If you look at the RBC study you're referring to and other studies by mostly banks, they would all point to an about 1% reduction in GDP should the U.S. withdraw from NAFTA. They are pretty much consistent.
The reason is simple. These are macroeconomic models, and they don't allow for what I explained at the very beginning. They don't have a lot of sectors in them. They might have two or three. They're very aggregate. They don't have the tariff changes in there. They impose a shock, and then you get a 1 percentage point reduction in GDP.
When you do it with a CGE model, the type of model we assess, you have 57 sectors, so when you make changes in tariffs, for example, then the sectors can adapt and workers that are having a harder time in that sector can move to another one. There is adaptation across sectors and over time.
Any studies by the banks that use macroeconomic models don't provide for that. They will point to very different results. So that's one thing.
In terms of the policy uncertainty and investments, there are certain things that we would have loved to take into account in the model because we do agree that a reduction in uncertainty is a good thing for business. That is partly why we took so much time; we re-did all the analysis a few weeks ago to try to do it the way the USITC has done it and see what it would provide for Canada.
That was very difficult to do. We didn't have the model or the data ex ante before putting it in the bigger model. We even tried to take USITC's coefficient. We thought, “The USITC does good work. We'll take its coefficient and put it in our model and see that we get.” We got results that did not make sense. We had an impact on Mexico that was much bigger than for Canada and the U.S.
So, there are other interventions done in the USITC report that I cannot explain just by looking at it. Even though it's 359 pages, I cannot tell you exactly what other interventions they might have made in the model.