Older people's participation in extra-cost disability benefits

https://doi.org/10.1016/j.jhealeco.2012.11.004Get rights and content

Abstract

The targeting of an UK extra-cost disability benefit for older people, Attendance Allowance, is analyzed using longitudinal data from the British Household Panel Survey. First, a binary model of benefit participation is used to investigate whether receipt is responsive to the onset of disability. Second, matching estimators are used to evaluate the consequences of missed participation on later financial wellbeing. Results indicate that participation is highly responsive to the onset of disability, although the chance of delays in receipt emerges. Personal characteristics unrelated to eligibility also appear to influence benefit receipt, translating into sizeable differences in the amount of cash support received. The comparison of recipients with observationally equivalent non-recipients confirms that timely participation reduces disabled older people's financial strain.

Introduction

Modern welfare systems encompass a range of cash programmes protecting against the consequences of disability. Some represent earnings replacement for working age individuals who have lost the capacity to work (Bound and Burkhauser, 1999); others are meant to provide for the additional needs that the onset of disability implies, such as care and additional goods and services purchased to compensate for disability (Jones and O’Donnel, 1995, Zaidi and Burchardt, 2005, Stapleton et al., 2008, Cullinan et al., 2011, Morciano et al., 2011).1 These ‘extra-cost’ benefits are the most relevant for older people, who generally no longer participate in the labour market and enjoy standards of living predominantly determined by previous working-age circumstances. Over the last few decades, several OECD countries have experienced a sustained growth in disability benefit rolls (Bound and Burkhauser, 1999, Disney and Webb, 1991, McVicar, 2008). A similar pattern, together with the anticipation of rising dependency ratios, is posing a strain on the continuation of current cash provision for disability in older age (OECD, 2005, Bolin et al., 2008).2

Despite the fact that several European countries offer extra-cost disability benefits, surprisingly little research has analyzed their effectiveness in reaching disabled older people and reducing their unmet needs. Targeting, i.e. whether benefits in payment are successfully and promptly reaching (only) the target population, is crucial in determining the effectiveness of this form of social care provision.3

Previous literature has mostly approached the issue as a problem of tagging truly disabled working age applicants, as opposed to individuals using disability benefits as earnings replacement in times of adverse economic conditions or as early retirement (Parsons, 1991, Parsons, 1996, Bound, 1989, Diamond and Sheshinski, 1995, Becker et al., 2005). However, in the case of people over retirement age, the issue stands in different terms. Receipt of disability benefit involves both the disabled elderly deciding whether to claim or not, and the administration assessing the case, once the claim is received (Pudney, 2009). Both the case of an eligible person not applying and that of a disabled claimant being denied the award represent targeting failures. From a claimant's perspective ex-ante eligibility is somewhat uncertain, and such uncertainty might in turn affect the decision to apply, as stressed in previous studies (Halpern and Hausman, 1986, Parsons, 1991, Kreider, 1998, Kreider and Riphahn, 2000) mostly based on working age individuals. Also, socio-economic characteristics unrelated to eligibility might make equally disabled individuals more or less likely to apply or to be judged eligible, compromising in practice the horizontal equity that schemes are meant to guarantee.

This paper analyses participation in an extra-costs disability benefit programme available to older people in the UK, known as Attendance Allowance (AA). AA was first introduced in 1971 and still represents the UK extra-cost benefit available to disabled people after the age of 65.4 AA is currently paid to about 1.7 million people. AA rolls have been constantly increasing since the benefit was first introduced; it has been argued that a similar pattern of growth could not be explained by the increased prevalence of disability only, and that trends in claiming behaviour or in administrative rigour might have played a role (Berthoud, 2009). Recent policy debate has questioned the role of AA (Bethoud and Hancock, 2008) and concerns have been raised about administrative inconsistencies in adjudication and severe delays in receipt after the onset of disability (Daly and Noble, 1996, Hirst, 1997, Banks and Lawrence, 2005, Nosowska, 2004, Pudney, 2010).

AA eligibility depends on whether the claimant needs care in order to perform daily activities as a consequence of a physical or mental impairment.5 The list of care needs covered includes: ‘getting into/out of bed’; ‘toilet needs’; ‘washing, bathing, showering’; ‘dressing/undressing’; ‘walking around indoors, going up/down stairs, getting in/out of a chair, transferring to/from a wheelchair’; ‘get up after a fall’; ‘cutting up food, eating or drinking’; ‘taking medicines’; ‘communicate with others’; ‘supervision to prevent danger’. Awards neither depend on past national insurance contributions nor are means tested. The programme delivers a sizeable additional income that recipients can spend as they wish: either at a lower rate (£220 per month), if care is needed throughout either day or night, or at a higher rate (£330 per month) if care is needed throughout both night and day. The claiming process needs to be initiated by the claimant; after disability has arisen, he or she can request an application form, collecting information and evidence on diagnosed chronic illnesses, health problems and the amount and type of care needed. Medical checks can be imposed in some cases, but are not a necessary component of the decision process. The administration is required to process the claim in 24 working days. The date of claim is relevant because AA starts to be paid only 6 months after the need for care is reported to have arisen.6 The length of an award is decided according to the particular disabling condition and care needs of the case. In fact though receipt of AA is to be regarded as an absorbing state as successful claims result in lifetime awards in the vast majority of cases. For example, according to a study conducted by the Department for Social Security (CHMSO, 1998) on a representative sample of recipients, virtually in all cases (the 99%) AA was awarded for life.7 Receipt of AA can be taken into account in calculating individual contributions towards the means-tested provision of care services arranged by Local Authorities, the other major component of the social care system for the elderly. However, AA does not reduce entitlement to means-tested benefits in cash.8

