A Deep Dive Into Education | Episode 54 | Everything is Everything

Notes

The scale of India's learning crisis

Amit opens by laying out just how bad the data looks, and stresses that this is not a lack of information problem anymore. ASER's surveys, running consistently for years, show that 57% of class five students cannot read a class two text, and a more recent survey of 14 to 18-year-olds found one in four still cannot read at that basic level, with half unable to perform simple division.

This shows up starkly in the job market. India has both a severe unemployment crisis and employers, especially in cities like Bangalore, who say they cannot find skilled workers, driving constant job-hopping among the few who are qualified. Amit points to a statistic from an episode with Kartik Muralidharan:

"When there was a recent advertisement for 368 jobs at the level of a peon, two and a half million people applied for it, including 250 PhDs. The figure came out to six and a half thousand applicants per peon job."

The conventional response, Amit notes, is to assume Nehru prioritized higher education (hence the IITs) while neglecting primary schooling, and that the fix is simply more money. But India already spends roughly 3% of GDP on education, and Amit argues the real problem is that this spending goes toward inputs, teacher salaries, infrastructure, and staff, without any correlation to outcomes. Research cited from Muralidharan shows that better infrastructure, better-ranked teacher hiring, and student inputs like midday meals do nothing to move learning outcomes, even though midday meals help in other ways. Amit also points to widespread teacher absenteeism in government schools, especially once a job is made permanent, as an incentive design failure baked into the system.

Amit connects this to a broader point about India's economic reality: most of the population lives in conditions comparable to Sub-Saharan Africa, while a small elite experiences a very different country. Failing to educate the vast majority compounds this divide, not just economically but cognitively, a gap he argues is as real and consequential as the visible gaps in nutrition and height across social classes.

Ajay adds two measurement tools essential to this conversation. First, ASER itself, which he calls one of the most influential interventions in Indian education precisely because it created reliable, consistent data where none existed before:

"In a policy pipeline, the story starts at information. You don't have the information, you can just thrash around for decades with various politically influential beliefs, but you're never making progress because you don't have the data."

Second, OECD PISA, a global assessment of 15-year-olds (sampled from households, not just those in school) that tests fluid general intelligence in science and math rather than rote memorization. Ajay describes how the Indian bureaucracy long resisted PISA participation, eventually allowing a limited pilot in Himachal Pradesh and Tamil Nadu, states expected to perform well. The results placed India among the worst performers globally, after which the bureaucracy largely withdrew from the exercise.

Yet Ajay notes a genuinely interesting wrinkle in the data: India's PISA distribution shows an unusual bump at the top end, a small population performing exceptionally well.

"You take that upper bump and multiply through the cohort size of 15-year-olds present in India, and you get an estimate of a scientific and engineering workforce of plausible capability. Lo and behold, you get an answer that India is the second largest or the third largest plausible scientific and engineering workforce in the world. That upper tail is amazing. But the main line is horrible."

Ajay also introduces PPE, per-pupil expense, calculated as government education spending divided only by children enrolled in government schools. This number has been rising sharply for years, even as outcomes on ASER and PISA stagnate or decline, underscoring that the problem is not a shortage of money but how it's being spent.

Why private schools work better and vouchers could fix incentives

Amit traces part of the dysfunction to historical hostility toward the profit motive in education, recalling Nehru's remark to JRD Tata, "do not speak to me of profit, it is a dirty word." This mindset shaped policy: for-profit schools were effectively banned, forcing private operators to route through trusts, while licensing and infrastructure mandates (playground size, toilet ratios) further constrained supply, built around an imagined utopia rather than the reality on the ground.

Despite this, Amit points to research showing that private schooling, even informal budget private schools in slums, is thriving because parents choose it. Studies by James Tooley and Pauline Dixon in Hyderabad, and by the Centre for Civil Society in Delhi among households earning under 5,000 rupees a month, found large numbers of poor families paying for private school rather than using free government schools.

"Parents know desperately, viscerally. Every parent wants their child to do better than they themselves did, to have opportunities they never had. So parents vote with their feet."

The outcome data backs this up: private schools deliver comparable or slightly better learning outcomes while spending roughly one-third as much per student as government schools. Amit argues this isn't a case for eliminating public spending, but for allowing markets to operate alongside it, drawing a parallel to how 1991 liberalization transformed once-scarce, expensive sectors like airlines and telecom.

Given that governments likely will keep spending on education, Amit proposes redirecting the money through school vouchers rather than direct institutional funding, an idea originally proposed by Milton Friedman.

