摘要：We investigate the effects of previous fertilizer subsidy program on local grain prices in Nigeria. The program has been considered ineffective in targeting and stimulating demand for fertilizer, with potentially rampant leakages. If the program has reduced food price, however, it can still be partly justified regardless of targeting efficiency. We exploit the panel structure of Living Standard Measurement Survey – Integrated Survey on Agriculture (LSMS) collected in 2010 post-planting season and 2011 post-harvesting season. Our methods use Euclidian distance between each district and state governors’ origin district in each state to identify fertilizer subsidy distribution. We also use proxy variable that accounts for both direct subsidy provision and indirect leakage effects to measure the effective size of subsidy. Fertilizer subsidy generally had no effect on maize and sorghum price. In northern Nigeria, fertilizer subsidy might have lowered district level price of local rice, but only to a limited extent. Low market orientation of many subsidy recipients, crowding out of commercial fertilizer, and political influence in subsidy allocations may explain such low impact. We also discuss how our methods minimize potential biases due to errors-in-variable and sample selection.
