Skip to content

← Economics

Why Countries Trade

Chapter 20 · The Whole Economy

In this chapter 4 sections
  1. Ricardo’s surprise
  2. Who loses at home
  3. Free trade or protection
  4. The rules Nepal trades by

Nepal grows only a little of the seeds that cooking oil is pressed from. Yet in 2025/26 its biggest export, by far, was soybean oil: Rs. 128.74 billion of it, refined in Nepal and sold to India. The beans’ oil came in crude, mostly from Argentina and Brazil, some 130 billion rupees’ worth, was refined in factories on the plains, and went out again across the southern border. With palm and sunflower oil, edible oils came to Rs. 146 billion, nearly half of everything Nepal sold abroad that year.

That looks like a puzzle about cooking oil, but it is really about rules. India taxes refined oil from most countries at 35.75% as it comes in. Refined oil made in Nepal can enter India without the tax. The trade lives in the gap between those two rules, and will last as long as the gap does. Most trade is not like that. Most of it would go on without any special rule, because countries gain from trading with each other in ways that surprise most people who meet them for the first time. The puzzle is why, and the next one is who loses even so.

Ricardo’s surprise

Adam Smith, whose pin workshop Chapter 9 visited, put the first answer in one homely sentence. “It is the maxim of every prudent master of a family, never to attempt to make at home what it will cost him more to make than to buy. The tailor does not attempt to make his own shoes, but buys them of the shoemaker.” And then: “What is prudence in the conduct of every private family can scarce be folly in that of a great kingdom.” With glass, hotbeds and heated walls, he wrote, Scotland could grow very good grapes and make very good wine from them, at about thirty times the cost of wine as good brought from abroad. It would be mad to try.

That much is easy. A country that makes something with less work than another has an absolute advantage in it. If each sells the other what it is better at, both are better off.

But suppose one country is better at everything. Should the other give up trading, since it has nothing to sell that its partner couldn’t make more cheaply? David Ricardo, whose theory of rent Chapter 12 met, took this up in 1817, in a chapter called “On Foreign Trade”. He chose two countries and two goods. In England, he supposed, making a year’s quantity of cloth took the work of 100 men for a year, and making a quantity of wine took 120. In Portugal the same cloth took 90 men and the same wine 80. Portugal was better at both.

Ricardo’s answer was that they should trade anyway, and that Portugal should buy its cloth from England. “This exchange might even take place,” he wrote, “notwithstanding that the commodity imported by Portugal could be produced there with less labour than in England.” The reason is the one Chapter 3 gave for every choice: the real cost of anything is what you give up to have it. In England, the men who make a load of cloth could instead have made five-sixths of a load of wine. In Portugal, they could have made a load and an eighth. So cloth costs less wine in England than in Portugal, even though it takes more men, and wine costs less cloth in Portugal. Economists later called this comparative advantage, तुलनात्मक लाभ: making something at a lower opportunity cost than someone else. The words aren’t Ricardo’s, but the idea is.

Try it

Cloth and wine

Ricardo’s own numbers of workers for a load. Alone, each country can make 40 loads of each.

EnglandCloth: 100 workers a loadWine: 120 workers a loadMakes 40 of eachNo tradeMakes 88 clothTrades 44 cloth awayMakes about 73 wineNo swap helps both4040ClothWinePortugalCloth: 90 workers a loadWine: 80 workers a loadMakes 40 of eachNo tradeMakes 85 wineTrades 44 wine awayMakes about 76 clothNo swap helps both4040ClothWine40: what each makes of each alone
What does England make?
World cloth
80 loads
World wine
80 loads

Without trade each country makes 40 loads of each. The world has 80 of each.

The gain doesn’t come from anyone working harder. It comes from the same workers doing the job they give up least to do. When England makes only cloth and Portugal only wine, the world has 88 loads of cloth and 85 of wine instead of 80 and 80. Swap 44 loads of cloth for 44 of wine and each country has more of both than the 40 and 40 it made alone. Put the two countries the wrong way round, though, and the world has less of both. Specializing, विशिष्टीकरण, pays only when each country gives up the thing it is relatively worse at.

