Comparative advantage explained with Ricardo's wine and English cloth

The most counterintuitive idea in economics is also one of the best-supported. It says that a country that is worse at making everything still gains from trade, and that its better-at-everything trading partner gains too. Most people hear that and conclude that economists have been out in the sun too long. Here is the claim, tested with arithmetic you can check on the back of an envelope.

The idea is called comparative advantage, and it was set out by the English economist David Ricardo in 1817 in his book On the Principles of Political Economy and Taxation. Conservatives argue about trade constantly, and they should. But they should argue from the same understanding of the theory, and then show exactly where it holds and where it breaks. I will do the first now and the second afterwards.

Terms first

Absolute advantage means being able to make something using fewer resources than another producer. If Portugal can make a bolt of cloth with 90 workers' labour for a year and England needs 100, Portugal has the absolute advantage in cloth.

Opportunity cost is what you give up to get something. If a worker spends a year on cloth, the opportunity cost is the wine that same worker could have made in that year.

Comparative advantage means being able to make something at a lower opportunity cost than another producer. This is the idea that matters, and it is different from absolute advantage. A country can have an absolute advantage in everything and still have a comparative advantage in only some things. That is the whole point.

The claim, then: trade is based on comparative advantage, not absolute advantage, and when each country specializes in what it gives up least to produce and trades for the rest, total output rises.

Ricardo's example, step by step

Ricardo used two countries, England and Portugal, and two goods, English cloth and Portuguese wine. His numbers, in the form in which they are usually repeated, were these. They are labour inputs, the number of men working for one year needed to make a given quantity of each good:

Look at the table. Portugal is better at both. It makes cloth with 90 men where England needs 100, and wine with 80 men where England needs 120. Portugal has the absolute advantage in everything. If absolute advantage were what mattered, Portugal would have no reason to trade with England at all.

Now compute the opportunity costs.

• In England, making one unit of wine takes 120 men, and the same 120 men could make 120 divided by 100, which is 1.2 units of cloth. So one unit of wine costs England 1.2 units of cloth.
• In Portugal, making one unit of wine takes 80 men, who could make 80 divided by 90, which is about 0.89 units of cloth. So one unit of wine costs Portugal about 0.89 units of cloth.

Wine is cheaper in Portugal, measured in cloth given up. Now look at it the other way:

• In England, one unit of cloth costs 100 divided by 120, which is about 0.83 units of wine.
• In Portugal, one unit of cloth costs 90 divided by 80, which is 1.125 units of wine.

Cloth is cheaper in England, measured in wine given up. So Portugal has a comparative advantage in wine and England in cloth, even though Portugal is better at both in absolute terms.

The gain from specializing

Without trade, suppose each country makes one unit of each good. England uses 100 plus 120, which is 220 men. Portugal uses 90 plus 80, which is 170 men. Total output in the world: 2 units of cloth and 2 units of wine.

Now let each country specialize.

• England puts all 220 men into cloth. That produces 220 divided by 100, which is 2.2 units of cloth.
• Portugal puts all 170 men into wine. That produces 170 divided by 80, which is 2.125 units of wine.

The world total is now 2.2 cloth and 2.125 wine, up from 2 and 2. No one worked harder. No one had a new machine. The same 390 men produced 0.2 more cloth and 0.125 more wine. That extra is the gain from trade, and it comes from nothing but arranging who does what.

Then they trade. They can divide the extra output in many ways, as long as each side ends up with more than it had without trade. England could swap some of its 2.2 cloth for wine at a rate between its own opportunity cost, 1.2 cloth per wine, and Portugal's, 0.89 cloth per wine. At a price of, say, one unit of cloth for one unit of wine, England gets its wine for a cloth cost of 1.0 where it would have cost 1.2 at home, and Portugal gets cloth for a wine cost of 1.0 where it would have cost 1.125. Both are better off. (That price, one for one, is my choice for illustration; the actual rate depends on bargaining and demand.)

That is the entire argument. Everything else is commentary.

A household example, labelled as hypothetical

If wine and cloth feel remote, try this. It is a hypothetical, with made-up numbers.

A lawyer bills $300 an hour. She is also the fastest typist in her office, so fast that she could type her own letters in half the time it takes her paralegal. The paralegal earns $50 an hour of the firm's money. Should the lawyer type her own letters?

The lawyer has the absolute advantage in both lawyering and typing. But every hour she spends typing is an hour she is not spending on billable legal work at $300. Her opportunity cost of typing is huge. The paralegal's opportunity cost of typing is small, because the alternative use of his time is worth less. So the paralegal should type and the lawyer should litigate, even though she types better. The firm makes more money, and the paralegal has a job. The lawyer has a comparative advantage in law, and the paralegal in typing.

The same logic applies to you in your own household. The best cook in the house may not be the one who should spend the most hours cooking, if that person's time is worth much more at work. That is not a political claim. It is arithmetic.

What the theory shows, and what it does not

Let me be clear about what the argument proves. It proves that, under its assumptions, free exchange between two countries raises total output and leaves each country with the potential to be better off. Notice the word potential. The theory shows that the gains are large enough that the winners could compensate the losers and still come out ahead. It does not say that they do compensate them. That is a political question, and conservatives should not skip over it, because the losers are real people.

