Defining income poverty out of existence?

We are about to enter the first milestone year of the Government's crusade against child poverty – a point at which it intends to have reduced the number of children living in poverty by a quarter. In December 2003 the Government also announced how it would measure child poverty after 2004/05. [Footnote 1] Here, Paul Dornan examines what progress has been made and the implications the measurement review may have on future policy.

Progress so far
The measurement review
Changes to the relative low income measure
Re-defining 'eradication'
Looking forward

Progress so far

Our historic aim, that ours is the first generation to end child poverty forever… It's a 20-year mission but I believe it can be done
Tony Blair, March 1999 [Footnote 2]

This goal, it later unfolded, was to be reached by first reducing the number of children in income poor households (defined as having needs-adjusted incomes below 60 per cent of the national median) by 2004/05. This target was written into the Public Service Agreement between the Departments for Work and Pensions (DWP) and the Treasury.[Footnote 3] Beyond 2004/05 we know rather less. There is a commitment to halve child poverty by 2010 and then to 'eradicate' it by 2020, but although we have new measures, we have no specific targets with which to measure these.

Approaching the first milestone year, the Government is making solid progress in tackling child poverty. The situation it inherited should not be forgotten: the proportion of children living in income poverty [Footnote 4] had increased from about one in ten in 1979 to one in three by 1996/97.[Footnote 5] By 2001/02 this stood somewhat lower at 30 per cent. The table below provides the latest Households Below Average Income data, together with the rates of income poverty necessary to meet the 2004/05 target.

The data shows various things. Child poverty has been declining; there were 500,000 fewer children in income poverty in 2001/02 than in 1996/97. Yet there is clearly still a distance to go before the 2004/05 target is reached. The data we have is old, from 2001/02, and so no reforms implemented after April 2002 have fed into the figures. The method adopted of assessing the effectiveness of policy to avoid the problem of old data has been to use existing data to project forward the effect of policy change. For this purpose we have analysis from both the Joseph Rowntree Foundation (JRF) [Footnote 6] and the Institute for Fiscal Studies (IFS). [Footnote 7] All of the analysis by these various authors pre-dates both the Chancellor's pre-Budget statement of December 2003 and the DWP's measurement review released in the same month. Further, the research reports do not agree with each other and so need careful examination to draw conclusions. Projecting the effect that policy change has on household incomes is complicated by the influence of other concurrent effects on families' incomes, such as employment levels (and earnings growth).

Based on the situation before the pre-Budget review, the JRF study suggested that the Government was on track to meet the first milestone on the 'before housing costs' measure, and just about on track to meet the 'after housing costs' measure. The IFS produced three separate analyses in 2003 (first in its 'Green Budget', [Footnote 8] updated after new data was made available from 2001/02 [Footnote 9] and then in the run up to the pre-Budget statement in December 2003 [Footnote 10] ). Each of these IFS analyses suggested that more needed to be done in order to meet the first child poverty target on the 'after housing costs' measure. The last report suggested the need for an additional £1 billion to be spent on child tax credit, [Footnote 11] to be implemented from April 2004, to ensure that the Government met its first child poverty reduction milestone. In the end, the pre-Budget report promised an additional £885 million, [Footnote 12] – substantial, though slightly less that the IFS suggested was necessary to reach the milestone. [Footnote 13] The amount provided also looks rather less when it is set against the income saved through freezing the family elements (hence the cost of these elements was shrunk in real terms) which occurred at the same point.

The upshot of all of this is that firstly, we do not know whether or not the Government will meet its target, but it may well be close. It is apparent that it will be easier to meet the 'before housing costs' than the 'after housing costs' measure since there is less far to go (see the table above). With this in mind, it is worth noting that the Department for Work and Pensions' measurement review conclusions, discussed below, will result in a shift away from reliance on both 'before' and 'after housing costs' to measure child poverty and towards 'before housing costs' only. Though important in judging progress, it is too easy to get caught up with the specifics of the first milestone. Progress has certainly been made: data drawn from the European Community Household Panel Survey released in the final measurement review document suggests a improvement in position from the worst in Europe to fifth from bottom. [Footnote 14] The rate of child poverty in the UK is still way above the EU average and still unacceptably high in comparison to the late 1970s. What is necessary now is a way looking at reductions towards 2010 and beyond.