The price of the swap matters too. England gains from any deal that gets it more than five-sixths of a load of wine for a load of cloth, since that is what making the wine itself would cost; Portugal from any that costs it less than a load and an eighth. Anywhere between, both gain, and where the price settles decides how the gain is shared. Ricardo put the oddest part plainly: “England would give the produce of the labour of 100 men, for the produce of the labour of 80.”

Why not move England’s money and workers to Portugal, where every job takes fewer men? Ricardo granted that it would pay England’s capitalists, but thought owners would rather take lower profits at home, held back by fears for their money abroad and by the “natural disinclination which every man has to quit the country of his birth”, feelings “which I should be sorry to see weakened”. Goods cross borders more easily than people. Chapter 32 tells what happens in Nepal when people go instead.

Out of this one idea come the other reasons for trade that Nepal’s Grade 10 book lists. A small country’s market is small, and selling abroad lets a factory grow big enough to make things cheaply. Trade brings in the raw materials a country lacks, puts people to work at what they do best, and carries new ways of doing things across borders, in the machines and know-how that come with the goods. Trade between countries is अन्तर्राष्ट्रिय व्यापार, international trade; what a country sells to others is its exports, निर्यात, and what it buys from them its imports, आयात.

As ships, planes and phones made it cheaper to move goods and messages, trade has spread round the world, and with it money, firms and ideas. That spreading is called globalization, विश्वव्यापीकरण. Much of it is carried by multinational companies, बहुराष्ट्रिय कम्पनी, firms with factories or offices in many countries, which can make a part in one country, assemble it in another and sell it in a third.

Who loses at home

The widget counts loads of cloth and wine for each country as a whole. Inside a country, the people who make cloth and the people who make wine are different people. When England stops making wine, its winemakers lose their work, and the gains go to others.

In 1941 two economists, Wolfgang Stolper and Paul Samuelson, showed how far this goes. Every good is made from some mix of land, labor and machines (Chapter 9’s factors of production), and the mixes differ: one good needs a lot of workers, another a lot of land. When trade raises the price of a good a country sells, the owners of whatever that good uses most gain, and the owners of what the other good uses most lose. They lose not just compared with the winners but in what their earnings will buy. The owners of what a country has little of, compared with its trading partners, tend to lose when it opens to trade, and to gain from protection. The country as a whole still gains, so the winners could in principle pay the losers in full and still be ahead. Whether they do is not a question economics can answer.

How big are such losses in practice? A study by David Autor and two colleagues, published in 2013, looked at what happened in the United States as imports from China soared between 1990 and 2007. The places whose factories competed most directly with Chinese goods had more unemployment, fewer people in work or looking for it, and lower wages than the places that didn’t. In their main estimate, competition from China explained about a quarter of the fall in American factory jobs over those years. And government payments to those places, for unemployment, disability, retirement and health care, rose sharply. The losses were real, and concentrated in particular towns.

Nepal’s Grade 10 book lists the dangers of open trade from a small country’s side. A home industry can be crushed by a bigger foreign one before it has grown, and jobs lost with it. A country that buys its fuel and medicines from one neighbor depends on that neighbor, and when the border is blocked, नाकाबन्दी, or supplies are cut, आपूर्ति अवरोध, its people suffer, as Nepal’s did in the autumn of 2015 (Chapter 8). And sales abroad rise and fall with markets a small country doesn’t control. The gain comes with costs that fall on particular people.

Free trade or protection

So countries argue about how open to be. A free trade policy, स्वतन्त्र व्यापार नीति, lets goods come and go without taxes or limits; the classical economists who followed Smith and Ricardo argued for it, and Nepal’s books also call it the policy of not interfering, अहस्तक्षेपकारी. Its defenders point to consumers, who get cheaper goods and more choice; to competition, which keeps home firms improving; and to the friendlier ties of countries that depend on each other. Its critics add that the big firms of rich countries can exploit poor countries’ markets and raw materials. A policy of protection, संरक्षण व्यापार नीति, shields some home industries from foreign competition.

