Dear Minister

An Open Letter To A New Education Minister

1. The most important job in government

Congratulations on your appointment. You now have one of the most important jobs in government. Education strongly predicts future economic growth and development. But there’s a catch: the relationship only holds for quality education.

Keeping that focus on quality is crucial, because it’s not where the attention of most governments and their development partners has been. Historically, most education systems combined high quality with limited access. Since the 1990s, that situation has reversed: access has improved a lot, and in most countries primary school is near universal. In 1950, the average adult in a developing country had about 2 years of schooling. By 2010, that had more than tripled, to more than 7 years of schooling. But the quality is so poor that children aren’t learning and overall levels of literacy and numeracy remain desperately low. According to one recent study of literacy in 51 developing countries, about 50% of women who reached Grade 6 left school unable to read a single, basic sentence. Few developing countries participate in the large cross-national assessments of learning outcomes that are routinely carried out in the world’s richest countries, but in those that do the inequality of learning outcomes between children from different countries is shocking. For example, in Zambia, just 2% of 15 year old schoolchildren achieve minimum proficiency in mathematics, compared to 77% in OECD countries, while in Senegal, just 9% achieve minimum proficiency in reading, compared to 80% in OECD countries. In short, getting out-of-school children into education is vital, but just pushing on access isn’t enough. You have to address what the World Bank has dubbed ‘the global learning crisis’.

Why the quality of education systems in many countries is so bad is a puzzle. Luckily, your profession makes you better placed than anyone to understand it: it’s politics! Delivering quality education is not technically difficult. It doesn’t require clever new technologies. Big improvements can be made through existing, proven, relatively inexpensive reforms. But the politics of making these reforms happen is really hard — and much harder than the politics of improving access was.

One reason is the cost-benefit equation — both when the costs and benefits fall, and upon whom. Expanding access is the ultimate quick win. By opening new schools or removing barriers to attending existing schools, you don’t just make parents and children happy: it also means new contracts for the construction companies that build schools, new photo opportunities for the local dignitaries that open them, new jobs for the teachers that staff them, and new opportunities to take a victory lap at global conferences for Presidents and Ministers. In the long run, the country suffers from an education system that isn’t equipped to cope with all the extra demand and isn’t able to deliver good value for money. But as the saying goes, in the long run, we’re all dead.

By contrast, getting serious about quality means creating losers now while the benefits only accrue over time. Eventually you’ll get the education system your country needs, but along the way you might have to win fights with teachers and administrators — respected, influential voices in their community and one of the largest and often most well-organised groups on the government payroll — or upset local communities, or embarrass other politicians. No wonder so many Ministers duck the challenge.

Even for Ministers brave enough to tackle the quality crisis, getting things done is difficult. In an ideal world, you could decide your policy and just trust the system to get on with implementing it. I’ve never met a Minister who’s found that to be the case. The best Ministers have to be firmly on top of the detail of policy implementation and delivery, not just its design. The trouble is, quality-focused reforms are harder to oversee from the Minister’s office than access-focused reforms. For example, it’s easier to verify whether 1,000 new teachers have been hired than it is to verify whether the classroom practices of 1,000 teachers have changed. Given this asymmetry, it is tempting — and maybe even logical — for Ministers to steer clear of quality reforms in favour of access reforms they can be more sure of having a grip on.

Finally, there’s another, deeper reason why reform is challenging. In most countries, teaching kids how to read and count is only one of the functions that schools and teachers perform. Politically, it may not even be the most important. Schools are also there to inculcate values, like obedience and respect for authority, that are important not just to governments but to families and communities. For the avoidance of doubt, there’s nothing wrong with these values — respect for authority and curiosity and critical thinking can go hand in hand. The point is that even if schools are doing an abysmal job of teaching, they may be doing a good job at inculcating these values, and so persuading local communities (or, for that matter, your colleagues in government) of the case for change may be harder than it looks.

In short, taking a system that was designed to deliver quantity (more children in schools) and making it deliver quality is the formidable task ahead of you. Turning things around will be a long-term mission. It won’t happen overnight, and it probably won’t happen during your tenure as Minister. Instead, your job is twofold: first, to take the first steps needed to get the country moving in the right direction; second, to change the national conversation about education, so that your successors follow the reform path you have set out. If you can accomplish that, it will be quite a legacy.

