This paper revisits the long-standing issue of the incidence of taxes in developing countries. Its central theme is that despite many decades of studies, tax incidence analyses for developing countries continue to be based upon the same shifting assumptions used in developed country studies, despite some obvious pitfalls. Taxes are assumed to be shifted forward to consumers, or backwards onto factor incomes, as has been the case for developed country tax incidence work from Bowley and Stamp to Peclunan and Okner. Developing countries typically have a much different non-tax policy and regulatory environment from developed countries, with higher protection, rationed foreign exchange, price controls, black markets, credit rationing and many other features. The paper argues that all these features can greatly complicate and even obscure the incidence effects of taxes in developing countries. For several taxes, taking such features into account can reverse signs and/or substantially revise estimates of incidence effects from conventional thinking and by substantial orders of magnitude. A final section sets out some implications for country lending programs, both by type of country and level of development, and comments on how the extent to which non-tax policy reform has already been implemented affects the significance of the points raised here.