The “Separated-but-Not-Broken" Families and the Intergenerational Social Closure in China Use Rank-Based Approach to Study Family Structure and Social Mobility

As one of the core axes of social stratification, family structure has been widely emphasized to lead to “diverging destinies for children” and shape social mobility. This paper examines the moderating effects of a special form of family structure: separated-but-not-broken families on Inter-Generational social Closure (IGC) across generations. Using the unique retrospective information of family structure from the CFPS 2010 baseline data, combined with the occupational information of the younger generation from the five follow-up rounds from 2012 to 2020, we formed a unique dataset that simultaneously includes childhood family structure, parental occupational status (origin), and children’s occupational status (destination), spanning multiple cohorts. We defined the separated-but-not-broken families as families where either the father or mother did not live with the child for more than three months when the child was 0 to 12 years old. To avoid the impact of the remarkable transformation of the occupational structure across generations and cohorts on the measurement of IGC, we used a new rank-based method, namely, LACOP (Latent Bivariate Copula regression framework, Tam, Wang and Xiang, 2023), to analyze IGC across family structures. This new method links the discrete observable data (such as class, occupational status, and education) with latent continuous distributions, enabling the IGC estimation as a margin-free Copula dependence. Our main findings include: (1) the occupational IGC is stronger in separated families compared to intact families; (2) the stronger occupational IGC in separated families is mainly observed for people who hold rural Hukou; (3) the stronger occupational IGC in separated families was mainly observed among males and in the cohort of 1958 to 1977.