2. The burning platform

It’s said the first step to recovery is to admit you have a problem. Too many countries fall at that first hurdle. If one doesn’t exist, your first step should be to commission a high quality, independently administered, nationally representative assessment of what students at key grade levels — e.g. grade 3 and grade 10 — can do. Ideally you should do this as soon as you come into office, both so you can make sure responsibility for the picture that emerges rests firmly on the shoulders of those who came before you, and to create a ‘burning platform’ and sense of urgency for the changes you want to make in the rest of your time as Minister. If you can, make sure the assessment is linked to relevant regional or global assessments like PISA, TIMMS and PIRLS, or SACMEQ. Nothing focuses minds so much as knowing that your country is lagging behind its neighbours and rivals. That’s one of the reasons the World Bank’s Human Capital Index, one of the first really systematic efforts to compare countries on measures of education that reflect quality not just enrollment, is already galvanizing action in some places.

Use these results to start a national conversation about the need to address the quality crisis. Visit schools. Engage the teacher unions. Talk to businesses about their challenges in recruiting skilled staff. Above all speak to students and parents about their experience. Sometimes we expect too little of parents. They know the current standards aren’t good enough. That’s why many who can afford it are voting with their feet and sending kids to private schools, and why many others end up taking the heartbreaking decision to put the child to work rather than keeping them in school. What they don’t always realise is that they can and should insist on better. Enroll them in your campaign for change.

3. To govern is to choose

Let’s suppose you succeed in moving the conversation from access to quality. What next? As Minister, you are responsible for what happens throughout education, from early years to universities, from schools to vocational education, from basic literacy to advanced STEM (science, technology, engineering and mathematics). Chances are, the more time you spend looking at each of these pieces, the more broken each will seem. You are going to be bombarded by people telling you that their piece of the system is the most important to fix. You need to decide where you are going to focus your energies, which battles to pick. You can’t and shouldn’t ignore the other stuff, but if everything is a priority, then nothing is. As a wise man once said, “to govern is to choose”.

This is in part a technical question — what does the evidence say — but it’s also a political question. Depending on where you choose to focus, different constituencies stand to win or lose.

My advice is to prioritise two things: early childhood provision that means every child goes to school ready to learn, and the achievement of universal foundational literacy and numeracy for every child before the end of basic education at the end of Grade 9 (approximately age 15). Let me take these in turn.

What happens in primary and early secondary school is highly dependent on what happens in the earliest years of a child’s life: 90% of a child’s brain development happens by their fifth birthday. Conversely, children who have not built strong foundations in the early grades find it incredibly difficult to catch up later. But that’s only part of the reason to think you should make early child development (ECD) one of your priorities. The bigger reason is political and institutional. One of the clearest findings in social science is that, once established, institutions tend to be ‘sticky’ and resistant to change — old habits really do die hard. In most countries ECD provision has not hitherto been a major part of the formal education system, so there is more of an opportunity for you to get in at the ground floor, shaping what that provision looks like from its inception rather than having to unwind entrenched models and approaches that are not working. That said, more and more attention is now being paid to ECD, and regrettably in the growing enthusiasm about its potential impact there has been a tendency to gloss over the nitty-gritty details of implementation. Generalising from a few studies of small, expensive, and exceptionally high quality pilots to the conclusion that ECD is a silver bullet is not a good way to make policy. How you design large-scale, centre-based ECD, and the quality of the organisations running those centres, really matters, as the long-term evidence from at-scale programmes like Quebec’s shows. But with careful thought as to what is provided, by whom, and how well, ECD can have an outsized impact on learning with fewer of the political challenges of reforming existing provision.

