Socioeconomic variation in child educational and socioeconomic attainment after parental death in Sweden

Abstract

In this study we use Swedish population register data to examine whether parental death differentially affects educational and occupational attainment according to the socioeconomic status of the parent who dies, and the socioeconomic status of the surviving parent and extended kin. That is, we examine whether parental death has an equalizing or exacerbating effect on offspring socioeconomic attainment, and also whether the socioeconomic status of the rest of the family plays a meaningful role in compensating for parental death. Using data on cohorts born 1973 to 1982 we examine five different outcomes, which are grade point average (GPA) at age 16 in high school, the transition from lower to upper-secondary education, the transition to tertiary education, overall educational attainment, and occupational status by age 30. We match families based upon antemortem parental socioeconomic trajectories. Overall, we find mixed results in our betweenfamily regression analyses adjusting for observables, with inconsistent evidence suggesting that losing a parent with very high socioeconomic resources is worse, and some evidence for a protective effect if the socioeconomic resources of the surviving parent and extended family members are at the top of the distribution. Using sibling fixed effects models that adjust for unobservable factors shared within the family, we see zero results for moderation by parents' occupations, but find consistent evidence that it is worse to lose a father at a younger age if grandparents have higher ranked occupations. We discuss possible interpretations of our findings. † barclay@demogr.mpg.de ‡

Lasting effects of parental death during childhood: evidence from Sweden

“…Following these theories, parental loss can also be considered an early-life shock with potentially lasting impacts throughout the life course. Losing a parent in early life has been found to negatively affect a variety of later-life outcomes, including mortality (Rostila et al, 2012), educational attainment (Barclay and Hällsten, 2019;Berg et al, 2014;Gertler et al, 2004;Kailaheimo and Erola, 2016;Prix and Erola, 2017) and occupational outcomes (Barclay and Hällsten, 2019). For the particular case of Sweden, research has shown that losing a parent in early life or childhood can have lasting effects on a educational attainment (Barclay and Hällsten, 2019;Berg et al, 2014), occupational choices (Barclay and Hällsten, 2019), mental health (Berg et al, 2016;Persson and Rossin-Slater, 2018), self-injury behaviour (Rostila et al, 2016), and mortality (Rostila and Saarela, 2011).…”

“…Second, the death of a parent might reduce both the financial resources of the household and the household's time budget as the surviving parent bears the responsibility for both childcare and earnings. Thus, losing a parent can lead to reduced parental investment into children with negative consequences for education, labour market outcomes or health (Barclay and Hällsten, 2019;Gertler et al, 2004). Several studies have reported a negative impact of parental loss on educational attainment at different levels (Barclay and Hällsten, 2019;Berg et al, 2014;Gertler et al, 2004; Kailaheimo studies exploit self-reported data on stress (Brodsky, 2016;Huizink et al, 2003;Lindström et al, 2012;Taylor et al, 2010), others look at stressful events taking place in administrative data (Pesonen et al, 2011), while yet others analyze medical measures of cortisol, a stress-related hormone (Huizink et al, 2003).…”

“…Moreover, step-parents did not compensate the negative impact of parental loss, which provides further evidence against the parental investment mechanism. Finally, Barclay and Hällsten (2019) find scattered evidence that the lost parent's educational or income levels regulate the effect of parental loss on the children's educational and occupational outcomes. They report some evidence for a protective effect of resources of the surviving parent or the extended family.…”

“…Third, it has been proposed that growing up in a non-traditional family structure may influence children's outcomes (Barclay and Hällsten, 2019;Corak, 2001), and thus the impact of parental loss can be likened to divorce (Amato and Anthony, 2014;Corak, 2001). Much of the impact of a nontraditional family structure on children's outcomes likely comes from differences in financial resources and parental time investments (McLanahan and Percheski, 2008), and thus there is considerable overlap between the parental investment mechanism and the family structure mechanism.…”

“…Disentangling these three mechanisms in practice is complicated, and few previous studies have provided empirical support for specific hypotheses (Barclay and Hällsten, 2019;Kailaheimo and Erola, 2016;Prix and Erola, 2017). In particular, identifying a direct effect of family structure independent of parental resources may not be possible.…”