Five Cars to Ecuador

In December 1975, a Korean group better known for pouring concrete than pressing steel began selling a four-door sedan with an Italian silhouette and a Japanese heart. It was called the Pony. In July 1976, five of them were shipped to Ecuador.

Five cars do not make an export industry. They make a declaration of intent.

That same year, on the banks of the Hooghly at Uttarpara, Hindustan Motors was building the Ambassador much as it had since 1957, on the lines of a Morris Oxford that Britain had long since abandoned. Buyers waited years for delivery. The Ambassador did not need to be exported. It scarcely needed to be sold. It needed only to be allocated.

Both cars were protected. This is the inconvenient fact that the familiar parable of Indian socialism versus Korean markets prefers to forget. Seoul sheltered its car industry more fiercely than New Delhi ever did. Korea admitted no foreign cars at all until 1987, and kept Japanese cars out until 1999. If tariffs and import bans were the disease, Korea ought to have been the terminal patient.

It was not. The firm that shipped those five Ponys now anchors one of the three largest automotive groups on earth. The Ambassador rolled off its last line in 2014, and its name was later sold to Peugeot.

The divergence lies in the terms. Korea protected the Pony so that it would learn to export. India protected the Ambassador so that it would never have to. Half a century on, the lesson remains open to India. The harder question, to which this essay will return, is how a democracy can enforce what a general once commanded.

The Sanctuary

The Ambassador was not born obsolete. When it reached Indian roads in 1957, its Morris lineage made it a respectable, even modern, sedan. Its misfortune was to live too long. Britain retired the design within a few years; India kept building it until 2014, a run of fifty-seven years that owed more to the stillness of policy than to the genius of Alec Issigonis.

The stillness was deliberate. The Industries (Development and Regulation) Act of 1951 placed private industry under licence. To build a factory, add a shift or expand a line, a firm needed the blessing of Delhi. Ford and General Motors, unwilling to manufacture on such terms, left in 1954. The car market settled into a gentlemanly duopoly of Hindustan Motors in Calcutta and Premier Automobiles in Bombay, the latter producing a Fiat that would become the Padmini.

What followed was protection of a peculiar kind. It shielded firms from foreign rivals, and then shielded them from one another. Capacity was rationed, prices were managed and waiting lists ran to years. A customer did not so much buy an Ambassador as qualify for one.

It would be unfair to call this system lawless. The Indian state attached a great many conditions to its protection. It simply attached the wrong ones. It forbade firms to grow too large, to import too freely or to produce beyond their licensed quota. It never required them to become good.

The ideology was sincerely held. A nation of famines and currency crises, the argument ran, could not squander scarce dollars on the motoring classes. A car was a luxury; a better car, an extravagance. The result was protection that preserved producers and improved none of them.

This is the crucial distinction. Protection was meant to be a nursery. India built a sanctuary instead. A nursery is a place one is expected to leave. A sanctuary is a place one is permitted to stay.

The Apprenticeship

History enjoys a private joke here. The Pony, too, had British blood. To engineer it, Hyundai hired George Turnbull, until recently managing director of Austin Morris at British Leyland, the very lineage that had given India its Ambassador. Two Asian republics drew from the same exhausted well of British motoring. One drank and grew stronger. The other drank and fell asleep.

Hyundai Motor was founded in 1967 to assemble Ford Cortinas from imported kits. It might have remained a screwdriver operation indefinitely, as many such ventures across the developing world did. The Korean state would not allow it. In 1973, the Ministry of Commerce and Industry issued a long-term plan for the motor industry. Its terms were exacting: local content of 95 per cent in small cars by 1975, annual output of half a million vehicles by the early 1980s and exports worth $150 million.

These were not aspirations but the price of admission. Firms that wished to enjoy a protected home market had to present credible plans for an indigenous model. The Pony was Hyundai’s answer: styled by Giugiaro in Turin, powered by a Mitsubishi engine under licence, built in Ulsan. In 1976, its first full year, it sold 10,726 units and took 43.6 per cent of the Korean market.

The same year, the five cars sailed for Ecuador.

The export obligation was enforced from the very top. From 1965, Park Chung-hee presided over a monthly export promotion meeting of ministers, bankers and industrialists. Annual targets were assigned to industries and individual firms, and performance was reviewed against them in the presence of the president. Park attended 147 of the 152 meetings held before his assassination in 1979. Few heads of government have taken such a personal interest in the order books of their manufacturers.