"Give vouchers to parents. Parents have their kids' best interest at heart. Let the parent decide which school. Now this does is that schools now start competing with each other, even the government schools."

Ajay pushes this argument further on grounds of fairness rather than just efficiency. He highlights how remarkable it already is that Indian parents choose to spend roughly 4,000 rupees a year of their own money on private school when the government alternative is free, spending around 10,000 rupees per pupil.

"That's a remarkable piece of exit. And I think there is no more revealing statement about the failures of the Indian state."

Ajay's position is that public per-pupil spending should be symmetric regardless of whether a child attends a government or private school, since punishing a family for exiting a failing system is unfair. He offers a lighter proposal for existing government school infrastructure: repurpose the buildings as midday meal canteens, letting children attend school of their choice and simply return to the government building for lunch, preserving the jobs of current government teachers while allowing the exit trend, clearly visible in ASER data over time, to continue unimpeded.

Ajay also notes that parental choice reflects considerations researchers and bureaucrats often miss, like safety and convenience, and points to the multi-generational shift among Indian families toward English-medium education as an example of parents making sophisticated long-term decisions about their children's future, even at real short-term cost to learning outcomes during the transition.

Does education actually cause development?

Amit introduces a idea he resisted when he first encountered it: that education does not straightforwardly cause economic development, and that the causal arrow may run the other way, from development toward more education. He references a debate sparked by his friend Murali Neelakantan following separate conversations with Kartik Muralidharan and Lant Pritchett.

Ajay grounds this in Lant Pritchett's 2001 paper, "Where Has All the Education Gone?", which he calls one of the most important papers in development economics. Pritchett's statistical analysis across many countries over long time spans found essentially no growth benefit from rising education levels.

"Lant loves to say things like the educational measures of Haiti today are comparable with those in France in 1970. France in 1970 was an extremely developed economy, and Haiti today is deeply stuck in its development journey."

Ajay applies this directly to India's own trajectory, imagining explaining today's roughly 75% literacy rate to Nehru in a time machine:

"He'd say, "Oh great, wow, then you must be at 60, 70% of the UK per capita GDP." And then we'd have to hang our heads in shame and say, no, things are really bad, we are basically a lower-middle-income country."

Ajay also notes that India's most significant growth period, 1991 to 2011, was followed by continued escalation of education spending and access after 2011, without corresponding growth acceleration, reinforcing his skepticism that education intensification by itself drives development.

He then lays out three possible explanations for this paradox, presented as rival hypotheses rather than settled conclusions:

First, that the failure is specifically about implementation, meaning Indian government schools are unusually bad. Ajay finds this only partly convincing, since many other countries have built more functional public education systems and still show a weak link between education expansion and growth, making India's case an outlier but not the whole explanation.

Second, that education functions mainly as a signaling mechanism rather than a transmitter of real capability.

"It's the Louis Vuitton bag. You pay the money, you close your nose and suffer the misery, and in return you get a diploma, and the diploma plays a signaling role in the labor market."

Third, and the explanation Ajay and Amit both find most compelling, is that modern industrialized education is fundamentally miscalibrated. It rewards conscientiousness, the ability to follow instructions and mechanically execute procedures like solving quadratic equations, rather than nurturing flexible general intelligence.

"Modern formal industrialized education only measures conscientiousness. It has nothing to do with the others. The kids that take orders, the kids who will obey, are the ones who will flourish in the industrialized education system. And that's really not what flexible general intelligence is."

Amit agrees with this third framing, describing the entire education system as a nineteenth-century machine built to produce industrial or bureaucratic workers, mismatched with what the twenty-first century actually requires.

"If you are going to test a monkey on how to climb a tree, you are also testing a fish on how to climb a tree. It is incredibly unfair. All kids are different. They need to grow in different ways."

Early childhood, family, and the real inputs to a good mind

Ajay introduces Jim Heckman's research on early childhood interventions, describing Heckman as an economist committed to rigorously measuring what actually delivers return on investment, stripped of "goody-goody" assumptions. Heckman's finding is that the window from roughly age zero to five, particularly zero to three, is where interventions matter disproportionately, especially for disadvantaged children, and this cannot be fully compensated for later in life.

Statistic: During this period, a child is estimated to form about one million neural connections per second.

Beyond direct engagement, Ajay highlights the importance of what children simply overhear.

"For a child to be in a room where adults do complex conversations, that use a rich vocabulary, that engage in complex discussion, not just about will you make tea, but conversations about the world and conversations about ideas. Children are flapping their ears and they are overhearing everything."

He recounts a half-joking idea attributed to Jim Heckman: pair underfunded graduate students with disadvantaged children simply to talk to them about ideas, since this exposure itself is a potent intervention.

