Education, Labour Supply and Marriage

Lead Research Organisation: Institute for Fiscal Studies
Department Name: IFS Research Team

Abstract

In most countries, inequality and child achievement are at the centre of public policy debate. Policies aimed at alleviating poverty or improving the opportunities of disadvantaged children often form the backbone of government welfare programmes. But such policies may fail to achieve their goals if not informed by a rigorous knowledge of what drives these outcomes.

However, poverty and underachievement are not well understood. They are the result of a complex net of decisions taken throughout the course of life by both the individual and the family. Three decisions are particularly crucial: education, labour supply and marriage. Their importance stems from the long-run effects they have for wellbeing, the outcomes of children and the distribution of resources within the family and across society as a whole. To a large extent, these effects are driven by the strong interconnections between education, labour supply and marriage over the lifetime.

For instance, education is understood to promote employment and the transmission of skills across generations. Since income and children's education are valued by families, education must be a valuable asset for marriage. Indeed, evidence shows that the education of wife and husband are strongly positively related. Spouses' education will then affect household resources and their allocation, particularly towards children. In turn, education investments are themselves determined by their likely returns in terms of employment, earnings, marriage and the distribution of household resources. Changes in any of these returns, maybe induced by a policy reform, lead to adjustments in education decisions and cause a chain response over the course of life.

The importance of some of these links has been demonstrated by past research. But they have not been investigated in a common framework capable of dealing with all their interconnections and explaining their consequences for the distribution of income or children's outcomes. We also know that changes in the tax and benefit system have a significant direct effect on some of these outcomes, but we know a lot less about their long-run effects triggered by indirect responses on forward-looking decisions like education.

In this project, we will contribute to a better understanding of the lifetime links between major decisions, their consequences for the wellbeing of spouses and children, and their implications for policy design. To this end, we will develop a model of education, labour supply, marriage and the distribution of resources within the household that explicitly deals with their lifetime links and consequences. The model will produce new systematic evidence on the nature of connections between major individual and family decisions and outcomes, their relevance for welfare policy design and their empirical significance in countries like the UK, US and Sweden.

Many aspects of a modern tax and benefit system will be explicitly included, such as progressive taxation or work-contingent benefits for families with children. The joint consideration of individual and family decisions over the course of life together with a description of the main features of a modern tax system will allow us to discuss formally the design of taxes and benefits for families for the first time. We expect the evidence produced by this research to make a significant contribution towards informing policymaking on how to balance key aspects of individual and family behaviour for the design of taxes and benefits for families.

Planned Impact

We propose investigating the role of marriage and marriage market conditions in determining choices and wellbeing over the course of life and the role of public policy in promoting investment in children and inter- and intra-household equity. The proposal lays the plans to contribute to the understanding of individual life-cycle behaviour, children's outcomes and the impact and design of welfare policies.

During the first two years of the project, we expect most of the impact to be academic as this project represents a first incursion into the proposed research theme. However, the project is eminently one of policy interest, and we will take the important steps into policy evaluation using the new framework. We expect to have an impact on three major non-academic groups:

1. Civil servants and policy-makers working in, or closely with, the HM Revenue and Customs, the HM Treasury, the Department for Work and Pensions and the Department for Education and responsible for policy design and assessment;

2. Think-tanks (such as the Resolution Foundation, the Institute for Public Policy Research and the Centre for Social Justice), other non-governmental organisations (such as Barnardos, the Child Poverty Action Group, the Family and Parenting Institute and the Joseph Rowntree Foundation), and international organisations supporting actions in the developing world (such as the World Bank, the European Bank for Reconstruction and Development and the OECD), interested in issues related with wellbeing, inequality and children's outcomes;

3. The media and the public in general, as we address issues at the centre of public debate and bring new insightful evidence to the table.

We foresee that the first two groups will benefit earlier from: (i) a better understanding of the most relevant decisions in life; (ii) a better understanding of the roots of inequality, not only between families but also within the family: (iii) a better understanding of the mechanisms promoting the inter-generational transmission of skills and child wellbeing; (iv) a better understanding of how taxes and subsidies may affect education, labour supply, marriage and the distribution of resources within the household, as well as their consequences for family wellbeing, particularly that of children; (v) the availability of a more adequate framework for tax policy analysis; and (vi) the empirical evidence we will bring to support our findings. All these have the potential to promote a more informed discussion of policy reform and to lead to improved welfare policy making.

