Measuring A consultation document child poverty · Measuring poverty and social exclusion in the European Union 17 The Irish Government’s anti-poverty target 18 Consultation options

A consultation documentwww.dwp.gov.uk

Measuring child poverty

ContentsForeword 3

Executive Summary 5

Introduction 7Why consult? 7Geographical scope 7How to respond 8What will happen next? 8

Our starting point 9Child poverty 9

Making a difference 11How we currently monitor progress 13

Opportunity for all indicators 13Public Service Agreement targets 14

The challenge we face 15International experience 16

The United States official poverty line 16Measuring poverty and social exclusion in the European Union 17The Irish Government’s anti-poverty target 18

Consultation options 19What makes a good indicator? 19Option One – multi-dimensional headline indicators 21

Advantages 21Challenges 21

Option Two – a child poverty index 21The Human Development Index 22Advantages 23Challenges 23

Option Three – combining relative low income and material deprivation –‘consistent poverty’ 24

The Poverty and Social Exclusion Survey 24Families and Children Survey hardship measure 25Advantages 25Challenges 26

Option Four – low income and ‘consistent poverty’ indicators 27Core indicators 27Advantages 27Challenges 28

Key consultation questions 28

Glossary 29

References 30

Measuring child poverty

The Prime Minister’s pledge to eradicatechild poverty in a generation is a significantand demanding undertaking. We arealready making progress year on year andthere is still more to do. But theGovernment alone cannot achieve thisambition without public support oversuccessive Parliaments. We need to ensurethat the right policies are in place to tacklenot just the symptoms of poverty but alsoits root causes. And we need to delivervisible results each year.

Tackling poverty is not only about improvingthe lives of individuals, it is an economicnecessity. Everyone pays the bill of povertyand everyone benefits when communitieshave more jobs, less crime, better healthand higher expectations for their children.

Our annual Opportunity for all reports setout our strategy to tackle poverty and socialexclusion and include a set of indicators tomonitor progress. We can already see realimprovements for children. Educationresults are improving for all age groups.

And there are 300,000 fewer children livingin workless households than in 1997. TheWorking Families’ Tax Credit, worth anextra £35 a week on average whencompared to the old Family Credit, isbenefiting nearly 1.3 million families with2.5 million children. Budget 2002 takesfurther steps to tackle child poverty.

To meet our long-term objectives we arebuilding on measures to raise the incomesof families – for example through the newtax credits. But poverty is about far morethan income. That is why our strategy isalso about delivering excellent publicservices in every neighbourhood: high-quality healthcare, a world-class educationsystem and the opportunity of decenthousing for all. And it is about raisingemployment – as a job remains the bestroute out of poverty.

As we extend Jobcentre Plus across thecountry, we will help more people to helpthemselves through work. We will continueto provide financial support for families anddeliver better public services for allneighbourhoods. We must make sure thatno child is left behind. This will be mosteffectively achieved by a partnershipbetween the Government and all sectors:voluntary, private, public, faith andcommunity groups. That is why we needyour views.

Foreword

3

We are a Government that is prepared tobe judged on what we deliver and to learnfrom the experience of others. We havealready engaged in wide public consultationon key issues, including the way we willprovide financial support in future tofamilies through the new tax credits. Nowwe need to make sure that we are trackingprogress towards our goal to eradicate childpoverty in the best way possible, so that wecan all see clear results.

It is our aim to build consensus for ameasure of child poverty that will endure inthe long-term. We are, therefore, nowasking for your views on how we shouldbuild on our existing indicators and targetsto measure child poverty in the long-term.

Rt Hon Alistair Darling MPSecretary of State for Work and Pensions

April 2002

4 Measuring child poverty

In March 1999, the Prime Ministerannounced the Government’s commitmentto eradicate child poverty within ageneration. As we move towards this goalwe want to be sure that we are measuringpoverty in a way that helps to targeteffective policies and enables theGovernment to be held to account forprogress. We have been following closelyimportant debates among experts in thefield on poverty measurement. We nowneed to consider how these can and shouldinform our approach to measuring childpoverty in the long term and wouldwelcome your views.

Our annual Opportunity for all reports andmore recently Tackling child poverty: givingevery child the best possible start in life setout our strategy for tackling child poverty.

Opportunity for all includes a set ofindicators to monitor annual progress.These indicators capture a range ofoutcomes, including:

• relative, absolute and persistent lowincome;

• worklessness;

• educational attainment;

• health inequalities; and

• housing standards.

The indicators are linked to Public ServiceAgreement targets, which are set as part ofthe regular Spending Reviews. But the focusof this document is a discussion of optionsto track long-term progress at tackling childpoverty. Some countries, such as the UnitedStates of America and Ireland, haveadopted headline measures of poverty andwe can learn from their experience andfrom research.

This paper highlights four differentapproaches to monitoring progress:

• using a small number of multi-dimensional headline indicators, such asthose already highlighted in Opportunityfor all, to track improvements in differentaspects of child poverty, for example lowincome, worklessness, education, healthand housing;

• constructing an index that combines asmall number of headline indicators toproduce a single figure to track progress;

• using a headline measure of ‘consistentpoverty’ – similar to the approach inIreland – that combines relative lowincome and material deprivation; and

• using a core set of indicators of lowincome and ‘consistent poverty’.

Executive Summary

5

Each approach has pros and cons. These arediscussed more fully in the paper. You mayhave suggestions about other ways tomeasure poverty in the long-term.

In particular, we would like your responseto the following questions.


1. What aspects of child poverty should be captured in a long-term measure?

2. Do you agree with the criteria for selecting a good indicator set out on page 19?

3. What do you think is the best summary or headline measure to track long-term progressof child poverty?

4. In particular do you have views on the four approaches?

• Option One – multi-dimensional headline indicators

• Option Two – a child poverty index

• Option Three – a headline measure of ‘consistent poverty’, combining relative lowincome and material deprivation

• Option Four – a core set of indicators of low income and ‘consistent poverty’

5. Does the approach you favour capture the factors you listed in response to question 1 andsatisfy most of the criteria that you have highlighted in response to question 2?

6. Do you have any particular views on the geographical coverage within the UK of the fourapproaches?

Details of the consultation arrangementscan be found on page 8 and on theenclosed response form.

Why consult?1. In March 1999, the Prime Minister

announced the Government’scommitment to eradicate child povertywithin a generation. We have alreadymade good progress but there is still along way to go to achieve our goal.As we move forward, we want to besure that we are measuring poverty in away that helps to target effectivepolicies and enables the Governmentto be held to account for progress.

2. This is far from straightforward.Debates about how to measure povertyhave been going on for many years.Experts in the field differ in their viewsand a range of approaches has beenadopted internationally. We havesought to learn from these differencesand encouraged experts to engage withus in thinking about how theoreticaldebates can be turned into a practicalblueprint for poverty measurement inthe UK.

3. Three events have been especiallyhelpful in this regard:

• a July 2000 workshop, which we co-hosted with the Centre for Analysisof Social Exclusion (CASE) at theLondon School of Economics (LSE)1;

• a November 2001 workshop hostedby the Institute for Public PolicyResearch (IPPR)2; and

• the launch of Tackling child poverty:giving every child the best possiblestart in life in December 20013.

