Singapore industrial rents have now risen for twenty-three consecutive quarters and the rental index sits at its highest level since 1996. On any normal reading that is a seller’s market. Yet strata industrial sales in the first quarter of 2026 fell to their lowest quarterly volume since 2020. Both things are true at once, and the reason they are true is the single most important thing to understand before buying an industrial unit in Singapore: the asset and the lease on it are two different investments, and only one of them is going up.
113.8
JTC rental index in 2Q2026 — the highest reading since 2Q1996
23
consecutive quarters of rental increases since the 3Q2020 trough
335
strata industrial units transacted in 1Q2026, the lowest quarter since 2020
The index that will not stop climbing
Start with what the market is actually doing, because the headline conditions are unusually clean. The JTC All Industrial rental index rose 0.5% quarter-on-quarter in 2Q2026, following 0.4% in 1Q2026. That is the twenty-third consecutive quarterly increase since the pandemic trough in 3Q2020, and cumulatively rents are up 27.2% over that run. The price index moved 0.6% in the quarter and 3.8% over the year. There is no ambiguity in the direction.
JTC All Industrial rental index
Quarterly, base Q4 2012 = 100
JTC quarterly market report, 2Q2026. Index rebased to Q4 2012 = 100.
What makes the run unusual is how even it is. Every quarter in the series above adds between roughly 0.4% and 0.7%. There is no spike to point at and no correction to wait out. That pattern is what a genuine supply constraint looks like rather than a speculative one: rents are being pushed by occupiers who need the space to operate, not by investors bidding each other up. The distinction matters when you get to the exit, because occupier-driven demand supports rent but does very little for resale price.
The segments beneath the headline do not move together. In 2Q2026 single-user factory rents rose 0.7%, multi-user factories 0.6%, warehouses 0.5% — and business park rents fell 0.1%. On the price side, single-user factory prices rose 1.1% against 0.4% for multi-user. The gap between the strongest and weakest sub-market in a single quarter is wider than the headline number itself, which is the first sign that “industrial is up” is not a usable statement for anybody actually writing a cheque.
The market data is the easy part
The lease is where the money is actually made or lost
You’ve seen where rents, tenure and demand are moving. The full report reveals what those numbers miss: how financing, lease decay and real ownership costs change the investment.
The financing cliff — the remaining-lease thresholds at which banks reduce, then withdraw, and what that does to your exit.
The amortisation drag that a gross yield never shows, with the full cost stack between gross and net.
A worked $800,000 purchase costed line by line, plus a live calculator for your own unit.
Open the full report
Enter your details for instant access — about 30 seconds.
You’re in.
The rest of the report and the calculator are open below.
Instant accessNo account requiredRelevant follow-up only
Where the money is quietly moving
Here is the finding that reframes the whole market. Split the 1Q2026 strata transactions by tenure and the average price of a 30-year leasehold unit fell 0.6% quarter-on-quarter, a 60-year leasehold unit rose 1.4%, and a freehold unit rose 2.9%. In a rising rental market, the shortest-lease assets were the only ones losing value.
Averages across caveated strata industrial transactions in 1Q2026. Sample sizes differ by tenure band and averages are sensitive to mix.
A freehold unit trades at roughly 2.5 times the price per square foot of a 30-year unit that may sit in the same estate and let for a similar rent. That spread is not about the building. It is the market pricing the number of years you are allowed to own it — and in a cautious operating environment, buyers have been paying up for the years rather than for the yield. Understanding why is most of the work of buying industrial well, and it is where the rest of this report goes.
Occupancy tells you which segment is real
Rent tells you what the last tenant paid. Occupancy tells you whether there will be a next one. In 2Q2026 multi-user factories ran at 90.5% occupancy, up 0.3 points; warehouses at 89.4%, flat; single-user factories at 89.3%, up 0.1 points. Business parks sat at 77.9% — improved from 76.7% but still more than twelve points below every other segment.
Occupancy rate by industrial segment
2Q2026
JTC quarterly market report, 2Q2026.
