This paper uses the twin births instrument to estimate the causal effect of fertility on labor market outcomes for first-time mothers in Colombia.The nature of the data used allows us to overcome previous concerns regarding the validity of this instrument.We can control for observed pre-pregnancy characteristics, and by including fixed effects, we can control for time-invariant unobserved heterogeneity.Results coming from the traditional instrumental variables approach suggest a reduction in female labor supply close to 5.2 percentage points (18%) due to a marginal increase in fertility.Additionally, after following Farbmacher, Guber and Vikström (2018) to eliminate the bias of twins resulting from mothers' characteristics and in vitro fertilizations (e.g., a non-exogenous event), the obtained results are of the same magnitude but less significant than those of the traditional instrument.By making additional assumptions on the correlation of the twin instrument and the error of the structural equation following Nevo and Rosen (2012), we find an upper bound estimate for the reduction in female labor supply close to -5 ppt.