This article examines the potential medium‐term causal relationship between changes in Gross Domestic Product (GDP) per capita and poverty in developing countries during the 1970s–1990s. For this purpose, we use panel data model evaluation techniques to test the out‐of‐sample forecasting performance of competing models. We conclude that the evidence supports the hypothesis that increases in GDP per capita cause unidirectional poverty reduction, measured by the $1/day poverty rate, in the period 1970s–1980s. The results are similar when analysing low‐ and middle‐income countries and mid‐high‐ and very high‐inequality countries separately. However, in the period 1980s–1990s, it is only statistically significant for low‐income countries.