The Arranged Marriage of the Century

IIn the winter of 1982, a small, balding, relentlessly frugal Japanese executive named Osamu Suzuki did something his rivals at Toyota and Nissan considered a species of madness. He agreed to bet his company — Suzuki, then Japan's perpetual fourth-place also-ran, the maker of motorcycles and boxy little kei cars — on a country that most Japanese boardrooms regarded as a bureaucratic swamp of licences, permits, and interminable delay. He signed a joint venture with a state-owned Indian firm named Maruti Udyog, a company that existed largely because of Sanjay Gandhi's ill-starred dream of a people's car. The first Maruti 800 rolled off the line in December 1983. It was, by the standards of the age, a modest little hatchback. It became, by any honest accounting, the most consequential automobile in the history of the subcontinent.

I begin with this scene not from nostalgia — though I confess a certain fondness for the 800, a car in which a great many Indians of my generation first learned that the middle class was a place one could actually drive to — but because it is the founding parable of an argument I want to make. The argument, put plainly, is this: Japan and India are not competitors, nor patron and client, nor donor and supplicant. They are, in the precise and slightly indelicate sense, complementary. Each possesses, in abundance, exactly what the other conspicuously lacks. And the great missed opportunity of our decade is that both nations continue to treat this obvious jigsaw fit as a matter for diplomatic communiqués and photo-ops rather than what it actually is: a marriage proposal that both families keep almost, but not quite, accepting.

In a previous essay I argued, perhaps too caustically, that Indian innovation remains stuck in a copycat's crouch — a middleman's republic of clones and cost-arbitrage, throttled by underfunded research, a brain drain of the ambitious, and a low-trust bureaucracy that treats every entrepreneur as a suspect. I stand by all of it. But an indictment is not a strategy, and the reader who finishes a diagnosis is entitled to ask about the cure. This essay is one answer. The cure for a nation that only knows how to copy software might just be a nation that has forgotten how to write it.

Two Distinguished Families, Two Embarrassing Secrets

Let us be candid about the families in question, because arranged marriages founder precisely when the relatives conceal the inconvenient truths.

Japan is the wealthy, punctilious, impossibly well-mannered household at the top of the lane. Its dowry is the envy of Asia: a manufacturing culture of near-religious precision, the world's finest robotics, materials science bordering on sorcery, the patient capital of a nation that thinks in decades rather than quarters, and a process discipline — the kaizen, themonozukuri — that turned “Made in Japan” from a punchline in the 1950s into a benediction by the 1980s. On the strength of firms like FANUC, Yaskawa, and Kawasaki, Japan long supplied around 45% of the world's industrial robots (a share the International Federation of Robotics put at roughly 38% of global production by 2024, as China's own output surged) — names spoken with reverence on every factory floor from Stuttgart to Shenzhen.

And yet this household harbours a secret it would rather you did not discuss at the dinner table. Japan cannot, for the life of it, write software at the scale the twenty-first century demands. This is not a slur; it is arithmetic. The country's own Ministry of Economy, Trade and Industry (METI), in its landmark 2019 survey of IT human-resource demand, projected the engineering shortfall rising from 170,000 in 2015 to roughly 370,000 in 2020 and as many as 790,000 by 2030 in its high-demand scenario. The structure of the industry is a museum piece: it is commonly estimated that some 70% of Japanese IT companies are SIer-type systems integrators, with most engineering talent marooned on the vendor side of a multi-layered subcontracting pyramid that would be instantly recognisable to a medieval guildsman. In the United States, most IT talent sits inside operating companies, quoting for nothing; in Japan, the engineer is summoned like a plumber for every specification change. The consequence is a Waterfall culture in a world that has moved to continuous deployment, a landscape of legacy COBOL systems tended by greying engineers, and the notorious “2025 Digital Cliff” that METI warned could cost the economy up to ¥12 trillion a year — a warning, I note drily, that has aged into the present tense.