The measurement review
The DWP has been reviewing the way in which it measures income poverty. The process has been an elongated one, launched in April 2002 with a consultation document, [Footnote 15] with interim conclusions formed in early 2003 [Footnote 16] and with final conclusions [Footnote 17] published as a press release seven days before Christmas and on the last day of Parliament. The importance of the review is that it provides us with a yardstick by which subsequent policy will be measured. It does not, however, provide us with targets for reducing child poverty: these are likely to follow with the Spending Review in 2004. The conclusions are a departure from the measures we have used thus far. From 1998/99 to 2004/05 the Government used a measure which defined families as poor if they had equivalised household incomes of below 60 per cent of the contemporary median (both 'before' and 'after housing costs' had been used simultaneously). [Footnote 18] The vehicle that will provide the data on which rates of poverty will be measured, the Family Resources Survey, [Footnote 19] is to remain the same, but beyond 2004/05 the Government proposes to use three separate measures: [Footnote 20]

• Absolute low income. Children are counted as poor on this measure if they live in household with a before housing costs equivalised income of less than 60 per cent of the 1998/99 median income, uprated for inflation.
  
  
• Relative low income. Children are counted as poor on this measure if they live in a household with a before housing costs equivalised income of less than 60 per cent of the contemporary median.
  
  
• Material deprivation and low income combined. Children are counted as poor on this measure if they live in a household with a before housing costs equivalised income of less than 70 per cent of the contemporary median and if they both lack material necessities and can not afford them.

Unpicking the detail behind these is vital to understanding the direction of the Government's plans for ending child poverty: the devil really is in the detail. The first point is that, at time of writing, we do not have specific targets that these measures are attached to, as we do with the current Public Service Agreement (PSA). [Footnote 21] Without these targets the trajectory of government policy is difficult to judge. As no PSA target exists, it is unclear as to how each of the three measures will relate to each other – will the Government have to halve each of them or will some be more important than others? From DWP conclusions we know that all have to be seen to be falling when all three measures show falls in the number of children in poverty, but we do not know how it will be judged if poverty is falling on one or two of the measures but not on the third.

The data on which progress will be measured is drawn from the Family Resources Survey and is collected within the benefit year (April to March). This does not align neatly with the intention to halve by 2010, and end by 2020, so halving child poverty by 2010 could therefore either mean 2009/10 or 2010/11. Further data does not exist on two of the three measures (the absolute low income [Footnote 22] and the material deprivation and low income combined measures) and so we do not know the number of poor children the Government will be seeking to remove from poverty in order to halve it. We will not know this until 2006 when data is released from the 2004/05 year of Family Resources Survey data.

The first measure will be comparatively easy to hit since absolute (real terms) incomes have tended to rise over time and so we would expect the proportion of children in poverty on this measure to reduce over time. The second measure is discussed in detail below. On the third measure, material deprivation and low income combined, the document implies that material deprivation will be judged to occur if families report not having and not being able to afford items from a list of material necessities. The suite of questions to be included in the Family Resources Survey is attached to the conclusion document as an annex. This list is wide ranging and includes being able to keep the home adequately warm, having a holiday away from home for one week a year, not staying with relatives, and the ability to save £10 per month for a rainy day. It is not clear, however, whether all of the questions (and if not which) will be used to determine the indicator. Neither is it made clear whether or not this list will change with time with social norms. Since living standards have improved over time it is also important that the list of necessities is also able to be adapted to account for this.

Changes to the relative low income measure
The imponderables discussed immediately above lead to caution when drawing conclusions from the review, but there are further elements of the measurement conclusion on which we can be clearer. First, the process of needs adjustment. The Households Below Average Income series adjusts household income to account for different need. This process, known as equivalisation, is based on the principle that a given amount of income will stretch further for certain households (say without children) than it will for others (with children). This calculation is being changed, the effect of which is to value the costs of younger children more highly, which will in turn increase the number of children measured as poor. To quantify the effect, under the old method of calculation in 2001/02 21 per cent of children (2.7 million) were measured as living in households with 'before housing costs' equivalised incomes below 60 per cent of the median. With the new measure, the percentage of children living in households with equivalised incomes below 60 per cent of the median rises to 23 per cent (2.9 million). [Footnote 23]

