3rd August 2018

Education Select Committee life chances inquiry: EPI written and oral evidence

In May 2018 the House of Commons Education Select Committee launched an inquiry into the impact that the early years of a child’s life have on their life chances.

The inquiry examines the Government’s current policies in this area – focusing on early years educational settings, but  also considering the role of wider services, including health services and services provided by the DWP.

The Committee invited evidence on:

  • The role of quality early years education in determining life chances and promoting social justice;
  • The importance of support for parents and families, and integration with other services, in prevention and early intervention; and
  • The importance of communication skills and language development.

Written evidence from the Education Policy Institute can be found below. In addition to this submission, EPI was also invited to give oral evidence to members of the Committee on the 12th June. Sara Bonetti, Associate Director of Early Years (pictured), gave evidence on behalf of EPI. A video recording of this session can be viewed here.


Written evidence submitted by the Education Policy Institute


Taking a broad view of the factors that influence life chances


  1. There is clear evidence about the ‘anatomy’ of socio-economic disadvantage in the English education system. The gap in life chances emerges early: at age 3, the social gradient is already present in cognitive and socio-emotional development.[1] Disadvantaged children are behind their more advantaged peers when they begin formal schooling, and the gap in school attainment widens as children progress through school, particularly during the secondary school phase.[2]
  2. However, the relationship between disadvantage and life outcomes is complex and the drivers are numerous. Research shows that adversity in very early life has a particularly harmful and lasting effect[3], influencing the architecture of children’s developing brains. Different factors explain different amounts of variation in the socio-economic gap in development at different stages. Key factors identified in early life include parental health and well-being, maternal age, paternal history of employment, and the home care and learning environment.[4] In primary and secondary school, maternal aspirations based on mother’s own experience of education and child’s behavioural problems at school appear to be important mediators of the attainment gap. Prior attainment was found to be the most important predictor of later attainment, emphasising the necessity of early intervention.4
  3. In order to design effective policy to combat disadvantage, it is essential to address the broad range of additive and interwoven risk factors that can combine to limit child development and educational achievement.[5] The Children’s Commissioner’s office has identified 32 groups of children at particular risk of poor outcomes.[6] These include children with physical and mental health difficulties, those with special educational needs and disabilities (SEND) and children who have experienced trauma or abuse. For many of these children, early identification of difficulties and intervention could help prevent or reduce the development of future problems.
  4. Taking a broad view of the determinants of disadvantage means considering the role of children’s services both vertically (the place of the early years in children’s journey through education towards adulthood, which we discuss under ‘programme interventions’ below) and horizontally (the place of pre-school education and school education within a wider sector of children’s services, and the position of policies for children with additional needs – i.e. those exposed to risk factors – alongside ‘mainstream’ education policy).
  5. Many additional needs in school are potentially identifiable in the early years, enabling early intervention and support. Children identified with Special Educational Needs and Disabilities (SEND) make up just over a fifth of school pupils in England, and they are disproportionately from disadvantaged backgrounds. Ofsted’s 2010 SEND review found that the system is characterised by inconsistencies in how young people’s needs are identified, barriers to parents and young people accessing services and wide variation in the provision of high quality support.[7] There is a shortage of evidence on how this has changed, and indeed whether it has changed, since the 2014 SEND reforms.
  6. Initial findings from ongoing research by EPI found that the prevalence of children identified with SEND is even higher, at one in four children recorded with SEND at some point over the course of school, from ages five to sixteen.[8] This may include some overidentification, but nevertheless it represents a measure of children who are at risk of struggling in school, and who may potentially benefit from early intervention before starting school. Findings from the Effective Provision of Pre-school Education project (EPPSE) suggest that pre-school provides an opportunity for effective intervention for the reduction of special educational needs, especially for the most disadvantaged children.[9]
  7. Findings from the EPPSE study suggest that specialised support in pre-schools could also be beneficial for children with English as an Additional Language (EAL).9 Recent EPI research has identified highly vulnerable subsets of children with EAL.[10] While the EAL group performs well against the national average for school attainment on average, this masks substantial variation within the group, with over one in ten children with EAL performing at a very poor level at GCSE. In assessing the adequacy of support for EAL, it is important to note that children sitting GCSEs in recent years experienced early years provision in the Sure Start era and ring-fenced local area funding during their primary education, under the Ethnic Minority Achievement Grant, which was mainstreamed in 2011. Future cohorts of children will have experienced different systems for which the long-term outcomes are not yet known.
  8. Any view of social justice that fails to take account of how we care for the most vulnerable children in society would be both incomplete, and would fail to address some of the most severe consequences of poverty and inequality. Of the approximately 389,000 children in contact with social services in England, 13 per cent are at risk of or currently suffering significant harm and 19 per cent are looked after by their local authority.[11] Since 2010 at least, there has been a rise in the number of children at the sharp end of social services: the rate of children on a Child Protection Plan has risen from 34.8 to 43.3 per 100,000 children and the rate of looked after children from 57 to 62 per 100,000. These children are at particular risk of low school attainment, educational instability and exclusion from school, as well as mental health difficulties and contact with the criminal justice system in adulthood.


