Skip to main content

Parental transfers, student achievement, and the labor supply of college students


Using nationally representative data from the NLSY97 and a simultaneous equations model, this paper analyzes the financial motivations for and the effects of employment on U.S. college students’ academic performance. The data confirm the predictions of the theoretical model that lower parental transfers and greater costs of attending college increase the number of hours students work while in school, although students are not very responsive to these financial motivations. They also provide some evidence that greater hours of work lead to lower grade point averages (GPAs).

This is a preview of subscription content, access via your institution.


  1. 1.

    Kalenkoski (2005) shows that a substantial portion of parents transfer less than their EPC towards their child’s postsecondary education, suggesting that students must either choose a lower cost schooling alternative or fund the higher-priced schooling some other way, perhaps through student employment.

  2. 2.

    Students may work to support living expenses when setting up a new household in a dorm or apartment. This study will not consider these effects, nor the costs of room and board, due to lack of data.

  3. 3.

    Including leisure directly in the model would add another endogenous variable and hence another simultaneous equation to our empirical analysis. However, we are unable to estimate a system of four simultaneous equations where some of the variables are censored, and we do not have data on leisure time.

  4. 4.

    Using time-use data on students from one college, Stinebrickner and Stinebrickner (2004) found a large positive relationship between study-time and first-year GPA.

  5. 5.

    There are other possible models where parental altruism is not assumed that could describe transfer behavior within families, such as an exchange model (Cox 1987).

  6. 6.

    In reality, some students do take out student loans and incur credit card debt to finance their postsecondary education. However, as discussed in the introduction, students may not be able to borrow enough to completely cover the cost because they face federally guaranteed student loan maximums and high credit card interest rates. Thus, amounts not covered by parental transfers would need to be paid by the student through his or her own earnings. However, if one were to add loans as a choice variable in the model described here, it would not change the signs of the predictions that we test in the empirical analysis. (Derivations are available from the authors upon request.) Hence, we abstract from this complexity here. One should note, however, that incorporating loans into the theoretical model would require adding a loan equation to our simultaneous model and would generate the expected additional prediction that students work less when loans increase. However, we are limited in the number and types of regressions we can estimate simultaneously using aML.

  7. 7.

    There are several ways the model could be extended to account for multiple children. A crude way would be to redefine M p as the portion of the parent’s income that is available for this particular child, and let it be a function of the number of siblings, e.g., M p = M p(N), dM p/dN < 0.

  8. 8.

    For the most part, using these specific functional forms does not change the predictions of the model that we use to motivate our empirical analysis. However, in a model with general functional forms and the assumptions of positive and diminishing marginal utility and marginal product, one is unable to determine the sign of the effect of the student’s hours of work on parental transfers. One is also unable to determine the sign of the effect of the student’s wage on his/her hours of work. However, this is also true in the model with specific functional forms, unless we make the assumption that the cost of full-time schooling is greater than the parental transfer amount.

  9. 9.

    This depends on the assumption that the net price of schooling is greater than the parental transfer.

  10. 10.

    If it is instead assumed that the student’s ability positively affects the marginal product of schooling-related time, i.e., μ enters the achievement function multiplicatively rather than additively, it would give the same key results with one exception. It would allow us to positively sign \(\partial L/ \partial P_{\rm s}\) as in the model that treats price as exogenous.

  11. 11.

    Similar to previous studies, this analysis assumes that the decisions whether and where to enroll in college have already been made. While one might wish to estimate an enrollment probit or ordered probit along with the other three equations estimated here, we are limited in the number of equations we are able to jointly estimate. Thus, our results may apply only to enrolled students.

  12. 12.

    IPEDS data are matched to the NLSY97 data using a college identification number (UNITID code) available in the geocode version of the NLSY97.

  13. 13.

    ASVAB scores are a composite measure of math and verbal aptitude percentile scores constructed by NLSY97 staff from the computer adaptive form of the Armed Services Vocation Aptitude Battery. This composite measure is similar to the Armed Forces Qualification Test (AFQT) score.

  14. 14.

    The Consumer Price Index for All Urban Consumers (CPI-U) was used to convert all monetary values into 1997 dollars.

  15. 15.

    This is higher than the average transfer of $2,944 (converted by the authors to 1997 dollars) in Oettinger’s (2005) single public university sample; however, our sample includes not only public universities and colleges but also private ones that are generally more expensive.

  16. 16.

