The association between basic pension insurance and fertility intentions among flexibly employed workers in China's new forms of employment.

INTRODUCTION: Following the relaxation of China's fertility policy, the nation-a rapidly growing group in the platform economy-are predominantly young, of reproductive age, and generally lack adequate social security coverage. Their fertility intentions therefore merit particular attention. METHODS: We conducted a stratified purposive survey across 23 provinces and municipalities, collecting 604 valid responses from NFE flexible workers. The analytical sample was restricted to ever‑married respondents with at least one child (N = 336). Using Logit, Probit, Poisson, and OLS regression models, complemented by Propensity Score Matching (PSM), we estimated the association between participation in basic pension insurance and two measures of fertility intention: willingness to have a second child (Plan1) and ideal number of children (Plan2). RESULTS: Participation in basic pension insurance is significantly negatively associated with both Plan1 and Plan2 across all models. The average marginal effect from the Logit model indicates a reduction of 8.5 percentage points in the probability of intending a second child (p < 0.01). The PSM‑estimated average treatment effect on the treated (ATT) shows that participants have, on average, about 23.5 percentage points lower second‑child intention than non-participants. Heterogeneity analyses reveal that the negative association is stronger for non-agricultural hukou holders and for those who engage in NFE as their sole or primary job, while it is weaker for secondary-job holders. DISCUSSION: These findings suggest that basic pension insurance may substitute for the traditional "raising children for old‑age support"motive, thereby dampening fertility intentions among NFE flexible workers. To balance old-age protection with fertility promotion, policy interventions could consider adjusting contribution rates, expanding social security coverage, and strengthening employment protection for female workers. Further research using panel or quasi‑experimental designs is needed to establish causality.