Cover image

Poverty, inequality and policy since 1997

John Hills, Tom Sefton and Kitty Stewart

25 February 2009

This study examines the impact of policy on poverty and inequality in Britain since 1997.

This research shows what effect policies introduced since 1997 have had on reducing poverty and inequality. It offers a considered assessment of impacts over a decade:

  • How did policies change, before 1997 and since then?
  • What evidence is there of impacts on key outcomes?
  • What gaps or problems remain or emerged?

The study covers a range of subjects, including public attitudes to poverty and inequality, children and early years, education, health, employment, pensions, and migrants. It measures the extent of progress and also considers future direction and pressures, particularly in the light of recession and an ageing society.

The research draws on extensive analysis of policy documents, analysis by government departments and research bodies, published statistics and evaluations, analysis of large-scale datasets, micro-simulation modelling and a long-running qualitative study with residents of low-income neighbourhoods. A related book, Towards a more equal society? Poverty, inequality and policy since 1997, is published by The Policy Press.


Download as PDF 0.1 MB

This study examines what has happened to different aspects of inequality in Britain, and how this relates to policies adopted since 1997.

Key points

  • There is no simple picture of success or failure in progress towards the objective of ‘a more equal society’. Compared to the period before 1997, trends have improved in more policy areas than they have worsened in, but what has happened has varied between areas and over time.
  • Notable successes in the last decade included:
    - reduced child and pensioner poverty;
    - improved educational attainment for the poorest areas and schools; and
    - a narrowing of economic and other divides between deprived and other areas.
  • However, health inequalities continued to widen, as did gaps in incomes between the very top and very bottom.
  • Where specific policy initiatives were taken, some progress was made, particularly in education, employment and neighbourhoods, and through tax-benefit reform. However, effects were often small in relation to the scale of the problems, and were sometimes offset by external changes.
  • In several policy areas there was a marked contrast between the period up to 2002 or 2003 and since then, when progress has been slower or has even stalled. As growth in living standards as a whole slowed and public finances became more constrained even before the current recession, policy momentum was lost.

The study concludes that the period surveyed was more favourable towards egalitarian objectives than seems likely in future, even beyond the immediate effects of recession: public attitudes have become less favourable towards redistribution and an ageing population will put pressure on public finances.

This study explores whether the Government has met its promise of making Britain 'a more equal society' than it was in 1997. The evidence now available allows a considered assessment of impacts over a decade, but does not present a simple picture. Table 1 (please see the PDF) summarises the areas surveyed under three headings:

  • How did policies change, both from before 1997 and during the period since then?
  • What evidence is there of impacts on key outcomes?
  • What gaps or problems remain or emerged?


As Table 1 shows (please see the pdf), outcomes varied between policy areas and over time since 1997. While the growth of inequality in earnings and incomes between those near the top and those near the bottom halted, inequalities including the very top and bottom continued to widen. Significant progress in reducing child and pensioner poverty was not matched for the working-age population. New evidence suggests indicators of child well-being improved significantly in the UK between 2001 and 2006 compared with other countries, but child poverty remains amongst the highest in Europe. Educational attainment improved faster for the poorest schools and areas, and other gaps between the most deprived and other areas narrowed. There were slow reductions in gaps between most minority ethnic groups and the white population in several respects, although recent migration has led to much greater diversity between groups, some of which are very disadvantaged. Health inequalities continued to widen.

As an overview, the study examines trends in the range of indicators chosen by the government itself to measure progress on poverty and social exclusion. Given the increase in national prosperity over a decade, it would be disappointing if some of the indicators had not improved. For others, an improvement reversing a previous decline is more impressive. Comparing trends in the period since 1997 as a whole with those in the same indicators before 1997, the overall balance is positive: for nearly half of the indicators, trends have improved, but trends worsened for a quarter.

What would have happened in the absence of policy change?

The extent to which explicit policy shifts can take the credit for any improvements since 1997 is often an open question. In the rather limited number of cases where specific rigorous evaluations have been carried out, individual initiatives do show positive effects – for instance some measures in schools, in employment promotion and in poor neighbourhoods – although these are often small by comparison with the scale of problems or the actual changes over the period.

A clearer comparison can be made between the distributional impact of the 2008/09 tax and benefit system and results if the 1996/97 system had been preserved. If, as was the general policy before 1997, benefit levels and tax allowances had been increased in line with price inflation, income inequality and poverty rates would both have been significantly higher than they actually were by 2006/07. The Gini coefficient measure of income inequality would have been 1 percentage point higher – a rise of 3 points since 1996/97, rather than the actual increase of 2 points. Strikingly, the ratio between the incomes of those near the top and those near the bottom would have increased by nearly a tenth, rather than remaining flat.

