FREE SCHOOL MEALS AND ROMA CHILDREN IN ENGLAND A review of recent evidence from twenty local authorities in England, June 2016

1

Roma Support Group acknowledges the support provided by Barrow Cadbury Trust in making possible the production of this report

2

FREE SCHOOL MEALS AND ROMA CHILDREN IN ENGLISH SCHOOLS 1. Poverty is not news throughout Roma communities in Central & Eastern Europe, or among the migrant Roma communities in the countries of the North and West of Europe. Indeed, many families have made the decision to migrate from the settlements of Eastern Slovakia and Transylvania to escape poverty and exclusion1. 2. With the accession of Central & Eastern European countries to join the European Union in 2004 (the ‘A8 countries’), it became possible for Roma and non-Roma to exercise their ‘treaty rights’ within the wider EU in seeking work. The UK, together with Ireland and Sweden, supported free movement by workers from the A8 countries. However, the UK government imposed a condition that all A8 nationals had to enrol on the Workers Registration Scheme (WRS); this made access to out of work benefits conditional on A8 nationals having completed 12 months continuous employment. This condition remained for seven years – until May 2011. Therefore prior to 20112, very few A8 nationals were claiming out of work benefits for themselves and their families. 3. Many Roma and non-Roma nationals from A8 countries have been able to find employment but some families have found jobs harder to come by. Before 2011, this meant that the initial weeks and months of settlement in the UK could be tough and families did not have access to the out of work benefits to support them. Whilst families could receive child benefit from day one of settlement, and could access inwork benefits (like tax credits) if they found work, the conditions of the WRS meant that claiming (e.g. Job Seekers Allowance) was extremely rare. As Olga Baba said about Bolton in 2011: I do not think that I will reveal a secret by saying that for many Roma parents here in Bolton their children’s school attendance is not a priority. Try to imagine that you arrive in a foreign country whose language you do not speak and you have to find a job to be able to provide food for your large family which is relying solely on you. Then think about the difficulty that you have when you want to find out what possibilities you have for support from local authorities. Then you attempt to contact HM Revenue for the twenty something time without any success to find out when you can finally be paid. Then you try to fill in a form the content of which you do not really comprehend…After all you would have to go through it could be quite

1

http://www.bbc.co.uk/news/in-pictures-34515451

2

This also included nationals from Romania and Bulgaria prior to the similar changes in January 2014 for A2 nationals. 3

understandable that education of your hungry children is not what you think of first3. 4. Many schools reported that Roma children were not eligible for free school meals (FSM), as their parents could not claim one of the so- passported out of work benefits which enabled them to claim FSM4. In 2012, the average price of a primary school dinner was £1.93 and at secondary school, £2.03p5. Children brought food from home, but sometimes they didn’t – and their families could not pay for the children to have school meals. Many schools looked after these children with the provision of meals for free. But this was not always the case – and children sometimes arrived at school hungry, and left even hungrier. 5. In 2012, the Children’s Society estimated that although there were about 2.2 million children living in poverty in the UK, only 1.5 million were eligible for FSMs; about 700,000 children in poverty lived in households where at least one adult worked and so were ineligible for FSMs6. 6. At the same time, the Coalition government introduced a new form of funding for schools. Pupil Premium (PP) funding was allocated to schools based on the number of children receiving FSMs. This was based on the government’s policy assessment of a growing gap between the attainment of pupils considered as ‘deprived’ (defined formally as receiving FSMs), and those ‘not deprived’. 7. On the one hand, it meant that children ineligible for FSM – for example, the ‘working poor’ or those denied out of work benefits – did not benefit from this funding. On the other hand, it meant that schools could take a more pro-active role in informing, encouraging and supporting parents who were eligible for FSM to make a claim. This might result in greater funding for the school, and potentially, in more support for children receiving FSMs. But it also meant more pressure on schools to help families through the maze of the benefit system. 8. In January 2014, the government took action to reduce the rights of EU migrants to a range of benefits, including out of work benefits like JSA and ESA. This again had consequences for eligibility to entitlement for FSMs. Roma Support Group had begun to receive news in 2012 about the number of Roma children who were not 3

http://equality.uk.com/Blog/Entries/2011/2/1_Roma_in_Bolton.html

4

http://www.childrenssociety.org.uk/news-and-blogs/our-blog/caring-roma-children

5

http://cft-staging-cdn.core-clients.co.uk.s3-eu-west1.amazonaws.com/2015/06/seventh_annual_survey2011-2012_full_report.pdf 6

https://www.childrenssociety.org.uk/sites/default/files/tcs/fair_and_square_policy_report_final.pdf

