After ten weeks of writing about Indian institutions, I want to do something different from what I’ve done in the other essays.
I want to say, as clearly as I can, where I actually stand — not just what the research shows, but what I think. Because I’ve been careful throughout this series to distinguish between what the evidence supports and what I’m inferring or extrapolating, and now that the series is ending, I think the honest thing is to drop the careful hedging and just say what my read of all of it is.
This will be more personal and more uncertain than the other pieces. That’s intentional.
What Genuinely Surprised Me
I went into this research expecting to find a story about corruption — individual bad actors, regulatory capture, political dysfunction driven by greed. I found something more interesting and more troubling: a story about rational people responding to badly designed systems.
The civil servant who avoids discretionary decisions because audit scrutiny follows transfers isn’t corrupt. They’re protecting themselves in a system that punishes initiative and excuses inaction. The politician who funds roads before elections instead of school quality isn’t cynical. They’re optimizing for the incentive structure they’re actually in.
This matters because the policy implication of “bad actors” is different from the policy implication of “bad systems.” If you fix the individuals, you haven’t fixed the system, and the system produces the same behavior from the next set of individuals. India has had many reform-minded politicians and administrators. The structural outcomes have been remarkably consistent.
The second thing that surprised me was how connected the problems are. I expected twelve separate domain analyses. I got twelve views of the same underlying dynamic: institutions that were designed or evolved to serve some purposes adequately, and that generate consistent dysfunctions across every domain where they’re the primary input.
Low trust raises transaction costs. High transaction costs keep firms small. Small firms don’t generate the tax base. A thin tax base means inadequate investment in courts and schools and hospitals. Inadequate courts mean contracts can’t be enforced, which keeps trust low. The same loop shows up slightly differently when you look at education, or healthcare, or the environment. But the underlying structure is similar.
What I Think Is Actually Strong
India has things that are genuinely impressive and shouldn’t be underestimated.
The digital public infrastructure is world-class. UPI is a better payments system than most rich countries have. Aadhaar, despite its exclusion problems, enabled financial inclusion at a scale that alternatives couldn’t have matched. The Account Aggregator framework and ONDC are genuine institutional innovations. Countries are studying how India built this because it’s worth studying.
The democratic resilience is more real than cynical analysis gives it credit for. The Supreme Court struck down the Electoral Bonds Scheme. State governments get voted out for underperformance — not always, not reliably, but it happens often enough to matter. The 2024 election produced a significant reconfiguration of power, with the BJP falling short of a majority and coalition politics returning. That’s a functioning accountability mechanism, imperfect and slow but operational.
Competitive federalism has produced genuine policy variation and learning. Kerala and Tamil Nadu have health outcomes that are categorically different from UP and Bihar — not because of different budgets, but because of decades of different institutional choices and social investments. This variation is information. It shows that the trajectory isn’t fixed.
India has an enormous and growing technical talent base. The IITs and IIMs produce graduates who compete globally. The STEM pipeline, whatever its quality problems at the primary and secondary level, generates real intellectual capital at the top.
None of these make the problems less real. But they’re assets on the balance sheet, and the balance sheet matters for how you assess the next decade.
Where I’m Genuinely Uncertain
I’m uncertain about two things in particular.
The first is the political economy problem. The six reforms that the research points to as highest-leverage — judicial vacancy filling, civil service tenure, outcomes data, early childhood development, urban fiscal reform, foundational learning — are all, to varying degrees, politically unrewarding. They produce diffuse, slow, invisible benefits. They don’t generate footage for the evening news.
The research shows these reforms are the right ones. It doesn’t show that the political system can be made to prioritize them. And I don’t have a confident model for how that changes.
The democratic accountability mechanism — if voters can see what’s happening, attribute it to specific officials, and believe their vote changes something — could theoretically drive this. Competitive federalism could drive it through state comparison. Civil society and media could drive it by making institutional quality a salient political issue.
But all of these mechanisms are weaker than they need to be. The information environment makes attribution difficult. Media concentration limits independent scrutiny of government performance. Civil society organizations face significant regulatory pressure. And the short electoral cycle keeps pulling attention back to visible, attributable, immediate outputs.
I don’t know how to bridge this. I have views about what would help — mostly around making outcomes data public and accessible, and around strengthening civic information infrastructure. But I’m genuinely uncertain whether these interventions are sufficient to shift political incentives in the direction of long-term institutional investment.
The second uncertainty is about timeline. The demographic dividend is often framed as an opportunity that will expire. That framing is right in a general sense but imprecise about when. India’s working-age population peak is in the 2030s. The window for maximizing the dividend is now through roughly 2040.
