Since they surprised the world with their stellar performance in the 2000 PISA test, Finland has been the edu-tourism destination of choice, with countless government delegations making pilgrimages to visit this education poster child. A few weeks ago, former education minister of Peru and now head of education at the World Bank, Jaime Saavedra, visited Finland’s schools and enthusiastically shared what he saw. Ministers from Eritrea have also visited, Haiti has signed a partnership agreement with the education department in Helsinki, and Argentina is drawing policy lessons from Finland as part of its “education revolution.”
But does this practice of “policy shopping” — visiting countries and picking policies to take home — actually work? Is there really anything happening in Nordic Finland, ranked 23 in the human development index, that is relevant or transferable for Eritrea, ranked 155 places lower on the index and with vast differences in its context and resources?
The consensus among almost everyone who works in education is that there isn’t; that the differences between countries means that taking one thing from another will almost always fail. And yet ... One thing that there is consensus on is that the neuroscience of children’s brains is universal. In other words, regardless of context, some things are the same when it comes to teaching and learning.
We argue that policy shopping — if done well — can be a sensible strategy. Rather than pretending that it’s best for every country to start completely afresh, it can be smart to look elsewhere to see what other countries have done that could address particular areas of underperformance. With the United Kingdom’s approach to international education now including an explicit focus on “exporting” British education expertise, it is timely to consider how this can best be done.
First, let’s look at a couple of examples where education policy shopping has been effective.
Ten years ago, mathematics performance in the U.K. was stagnating and policymakers and practitioners looked abroad for inspiration to break this cycle of underachievement. Struck by Singapore’s success, they decided to run some pilots to see whether teaching “Singapore math” could work for British children. Initial data showed children who were taught the Singapore math learned faster than their classmates — making, on average, an extra month of progress in a calendar year. The program has since been rolled out to almost half of the U.K.’s primary schools.
Ten years ago, a Vietnamese delegation visited Colombia to look at their Escuela Nueva, a pedagogical model using a participatory learning methodology that started in the 1970s in a handful of primary schools in rural Colombia. Impressed by the model, Vietnam piloted the approach in 24 schools in six provinces. It was enthusiastically embraced by teachers, communities, and local government officials and was scaled, eventually reaching 500,000 children across the country. The evaluation showed early signs of positive impact on cognitive and noncognitive achievement, at a cost of roughly $40 per student per year, about 2.5 percent of total expenditure.
But, as many folks in education would tell us, policy shopping can go wrong. After the end of apartheid in South Africa, the new government looked to Australia for inspiration and imported their model of “outcomes-based education.” The approach did not achieve expected results in South Africa, mostly because the decentralized approach led to additional work and expected too much of teachers, given their capacity and the limited resources they had at their disposal. Outcomes-based education was eventually abandoned 15 years later.
Without wanting to oversimplify complex matters, what might the drivers of successful policy shopping look like?
1. Countries need to be comparable.
Policy shopping seems to work best when the “buyer” and the “seller” are relatively similar in terms of wealth, resourcing, and system functionality. If Singapore can afford something that has a transformative impact on learning outcomes, England probably can too. Australia and South Africa, on the other hand, are vastly different in terms of the resources they have at their disposal and their system’s capacity to absorb and execute new models and interventions.
2. Buy a policy, not a set of outcomes.
Education ministers wanting to buy and replicate Finland’s success will almost certainly fail. Buyers need to know what it is that they’re buying. Seller countries need to show good evidence on the “what”, the “why” and critically also the “how” of their models, not just a set of system-level outcomes. Of course, evidence will always be contested. But shopping seems to work better when the buyer is looking for something explicit, and the seller has data on the drivers of success, and can explain the method of implementation of a particular model.
3. Context matters.
Contextualizing the policy to make it fit the local context is vital. Part of the reason outcomes-based education failed in South Africa was down to the unique and challenging post-apartheid context. Importing an Australian curriculum, without taking sufficient account of the realities of South African schools and teachers’ capacity made the approach, in the words of Professor Jonathan Jansen, destined to fail. Good policymakers act as tailors as well as shoppers: Picking a model that meets their country’s particular needs, and then altering and adjusting until it’s a great fit.
4. But capacity is important too.
We argue that reforms that fail generally do so because of lack of capacity, not context. Context does not affect the way a child’s brain “learns”: the fundamental neuroscience is the same in Perth, Australia, as it is in Pretoria, South Africa. Why, then, do some programs not translate? It is likely that the Australian curriculum example, which relies on sufficient ability and willingness amongst teaching staff to design their own resources and plan curriculum sequencing, struggled to be replicated in South Africa, with its very different financial and human resources — in other words, not because South African children learn differently to Australian children.
Simply put, it’s difficult to design programs for high fidelity and reliable implementation in weak public education systems
Policymakers should indeed be cognizant of context, and must scrutinize their policy purchases carefully. But that should not stop them from looking at options and innovations from elsewhere, particularly in systems with comparable capacity. It’s often an unpopular perspective, but there is a risk in taking the argument around contextual differences too far. It can even be used to mask difficult but important conversations about system underperformance. Admitting that there is something to learn from elsewhere isn’t always easy. But it may, ultimately, lead to quicker system improvements than reverting to contextual differences as an objection when presented with alternatives that could work.