What about the focus on achieving universal literacy and numeracy by the end of Grade 9? To build the economy of the future, you might think it’s more important that children in your country learn coding or advanced computational statistics. But if they master the basics by Grade 9 there are more and more ways to learn those higher order skills, in many cases from some of the best teachers in the world and for free, via the explosion in MOOCs (massive open online courses) from the likes of Coursera, EdX and Udacity. Conversely, if they can’t read or do basic maths, no amount of investment in STEM will make a difference. As the South African researcher Nic Spaull puts it:

You cannot leap-frog literacy. Learning to work collaboratively is not more important than learning to read. Learning to code is not more important than learning to read.That’s because you can’t do any of these other things properly if you don’t learn to read first. Some might say we can do both, together, reinforcing each other, but this neglects the fact that there are real trade-offs.

Prioritising your investment this way is also fairer. The students making it to the end of secondary school and especially those making it to university tend to be drawn from the ranks of the better off. These barriers aren’t just economic, though these obviously matter: they are about the quality of provision and the opportunities to develop the skills needed to progress to further study. By as early as age 8, the gap in literacy skills between children from the top fifth of the income distribution and the bottom fifth is 25 percentage points or more. Yes, SDG4 says countries should focus on 12 years of free quality education. Yes, the world’s most compelling and persuasive education advocate also says that’s what you should be doing. The problem is, the cost of providing education at this level is much higher than at earlier grades, and the profile of the students who get to benefit much more skewed towards the better off. If you’re not careful, you may end up directing your time and money in ways that are not very progressive. 12 years of free quality education is a good place to end up, but it’s not a good place to start.

Finally, focusing your energies on the parts of the system where the largest number of students participate is good politics. Education is becoming more and more of a priority for the public — surveys show it in their top 2 or 3 biggest concerns, behind only jobs. By improving education for the many not the few, you are enlarging the constituency that will feel the impact of your efforts.

4. Avoid distractions

Now that you have established your priorities, what policies and programmes will actually move the needle?

First of all, you can stop your Ministry and its development partners from doing the wrong things. One of the challenges in education is that there a lot of policy ideas which sound great on paper but which, the evidence suggests, aren’t necessarily the best way to spend your scarce resources. Worse, many of these ideas have an intuitive appeal that makes them very hard to kill off. Instead, they roam around education ministries like zombies, preying on unsuspecting Ministers.

It’s not (necessarily) that these policies are bad or harmful. It’s that they are distractions. There are better ways to spend your time, money and political capital which will easily get crowded out if you’re not careful.

Here are 8 distractions to watch out for:

