The IHDS Forum is a monthly update of socio-economic developments in India by the IHDS research community, based on the India Human Development Survey, jointly conducted by NCAER and the University of Maryland. While two earlier rounds of the survey were completed in 2004-05 and 2011-12, respectively. Fieldwork for the third round was undertaken in 2022-24 and the data is currently being cleaned and processed.
Click here for previous issues.
While enhancing online safety, opportunities to learn, connect and develop digital capabilities must be preserved
The debate over children’s access to social media has entered a new phase worldwide. Governments are moving towards age restrictions, while technology companies are being pushed to strengthen age assurance, parental controls and safer platform design. Meta, in a US lawsuit, has agreed to introduce features for teens to address concerns around addiction and excessive screen time. In India, the Supreme Court recently sought the Centre’s response to a petition seeking safeguards against minors independently maintaining social-media accounts.
The concern is legitimate: children need protection online. But as India considers its response, it should also consider whether a blanket social media ban could deepen existing digital inequalities and widen gaps in children’s opportunities to participate meaningfully online.
Use of social media
For much of the past decade, India’s digital agenda has focused on narrowing the access divide through affordable mobile internet and smartphones. A recent report (Ganguly et al., 2026), based on the India Human Development Survey (IHDS-3, 2022-24), shows that 95.1 per cent of households own a mobile phone. Yet access to, and participation in, the digital world remain unequal among children and teenagers.
For many young people, social media and other online platforms are spaces for communication, information, creativity and social connection.
At the same time, India lacks evidence on how children use social media and which online experiences are harmful. ASER 2024 reports that 57 per cent of rural adolescents aged 14-16 who can use a smartphone used it for an education-related activity in the preceding week, but it does not identify the platforms used. The same platforms can serve educational and recreational purposes, making the boundary between “useful” and “harmful” online experiences less clear-cut.
The inequalities become clearer in social media use. The IHDS-3 report shows that among children aged 13-16 years, 65.3 per cent have access to a mobile device, 37.8 per cent use the internet, and 21.3 per cent use social media. Social media use is higher among boys than girls (26.4 per cent versus 16 per cent) and declines sharply from metropolitan areas (30.7 per cent) to rural India (18.5 per cent). It increases steadily with parental education.
Children do not begin their digital lives from the same starting line. A teenager from a metropolitan, higher-educated household inhabits a different digital environment from a first-generation internet user in a rural household. A blanket restriction could therefore reduce exposure to online harms for some children while simultaneously removing opportunities for digital participation for others. Uniform regulation may not produce uniform outcomes in an unequal society.
None of this diminishes the harms associated with social media. Protecting children from cyberbullying, harmful content, addictive platform design, online exploitation and excessive use remains an important policy objective. But protection and participation should not be treated as competing goals.
The emerging policy landscape offers more tools than a simple choice between unrestricted access and a blanket ban. Age assurance, parental controls, safer defaults, limits on harmful platform design and greater accountability for technology companies are increasingly part of the policy toolkit. For India, the challenge is to determine which combination of safeguards can reduce harm while preserving opportunities for children to learn, connect and develop digital capabilities. India should also assess the distributional consequences of any restriction — who is likely to lose access, who can compensate through other devices or support, and which children are already least able to participate online. These questions are central to equitable digital policy.
India’s first digital divide was about who could get online. The next is about who can participate meaningfully and safely once connected. The goal should be to protect children online while keeping the door open to opportunities for learning, skill development and connection that the digital world can offer.
Ganguly is Associate Fellow, NCAER–National Data Innovation Centre, and Agarwal is Policy Analyst at The Quantum Hub. Views are personal.
India’s digital agricultural transformation should focus on building institutional and economic links between data and action
Indian agriculture has always suffered from an information problem. Governments often did not know precisely what was being grown where, farmers had limited information about weather, soil and markets, and financial institutions struggled to assess agricultural risk. Agricultural policy, therefore, often had to work with incomplete, delayed or fragmented information.
That is beginning to change. Satellite imagery, remote sensing, geographic information systems, drones, digital crop surveys and other technologies are making it possible to observe agriculture at a much finer level than before. We can increasingly identify where crops are being grown, assess their condition, map soils and water resources, monitor weather-related risks and link this information with farmers, land records and markets. This is an important technological change. But its real significance lies elsewhere.
The value of agricultural information begins when it changes a decision.
A satellite image, for example, does not by itself increase agricultural productivity. It becomes economically valuable when the information derived from it leads to a better decision about irrigation, fertiliser, crop insurance, credit, procurement or logistics.