4

receiving free school meals, though this was mainly to do with the Workers Registration Scheme and families not being able to claim any of the qualifying benefits7. We decided to submit Freedom of Information requests to a number of local authorities to assess recent trends in eligibility for free school meals according to the ethnicity of the child. We were also aware that the apparent low take-up of free school meals by bi-lingual children has been the subject of exploration recently by others8. 9. We asked for information about the eligibility for free school meals by ethnicity annually - from 2012 to 2015. This information is gathered every January by the Department for Education in the annual school census, and the resulting information, Schools, pupils and their characteristics, is published every June9. 10. The definition of ethnicity is not straightforward. For the four years we surveyed, the DfE advised schools that the definition of ‘Gypsy/Roma’ pupils in the annual census should include: ...pupils who identify themselves as Gypsies and or Romanies, and or Travellers, and or Traditional Travellers, and or Romanichals, and or Romanichal Gypsies and or Welsh Gypsies / Kaale, and or Scottish Travellers / Gypsies, and or Roma. It includes all children of a Gypsy ethnic background or Roma ethnic background, irrespective of whether they are nomadic, semi nomadic or living in static accommodation. It should not include Fairground (Showman's) children; the children travelling with circuses; or the children of New Travellers or Bargees unless, of course, their ethnic status is that which is mentioned above10. 11. As Roma Support Group has consistently argued, the juxtaposition/conflating of ‘Gypsy’ and ‘Roma’ does not encourage (‘migrant’) Roma families to ascribe their children as ‘Gypsy/Roma’. It also means that the numbers of pupils described as ‘Gypsy/Roma’ would naturally include pupils of English Gypsy (sometimes known as ‘indigenous Roma’), as well as pupils of (‘migrant’) Roma heritage, born in UK or in 7

See appendix 1 for a resume of the issues identified in 2012

8

“EAL and the pupil premium”; (Judith Longstreth, no date) http://www.naldic.org.uk/Resources/NALDIC/Research%20and%20Information/Documents/EAL%20and%20T he%20Pupil%20Premium%20JL.pdf 9

https://www.gov.uk/government/statistics/schools-pupils-and-their-characteristics-january-2015

10

“School census 2015 to 2016 guide”; https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/495464/2015_to_2016_Sch ool_Census_Guide_V_2_5.pdf

5

CEE countries. Roma parents in some schools have requested that their children’s ethnicity be ‘any other white’, or as ‘white East European’, to avoid fear of stigmatisation. In the 2016 school census, the DfE have advised that schools offer ethnic categories which are now distinct - ‘Gypsy’ and ‘Roma’. But for the data we have from 2012 – 2015, we have to accept that the definition of ‘Gypsy/Roma’ is a rough, and probably severely under-estimated, number of the (‘migrant’) Roma pupils in English schools. An education worker from Manchester said, …with some of the parents, they are very happy to say that they are Roma if they are approached in the right way and if they trust the person that they are talking to. The first instinct would be probably to say their country, that they Romanian or they are Slovakian or Czech and not put the Roma bit down. There is obviously....a worry about letting the authorities know that that’s their true ethnicity.”11 12. The national data provides clear evidence since 2014 of the greater proportional decline of Gypsy/Roma pupils receiving FSMs. In tables 1 and 2, we have used as a comparator with Gypsy/Roma, the proportions of ‘white British’ pupils; ‘Travellers of Irish heritage’ pupils; ‘any other white’, and of all pupils. Table 1: primary pupils eligible for and claiming free school meals, by ethnicity, 2012 – 2015, England

White British Traveller of Irish heritage Gypsy/Roma Any other white All

2012

2013

2014

2015

16.6 65.6

% change 2013/4 -4.6 -0.2

15.5 60.6

% change 2014/5 -6.6 -7.6

17.4 63.6

17.4 65.7

42.4 13.7

47.9 13.2

45.0 11.5

-6.1 -12.9

38.2 9.6

-15.1 -16.5

19.3

19.2

18.1

-5.7

16.6

-8.3

(Source: Department for Education, FoI request, 23 February 2016)

