Comparative advantage and trade

Portugal in 1817 could make both wine and cloth using fewer hours than England. David Ricardo used that example to argue that Portugal should nonetheless buy its cloth from England, and that both countries would end up with more. Two centuries later the result is still the one people find hardest to accept, because it says being better at something is not a reason to do it.

Two questions that sound the same

Who makes it in fewer hours? That is absolute advantage, and it is the question everyone asks first.

Who gives up less to make it? That is comparative advantage, and it is the one that decides who should.

Portugal makes a cloth in 2 hours, which is 2 wines it did not make. England makes a cloth in 4 hours, which is only two thirds of a wine it did not make, because England is slow at wine anyway. England is worse at cloth in hours and cheaper at cloth in what it costs to make, and the second one is what matters.

Try making one country better at everything by a factor of ten. The trade survives. Then set both countries' ratios equal, so each gives up the same to make cloth, and watch the whole thing collapse. That is the actual condition, and it has nothing to do with which country is richer.

The example is Ricardo's, from chapter 7 of On the Principles of Political Economy and Taxation, 1817, which is out of copyright and freely readable. The arithmetic above comes from the model on this page, which is unit tested, including the case where specializing lowers the output of one of the two goods.

ConventionComparative advantage says which pattern of production makes the most, and says nothing about how the gains get divided. Those are separate questions and they get answered by separate things: technology settles the first, and bargaining power settles the second. A lot of argument about trade is really argument about the second while quoting results about the first.

Where the simple version stops working

The model has two countries, two goods, no transport costs, no economies of scale, and labor that moves instantly from one industry to another. Every one of those is false, and the interesting question is which falsehood matters.

• Labor does not move instantly. A cloth worker in a town with one mill does not become a wine worker. The gains arrive in a decade and the losses arrive in a year, and they arrive to different people.
• Advantage is made, not found. Ricardo treats the hours as given. In practice a country that starts making semiconductors gets better at making semiconductors, which is the argument for industrial policy and the reason it keeps coming back despite being easy to do badly.
• Scale changes the answer. When making more of something lowers its unit cost, whoever gets there first can stay there whether or not they had any advantage to begin with. That is most of why a handful of countries make almost all of certain things.

What to do with this

When you see a claim that a country cannot compete, check whether it means absolute or comparative. Almost every version of that claim is about absolute advantage and is therefore answering the wrong question. A country with no absolute advantage in anything still has a comparative advantage in something, because the ratios cannot all be equal.

And apply it to your own hours, which is where it will actually pay. The task you are best at and the task you should be doing are different questions, and the gap between them is whatever your time would otherwise have produced.