Spatial patterns of large-scale land transactions and their potential socio-environmental outcomes in Cambodia, Ethiopia, Liberia, and Peru

Recent large-scale land transactions, often framed as 'land grabbing,' are historically unprecedented. Millions of hectares of land have changed hands for agriculture-driven development over the past decade, and their implementation generates substantial risk of land degradation. This paper investigates land transaction patterns and evaluate their potential socio-environmental impacts in Cambodia, Ethiopia, Liberia, and Peru. We undertake a novel spatially explicit approach to quantify land transactions and conduct scenario-based analyses to explore their implementation consequences on people, land, and carbon emission. Our results demonstrate that existing global datasets on land transactions substantially underestimate their incidence but can either exaggerate or underreport transacted areas. Although confirming that land transactions are more likely to occur in sparsely populated, poorer, and more forested areas, our scenario-based analyses reveal that if fully implemented for agricultural development, land transactions in the four countries will affect more than one million people, yield over 2 Gt of carbon emissions, and disrupt vast swathes of forests. Our findings refute the 'empty land' discourse in government policy and highlight the consequences of land degradation that can occur at an unexpected scale in the 'global land rush.' Future policymaking needs to anticipate the risk of land degradation in terms of deforestation and carbon emission while pursuing agriculture-driven development through land transactions.