Gabriel Heller Sahlgren is Research Director at the Centre for Market Reform of Education, and Affiliated Researcher at the Research Institute of Industrial Economics in Stockholm, Sweden.

While the British economy has made important strides in the past months, there’s still some way to go before we can talk about recovery proper. Indeed, GDP per capita remains under pre-crisis levels. But even when it finally recovers, the illusory question of how to increase long-term economic growth remains.

For many, including myself, human capital will play an important role. Better skilled workers will allow us to thrive amidst strong competition in the global economy. And in order to produce better skilled workers, the education system is key.

But there’s education and there’s education. Shuffling more people into upper-secondary school and higher education offers a simple solution to the problem. With more schooling years, the story goes, people will acquire the skills necessary to create a competitive economy. Indeed, this idea was partly behind the 2008 Education and Skills Act, which increased the compulsory schooling age to 18.

But simple solutions are rarely good solutions, and this certainly applies here. My paper, published today by the Adam Smith Institute, reviews the existing empirical research and provides new statistical evidence on the impact of education quantity and quality on growth. And, in fact, support for the idea that schooling years per se raise growth is thin, at least in developed economies.

On the other hand, a growing literature displays a strong effect of higher educational achievement – measured by international test scores – on economic growth. This research has become more methodologically sophisticated recently, suggesting that the relationship is indeed causal. In my own econometric analysis – using slightly different data, growth period, and methodology, while also carrying out more sensitivity analyses, compared to previous papers – I find results remarkably similar to earlier analyses: there is no impact of average years of schooling on growth, while test scores have a strong effect.

Indeed, the benefits of higher education quality are very large. My calculations suggest that if the UK had performed at the level of Taiwan in international tests since the mid-20th century, average annual per-capita GDP growth would have been about 0.8 percentage points higher in the period 1960-2007. This in turn means that our per-capita GDP would have been over £5,000 higher in 2007 than it was.

This raises the question: how can we improve international test scores in a cost-effective way? While input-based policies, such as class size reductions, can sometimes (but far from always) provide some benefits, those policies are rarely cost-effective. In other words, the return on such an investment in the form of higher achievement in international tests tends to be low.

Instead, research consistently finds that independent-school competition raises countries’ international test scores. The strongest available study suggests that the impact applies to independent school and state school pupils equally, and that competition also reduces costs. The productivity gains are therefore quite substantial.

The discrepancy between these findings and the widely reported result in the PISA report’s fourth chapter – which finds no impact of independent-school competition – is due to the latter’s methodological problems. The PISA report’s fourth chapter amounts to little more than a firework of correlations, and it can’t tell us anything about what works in education. In academia, few take the chapter seriously; any serious evidence-based policy discussion should ignore it.

So, independent-school competition appears to be a cost-effective way to raise international test scores, which in turn helps to spur growth. Indeed, my calculations suggest that if the UK had matched the Netherlands’ long-term independent-school enrolment rate in 1960-2007, average annual growth during this period would have been almost 1 percentage point higher because of the positive effect on educational achievement. This means that our GDP per capita would have been about £6,000 higher in 2007 than what it was.

The policy implications are the following: first, investing in education quality is probably a better growth strategy than merely making more people taking higher qualifications. Second, expanding school choice and competition should remain a priority in order to produce higher education quality.

In theory, parents have had the right to choose a school since the 1988 Education Reform Act. But this hasn’t been the case in practice. This is because most reforms focused on demand, while ignoring the supply-side of the equation. Yet without new supply of schools, choice becomes a zero-sum game. And in a system where proximity is the king of tiebreak devices, rich people inevitable comes out victorious in this game, simply because they can afford to move closer to the good schools. Among the poor, where it’s most desperately needed, choice and competition remain quite low. The free schools programme has certainly been a start in the right direction, but it has proven too feeble and expensive to provide the impetus we need.

The solution is therefore to liberalise the free school programme from its administrative shackles. Get rid of the current procurement-like process in which the government attempts to pick winners, and instead approve all providers, for-profit or non-profit, that meet stipulated minimum requirements. At the same time, phase out capital funding, and instead let the market determine which schools should get financial backing. This can be backed up by a voucher scheme to existing independent schools that meet certain conditions. In one stroke, England would instil significantly more growth-inducing competition in its education system.

While the foundations for the market are in place, it’s time to build proper structures and take a significant leap towards a well-functioning competitive system that target quality deficiencies specifically. Indeed, as my research shows, the economic benefits accruing from this are simply too important to ignore.

> Incentive to Invest? How education affects economic growth is published today by the Adam Smith Institute.