  • Reducing class sizes. Intuitively, it makes sense that children will do better in small classes. But overall, it turns out that the quality of the teacher is much more important than the number of students they are teaching and that changing the composition of a class (so that students are at a more similar level of ability) is much more important than reducing its size. Classes have to be very large indeed to see any real effect from class size reductions, and even then the relatively high costs (of hiring the extra teachers required to bring ratios down) make this a poor way to spend scarce resources.
  • Updating the curriculum. National curricula are often terribly out-of-date (I know of one that calls for children to name the parts of a typewriter) so it feels like a no-brainer to bring it into line with the demands of the 21st century. The problem is that teachers and schools that are failing to teach students to read and write will also fail to teach 21st century skills, coding, STEM or anything else. Spending time and money hiring consultants or organising consultative workshops won’t change that fact. The biggest problem with most national curricula is not that they are out-of-date but that they are much too hard, expecting children (and, for that matter, their teachers) to master too much content too quickly before they have mastered the basics. When you have addressed the underlying effectiveness of your education system, making the curriculum more modern and relevant is a good idea. Until then, it’s a distraction.
  • Raising the qualification requirements for teachers. It’s been said that “the quality of an education system cannot exceed the quality of its teachers”, and there’s a lot of truth to that. But it doesn’t follow that you should require teachers to acquire additional qualifications before they are let into the classroom. There are two big reasons for this. First, paper qualifications are a relatively poor predictor of how well teachers will perform in the classroom; adding new requirements won’t help. One recent study found that when you adjusted teachers’ qualification levels for the actual quality of the education they had received and the knowledge they had gained, you would need teachers to be educated to advanced degree level just to be confident that they had mastered the subject knowledge to teach Grade 4 curriculum. Second, unlike in OECD countries where there are big differences between countries in the selectivity of the teaching profession (high performing systems tend to recruit teachers at the upper end of the skills distribution, whereas lower performing systems tend to be less picky), teachers in low income countries tend to be relatively well qualified and relatively high skilled compared to their peers already, so there are limits to what can be achieved by trying to make the profession more selective. On this one, the siren voices calling for 60% of all university graduates to be recruited into teaching have just got it wrong.
  • Buying laptops and other edtech. Edtech is the future — and it always will be. OK, maybe that’s a little unfair. But around the world, the track record of edtech in improving learning outcomes is pretty dismal and in low income settings it’s been particularly bad. The much vaunted One Laptop Per Child initiative has failed wherever it has been tried. A study in Lagos found that providing students with e-readers did not significantly improve learning outcomes. The reason is fairly simple. Edtech creates additional logistical complexity, and requires such careful thought in how it is integrated into broader classroom practice, that the costs almost always outweigh the benefits. There are some promising applications, and in future edtech is bound become a more important part of the education system than it is today, but for now hoping that edtech will allow you to leapfrog challenges in your existing system is wishful thinking.
  • Buying textbooks. This one is particularly counter-intuitive. Surely buying textbooks must be a good idea? Not according to four different ‘gold standard’ randomised controlled trials, which all found that textbook distribution had failed to improve learning outcomes. In some cases the material was pitched at the wrong level because students were so far behind where they were expected to be. In other cases, students didn’t see much benefit from textbooks because teachers and school leaders allowed other priorities to overshadow their students’ learning. In still others, getting low skilled teachers to turn the content in textbooks into engaging lessons proved easier said than done. It’s not that having textbooks isn’t important. It is. But the best evidence we have is that textbooks only work when two more fundamental conditions are in place: when the curriculum content is at a level students and teachers can access, and when teachers and school leaders are incentivised and managed to act in ways that promote student learning. Fixing these two issues should be your first order priority, and I would be cautious about spending lots of money on textbooks before you have.
  • Upgrading school infrastructure. Building and fixing up schools is popular with local communities and local politicians and it creates jobs for local businesses. Unfortunately it is expensive and does little to improve learning. Seeing the conditions in which students are asked to learn and teachers are asked to teach can be heart-wrenching, but you weren’t hired to make the easy decisions. With limited resources available, this isn’t going to help you fix the learning crisis.
  • Subsidising school feeding and school uniforms. There are three good arguments for these sorts of policies, but three (in my view, decisive) arguments against (Ministries of Education) adopting them. First the pros, starting with the obvious: feeding and clothing poor children is not just good retail politics, it’s a moral imperative. Second, there is at least some evidence that these programmes help increase student attendance as well as learning outcomes, particularly for the most disadvantaged students. Third, this is something governments, even those facing significant capacity gaps, have shown they can do and do well. Even if this isn’t the best policy for improving outcomes, isn’t it better to implement a sub-optimal policy well than a more optimal policy badly? Well, maybe. But here’s why I think you should be cautious. First, the the short term impacts you see from these policies tend to fade out over time once bigger, structural factors driving student drop-out kick in: in the end, students who aren’t learning won’t stay in school. Second, school feeding programmes in particular are not cheap, averaging $50–60 per child per year in low income countries. For most education ministries, that is a very large slice of per student education spend, limiting your fiscal space for other quality-enhancing reforms. Third, direct cash transfers appear to be as or more effective in tackling barriers to school participation, and with the considerable benefit that they can come out of a different Ministry’s budget, freeing up yours to spend on the programmes your Ministry alone is equipped to implement.
  • Eliminating ghost teachers. Cleaning up the payroll by getting rid of teachers who don’t exist or aren’t turning up for work will save some much needed cash and win you points with your Finance Minister and the World Bank. The trouble is, we know the problem isn’t restricted to the classrooms without a teacher. It’s also in the classrooms where there is a teacher but one who doesn’t have the subject or pedagogical knowledge to teach well. Getting rid of ghost teachers therefore won’t, by itself, actually lead to better learning outcomes. What it will mean is taking on the vested interests that have a stake in the status quo and whose support you might need to implement your other reforms. I’m not sure that’s the right fight to pick, at least not straight away. If you think you can do it relatively easily without expending a lot of political capital, go for it. If not, kick that can down the road.