One in five business park square feet is empty, and that segment is also the only one where rents fell this quarter. Vacancy and falling rent arriving together is the combination that usually precedes a price adjustment, and it is a reasonable read that the hybrid office-lab product has structurally lost demand to conventional offices that de-rated after 2020. The multi-user factory segment — which is where nearly all strata units for individual buyers sit — is the tightest of the four. That is the good news in this report, and it comes with the large caveat in the next section.
The financing cliff
Every leasehold industrial unit has a date after which most banks stop lending against it. That date is not the end of the lease — it arrives roughly two to three decades earlier, and it is the single mechanical fact that explains the tenure spread on the previous page.
What lenders will do, by remaining lease
Remaining lease
Typical loan tenure
Typical LTV
What it means in practice
60 years or more
25–30 years
70–80%
Financing is routine
40–60 years
25–30 years
70–80%
Most lenders still comfortable
30–40 years
20–30 years
60–70%
The pool of lenders starts thinning
25–30 years
10–20 years
Reduced
Cash-flow stress becomes the binding test
Under 25 years
Minimal or none
—
Cash purchase only
mastplan
Indicative lender behaviour rather than a published rule. Individual banks set their own policy and will underwrite the borrower and the property as well as the lease.
Read that table from the seller’s side rather than the buyer’s and it changes shape. If you buy a 30-year leasehold unit today with twenty-eight years left and hold it for eight years, you are trying to sell an asset with twenty years remaining. At twenty years the buyer pool is no longer “anybody who wants an industrial unit” — it is “anybody who wants an industrial unit and has the entire purchase price in cash”. That is a fraction of the original market, and a fraction of a market is what a discount is made of.
This is why post-2012 government industrial land sales matter so much. JTC capped new industrial land sales at 30-year leases from that point, against the 60-year leases that were standard for the strata stock built through the 1980s and 1990s. Every 30-year unit sold after 2012 is therefore already inside the second half of its financeable life, and every year that passes moves it closer to the cash-only band rather than further away. The 60-year stock, by contrast, still has decades of ordinary financing ahead of it. The 1.4% versus −0.6% divergence in the tenure table is not a market mood. It is a countdown.
What the lease is actually worth as it shortens
Leasehold value does not decline in a straight line. It holds up well for decades, then falls away sharply once the financeable window closes. The shape below is the conventional valuation view of leasehold value expressed as a percentage of an equivalent freehold — the same logic professional valuers apply, and close to what transaction evidence shows.
Leasehold value as a share of freehold value
By years of lease remaining
Share of freehold value
Conventional leasehold valuation curve, indicative. Illustrative model — actual discounts vary by property, estate and demand.
The important feature is not the level at any point — it is where the slope changes. From ninety years down to about fifty, the curve costs you roughly ten percentage points across four decades: slow, tolerable, easily outrun by rental income. From forty years down to twenty, it costs more than thirty points across two decades. The cliff and the curve are the same event seen from two angles. Value falls off precisely where financing does, because financing is what creates the buyer pool that sets the price.
Apply that to the two live products. A 60-year leasehold unit bought today with, say, forty-five years left sits on the flat part of the curve for the whole of a normal ten-year hold. A 30-year unit bought today with twenty-eight left spends that same decade travelling the steepest section there is. Identical rent, identical building, opposite capital outcomes — and the gross yield quoted in the listing is blind to all of it.
The number a gross yield hides
Industrial is marketed on yield because the yield looks good next to residential. B1 light industrial units generally quote gross yields around 5.5% to 6.8%, with B2 a little higher. Residential rarely clears 3.5%. If the comparison stopped there, industrial would be the obvious trade. It does not stop there, for two reasons that compound.
Where a gross industrial yield actually goes
Indicative, on a leased strata unit
Net to ownerTax, maintenance, agency, vacancy
Illustrative model on quoted gross yield ranges. Costs include property tax at 10% of annual value, maintenance and sinking fund, management and agency, and a vacancy allowance.
