Empirical Economics

, Volume 44, Issue 1, pp 135–161 | Cite as

College education and wages in the U.K.: estimating conditional average structural functions in nonadditive models with binary endogenous variables

  • Tobias J. Klein
Open Access


Recent studies debate how the unobserved dependence between the monetary return to college education and selection into college can be characterised. This paper examines this question using British data. We develop a semiparametric local instrumental variables estimator for identified features of a flexible correlated random coefficient model. These identified features are directly related to the marginal and average treatment effect in policy evaluation. Our results indicate that returns to college systematically differ between actual college graduates and actual college non-graduates. They are on average higher for college graduates and positively related to selection into college for 96% of the individuals. The dependence between selection into college and returns to college education is strongest for individuals with low math test scores at the age of 7, individuals with less educated mothers, and for working-class individuals.


Returns to college education Correlated random coefficient model Local instrumental variables estimation 

JEL Classification

C14 C31 J31 



This paper is based on the second chapter of my Ph.D. thesis, which I defended in July 2006 at the University of Mannheim. I am especially grateful to Erich Battistin, Richard Blundell, Pedro Carneiro, Andrew Chesher, Johannes Gernandt, Stefan Hoderlein, Enno Mammen, Melanie Schienle, and Edward Vytlacil for stimulating discussions and helpful comments on earlier versions of this paper. Moreover, I would like to thank Richard Blundell, Lorraine Dearden, as well as Leslie McGranahan for sharing their data. I would like to thank UCL for its hospitality during the academic year 2003/4, the European Commission for financial support through the Marie Curie program, and the Deutsche Forschungsgemeinschaft for financial support through SFB/TR 15. Furthermore, I would like to thank the audiences of various conference and seminar presentations for valuable comments. Finally, I would like to thank the three referees and the editor for their detailed comments and suggestions.

Open Access

This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.

Supplementary material

181_2010_355_MOESM1_ESM.pdf (25 kb)
ESM 1 (PDF 26 kb)