But institutional change compounds slowly. If the right investments in early childhood development, foundational learning, and judicial capacity begin in 2026–2028, the full productivity effects arrive in 2045–2060 — after the demographic peak. The compounding logic works, but the timing is tighter than the framing suggests.
This means the cost of delay isn’t constant. Every year without foundational learning reform is a cohort of children who enter the labor market without functional literacy. Every year without judicial reform is a cohort of businesses that stayed informal because formal contracting was too unreliable. The losses accumulate.
Where I’m More Confident
I’m more confident that the diagnosis is roughly right. The structural bottlenecks — electoral time-inconsistency, judicial incapacity, bureaucratic dysfunction, education quality crisis, inequality, missing middle in firm size — are well-documented and consistent across multiple independent research traditions. I haven’t found credible counter-evidence that these aren’t real problems.
I’m also more confident that the solutions are tractable in principle. None of the six highest-leverage reforms require resources that India doesn’t have, technical knowledge that hasn’t been developed, or political consensus that’s unachievable in principle. They require political will and implementation capacity. These are the hard parts, but they’re not the impossible parts.
Estonia transformed from Soviet-era institutional dysfunction to European-standard governance in roughly thirty years. South Korea built an educational system from low base to high performance in two decades through specific, committed investments. These aren’t comforting analogies — India’s scale is far larger, its diversity far greater, its political complexity far deeper. But they demonstrate that the transformation isn’t theoretically impossible.
I’m also more confident that the direction of reform matters more than the pace. Getting the direction right now — investing in the invisible, slow-yield things rather than the visible, fast-yield things — sets up compounding effects that matter enormously over 20–40 years. Getting the direction wrong and then reversing it is costly not just in time lost but in institutional capacity destroyed.
What I Think Will Actually Happen
I’ll try to say this carefully, because prediction in complex systems is genuinely hard.
My read is that India’s institutional quality will improve, but unevenly and slowly. The trajectory is positive on several dimensions — digital infrastructure will continue to develop, competitive federalism will drive some state-level progress, the democratic mechanism will continue to produce occasional accountability moments.
The structural problems — judicial congestion, bureaucratic dysfunction, education quality, urban fiscal capacity — will improve at the margin but probably not at the rate the demographic window requires. The political incentive structure that discourages investment in invisible long-term goods is deep and self-reinforcing.
The inequality trajectory is the one I’m most uncertain about. There are forces pushing toward more concentration (network effects of capital, political economy of large donors, elite capture of quality institutions) and forces pushing toward less (mobile internet enabling skill development outside traditional networks, digital financial inclusion expanding credit access, urbanization mixing previously separate communities). Where this resolves is genuinely unclear.
The environmental trajectory is concerning in ways the governance research understates. India’s air quality in major cities is a public health crisis with significant economic costs — productivity loss, health spending, talent repulsion. Groundwater depletion is a long-timeline but high-stakes risk to agricultural viability. These aren’t problems that respond well to market or institutional reform on the usual timeline.
The Honest Bottom Line
India in 2036 will almost certainly be richer than India in 2026. The economy’s structural drivers — large domestic market, technical talent, improving infrastructure, digital adoption — are real. GDP growth is likely to be substantial.
Whether India in 2036 will have meaningfully better institutions than India in 2026 is less certain. Institutional quality doesn’t automatically follow economic growth. It follows deliberate investment in the right things, sustained over long enough timelines to produce compounding effects.
The specific things I’d watch as indicators:
Whether the civil service tenure minimum gets legislated. It’s a small, cheap, testable reform with large leverage. If it happens, it signals that the political system can prioritize invisible institutional investment. If it doesn’t, that’s informative too.
Whether the National Outcomes Dashboard gets built and made publicly accessible. This is the informational infrastructure that everything else depends on. Without it, accountability mechanisms can’t function at the precision they need to.
Whether foundational learning outcomes improve measurably. The ASER data is the clearest, most regularly updated signal of whether India is building or burning its demographic dividend.
These three aren’t sufficient for the full transformation. But they’re necessary, and they’re early enough in the causal chain to tell you something about whether the direction is right.
I don’t know how they’ll resolve. What I’m more sure of is that watching the signals matters, and that the question of India’s institutional trajectory is one of the most consequential questions in global development over the next decade.
Worth paying attention to. Even if — especially if — the answers are uncertain.
This is the closing essay of a 10-week series on India’s institutional systems. The research draws on 12 empirical domain studies. Thank you for reading.