The rewards were real. Credit, in a capital-starved economy, was the scarcest commodity of all, and the state controlled the banks. Export performance unlocked cheap loans; failure closed the tap. When the industry overreached at the end of the 1970s, Seoul did not hesitate to rearrange it. In 1981, Kia was ordered out of passenger cars altogether and confined to trucks and buses.

Here is the essential contrast. The Indian licence told a firm what it might not do. The Korean bargain told a firm what it must do. One was a fence. The other was a racetrack, with a finish line, a stopwatch and a referee who could disqualify you.

Nothing Given for Free

The economist who gave this contrast its theory was Alice Amsden, who taught at MIT and mined the loan files of development banks across the late-industrialising world. Her reading of those ledgers produced one of the more useful ideas in development economics, and one of the least fashionable.

Amsden began where free-market economists would agree. In an open economy, competition disciplines the firm: those who make poor products at high prices lose customers, then capital, then existence. The market is an examiner who never sleeps. Protection suspends the examination. Here Amsden parted company with planners and liberalisers alike. The planners assumed protection was sufficient; the liberalisers assumed it was fatal. Amsden argued that it was neither, provided the state replaced the discipline it had removed.

She called this replacement the reciprocal control mechanism. Every subsidy, every concession, every year of shelter was exchanged for a monitorable result. Her verdict on the successful latecomers was blunt: nothing was given away for free. In the loan files of the Korea Development Bank, she found conditions that read less like industrial policy than like a demanding father’s instructions. Hire a finance director from outside the family. Buy the latest machinery. Bring in foreign technical consultants. Export, or explain why not.

Three properties made the mechanism work. The standard had to be measurable, so that success could not be faked. It had to be bounded in time, so that infancy did not become a permanent condition. And it had to be revocable, so that the threat of withdrawal was believed. Even Thailand’s Board of Investment, hardly a byword for dirigisme, stripped benefits from around 7 per cent of its clients each year for breach of terms. Remove any one of the three and the bargain decays into a gift.

A subtlety deserves attention here, because it saves the argument from caricature. Amsden did not place India among the states that attached no conditions. Delhi imposed conditions in abundance. But they governed inputs and permissions rather than outcomes: how much a firm could make, where it could build, what it could import. Korea’s conditions governed results: what a firm achieved, how much it sold abroad, how fast it localised.

The moral is precise. A state that protects without judging has abolished the examination while continuing to collect the fees. Every successful industrial nation, from Hamiltonian America to Bismarckian Germany, sheltered its infant industries. The question is whether the infant is ever asked to stand.

Why the Referee Never Arrived

Why did India not learn the lesson? Ideology explains the beginning. The planners of the 1950s were export pessimists, doubtful that a poor country could sell manufactures to the rich and suspicious of profit as a guide to national purpose. A state that regards ambition with suspicion is unlikely to demand it of its firms. Ideology does not, however, explain the persistence. Protection creates rents, and rents create constituencies. Over three decades, the sanctuary acquired its own priesthood, and those who prosper from scarcity rarely campaign for abundance.

The sociologist Peter Evans offered a more structural answer. Comparing Korea, Brazil and India, he argued that successful developmental states possess embedded autonomy: they are close enough to business to understand it, and independent enough to discipline it. For much of its first four decades, the Indian state erred on the side of distance, treating industrialists as suspects to be licensed rather than partners to be tested. A liberalised economy faces the opposite hazard: a state grown so close to its favoured firms that it can no longer bring itself to judge them.

Candour requires two qualifications. The first concerns the source of Korean discipline. Park did not persuade the chaebol; he commanded them. Soon after seizing power in 1961, his regime arrested leading businessmen for illicit wealth accumulation and released them in return for investment in his priorities. A democracy cannot borrow that method. Its discipline must come from rules and transparency instead.

The second concerns the perishability of discipline. Korea did not keep its referee for ever. As the chaebol grew too large to fail, the state grew too timid to judge. Daewoo, once the country’s second-largest conglomerate, borrowed its way through the Asian financial crisis and collapsed in 1999 with debts of around $50 billion. Reciprocity is a muscle, and it atrophies without use.

