Work over welfare:
lessons from America

Work over welfare: lessons from America?

According to new media-received wisdom, the controversial 1996 Clinton welfare reforms – despite outright opposition from liberal America – have been shown, in practice, to be an unqualified success. Ten years on, it is argued, employment rates are up and child poverty is down. If this is true, why are we not prepared to bite the bullet and accept that the compulsory work-first approach is needed to reduce UK worklessness? Could US-style ‘tough love’ be the answer to UK child poverty? Have we anything to teach the US?

The Clinton reforms

It has been more than ten years since the introduction of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996. Bill Clinton, who had proposed time limits to social assistance along with work programmes for those who reached the time limits, was eventually strong-armed into agreeing to a much more radical piece of legislation that ended entitlements to assistance for poor families with children and allowed states to cut off and time-limit assistance to families. The centrepiece of the reforms was the replacement of Aid to Families with Dependent Children (AFDC) with the Temporary Assistance for Needy Families (TANF) programme. For ‘families’, read ‘one-parent families’, as they are the main recipients of AFDC/TANF. As Ron Haskins, a key staffer on the Republican House Ways and Means Committee and one of the architects of the 1996 reforms, says: ‘The pattern is clear: earnings up, welfare down. This is the very definition of reducing welfare dependency.1 Clinton, whilst initially reluctant to sign the Bill, was eventually persuaded to do so.

Lessons for the UK?

Does the US experience have any resonance here? Recent interest in the so-called Clinton welfare reforms, seems so far to be media led, rather as interest in Charles Murray and the underclass theorists was in the 1980s. Undoubtedly, belief in the moral and social benefits of work bind US and UK social policy thinking historically, dating back to our export of Elizabethan poor law and returning to us reinvigorated through present day neo-liberal policies. Nevertheless, we should be careful both about comparing policy outcomes and attempting to uproot policy and practice from one country to another with quite different cultural and social histories.2 Clearly, in order to assess the success or failure of the 1996 US welfare reforms, we also need to analyse the context of the reforms, what they sought to achieve, the economic and social background to the reforms and the comparability of the results.

American welfare benefits

The US welfare system is not particularly effective at lifting people out of relative income poverty. In most OECD countries the combined effect of taxes and benefits is to lift, on average, more than half of the population out of poverty.3 But in the US the effect is only to lift out one-quarter of those below the threshold – compared with more than two-thirds in Denmark and more than one-half in the UK.

Social security

The focus on recent welfare reform perhaps risks misrepresenting the US welfare system. Whilst social assistance in the form of TANF is available almost exclusively to poor, lone parents, this is not the sole form of social protection in the US. There is also the system of insurance-based benefits for old age, survivors and disability, commonly referred to in the US as ‘social security’. Introduced as a centrepiece of the 1935 Social Security Act – Franklin Delano Roosevelt’s ‘New Deal’ package – it remains a popular feature of US income maintenance and has been defended successfully against numerous attempts to cut it back, due mainly to its popularity. Observers might note with some irony that whilst in Britain the term ‘welfare’ is now commonly substituted for what was once referred to as ‘social security’, in the US, social security remains the reference point for this most popular form of insurance-based, income maintenance. An unemployment insurance system also provides support for up to 26 weeks for unemployed workers who meet specified requirements. Roosevelt’s New Deal also introduced the idea that to tackle poverty there would need to be continual policy experimentation and development – a piloting approach arguably adopted only recently in the UK.

Welfare

A supplemental security income (SSI) provides means-tested assistance for elderly and disabled people. But the key policy instrument developed to protect families with children was AFDC, a scheme introduced in 1935 and now replaced by TANF – commonly referred to as ‘welfare’. Intended originally to prevent poor lone mothers from having their children taken into care and to provide support without being in the labour market, it unintentionally became a piecemeal social assistance safety net mainly for one-parent families. Couples can claim, but income tests are generally too low for even unemployed families to qualify. There is no further means-tested income support outside of these programmes. The US system is split not only between social insurance and means-tested programmes, but also between federal and state-level legislation. There is massive variation within the US in both the type of schemes implemented and the amounts paid in assistance. Therefore, the results of policy interventions vary state by state.