Consider water. If remote sensing can identify areas experiencing crop stress, that information can help target irrigation more precisely. The benefit then comes not from the satellite itself, but from using less water, reducing crop stress and potentially improving yields. The same principle applies to fertiliser. Better information about soil and crop conditions can allow inputs to be applied more precisely, reducing waste and costs while improving resource efficiency. But this is where the challenge begins.
Information does not automatically become action.
A useful way to look at the emerging digital agricultural ecosystem is as a chain:
Data → Information → Intelligence → Decision → Implementation → Economic outcome.
There can be a gap at every stage. The first is a data gap. Do we know what is happening and where? Which crop is being cultivated on which plot? What is its condition? What are the soil and water conditions? Who is actually cultivating the land? New digital and geospatial systems are helping address many of these questions.
The second is an information-to-intelligence gap. Having data is not the same as understanding it. A crop map may tell us where wheat is being cultivated. But can that information tells us how much wheat is likely to be produced, what the crop may require in terms of water and inputs, or where production risks are emerging? This is where analytics, artificial intelligence and the integration of multiple datasets become important.
The third is the intelligence-to-decision gap. Even when we know what is happening, does somebody act on that information? Suppose a system identifies water stress in a particular area. The economic value arises only if the information reaches the appropriate decision-maker in time and results in a change in irrigation behaviour or water management.
The fourth is the decision-to-implementation gap. A digital system may recommend a particular input, irrigation schedule or crop practice. But does the farmer have access to irrigation? Is the required input available? Can credit be obtained? Is the extension system capable of supporting the farmer? Digital information cannot substitute for missing physical and institutional infrastructure. This is perhaps one of the most important points in the debate over digital agriculture. Technology can improve information, but it cannot by itself remove every constraint on agricultural production.
The final gap is between implementation and economic outcomes. A digital crop survey may improve the government’s ability to identify beneficiaries. Better crop-loss information may make insurance assessment faster. Digital land and crop records may make agricultural credit easier to process. These are valuable improvements. But they are largely improvements in administrative or transactional efficiency.
The larger economic question is whether they ultimately affect the farm economy. Are input costs falling? Are yields increasing? Are crop losses declining? Is water being used more efficiently? Is access to credit improving? Are farmers receiving better prices? Is income becoming more stable? This distinction between administrative efficiency and economic transformation deserves much greater attention.
The possibilities also extend beyond the farm.
Better information about the location, condition and likely harvest of crops can improve procurement planning, storage, transportation and processing. A retailer or food processor could potentially use agricultural intelligence to anticipate where supplies will come from, when they will reach the market and how much infrastructure will be required.
Thus, some of the economic value of agricultural intelligence may emerge well beyond the farm gate. And this raises another important question:
Who captures that value?
Better information can reduce information asymmetry. But it can also create new forms of asymmetry. A farmer may gain from knowing more about soil conditions or weather. At the same time, a large buyer may gain from knowing more about expected production and local market conditions. A bank may assess agricultural risk more accurately. An insurer may price risk more precisely. A retailer may optimise procurement and logistics.
All of these can create economic value. But the distribution of that value will depend on access to information, market structure and bargaining power. This is why the debate about digital agriculture should not stop at the question of how much data India can generate. The more important question is how far that information travels through the agricultural system.
• Does it reach the farmer?
• Does it reach the extension worker?
• Does it improve a government decision?
• Does it change the behaviour of a bank or insurer?
• Does it improve procurement and logistics? And, ultimately,
• Does it improve the economic position of the farmer?
India is, therefore, moving from an era of limited agricultural visibility towards one in which it can increasingly observe agriculture at the level of the field and, potentially, the individual plot. That is a remarkable capability. But better visibility is not the same as better outcomes.
The next phase of India’s digital agricultural transformation should, therefore, focus on building the institutional and economic links between data and action. We need to know not only whether information exists, but whether it is usable, accessible and acted upon. The ultimate test should not be the number of hectares mapped, databases created or applications developed. It should be what these systems change on the ground.
The journey we need to complete is straightforward: From data to information, from information to intelligence, from intelligence to better decisions, and from better decisions to better economic outcomes. The satellite image is not the outcome. The better decision is the outcome.
And the better decision itself is not the end. For Indian agriculture, the real measure of success will be whether this new information architecture ultimately produces higher productivity, greater resilience, more efficient use of resources and, most importantly, better and more stable incomes for farmers.