Expenditure on children’s services


  1. Analysis and projections by the Institute for Fiscal Studies found that public expenditure per child on early years provision rose rapidly during the 2000s as national early education and childcare entitlements were introduced and expanded.[12] However, more recently, expenditure has not increased in line with participation (demand from parents) and population size. Expenditure per child has been cut by 17 per cent since 2008, mainly due to reductions in local authority early years expenditure.12
  2. Average expenditure per head is set to increase again following the introduction of the longer 30-hour entitlement for working parents12, but early evaluation of the policy indicates that this additional spend is tilted in favour of more affluent families[13], as EPI warned in 2016 before the policy was rolled out, based on modelled estimates. [14] It is not safe to assume that these additional hours are ‘merely childcare’ and convey no educational benefit that less advantaged children may miss out on.
  3. Given the broad range of risk factors, outlined above, that can undermine social justice in education, expenditure on services to support children’s development and education should not be assessed on a piecemeal service-by-service basis. Indeed, concerning trends in government spending on children are apparent across a range of budget areas that are relevant to the life chances of vulnerable children.
  4. Looking vertically within the education system, overall school spending has steadily grown since the 1990s in England – with the figure around £43.7bn in 2018-19. Additionally, Pupil Premium funding (totalling £2.4bn), set aside for disadvantaged pupils, stood at £1,320 for primary school pupils, and £935 for secondary school pupils in 2017-18. However, in recent years, pressures in the Dedicated Schools Grant led to deteriorating school budgets as a result of flat cash per pupil settlements from 2010-11 to 2014-15, followed by real terms increases that have been too small to fully offset significant cost pressures for staffing during the last three years.[15],[16] The intended ‘additionality’ of the Pupil Premium over and above the Dedicated Schools Grant has clearly been compromised by dwindling overall budgets, with 30 per cent of leaders in 2017 reporting to NfER / The Sutton Trust that they use some of their Pupil Premium budget to offset cuts to wider funding and 6 per cent reported that this was their main priority for the Premium.[17]
  5. The challenge of increasing social justice in early years providers and schools becomes more acute to the extent that they must tackle disadvantage in a vacuum. Looking horizontally across children’s services, an analysis of local authority trends in expenditure on all children and young people’s services between 2010 and 2015 found a decrease of 38 per cent in non-looked after children and non-safeguarding spending, with a more pronounced decrease in the most deprived third of areas (46 per cent v. 28 per cent in the least deprived).
  6. Similarly, the National Audit Office reports a modest overall rise in spend on children’s services from 2010 to 2017 of 3.2 per cent, accounted for by late-intervention (LAC and safeguarding) services, while non-statutory early support service spending has fallen substantially. But despite the increasing prioritisation of acute social care, there are indications that funding to meet needs may have reached a precarious stage: the number of authorities overspending on social care has increased steadily since 2011. Last year, 80 per cent of LAs overspent on social care, and 66 per cent drew down their reserves. A third of councils surveyed reported that children’s social care is their top immediate spending pressure, up from 7 per cent last year.
  7. Looking horizontally at support for children with mental health difficulties, it is unclear how much of the funding allocated to the ‘transformation’ of children and adolescent mental health services is reaching frontline providers. The charity Young Minds sent out Freedom of Information request in 2016/17 and found that half of the 199 Clinical Commissioning Groups surveyed had used some or all the additional funding to backfill cuts or on other priorities; fewer than half could provide full information on their CAMHS budget.[18]
  8. The significance of the mounting financial pressures on local authority budgets that historically provided for greater early years spend and more preventative or early intervention services for vulnerable children and their parents is that the loss of such services is likely, over time, to translate to even greater pressures on acute late-intervention services such as children’s social care, CAMHS and SEND support, more unaddressed problems left at the door of mainstream schools, and more children educated outside of mainstream education as schools struggle to respond. Spending less now on prevention probably means paying more later.

The early years and children’s workforces:


  1. The first five years of a child’s life are crucial for their social, emotional and cognitive development. A supportive family, access to high quality health visitors and other health services, positive and stimulating interactions in safe and stimulating environment with other children and highly qualified teachers, are all key contributors to positive child development. From the day they are born, children come into contact with a wide variety of professionals that help and support them and their families: midwives, doctors, nurses, health visitors and early years educators.
  2. Children from disadvantaged families are more likely to have additional needs that require specialist support from professionals, such as social workers, educational psychologists and speech and language therapists. These needs are most likely to be identified and referred for additional support at an early stage if the early years workforce is stable, highly trained and experienced. Yet the capacity and stability of children’s workforce across early years, schools and social work is increasingly precarious.[19]
  3. We have recently reported on the state of the early years workforce in England, and found several causes for concern, including the apparent beginnings of a downward trend in qualifications at the critical level 3 threshold.[20] Seventy-nine per cent of group-based staff, 77 per cent of nursery staff, 74 per cent of reception staff and 69 per cent of childminders have at least a level 3 early years qualification.20 However, recent survey data from the National Day Nursery Association (NDNA)[21] suggests that the proportion of level 3 qualified staff in early years settings fell from 83 per cent in 2015 to 75 per cent in 2016. A separate study carried out by the Professional Association for Childcare and Early Years (PACEY) reported an increase in the cost of level 3 courses from £250 in 2012 to £1,900 in 2015.[22]
  4. Secondly, the age distribution of the most highly qualified early years staff suggests that we may soon be facing problems in retaining current levels of expertise: Almost half of highly qualified staff (level 6 and more) are aged over 40, with 21 per cent aged over 50 and approaching retirement in the next 10 to 15 years.20 Findings around financial incentives[23] and current enrolment levels in Early Years Teacher Training Courses in 2017 cast doubt on the capacity of the younger workforce to keep the proportion of graduates steady, let alone for it to increase. This potentially means that the early years workforce in the future could be even less qualified than today.
  5. Thirdly, considering the government’s policy of providing free early education to disadvantaged two-year-olds as a key plank in its approach to social mobility, it’s worth noting that the proportion of these children benefitting from the presence of a graduate teacher in the classroom is less than half, and has decreased slightly between 2014 to 2016 (from 45 to 44 per cent).20 While this change is small on a national basis, it has been particularly sharp in London and Yorkshire and the Humber, where the proportions fell respectively from 47 to 41 per cent and from 48 to 44 per cent.
  6. Fourthly, over recent years there have been increases in the use of unpaid and volunteer staff in early years providers, especially in full day care settings, where a sixty per cent increase was reported between 2008 and 2013.[24] Very little is known about these staff beyond how many there are. Reliance on volunteers raises many questions about the skills and qualifications of those working with young children, as well as the financial viability and sustainability of current staffing models.
  7. Fifthly, pay in the early years sector is so low that the recent increases in the National Living Wage and National Minimum Wage have had a significant impact on costs and the viability of providers. A Survey by the National Day Nurseries Association in 2015 estimated that payroll bills would be pushed up by at least 34 per cent between by 2020. [25] For those providers that manage to stay in business, the quality of the staff they can afford to employ will become a key issue, with 58 per cent of the survey respondents reporting that they would look at recruiting younger and less qualified staff.
  8. The mixed market of private, voluntary, and independent, group settings and local authority or school-based early years provision results in large differences in pay between provider types, linked to differences in staff qualifications. On average, staff in group-based providers earn £8.30 per hour, compared with £14.40 for nursery staff in school-based providers, and £15.10 for school reception staff.20 The wage differential between senior and non-senior staff is also greater in school-based settings than in group-based settings. Moreover, the mean hourly pay for senior staff in group-based settings (£11.20) is lower than the mean hourly pay received by non-senior staff in school-based settings (£11.90 and £13.00 respectively for nursery and reception teachers).
  9. Staff turnover has reached concerning levels across several parts of the children’s workforce, but instability is greatest for early years staff working in private and voluntary nurseries and play groups, one in sevenof whom left the sector in 2016.20 This is followed closely by turnover for children’s social care workers, one in eight of whom left in 2017[26]; and school teachers, of which one in 10 left in 2016.[27] Among school-based early years staff, turnover in 2016 was a little lower, at just under one in 1220
  10. The government’s own evaluation of the early stages of the 30-hour childcare entitlement for working parents found that while the expansion helped to increase parental working hours (its primary stated objective) and improved household finances for those families that benefitted (typically more affluent families)[28], this came at a price of concerns about its impact on staff recruitment and retention13, and therefore capacity to meet the increased total demand for childcare. Perhaps most worryingly, 40 per cent of voluntary providers and more than a third of childminders reported a decrease in profits13, with potential consequences for the quality of future provision. These financial pressures pose a risk that some providers may decide to prioritise children eligible for 30 hours over those only eligible for fifteen, due to their parents working less than 16 hours each per week.