Amit describes this as revealing an additional, less obvious layer of privilege beyond income: growing up around rich cognitive discourse between ages zero and five confers an advantage that, according to this research, cannot be made up for later.

Ajay extends this into a critique of age-segregated schooling, arguing that psychologists have long flagged it as harmful, since a ten-year-old has little to teach another ten-year-old.

"In hunter-gatherer tribes, the beautiful model of education was where kids grew up together, where there was a diverse range of ages which hung around as a mad pack. The whole modern prissy attitude that kids should be in school with age-segregated people is held by people who have not thought deeply about education."

He also points to evidence that family remains a powerful predictor of life outcomes regardless of the broader welfare state, noting that the predictive power of family background is comparably strong in both Denmark's expansive welfare system and the United States' minimal one. For Ajay, this affirms that intergenerational transmission of knowledge inside the family cannot simply be replaced by schools.

What this means for parents, individuals, institutions, and policy

Translating this body of research into practical guidance, Amit argues that parenthood is a serious responsibility that too many people enter without enough thought.

"You have to create an enabling environment. Do not think of the role that schools play in their life as a daycare center. Too many parents think of it as convenient. And I'm like, no, you need to think harder about this."

Ajay reframes this slightly, arguing parents should explicitly treat school as daycare, freeing them to focus on the real work of education happening at home, through books, conversation, and modeling genuine engagement with ideas rather than merely owning books as décor.

For young people, Ajay warns against treating credentials like IIT or elite degrees as an endpoint, since the signaling value of a credential is real but distinct from actual capability.

"Don't drink the Kool-Aid. The greatest victims of credentialism are the credentialed ones. And don't look at other people through the lens of their degrees either."

Amit agrees but adds nuance: pursuing credentials for their signaling and network value is rational, but should be paired with deep, sustained immersion in subjects of genuine interest rather than shallow, scattershot engagement, and paired with openness to good-faith disagreement rather than treating challenged beliefs as identity attacks.

Ajay closes this section with a practical heuristic on changing one's mind, describing a spectrum from people who never update their views to those who can quickly and graciously acknowledge being wrong, with a workable minimum being a willingness to shift one's position modestly upon quiet reflection after an argument. He also recommends structured practices like book clubs, and references the "Tank Scrub" format at the Takshashila Institution, where participants present weekly what they've learned, as a way of embedding this kind of intellectual exchange into a community.

For educational leaders, Ajay urges more self-doubt about the industrialized, exam-centered model many have built their identities around, and more willingness to depart from age-sorted, conscientiousness-rewarding structures.

For policymakers, Ajay argues for skepticism about the returns on government education spending, for equalizing per-pupil spending regardless of school type, for reducing reliance on competitive exams as the primary sorting mechanism (citing the IIMs' move toward interviews and group discussions as a possible model for IITs), and for decentralizing control so that institutions like IITs, IIMs, and central universities can develop distinct identities rather than being run uniformly from Delhi.

For philanthropists, Ajay holds up funding for ASER as the most effective use of philanthropic capital in this space precisely because it built foundational data and shifted the wider conversation, in contrast to philanthropy that merely intensifies or digitizes the existing government system without questioning its underlying paradigm.

"There is a great deal of bias for action in philanthropy, and that's not useful."

Finally, Amit pushes back on any suggestion that "bias for action" is inherently bad, distinguishing between its harmful use in wielding state coercive power carelessly and its necessary role in individual creative and learning pursuits, where skill is built through iteration rather than indefinite planning. Ajay agrees, clarifying that his critique is aimed specifically at hasty, under-considered use of state power, not at individuals learning by doing.

Supplementary Resources

• Addressing the learning crisis: an emergent consensus by Lant Pritchett (Book chapter)
• Rethinking NEET, and medical education by Vijay Kelkar and Ajay Shah (Resource)
• Doing better than a competitive exam by Nitin Pai and Ajay Shah (Resource)
• Carrying the ideas of NEP to execution by Ajay Shah (Resource)
• The Rebirth of Education: Schooling Ain't Learning by Lant Pritchett (Book)
• Education in India at the crossroads by Ajay Shah (Resource)
• Accountability in education by Jeff Hammer (Resource)
• The first PISA results for India: The end of the beginning by Lant Pritchett (Resource)
• Geniuses and economic development by Ajay Shah (Resource)
• The mess in education by Ajay Shah (Resource)
• Where Has All the Education Gone? by L. Pritchett (Journal article)
• The Right to Education Act: A critique by Parth Shah (Resource)

The complete transcript file is available to download below.