Skip to main content

A discrete choice model for labor supply and childcare


A discrete choice model for labor supply and childcare for mothers of preschoolers is presented. The mothers are assumed to make choices from a finite set of job possibilities and from a finite set of childcare options. Options in the markets for childcare are characterized by opening hours, fees and quality attributes. Similarly, jobs are characterized by a fixed wage rate, working hours and a number of variables related to job satisfaction. In the estimation of the model, we take into account that the number of options available might vary across work/care combinations and that some mothers are rationed in the market for care at day care centers. The model is employed to simulate the female labor supply effects of the Norwegian home care allowance reform.

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

Fig. 1
Fig. 2


  1. 1.

    These figures apply to children 1–5 years old. Thanks to generous maternity leave provisions, very few children attend day care centers in their first year. Children start school in their sixth year.

  2. 2.

    The term “home care allowance” is in accordance with the terminology used in Ilmakunnas (1997). However, the transfer might rather be characterized as an “out-of-childcare center allowance”. Note also that Schøne (2004) names the transfer as the “cash-for-care subsidy”. The figure in the text applies to 1998, while the (maximum) benefit for 2004 is NOK43,884 ($6,300).

  3. 3.

    There are, however, signs of convergence between the sexes in Norway regarding childcare and market work, which could mean that male income should also be treated as endogenous in future model proposals.

  4. 4.

    There are differences across regions with respect to price levels at childcare centers and to what extent rebates are given for care of the second and third child of the family. The figure refers to the fee of first child. Note also that expenses are deductible in the ordinary income tax base, which means that the government covers 28% of costs, up to a threshold. Users of center-based care and users of care by childminders that report their income to the tax authorities are eligible for this deduction.

  5. 5.

    This is an a priori assumption about grandparents.

  6. 6.

    It is fair to say that the reform gave rise to a fierce exchange of views on the direction of the modern welfare state.

  7. 7.

    There have been some minor adjustments in the transfer system since it was introduced. The system for 2000 is presented here, which cost about NOK2.8 billion in total expenditure.

  8. 8.

    At least, equal support was the initial intention. State budgets in recent years show that this is not strictly the case, as subsidies to centers are separated from the home care allowance rate system.

  9. 9.

    As a consequence, we only include families with a full-time working male in the estimations. Unemployment rates are in general low in Norway, and most fathers of preschoolers are full-time employed. Families that manage to combine two careers with parental care are observed in the data. However, it is difficult to find a good representation of the costs involved in such arrangements.

  10. 10.

    Note that two full-time working parents might not employ non-parental care because of shift-work arrangements. However, according to the data, this is not common.

  11. 11.

    The participation elasticity estimates in a non-rationing situation are 0.29 (wage), −0.05 (non-labor income), −0.11 (childcare costs).


  1. Aaberge R, Dagsvik JK, Strøm S (1995) Labor supply responses and welfare effects of tax reforms. Scand J Econ 97:635–659

    Article  Google Scholar 

  2. Averett SL, Peters HE, Waldman DM (1997) Tax credits, labor supply, and child care. Rev Econ Stat 79:125–135

    Article  Google Scholar 

  3. Bendiksen F, Hovland KA (1998) Foreldrebetalingsundersøkelse, spring 1998. Notater 98/21, Statistics Norway (in Norwegian)

  4. Blau DM, Hagy AP (1998) The demand for quality in child care. J Polit Econ 106:104–146

    Article  Google Scholar 

  5. Blau DM, Robins PK (1988) Child-care costs and family labor supply. Rev Econ Stat 70:374–381

    Article  Google Scholar 

  6. Blundell R, Duncan A, McCrae J, Meghir C (2000) The labour market impact of the working families' tax credit. Fisc Stud 21:75–104

    Article  Google Scholar 

  7. Chevalier A, Viitanen TK (2002) The causality between female labour supply participation and the availability of child care. Appl Econ Lett 9:915–918

    Article  Google Scholar 

  8. Chiuri MC (2000) Quality and demand of child care and female labour supply in Italy. Labour 14:97–118

    Article  Google Scholar 

  9. Connelly R (1992) The effect of child care costs on married women's labor force participation. Rev Econ Stat 74:83–90

    Article  Google Scholar 

  10. Connelly R, Kimmel J (2003) Marital status and full-time/part-time work status in child care choices. Appl Econ 35:761–777

    Article  Google Scholar 

  11. Dagsvik J (2000) Aggregation in matching markets. Int Econ Rev 41:27–57

    Article  Google Scholar 

  12. Dagsvik J (2001) The asymptotic distribution of maximum of independent and nonidentically distributed random variables, unpublished note, Statistics Norway

  13. Dagsvik J, Strøm S (2005) Sectoral labor supply, choice restrictions and functional form. J Appl Econ (forthcoming)

  14. Del Boca D (2002) The effect of child care and part time opportunities on participation and fertility decisions in Italy. J Popul Econ 15:549–573

    Article  Google Scholar 

  15. Del Boca D, Locatelli M, Vuri D (2004) Child care choices by Italian households. IZA DP No. 983, The Institute for Studies of Labor

