Skip to main content
SpringerLink
Log in

Timing of family income, borrowing constraints, and child achievement

  • Original Paper
  • Published:
Journal of Population Economics Aims and scope Submit manuscript

Abstract

I investigate the effects of the timing of family income on child achievement production. Detailed administrative data augmented with Programme for International Student Assessment test scores at age 15 are used to analyze the effects of the timing of family income on child achievement. Contrary to many earlier studies, the results suggest that the timing of income does not matter for long-term child outcomes. This is a reasonable result given the setting in a Scandinavian welfare state with generous child and education subsidies. Actually, later family income (age 12–15) is a more important determinant of child achievement than earlier income.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. See Table 5.21b in OECD (2007).

  2. To simplify exposition, this constraint completely rules out borrowing and saving.

  3. I choose to use actual income as there are no better alternatives, and it is the approach used in the existing literature.

  4. Certain public transfers are included in the disposable income: child benefits, housing benefits, and early retirement supplements.

    In 2000 and 2001, the disposable income does not include the rental value of own home. To adjust for this break, I use the cash value of real estate owned and a rough version of how the tax authorities calculated the rental value of own home in previous years.

    As a robustness test, estimations excluding the rental value of own home from disposable income prior to 2000 were run. Results were qualitatively the same, and the coefficients were statistically indistinguishable from the coefficients reported here.

  5. All the empirical analyses take the complex sampling design into account by employing the final student weights and the balanced repeated replications technique with Fay’s modification to obtain correct parameter estimates and standard errors. The test score for each individual is represented by five plausible values (PV), and the guidelines given in OECD (2005) are followed in order to obtain valid estimates and standard errors.

  6. Hanushek (1986) doubts the appropriateness of using test scores as outcome measures because they may not be strong predictors of outcomes outside the education system. The PISA test scores should be less likely to suffer from this shortcoming since they are designed to measure competencies needed not only inside but also outside the education system.

  7. I also find a positive association between the PISA reading score and the high school GPA.

  8. Individuals who are missing in the Danish registers are primarily individuals who were not present in Denmark in the year in question. As a results, only a small number of immigrants will remain in the sample after restricting the sample.

  9. The definition uses the legal parents (not necessarily the biological parents), implying that in the extreme, children can “change” parents over time, e.g., in the event of adoption of the child by a new step dad.

  10. r = 0.05 is also used in, e.g., Carneiro and Heckman (2003) and Caucutt and Lochner (2005). Section 5.2.5 takes a closer look at the consequences of changing this discount rate.

  11. The slope of the income profile could also be defined using the logarithm of income. The results are qualitatively similar in this case.

  12. To account for the fact that some families might temporarily appear to be single-parent families even though they are not, I further make the assumption that a child cannot be in a single-parent family in year t if the child lived with both parents at times t − 1 and t + 1. Thus, a breakup is conditioned on the parents living apart for at least two consecutive years. Specifically, for 1984, children did not incur a breakup if parents were living together the following year.

  13. I use the cumulative unemployment degree to quantify parents’ unemployment. Since this variable is only available for 1984 onwards, unemployment in a given year can only be quantified from 1985 onward.

  14. I have considered other thresholds (6, 9, and 12 months) as well as separate indicators for maternal and paternal unemployment. Although there are some differences in the significance and the size of the coefficients, all coefficients turn insignificant when all controls and timing variables are included in the specification.

  15. I also look at a couple of other outcomes. From the administrative data: completion of youth education and high school GPA. From the PISA survey: academic self-concept, mathematical self-concept, and an index of effort and perseverance. The PISA variables are indices with mean 0 and standard deviation 1 in the entire OECD sample. A thorough description of PISA 2000 questions and constructed indices is given in OECD (2002).

  16. Statistics Denmark, www.statistikbanken.dk.

  17. Similar transition matrices for the combination of different income stages yield similar results although, of course, the persistence tends to be higher for family incomes in stages closer to each other in time.

  18. The following section will consider the importance of the timing of family income in more detail.

  19. There is likely to be some interrelated effects of the age of the parents and their education, but the exact effects of these variables are not of interest in the present paper and will not be discussed further. In addition, there could be interaction effects between parents’ age at birth and the timing variables. I abstract from this and simply estimate the effect of the timing variables on the achievement outcome conditional on age.

  20. Excluding immigrants entirely from the sample does not change the results, although results where immigrant status is interacted with the other explanatory variables do suggest that the estimated coefficients for immigrants are significantly different for some of the explanatory variables. However, this is not the case for the family income variables.

  21. A fully flexible specification would just include permanent income and family income in each year except the last.

  22. Using the logarithm of income in the definition of the slope of the income profile or defining separate variables for different intervals of the slope does not change this result.

  23. Results summarized in this section are not presented, but they are available from the author upon request.

  24. The PISA 2006 survey sampled children born in 1990. It differs from the PISA 2000 survey on a few points. Most importantly, the focus area of PISA 2006 was science, while the focus area of PISA 2000 was reading. However, the 2006 data set includes test scores on both reading, math, science, and problem solving. Also, some variables are not available in the 2006 data set. Specifically, some survey questions about family structure, birth order, and the number of siblings are not included in the 2006 survey, nor is postoutcome income available. The 2006 survey also includes some questions that were not in the 2000 survey. These were mainly science-related questions.