Nor is India’s record uniformly bleak, and the most instructive exception is the industry with which this story began. In 1982, the government formed Maruti Udyog with Suzuki and insisted on progressive localisation. When the sector opened in the 1990s, Delhi went further, and came closer to Seoul than it is usually given credit for. Under a public notice of 1997, every carmaker importing kits had to sign a memorandum committing it to 50 per cent local content by its third year and 70 per cent by its fifth, and to balance the foreign exchange spent on imported components with exports of cars and parts. This was Amsden’s bargain in all but name: access to a protected market, in exchange for measurable, time-bound results. By May 2001, nearly every major manufacturer had crossed the 70 per cent threshold.

The bargain did not survive Geneva. Acting on complaints from the United States and the European Communities, a WTO panel ruled in 2001 that both conditions breached India’s obligations, and the policy was withdrawn. The capability, however, outlived the rule. Maruti’s cars are now more than 95 per cent locally sourced. In 2022, India overtook Japan to become the world’s third-largest vehicle market, and in 2025 it exported a record 8.63 lakh cars, to buyers across the Middle East, Africa and Latin America. The car industry did not refute the Korean lesson. It confirmed it twice: first by its long sleep, and then by its awakening.

In pharmaceuticals, Amsden herself observed that price controls, paired with loosely enforced foreign patents, pushed Indian firms towards cost-saving innovation and exports. The pharmacy of the developing world was, in part, a product of disciplined protection.

India, in other words, does not lack the knowledge. It lacks the habit.

The Test, Applied

The habit is now being tested again. Since 2020, India’s flagship industrial instrument has been the production-linked incentive, a programme of ₹1.97 lakh crore spread across fourteen sectors. Its most visible success sits in millions of pockets: smartphones became India’s single largest export in 2025.

In one respect, the scheme is the most Amsdenian instrument Delhi has ever designed. It pays for results, not promises: firms receive nothing until incremental sales materialise, and only for a fixed term. Critics note that less than a fifth of the outlay had been disbursed by mid-2026, and read this as failure. Amsden might have read it, at least partly, as discipline: a state declining to pay for performance that never arrived.

The difficulty lies in what the scheme chose to measure. Incremental sales are a metric of volume, not of capability. A firm can satisfy them by assembling imported kits at scale, which is precisely the Cortina stage from which the Korean state forced Hyundai to graduate. Domestic value addition in Indian mobile phones stands at roughly 22 to 23 per cent, against an original ambition of 40. Under the first scheme, value addition was monitored but not required for payment. Monitored but not required: there, in four words, is the gap between a nursery and a sanctuary.

There are signs that the lesson has been absorbed. The government has said it wants value addition in phones to reach 35 to 40 per cent, and the successor schemes are expected to tie incentives to domestic content. If that is done with measurable targets, fixed horizons and credible clawbacks, India will have rediscovered, half a century late, the bargain that built the Pony.

The same test applies to the national champions. A large economy needs large firms capable of competing abroad, and a state that helps to build them need not apologise. What it cannot defend is help without terms. Any firm, in any sector, that receives cheap land, protected markets or privileged access to public contracts should be able to answer three questions in public. What has it been asked to deliver? By when? And what happens if it does not? Where the answers are vague, the arrangement is not industrial policy. It is patronage with a business plan.

The external weather makes these questions only more urgent. The American tariff of 50 per cent imposed on Indian goods in August 2025 lasted barely six months before a trade framework cut it and the United States Supreme Court struck down its legal basis. A strategy built on tariff arbitrage is a strategy built on another country’s politics.

India spent the late 2010s raising its tariffs, and has since changed course: average customs duties have been cut since 2025, the finance minister now promises single digits on most items, and free trade agreements are being signed at pace. Opening can be defended as readily as closing; neither, by itself, builds capability. Every rupee of incentive and every point of protection ought to come with the question Park asked each month: what have you sold abroad?

The Referee a Democracy Can Afford

There is a difficulty with Park’s question. Asked as a condition of subsidy, it is now unlawful.

The WTO’s Agreement on Subsidies and Countervailing Measures prohibits two kinds of support outright: subsidies contingent on export performance, and subsidies contingent on the use of domestic over imported goods. These were the twin engines of the Korean bargain. Both are contraband today.

India has learnt this in Geneva more than once. The car case of 2001 was the first lesson. In 2016, the Appellate Body upheld a ruling that the domestic content requirements of its National Solar Mission breached its obligations. In 2019, a panel held that India, having crossed the threshold of $1,000 in income per head, could no longer offer export subsidies; its appeal waits before an Appellate Body that has ceased to function. In October 2025, China requested consultations over the production-linked incentives for batteries and automobiles, arguing that their domestic value addition conditions amount to prohibited import substitution.

