Skip to main content

Technical Solutions to Social Problems?: Some Implications of a Computer-Based Welfare Benefits Information System*

  • Michael Adler and David Du Feu

Technical experts frequently prescribe simple technical ‘solutions’ to complex social problems. In this article, the authors describe an experimental project which was designed to increase the take-up of means-tested benefits. The project incorporated a computer-based information system which provided individuals who supplied details of their financial and other household circumstances with personalized information about their entitlement to a wide range of benefits. Although the project was a technical success, it made little impact on take-up rates. In attempting to explain why this should have been so, the authors point to the complexity of the problems involved and advocate the need to examine and simplify existing administrative procedures at the same time as investigating the potential of computerization, rather than assuming that a technical device can of itself side-step the problems. They also point out that administrative simplification is itself no easy task and is fraught with political and organizational problems.

Hide All

1 Computers and Privacy, Home Office, Cmnd 6353, HMSO, London, 1975.

2 These and other computer applications in the public sector are listed in Computers: Safeguards for Privacy, Home Office, Cmnd 6354, HMSO, London, 1975.

3 Rule, James B., Private Lives and Public Surveillance, Allen Lane, London, 1973.

4 Under the Comprehensive Employment and Training Act 1973 the United States government is committed to ‘establish and carry out a nationwide computerised job bank and matching program’, and it is anticipated that all government job centres will provide such a service by 1980.

5 For a prototype scheme see Computer Based Housing Letttngs, City Treasurer's Department, Newcastle-upon-Tyne, October 1973. The Greater London Council won the British Computer Society's John Player award (for the computer application which has made the biggest contribution to society) in 1976 for such a scheme.

6 Butler, A. M. and Dowsey, M. W., Interactive Careers Guidance System: Final Report of the IBM/Cheshire Project, IBM (UK) Scientific Centre, Peterlee, 04 1975.

7 A means-tested benefit is one for which entitlement is based in part on an assessment of household income.

8 Both the figures are rough approximations based on statistics for the most important means-tested benefits.

9 For example, see Kincaid, J. C., Poverty and Equality in Britain, Penguin Books, Harmondsworth, 1973.

10 Finland and Italy are two European countries which have recently introduced computerized systems of this type.

11 Hansard, House of Commons, 7 April 1975, cols 344–5.

12 Taylor-Gooby, Peter, ‘More Welfare for Less Cost: The Logic of Combined Assessment’, Poverty, 34 (1976), 1620. National Consumer Council, Means Tested Benefits, Discussion Paper 3, HMSO, London, 1976, which was published after this paper was written, also contains a critical assessment of means-tested benefits. This paper looks at means-testing from a historical and comparative perspective and examines the characteristics of meanstested schemes and proposals for reform.

13 As a household's gross income rises, income tax and national insurance contributions increase and the household progressively loses its entitlement to means-tested benefits. People caught in the poverty trap require a substantial increase in take-home pay (which may be as much as £10–15 per week) if they are to be significantly better off in netterms after an increase in earnings. This calculation assumes that the household is claiming all the benefits to which it is entitled. The Department of Health and Social Security has developed a computer model to show how taxation and entitlement to benefits affect households of different composition at different levels of earnings. The poverty trap continues to exist despite attempts to stagger cut-off points and cut-off rates because of the large number of separate means-tested schemes now in existence.

14 According to figures published in Supplementary Benefits Commission Annual Report 1975 it would appear that the cost of administration per pound of benefit dispensed is six to seven times greater for supplementary benefit than for national insurance benefits. This difference is, however, only partially due to the fact that supplementary benefit is means-tested; it is also due to the substantial element of discretion in the supplementary benefit scheme.

15 Lister, Ruth, Take-up of Means-Tested Benefits, CPAG, Poverty Pamphlet 18, London, 1974; and Lister, Ruth, ‘Take-up: The Same Old Story’, Poverty, 34 (1976), 38.

16 Lister, Take-up of Means-Tested Benefits.

18 For an example, see Jenners, Roger, ‘Welfare Benefits: The Experience of One Family’, Poverty, 34 (1976), 8–1.

19 Lister, , Take-up of Means-Tested Benefits.

20 A sub-department of the Scottish Office responsible for the provision of social work services in Scotland.

21 Adler, M. and Feu, D. du, ‘Welfare Benefits Project: Final Report’, Bulletin of Charing House for Local Authority Social Services Research, 2 (1974), 151.

22 Adler, M. and Feu, D. du, A Computer Based Welfare Benefits Information System: The Inverclyde Project, IBM (UK) Scientific Centre, Peterlee, 1975.

23 Available from the Department of Social Administration, University of Edinburgh.

24 Feu, D. du, A Computer Based Welfare Benefits Information System: System Documentation, IBM (UK) Scientific Centre, Peterlee, 1977.

25 The only scheme which contains a large element of discretion is the supplementary benefit scheme. The program does assess entitlement to dietary and heating additions, since, although these are discretionary in theory, entitlement is in fact determined through the application of fairly simple rules. In all other cases, the rules are either too complex or the reliance on personal judgement too great to enable the computer to simulate the decision-making process with any degree of reliability. If the supplementary benefit system can be simplified, as has been advocated by its chairman (see Donnison, David, ‘Supplementary Benefit: Dilemmas and Priorities’, Journal of Social Tolicy, 5: 4 [1976], 337–58), the scope for computerized assessment will be greater (see Hoshino, George, ‘Computerised Supplementary Benefits’, Social Work, 27: 3 [1970], 21–3).

26 It is very difficult to estimate the cost of providing and running a computerized information system, since this depends on how the system is implemented and on the costing procedures of the local authority providing the service. However, some guide-lines on cost, detailing the various factors that have to be considered in trying to make an estimate, are available from the Department of Social Administration, University of Edinburgh.

27 In multiple-assessment schemes separate assessments of entitlement are made by the departments which administer benefits, on the basis of the information recorded on a single multi-purpose claim form. In combined-assessment schemes assessment of entitlement to a number of benefits is made by a single combined-assessment unit.

28 For example, assessment of both benefits from a single form produced a 277 per centincrease in rate rebate claims among council tenants in Manchester in 1974–5.

29 Taylor-Gooby, op. cit.

30 For a very illuminating discussion of such problems see Habermas, Jürgen, Towards a Rational Society, Heinemann, London, 1971, chs 4–6.

* An earlier draft of this article was presented to ‘Computing and People '76’, a conference on human factors in computing, held from 20 to 22 December 1976 at Leicester Polytechnic (proceedings to appear).

Lecturer, Department of Social Administration, University of Edinburgh.

SSRC Fellow, Department of Social Administration, University of Edinburgh.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

Journal of Social Policy
  • ISSN: 0047-2794
  • EISSN: 1469-7823
  • URL: /core/journals/journal-of-social-policy
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Altmetric attention score

Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed