Skip to main content

Progress on poverty, 1997 to 2003/4

A detailed exploration of how poverty has changed since 1997.

Written by:
Holly Sutherland, Tom Sefton and David Piachaud
Date published:

How and why has poverty in Britain changed? What has the Government achieved in terms of reducing poverty? What progress has it made toward its goal of abolishing child poverty in a generation?

Poverty in Britain is a detailed exploration of how poverty has changed since 1997. It examines survey material and uses micro-simulation techniques to review the possible effects of recent government policy. It reviews the impact on poverty of alterations in the wider economy, including falling unemployment, the inequality of earnings, and changing household types.

The authors consider changes in social security benefits and taxes, including tax credits, and also explore the impact of changes in indirect taxes – neglected in most previous studies. The study shows how great a challenge the reduction of poverty represents. Poverty would now be much worse without the improvements that have been made in the tax and benefit system; but to stay on track to reach the goal of halving child poverty by 2010 will require substantially more redistribution to the poorest families.


The Government has set the target of ending child poverty in a generation and reducing it by one-quarter by 2004. A study by Holly Sutherland, Tom Sefton and David Piachaud from the University of Cambridge and the London School of Economics gives the most detailed analysis to date of the progress so far towards these goals and the prospects of reaching the short-term target. The study used the same definition of 'relative poverty' as that used by the Government: household income below 60 per cent of the median income level in that year. It found:

  • Between 1996/7 and 2000/1 relative poverty fell, largely as a result of improvements in employment rates and in the level of some benefits. Overall poverty fell by about one million, including about half a million fewer children in poverty.
  • The effect of policy changes introduced between 1997 and 2003/4 would have been to reduce child poverty by about 1.3 million children, other things being equal.
  • Because incomes generally continue to rise, raising the relative poverty threshold, the actual reduction in child poverty is lower: there are likely to be about one million fewer children in poverty in 2003/4 than in 1997.
  • Pensioner poverty will be lower in 2003/4 than in 1997 but the size of the fall is very sensitive to the measure of income used: a one million fall based on income after housing costs but only one quarter of this based on income before housing costs.
  • Indirect taxes and duties have increased in real terms since 1997. This has affected poorer households disproportionately, because most indirect taxes are regressive. The effect of this (not included in official poverty statistics) would, however, be relatively small - equivalent to raising poverty between 1997 and 2002/3 by 0.4 percentage points.
  • The researchers conclude that the Government could just succeed in reaching its first milestone of reducing poverty by a quarter by 2004. However, achieving its longer-term targets for child poverty is likely to be more difficult and will require the Government to continue to give priority to poverty reduction.


Tackling child poverty has come to be a key policy goal of the Government with a specific target of reducing it by one quarter from the 1998/9 level by 2004. It is a central element in the Government's programme to promote 'Opportunity for All'.

The purpose of the research reported here was to examine the evidence on what had been achieved between 1997 and 2000/1 and what the prospects were up to 2003/4.

The study examined the same data using the same definition of poverty as used by the Government (see Box 1). Future prospects were explored using POLIMOD, the University of Cambridge microsimulation model. In addition, the study assessed the effects of changes in indirect taxation, which are not normally considered when evaluating changes in poverty.

Box 1: Defining poverty
The study uses the same definition of poverty as used by the Government in Opportunity for All (DWP, 2002).

The poverty line is 60 per cent of median income level – where the median is the level of income after direct taxes and benefits, adjusted for household size, such that half the population is above the level and half below it. This definition is a standard that changes as median income levels change; it is a measure of relative poverty.

Because housing costs vary greatly and do not necessarily change in line with incomes generally, the study assessed incomes both before and after housing costs.

The researchers estimate that child poverty will be a quarter below its 1998/9 level by 2004 after taking housing costs into account and a third lower before taking housing costs into account. This suggests that the Government's target should be met unless other factors such as employment levels change in an unfavourable direction in which case further redistributive measures will be needed.

Pensioner poverty
Pensioner poverty showed little change between 1996/7 and 2000/1 but a substantial reduction between 2000/1 and 2003/4; the size of this reduction is very different - between one million and a quarter of a million - depending on whether or not housing costs are taken into account (housing costs are on average lower for pensioners than for the rest of the population) (see Table 1).

Table 1: Reductions in poverty before and after housing costs (children and pensioners)
Children Pensioners
Before housing costs After housing costs Before housing costs After housing costs
Proportional reduction in poverty 33% 24% 13% 45%
Fall in number in poverty (thousands) 1,030 1,040 270 1,220

Policy changes 1997 to 2003/4

The principal policy changes directed at reducing poverty since 1997 have been the introduction of or increases in:

  • National Minimum Wage;
  • Working Families Tax Credit and Income Support changes targeted on children (replaced and enhanced by Child Tax Credit and Working Tax Credit);
  • Minimum Income Guarantee for Pensioners (replaced and enhanced by the Pension Credit);
  • Winter Fuel Payments.

Changes to income tax and National Insurance contributions as well as some benefits not listed above have also affected poverty. Some changes increase (or decrease) poverty directly, and some shift the poverty line up (or down).
The effect of employment policies such as the introduction of New Deals for young people, the long-term unemployed, lone parents, disabled people and others will have had an additional impact.

