9.13 — How Japan, Korea, Taiwan and China Got Rich

In 1960, South Korea's income per person was lower than Ghana's. It had been under Japanese colonial rule until 1945, had lost most of its industry — which was in the north — and had been destroyed by a war that killed around a tenth of its population.

By 2020 it was a member of the rich-country club, with income per person comparable to Japan's and above Spain's, and with global brands in electronics, cars, shipbuilding and steel.

That is roughly a fifteenfold increase in income per person in sixty years. Nothing on that scale had happened before, and then it happened four times in the same region.

This chapter is about how, and it is written to be usable as a comparison, because Chapter 9.14 is India.

The common recipe

Six elements. Every one of the four did all six, in roughly this order, and the order matters.

1. Land reform first

This is the least glamorous element and it may be the most important.

Japan, South Korea and Taiwan all conducted thorough land reforms between 1946 and 1953, in Japan and Korea under American occupation authority, in Taiwan by a Kuomintang government that had just lost China and had no local landlords to protect.

What was done. Ceilings on landholding, compulsory purchase of the excess, and sale to the cultivating tenants on long repayment terms. In Japan roughly a third of cultivated land changed hands. Tenancy rates fell dramatically in all three.

Why it mattered so much, and there are four separate mechanisms.

Productivity rose, because an owner-cultivator invests in the land and a tenant paying half the crop does not.

Rural demand appeared. A farming population with income buys manufactured goods, which gave the first factories a domestic market.

Inequality fell at the start, which meant growth began from a relatively equal base — and East Asia grew fast while remaining among the more equal regions in the developing world, which is unusual.

And the landlord class was destroyed as a political force. This is the decisive point. A landed elite is the standard obstacle to industrialisation everywhere — it opposes land taxes, opposes rural education because educated workers leave, and opposes industrial policy that raises rural wages. All three countries removed it before they industrialised, and in each case it was removed by an outside power or by a party with no local base. That is not a policy other countries can simply choose.

2. Universal primary education, before wealth

Japan had over 90 percent primary enrolment by 1900, when it was still poor. South Korea and Taiwan achieved near-universal primary education in the 1950s, before industrialisation.

The sequence is the point. They educated first and got rich afterwards, not the reverse. Chapter 9.4 identified human capital as the mechanism through which institutions produce growth, and this is where the claim is clearest.

And the emphasis was on breadth before depth. Universal basic literacy and numeracy first; universities later. India did the opposite and Chapter 9.14 counts the cost.

3. Export discipline

This is the element that made industrial policy work here when it fails almost everywhere.

All four protected domestic industries and subsidised them heavily — cheap credit, tariff protection, licensing, cheap foreign exchange, and direct government direction of investment.

And the support was conditional on export performance.

Why that single condition changes everything. A domestic market can be captured by a well-connected firm producing a poor product at a high price. A foreign market cannot. Selling in Germany or the United States requires meeting world quality at world prices, which means the export figures are an objective test of whether the subsidy is producing capability or waste.

And firms that failed the test lost their support. South Korea's government withdrew credit from failing chaebol and let some go under. Firms were given targets and their performance against them determined their access to subsidised finance.

Compare Chapter 6.28's Licence Raj, where protection was permanent, unconditional and evaluated against no external benchmark. The difference is not whether the state intervened. It is whether the intervention had a test attached that the state was willing to enforce.

4. Very high saving and investment

Investment rates of 30 to 40 percent of GDP, sustained for decades. China's exceeded 40 percent for years.

How it was achieved: financial repression — deposit rates held low, so savers earned little and capital was cheap for industry; restrictions on consumer credit and on capital leaving the country, so savings had nowhere to go but domestic banks; and government-directed lending.

And it is a genuine transfer from households to industry. Chinese households in particular subsidised Chinese industry for three decades through low deposit rates, which is why Chinese consumption is such a low share of GDP and why rebalancing is so difficult.

5. Moving up the value chain deliberately

None of them stayed where their comparative advantage was in 1960.

The Korean sequence is the clearest. Wigs and plywood and textiles in the 1960s. Steel, shipbuilding and chemicals in the 1970s. Cars and consumer electronics in the 1980s. Semiconductors and telecommunications in the 1990s. And now displays, batteries and cultural exports.

POSCO is the illustration. South Korea decided in the 1960s to build an integrated steel plant. The World Bank advised against it, on the grounds that Korea had no iron ore, no coking coal and no comparative advantage in steel. It was built anyway, financed partly with Japanese reparations money, and became one of the most efficient steel producers in the world.

This is Chapter 9.12's point about the static nature of comparative advantage, demonstrated.

6. Competent, insulated bureaucracies

Japan's Ministry of International Trade and Industry, Korea's Economic Planning Board, Taiwan's economic ministries.

