This paper uses a reasonable model and recent cross-country data to study empirically the effects of income and equality on poverty. Three main points are noted. First, the estimates show highly significant roles of income and equality in poverty reduction, and the effects of increased income and lower inequality are both substantial. Second, the elasticity of poverty with respect to inequality is substantially larger than that relative to income. Third, the estimates suggest a 'growth elasticity' of poverty that is much smaller than the values used in almost every study. Therefore, most of the well-known and influential recent research seems to have overstated the impact of income growth on poverty alleviation by de-emphasising the role of inequality, to which poverty is highly responsive, and by using an income (growth) elasticity of poverty that is much larger than what seems reasonable. Copyright © 2007 John Wiley & Sons, Ltd.