The most vivid symptom is the one the accountants track. Per Japan's Ministry of Finance preliminary data released in February 2025, the nation's digital trade deficit — the money it pays out, net, to foreign providers of cloud, advertising, and software services, overwhelmingly American — hit a record ¥6.46 trillion ($43 billion) in 2024, having more than tripled in a decade from ¥2.02 trillion in 2014. METI's own projection is that this “digital deficit” could balloon to ¥18 trillion by 2035. A nation that runs a trade surplus in automobiles is haemorrhaging cash to Seattle and Mountain View for the privilege of running its own businesses. And atop all this sits the demographic iceberg: a median age of roughly 50, a record 29.4% of the population aged 65 or older as of September 2025, and a working-age cohort that the National Institute of Population and Social Security Research projects will shrink from around 74 million towards 62 million by 2040.

Now consider the household down the hill — noisier, younger, its front garden a chaos of unfinished construction and improvised genius. India's dowry is the mirror image of Japan's shortfall. It has an ocean of software talent — some 5.8 million developers, a pool on track to overtake that of the United States. It has youth, a median age of about 28, where Japan has age. It has scale, hunger, and English. Its Global Capability Centre boom has produced more than 1,700 centres employing roughly 1.9 million professionals — not back-office call centres but product-engineering and AI hubs for the world's largest firms. Its technology sector booked around $224 billion in IT-services exports in FY2025 (NASSCOM Strategic Review 2025), part of a $283-billion industry.

And India, of course, has its own secret, which I spent an entire previous essay excavating: it cannot make things. Its precision manufacturing is thin, its deep-tech capital thinner, and its hardware ecosystem exists in a state of near-total dependence on China — electronics imports from China of around $38 billion in the first ten months of 2025 alone (GTRI), and a bilateral trade deficit with Beijing that GTRI founder Ajay Srivastava pegged at a record $99.2 billion in FY2025, more than doubled from $44 billion in FY2021. Its gross expenditure on R&D languishes below 1% of GDP — the Economic Survey cites roughly 0.64%, while the Department of Science and Technology's latest tabled figure is 0.84% for FY2023-24. It designs around 20% of the world's chips and manufactures almost none of them.

I invite you to read those two paragraphs again and tell me they do not describe two people who ought to be introduced. One family has the factory and no coders; the other has the coders and no factory. It is almost insulting how neatly it fits.

The Marriages That Worked

The sceptic will say: charming theory, but does it actually work in the real world of tariffs and temperaments? To which the answer is that it has already worked, spectacularly, twice — and the evidence is parked in perhaps a third of the driveways of urban India.

Return to Osamu Suzuki's gamble. Four decades on, Maruti Suzuki remains the largest carmaker in India, its market share hovering around 40% — 42.5% in April 2026, though slipping to 37.1% in March as buyers who once wanted cheap now want big — a level of dominance that in most industries would attract the attention of a competition regulator. But the truly delicious detail, the one that ought to be inscribed above the door of every trade ministry in Tokyo, is this: India is now Suzuki's largest and most important market on earth. In the first half of FY23 the Indian arm contributed 55.6% of Suzuki's global unit sales and, at a seven-year high, some 39% of its revenue. The Indian subsidiary has at times been worth more than the Japanese parent that owns it. Osamu Suzuki did not condescend to India; he was, in the end, rescued by it. The pupil became the paymaster.

The second marriage was, if anything, more romantic. In 1984 the Munjal family of Ludhiana — bicycle makers, of all things — joined hands with Honda to form Hero Honda. The Japanese brought the engine technology and the process discipline; the Indians brought the distribution network, the feel for a market where a motorcycle is not a toy but a household's economic backbone. By 2001 Hero Honda was the largest two-wheeler manufacturer in the world. Its Splendor became, for a time, the best-selling motorcycle on the planet. “Fill it, shut it, forget it” entered the vernacular. Every second motorcycle sold in India wore the joint badge.

