This paper deals with forest trade modelling from a theoretical, analytic and empirical perspective. An integrated dynamic log-lumber trade model is developed and then used to examine two trade issues, namely, a reduction of Russian taxes on log exports and removal of the taxes on Canadian lumber destined for the United States. To demonstrate the dynamic aspect of the model, both sets of taxes are lowered over a period of time. The trade model consists of five Canadian regions, three U.S. regions, New Zealand, Australia, Chile, Rest of Latin America, Russia, Sweden, Finland, Rest of Europe, Japan, China, Rest of Asia, and Rest of the World – a total of 20 regions. It concerns only coniferous logs and softwood lumber, ignoring hardwoods. The model is also calibrated on 2010 observed bi-lateral flows of logs and lumber using positive mathematical programming. The forest trade model is written using an Excel-GAMS interface, with input data retrieved by GAMS from Excel and GAMS output written to Excel, where final calculations are made.