Department of Agricultural, Food, and Resource Economics Food Security Group

IDWP 155. Veronique Theriault, Melinda Smale, and Hamza Haider. 2017. Maize Yield Response to Fertilizer under Differing Agro-Ecological Conditions in Burkina Faso 

EXECUTIVE SUMMARY:
Achieving food security in Sub-Saharan Africa depends on raising the productivity of smallholder farmers, and in Burkina Faso, there is no option for enhancing crop productivity other than intensification. The soils in the Sahel and Savanna of West Africa are old, deep and poor in soil organic matter, with low capacity to retain nutrients, while this region is also the most densely populated in the continent. Yet, as in other countries of Sub-Saharan Africa, the national agricultural research system formulated fertilizer recommendations during the 1970s and 1980s, but these did not, and still do not, take differing agro-ecologies into account. The heterogeneity of agro-ecological and soil conditions has led to a diversity of farming systems and cropping patterns. This heterogeneity, along with incomplete input markets, creates highly variable economic incentives for smallholder farmers. Thus, there is a need to understand of farmers’ incentives to use intensification strategies, including fertilizer. We use farm household survey data from Burkina Faso to examine how agro-ecological factors, measured at several scales of analysis (plot, village, and zone), affect the yield response of maize and the profitability of fertilizer use on maize. We focus on maize because it is the only dryland cereal with significant fertilizer use. We estimate the maize yield response function with a two-step procedure that combines correlated random effects with the control function approach in order to handle time-invariant unobserved heterogeneity and the endogeneity of fertilizer use. We then analyze the profitability of fertilizer use by calculating the marginal and average value-cost ratios based on coefficient estimates. We explore the sensitivity of profitability to different assumptions about maize price and fertilizer costs, including fertilizer subsidies and adjustments for transactions costs. Results indicate that agro-ecological factors measured at the scale of plot, village and zone significantly affect maize productivity. The agronomic optimum lies outside the range of observed data, consistent with evidence of long-term nutrient depletion and suggesting that all maize plots would benefit from use of additional fertilizer. The marginal effect of nitrogen is stronger on less fertile soils and in the Sudano-Sahelian zone. At full market prices, fertilizer use is unprofitable, whereas it is profitable with the 50% fertilizer subsidy. However, transaction costs diminish the benefits of the subsidy. In some cases, fertilizer use is not profitable. Findings underscore the need to be cautious when generalizing across regions or formulating policies based on findings from a single region. Policies that consider heterogeneity may be more effective in promoting sustainable input use by making it more profitable. As currently designed, the fertilizer subsidy program promotes maize, which is not well suited to all agroecologies in Burkina Faso. Programs targeted to a single crop may not be desirable, especially in the context of climate change. Although the subsidy enhances profitability (to the extent that it covers transactions costs), there may be more effective ways to make fertilizer more affordable to farmers, such as investments in road infrastructure and removing illicit tax collection.