The writer is senior fellow at the National Council of Applied Economic Research, New Delhi. Views are personal.
The strategic question is not simply whether India possesses rare-earthresources or whether imports from China can be reduced. The real policychallenge is understanding what happens in between
Most Indians know the parable of several blind men trying to describe anelephant. One touches its trunk and believes it is a snake. Another feels itsleg and insists it is a pillar. None is entirely wrong, yet none sees the whole animal.
India’s strategy for advancing its cutting-edge technologies, aimed at becoming a global manufacturing leader, carries the risk of making a similar mistake. From EV motors and semiconductor fabrication facilities to precision manufacturing machineries, at the heart of each lies an unassuming component: the high-performance permanent magnet. They serve as the silent operating system of industrial acceleration.
Magnets are not all the same. Ferrite, Alnico and Samarium-Cobalt magnets continue to serve important industrial applications. Yet Neodymium-Iron-Boron (NdFeB) magnet shave become the backbone of energy transition and advanced manufacturing because no other commercially available permanent magnet combines comparable magnetic strength with such a high power-to-weight ratio. Recognising its importance, India has begun strengthening its critical minerals and rare earth element strategy through initiatives such as the National Critical Mineral Mission, overseas mineral acquisitions, expanded geological exploration, and PLI schemes.
However, China’s tight export controls on rare-earth magnets and materials in April 2025not only revealed the deep-seated vulnerabilities in global industrial value chains, but also the limits of our understanding. For India, the primary vulnerability does not stem from a shortage of critical minerals; rather, it lies in the absence of a comprehensive framework to pinpoint where strategic technological dependence is cultivated, accumulated and propagated throughout the permanent magnet value chain.
The journey of a magnet
Consider a simple example. India’s Annual Survey of Industries estimates the domestic permanent magnet market at around ₹750 crore. Yet international trade statistics indicate import values several times larger than that entire reported market. The gap may reflect differences in statistical coverage, industrial classification or supply-chain accounting. Whatever the explanation, policymakers face a troubling reality: they cannot confidently explain where all these magnets enter the economy or how they move through it. In other words, parts of India’s permanent magnet economy appear to exist without being fully visible in statistics.
Unlike other products, permanent magnets pass through a long technological journey. Geological exploration leads to mining. Mining feeds mineral processing. Processing enables chemical separation. Separation produces oxides, which are refined into metals, transformed into alloys, engineered into magnetic materials, and finally manufactured into finished magnets. Each stage demands different scientific knowledge, industrial capability and technological maturity. This is why discussions on critical minerals frequently miss the real point. The strategic question is not simply whether India possesses rare-earth resources or whether imports from China can be reduced. The real policy challenge is understanding what happens in between.
Although India has developed capabilities across several stages of permanent magnet manufacturing, the country still lacks a systematic way of identifying where those capabilities are globally competitive, where critical gaps persist, and how dependence accumulates across production stages. India’s statistical system provides only a partial picture of what is mined, imported, and manufactured.
One way to bridge this gap is through an Integrated Techno-Economic Mapping (ITEM)framework, which presently is a missing piece in India’s industrial policy toolkit. Such a framework brings together information on engineering and economic measurement to show how a permanent magnet is built, from minerals in the ground to the finished products that power advanced manufacturing. More importantly, it would identify where industrial capability should be built, where technological partnerships become essential, and where domestic investment would yield the greatest strategic return. A country may secure mineral resources yet remain dependent if it lacks processing and manufacturing capabilities.
Bridging this measurement gap is not an academic exercise; it is an industrial imperative.
Suvajit Banerjee is a Fellow at the National Council of Applied Economic Research (NCAER), New Delhi, in the research vertical of Computable General Equilibrium Modelling and Policy Analysis. Sovini Mondal is a Research Analyst at the NCAER in the research vertical of Computable General Equilibrium Modelling and Policy Analysis. Sanjib Pohit is a professor and Thematic Lead at the NCAER in the research vertical of Computable General Equilibrium Modelling and policy analysis. Views are personal.
Gig workers in India face a financial literacy gap impacting their ability to manage income, insurance, and taxes, calling for targeted policy, education, and platform-based solutions to enhance their financial well-being.
A delivery rider is paid through an app, buys his fuel and sends money home via UPI, and can tell his customer rating to the last decimal. Yet he might only quote what the platform paid him rather than what he makes, and merely guess about any insurance coverage. In other words, he is thoroughly banked, yet unable to comprehend his own finances fully and maximise financial well-being.
