Dholera plot loan: financing land that pays you nothing

The financing question almost always arrives late in the conversation, after the map has been folded away and the number has been said out loud. Can I take a loan for this? It gets asked casually, in the tone of somebody checking a formality, because most of us have watched a home loan work at close range and quietly assume land behaves the same way with a different address on the form. It does not. The gap between those two assumptions is where a great deal of avoidable financial pain lives, and in a market as young as this one the pain tends to arrive years after the decision that caused it.

Let me be honest about the limits of this essay before it starts. I am not going to hand you an interest rate, a loan to value ratio, a tenure, a margin requirement, or the name of a product at a named bank. Not out of coyness, but because none of it is reliably sourceable in a form I would put my name to, and lending terms shift with the lender, the borrower's profile, the classification of the land and the month you walk in. Anyone publishing a confident table of plot loan terms for Dholera is describing either one deal they heard about or an ambition they have. What I can do, and what turns out to be more useful anyway, is explain why land lending is structurally different from home lending, why leverage sits badly on this particular asset, and exactly what to put to a lender in writing so that the answer you act on came from the institution that will hold your paper rather than from the person selling you the plot.

A plot loan is not a home loan wearing different clothes

Start on the lender's side of the table, because everything else follows from it. When a bank lends against a completed home it is lending against something with three convenient properties. It can be valued against many nearby sales. It can be occupied. And if the borrower stops paying, it can be sold to an ordinary family who wants to live in it. Bare land offers none of the three. There is nothing to occupy. In a young market there are few genuinely comparable transactions to value against, which is the same data vacuum I keep running into whenever anyone asks me what a plot is worth. And the pool of people willing to buy a repossessed plot in a city still under construction is smaller and far more specialised than the pool who will buy a flat in a functioning suburb.

So lenders treat unconstructed land as its own category, with its own appetite, its own documentation and its own conditions. That is why the answer to "will a bank fund this" is genuinely lender by lender rather than a fact about Dholera that I could look up and report. Some institutions do not lend against bare plots at all. Some will, with conditions attached about construction timelines, about the land's classification, or about where exactly the parcel sits. Some will weigh your income and your existing assets far more heavily than the land itself. I cannot tell you which is which, and I would distrust any article that pretends to, particularly one written by somebody who benefits from your yes.

There is a second reason the two categories diverge, and buyers rarely think about it: the lender's lawyer is going to run the same title file I keep telling you to run yourself. The mother deed and the full chain of transfers after it, a thirty year Encumbrance Certificate, the 7/12 extract for the survey number, property tax receipts, registration executed in person at the sub registrar with biometric verification, and mutation so that the Khata carries your name. That is the standard Gujarat process, it does not bend for enthusiasm, and a bank has no emotional stake in your excitement about a semiconductor fab.

A refusal is free diligence

Which leads to the most underrated line in this essay. If a lender declines to fund a particular plot, that decision is information, and you just received it free of charge from a professional with money at stake and no commission riding on your purchase.

Sellers have a ready explanation for refusals. Banks are slow, banks do not understand emerging markets, everybody here buys with cash anyway. Sometimes that is even partly true, because thin markets and unusual land classes really do fall outside standard lending boxes. But sometimes a refusal means the title chain has a gap in it, or the land's classification is not what the brochure implies, or the scheme's regulatory position is murky, or the parcel cannot be located confidently against the sanctioned layout. You do not get told which. What you get is a no, and the correct response is to find out why rather than to route around the no with a friendlier source of money.

The same instinct applies to any scheme that gently discourages you from involving a bank at all. Nobody is obliged to make your financing convenient, and small transactions in young markets often do run on cash. Still, an argument that reduces to paperwork being unnecessary sits badly beside the GUJRERA check, which takes minutes, and beside the plain rule that a marketed project including a plotted development must register, with Gujarat exempting some plot only schemes in a way that shifts the entire burden onto title work. Where a regulator's file is absent, a lender's file is one of the very few independent reads left on the asset. Refusing both at once is how people end up owning a dispute instead of a plot.

The asset pays you nothing while the loan charges you monthly

Now the structural heart of it. A house bought with borrowed money can be rented, so a part of the loan services itself. A business bought with borrowed money produces revenue. A plot of land produces nothing at all. No rent, no dividend, no coupon, no yield of any description. It sits there, it accrues property tax and attention, and it waits.

