Abstract：
The efficient use of nitrogen fertilisers is a global priority to optimise the economic and environmental outcomes of farming. This paper is the first to consider pollution in the form of nitrous oxide emissions and excess nitrogen to analyse technical efficiency (TE) in farming. This is done by extending the two-stage double bootstrap Data Envelopment Analysis to explicitly model nitrogen pollutants as undesirable outputs. We compared green TE (when undesirable pollutants are considered) and conventional TE (without pollutants) using a case study of 33 rice-producing provinces in the Philippines. Provinces in Mindanao, Luzon, and Visayas islands experienced improvements in green TE but stagnant conventional TE from 2006 to 2017. Although transplanting rice seedlings (rather than direct sowing of seeds) improved both green and conventional TE, seed quality was also identified as an important factor for green TE but not for conventional TE. Our analysis has implications for sustainable rice production and such analysis can be extended to other crops. To advance the effective green transformation of rice production, future research should analyse farm-level data to understand farmers\' decisions regarding seed quality, crop establishment method and nitrogen fertiliser application to devise comprehensive farm integrated management plans.
摘要：
有效利用氮肥是优化农业经济和环境成果的全球优先事项。本文首次考虑了一氧化二氮排放和过量氮形式的污染，以分析农业中的技术效率（TE）。这是通过扩展两阶段双引导数据包络分析来明确地将氮污染物建模为不期望的输出来完成的。我们使用菲律宾33个水稻生产省的案例研究，比较了绿色TE（考虑到不良污染物时）和常规TE（不含污染物）。棉兰老岛各省，吕宋,从2006年到2017年，维萨亚斯群岛的绿色TE有所改善，但传统TE停滞不前。尽管插播水稻幼苗（而不是直接播种种子）改善了绿色和常规TE，种子质量也被确定为绿色TE的重要因素，而不是常规TE。我们的分析对可持续水稻生产有影响，这种分析可以扩展到其他作物。推进水稻生产的有效绿色转型,未来的研究应该分析农场层面的数据，以了解农民关于种子质量的决定，作物建立方法和氮肥施用，以制定全面的农场综合管理计划。