Two more clarifications. First, comparative advantage does not say that trade is the only consideration. It is a statement about efficiency, about how much stuff a country can produce. Conservatives care about more than stuff. Second, it is not a claim that every trade deal is a good deal. Specific agreements can be well or badly written. The principle is about the exchange itself.

Where the assumptions break

A good argument states its assumptions and tells you where they fail. Ricardo's model assumes several things, and the three most important for a conservative debate are these.

1. Capital does not move. Ricardo explicitly assumed that capital stayed within the country. He thought a merchant would prefer to invest at home, with its familiar laws and people, and he noted the reluctance of the owner of capital to leave his own country. Now capital moves around the globe in a click. A factory can be built in whichever country has the lowest costs, even if the country has no comparative advantage in the sense Ricardo described. When capital moves, a country can lose an industry that its own workers could have done well, to a place that has an absolute advantage in cost, because of low wages, subsidies or lax rules. The classic model does not cover this, and the tariff-favouring conservative has a real point here.

1. Resources shift smoothly from one industry to another. In the model, the men who made wine in England move to cloth with no friction. In real life, a worker who loses a job in a factory takes months or years to find another, and may never earn as much. Whole towns are built around a plant. The theory says the nation as a whole gains; it says nothing about the town that loses. I looked at the evidence on this in The China shock was real, and what free traders and tariff hawks each get right.

1. The goods traded are simply goods. The model treats cloth and wine as interchangeable commodities that are good to have more of. It does not distinguish between a bolt of cloth and a guided missile, or between a bottle of wine and a supply of antibiotics. National security is outside the model entirely. A country that depends on a rival for something it would need in a war has a kind of risk that no efficiency calculation captures. Adam Smith himself, the great defender of free trade, accepted that defence was more important than opulence. I explored that point in Hamilton's Report on Manufactures and the conservative case for tariffs.

There are other limits worth naming. Some industries have learning effects: a country that starts early and builds expertise can gain a lasting advantage that was not present at the start, which is why Alexander Hamilton argued that a young nation needed protection to catch up. And some trade partners do not play by the rules. If a government subsidizes an industry heavily, steals designs, or manipulates its currency, the price signals the model relies on are distorted.

The fair hearing for the tariff conservatives

Conservatives who favour tariffs are not ignorant of this theory, and I will not caricature them. Their strongest objections run like this:

• The theory is about a world of friendly traders. We do not live in it. Some governments pursue trade as a tool of national power, and a rule that treats their exports like those of an ally is naive.
• Efficiency is not the only value. A nation that makes things has a skilled workforce, a base for defence and a stake in its own future. Losing those for cheaper goods is a bad bargain.
• The gains are diffuse and the losses concentrated. The savings are a few dollars on each of millions of purchases. The losses are a closed plant and a broken town. A government is right to care about the second as well as the first.

I agree with the premise of every one of those. I disagree that tariffs are the best answer to them. A tariff is a tax on the American buyer. It raises the cost of the goods that American factories need as inputs. It invites retaliation against American exporters. And it hands a power over prices to whichever industries are best at lobbying. A narrow, targeted measure for a real security interest is defensible. A broad tariff as a general strategy is not, and the reasoning in this post is why. The accounting side of the debate, about what a trade deficit does and does not mean, is in a trade deficit is not a loss, and the accounting identity proves it.

What it means for a family

Return to the kitchen table. A family does not make everything it consumes, and no one thinks it should. The father is not a farmer, the mother is not a weaver, and the family trades its labour for a paycheque to buy what others make better and cheaper. That is trade, and the family is richer for it. A country is a large family in this respect. A tariff on a good the family buys is a tax on that family. If the tax protects a job that a neighbor holds, there is a fair argument to be had about whether the job is worth the price. But the price should be stated, and the family should see the bill.

The founding generation understood trade as a matter of liberty as well as wealth. The Constitution's Article I, Section 9 forbids a tax on exports, and Section 10 forbids the states from taxing imports without congressional consent, so that the country would be one market. Free exchange within the country is why a farmer in Ohio can sell to a buyer in New York. Ricardo's logic is the same logic, applied across a border.

You can find the official trade figures on the numbers page, and you can read the text that gives Congress the power to regulate foreign commerce on the Constitution page.

The case in brief

1. David Ricardo showed in 1817 that trade is governed by comparative advantage, the lower opportunity cost, and not by absolute advantage.
2. In his example, Portugal is better at making both wine and cloth, yet specializing and trading raises the output of 390 men from 2 cloth and 2 wine to 2.2 cloth and 2.125 wine.
3. The same logic works in a household: the lawyer should not type her own letters, however fast she types.
4. The theory has limits: capital moves across borders, workers do not shift between industries without cost, and goods vital to national defence are not interchangeable with cloth.
5. The tariff conservatives are right about those limits and wrong that a general tariff is the cure; the better answer is targeted measures for real security interests and honest help for displaced workers.

What you can do: pick three things in your home that were made abroad, find out what each would cost if made only by American producers, and bring the figures to the next town hall where trade is discussed.