This paper diagnoses the symptoms of the Dutch disease in a two-sectorstochastic endogenous growth model. A productive, low-skill-intensiveprimary sector causes the currency to appreciate in real terms, thushampering the development of a high-skill-intensive secondary sector andthereby reducing growth. Moreover, the volatility of the primary sectorgenerates real-exchange-rate uncertainty and may thus reduce investment andlearning in the secondary sector and hence also growth. Cross-sectional andpanel regressions based on data for 125 countries in the period 1960–1992confirm a statistically significant inverse relationship between the size ofthe primary sector and economic growth, but not between the volatility ofthe real exchange rate and growth.