How International Trade Works: Comparative Advantage and Tariffs

In 1817, David Ricardo published On the Principles of Political Economy and Taxation and used a short example about cloth and wine to make a point that still surprises people. Even if Portugal could make both goods with fewer hours of labor than England, both countries would end up better off if each specialized in what it did relatively better and traded. That idea, comparative advantage, is the core of the economic case for trade. This article works through the arithmetic, then covers deficits, tariffs, supply chains, and the evidence on who gains and who loses.

Educational only, not financial advice. Investing involves risk of loss; past performance does not guarantee future results. Consult a licensed financial advisor before making investment decisions.

The theory is sound, and it is also incomplete. It explains why open trade raises total output, but it says little about who inside each country absorbs the costs. Understanding both halves is the difference between treating trade as a miracle and treating it as a villain.

Key takeaways

• Comparative advantage, not absolute advantage, drives trade. A country gains by specializing where its opportunity cost is lowest, even when another country is more productive at everything.
• Trade is positive-sum in total but not for everyone. Economists have documented concentrated job losses in import-exposed regions alongside diffuse gains for consumers.
• A trade deficit is not a loss. It is the other side of capital inflows, and its size depends on saving and investment as much as on competitiveness.
• Tariffs are taxes paid mostly at home. Studies of the 2018-2019 US tariffs found they raised prices for US buyers and hurt, on net, manufacturing employment.
• Supply chains trade resilience for efficiency. The CHIPS and Science Act of 2022 is a deliberate decision to pay for more domestic chip capacity.

Ricardo's example, worked through

Ricardo's illustration uses two countries and two goods. The numbers below are the hours of labor needed to make one unit of each good.

	Hours to make 1 cloth	Hours to make 1 wine
England	100	120
Portugal	90	80

Portugal has an absolute advantage in both goods: it needs fewer hours for each. The intuitive conclusion is that Portugal should make everything and England has nothing to offer. Ricardo showed that conclusion is wrong, because what matters is not how many hours a good takes but what you give up to make it. That sacrifice is the opportunity cost.

• In England, an hour spent on cloth is an hour not spent on wine. One cloth costs 100 hours, which could have made 100 / 120 = 0.83 units of wine.
• In Portugal, one cloth costs 90 hours, which could have made 90 / 80 = 1.125 units of wine.

England gives up less wine to make cloth, so England has the comparative advantage in cloth. Portugal gives up less cloth to make wine, so Portugal has the comparative advantage in wine. Here is what the gain looks like in numbers.

	England	Portugal	Total
Labor available (hours)	220	170	390
Without trade: each makes 1 cloth and 1 wine	1 cloth, 1 wine	1 cloth, 1 wine	2 cloth, 2 wine
With specialization	2.2 cloth	2.125 wine	2.2 cloth, 2.125 wine

Without trade, each country uses its 220 or 170 hours to make one cloth and one wine. With specialization, England puts all its labor into cloth (220 / 100 = 2.2 units) and Portugal puts all of its into wine (170 / 80 = 2.125 units). The world now has more of both goods than before: 2.2 cloth instead of 2, and 2.125 wine instead of 2.

Now let the countries trade at one cloth for one wine, a ratio that sits between the two opportunity costs. England sends 1.1 cloth to Portugal and receives 1.1 wine. England ends with 1.1 cloth and 1.1 wine, and Portugal ends with 1.1 cloth and 1.025 wine. Both countries have more than the 1 cloth and 1 wine they had alone. Nothing was taken from anyone. The gain came from using each country's labor where it was relatively most productive.

Any exchange ratio between 0.83 and 1.125 wine per cloth leaves both countries ahead, and the exact ratio decides how the gain is split. That is why trade negotiations are fights over terms and not only over access.

Where the idea came from, and why it is hard to accept

Ricardo did not work alone. According to the summary on Wikipedia's comparative advantage entry, an anonymous 1814 pamphlet, Considerations on the Importation of Foreign Corn, contains the earliest recorded formulation, and Robert Torrens published a version in his 1815 Essay on the External Corn Trade. Ricardo's 1817 treatment is the one economists remember because his cloth-and-wine example made the logic concrete.

