Migrant women in industrialised countries experience high caesarean section (CS) rates but little is known about the effect of a previous delivery in the host country. This study set out to investigate this effect among migrant women in France, using data from the DSAFHIR study on healthcare access of migrant women living in emergency housing hotels, collected in the Paris Metropolitan area in 2017. Respondents reported life-long history of deliveries. We focused on deliveries occurring in France in 2000-2017: 370 deliveries reported by 242 respondents. We conducted chi-square tests and multivariate logistic regressions, adjusting for the clustering of deliveries among respondents by computing standard errors allowing for intragroup correlation. Mode of delivery was associated with duration of residence among multiparous women with no prior CS, with a higher CS rate with shorter duration of residence (16% vs. 7%, p = 0.04). In this group, a previous delivery in France was associated with a lower CS rate (5% vs. 16%, p = 0.008). In multivariate analysis, compared with women with previous birth in France, women giving birth in France for the first time had a higher risk of CS, regardless of duration of residence (aOR = 4.0, 95% CI = 1.3-12.1 for respondents with short duration of residence, aOR = 4.7, 95% CI = 1.2-18.0 for respondents with longer duration of residence). Efforts directed at decreasing the CS rate among migrant women should target women giving birth in the host country for the first tim