Elsevier

Labour Economics

Volume 17, Issue 3, June 2010, Pages 455-465
Labour Economics

A dynamic analysis of informal care and employment in England

https://doi.org/10.1016/j.labeco.2010.01.001Get rights and content

Abstract

This paper analyzes the dynamics in employment and informal care outcomes of women in England. To this end, we develop a dynamic model to describe pathways leading to a negative correlation between informal care and employment in a cross-section. The model allows for different types of caregiving, correlated permanent unobserved heterogeneity and initial sorting. The model is estimated on data from 6 waves of the BHPS 2000–2005. Our findings suggest modest feedback effects. We find a negative effect of co-residential caregiving on future employment and a negative effect of employment on future co-residential and extra-residential caregiving. We also find evidence of positive state-dependence in caregiving although most of the persistence in such activities is related to unobserved heterogeneity rather than state-dependence.

Introduction

The Office for National Statistics (ONS) estimates the total number of caregivers to be 6.8 million for the whole of Great Britain in 2000 (ONS, 2002). This number has been growing steadily over recent decades, in part explained by population aging and increased longevity. At the same time, there has been a trend away from co-residency of parents with children towards independent living and a preference for care in their own homes either purchased (formal sector) or provided by children (Grundy, 2000, OECD, 2005).

Increasing employment of particular groups (female and older workers) in Europe is still at forefront of the political agenda and is likely to remain so for some time. An important question to ask is therefore how responsive labor supply will be to changes in the age structure of the population and the demand for informal care. Many may face a “time crunch”, their time being more and more diverted from leisure towards work, caring for children, and caring for sick relatives, mostly parents and spouses.

The evidence for a link between caring and labor market outcomes is however mixed. Using U.K. data, Carmichael and Charles, 1998, Carmichael and Charles, 2003 document a large cross-sectional difference in the employment and wage rates of caregivers and non-caregivers. Heitmueller (2007) shows that the differences remain for co-residential caregivers (caregivers who live with the care recipient), using both IV and panel quasi-fixed effect methods. The U.S. literature is less conclusive on the link between employment and care. Early studies using an akin methodology to Carmichael and Charles, suggested a similar trade-off to that found for the U.K. (McLanahan and Monson, 1990, Stone and Short, 1990). However, Ettner (1995) tackles the simultaneity between these outcomes and finds an association between employment and co-residential care while no association for extra-residential care. Wolf and Soldo, 1994, Stern, 1995 find little negative association using alternative IV strategies and different datasets.

To our knowledge, no studies looked at the dynamics of caregiving and employment in the U.K. Dynamics ought to be important because time away from the labor market is likely to make it more difficult to return to the labor force. Also, caring may become easier or harder with time depending on whether learning is involved or health of the care recipient or caregiver deteriorates. In this paper, we study employment and caring trajectories of respondents in the British Household Panel Study (BHPS).

We follow a reduced-form approach of looking at the sequence of events that lead individuals to be out of the labor force and caring or vice versa. This involves modeling the sequence of caring and employment spells and looking for systematic differences in transition rates conditional on current states. In this process, we allow for correlated permanent unobserved heterogeneity which allows for permanent individual effects such as taste for care giving or attachment to the labor force. Since the static results tend to suggest that the relationship of different types of care and employment might be different (Ettner, 1995; Heitmueller, 2007), we distinguish between two types of care: extra-residential and co-residential care (i.e. people living either away from or with the person they care for). Caregivers living with the recipient are likely to provide different types of care compared to those providing care to external people, and therefore the dynamics may be different.

The structure of the paper is as follows. In Section 2, we briefly survey the theoretical foundation for the dynamic relationship between work and caring. Section 3 presents the data while Section 4 presents the empirical strategy used. In Section 5 we discuss the results, their robustness and their implications. Section 6 concludes.

Section snippets

Theoretical foundations for a dynamic model

Since we adopt a reduced-form dynamic approach to model caring decisions, it is important to discuss what kind of theoretical models would give rise to dynamics. Most of the literature on long-term care focuses on caregiving to disabled parents by children. There are two strands of models in this literature. The first strand models the interaction between one parent and one child. Wolf and Soldo (1994) (SW) present a simplified static unitary model of caregiving and labor supply in which

Data

The paper is based on longitudinal data from the British Household Panel Study (BHPS). We use annual records from the period 2000 to 2005.3

Econometric approach

We define the employment outcome as hit = I(hit* > 0) for respondent i in wave t where hit* is some loosely specified latent index measuring the utility from working. This latent index is not directly derived from a theoretical model. In a life-cycle model, it would be the difference in expected utility from working and not working, i.e. the function δh() in Eq. (1). Rather, the index is taken as:hit*=xitβ+jitγ+ϕhhhit1+ϕhd1dit1,1+ϕhd2dit1,2+uit,ht=1,...,Ti.

The vector xit includes a number of

Estimates

In Table 4a, Table 4b we present the parameter estimates and t-statistics for index parameters and covariance terms respectively.

We first focus on the dynamic effect of caregiving on employment. Current co-residential caregiving, but not extra-residential caregiving, seems to be negatively associated with future employment. This is consistent with caregiving depriving employment opportunities of caregivers who care for someone at home. The effect of extra-residential care on future employment

Conclusion

In this paper, we performed a dynamic analysis of the pathways through which caregiving and employment interact, allowing for different types of care, as well as correlated heterogeneity affecting each outcome. We find a small but statistically significant effect from co-residential caregiving to future employment and insignificant effect from extra-residential caregiving. Furthermore, we have found evidence of a dynamic link between employment and future caring spells, which would suggest that

Acknowledgements

We thank the editor and two referees as well as Julie Zissimopoulos, Arie Kapteyn, Arthur van Soest and seminar participants at various institutions for their comments and suggestions. Nazarov acknowledges funding from the Population Research center at RAND. The views expressed do not necessarily reflect the opinion of the Prime Minister's Strategy Unit.

References (24)

  • F. Carmichael et al.

    The labour market costs of community care

    Journal of Health Economics

    (1998)
  • F. Carmichael et al.

    The opportunity costs of informal care: does gender matter?

    Journal of Health Economics

    (2003)
  • A. Heitmueller

    The chicken or the egg? Endogeneity in labour market participation of informal caregivers in England

    Journal of Health Economics

    (2007)
  • D. Bell et al.

    Did the introduction of free personal care in Scotland result in a reduction of informal care

  • A. Borsh-Supan et al.

    Health, children, and elderly living arrangements: a multiperiod–multinomial probit model with unobserved heterogeneity and autocorrelated errors

  • Community Care Statistics, 2000–2005. Home Help/Care services for adults, England, various years (2000–2005) available...
  • B. Dostie et al.

    The living arrangement dynamics of sick, elderly individuals

    Journal of Human Resources

    (2005)
  • M. Engers et al.

    Long-term care and family bargaining

    International Economic Review

    (2002)
  • S. Ettner

    The Impact of “Parent Care” on Female Labor Supply Decisions

    Demography

    (1995)
  • E. Grundy

    Co-residence and mid-life children with their elderly parents in England and Wales: changes between 1981 and 1991

    Population Studies

    (2000)
  • J.J. Heckman

    The incidental parameters problem and the problem of initial condition in estimating a discrete-time data stochastic process

  • B. Heidemann et al.

    Strategic play among family members when making long-term care decisions

    Journal of Economic Behavior and Organization

    (1999)
  • Cited by (68)

    • The effect of unemployment on care provision

      2022, Journal of the Economics of Ageing
    View all citing articles on Scopus
    View full text