The political economy of stagnant learning outcomes: Insights from Rafiullah Kakar
Rafiullah Kakar is a public and international development expert currently serving as the Member Social Sector & Devolution at the Planning Commission of Pakistan. Prior to this position, he served as an Education Policy Advisor and Specialist for the Government of Balochistan and has worked with organisations like the World Bank, United Nations Development Fund, United Nations Educational, Scientific and Cultural Organization, and The Commonwealth. A Rhodes Scholar, he holds two Master’s degrees in Global Governance and Diplomacy and Public Policy from the University of Oxford. Rafiullah has a long history of advocating for social and youth development, education reforms, and conflict management in Pakistan, and specifically in Balochistan.
Photo Credit: Tom Jamieson for The New York Times
In his paper “School Education in Post-18th Amendment Balochistan: A Political Economy Perspective”, Rafiullah Kakar examines the proximate and deeper causes of stagnant education outcomes in Balochistan. He uses Mushtaq Khan’s political settlement framework, combining his theoretical expertise in political economy and his practical experience in government to inform his analysis. In this interview, Rafiullah Kakar talks to our Junior Associate, Research, Manahil Naeem, about his paper and dives into some of the issues driving low learning outcomes.
Manahil: Hi Rafiullah sb, thank you for taking out the time to speak to us about your paper on “School Education in Post 18th Amendment Balochistan”. I would like to start by talking a little bit about you. We all know the facts, we know you were a Rhodes scholar and are currently a member of the Planning Commission, but tell us a little more about yourself and when and how it is that you became interested and invested in education policy and planning, specifically concerning Balochistan.
Rafiullah Kakar: Thank you for reaching out to me! My interest in public policy and development is rooted in my childhood experiences in a remote village on the Pakistan-Afghanistan border in the Qila Abdullah district of Balochistan. It was here that I experienced first-hand the damaging consequences of deficiencies in development and public service delivery.
I was fortunate to escape the conflict trap and have the opportunity to study at some of the world's best universities, thanks to the scholarships I received. From my own experience, I realised the transformative potential of education and the upward mobility that high-quality education brings to families who otherwise have little chance of moving up in the world. There was no other major change in our family other than the fact that I got the opportunity to study at some of the world's best institutions. I wanted our public education system to be a path to transformation for others, too. That’s what sparked my interest in education policy, though at the time, I wasn’t sure if I wanted to work in the public sector or somewhere else.
After completing two Master’s degrees at Oxford, I worked with international organisations in both the United Kingdom and Pakistan. Through my roles in various organisations and capacities, I quickly realised that these international organisations and donors, while valuable, weren’t the ones actually implementing the plans and policies they recommended. At best, they provided strong policy research and strategic frameworks, but the real implementation happened within the public sector. In other words, the outcome and success of all this quality research and planning ultimately depended on the average policy makers within the government. It was this realisation and the desire for impact that inspired me to look for entry points into public service soon after I returned to Pakistan.
I started off working with the Government of Balochistan, where I was involved in various reform efforts, including sectoral planning and reforms in the education sector. However, after a few years, I came to realise that the core issues in education weren’t really about the lack of solid or technically-sound policies or even insufficient public funding, which is what many outside the policy world still think. To borrow the famous phrase from American politics, it was all about implementation—'it was the implementation, stupid’. Since then, I have been trying to understand the systemic drivers of the inefficient and ineffective implementation in education and other social sectors.
Manahil: Could we start with a broad overview of your paper and then we can move on to some of the more specific questions I have on the proximate and deeper causes contributing to low learning levels in Balochistan?
Rafiullah Kakar: So, like many other well-meaning outsiders, I also thought that the two biggest problems with our education system were a) we don't spend enough on education and b) the government doesn't have technically-sound policies for education. I would like to clarify that these statements are accurate, but my contention is that they don't represent the most pressing issues of our education system. In Balochistan, I had the chance to work with an experienced and talented team to develop a comprehensive, research-based Five-Year Education Sector Plan for the provincial government. After that, I spent the next two to three years trying to ensure this plan was actually implemented, and it became clear to me that the main issue wasn’t a lack of good, technically sound policies
So, that’s how I got drawn into the whole puzzle of implementation and the politics around it. As a researcher, the big question for me was this: if we have solid policies in place and the public financing data also shows significant increases over time, then why aren’t these policies and increased spending translating into better education outcomes? First, I looked into the immediate technical drivers behind this slow progress. I realised that the planning processes were weak, often discretionary, and heavily influenced by political considerations, and that the monitoring and accountability systems weren’t effective. But then I had to ask: who’s responsible for this? Is it the politicians, the bureaucracy, the teachers, the parents, or the larger political culture? In my quest for understanding the deeper drivers, I found myself needing to study further and update my own academic understanding as well.
