Second Peter Townsend Memorial Conference

Measuring Poverty: The State of the ArtUniversity of Bristol

22 and 23 January 2011

No reduction in the UK since 2004/5 – 2010 target missed

All but seven OECD countries increased their child poverty 1995-2005

In EU 2005-2009 child poverty increased in swe, ger, fra, ita, gre and ire – flat in many others and higher than pensioner poverty in half

In developing world despite economic growth – child poverty flat lining.

Measurement of child poverty in the UK EU (and OECD) Developing countries

Shift from income to deprivation Shift from poverty to material well-being Shift from material well-being to

Social exclusion well-being

Asking children about poverty in The Children’s Society Survey

Pilot work for PSE 2010 on child deprivation

Household surveys (or registers) though some administrative stats in spatial analysis (ie ID) and international (SOWC)

Child the unit of analysis (only recently by WB) Adult the informant Mainly indirect - consumption/income Arbitrary thresholds –

Too low in poor countries ?Too high in rich countries

Equivalence scales with no scientific base Usually relative

Relative low income = Equivalised net household income less than 60% median 2020 target: <10% of children

Combined low income and material deprivation = Material deprivation >20% and equivalised net household income less than 70% median 2020 target: <5% of children

‘Absolute’ low income= Equivalised net household income falling below 60% of the ‘adjusted base amount’ 2020 target: <5% of children

Persistent poverty= Equivalised net household income less than 60% of medianfor 3 years prior to current year

2020 target: not yet set

+Frank Field

Sources: OECD Growing unequal - every five years Luxembourg Income Study - every five years EU SILC now annually

EU social inclusion indicators from SILC – child poverty key focus At risk of poverty rate<40,50,60,70% of national median At risk of poverty gap 60% <60% anchored Lacking 3+ (out of 9) deprivation items Persistent – to be developed Now 2020 target =<60% median or lacking 4+ deprivation items or

workless SILC 2009 includes special module on child poverty/well-being University of York work on extreme poverty recommends

‘consensual’ overlaps of deprivation and a budget standards income threshold

Relative income poverty thresholds too low in EU10

World Bank dominates survey data and analysis

Child rarely the unit of analysis Very low thresholds $x per day per capita

completely arbitrary Best work

Bristol method – lacking in one or more of seven domains

Mekonen Child Friendly Governments in Africa

Severe Food Deprivation– children whose heights and weights for their age were more than -3 standard deviations below the median of the international reference population, i.e. severe anthropometric failure (Nandy et al, 2005).

Severe Water Deprivation - children who only had access to surface water (e.g. ponds, rivers or springs) for drinking or who lived in households where the nearest source of water was more than 15 minutes away.

Severe Deprivation of Sanitation Facilities – children who had no access to a toilet of any kind in the vicinity of their dwelling, including communal toilets or latrines.

Severe Health Deprivation – children who had not been immunised against any diseases or young children who had a recent illness causing diarrhoea or acute respiratory infection (ARI) and had not received any medical advice or treatment.

Severe Shelter Deprivation – children living in dwellings with five or more people per room (severe overcrowding) or with no flooring material (e.g. a mud floor).

Severe Education Deprivation – children aged between 7 and 18 who had never been to school and were not currently attending school (no fessional education of any kind).

Severe Information Deprivation – children aged between 3 and 18 in households which do not possess a radio, television, telephone or computer.  

“one which is making the maximum effort to meet its obligations to respect, protect and fulfil child rights to ensure child well-being” Mekonen, Y. Child Ind. Res. (2010)

Three dimensions based on indicators Laws and policies Allocation of resources Child Outcomes: access to education, health, water,

sanitation, survival, nutrition, poverty

Overall

Peter Townsend Poverty in the UK Mack and Lansley Breadline Britain PSE x 2 Child deprivation index in PSE 1999 not very

discriminating Every item a necessity Lacking 1 or more the threshold

Must do better this time

Bristol SEM Cabinet Office Social Exclusion Task Force Social

Inclusion Over the Life Course B-Sem operationalised for families with children

(Oroyemi et al 2009). Used data from the Families and Children Study

(FaCS). Eighteen markers of risk were constructed from the

data, ranging from income poverty to lack of social contact to overcrowded accommodation.

They found that 45% of families with children were exposed to multiple risk markers (i.e. two or more markers of risk) in 2006,

less than 2 per cent experiencing 10 or more risks. Cluster analysis was used to group families into nine

relatively homogeneous ‘clusters’. :

Material well-being combines Income poverty

Rates and Gaps

Deprivation Parental worklessness

Table 3.11: Correlations between the domains of well-being in the European Union.Domain Correlation coefficient

Education 0.64***

Risk 0.59**

Housing 0.56**

Health 0.52**

Subjective 0.45*

Relationships 0.20

Table 3.5: Spatial association between poverty and other domains of well-being at LSOA level (Spearman rank correlations - all coefficients are statistically significant at the <0.01 level)

Domains Correlation coefficient

Education 0.80

Housing 0.63

Health 0.56

Crime 0.55

Environment 0.07

How do children understand child poverty?

Can children be a useful data source in measuring child poverty?

How can child poverty be measured at the level of the child, rather than the family?

How useful are deprivation indicators in the measurement of child poverty?

the extent to which children seek to protect their parents from their own feelings of deprivation, sometimes including hunger

the extent to which they feel unable to invite their friends for meals or to stay

the importance of grandparents and other relatives in providing extras that mitigate the deprivation in their lives

the costs and inconvenience of public transport, particularly in rural areas, which restrict their lives

the value of holiday schemes that give children the chance to get away

the sense of shame and embarrassment when they are unable to dress like their peers and

their experience of schools as exclusionary – their inability to go on trips and outings, to contribute to school funds, to dress well and the frequent identification as ‘free dinner’ children.

Asked Number of workers Free school meals Pocket money

Very weak association with subjective well-being

Stronger with Happiness with possessions 11 point scale How well off do you think your family 5 point scale

Worry about reliability of responses

Ran focus groups with children asking them to identify socially perceived necessities

Then piloted them with 300 children and their parents

Answer codes: Have, don’t have, don’t have and don’t want

93.6% parent-child couples agreed on the number of adults in paid employment

98.6% parent-child couples agreed on whether the child received free school meals

50.7% parent-child couples agreed in their subjective assessment of the family’s wealth

For deprivation items, agreement varied between 82% (family day trips on a monthly basis) to 99% (computer with an internet connection at home)

Conclusion 11-16s can report material well-being

The 20 item deprivation scale: Moderate correlation with parental report on household

income (r=0.4**) No significant relationship with number of adults in paid

work Children not receiving free school meals owned on

average 3.3 more items than those receiving free meals (t=2.7**)

Significantly associated with children’s perceptions of their family’s wealth (F=21.6**)

Scalability within acceptable range (Cronbach’s Alpha=0.86)

Shorter 10 item scale devised and tested with minimal change to Cronbach’s Alpha (0.82)

Adopted for survey of 5000 8-15 now in field

Children’s reporting of objective poverty measures highly similar to parental reports, supporting further research with children as respondents.

Within deprivation items that children lacked, differences between parents and children tended to be in the form of parents reporting that children did not want an item whilst children reported that they did want it.

Support for the idea that children may conceptualise poverty differently to adults, or hide their experiences of poverty from parents.

Improve the socially perceived necessities items for children

Children’s Society work

Review of questions in other surveys