Teen childbearing is commonly believed to cause long-term socioeconomic disadvantages for mothers and their children. However, earlier cross-sectional studies may have inadequately accounted for marked differences in family background among women who have first births at different ages. We present new estimates that take into account unmeasured family background heterogeneity by comparing sisters who timed their first births at different ages. In two of the three data sets we examine, sister comparisons suggest that biases from family background heterogeneity are important, and, therefore, that earlier studies may have overstated the consequences of teen childbearing.