The Larger World ArchiveClaims · Sources · Evidence
States, secrets and documentsLive debateDevelopment economics · 1950s – present

The Resource Curse

Saudi Arabia and Nigeria both sit on enormous oil reserves. One has a GDP per head around twenty times the other. The difference is not geology.

The crude distillation column at the Dangote refinery in Nigeria. Oil wealth in a state wi
The crude distillation column at the Dangote refinery in Nigeria. Oil wealth in a state with weak checks on executive power is the classic case the resource-curse literature describes. Image: FrankvEck · CC BY-SA 4.0 — Wikimedia Commons.

The pattern

Sachs and Warner's 1995 study found that countries with a high ratio of natural resource exports to GDP grew more slowly than resource-poor countries over the preceding two decades, controlling for other factors. The finding launched a research programme and a policy vocabulary.

The proposed mechanisms are economic and political. Dutch disease: a resource boom raises the exchange rate and pulls labour into the export sector, making manufacturing and agriculture uncompetitive. Rentier politics: when the state's revenue comes from resource rents rather than from taxing citizens, the government has less reason to be accountable, and capturing the state becomes the fastest route to wealth.

The statistical association extends to conflict: oil and mineral dependence correlates with civil war onset, particularly where resources are lootable and the state is weak.

The counter-evidence

Later work questioned the strength of the growth result. Using commodity price shocks as an instrument, some studies find that a resource boom raises incomes in the short run without the predicted long-run drag. Others find the curse holds for some minerals and not others — point resources like oil and diamonds behave differently from diffuse ones like alluvial gold.

There are clear exceptions: Norway, Botswana, Chile and Australia are resource-rich, growing and democratic. Botswana is the standard counter-example — diamonds, sound institutions, and decades of growth — and it is usually explained by institutions predating the diamonds and by a renegotiated deal with De Beers.

The institutional reading, associated with Acemoglu and Robinson, reverses the causality: it is not resources that determine institutions but the interaction of the two, and the same commodity can be a blessing or a curse depending on who controls the revenue and how it is taxed.

What follows for policy

Supported: dependence on a narrow resource base makes a state vulnerable to price shocks and creates strong incentives for corruption. Transparency in payments — the EITI, the Extractive Industries Transparency Initiative — exists because of this evidence.

Not supported: the deterministic claim that resources make a country poor. Geology is not destiny; the variation within resource-rich countries is larger than the average effect.

The honest summary: the curse is real as a risk, conditional on institutions and commodity type, and strongest for oil and for states with weak checks on executive power.

The archive labels this live debate because the correlation is documented and the causal story has been repeatedly revised.

Why it is in the archive

Because 'they are poor because they have oil' and 'the resource curse is a myth' are both widespread, and the evidence supports neither as stated.

Related: geographic determinism, the great stagnation.

Sources for this page

  • Wikipedia, “Resource curse” — theory, evidence, criticismen.wikipedia.org
  • Sachs & Warner (1995), NBER w5398 — natural resource abundance and economic growthwww.nber.org
  • Mehlum, Moene & Torvik (2006), Economic Journal — institutions and the resource cursedoi.org
  • Extractive Industries Transparency Initiative — the disclosure standardeiti.org