Protection has a handful of tools. The commonest is a tariff, भन्सार महसुल, a tax on goods coming in, like India’s 35.75% on refined oil from outside South Asia. It raises the foreign good’s price, so home makers sell more, and the government collects the tax (Chapter 18 shows how much of Nepal’s revenue comes in at the border). A quota, कोटा, limits how much may come in at all, whatever the price. A subsidy pays home makers, so they can sell more cheaply. An embargo bans trade with a country altogether, usually as a punishment. Countries also defend themselves against dumping, a foreign firm selling below its price at home, or below its cost, to take a market. And a tariff often brings retaliation: the country hit taxes the first country’s goods in turn, and both lose sales.

The arguments for protection are old. An infant industry, शिशु उद्योग, might one day make things as cheaply as foreigners do, but not while it is small, and may never get the chance if cheap imports kill it first. Some industries are strategic, needed in a war or a blockade, and a country may want to keep its natural resources from being sold off cheaply. Protection keeps jobs in the protected industry, keeps money at home, and a tax can make people buy less of something harmful, such as cigarettes. Against all this, Nepal’s books set the costs. Consumers pay more. A protected firm with no competitors can behave like the monopoly of Chapter 11, with little reason to improve. Some firms come to depend on protection and never grow up. And the favors go to those with the government’s ear.

The infant industry is the argument that matters most for a poor country, and here economists really do disagree. In 1841 a German economist, Friedrich List, accused Britain of having grown rich behind tariffs and then preaching free trade to others: of “kicking away the ladder” it had climbed. The economist Ha-Joon Chang, who grew up in South Korea, borrowed the phrase and tells Korea’s story to make the same case. Chapter 4 met Korea’s government steering its economy; Chang’s telling is about trade.

In 1961, Chang writes, Korea’s income was $82 a head, less than half of Ghana’s $179. Its main exports were tungsten, fish and wigs made of human hair, and in the 1950s a report of the American aid agency had called it a “bottomless pit”. A general who seized power in 1961 set five-year plans, and from 1973 pushed heavy and chemical industries. In Chang’s account, Korea banned many imports and taxed others heavily, the government owned all the banks and told them whom to lend to, and it controlled foreign money so tightly that breaking the rules could carry the death penalty. What Korea did, he writes, was “to nurture certain new industries, selected by the government in consultation with the private sector, through tariff protection, subsidies and other forms of government support… until they ‘grew up’”. The protection was “not there to shield industries from international competition forever”, only until they could compete in the world market, while simple exports such as garments earned the money to buy the machines. Korea became rich, he argues, through “a clever and pragmatic mixture of market incentives and state direction”, not because, as he puts the view he attacks, it “followed the dictates of the free market”. He doesn’t hide the cost: a dictatorship, slums cleared by the police, and working weeks of more than 50 hours.

Others who have measured find less. Lee Jong-Wha looked at 38 Korean industries from 1963 to 1983. The ones protected most from imports raised their output per worker more slowly, not faster, and cheap loans and tax breaks didn’t go with faster productivity growth either. Less interference in trade went with faster productivity growth. But a newer study by Nathan Lane, published in 2025, looked only at the heavy and chemical industry push of 1973 to 1979 and found that the industries it targeted grew more, gained an edge in world markets, and helped the industries that bought from them, with benefits that lasted after the push ended.

The two studies ask different questions. Lee asks whether protection made industries better at what they did; Lane, whether a few years of targeted help made the chosen industries bigger and stronger sellers abroad. A country could get Lane’s result without Lee’s. Nobody can see the Korea that never protected anything, so the argument goes on. Careful comparisons of industries pushed and left alone in the same years, and of why the same tools worked in Korea and failed elsewhere, are what could narrow it.

The rules Nepal trades by

A three-storey building with a red roof and a sign in Nepali and English reading Birgunj Customs Office, a car parked in front.
Figure 20.1 The customs office at Birgunj’s inland container depot, on the border with India, in 2025. Photo: Saddam19, CC BY-SA 4.0

Since 1995 most of the world’s trade has run on rules agreed at the World Trade Organization, in Geneva, whose 166 members carry on 98% of the world’s trade. Nepal has been one since 23 April 2004. Its first rule is that a favor to one member is a favor to all: a country that cuts its tariff for one partner must cut it for every member, treating each as its most favored nation. There are two big exceptions. Countries may agree to free trade among themselves, and they may give the poorest countries special access.

