Understanding the Growth in Welfare Benefit Receipt in Britain: A Review of the Evidence

Walker, Robert; Howard, Marilyn; Maguire, Sue; Youngs, Rachel Working Paper Understanding the Growth in Welfare Benefit Receipt in Britain: A Review of the Evidence New Zealand Treasury Working Paper, No. 00/23 Provided in Cooperation with: The Treasury, New Zealand Government Suggested Citation: Walker, Robert; Howard, Marilyn; Maguire, Sue; Youngs, Rachel (2000) : Understanding the Growth in Welfare Benefit Receipt in Britain: A Review of the Evidence, New Zealand Treasury Working Paper, No. 00/23, New Zealand Government, The Treasury, Wellington This Version is available at: https://hdl.handle.net/10419/205438 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Disclaimer: The views expressed are those of the author(s) and do not necessarily reflect the views of the New Zealand Treasury. The Treasury takes no responsibility for any errors or omissions in, or for the correctness of, the information contained in these working papers. TREASURY WORKING PAPER 00/23 Understanding the Growth in Welfare Benefit Receipt in Britain: A Review of the Evidence Robert Walker, Marilyn Howard, Sue Maguire, Rachel Youngs ABSTRACT This paper summarises the conclusions from a more extensive report commissioned by the Treasury and the Ministry of Social Policy to provide an explanation of growth in welfare benefit receipt in Britain in the period 1971 to 1997. The study is structured around a simple heuristic model of four sets of influences, ‘drivers’, the interaction of which helps to explain trends in claimant caseloads: the economy, demography, welfare institutions and belief systems. Four sets of claimants are considered: unemployed people, disabled people, retirement pensioners and children and families. It assembles a widely dispersed literature, in the first comprehensive review of the evidence on this issue. The conclusions reported here are supported in the full report by a wide range of data and analyses. The full report has been extensively edited, updated, reformatted and published in September 2000 by the Policy Press at the University of Bristol, under the title The Making of A Welfare Class? Benefit receipt in Britain, by Robert Walker with Marilyn Howard. Details of this publication may be obtained through the link to the Policy Press website provided below. The study concludes that, while caseloads increased across the four domains, they did so for very different reasons. While the process of de-industrialisation provided an important back-cloth to all the changes, it was only directly implicated as a major influence in the growth of unemployment related benefits. The upward trends in disability benefits were principally a reflection of the greater social awareness of the personal costs of disability, while the growth in benefits pertaining to the family was very largely a response to changing social attitudes and sexual behaviour. Demography, notably increased longevity, explains the observed growth in pensioner caseloads although the balance between contributory and means-tested pensions was the result of foresighted policy decisions made between the 1940s and 1970s. CRSP 7-2 UNDERSTANDING THE GROWTH IN WELFARE BENEFIT RECEIPT IN BRITAIN: A REVIEW OF THE EVIDENCE CRSP372 Robert Walker Marilyn Howard Sue Maguire Rachel Youngs June 1999 Centre for Research in Social Policy Loughborough University Loughborough, LE11 3TU United Kingdom Tel: 44 (0) 1509 223618 Fax: 44 (0) 1509 213409 [email protected] CRSP 7-3 EXECUTIVE SUMMARY 0.1 Introduction New Zealand is experiencing a long-term growth in the number of people claiming welfare benefits. The New Zealand Treasury and Ministry of Social Policy have invited CRSP to undertake a review of developments in the United Kingdom. The review has three objectives: 1. to provide a coherent analytic account of reasons for the upward trend in welfare receipt; 2. to offer empirical documentation to support the above account; 3. to assess the impact of current and past policy on these trends. British usage is adopted throughout. The term social security is used to cover means-tested benefits (social assistance in international parlance), contributory based benefits (social security) and non-contributory benefits. ‘Claimants’, ‘benefit recipients’ and ‘welfare recipients’ are used interchangeably. The literature is widely dispersed and has not been reviewed in detail before to address the issues at hand. It has therefore been necessary to engage in a process of assembly rather than simply conduct a review of established facts. The report is structured a round a simple heuristic model of four sets of influences, ‘drivers’, the interaction of which helps to explain trends in claimant caseloads: the economy, demography, welfare institutions and belief systems. Four sets of claimants are considered: unemployed people, disabled people, retirement pensioners and children and families. The resulting relationships are set out in Figures 1 to 4. 0.2 The British Context Britain experienced three major recessions between 1971 and 1997 and a marked trend towards de-industrialisation with declines in manufacturing employment and growth in the service sector. After seven years of economic recovery unemployment is still above levels in the early 1970s. Economic participation rates of women have risen while those of men are near a fifty year low. There has been an increasing polarisation between ‘work rich’ families with more than one earner and ‘work poor’ families with no one in paid employment, and in wages - those of low skilled workers have not increased in real terms for over a decade. There have been significant increases in lone parenthood, reflecting upward trends in cohabitation and divorce (now ended), in the number of disabled people living in families without work and in the number of pensioners. Casual employment has increased and