4. The feedback from these events, andfrom commentators more widely, is thatpoverty is about much more than justlow income at one point in time.This supports the multi-dimensionalindicators approach in our annualOpportunity for all reports4. However,at the same time some commentatorshave highlighted the merit in having asingle measure to monitor progresstowards the eradication of childpoverty.

5. This consultation paper is for anyonewho is interested in measuring progressin tackling child poverty. There is clearlyan important debate to be had on thisissue and we would now like to hearyour views more formally.

Geographical scope 6. This paper seeks views about a long-

term measure of child poverty for theUK Government. However, theprovision of services for children inScotland, Wales and Northern Ireland,

Introduction

7

such as education and health, aregenerally matters for the DevolvedAdministrations. Consequently,references in the paper to theinformation and statistics for theseservices apply to England only. However,the discussion on the four suggestedways of measuring child poverty brieflyoutlines how the indicators would workacross the UK.

How to respond7. Please send your responses on the

enclosed form to:

The Child Poverty MeasurementConsultation Team Department for Work and PensionsRoom 521The Adelphi1–11 John Adam StreetLondon WC2N 6HT

8. Copies of the consultation documentand response form can also be found at:http://www.dwp.gov.uk/consultations/2002

You can also send responses using theelectronic form to the following e-mailaddress: [email protected]

Comments should be received by10 July 2002. We will acknowledgereceipt of your comments.

What will happen next?9. Responses will usually be available to

the general public unless youspecifically ask us to keep your viewsconfidential.

10. We hope to host events to discuss theissues outlined in the document andwill announce further details in duecourse.

11. We expect to make known ourpreliminary conclusions later this year.Further technical work, involvingexperts in this field, will be necessarybefore finalising the precise details ofany new measures.


12. This section summarises our strategy fortackling child poverty and our currentapproach to monitoring progress.It then discusses the challenge we faceand highlights some internationalapproaches to poverty measurement.

Child poverty 13. By the mid-1990s, the UK had one of

the highest proportions of childrenliving in low-income households among

the member countries of theOrganisation for Economic Co-operation and Development(OECD)5 and more than double thelevel two decades earlier (see chart)6.

14. Low income is a key aspect of childpoverty. It influences outcomes not onlyduring childhood but also intoadulthood. It will therefore remain acentral part of any long-term povertymeasure.

Our starting point

9

Percentage of children living in low-income households

Year

Before Housing Costs After Housing Costs

Source: Family Expenditure Survey 1979–1993/94, Family Resources Survey 1994/95–2000/01. Note: Low income is defined as below 60 per cent of contemporary median equivalised household income.

Perc

enta

ge o

f ch

ildre

n

0

5

10

15

20

25

30

35

40

1999

/2000

2000

/01

1998

/99

1997

/98

1996

/97

1995

/96

1994

/95

1993

/94

1992

/93

1991

/92

1990

/91

1988

/8919

8719

8119

79

15. The length of time spent in low incomeis also important, with those sufferinglonger spells finding it more difficult tomove out of low income and beingmore likely to suffer adverse outcomes.Research shows that most peopleentering low income could expect tosuffer relatively short spells, but asignificant minority remains in lowincome for longer spells. Movements inand out of low income are associatedwith key trigger events, such asentering or leaving the labour market,the birth of a child or maritalbreakdown7.

16. However, other influences also matter,such as neighbourhood environments,lack of opportunities to acquireknowledge and skills, and inequalities inhealth. Those living in poorerneighbourhoods often have to put upnot only with a rundown physicalenvironment and limited opportunities,but also the worst public services,including education, health andhousing.

17. A growing body of research indicatesthat children’s life chances are affectedby the circumstances in which theygrow up as well as by their own talentsand efforts.

• Poverty in childhood increases thelikelihood of low income inadulthood. There is a strongassociation between children’ssubsequent earnings and those oftheir parents8. Those who grow up inpoor families are more likely to beeconomically inactive in young

adulthood. This is particularly true ofthose who experienced poverty eitherin pre-school years (ages 0–5) or inadolescence (ages 11–15)9.

• There is evidence of a strongrelationship between parental incomeand early childhood educationalattainment. Early age test scores arepositively correlated with subsequenteconomic success10.

• Boys growing up in poor families aremore likely to have contact with thepolice than those who are not frompoor families. Contact with the policeduring childhood is one predictor ofoutcomes during adulthood11.

• Children from poor families havelower expectations about theirfuture12.

• Children who grow up poor are morelikely to have lower self-esteem, playtruant and plan to leave school atage 1613. Individuals who leaveschool with low levels of educationalattainment are at higher risk ofexperiencing social exclusion as adultsand have significantly lower lifetimeearnings14.

• There is a clear intergenerationaltransmission of poverty throughworklessness. Those people growingup in a family experiencingunemployment are about twice aslikely to have prolonged spellsthemselves. Those from poorerbackgrounds are significantly morelikely to end up not in work thanthose with higher income fathers15.


18. Action to tackle child poverty musttherefore raise the incomes of thepoorest families and invest in highquality public services to enable childrento reach their full potential as adults tobreak the cycle of poverty anddisadvantage. And extra resources needto be targeted at neighbourhoods thatneed most help – it is our aim thatwithin 10 to 20 years no one shouldbe seriously disadvantaged by wherethey live16.

19. Eradicating child poverty is a significantand demanding undertaking. We knowthat it will take time, but we havealready taken significant steps. Thefollowing box highlights key policies wehave introduced since 1997 to tacklechild poverty.

11Our starting point

Making a difference

Since 1997 we have made important inroads into the problems of poverty and socialexclusion that too many children in this country were facing. We have put into place acomprehensive strategy to improve the living standards of today’s poorest children and tobreak the cycle of disadvantage so that future generations will not grow up in poverty.

At the centre of our strategy is the belief that, for most families, work is the best route out ofpoverty. A stable and growing economy is important. And our active labour market policies,such as the New Deals and policies to improve adult skills, are ensuring that parents have morehelp than ever before in their search for work. And policies such as the Working Families’ TaxCredit and the National Minimum Wage ensure that families are better off when they get intowork. The Working Families’ Tax Credit, worth an extra £35 a week on average when comparedto the old Family Credit, is benefiting nearly 1.3 million families with 2.5 million children.

Alongside making work pay we are helping all families with children through the tax andbenefit system, in particular targeting help at those who need it most, when they need itmost. For example:

• we have increased the child allowances in Income Support and other income-relatedbenefits, with rates for children aged under 11 rising by 80 per cent in real terms;

• as a result of personal tax and benefit reforms announced in the last Parliament, familiesin the poorest fifth of the population are now on average £1,700 a year better offcompared to 1997; and

• the Sure Start Maternity Grant paid to recipients of Income Support, income-basedJobseeker’s Allowance, Disabled Person’s Tax Credit and Working Families’ Tax Credit,increased to £500 for each child in April 2002, up from £300.

