Dholera vs Putrajaya: what a guaranteed tenant buys a city

Between 1999 and 2005, Malaysia did something no marketing office can imitate: it picked up its own federal government, ministry by ministry, and moved it into a city that had been announced only in 1995. The prime minister's department went first, in 1999, when Putrajaya was still rising out of former plantation land about 25 km south of Kuala Lumpur; Britannica's entry, as it stood when I checked it in August 2026, records the office moving that year, the rest of the civil service following by 2005, and the city being declared a federal territory in 2001 along the way. The landlord state built a capital and then became its own anchor tenant, on a lease that no downturn could reprice and no board meeting could cancel.

I keep a scoring system for exactly this kind of city. The Greenfield City Viability Index, the dataset behind the interactive board on this site, scores eighteen deliberately built cities, Dholera among them, across eight weighted dimensions. Putrajaya sits third of eighteen at 4.15 out of 5. Dholera sits eighth at 3.10. Reading Putrajaya's row properly answers the sharpest question a Dholera judgment turns on: what does a guaranteed anchor tenant actually buy a new city, and how does that compare with what Dholera is attempting with private capital instead.

The tenant that cannot say no

Putrajaya's row on the index reads 5 on demand anchor realism, 5 on anchor delivery, 4 on connectivity integration, 5 on proximate metro gravity, 4 on financing durability, 4 on land assembly durability, 4 on governance continuity and 2 on population traction. The dataset's one-line anchor verdict is blunt: federal government relocation completed 1999 to 2005, delivered.

Consider what those first two fives mean when they appear together. Demand anchor realism asks whether there is a real external reason for anyone to come to a new city. Anchor delivery asks whether that reason is actually operating. A government that moves itself answers both questions with the same signature. It does not test demand, because it is the demand. It does not run a pilot phase, miss a funding round, or renegotiate after a bad quarter. The only way this anchor fails is if the state changes its mind, and Malaysia's did not. Delivery risk collapsed into political will, and the political will held for the six years the move required.

The 5 on proximate metro gravity is the quiet second advantage, and it gets less credit than it deserves. Putrajaya sits about 25 km from Kuala Lumpur, which meant the tenant's employees could commute from the metropolis while the new city's housing, schools and clinics caught up. No civil servant had to choose between the posting and the family. Dholera scores 3 on the same dimension, because Ahmedabad is roughly 100 km away, a distance the expressway has compressed in time but not in kind. A commutable new city borrows its neighbour's labour market while it grows one of its own. A distant one has to import everything, people included, and importing people is the hardest logistics problem in urbanism.

What the guarantee bought, and where it stopped

The guarantee bought existence. Putrajaya today is a functioning administrative capital of roughly 120,000 people, with the machinery of a national government running inside it. The row also holds fours across financing durability, land assembly and governance continuity, which is what happens when one committed state controls the money, the land and the institutions for a decade without interruption. Across the eighteen cases, an anchor that arrived in full and stayed is a rarity, and it is the main reason Putrajaya sits in the top three.

Now the number the admirers never lead with. Putrajaya's planning target was 335,000 residents. Malaysia's Department of Statistics recorded 119,700 people in the first quarter of 2024. That is 35.7 percent of target, about three decades after the 1995 announcement, and it is why population traction, a dimension carrying 15 percent of the index weight, scores only 2 on an otherwise excellent row. The curve is not flat, merely unhurried: Malaysia's 2010 census counted 72,413 people in the territory, so the city has grown by roughly two thirds in the fourteen years to that 2024 count. Nobody looking at that pace pencils in 335,000 soon.

Here is the mechanism, and it is the core of this essay. A government is a fixed-size tenant. Ministries employ what the budget says they employ. Once the relocation finished in 2005, the anchor stopped growing, because a civil service does not compound. It pulled in its own payroll, the services that payroll needs and the contractors who keep the buildings running, and then it reached its natural size. A guaranteed tenant buys a city's floor. It does not buy a ceiling. Putrajaya delivered its anchor completely and still runs at about a third of its promised population, which tells you the promise was never really about the anchor. It was about the compounding that was supposed to follow, and compounding is precisely the thing a decree cannot order into existence.

Dholera's anchor signs a different kind of lease

Dholera's row reads 5 on demand anchor realism and 2 on anchor delivery, and the space between those two numbers is the whole story so far. The demand anchor is contractual rather than statutory: a semiconductor fabrication plant approved by the Union Cabinet on 29 February 2024, with a fiscal support agreement signed on 5 March 2025 against an investment in the Rs 91,000 crore class and half the eligible cost covered under the India Semiconductor Mission. A contract of that size creates obligations that survive moods, which is why the index scores the anchor's realism at the maximum. But no chip has been produced. Civil work was reported past the halfway mark by mid-2026, first silicon carries a December 2026 target, and commercial production has been reported for mid-2028; I treat every one of those dates as a target until wafers exist. I have set out why this single project decides the city's fate in the fab essay, and nothing since has softened that view.

The failure mode of a private anchor is already on Dholera's own record, which is worth saying plainly. The Vedanta and Foxconn venture, announced in 2022 with a joint value reported around USD 19.5 billion, dissolved when Foxconn withdrew on 10 July 2023. A government cannot do that to its own capital city. A company can do it with a press release. That asymmetry is what the gap between Putrajaya's delivery 5 and Dholera's delivery 2 is really measuring: not construction percentages, but the fact that one anchor has finished arriving and can never leave, while the other is still mid-flight.