unemployment remains high. Child poverty has tripled since 1979. CRSP 7-4 The British social security system is highly centralised and uniform. Virtually all benefits are administered by the Benefits Agency, an executive agency of the national Department of Social Security which has policy responsibility for all benefits. Benefit expenditure has risen eight-fold in real terms since 1949/50, and by about quarter during the 1990s. Social security expenditure equates to 13 per cent of GDP and is predominantly funded through direct taxation. National Insurance contributions maintain the fiction of funding. The main contributory benefits are Retirement and Widows’ Pensions, Incapacity Benefit and the first six months of Jobseeker’s Allowance (for the unemployed). Contributory benefits are comparatively low by international standards. About 6 million households - a little under a fifth - receive Income Support, the principal means-tested benefit. Income Support (Jobseeker's Allowance for unemployed claimants) is universally available on the basis of need. Two important means-tested benefits, Housing and Council Tax Benefits, are administered on an agency basis by local government. Family Credit1 is a means-tested benefit payable to working people with dependent children. The main non-contributory benefits are Child Benefit, a universal benefit payable for dependent children, Invalid Care Allowance paid to carers in lieu of paid employment, Disability Living Allowance to meet the extra costs of disability, and Severe Disablement Allowance for disabled people who do not qualify for Incapacity Benefit. The major policy goal has always been the alleviation of poverty while ensuring perverse incentives not to work or save have been contained. The ideas of the radical right, especially those of US thinkers, were particularly influential in policy circles in the late 1980s and 1990s and the concept of benefit dependency was first used by a British minister in 1987. In the early 1990s a political consensus began to develop that social security had to be adapted to meet the needs of a flexible labour market, that employment provided the best defence against poverty and that social security polices should be active rather than passive. The new Labour Government has carried forward the emphasis on employment and personal responsibility in a series of welfare to work schemes, the New Deals. 0.3 Unemployment 0.3.1 The economy Unemployment in Britain has risen dramatically since 1971 and, although levels have fallen steadily over the last five years, unemployment remains 1 Family Credit was replaced by Working Families’ Tax Credit in 1999, a more generous scheme for low income working families delivered through the tax system. CRSP 7-5 higher than in 1980. In order to understand the reasons, it is necessary first to take account of changes in the economy and in the goals of public policy. High inflation triggered by the oil crises of the 1970s led to tighter control of public spending and, with the election of the Thatcher government in 1979, an explicit belief that the control of inflation was more important than a commitment to full employment. Unemployment was used as a policy instrument in the fight against inflation. The commitment to prudent financial planning is also a feature of the new Labour government. Rising unemployment has coexisted with increasing employment over most of the period since 1971. A decline in manufacturing was more than offset by increased employment in the service sector, providing a mix of low and high paid jobs that recently have been increasingly of a part-time and/or short-term nature. Educational standards have risen, financial returns to qualifications have increased and wage dispersal has grown; the earnings of the low paid have remained unchanged in real terms, while median earnings have risen by a third and higher earnings by more. People who become unemployed are disproportionately male, young, poorly qualified and with little experience or a poor work record. The same groups predominate among the stock of unemployed although older people and members of ethnic minorities, who find it more difficult to return to work, are over represented. On the other hand, young people seem more able to find work although they are also more likely to suffer repeated spells of unemployment. Unemployment has risen most among the unskilled and poorly educated. The average duration of unemployment has also increased since the 1970s, particularly affecting older workers, those with health problems, tenants (often used as an index of material deprivation) and, possibly, minority ethnic groups. So, too, has the ‘no pay low pay’ phenomenon of people experiencing repeat spells of unemployment. 0.3.2 Demography The trend towards higher unemployment was exacerbated by demographic changes including an increasing potential work-force, influxes of young workers in the late 1970s and early 1980s and increased employment participation - notably among mothers of young children. Perhaps as a reflection of labour market pressures among others, male employment rates fell substantially as, to a lesser extent, did those of lone parents. Older men were the worst affected with rates of economic inactivity quadrupling among those aged 5059 and rising to 57 per cent among 60 to 64 year olds. Even so, unemployment rates remained highest among the young and did not fall so much (and, indeed actually rose for 16 and 17 year olds) during the economic recovery of the 1990s. Nor did unemployment fall proportionately among ethnic minorities. CRSP 7-6 0.3.3 Welfare institutions Labour market and social security policies also changed radically between 1971 and 1997 in ways that were often explicitly designed to influence the size of the unemployed caseload. The 1970s witnessed the embryonic development of schemes to provide training and work experience which were expanded on a large