5. Challenge and support

If those are the don’ts, what are the dos? Fundamentally, the education system needs support and challenge: support to raise the quality of teaching and improve the productivity of the teacher workforce; challenge to ensure there is the right accountability at every level of the system to make sure this support is being provided and to drive progress.

There’s good evidence that challenge and support have to go together for either to be effective. With accountability but not support, you may see schools adopt strategies that inflate their results but don’t change the underlying quality of teaching and learning and may even harm some children, like the practice of systematically excluding weaker students just before public exams in order to boost scores. With support but no accountability, well-meaning policies won’t have the desired effect. In Sierra Leone, a programme to distribute textbooks to primary school students ended up having zero impact. It turned out the books had got to schools, but the school principals considered it more important to preserve the books in good condition so kept them locked up in their office, rather than enabling their current students.

The question is how to bring challenge and support together and in a way that works with, not against, the grain of the politics of your education system. Rather than a fixed blueprint, that suggests the best approach is to give you some broader reform orientations and let you figure out how best to apply them in your context.

6. Getting more from your teachers

Let’s start with teaching. And I use the word teaching not teachers deliberately. There’s too much emphasis on changing the composition of the teacher workforce, which is both practically and politically difficult, and too little on getting more out of the teachers you’ve got. We now have pretty good evidence to support three reform orientations when it comes to improving teaching.

The first is what’s come to be known as Teaching at the Right Level (TaRL). Children can only learn if the content of each lesson is pitched appropriately. But in most countries, the distribution of knowledge and skills among children in a given grade doesn’t look how the system expects it to look. First, the average skill level is far below what, according to the national curriculum, students in that grade are expected to be able to know and do. Second, there is very wide variation around this average, with some students exceeding grade expectations and some miles behind. This makes life incredibly difficult for teachers, who invariably end up teaching at a level that works for more advanced students but leaves other students struggling. TaRL simplifies things by grouping students for at least part of the school day by ability level rather than age. Simple as it sounds, this has been shown to deliver large improvements in student learning and at very minimal cost. Implementation requires careful thought, in particular to secure buy-in from parents, but TaRL programmes have been successfully adopted and scaled up in a number of countries.

The second is what goes by the clunky phrase of ‘structured pedagogy’ but is essentially about providing teachers with a lot more guidance on how they should deliver their lessons, often in the form of a ready-prepared teacher guide or lesson plan. We know that both teachers’ subject knowledge and their pedagogical knowledge is very low. In a recent study of teachers in 7 countries covering 40% of the Sub-Saharan African school population, 90% of primary school language arts teachers had mastered at best the subject knowledge expected of students in Grade 5. If this is the starting point, expecting teachers to be able to translate their subject knowledge into effective, engaging lessons is unrealistic. By removing this burden and allowing teachers to focus on how they teach rather than what they teach, lesson plans seem to be an effective way to raise the quality of teaching.

The third is coaching. Initial teacher training matters, but learning something in training and applying it in real life are not the same thing. That’s where coaching, either from more highly skilled teachers within the school or from specialist instructional coaches that work with a number of schools, can help. There is promising evidence that providing teachers with frequent, regular, on-the-job feedback about their teaching helps improve the quality of their practice and, with it, student learning.

7. Incentives matter

None of these approaches — Teaching at the Right Level, structured pedagogy or coaching — are technically that complicated for schools or administrators to implement. But until there are consequences for what happens in schools — at every level of the system — things won’t change. Incentives do matter.

In principle, parents ought to be a key source of accountability. In practice, when the power and resources to fix problems sit with local officials, principals and teachers — individuals who are typically more educated and higher status than themselves — parents may struggle to make change happen. There are some examples of reforms that have successfully empowered parents, but in general bottom-up accountability is an unreliable route to higher performance.

‘Top-down’ accountability needn’t be a dirty word. The trick is to recognise that effective accountability requires both bite and brain. Accountability must have teeth or you end up with a lot of “cheap talk”: school leaders and administrators going through the motions without really embracing change. But dumb accountability, that delivers the ability to make judgments of performance but not the capacity to support its improvement, can be just as harmful.

