农业技术采用与森林砍伐：来自一项随机对照试验的证据
Agricultural technology adoption and deforestation: Evidence from a randomized control trial
Journal of Development Economics · 2025 · [{"name": "Jeffrey R. Bloem", "affiliation": ["International Food Policy Research Institute"]}, {"name": "Clark Lundberg", "affiliation": ["San Diego State University"]}]
中文摘要
我们利用尼日利亚的一项随机对照试验，研究采用改良农业投入品对森林砍伐的影响，该试验引入了一种更高效且更环保的氮肥。我们将干预中的调查数据与地球观测数据相结合，提出一种可推广的方法，用于评估集群层面干预对景观层面结果的影响。我们发现处理暴露存在集约化反应的证据，这种反应在不同土地覆盖类型上呈现出显著的异质性：在干预前树木覆盖相对稀疏的土地上，处理暴露增加了森林砍伐；而在森林更密集的地区，干预减少了森林砍伐。我们还发现相应的效应表明，处理暴露提高了农业生产率。我们的结果反映出一种针对改良农业技术的集约化反应，该反应将农业活动从森林转向现有耕地。
Abstract
We study the effect of the adoption of improved agricultural inputs on deforestation using a randomized control trial in Nigeria which introduced a more efficient and environmentally-friendly nitrogen fertilizer. We combine survey data from the intervention with earth observation data to develop a generalizable method for evaluating the effects of cluster-level interventions on landscape-level outcomes. We find evidence of an intensification response to treatment exposure that reflects significant heterogeneity across land cover. On land with relatively sparse pre-intervention tree cover, treatment exposure increased deforestation while in denser forest areas the intervention reduced deforestation. We find corresponding effects showing treatment exposure increases agricultural productivity. Our results reflect an intensification response to improved agricultural technology that redirects agricultural activity away from forests and towards existing cropland.