The idea has an unusual reputation. The mathematician Stanislaw Ulam is reported to have challenged Paul Samuelson to name a proposition in the social sciences that is both true and non-trivial. Samuelson's answer, in a 1969 address, was comparative advantage. He argued that it is logically true, and that its non-triviality is shown by how many intelligent people have been unable to believe it after it was explained to them. Sources differ on when Ulam first posed the challenge.

The reason it feels wrong is that people think in absolute terms. A country that is better at everything seems to have no reason to buy from anyone. The reply is that every country has only so much labor, capital, and time. Using scarce resources to make something a partner can make at lower sacrifice is waste, no matter how good the first country is at it.

How much trade there actually is

The world trades far more, as a share of output, than it did fifty years ago. The World Bank measures trade as the sum of exports and imports of goods and services as a percentage of GDP.

Year	World trade (% of world GDP)
1970	25.8
1990	38.1
2000	50.1
2008	60.3
2019	55.8
2022	62.2
2024	56.7

Source: World Bank, World Development Indicators, series NE.TRD.GNFS.ZS (World, US, and China values).

Trade as a share of world GDP roughly doubled between 1970 and 2008, stalled after the 2008 financial crisis, and peaked at 62.2% in 2022. The United States is less trade-dependent than most large economies: trade equaled 19.8% of its GDP in 1990 and 25.0% in 2024, compared with 24.2% and 38.0% for China over the same period.

Two forces made the rise possible. One was policy: successive rounds of tariff cuts under the General Agreement on Tariffs and Trade, and from 1995 the World Trade Organization. The other was cheaper shipping. On April 26, 1956, the converted tanker Ideal X sailed from Port Newark to Houston with 58 containers on deck. Accounts based on the shipping entrepreneur Malcom McLean's own calculations put loading costs at about $5.83 per ton for loose cargo against under $0.16 per ton for containers. Those figures are reconstructions and not audited measurements, but the direction is not in dispute: standard boxes made it far cheaper to move goods between ships, trains, and trucks.

Our look at why the world is deglobalising covers what has changed since the late-2000s peak.

What the theory gets right

The empirical record supports the core claim. Economies that opened to trade, including South Korea, Taiwan, and later China and India, grew faster than comparable economies that stayed closed, though trade was one factor among many. The textbook model also predicts that opening raises total income, because consumers get cheaper goods and a wider variety, and firms get access to bigger markets and better inputs.

Economists have refined the original idea over two centuries. Later models, grouped under the Heckscher-Ohlin approach, explain comparative advantage by differences in resources: a country with plentiful land exports farm goods, and one with plentiful skilled labor exports goods that need skills. Other models add scale economies and product variety, which explain why rich countries trade heavily with one another in similar goods, such as cars exchanged for cars. The Ricardian logic survives in all of them. Different countries do different things at different costs, and exchange lets each use its resources better.

What the theory misses: winners and losers

Ricardo's model predicts a bigger pie. It does not say who eats the slices.

The Stolper-Samuelson result. In a 1941 paper, "Protection and Real Wages," Wolfgang Stolper and Paul Samuelson showed that, under standard assumptions, a rise in the relative price of a good raises the real return to the factor used most intensively in making it and lowers the return to the other factor. Applied to trade, the factor that is scarce in a country tends to lose when the country opens up. For a country like the United States, which is rich in capital and skilled labor relative to its low-wage trading partners, the model predicts a loss for workers without those skills. Real economies have more than two factors, and later economists have found that attempts to generalize the theorem to many goods and factors give qualified results, so it is a warning about direction, not a forecast of size.

The China shock. David Autor, David Dorn, and Gordon Hanson tested the idea in a 2013 American Economic Review paper that studied how rising Chinese import competition between 1990 and 2007 affected local US labor markets. They found that markets with more exposure saw higher unemployment, lower labor force participation, and lower wages. In their conservative estimates, the trade shock accounted for about a quarter of the aggregate decline in US manufacturing employment over the period, and government transfer payments for unemployment, disability, retirement, and health care rose in the most exposed places.

The key finding is not that trade destroys jobs overall. It is that the losses were concentrated in specific communities, while gains arrived as slightly lower prices spread across hundreds of millions of shoppers. Concentrated losses are visible and personal; diffuse gains are invisible. Politics follows that pattern.