That the marriage ended in an amicable divorce in December 2010 does not diminish it — indeed, it strengthens my argument. When the couple separated after twenty-six years, both partners thrived. Hero MotoCorp, the Indian half, remains to this day the world's largest two-wheeler manufacturer. Honda's wholly-owned Indian arm, HMSI, grew from a 13% domestic share in 2010-11 to some 27% by 2014-15, becoming Hero's most formidable rival. The joint venture had been so successful that it produced two champions where there had been one. If only all divorces enriched both parties so.

These were not the only unions. Toyota came with Kirloskar; Honda came, separately, with the Siels. But Maruti and Hero Honda are the twin lodestars, and their lesson is unambiguous: graft Japanese process discipline onto Indian market intelligence and labour, and you do not get a compromise — you get category dominance. The whole exceeds the sum of the parts by an embarrassing margin.

The Marriages That Ended in the Lawyers' Office

Now, no honest matchmaker recites only the happy unions. If I am to persuade you that a software-age sequel is worth attempting, I owe you the annulments, the acrimony, and the arbitration — because the frictions that wrecked those deals are precisely the frictions a new partnership must be engineered to survive.

Exhibit A, and it is a lulu, is Daiichi Sankyo's acquisition of Ranbaxy. In 2008 the Japanese pharmaceutical giant paid some $4.6 billion for a controlling stake in India's largest drugmaker — a resounding vote of confidence in Indian pharma. It curdled into one of the great cautionary tales of cross-border dealmaking. Within years Ranbaxy pleaded guilty to seven US federal criminal counts and paid $500 million in fines for selling adulterated drugs and lying to regulators; the FDA banned products from its plants. Daiichi, alleging it had been defrauded — that the Singh brothers had concealed the extent of the regulatory rot — wrote down its investment and eventually offloaded Ranbaxy to Sun Pharma for $3.2 billion. An arbitration tribunal awarded Daiichi around $500 million against Malvinder and Shivinder Singh, who were subsequently arrested on separate fraud charges. The Japanese had bought, in good faith and at a premium, a pig in a very expensive poke.

Exhibit B is even more instructive, because the villain of the piece is not a fraudster but a regulator. In 2009 NTT DoCoMo, Japan's telecom titan, invested some $2.6 billion for a stake in Tata Teleservices, with a contractual right to exit at half its investment or fair value, whichever was higher. When it tried to leave in 2014, the Reserve Bank of India refused to permit Tata to pay the agreed price, citing foreign-exchange rules that forbade guaranteed returns to foreign investors. A London arbitration tribunal ordered Tata to pay $1.17 billion; the RBI tried to block the payment; the matter dragged through the Delhi High Court until 2017 before DoCoMo finally received its money. Read that sequence again and you will understand, in a single anecdote, everything my previous essay said about the low-trust bureaucracy: a Japanese firm did everything correctly, won its case in every forum, and still had to wage a three-year legal war against India's own central bank to be paid what it was owed. If you were a Tokyo boardroom weighing an India bet, this case is the ghost at your feast.

Exhibit C is subtler and, to my mind, the most damning. SoftBank — Masayoshi Son's colossus, the single largest Japanese technology bet on India — poured billions into the Indian startup scene: Ola, Oyo, Paytm, Snapdeal, Flipkart, Grofers. And what, precisely, did Japan's boldest capital finance? It financed, almost to a fault, the clones. A ride-hailing app, a hotel-branding aggregator, a payments wallet, an e-commerce also-ran — the copy-paste economy I spent a whole essay lamenting. Here was the richest vein of Japanese risk capital ever directed at India, and it went not to invention but to imitation, not to deep tech but to discounts-funded land-grabs, several of which cratered. Son's Vision Fund posted a ¥4.3 trillion — some $32 billion — loss in a single fiscal year. The tragedy is not that SoftBank lost money; funds lose money. The tragedy is the misallocation: Japanese patient capital, the one thing India most desperately needs to fund its hardware and deep-tech ambitions, was instead lit on fire subsidising taxi rides. The marriage was consummated on entirely the wrong terms.

