Poverty is still a major problem in Pakistan. Worse, the excellent progress made in poverty reduction in the 1970s and 1980s has been reversed in the 1990s. That is the bad news. The good news is that Pakistan is unusually well placed to return to rapid reduction in poverty. We have long known that agricultural growth is closely related to poverty reduction. Recent studies by Peter Timmer and by Martin Ravallion and their colleagues provide massive statistical evidence of this relationship. Rural growth and agricultural growth have a major effect on poverty reduction; urban growth and manufacturing growth do not. At first glance that is strange because farmers are not the poorest rural people, and the direct benefits from agricultural growth are distributed roughly proportionately to size of landholding. The poor in rural areas are heavily concentrated in the rural non-farm sector. They produce non-tradable goods and services. That is, local demand is essential to their growth. It is rising agricultural incomes that provide that growth in local demand. Thus, agriculture’s massive impact on poverty is indirect, working through expenditures on the rural non-farm sector. The bulk of those expenditures are for consumption goods.