Advantage changes. Comparative advantage is measured at a point in time, but advantage can be built. South Korea's exports shifted from low-skill goods in the 1960s toward ships and semiconductors, and government industrial policy played a role in that shift. Whether advantage is something countries discover or something they create is an open argument. The economists James Brander and Barbara Spencer have argued that, in markets with few competitors, strategic export subsidies and import restrictions can raise a country's welfare, while critics of the classic theory argue that it depends on assumptions such as constant returns to scale, which often do not hold.

Infant industries. The argument that a young industry may need temporary protection until it reaches competitive scale is theoretically sound and has historical support. The practical problem is political: governments are poor at picking winners, and protected industries tend to lobby for protection to continue.

Understanding trade deficits

A trade deficit means a country imports more goods and services than it exports. The Bureau of Economic Analysis (BEA) reported that the US goods and services deficit was $918.4 billion in 2024, up $133.5 billion from $784.9 billion in 2023, or 3.1% of GDP in 2024 against 2.8% in 2023. The goods deficit was $1,211.7 billion, partly offset by a $293.3 billion surplus in services.

Political discussion often treats a deficit like a company's loss. The accounting does not support that. A country's balance of payments must add up: if it buys more from the world than it sells, the difference must be financed by foreigners acquiring claims on that country, such as bonds, stocks, real estate, and business ownership. The BEA's broader current account, which adds income flows to the trade balance, ran a deficit of roughly $1.1 trillion in 2024, and the same logic applies to it. The deficit is, from the other side, an inflow of foreign investment.

Another way to see it is through saving and investment. A country that invests more than it saves at home must borrow the difference from abroad, and that borrowing shows up as a current account deficit. The US has been a magnet for foreign capital partly because the dollar is the main reserve currency, a point covered in our article on why the US dollar is the world's reserve currency.

A deficit can reflect strength or weakness.

• Strength: strong demand pulls in imports, and a fast-growing economy attracts investment.
• Weakness: uncompetitive industries, or an exchange rate held artificially low or high, can push the balance in unhelpful directions. Our explainer on why currency exchange rates change shows how rates respond to these pressures.

The deficit alone cannot tell you which. Economists look at what the borrowed money is used for, how large the foreign debt is relative to income, and whether the pattern is stable.

Bilateral deficits are a poor guide

Headlines often focus on the deficit with a single country. The table shows the largest US goods deficits by partner in 2024, on the Census basis used by the BEA.

Partner	US goods deficit, 2024 (US$ billion)
China	295.4
European Union	235.6
Mexico	171.8
Vietnam	123.5
Ireland	86.7
Germany	84.8
Taiwan	73.9
Japan	68.5
South Korea	66.0
Canada	63.3

Source: BEA, U.S. International Trade in Goods and Services, December and Annual 2024.

Bilateral figures are misleading because of how goods are made. A phone assembled in one country may contain parts from several others, yet trade statistics attribute the full value of the finished product to the country where it was made. Shifting final assembly from one country to another can move the bilateral deficit without changing the total. Only the overall balance, which reflects saving and investment, tells you about the country's position.

Tariffs: what they actually do

A tariff is a tax on imported goods. A 25% tariff on steel means steel entering the country costs 25% more at the border. The effects spread out from there.

Domestic producers benefit. Foreign competitors are more expensive, so domestic steelmakers can raise prices or gain market share.

Downstream industries pay. Firms that use steel as an input, such as carmakers, appliance makers, and builders, face higher costs, which can reduce their output and employment.

Consumers pay. Prices rise for imported goods and for domestic goods that no longer face the same competition.

Partners may retaliate. Trading partners often respond with tariffs on the first country's exports, harming its exporters.

The best-known evidence comes from the 2018-2019 US tariffs. Mary Amiti, Stephen Redding, and David Weinstein, in the Journal of Economic Perspectives in 2019, found that after the tariffs, US prices rose for both intermediate inputs and final goods, supply chains shifted, fewer imported varieties were available, and tariffs were fully passed through to domestic prices of imported goods. In their words, the full incidence fell on domestic consumers and importers. They estimated that US aggregate real income fell by about $1.4 billion per month by the end of 2018.

