School closures carry a heavy economic price
For most children, and their parents, the novelty of home schooling has long since worn off. On top of the deficit in formal education and structured routine, children are missing their friends and all the social and sporting activities that are part and parcel of normal school life. This is adding to worries about mental health and food insecurity. Many parents, especially working mothers, are having to take on an even larger burden and step in as teachers too.
What’s more, there is now a growing awareness of the damage that prolonged closures are doing to the economy. This affects us all and could continue to do so long into the future. These economic costs can be thought of in three ways.
Let’s start with the long-term damage caused by young people missing out on education that could have increased their productivity, allowing them both to contribute more, and to earn more, for the rest of their lives. This may be a particular problem for poorer households which might benefit the most from a better start but are least likely to be able to study well at home, thus widening educational inequality.
However, quantifying this is difficult. Indeed, even the economics is controversial. There does appear to be a strong correlation between years of education and earning power, both when comparing individuals and countries (higher average levels of education tend to be associated with higher levels of per capita GDP). But there are two problems here.
First, beyond a certain point, education may simply be a “positional good”. In other words, it provides a signal which allows someone to distinguish themselves from other job candidates and thus earn more, without actually improving their individual productivity. This would go a long way towards explaining why the value of a university education appears to be falling in the UK, now that many more people are going into higher education.
Second, the fact that education levels tend be higher in richer countries does not necessarily prove anything about causation. It may only be richer countries that can afford to have large numbers of people taking time out to study not-very-useful degrees at universities.
Nonetheless, even sceptics of the value of some forms of further and higher education – such as Professor Alison Wolf – have noted the importance of a good start in life, and especially at primary school. There is also still plenty of evidence that the positive impact on productivity (enhancing human capital) is more important than the screening effect. And even if the main benefit of more years of education is to provide a signal to potential employers, weakening that signal would increase costs for businesses as they try to find other ways of screening candidates instead.
On balance, then, it does seem reasonable to conclude that the loss of education, especially in the early years, will have some negative impact on the economy as a whole – and that this can be proxied by the loss of future earnings.
A conservative assumption is that each additional year of schooling might raise incomes by eight per cent. (Many studies suggest a higher figure than this, although they often include developing countries where education may be scarcer and the returns therefore higher.) There is also plenty of evidence of a strong relationship between the number of hours of teaching received and exam performance.
If we then assume that, on average, children are effectively missing out on three months of schooling, or a quarter of a full year, that could mean that a generation might now earn two per cent less than they would otherwise have done, throughout their working lives.
That might not sound like a lot. But if we could think of any other intervention that could boost the UK’s annual productivity by two per cent, we would surely jump at it. Put another way, if we assume that someone might earn £1.2 million over their lifetime, two per cent of that is a loss of £24,000, or £24 billion in total for every million young people who are currently missing out.
There will also be two types of short-term cost. One is the value of the lost output in the education sector itself. Again, we need to be careful here. It wouldn’t be right to add the cost of lost output in the education sector to the long-term hit on future earnings. At best this would be double counting, because the benefits of a good education are presumably related to the value of the resources that are used to provide them.
Indeed, education is actually an input to achieve the output of additional earnings, so there is a case for deducting it from the future value of benefits. But we can use the output of the education sector as an alternative proxy for the value of those benefits.
For example, one study of the costs of the lockdown (led by Professor Richard Layard) has noted that the UK spends about £7 billion a month on schools. The authors of that study assumed that the present value of benefits equals double the costs, but that half these benefits are still being achieved. On this basis, each month of school closures reduces welfare by £7 billion, or £21 billion over three months.
Approached another way, the education sector accounts for a little under six per cent of the UK economy, but this figure includes nurseries and universities as well as schools. Schools are, of course, also still providing some education, either in person (mainly to children of key workers) or online. If we assume that the equivalent of two per cent of national output is lost for three months, that would be a total hit of about £11 billion over those three months (smaller than the Layard study, but still significant).
We also need to think about the negative impact on the productivity of working parents who are stuck at home. For example, if three million parents (out of about 33 million in employment) can only earn half of what they would normally do for three months, that could be another £11 billion in lost income. In practice, there could be several million parents who will not be able to work at all.
In total, it’s reasonable to suggest that every three months that schools are shut down could cost between £22 billion (about one per cent of a normal year’s GDP) and £33 billion (about one and half per cent). And in the longer term, every child affected could contribute at least two per cent less to the economy.
Of course, these are all back-of-the-envelope numbers. But the key point is that keeping schools closed will have substantial economic costs in both the short and the long term, even without taking account of the less tangible impacts on wellbeing.
Indeed, this is a good example of what is sometimes called the “identifiable victim” problem, where policymakers focus too much on harms that are relatively easy to identify and put too little weight on less visible costs.
This all might matter less if there were good evidence in favour of keeping schools closed to help the fight against Covid-19. But the scientific consensus (summarised here and here) is that children themselves are at a minimal risk from the virus and that transmission rates to adults, including teachers, are low. Consistent with this, many studies, including a large literature review, suggest that school closures will be relatively ineffective at reducing Covid-19 deaths.
In summary, this is something that should worry everyone. In my view, the government should continue to prioritise the reopening of schools and be willing to spend more to make that happen, including increased funding for summer schools, and investment in new infrastructure and equipment so that schools have the capacity to teach more children and still meet social distancing rules.
We could knock up a “Nightingale hospital” in a few weeks. Why not a new school building?
This post was originally published on Julian Jessop’s blog.