Lin LHUNG-HAO CHANG2021-07-262021-07-26202120440839https://www.scopus.com/inward/record.uri?eid=2-s2.0-85100232740&doi=10.1108%2fJADEE-07-2020-0157&partnerID=40&md5=a4b923ce88fc2c1a1a617162d8e03399https://scholars.lib.ntu.edu.tw/handle/123456789/573090Purpose: The purpose of this paper is to identify the factors associated with the adoption of agro-processing methods and to estimate their impact on farm income and farm diversification. Design/methodology/approach: Using a large-scale sample of 12,122 special crop farm households drawn from the 2015 Agricultural Census Survey in Taiwan, the semiparametric multivalued treatment effect model was estimated. Findings: The authors found that agro-processing farm households obtain higher farm incomes than non-agro-processing farm households. Among the agro-processing methods, self-processing generates higher farm income than outsourced-processing. Moreover, farm households that adopt either agro-processing method are more likely to diversify into agritourism and other agribusinesses than non-agro-processing farms. Research limitations/implications: The authors could only access data on farm income and not on agro-processing costs. Future studies may address the impact of agro-processing on farm profitability if relevant data are available. Originality/value: Very few studies have examined the relationship between agro-processing, farm income and farm diversification. To the best of the authors’ knowledge, this is one of the first papers to examine the impact of different agro-processing practices on farm income and farm diversification. ? 2020, Emerald Publishing Limited.[SDGs]SDG2[SDGs]SDG8journal article10.1108/JADEE-07-2020-01572-s2.0-85100232740