Pardon the wonkiness, but this point is an important one to get correct in discussions of the economics of trade and of trade policy.
You write: “Ricardo’s theory of comparative advantage is bedrock for the market fundamentalist case for free trade. But [Oren] Cass is unquestionably right that Ricardo [in explaining comparative advantage ] assumed that capital stayed at home. That is a big reason why classical economists would object to the globalization we’re in today.”
I’m afraid that Oren’s and your understanding of Ricardo is no more valid than is his and your understanding of Adam Smith.
To explain comparative advantage , Ricardo did indeed assume that capital is not invested abroad. But he made this assumption to demonstrate that even when, say, Portugal can produce all goods using fewer resources for each good than is required for such production in, say, England, Portugal can nevertheless gain by importing some goods from England – and England can gain by exporting some goods to Portugal. This counterintuitive result exposes the error of those who insist that the ‘more efficient’ Portuguese can’t possibly gain from freely trading with the English, and the error of those who claim that the ‘less efficient’ English, were they to trade freely with the Portuguese, would have no demand from Portugal for English exports.
What matters for trade, said Ricardo, is not how many resources are used in one country to produce wine and how many resources are used to produce cloth compared to the amount of resources used in another country to produce these outputs. What does matter (to use Ricardo’s example) is how much the production of cloth in Portugal falls when the Portuguese produce an additional pipe of wine compared to how much the production of cloth in England falls when the English produce an additional pipe of wine. If the Portuguese sacrifice less cloth to produce a pipe of wine than do the English, the Portuguese have a comparative advantage at producing wine and the English have a comparative advantage at producing cloth. Therefore, people in both countries gain from trading if the Portuguese specialize in producing wine and the English specialize in producing cloth.
Furthermore – and this point is especially important – allowing for unrestricted cross-border investment does nothing to change the essence of the matter. In Ricardo’s example, if cross-border investment were allowed, capital would flow to Portugal. Such investment might well change which country has a comparative advantage at producing wine and which at producing cloth. But as long as the amount of cloth sacrificed in England to produce a pipe of wine differs from the amount of cloth sacrificed in Portugal to produce a pipe of wine, one country will have a comparative advantage at producing wine and the other country a comparative advantage at producing cloth. And so both countries will still gain from freely trading with each other.
To be fair to Oren, he’s not the first person to be led by Ricardo’s exposition to the mistaken conclusion that a policy of free trade is beneficial only in the absence of international investment flows. As you know, in this paper  from long ago I tackled some of Oren’s mistaken predecessors on this front. But that an error is commonplace doesn’t make it less of an error.
The complete case for free trade, far from prohibiting international capital flows, recommends the removal of all such prohibitions. To assert or suggest that international capital flows nullify the principle of comparative advantage and the case for free trade is to reveal a failure to understand comparative advantage and the economics of trade.
Donald J. Boudreaux
Professor of Economics
Martha and Nelson Getchell Chair for the Study of Free Market Capitalism at the Mercatus Center
George Mason University
Fairfax, VA 22030