WHY THIS MOMENT MATTERS

India’s education story over the past two decades has been, in many respects, a remarkable achievement. Near-universal enrolment has been secured: ASER 2024 found that over 97% of children aged 6–14 were attending school, with government school enrolment growing for the fourth consecutive year. The access problem, which consumed the system’s energy and resources for a generation, has largely been solved.

But access without learning is an incomplete victory. The same evidence that documents enrolment success also reveals the depth of the learning challenge. ASER 2024 found that only 23.4% of Class 3 students in government schools could read a Class 2-level text — an improvement from 16.3% in 2022, but a figure that still means more than three in four children at this stage are not reading at the expected level. The National Achievement Survey (NAS) 2021 paints an equally sobering picture: average Mathematics scores fell from 57% in Class 3 to just 32% in Class 10, pointing to a system that struggles not merely to build foundational skills but to sustain them as children progress.

Five years after the launch of the National Education Policy (NEP) 2020 and nearly five years since NIPUN Bharat placed foundational learning at the centre of India’s education agenda — the system is beginning to show early signs of directional change. The question now is whether these gains can be deepened, institutionalised and sustained at scale. 

ASER 2024’s most striking finding was that government schools outpaced private schools in learning recovery for the first time in the survey’s history — a reversal attributed by analysts to NEP 2020 and NIPUN Bharat’s focus on foundational learning. This is the first time in 20 years of ASER data that improvements of this magnitude have been observed. It suggests that deliberate systemic effort works.

The central question:  Can India now do deliberately and at scale what has so far occurred partially and unevenly — and can it do so before the political and institutional momentum of NEP begins to dissipate?

The five-year milestone of NEP 2020 and NIPUN Bharat makes this an especially important moment for stocktaking — not only of policy intent, but of whether implementation structures across states are becoming meaningfully aligned to learning outcomes. 

Implementation has been uneven across states. The financing window under SAMAGRA SHIKSHA is constrained. And the political economy of education reform — where gains are slow and diffuse, while resistance is organised and immediate — does not favour delay. The decisions made in the next two to three years will determine whether this generation of reforms produce lasting change or fades, as previous reform cycles have, into the background of a system that reverts to its default logic.

DIAGNOSING THE SYSTEM

What the RISE Framework Tells Us About India

The RISE (Research on Improving Systems of Education) diagnostic framework offers a precise and useful lens for this discussion. It identifies four orientations around which education systems can be aligned:

1. Access — ensuring schooling is available to all children
2. Selection — filtering students for advancement or further opportunity
3. Process compliance — ensuring system actors follow prescribed inputs and procedures
4. Learning — placing children’s learning outcomes at the centre, with all system functions and incentives aligned towards that objective

Applied to India, the diagnostic is clear: the system is currently aligned primarily around access and process compliance. Inputs are tracked and reported. Attendance, enrolment and infrastructure are monitored. Teachers are expected to complete the prescribed syllabus. But the question of whether children are actually learning — and whether every function of the system is oriented towards that end — has historically been secondary.

This is not a failure of intent. It is, in large part, a structural inheritance. A system built for a generation around the task of bringing children into school cannot instantly reorient towards the harder and more diffuse goal of ensuring they learn once there. The incentives, accountability structures, data systems and institutional cultures of access-alignment do not automatically become learning-alignment simply because a policy document declares it so.

The reform challenge:  NEP 2020 has articulated the vision. The task now is to re-engineer the system’s operational logic — its incentives, measurements, support structures and its accountability mechanisms — so that learning, not process compliance, becomes the system’s actual organising principle.

THE CRITICAL LEVERS AND THE DILEMMAS WITHIN THEM

Reorienting a system of India’s scale — across 28 states, over 1.5 million schools and nearly 10 million teachers — requires deliberate action on a set of interconnected levers. What follows is not a list of recommendations. It is a map of the genuine policy dilemmas embedded in each lever — tensions that cannot be resolved by technical design alone and that require deliberate choices at the highest level of leadership.

