What affects MFP in the long-run? Evidence from Canadian industries

In this paper we explore variables that may have an impact on multifactor productivity (MFP) in the long-run using the KLEMS database for Canada. We estimate a dynamic heterogeneous panel error-correction model of twelve 2-digit level industries. Variables investigated include ICT capital, outsourcing, competition, trade openness, public infrastructure, and R&D. Results suggest that over the 1976 - 2003 period ICT capital services, outsourcing and trade all had a positive impact on the level of industry MFP. The speed of adjustment varies significantly by industry.