Research on divorce or union dissolution demonstrates that the presence of children, the number of children and/or the age of the youngest child is negatively correlated with the risk of breaking up. One line of argument is that children stabilize unions. Different mechanisms might be at work such as stronger feelings of obligation, stronger economic dependencies of mothers due to family specific investments leading to low or no labour market investments or due to reduced working hours as a consequence of insufficient external child care and so on. Another line of argument is that partners being less satisfied with or obligated to their relationship are less prone to have any children at all or to have further ones. We can also put it the other way round: A more promising relationship increases the chance of having (more) offspring as well as the chance of having a stable union.
To get new insight into this controversial topic of protection vs. selection we test the consequences of births resulting from unplanned pregnancies on the stability of (marital and cohabitating) unions. We frame women with an unplanned pregnancy as the treatment group. As unplanned pregnancies are highly selective we use the technique of propensity score matching (PSM) to identify one control group of women with planned pregnancies and one control group of women without a pregnancy (birth). Then we compare the union stability of these three groups using survivor functions. We use data from the German Socio-Economic Panel Study (SOEP), as since 2003 this yearly panel survey asks mothers of new-borns if the pregnancy had been planned or not.