Most of these benefits will be realised during the third year and after the end of the project, as we finish our academic investigation into the subject, study the policy implications of our framework, disseminate our results through conferences and seminars and release the tools allowing for a wider range of policy questions to be addressed in the future.

In the longer run, we anticipate outputs from our project will bring further insights on topics firmly at the centre of the public discussion arena, such as household behaviour, inequality, children's outcomes and their interactions with public policy, particularly the transfer system. The media and the public will benefit from the deeper understanding of these topics. Hopefully, the public will also benefit from improved policy design, via the impact we have on policymakers and civil servants and on motivating further research in this still largely unexplored field.
 
Description This aim of this project was to study the relationship between crucial lifetime decisions such as education, marriage and labour supply to understand their determinants, their consequences for individual wellbeing, and their implications for policy design. We investigated the interaction between marriage market conditions and the tax and benefit system, as well as their joint influence on education investments, life-cycle employment and the intra-household division of resources. More specifically, our research was driven by two overarching goals:
1. To formalise the links between education, marital prospects and labour supply in an integrated model of the lifecycle, and to use such a framework to understand theoretically and empirically: (i) the determinants of these decisions and (ii) their consequences for the intrahousehold allocation of resources and overall inequality.
2. To better understand the ability of taxes and benefits to redistribute resources and improve wellbeing when individual decisions are guided by household considerations.

Work on this project started by focusing on women life-cycle choices as these are more challenging to model than those of men given that selection into work and the choice of working hours is more critical for women. In a model of the women's education, family formation, labour supply and human capital formation, we studied the importance of working experience for earnings and the impact of the Working Families Tax (WFTC). The model is estimated on data from the British Household Panel Survey.

We found that working experience in full time jobs is critical for the wage progression of University graduated women and, to a less extent, High-School graduates. But experience of the job has little value for women leaving education at 16, without formal qualifications other than the GCSEs. We then find that the WFTC reform provided strong incentives for those with low qualifications to work, especially as lone mothers and particularly in part-time jobs. Since experience accumulated on the job is of little importance for the wages of these women, the programme has no long-term effects on the employment or earnings of these women. However, it does provide extra insurance against bad wage shocks of the low skilled and shifts resources from high-skilled and high-paid workers towards low paid workers with children. For that reason, it reduces the incentives to invest in education early in life, and we do detect a small negative impact of the reform on the probability of investing in education beyond the GCSE level. This research is now published in Econometrica (September 2016).

We then moved to formalise endogenous marital choices within an equilibrium model of marriage. Our model develops the life-cycle framework described above and extends it to account for marriage in a matching market, where individuals value the human capital of their prospective spouses. Equilibrium in the marriage market determines the intra-household allocation of resources in a transferable utility framework.

The model considers three stages of life and assumes Pareto efficiency and full commitment. We abstract from issues relating to divorce and our full commitment assumption precludes renegotiation; these are important questions we wish to address in future research. Agents first independently invest in human capital; their decision is driven by their idiosyncratic ability, their idiosyncratic cost of investment (which may for instance reflect borrowing constraints), and the expected returns on investment - which is itself determined by the equilibrium prevailing on the labour and marriage markets. In the second stage, individuals match on the marriage market, based on their human capital and their idiosyncratic preferences for marriage. Finally, the last period is divided into T sub-periods, during which couples or singles consume private and public goods, save and supply labour subject to permanent and transitory wage shocks, very much like standard lifecycle models.

Our model has several, original features. Following a Beckerian tradition, we model marriage as a frictionless, matching game in a transferable utility framework with risk averse agents. Individual utilities have an economic and a non economic component. The economic gain from marriage is twofold: spouses share a public good, and also insure each other against productivity shocks. In addition, marriage provides idiosyncratic, non-monetary benefits, which are additively separable and education-specific. The transferable utility property implies that, once married, households behave as a single decision-maker (unitary household). Despite its obvious shortcomings, this property considerably simplifies the analysis of the couple's dynamics of consumption and labour supply. We estimate the model using the British Household Panel Survey.

