9.4 — What Actually Makes a Country Rich

Nogales is one town divided by a fence. On the north side, in Arizona, average household income is several times higher than on the south side, in Sonora. Life expectancy is higher. Schooling is longer. Infant mortality is lower. Crime is far lower and public services work.

The people on both sides are the same people. Same ancestry, same culture, same food, same religion, same climate, same soil, same diseases, same latitude. Many of them are related to each other.

The only variable that differs is which set of institutions they live under.

This chapter goes through every serious candidate explanation for why some countries are rich, tests each, and states what the evidence supports. It is the framework for the rest of this Part, and specifically for Chapters 9.13 and 9.14.

First, what "rich" means

GDP per capita — the total value of goods and services produced, divided by the population — is the standard measure and it has known defects that should be stated before it is used.

It counts activity, not welfare. A car crash raises GDP through repairs and hospital treatment. Pollution raises it twice: once producing it and once cleaning it up.

It misses unpaid work — childcare, housework, subsistence farming — which is a systematic downward bias for poorer countries and an erasure of women's work everywhere.

It says nothing about distribution. A country where a few own everything can have a high average.

And comparison across countries requires adjusting for what money buys locally — purchasing power parity — because a haircut costs a fraction as much in Patna as in Zurich.

Better measures exist and are used alongside it: the Human Development Index combining income, life expectancy and schooling; median rather than mean income; and direct measures of health, literacy and access to water and electricity.

The scale of the difference is what needs explaining. Around 1800, the richest country was perhaps four or five times richer per person than the poorest. Today the ratio between the richest and poorest countries is on the order of a hundred to one. That gap opened in two centuries and the whole of development economics is an attempt to explain it.

Candidate 1: natural resources

Fails badly.

Japan, South Korea, Taiwan, Singapore and Switzerland have almost no natural resources and are rich. The Democratic Republic of Congo has extraordinary mineral wealth and is among the poorest countries on Earth.

And Chapter 1.12 showed the resource curse mechanism — a government funded by an oil terminal does not need to bargain with its citizens, so it never builds the institutions that bargaining produces. Norway and Botswana are the counter-examples and they are counter-examples because of what their institutions were like when the money arrived.

Resources are neither necessary nor sufficient.

Candidate 2: geography

Real, and it explains the starting position rather than the outcome.

What is genuinely true. Tropical disease burden is a serious drag — malaria alone is estimated to cost affected economies a meaningful share of growth annually. Landlocked countries face permanently higher trade costs; almost every landlocked developing country is poor, and Switzerland's exception depends on being surrounded by rich neighbours with excellent infrastructure. Chapter 3.7's east-west axis argument and the distribution of domesticable species explain a great deal about 1500.

Where it fails. It cannot explain divergence between neighbours. North and South Korea have identical geography and a roughly twentyfold income difference. Botswana and Zimbabwe, Haiti and the Dominican Republic, and the two Nogales all share a climate. And it cannot explain reversal: some of the richest parts of the world in 1500 — Mughal India, the Aztec and Inca zones — are among the poorer parts today, and some of the poorest — northern Europe, North America — are among the richest.

Geography sets the hand. It does not play it.

Candidate 3: culture

This is the explanation with the worst track record and it deserves a full hearing precisely because it keeps returning.

The pattern of its failures is instructive.

Max Weber's Protestant ethic (Chapter 8.7) explained why Protestant Europe was rich. Then Catholic Bavaria, Italy and Ireland got rich.

Confucianism was, in the early twentieth century, the standard explanation for why China and East Asia could not modernise — its supposed conservatism, veneration of the past, and low status for merchants. Then Japan, South Korea, Taiwan, Singapore, Hong Kong and China grew faster than anywhere in history, and the same commentators cited Confucian values — discipline, education, saving, family — as the explanation.

The same tradition was used to explain both outcomes, which is a clear sign that the explanation is being fitted to the result.

What survives from the cultural argument is narrower and real. Attitudes to education, to saving, to trust between strangers and to corruption do vary and do matter. And they are themselves largely produced by institutions and history rather than being fixed properties of a people. Trust is high where contracts are enforced. Chapter 8.6 showed how the slave trade destroyed trust in the regions it drained, with measurable effects today.

And the strongest reason to be sceptical of culture as a first-order explanation is the speed of the transitions. South Korea went from poorer than Ghana in 1960 to a member of the rich-country club in one lifetime. Culture does not change in thirty years. Institutions and policy can.

Candidate 4: colonies and exploitation

Real for the colonised, insufficient for the colonisers.

Chapter 6.17 established the effect on India and it was substantial: deindustrialisation, an annual fiscal transfer, and public expenditure directed away from human capital.

And as an explanation for European wealth it is weaker than it first appears, for three reasons.

The largest colonial powers are not the richest countries. Spain and Portugal had the largest early empires and were among western Europe's poorer countries by 1900. Switzerland, Sweden and Denmark had negligible or no colonies and are among the richest.

The size of the flows was modest relative to the metropolitan economies. Estimates of colonial profits as a share of British GDP in the eighteenth and nineteenth centuries are generally in the low single digits per year.

And the timing is imperfect. Britain's industrial takeoff begins before the largest colonial revenues arrive.