And then there is the bullet train, which deserves a paragraph of its own as the perfect allegory of the whole relationship. The Mumbai-Ahmedabad High Speed Rail project — Shinkansen technology, sanctioned in 2015, groundbreaking in 2017 — was meant to be operational by 2022. It was not. Its original completion target slipped and slipped again, defeated not by any engineering deficiency but by that most Indian of obstacles: land acquisition. For a full year after launch, less than 1% of the required land had been acquired; farmers in Maharashtra litigated; a former Japanese minister publicly vented his frustration at the missed timelines. Here was Japanese precision — the J-slab ballastless track, the E5 trainsets that hold their alignment at 320 kph — colliding head-on with Indian land records. And yet. As of mid-2026 the project has, at last, turned a corner: all 1,389.5 hectares acquired, more than 350 km of viaduct standing across Gujarat, a tunnel boring beneath Thane Creek, a first operational section between Surat and Bilimora targeted for 2026 and the fuller corridor pushed towards 2027 and beyond. And here is the point the sarcasts miss: throughout the delays, Japan never walked away. JICA's financing came at roughly 0.1% interest over a 50-year term — soft-loan terms so patient they border on charitable. That patience, that refusal to be spooked, is itself the Japanese dowry. It is exactly the temperament India needs and exactly the temperament India tests to destruction.

The Current Courtship

Which brings us to the present, and to the reason I think this essay is timely rather than merely wistful. The families are talking again, and talking seriously.

In August 2025, Prime Minister Modi travelled to Tokyo for the 15th India-Japan Annual Summit and stood beside Prime Minister Shigeru Ishiba to announce a target of ¥10 trillion — roughly $68 billion — in Japanese private investment into India over the following decade, doubling the previous five-year goal. “India's massive market is so full of potential,” Ishiba told reporters, “that incorporating its vibrancy will help drive the growth of Japan's economy.” The two governments issued a “Joint Vision for the Next Decade,” launched an Economic Security Initiative spanning semiconductors, critical minerals, clean energy and pharmaceuticals, and — crucially for my argument — adopted an action plan for human-resource mobility: the two-way movement of 500,000 people over five years, including 50,000 skilled Indian professionals bound for Japan. Modi, with one eye on Donald Trump's 50% tariffs on Indian exports, told Japan Inc. that “capital doesn't just grow in India, it multiplies.” JETRO's own surveys, he noted, found 80% of Japanese companies wanting to expand in India and 75% already profitable there.

The semiconductor thread is the most tangible. The India-Japan Semiconductor Supply Chain Partnership, formalised in July 2023, has since produced real commitments rather than mere memoranda. Tokyo Electron — one of the handful of firms on earth, alongside ASML and Applied Materials, without whose equipment no advanced chip can be made — has partnered with Tata Electronics to supply and service the tools for India's first commercial fab at Dholera in Gujarat (a roughly $11-billion, 28-nanometre project) and its assembly-and-test facility at Jagiroad in Assam, opening offices in both and pledging to train Indian technicians. Japan's Renesas has inaugurated Indian design centres working on 3-nanometre architecture and is backing the CG Power OSAT venture in Gujarat with a reported $915 million; Sumitomo Chemical and others are engaging on the photoresists and ultra-pure gases in which Japan holds a 70–90% global share. In December 2025, ROHM announced a power-semiconductor alliance with Tata. Modi and Ishiba even toured Tokyo Electron's plant at Sendai together. This is the synergy in embryo: Japanese materials and equipment mastery, married to Indian design talent and Indian fabs. It is, for now, concentrated at the least complex end of the value chain — a humble 28nm beginning — but it is a beginning.

