Every industrial revolution has been won by whoever owned the underlying knowledge, not whoever assembled the most units. Britain owned the steam patents. Germany owned industrial chemistry. The US and, more recently, China own the semiconductor and AI stack the next one will run on. India wants a seat at that table while spending 0.64% of its GDP on research and development, a figure the Economic Survey itself admits has barely moved in two decades. Before asking whether India can compete in whatever comes next, it is worth understanding exactly how wide that gap actually is, because the headline numbers the government points to and the numbers underneath them are telling two very different stories.
Chapter 1What does the next industrial revolution actually require, and is India built for it?
The pattern across every genuine industrial transition, steam, electricity, computing, has been the same. Adoption spreads fast to whoever can afford the equipment. Ownership of the next wave concentrates permanently with whoever did the underlying research first, because patents, standards, and manufacturing know how compound for decades once established. The current wave, built on AI, robotics, and advanced semiconductors, is following the same pattern faster than any before it, and it rewards deep, sustained R&D spending specifically, not manufacturing capacity on its own.
This is where India's own current growth numbers already hint at the shape of the problem. BCG's India Economic Monitor for June 2026 shows real GDP growing 7.8% year on year in the January to March quarter, with services leading at 10% and industry, manufacturing and construction combined, at 7.3%. That is a strong number in isolation. It is also, structurally, a services led number in an era where the technologies that will define the next decade are being built in labs and fabrication plants, not call centres and IT services offices, which remain India's largest and most competitive export categories.
Chapter 2Is India's patent boom real, or is it mostly noise?
On paper, this looks like the one place India is genuinely winning. Patent filings nearly doubled between FY20 and FY25, rising from 58,503 to 110,375 applications, an annualised growth rate of 17.2%, and India climbed from 14th to 6th globally in patent applications over the past decade, a 215% increase the government cites often and accurately.
The number underneath that headline tells a much less flattering story. As of April 2025, 230,480 patents were in force in India, and only 1.59% of them had ever been commercialised, brought to market as an actual product, licence, or manufacturing process. Even generously including every unverified self reported claim of commercial use filed under Form 27, that figure only rises to 6.39%. A patent nobody builds anything from is a filing fee, not innovation.
The filing boom is also concentrated in places that should raise questions rather than confidence. Lovely Professional University alone filed 7,096 patent applications between 2020 and 2023. Galgotias University filed 1,752 in the same window. Between them, these two private universities outpaced the combined patent output of every Indian Institute of Technology, which filed 2,333 applications across the entire five year period from 2020 to 2025. Patent attorneys quoted in recent reporting describe founders routinely filing provisional patents right before a fundraising round purely as an investor signalling tool, with no resources or intention to see the invention through to grant, let alone commercial use. A rising patent count driven by university metric gaming and pre fundraise signalling is not the same evidence of innovation capacity that the same number would represent if it came from IIT labs and private sector research divisions instead.
Chapter 3Why is R&D spending stuck, and is Atmanirbhar Bharat making this worse rather than better?
This is the number that actually explains the patent problem, not just accompanies it. India's gross expenditure on R&D sits at approximately 0.64% of GDP, according to the Economic Survey 2025-26 itself, against a global average of roughly 2.67%. The comparison set is not close. The United States spends 3.59%. China spends 2.56%. South Korea spends 5.21%. India's 0.64% ranks it 55th in the world on R&D intensity, behind Rwanda, Egypt, Kenya, Tunisia, and Iran, none of which anyone currently expects to lead the next industrial revolution.
The deeper problem sits in who is actually paying for what little R&D happens. India's private sector contributes only 36% of total R&D spending, with government funding covering the rest. In the US, private industry funds roughly 79%. In China, roughly 77%. Israel's private sector funds 93%. The comparison that should sting most is Vietnam, a country the World Bank classifies at the same lower middle income level as India, where private industry funds 90.5% of R&D. The reason is specific and directly relevant to the premise of this piece: Vietnam runs a genuinely export driven economy with essentially no tariff protection for its domestic industry, meaning Vietnamese firms have to actually out innovate foreign competitors to survive, with no wall behind which to hide.
That is the mechanism worth sitting with plainly. Atmanirbhar Bharat's entire logic runs on raising tariffs and localisation requirements to shield Indian industry from foreign competition, on the stated theory that protection buys domestic manufacturers the breathing room to build real capability. The R&D funding data suggests the opposite has happened. A protected domestic firm facing limited foreign competition has correspondingly limited pressure to invest in the kind of research that only becomes commercially necessary when a cheaper or better foreign competitor is genuinely allowed to undercut you in your own market. Vietnam's 90.5% private R&D funding, achieved from a similar starting income level and with zero tariff shelter, is close to a natural experiment showing what the alternative path produces.
Chapter 4What does this cost India in numbers that actually show up on the balance of payments?
India's limited rate of patent commercialisation is not just an abstract innovation policy failure, it shows up directly as a hard currency outflow. India runs a net outflow of roughly 14.5 billion dollars a year in IP charges, royalties and licence fees paid to foreign patent holders for technology India has not developed enough of domestically to avoid licensing from abroad. That figure sits inside the same current account the rest of India's macro policy spends enormous effort trying to manage on the trade side, and it is a direct, measurable consequence of the R&D gap rather than a separate problem.
