“Why ‘Make in India’ isn’t making India.”
On a factory floor outside Chennai, a worker opens a box shipped from Vietnam, lifts out a semiconductor chip made in Taiwan, a display panel from South Korea, a battery cell from China, and fits them into a chassis stamped “Made in India.” The phone that leaves that factory will be counted as a triumph of Indian manufacturing. It will show up in export data, in ministry press releases, in the year’s Independence Day speeches. Almost none of what makes it valuable was made here.
This is not a fringe observation. It is, more or less, what Raghuram Rajan — a former Governor of the Reserve Bank of India, not a professional critic of the government — and his co-authors have been showing for two years: that flagship schemes like the Production Linked Incentive programme amount to “merely assembling imported parts,” not building an industrial base. Their numbers are worth sitting with, because they are more damning than a simple “we still import a lot” complaint. Look only at finished phones, and the PLI years look like a genuine success story: net exports of phones went from -$3.3 billion in FY17 to +$9.8 billion in FY23 — a real, measurable achievement. But phones don’t get made from nothing. India imported $32.4 billion worth of semiconductors, display panels, camera modules, batteries and circuit boards in FY23 alone to make that export number possible. Add those component imports back into the picture, and the combined net position — phones plus the parts that go inside them — actually worsened, from -$12.7 billion in FY17 to -$21.3 billion in FY23. We did not become less dependent on imported technology. We became more dependent, at a larger scale, with a better export chart to show for it.
I want to be careful about what I am arguing, because it is easy to be misread here. This is not an argument against the idea of manufacturing in India, against Apple’s suppliers setting up in Tamil Nadu, or against the very real jobs this activity creates. It is an argument about what we are choosing to call success, and why that choice matters more than it looks like it does.
Two different things wearing the same name
“Make in India” and “Made in India” sound like the same sentence with the words reordered. They are not. One is a location. The other is a claim about capability.
A location is where the last step happens — where components arrive, get screwed together, boxed, and shipped. It is real economic activity, it employs people, and a country is better off having it than not. But it is also, by design, replaceable. If wages rise in Tamil Nadu, the same assembly line can move to Vietnam or Bangladesh with minimal disruption, because the expensive, hard-to-replicate work — chip design, precision optics, battery chemistry, the machine tools that make the machines — never happened here in the first place.
Capability is different. It is the ability to design the product, not just fit its parts together; to make the components, not just import them; to set the standard, not just meet someone else’s. Capability is sticky. It compounds. It is why Germany still exports precision machinery a century after building its reputation for it, and why South Korea, starting from a poorer base than India in the 1960s, now owns global leadership in memory chips, shipbuilding, and batteries rather than renting factory space for someone else’s.
Our industrial policy over the last decade has been very good at attracting the first kind of activity and has mostly assumed the second would follow automatically. It has not. Studies of the PLI scheme’s own value-addition requirements have found they “often fail to significantly increase domestic value creation.” We have subsidised the box. We have not built what goes inside it.
The China comparison, without the mythology
China is the benchmark in this conversation, and it is worth being precise about why, because the popular version of the China story — cheap labour, an authoritarian government that can build anything fast — misses the actual mechanism.
China spent three decades doing something India has never systematically done: it used its market access as leverage. Foreign manufacturers who wanted to sell to Chinese consumers were, for years, required to do so through joint ventures with domestic firms and, in many sectors, to transfer technology as part of the deal. It was not always fair, and it was not always welcomed by the foreign partners forced into it. But it meant that when the joint venture ended, or when the market shifted, China had absorbed the capability — the engineering knowledge, the supply chain, the component base — rather than just having hosted the factory.
Layered on top of that: sustained, unglamorous state investment in the unsexy middle of the supply chain — the machine tools, the specialty chemicals, the industrial materials that nobody photographs for a press release but without which nothing downstream can be built domestically. By one recent estimate, China now accounts for close to a third of global manufacturing output and an even larger share of manufacturing growth over the past five years. That is not a labour-cost story. Labour in coastal China has not been cheap for a long time. It is a capability story, compounded over thirty years, in sectors — electric vehicles, batteries, solar panels, high-speed rail, drones — where China now sets the pace that everyone else, including established manufacturing economies, is reacting to.
India has taken a different, faster-looking shortcut: pay companies to relocate final assembly here, count the export numbers, declare victory. It produces better headlines in less time. It does not produce the underlying thing China spent three decades building.
What actual manufacturing sovereignty would require
If the goal is genuinely “Made in India” rather than “assembled in India,” the policy conversation needs to move to less glamorous ground: a domestic component ecosystem deep enough that a display or a battery cell doesn’t have to cross three borders to reach a factory in Noida. Serious, patient investment in semiconductor fabrication rather than only assembly and testing. R&D spending that currently sits well behind our manufacturing ambitions — most industrial R&D in India is still done by a handful of large firms, not the wide base of mid-sized manufacturers that actually absorb new technology into an economy. Vocational and engineering education aimed at precision manufacturing, not just IT services. And, perhaps hardest politically, the patience to subsidise capability-building for a decade before it shows up in an export chart, rather than subsidising activity that shows up in next quarter’s numbers.
None of this is a criticism of any particular administration; every government since the 1991 reforms has, in different ways, preferred the version of industrial policy that produces a ribbon-cutting over the version that produces an industrial commons. That is a systemic bias, not a partisan one, and it will outlast whoever is in office unless the incentive structure itself changes.
On a day like this
Independence Day is a strange occasion to write about supply chains, and yet it is exactly the right one. The independence being marked was political — the removal of a colonial administration. What has never fully arrived is a second, quieter kind of independence: the ability to design and build the technology a modern economy runs on, rather than assembling it under license from wherever the intellectual property happens to live.
A country can be sovereign on a map and still be a subcontractor in its own supply chain. That is not a failure to be embarrassed about — Korea and China both spent decades as subcontractors before they weren’t. It is a stage. The only mistake is mistaking the stage for the destination, and calling the assembly line the finish line before the harder work — the unglamorous, decade-long work of owning the parts, not just the box — has properly begun.
Part of an ongoing series under Eyes Wide Shut, observations on technology, business, systems and modern India.
