![[ Visit Client Website ]](images/banner.gif)
We collected 190 provincial-level urban Dibao policy documents from legislative databases and government websites, developing a coding scheme to transform textual eligibility rules into a quantifiable format. Using this scheme, we constructed an original database of Dibao eligibility rules, encompassing 31 provincial regions over the period from 2004 to 2023. The 36 eligibility rules analyzed span six domains: income, assets, lifestyle, conduct, family composition and obligations, and status. To explore provincial variations and temporal evolution, we first employed latent class analysis (LCA) to identify typologies of eligibility conditions, followed by sequence analysis (SQ) to trace the trajectory patterns of these typologies over time across provinces (N=617). The LCA revealed five distinct classes of provincial Dibao eligibility regimes across two decades: stringent on lifestyle (21.56%), stringent on family composition and obligations (15.88%), stringent on status (14.42%), lenient on assets and lifestyle (23.82%), and stringent on assets but lenient on income (24.31%). The SQ analysis shows five clusters of transition patterns, exhibiting a converging trend from four classes in 2004 to three classes in 2023, with the majority (90.3%) being in the most lenient class—stringent on assets and lenient on income.
Our findings demonstrate that while local governments have exercised significant discretion in shaping Dibaoeligibility criteria—leading to the emergence of distinct regime types—a converging trend is evident toward greater leniency on income and status criteria, coupled with increased stringency on asset and lifestyle requirements. This study contributes to the conceptualization and operationalization of welfare conditionality in China by constructing multidimensional eligibility indicators. Furthermore, it enriches the understanding of subnational welfare regimes in China by providing a longitudinal analysis of their evolution.