Abstract
This study examines the effects of international trade and investment on output and tests the null hypothesis of Granger non-causality among trade, investment and economic growth in Canada. The long-run model is estimated using several single-equation and system estimators to assess the robustness of results across methodologies. The single-equation, OLSEG, GMM, DOLS, NLLS and FMOLS, estimates of the model provide consistent support for the positive and significant long-run effects of exports and investment on output. The ML system estimates cross-validate the cointegrating relationship and reinforce the positive effects of exports and investment and the negative effects of imports on output. The over-parameterized level-VAR estimates suggest unidirectional Granger-causality from exports, imports and investment each to output. The estimates of the model with structural breaks support the long-run relationship, though the evidence is not unambiguous ubiquitously across all the tests. The evidence supporting the positive and significant long-run effects overwhelms the evidence providing weak or no support for the effects of trade on output. The results underline the need for the acceleration of exports (and investment) to offset the demand-reducing effects of imports and escalate the altitudes of output and economic growth.