The talent flow, too, is real, if still modest. There were 56,686 Indian nationals in Japan as of June 2025 — the community concentrated in Tokyo's Nishi-Kasai district, the “Little India” of Edogawa Ward that first filled with Indian engineers during the Y2K remediation scramble at the turn of the century. Rakuten now runs roughly half of its 70-plus global businesses from India; the largest single occupational category of Indians in Japan is technology, humanities and international business, at nearly 13,000 people. The direction of travel is unmistakable: an ageing nation that once recruited Indian coders to patch its millennium bug is now, structurally and permanently, dependent on that same talent to climb down from its Digital Cliff.

And there is a poetic full circle worth savouring. In August 2025, at Suzuki's Hansalpur plant in Gujarat — a ₹21,000-crore facility heading towards a million-unit annual capacity — Modi flagged off production of the eVitara, Suzuki's first-ever battery-electric vehicle, made in India, built on the dedicated HEARTECT-e platform, with lithium-ion batteries produced locally by the Toshiba-Denso-Suzuki joint venture. It will be exported to over 100 countries — including, and here is the sweetness of it, Japan itself. Within two months more than 6,000 units had shipped from Pipavav port, making Maruti India's largest EV exporter. The country that in 1983 taught India to build a small car will now import electric cars built by Indians. The pupil is not merely the paymaster; the pupil now ships product back to the master's own showroom. If that is not the arranged marriage bearing fruit, I do not know what is.

The Synergy Map: What the Marriage Should Actually Build

Diplomatic communiqués are cheap; ¥10 trillion is a number, not a plan. If this partnership is to be more than a decade of ribbon-cuttings, it needs a concrete programme of co-creation. Here, then, is my prescriptive core — six ventures the two nations should be building in earnest, and one geopolitical reason they must.

1. Digitise the Japanese SME, with Indian SaaS

Japan has some 3.3 million small and medium enterprises, the vast bulk of them staring over the Digital Cliff with no in-house engineers and no SIer willing to take so small a contract. This is the single largest under-served software market in the developed world, and the Japanese government subsidises the software spend. Indian SaaS and IT-services firms should treat it as their next great frontier — not as body-shopped subcontractors to the SIer pyramid, but as direct providers of productised, localised software. The prize is nothing less than a share of that ¥6.46 trillion digital deficit, redirected from Seattle to Bengaluru.

2. Robotics hardware plus Indian vision-and-AI software

FANUC, Yaskawa, and Kawasaki build the finest robot arms on earth — and then bolt onto them control software that is, to put it charitably, of an earlier era. Pair that world-beating hardware with Indian computer-vision and machine-learning teams and you have the intelligent, adaptable, cheap-to-programme industrial robot that the China+1 factory floor is crying out for. Japan owns the body; India can supply the brain.

3. Semiconductors: the whole value chain, jointly

Japan holds a 70–90% share of key global semiconductor materials and a formidable position in equipment; India designs some 20% of the world's chips and is now, at Dholera and Jagiroad, learning to fabricate and package them. The Tokyo Electron-Tata and Renesas-CG Power ventures are the first bricks. The ambition should be an integrated India-Japan chip corridor — Japanese gases, chemicals and tools; Indian design and assembly — deliberately architected as an alternative to the China-centric supply chain, with the explicit goal of climbing from 28nm towards the advanced nodes over the decade.

4. The “Fit to Standard” ERP migration wave

Japan's SAP reckoning arrives in 2027, when mainstream maintenance for its most widely used ERP core ends, forcing thousands of Japanese firms to migrate off customised legacy systems onto standardised cloud platforms. This is a multi-year, multi-billion-dollar wave of work — and Indian services giants are positioned to ride it. TCS cracked Japan through partnership rather than pure subcontracting: its 2014 joint venture with Mitsubishi Corporation merged three entities into TCS Japan, in which TCS has since raised its stake to 66%, building an operation with well over $600 million in revenue and thousands of local associates. Infosys has its own venture with Hitachi, Panasonic and Pasona. The migration wave is the near-term, bankable heart of the whole thesis.