  16. Duncan A, Giles C (1996) Labour supply incentives and recent family credit reforms. Econ J 106:142–155

    Article  Google Scholar 

  17. Duncan A, Paull G, Taylor J (2001) Mothers' employment and the use of child care in the United Kingdom. IFS Working Papers WP01/23, Institute for Fiscal Studies

  18. Eissa N, Liebman J (1996) Labor supply response to the earned income tax credit. Q J Econ 111:605–637

    Article  Google Scholar 

  19. Ermisch J (1989) Purchased child care, optimal family size and mother's employment. J Popul Econ 2:79–102

    Article  PubMed  Google Scholar 

  20. Gustafsson S, Stafford F (1992) Child care subsidies and labor supply in Sweden. J Hum Resour 27:204–230

    Article  Google Scholar 

  21. Håkonsen L, Kornstad T, Løyland K, Thoresen TO (2001) Kontantstøtten–effekter på arbeidstilbud og inntektsfordeling. Rapporter 2001/5, Statistics Norway (In Norwegian)

  22. Heckman JJ (1979) Sample selection bias as a specification error. Econometrica 47:153–161

    Article  Google Scholar 

  23. Heckman JJ, MacCurdy T (1980) A life cycle model of female labor supply. Rev Econ Stud 47:47–74

    Article  Google Scholar 

  24. Hofferth S, Wissoker D (1992) Price, quality, and income in child care choice. J Hum Resour 27:70–111

    Article  Google Scholar 

  25. Ilmakunnas S (1997) Female labour supply and work incentives. Labour Institute for Economic Research, Finland

    Google Scholar 

  26. Jenkins SP, Symons EJ (2001) Child care costs and lone mothers' employment rates: UK evidence. The Manchester School 69:121–147

    Article  Google Scholar 

  27. Kimmel J (1998) Child care costs as a barrier to employment for single and married mothers. Rev Econ Stat 80:287–295

    Article  Google Scholar 

  28. Kornstad T, Thoresen TO (2004) Means-testing the child benefit. Rev Income Wealth 50:29–49

    Article  Google Scholar 

  29. Kreyenfeld M, Hank K (2000) Does the availability of child care influence the employment of mothers? Findings from Western Germany. Popul Res Policy Rev 19:317–337

    Article  Google Scholar 

  30. Lanot G, Walker I (1995) Childcare and the labour market behaviour of UK mothers. Mimeo, Keele University, Keele

  31. Løyland K, Thoresen TO (2004) En undersøkelse av den registrerte dagmammavirksomheten. Notater 2004/41, Statistics Norway (In Norwegian)

  32. McFadden D (1974) Conditional logit analysis of qualitative choice behavior. In: Zarembka P (ed) Frontiers in econometrics. Academic, New York 105–142

    Google Scholar 

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

    Google Scholar 

  34. Michalopoulos C, Robins PK (2000) Employment and child care choices in Canada and the United States. Can J Econ 33:435–470

    Article  Google Scholar 

  35. Michalopoulos C, Robins PK (2001) Employment and child-care choices of single-parent families in Canada and the United States. J Popul Econ 15:465–493

    Article  Google Scholar 

  36. Michalopoulos C, Robins PK, Garfinkel I (1992) A structural model of labor supply and child care demand. J Hum Resour 27:166–203

    Article  Google Scholar 

  37. Naz G (2004) The impact of cash-benefit reform on parents' labour force participation. J Popul Econ 17:369–383

    Article  Google Scholar 

  38. Powell LM (1998) The impact of child care costs on the labour supply of married mothers: evidence from Canada. Can J Econ 30:577–594

    Article  Google Scholar 

  39. Powell LM (2002) Joint labor supply and childcare choice decisions of married mothers. J Hum Resour 37:106–128

    Article  Google Scholar 

  40. Ribar DC (1992) Child care and labor supply of married women. J Hum Resour 27:134–165

    Article  Google Scholar 

  41. Ribar DC (1995) A structural model of child care and the labor supply of married women. J Labor Econ 13:558–597

    Article  Google Scholar 

  42. Scholz JK (1996) In work benefits in the United States: the earned income tax credit. Econ J 106:156–169

    Article  Google Scholar 

  43. Schøne P (2004) Labour supply response to a cash-for-care subsidy. J Popul Econ 17:702–727

    Article  Google Scholar 

  44. Statistics Norway (2000) NOS C583 Kindergartens 1998

  45. Statistics Norway (2003) Most children with full-time spaces. Today's Statistics, 16th of June 2003

  46. Stølen NM, Svendsen I (1999) Skatt og arbeidstilbud. Appendix 3 in NOU 1999:7 Flatere skatt. Akademika, Oslo (In Norwegian)

  47. Walker J (1991) Public policy and the supply of child care services. In: Blau D (ed) The economics of child care. Russell Sage Foundation, New York pp. 51–86

    Google Scholar 

  48. Wrohlich K (2004) Child care costs and mothers' labor supply: an empirical analysis for Germany. DIW Discussion Papers 412, German Institute for Economic Research

Download references


We are in particular grateful to John K. Dagsvik and three anonymous referees for helpful suggestions. Valuable comments from seminar participants at the 57th Congress of the International Institute of Public Finance, Linz (Austria), August 2001, and the 13th Annual Conference of the European Association of Labor Economists, Jyväskylä (Finland), September 2001, are acknowledged. We have benefited from research assistance by Bård Lian. Financial support has been received from the Norwegian Research Council.

