In 1817, the English political economist David Ricardo published the most counterintuitive idea in economics. Imagine that Portugal can produce both wine and cloth more efficiently than England. Common sense says Portugal should make both, England nothing. Ricardo's comparative advantage says: both countries gain if Portugal specializes in whichever it produces relatively better and trades for the other. Even when one party is worse at everything, mutually beneficial trade is possible, because the relevant question is not who is better in absolute terms but who has a lower opportunity cost. The result has been called 'the only proposition in social science that is both nontrivial and true.'
The mathematical proof is brief and the conclusion is robust to nearly all the obvious objections. The gains from trade arise from specialization according to opportunity cost; the gains exist regardless of absolute productivity differences; the gains accrue to both trading partners (though not necessarily evenly). The standard economic argument for free trade runs on Ricardo, and it has been the consensus position of professional economists for two centuries. The complications are also well-known. Comparative advantage assumes full employment of factors — but if displaced workers cannot easily move into expanding sectors, trade can produce localized devastation alongside aggregate gains. It assumes complete contracts and stable property rights — but international supply chains have introduced new failure modes (national-security dependencies, intellectual-property leakage, weaponized chokepoints). It assumes static comparative advantage — but advantage can be built through industrial policy (Korea, Japan, China), making the early-stage protection of infant industries a defensible deviation. The distributional consequences within trading countries — Stolper-Samuelson tells us that trade tends to hurt scarce factors, which in rich countries means low-skill labour — were systematically downplayed by free-trade advocates for thirty years and are now central to the political economy of trade.
The China shock — the 2001-onward acceleration of Chinese manufacturing exports and the displacement of American mid-skill manufacturing — has been one of the most studied recent natural experiments in trade economics. The empirical finding (Autor, Dorn, Hanson) is that aggregate gains were real but local losses were severe, and that the political backlash to those losses has reshaped American politics. The de-globalization of the 2020s — tariffs, industrial policy, supply-chain reshoring — is, in part, a political revolt against a trade orthodoxy that emphasized aggregate welfare while neglecting distributional consequences. Comparative advantage as a theorem is unaffected; comparative advantage as a policy slogan is in retreat.