Working Paper: NBER ID: w25165
Authors: Anna Aizer; Paul J. Devereux; Kjell G. Salvanes
Abstract: Women who give birth as teens have worse subsequent educational and labor market outcomes than women who have first births at older ages. However, previous research has attributed much of these effects to selection rather than a causal effect of teen childbearing. Despite this, there are still reasons to believe that children of teen mothers may do worse as their mothers may be less mature, have fewer financial resources when the child is young, and may partner with fathers of lower quality. Using Norwegian register data, we compare outcomes of children of sisters who have first births at different ages. Our evidence suggests that the causal effect of being a child of a teen mother is much smaller than that implied by the cross-sectional differences but that there are still significant long-term, adverse consequences, especially for children born to the youngest teen mothers. Unlike previous research, we have information on fathers and find that negative selection of fathers of children born to teen mothers plays an important role in producing inferior child outcomes. These effects are particularly large for mothers from higher socio-economic groups.
Keywords: Teen motherhood; Child outcomes; Paternal characteristics; Sister fixed effects
JEL Codes: I3; J13; J24
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Negative selection of fathers (J12) | Child outcomes (J13) |
| Teen motherhood (J13) | Child outcomes (J13) |
| Family background (J12) | Child outcomes (J13) |
| Teen motherhood (J13) | Cognitive test scores (C12) |
| Teen motherhood (J13) | Years of schooling completed (I21) |
| Teen motherhood (J13) | Earnings at age 30 (J31) |
| Teen motherhood (J13) | Educational and labor market outcomes (J24) |