In many countries, along with rising rural labor transfer, the problem of rural domestic sewage discharge is becoming increasingly serious due to labor shortages in the villages. It is urgent to solve the environment pollution and health problems of residents which is caused by the massive discharge of domestic sewage in rural areas. Based on the survey data collected from Nanjing Agricultural University in 2020, this paper employs the ordered probit model and the CMP method, to empirically test the impact of non-agricultural employment and regional choice on farmers’ domestic sewage discharge behavior and the moderating effect of environmental cognition and the social network. The results show that: (1) There is a significantly positive correlation between non-agricultural employment and farmers’ sewage treatment behavior. (2) Environmental cognition significantly improves the participation of urban non-agricultural employment farmers in sewage treatment, and the social network has a significant role in promoting the adoption of sewage treatment behavior of local non-agricultural employment farmers. (3) Further heterogeneity analysis results show that the inhibitory effect of urban non-agricultural employment on random sewage discharge is more pronounced than that of local non-agricultural employment. Therefore, in order to effectively solve the problem of rural domestic sewage discharge, it is necessary to actively guide the sewage discharge behavior of non-agricultural employment households, strengthen the social network interaction within the village, and increase the publicity for sewage discharge knowledge.

This study examines the spillover effects of China’s New Rural Pension Scheme (NRPS) implemented in 2009 as a cash transfer program for agricultural production. Based on the data collected by the China Health and Retirement Longitudinal Survey (CHARLS) in four periods (2011, 2013, 2015 and 2018), we employ Seemingly Unrelated Regression to explore how China’s NRPS affects agricultural production. Our findings show that NRPS pensions reduce household operating areas by 1.99 mu and agricultural investment by 1150 yuan, while increasing the labor time of their own agricultural production by 168 h, and farmers in the payment period have a similar impact. This finding is still reliable after a series of robustness tests. Gender heterogeneity analysis indicates that male participation in NRPS is more likely to reduce the actual operating area and increase the labor input of the family, while female participation in NRPS is more likely to reduce the agricultural capital input of the family. Moreover, the in-depth study of agricultural performance shows that the implementation of NRPS helps increase the average output value per mu by 700 yuan and technical efficiency by 0.2%, although this is at the cost of declining labor productivity. This study links the joint decision-making of agricultural production factor inputs with pension schemes and contributes to the development of relevant research, which may provide policy implications for how cash transfer schemes affect agricultural production and agricultural performance in other countries.