Our second approach is to consider the impact of parental death. Parental death
is much less correlated with other measures of family background than is parental
absence. Examining the impact of this more exogenous source of family dissolution
on children's outcomes may therefore tell us a great deal about the true impact of
other causes of separation. Using this approach, we conclude that the most important
effect is a strong impact of father's presence on son's probability of being married.
Overall, we find little evidence that parental presence early in life affects economic
well-being in adulthood.