Policymakers in Southern Africa have long viewed agricultural cooperatives as a way to reach rural households more easily, but because not all rural households can afford to join cooperatives, this raises questions about membership inclusivity and whether cooperatives advantage some smallholders over others. Agricultural cooperatives can serve as an institutional vehicle for policymakers to deliver direct benefits to smallholder farmers in the form of subsidized agricultural inputs, usually improved seed varieties and fertilizers. They can also serve as platforms for collective action through which smallholders can reduce transportation and transaction costs or disperse the costs of marketing. In Zambia farmers are required to join a cooperative to qualify for seed and fertilizer support through the national Farmer Input Support Program (FISP), which is a structural incentive for cooperative membership and makes it complicated to determine if cooperative membership increases use of inputs and yields on its own. Using the Inverse Probability Weight Regression Adjustment (IPWRA) technique, we find Zambian smallholders participating in cooperatives achieve significantly higher maize yields and use significantly more hybrid-maize seed and fertilizers, even when controlling for FISP eligibility. Furthermore, across all selected productivity and input variables we find greater average treatment effects from cooperative membership than FISP participation. Cooperative membership is also associated with more assets, more land, and higher maize yields, which suggests that the poorest rural households are often not cooperative members. Cooperatives can be a powerful poverty policy tool to deliver direct benefits to small farmers, but not including the poorest households likely widens the rural poverty gap, both in terms of limited direct benefits of input support programs and in terms of farmers missing out on cooperative-specific benefits.

南部非洲的政策制定者长期以来一直将农业合作社视为更容易接触农村家庭的一种方式，但由于并非所有农村家庭都有能力加入合作社，这引发了关于成员包容性以及合作社是否比其他小农户更有优势的问题。农业合作社可以作为政策制定者的制度工具，以补贴农业投入品的形式向小农提供直接利益，通常是改良的种子品种和肥料。它们还可以作为集体行动的平台，小农可以通过这些平台降低运输和交易成本或分散营销成本。在赞比亚，农民必须加入合作社才有资格通过国家农民投入支持计划 (FISP) 获得种子和肥料支持，这是对合作社成员资格的结构性激励，并使确定合作社成员资格是否会增加投入和产量的使用变得复杂。使用逆概率权重回归调整 (IPWRA) 技术，我们发现参与合作社的赞比亚小农实现了显着更高的玉米产量，并使用了更多的杂交玉米种子和肥料，即使在控制 FISP 资格的情况下也是如此。此外，在所有选定的生产力和投入变量中，我们发现合作社成员的平均处理效果比 FISP 参与更大。合作社成员身份还与更多资产、更多土地和更高的玉米产量相关，这表明最贫困的农村家庭通常不是合作社成员。