Programme Interventions


  1. There is now abundant evidence pointing to the role of early childhood education in improving children’s outcomes both in the short and the long term. In a seminal study, James Heckman established that the returns to investment (ROI) in preschool programmes for three- and four-year olds amount to 7-10 per cent return on investment.[29] Later work by Heckman established a larger ROI of 13 per cent for longer high-intensity programmes including home visits from birth to age five, and targeted at children whose family scored highly on a child risk index.[30]
  2. It is important to note that in contrast to UK cohort studies such as EPPSE[31], and the National Evaluation of the Sure Start programme[32], these US evaluations measured outcomes up to age 30, and extrapolated from these to estimate full life-cycle benefits across a range of human development outcomes. These included health benefits measured via quality-adjusted life years, and reductions in the risk of involvement in crime, as well as increases in earnings from employment for parents and children who received the programmes, and reductions in state transfer benefits. Comparison of the UK and US studies suggests that measures of short-term educational impact may not capture the full range of human capital benefits from early years programmes, which are wider than school attainment and many of which only become apparent much later in life.
  3. The EPPSE study examined the benefits of early years provision, measuring outcomes from age three to sixteen, and found that attending pre-school enhanced children’s development, especially where the quality of the processes and interactions in the setting was high, and especially for disadvantaged children.[33] Additionally, it found that the duration (in months) of pre-school is important, and that starting pre-school before age three is beneficial for children’s intellectual development. Disadvantaged children benefited particularly from provision that was attended by children from a mix of social backgrounds, as well as from higher quality provision.
  4. Results of the National Sure Start Evaluation and the later reports on the EPPSE cohort focusing on school attainment were met with disappointment in many quarters due to the phenomenon of ‘fade-out’, whereby larger initial benefits from early years provision were followed by smaller attainment effects the later in childhood these outcomes were measured. Fade-out of cognitive academic effects is also common in the wider international literature on early years programmes[34], but Heckman presents evidence suggesting that the explanation for fade-out during school yet large adulthood effects may have two key components. The first is the multidimensional nature of human capital and skills[35] (including health and non-cognitive character and social and emotional skills[36]) which is not well captured and measured by academic attainment alone.
  5. The second reason proffered by Heckman for the pattern of fade-out in school attainment effects is that while the early years are a highly sensitive period for child development[37], and interventions during this period have the highest rates of return, they do not provide ‘inoculation’ against poor quality educational settings later in childhood nor against continued adversity due to socio-economic deprivation. The pattern of investment which is most likely to produce socially just outcomes is to begin with large early investments, but then continue to invest in additional support for disadvantaged children throughout their childhood.37 The economic argument for this is that early investments increase the effectiveness or value-for-money of later investments35, and that human skills are built dynamically over the life course, with each new level of skill depending on the last.
  6. Evidence suggests that very early investment (before the age of three) has the largest returns in terms of cognitive skills or academic attainment37, whereas non-cognitive or character, social and emotional skills continue to respond to investment and intervention during adolescence.[38] In England, these two sensitive periods correspond closely with the periods when the disadvantage gap in attainment is observed to grow at its fastest.