We found that the surplus from marriage is super-modular nearly everywhere, pushing towards positive assortative matching on human capital, with any departures from perfect sorting being driven by random preferences for mates. We also find that the human capital of women is a very strong determinant of marital surplus, more so than the human capital of men. The model is able to replicate matching patterns very well, despite the fact we do not allow for frictions.

High human capital women get more than half of the marital surplus, while men marrying low human capital women get most of the surplus. These shares reflect the existing equilibrium in the data. However, the share of welfare is endogenous and changes in the supply of men and women of different levels of human capital can change them. In a counterfactual simulation of the effects of a subsidy to university education we find that the surplus share of low human capital women increases, while the share of low ability college graduate women declines.

Lastly, our model sheds light on the determinants of human capital investments. Two conclusions emerge. First, non-economic factors play an important role in both the decision to marry and the marital patterns conditional on marriage and therefore indirectly affect the return to education. Second, and more importantly, we can explicitly decompose the returns to education into those perceived on the labour market and those reaped in addition on the marriage market (the \marital college premium"). Our results are unambiguous: the benefits perceived through marriage (through risk sharing and the joint consumption of public goods) are dominant. This paper was accepted for publication in the Journal of Political Economy.

Finally, work and discussion on both papers described above led us to investigate the gender pay gap and how it opens over the course of life. We used both the British Household Panel Survey and the Labour Force Survey for this study. In a widely discussed IFS Briefing note we found that, once we look by education groups, the gender pay gap did not close over the last 2 decades except for those without qualifications above the GCSE level. The gender wage gap widens gradually but significantly from when individuals reach their late 20s and early 30s. Men's wages tend to continue growing rapidly at this point in the life cycle (particularly for the high-educated), while women's wages plateau. The opening of the gap is clearly associated with the arrival of the first child and the diverging labour supply choices that mothers and fathers make. The gap is highly persistent: by the time their first child is aged 20, women have on average been in paid work for four years less than men and have spent nine years less in paid work of more than 20 hours per week.

Looking at women who leave paid work, hourly wages for those who subsequently return are, on average, about 2% lower for each year that they have taken out of employment in the interim. This relationship is stronger, at 4% per year, for women with at least A-level qualifications. We do not see such a relationship for the lowest-educated women, which is likely because they have less wage progression to miss out on or fewer skills to depreciate. Finally, there is no immediate hourly wage drop on average when women reduce their hours to 20 or fewer per week. But women working no more than that amount see less growth in wages, on average, than other women. Hence, the lower hourly wages observed in lower-hours jobs appear to be the cumulative effect of a lack of wage progression.
Exploitation Route This project constituted the first step towards a wider research agenda aimed at understanding the drivers of family decisions, their consequences for the wellbeing of individuals, for inequality within and across families and for the design of policy for families. It has already developed into at least two other research projects. The first project looks at the allocation of resources in families towards children and how parental human capital and marital choices affect child outcomes. The second project looks more carefully at the gender wage gap to discern how much of the divergence in wages is due to the loss in human capital from career intermittency on the part of women, and how much is explained by labour market conditions and the expectations from both firms and workers that women interrupt their careers or select into less demanding paths once they become mothers.

Our research on both these areas is being regularly discussed with groups from the DfE and the DWP working on child development and gender inequality issues.

The resources developed for the current project, including data and code, are now available for all to use. Many have already started to use the code developed for the first part of the project in their own work, e.g. to look at issues related to fertility and family labour supply.
Sectors Education,Government, Democracy and Justice

 
Description Our work on the links between education, family formation, labour supply and pay progression has received a strong attention from both government departments such as the DWP and the DfE, the Office for National Statistics, and non for profit organisations such as the The Fawcett Society and the Royal Society. In particular, we have discussed our approach and findings in various occasions with a group from the Government Equalities Office, and continue doing so regularly given their ongoing interest and work on gender inequalities. We have also organised a meeting with representatives for the Labour Market Analysis group of the DWP, to discuss their and our work on wage progression. Our reports on the gender pay gap received strong attention from the media, with coverage on the Financial Times, Guardian, BBC and other media outlets. We have been asked to look more closely at the drivers of the gender inequalities and their relation to family formation by various parties including charities like Oxfam, which we partly and are continuing to work on. Our work on education, marriage and labour labour supply caught strong attention from the academia and is generating further work on the implications of human capital and marital sorting for inequality and intergenerational mobility and the role of policy in protecting families against poverty. This work will lead to further impact.
First Year Of Impact 2016
Sector Education,Government, Democracy and Justice
Impact Types Economic,Policy & public services

