I’ve worked in Westminster for over sixteen years now, yet the past couple of years since the EU Referendum and the election of President Trump have doubtlessly been the most fascinating in my career. 

The common theme that seems to run through many of the political ‘shocks’ during this time is an ever-increasing disconnect between evidence and government policy. Michael Gove’s, now infamous, quote during the 2016 Leave Campaign that the country has “had enough of experts” may have angered the academic community, but given the outcome of the EU Referendum, it seems to have resonated with more than half of voters. But, if the country has had enough of experts, when did this swing occur and what might have led to it?

While I’m not an expert in either domestic or international politics, I can give my view on what policy-making looked like from within the Civil Service during the Blair, Brown and Cameron years and now, almost three years out of government.

One of my first roles in the civil service was as assistant private secretary to a Minister of State in what was then the “Department for Children, Schools and Families”. Back then, in the early 2000s, Sure Start Local Programmes were being rolled out in the poorest neighbourhoods across the country in an aim to mirror the US “Head Start” programme. At the time, there was a body of evidence of the positive effects for 3 year olds in relation to pre-reading, pre-vocabulary and parent reports of children's literacy skills as well as longer term effects on high-school retention rates, college attendance and earnings in adulthood.[1]  The popularity of Sure Start, combined with the economic context at the time, meant that the Blair Government was keen to extend the programme beyond the most deprived communities and to, ultimately, have a Sure Start Children’s Centre in every community. However, emerging evidence from the National Evaluation of Sure Start found mixed effects on outcomes.[2] While there were some positive effects (mainly in relation to parents behaviours), there were also negative effects in relation to maternal depression and engagement with schools. There were no effects in relation to school-readiness indicators.

These findings raised serious questions about the effectiveness and value for money of the programme. Indeed, when the Coalition Government came to power in 2010, the “ring-fenced” funding for Sure Start was removed and eventually consolidated into local authorities core funding. As a result of depleting local authority budgets and the lack of government prioritisation, a recent study by the Sutton Trust estimated that more than one thousand children’s centres have closed since 2009 – potentially affecting families up and down the country.[3]

So, what can we take from this? First, that great care must be taken when attempting to convert a targeted programme to a universal one. The Labour Government (Blair and Brown) was so keen to demonstrate that it prioritised early years, that it strayed away from the evidence and expanded the programme too fast and too soon. That isn’t to say that the expansion was a complete mistake. There is very little evidence that targeted programmes are more effective than universal ones. Indeed, where the evidence has been conclusive, it suggests that universal programmes have a greater benefit from a societal perspective – largely because they reduce stigma and segregation.

The second lesson can be observed through the Coalition Government’s side-lining of the policy, due to a combination of the evidence of its impact and the need to make austerity savings. In my view, this decision very much suited the analogy of “throwing the baby out with the bath water”. In the face of genuine concerns that children’s centres were not supporting the most disadvantaged families, the Coalition Government should have critically reviewed the purpose of centres and remodelled them based on the latest evidence of what works, putting in place greater oversight and possibly, accountability arrangements. Instead, it simply de-prioritised the programme and left it to wither away with little to replace it. At the same time, the early years and schools ‘field-force’, the National Strategies, was also dismantled, along with targeted programmes designed to improve early language and social development. Once again, there was nothing left to replace these services.

Another example of where government policy has strayed from the evidence recently is the free early years entitlement. The free entitlement was originally introduced in 1998 under the Blair Government. It initially offered 12.5 hours a week (taken over five 2.5 hour sessions), for 33 weeks per year, for all four year olds. During the rest of the Labour Government, the entitlement increased 38 weeks per year and extended to include all 3 year olds.[4] The extension of the entitlement was, in part, due to findings from the Effective Provision of Pre-School Education (EPPE) Project, which found positive benefits of (high quality) early years provision on children’s intellectual and social / behavioural development by the end of Key Stage 1 (age 7), particularly so for the most disadvantaged children.[5] 

By 2014, the study (now called the “Effective pre-school, primary and secondary education project (EPPSE)) had looked at outcomes for secondary aged pupils and beyond and found that while there were still lasting effects of high quality pre-school (again, particularly for the most disadvantaged children), the influence weakened over time. [6]

In response to these findings, successive governments have attempted to increase the number of graduate-level staff in the early years through new qualifications and targeted investment. However, as EPI research finds, the proportion of highly qualified staff in early years settings is declining and, as almost half of staff with a Level 6 qualification or higher are aged over 40 (with over a fifth aged over 50), there are genuine concerns that the early years workforce in the future could be even less qualified than today. The likelihood is that without pay and parity that is comparable to teachers, the early years graduate route will inevitably be unappealing for many.

