Comparative advantage
A Socratic walk-through of comparative advantage — reasoned out one step at a time, not lectured.
The question we started with
THE QUESTION #How can two countries both gain from trade when one of them produces everything more cheaply?
Suppose one country makes everything more efficiently than another — less labour per bolt of cloth, less labour per cask of wine, no exception anywhere. Common sense says it should make everything itself and buy nothing. Yet economists insist both countries gain from trading, and that the weaker one is not being pitied in the process. That sounds like arithmetic being smuggled past you. So let us not take it as a slogan; let us build the numbers and see whether they work.
Reasoning it through
REASONING #Two countries, Alpha and Beta; two goods, cloth and wine; labour hours the only cost. Alpha needs 2 hours per bolt of cloth and 4 hours per cask of wine. Beta needs 12 hours per bolt and 8 per cask — so Alpha is better at both, six times better at cloth and twice as good at wine. Give each 240 hours, split evenly to start:
| | Alpha | Beta | World | |---|---|---|---| | Cloth (120 hours each) | 120 / 2 = 60 | 120 / 12 = 10 | 70 | | Wine (120 hours each) | 120 / 4 = 30 | 120 / 8 = 15 | 45 |
Now the question that turns the puzzle. Ask not "who is better?" but "what does each country give up?" A cask of wine costs Alpha 4 hours, and those hours would have made 2 bolts of cloth — so Alpha's wine costs 2 cloth. A cask costs Beta 8 hours, which would have made 8/12 of a bolt — so Beta's wine costs two-thirds of a cloth.
Read that again, because it is the whole idea. Beta is worse at wine in every absolute sense, and yet wine is cheaper for Beta — cheaper in the only currency that matters when a country decides what to do with an hour, which is the other thing that hour could have made. Alpha's superiority at cloth is so overwhelming that every hour it spends on wine is an expensive hour; Beta, uniformly bad, sacrifices little by choosing wine.
So let each move toward what it gives up least. Beta puts all 240 hours into wine: 30 casks. Alpha makes wine only up to what is still needed — 60 hours for 15 casks — and puts the remaining 180 hours into 90 bolts of cloth.
| | Before | After | |---|---|---| | Cloth | 70 | 90 | | Wine | 45 | 45 |
The same wine and twenty more bolts of cloth, out of the identical 480 hours. Nobody worked harder and no technology improved; the gain came only from hours moving to where their sacrifice was smallest.
Does each country get a share? That depends on the trade ratio, and any ratio strictly between the two opportunity costs — between two-thirds and 2 bolts per cask — leaves both better off. Take 1.2 bolts per cask: Beta ships 15 casks to Alpha and receives 18 bolts.
| | Alpha before | Alpha after | Beta before | Beta after | |---|---|---|---|---| | Cloth | 60 | 72 | 10 | 18 | | Wine | 30 | 30 | 15 | 15 |
Both consume more cloth than before and no less wine. The twenty extra bolts split twelve to Alpha and eight to Beta — and note that Beta, the country worse at everything, gained.
The analogy
THE ANALOGY #A surgeon who happens to be the fastest typist in the hospital still hires a secretary. Not because she cannot type, but because an hour of her typing costs an operation, while an hour of the secretary's costs almost nothing else.
the surgeon and the secretary have very different pay, which makes the choice obvious in money terms, whereas between countries the wage levels and exchange rate are themselves outcomes of the trade rather than fixed facts standing outside it — so the ratios settle the direction of trade, but not who captures how much of the gain.
Clarifying the model
THE MODEL #Three honest qualifications. First, the gains are aggregate: the countries as wholes consume more. Inside Alpha, the workers and owners in wine watch their industry shrink, and "the nation gained twelve bolts of cloth" is no consolation to someone whose skills, town and pension were built around wine. Those distributional losses are real, concentrated and often permanent for the people concerned, while the gains are thin and spread across everyone — which is why trade politics sounds nothing like trade theory.
Second, the argument assumes labour moves freely within a country and not at all between them, with no transport costs, no economies of scale and full employment. Relax those and the conclusion softens rather than reverses.
Third, it explains the direction of trade, not a country's destiny. Which goods you have low opportunity costs in is not fixed by nature; it shifts with education, capital and technology, which is the serious argument behind industrial policy.
A picture of it
THE PICTURE #How to readThe two left-hand blocks are what each country produces after specialising, and each ribbon flows to whoever ends up wearing or drinking it — so a ribbon crossing between the families is an export. Thickness is the number of units, so compare cloth with cloth and wine with wine, since a bolt and a cask are not the same thing. The point to notice is that Beta's block contains no cloth at all, and yet Beta consumes 18 bolts, nearly double the 10 it made for itself before trade.
What became clearer
WHAT CLEARED #Being worse at everything is no disqualification, because production is not a contest of skill but a choice about what to give up. Every hour spent on one good is an hour not spent on the other, and the country that sacrifices least in making a good should make it — even if it makes it badly. Trade then converts that reallocation into more goods for both sides. The arithmetic is genuinely sound; what it does not tell you is who inside each country bears the cost of the reallocation.
Where to go next
ONWARD #- What sets the actual exchange ratio between the two limits, once demand rather than cost is doing the deciding.
- Why so much real trade runs between similar rich countries in similar goods, which comparative advantage alone does not explain.
Key terms
TERMS #| Term | What it means |
|---|---|
| Opportunity cost | what must be given up to produce something, measured in the next best use of the same resources. |
| Absolute advantage | producing a good with fewer inputs than another producer; notably not what decides who should specialise in it. |
| Terms of trade | the ratio at which the goods are actually exchanged, which decides how the gains are divided. |
Every term the collection defines is gathered in the glossary.