Budget 2002 takes further steps to tackle child poverty, in particular through the new taxcredits. Details of rates and thresholds are set out in Budget 2002.


We recognise that if we are to improve the quality of life for all children and young people,we must marry measures to improve family incomes with increased investment in services.Improving public services for children and their families, regardless of where they live, remainsat the core of our agenda for change.

All children deserve the opportunity to benefit from a high quality education. Providingchildren with a high standard of education, irrespective of family background and where theylive, is essential and underpins our central strategy to tackle the root causes of poverty. Weare investing record amounts in education and have schemes in place to help schools in themost deprived areas. We have increased investment in both housing and the health service.Whilst these investments are long-term plans to ensure that our public services are fit for thetwenty-first century, we have linked them to short-term targets to ensure continuousimprovements. We are already seeing results and will build on this progress.

• We have introduced our Sure Start programme in deprived areas to help ensure that childrenin these areas receive the best possible start in life and are not already disadvantaged by thetime they start school. By 2004 there will be 500 Sure Start areas up and running.

• There have been significant improvements in educational attainment for all age groups.In particular, the proportion of 11-year-olds achieving the expected standards in literacyis up from 63 per cent in 1997 to 75 per cent in 2001, and in numeracy from 62 per centto 71 per cent. The number of children permanently excluded from school has fallen from12,700 in 1996/97 to 8,300 in 1999/2000.

• Housing is an important element of tackling poverty and social exclusion. We have begunto address the huge backlog in repairs in social housing, and we are on course to bring allsocial housing up to set standards by 2010. Between 2001 and 2004, we will reduce byone third the number of households living in social housing that does not meet thesestandards. This work will complement programmes in health, education and crime.

• Making the transition to adulthood is often a difficult time for young people. As part of ourstrategy we will put in place the help needed to make the transition a success. Building onthe work of the Social Exclusion Unit we have implemented programmes to help tackle theproblems of youth crime, homelessness, drug misuse and teenage pregnancy. And the mostrecent data shows an encouraging fall in the teenage conception rate.

• Over the next few years, we will continue to overhaul the child support system, introducethe Child Tax Credit, increase our investment in education, health and housing, and roll outthe Children’s Fund and the Connexions service to help young people make the transitioninto adulthood.


How we currentlymonitor progress

20. In September 1999 we published ourfirst annual Opportunity for all reportsetting out our strategy to tacklepoverty and social exclusion. This wasthe first time a UK Government hadproduced an annual poverty report with

indicators of progress. The reportincluded 13 indicators to monitor ourprogress in tackling child poverty.We use clear criteria to review theseindicators regularly and now track ourprogress against 15 indicators17.

Opportunity for all indicators

Improving family incomes by tackling worklessness and increasing financial support for families

• Children in workless households

• Low income (relative, absolute and persistent low income)

Investing in the crucial early years and education to break the cycle of disadvantage

• Key Stage 1 (7-year-olds) attainment in Sure Start areas

• Key Stage 2 (11-year-olds) attainment

• 16-year-olds with at least one GCSE

• 19-year-olds with at least a Level 2 qualification

• Truancies and exclusions

• Educational attainment of children looked after by local authorities

Improving the quality of the lives of children and young people

• Housing standards

• Infant mortality

• Smoking rates for pregnant women and children aged 11–15

• Serious unintentional injury

• Re-registrations on the Child Protection Register

Supporting young people in the transition to adult life

• Teenage conceptions and teenage parents in education and employment

• 16–18-year-olds in learning

21. This set of indicators captures the manydifferent aspects of poverty and socialexclusion for children and youngpeople. However, we acknowledge thatit is difficult to summarise overallprogress using such a comprehensiveset of indicators. Therefore in oursecond annual Opportunity for allreport we highlighted a smaller set ofchild poverty headline indicators: lowincome, worklessness, educationalattainment, health inequalities andhousing standards. In the third annualreport we included a table of trends toprovide a summary of progress.

Public Service Agreementtargets

22. As part of Spending Review 2000 weset a number of Public ServiceAgreement (PSA) targets including thefollowing which will contribute toprogress in tackling child poverty18:

• to make substantial progress towardsthe eradication of child poverty, byreducing by a quarter the number ofchildren in low-income households(below 60 per cent of contemporarymedian equivalised householdincome) by 2004 (Great Britain);

• to reduce the number of children inhouseholds with no one in work overthe three years to 2004 (GreatBritain); and

• to ensure that all social housingmeets set standards of decency by2010, by reducing by a third thenumber of households living in socialhousing that does not meet thesestandards between 2001 and 2004,

with most of the improvementstaking place in the most deprivedlocal authority areas as part of acomprehensive regeneration strategy(England).

23. The Secretary of State for Healthannounced two health inequality PSAtargets in February 2001, one relatingto children:

• starting with children under one year,by 2010 to reduce by at least 10 percent the gap in mortality betweenmanual groups and the population asa whole (England).

24. On 13 March 2002, the Secretary ofState for Education and Skillsannounced new targets for literacy andnumeracy beyond 2002:

• to increase the percentage of 11-year-olds who achieve Level 4in the Key Stage 2 English andmathematics tests to 85 per cent by2004; and

• to close the achievement gapbetween different parts of thecountry by expecting all localeducation authorities to have at least78 per cent of their 11-year-olds atLevel 4 or above in English andmathematics by 2004 (England).

25. PSAs provide clear commitments to thepublic and each agreement sets outexplicitly which Departments areaccountable for delivering the targets.This paper does not seek views onthese PSAs, which we are committedto deliver.


The challenge we face26. Our current approach to measuring

poverty presents a key challenge, whichunderlies this consultation.

27. This was summarised by Lisa Harker,Deputy Director of the IPPR19:

“There are a range of indicators ofpoverty and social exclusion set out inthe Government’s annual reportOpportunity for all. This is important fora number of reasons … On the otherhand a single indicator is appealingbecause of clarity. A headline or priorityindicator is needed in order to monitorprogress towards the pledge toeradicate child poverty. In the absenceof any other measure the HouseholdsBelow Average Income measure will beused by politicians, the media andothers as the principal measure ofprogress towards reducing child poverty… The headline indicator would needto be supplemented by other measuresin order to capture broader aspects ofdeprivation and social exclusion.”

28. In the absence of a single measure ofchild poverty, the relative low-incomemeasure receives most attention.But relative low income has somedrawbacks as an exclusive long-termmeasure of child poverty. These werewell summarised by Professor John Hills,Director of CASE and Professor of SocialPolicy at the LSE, at the IPPR conferencein November 2001. While acceptingthat perceptions of poverty are largelyrelative, he identified as problems:

• particular cut-offs (for example,60 per cent of median) are largelyarbitrary;

• ‘income’ can be defined in manydifferent ways;

• ‘depth’ of low income matters (totake account of how far below thelow-income line people fall) as well ashead counts;

• persistence (a long spell in lowincome) matters;

• low income and deprivation overlapbut are not identical; and

• the ‘relative low-income paradox’ – attimes when incomes in general arerising unusually rapidly (for example,in the economic boom in Irelandduring the 1990s), public views ofwhat constitutes poverty may not riseas fast. At such times, the numbersof people with relative low incomemay rise, even though real livingstandards for the poor are increasingsignificantly. The opposite canhappen in a recession (for examplein some Central European countries,such as Hungary or Poland, in thefirst half of the 1990s) when arelative low-income threshold can fall,even though public views of povertyhave not changed20. Consequently,the number of people in relative low-income may fall, giving theimpression of progress, at a timewhen real living standards are notimproving.