Which means every rupee of interest on a plot loan comes out of your income, every single month, for as long as you hold the loan and the land together. That is not an argument against land as an asset. Unbuilt land inside a corridor that is genuinely being built has an honest case, and I have made parts of that case elsewhere. It is a description of what leverage does here. Borrowing amplifies whatever the underlying asset does, and while the underlying asset does nothing whatsoever in cash terms, the amplification runs in one direction only: against you, on a schedule, in money you have to earn somewhere else.

Compare that with the mental model most first time buyers import from the home loan world, where the borrower moves in and stops paying rent, so the loan replaces an existing outflow. There is no rent being replaced here. The loan is a pure addition to your monthly obligations, secured against something that cannot pay a rupee towards it until you sell. Anybody who tells you the appreciation will take care of the interest is making two forecasts at once, one about the market and one about your income, and I would not endorse either. On the returns side I have set out at length why no honest return figure exists for this market, and a projection that fails on its own is not improved by borrowing against it.

The horizon problem, read off a calendar

The other half of the leverage problem is time, and time here is not a matter of opinion. It is printed on public documents, so you can read your own horizon off a calendar rather than off a sales deck.

Some things have already happened. The Ahmedabad to Dholera expressway, roughly 109 km of greenfield access controlled road, was reported inaugurated on 31 March 2026 and operational, cutting the journey from about two hours to somewhere between 40 and 60 minutes depending on which account you read. Trunk infrastructure works inside the Activation Area, about 22.5 sq km within TP2, were recorded complete in the NICDC Delivery Monitoring Unit report to DPIIT dated 30 June 2026. Those are facts on the ground today, which also means they already sit inside whatever price you are being quoted.

Almost everything else that would transform the place is dated in the future and carries a tier. The airport is reported at roughly 80 percent completion with operations targeted for September or October 2026, after an AAI trial and calibration landing on 4 June 2026, and this project has been slipping its dates since about 2010, so a target is exactly that. The Tata and PSMC fab, approved by Cabinet on 29 February 2024 with an investment of Rs 91,000 crore, has first silicon targeted around December 2026 and commercial production reported for mid 2028, and no chip has been produced there yet. The semi high speed rail line was approved by CCEA on 13 May 2026, Rs 20,667 crore for about 134 km, with completion targeted up to 2030-31. The sanctioned plan's own Phase II window, covering TP3 and TP4 across 126 sq km, runs to 2032, and Phase III runs from 2033 to 2042.

Line those dates up and the honest holding period for a Dholera plot is measured in years, quite plausibly the better part of a decade, and the deepest transformation the brochures describe sits at the far end of that. Now put a loan across the same span. You are not simply betting that Dholera delivers. You are betting that your income remains stable, your job or your business survives whatever the next several years contain, and your appetite for a monthly outflow against a silent asset holds up through at least one stretch where nothing visible happens on the site. Most people underestimate that last one. Construction timelines have long quiet periods, and a quiet year is much harder to sit through when a payment leaves your account every month.

The costs a loan does not make disappear

There is an arithmetic layer underneath the loan that borrowed money tends to hide rather than solve. Gujarat's effective stamp duty is 4.9 percent, made up of 3.5 percent basic duty plus a 1.4 percent surcharge, and the registration fee adds 1 percent on top. That is Rs 5.90 for every Rs 100 of value the registrar recognises, paid once, at the front, and never recovered. There is a documented registration fee waiver where the property is registered in a woman's sole name. A further 1 percent stamp concession is claimed in places and I have not been able to corroborate it, so treat it as unverified until the sub registrar confirms it for your document. I have worked the whole ledger through in the transaction cost essay, and it is worth reading before you decide how much of your own cash the purchase actually needs.

Do not assume any of that can be folded into a loan. Whether a lender funds only the recognised value of the land, and how it treats duty, registration, brokerage and legal fees, is a term of that specific facility and has to be asked rather than guessed. The reason this matters is simple. Borrowing makes the headline number look reachable while the cash costs stay exactly where they were, and a buyer who has planned only for the plot price can find themselves short at the registrar's counter, which is the worst possible moment to be improvising.

The other cost is the one that never appears on any statement. Brokerage tends to show up on both sides of a transfer in this market, the land pays nothing while you hold it, and the duty is paid again by whoever buys from you. Every one of those frictions has to be cleared before a leveraged position is even level, which is another way of saying that borrowing to buy land shortens your patience precisely when the asset demands more of it.

What to put to a lender in writing

Here is the practical core of the essay. Everything in the table below is a question for the institution that will actually hold your paper, and every answer belongs in writing, on their letterhead or in an email from their address, before you pay a token to anybody. Verbal assurances from a seller's preferred contact at a bank are not terms. They are conversation.