scale in the first half the 1980s to provide surrogate employment and reduce the claimant count (by over 500,000). In 1973 employment services were separated, physically and administratively, from the provision of benefits, a policy that was reversed in 1997 with the introduction of Jobseeker’s Allowance. By the late 1980s the policy emphasis had changed away from job creation to policies to encourage claimants to find work, sometimes through the threat of sanction. Jobseeker’s Allowance, which made benefit receipt conditional upon signing and following a Jobseeker’s Agreement, was the culmination of this policy trend towards increasing compulsion. It has now been complemented by the Labour government’s commitment to offer quality training and work experience which is obligatory for some groups. After a period when unemployed claimants were financially encouraged to leave the labour market, new measures to tighten the eligibility conditions for benefit helped to support the emphasis on flexible job-search while also reducing expenditure and claimant numbers. However, these policies, combined with lengthening spells of unemployment, had the effect of radically shifting the ratio of means-tested to insurance based support from 1:2 in 1971 to 4:1 in 1997 (based on claimant numbers). This process was also fuelled by policies to increase rents for tenants in social housing, thereby increasing the Housing Benefit caseload. Increased means testing focused policy attention on work incentives which, together with evidence of low pay, led to the introduction of in-work benefits for families to adopt the rhetoric of the current government, ‘to make work pay’. The coverage of these schemes was extended in the late 1980s and early 1990s, partly with the intention of encouraging lone parents into employment, partly in recognition of the growth in low pay. However, means-tested, in-work benefits create a ‘poverty trap’ in which claimants gain little return from increased earnings because of the interaction of taxation and the withdrawal of benefit with rising income. Subsequent policy revisions, including the 1999 introduction of a Working Families Tax Credit, were in part responses to this problem which is thought to inflate caseloads and expenditure as a result of inhibiting movement off benefit. 0.3.4 Belief systems The effectiveness of policy in reducing numbers on benefit is likely to be affected by the degree of correspondence between assumptions on which the policy is based and empirical reality. Many of the policies pursued since the late 1980s were premised on the assumption that people need a degree of CRSP 7-7 coercion to find work. The empirical evidence is that few claimants prefer not to work or fail actively to engage in job search. The majority of the long-term unemployed appear not to become dependent as result of receiving benefits, rather they become long-term unemployed because they do not have the attributes that would help them secure employment. Another premise of recent policy concerns the prime importance of financial incentives. Again the evidence is equivocal suggesting that financial disincentives may be less important than uncertainty, risk aversion and the practical constraints imposed by complex systems and poor administration. Although the employment rate of the partners of unemployed claimants receiving benefits on the basis of a household means-test is only a third that of those with employed partners, only a fifth of the difference has been found to result from financial disincentives. On the other hand, there is some evidence of people on in-work benefits not seeking better paid work. In sum, unemployment caseloads have been driven upwards by fundamental economic processes of de-industrialisation occurring in a more open international trading environment and by the priority given to the control of inflation over full employment as a policy goal. Social security and labour market polices have sought to contain the growth in caseloads, initially by diverting people out of the labour market or into training in large numbers. Subsequently tightening of the benefit regulations may have reduced the total claimant count by, perhaps 350,000. It also shifted the balance of provision radically towards means-testing. Concern about work incentives has led to the provision of in-work benefits that have created a new class of benefit recipient. In 1997 the number of employed people receiving means-tested benefits fell only a little short of the number receiving unemployment related benefits. 0.4 Disability 0.4.1 Welfare institutions Disability benefits in Britain can usefully be categorised by their principal objective, either to maintain income, to compensate for additional costs or to supplement wages. In 1971, benefits of the latter two kinds were either rudimentary or non-existent but are now received by over a million and a half people. The numbers of people receiving income maintenance payments on grounds of disability approximately tripled between 1971 and 1997. Many disabled people receive more than one kind of benefit. Most of the growth in income maintenance benefits was due an increase in take up, various rational and perverse incentives to move from other benefits and, possibly less important, an increase in the incidence of disability. Flows onto benefit decreased as take-up rose and policies to encourage people to move from unemployment to disability benefits were given less prominence. Average durations of receipt lengthened due to maturing of the benefits and the slackness of labour demand. CRSP 7-8 The growth in benefits to meet extra costs was ten times that expected and led government to attempt to curtail their expansion in the mid 1990s. Caseloads probably grew primarily because around half of the annual number of new claims were payable for life. The depressed labour market did not help but official concern has recently