In February 2013 Runnymede responded to the then-Government's Measuring Child Poverty Consultation. With the Prime Minister suggesting it is time to re-define the measurement of child poverty, we re-publish our response, indicating why a changed definition could result in a misleading discounting of child poverty among Black and minority ethnic households, and a diversion of resources away from those families.



A summary of Runnymede's response is:

• We agree with other responses (CASE, ISER, JRF, CPAG) on the difficulties involved in creating a new multidimensional measure of poverty;
• Ther are much higher rates of child poverty among Black and minority ethnic households, for reasons we explain below. This includes an ‘ethnic penalty’ independent of any other ‘dimensions’;
• Many of the new proposed ‘dimensions’ of poverty do not account for the higher rates of child poverty in BME households;
• We suggest ‘experience of discrimination’ as an alternative or additional dimension if the government goes ahead with a multidimensional measure;
• We highlight some ‘messages for policy’ on ethnic minority child poverty, taken from earlier DWP-commissioned research on this topic;
• Any changes in the measurement of child poverty, and any policies that are thereby adopted, must be assessed for the impact on ethnic minority children. Given the already large gap between ethnic minority and white British rates of child poverty, any policy changes must demonstrate they do not increase these gaps further.

Rather than expand on the good responses to the challenges facing a multidimensional measure (on which we agree with CASE at the LSE and ISER at University of Essex), or on the confusion between causes, consequences and indicators (see also Joseph Rowntree Foundation’s and CPAG’s response), Runnymede’s consultation response rather highlights the specific experiences and reasons for child poverty among ethnic minority groups.



We are open to different ways of intervening to reduce child poverty, but believe that many of the proposed changes work against understanding the cause and nature of child poverty for ethnic minority children in the UK. We of course agree that more policy interventions are needed in a range of areas including family breakdown and alcohol abuse. However, we anticipate that any antipoverty measure developed in this area will fail to respond to many ethnic minority communities, and so will increase the already large inequality between white and ethnic minority groups in terms of their risk of child poverty. That is, any diversion of antipoverty resources to take into account these ‘dimensions’ of poverty runs the risk of increasing the already large gap between white and BME rates of child poverty.



In a 2009 review of ethnic minority child poverty for the DWP, Lucinda Platt found ‘all minority groups have higher rates of child poverty than the majority and the poverty rate for Bangladeshi children approaches two-thirds, compared to an average of one fifth’.[1] Analysis of the DWP’s Households Below Average Income survey (2010/11) found that whereas 26% of white British children live in poverty, 47% of Black British and 58% of Pakistani and Bangladeshi children live in poverty.



The main reason for these higher rates of child poverty is the higher rates of income poverty within these groups. The employment gap between ethnic minorities and white British people is 11%, and tackling discrimination in the labour market remains an effective lever for lowering the child poverty rate amongst Black and minority ethnic children. And where ethnic minority people are in employment, they are much more likely to have low wages and more precarious work. If the government is convinced that there must be new ‘dimensions’ in its measurement of child poverty, we believe that ‘experiencing discrimination’ is at least as compelling a dimension as the proposed alternatives.



Conversely, those groups with the highest rates of child poverty (Bangladeshi and Pakistani) are also least likely to experience some of the proposed new ‘dimensions’ of child poverty, namely drug and alcohol abuse and family breakdown. Any approach that prioritises these considerations over income will have the effect of misleadingly reducing the proportional share of Bangladeshi and Pakistani children living in poverty, and will further offer policy solutions that will have little to no effect on raising these children out of poverty. This is because there remain ethnic penalties in poverty: ‘higher poverty risks for minority ethnic groups are not simply the result of higher proportions of families which we know are at higher risk of poverty, such as lone parent families, large families or workless families.’ (DWP, 2009)



We do of course recognise that there are significant differences between ethnic groups, but this is one reason why we believe a focus on income or resources (including assets) remains the best approach to measuring poverty. In general, ethnic minority groups not only have a higher number of families at risk of poverty, ‘the risks of poverty associated with living in different types of family’ differ across ethnic groups: ‘children in white couple parent families had a 14 per cent risk of being in poverty, but Bangladeshi children in couple families had a 66 per cent risk’ (DWP, 2009: 3). In terms of number of children, whereas white families with one or two children had a 17 per cent chance of living in poverty, this rises to nearly half (49%) for Pakistani children in small families. Lone parent families also vary in terms of the risk of child poverty across ethnic groups: 36 per cent of white children in such families are in poverty, compared to 49% of Pakistani and 46% of black African children in lone parent families.