29. These comments indicate theimportance of a broad range ofindicators with income at their heart.However, with multiple measures it isdifficult to track overall progress clearly.


International experience31. Before moving on to specific

consultation options for the UK, it isworth reflecting on the measures thathave been adopted internationally.


The United States official poverty line

The US Government adopted the official poverty line in 1968 drawing on research by MollieOrshansky21. She determined a minimum food budget for a family of four and noted thatthis amounted to around a third of family expenditure. The food budget was thereforemultiplied by three to obtain the overall minimum budget, and those with incomes belowthis were counted as poor. Since then the US poverty line has been uprated in line withprices. It is therefore an absolute measure, with a fixed real value, that has not taken accountof changing consumption needs. By the mid-1990s the poverty line had fallen to anequivalent of less than a third of median household income.

A recent review of the official poverty measure convened by the National Research Councilsuggested that it should be revised to reflect not only price change, but also changes in theconsumption of basic necessities22. Using that report as its start, in 1999, the Bureau of theCensus issued an ‘experimental’ new version of poverty figures23. However, changing the USpoverty line is not a straightforward matter. Because the funding of many social programmesis linked to poverty rates methodological changes would have different financial implicationsfor different States24.

30. The challenge that we must meet is tofind the right balance between thedesire for clarity, which argues forone clear headline indicator, andcomprehensiveness, which arguesfor a broader range of indicators.International and UK experts advocatedifferent approaches to achieve thisbalance.

contemporary median equivalisedincome before housing costs) as anindicator to compare Member States ofthe EU. Other indicators of low income,such as persistent low income anddistribution of income have also beenagreed by the EU.

32. In contrast to adopting a singleabsolute poverty measure, MemberStates of the European Union (EU) haveagreed a multi-dimensional range ofindicators. In the domain of lowincome, Eurostat has adopted a relativemeasure (based on 60 per cent of


Measuring poverty and social exclusion in the European Union

During 2001 the indicators’ sub-group of the EU’s Social Protection Committee worked toproduce a set of commonly agreed indicators of poverty and social exclusion for MemberStates of the EU. In December 2001, the report of this group was endorsed by the LaekenEuropean Council25.

These indicators focus on social outcomes rather than the means by which they are achieved.Because of the decision to select a large number of indicators to properly assess the multi-dimensional nature of poverty and social exclusion, the indicators are prioritised into threelevels.

• Primary indicators would consist of a restricted number of lead indicators.

• Secondary indicators would support the lead indicators and describe other dimensions.Both primary and secondary indicators would be commonly agreed and defined indicators,used in future National Action Plans.

• A third level of indicators would be selected by Member States themselves to support theprimary and secondary indicators or highlight progress in particular aspects of poverty andsocial exclusion.

The primary and secondary indicators encompass different measures of low income,employment, educational attainment and health outcomes.

33. The EU indicators will, however, bemost useful at facilitating comparisonbetween the different Member States.This does not preclude the developmentof other purely national indicatorstailored to best monitor the situation inthe UK.

34. The Irish Government has adopted adifferent approach to measure poverty,which combines relative low incomeand material deprivation into a conceptof ‘consistent poverty’.


The Irish Government’s anti-poverty target

The Irish Government has framed its official headline poverty target in terms of a measure of‘consistent poverty’ developed at the Economic and Social Research Institute (ESRI), Dublin.People who are ‘consistently poor’ suffer both low income (in a relative sense) and materialdeprivation. This headline indicator is also supplemented by a set of indicators capturing thewider aspects of poverty, such as education.

The Irish Government bases its choice of deprivation indicators on an analytical techniquecalled factor analysis. This uses statistical analysis to establish a cluster of deprivationindicators that correlate well with what we mean by being ‘poor’. This analysis does notattempt to list all the items that people should have in order to avoid poverty. Instead it seeksto identify a set of key items that can be used to identify those who suffer materialdeprivation and have low incomes. The items identified by the analysis as acting as a proxyfor material deprivation in Ireland are:

• unable to afford new (not second-hand) clothes, two pairs of shoes, meat, chicken, fishevery second day, warm overcoat, roast once a week;

• went without substantial meal;

• went without heating; and

• in debt/arrears to meet ordinary living expenses.

People are classified as deprived if they go without at least one of these basic items becausethey want it but are unable to afford it. Not having an item because of choice wouldtherefore not constitute deprivation26.

Professor Brian Nolan, Research Professor at the ESRI in Dublin, discussed the IrishGovernment’s poverty targets at workshops in London in 2000 and 200127.

The following data compares movements in ‘consistent poverty’ counts with absolute andrelative low income counts for Ireland.

1987 1994 1997 2000Relative low income 19% 21% 22% 26%

‘Consistent poverty’ 16% 15% 10% 6%

Absolute low income – 15% 8% 3%

Note: Relative low income is measured below 50 per cent of contemporary mean. ‘Consistent poverty’ is measuredas income below 60 per cent of contemporary mean and suffering material deprivation as described. Absolute lowincome is measured using 1994 as the base year and uprating the 1994 low-income threshold by prices (rather thanincomes) to obtain the low-income thresholds in 1997 and 2000.

These figures highlight apparently paradoxical results of a relative low-income measure ofprogress during the sort of economic boom experienced in Ireland. Because incomes at thebottom of the distribution have not kept pace with the average, the relative low-incomeindicator does not reflect the substantial material gains among those on low incomes. Incontrast, the absolute low-income measure shows progress between 1994 and 2000, and the‘consistent poverty’ measure captures progress in tackling deprivation over the same period.

19

35. This section sets out options for a long-term measure of child poverty for theUK Government. Before looking atoptions on which we wish to consultwe set out the desired qualities of along-term measure of child poverty.

What makes a goodindicator?

36. A detailed discussion of the underlyingprinciples of poverty and social exclusionindicators is found in the 2001 reportby Professor Sir Tony Atkinson andcolleagues28. Professor John Hillssuggests that official measures shouldsatisfy four criteria: political credibility;policy credibility; public credibility; andtechnical credibility29.

37. Drawing on these papers a goodmeasurement approach should aim to:

• encompass the different dimensionsof child poverty;

• be readily summarised whereappropriate so that overall progresscan be identified and explained;

• be based on child outcomes ratherthan processes – for example, thenumber of children living inhouseholds where no adult worksrather than the number of loneparents being helped to find workthrough the New Deals;

• be unambiguous in interpretation – ifpolicies are working there should bean improvement in the indicator; and

• have longevity, being relevant nowand to track long-term progress.