focused on incorrect assessments by staff, overstatement of need, avoidable disablement and effective welfare rights activity promoted by disability organisations. In contrast to the unanticipated growth in income maintenance and extra cost benefits, the uptake of the wage supplementation benefit introduced in 1992 has been minimal. Receipt of disability benefits is long term, almost three-quarters of all disability claims are for two years or more. Increases in the duration of income maintenance benefits have been linked to a decline in employment and services available to disabled people, to the increased prevalence and duration of certain health conditions, delay in treatment due to increased pressure on medical facilities, and changes - and sometimes delays - in the administration of benefits. The length of claim may also be linked to the pathway onto benefit - whether it is led by the health condition, prompted by the inability to continue work or some complex combination of factors that impinges on the motivation to work. 0.4.2 Belief systems The provision of additional benefits reflects a change in social attitudes to disability that may in turn interact with the beliefs and attitudes and behaviour of claimants, potential claimants and gatekeepers to increase caseloads. It may be more acceptable for people to define themselves as ‘disabled’ and to seek out appropriate benefits. Early retirement on health grounds has become more common, perhaps to the mutual benefit of employers and employees. Fraud may be involved in a seventh of claims but whether this is the outcome of an upward trend cannot be established. Employment agency staff and medical doctors may have encouraged claimants in areas of high employment to turn to the more generous disability benefits although the evidence for this is ambiguous. However, despite successes in promoting benefits, as yet takeup is rarely much in excess of 50 per cent. The response of policy makers to increasing caseloads has been to suspect malpractice and to seek to improve benefit security and avoid double provision. 0.4.3 The Economy There is evidence that industrial change and new production practices have lowered the threshold of employability down the severity scale, excluding people with medium severe impairments who might well have found work in CRSP 7-15 by persuading them to take unsuitable jobs with limited prospects, and by creating high marginal benefit deduction rates that deter lone parents from seeking extra hours or better paid work. More recent polices aimed at encouraging lone parents into work include childcare premiums - the take-up of which has been poor, seemingly because support is restricted to officially registered child care - and access to jobsearch and training facilities. The second was piloted in 1985 by the Conservative Government and re-launched by the incoming Labour Government as the New Deal for Lone Parents in 1997. Lone parents have never been obliged to seek work - when policies were originally evolved, fewer married women worked and it was widely accepted that it was preferable for mothers to stay at home to look after their offspring. Nor, however, did they have access to the resources of the Employment Service. New Deal changed the latter but, to date, the take-up has been much below expectations and proposals have been announced to make at least a work focused interview compulsory for lone parents claiming benefit. Until 1991 official interest in the absent parents of children in lone parent families was restricted to ensuring that they were not resident and thereby illicitly benefiting from social security payments. With the creation of the Child Support Agency, the state sought to enforce maintenance payments and to recover them from benefit. Substantial benefit savings have been achieved - no research exists on the impact on caseloads - but the lone parents affected received no financial benefit. This is to be changed with a reform of the Child Support Act that is currently under consideration. Child Benefit which is received on behalf of every child was introduced as a tax allowance, to maintain work incentives and as part of a post war, pro-natalist policy. It was converted into a cash benefit in 1979 in response to concerted lobbying to tackle child poverty and to redirect the benefit to mothers as the principal carers. Unlike the family allowances that it replaced and which were available for the second and subsequent child, Child Benefit was payable to every child. This immediately more than doubled the number of recipients, although numbers have since fallen back due to the decline in fertility. Child Benefit has survived several attempts to abolish it, tax it or reintegrate it into the tax system. The Labour government has significantly increased its value since coming to power, both to support its welfare to work agenda and its emerging campaign to eradicate poverty within 20 years. The pressures driving the increased numbers of families and children claiming benefit are complex and partly a necessary corollary of increasing caseloads elsewhere across the social security system. However, the driving force of individual and collective beliefs and behaviour is more evident than in any other domain of social security except disability. The ‘child benefit’ caseload doubled as the result of specific anti-poverty and child care or feminist policy agendas. The growth of lone parenthood was first and foremost the result of CRSP 7-16 changed social and public mores with respect to marriage, gender equality and sexual behaviour. The policy response was first to meet the financial needs created by these new forms of living while having regard to the maintenance of equity. Thereafter, policy moved to constrain growth in expenditure, to remove apparent inequity that might be construed by some to undermine traditional family values, and to encourage employment among lone