Simulating the Reform of Means‐tested Benefits with Endogenous Take‐up and Claim Costs*
| Published date | 01 April 2006 |
| DOI | http://doi.org/10.1111/j.1468-0084.2006.00156.x |
| Date | 01 April 2006 |
| Author | Ruth Hancock,Holly Sutherland,Stephen Pudney |
PDFlib PLOP: PDF Linearization, Optimization, Protection
Page inserted by evaluation version
www.pdflib.com – sales@pdflib.com
Simulating the Reform of Means-tested
Benefits with Endogenous Take-up and
Claim Costs*
Stephen Pudney,Ruth Hancockand
Holly Sutherland§
Institute for Social and Economic Research, University of Essex, Essex, Colchester,
UK (e-mail: spudney@essex.ac.uk)
Department of Health and Human Sciences, University of Essex, Essex, Colchester,
UK (e-mail: rhancock@essex.ac.uk)
§Microsimulation Unit, Institute for Social and Economic Research, University of
Essex, Essex, Colchester, UK (e-mail: hollys@essex.ac.uk)
Abstract
Microsimulation models are commonly used to examine the distributional
impact of reforms of the means-tested benefit system. Take-up behaviour is
related to the level of entitlement, so reform may induce changes in take-up.
We develop a stochastic simulation method and apply it to a probit model of
‘income support’ take-up by the UK pensioners. The method allows us to
adjust net income for the welfare losses because of tangible or intangible claim
costs. Endogenous take-up and claim costs both have an important impact on
the simulated outcomes of the policy reform.
*We are grateful to two anonymous referees whose comments improved the exposition and helped
us spot an error in the first draft. Participants in the NATSEM International Conference on Micro-
simulation, Canberra, December 2003, made valuable comments. Geraldine Barker and Monica
Hernandez provided valuable assistance. The Economic and Social Research Council provided
financial support for this research, under project grant R000239105. Material from the Family
Resources Survey, made available by the Department for Work and Pensions via the UK Data
Archive, has been used with permission. All responsibility for the analysis and interpretation of the
data presented here lies with the authors.
JEL Classification numbers: C25, D63, H55, I32, I38.
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 68, 2 (2006) 0305-9049
135
Blackwell Publishing Ltd, 2006. Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK
and 350 Main Street, Malden, MA 02148, USA.
I. Introduction
Means-testing is an obvious way of focusing welfare spending on those most
in need and therefore of controlling the burden on public finances. One of the
drawbacks of means-testing is that people who are entitled to receive welfare
benefits may not come forward to claim them and there is evidence that non-
take-up of welfare benefits can be widespread (Department for Work and
Pensions, 2004). The take-up issue has been the subject of a great deal of
applied research aimed at estimating the incidence and magnitude of claim
costs, informational problems and attitudinal barriers to claiming entitlements.
The studies by Ashenfelter (1983), Moffitt (1983), Blundell, Fry and Walker
(1988), Duclos (1995), Blank and Ruggles (1996), Bollinger and David
(1997), Kim and Mergoupis (1997) and Keane and Moffitt (1998) are
examples of the development of this literature.
The potential endogeneity of take-up is not routinely accounted for in the
simulation of policies or policy reforms. Frequently, simulations assume
100% take-up (Clark, 2002) and this tends to over-emphasize the efficiency
of means-tested benefits in meeting the needs of the poor. In some countries
the assumption of complete take-up gives rise to estimates of benefit
entitlement that are considerably larger than the receipt that is known
from administrative data (Sutherland, 2001). A simple solution is to tie
entitlement to observed receipt on a case-by-case basis. This has the
advantage of producing results that are easily reconciled with other statistics,
but it severely limits the scope of policy changes that the model can
simulate. An alternative simple approach is to assume a fixed probability of
taking up, which depends on personal characteristics but is not affected by
changes in the benefit system (Hancock and Sutherland, 1997; Atkinson and
Sutherland, 1998). This does capture the advantage of non-means-tested
benefits to the extent that they are subject to fewer take-up problems. But it
only provides an approximate guide to the cost and distributional effects of
reforms that extend means-testing to new groups or that change the design
or level of generosity of the means-tested benefit.
In this paper, we demonstrate the importance of modelling take-up as an
endogenous component of a policy simulation model. Take-up behaviour
changes in response to changes in benefit entitlements and prebenefit income.
This affects the aggregate cost, the broad distributional effects and the impact
of reforms on people in particular situations.
II. Take-up as a factor in policy reform
Our analysis is based on the UK, where there are three main types of state
benefits for pensioners: the flat-rate basic state pension, an earnings-related
136 Bulletin
Blackwell Publishing Ltd 2006
To continue reading
Request your trial