This gap in India’s gig economy needs policy attention. A National Centre for Financial Education survey found only 27% of Indian adults are financially literate. For a salaried worker, provident fund, insurance, and a payslip absorb much of the financial-management burden. But gig workers need to manage all three themselves.
A study by Dvara Research in Bengaluru found platform workers grossing Rs 35,075 a month and spending Rs 18,470 on fuel, maintenance, and other running costs. Most workers perceive the gross income as net. The Economic Survey 2025-26 records about 40% of gig workers earning below Rs 15,000 a month. Of 11 the platforms assessed in the Fairwork India ratings for 2024, only two could evidence workers earning the local minimum wage once costs were deducted.
The behavioural evidence supports the financial literacy deficit in gig workers. A study by LEAD at Krea University found 62% lacked emergency fund, and close to half had no insurance. A 2024 survey across 40 cities by Borzo found 65% lacking life insurance, 77% holding no mutual funds, and 61% being unable to state the income tax slabs. Two-thirds of gig workers had never filed a return. Only 20% earned Rs 2.5 lakh to Rs 5 lakh annually. For ride hailing services, while GST is managed by the platform, the workers are supposed to monitor their tax liabilities themselves. Lack of financial knowledge can result in self-inflicted exclusion from credit.
It has also become a constraint for existing schemes. The Atal Pension Yojana crossed 90 million gross enrolments in April; it pays Rs 1,000-5,000 a month for those aged 60 years or above. The PM Shram Yogi Maandhan similarly offers Rs 3,000. Registration on e-Shram is now the gateway to labour code protections. Yet by June, against an estimated 12 million gig workers and a Budget commitment to cover roughly 10 million, the labour ministry put the number registered at between 1 and 1.2 million.
Kenya offers a warning signal. By 2024, 84.8% of Kenyan adults had access to formal financial services compared to 26.7% in 2006. However, the FinAccess surveys, held every two to three years, revealed that only 18.3% were financially healthy. Inclusion produces borrowers. India’s instant loan apps are advertising to precisely this demographic.
Platforms record every trip, incentive, and deduction. A monthly statement in the worker’s language, setting out gross earnings, work-related costs, and net take-home, would hand millions of people data they currently lack. A worker who can see his net income can plan around it.
The government or NGOs can make a short financial capability module a condition for aggregator registration in the workers’ own language. Financial education cannot be an induction exercise; it has to travel alongside working life cycles. Platforms can use short, vernacular, mobile-first modules, interactive simulations, and gamified learning to explain budgeting, insurance, pensions, mutual funds, credit, taxation, and digital fraud. Workers should receive timely nudges when they get an incentive, take a loan, make a large withdrawal, or become eligible for a government scheme, directing them to the regulated financial product or official source. Periodic financial-literacy camps at delivery hubs, transport centres, and gig-worker congregation points, supported by banks, regulators, insurers, and pension providers, can complement in-app learning. Four topics carry the most immediate value: budgeting on an income that changes weekly, the difference between gross and net, how to enrol in a pension and an insurance scheme, and why an instant loan app quoting a daily rate is the most expensive money they will borrow.
Gig workers should know the simple difference between a bank deposit, insurance, pension, mutual fund, and credit product; who regulates each of them; how to check if a provider is legitimate; and which government schemes they are eligible for. A “financial toolkit” embedded in every gig platform could provide links to regulated products and official portals, explain eligibility in plain language, and allow workers to compare costs, risks, and benefits before they commit their money.
Once workers understand the financial products and schemes available, platforms should use timely nudges. For example, reminding a newly eligible worker to enrol in a pension or insurance plan, or showing the true cost of an instant loan before it is taken. Auto-enrolment in the Atal Pension Yojana at onboarding, with a visible opt-out and the contribution met from the levy aggregators already owe, will enrol more gig workers in a quarter than a decade of awareness campaigns.
Financial awareness is worth little if the system cannot see the worker back. India already has the Account Aggregator framework. Platform earnings history should flow into it with the worker’s consent. Besides, literacy surveys can consider carrying a gig worker module as there is no separate national measurement of financial capability.
Riders know their rating as it matters to them as well as the platform, but they aren’t aware about what they net in a month. A design choice made by platforms can remove this asymmetry of knowledge.
The writers are Respectively IEPF Chair Professor and Senior Research Analyst, National Council of Applied Economic Research, New Delhi. Views are personal.