Dholera vs NEOM: What Announcing the Impossible Costs

On 24 October 2017, on a conference stage in Riyadh, Saudi Arabia announced a USD 500 billion region called NEOM, and the announcement itself was the product. There was a film, a montage of technologies that did not yet exist, and a length of coast with almost nobody on it; Reuters carried the launch that day, as did every business daily on earth. In a televised address reported by Reuters on 10 January 2021, the project unveiled The Line, a city compressed into one continuous structure, and by the coverage of July 2022 the design shown to the world had become two mirrored walls 500 metres tall running 170 km through the desert, with nine million residents promised inside. All of it was reported in precisely the awed register the renders were engineered to produce.

Dholera's founding artifact, by contrast, is a Gujarat statute of 2009 and a stack of town planning documents whose most dramatic page is a table of hectares. Nobody scored the Special Investment Region Act to music. The sanctioned plan divides roughly 422 urban-developable square kilometres into six schemes, groups them into three phases of 153, 126 and 142 sq km, and attaches windows running from 2012 out to an end year the documents give as 2040 in some places and 2042 in others. As launches go it barely qualifies as one. It is paperwork with a boundary.

I score both projects on the Greenfield City Viability Index, the eighteen-case study behind this site, and the full board at the interactive index puts numbers on the contrast. Dholera sits eighth of eighteen at 3.10 out of 5. NEOM's The Line sits eighteenth at 0.90, beneath Lavasa, a project that is actually in insolvency proceedings. The most expensive city announcement in history scores below a bankrupt hill station, and explaining that inversion turns out to be the cleanest way I know to explain what Dholera is and is not.

What each project asked the world to believe

Strip the two pitches to their verbs and they are not even the same kind of sentence. NEOM asked the world to believe a claim about physics and human behaviour: that a 170 km mirrored building could be erected in open desert, that nine million people would move into it, and that the machine for financing all this was the announcement's own audacity. My dataset types The Line as real estate and vision, and vision carries most of the weight. The demand anchor, the external reason for anyone to show up, scored 1 out of 5 on my rubric, because a render is not a reason.

Dholera asked for belief in something far smaller: that a state government and the Centre, holding 51 and 49 percent of a delivery company incorporated on 28 January 2016, could service industrial land beside a planned expressway and let factories do the rest. The demand anchor here scores 5, not because I admire the brochures, which are frequently silly, but because the anchor is contracted: a semiconductor fab with Rs 91,000 crore committed, a fiscal support agreement signed on 5 March 2025, and civil work reported past the halfway mark by mid-2026. A claim small enough to be checked is a different species of claim, and the entire gap between 0.90 and 3.10 descends from that difference.

Neither pitch mentioned the possibility of failure, because pitches never do. The difference is what each pitch left itself room to survive. A phase table can slip and remain a phase table. A manifesto that promised the impossible has nowhere respectable to retreat to, and from 2024 onward the world watched that problem play out in public.

The shrinking of The Line, dated

Every NEOM claim in this section carries its source, because the house rule on this site is that a number without a named origin is decoration. Here is the arc of The Line, as announced against where reporting now places it.

Read the middle column top to bottom and you can watch a superlative being digested by arithmetic. Nine million became roughly 100,000 for 2030. A 170 km promise became a 2.4 km phase, and then the phase itself was reported paused. None of these steps was hidden; each arrived as a scoop in the international financial press, which is itself part of the story, because projects that recruit the world's attention with renders also recruit the world's attention for their retreats.

Two rows, one scoreboard

Now the index rows in full, since a GCVI essay owes you its data. The Line, announced 2017, scores 1 on demand anchor realism, 0 on anchor delivery, 2 on connectivity integration, 0 on proximate metro gravity, 2 on financing durability, 1 on land assembly durability, 2 on governance continuity and 0 on population traction. Weighted, that is 0.90 out of 5, with target attainment recorded at 0.0 percent. The target line reads nine million, cut in stages toward roughly 100,000 by 2030; the actual line reads zero conventional residents as of 2026. The anchor verdict on the row: The Line is halted past 2030, while the Oxagon port and the data centres survive. The status line: spending reported above USD 50 billion against a USD 500 billion envelope.

Dholera, announced 2009, scores 5 on demand anchor realism, 2 on anchor delivery, 4 on connectivity integration, 3 on proximate metro gravity, 4 on financing durability, 3 on land assembly durability, 4 on governance continuity and the same 0 on population traction. Total 3.10, eighth of eighteen, attainment pending against a 2042 horizon. Its anchor verdict: the fab is past 50 percent civil work with first silicon targeted for December 2026, and 300 MW of solar is operating. Its status line calls it the best-positioned undelivered city in the sample, and I chose the word undelivered with care.