I studied political economy literature to understand the deeper determinants of weak implementation of public policies, and I stumbled upon Mushtaq Khan’s political settlement framework. I felt Khan’s framework applied very well to our context because it took into account the influence of informal structures as well as politics in shaping education service delivery. Political settlement has since become my favourite topic. I can talk for hours and hours about it but that's the background of why I conducted this research.
Manahil: I find it very interesting, and you just spoke about this also, that the Balochistan education budget has increased significantly since the devolution, but we don't see a corresponding increase in the learning gains or even improvements in access. And like you said, the most commonly identified cause of our education problems is usually a low budget. However, your findings challenge the simplicity of that statement and draw our attention to more complex factors. Could you speak a little bit about some of the proximate causes of stagnant education outcomes in Balochistan?
Rafiullah Kakar: I think one of the biggest challenges in advancing the education agenda is that learning itself is hardly ever a priority in our education policy—and that’s before even touching on implementation issues. You rarely, if ever, hear the finance minister, the education minister, or the chief minister talk about learning. Occasionally, you might see a mention of learning and education quality in documents prepared by external consultants, but that’s about it. On top of that, our education data system doesn’t measure and monitor learning outcomes. So, even though we have real-time school monitoring and EMIS systems, we’re still missing data on actual learning. And what doesn't get measured doesn't get improved. That’s where we are in terms of prioritising learning.
The bigger puzzle for me was why access indicators weren’t improving. The government wasn’t just making big promises about expanding access; they were actually backing it up with significant public investments in education infrastructure. In some years, Balochistan allocated 20% of the total development budget to education, which was pretty impressive. But I came to realise that the real trick lied in how and where that money was spent.
The development planning process in Balochistan is almost entirely politically driven, with no needs assessment involved. From identifying projects to implementing and contracting for infrastructure work, the whole process is controlled by the relevant MPA. There’s minimal input from the education department’s technical experts or the local school managers. Members of the provincial assembly decide where new schools will be built or which existing buildings will be renovated, often leading to development funds being spent on schools in areas where they’re not even needed. Even more concerning, many of these newly constructed schools remain non-functional for years after completion because the necessary teaching and non-teaching staff aren’t hired. The Finance Department resists creating new posts, seeing it as an added burden on the budget, especially given that most of these schools were built for political reasons. Interestingly, the MPAs themselves often have little interest in making the schools operational since they don’t control the hiring process for teachers, which is managed independently through testing. As a result, even our access indicators are stagnant because our development projects are weakly aligned with the most pressing access-related needs of the education system.
Another issue with how the education budget is spent is the large sums allocated to big campuses for both university branches and residential boarding schools and cadet colleges. The argument against investing in costly boarding schools is a nuanced one because, in fairness, some of the semi-decent quality education we see does come from these schools. However, the real question is whether these schools provide value for money. Boarding schools, which can serve only around 200-400 students per year, simply aren’t scalable. Moreover, their operational costs are seven to eight times higher than those of regular public schools.
This means that a significant portion of our limited public funds is spent not only on constructing these schools but also on the ongoing expenses required to maintain them. Yet, despite this heavy investment, these schools barely address a tiny fraction of the out-of-school children crisis. Ultimately, these resources could be redirected toward more sustainable solutions that could have a broader impact across the education sector
Lastly, I was surprised to find that while we have rich data on missing facilities in schools, student and teacher attendance, and other aspects of service delivery, this data isn’t being used. It’s enough to give the government a sense of the gaps, but if it isn’t acted upon, it doesn’t help. This is why I’m usually sceptical when donors and development partners stress the importance of data. Data is indeed essential but it does not mechanically and automatically translate into improvements. Investing heavily in data systems without government willingness to use it leads to nowhere. That is why I think that development partners should work more on making data accessible to civil society and parents. If you can’t influence the supply-side to take the necessary steps, why not empower the demand side to ask the right questions? To sum it up, if you look at the entire education delivery chain, the weakest link by far is the monitoring and accountability system.
