This study explores the linkage between nine policy indicators of public childcare provision and maternal employment in terms of employment propensity and (conditional) working hours. We apply different identification strategies with a two-way fixed effects specification with individual and macro-level confounders as well as year and state fixed effects as our most ambitious specification. Based on German microcensus data for waves 2006-2014, our findings show that identification, particularly in terms of state fixed effects, is crucial for the estimated effects. For three indicators however, we are left with significant associations even in the most complex model: For 1-2 year old children cared for by a childminder (3-5 year old children in daycare centres), an increase in the share of children taken care for less than 25 weekly hours on all same-age children attending public care by 10 percentage points is associated with a decrease of maternal employment propensity by 2 (4) percentage points. Thirdly, the existence of a legal claim on childcare from the age of one is associated with an increase in weekly working hours by 4.3 %, compared to a situation without such an entitlement. Compared to medium-level educated mothers, associations with respect to employment propensity are stronger (weaker) for mothers with a high (low) educational level whereas hours associations are weaker for highly skilled mothers. Compared to mothers in couples, single mothers respond less sensitively concerning both the extensive and the intensive margin of employment.