The first reason is the ordinary one: roughly a third of the gross rent never reaches you. Non-residential property tax runs at 10% of annual value with none of the owner-occupier concessions residential enjoys, maintenance and sinking fund contributions on strata industrial are meaningful, and industrial vacancy between tenants is measured in months rather than weeks because the pool of businesses that need your exact specification is small. A 6% gross becomes something in the low fours.
The second reason is the one almost nobody prices. On a leasehold asset, part of what you collect as rent is not income at all — it is the lease being consumed. Straight-line that over the remaining term and a unit with thirty years left is writing down roughly 3.3% of its own value every year. Set that against a 4.2% net yield and the genuine return on capital is under one per cent before you have accounted for a single vacancy. On a sixty-year unit the same drag is 1.7%, and on freehold it is zero. This is the arithmetic behind the tenure spread, and it is why the same rent supports wildly different prices.
Amortisation drag by remaining lease
Remaining lease
Annual lease consumption
Net yield of 4.2% becomes
Freehold
0.0%
4.2%
60 years
1.7%
2.5%
40 years
2.5%
1.7%
30 years
3.3%
0.9%
20 years
5.0%
−0.8%
mastplan
Illustrative model, straight-line amortisation of the purchase price over the remaining lease. Real-world decay is not linear — see the curve above — but the direction and rough magnitude hold. Not a forecast or a projection of returns.
Price per square foot is mostly a map
Specification varies less than location does
Industrial pricing in Singapore is dominated by zoning and distance from the centre, not by building quality. B1 sits where light, clean operations can coexist with housing — Ubi, Kaki Bukit, Kallang, Tai Seng, generally within reach of an MRT station. B2 is deliberately pushed to Jurong, Tuas, Woodlands and Senoko, where heavier processes have buffer distance from residential areas. The price gap between the two is roughly a factor of three.
Indicative strata price by estate
2026, per square foot, mid-range
Indicative 2026 ranges compiled from market guides; individual transactions vary widely with tenure, floor, size and specification.
Two things follow from that map. First, the psf you pay in a B1 inner estate is buying accessibility to staff and customers, which is what keeps those units let — B1 inner-estate capital appreciation over the past five years has run in the high thirties to mid forties per cent against roughly twenty-five to thirty-five per cent for B2 outer estates. Second, that same accessibility is why the B1 gross yield (4.5% to 6%) sits below the B2 gross yield (5.5% to 7%). You are being paid more, in yield terms, to own the harder asset in the harder location. Whether that trade is worth taking depends entirely on how long you intend to hold and how confident you are of re-letting.
An $800,000 unit, costed line by line
Abstractions are where these decisions go wrong. Here is the same purchase — a 1,000 square foot B1 multi-user strata unit at $800 psf, letting at $2.10 psf a month — run twice, once with sixty years of lease remaining and once with twenty-eight. The rent is identical. Everything that matters is not.
Getting in
Line
60-year lease
28-year lease
Purchase price
$800,000
$800,000
Loan available
$640,000 (80%)
$440,000 (55%)
Cash required at completion
$160,000
$360,000
Buyer’s Stamp Duty
~$18,600
~$18,600
Legal, valuation, incidentals
~$6,000
~$6,000
Total cash to complete
~$184,600
~$384,600
mastplan
Illustrative. Buyer’s Stamp Duty on non-residential property; no Additional Buyer’s Stamp Duty applies to industrial. CPF cannot be used for industrial property, so every dollar of the cash line is genuinely cash. GST at the prevailing rate may apply where the seller is GST-registered and is recoverable only if you are too.
The first row that should stop you is the cash line. The short-lease unit is not cheaper to buy — it is cheaper per square foot and more than twice as expensive in cash, because the lease band determines the loan-to-value. Buyers who shop on psf routinely discover this at the point of loan approval rather than before it, and by then the option cost has already been paid in time and in deposit.
Holding it for a year
Line
60-year lease
28-year lease
Gross rent (1,000 sqft at $2.10 psf)
$25,200
$25,200
Property tax, maintenance, agency, vacancy
−$7,600
−$7,600
Net rent
$17,600
$17,600
Gross yield on price
3.15%
3.15%
Net yield on price
2.20%
2.20%
Annual lease consumption
−$13,300
−$28,600
Net of lease consumption
+$4,300
−$11,000
mastplan
Illustrative model. Lease consumption is the straight-line write-down of the purchase price over the remaining term; it is an accounting view of decay rather than a cash cost, and real decay follows the curve shown earlier rather than a straight line. Not a forecast.
At $800 psf this particular unit is expensive for its rent — a 3.15% gross yield is well below the 5.5–6.8% band the segment usually quotes, which tells you the asking price is priced for capital rather than for income. Run the same building at $550 psf and the gross yield lands at 4.6%; at $450 it clears 5.6%. The point of the two columns is not the absolute return, which will differ for every unit. It is that the entry price and the remaining lease are the two variables that decide the outcome, and the rent — the number every listing leads with — is the same in both columns.
The rule that decides what you may do with it
Industrial property in Singapore is regulated as productive space, not as an investment vehicle, and that regulation reaches into how you may use what you own. The core condition on JTC-derived industrial space is an occupier requirement: at least 60% of the gross floor area must be occupied by the owner or a qualifying anchor tenant for an approved industrial use, leaving only 30% to 40% available to sublet.
For an owner-occupier running a business, this is invisible — you are the anchor. For a passive investor it is the whole design of the deal. You cannot carve a unit into six small tenancies and run it as a yield product. You need one substantial tenant whose activity qualifies as industrial, and you carry the concentration risk that comes with a single tenant: when they leave, the whole income stops at once, not a sixth of it.
A second condition shapes the space itself. Both B1 and B2 permit up to 40% of gross floor area for ancillary office use, with the balance required to serve genuine industrial functions. Units marketed with a large fitted office component are the ones to check against this, because a tenant who wants mostly office space is a tenant you may not be permitted to house — and a unit fitted out beyond the ancillary allowance can be an enforcement problem you inherit at completion rather than a bonus you bought.
What applies to industrial that does not apply to residential
Item
Position on industrial property
Additional Buyer’s Stamp Duty
Does not apply, at any count of properties
CPF
Cannot be used — cash and bank financing only
Seller’s Stamp Duty
15% in year 1, 10% in year 2, 5% in year 3, nil thereafter
Property tax
10% of annual value, no owner-occupier concession
GST
Chargeable where the seller is GST-registered; recoverable only if you are
Occupier requirement
At least 60% of GFA to owner or qualifying anchor tenant
Ancillary office
Capped at 40% of gross floor area
mastplan
Positions stated as at publication and subject to change. JTC and URA conditions vary by site, lease and estate — always check the specific title and lease documents.
What is being built, and where it lands
Roughly 4.4 million square feet of new industrial space is scheduled for completion in the second half of 2026, equivalent to about 0.7% of existing stock. That is a modest number in aggregate, and if it were spread evenly it would barely register. It is not spread evenly.
2H2026 industrial completions by type
Million square feet
JTC quarterly market report, 2Q2026. Shares of 53%, 46.9% and 0.1% of 4.4M sq ft.
Single-user factories take 53% and warehouses 46.9%. Multi-user factories — the segment that contains essentially every strata unit an individual can buy — take 0.1%. Four hundred thousand square feet of new supply would be a rounding error; four thousand is effectively nothing.
That is the strongest structural argument in favour of the multi-user segment and it should be weighed honestly rather than repeated as a slogan. Near-zero new supply into a segment already running at 90.5% occupancy is a genuine constraint on rent for the next several years. What it is not is an argument about price, because the multi-user stock that exists is overwhelmingly the 30- and 60-year leasehold stock whose value curve you have already seen. Scarcity supports the rent. The lease still governs the capital. Both remain true.