Overall poverty rates would have been 6–7 percentage points higher than the actual outcome, corresponding to a rise in poverty of 4–5 points over the period, rather than the actual fall of 1–3 points by 2006/07. Rather than falling, the child poverty rate would have risen by 6–9 points and the pensioner poverty rate by 7 points. Against an alternative benchmark – with the 1996/97 system uprated in line with earnings growth – the redistributive impact of the policy reforms looks more modest, although actual overall poverty rates are still 3 points lower than they would have been in this scenario. Against this tougher benchmark, tax and benefit changes between 2004/05 and 2008/09 were actually mildly regressive.

The extent of progress

In some areas, including poverty reduction, there are clear differences between progress in the first half of the period, and less progress or even regression since then. In several, this reflects an apparent loss of policy momentum, particularly after 2004. These include:

  • the overall direction of tax and benefit changes since 2004/05;
  • slowing of growth in (generally pro-poor) spending on education and health;
  • plateauing of child-related spending after 2004;
  • the impact of the New Deals for young people and long-term unemployed;
  • the ending of specific area-based initiatives.

In other areas, there was little momentum to lose. Conspicuously, policy since 1997 has not focused on inequalities at the very top of society, which increased. Also, the assumption was that general expansion of health services would narrow health inequalities, but there is no evidence this occurred. Devolution, although accompanied by much rhetoric on social justice and inclusion, has so far had little impact on income and other inequalities. One reason for lost momentum is the slower economic growth and greater pressures on the public finances after 2003. Not only have there been fewer resources for new initiatives, but the losers from ‘quiet redistribution’ have become more visible. If incomes in general are rising slowly, redistributive policies lead to potential falls, rather than just slower growth, in living standards for some.

By contrast, there are areas where policy was intensified or where initiatives’ impacts will be long term, and not yet visible. These include early years policies in particular, as well as aspects of education policy, recent policies on economic inactivity, pensions and the equality and human rights agenda.

Future direction and pressures

The decade from 1997 was favourable to an egalitarian agenda in several ways: the economy grew continuously; the government had large majorities and aspired to create ‘a more equal society’; and public attitudes surveys suggested pent-up demand for more public expenditure.

This environment looks very uncertain for the immediate future. In the medium term, fiscal pressures from an ageing society could further constrain the resources available for redistribution, particularly if the overall share of taxation in national income is fixed or intended to fall. Without compensating those on low incomes through changes in taxes and benefits and direct action to improve energy efficiency, putting a higher price on carbon emissions could be regressive. While the party-political debate is more accepting of reducing relative poverty as an objective, public attitudes towards the benefit system have hardened and support for redistribution has declined. This period may have been ‘as good as it gets’ for egalitarian aims, for some time to come.


In several key respects the UK has become a somewhat more equal society than it was in 1997, but the judgement depends on which inequalities are being examined, between whom, and over which time period. Where significant policy initiatives were taken, the outcomes generally moved in a positive direction. It is not the case that nothing was tried or nothing worked. Rather, many things were tried, and most worked.

Setting objectives in relative terms or for gaps between groups creates potentially difficult ‘moving targets’, rising with national prosperity. However, the Government actually found it harder to make progress after 2003, when overall living standards grew much more slowly, even before the ‘credit crunch’. The fiscal position became tighter, and rapid increases in public spending ended. The politics of redistribution with growth appear far easier than those of redistribution without growth.

The UK’s experience in the 1980s and 1990s showed that the strategy of hoping that growth in living standards at the top would ‘trickle down’ to those at the bottom did not work. The last decade has shown that a more interventionist ‘pump up’ strategy is hard in an unequal society. Gains are possible, but require intensive and continuous effort to be sustained.

About the project

This study was carried out by a team based at the LSE’s Centre for Analysis of Social Exclusion, the Universities of Essex and Southampton, Institute of Education, and Institute for Public Policy Research. It draws on extensive analysis of policy documents, analysis by government departments and research bodies, published statistics and evaluations, analysis of large-scale datasets, micro-simulation modelling and CASE’s long-run qualitative study with residents of low-income neighbourhoods. It builds on an earlier book, A More Equal Society? New Labour, poverty, inequality and exclusion edited by John Hills and Kitty Stewart (The Policy Press, 2005).

Recommend to a friend via email: