How can UK policy-makers meet the increased need for tutoring?

Millions of children in the UK fell behind in their education while schools were closed in 2020. One-to-one tutoring is effective to make up lost learning, but costly. Offshoring online tutoring to lower-wage countries could allow for much wider access.

Since March 2020, children in the UK have missed almost half a year of in-school lessons. In England, schools were closed for 21 weeks across the two lockdowns and during further Covid outbreaks, with hundreds of classes and year groups sent home to isolate for further days and weeks.

The resulting loss of learning is a significant challenge for teachers and policy-makers to overcome. Additional tutoring – either one-to-one or in small groups – may be an effective solution. But such support is costly, and the supply of tutors is limited. Outsourcing, which may help to keep costs down, raises questions about teaching quality, particularly as the evidence on effectiveness of online tutoring is weaker.

What are the effects of school closures?

While schools were closed during the peaks of the Covid-19 crisis, lessons moved abruptly online. But teachers struggled to keep up with the curriculum, with 80% reportedly covering less material than usual (National Foundation of Education Research, 2020).

Pupil and parental engagement have also been significantly lower than normal, particularly among the most disadvantaged pupils and in the most deprived schools. Teachers report that, on average, just over half of their pupils’ parents were engaged with their children’s home learning. But the rate was significantly lower for pupils in the most deprived schools.

This meant that by the end of 2020, children in England were roughly two months behind where they should have been. This lag was present across virtually all subjects and year groups, according to separate research by the Department for Education, National Foundation for Educational Research and RS Assessment.

This matters as education is an important determinant of a person’s knowledge and skills (their ‘human capital’). It also influences how productive they are in their job (their ‘labour productivity’) and even their future employment and earnings.

Learning loss during the pandemic – due to almost four months of UK school closures – could equate to £350 billion in lost lifetime earnings across the 8.7 million school children in the UK, according to predictions from the Institute of Fiscal Studies.

Can tutoring help students to catch up?

Empirical studies show that tutoring, either individually or in small group settings, can be very beneficial to help students to catch up. There is also evidence that one-to-one tutoring is effective even when delivered by volunteers or peers. The Education Endowment Foundation (EEF) systematic review finds that one-to-one tutoring has ‘moderate impact for high cost, based on extensive evidence’.

It is therefore promising that in June 2020 the UK’s Department for Education announced a new, centrally subsidised National Tutoring Programme (NTP) to help children catch up. Under the initiative, £250 million has been allocated for pupils aged 5-16 and about £100 million for pupils aged 16-19. But an important constraint to the scalability of the programme are the high costs, as well as a lack of tutors in the UK.

Are there enough tutors in the UK in meet the increased demand?

Even before the pandemic, the UK faced a shortage of teachers in primary and secondary schools. The number of unfilled vacancies in state schools has been rising since 2010, as has the rate of teachers leaving. Attracting young people to teaching is a long-established structural problem, particularly in science, technology, engineering and mathematics (or STEM subjects).

Recruitment to teacher training in physics, maths, and chemistry is significantly lower than the required targets but did increase during the pandemic. This constrained supply, coupled with a 46-hour work week and the public sector pay freeze, leaves the UK education sector facing numerous challenges and in need of innovative interventions to manage the learning loss.

To make up for school closures, one estimate is that children may need two additional hours of teaching time per week over a year (Centre for Economic Performance). If a single tutor could teach 20 pupils over a week, and with around 8.9 million school children in the UK, this would imply a need for 445,000 private tutors. If we consider only the 1.4 million disadvantaged pupils who are potentially eligible for the National Tutoring Programme, delivering the extra tuition would require 70,000 full-time private tutors.

As private tutors do not need to be formally qualified or registered, there is not a precise figure for their present numbers. The Tutor’s Association estimates that there are up to 100,000 private tutors in the UK. Further, the Sutton Trust has estimated that nearly half of teachers (200,000) are either currently tutoring or have previously tutored. In addition, university students and recent graduates have also begun tutoring – often on a voluntary basis – since the start of the pandemic.

The supply of private tutors could expand further if the UK entered a recession and unemployment rose, as people may look for alternative sources of income. But overall, it is unclear whether there are enough tutors in the UK to accommodate a large increase in demand.

How affordable is private tutoring?

While the National Tutoring Programme is a step towards bringing affordable personalised tutoring to disadvantaged children, the allocated £250 million can buy only about six hours of tutoring time for each of the 1.4 million disadvantaged students in the UK (IFS Report, 2020). This is unlikely to be enough.

In addition to government funding, private tutoring is also widespread and on the rise. Due to the costs, pupils from more affluent households (34%) are substantially more likely than those from less affluent (20%) to use private tutoring (Sutton Trust, 2019).

In addition, the cost of tutors in the UK is rising. Economist William Baumol made the observation in the 1960s that prices tend to rise in the service sectors where productivity grows most slowly. He gave the example of a live string quartet – which takes the same amount of time to play a piece of music today as it did 100 years ago. As productivity rises in other sectors (such as manufacturing), wages – and therefore prices – also rise in the service sector with low productivity growth (Helland and Tabarrok, 2019). This is known as the ‘cost disease’.

Education is similar and the pattern predicted by Baumol’s cost disease theory is clear in the data (see Figure 1). Overall, the price of all goods in the UK has increased by 60% since 1988. But the price of services has increased by 346%, and the price of education by 938%.

Figure 1: Change in price by sector, 1988- 2020

Source: Office for National Statistics, Consumer Price Inflation data, selected sectors

How can trade reduce the price of education?

While the price of services has gone up in the last 30 years, the price of some goods, such as clothing, has actually fallen. To some, the reason for this decrease is clear – globalisation and the movement of production to lower-wage economies such as China (Barrell et al, 2006).

While there are downsides to globalisation, such as concerns about workers’ rights and environmental conditions, the evidence supports the basic economic theory that the benefits are large, including falling prices in developed countries and rising wages in poorer countries. Trade in services is less straightforward than trade in goods, but it is growing and becoming increasingly feasible with improvements in communication technology.

Global online tutoring, via platforms such as Zoom or Skype, could therefore offer a potential solution to help make up learning loss and reduce teacher workloads, particularly in disadvantaged settings. The average hourly price of one hour of one-to-one tutoring is estimated at £50 for the 22 NTP listed providers. One of these providers, a company called Third Space Learning hires its maths tutors from India and Sri Lanka, and offers tutoring at less than one-third of this price (£18/hour) under the NTP. We don’t yet know whether teaching standards are comparable.

In addition to questions around quality, there are concerns that outsourcing tutors from emerging economies could be exploitative. But at the aggregate level, it is argued that service exports are important for developing countries, particularly India. For individual workers, wages may be low by global standards but are higher than local average wages. Outside of migrating to a high-wage economy, workers in low-wage economies have little option to increase their earnings. Working directly for people in high-wage economies may be one way for people in poorer countries to increase their income levels.

How effective is online tutoring?

There is a strong empirical case for the effectiveness of one-to-one and small group tuition in general. But the evidence is less clear for online tuition with international teachers. An Education Endowment Foundation (EEF) study of Third Space Learning found disappointing results, with students experiencing no gain in learning from this tutoring. In response, the company argues that the recruitment and training model has changed and improved since that study was completed.

Another EEF trial of online tuition with four UK-based tutoring firms conducted in 2020 found that although students and tutors preferred in-person to online tutoring, the latter was feasible and students did still value it. Observational analysis of schooling during the lockdown also found that student engagement was higher in schools in which teachers had online conversations with students.

Studies from other countries during the pandemic also show good results. A large randomised control trial in Italy found that online tutoring (by Italian tutors) had substantial benefits for learning outcomes, and a smaller trial in the United States also found large positive effects.

Whether tutoring is delivered in person or remotely, its success depends on the effectiveness of the individual tutor and on the materials that they have to deliver. An earlier study from the United States shows that online learning can be at least as effective as in-person instruction, but there is a lot of variation between individual programmes.

Evidence on the effectiveness of online tutoring in general is therefore mixed, and extremely limited when it comes to international tutors. What is clear is that there is a large cost differential between UK and international tutors, and this is only likely to rise in the short to medium term. More evidence on the benefits and overall cost-effectiveness of international online is needed and would help policy-makers assess the viability of this policy option.

Where can I find out more?

How can we make up the learning losses from lockdown? Simon Burgess makes an early case for a national tutoring programme (in June 2020)
Learning loss: the National Tutoring Programme for England is a valuable step – but may not go far enough Simon Burgess and Hans Sievertsen review progress on the national tutoring programme (in November 2020)
One to one tuition The Education Endowment Foundation summarises the evidence on the effectiveness of one-to-one tutoring
Offshoring: opportunities and threats as services go global Dirk Willem te Velde outlines the opportunity to developing countries from service exports