parents as a route from poverty and social exclusion. The economy is a powerful indirect driver. Lower real wages and higher unemployment, mean that families confronted with the financial responsibility of supporting children are disproportionately brought into the benefit system and perhaps into poverty. CRSP 7-17 7 UNDERSTANDING THE PATTERN OF GROWTH The objective of this review is to provide a coherent analytic account of the upward trend in social security recipients, taken to include contributory, noncontributory and means-tested claimants, to offer evidence and to assess the impact of current and past policy. The very complexity of the British social security system makes it impractical to consider all recipients together as a single set. System is not the right word to describe a collection of at least 39 separate schemes that have disparate objectives, are targeted on different groups of recipients, were implemented at various times over 50 years and have reached different stages of administrative maturity. The system has coherence when viewed at a distance: it is principally concerned with the relief of poverty; means-testing is more widely used than in most Continental European countries and the system is heavily centralised, although benefits are increasingly being delivered by interlocking agencies including private and not for profit concerns. When studied closely it has the form of an amoeba. The approach adopted, therefore, is to take an intermediate focus, directing attention on four categories of recipient: the unemployed, disabled people, pensioners and families with children. Although there is a degree of overlap between categories, in a British context they can serve as different subdomains within global social security policy. This has the advantage of avoiding unnecessary detail and, since the groups have featured in official publications for somewhat over a decade, some coherent information is available. In similar mode, the multifarious factors that conspire to influence the claimant caseloads have been reduced to four sets of ‘drivers’ appertaining to the economy, demography, welfare institutions engaged in the design and delivery of benefits and the belief systems that forge politics and influence individual behaviour. Given that the focus of the analysis is on change and that change is the outcome of process, wherever possible account has been taken of the dynamics of benefit schemes and of the system in which they are embedded. The literature reviewed is widely dispersed, seldom provides long range trend data, and is rarely based on methods of data collection and analytic rigour that would pass an entry test for good science. The commission was not to collect raw data - although much of that has been necessary, nor to engage in secondary analysis and neither to develop and/or test a formal analytic model. Within these constraints, the strategy adopted was to assemble the literature to begin to provide accounts of the main factors driving trends. It has not, however, proved possible to develop detailed and standardised visual models of the processes involved. CRSP 7-18 The aim in this final section is briefly to recap on the processes involved in each domain but, importantly also to draw comparison between them so as to approach a coherent account of the system as a whole. The division of the social security system into the four domains is legitimated by the fact that different drivers are pre-eminent in each case. Perhaps inevitably, in some respects, the economy is the most important driver. Over time it more or less efficiently delivers the resources that society can choose to spend on social security. Moreover its inefficiencies generate many of the needs to which social security provisions are a response. However, its direct influence on the claimants caseloads was most apparent with respect to unemployment. 7.1 Unemployment Over the last almost 30 years Britain has been engaged in a process of deindustrialisation. Triggered, or highlighted, by the oil crises of the 1970s, traditional production industries declined, manufacturing employment fell by 49 per cent and the service sector grew to provide 67 per cent rather than 47 per cent of all jobs. Productivity levels rose and production methods changed. Short term and part time employment increased and job openings that predominately go to women rose relative to those available to men. The skills required of employees changed and increased. Returns to education grew and median and above median wages rose by at least a third in real terms while those at the bottom did not change. Throughout most of this period total employment continued to grow, but many of the jobs created were low waged. Nevertheless, unemployment was driven to unprecedented levels and changed in nature. The proportion of long term unemployment rose and it is probable that ‘churning’, the process by which people move in and out of short-term unemployment, increased. Caseloads of unemployment related benefits naturally rose and the ratio of contributory to means-tested benefits began to change from 2:1 in 1971 to 1:2 in 1997 as people’s entitlement to the former was exhausted and not replenished (and eligibility criteria were altered). Demography did not ease the transition to a postindustrial society. The potential labour force increased by eight per cent. But as important, if not more so, an influx of a large cohort of young people coincided with the recession at the beginning off the 1980s, triggering the growth of long term unemployment comprised of people with little work experience. A disproportionate number of the newly created jobs were taken by women, the result of changing aspirations and priorities, especially among mothers of young children. A further factor may have been the increasing requirement to have two jobs in order for a family to make ends meet. These events drove the policy agenda