“…Grounded in prior research, this study controls for individual, household, and city-level characteristics to mitigate omitted variable bias (Luttmer, 2005;Dynan and Ravina, 2007;Campante and Yanagizawa-Drott, 2015;Ding et al, 2021;Qiao and Cai, 2023). The individual-level characteristic variables include: Gender, with males coded as 1 and females as 0; Age, where respondents outside the age range of 15-65 years are excluded based on survey guidelines; Marital status, with married individuals coded as 1 and others as 0; Ethnicity, with Han ethnicity coded as 1 and ethnic minorities as 0; Self-rated health status, respondents who consider themselves as "healthy" or "very healthy" are coded as 1, while others are coded as 0.; Political affiliation, with party members coded as 1 and non-party members as 0; Weekly working hours; Education, with values ranging from 1 to 9 representing no schooling, primary or private school, junior high school, regular high school, vocational high school, technical school, secondary specialized school, junior college, and undergraduate or higher, respectively.…”