Table 2: secondary pupils eligible for and claiming free school meals, by ethnicity, 2012 – 2015, England

White British Traveller of Irish heritage

2012

2013

2014

14.0 67.3

14.3 67.8

14.1 68.0

11

% change 2013/4 -1.4 +0.3

2015 13.6 63.4

% change 2014/5 -3.5 -6.8

http://usir.salford.ac.uk/35793/1/What's%20working%20-%20FINAL%20REPORT%20March%202013.pdf (p35)

6

Gypsy/Roma Any other white All

44.1 15.2

49.2 15.1

48.4 13.9

-1.6 -7.9

40.0 12.2

-17.4 -12.2

16.4

16.7

16.2

-3.0

15.4

-4.9

(Source: Department for Education, FoI request, 23 February 2016)

13. Tables 1 and 2 show that since 2012 (for primary pupils) and since 2013 (for secondary pupils) there has been a decline in the number of all pupils receiving FSMs; and that this decline is slowly accelerating. For example in 2013-2014, the decline was 5.7% in primary age groups and 3.0% in secondaries. In 2014-2015 the decrease was 8.3% for primary and 4.9% for secondary. 14. In terms of particular ethnic groups, between 2013 and 2015, the major decline in eligibility in FSM was amongst Gypsy/Roma at both primary and secondary levels. However, between 2014 and 2015, both the ‘Gypsy/Roma’ and ‘any other white’ ethnic groups witnessed the sharpest decline in FSM eligibility. This corresponds to the impact of the so-called welfare reforms from January 2014, and the impact on EU migrants – either those declaring as ‘Gypsy/Roma’, and/or those described as ‘any other white’ (almost all of whom are EU citizens). 15. In September 2014, the Coalition government introduced Free School Meals for all pupils in the infant stage (from reception class to year 2). This was known as the universal Infant Free School Meals (UIFSM). There was no means test attached to eligibility - all infant stage children irrespective of status received a free school meal. However, schools still encouraged parents who would qualify for FSM (for instance, because the parents received JSA) to apply as this would still be the basis for schools to receive Pupil Premium funding. Table 3: Infant pupils eligible and claiming free school meals, by ethnicity, 2014 – 2015, England

White British Traveller of Irish heritage Gypsy/Roma Any other white All

2014

2015

17.1 62.1

15.7 53.4

% change 2014-2015 -8.2 -14.0

44.4 10.2 18.2

36.9 8.4 16.4

-16.9 -17.6 -9.9

(Source: Department for Education, FoI request, 23 February 2016)

16. Table 3 shows the decline in eligibility for FSM by infant pupils declined much more rapidly between 2014 and 2015 than for all primary children. And that this decline in numbers was least amongst ‘white British’ children (8%) and greatest amongst both 7

‘Gypsy/Roma’ (17%) and ‘any other white’ (18%) pupil groups. The extent of decline of Traveller children (14%) is also of note. 17. We have already noticed that the definitions and terminology of ‘ethnicity’ is inexact. In the expectation that we might be able to more clearly distinguish between (‘migrant’) Roma, and (‘indigenous’) Gypsy pupils and their families, we identified a number of education authorities around England where we knew there to be substantial settlement of Roma families. This also included sizeable numbers of historic and present Gypsy/Traveller settlement too (e.g. in Kent and Bradford). But it might show how there are differences and similarities between the fortunes of Roma and Gypsy pupils and their families. 18. So how is this reflected at a local authority level? We submitted twenty Freedom of Information requests to twelve local authorities outside of London and eight London boroughs. All the authorities outside London responded fully. These were:            

Kent Sheffield Bradford Leeds Leicester Birmingham Newcastle Derby Liverpool Manchester Rotherham Peterborough

19. However, of the London boroughs, only LB Waltham Forest, LB Redbridge and LB Enfield responded with the information. LB Haringey, LB Tower Hamlets, LB Hammersmith & Fulham and LB Barking & Dagenham did not collect this data12. LB Newham failed to reply to both the initial request and two subsequent reminders.

12

This raises a concern as to why some London boroughs do not routinely collect and analyse information about how a measure of poverty (eligibility for free school meals) might relate to a significant demographic (ethnicity).

