Low investment in profitable technologies contributes to persistent poverty. Many farmers in developing countries invest too little in fertilizer despite evidence that fertilizer is profitable. This field experiment investigates a two-part explanation: (1) farmers are reluctant to invest without farm-specific evidence of profitability, possibly because of heterogeneous returns, and (2) information is not sufficient to increase investment because of financing constraints. Farmers in one arm of the experiment receive fertilizer recommendations based on tests of their soils, others receive recommendations paired with an input subsidy, and others receive only the input subsidy. Only farmers who receive recommendations and the subsidy increase fertilizer application and yields relative to the control group. The financing constraint may explain limited response to heterogeneous fertilizer recommendations. The approximate net benefit of increased yields, accounting for the full cost of inputs and soil tests, is equivalent to average wages for seven days of work.

Read the paper in the Journal of Development Economics.