The irony is hard to miss. The very correction this essay commends, tying incentives to local content, is the one most exposed to challenge. Another door closed at the end of 1999, when a safe harbour for research subsidies lapsed because members declined to renew it. The rules are not enforced even-handedly; Washington’s own climate legislation of 2022 rewarded domestic content generously. But a strategy that depends on selective enforcement is a wager rather than a plan.

What remains is more than the pessimists suppose. Three instruments survive. The first is conditionality on conduct rather than destination. Amsden’s techno-standards govern how a firm is run, not where it sells or what it buys: professional management, plants of efficient scale, prudent leverage and spending on research and skills, each tied to time-bound support and clawed back on failure. The second is the tariff itself, within limits India has already negotiated: its average bound rate on non-agricultural goods is about 36 per cent, and roughly three in ten of its industrial tariff lines carry no binding at all. The third is the state’s own purse. India has remained outside the WTO’s Agreement on Government Procurement, as an observer since 2010, and may lawfully prefer domestic suppliers in its own purchasing.

The deeper answer lies elsewhere, and it is one Park could never have offered. A democracy cannot summon its industrialists to a monthly audience with a general. It can do something more durable: publish. Every firm receiving public support could be bound to report its performance against agreed milestones in an open ledger, available to Parliament, the press and competitors alike. Rivals are the most diligent auditors ever devised. Sunlight is the democratic substitute for the presidential glare.

The export market, meanwhile, can still serve as examiner, even if exports cannot be the formal condition. The state need not require firms to sell abroad. It need only make shelter finite: legislate a sunset for every tariff and incentive at the outset, and decline to renew it. When the date arrives, the firms left standing will be those able to compete with the world. The verdict will have been delivered by a buyer in Riyadh or Johannesburg who cannot be lobbied, flattered or bribed.

A sunset, of course, is only as credible as a government’s willingness to let the sun set. That is a test of political character, and no treaty can sit it on a nation’s behalf. Park’s referee wore a uniform. A democracy’s referee is a stranger, somewhere abroad, choosing between two cars.

A Loan, Not a Gift

One objection remains, and it is the most fashionable. Perhaps the factory is the wrong prize altogether. Automation is thinning the workforce on every assembly line, the argument runs, and India’s future lies in services. The objection deserves a reply. Automation reduces the number of workers a factory needs; it has not reduced the importance of factories. Across the developing world, manufacturing’s share of output has held at around a fifth for decades, and at constant prices it has risen. Manufacturing has not disappeared. It has changed address. Meanwhile, the services India sells most readily abroad, routine coding and back-office processing, are precisely those most exposed to artificial intelligence. The machines are an argument for building capability faster, not for abandoning the attempt.

Every tariff is a transfer. Consumers pay more today so that producers may become competent tomorrow. The Korean motorist of the 1970s paid handsomely for his Pony, denied the cheaper and better Japanese cars a short sea crossing away. He was repaid, eventually, with Hyundai.

The Indian buyer who waited years for his Ambassador made the same payment. He was repaid with nostalgia.

Both publics made the same sacrifice. Only one state imposed an obligation in return. In Seoul, the state behaved like a creditor: it lent shelter, set terms and called in the debt. In Delhi, it behaved like a benefactor, and benefactors rarely ask to be repaid.

History has since supplied an epilogue of almost theatrical irony. In 1998, Hyundai opened its first fully integrated plant outside Korea, at Sriperumbudur, on the outskirts of Chennai. It is now India’s largest exporter of passenger cars by cumulative volume, having shipped more than 3.9 million vehicles to over 150 countries. Not far away, at Thiruvallur, Hindustan Motors kept its second plant. There, in its later years, the maker of the Ambassador assembled Mitsubishis under licence, products of the same Japanese house that had once supplied the Pony’s engine. The firm that sent five cars to Ecuador now exports from the doorstep of the champion India protected.

None of this argues for a return to the licence raj, nor for transplanting Korean methods wholesale. The argument is narrower, and more demanding. India will protect and subsidise its industries in the decades ahead, as every serious industrial nation has done. The only question is whether it will remember to collect.

Protection is a loan, not a gift. The Pony repaid its loan with interest. The Ambassador was quietly forgiven. India’s next generation of industries will follow one or the other, and the choice will be made in the fine print of the bargains the state is willing to enforce.