These findings suggest that the government should prioritise access to high quality early years provision, particularly for the most disadvantaged pupils, and increase its efforts (and probably investment) in growing the number of qualified staff working in this area. In addition, the government should also consider ways to support parents to improve the home-learning environment, reflecting another key finding from EPPSE about the importance of the home and family background. On the issue of the home-learning environment, the DfE has started to acknowledge its importance and has recently announced a £5m fund for the Education Endowment Foundation to trial new schemes to support parents with their children’s learning.[7]

However, in other areas we once again find that government policy is going against, rather than in-line with the evidence. The Conservative Government has increased the free entitlement to 30 hours per week for working parents (who meet the eligibility requirements) – delivering on a manifesto commitment set out ahead of the 2015 General Election. The promise of more, free ‘childcare’ is, understandably, appealing to many parents across the country but also a clear example of how evidence can be side-lined in favour of populism.

And we are already starting to see evidence that the 30 hour entitlement could have an adverse impact on disadvantaged children and their families. Research conducted by EPI found that, once all new government subsidies (including the 30 hour offer and Universal Credit) are rolled out, the government’s policies are likely to have a greater benefit for those on higher incomes. A two-parent family earning £19,000 per year could receive 20 per cent less than a two-parent family with annual earnings of £100,000.

The evaluation of early implementation of the 30 hour offer also highlighted some concerns. Almost 50 per cent of private providers, 40 per cent of voluntary providers and more than a third of childminders reported a decrease in profits during the first year of the offer. With tighter incomes, providers could be forced to favour children eligible for 30 hours over those eligible only for 15 (i.e. the more affluent over the most disadvantaged) and to compromise on the qualification and professional development of staff. The evaluation also found that day nurseries and playgroups cited concerns about staff recruitment and retention - meaning that there is a greater strain on workforce in a sector that was already struggling.

Finally, more than a third of families who took up the 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. As predicted, the policy is benefiting higher earners the most and providing a transfer of cost from higher-income families to the tax-payer.

Often, the government can also be conflicted about how to use evidence and data. There has been much debate recently about the reliability of Ofsted’s “Outstanding” judgement, sparked by a recent National Audit Office Report which found that, as at August 2017, 1,620 schools had not been inspected for six years or more (including 296 schools that had not been inspected for 10 years or more).[8] In 2016, EPI research showed that Good and Outstanding schools whose performance deteriorated substantially after inspection were not routinely prioritised for re-inspection.

This has led for calls for the DfE to remove the exemption to re-inspect Outstanding schools.

But ministers are in a difficult position. They relentlessly use the line that 1.9 million children are now in good or outstanding schools, despite it being somewhat misleading. If they go with the evidence that Outstanding schools should be re-inspected more frequently, then they risk jeopardising their go-to good news line. If they ignore it, they risk thousands of children’s continuing in schools that may be below-par and not accessing the support they need.

In conclusion, evidence should be used to support policy-making. As we observed with Sure Start Children’s Centres, accelerating policy beyond the evidence can risk diluting its impact. It is also rare for any research or evaluation to show outcomes that are 100 per cent positive, long-lasting or conclusive. Crucially, the US studies into Head Start measured long term outcomes into adulthood. Indeed some of the biggest effects were in relation to employment and crime rather than school achievement measures. EPPSE meanwhile only measured shorter-term outcomes. Governments need to consider the important human capital effects of early years intervention, rather than narrow, short-term effects.

Governments should pay close attention to approaches that are proven to work and build on them in a consultative and evidence-based way. Research must be used as part of a wider assessment about how to improve public-policy making, alongside engagement with the sector, academia and service-users. High quality, robust research is vital and should not be ignored but neither should it be used as a Trojan Horse to justify populist policies.

The failure of successive governments to draw on such research in a balanced way has meant that we now have a fragmented early years sector and national policy. The result of this is that there is a real danger of widening the disadvantage gap at age 5, which already constitutes 40 per cent of the gap at age 16.  If such risks are to be avoided, stemming the growing disconnect between government policy, and what the available evidence tells us, is essential.