A Federal Reserve staff paper by Aaron Flaaen and Justin Pierce looked at manufacturing directly. It found that industries more exposed to the tariff increases lost employment relative to less-exposed industries. The paper identified three channels: protection from imports (a gain), higher input costs (a loss), and retaliation abroad (a loss), and concluded that the gain was outweighed by the two losses. Producer prices also rose, mainly because of input costs. The paper is a preliminary Fed staff working paper, so the exact estimates deserve caution, but its direction matches the Amiti study.

There are cases for tariffs, and serious economists make them.

1. Leverage in negotiations. A credible threat can persuade a partner to lower its own barriers, though the threat is costly if it has to be carried out.
2. National security. Protecting industries judged essential to defense can justify paying more.
3. Infant industries. Temporary protection can help a new industry reach scale, if the government can end it.
4. Revenue. Tariffs raised much of government revenue before income taxes existed.

Most economists accept the first three in specific, limited cases and doubt that tariffs help broadly. Our explainer on game theory and the Nash equilibrium shows why tariff wars are hard to win: each side's best individual move, raising its own tariff, leaves both worse off if both do it.

The Smoot-Hawley episode

The standard cautionary tale is the Smoot-Hawley Tariff Act of 1930. Douglas Irwin's 2011 book Peddling Protectionism, a history of the law, argues that it began as an effort to win farm votes and expanded through logrolling into higher duties across many goods. After it was signed, US imports fell sharply, other countries retaliated, and US exports declined. Irwin's conclusion is more careful than the popular story. He argues that the tariff did not cause the Great Depression, but that it contributed to the decline in world trade and led to discrimination against US exports that lasted for decades. Other forces, such as the collapse in demand and Britain's departure from the gold standard, also played a role in the fall in trade.

Supply chains: efficiency and fragility

The growth of trade after the Second World War, together with containerization and cheap communication, let firms spread production across many countries. A product's components may be made where each costs least and assembled somewhere else. "Made in" a country tells you where the last step happened, not where most of the value was added.

The benefits are lower costs, scale economies, and competition among suppliers. The costs show up when something breaks. During the COVID-19 pandemic, shortages of a few critical components disrupted production of finished goods in many industries, and the episode made firms and governments rethink how much slack to hold. Just-in-time production keeps inventories low, which saves money until a single supplier fails. Our article on how the global financial system works covers the financial side of the same interdependence.

One response has been to pay for domestic capacity in strategic sectors. The CHIPS and Science Act, signed on August 9, 2022, appropriated $52.7 billion for domestic semiconductor research and manufacturing, including about $39 billion in manufacturing subsidies and about $13 billion for research and workforce training, plus a 25% investment tax credit for chip-manufacturing equipment. Subsidized companies face restrictions on expanding advanced chip production in China. Economists generally accept national security as a legitimate reason to pay such a premium, while pointing out that the premium is real: chips made in the United States may cost more to produce than chips made elsewhere. Our explainer on why semiconductors are so hard to make shows why capacity is so costly to build and so difficult to replace.

The trade-off between efficiency and resilience has no free answer. Redundancy costs money, and a country that insists on making everything at home gives up the gains Ricardo described.

The rules: the WTO and its dispute system

The World Trade Organization, founded in 1995, provides a forum for negotiating tariff limits and a system for settling disputes between members. Members agree to bound tariff levels and to treat other members' goods without discrimination.

The dispute system has been weakened. The WTO's Appellate Body, a standing seven-member court created in 1995 to hear appeals from panel rulings, stopped functioning in December 2019 because it no longer had the three judges needed to hear a case. The United States refused to approve new appointments, a blockade that began with Obama administration vetoes in 2016, became a standing refusal under the first Trump administration, and continued under the Biden administration. As a result, a party that loses a panel ruling can "appeal into the void," and the ruling never becomes final. According to the WTO's count as reported on Wikipedia, 32 panel rulings had been appealed into the void as of April 2025.

In March 2020, the European Union and 15 other members created the Multi-Party Interim Appeal Arbitration Arrangement, a voluntary alternative. As of April 2026 it had 59 of the WTO's 166 members and had issued three rulings. The practical effect is that trade rules still exist but are harder to enforce, which makes retaliation more likely to replace adjudication.