There are many possibilities here, but the most promising approaches to strengthening accountability seem to involve:

  • Setting clear, aspirational, but achievable standards that schools are expected to help children reach;
  • Setting clear expectations about the teaching practices and approaches schools are expected to adopt in order to achieve these standards;
  • Intensively monitoring schools to track whether these expectations are being met;
  • Linking this monitoring to the provision of practical support where schools and teachers are falling short;
  • Building the system’s capacity not just to do the monitoring but also to do the practical support;
  • Taking the time to test these changes on a small scale before rolling them out across the wider system.

This might sound obvious but the devil is absolutely in the detail of how these reforms are implemented, and how thoughtfully they approach questions of motivation.

The extraordinary success of the Tusome Early Literacy programme in Kenya, for example, seems attributable to the way it developed a cadre of local Ministry staff (‘Curriculum Support Officers’ or CSOs) able to coach, support and challenge schools on their teaching and learning. But this didn’t happen by accident: the reform process included rewriting the job description for this role, rehiring all staff against this new job description, and providing both regular training and on-the-job coaching for CSOs on how to perform it. And there were harder incentives too: the frequency of monitoring visits by CSOs was monitored using GPS tracking, with travel expenses reimbursed only if the target number of visits was hit. The result was dramatic increases in the frequency of visits to schools, and large, cost-effective impacts on the quality of teaching and learning.

By contrast, a recent study of a large-scale school management reform programme in India shows that it can be easy for accountability reforms to look like they are focusing on the right things when they aren’t. More than 2000 schools were randomly assigned into a treatment group and a control group. The treatment group received a comprehensive school improvement programme modelled on international best practices; the control group received business-as-usual management by the local and state authorities. After 3 years, researchers found the programme had comprehensively failed to improve learning outcomes. The problem was the accountability was too weak. The treatment schools conducted learning assessments of students, received school ratings based on these assessments, and developed sensible improvement plans to improve these ratings. But when it came to the hard graft of improvement on the ground — by school leaders in implementing their plans, but more crucially by local inspectors in following up with schools to make sure they were implementing their plans — it just didn’t happen. The researchers’ conjecture is that the programme added to the responsibilities of implementers without changing their incentives or capacity. “Implementation quality”, they concluded, “is a first-order issue in developing countries and worthy attention in and of itself.”

8. Getting things done

This difficulty of implementation highlights the last — but maybe the toughest — question of all: capacity. The most important decisions you make, argues business guru Jim Collins, “are not what decisions but who decisions”. Where’s the capacity to get this done going to come from? It’s a hard question to answer for most Ministers. Often the experience of going into government feels like entering an unwinnable race between expectations and capacity. As Liberia’s Nobel Peace Prize-winning former President Ellen Johnson Sirleaf once put it, “everyone wants to see change take place right away. After all, they voted for you because of your capacity to deliver — immediately. Only you cannot. Not because of the lack of financial resources but simply because the capacity to implement whatever change you have in mind does not exist.”

If like most Ministers you find yourself grappling with the challenge of low capacity, you have two options: to create new capability within the Ministry, and/or to pull in additional capacity outside the Ministry by leveraging the non-state sector.

First, you need to set about creating a culture within the Ministry that is focused on reforming not maintaining the status quo. One way to start is to invest in a specialised delivery unit or team working directly to you as Minister. In the first instance, this team can help you figure out what the implementation process needs to look like and where the potential weak links in the chain might be. It can then act as your eyes and ears in making sure this process is actually being followed. This calls for quite a different, more execution-focused skillset than the more traditional policy formulation skillset in which ministries have historically specialised. Governments around the world have experimented with delivery units, and there’s increasingly good evidence about what’s required to make them a success, both when it comes to delivery in general and in education in particular. The key point is that “set it and forget it” won’t work here; these units are only as useful as you enable them to be.