and initially forged policy responses that were pragmatic. Subsequently a bold experiment was initiated in the practical CRSP 7-19 application of monetarist theory and latterly, an iterative process developed in which policy and events interacted with policy makers sometimes having to respond to the unanticipated consequences of their own policies The stagflation that gripped Britain in the 1970s caused the Labour government to drop its commitment to full employment and to adopt unwelcome policies of fiscal restraint. Then the Thatcher government explicitly sought to use unemployment as a weapon against wage inflation, initially with little regard to the benefits bill; although earnings related Unemployment Benefit was abolished very quickly after the Conservatives assumed power, this yielded few savings in benefit expenditure. Rather, it was primarily an ideological gesture, emphasising the importance that the government attached to financial incentives. Later, repeated paring of contributory benefits did cut their cost (and value) but added to the meanstested caseloads in almost equal measure. Employment advice and benefit administration were separated and small scale training and work experience schemes, started in the 1970s, were rapidly expanded in the early 1980s. Older workers were also encouraged to leave the labour market. Both the latter strategies were used to reduce benefit caseloads. In fact, economic activity among men, especially those over 55, fell steadily throughout the period between 1971 and 1997. Ill-health, it was discovered, was the major reason in the early 1990s, perhaps a reflection of the changing demands of work in a flexible labour market. For reasons of cost, new ministerial leadership, and changing perceptions of the nature of unemployment, job creation measures were cut back from the middle 1980s and replaced by policies that encouraged and cajoled the unemployed to be more enthusiastic and effective in their search for work. The introduction of six monthly interrogatory ‘restart’ interviews with unemployed claimants in 1968 and of Jobseeker’s Allowance with the reintegration of employment and benefit administration and a ‘stricter benefit regime’ in 1996 were all credited with reducing claimant numbers but whether the impact was sustained is unclear. Given that the empirical evidence is that the vast majority of unemployed are positively committed to working and that the long term unemployed are trapped by the disadvantages - low skills, poor education, ill health, age - that they take with them into unemployment, the impact of such schemes is likely to have been marginal. However, the strategy - which emphasises a contract between state and benefit claimant - was taken forward by Labour following its election to power in 1997. Labour added compulsory training and work experience for significant numbers of claimants. Policy increasingly focused, from the mid-1980s onwards, on the need to provide and enhance in-work benefits to address the problem of work incentives. This problem was itself in part a consequence of policies that promoted means testing and in part the result of the increase in low paid jobs. As a measure of ‘success’ the Family Credit caseload rose from 350,000 in 1992 to 725,000 in 1996, partly helped by lone parents being encouraged to CRSP 7-20 enter the labour market. (In 1995/6, though, the proportion of eligible claimants in receipt, another measure of success, was still less than 70 per cent). The numbers in receipt of Housing Benefit also rose due largely to a policy of higher rents. The new Labour government shares the same enthusiasm for in-work benefits although it is to provide them through the tax system. They certainly boost inwork incomes and provide a ‘parachute’ for two earner families’ that lose one job. There is, though, little evidence that wage levels are a major constraint on people taking jobs but some research that suggests that in-work benefits may have “trapped” people in low paid work. Either way, the policy has created a new class of workers whose income derives from a mix of wages and benefits. 7.2 Pensioners If unemployment caseloads were driven by economic change, pension caseloads were determined largely by demography. The influence of the economy was indirect, fixing the resources that workers and employees could accumulate for retirement and constraining the amount that one generation was prepared to pass to another in the form of ‘pay as you go’ pensions. The pensioner population grew by 42 per cent between 1971 and 1997 despite a falling birth rate 60-65 years earlier that should have generated falling caseloads. The principal reason, in volume terms, was an increase in the number of people surviving to retirement age, closely followed by the 20 per cent increase in life expectancy after retirement. The second major feature of pension provision in the last 30 years was the halving of the proportion of pensioners dependent on means testing - the mirror opposite of the change in unemployment related benefits. The principal explanation is the rise in pensioners’ incomes as a result of the development and maturing of occupational pensions since the 1950s together with the introduction of SERPS in 1978 for those without access to a non state pension. To determine which was the more important would require further research since pensions from SERPS (an average £21 per week in 1997) were less than occupational pensions (£933) but directed at those with less other income. A secondary factor was the differential improvement in longevity that benefited men and which meant that less women were widowed and/or were widows for less time. The shift away from means-testing would have been even greater had rents not risen in social housing. The improvement in pensioner incomes was primarily the result of foresight by policy planners, and employers providing occupation pensions, rather than judicious decisions of ordinary citizens. The latter find British pension provisions complex and confusing, planning is limited and often belated. 