1.  Learning Targets: Ambition vs. Political Risk

Setting clear, measurable learning goals at each stage of schooling is foundational to a learning-aligned system. NIPUN Bharat’s target of universal foundational literacy and numeracy by Grade 3 by 2026–27 is precisely this kind of commitment. But targets without consequences are aspirations, not accountability mechanisms.

The harder question is whether the system is prepared to make those targets genuinely public, trackable and consequential. Visible targets create accountability — for states, districts and political leadership. They also create political exposure when progress is slow or uneven. The experience of ASER data over two decades illustrates the tension clearly: the same data that galvanised reform attention has also, in some states, been resisted precisely because it made underperformance visible.

The dilemma for discussion is not what targets should be set, but how much transparency and consequence the system is prepared to attach to them and how national commitments are translated into meaningful — not merely ceremonial — ownership at state and district levels.

2.  Programme and Financing: Standardisation vs. Contextual Flexibility

Centrally designed programme structures and financing norms provide consistency and equity across states. They also risk becoming strait-jackets for states with very different starting points, capacities and contextual needs. A state with 40% foundational literacy rates and a state with 75% are both asked to work within broadly similar frameworks — a design logic that privileges uniformity over effectiveness.

The NAS 2021 data illustrated the scale of inter-state variation starkly: Mathematics scores at Class 3 ranged dramatically across states, with high-performers and laggards operating in what are effectively different systems within the same national framework. The financing and programme architecture of Samagra Shiksha  has not kept pace with this reality.

The dilemma is not whether flexibility is desirable — it clearly is — but where exactly the line between national standardisation and state adaptation should be drawn and how financing structures can create genuine incentives for learning outcomes rather than compliance with input norms.

3.  Curricular and Academic Support: Coverage vs. Effectiveness

There is broad consensus that teachers need better support. The harder question is what kind of support actually moves learning outcomes. The evidence here is, by any honest reading, uncomfortable.

Large-scale, generic teacher training — the dominant model in India and across comparable systems — has a persistently weak track record. Studies cited in the broader research on improving learning outcomes, including Muralidharan (2019), confirm that “efforts made in improving infrastructure, providing teacher training and teacher support and revamping syllabi have yielded limited results.”

What the evidence does support is more intensive and targeted: sustained, subject-specific, on-site instructional support tied to real-time data on children’s learning levels. J-PAL’s work on Teaching at the Right Level (TaRL) — which has now reached over 60 million children across India and Africa — found that when teachers were guided by clear learning goals, supported by strong mentors with ongoing on-site help and brought together regularly to review data and share learning, outcomes improved substantially. But this model is expensive, difficult to standardise at scale and structurally different from the training-event model that currently dominates.

The dilemma is whether the system is willing to make the structural and financial commitments that effective academic support actually demands, rather than continuing with high-coverage, low-impact models that are easier to report but harder to defend on outcomes.

4.  Assessment: Measurement vs. System Gaming

Reliable, regular assessment of learning levels is essential for a learning-aligned system. India has both ASER — the world’s largest citizen-led household survey of learning, which has been running since 2005 — and NAS, the government’s own national achievement survey. Both are valuable; neither is sufficient alone, and their limitations are not always acknowledged in policy discussions.

Research published in peer-reviewed journals has found that NAS state averages are “unrealistically high and contain little information about relative state performance.” ASER, while more reliable for comparing state averages, is less suited to tracking changes at district level. These are not minor technical caveats: they bear directly on how assessment data should — and should not — be used for accountability and planning.

More fundamentally, assessment systems in large public systems carry a well-documented pathology: once consequential, they tend to be gamed. Schools teach to the test. Data gets adjusted. The signal degrades. The dilemma is how to design assessment that is meaningful enough to drive action, granular enough to inform instruction, and insulated enough from high-stakes consequences that it retains integrity. This requires deliberate architectural choices — about what is assessed, by whom, at what level, and with what stakes attached — that are currently unresolved.

5.  Data and Information Systems: Connectivity vs. Overload

India has invested significantly in education data infrastructure. UDISE+ now covers over 1.5 million schools and tracks hundreds of input and process variables. NAS and ASER generate large national datasets. NIPUN Bharat has created new state-level tracking mechanisms. The problem is not data volume.