38. In addition, the detailed indicators andstatistics that lie behind any approachshould be:

• timely;

• open and robust to statistical scrutinyfrom experts;

• be credible with the public;

• be capable of generating a long-termrobust time series;

and if possible:

• be capable of disaggregation bygroup and by locality; and

• be internationally comparable.

39. Whilst these are ideal qualities of along-term child poverty measure, somemay be difficult to achieve at the sametime. For example, a comprehensiveapproach may not be readilysummarised. The most sophisticatedstatistics that capture the mostimportant aspects may not be the mosttimely. The most robust measuresnationally may not be readily availablefor smaller areas.

Consultation options

40. In practice it is possible that we may needto compromise and find an approachthat satisfies most of these criteria.

41. However, it is important to consider thispaper in the context of the widerGovernment strategy for children anddata development. As a response to thePolicy Action Team 18 report, theNeighbourhood Renewal Unit inpartnership with the Office for NationalStatistics, central governmentdepartments and the Local GovernmentAssociation is taking forward work toimprove data for small areas. Thisinitiative aims to enable more effectiveidentification of deprived areas, therebyensuring that Government action,centrally and locally, is appropriatelyfocused and monitored30. This data andsubsequently developed indicators willcomplement national measures.

42. And in Building a Strategy for Childrenand Young People31, the Children andYoung People’s Unit (CYPU) consultedabout a framework for monitoringoutcomes for children and youngpeople. The outcomes were provisionallygrouped into six key areas: health andwell-being; achievement and enjoyment;participation and citizenship; protection;responsibility; and inclusion. Theconclusions of this consultation willcomplement the work already beingtaken forward by the CYPU.

43. Having considered the criteria on page19, we welcome your views on thefollowing four approaches to measuringchild poverty:

• Option One – multi-dimensionalheadline indicators;

• Option Two – a child poverty index;

• Option Three – a headline measureof ‘consistent poverty’, combiningrelative low income and materialdeprivation; and

• Option Four – a core set of indicatorsof low income and ‘consistentpoverty’.

44. The first approach draws closely on theexisting Opportunity for all indicators,and subsequent approaches build onthis or would require the developmentof new measures. You may have otherideas about how we should measurepoverty and we would also beinterested to hear about theseapproaches. Whichever approach youfavour we would like you to considerthe following key issues.

What key aspects should a long-termmeasure of child poverty capture? Doesyour preferred measure capture them?

Do you agree with the criteria set out onpage 19? Does the approach you favoursatisfy most of these criteria?


21Consultation options

Option One – multi-dimensionalheadline indicators

45. We could refine our existing approach.In the second Opportunity for all reportpublished in September 2000, wehighlighted five headline indicators in,what we consider to be, key areas:

• low income;

• worklessness;

• educational attainment;

• health inequalities; and

• housing standards.

46. These headline indicators are linked toPSA targets, which quantify theimprovements to be achieved by 2004or 2010 (see page 14). We arecommitted to deliver the resources tomeet these PSA targets and are notconsulting on them. But we could nowthink about how a set of headlineindicators might be used to monitorchild poverty in the longer-term.

Advantages 47. The headline indicators aim to capture

the different aspects of poverty andsocial exclusion that affect children’slives. The indicators measure outcomesrather than processes. They capturedimensions that affect current livingstandards and factors that are likely tolead to worse outcomes duringadulthood. The approach of using arange of indicators to monitor progress,and prioritising some as lead indicatorsis consistent with the approach agreedby Member States of the EU (seepages 16 and 17).

48. The headline indicators approach isbased on existing data sets withestablished baselines and historical timeseries. The indicators providetransparent measures of progress inparticular aspects of poverty.

49. This approach works well to distinguishbetween devolved and reservedmatters. Some of the headlineindicators, such as low income andworklessness, would extend to GreatBritain as a whole. Other indicatorsrelate to England only.

Challenges 50. The headline indicators approach does

not produce a single figure to trackprogress. In the absence of a singlemeasure it is difficult to decide howlong-term success would be measuredusing multiple indicators.

Option Two –a child poverty index

51. A child poverty index could be used toreduce the headline indicators ofOption One into a single measure. Suchan index would assign weights to eachindividual indicator and combine theminto a single composite measure.Changes in the indicators could then betranslated into proportional changes inthe index. Analysis of each individualindicator would still be important tounderstand where progress has beenmade.

52. The use of summary measures to trackprogress was discussed at the thenDepartment of Social Security and CASEworkshop in July 2000. An analysis ofthe advantages and disadvantages of

this approach were presented in papersby John Micklewright (Head of Researchat the UNICEF Innocenti ResearchCentre, Florence) and Helen Barnes(now at the Policy Studies Institute)32.

53. A child poverty index could provide asingle measure of progress. Perhaps themost well-known index in a similar field(though designed for a differentpurpose) is the United NationsDevelopment Programme’s HumanDevelopment Index.


The Human Development Index

Extracts from John Micklewright’s paper:

“I don’t think anyone would seriously suggest that progress in realising the UK Government’scommitment to reduce poverty and social exclusion should be measured only by a compositeindex – a single number summarising all aspects of these two problems. At best, an indexcould complement the presentation of a range of separate indicators and my comments onthe subject of indices should be interpreted in this light …

“it is useful to bear in mind the most obvious example of where an index of well-being hasbeen used to great effect, the United Nations Development Programme’s (UNDP) ‘HumanDevelopment Index’ (HDI).

“It is important to note that the HDI has always been seen by its proponents as only thestarting point for debate on the progress of human development – as a device to catch theeye and force attention on what lies behind the rankings, i.e. on the separate indicators thatgo into its construction. Amartya Sen recounts how he was originally a sceptic whenconfronted with the enthusiasm of the creator of the Human Development Report, the lateMahbub ul Haq, for the use of a composite index. But he was converted when he saw thepower of the HDI to channel public interest into the different dimensions of well-beingcovered by the report and away from the default of falling back on GNP as a summarymeasure of development (UNDP 1999, p23).

“The HDI is arguably most effective when used to summarise differences among countriesrather than differences across time.“


Advantages 54. A child poverty index would capture the

different dimensions of child poverty.Arguably this approach, by producing asingle number that could be used totrack progress over time, haspresentational and communicationadvantages. This is well illustrated bythe Retail Price Index (RPI), which isused by the Bank of England to monitorinflation. The RPI is calculated using theweights from the Family ExpenditureSurvey, and components of the indexare also published.

Challenges55. However, as highlighted above, some

would argue that it is not meaningfulto combine different aspects of povertyand social exclusion, purely to producea single figure to track progress. Thereare a number of other issues that couldcomplicate the construction of anindex.

• When constructing an index,choosing the relative importance(weight) of each component issomewhat arbitrary. In particular,direction of movement of an indexmay not be robust to the choice ofweights. And if the individualindicators capture overlapping orcorrelated factors, such as lowincome and worklessness, double-counting needs to be considered.

• It is also necessary to decide thedegree of substitutability to allowbetween the elements that areselected for inclusion in an index.

This issue is distinct from that ofweighting. For example, should therebe full or partial trade-off between apercentage point improvement in onecomponent with a percentage pointworsening of another.