Two habits make that table work. Ask more than one institution, because the answers will differ and the differences are themselves a map of how the asset is viewed. And keep the correspondence, because a lender's written assessment of the land is the closest thing to an independent second opinion that most buyers will ever obtain on a plot.

What leverage does to concentration risk

Dholera's investment case leans heavily on one anchor. The Tata and PSMC fab is the largest single commitment in the region by a wide margin, Rs 91,000 crore approved by Cabinet on 29 February 2024, with the Fiscal Support Agreement signed on 5 March 2025 citing Rs 91,526 crore. Civil work was reported at more than half complete by the middle of 2026 with cleanroom fit out underway. It is a serious project, visibly under construction, and I have no interest in talking it down.

But the region also carries a documented reminder of what an anchor looks like when it fails. The Vedanta and Foxconn joint venture, announced in 2022 with a headline value near 19.5 billion dollars, collapsed when Foxconn withdrew on 10 July 2023. That capital was never committed and should never be counted. If you had bought unleveraged land in 2022 on the strength of that announcement, you would have absorbed a disappointment and waited. If you had bought the same land with a loan, the announcement's collapse would have arrived as a monthly payment against an asset whose story had just changed, with no rent to soften it and a thin market to sell into.

That is the whole point about leverage and concentration. Unleveraged, a delay costs you time and opportunity. Leveraged, a delay costs you cash on a fixed schedule, and it removes your ability to simply wait, which in a market this young is the only real advantage a patient buyer has. The same applies to gentler slippage. Airport dates have moved repeatedly since about 2010. The old target of roughly 120,000 residents and 80,000 jobs by 2020 lapsed unmet. Land acquisition for the SIR was stayed by the Gujarat High Court in 2015 after farmer petitions, and a 2017 Business Standard report noted that only about 290 sq km of the 900 plus had then been secured. None of that history is fatal to the thesis. All of it is the sort of thing a borrower feels more sharply than an owner.

If you are buying from abroad

Non resident Indians and OCI holders may buy residential and commercial property, but not agricultural land. That single rule does more work in Dholera than anywhere else I write about, because agricultural land marketed as being in the SIR is one of the documented red flags here, and a parcel can sit inside the notified region while remaining agricultural in classification. If you are financing from outside India, the rule interacts with your loan as well as your purchase, because a lender's eligibility check will hinge on the same classification your ownership does.

I will not state cross border banking or remittance rules here, because I cannot source them to the standard the rest of this site holds. Put them to your own bank and to a qualified professional, in writing, and get the answers before you commit to a schedule of payments across time zones. The rest of the remote buying protocol, including the care a power of attorney deserves, sits in the NRI essay. The financing lesson is narrow and firm: never let a payment deadline set abroad outrun a verification that has to happen in Gujarat.

If you borrow anyway, borrow like this

Plenty of sensible people will read all of the above and still take a loan, and there is no rule that says they are wrong. So here is the discipline I would hold myself to, stated without a single number because the numbers belong to your situation and your lender.

Borrow only an amount you could keep servicing comfortably if the land were worth nothing for a decade, because for cash flow purposes it is worth nothing for a decade. Keep a reserve sized against the remaining schedule rather than against a few months, since the periods when nothing visible happens on site are exactly the periods when incomes wobble. Never borrow against a plot you have not verified yourself, since leverage magnifies a title defect the same way it magnifies everything else. Get every term in writing before any token money moves, because a booking deadline is a sales instrument and it should never be allowed to set your financing timeline. And decide in advance, on paper, what you will do if you cannot service the loan, since a plan made calmly beats a distress sale into a market with no reliable price discovery.

One more, and it is the one people skip: know where your plot sits before you finance it. The Activation Area is the map that matters, roughly 22.5 sq km inside TP2 where trunk works are recorded complete, and land far outside it is a longer bet by construction rather than by opinion. A longer bet financed on a shorter schedule is where the mismatch turns painful. And whichever way you fund it, verify the GUJRERA registration where it applies and satisfy yourself of clear, marketable title inside the SIR boundary before money moves.

The through line of everything above is that a loan does not change the asset. It changes you. It converts a patient position into a scheduled obligation, and patience is close to the only edge available in a market that has an operating expressway, a fab under construction, 2,779 people recorded in the 2011 census and a resident city still years from arriving. If you can buy this land without borrowing, the waiting costs you only opportunity. If you borrow, the waiting sends you a bill every month, and the bill does not care what stage the cleanroom is at. Ask your bank the six questions, read the sanction letter twice, and let the answers, rather than the brochure, decide how much of this you take on.

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