From time to time, when I write a blog post I feel as though I’m setting a trap. Setting a trap is never my intention, but on such occasions I have a pretty good sense of the specific contents of the retorts that my post will elicit – retorts that will give me the opportunities to explain just why the retorts are mistaken.
I had such a sense when writing this post  earlier today and, sure enough, the retort that I felt confident would come actually came (as it happens in this case from Bret Wallach). My post made the point that the logic of protectionism, if it were correct, would suggest that individual households would maximize their material prosperity by refusing to trade with each other. Each household would be autarkic.
Here’s Bret Wallach’s comment , in full, on my earlier post:
And an anti-protectionist who makes such a claim has no concept that scale can matter, that while going from one person doing everything himself to a thousand people specializing is overwhelmingly more productive, going from a billion people trading to a trillion people trading might not be more productive at all, and even if it is more productive, will not yield nearly the increase as going from 1 to 1,000 people.
Scale does indeed matter, but contrary to Mr. Wallach’s evident supposition, this reality does not strengthen the protectionist’s case for trade restrictions. I have two responses to Mr. Wallach’s point, with the latter being the more fundamental – the one that I anticipated being given the opportunity to make.
First, scaling up from an individual being completely self-sufficient to that individual trading with a thousand people might well produce a larger increase in that individual’s material standard of living than would be produced if, as one of a million people who as a group start off trading only with each other, the group opens up trade such that the trading group now consist of a billion people. But maybe not. Suppose that when the group of a million people opens up trade with 999,000,000 other people, one of the 999,000,000 other people is an Alexander Fleming  who invents antibiotics or a Norman Borlaug  whose research greatly expands crop yields. Is it obvious that the additional lives saved and improved by having access to the fruits of such a person’s genius, creativity, and hard work would be a lesser improvement in living standards than that which would occur when one individual begins trading with 999 other individuals?
Remember that the number of possible ways for individuals to connect and to cooperate with each other – opportunities (as Matt Ridley says ) for ideas to “have sex” with each other – increases exponentially with the size of the population. These increasing returns to scale work against Mr. Wallach’s point. (See also this important article  by my late Nobel-laureate colleague Jim Buchanan and his co-author Yong Yoon on Adam Smith’s increasing-returns-based argument for free trade.)
Second and more fundamentally, even if there unambiguously are decreasing returns to the scale of the trading population, this fact means neither that the optimal scale is the population of the nation-state nor that the increased benefits to be gotten by expanding trade beyond the borders of the nation-state are somehow trivial or unworthy simply because the magnitude of these increased benefits aren’t as large as is the magnitude of the benefit of scaling up the size of the trading group from some unit smaller than the nation-state to the full nation-state. Decreasing returns to scale does not mean negative returns to scale. And decreasing returns to scale does not mean that the benefits of increasing the scale are necessarily less than the costs of doing so.
One benefit of free trade is that such trade allows the ‘optimal’ scale of the trading group to be discovered through actual, competitive market activities – with buyers and sellers spending their own money – rather than through political machinations which invariably – and invariably mistakenly – treat the nation-state as if it is economically something special or even meaningful.
Put differently, the case for free trade can be stated as a case for allowing markets rather than government to discover the ‘optimal’ scale of the trading group – which, do note, almost certainly differs for different goods and services.
Whatever the case, there is absolutely no good reason to accept the protectionist’s implicit assumption that the size and population of whatever nation-state he or she happens to be in somehow is the optimal scale of the trading group.