The strongest version of the argument survives all three points and is worth stating precisely. The Atlantic system did not fund European growth by transferring a large share of GDP. It did something more specific: it supplied cheap raw material inputs — cotton above all — a protected market for manufactures, and a large stimulus to shipping, insurance, banking and port cities, which are exactly the sectors that industrialised. And it did so in exactly the countries — Britain, the Netherlands, France — that industrialised first. Chapter 9.5 weighs it.

Candidate 5: institutions

This is where the evidence points, and the finding is strong enough that its principal researchers received the Nobel Prize in Economics in 2024.

The claim. The rules that govern economic and political life — whether property is secure, whether contracts are enforced, whether the state is constrained, whether markets are open to entrants, and who has political power — are the main determinant of long-run prosperity.

Daron Acemoglu, Simon Johnson and James Robinson distinguish two types.

Inclusive institutions secure property rights, enforce contracts impartially, permit entry into markets and occupations, provide public education and infrastructure, and constrain those in power. They reward invention and effort, because the inventor keeps the reward.

Extractive institutions concentrate power and direct resources to a narrow elite. They can produce growth for a period — the Soviet Union grew rapidly for three decades, and China has grown enormously — and they cannot sustain it, because they cannot permit the creative destruction that continuous innovation requires. New industries destroy old ones, and where the old ones are owned by the people holding power, they are protected.

The evidence is unusually good for a claim of this kind.

The Korean natural experiment. One people, one language, one culture, one history until 1945, then two sets of institutions. North and South Korea's income ratio is on the order of twenty to one. It is the cleanest experiment in the social sciences and the result is not ambiguous.

The colonial reversal of fortune. Among former colonies, the places that were richest and most densely populated in 1500 are on average the poorest today, and the mechanism proposed is that dense wealthy populations invited extractive institutions — encomienda, plantation, mining — while sparsely populated regions with high European settlement got institutions designed for settlers, meaning property rights and representative bodies.

Settler mortality. The most-cited study uses the death rates of European soldiers and bishops in each colony in the nineteenth century as a variable that affects institutions and has no other plausible route to modern income. Where Europeans could survive, they settled and built inclusive institutions; where they died, they built extractive ones and left. The correlation with modern income is strong.

The criticisms are real and should be stated. The settler-mortality data have been challenged on quality. The direction of causation is arguable — rich countries can afford good institutions. And "institutions" is a broad category that can be defined after the fact to fit. The Korean and Nogales comparisons are the strongest evidence and they are hard to explain any other way.

Candidate 6: human capital

Education and health, and this may be the most under-weighted factor in public discussion.

The correlation is very strong. No country has become rich with a poorly educated population, and none has stayed poor with a well-educated one for long.

The East Asian sequence is the clearest evidence (Chapter 9.13): Japan, South Korea and Taiwan all achieved near-universal primary education before they got rich, not after. Japan had over 90 percent primary enrolment by 1900, when it was still a poor country.

Health compounds with it. A child with intestinal parasites or chronic malnutrition learns less from the same schooling; stunting in the first thousand days has permanent cognitive effects (Volume V). India's persistent stunting rate is therefore not only a health failure but an economic one.

And what is measured matters. India achieved near-universal primary enrolment and did not achieve universal learning (Chapter 6.28). Years in school and skills acquired are different variables, and only the second correlates with growth.

Candidate 7: technology and its adoption

Growth over the long run is overwhelmingly about total factor productivity — output rising faster than the inputs of labour and capital, which means getting more from the same.

And the interesting question is not who invents but who adopts. Most countries do not need to invent anything; the technology exists and can be licensed, copied or bought. Catch-up growth is far easier than frontier growth, which is why fast growth is available to poor countries and why no rich country grows at 8 percent.

What stops adoption is usually not the technology. It is the absence of the skills to operate it, the infrastructure to power it, the finance to buy it, or the incentive to bother — and incumbents with political power frequently prefer that it not be adopted.

The synthesis

The honest summary of the current state of knowledge.

Geography and biological endowment set the starting position and explain a great deal about the world of 1500.

Institutions are the main determinant of what happens after that — whether effort and invention are rewarded, whether power is constrained, whether markets are contestable.

Human capital is the mechanism through which institutions produce growth, and it can be built deliberately and relatively quickly.

Technology adoption is the proximate source of growth, and it is gated by the three above.

Culture matters at the margins and is largely endogenous — a product of institutions and history rather than an independent cause.

Colonial extraction damaged the colonised substantially and does not explain the colonisers' wealth.

And contingency is real. Chapter 8.10 argued that Europe's advantage was an accidental configuration. Nothing here is deterministic, which is the point — the whole practical value of the institutional finding is that institutions are the thing a country can change.

Where this shows up in your life

Every argument about Indian economic policy is a version of this chapter — whether to focus on infrastructure, on education, on regulation, on manufacturing incentives, or on institutional reform of courts and land records.

And the Nogales comparison is worth keeping. When someone explains a country's poverty by the character of its people, ask them why the same people prosper across a border. Indians in the Gulf, in Britain and in the United States earn multiples of what they earn at home, doing the same work with the same abilities. The variable is not the person.

What the next page covers

Around 1760, in a corner of northwest Europe, output per person began growing continuously for the first time in human history and has not stopped since. Chapter 9.5 covers the Industrial Revolution — what actually happened technically, why it happened in Britain and not in China or India or France, what coal and high wages had to do with it, what it did to the people who lived through it, and why it is the single most important discontinuity in this entire volume.