Nepal uses both. Under the South Asian Free Trade Area, signed in Islamabad on 6 January 2004 and begun on 1 January 2006, South Asia’s countries agreed to cut their tariffs on each other’s goods to between 0 and 5%. The poorer members, Nepal among them, were given longer to do it: down to 30% within two years, then to 0–5% within eight more. Each country keeps a list of sensitive goods it doesn’t cut, reviewed every four years.

What matters most to Nepal, though, is its treaty with India, which takes most of what Nepal sells. The treaty’s 2009 text, signed in Kathmandu on 27 October that year, runs for seven years at a time and renews itself unless either side objects. It has renewed itself, unchanged, in 2016 and again in 2023. Under it, farm and forest goods cross both ways free of tariffs and quotas. And India lets in goods from Nepal’s factories free of tariffs and quotas, without asking the same in return, provided they really are made in Nepal: wholly from Nepali or Indian materials, or worked enough to change their customs heading, with materials from other countries no more than 70% of the price and the last stage done in Nepal.

The text also names four goods India limits: vegetable ghee, वनस्पति घिउ, to 100,000 tonnes a year free of tariff, acrylic yarn and copper goods to 10,000 tonnes each, and zinc oxide to 2,500 tonnes. Those quotas have a story. An earlier treaty, of 1996, had opened India to Nepal’s factory goods with few conditions beyond a certificate of origin, and factories in Nepal began turning out exactly those four goods. India complained of surges of imports with little made in Nepal, and in 2002 the treaty was rewritten with rules on where goods come from, and the quotas.

The refined oil of this chapter’s first paragraph goes to India free of tariff under SAFTA and the treaty, and it is the same story again. India’s refiners say the oil gains only 5–7% of its value in Nepal, far short of the 30% they say SAFTA’s rules require. Nepal’s oil makers say they keep to the rules, and their 27 refineries employ some 15,000 people, directly and through their suppliers. A Nepali economist warned that India could put on tariffs or limits at any time. A trade built on a difference in tariffs, rather than on anything a country makes more cheaply, lasts only as long as the other country’s patience.

All this is the setting for figures Chapter 5 first showed. In 2025/26 Nepal sold Rs. 315 billion of goods abroad and bought Rs. 2,096 billion, so its exports paid for only 15 of every 100 rupees of its imports. Ten years earlier, in 2015/16, a year when trade across the border with India was disrupted for months, it had sold Rs. 70 billion and bought Rs. 774 billion. Petroleum alone cost Rs. 350 billion, more than every export put together. About 82% of what Nepal sold went to India, and 58% of what it bought came from India, with 20% from China. How a country pays for so much more than it sells, mostly with money its workers send home, belongs to Chapter 32 and Chapter 33.

Nepal has had one more kind of special access since 1971: the United Nations counts it among the least developed countries, अति कम विकसित देश, and many richer countries let such countries sell to them free of tariffs and quotas. The UN judges a country on its income per person, its people’s health and schooling, and how exposed its economy is to shocks, and recommends that one passing two of the three tests at two reviews in a row leave the list. Nepal passed on health and schooling and on exposure; its income, $1,300 a head in the 2024 review, sat just under the $1,306 line. Its date to leave was set as 24 November 2026. In May 2026 Nepal asked for more time, pointing, among other worries, to the duty-free, quota-free access it would lose, and in October the General Assembly moved the date to 24 November 2029. Whether leaving is an honor or a loss is a question for Chapter 36.

Ricardo’s idea still holds: two countries gain by each doing what it gives up least to do, even when one is better at everything. But the gain is shared out by prices, by rules written between governments, and by luck: some people gain, some lose their work, and some, like Nepal’s oil refiners, find their living rests on a rule another country can change. The argument about trade is rarely about whether it pays. It is about who gets the gain, who pays the cost, and how long a poor country should shelter what it hopes to grow.

Found a mistake? Please tell us.