• The time period measured by data fordifferent indicators varies. Forexample, low incomes are monitoredusing data for financial years, worklesshouseholds are monitored using springquarter data, educational attainment ismonitored using assessments duringthe academic year.

56. Another challenge relates to thegeographical coverage of a childpoverty index within the UK. Some ofthe headline indicators would extend tothe UK as a whole, whilst others wouldnot. It might, therefore, be necessary toconstruct separate child poverty indicesfor each of Scotland, Wales, NorthernIreland and England. We would need todiscuss this possible approach with theDevolved Administrations.

57. In addition to challenges with theconstruction of an index, there could beproblems with interpretation. The trendof an index is only meaningful if allindicators move in the same direction,otherwise improvements in oneindicator can be masked by negativemovements in others.

Option Three – combiningrelative low income andmaterial deprivation –‘consistent poverty’

58. Research shows that those who sufferlonger spells of poverty will suffer moreadverse outcomes than those who arein poverty for short periods of time33.In the domain of low income, this ideaof persistence is already captured in theOpportunity for all indicators, whichincludes a measure of those in lowincome in three out of four years. Manywould argue that a snapshot of incomeat a particular moment does notprovide the best measure of people’sliving standards.

59. There are other ways to monitor theidea of persistence. Those sufferingrelatively short spells of low incomemay be able to smooth outconsumption over a longer time-scale.

Those who suffer longer spells of lowincome are more likely to go withoutessential items. This can be monitoredusing material deprivation measures.These measures establish a selection ofitems that are ‘necessities’. People arethen classified as ‘deprived’ if they gowithout some of these items becausethey are unable to afford them.

60. Interest in deprivation indicators hasbeen growing since Townsend’s study in197934. His intention was to developindicators of objective deprivation tomeasure when individuals lack anamenity or do not participate in anactivity that the majority of thepopulation possess or participate in.An alternative approach was developedby Mack and Lansley in 1985 usingsocially prescribed necessities asdeprivation indicators35. They usedthese to directly identify the poor36.


The Poverty and Social Exclusion Survey

More recently the Joseph Rowntree Foundation published analysis of the Poverty and SocialExclusion Survey37. The study largely repeats the 1983 and 1990 Breadline Britain surveys,albeit with different sampling methods.

The analysis sought to identify enforced lack of necessities by presenting respondents with alist of items and asking them to distinguish items they thought were necessary in Britaintoday (i.e. that all adults should be able to afford and not have to go without) and those thatwere not. They were then asked to sort the same items into three groups: those they had,those they did not have but did not want, and those they did not have and could not afford.

Items were selected into the list of necessities if over 50 per cent of people deemed themnecessities. There were 35 items in this basket ranging from beds and bedding, refrigerator,two meals a day to an outfit for social occasions and a holiday away from home once a year.Unweighted deprivation scores were ranked against income to produce a poverty threshold.The authors consequently defined ‘poverty’ as lacking two or more items.

61. In addition to this and the IrishGovernment’s measure, the workconducted by the Policy Studies

Institute using Families and ChildrenSurvey data provides an alternativeillustration of deprivation indicators38.


Families and Children Survey hardship measure

The Families and Children Survey (FACS) started in 199939. It was initially designed to berepresentative of all lone parent families and low- to moderate-income couples with children.From 2001, the third wave, the survey was extended to be representative of all families withchildren across the whole income distribution. FACS provides a good source of annuallycollected deprivation data for families with children in Great Britain. It includes 34 questionscovering four areas of spending – clothing, food, leisure and consumer durables. As well asthese items, FACS collects information about housing conditions, debts and moneymanagement. These are used by the Policy Studies Institute to derive a hardship index withinthe FACS report.

The FACS hardship measure uses nine indicators covering:

• accommodation and warmth;

• money and debt; and

• material hardship scores for clothing, food, leisure and consumer durables.

Prevalence weightings are assigned to items in the material hardship indicators. Eachrespondent who cannot afford but wants an item is given a score equivalent to thepercentage of families who said they owned the item.

The index was designed to tap into different dimensions of living standards and to identifythose with multiple problems and those who are consistently worse-off – in particular toidentify differences between those in and out of work and to highlight problems at the lowerend of the income distribution.

62. Our current approach to monitoringchild poverty in the UK does not includea measure of material deprivation. TheIrish Government’s headline anti-povertytarget incorporates a measure of materialdeprivation (see pages 17 and 18).This defines people as ‘consistentlypoor’ if they suffer both relative lowincome and material deprivation.

63. Currently there is no combined annuallycollected income and deprivation datafor the whole population of GreatBritain. The most robust source ofincome statistics for Great Britain arethose published annually in Households

Below Average Income40. This analysisis based on the Family Resources Survey(FRS), which does not currently includedeprivation indicators. However, ifdeprivation indicators were added tothe FRS it would be possible toconstruct a ‘consistent poverty’ measurein the future.

Advantages 64. Material deprivation measures resonate

well with the perception that povertyshould encompass some idea of thepractical effects that result from lowincome.

65. It is widely accepted amongst incomedistribution statisticians that measuredincomes do not always reflect livingstandards, especially at the bottom ofthe income distribution41. This mayreflect a range of factors, including:incorrect reporting of incomes; timelags between changes in incomes andliving standards; and the influence ofnon-monetary factors, such as prioritiesfor access to social housing. Arguably abetter measure of living standards, atany given point, can be obtained bymeasuring both low income andmaterial hardship.

66. A ‘consistent poverty’ measure,combining relative low income andmaterial deprivation, could provide aheadline measure to monitor the long-term pledge to eradicate child poverty.As discussed in Option Four, this couldsit amongst the set of multi-dimensional Opportunity for allindicators that would continue to beused to monitor progress in the otherdimensions of poverty and socialexclusion.

Challenges67. The construction of such an indicator

would clearly generate debate. Theprocess for deciding items to beincluded in the deprivation measurewould be key. It is also important toconsider carefully how the measuremight be updated to reflect currentneeds and how broad the selection ofitems should be to reflect social andcultural as well as physical needs.

68. There are a number of methodologiesthat could be used to select deprivationitems. One is the statistical approach offactor analysis, which is a data-drivenapproach to group items according tounderlying factors. For example, analysisconducted by ESRI in Dublin shows thatdata for Ireland clustered into threedomains of basic deprivation, secondarydeprivation and housing deprivation.The measure adopted by the IrishGovernment focuses on basicdeprivation42. A second approach is toask people which items, in theiropinion, are necessary to avoiddeprivation and then selecting thosewhich the majority view as necessities43.

69. Statistical approaches are lesstransparent than asking people theiropinion. However, there are alsodrawbacks to just taking a majorityview. For example, some of the itemsdeemed as necessities by just over halfof those asked are still considered notto be necessities by a significantproportion of respondents. It isimportant to bear in mind that the aimis not necessarily to obtain acomprehensive list of items that peopleneed but to identify a set of items thatindicate deprivation.