Effects 1997 to 2000/1

Analysis of the Family Resources Survey showed that between 1996/7 and 2000/1 there was some considerable progress with regard to children, but less progress with regard to poverty among other groups.

The overall explanation for the changes in relative poverty that occurred between 1996/7 and 2000/1 is fairly clear. The relative poverty line rose by about one-tenth in real terms. There was little change in family composition or in the shape of the earnings distribution. Two things did change. Most significantly, unemployment fell and more households had someone in paid employment. Second, policy on benefits and tax credits clearly disadvantaged some and helped others. Those with benefits falling relative to incomes generally were more likely to be losers. They included those on the basic state pension, Jobseeker's Allowance and Incapacity Benefit and those on Income Support who did not have children. (In addition, lone parent benefits were abolished.) Those more likely to gain were those with children, particularly low earners in employment, and this was a major factor in the reduction by some half a million in the number of children in poverty.

Changes in poverty up to 2003/4

The combined effect on poverty of all the changes to tax and benefit policies between 1997 and 2003/4 was assessed using the microsimulation model, POLIMOD.

Child poverty
In 1997 between one-quarter and one-third of children - 3 to 4 million children, depending on the measure of income used - were living in households in relative poverty. This was triple the proportion in poverty in 1979.

The prospect is that the policy measures taken between 1997 and 2003/4 will have resulted in 1.3 million fewer children being in poverty than would otherwise have been the case. This does not mean that actual poverty will have fallen by this amount. With a relative poverty line that rises with the growth in incomes, child poverty in 2003/4 is estimated to be one million less than in 1997. Figure 1 shows how the policy changes reduce the proportion of children with low household incomes, but also how the poverty line shifts to higher levels as incomes in general rise. For example, in 1997 24 per cent of children were in households with below 60 per cent of median 1997 income. Looking downwards along this line shows that six years later only 11 per cent of children are in families with incomes below this level. However, moving to the right to adjust for rising overall incomes, shows that 16 per cent were below the higher poverty line of 2003/4.

The impact of indirect taxes
On balance, indirect taxes and duties have increased in real terms since 1997 (though they fell between 2000/1 and 2002/3). This has affected poorer households disproportionately, because indirect taxes generally represent a higher proportion of income for those on lower incomes. Lone-parent households have been hit hardest by the above-average increase in tobacco duties. Pensioners have been least affected by changes in indirect taxes, largely because those aged over 75 were exempted from TV licence fees during this period.

Overall reductions in poverty would only be marginally lower as a result of recent increases in indirect taxation if these were reflected in official poverty statistics (which they are not). For lone-parent households, who were most adversely affected, the impact of higher indirect tax rates is equivalent to an increase in their poverty rate of around one percentage point.

What are the prospects beyond 2003/4?

For poverty to remain at the same level, let alone fall further, it is necessary both for employment levels to be maintained and for benefits and tax credits to rise in line with median incomes. This necessarily requires increasing expenditure in real terms, although not as a proportion of national income. In terms of overall poverty, it remains the case that Income Support levels are substantially below the 60 per cent of median income poverty level used by the Government. The failure in recent years to increase these benefit levels in line with incomes generally has made this situation worse.


The findings of this study show the challenge that tackling poverty represents. Between 1996/7 and 2000/1 relative poverty fell, largely as a result of improvements in employment rates and in the level of some benefits. But the overall impact was less substantial when compared with a poverty line which rose with median incomes. The tax and benefit changes analysed in the study do have a major effect - but most of this is needed simply to keep up with a moving target.

What the study shows is that greater employment, or 'work for those who can', has made a real contribution to reducing poverty. But there is a limit to how much further employment measures can contribute to reducing poverty. The study also shows that without the improvements which have been made to the tax and benefit system for those with low incomes, poverty would be much worse. Changes in indirect taxation have had only a small effect on poverty but it is important for the future that their regressive effects are taken into account. Whatever its form, more redistribution will be needed each year simply to maintain the progress which has been made.

In 1999 a goal was set of halving child poverty by 2010 and a relative definition of poverty was clearly adopted. Even if this goal is achieved, relative poverty will still be higher than in 1979. It will be possible to stay on track to achieve this goal but to do so will require substantially more redistribution to the poorest and continuing priority to be given to the goal of ending child poverty. Further reductions in child poverty are likely to be increasingly hard to achieve. The task of ending poverty more generally remains to be tackled.

About the project

The study was carried out by Holly Sutherland, Director of the Microsimulation Unit, in the Department of Applied Economics at the University of Cambridge, Tom Sefton, Research Fellow at the ESRC Research Centre for Analysis of Social Exclusion (CASE) at the London School of Economics, and David Piachaud, Professor of Social Policy at the London School of Economics and an Associate of CASE.

The analysis was largely based on data from the Family Resources Survey; data from the Family Expenditure Survey were used for the analysis of indirect taxes. The study was carried out between April 2002 and March 2003.


043.pdf (80.11 KB)
1859351522.pdf (266.3 KB)