[1] Dearden, L. Sibieta, L. & Sylva, K. (2011). The socio-economic gradient in early child outcomes: Evidence from the Millennium Cohort Study. London: Institute of Fiscal Studies.

[2] Andrews, J. Robinson, D. and Hutchinson, J. (2017). Closing the gap? Trends in educational attainment and disadvantage. London: Education Policy Institute.

[3] Shonkoff, J. & Garner, A. (2011). The Lifelong Effects of Early Childhood Adversity and Toxic Stress. Pediatrics. 2012 Jan;129(1):e232-46. doi: 10.1542/peds.2011-2663. Epub 2011 Dec 26.

[4] Goodman, A. & Gregg, P. (2010). Poorer children’s educational attainment: how important are attitudes and behaviour? York: Joseph Rowntree Foundation.

[5] Sabates, R., & Dex, S. (2013). The Impact of Multiple Risk Factors on Young Children’s Cognitive and Behavioural Development. Children & Society, 29, n/a–n/a. https://doi.org/10.1111/CHSO.12024

[6] Children’s Commissioner for England (2017). On measuring the number of vulnerable children in England. London: Office of the Children’s Commissioner.

[7] Ofsted (2010). The special educational needs and disability review: A statement is not enough. Manchester: The Office for Standards in Education, Children’s Services and Skills.

[8] Hutchinson, J. (2017). Policy Analysis Paper: How many children have SEND? London: Education Policy Institute.

[9] Sylva, K. et al. (2004). Technical Paper 12 The Final Report: Effective Pre-School Education. London: Institute of Education.

[10] Hutchinson, J. (2018). Educational Outcomes of Children with English as an Additional Language. London: Education Policy Institute.

[11] Crenna-Jennings, W. (2018). Vulnerable children and social care in England: a review of the evidence. London: Education Policy Institute.

[12] Belfield, C. & Sibieta, L. (2017). Long-Run Comparisons of Spending per Pupil across Different Stages of Education. London: Institute of Fiscal Studies.

[13] Paull, G. & La Valle, I. (2017). Evaluation of Early Implementation of 30 Hours Free Childcare. London: Department for Education.

[14] Johnes, R. & Hutchinson, J. (2016). Widening the gap? The impact of the 30-hour entitlement on early years education and childcare. London: Education Policy Institute.

[15] Perera, N., Andrews, J., & Sellen, P. (2017). The implications of the National Funding Formula for schools. London: Education Policy Institute.

[16] Andrews, J. & Lawrence, T. (2018). School Funding Pressures in England. London: Education Policy Institute.

[17] National Foundation for Educational Research (2017). Pupil Premium Polling 2017. London: Sutton Trust.

