The child penalty is widely regarded as a main driver of the gender wage gap. This paper contributes to the literature by combining an event study framework with instrumental variables (IV) using the outcomes of in vitro fertilization (IVF) in Norway as instruments. We find that the endogeneity of fertility timing introduces substantial omitted variable bias in a regular OLS event study, while delayed fertility and intensive margin fertility responses lead to bias in previous IV estimates using IVF treatments. Our approach addresses these issues by centering time on the age of child rather than on time of the IVF attempt, and by instrumenting first and second births. The resulting estimates show that (i) fertility has short-lived negative effects on mother's earnings that mostly disappear after two to three years; (ii) effects are similar for the extensive and intensive margin of fertility; and (iii) partners are shielded from any negative effects