Malawi has attracted global attention following enormous gains in crop yields following the introduction of a fertilizer subsidy in 2005, but the agronomic and political stability of this strategy are increasingly being called into question as Malawi faces repeated weather-related crop losses and a similar program is discontinued in Zambia due to the financial strain on government funds. Agroforestry offers an alternative method to improve crop yields and resilience to climate and weather at a relatively minimal cost, but adoption has been slow, especially in comparison to farmers’ response to fertilizer subsidies.  We consider and reject hypothesis that this is because agroforestry systems do not actually increase yields outside of experimental or researcher-lead settings.  Data were collected during a household survey of 390 farms in Malawi to assess the impact of the tree species Faidherbia albida on maize yield. Regression results indicate an increase of 12% to 14% (169-201 kg) per hectare, much less than in experimental settings but greater than or equivalent to other crop management strategies assessed in the same communities, including fertilizer purchased under the subsidy. We then combine regression results with data collected on the financial and labor demands of fertilizer and agroforestry and farmer preferences to better elucidate how resource constraints affect farmers’ choices between alternative crop management strategies. Agroforestry implementation is expected to increase on-farm labor demand by 11% to 14% at establishment and 1% in maintenance years but financial inputs are minimal, making the system a superior alternative in land or credit constrained areas, but difficult in land constrained communities. A ranking exercise identified system flexibility and compatibility with existing systems as the most important decision criteria in agroforestry adoption. Efforts to expand the use of the system must focus on these two areas rather than the existing focus on economic profitability.