  1. Battistin E, Rettore E (2008) Ineligible and eligible non-participants as a double comparison group in regression-discontinuity designs. J Econom 142(2): 715–730CrossRefGoogle Scholar
  2. Berger MC, Black D, Smith J (2001) Evaluating profiling as a means of allocating government services. In: Lechner M., Pfeiffer F (eds) Econometric evaluation of labour market policies. Physica, Heidelberg, Germany, pp 59–84CrossRefGoogle Scholar
  3. Björklund A, Moffitt R (1987) The estimation of wage gains and welfare gains in self-selection models. Rev Econ Stat 69(1): 42–49CrossRefGoogle Scholar
  4. Blundell R, Dearden L, Sianesi B (2005) Evaluating the impact of education on earnings in the UK: models, methods and results from the NCDS. J R Stat Soc Ser A 168(3): 473–512CrossRefGoogle Scholar
  5. Blundell R, Powell JL (2003) Endogeneity in nonparametric and semiparametric regression models. In: Hansen L (eds) Advances in economics and econometrics. North Holland, AmsterdamGoogle Scholar
  6. Bound J, Jaeger DA, Baker RM (1995) Problems with instrumental variables estimation when the correlation between the instruments and the endogeneous explanatory variable is weak. J Am Stat Assoc 90(430): 443–450Google Scholar
  7. Card D (2001) Estimating the return to schooling: progress on some persistent econometric problems. Econometrica 69(5): 1127–1160CrossRefGoogle Scholar
  8. Carneiro P, Lee S (2009) Estimating distributions of potential outcomes using local instrumental variables with an application to changes in college enrollment and wage inequality. J Econom 149(2): 191–208CrossRefGoogle Scholar
  9. Carneiro P, Meghir C, Parey M (2007) Maternal education, home environments and the development of children and adolescents. CEPR Discussion Paper No. 6505, CEPR, London, U.K.Google Scholar
  10. Cleveland WS, Grosse E, Shyu WM (1991) Local regression models. In: Chambers JM, Hastie TJ (eds) Statistical models in S. Wadsworth/Brooks-Cole, Pacific Grove, CA, pp 309–376Google Scholar
  11. Currie J, Moretti E (2003) Mothers education and the intergenerational transmission of human capital: evidence from college openings. Q J Econ 118(4): 1495–1532CrossRefGoogle Scholar
  12. Fan J (1992) Design-adaptive nonparametric regression. J Am Stat Assoc 87(420): 998–1004CrossRefGoogle Scholar
  13. Fan J, Zhang W (1999) Statistical estimation in varying coefficient models. Ann Stat 27(5): 1491–1518CrossRefGoogle Scholar
  14. Frölich M (2007) Nonparametric IV estimation of local average treatment effects with covariates. J Econom 139(1): 35–75CrossRefGoogle Scholar
  15. Goldberger AS (1989) Economic and mechanical models of intergenerational transmission. Am Econ Rev 79(3): 504–513Google Scholar
  16. Griliches Z (1977) Estimating the returns to schooling: some econometric problems. Econometrica 45(1): 1–22CrossRefGoogle Scholar
  17. Hastie T, Tibshirani R (1993) Varying-coefficient models. J R Stat Soc Ser B (Methodological) 55(4): 757–796Google Scholar
  18. Haveman R, Wolfe B (1995) The determinants of children’s attainments: a review of methods and findings. J Econ Lit 33(4): 1829–1878Google Scholar
  19. Heckman JJ, Urzua S, Vytlacil E (2006) Understanding instrumental variables in models with essential heterogeneity. Rev Econ Stat 88(3): 389–432CrossRefGoogle Scholar
  20. Heckman JJ, Vytlacil EJ (1998) Instrumental Variables methods for the correlated random coefficient model: estimating the average rate of return to schooling when the return is correlated with schooling. J Hum Resour 33(4): 974–987CrossRefGoogle Scholar
  21. Heckman JJ, Vytlacil EJ (1999) Local instrumental variables and latent variables models for identifying and bounding treatment effects. Proc Natl Acad Sci 96: 4730–4734CrossRefGoogle Scholar
  22. Heckman JJ, Vytlacil EJ (2000) The relationship between treatment parameters within a latent variable framework. Econ Lett 66(1): 33–39CrossRefGoogle Scholar
  23. Heckman JJ, Vytlacil EJ (2001) Local instrumental variables. In: Hsiao C, Morimune K, Powell J (eds) Nonlinear statistical modeling. Proceedings of the thirteenth international symposium in economic theory and econometrics: essays in Honor of Takeshi Amemiya. Cambridge University Press, Cambridge, pp 1–46CrossRefGoogle Scholar
  24. Heckman JJ, Vytlacil EJ (2005) Structural equations, treatment effects, and econometric policy evaluation. Econometrica 73(3): 669–738CrossRefGoogle Scholar
  25. Hirano K, Imbens GW, Ridder G (2003) Efficient estimation of average treatment effects using the estimated propensity score. Econometrica 71(4): 1161–1189CrossRefGoogle Scholar
  26. Imbens GW, Angrist JD (1994) Identification and estimation of local average treatment effects. Econometrica 62(2): 467–475CrossRefGoogle Scholar
  27. Klein TJ (2010) Heterogeneous treatment effects: instrumental variables without monotonicity?. J Econom 155(2): 99–116CrossRefGoogle Scholar
  28. Mincer J (1974) Schooling, experience, and earnings. Columbia University Press, New YorkGoogle Scholar
  29. Newey WK (1997) Convergence rates and asymptotic normality for series estimates. J Econom 79: 147–168CrossRefGoogle Scholar
  30. Rosenbaum PR, Rubin DB (1983) The central role of the propensity score in observational studies for causal effects. Biometrika 70(1): 41–55CrossRefGoogle Scholar
  31. Solmon LC, Taubman PJ (1973) Does college matter? Some evidence of the impacts of higher education. Academic Press, New YorkGoogle Scholar
  32. Staiger D, Stock JH (1997) Instrumental variables regression with weak instruments. Econometrica 65(3): 557–586CrossRefGoogle Scholar
  33. Vytlacil E (2002) Independence, monotonicity, and latent index models: an equivalence result. Econometrica 70(1): 331–341CrossRefGoogle Scholar
  34. Willis RJ, Rosen S (1979) Education and self-selection. J Polit Econ 87(5, Part 2: Education and income distribution):S7–S36Google Scholar
  35. Wooldridge JM (2007) Instrumental variables estimation of the average treatment effect in correlated random coefficient models. In: Millimet D, Smith J, Vytlacil E (eds) Advances in econometrics: modeling and evaluating treatment effects in econometrics, vol 21. Elsevier, AmsterdamGoogle Scholar
  36. Xia Y, Li WK (1999) On the estimation and testing of functional-coefficient linear models. Stat Sinica 9: 735–757Google Scholar

Copyright information

© The Author(s) 2010

Authors and Affiliations

  1. 1.Department of Econometrics and OR, Netspar, CentERTilburg UniversityTilburgThe Netherlands

Personalised recommendations