Staffed by the top graduates, with real authority, insulated from day-to-day political pressure, and — the term used in the literature is embedded autonomy — close enough to business to have accurate information, and independent enough not to be captured by it.

Corruption existed and was not absent, and Korea's presidents have a striking record of prosecution after leaving office. What distinguished the arrangement was that the corruption did not prevent performance targets from being enforced.

The differences

Japan industrialised first, from the Meiji Restoration of 1868, which is the earliest and most remarkable case. Confronted by American warships in 1853 and forced to open (Chapter 7.9), Japan's response was to send missions abroad to study Western institutions systematically and to import wholesale — a German-modelled constitution and army, a British-modelled navy, a French-modelled legal code, and an American-modelled education system. Its slogan was rich country, strong army. It industrialised, defeated Russia in 1905, and then took the militarist path that ends in Chapter 11.8.

Post-war Japan rebuilt under American occupation, with land reform, the keiretsu corporate groupings, MITI's coordination, and lifetime employment in large firms. It became the world's second largest economy by 1968. Its asset bubble burst in 1990 and it has had slow growth since, which is the standard cautionary case about what happens after catch-up ends.

South Korea did it under military dictatorship. Park Chung-hee, who took power in a coup in 1961 and ruled until his assassination in 1979, is the architect, and the record includes both the growth and the repression: political prisoners, torture, censorship, and the suppression of labour. Democracy came in 1987, after mass protests, and only after most of the growth had happened.

Taiwan followed a similar path under Kuomintang one-party rule, with more emphasis on small and medium enterprises than Korea's giant conglomerates. It democratised in the 1990s. TSMC, founded in 1987 as a government-backed venture, now manufactures a majority of the world's advanced semiconductors — which is why Chapter 12.10 treats Taiwan as the most dangerous place on the world map.

China started later and moved faster.

China specifically

Chapter 12.7 covers the politics. Here is the economics.

Deng Xiaoping's reforms from 1978, and their sequencing is the interesting part.

Agriculture first. The household responsibility system dismantled the communes and let households farm assigned land and keep the surplus after meeting a quota. Grain output rose enormously within a few years, and hundreds of millions of people moved out of extreme poverty largely through this single change.

Then special economic zones — Shenzhen and others from 1980 — where foreign investment, foreign management and export production were permitted under different rules from the rest of the country. A controlled experiment in one district, expanded when it worked.

Then township and village enterprises, locally owned rural industry, which absorbed labour and produced consumer goods.

Then a long, deliberate opening to foreign investment, with technology transfer requirements attached, and WTO accession in 2001, which locked in access to world markets.

Deng's stated method — crossing the river by feeling the stones — describes it accurately: pilot, evaluate, expand. No comprehensive blueprint, and a willingness to reverse what failed.

The result: on the standard measures, around 800 million people moved out of extreme poverty, the largest and fastest such reduction in recorded history.

And the costs are substantial. An authoritarian system with no political liberalisation, severe environmental damage, the hukou household registration system that made hundreds of millions of internal migrants second-class residents in the cities they built, a demographic problem created by the one-child policy, enormous debt accumulation particularly in property and local government, and rising inequality.

Can it be copied

Three honest qualifications, because the recipe is not simply available.

The external conditions were unusually favourable. Japan, Korea and Taiwan were American Cold War allies, which brought aid, security guarantees, technology transfer and — crucially — open access to the American market at a time when the United States tolerated large trade deficits for strategic reasons. That geopolitical bargain is not on offer now.

The land reforms were carried out by occupying powers or by parties with no landed base. A democratic government facing an organised landlord class cannot simply choose to do this, and India's experience proves it (Chapter 9.14).

And export-led manufacturing growth may be a narrowing path. Automation is reducing the labour intensity of manufacturing, so the same output supports fewer jobs, and the ladder that carried hundreds of millions out of poverty may have fewer rungs than it did.

What does transfer: educate everyone first; make support conditional on measurable performance; invest in infrastructure ahead of demand; keep the currency competitive; build state capacity; and be willing to let failing firms fail. None of that requires an authoritarian government — Japan, Korea and Taiwan all became democracies and continued growing, which disposes of the claim that development requires dictatorship.

Where this shows up in your life

The products. Samsung, LG, Hyundai, Toyota, Sony, TSMC, Huawei, and the components in every device you own.

And the comparison is the point of the chapter. In 1960 India's income per person was comparable to South Korea's and higher than China's. Today South Korea's is roughly ten times India's and China's is roughly five times.

That gap is not about ability, resources or culture (Chapter 9.4). It is the compound effect of policy choices made over sixty years, and the next chapter goes through them one by one.

What the next page covers

Chapter 9.14 is the honest comparison with India — what India did differently at each of the six steps, why land reform failed, why primary education was neglected, why manufacturing never took off, what India got right that the East Asian states did not, and what the realistic path forward looks like given that the world of 2026 is not the world of 1965.