3 A mean of £91 was recorded for 1996/97 for all pensioner units, inflated in the text to 1998 prices. CRSP 7-21 Indeed, many people are unable to predict their likely incomes and assets in retirement even when they are within five years of retirement age. In fact, in the past little planning was necessary since occupational pensions were a condition of employment and those without one could look forward to increasingly generous pension provision. This has no longer been true since the Thatcher government downgraded SERPS, fearful that it would not be affordable when the dependency ratio worsened in 2020. In its place the government promoted personal pensions, provided by insurance companies, but, despite tax incentives, these proved inappropriate for the poorly paid workers including disproportionate numbers of women, for which SERPS catered. (Mis-selling by the insurance companies also lessened the public’s faith in personal pensions.) In effect, unless the proposals put forward by the current Labour government come to fruition, the recent movement in the balance between contributory and means-tested provision will be reversed. Although occupational pensions contributed significantly to the improvement in pensioners incomes between 1971 and 1997 they also increased income inequality in retirement, carrying over and exaggerating differentials originating in the labour market. Indeed the imprint of personal history is very strong in old age - the decline in incomes observed among older people, and the greater reliance on means-tested supplementation, results more from the improved working conditions and incomes of later generations than from the depreciation of pension income or the depletion of assets in old age. The interaction of occupational pensions with means-testing generated an occupational pension trap that, in 1997, meant that at least 850,000 occupational pensioners were not recouping the full value of their pension contributions. (From the perspective of the state, of course, the same meanstesting generates savings in public expenditure.) Finally, the marked trend towards economic inactivity observed among men over 50 serves to extend the boundary of retirement to younger ages but, because state pensions are only payable at retirement age, this has no effect on Retirement Pension caseloads. Younger generations aspire to retiring at ever younger ages, and occupational pensions often encourage this by offering variable retirement ages, but whether this enthusiasm translates into the planning that earlier generations sought to avoid remains to be seen. CRSP 7-22 7.3 Disability In Britain over the last five years concern has focused on the extent to which it is the economy, perhaps interacting with or generating declining motivations to work, that is responsible for the markedly increased number of people receiving incapacity benefits. Taking a longer time horizon it is clear that the driving factor is a changed level of public awareness and acceptance of disability and a more heightened appreciation of the personal financial cost involved. The economy may well be important, though, in terms of increasing the psychological and physical pressures in the workplace although the evidence in this regard is largely circumstantial. Of the three kinds of disability benefit in Britain, income maintenance, additional costs and wage supplementation, the latter two were either rudimentary or non-existent in 1971. In 1997 they were received by over 1.2 million people. The number of people claiming income maintenance benefits tripled between 1971 and 1997. Prior to the 1970s there had been no social recognition of the impact of longterm sickness or provision for it within the social security system. Disability, particularly that resulting from congenital conditions, was difficult to fit within a social insurance framework and had been largely overlooked. Growing social awareness, in part stimulated by the emergence of disability rights groups, and a policy concern in the early 1980s to reduce reliance on means-testing led to the creation of a new kind of non-contributory, non-means-tested benefit. Reflecting social attitudes and understanding of the day they were directed first exclusively on physical impairment. Only later, as part of the expansion of coverage, were other forms of disability recognised. The 1980s policy agendas developed in response to the rise in unemployment and a political preference for the approaches of business, embraced concerns about financial disincentives and introduced reforms to reduce administration costs and pressures on business. Concerns to tackle sex inequality were also in evidence, sometimes forced on the British government against its inclinations. The European Court insisted that Invalids Care Allowance – a benefit that acknowledges earnings lost by carers – should be available to women to care for their partners. Most importantly though, decisions were deferred until the outcome of large-scale empirical studies of the nature and prevalence of disability. The research results pointed to the need to extend coverage to groups previously excluded such as those with learning difficulties. This was done in the 1990s despite great pressures to restrain expenditure, which, in practice, meant tightening eligibility criteria and greater means testing. The finding that many disabled people expressed the wish to work chimed with the emerging policy perspective with respect to unemployment that paid work provided the best route to self sufficiency and that this could be facilitated CRSP 7-23 with in-work benefits. This led to the introduction of Disability Working Allowance in 1992. With the exception of Disability Working Allowance the caseloads for the newly introduced benefits much