  25. I also compute per capita income using information about the number of adults and children in the household and find qualitatively similar results (not reported). The family structure variables and the number of siblings will partially account for household size in the reported analysis.

References

  • Aakvik A, Salvanes KG, Vaage K (2005) Educational attainment and family background. Ger Econ Rev 6(3):377–394

    Article  Google Scholar 

  • Blau DM (1999) The effect of income on child development. Rev Econ Stat 81(2):261–276

    Article  Google Scholar 

  • Carneiro P, Heckman JJ (2003) Human capital policy. In: Heckman JJ, Krueger AB, Friedman BM (eds) Inequality in America. What role for human capital policies? MIT, Cambridge, pp 77–239

    Google Scholar 

  • Caucutt EM, Lochner L (2005) Early and late human capital investments, borrowing constraints, and the family. Unpublished manuscript

  • Cunha F, Heckman JJ (2006) The technology of skill formation. Am Econ Rev 97(2):31–47

    Article  Google Scholar 

  • Cunha F, Heckman JJ, Lochner L, Masterov DV (2006) Interpreting the evidence on life cycle skill formation. In: Hanushek EA, Welch F (eds) Handbook of the economics of education, vol 1. North-Holland, Amsterdam, pp 697–812

    Chapter  Google Scholar 

  • Dahl GB, Lochner L (2005) The impact of family income on child achievement. NBER working paper 11279. NBER

  • Danish Economic Council (2006) Danish economy fall 2006 (in Danish: Vismandsrapporten). Danish Economic Council, Copenhagen

    Google Scholar 

  • Duncan GJ, Brooks-Gunn J (eds) (1997a) Consequences of growing up poor. Russell Sage Foundation, New York

    Google Scholar 

  • Duncan GJ, Brooks-Gunn J (1997b) Income effects across the life span: Integration and interpretation. In: Duncan GJ, Brooks-Gunn J (eds) Consequences of growing up poor. Russell Sage Foundation, New York, pp 596–610

    Google Scholar 

  • Duncan GJ, Yeung WJ, Brooks-Gunn J, Smith JR (1998) How much does childhood poverty affect the life chances of children? Am Sociol Rev 63(3):406–423

    Article  Google Scholar 

  • Ermisch J, Francesconi M, Pevalin DJ (2004) Parental partnership and joblessness in childhood and their influence on young people’s outcomes. J R Stat Soc, A 167(1):69–101

    Article  Google Scholar 

  • Haider S, Solon G (2006) Life-cycle variation in the association between current and lifetime earnings. Am Econ Rev 96(4):1308–1320

    Article  Google Scholar 

  • Hanushek EA (1986) The economics of schooling: production and efficiency in public schools. J Econ Lit 24(3):1141–1177

    Google Scholar 

  • Haveman R, Wolfe B (1995) The determinants of children’s attainments: a review of methods and findings. J Econ Lit 33(4):1829–1878

    Google Scholar 

  • Jenkins SP, Schluter C (2002) The effect of family income during childhood on later-life attainment: evidence from Germany. Working Papers of the Institute for Social and Economic Research 2002-20. Colchester, University of Essex

  • Knighton T, Bussière P (2006) Educational outcomes at age 19 associated with reading ability at age 15. Culture, tourism and the centre for education statistics research papers. Statistics Canada, Ottawa

    Google Scholar 

  • Levy D, Duncan GJ (2000) Using sibling samples to assess the effect of childhood family income on completed schooling. Joint Center for Poverty Research working papers 168

  • Mayer SE (1997) What money can’t buy. Family income and children’s life chances. Harvard University Press, Cambridge

    Google Scholar 

  • OECD (2001) Knowledge and skills for life. First results from the OECD Programme for International Student Assessment (PISA) 2000. OECD, Paris

    Google Scholar 

  • OECD (2002) Manual for the PISA 2000 database. OECD, Paris

    Google Scholar 

  • OECD (2005) PISA 2003 data analysis manual. OECD, Paris

    Google Scholar 

  • OECD (2007) PISA 2006. Vol 2 data. OECD, Paris

    Book  Google Scholar 

  • Ruhm CJ (2004) Parental employment and child cognitive development. J Hum Resour 39(1):155–192

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. School of Economics and Management, Aarhus University, Building 1322, Bartholins Allé 10, 8000, Aarhus C, Denmark

    Maria Knoth Humlum

Authors
  1. Maria Knoth Humlum
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Maria Knoth Humlum.

Additional information

Responsible editor: Deborah Cobb-Clark

I thank the Danish Strategic Research Council Project: “Intergenerational Transmission

of Human Capital” for financial support. I am indebted to Helena Skyt Nielsen and Rune Majlund Vejlin for their many useful comments and suggestions. I thank Michael Svarer, Marianne Simonsen, Astrid Würtz, seminar participants at DGPE, and two anonymous referees for helpful comments. The usual disclaimer applies.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Humlum, M.K. Timing of family income, borrowing constraints, and child achievement. J Popul Econ 24, 979–1004 (2011). https://doi.org/10.1007/s00148-010-0309-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00148-010-0309-9

Keywords

JEL Classification

Search

Navigation