[18] Young Minds (2017). Children’s mental health funding not going where it should be. Retrieved from: https://youngminds.org.uk/about-us/media-centre/press-releases/children-s-mental-health-funding-not-going-where-it-should-be/

[19] Crenna-Jennings, W. (2018). We can’t keep papering over the cracks. Vulnerable children need stable support. Article published by The Guardian Social Care Network on Tuesday 1st May 2018. Retrieved from: https://www.theguardian.com/social-care-network/2018/may/01/vulnerable-children-need-stable-support-education-policy-institute

[20] Bonetti, S. (2018). The early years workforce: a fragmented picture. London: Education Policy Institute.

[21] National Day Nursery Association, 2016, ‘Early Years Workforce Survey: England 2016’. NDNA survey results are slightly different from statistics available through administrative survey. However, trends have always been consistent with the latter.

[22] Susanna Kalitowski, 2016 ‘Towards and Early Years Workforce Development Strategy for England’, PACEY, p. 4.

[23] Data from a survey by Save the Children in collaboration with Nursery World. Initial findings can be found

at https://www.nurseryworld.co.uk/nursery-world/news/1163363/exclusive-survey-reveals-barriers-totraining-and-employing-eyts

[24] It is not possible to provide fully comparable figures for more recent years due to changes to the government’s Survey of Childcare and Early Years Providers.

[25] NDNA, 2015, ‘National Living Wage funding factor is vital ahead of 30 hour free childcare offer’, August 2015 Press Release, http://www.ndna.org.uk/NDNA/News/Press_releases/2015_press_releases/National_Living_Wage_funding_f actor_is_vital_ahead_of_30_hour_free_childcare_offer_.aspx

[26] DfE (2018). Experimental statistics: Children and family social work workforce in England, year ending 30 September 2017. London: Department for Education SFR09/2018.

[27] DfE (2017). School Workforce in England: November 2016. London: Department for Education SFR25/2017.

[28] More than a third of families who took up the 30-hour offer earned between £32k and £52k. An additional 32 per cent of families earned more than £52k, while take-up for families earning less than £15.6k was lower than 10 per cent. Indeed, families earning more than £52k increased their use of childcare by around 38 per cent, but only around 18 per cent of mothers and around 7 per cent of fathers reported an increase in working hours. It is clear that to date, this policy has resulted in a transfer of cost from higher-income families to the taxpayer.

[29] Heckman, J., Moon, S., Pinto, R., Savelyev, P., & Yavitz, A. (2010). The Rate of Return to the High/Scope Perry Preschool Program. Journal of Public Economics, 94(1-2), 114–128.

[30] Garcia, J., Heckman, J., Leaf, D., & Prados, M. (2017). Quantifying the Life-Cycle Benefits of a

Prototypical Early Childhood Program. Bonn: IZA Institute of Labor Economics.

[31] Cohort Study web page: https://www.ucl.ac.uk/ioe/research/featured-research/effective-pre-school-primary-secondary-education-project

[32] National Evaluation web page: http://www.ness.bbk.ac.uk

[33] Sylva, K. et al. (2014). Students’ educational and developmental outcomes at age 16 Effective Pre-school, Primary and Secondary Education (EPPSE 3-16) Project. London: Department for Education.

[34] Elango, S., Garcia, J., Heckman, J., & Hojman, A. (2015). Early Childhood Education. NBER Working Paper No. 21766 retrieved from: http://www.nber.org/papers/w21766

[35] Heckman, J. (2015). Creating Flourishing Lives: The Dynamics of Capability Formation. Retrieved from: https://heckmanequation.org/assets/2017/01/HDCA-SenLecture_091115.pdf

[36] Goodman, A. et al. (2015). Social and emotional skills in childhood and their long-term effects on adult life: A review for the Early Intervention Foundation. London: Institute of Education.

[37] Heckman, J. & Mosso, S. (2014). The economics of human development and social mobility. NBER Working Paper No. 19925 retrieved from: http://www.nber.org/papers/w19925.pdf

[38] Gutman, L. & Schoon, I. (2013). The impact of non-cognitive skills on outcomes for young people: Literature review. London: Education Endowment Foundation.