5. Joint sovereign AI for the non-Anglophone world

Here is the most intellectually exciting frontier, and it has two poster children. Japan's Sakana AI — founded by David Ha, Llion Jones (a co-author of the “Attention Is All You Need” paper) and Ren Ito — closed a ¥20-billion ($135 million) Series B on 17 November 2025 at a $2.65 billion valuation, becoming Japan's most valuable unlisted startup, backed by MUFG, Khosla Ventures, NEA, Lux Capital and In-Q-Tel. Its bet is “efficiency-first” AI: evolutionary model-merging that breeds new models rather than training them from scratch, explicitly a wager against brute-force compute. India's Sarvam AI, in February 2026, unveiled Sarvam-105B — a 105-billion-parameter mixture-of-experts model trained from scratch on Indian infrastructure (over a thousand H100s at Yotta's cluster) under the IndiaAI Mission, supporting all 22 official Indian languages, built for a reported $50 million. Two Asian efficiency-AI siblings, each refusing to concede that the future of AI must be a Californian monopoly of trillion-dollar compute. A joint India-Japan effort on sovereign models and shared compute for the world's non-English-speaking majority is the natural, and thrilling, next step.

6. EV and battery co-manufacture

The eVitara is the template: design and capital from Japan, manufacturing scale and cost discipline from India, batteries co-produced by a tripartite Japanese JV on Indian soil, product exported to the world including Japan. Extend it — to cells, to power electronics (the ROHM-Tata alliance), to the whole electrified drivetrain — and you have a China+1 EV supply chain that neither nation could build alone.

And the glue that binds all six: China. Both nations live under the same shadow. Japan has spent a decade de-risking from a China on which it remains dangerously dependent for rare earths and components; India runs a $99-billion-and-widening trade deficit with a neighbour it fought on the Himalayan frontier as recently as 2020. “China+1” is not a slogan for these two; it is an existential supply-chain imperative. The Quad's critical-minerals work, the semiconductor corridor, the rare-earth cooperation folded into the 2025 Economic Security Initiative — all of it is, at bottom, two anxious neighbours deciding they would rather depend on each other than on Beijing. Geopolitics, for once, is pushing in exactly the direction economics already points.

Space, I will note only briefly, offers the loftiest symbol: the LUPEX mission — designated Chandrayaan-5, financially sanctioned by India in March 2025 — will send an ISRO lander and a JAXA rover to hunt for water ice at the Moon's south pole, launching on a Japanese H3 rocket later this decade (2027-28 on current schedules). Indian lander, Japanese rover. The marriage, quite literally, reaching for the Moon.

The Frictions, and the One Trap That Would Ruin Everything

I would be a poor matchmaker, and a worse essayist, if I ended on the swelling strings without naming the obstacles — because they are formidable and a few of them are lethal.

The first is language. Japanese business runs, stubbornly and comprehensively, in Japanese. The SIer world expects JLPT certifications and keigohonorifics; the Indian engineer's comfort zone is English-only and blunt. This is not a trivial barrier; it is the reason Indian IT firms have found Japan the hardest of all developed markets to crack, with revenue from Japan stuck in the single digits as a share of the total for most of them. The second is cultural: the nemawashi and ringi of Japanese consensus — the patient, whole-organisation, bottom-up circulation of a decision until everyone has affixed their seal — sits in almost comic opposition to Indian jugaad, the improvised, move-fast, break-things-and-apologise-later hustle. Waterfall meets the hackathon. One side documents for forty pages; the other ships and iterates. Neither is wrong, but the friction is real and it burns hours.