8

Table 4: proportion of Gypsy/Roma pupils’ eligible and receiving free school meals, 20132015, by primary and secondary, selected education authorities, and in England; and the total number of Gypsy/Roma pupils in 2015 2013 Primary

2014

Secondary

Primary

2015

Secondary

Primary

2015

Secondary

% Roma pupils receiving free school meals Kent Sheffield Bradford Leeds Leicester Newcastle Derby Liverpool Manchester Rotherham Peterborough Waltham Forest Redbridge Enfield (Sub-total) Total: England

49 64 57 60

43 68 54 70

41 41 55 55

32

42 45 57 64

38 23 48 42

18

40 23 48 56 11

56 71 26 8 64 85 43 46 42

52 55 68 20 53 68 38 48 46

56 63 26 9 40 81 35 43 24

62 57 83 27 40 57 18 31 38

46 33 26 21 23 45 8 39 27

54 42 80 24 29 30 15 29 31

48

49

45

48

38

40

Primary

Second

Total pupil numbers 1149 682 728 552 770 478 412 269 127 109 229 214 425 258 39 44 168 76 344 250 129 148 49 48 85 56 41 36 (4695) (3220) 12936 6990

20. These twenty authorities in Table 4 represent just over a third of all Gypsy/Roma primary school age pupils in England, and just under one half of all Gypsy/Roma secondary school students. 21. Table 4 shows that in nearly all authorities, the rate of decline in eligibility for free school meals for ‘Gypsy/Roma’ pupils was higher than the national data (in tables 1 and 2). This would suggest that those authorities with known, sizeable Roma settlement - and so more ‘Roma’ pupils and fewer ‘Gypsy’ pupils - are experiencing a more significant decline in eligibility. In other words, this decline in eligibility is more about the circumstances of Roma pupils and their families, than about Gypsy pupils and families. 22. Table 4 seems to show three groups of authorities: a) A group of cities/towns with substantial reductions in the number of eligible Gypsy/Roma pupils from 2013 to 2015, and particularly in 2014 to 2015 – e.g. Sheffield, Rotherham, Derby, Peterborough and Waltham Forest 9

b) Another group of councils which show a decline in eligibility, but not so steep a decline for Gypsy/Roma pupils – e.g. Bradford, Leeds, Kent, Newcastle and Enfield c) And then a third group of councils where a change in the eligibility for Roma/Gypsy pupils was not evident at all – e.g. Manchester (where eligibility in fact increased) and Liverpool. 23. The most significant explanation for these differences might be the nature of the Roma settlement. For example, most Roma pupils in Manchester are of Romanian Roma heritage13; whereas in Sheffield, the Roma families are mainly from eastern Slovakia14. The circumstances governing EU migrants from Romania and Bulgaria – the so-called A2 accession states – meant that they were largely self-employed, and also unable to qualify for FSM until 2014. This meant that very few Roma children (from Romania and Bulgaria) would be eligible for FSM in Manchester in 2013 and 2014. And despite the tightening up on access to out-of-work benefits for EU migrants after 2014, actually more Gypsy/Roma pupils received free school meals in Manchester 2015 than they did in 2014. 24. Despite the differences between various counties, towns and cities, it is clear that what is happening to ‘Gypsy/Roma’ families is markedly different to other minority ethnic or non-British communities. They had very high levels of eligibility for free school meals in 2012 and 2013, a slight decline in 2014, and a much steeper and swifter decline in 2015. This trend is also evident amongst ‘any other white’ pupils, although the extent of eligibility for free school meals was much lower – lower than amongst ‘white British’ children and young people. 25. Does this mean that Roma children are going hungry at school, as they no longer receive free school meals, but their families are unable to pay as they do not receive any benefits? Or that Roma families are suddenly enjoying a wave of prosperity, which takes them out of the lowest earning households and enables families to pay the £10 a week per child for school meals? 26. It is more likely that as families contend with the tightening up of benefits, especially for newly arriving families, the adults are more prepared to take any work – and in so doing, become part of the working poor who are denied access to free school meals. All schools with substantial numbers of Roma students report that Roma

13

http://romani.humanities.manchester.ac.uk/downloads/2/report.pdf

14

http://www.thefreelibrary.com/The+health+needs+of+the+Slovak+Roma+community+in+Sheffield.a0194699205