The gap is decision-relevant connectivity. Outcome data, process data, and input data sit in separate silos and rarely inform each other in real time at the levels where consequential decisions are made. A district officer or state planner typically cannot look at a single dashboard and understand the relationship between what is being spent, what pedagogical support is being provided, and what children are actually learning in a given block or cluster.

The dilemma is not whether to build more data systems — the risk of doing so is well-established — but how to restructure existing ones around the decision-making needs of the actors closest to schools. This is as much a question of institutional design and political will as it is of technical architecture.

6.  Local Accountability and Governance: Community Voice vs. System Capture

School-level governance structures — School Management Committees, gram sabhas, elected local bodies — exist across the country. In principle, they represent the most direct accountability mechanism available: communities holding schools to account for whether children are learning.

In practice, the evidence on whether these structures meaningfully shift learning outcomes is mixed at best. The challenge is not the absence of community structures but the absence of the conditions that would allow them to function as genuine accountability mechanisms. Three things are consistently absent: the information needed to assess school performance (most community members have no access to reliable learning outcome data for their school), the authority to act on that information (SMC powers are often advisory rather than binding), and protection from retaliation for community members who challenge school or system actors.

The dilemma is not whether local accountability is desirable but what it actually requires to function — and whether the system is prepared to make the structural commitments, in terms of information sharing, devolved authority, and grievance mechanisms, that genuine community accountability would demand.

THE GOVERNANCE QUESTION

What the Centre-State Dynamic Must Look Like

Education in India sits on the Concurrent List, and no discussion of learning system reform is complete without confronting the federal question directly. The six levers described above are each, in different ways, subject to the tension between national ambition and state capacity, between central accountability and state autonomy, and between the government’s need for comparability and states’ need for contextual adaptation.

The trajectory of reform under NEP has, so far, been primarily one of national articulation followed by variable state interpretation. Some states — in foundational learning, in assessment design, in teacher support models — are moving with intent and at pace. Others are moving slowly, for reasons that range from capacity constraints to political economy. The centre’s leverage over state implementation is real but limited: it flows primarily through financing conditionalities and technical assistance, neither of which has, historically, been sufficient to drive deep systemic change.

The governance question that must be resolved is this: what is the centre’s theory of change for accelerating reform in lagging states, and what combination of incentives, support, peer learning, and accountability will make that theory operational? This is not a question that can be answered through programme design alone. It requires a political commitment, at the highest level, to a model of cooperative federalism in education that goes beyond rhetoric.

For discussion:  What mechanisms — financial, technical, political — can the Government of India deploy to accelerate learning system reform in the 10–12 states where progress is slowest, without undermining the autonomy and ownership that sustained reform ultimately requires?

CONCLUSION: THE DECISIONS THAT CANNOT BE DELEGATED

The shift from an access-aligned to a learning-aligned education system is not primarily a technical problem. The broad contours of what needs to happen — clearer targets, more effective teacher support, better assessment, connected data, genuine local accountability — are reasonably well understood. What has been missing is not knowledge but resolve: the willingness to make choices that are genuinely difficult, that create political exposure, and that require sustained commitment across electoral cycles.

The six dilemmas outlined in this paper are presented not as problems requiring more analysis but as decisions requiring ownership. Each one has been deferred, in different ways, across successive reform cycles. The evidence base for what works is now sufficiently strong. Five years since the launch of NEP 2020 and NIPUN Bharat, India stands at an important inflection point in its education reform journey. The political window created by NEP 2020 and NIPUN Bharat is real but not indefinite. The next phase must move beyond policy articulation towards deeper institutional alignment, stronger implementation discipline and sustained accountability for learning outcomes. 

As NEP 2020 and NIPUN Bharat complete five years, the question is no longer whether India has the right vision for education reform. The question is whether the system is prepared to make the difficult governance choices required to translate that vision into quality learning for every child, paving the way for a Viksit Bharat.