It’s so much easier to educate students for our past than for their future. Schools are inherently conservative social systems; as parents we get nervous when our children learn things we don’t understand, and even more so when they no longer study things that were so important for us. Teachers are more comfortable teaching how they were taught than how they were taught to teach. And, as a politician, you can lose an election over education issues, but you can rarely win one, because it takes far more than an election cycle to translate intentions into results.

So changing education bureaucracies seems like moving graveyards: it’s often hard to rely on the people out there to help, because the status quo has so many protectors. The biggest risk to schooling today isn’t its inefficiency, but that our way of schooling is losing its purpose and relevance. And when fast gets really fast, being slower to adapt makes education systems really slow and disoriented.

We live in a world in which the kind of things that are easy to teach and test have also become easy to digitise and automate. Education has won the race with technology throughout history, but there is no guarantee it will do so in the future. Students growing up with a great smartphone but a poor education will face unprecedented risks. When we could still assume that what we learn in school will last for a lifetime, teaching content knowledge and routine cognitive skills was rightly at the centre of education. Today, the world no longer rewards you for what you know – Google knows everything – but for what you can do with what you know. If all we do is teach our children what we know, they may remember enough to follow in our footsteps. But it is only if we help them build a reliable compass and navigation skills that they will be able to go anywhere and find their way through this increasingly complex, volatile and ambiguous world.

One of the reasons why we get stuck in education is that our thinking is framed by so many myths. So I start my new book, World Class: Building a 21st-century school system, by debunking some of the most common.

• “The poor will always do badly in school.” That’s not true: the 10% most disadvantaged kids in Shanghai do better in maths than the 10% most advantaged students in large American cities.

• “Immigrants will lower the performance of a country on international comparisons.” That’s not true: there is no relationship between the share of immigrant students and the quality of an education system; and the school systems in which immigrant students settle matter a lot more than the country where they came from.

• “Smaller classes mean better results.” That’s not true: whenever high-performing education systems have to make a choice between a smaller class and a better teacher, they go for the latter. Often it is small classes that have created the Taylorist culture where teachers end up doing nothing other than teaching, and don’t have the time to support individual students, collaborate with other teaching professionals or work with parents – activities that are hallmarks of high-performing education systems.

• “More time spent learning always means better results.” That’s not true: students in Finland spend little more than around half the number of hours studying than what students in the United Arab Emirates spend; but students in Finland learn a lot in a short time, while students in the United Arab Emirates learn very little in a lot of time.

• “The results in PISA are merely a reflection of culture.” That’s not true: rapidly improving education systems did not change their culture but their education policies and practices.

Why is our thinking so captured by myths and past practice? Because education systems have a habit of building “walls” that separate teachers, schools or the systems themselves from peer learning. When I started PISA, the central idea was to break those walls. The idea was to count what counts – that is, to collect high-quality data and combine that with information on wider social outcomes; to analyse that data to empower educators and researchers to make more informed decisions; and to harness collaborative power to act on the data, both by lowering the cost of political action, and at times by raising the cost of political inaction, as well.

The good news is that our knowledge about what works in education has improved vastly. In my book, I write extensively about what makes school systems successful, and what makes high-performing school systems different.

Still, knowledge is only as valuable as our capacity to act on it. To transform schooling at scale, we need not just a radical vision of what is possible, but also smart strategies to help drive change. The road of education reform is littered with good ideas that were poorly implemented. The laws, regulations, structures and institutions on which education leaders tend to focus are just like the small tip of an iceberg.

The reason it is so hard to move school systems is that there is a much larger invisible part under the waterline. This invisible part is about the interests, beliefs, motivations and fears of the people who are involved in education, including parents and teachers. This is where unexpected collisions occur, because this part of education reform tends to evade the radar of public policy. That is why education leaders are rarely successful with reform unless they build a shared understanding and collective ownership for change; and unless they build capacity and create the right policy climate, with accountability measures designed to encourage innovation rather than compliance.

Our task is not to make the impossible possible, but to make the possible attainable.

Many teachers and schools are ready for change. To encourage their growth, policy needs to shift towards inspiring and enabling innovation, identifying and sharing best practice. Such a shift will need to be built on trust: trust in education, in educational institutions, in schools and teachers, in students and communities. Trust is an essential part of good governance in all public services, and a key determinant of where great people want to work. But trust cannot be legislated and mandated; that is why it is so hard to build into traditional administrative structures. And trust is always intentional. Trust can only be nurtured and inspired through healthy relationships and constructive transparency. That is the lesson we can all learn from Finland, where opinion polls consistently show high levels of public trust in education. At a time when command-and-control systems are weakening, building trust is the most promising way to advance and fuel modern education systems.

In the face of all these challenges, we don’t need to be passive. While technology and globalisation have disruptive implications for our economic and social structure, these implications are not predetermined. Their outcomes will be determined by our collective response to these disruptions – the interplay between the technological frontier and the cultural, social, institutional and economic agents that we mobilise in response. We have agency, the ability to anticipate and frame our actions with purpose, and to devise and execute a plan to achieve that purpose.

I decided to write this book when I saw children from the poorest neighborhoods of Shanghai learning from Shanghai’s best teachers. It was then that I realised that universal, high-quality education is an attainable goal, and that our task is not to make the impossible possible, but to make the possible attainable. This is not rocket science; it is entirely within our means to deliver a future for millions of learners who currently don’t have one.

World Class: Building a 21st century school system is available as a free download here.