Manahil: So, we’ve gone over some of the proximate causes of stagnant education outcomes in Balochistan. Could you now tell us about the deeper determinants of stagnant education outcomes in Pakistan?
Rafiullah Kakar: In Balochistan, conflict adds another layer that we can’t ignore. I’ve often asked political parties and their MPAs why they aren’t prioritising education, especially when the people have given them this mandate. Their response is usually something like, 'Look, our constituents judge us based on two main things.' First, they want us to deliver visible, tangible development, which translates to infrastructure projects and job creation. Second, the educated middle-class voters, especially those from ideologically-driven parties, focus on issues like the right to life, the issue of missing persons, the use of force, human rights violations, and security checkpoints. Politicians respond to these demands of their political base, and no one is asking them about learning outcomes. In public discourse and agenda-setting in conflict zones, issues like the right to life often take priority over what might seem like 'mundane' service delivery issues. These are the realities shaping the political focus away from education.
Fragility and elite fragmentation are also key features of Balochistan’s politics. So, in my paper, I have also argued that the nature of elite cohesion also makes a difference. Balochistan has a severely fragmented polity. We do not have a political party in the real sense of the word. Instead there are power blocks who are loosely bound in a coalition and every coalition partner behaves like a de facto chief minister. Political parties are artificially created by powers that-be and they have very short-term interests and time horizons. So, as a result, whatever they do is very constituency-centric and extremely short-term, more so than any other province in Pakistan. These leaders use development funds to provide targeted benefits to their loyalists and consolidate their patronage networks. Ironically enough, I think that if there is any place that mirrors Balochistan in terms of political fragmentation, it's the federal government because it has the same issues of political factionalism, coalition politics, and political discontinuity as Balochistan. No wonder the federal government hasn't done a good job either.
Next I looked at the critical role of tribal and ethnic networks in accessing services. Those of us from the peripheries basically rely on our informal social and tribal networks because the state is weak where we come from. And I don't want to completely break away from my networks because they're my last resort of protection when all else fails. Reciprocity is a major feature of these networks. Just to give you a sense of what I'm talking about, as a member of the Planning Commission I can possibly interfere in the school affairs of the village of the chief of army staff or even the prime minister. However, I can't do much in my own village without first consulting the local tribal leader. This is because interfering in my own village may bring me in direct confrontation with our tribal elders. Because I depend on them for so many other things, I'll probably opt for the wiser course and leave them be.
Lastly, I've also talked about how development funds shape elite incentives and are used by them to reward their support-base. The time, attention, and resources that should have gone into education planning, monitoring and implementation is simply hogged and taken away by this interest and focus on development funds. As a result, no wonder that other than creating new buildings, the mundane aspects of education like learning outcomes, classroom environment don't get the kind of senior high-level policy attention that they should get.
Manahil: Back to what you said about how informal networks affect the chain of service delivery. So, anywhere you go, teachers are important key stakeholders in the education process. In your paper, you also say they play a very important role, but in a different way, where you mentioned the politician-teacher nexus. Now, we've already spoken about that a little bit, but can you speak about the possible implications of the politicisation of such an important stakeholder in the education system and network?
Rafiullah Kakar: That's a very good question. As you know, the teaching workforce in Pakistan is involved in performing election duties and data collection for the census. These are also the most politically sensitive activities in the country because it is on their basis that you get to decide who gets to rule Pakistan. That has prompted politicians to forge these partnerships with teachers leading to the politician-teacher nexus. No wonder, more than two-thirds of the teaching workforce were workers of political parties before the introduction of the merit-based teacher recruitment mechanisms. And that had several damaging implications because you can't hold them accountable when they don't perform their duties. So, the teacher-politician nexus is corrupting education delivery by impeding accountability and allowing politically driven transfer postings to urban centres. And as a result, there are fewer left to teach in the rural areas.
Manahil: My next question is about what you call the “soft side of education” in your paper. You’ve gone into such intricate detail and described structural causes impacting learning and access in Balochistan so comprehensively, but you could please elaborate on what the soft side of education looks like and what are ways to encourage policymakers to engage more with that?