How to read a trade claim

• Ask who gains and who loses. A statement about "the economy" hides distributional effects, and most trade arguments are really about distribution.
• Separate total from bilateral. The deficit with one country says little about the nation's overall position.
• Check who pays the tariff. The evidence from 2018-2019 is that domestic buyers paid, at least in the short run.
• Look at time horizons. Gains from trade are spread out and arrive gradually, while job losses arrive suddenly and locally.
• Distinguish theory from policy. Comparative advantage shows that trade can raise total output. It does not prove that every specific trade agreement does, and it does not rule out transition support for people who lose.

FAQ

Q: What is comparative advantage in simple terms? A: It is the ability to make a good at a lower opportunity cost than another producer, meaning you give up less of other goods to make it. A country should specialize in what it gives up least to make and trade for the rest, even if it is more productive than its partner at everything. Ricardo's 1817 example shows both countries ending up with more.

Q: Does trade really create more jobs than it destroys? A: Economists generally find that trade changes the mix of jobs more than the total number, and the effects are uneven. Autor, Dorn, and Hanson found that Chinese import competition between 1990 and 2007 explained about a quarter of the decline in US manufacturing employment, concentrated in exposed local labor markets. Gains from lower prices were wider but less visible, and the transition costs for affected workers were large.

Q: Is a trade deficit bad? A: Not by itself. A deficit means a country is financing imports with foreign capital, and it can reflect strong demand or an attractive place to invest. The US goods and services deficit was $918.4 billion in 2024, or 3.1% of GDP. What matters is what the borrowing funds and whether the pattern is sustainable.

Q: Who pays for tariffs? A: Mostly the importing country's buyers. A 2019 study by Amiti, Redding, and Weinstein found complete pass-through of the 2018 US tariffs to domestic import prices. A Federal Reserve paper found that manufacturing industries more exposed to the tariffs lost employment relative to others, because higher input costs and retaliation outweighed the protection.

Q: Why did the WTO's dispute system stop working? A: The Appellate Body, which hears appeals of panel rulings, lost its quorum in December 2019 because the United States blocked the appointment of new members. Since then, appealed rulings can be left in limbo. The European Union and 15 other members created an interim arbitration arrangement in March 2020, which had 59 members as of April 2026.

Q: Are trade wars ever won? A: Rarely by ordinary measures. Retaliation reduces trade, raises prices, and creates uncertainty, as the 2018-2019 evidence and the 1930 Smoot-Hawley episode both suggest. Tariff threats have sometimes produced concessions, but the cases where one side clearly came out ahead are hard to find, and each round of retaliation tends to raise costs for both.

The takeaway

Ricardo's arithmetic is correct and surprisingly robust: when countries use their labor and capital where they sacrifice least, the world makes more of everything, and every partner can end up with more than it had alone. What the arithmetic leaves out is who inside each country pays the adjustment. The China-shock research shows that trade losses can concentrate in specific towns for years, while the gains are spread too thinly to notice, which is why trade politics is hard. Tariffs, meanwhile, behave less like a tax on foreigners than like a tax at home, and supply-chain policy is a decision about how much efficiency a country will give up for security. The useful question is not whether trade is good or bad, but who gains, who pays, and what a country does about the people who lose.

Sources

• David Ricardo, On the Principles of Political Economy and Taxation (1817), as described in Wikipedia: Comparative advantage
• Autor, Dorn, and Hanson, The China Syndrome: Local Labor Market Effects of Import Competition in the United States, American Economic Review 103(6), 2013
• Amiti, Redding, and Weinstein, The Impact of the 2018 Tariffs on Prices and Welfare, Journal of Economic Perspectives 33(4), 2019
• Flaaen and Pierce, Disentangling the Effects of the 2018-2019 Tariffs on a Globally Connected U.S. Manufacturing Sector, Federal Reserve FEDS 2019-086
• BEA, U.S. International Trade in Goods and Services, December and Annual 2024
• World Bank, trade as a share of GDP (NE.TRD.GNFS.ZS)
• Douglas Irwin, Peddling Protectionism: Smoot-Hawley and the Great Depression, Princeton University Press, 2011
• Wikipedia: Appellate Body and CHIPS and Science Act

---

Read next:

• Countries with the Biggest Trade Surpluses (2025 Ranked)
• How Inflation Works — and Why It's So Hard to Stop
• How Inflation Works: The Economics Explained
• Countries Attracting the Most Foreign Direct Investment (2025)