Do Men Care? Estimating Men’s Preferences for Spending Time with Their Children

ABSTRACT

Is the time men use on childcare and household work the result of preferences or cultural, institutional and economic constraints? Can such constraints be measured when we only observe men’s choices (functionings) but not their capabilities? Using a random utility model together with stochastic specifications of the probability of having different capabilities, this paper shows that it is possible to distinguish between preferences and capabilities. Utilising time use data for Spain, we find that even though men do relatively little childcare, it is important to them. So, men do care to care. Our estimates show that, given our model, only about 9% of men with children have the full capability set, while 58% of them are constrained to a low level of care and housework. According to our model, many of these would not change behaviour if they had the full capability set, but about 20% of fathers would choose to provide more childcare and housework.

KEYWORDS:

• Household production
• random utility models
• time use
• capabilities
• childcare

JEL CLASSIFICATION:

• C25
• D13
• J13

Acknowledgements

The project has been carried out thanks to the “Progetto di Ateneo 2012” grant scheme funded by Compagnia di San Paolo. We are very grateful for the valuable work done by Marco Fuscaldo on the data and for his help in the initial modelling stages of the project. We also thank for valuable suggestions and comments from participants to the seminar at the Frisch Center for Economic research, University of Oslo, and to participants to the conferences organised by the European Society for Population Economics, by the Human Development and Capability Association, and by the International Association for Feminist Economics.

This document presents results drawn from the Multinational Time Use Study (MTUS), but the interpretation of this data and other views expressed in this text are those of the author. This text does not necessarily represent the views of the MTUS team or any agency which has contributed data to the MTUS archive. The author bears full responsibility for all errors and omissions in the interpretation of the MTUS data.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 A Time Use Survey for Spain was collected also in 2009, but with a much smaller sample size than the 2002 survey (approximately 23,000 vs. 47,000 observations). Given the complexity of the model estimated in this paper, we have preferred to use the data for 2002.

2 The original data are challenging in that they only have one observation per household and include only household income divided into eight different categories. We describe in detail in Appendix A and the supplemental material how we have transformed the data using different estimation techniques.

3 Our discussion here falls within the collective model of the household, where each person in the family will act to maximize their welfare, given the predicted behaviour of others. There is a large literature on bargaining in the family. A useful introduction is given in Ermisch (2003).

4 Appendix A gives a more detailed description of how we interpret the household dynamics and how the men’s and women’s time use in different states are calculated. Supplemental Material A gives a simple example of this type of household dynamics. This implies predicting the time use and income they could have in the states that they have not chosen. We do this based on estimated equations for how each activity is shared between couple in different states. Supplemental Material B describes these estimations.

5 We shall later see that in our chosen econometric specification (a random utility model with a log linear utility function) these economy of scale parameters are not identified, but we include them here for completeness.

6 We also explored the possibility of including time spent on job search activities as an additional time category for people in low level of paid work. Unfortunately, only 12 men in states 3 and 4 in our sample reported any time dedicated to this activity. There is therefore not enough variability in the variable to include it in the model.

7 These shares were defined in Equations A2 to A4 in Appendix A.

8 Not all men that declare themselves as unemployed in our survey are observed to work 0 h, and some unemployed men are actually observed in a high level of paid work (i.e. 6 out of the 90 unemployed men observed in our sample).

Additional information

Funding

This work was supported by Compagnia di San Paolo [Grant Number Progetto di Ateneo 2012].

Notes on contributors

Leif Andreassen

Leif Andreassen has a Ph.D. in Economics from the University of Oslo. He has been mainly employed at Statistics Norway since 1986. From 1996 to 1999, he held a part-time position as Associate Professor II at the University of Oslo. During the academic year 1999–2000, he was a visiting scholar at New York University and from 2008 to 2013 he was a visiting scholar at the University of Turin. He led the development of the first version of the MOSART microsimulation model from 1988 to 1990 and headed the project until 1998. His research interests are mainly in demographics, labour economics, Sen's capability approach and econometrics, including work on microsimulation models, overlapping generation models and econometric analysis of income and labour supply.

Maria Laura Di Tommaso

Maria Laura Di Tommaso is Full Professor of Economics at Department of Economics and Statistics “Cognetti de Martiis” of the University of Turin in Italy, Fellow of Collegio Carlo Alberto and Associate Researcher of the Frisch Center for Economic Research in Oslo. Previous positions include: College Lecturer in Economics and Fellow at Robinson College, University of Cambridge and Research Associate in the Department of Applied Economics, University of Cambridge. She has extensively published in international journals and her research interests are in the field of microeconomics applied to gender and feminist issues. She was Associate Editor of the journal Feminist Economics from 2012 to 2016, she is the Director of the PhD program in Economics Vilfredo Pareto and she was a member of the National Scientific Qualification committee (ASN 2018-2021) in econometrics.

Anna Maccagnan

Anna Maccagnan holds a Ph.D. in Labour Economics from the University of Modena and Reggio Emilia. This work has been carried out while working as a Research Fellow at the University of Turin. Anna has previously held research posts at the University of Modena and Reggio Emilia, at the University of Bologna and at the University of Exeter, working in the fields of happiness and wellbeing, measurement of capabilities, economics of education, household economics and labour economics.