 Elsevier Science B.V. All rights reserved.