Where redistribution and anti-poverty policies consist of cash transfers allocated according to some pre-specified rules, evaluating their impact on the distribution of living standards and poverty might seem straightforward. It seems sufficient to apply the transfer rules to some representative sample of households. This is the essence of 'incidence analysis' and micro-simulation techniques used in many countries. In practice, however, things are not so easy. There are various reasons for this: a) cash transfers are likely to modify behavior, which in turn can generate economy-wide changes through general equilibrium effects; b) in most developing countries, transfers are made only indirectly, through public spending or indirect taxation, with allocation rules which are often far from transparent and may themselves depend on behavior; and c) implementation may be partial or distorted. More fundamentally, poverty reduction policies often go through both macro-economic and structural instruments aimed at enhancing economic activity and growth The actual change in individuals