This special issue focuses on the poor and their role in forest degradation. The idea that the poor are both agents of forest degradation and victims of forest loss is not new. Neither is the observation that unsustainable rates of forest use by smallholder farmers arises as a result of a complex interplay of incentives, constraints, and institutional forces. For researchers and policy makers concerned with natural resource use in low-income settings, theory is often useful, but rarely adequate to explain behavioural patterns. Unambiguous theoretical findings tend to flow only from substantial sets of simplifying assumptions, and such assumptions, measured against the observed facts of smallholder agriculture, seem largely untenable. Unfortunately much empirical research on the topic suffers from an opposite problem. In settings where, at any point in time, everything matters and nothing is held constant, econometric findings are frequently weak. The combination of heterogeneity of circumstance, homogeneity of market signals, and limitations on smallholder response generally overwhelms the statistical power of small data sets, often collected with different purposes in mind.