Conventional estimates of the economic return to agricultural research use market prices for the values of products and inputs; on this basis, economic rates of return are typically well above the cost of capital, suggesting that more investment in research would be socially desirable. But these estimates may be incorrect if, as is often the case, market prices are distorted by market failures or government policies and hence do not reflect social values. This paper presents a simple, partial-equilibrium methodology with which to improve the measurement of social returns to research by taking account of multiple distortions in the market prices of products, inputs and foreign exchange. The method also takes account of variation in domestic and world prices, making a product tradable in some years and nontradable in others. The method is applied to the case of Hageen-Dura 1 (HD-1), Sub-Saharan Africa's first commercially successful hybrid sorghum. HD-1 was released in the Sudan in 1983. From the start of research in 1979 to 1992, the HD-1 breeding program had an estimated IRR of 97% when all major policies in the sorghum market, the fertilizer market, and the exchange rate are taken into account. The high rate of return to HD-1 research was due to the program's low cost and rapid payoff, pointing to the potential value of small adaptive research programs, taking full advantage of foreign technology and genetic material to produce locally-appropriate crosses in a short period of time. Even in the highly distorted economies of Africa, such programs can yield very high payoffs.