However, while the risk of being in poverty is higher for some groups, the proportion of families in a particular circumstance also varies by ethnicity. So while the risk of being in poverty for black Caribbean children in lone parent families is actually lower than for Pakistani families in a similar situation, more black Caribbean families are lone parent households. This means measures to tackle poverty within lone parent families are likely to do more to raise black Caribbean children out of poverty. A final point worth highlighting is that among most ethnic minority groups, those living in deprivation (but not poverty) share many experiences with those living in poverty. This implies that ethnic minorities have lower standards of living even on the same income as white British households.



This moves the discussion from analysis to policy. We have already suggested that tackling discrimination in the labour market will be a good way of reducing child poverty among ethnic minorities. Here we again cite the DWP’s report on this topic to indicate which factors explain the differences in poverty across ethnic groups, and to endorse its ‘overarching messages for policy’. The factors included ‘employment rates, hours and pay, non-take-up of benefits and credits, numbers of adults in employment relative to dependants within the household and lack of additional buffers such as savings, and alternative incomes.’ These do not significantly overlap with the proposed dimensions in the consultation document, which motivates our concern that any policy changes based on these dimensions will not effectively response to the already high rates of child poverty among ethnic minorities.



The following ‘messages for policy’ are pasted in their entirety from the 2009 DWP document:

• consideration of ethnicity should be explicitly incorporated within the child poverty agenda to enable the impact of policies across and within groups to be evaluated and to ensure that certain groups are not left behind;
• when evaluating impact of policy, it may be important to consider how different measures tell different stories across ethnic groups;
• attention should be paid to the amelioration of poverty among those who are potentially the most severely affected;
• at the same time, much strategy focuses on the currently poor, but some groups have much greater risks of falling into poverty or into a workless household. Protecting those children not in poverty against becoming poor could potentially play an important role in reducing poverty differences between ethnic groups;
• differences in child poverty between groups can be as great or greater than those between minorities and majority. Policy needs to be sensitive to these differences and to ways of targeting the different risks across groups. (DWP, 2009: 9)

Policymaking should of course be evidence-based and we agree that more should be done in a range of policy areas to combat poverty and its consequent effects. But responding to the consequent effects of poverty, or indeed correlates for high risk of poverty, is not the same as how we measure poverty. There are technical, analytical and indeed impact measurement reasons why a multidimensional measure that throws together all sorts of social bads into a measurement of poverty is unlikely to bear much fruit. Of course it is bad for children when their parents divorce (though still less bad than living in a household with abusive parents); but a well-off family who experiences family breakdown is not, however, poor on a ‘dimension’ of poverty. This is just to say that policymakers must do more than simply respond to poverty.



Changing the measurement and definition of poverty in this way requires strong justification, as it goes against the grain of millennia of understanding and most international policy approaches. The main problem of the consultation is the attempt to shoehorn in a multiplicity of socially bad or unjust outcomes within the conception of ‘poverty’ and to design a precise multidimensional indicator on that basis. But someone whose family breaks down is not thereby poor; nor is someone who lives in sub-standard accommodation; nor are the many people whose lives are made worse through drug and alcohol addiction. We are keen that government develops better policies to respond to these social problems but believe the proposed changes in the consultation confuses the issue of how we measure poverty.



To the extent that antipoverty resources are committed to policies in these new proposed ‘dimensions’, the already large gap in the relative risk of white and ethnic minority children living in poverty is likely to increase. There are always winners and losers in any policy change, but we are concerned that the proposed changes will make losers out of one of the groups most likely to experience child poverty – ethnic minorities. So if the government goes ahead in their new measurement framework, and more importantly if they use that measurement framework to develop new policies, they must minimally show how this will impact ethnic minority children currently living in poverty, including how these interventions will reduce (or increase) the gap in child poverty rates between white British and ethnic minority children.



[1] DWP (2009), Ethnicity and Child Poverty. DWP research report 576, p. 2: http://dera.ioe.ac.uk/11016/1/rrep576.pdf