The impact of the pandemic on education means that pupils in England are on course for lower GCSE grades well into the 2030s. Lee Elliot Major, Andrew Eyles, Esme Lillywhite and Stephen Machin say that the school system needs rebalancing to support pupils' cognitive and socio-emotional skills and thereby improve the education and life outcomes for the Covid generation.

Could learning loss suffered by pupils in England and across the UK end up being the most enduring and inequitable legacy of the Covid-19 pandemic? Our research seeks to provide the most authoritative answer yet to this question. By deploying the latest econometric methods to understand how children develop skills over time, using data from UK cohort studies, we estimate the educational and lifetime impact of mass school closures and deepening social divides on a generation of children.

In previous work, we highlighted the likely devastating impact that school closures would have on young people's life prospects (Elliot Major et al, 2021). We estimated that a quarter of pupils - around two million children - received no schooling during the first lockdown in early 2020. When the equalising force of the classroom is removed, home learning inequities are exposed. Pupils eligible for free school meals, those educated at state schools and, more generally, those from less affluent backgrounds, suffered learning losses at a much greater rate than their more affluent peers. Disruption to schooling between March 2020 and April 2021 was unprecedented in scale, with most pupils missing over half of their expected days in the classroom.

In our latest research, supported by the Nuffield Foundation, we produce updated estimates of learning losses and the likely post-pandemic fall in levels of income mobility levels. A fall in income mobility means that poorer children are more likely to become relatively poor adults while richer children stay rich. These estimates consider not only cognitive learning losses but also declines in socio-emotional skills and are based on more detailed calculations of the impacts experienced at different ages and for different pupils. Even our previous gloomy predictions did not foresee the perfect storm of factors that would exacerbate inequalities inside and outside education in the wake of the pandemic.

Like other countries, England is facing a national crisis in persistent school absence, defined as a pupil missing 10% or more of their lessons. An analysis undertaken as part of this project calculated that 28% of primary and 40% of secondary pupils who qualified for free school meals were persistently absent during the 2021/22 autumn term. In any other era, these damning figures might constitute a national crisis. The closure of schools deprived children of access not only to face-to-face learning but to the many other things that schools provide from emotional support to social interaction to, for some, regular meals. Some claim we now face a tsunami of mental health problems among children in the wake of Covid-19, while teachers have reported greater behaviour problems in classrooms since the pandemic. Our latest estimates confirm a significant decline in socio-emotional skills for successive year groups or cohorts while international comparisons show that life satisfaction scores for pupils remain low compared with most other countries.

Socio-emotional skills, including attention and resilience, play an important role in academic attainment

Catch-up comparisons

We need to tread carefully when inferring conclusions from international comparisons: differences in pupils' test scores between countries can be driven by social factors outside schools rather than what is happening in the classroom; results can be biased by skewed samples of children taking tests and when tests are taken. Countries also faced varied pandemic experiences. Factors such as the severity of the disease spread, lockdown stringency and the duration of school closures, contributed to varying degrees of learning loss.

Nonetheless, an emerging consensus indicates three broad findings. First, children across a range of countries suffered significant learning loss during the pandemic - defined by lower scores in tests compared with previous cohorts of pupils - alongside a significant decline in wellbeing. Second, children from under-resourced backgrounds have fallen further behind their more privileged peers in average test scores. And third, these persistent and growing achievement gaps are the consequence of long-term trends, not solely disruptions due to the pandemic. England, and the UK, typify these patterns.

Children in the UK suffered significant Covid-19 learning loss - up to six months - with children from low-income backgrounds experiencing an extra two months of learning loss compared with their peers. Children also suffered a significant decline in socio-emotional skills. These two factors combined to limit pupils' development at every life stage. The pandemic amplified long-term persistent education gaps across a range of Organisation for Economic Cooperation and Development countries, including the UK. Compared with most other nations, England's pandemic response was heavily focused on academic catchSocio-emotional skills, including attention and resilience, play an important role in academic attainment up with less emphasis on socio-emotional skills, extracurricular support and wellbeing. Our results suggest that to improve child outcomes, much greater emphasis is needed in schools on activities that improve both socio-emotional as well as cognitive skills.

The social side of academic success

This work is the first to assess the damage wrought by school closures during the pandemic on England's children in a multi-dimensional, dynamic way. Our economic model of skill formation enables us to understand the effect of the pandemic on the attainment of the Covid-19 generation in more detail, across different ages of children, considering gender and socio-economic backgrounds, and predicting the full consequences for the cohorts of children affected by the pandemic response.

Critically, we assess the impact of declines in socio-emotional skills alongside falls in cognitive skills on children's development, considering these over three distinct life stages - early childhood (ages 5-11), middle childhood (ages 11-14) and later childhood (ages 14-16). Our analysis uses data from the Millennium Cohort Study, which is following the lives of around 18,818 individuals born at the turn of the century. It covers all four UK nations and contains rich data on educational, family background, health and other variables from birth until age 17. These young people were educated in a similar schooling environment to the Covid-19 cohorts, many of whom were born only a few years later.

We find cognitive levels at age 5 are highly predictive of cognitive skills at age 11, while socio-emotional skills are even more persistent during this life stage; 4CentrePiece Summer 2024 socio-emotional skills and parental time investments, such as time spent reading with a child or visiting museums, also play a significant role in predicting cognitive skills at age 11.