    These states include California, Colorado, Florida, Idaho, Indiana, Iowa, Kansas, Kentucky, Michigan, Minnesota, Nevada, New Mexico, Ohio, Pennsylvania, Rhode Island, Utah, and Washington. California discontinued their program in August 2003 and Rhode Island and Iowa in July 2001. Iowa’s program was restarted in June 2005.

  17. 17.

    High school grades are self-reported and measured on a 8.0 scale with 1.0 being mostly below D’s and 8.0 being mostly A’s.

  18. 18.

    Full coefficient estimates are available from the authors upon request.

  19. 19.

    The logical consistency condition for this model, 1 − γ 1 γ 2 > 0, is satisfied.

  20. 20.

    The number of siblings in the household is also used to help identify parental transfers in the hours of work equation, although it is not statistically significant.

  21. 21.

    As a robustness check, we estimated a specification with the net price of schooling included in the GPA equation. This was done because the net price of schooling may reflect the quality of the institution attended, and this quality may affect student GPAs. However, the net price of schooling was statistically insignificant in the GPA equation, and the other key results remained unchanged.

  22. 22.

    We also estimated a version of this simultaneous model where household siblings and the state work study program indicator are included in all equations rather than excluded from their respective transfer and work equations. This was done because one might challenge the former variable as previously discussed, while the latter variable has not been used by other researchers. In the model using 2-year college students, the effect of hours on GPA was insignificant, and work study had a positive significant effect on GPA.

  23. 23.

    Curiously, the estimated effect of the state minimum wage on parental transfers is positive and statistically significant, rather than negative, as expected. It is possible, however, that even though we intended for this variable to proxy for the student wage, it is capturing something else, perhaps economic conditions or general state support for youth.

  24. 24.

    The marginal effect for the net price of schooling accounts for both the linear and the squared term.

  25. 25.

    Students were asked to provide information on loans from relatives and friends as well as government-subsidized and other types of loans. We include all types of loans.

  26. 26.

    There were two few students (279) who attended school part-time to provide separate estimates.


  1. Becker GS, Tomes N (1976) Child endowments and the quantity and quality of children. J Polit Econ 84(4):S143–S162

    Article  Google Scholar 

  2. Cox D (1987) Motives for private income transfers. J Polit Econ 95(3):508–546

    Article  Google Scholar 

  3. Dustmann C, Van Soest A (2007) Part-time work, school success and school leaving. Empir Econ 32(2–3):277–299

    Article  Google Scholar 

  4. Dustmann C, Rajah N, Smith S (1997) Teenage truancy, working habits and wages. J Popul Econ 10:425–442

    Article  Google Scholar 

  5. Dustmann C, Micklewright J, Van Soest A (2009) In-school work experience, parental allowances, and wages. Empir Econ 37(1) (in press)

  6. Ehrenberg RG, Sherman DR (1987) Employment while in college, academic achievement, and postcollege outcomes: a summary of results. J Hum Resour 22(1):152–164

    Article  Google Scholar 

  7. Hotz VJ, Xu LC, Tienda M, Ahituv A (2002) Are there returns to the wages of young men from working while in school? Rev Econ Stat 84(2):221–236

    Article  Google Scholar 

  8. Jones EB, Jackson JD (1990) College grades and labor market rewards. J Hum Resour 25(2):253–266

    Article  Google Scholar 

  9. Kalenkoski CM (2005) Parents who won’t pay: expected parental contributions and postsecondary schooling. Public Financ Manag 5(1):178–236

    Google Scholar 

  10. Kalenkoski CM (2008) Parent-child bargaining, parental transfers, and the postsecondary education decision. Appl Econ 40(4):413–436

    Article  Google Scholar 

  11. Kane TJ, Rouse CE (1999) The community college: educating students at the margin between college and work. J Econ Perspect 13(1):63–84

    Article  Google Scholar 

  12. Keane MP, Wolpin KI (2001) The effect of parental transfers and borrowing constraints on educational attainment. Int Econ Rev 42(4):1051–1103

