Investments in the Education of Children Growing Up in Jobless Household: Causal Evidence From the Millennium Cohort Study

• In Event: Unequal Opportunities and Unequal Outcomes: International Evidence

Abstract

Objectives
I focus on children growing up in workless households (households in which no parent works for money) and how parents invest money and time in their children’s education.
Research Questions
How do educational investments differ between workless and non-workless households at different ages? The MCS data contains several measures for educational investments from the age of nine months up to the age of 17 that I am exploring in this study.
Is there a causal link between worklessness and educational investments? While finding an association between worklessness and investments is important to understand the disadvantage this group faces, seeing whether this link is causal helps understanding the mechanics of this particular disadvantage.
Theoretical Framework
Previous theoretical work (e.g., Becker & Tomes, 1986; Solon, 2004) models investments in the human capital of children by parents. These theories suggest that parents make decisions on their money and time investments based on their wealth and abilities.
Workless parents are likely to have stronger constraints on spending money on their child’s education. However, they might have more time with their child that they could spend on educating their children. On the other hand, workless parents might have smaller resources to effectively assist their child’s education, making time investments inefficient.
Data
I use data from the Millennium Cohort Study, a British birth cohort study that follows around 19,000 children born in the United Kingdom in the years 2000-2002.
Methods
To understand the difference between workless and non-workless background children and the investments made in them at different ages, I use both simple descriptive statistics as well as regression analyses to account for background characteristics. The causal analysis is based on more advanced panel data methods such as fixed effects regression and a combination of matching and difference in differences. Also, I apply a method described by Gottschald (1996) in which future worklessness is used as an instrument to account for unobserved factors at the time of interest.
Scientific Significance
Workless households face some unique challenges both financially and socially. Understanding how this affects children’s educational careers can help shape educational policy and benefits.
In light of the current word-wide pandemic and the resulting economic downturn with soaring unemployment, understanding the ramifications of worklessness becomes more important. Information about how parental worklessness affects children’s education can help prevent the attainment gap between advantaged and disadvantaged children to widen.
Results
Preliminary results indicate that workless parents invest less money on education and might not compensate by investing more time. For example, while 19% of non-workless background children and 12% of workless background children take an entry level exam (usually taken at age 11 to apply for secondary schools), 41% of non-workless background children receive paid-for lessons in order to prepare for that exam compared to only 13% of workless background children. On the other hand, average workless parents tend to read to their child less frequently and are more likely to not help with homework, even though they report having more time available with their child.

Author

• Silvan Häs, University College London