The second change relates to the way in which housing costs are dealt with. Studies of income distinguish between when income before and after housing costs have been accounted for. The reason for doing this is that housing is usually the largest item of spending and is one which does not tend to vary. In consequence, a more accurate understanding of income is available after having accounted for housing costs (often called disposable income). Less children are counted as being in poverty before housing costs have been accounted for than after housing costs have been accounted for (see the table above) since housing costs bear particularly heavily on families with children. Following this review, child poverty is to be measured on the 'before housing costs' measure only. In 2001/02 Households Below Average Income reported that 2.7 million children were in poverty on the 'before housing costs' measure and 3.8 million on the 'after housing costs' measure. [Footnote 24] By settling on the 'before housing costs' measure, the Government has reduced the headline figure for the number in poverty substantially. Taken with the change in the equivilisation, discussed above, this alters the implied (maximum) headline relative income poverty figure from 3.8 million to 2.9 million children living in poverty in Great Britain in 2001/02. At a stroke, the measurement change has 'removed' 900,000 children from poverty. By doing so it has reduced the number of poor children and therefore the number to be taken out of poverty.

Re-defining 'eradication'
As well as changes in the measures, the review also refers to how success in eradicating child poverty could be judged:

Success in eradicating [child] poverty could, then, be interpreted as having a material deprivation child poverty rate that approached zero and being amongst the best in Europe on relative low incomes. [Footnote 25]

And further, in a footnote:

Possible ways to define being 'amongst the best in Europe' could include: having a relative child poverty rate no higher than the average of the best three countries in Europe; having a relative child poverty rate no higher than the average of the best four countries in Europe; and, having a relative child poverty rate that was within 2 percentage points of the average of the best three counties. Achieving any of these on current definitions would mean having a poverty rate of between that of Sweden and Denmark. [Footnote 26]

The goal of 'eradicating child poverty' now appears not to mean removing every child from poverty at all, but reducing the relative child poverty rate to meet 'the best in Europe'. Using Eurostat data from 2001 published with the review conclusions, [Footnote 27] this implies the following proportion and number of children living in relative income poverty:

• Based on the average of the best three countries in Europe, this suggests an end point - around 7 per cent of children living in poverty (900,000 children).
  
  
• On the average of the best four, this would be about 8.5 per cent (1.1 million children).
  
  
• On the third (within 2 percentage points of the average of the best three) about 9 per cent (1.2 million children).

The changes in the way in which child poverty is to be measured are a long way short of the eradication of child poverty. They imply firstly that the income poverty to be targeted is not as high as has until now been presumed; raising the bar and reducing the number of children whose family income must rise to escape income poverty. Further, they imply that we could have up to 1.2 million children living in relative income poverty and have eradicated child poverty.

Looking forward
So we return to the central concern: what does the DWP review, together with the policy climate, mean for the 2004 Spending Review, and so for policies designed to halve and to 'eradicate' child poverty by 2020. The effort currently going into poverty reduction strategies is certainly to be welcomed since it is focused on the methods by which child poverty should be reduced and not the rightness of the ambition. That much is to be commended, as is the progress that has already been made. Analysis of the progress shows us two useful things. Firstly, that progress has been made and secondly, the scale of the task still to do. The progress that has been made should show the Government that more can be achieved. With that in mind, the implied timidity of the new relative income indicator, which appears to make it easier to 'eradicate' child poverty, is to be regretted.

The measurement review contains within it two particularly concerning elements. First, that the 'eradication' of child poverty no longer means what most would understand by the word. Instead we are now told that this is to mean being among the best in Europe. Though this implies a much-improved situation on what we have currently, it is not what we were promised. Secondly, the shift from both 'before' and 'after housing costs' to relying on just 'before housing costs' measures in the relative income measure has the effect of substantially reducing the number of children currently measured as being in poverty. The consequence is not only to raise the bar (fewer children are defined as in income poverty based on the 'before' than on the 'after housing costs' measure) but to make the targets easier to attain. This is not only a lamentable step: it lacks the ambition that the children of the UK have a right to expect.

Setting this to one side and looking towards the future, there is a real need to make, as Gordon Brown said in Budget 2003, faster progress [Footnote 28] against the milestones. The 2004 Spending Review offers the possibility to achieve this improvement. Here there is a need for a wider strategy – backed by targets – to deliver further on the pledges. Both the Treasury child poverty review and the Work and Pensions Committee reviews provide useful vehicles for encouraging spending departments to recognise the role that they should play in ensuring that policies are both child-friendly generally and specifically assist in the fight against poverty. Beyond this, however, there is a need for a wider public debate than we are having at the moment. Ending child poverty is a serious challenge, but one of which the Government ought to be very proud. The level of discussion had on this issue by the Government does not match either the importance of the issue or the future challenge of committing the increasing level of resources needed to keep up with the poverty targets. This requires a wider debate

