Abstract
Investment and market conditions across much of Africa are unfavorable. With scarce empirical evidence, the peculiar choice of investing in land in the African frontiers has been the subject of many conjectures. We address this gap by reconstructing the underlying logics of investment location choices in a Bayesian network with firm and actor level interview and spatial data from 37 transnational agriculture and forestry investments across 121 farm and plantation sites in Mozambique, Zambia, Tanzania, and Ethiopia. Drawing from rent theory, we characterized the appeal of investment locations in terms of resource frontiers (i.e., where land resources are abundant relative to other inputs) and agglomeration economies (i.e., where scale economies external to the individual firm exist). Our sample captured a diversity of investments, made up of investors with varying farming or forestry and regional skill sets and product market control, resulting in varied investment priorities and location choices. Experienced investors focusing on high-value crops invested preferentially in remote subsistence frontiers, seeking land with specific agroecological conditions. In contrast, newcomers - such as the typical speculators - focused more on proximity to infrastructures and markets. We identified four comparative advantages of the African frontiers: large tracts of agroecologically suitable land; agroecologically unique land, potential market access, and location advantages conditional on investor track records. Development efforts founded upon sustainable land-based investments would benefit from targeting investors who can survive unfavorable investment contexts to induce scale economies that extend beyond individual firms.