10

families are strong supporters of breakfast clubs. We have no evidence that Roma communities are enjoying a wave of prosperity in England at present. 27. In Sheffield for example in 2015, the proportion of Roma young people (22%) eligible for free school meals is still higher than for white British young people (16%), and slightly higher than for all Asian students (21%), but lower than for all black/minority ethnic children/young people (28%), students of ‘mixed heritage’ (30%), and much lower than for Somali pupils (53%). 28. What might be the wider implications for this decline in receiving free school meals? Pupil Premium (PP) funding is linked to the number of students eligible for free school meals at any point in the last six years. Potentially, the decline in Roma/EU migrant children being eligible now for free school meals due to benefit changes could have repercussions for their schools in six years time. 29. But as the Children’s Society report argues, there are also other factors that might dissuade parents from claiming FSMs. This was a point that a Tory MP, and no great supporter of Roma rights, made in February 2011; Stewart Jackson MP for Peterborough said in an adjournment debate when the PP funding was being consulted on as the Coalition government way of targeting education spending on the ‘most deprived’ children: “Does my hon. Friend concede....that, in the case of east European migrants particularly and other migrant groups, there is a cultural predisposition against claiming free school meals? Therefore, some of the children who would most benefit from extra funding are not able to do so, and that obviously has an impact on the overall educational attainment in their school” (our emphasis). 30. The schools minister, Nick Gibb replied: “.....we believe that the best pupil level measure available for identifying and targeting underachievement is free school meals, or some component of it. We expect that more parents will apply for free school meals once it is made clear that pupils will attract additional funding for the school. Indeed, we are receiving reports even now that more parents are doing so, and we hope that that will help to solve the problem of under-claiming. The parents my hon. Friend referred to who do not claim for stigma reasons may well feel that they should claim because it helps the school more broadly if they do” (our emphasis). 31. So the Coalition government are (rightly) encouraging EU migrants to claim benefits, although many are unable to do so as the same government removed their right to 11

eligibility. However, the decision by the same government to clamp down on what were seen as unnecessary rights to benefits by EU migrants in 2014 has clearly meant that, whether or not there is a stigma attached to claiming free school meals, if you are poor but not eligible for benefits, there is no possibility of solving the problem of under-claiming. By definition; some poor people are not in the ‘right’ kind of poverty. 32. It might be that the decline in claiming FSMs for some groups, notably Roma, is an unintended consequence of the political rhetoric against EU migrants that the Coalition government focused on from late 2013 onwards. Was the denial of access to, for example, Job Seekers Allowance intended to penalise children of Slovak families? And also the schools attended by these same Slovak children, where extra resources via PP funding would have helped integrate and support newly arrived students? 33. The Department for Education is now consulting on proposals to alter schools funding, “Schools national funding formula – government consultation stage 1”. Within one of the four building blocks of the school grant, ‘deprivation’ will still continue to determine an element of school funding, based upon pupils’ characteristics. The consultation papers say, We are proposing to use Ever6 FSM as our primary measure of deprivation at pupil-level. But we recognise that when pupils are entitled to FSM, schools are responsible for funding those free meals, which creates an extra cost pressure. We are therefore proposing to use both Ever6 FSM and current FSM data in the national funding formula so that schools’ budgets reflect both the number of pupils for whom they are providing FSM, and the number of pupils who have some history of socioeconomic deprivation.15 34. Moreover, recognising that as Universal Credit is introduced (replacing the variety of benefits that previously enabled eligibility for FSMs), the Department for Education says, We are working closely with the Department for Work and Pensions to determine the level of household earned income at which children in households in receipt of Universal Credit will become entitled to FSM. We will set out our proposals in due course, but we anticipate that the new earnings

15

“Schools national funding formula”, Dept for Education (March 2016) https://consult.education.gov.uk/funding-policy-unit/schools-national-fundingformula/supporting_documents/Schools_NFF_consultation.pdf; #2.16

12

threshold will – over time – further strengthen the FSM measure by relating it more directly to household income16. But this ignores a central claim by the Government – that newly arrived EU migrants will be ineligible for Universal Credit17. 35. As long as core funding for schools is based upon current or historic (the ‘Ever6FSM’) receipt of free school meals, children of families who are increasingly being denied benefit will suffer. Potentially, both in terms of not having a meal at dinner time; and as their school will not attract the correct amount of funding that supposedly directed at schools which deprived children attend.

Andy Shallice RSG June 2016

16

Ibid, #2.17 https://www.gov.uk/government/news/eu-jobseekers-barred-from-claiming-universal-credit

17

13