The semiconductor sector is the clearest live example of what this costs strategically, not just financially. As of December 2025, ten projects worth a combined 1.6 lakh crore rupees in approved investment have been cleared under the India Semiconductor Mission, backed by a 76,000 crore rupee incentive framework, and the government has now taken this further with a second phase, Semicon 2.0, cleared by the Finance Ministry's Expenditure Finance Committee in July 2026 at a proposed 1.25 lakh crore rupees. None of this has yet produced a commercial fabrication plant below the 28 nanometre node, and the country's own roadmap only targets the ability to design and manufacture chips for 70 to 75% of domestic applications by 2029, keeping India dependent on imports for the advanced logic chips that will sit at the centre of essentially every AI and robotics application the next industrial revolution is built around. Defence tells the same story with sharper edges right now. India domestically produces roughly 65% of its defence equipment by value, a genuinely large and improving share, but the Tejas Mk1A fighter programme is living through an acute version of this exact gap in real time. Of the 99 GE F404-IN20 engines contracted from General Electric in 2021, only a handful, somewhere between six and ten, had actually been delivered by mid-2026, leaving roughly 30 fully assembled Tejas airframes sitting grounded, complete in every respect except the one component India still cannot build itself. The domestic alternative, Kaveri 2.0, was only formally relaunched in February 2026 after being delinked from the Tejas programme back in 2008, and it remains years away from certification for a manned fighter, still targeting unmanned platforms first. A fighter jet programme that is otherwise a genuine indigenous manufacturing success story remains hostage to a single foreign supplier's production line specifically because the multi decade R&D investment needed to close the engine gap was never sustained long enough to finish the job.
Chapter 5Does India's own regulatory unpredictability compound the R&D problem?
It does, and this is the piece that connects the trade policy critique above to something even more specific to research investment. R&D spending is a bet with an unusually long payback horizon, often a decade or more between the initial investment and any commercial return, which makes it exceptionally sensitive to exactly the kind of policy unpredictability India has a documented history of producing.
The clearest historical example remains the Vodafone case. In 2012, the government retrospectively amended tax law specifically to reverse a Supreme Court ruling India had already lost, creating a 20,000 crore rupee tax demand on a transaction completed years earlier, a move that became the textbook global example cited by investors of a country changing the rules after the fact rather than before it. India formally repealed that retrospective taxation provision in 2021, a genuine and welcome correction, but the reputational half life of a single retrospective law change runs far longer than the law itself. The US State Department's own 2025 Investment Climate Statement on India, an independent assessment rather than an Indian government publication, still flags sudden policy changes, inconsistent application of regulations, judicial backlogs, and slow payment of arbitral awards as live concerns for foreign investors navigating India's regulatory environment today, alongside genuine acknowledgement of the reform progress made over the past decade.
That combination matters specifically for R&D in a way it does not for, say, a factory built to assemble an already proven product. A company deciding whether to build an assembly line can plan around a five year horizon and adjust relatively quickly if policy shifts. A company deciding whether to fund a decade long chip design programme or a fighter engine development effort needs confidence that the regulatory, tax, and trade environment will hold roughly steady for the entire life of that investment, since abandoning it halfway through destroys the capital already spent with nothing to show for it. India's government has visibly worked to address this since 2025, reducing over 47,000 individual compliance requirements, rationalising GST into a simpler two rate structure in September 2025, and proposing further income tax certainty measures in the 2026 to 27 Budget specifically framed around reducing litigation and providing predictability. The reforms are real. Whether they land the way ISM 2.0's chip roadmap or Kaveri 2.0's engine programme actually need them to, sustained, decade spanning policy stability rather than periodic correction after damage is already done, is the open question the R&D numbers in this piece are still waiting on an answer to.
Chapter 6Is this actually visible in the newest data, or is it a lagging concern?
It is visible right now, in the same June 2026 report cited earlier. Wholesale price inflation reached 9.7% in May 2026, driven specifically by higher crude, natural gas, and manufactured product prices, a pattern more consistent with an economy absorbing input cost shocks than one generating productivity gains fast enough to offset them. The Research, Development and Innovation Scheme, a 1 lakh crore rupee fund launched in November 2025 specifically to incentivise private sector R&D, is a genuine and recent policy acknowledgement that the government itself now recognises the 0.64% figure as a structural liability rather than a rounding error, which is a more honest signal than the patent filing statistics the same government tends to cite in the same speeches.
Chapter 7So, can India actually compete in the next industrial revolution?
Not on the current trajectory, and the gap is not a lack of talent or ambition, both of which the raw filing numbers and India's rise to third largest producer of scholarly publications globally genuinely demonstrate. It is a gap between research and deployment, what the Economic Survey itself calls the valley of death between early stage discovery and a market ready product, and it is precisely the stage that requires sustained private capital willing to bet on genuine competition rather than government backed protection, held over a policy horizon stable enough to survive a decade of tax, trade, and regulatory changes without being abandoned halfway through. Atmanirbhar Bharat, in its current tariff and localisation heavy form, treats self reliance as a trade policy question, insulating domestic firms from imports. The R&D and patent commercialisation data suggests self reliance is actually an innovation intensity question, compounded by a regulatory track record, retrospective tax changes, inconsistent enforcement, slow arbitral payments, that a State Department assessment still flags as a live concern even after genuine reform progress. Insulating firms from competition has so far correlated with less of exactly the R&D investment self reliance was supposed to produce, not more. A country cannot patent its way into the next industrial revolution while its own patents sit uncommercialised at a 1.59% rate, cannot build engines or chips on multi decade research timelines while investors price in policy risk on a much shorter one, and cannot protect its way into deep technological capability when the closest available comparison, an economy at India's own income level running the opposite policy, gets more than double the private R&D participation as a direct result.