Linked to this is the tricky issue of political consensus. We have a pledge to end child poverty by 2020, but the maximum tenure of each administration in the UK is five years, so between now (2004) and then (2020) we will have a minimum of three general elections, each of which may cause some alteration in the plans. Although Iain Duncan-Smith, then leader of the official opposition, was clear of the importance of poverty, employing Beveridgeite language saying five new giants blight Britain today… the fourth giant is child poverty, [Footnote 29] his successor, Michael Howard, has not as yet echoed these words. The use of the term does suggest that the Government has managed to shift the centre of the debate, so that all parties have to address child poverty, but we need further cross-party consensus. The interest of the cross-party Work and Pensions Committee in child poverty offers some hope of a level of consensus, but more is needed to ensure progress is made in reducing child poverty irrespective of the colour of the administration.

Paul Dornan is CPAG's policy officer

Footnotes
1. Department for Work and Pensions, Measuring Child Poverty, 2003, can be found at www.dwp.gov.uk/consultations/consult/2003/childpov/final.asp [back to text]
2. Reproduced in R Walker (ed), Ending Child Poverty: popular welfare for the 21st century?, The Policy Press, 1999 [back to text]
3. Department for Work and Pensions, Public Service Agreement (PSA) Technical Note for the Department for Work and Pensions, DWP and HM Treasury, 2002 [back to text]
4. Defined as living below 60 per cent of the median 'after housing costs' equivalised income [back to text]
5. See T Clark and A Goodman, Election 2001 Living Standards under Labour, Election Briefing Note 4, May 2001 [back to text]
6. H Sutherland, T Sefton and D Piachaud, Poverty in Britain: the Impact of government policy since 1997, Joseph Rowntree Foundation, 2003 [back to text]
7. M Brewer and G Kaplan, 'What do the Child Poverty Targets Mean for the Child Tax Credit?' in R Chote, C Emmerson and H Simpson (eds), The IFS Green Budget, Institute for Fiscal Studies, 2003; M Brewer, A Goodman and A Shephard, How has Child Poverty Changes under the Labour Government? An update, Institute for Fiscal Studies, 2003; M Brewer, What do the Child Poverty Targets Mean for the Child Tax Credit? An update, Briefing Note 41, Institute for Fiscal Studies, 2003 [back to text]
8. M Brewer and G Kaplan, 'What do the Child Poverty Targets Mean for the Child Tax Credit?' in R Chote, C Emmerson and H Simpson (eds), The IFS Green Budget, Institute for Fiscal Studies, 2003 [back to text]
9. M Brewer, A Goodman and A Shephard, How has Child Poverty Changes under the Labour Government? An update, Institute for Fiscal Studies, 2003 [back to text]
10. M Brewer, What do the Child Poverty Targets Mean for the Child Tax Credit? An update, Briefing Note 41, Institute for Fiscal Studies, 2003 [back to text]
11. See note 10 [back to text]
12. See HM Treasury, Pre-Budget Report, Cm 6042, December 2003, Table B4 [back to text]
13. See note 8 [back to text]
14. On the basis of the proportion of children in households with 'before housing costs' equivalised incomes of less than 60 per cent [back to text]
15. Department for Work and Pensions, Measuring Child Poverty: a consultation document, April 2002 [back to text]
16. Department for Work and Pensions, Measuring Child Poverty: preliminary conclusions, May 2003 [back to text]
17. Department for Work and Pensions, Measuring Child Poverty, December 2003 [back to text]
18. See note 3 [back to text]
19. Up until this year the Family Resources Survey has excluded Northern Ireland, meaning that previous analysis relates to Great Britain only. As data collection has now been extended to Northern Ireland, future data will relate to the United Kingdom. [back to text]
20.See note 17 [back to text]
21. See note 3 [back to text]
22. It would be possible to rework the data to produce this absolute low income data, but not that which relates to material deprivation as this requires new questions to be asked in the Family Resources Survey. [back to text]
23. National Statistics, Households below Average Incomes: an analysis of the income distribution from 1994/95 - 2001/02, DWP, 2003 [back to text]
24. See Table [back to text]
25. See note 17 [back to text]
26. See note 17 [back to text]
27. See note 17 [back to text]
28. HM Treasury, Budget 2003, HC 500, 2003, para 5.10 [back to text]
29. I Duncan-Smith, 'Defeating the Five Giants' speech at Toynbee Hall, 13 September 2002 [back to text]

Poverty 117, Winter 2004