Children's cognitive skill levels at age 14 are primarily shaped by cognitive skills and parental investments at age 11 while socio-emotional skills at age 11 are highly predictive of socio-emotional skills at age 14.

On average, children's cognitive and socio-emotional skill levels at age 14 are equally important in predicting whether they will go on to achieve five good GCSEs including English and maths at age 16. Socio-emotional skills, including positive social skills, attention and resilience, play an important role in academic attainment.

These results may seem surprising in light of the previous findings. We have already shown that socio-emotional skills at age 11 have little predictive power for age 14 cognitive skills once age 11 cognition is controlled for (along with other factors). But we view GCSEs as being different to the previous cognitive measures used, which are aimed at narrowly capturing cognitive capacity. As the previous measures are low stakes and pupils can't prepare for them, there is less of a role for socio-emotional skills to drive these results. This is not true for GCSEs where individuals can revise and where the stakes are high (Machin et al, 2020). This opens the door for the kind of socio-emotional skills that we measure to play a large role in shaping outcomes.

Across all our analyses, we find that GCSE results are strongly related to socio-emotional skill levels, encompassing resilience, study skills, positive social skills and attention - probably to a greater extent than is commonly recognised. And while for boys, cognitive skill levels at age 14 are more important than socioemotional skills in determining future GCSE prospects; for girls, the opposite is true, with socio-emotional skills at age 14 playing a bigger role.

Lower grades, widening gaps

We estimate that Covid-19 induced learning losses and declines in socioemotional skills will significantly damage the educational prospects of those children who were aged 5 at the time the school closures began, with boys up to 4.4 percentage points less likely to achieve five good GCSEs and girls 4.8 percentage points less likely to do so. The socio-economic gap in basic GCSE grades for 11-year-olds at the time of school closures is set to widen significantly - by 4.5 percentage points for boys and by 4.3 percentage points for girls.

The government's preferred benchmark in recent years is the percentage of pupils achieving grade 5 or above in English and maths GCSEs. In 2018/19, before the pandemic this stood at 43.2% for pupils in all state schools. Our research The disruption of the pandemic amplified persistent and long-term educational gaps among children from different economic backgrounds suggests that this could drop to 38.8% for boys and 38.4% for girls taking GCSEs in 2030. In 2022/23, government statistics showed that 45.3% of pupils achieved this benchmark.

These results represent a double whammy to the educational progress of Covid-19 cohorts: they are on course for the biggest overall decline in basic GCSE achievement for at least two decades, and a significant widening of the socioeconomic gap in GCSE prospects.

We estimate that the educational damage from Covid-19 will have an economic cost to the national economy, due to lower lifetime earnings because of falling GCSE achievement, of £31.4bn in today's prices.

The disruption of the pandemic amplified persistent and long-term educational gaps among children from different economic backgrounds

Growing socio-economic divides point to a step change downwards in the UK's relative income mobility levels. We estimate that for children aged 11 - who experience greater inequality in attainment due to Covid-19 - the relationship between parental and child income will increase by 15% for women and 12% for men. By international standards, this equates to a significant decline in income mobility.

Quick wins for long-term recovery

Over the long term, we would advocate for more investment in education overall. But in the current context, we believe a focus on low-cost and cost-effective policies that have the potential to level the education playing field is more useful for policymakers.

Given budget constraints facing governments, our aim here is to provide policymakers with some immediate "quick wins". The cost of these reforms pales in comparison with the economic costs from learning losses. We emphasise approaches that improve children's academic progress but also enhance their socio-emotional learning, and on equalising policies - those with the potential to benefit children and young people from underresourced backgrounds who have suffered disproportionately during the pandemic.

We recommend:

• A national programme of trained undergraduate tutors delivering academic and mentoring support to help boost pupils' foundational skills.
• An enrichment guarantee in schools so that all children benefit from wider activities outside the classroom that nurture socio-emotional skills.
• A national programme to measure pupils' wellbeing to create greater focus on wellbeing that is strongly linked to children's evolving skills.
• A dedicated research programme to develop evidence-informed approaches to school-parent and school-community partnerships.
• Rebalancing Ofsted inspections to explicitly recognise disadvantage and credit schools that are excelling when serving disadvantaged communities.
• A new deal for teachers clarifying working hours and creating a more balanced school calendar with a shorter summer holiday to help improve teacher and pupil wellbeing.

The proportion of pupils getting five good GCSEs in 2030 could drop to less than 40%; last year, 45% of pupils achieved this benchmark

Delivering on these low-cost reforms, we believe, would go some way to offset the falling achievement levels and stark educational divides that our analysis predicts.

While this research is concerned with the extra learning losses suffered by children and young people during the pandemic, it should be emphasised that this disruption amplified long-term persistent education gaps in England and other countries. The policies we propose aim to offer a strategic response to rebalance the education system so that it supports all children irrespective of their backgrounds.

Failure to rebalance the education system will risk failing a generation damaged from learning losses and declines in socioemotional skills. The likely fall in income mobility levels will cast a long shadow over our society for decades to come.

---

20 June 2024     Paper Number CEPCP678

Download PDF - A generation scarred by the Covid-19 pandemic

Download Press Release

This CentrePiece article is published under the centre's Education programme.

This publication comes under the following theme: Inequality and social mobility