Second, you should leverage the non-state sector, both as a private complement to public provision and, where appropriate, as a direct partner in public provision. On the former, significant minorities of children are already being educated outside the government system in Africa; by 2021, this proportion is expected to rise to 1 in 4, and in urban areas it is already much higher. Of course, even at this scale the non-state sector by itself is not going to solve the learning crisis — not least since on average the quality in private schools tends only to be marginally higher than in government schools, once differences in student selection are accounted for. But given that this sector doesn’t cost the government money (indeed, if anything it saves money by reducing burdens on the public system), and given that for the foreseeable future significant minorities of children will be in the non-state sector, you should still look for low-cost, high-leverage ways to maximise its impact. This really comes down to the regulatory framework. Modern approaches tend to be flexible and principles-based, focusing on key outcomes like health, safety and learning rather than prescriptive rules about how many acres of playing fields schools are required to have.

Engaging the non-state sector directly as a partner is a more complex but potentially higher impact endeavour. A number of countries are experimenting with separating government’s fundamental responsibility to equitably fund public education from its historical role as a monopoly provider of it. There are two main rationales for these public-private partnerships (PPPs). One is to rapidly increase the supply of school places where existing public provision cannot meet demand and where it is costly or logistically challenging for government to do this directly. For example, government can give parents vouchers to cover all or part of the cost of attending non-state schools or can pay non-state operators to establish and run new publicly-funded schools. The other is to improve the quality of existing schools, by contracting out their management to non-state operators.

Many of these reforms are still relatively new and the evidence base on PPPs remains quite limited. It is also true that for these programmes to work the ‘public’ bit of the PPP equation matters as much as the ‘private’: governments need to be able to design, contract for and manage programmes effectively if they are to see the results they want. Experience suggests this is a challenge for governments all over the world never mind just in developing contexts.

Nevertheless there is some promising evidence to suggest that, with the right checks and balances in place and where PPPs manage to attract (or promote the emergence of) high quality non-state school operators, they can make a big impact at speed and at scale, and bring additional capacity, energy, expertise and innovation into the system.

9. It’s not (all) about the money

At this point, you are probably asking yourself: what about money? These things all sound costly, so should you be picking a fight with the Minister of Finance for a bigger slice of the budget? The answer is obviously: yes. Education is chronically underfunded in most countries, and countries with better education systems tend to spend more on them. In the end it won’t be possible to achieve the progress you want without a bigger budget.

But you have to be realistic that the reason finance ministries and development partners have been reticent about funding education to the level required is partly because they haven’t been convinced the money will be put to good use. And their scepticism is justified. Spending more money doesn’t tend to yield significantly improved outcomes. On average, a 10% increase in spending yields less than a 1% improvement in outcomes. For most low income countries, doubling the amount you spend per child will only raise the “learning-adjusted years of schooling” (LAYS) children receive by 0.5 years. (LAYS is a measure that adjusts the amount of schooling children receive to reflect the quality of the learning happening along the way.) There are some notable examples of very large increases in budgets completely failing to produce results. In Indonesia, a tripling of government spending on education over 14 years delivered essentially no improvement in the percentage of young people leaving school with a mastery of basic numeracy.

The attitude of one major funder is instructive. After many years of calling for a “Bill Gates for Education”, some campaigners got their wish when the world’s biggest philanthropist announced plans for his Foundation to move into global education. But unlike in global health, where the Gates Foundation has put billions of dollars to work, the Foundation is approaching global education much more cautiously, with (for the time being) a fraction of the budget and a focus on supporting research and data to help identify promising reforms rather than going big on a reform template that already exists.

Instead of being frustrated by this, consider it another part of your legacy: proving that education can be a great investment, and thus securing more financial headroom for your successor than you have had to make do with.

10. Conclusion: not easy, but simple?

The scarcest commodity a Minister possesses is time, so I won’t take any more of yours. As an American president once said, “there are no easy answers, but there are simple answers”. The answers in global education are not complex, they are simple: provide the practical support teachers need to improve their teaching, and hold schools and administrators accountable for making sure this support translates into better results. What is complex is the politics of reform, which is why the first step for any country hoping to change things is to have the right Minister in charge. You’ll find no shortage of advice on what to do. But the what is the easy part. It’s figuring out the how that will make you a great minister.

If there’s anything I can do to help, I would be delighted to try.

Paul Skidmore is the founder and CEO of Rising Academies. He writes here in a personal capacity.