70. Policies underlying some deprivationfactors that could be included in thissort of measure are the responsibilityof the Devolved Administrations.Therefore, when constructing anindicator of this nature, the detailswould need to be discussed with theDevolved Administrations.



Option Four –a core set of low incomeand ‘consistent poverty’indicators

71. A ‘consistent poverty’ measure, asoutlined in Option Three, could sitamong the existing set of Opportunity

72. The UK has a poor child poverty record(measured in terms of low income)compared to OECD countries, falling inthe bottom four of a relative childpoverty league table, using data for theearly and mid-1990s and the bottom sixof an absolute child poverty leaguetable45. Using international comparisonsto provide a benchmark for long-termprogress of relative and absolute lowincome would be one way ofpresenting progress in context.

73. Because this approach encompassesOption Three, readers are referred tothe discussion of the advantages and

Core indicators

At the workshop hosted by IPPR in November 2001, Professor John Hills presented anapproach to measuring child poverty that builds on the Government’s existing approach. Thisincludes three main tests of long-term progress44:

• an absolute low-income indicator, fixed in real terms, against which there must beprogress;

• a measure similar to that used in Ireland, combining indicators of material deprivation withrelative income, to show good progress, particularly during times of fluctuating growthrates; and

• a relative income measure to take account of rising living standards and changingexpectations across society over time. EU monitoring of households below 60 per cent ofmedian income could offer an international benchmark.

As currently set out in Opportunity for all there would be indicators of other dimensions,such as employment, education, health and housing.

for all indicators (listed on page 13 anddiscussed in Option One). Long-termprogress in child poverty could then bemonitored with reference to a core setof low income and ‘consistent poverty’indicators.

challenges of constructing ‘consistentpoverty’ measures (pages 25 and 26)as well as the advantages andchallenges discussed here.

Advantages74. This tiered approach would show a

gradient of progress, from the absolutelow-income indicator through to therelative income measure, to takeaccount of rising living standards overtime. The consistent poverty measurecould better reflect progress duringtimes of fluctuating growth rates.


Challenges75. One risk of this approach is that it may

not provide a clear single headlinefigure to track progress. As now, it islikely that there will continue to begreater lags in producing internationallycomparable low-income data thannational data. This will delay the abilityto make international comparisons oflow-income indicators.

1. What aspects of child poverty should be captured in a long-term measure?

2. Do you agree with the criteria for selecting a good indicator set out on page 19?

3. What do you think is the best summary or headline measure to track long-term progressof child poverty?

4. In particular do you have views on the four approaches?

• Option One – multi-dimensional headline indicators

• Option Two – a child poverty index

• Option Three – a headline measure of ‘consistent poverty’, combining relative lowincome and material deprivation

• Option Four – a core set of indicators of low income and ‘consistent poverty’

5. Does the approach you favour capture the factors you listed in response to question 1 andsatisfy most of the criteria that you have highlighted in response to question 2?

6. Do you have any particular views on the geographical coverage within the UK of the fourapproaches?

Details of the consultationarrangements can be found on page 8and on the enclosed response form.

Key consultationquestionsThe options above explore somedifferent approaches to measuring childpoverty. In particular, we would likeyour response to the followingquestions.

29

Absolute low-income indicators monitorprogress against a fixed real-terms low-income threshold. In the Opportunity for allindicators the low-income thresholds(50, 60 and 70 per cent of median and40, 50 and 60 per cent of mean) are fixedat 1996/97 levels. The low-incomethresholds for subsequent years are thenobtained by uprating these thresholds byprices (rather than incomes).

Consistent poverty is the definitionelaborated by the Economic and SocialResearch Institute, Dublin, and used inthe Irish National Anti-Poverty Strategy.It identifies as poor those households thatare below the relative income thresholdand are also experiencing basic deprivation(enforced lack of one or more items in thebasic deprivation dimension – see Factoranalysis).

Material deprivation consists in theenforced lack of items identified asnecessities. The absence is enforced in thesense that it is due to lack of resourcesrather than a matter of choice.

Factor analysis is a statistical techniquethat seeks to cluster variables (in thisinstance, deprivation items) according to alimited number of underlying factors. In thisway, for example, the Irish Economic and

Social Research Institute was able to identifythree different dimensions of deprivation,i.e. basic, secondary, and housingdeprivation, and to group the deprivationindicators accordingly.

Persistent low income refers to a situationwhereby households remain below the low-income threshold over a prolonged periodof time. In Opportunity for all persistentlow-income indicators measure theproportion of children, working-age adultsand pensioners who have lived in householdswith low income in at least three out offour years, based on a before housing costmeasure for 60 and 70 per cent of medianincome.

Relative low-income indicators measurethe percentage of the population whoseincome is lower than a certain proportionof average income. In Opportunity for allthe relative low income thresholds used are40, 50 and 60 per cent of mean incomeand 50, 60 and 70 per cent of medianincome, based both on before and afterhousing costs measures.

Glossary


11 Hobcraft J, 1998, Intergenerational and life-coursetransmission of social exclusion: influences ofchildhood poverty, family disruption, and contactwith the police, CASE paper 15, LSE.

12 Shropshire J and Middleton S, 1999, Smallexpectations: learning to be poor? Joseph RowntreeFoundation.

13 Ermisch J, Francesconi M and Pevalin D, 2001,Outcomes for Children of Poverty, DWP ResearchReport No. 158, Corporate Document Services.

14 Sparkes J, 1999, Schools, Education and SocialExclusion, CASE paper 29, LSE. Dearden L,McIntosh S, Myck M and Vignoles A, 2000, TheReturn to Academic, Vocational and Basic Skills inBritain, Skills Task Force Research Paper 20, DfEE.

15 Gregg P and Machin S, 1998, Child Developmentand success or failure in the Youth Labour Market,Centre for Economic Performance Discussion Paper397. Johnson P and Reed H, 1996, IntergenerationMobility among the Rich and the Poor: results fromthe NCDS, Oxford Review of Economic Policy (12.1).

16 A New Commitment to Neighbourhood Renewal:National Strategy Action Plan, 2001,http://www.neighbourhood.dtlr.gov.uk/actionplan/index.htm

17 Endean R, 2001, Opportunity for all: Monitoringthe Government’s strategy to tackle poverty andsocial exclusion (in Indicators of Progress,A discussion of approaches to monitor theGovernment’s strategy to tackle poverty and socialexclusion, Report of the workshop held on 19 July2000 organised by DSS and CASE, LSE).http://www.dwp.gov.uk/publications/dwp/2001/iop/iop.pdf

18 See http://www.hmt.gov.uk/Documents/Public_Spending_and_Services/Public_Service_Agreements_2001-2004/pss_psa_whitepaper.cfm

References1 See http://www.dwp.gov.uk/publications/

dwp/2001/iop/iop.pdf

2 See http://www.ippr.org/research/index.php?project=73&current=24

3 Tackling child poverty: giving every child the bestpossible start in life, A Pre-Budget ReportDocument, 2001, HM Treasury.