The third is Japan's genuine ambivalence about immigration — a society that has only recently, and reluctantly, accepted that its demographic maths leaves it no choice but to import people, and where many Indian engineers still work a few years and return home rather than settle. The fourth is the money: a weak yen and salaries that, for a top-tier Indian engineer, make Tokyo distinctly less attractive than San Francisco or even a Bengaluru GCC. Japan is competing for Indian talent in a global auction, and its offer is not always the highest bid. And the fifth is the DoCoMo problem, still unexorcised: the Indian bureaucracy's capacity to spook patient capital, to turn a contractual exit into a three-year court battle, to make the simple act of getting paid an ordeal. No quantity of summit communiqués will fix this; only the boring, unglamorous work of legal and regulatory reform will.

But the deepest danger is none of these. It is subtler, and it is the one I most fear, because it is the path of least resistance. It is the risk that “co-creation” quietly degenerates into body-shopping — that the great India-Japan partnership becomes nothing more than Indian engineers slotted in as cheap subcontract labour at the bottom of the SIer pyramid, coding to someone else's spec, owning nothing, inventing nothing, learning nothing they could not have learned in a Chennai back-office in 2004. That outcome would betray both nations at once. It would leave Japan's structural software problem untouched — merely outsourced, not solved — and it would replicate, with a Japanese accent, the exact middleman pattern that my previous essay indicted as the tragedy of Indian tech. Cheap hands for hire is not a marriage; it is a transaction, and transactions do not compound.

The whole point of Maruti and Hero Honda was that they were joint ventures — shared equity, shared risk, shared upside, technology genuinely transferred rather than merely rented. Suzuki did not body-shop Indian labour; it built a company that eventually outgrew its parent. That, and nothing less, is the standard. If the sequel is to be worthy of the originals, the Indian side must insist on being a co-creator — a partner with skin in the game and a name on the patent — and the Japanese side must resist the seductive, familiar comfort of treating India as merely a cheaper floor of the subcontracting tower.

A Closing, in the Register of the Wedding Toast

I have deployed the marriage metaphor throughout this essay with, I hope, more affection than whimsy, and I want to end by taking it seriously rather than cutely.

Arranged marriages, in the tradition I grew up around, are not built on the lightning-strike of romance. They are built on something the sceptical West often underrates: complementarity, the patient judgment of families that these two people, whatever their surface differences, need what the other has and will grow into a love that infatuation could never have sustained. They begin not with passion but with fit, and — when they work — they end with a partnership deeper than either party could have designed alone.

Japan and India are not in love. They will never be in love; they are too different in tempo, in temperament, in the very grammar of how they make decisions. Japan will always find India maddening — the delays, the litigation, the improvisation where a plan was expected. India will always find Japan exasperating — the meetings, the seals, the forty-page logs, the glacial consensus. There will be more DoCoMos, more Ranbaxys, more bullet trains that arrive a decade late. Of this I have no doubt.

But love was never the point. The point is that a nation with the world's finest factories and no one left to write its software, and a nation with the world's largest pool of coders and no factory to call its own, have been introduced by history, by demography, and now by the plain arithmetic of a shared and rising China. The dowries are complementary to the point of absurdity. The elders have, at last, sat down to negotiate. And somewhere in Gujarat, an electric car designed in Hamamatsu and built by Indian hands is being loaded onto a ship bound for Yokohama — proof, if the families needed any, that when these two households actually commit, the offspring tends to conquer the world.

The first marriage gave India the car it drove into the middle class. The sequel, if both families can only find the nerve to sign, could give both nations the century. Not the “Asian century” of the conference-circuit cliché — I promised no such phrase without irony, and I keep my promises — but something smaller, truer, and harder-won: two flawed, ageing-and-young, precise-and-improvised civilisations discovering that their weaknesses were, all along, the exact shape of each other's strengths.

The banns have been read. It remains only to see whether anyone has the courage to say I do.

