BIG tax change announced today. A “productivity mega deduction” broadens how much is eligible for immediate expensing. That cuts the effective tax on new investment in half. More to do, but this is big. Details: canada.ca/en/department-fina… #cdnecon

Pay more attention to people (or models) who forecast well, period. Reject rationalizations for why their past success might be temporary. These rationalizations always exist, but empirically, forecasting success is highly persistent over time.

I want to flesh out the basis for my argument that we shouldn't retaliate against US tariffs. Start by thinking of it as a cost-benefit analysis. The cost - higher prices for Canadian consumers and a weaker Canadian economy as a result - is pretty clear, and I don't think anyone is arguing that cost doesn't exist. We can spread it around through redistribution, but someone in this country has to bear the cost. So the question is: what exactly is the benefit, and why is it larger than the cost? The benefit must be that retaliating will change US behavior in some way that we like: repealing Section 338 tariffs, lowering Section 232 tariffs, agreeing to stop the 51st state nonsense, etc. I have not seen anyone actually make the theory of the case for how this is going to work. Just saying "game theory" doesn't cut it. (And the idea that Trump is a rational actor to whom you can apply game theory at all is ludicrous). Yes, I know the midterms are coming up. So what? Even if the Democrats take both houses of Congress, and even if they pass legislation to repeal Trump's tariff authorities, he can always veto that legislation. The probability that the Democrats will have a veto-proof majority is zero. Concretely, if you support retaliation, you need to make explicit and justify your assumptions about the probability that retaliating will affect some margin we care about. For example, why is the probability that Trump will reverse S338 increasing in the intensity of our retaliation? You also need to account for the fact that, instead of convincing Trump to do what we want, retaliating may instead cause him to escalate, which would cause more damage to our economy. (He announced that he's going to do exactly that in January.) That possibility has to be factored into the cost as well. And again, make your assumptions explicit: Why is the probability that retaliating will cause Trump to treat us better greater than the probability that it will make him treat us worse? The bottom line is that I have yet to see anyone make a convincing argument that the expected benefit from retaliating exceeds the expected cost. And I think Mark Carney should have to make that argument if he's going to retaliate. What are the expected goals of the retaliation package? On what timeline should progress toward these goals be measured? If no progress materializes after X number of years, how do we proceed? Do we just keep retaliating forever, even if the US keeps its current course? Surely there is no rational basis for doing that. Many people point out that Canadians strongly support retaliation, and that we have the stomach to tolerate any economic pain retaliation might cause. I agree that's true. But that pain is still for naught if it doesn't accomplish anything. Resolve alone doesn't yield outcomes. I agree that Canadians can tolerate a protracted trade war. That's not what matters. What matters is the likelihood that such a situation going to end with outcomes we want to achieve. Again, lay out the theory of the case. @stephenfgordon makes an interesting and compelling argument, namely that by the revealed preference theorem, strong support for retaliation implies that the value people assign to sticking it to Trump is greater than the cost, even if that value is purely subjective. That is far and away the best pro-retaliation argument I've seen so far. However, I don't think the job of a policymaker is to simply enact the policies that his or her constituents claim to prefer. Surveys show that strong majorities are in favor of economically stupid ideas like price controls, internally inconsistent fiscal policies (e.g. cutting taxes while increasing spending), and so on. Doing things just because people support them is folly. Policymakers should care only about the outcomes their policies lead to - and how people eventually judge those outcomes - not the content of those policies per se. And, on balance, my assessment is that retaliating is not likely to lead to better outcomes.