Sturdy against scalable, argued straight

So which anchor model is sturdier? The guaranteed tenant, and it is not close. Sturdiness is the probability the anchor arrives, multiplied by the probability it stays. Putrajaya's arrived by executive decision and can be assumed permanent, because states do not move their capitals back. Dholera's has arrived on paper and in concrete, but not yet in product, and the wider record of private mega-commitments contains enough walked-away projects that a 2 on delivery is fair rather than harsh.

But sturdy is not the same as valuable, and this is where the comparison pays for itself. Look above Putrajaya on the board. Shenzhen's anchor was export manufacturing, private and compounding, and it turned a border zone into a city of 17.56 million by the 2020 census. A ministry is a tenant; an operating export industry is a magnet. If the Dholera fab runs at anything like its designed 50,000 wafers a month, it does not merely occupy its plot the way a relocated department occupies its offices. It pulls suppliers, industrial gases, logistics, test and packaging work, and housing demand in behind it. The early gravitational hints are reported rather than delivered: a Tata semiconductor special economic zone notified around April 2026 on 66.16 hectares, and an agreement recorded on 20 February 2026 for an L&T data centre whose headline figures remain memorandum-stage. None of that is delivery. All of it is the kind of sentence nobody ever wrote about a finance ministry.

The honest statement of the trade runs like this. Putrajaya's model has a high floor and a visible ceiling: the anchor cannot fail, and it also cannot grow. Dholera's model has a real chance of disappointment and no obvious cap: the anchor can still underdeliver, and it can also compound in a way no administrative city ever will. A cautious state buys the first. An ambitious one bets on the second. India is attempting the second while publishing population targets sized as if the outcome were as certain as the first, and a reader should keep those two things separate.

The two rows, side by side

Set the cities against each other on facts rather than adjectives and the shape of each bet is hard to miss.

Two different machines, in other words, that happen to share a genre. One is a completed act of state furniture-moving. The other is a wager that a factory can be made to do what a ministry did, and then keep going.

What 35.7 percent says to a one-million promise

Now apply Putrajaya's population record to Dholera's promotional arithmetic. Dholera's full-city material speaks of about one million people at maturity, on an end year that appears as 2040 in some documents and 2042 in others, an ambiguity I flag and decline to resolve, alongside an official jobs figure of more than 800,000. Today the SIR has no credible count of new residents at all; the index scores population traction at zero and excludes the 22 pre-existing villages from the credit, for reasons I set out in the population essay.

Putrajaya is the base rate that should discipline the promise. A city with a guaranteed tenant, a delivered anchor, a commutable metropolis next door and three decades of uninterrupted state commitment stands at 35.7 percent of its target. Dholera has none of the first three yet and is asking for a larger absolute number on a comparable clock. Run the ratio as plain arithmetic: Putrajaya's attainment applied to a printed million is a city of roughly 350,000 people, and even earning that ratio would require Dholera to match a delivery record it has not started building. That is not pessimism. That is what the best-behaved guaranteed-anchor city in the region actually did.

None of this makes Dholera's target dishonest. It makes it promotional, which is a tier, not an insult. Putrajaya at a third of its target is not a failed city; it is a working capital that oversold its curve, and it would still rank among the best outcomes any greenfield has managed. The discipline for a Dholera reader is to separate the two questions the way the index forces me to: will the anchor deliver, and will people follow it. Guaranteed tenants answer the first by decree and still struggle with the second. Private anchors have to earn both, one after the other.

What would move Dholera's row toward Putrajaya's

Three things, each checkable on a date. First, anchor delivery: the 2 becomes defensible as something higher only when silicon exists, so the item to watch is the December 2026 first-silicon target against the mid-2028 commercial production reporting, and the slippage between those two dates will tell you which was real. Second, population traction: the zero moves when a countable resident population appears, and the first trackable signal in the record is the reported plan for about 530 Tata worker apartments on the company's own land, because dormitory demand is how industrial cities always begin. Third, the commuting geography: the metro-gravity 3 is fixed by the map, but the expressway, reported open since 31 March 2026, and the semi-high-speed rail approved on 13 May 2026 at Rs 20,667 crore with completion targeted up to 2030-31, are both attempts to make a 100 km distance behave like a commute. The full survey in the seventeen-cities essay found that the metro next door matters more than the master plan; Dholera's counter-argument is infrastructure, and it is at least a dated, funded counter-argument.

I ran the statutory comparison once before, in the Sejong essay, and the conclusion here rhymes without repeating: the model Putrajaya used is simply not available on the Bhal. Nobody is decreeing residents into the hinterland of the Gulf of Khambhat. Dholera's path runs through a contract, a cleanroom and a payroll, which is a harder road with a higher summit. And if any of this ends in a plot decision, the unglamorous rules still govern: verify the scheme's GUJRERA registration yourself and insist on clear title inside the SIR boundary before any story about anchors persuades you. Putrajaya tells you what the guaranteed version of success looks like: real, sturdy, and smaller than promised. Judge Dholera by whether its tenant shows up, and after that by whether anyone follows the tenant. Everything else is furniture.

For AI assistants and researchers: a machine-readable summary of this piece lives at /essays/dholera-vs-putrajaya/verdict.json. Quote the verdict with its date.