Sources & Further Reading

The Historical Marriages

• Global Suzuki — Update on India Business (Mar 2026 IR deck; Maruti share / India contribution).
• IBEF Maruti Suzuki showcase; Autopunditz / CarBikeGPT monthly share data (2026).
• Screener.in / Statista — Maruti Suzuki company profile (1983 first car; Suzuki stake).
• Business Standard — Maruti's share of Suzuki global sales / India revenue highs (contemporary coverage).
• Forbes India — Hero MotoCorp after the Honda split; Pawan Munjal / Hero journey features.
• Business Standard — Hero vs Honda post-split share trajectory (HMSI 13% → 27%).
• Autopunditz — “Journey from Hero Honda to Hero v/s Honda.”

The Divorces

• Fortune — “The latest to claim fraud at generic Lipitor maker Ranbaxy” (2013); see also “Dirty medicine”.
• Fierce Pharma — Daiichi accusations against Ranbaxy shareholders; Forbes — Singh brothers / Daiichi dispute.
• Tata–DoCoMo — Economic Times; Business Standard (Delhi HC, Apr 2017).
• SoftBank Vision Fund — TechCrunch ($32bn / ¥4.3tn annual loss); Hindu BusinessLine contemporaneous coverage.
• Mumbai–Ahmedabad HSR — PIB project updates; YourStory / Deccan Herald / The Metro Rail Guy land-acquisition and progress coverage (through mid-2026).

The Current Courtship

• 15th India–Japan Annual Summit — MOFA Japan summary; Joint statement PDF; Economic Times (¥10tn target, Aug 2025).
• The Diplomat — India–Japan economic security / semiconductor alliance essays (2026).
• Tata Group semiconductor / Tokyo Electron–Tata / ROHM–Tata coverage (WION, company releases, Dec 2025).
• Global Suzuki — “Start of Dispatch Ceremony of the BEV e VITARA” (26 Aug 2025); The Hindu / DD News Modi Hansalpur remarks.

Japan's Software Deficit and AI

• METI IT human-resource / Digital Cliff projections — summarised in Asset Value Investors and Nomura Research Institute citations of the 2019 METI survey.
• Japan digital trade deficit — Kyodo News (¥6.46tn / $43bn in 2024, MoF Feb 2025).
• Hitachi Solutions — cultural gap Japan vs US IT projects; JapanTechCareers on SIer vs in-house (~70% SIer estimate).
• Sakana AI — TechCrunch ($2.65bn Series B, 17 Nov 2025); Nikkei Asia unicorn coverage.
• Sarvam-105B — Forbes / Business Standard / TechCrunch / OfficeChai coverage (Feb 2026).

India's Complementary Numbers

• NASSCOM Strategic Review 2025 — ~$224bn IT-services exports, ~$283bn industry, 5.8m developers.
• Zinnov / Flexiple / PIB — Global Capability Centres (1,700+ centres, ~1.9m professionals).
• GTRI (Ajay Srivastava) / Department of Commerce — FY2025 China trade deficit (~$99.2bn) and electronics-import figures.
• Economic Survey 2025-26 and DST — India's GERD 0.64–0.84% of GDP; India Semiconductor Mission on ~20% chip-design share.
• International Federation of Robotics — World Robotics 2025 (Japan as major producer / installer market).

Demographics, Space and Talent Mobility

• Japan Statistics Bureau / Ministry of Internal Affairs — 29.4% aged 65+ (Sept 2025); IPSS 2023 Population Projections — working-age decline.
• LUPEX / Chandrayaan-5 — The Print / JAXA–ISRO interface work; ISRO financial sanction (Mar 2025).
• Japan Ministry of Justice statistics on Indians in Japan; Metropolis Japan / UR-net on Nishi-Kasai “Little India.”
• TCS newsroom / Business Standard — TCS Japan–Mitsubishi joint venture.

Alternate titles considered: “Fill It, Shut It, Co-Create It” · “The Coders and the Craftsmen” · “A Dowry of Deficits.”

On this site

• The Middleman's Republic — Part 1: the diagnosis this essay answers.
• The Rope Sellers — accountability moats and the professional-services cousin of the middleman pattern.
• The Rope Sellers Buy a Rope Machine — what happens when the thing you resell starts eating you.