Race and gender

There is also a race and gender dimension to these benefits. The social security entitlement based on a family wage model means those least likely to qualify include lone parents, widows and disabled people, hence the reliance of many lone parents on welfare and the high risk of poverty this entails. Numbers claiming AFDC increased significantly in the 1970s and 1980s, a ‘welfare explosion’ often credited to the civil and welfare rights movements of the 1960s which enfranchised many black people in the welfare state for the first time.4 Significant cuts to eligibility and benefits were made in the Reagan era in the run-up to the PRWORA, and the demonisation of the (often black) ‘welfare mom’ played a key role in these developments.

Work first

From the 1960s onwards, work requirements began to be introduced in some states. Starting in 1988, all states were required to operate work-related services for AFDC families with children aged three and over. Federal matching funds were made available to support childcare. In the 1980s and 1990s, there were early indications that more sustained impacts were likely from programmes emphasising a range of services including job search, work experience and education and training. Nevertheless, evidence from random assignment trials showed schemes relying chiefly on one service such as job search or basic education alone showed weak impacts. This evidence was used to dismiss all education and training schemes in favour of a ‘work-first’ approach with rapid labour market attachment and strong job search requirements,5 despite evidence of mostly very modest rises in employment rates due to work-first programmes. Nevertheless, it was the cheapest and most politically expedient conclusion to draw. From 1992, President Bush allowed states to apply for waivers to alter programmes in favour, for example, of more work requirements and time limits. This approach continued with Clinton.

Mixed programmes perform best

Under TANF, states were given broad discretion to use their block grant funds to operate programmes to help lone parents find employment (among other purposes), but states generally focused on immediate employment rather than skill-building strategies; less than 1 per cent of funds were used for education and training by 2000.6 This, despite the fact that the strongest employment results came from mixed welfare-to-work programmes such as the Portland programme, which places a strong emphasis on employment, but also on job search, education and training. Career interests and job skills were identified and opportunities provided to train for high-demand jobs and job search to match the new skills. Rather than take the first available job, participants were urged to wait for a good job, one paying at least 25 per cent above minimum wage and offering a chance for stable employment.7

The 1996 reforms:
the Personal Responsibility and Work Opportunity Reconciliation Act

Key changes

• Replacement of AFDC with TANF block grant.
• Block grants mean a fixed sum (with no inflationary increase since 1996 – a real-terms cut of 22 per cent 8) to provide:
  o social assistance to needy families;
  o programmes to ‘end dependence’ through job preparation, work and marriage;
  o reduce extra-marital childbearing; and
  o promote marriage.
• Maintenance of effort. States are allowed to reduce their own spending to 75–80 per cent of the amount they had been spending in 1994.
• Individual states can determine eligibility for cash assistance, who for and how much – but certain groups are prohibited and there is no duty to assist any family.
• Time limits for assistance. States cannot use federal funds to provide assistance for more than 60 months – exceptions for up to 20 per cent of the caseload; states can choose to impose shorter time limits.
• Work requirements for families receiving assistance. States will be penalised unless a specified percentage of families are participating in work activities, but the requirements are reduced if state caseloads fall, resulting in a strong incentive to simply cut caseloads.
• Education and training can only count to meet requirements for a limited period.
• Childcare. TANF funds can be transferred into childcare programmes.
• And, $55 billion cuts in:
  o Food Stamp Programme ($28 billion in cuts);
  o income assistance;
  o nutritional aid and Medicaid to legal immigrants ($22 billion in cuts, including food stamps, over six years – although cuts for elderly and disabled migrants later reversed);
  o restrictions in eligibility for SSI programme for disabled children.