 
Description A Celebration of the Life and Work of Gary S. Becker 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact Presentation at a conference entitled "A Celebration of the Life and Work of Gary S. Becker" at the University of Chicago, October 2014.
Year(s) Of Engagement Activity 2014
URL http://www.ifs.org.uk/uploads/publications/conferences/Chicago-PAC_MCD_CM-oct2014.pdf
 
Description Barcelona GSE Summer Forum 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact Workshop at the Barcelona GSE Summer Forum, June 2015.
Year(s) Of Engagement Activity 2015
URL http://www.ifs.org.uk/uploads/publications/conferences/GSE-Barcelona-PCC_MCD_CM-jun2015.pdf
 
Description Child development, parental investments and maternal labour supply 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Postgraduate students
Results and Impact Research meeting involving academincs, post-graduate students, researchers and practitioners.
Year(s) Of Engagement Activity 2017
 
Description Dynamic economics in Practice 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact Presentation: Dynamic economics in Practice at a seminar at FRB Chicago on the gender wage gap
Year(s) Of Engagement Activity 2018
 
Description Early parental investments and child development 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Industry/Business
Results and Impact Sarah Cattan gave a presentation at the workshop on 'Aging, Family and Social Insurance' on 28 February 2019 at Bergen University.
Year(s) Of Engagement Activity 2019
URL https://www.uib.no/sites/w3.uib.no/files/attachments/schedule_-_final.pdf
 
Description Education, marriage and child development 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Industry/Business
Results and Impact Presentation in Leuven by Costas Meghir at HCEO Family Inequality Workshop, June 19
Year(s) Of Engagement Activity 2018
URL https://hceconomics.uchicago.edu/events/family-inequality-workshop-spring-2018
 
Description Female Labour Supply, Human Capital and Tax Reform 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other audiences
Results and Impact presented at BCMS labour supply workshop, audience: academics
Year(s) Of Engagement Activity 2014
URL https://www.ifs.org.uk/uploads/Presentations/BCMS_Copenhagen_Slides_November_5_2014.pdf
 
Description Female Labour Supply, Human Capital and Welfare Reform 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact This presentation as given at:
- NBER TAPES (Trans-Atlantic Public Economics Seminar) conference on 16 to 18 June 2014 (Audience: academics)
- HM Treasury Microsimulation Seminar on 17 July 2014. (Audience: policy makers, civil servants)
- EEA-ESEM conference in Toulouse on 25th August 2014. (Audience: academics)
- University of Copenhagen Departmental Seminar on 5 November 2014. (Audience: academics)
- Choices workshop, May and November 2016 (audience: academics)
Year(s) Of Engagement Activity 2014,2016
URL https://www.ifs.org.uk/publications/7807
 
Description Human Capital, Marriage and Public Goods 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Conference presentation
Year(s) Of Engagement Activity 2017
URL https://www.economics.northwestern.edu/events/nemmers/conference.html
 
Description Income Dynamics, Income Risk, and Income Risk-Sharing 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact This workshop was held at the University of Atlanta in September 2013.
Year(s) Of Engagement Activity 2013
URL http://www.ifs.org.uk/uploads/publications/conferences/Atlanta-PAC_MCD_CM_sep2013.pdf
 
Description Marriage Market, Labor Supply and Education Choice 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact Seminar at the University of Chicago in February 2015.
Year(s) Of Engagement Activity 2015
URL http://www.ifs.org.uk/uploads/publications/conferences/Chicago-PAC_MCD_CM-feb2015.pdf
 
Description Marriage market, labor supply and education choice 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact This presentation was given at the following meetings:

This presentation was held at:

- Barcelona Graduate School of Economics, March 2016
- Society for Economic Dynamics, Annual Meeting, Toulouse School of Economics, France, June 2016
- Conference in Honour of Martin Browning, Oxford, UK July 2016
- Pennsylvania State University, October 2016
- Rice-PAC, MCD CM, February 2017
- ISCTE-PAC-MCD-CM February 2017 CCM
Year(s) Of Engagement Activity 2016,2017
URL https://www.ifs.org.uk/publications/8356
 
Description Marriage market, labor supply and education choice 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact This presentation was held at:

Barcelona Graduate School of Economics, March 2016,
Society for Economic Dynamics, Annual Meeting, Toulouse School of Economics, France, June 2016,
Conference in Honour of Martin Browning, Oxford, UK July 2016,
Pennsylvania State University, October 2016,
Rice-PAC, MCD CM, February 2017,
ISCTE-PAC-MCD-CM February 2017,
Royal Holloway Economics PhD Conference, May 2017,
Oslo, 2017,
Nemmers Prize Conference in Honour of Richard Blundell,
Seminar at the University of Chicago in February 2015.
Year(s) Of Engagement Activity 2015,2016,2017
 
Description Measuring and comparing assortativeness 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other audiences
Results and Impact Monica Costa Dias, "Measuring and comparing assortativeness" at:
- conference on "The Economics of Divorce", University of Oxford, Nov 18-19 2019.
- seminar at University of Essex, Feb 13, 2020
- seminar at University of Bristol, Feb 20, 2020
- seminar at University of Zurich, Nov 17, 2020
Year(s) Of Engagement Activity 2019,2020
 
Description Presentation: The Marriage market, labor supply and education choice 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact The Marriage market, labor supply and education choice: presented by Pierre-Andre Chiapporiv, Monica Costa Dias and Costas Meghir, held at:

• GSE Barcelona, March 8, 2016
• SED Annual Meeting, Toulouse, June 2016
• Conference in Honour of Martin Browning, Oxford July 2016
• Penn State University, October 2016
• Rice-PAC, MCD CM, 23 February 2017
• ISCTE-PAC-MCD-CM February 2017 CCM
Year(s) Of Engagement Activity 2016,2017
URL http://www.ifs.org.uk/uploads/Presentations/Monica%20Costa%20Dias_SED%20Toulouse_PAC_MCD_CM%20SED201...
 
Description The Gender Wage Gap and the Career Patterns of Men and Women 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other audiences
Results and Impact presented at BDEM-Greenwich-Jan2016, Audience: Academics
Year(s) Of Engagement Activity 2016
URL https://www.ifs.org.uk/uploads/publications/conferences/BDEM-Greenwich-Jan2016.pdf
 
Description The careers and wages of women 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact Various resentations at:
- Gorman conference 2020, on Oct 12-13
- Georgetown University, Mar 2, 2021
Year(s) Of Engagement Activity 2020,2021
 
Description The gender pay gap in the UK: children, working experience and location 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Seminar at the Federal Reserve Bank, Chicago
Year(s) Of Engagement Activity 2018
 
Description Wage progression in the UK 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact Presented at DWP meeting, Audicen: Policy makers
Year(s) Of Engagement Activity 2017
URL https://www.ifs.org.uk/publications/9263
 
Description Workshop in family economics 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact Workshop in Family Economics at the University of Copenhagen, September 2015.
Year(s) Of Engagement Activity 2015
URL http://www.ifs.org.uk/uploads/publications/conferences/Copenhagen-PAC_MCD_CM-Sep2015.pdf
 
Description Workshop: Female labour supply, human capital and tax reform 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other audiences
Results and Impact Female labour supply, human capital and tax reform, prepared for Choices workshop, delivered May 2016 by Richard Blundell, Monica Costa-Dias, Costas Meghir and Jonathan Shaw
Year(s) Of Engagement Activity 2016
URL http://www.ifs.org.uk/uploads/Presentations/BCMS_Copenhagen_Slides_November_5_2014.pdf
 
Description course on Computational Methods in Dynamic Economics 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact course on Computational Methods in Dynamic Economics at the FRB Chicago (Jan 16-19, 2018) and at IFS (Feb 8-9).
Year(s) Of Engagement Activity 2018