Forecasts for the year sit in a narrow band and none of them are dramatic: rental growth of roughly 1% to 3% across the index, capital values up 3% to 5%, with multi-user factories and business parks projected at 0% to 2% rental growth and warehouse and logistics at 0% to 1%. The 2025 outturn was 5.0% on prices and 2.4% on rents. Nobody credible is forecasting a break in either direction.
Where this asset class disappoints people
Industrial is sold on the absence of Additional Buyer’s Stamp Duty and a headline yield that beats residential. Both are real. Neither is the reason most industrial purchases underperform, and the reasons that do are consistent enough to list.
It is not a liquid asset, and the illiquidity is structuralThree hundred and thirty-five strata units changed hands across all of Singapore in 1Q2026 — the entire market, every estate, every tenure. Residential resale runs roughly ten times that in a quarter. When you want out, there may be a handful of active buyers for your specification, and marketing periods of six to twelve months are ordinary rather than a sign that something is wrong.
It is not a CPF purchase, so the cash requirement is absoluteEvery dollar of downpayment, stamp duty and legal cost on an industrial unit is cash out of a bank account. On the short-lease unit in the worked example that was $384,600 before a single month of rent arrived. Buyers who mentally price industrial against a residential downpayment consistently underestimate it by a factor of two.
It is not passive, because the occupier rule makes it a single-tenant assetThe 60% anchor requirement means one tenant carries your income. Finding a replacement is a specification-matching exercise — ceiling height, floor loading, power supply, permitted use — and not a matter of listing and waiting. Vacancy is measured in months.
A short lease is not a discount, it is a shorter runwayThe gap between $353 and $876 psf is priced, not free. Buying at the cheap end means buying the section of the value curve that falls fastest and selling into the buyer pool that shrinks every year. The unit is cheap because the exit is hard, and the exit gets harder for as long as you own it.
Business park exposure is not a defensive positionIt reads as the premium segment because it commands the highest rents, but it is running at 77.9% occupancy with rents falling while every other segment tightens. Whatever demand hybrid office-lab space had before 2020 has not fully returned, and rent with vacancy against it is not the same asset as rent with a waiting list behind it.
How to actually run the decision
The sequence below is deliberately ordered so that the cheapest checks eliminate the most candidates. Almost every expensive industrial mistake is made by doing these in the wrong order — falling for a unit, then discovering the lease band, then discovering the loan.
1
Decide whether you are an occupier or an investor, and answer honestlyThis determines everything downstream. An occupier is buying control of premises and escaping rent escalation; the 60% rule is free and the lease matters mainly because it must outlast the business plan. An investor is buying a single-tenant income stream on a depreciating right, and needs the lease to outlast the hold plus the next buyer's financing window. They are different assets and they should never be shortlisted together.
2
Set the lease floor before you look at a single listingTake your intended hold period, add the fifteen to twenty years your eventual buyer will need in order to obtain financing, and that sum is your minimum remaining lease. A ten-year hold implies roughly forty years remaining at purchase. Anything shorter is a decision to sell into a cash-only market, which can be the right call, but should be made deliberately and priced accordingly.
3
Get the financing position in writing before you shortlistTake the lease floor from the previous step to a lender and establish the actual LTV and tenure available to you on that band, on your income and your existing commitments. The difference between 80% and 55% on an $800,000 unit is $200,000 of cash. Discovering it after an option fee is paid is the most common and most avoidable loss in this market.
4
Underwrite on net yield after lease consumption, never on grossTake the asking rent, strip roughly 30% for property tax, maintenance, agency and a vacancy allowance, then subtract the annual lease write-down. If the resulting number is negative, the unit is a capital bet with a rental subsidy rather than an income asset, and it needs to be justified as a capital bet — on estate, specification and scarcity — not on its yield.
5
Check permitted use and the office component against the actual fit-outConfirm the URA zoning, the JTC or lessor conditions on the specific title, and whether the existing office fit-out sits inside the 40% ancillary allowance. Verify that the tenant you have in mind qualifies as an industrial user for that zone. An unauthorised fit-out or a non-qualifying tenant becomes your problem at completion, not the seller's.