4 Opportunity for all – tackling poverty and socialexclusion, 1999 (Cm 4445), Opportunity for all –one year on: making a difference, 2000 (Cm 4865),Opportunity for all – making progress, 2001(Cm 5260), http://www.dwp.gov.uk/publications/dwp/2001/oppall_third/index.htm

5 UNICEF Innocenti Report Card Issue No. 1, 2000,A league table of child poverty in rich nations.http://www.unicef-icdc.org/publications/pdf/repcard1e.pdf

6 Households Below Average Income 1994/5 –1999/00, 2001, DWP, Corporate DocumentServices. Low income for this statistic is defined asliving in a household with equivalised householdincome below 60 per cent of median income afterhousing costs.

7 Jenkins S and Rigg J, 2001, The Dynamics ofPoverty in Britain, DWP Research Report No. 157,Corporate Document Services.

8 Johnson P and Reed H, 1996, IntergenerationMobility among the Rich and the Poor: results fromthe NCDS, Oxford Review of Economic Policy(12.1).


10 Machin S, Childhood Disadvantage andIntergenerational Transmissions of Economic Statusin Atkinson A B and Hills J, 1998, Exclusion,Employment and Opportunity, CASE paper 4, LSE.

31References

19 Harker L, 2001, Measuring wider aspects of povertyand social exclusion (in Indicators of Progress,A discussion of approaches to monitor theGovernment’s strategy to tackle poverty and socialexclusion, Report of the workshop held on 19 July2000 organised by DSS and CASE,. LSE). This wasalso discussed at the launch of Tackling childpoverty: giving every child the best possible start inlife, A Pre-Budget Report Document in December2001.

20 See also the box ‘Poverty Paradox’ in UNICEFInnocenti Report Card Issue No. 1, 2000, A leaguetable of child poverty in rich nations.http://www.unicef-icdc.org/publications/pdf/repcard1e.pdf

21 Orshansky M, 1963, Children of the Poor, SocialSecurity Bulletin, 26 (July): 3–13. Orshansky M, 1965,Counting the poor: another look at the povertyprofile, Social Security Bulletin, 28 (Jan): 3–29.

22 Citro C and Michael R (eds), 1995, MeasuringPoverty: A New Approach. National Assembly Press.

23 Short K, Garner T, Johnson D and Doyle P, 1999,Experimental Poverty Measures, 1990–97, USBureau of the Census, p60–205. US GovernmentPrinting Office.

24 Glennerster H, 2000, US Poverty Studies andPoverty Measurement: The past twenty-five years,CASE paper 42, LSE.

25 See http://register.consilium.eu.int/pdf/en/01/st13/13509en1.pdf. The group commissioned areport to contribute to its work, which waspresented at a Conference on Indicators for SocialInclusion: Making Common EU Objectives Work inAntwerp on 14/15 September 2001. Atkinson T,Cantillon B, Marlier E and Nolan B, 2001, Indicatorsfor Social Inclusion in the European Union, OxfordUniversity Press.

26 Nolan B and Whelan C, 1996, Resources.Deprivation and Poverty, Clarendon Press.

27 See http://www.ippr.org/research/files/team24/project73/brainnolanpres.pdf. See also thediscussion in Nolan B, Measuring and Targetingpoverty: An Irish Example (in Indicators of Progress,A discussion of approaches to monitor theGovernment’s strategy to tackle poverty and socialexclusion, Report of the workshop held on 19 July2000 organised by DSS and CASE, LSE).

28 Atkinson T, Cantillon B, Marlier E and Nolan B,2001, Indicators for Social Inclusion in theEuropean Union, Oxford University Press.

29 Hills J, Measurement of income poverty anddeprivation: the British approach (in Indicators ofProgress, A discussion of approaches to monitorthe Government’s strategy to tackle poverty andsocial exclusion, Report of the workshop held on19 July 2000 organised by DSS and CASE, LSE).Papers by Jonathan Bradshaw and Rebecca Endeanalso consider the criteria for good indicators.

30 See http://www.neighbourhood.dtlr.gov.uk/info/index.htm

31 Building a Strategy for Children and Young People– Consultation Document, 2001. Note:Consultation period closed on 1 March 2002.

32 Micklewright J, Should the UK Governmentmeasure poverty and social exclusion with acomposite index? and Barnes H, SummaryMeasures of Child Well-Being (in Indicators ofProgress, A discussion of approaches to monitorthe Government’s strategy to tackle poverty andsocial exclusion, Report of the workshop held on19 July 2000 organised by DSS and CASE, LSE).


34 Townsend P, 1979, Poverty in the United Kingdom,Penguin.

35 Mack J and Lansley S, 1985, Poor Britain, Allen andUnwin.

36 Other work in this area includes: for Britain,Townsend P and Gordon D, 1989, Memorandumsubmitted to Social Security Committee of theHouse of Commons, in Minimum Income:Memoranda Laid Before the Committee, Session1988–89, London, HMSO; for the USA, Mayer Sand Jencks C, 1988, Poverty and the distribution ofmaterial hardship, Journal of Human Resources,24 (1): 88–114; and for the Netherlands, Muffels Rand Vrien M, 1991, The Comparison of Definitionsof Consumption Deprivation and IncomeDeprivation, mimeo, Tilburg University.

37 Gordon D, Townsend P et al., 2000, Poverty andSocial Exclusion in Britain, Joseph RowntreeFoundation.

38 Vegeris S and McKay S, 2002, Low/moderate-income families in Britain: Changes in LivingStandards 1999-2000, DWP Research ReportNo. 164, Corporate Document Services.

39 Formerly called the Survey of Low Income Families(SOLIF).


40 Households Below Average Income 1994/5–2000/01, 2002, DWP, Corporate DocumentServices.

41 Expert Group on Income Statistics: The CanberraGroup – Final Report and Recommendations,Ottawa 2001, http://lisweb.ceps.lu/links/canberra/finalreport.pdf

42 Nolan B and Whelan C, 1996, Resources.Deprivation and Poverty, Clarendon Press.

43 Gordon D, Townsend P et al., 2000, Poverty andSocial Exclusion in Britain, Joseph RowntreeFoundation.

44 See http://www.ippr.org/research/files/team24/project73/johnhillspres2.pdf. This approach buildson the discussion in Layte R, Nolan B and WhelanC, 2000, Targeting Poverty: Lessons from Ireland’sNational Anti-Poverty Strategy, Journal of SocialPolicy, 29 (4), 553–75.

45 UNICEF Innocenti Report Card Issue No. 1, 2000,A league table of child poverty in rich nations.http://www.unicef-icdc.org/publications/pdf/repcard1e.pdf

Further copies of this consultation document areavailable free of charge from:Welfare Reform (Poverty)Freepost (HA4441)Hayes UB3 1BR

Tel: 020 8867 3201

A service for textphone users is available on:

020 8867 3217

The lines are open Monday to Friday, 9am – 5pm.Please quote code MCP1

Copies are also available in Braille, on audio cassetteand in Welsh, from the above address.

This consultation document can also be accessed onthe internet at: www.dwp.gov.uk/consultations/2002

©Crown copyrightProduced by the Department for Work and PensionsPrinted in the UKApril 2002MCP1

ISBN: 1 85197 981 6