Don Boudreaux favorably quotes Doug Irwin over at Cafe Hayek. Below is a slightly edited comment I left:
What Iriwn, like Hayek and Coase before him, points out I think is just brilliant: the scarcityists’ arguments are one long exercise in begging the question. They’re assuming they have the very knowledge they’re trying to show they can acquire. Yes, if one just happened to know the complete set of preferences and demand curves for all people in the nation, then one could create an optimal tariff or policy for that given moment in time. But it’s in getting that knowledge where the trick lies.
Note that the key word here is “knowledge,” not “information.” You don’t need data points to feed into a machine, but precise observations about the nature and time and place of each individual, observations the individual himself does not necessarily know. Collecting these necessary observations are impossible.
But there is another thing to keep in mind: Bastiat. Let’s grant the scarcityist’s assumptions and say we can set an optimal tariff policy. Such a policy is optimal in name only; it’s optimal only through incomplete accounting. It’s optimal only from the point of view of the country levying the tariff. But economics is not about only looking at one person in one time period (the seen). We must look at all people over all time periods (the unseen). An optimal tariff in the US may temporarily raise US net welfare, but at the same time, the world as a whole is made worse off. A poorer world means fewer buyers of US products and fewer sellers of goods to the US (exacerbated by the tariff). A poorer world also means increased instability, and likely more war. Both these effects will rebound on the US, leading to a poorer US as well over time. In short, even if we grant the assumptions of the scarcityists, the outcome from tariffs, when explored across all people in all time periods, is still negative.