6
Test the exit before you test the entryPull the actual transaction history for the estate over the past three years. Count the deals, not the price. If fewer than a handful of comparable units have changed hands, you have your answer on liquidity regardless of how the yield looks, and you should price the illiquidity into the offer rather than discover it at the exit.
Run your own unit
Your unit, not the index
Price the lease, not just the rent
Enter the asking price, the size, the rent and — most importantly — the years remaining. The calculator strips the cost load out of the gross yield, then subtracts what the lease itself is consuming, and shows you which financing band your eventual buyer will be shopping in.
Net yield after lease consumption
—
—
Price per square foot—
Gross yield—
Net yield (after ~30% cost load)—
Annual lease consumption—
Indicative loan-to-value—
Indicative loan tenure—
Cash required to complete—
Monthly repayment (at 4.0%)—
Monthly net cash flow—
Years until the unit reaches the cash-only band (25 years left): —
—
Indicative only, and a guide rather than financial advice. The cost load assumes roughly 30% of gross rent absorbed by property tax at 10% of annual value, maintenance and sinking fund, management and agency, and a vacancy allowance. Lease consumption is a straight-line write-down of the purchase price over the remaining term and is an accounting view of decay rather than a cash cost. Loan-to-value and tenure are indicative lender behaviour, not an offer — every bank sets its own policy and underwrites the borrower as well as the property. CPF cannot be used for industrial property. Stamp duty is Buyer’s Stamp Duty on non-residential property; GST may apply separately where the seller is GST-registered.
What the 2026 data actually supports
The rental story is real and the price story is conditional. Twenty-three quarters of increases and a 1996-high index are not in dispute. What that rent is worth to you depends entirely on the term of the right you bought it with.
Tenure is now the market’s main dividing line. In one quarter, 30-year units fell 0.6% while freehold rose 2.9%. The building did not change; the countdown did.
Financing sets the exit, and the exit sets the price. Below roughly 30 years remaining lenders retreat and below 25 they largely stop, which collapses the buyer pool long before the lease runs out.
A gross yield that ignores amortisation is not a yield. Six per cent gross on a 30-year lease is under one per cent once the cost load and the lease write-down are taken out.
Multi-user scarcity is the strongest structural argument in the market — 0.1% of 2H2026 completions into a segment at 90.5% occupancy — but it supports rent, not capital, and it does not repeal the lease curve.
mastplan
Your unit, your lease, your numbers
Get an industrial read on the specific deal.
A run through the actual lease position and financing band on the unit you are looking at, the net yield after cost load and amortisation, the transaction depth in that estate, and what the exit realistically looks like at your intended hold period. No obligation — you will leave with a clear next step either way.
About the figuresRental and price indices, occupancy rates, segment breakdowns and the 2H2026 supply pipeline are drawn from JTC’s quarterly market report for 2Q2026. Strata transaction counts and average prices by tenure are from Savills Singapore’s 1Q2026 strata industrial review. Full-year 2025 price and rental growth and 2026 forecasts are from published market outlooks. Zoning definitions and the ancillary office allowance follow URA Master Plan classifications; occupier requirements follow JTC lease conditions; Buyer’s and Seller’s Stamp Duty rates and the non-residential property tax rate follow IRAS. Indicative price-psf ranges by estate, gross and net yield bands, lender loan-to-value and tenure behaviour, the leasehold value curve, the amortisation-drag table, the worked $800,000 example and the calculator are illustrative models compiled by mastREplan on the stated assumptions; they are not forecasts, valuations, loan offers or projections of returns, and actual outcomes vary widely and can be negative. Rules and rates are stated as at publication and are subject to change. Nothing here is financial advice — please check with a qualified professional before making any property decision. See our full Disclaimer.
Create your account or sign in
One account unlocks every gated resource. New members create an account here; returning members use the same form to sign in.
