Biodiversity Net Gain (BNG) in England is a recently legislated mechanism for ensuring that the biodiversity impacts of new developments are appropriately mitigated. Despite the assumption that some elements, such as the preference for locally implemented offsetting of impacts, should provide benefits for people, the policy's focus is on ecological outcomes. The social feasibility of BNG guidelines has not been properly tested, nor has their generalisability across people and places. Understanding the preferences of local project-affected people for Biodiversity Net Gain and incorporating this into both policy and project-level decision-making is a critical step for managing trade-offs ex-ante, thereby maximising the likelihood that BNG projects benefit people’s wellbeing. Using a choice experiment of hypothetical BNG projects in the context of housing development, we examine the trade-offs between the features of the BNG project: distance from home; biodiversity level (species richness); off-site vs on-site biodiversity provision; public access to the offset site; and a non-biodiversity feature (provision of affordable housing). We found that public access and species richness were proportionally more important than proximity and the percentage provision of affordable housing. These preferences were of course, heterogeneous and determined by sociopsychological variables, e.g., captured in the notions of "attachment to place", connectedness to nature, socio-economic variables and rural versus urban location. The preferences expressed identify a range of BNG approaches that respect peoples' preferences and trade-offs, noting that acceptance depends to a great degree on outcomes that are either not an explicit priority (i.e., species richness) or are disincentivised (i.e., public access) by current BNG policy. For BNG to be publicly acceptable and socially sustainable, the study concludes that policy and practice must be flexible enough to incorporate place-specific preferences, especially relating to aspects of access to nature, localised notions of biodiversity, and broader cultural and aesthetic consequences of the development. • Choice experiments examine local people’s acceptance of biodiversity offsetting. • Local people have clear preferences for Biodiversity Net Gain compensation outcomes. • Public accessibility and species richness are the most influential offset outcomes. • Preferences varied according to age, income, geography, and place attachment. • Acceptance depended on outcomes not currently prioritised by net gain policy.

Conservation Agriculture (CA) can promote sustainable smallholder farming by improving yields, enhancing climate resilience and reducing the adverse environmental impact of production. Although the economic dimensions of CA have been extensively studied, its contributions to social and environmental outcomes, such as labour demand and soil health, remain largely unexplored. This study investigates the implementation of CA practices on cultivated plots and their contributions to labour use, productivity, and soil erosion. Using a nationally representative household plot-panel dataset of 2117 households and 10,075 cultivated plots in Zimbabwe, multinomial endogenous switching regression models, incorporating an ordered probit selection rule to address selection bias, were employed to analyse the uptake and associations of varying levels of CA adoption on labour use, productivity, and soil erosion. CA adoption packages (compared to non-adoption) are associated with improved cereal productivity under normal rainfall but provide limited gains for legumes and under flooding. CA packages characterized by increased minimum tillage and herbicide use are linked to labour savings and a reduced family labour burden, yet also to a greater dependence on external ( hired/exchange ) labour. Minimal CA adoption (using only one practice) is linked to higher labour use, lower productivity, and increased soil erosion compared to non-adoption. CA adoption on flood-exposed plots is correlated with increased soil erosion. Exposure to climate shocks, poor soil fertility, and better resource endowments are associated with higher CA adoption, while insecure land tenure and limited market access are linked to lower adoption. Tailoring CA interventions to local conditions and structural realities is essential for achieving overall sustainability. • We evaluate CA practices on cultivated plots and their effects on labour, productivity, and soil erosion. • CA package adoption improves cereal productivity with moderate rainfall but has limited benefits for legumes. • CA packages with minimum tillage and herbicides reduce family labour but rely more on hired/exchange labour. • CA’s ability to limit soil erosion is limited, especially under extreme rain and low CA adoption. • Climate shocks and resource availability boost CA uptake, while insecure land tenure and poor market access hinder it.