exceeded expectations. Because, many of the benefits are payable for life subject to person’s circumstances remaining unchanged, it was inevitable that caseload numbers would rise and durations lengthen as long-term recipients accumulated in the caseload. The growth in the caseloads of established income maintenance benefits for disabled people grew initially from improvements in take-up, perhaps because social attitudes meant that people were more ready to recognise and accept their own status as disabled people. As take up increased, caseload growth arose less from a flow of new cases and more from extending periods of receipt. This reflected the stage in the maturation of benefits designed to meet the needs of long-term conditions but raised concerns, at a time of high unemployment and fiscal pressure, that disability benefits were becoming a safe haven for the unemployed. In fact the empirical evidence on a link between unemployment and the receipt of disability is equivocal. Economic inactivity is rising quickly among older men. Whether this is de facto early retirement or unemployment, ill-health is often involved which puts upward pressure on disability benefit caseloads. Caseloads for disability benefits are larger in areas of high unemployment but do not appear to be susceptible to cyclical economic fluctuations. It seems likely that disabled people are marginalised by employers who enjoy a surplus of labour supply rather than themselves being deterred from wanting work by a deficiency of vacancies. In such circumstances it might be that gatekeepers, medical doctors and Employment Service staff, direct people with health problems to disability rather than unemployment related benefits. The evidence of explicit discrimination by employers is slight. This complex interaction between beliefs, behaviour and the labour market presented dilemmas for policy. Malpractice and abuse was suspected, but large-scale reviews of claims did not lead to sizeable reductions in caseload. Moreover, take-up rates remain at only fifty per cent. The current policy response is to emphasise employability, what a person can do rather than what they cannot do. New tests of incapacity are being introduced and, under the New Deal for Disabled People, specialist counsellors are being deployed to assist disabled people to enter the labour market. Demography has been a less important driver of disability benefit caseloads than the change in attitudes. However, the population did age over the period 1971 to 1997 and this was important in the growth of Severe Disablement Allowance which has a caseload dominated by people who are in their seventies or older. The closure of long-stay hospitals, the increased prevalence of older people living alone and medical advances in keeping people alive for longer have all added upward pressure to caseloads. Finally, CRSP 7-24 there is also growing evidence that growing economic inequalities may be generating greater health needs among poor people and in poor areas. 7.4 Children and Families The processes that drove the growth in unemployment and disability benefit caseloads necessarily affected the circumstances and interests of families with children. This is inevitable, of course, because many disabled and unemployed people have children and some children are disabled and receive special benefits as a consequence. It reflects the fact that the domains of social security used in Britain are not mutually exclusive. Nevertheless, there are separate benefits payable to children in their own right and as dependants, and distinct issues that arise from their receipt of other benefits. The principal driving force behind caseload growth in this domain between 1971 and 1997 was the change in individual and collective beliefs about ways of living and the nature and role of the family. The economy, too, was important, making it increasingly difficult for families to be sustained by one income and increasing the risk of unemployment and worklessness and of the associated poverty. Indeed, one striking social development that occurred during the period was the replacement of pensioner households by families with children in the bottom quintile of the equivalised income distribution. In terms simply of caseloads, the biggest growth occurred with the transformation of the family tax allowance into a cash benefit, Child Benefit, in 1979 and at the same time extending coverage to the first-born child. This more than doubled the number of children for whom benefit was received and reflected a re-focusing of policy objectives. Family allowances were originally introduced to enhance work incentives with the creation of the first national means-tested safety net. The financial saving achieved by limiting the allowance to second and subsequent children was also consistent with the pronatalist policies of the time. Work incentives were still an important consideration in the introduction of Child Benefit but the prime objectives were to tackle child poverty, particularly among working families, and to redirect control of the income to the principal carer, usually the mother. Declining fertility has since caused the Child Benefit caseload to fall by almost 10 per cent. The other area of major caseload growth was a reflection of the more than twofold growth in lone parenthood. Over one in five families with dependent children was headed by a lone parent in 1997. Originally driven by separation and, following divorce law legislation in 1971 by divorce, the proportion of never married lone mothers, either separated from a partner or never having lived with one subsequently increased. Moreover, at a time in the 1980s and early 1990s when economic activity rates among married mothers were rising, the proportion of lone mothers in work fell and the proportion claiming meanstested Income Support rose. In 1997 lone parents comprised one in five Income Support claimants.