Author information



Corresponding author

Correspondence to Tom Kornstad.

Additional information

Responsible editor: Deborah Cobb-Clark


Appendix 1

Table 6 Second-stage estimation results for wage regression model, Heckman's selection model. Log wage as the dependent variable

Appendix 2

Specification of the likelihood function

The likelihood function used in the estimation of the model is given by

$$\begin{array}{*{20}c} {{\log L = {\sum\limits_{h = 1}^N {\left( {{\sum\limits_{m \in \Omega _{h} } {{\sum\limits_{j = 1}^5 {y_{{hjm}} } }} }\log {\sum\limits_{t = 1}^{30} {\frac{{e^{{v{\left( { \ifmmode\expandafter\tilde\else\expandafter\~\fi{H}_{j} , \ifmmode\expandafter\tilde\else\expandafter\~\fi{C}_{{hjmt}} ,X_{h} } \right)} + d_{{jm}} Z_{{jm}} }} }}{{e^{{v{\left( { \ifmmode\expandafter\tilde\else\expandafter\~\fi{H}_{1} , \ifmmode\expandafter\tilde\else\expandafter\~\fi{C}_{{h13t}} ,X_{h} } \right)} + d_{{13}} Z_{{13}} }} + {\sum\limits_{l \in \Omega _{h} } {{\sum\limits_{i = 1,2, \ldots ,5} {e^{{v{\left( { \ifmmode\expandafter\tilde\else\expandafter\~\fi{H}_{i} , \ifmmode\expandafter\tilde\else\expandafter\~\fi{C}_{{hilt}} ,X_{h} } \right)} + d_{{il}} Z_{{il}} }} } }} }}}} }} \right.} }}} \\ {{\left. { + y_{{h13}} \log {\sum\limits_{t = 1}^{30} {\frac{{e^{{v{\left( { \ifmmode\expandafter\tilde\else\expandafter\~\fi{H}_{1} , \ifmmode\expandafter\tilde\else\expandafter\~\fi{C}_{{h13t}} ,X_{h} } \right)} + d_{{13}} Z_{{13}} }} }}{{e^{{v{\left( { \ifmmode\expandafter\tilde\else\expandafter\~\fi{H}_{1} , \ifmmode\expandafter\tilde\else\expandafter\~\fi{C}_{{h13t}} ,X_{h} } \right)} + d_{{13}} Z_{{13}} }} + {\sum\limits_{l \in \Omega _{h} } {{\sum\limits_{i = 1,2, \ldots ,5} {e^{{v{\left( { \ifmmode\expandafter\tilde\else\expandafter\~\fi{H}_{i} , \ifmmode\expandafter\tilde\else\expandafter\~\fi{C}_{{hilt}} ,X_{h} } \right)} + d_{{il}} Z_{{il}} }} } }} }}}} }} \right)}} \\ \end{array} $$

where d jm denotes parameters, subscript h denotes female h and subscript t refers to the drawing of the error term in the wage rate equation. The dummy variable y hjm is defined as

$$ y_{{hjm}} = \left\{ {\begin{array}{*{20}l} {1 \hfill} & {{{\text{if}}\;{\text{female}}\;h'{\text{s}}\;{\text{labor}}\;{\text{supply}}\;{\text{is}}\;{\text{in}}\;{\text{group}}\;j\;{\text{and}}\;{\text{her}}\;{\text{choice}}\;{\text{of}}\;{\text{childcare}}\;{\text{is}}\;{\text{in}}\;{\text{mode}}\;m} \hfill} \\ {0 \hfill} & {{{\text{otherwise}}.} \hfill} \\ \end{array} } \right. $$

The dummy variables Z jm are introduced to adjust the probabilities for variations in number of opportunities across states. We assume that there might be more jobs and childcare possibilities within the groups long part-time work/childcare center (Z 41), full-time work/childcare center (Z 51) and not working/parental care (Z 13). The dummy variables are therefore activated only for j∈{4, 5} and m=1, and for j=1 and m=3. They are specified as

$$ Z_{{jm}} = \left\{ {\begin{array}{*{20}l} {1 \hfill} & {{{\text{if}}\;{\text{labor}}\;{\text{supply}}\;{\text{is}}\;{\text{in}}\;{\text{group}}\;j\;{\text{and}}\;{\text{choice}}\;{\text{of}}\;{\text{childcare}}\;{\text{is}}\;{\text{in}}\;{\text{mode}}\;m} \hfill} \\ {0 \hfill} & {{{\text{otherwise}}.} \hfill} \\ \end{array} } \right. $$

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kornstad, T., Thoresen, T.O. A discrete choice model for labor supply and childcare. J Popul Econ 20, 781–803 (2007).

Download citation


  • Female labor supply
  • Childcare
  • Discrete choice
  • Rationing

JEL Classification

  • J13
  • J22
  • C25