By Sir Michael Barber, Chief education advisor, Pearson

Over the last decade, international benchmarking of education systems has become ever more prevalent. More importantly, it has become increasingly influential in shaping education policy at local, regional and national levels. As studies by OECD-PISA and TIMSS become more sophisticated and longitudinal time sequences develop there is ever more to learn about what successful education systems look like and how success can be achieved. 

In the early days of international benchmarking, education ministers and other leaders tended to worry more about the media impact than the implications for policy. However, once the regular routine of published PISA results was established, in 2001, this changed. Germany, for example, found itself much further down the first PISA rankings than it anticipated. The result was a profound national debate about the school system, serious analysis of its flaws and then a policy response to the challenges that were identified. A decade later, Germany’s progress up the rankings is visible to all. 

Now, in fact, we are beyond the phase of individual country reactions. Increasingly what we see is a continuous dialogue among education ministers and top officials around the world about the evidence from international benchmarking and the implications for education reform. Education ministers in places such as Singapore are constantly monitoring and visiting other countries to learn what they might do better. Arne Duncan organised a series of international dialogues with fellow ministers and union leaders about the future of the teaching profession around the world. Meanwhile Michael Gove, Secretary of State for Education in the United Kingdom, has shown more interest in international benchmarking than any of his predecessors. 

The continuous benchmarking series also enables more sophisticated analysis of what works in education, which leaders from around the world can draw upon. I have been involved in a series of three publications which have explored the lessons in depth. The first of these, written with colleagues at McKinsey, How the World’s Best Performing Schools Come Out on Top, examined the lessons from the most successful school systems, and highlighted the importance of recruiting, training and developing great teachers. 

The second, also written with colleagues at McKinsey, How the World’s Most Improved School Systems Keep Getting Better, took a new angle and looked not at what great systems do, but at how, over time, systems come to be successful. The third, published earlier this year with colleagues from Pearson, Oceans of Innovation, went a step further and asked whether achieving educational success as measured by PISA and TIMSS was sufficient to ensure a country was on track for economic and social success in the 21st century. The work of Eric Hanushek has likewise connected PISA and TIMSS outcomes to the wider goals of society, especially GDP growth. Eric has demonstrated a strong correlation between the quality of school systems and economic growth. 

His work points directly to the reason we supported the Economist Intelligence Unit (EIU) in the development of The Learning Curve. Here we have assembled in one place a wide range of data sets which will enable researchers and policymakers to correlate education outcomes with wider social and economic outcomes more easily than ever before. 

In assembling these data sets we looked at a wide range of correlations and have been, judiciously I believe, cautious in interpreting the results. This avoids significant pitfalls, including, of course, the fact that correlation does not imply causality. Nevertheless there are some clear messages. For example, the report highlights the importance of culture and teacher quality in education. We should note that even if we can be clear, for example, that better education leads to less crime, there is still an issue about how long after the school system improves we would see the reduction in crime. 

And of course the data sets themselves are by no means perfect. One of the reasons we are making them available in this format is that we believe this will encourage those responsible to address the data quality issues that are raised. Our intention is that the data sets available through The Learning Curve will be updated as new data appears. We are therefore making available an open, living database which we hope will encourage new research and ultimately enable improved education policy. In this way, we hope to promote a growing and welcome trend around the world towards evidence- informed education policy. The challenge then for policymakers is less knowing what they should do than having the courage to act on the evidence. For example, acting on the clear message that reducing class size is expensive and has little or no impact on system performance. 

This report includes a number of country rankings. These always generate interest and should be seen in the context of the issues raised here about the quality of data available. This is particularly the case with graduation rates which for the moment are based on national data sets involving a range of different definitions. We hope by publishing this particular ranking we will generate debate about how to improve data consistency as well as about the underlying policy issues.

We hope this research programme prompts a lively conversation on how we learn more about learning.  If you have any comments or reflections on the issues raised in this report, please send them using our suggestion box or via email at
Sir Michael Barber
Chief education advisor

Download the report

Download PDF

The Learning Curve videos

View videos

Want the raw data?

Request the data

Suggestion box

Please contribute any comments, thoughts, feedback or suggestions

Make a suggestion