    Article  Google Scholar 

  13. Light A (1999) High school employment, high school curriculum, and post-school wages. Econ Educ Rev 18(3):291–309

    Article  Google Scholar 

  14. Light A (2001) In-school work experience and the returns to schooling. J Labor Econ 19(1):65–93

    Article  Google Scholar 

  15. Loury LD, Garman D (1995) College selectivity and earnings. J Labor Econ 13(2):289–308

    Article  Google Scholar 

  16. Maddala GS (1983) Limited-dependent and qualitative variables in econometrics. Cambridge University Press, Cambridge

    Google Scholar 

  17. Michael RT, Tuma NB (1984) Youth employment: does life begin at 16? J Labor Econ 2(4):464–476

    Article  Google Scholar 

  18. Molitor CJ, Leigh DE (2005) In-school work experience and the returns to two-year and four-year colleges. Econ Educ Rev 24(4):459–468

    Article  Google Scholar 

  19. Neumark D, Joyce M (2001) Evaluating school-to-work programs using the new NLSY. J Hum Resour 36(4):666–702

    Article  Google Scholar 

  20. Oettinger GS (1999) Does high school employment affect high school academic performance? Ind Labor Relat Rev 53(1):136–151

    Article  Google Scholar 

  21. Oettinger GS (2005) Parents’ financial support, students’ employment, and academic performance in college. University of Texas, Austin

    Google Scholar 

  22. Pabilonia SW (2001a) Evidence on youth employment, earnings, and parental transfers in the national longitudinal survey of youth 1997. J Hum Resour 36(4):795–822

    Article  Google Scholar 

  23. Pabilonia SW (2001b) Youth employment and parental transfers. Ph.D. Dissertation, University of Washington

  24. Rothstein DS (2006) High school employment and youths’ academic achievement. J Hum Resour 42(1):194–213

    Google Scholar 

  25. Ruhm CJ (1995) The extent and consequences of high school employment. J Labor Res 16(3):293–304

    Article  Google Scholar 

  26. Ruhm CJ (1997) Is high school employment consumption or investment? J Labor Econ 15(4):735–776

    Article  Google Scholar 

  27. Stephenson SP (1981) In-school labour force status and post-school wage rates of young men. Appl Econ 13(3):279–302

    Article  Google Scholar 

  28. Stinebrickner R, Stinebrickner TR (2003) Working during school and academic performance. J Labor Econ 21(2):473–492

    Article  Google Scholar 

  29. Stinebrickner R, Stinebrickner TR (2004) Time-use and college outcomes. J Econom 121(1–2):243–269

    Article  Google Scholar 

  30. Stratton LS, O’Toole DM, Wetzel JN (2008) A multinomial logit model of college stopout and dropout behavior. Econ Educ Rev 27(3):319–331

    Article  Google Scholar 

  31. Tyler JH (2003) Using state child labor laws to identify the effect of school-year work on high school achievement. J Labor Econ 21(2):353–380

    Article  Google Scholar 

  32. U.S. Department of Labor (2007) College enrollment and work activity of 2006 high school graduates. 26 April. Bureau of Labor Statistics, Washington, DC.

    Google Scholar 

  33. U.S. Department of Education (2007) Digest of education statistics (1997–2006). National center for education statistics, Washington, DC.

    Google Scholar 

  34. Van Dyke R, Little B, Callender C (2005) Survey of higher education students’ attitudes to debt and term-time working and their impact on attainment. Higher Education Funding Council for England, Bristol

    Google Scholar 

  35. Wolff FC (2006) Parental transfers and the labor supply of children. J Popul Econ 19(4):853–877

    Article  Google Scholar 

Download references


The authors would like to thank Kweku Opoku-Agyemang, Anastasiya Osborne, Tatevik Sekhposyan, and Judy Yang for research assistance. The authors would also like to thank Alison Aughinbaugh, Michael Giandrea, Mark Long, Peter Meyer, David Ribar, Larry Rosenblum, Donna Rothstein, Leslie Stratton, Leo Sveikauskas, Bruce Weinberg, Cindy Zoghi, and three anonymous referees for comments and suggestions. The authors are especially indebted to Stan Panis and David Ribar for their aML programming assistance.

Author information



Corresponding author

Correspondence to Sabrina Wulff Pabilonia.

Additional information

The views expressed in this paper are those of the authors and do not necessarily reflect the policies of the U.S. Bureau of Labor Statistics.

Responsible editor: Christian Dustmann

Appendix 1

Appendix 1


Table 10 Sample sizea

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kalenkoski, C.M., Pabilonia, S.W. Parental transfers, student achievement, and the labor supply of college students. J Popul Econ 23, 469–496 (2010).

Download citation


  • Employment
  • Transfers
  • GPA

JEL Classification

  • D1
  • I2
  • J2