Rafiullah Kakar: That's a good but tough question to tackle. My first recommendation is to measure learning outcomes, make that data public, and break it down by political constituency. As a Member of the Planning Commission, I have been pushing for publication of PSLM and other data by constituency, but there are very few stakeholders showing interest. The District Education Performance Index is one approach I’m using to try and engage policymakers with these core aspects of education. Measuring and publishing learning outcomes by constituency can get politicians' attention, aligning their accountability with educational improvement. Blaming them outright isn’t just unhelpful; it’s also inaccurate. Instead, we need to think about how to align their incentives with the public’s interests.
Secondly, we need a more informed public conversation on education, especially in the media. The current quality of public discourse on education is disheartening. Education isn’t even part of the mainstream discussion, which is a major tragedy. A bigger challenge, and a tough one to address, is that no matter what we do in the public education sector, we're unlikely to capture the attention of elites and the middle class who dominate the media. Over the past few months, I’ve been reflecting on why this is the case. Health and education aren't prioritised topics in the media because the media’s dominant class doesn’t rely on public schools and hospitals. For the elites and middle class, private sector alternatives are easily accessible. Since they have opted out of public education and healthcare, they have little incentive to prioritise these issues in the media.
In the long run, it’s critical to democratise the media and political parties, creating a space where the voices of those who rely on public education can be heard. We need more vertically inclusive and representative political parties and media so that those without the luxury of opting out of public schools can have a voice.
There’s no quick fix here, but these steps, I believe, could help align elite interests with the ‘softer’ aspects of education and learning.
Manahil: Finally, one last question. You’ve been working in government for a while now. How has that experience been for you, and more specifically how do you bring ideas and findings from your research and reports like the recent District Education Performance Index into governance to ensure policy is informed by research and evidence?
Rafiullah Kakar: That’s a great question, and honestly, there isn’t a one-size-fits-all answer or a quick solution. It varies depending on the context, but I’ve given this a lot of thought and come to a few conclusions. There are basically two paths that people like us can take.
Option A is to use our privileged position and resources to opt out of public education, complain about government and policymakers from the comfort of our homes, and withdraw from the system. We can look for our own easy alternatives instead. The truth is, changing the system is hard—it demands commitment, sacrifice, and persistence. So, opting out is the easier, less demanding route.
The other path, which I think is far more challenging and laborious but also more meaningful, is to engage with the system and harness whatever space is available to make small but valuable contributions while simultaneously striving for changing the system for better. This involves a range of activities from organised political advocacy to public service. If somebody thinks the system is incorrigible, then he or she should become part of an organised political effort to change the system for the better. But giving up on reforming the system is not a luxury we can afford.
For those like me who are passionate about public service and impact, there are ways to make small but meaningful contributions while simultaneously driving for long-term change. I don’t go in thinking I can overhaul the whole system; instead, I focus on where I can make a difference. One big difference I have with many others in this field is the perception that politicians and bureaucrats are inherently corrupt or ineffective and that there’s no hope for them while these people are around. I think we need to rethink that. For anyone working in policy, I’d say: let go of any preconceived notions you might have about the system being ‘broken’ or the people in it being ‘bad.’ Instead, take the time to understand the system, be humble, and come ready to collaborate with whoever is there, accepting the reality of an average-functioning bureaucracy. Don’t bring a sense of entitlement or self-righteousness to the table.
Once you build that understanding, you can start actively looking for windows of opportunity. Don’t see the civil service or politicians as a single, uniform entity. Seek out allies, identify areas where your goals align with theirs, and find those crucial openings where you can introduce change. When you identify these opportunities, come prepared—with research-based insights and well-grounded ideas. Show how your recommendations can align with their goals while serving the public interest. Assign primacy to advancing and protecting the public interest. This is something that our bureaucracy in particular needs to improve. They are rarely concerned with public interest and instead are mainly interested in pleasing their bosses and ensuring compliance with rules.
One of the challenges in our bureaucracy right now is this outdated, hierarchical relationship between politicians and civil servants, where the bureaucrat plays the role of an obedient servant, and the politician is seen as the master. When a directive is issued, the bureaucratic machinery rushes to implement it without offering technical input or feedback. Bureaucrats should serve as the technical advisors, providing policymakers with honest, evidence-based feedback. Yet, too often, the insights people like us generate within ministries never make it up to the decision-makers. Improving this aspect—where bureaucrats feel empowered to offer informed, constructive advice—could lead to incremental change in our governance system.
Manahil: Thank you so much for speaking to us, Rafiullah sb! I really enjoyed the paper, and quite appreciate the comprehensive structural and historical focus it had.