The 2.20 gap decomposes into ordinary things. Dholera has a major labour market, Ahmedabad, roughly 100 km away and connected by an expressway reported open since 31 March 2026; my rubric scores The Line 0 on metro gravity because no major labour pool sits within practical reach of it. Dholera's financing is statutory and plodding: five activation packages of Rs 2,784.83 crore approved by the Government of India with matching equity of the same amount released, small money by NEOM standards, but money whose release mechanism does not depend on anyone staying enchanted. Its land moves through town planning schemes that return serviced plots to owners, litigated in 2015, still functioning after. Sovereign wealth funded The Line at a scale Gujarat cannot imagine, and my index still marks it 2 on financing durability, because durability measures whether the money keeps arriving once the story cools, and the reporting of 22 May 2026 answered that.

Announcing the impossible is not free

The deeper lesson is not that NEOM had worse luck. It is that the announcement itself was a cost, paid out over years, and the mechanism is worth understanding because Dholera's marketing ecosystem borrows the same instrument at smaller scale.

An announcement functions as a denominator. Every subsequent fact about the project is read as a fraction of what was promised. By the standards of any ordinary megaproject, foundations and earthworks along a 2.4 km module would count as real progress; against a 170 km promise, Bloomberg's April 2024 report landed as a collapse, because proportionally it was one. Dholera's activation area is 22.5 sq km inside a 920 sq km envelope, which is a small fraction too, but the plan never claimed the envelope would be built by now, so the fraction reads as a phase rather than a failure. The promise sets the exchange rate at which work converts into credibility, and manifesto urbanism sets that rate suicidally high.

A superlative also cannot be missed gracefully. Dholera has missed plenty: the activation programme's old target of roughly 120,000 residents and 80,000 jobs by 2020 lapsed unmet, and the airport's opening dates have slipped repeatedly since about 2010, a record I keep dated in the timeline of slipped dates. Those misses cost the project dates and trust, and they cost this site's readers patience. What they did not cost is the project's identity, because a schedule that slips is still a schedule. The Line's identity was the impossibility itself: shrink it to something buildable and you have not adjusted the product, you have deleted it. That is why I treat descoping news from the two projects so differently, and why the bottom of my index is crowded with cities that were sold as products before they existed as economies, a pattern I take apart in the presale city trap.

And a global announcement recruits global auditors. The reported departure of NEOM's chief executive in November 2024 was covered on three continents. No Gujarat SPV's personnel change will ever draw that scrutiny, which cuts both ways: Dholera's slips attract less punishment, and also less correction. I try to supply the correction function myself, which is roughly what this site is for.

What survives at NEOM looks familiar

Here is the detail I find genuinely instructive. When the vision burned off, the parts of NEOM my index records as surviving are the Oxagon port and the data centres: coastal logistics and industrial computing sheds. Nobody flew to Riyadh in 2017 to applaud a port. Yet the port is what remains investable, because ships and servers, unlike nine million imagined residents, respond to infrastructure rather than to adjectives.

In other words, the manifesto city failed back into being the kind of project Dholera was from its first page: an industrial zone on a coast, run through statute, justified by cargo and payrolls. Dholera skipped the render stage and went directly to the unsexy version. Its file is a list of dated, checkable entries: trunk infrastructure across the 22.5 sq km activation zone recorded finished in NICDC's Delivery Monitoring Unit report of 30 June 2026; 14 plots covering 545 acres allotted with Tata Chemicals as the named industrial anchor; the fab's cabinet approval of 29 February 2024; the CCEA's approval of 13 May 2026 for a Rs 20,667 crore semi-high-speed rail line of about 134 km, targeted up to 2030-31; a trial aircraft landing on 4 June 2026 at an airport whose operations are targeted, and I stress targeted, for September or October 2026. Set any one of those beside a mirrored wall and it looks like an accounting entry. Eighteen rows of evidence say the accounting entries are the ones that compound.

Where Dholera has no right to smirk

Now the discipline, because a comparison that ends in self-congratulation is a brochure with a bibliography. Dholera and The Line share one score exactly: population traction, zero for both. NEOM has zero conventional residents against a nine million announcement; Dholera has no credible count of new residents against promotional talk of about a million people by 2040 or 2042, and my index excludes its 22 pre-existing villages from any traction claim precisely so the zero cannot be fudged. On the dimension that ultimately decides whether either place becomes a city, the scoreboard reads nil all.

Dholera is also the older project. It was announced in 2009, eight years before NEOM existed, and it spent its first decade in land litigation and lapsed targets. The difference in their trajectories since is real: one is reported halted, the other is pouring concrete against named dates. But a 3.10 is a description of position, not a prophecy, and the position decays if the dates fail. The entries I am watching are few and specific: first silicon at the fab against its December 2026 target, airport operations against September or October 2026, rail progress against 2030-31, and above all the first evidence of households arriving, the subject I keep separate in the essay on Dholera's own zero.

The honest summary fits in three sentences. NEOM demonstrates that no quantity of money converts an announcement into a demand anchor, and my index prices that demonstration at 0.90. Dholera demonstrates that statute, corridors and one contracted factory can hold eighth place on an eighteen-city board without a single new resident, which is both an achievement and a warning. The remaining distance, for both of them, is people, and on that subject the only defensible render is a blank page: the wider evidence base for that claim is in the seventeen-city survey.

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