Why we should invest in education and how to measure the return

Not all skills are equally valuable. For example, my early work has shown that people who gain good mathematical skills, analytical skills and numeracy do much better in the labour market in terms of earnings. When we're thinking about this notion of education as an investment, we also need to think of education beyond a “one size fits all” generic concept. We should consider that individual decisions about what to study and what skills to gain will have an impact on how well one does in the labour market.

Some of the big issues are: whether to have a degree, what to study and where to study. In the UK, for example, there are many universities, but there's also a hierarchy. We found that studying at some elite, prestigious universities tends to give you a higher return on your investment. In other words, it gives you more earnings in the labour market when you graduate. These are all decisions that the individual will have to make when thinking about an investment in education.

Since education holds value for society, it matters beyond the individual. In general terms, as a society, we have decided that we should have several years of compulsory education. The reason for that is very straightforward: everybody gains if everybody is literate. For example, public health is a critical topic at the moment. Individuals’ understanding of public health issues and the ability to follow instructions will be determined by their literacy. As a society, we think it's essential that we ensure everybody has a minimum level of education, which, in a way, is the argument for compulsory education and for the State to pay for it. It wouldn't be enough to make education compulsory if individuals had to pay for their children to go to school. We would inevitably see that the rich would be able to pay for their children to go to school, whereas the poor wouldn’t.

There are some precise economic terms here. We call education a public good because the benefit goes beyond the individual. There are externalities, in economic jargon, which means that the benefits of investing in education from a society perspective are greater than the pure returns that an individual gets, and that's why we must have both compulsory and state-funded education.

When we measure the return on education, we tend to think in quite crude terms of what the value of an extra year of schooling is. Historically, it’s been between 5 and 10 % higher earnings per year of schooling. While it varies by context, we could use that as a rough indicator. What that doesn't tell you is how much that extra year of schooling costs, and also whether or not we could become more efficient in our school system so that the extra year of schooling or the extra pounds spent on it gives you a better return. We've spent a lot of time trying to figure out whether education systems are productive. For example, we use the OECD PISA tests to survey 15-year-olds regularly and to determine which education systems are better in terms of the skills that 15-year-olds have.

That is one way to measure the efficiency of a system. Still, we must understand that many factors have an impact on the efficiency of an education system. For example, in unequal societies, where you have a large number of low-income families that may be struggling to support their children, you would expect to have big gaps in achievement between rich and poor students. The education system does not necessarily determine that sort of feature of the system; the wider society determines it. Therefore, when we're thinking about efficiency in education systems, we need to be very careful not to attribute all the differences that we see across countries simply to the efficiency of the education system in a particular country.

First of all, we need to acknowledge that many of the differences in achievement levels that we see between children are due to other factors outside the education system, such as family background. The second thing we need to know is that differences in pupil achievement across schools are less important than differences in pupil achievement within schools. That means that there are children who are going to be doing very well and children who are going to be doing less well in the same school. Some of the factors behind that is the quality of teaching that they receive. We know that teachers and teaching quality is more important than just which school a child attends.

We generally measure this through test scores. We look at the gains that a child experiences in terms of their test scores and then examine whether children in one group who are taught by one teacher experience more rapid gains in their achievement than the children in another group who are taught by a different teacher. When we do that – and we've done that across Western systems of education and systems all over the world – we generally find that the teacher the child has is critical in determining their achievement. Teachers and teaching really matter.

It turns out that some of the apparent characteristics, such as the level of the teacher's qualification, aren't necessarily a good indicator of their teaching quality. Some researchers have gone further than that, and have gone into classrooms, observed teaching practices and then tried to score teachers in terms of their quality and effectiveness. You can identify some characteristics and pedagogical approaches that seem to be beneficial for pupil achievement, but it's quite a difficult area. Many of the things that we think we know about teaching quality are unproven. That's why it's important to assemble evidence based on observations and randomised control trials to try to determine what's likely to improve teaching quality.

It's also worth saying that what works in one context may not be as effective in another, which begs the question: how is it that schools in China managed to do so well for their pupils, even though the class sizes are enormous and they use very different teaching approaches to schools in, say, Scandinavia, France or Germany? Understanding why something works in one context and not in another is as important as working out what makes for high-quality teaching.