The broader context in which TANF was implemented

• Near full-employment economy.
• Large expansion of earned income tax credit in 1990 and 1993.
• Minimum wage increases in 1996 and 1997.
• Tripling of childcare spending – increasing from $4 billion in 1997 to $11.9 billion in 2004.
• Work supports – including transport and work expenses – expanded, though available less widely than childcare.
• Expansions of Medicaid and the State Children’s Health
  Insurance Programme in the late 1980s and 1990s.
• Stronger child support enforcement – up from 20 per cent receiving any in 1996 to 51 per cent by 2004 (an 82 per cent increase in collections).
• More money to states through TANF.

Since 1996 federal funding of in-work support has increased. Essentially, resources have been switched from social assistance benefits to support for low-income working families. See, for example spending increases in Wisconsin under the block grants.

Although the numbers look different for different states, in general, cash benefits fell and work supports increased, but the growth in use of TANF to support work entry was not as big as the overall decline in cash spending.

US poverty measurement

The official poverty line in the US is an absolute threshold – often referred to as the Orshansky measure. Over forty years ago, it took the costs of an ‘economy’ diet, multiplied by three (because at that point, food was around one-third of spending) and has since only been adjusted annually for inflation. For a family of four, this represents less than 30 per cent median disposable income. In the early 1960s it was about half median disposable income. Arguably, it now fails to represent anything like adequate living standards and measures only those in the most severe poverty. In cash terms, the poverty line for a family of four in 2006 was $20,444 or about £10,423 per year (£200 per week).10

According to the official US definition, 17.6 per cent of children under 18 were below the official poverty line in 2005, nearly 13 million – up from 16 per cent in 2001.11 And the child poverty rate for children in female-headed households was 39 per cent. More than half of all poor children live in these lone-mother households.12 Even by comparing trends in absolute poverty, the US results do not compare well (see Figure 2).

According to the US poverty measure, in 1993, 22.7 per cent of children were poor. The same year, child poverty rates began to fall and reached 16.2 per cent by 2000.13 By 2004, the rate was 17.8 per cent, with an extra 1.4 million poor children since 2000.

Notes

1. R Haskins, Testimony of Ron Haskins to Committee on Ways and Means, 19 July 2006, available at www.brookings.edu/views/testimony/haskins/20060719.htm
2. For a useful and cautionary account of the history of US ‘welfare’ reform and its influence over UK policy, see D Bull, ‘Foreword’ in R J Link, A A Bibus and K Lyons, When Children Pay: US welfare reform and its implications for UK Policy, CPAG, 2000
3. M Forster and M d’Ercole, Income Distribution and Poverty in OECD Countries in the Second Half of the 1990s, OECD Social, Employment and Migration Working Papers, No. 22, OECD, 2005
4. N Ginsberg, Divisions of Welfare, Sage,1992
5. M Greenberg, ‘The US: rising employment, explanations, implications, and future trajectories’, in J Millar and M Evans, Lone Parents and Employment: international comparisons of what works, CLASP/DWP, 2003
6. See note 5
7. See note 5
8. S Parrott and A Sherman, TANF at 10: program results are more mixed than often understood, Centre on Budget and Policy Priorities, 2006, available at www.cbpp.org/8-17-06tanf.pdf
9. Wisconsin Department of Workforce Development website at www.dwd.state.wi.us/dws/tanf/default.htm
10. M Greenberg, What Now for Welfare Reform and Poverty: recent US developments, presentation to DWP conference on welfare reform, ‘Challenges, Choices and International Insight’, 26 March 2007, available at www.dwp.gov.uk
11. US Census Bureau, Income, Poverty and Health Insurance Coverage in the United States: 2005, US Census Bureau/Economics and Statistics Administration, 2006
12. See note 5
13. See note 8

Contents

Executive summary

Section 1
The Clinton reforms
Lessons for the UK
Americal welfare benefits
The 1996 reforms: the Personal Responsibility and Work Opportunity Reconciliation Act Key changes
The broader context in which TANF was implemented
US poverty measurement

Section 2
What are the key US findings?

Setion 3
The UK context
What are the key UK findings?

Notes

Print the entire document