AA rejection rates are unsurprisingly higher than for benefits where eligibility can be determined ex ante in a more deterministic fashion. About one in five AA claims is rejected (DWP, 2008). It is then possible to appeal against the administrative rejection. In fact, more than 40% of appeals following a rejected claim are subsequently re-assessed in favour of the appealing claimant (NAO, 2003), even in the absence of additional supporting evidence.

The analysis that follows is carried out using the eighteen waves of British Household Panel Survey (BHPS) data, presented in the following section 2. A binary model of entry into AA is estimated in section 3 to assess on the one hand whether AA receipt is promptly responsive to changes in the underlying disability status, and on the other hand the role of other personal characteristics, irrelevant to eligibility, in confounding the benefit assignment mechanism. Semi-parametric methods are instead used in section 4 to gauge the consequences of targeting failures, i.e. how AA receipt affects later perceived financial wellbeing. Section 5 discusses the results.

Section snippets

Data

The BHPS has collected yearly information on individuals living in private households in Britain since 1991.9 The BHPS offers a unique source of

Empirical method and estimation

Notation is defined as follows: ri,t is a binary indicator for first entry of individual i into AA in time t, where ri,t = 0 if i has not yet entered AA in t and ri,t = 1 if i enters AA in t. Observations from periods after first receipt are discarded because receipt of AA is an absorbing state.20 For each individual i we observe a vector of time varying disability indicators ei,t, reflecting disability/possible eligibility to AA, and a vector xi,t of other

Empirical method and specification

What are the consequences of a missed or postponed claim? While receipt of AA obviously implies an increase in available financial resources, it is interesting to see whether it also results in an improvement of recipients’ financial situation actually perceived.25

Discussion

This paper has offered new evidence on participation in AA, the UK extra-cost disability benefit for older people. The availability of suitable panel data has allowed to study AA receipt accounting for dynamic aspects of the process. Information about health and disability prior to first receipt has allowed both the investigation of targeting, in terms of responsiveness to earlier disability onset, and the circumvention of the potential justification bias that the lack of disability indicators

Acknowledgments

The author has benefited from the helpful comments of two anonymous referees, Stephen Pudney, Ruth Hancock, Monica Hernandez, Marcello Morciano, Mike Brewer, David Bell, Richard Berthoud, the trustees and referees of the Nuffield Foundation and participants of the 2009 BHPS Research Conference, the 2011 New Directions in Welfare Conference and of a seminar held at the University of York. The BHPS data were originally collected by the ESRC Research Centre on Micro-social Change at the University

References (55)

  • P. Banks et al.

    Transparent or opaque? Disabled people in Scotland describe their experience of applying for disability living allowance

    Journal of Social Work

    (2005)
  • H. Benitez-Silva et al.

    How large is the bias in self reported disability?

    Journal of Applied Econometrics

    (2004)
  • Berthoud, R., 2009. The impact of disability benefits. A feasibility study. DWP Working Paper No....
  • Bethoud, R., Hancock, R., 2008. Disability benefits and paying for care. ISER Working Paper 2008-40, Institute for...
  • K. Bolin et al.

    Informal and formal care among single- living elderly in Europe

    Health Economics

    (2008)
  • J. Bound

    The Health and Earnings of Rejected Disability Insurance Applicants

    American Economic Review

    (1989)
  • J. Bound

    Self-Reported versus Objective Measures of Health in Retirement Models

    Journal of Human Resources

    (1991)
  • J. Bound et al.

    Economic analysis of transfer programs targeted on people with disabilities

  • Commission on Funding of Care and Support

    Fairer Care Funding

    (2011)
  • Controller of Her Majesty's Stationary Office

    A Study of Disability Living Allowance and Attendance Allowance Awards

    (1998)
  • Controller of Her Majesty's Stationary Office

    Shaping the Future of Care Together

    (2009)
  • J. Cullinan et al.

    Estimating the extra-cost of living for people with disabilities

    Health Economics

    (2011)
  • Currie, J., 2004. The take-up of social benefits. IZA Discussion Paper No. 103,...
  • M. Daly et al.

    The reach of disability benefits: an examination of the Disability Living Allowance

    Journal of Social Welfare and Family Law

    (1996)
  • R. Disney et al.

    Why are there so many long term sick in Britain?

    Economic Journal

    (1991)
  • Department for Work and Pensions

    Attendance allowance quarterly statistics: February 2005

    Online Statistical Bulletin

    (2008)
  • B. Gannon

    The influence of economic incentives on reported disability status

    Health Economics

    (2009)
  • Cited by (6)

    A previous working paper version of this work appeared as: Zantomio, F., 2010. Older people's participation in disability benefits: targeting, timing and financial wellbeing. ISER Working Paper 2010-23, Institute for Social and Economic Research, University of Essex, Colchester.

    View full text