Productivity of Canadian forest product industries: panel data evidence from five major provinces

Most studies of forest productivity in Canada have focused on a single product or a few provinces and most are outdated and do not reflect recent structural and technological changes. This paper addresses this gap by analyzing annual panel data from 2004 to 2020 for three subsectors of forest products and five main provinces. Accordingly, the authors applied the fixed effect for estimating the static model. A more flexible and advanced linear dynamic panel data method based on the Generalized Method of Moments was used to estimate the dynamic model, addressing issues such as the endogeneity of the lagged dependent variable, the small cross-sectional sample size and allowed us to restrict the number of instrumental variables. Both the Cobb-Douglas and the translog theoretical production functional forms are estimated. The financial shock in 2008 negatively impacted productivity. Productivity increased in the pulp and paper and wood products industries during the post-shock period compared to the pre-shock phase. The findings indicate disproportionate factor earnings for capital compared to its contribution to production, which warrants policy attention.