International donors invest billions of dollars to conserve ecosystems in low-income nations. An emerging debate rages among academics and practitioners as to the most effective forms of conservation investment. Among the more popular initiatives to achieve this objective is the use of development interventions in the peripheral areas of endangered ecosystems. Such interventions indirectly provide desirable ecosystem services through two mechanisms: (1) by re-directing labor and capital away from activities that degrade ecosystems (e.g., agricultural intensification); and (2) by encouraging commercial activities that supply ecosystem services as joint products (e.g., ecotourism). We contrast this dominant approach with an approach that pays for ecosystem protection directly. Based on theoretical and empirical analyses, we argue that investments aimed at making payments that are conditional on conservation performance are likely to be far more cost-effective than the currently popular indirect approaches to conservation investment. Although direct payment initiatives have imposing institutional requirements, we argue that all conservation initiatives face similar challenges. An empirical example from Africa illustrates the substantial cost savings that can be realized by direct payment initiatives.