Mapletree Industrial Trust: A 7.1% Yield, or a Warning Sign?
MIT trades at S$1.77 on a 7.13% yield and 1.07x book, but DPU has now fallen two years running. We weigh the payout against the risks.
Mapletree Industrial Trust (SGX: ME8U) closed at S$1.77 on 5 November, up 2.31% on a session where 14.7 million units changed hands, about 1.5 times its 20-day average of 9.6 million. That is a lively day for a REIT that has had a hard year.
Zoom out and the picture is less cheerful. The units sit 2.91% above their 52-week low of S$1.72 and 16.11% below the 52-week high of S$2.11. Year to date they are down 14.49%, while the Straits Times Index is up 13.75%. Price is below both the 50-day moving average of S$1.846 and the 200-day of S$1.9426, and the 14-day RSI of 46.2 is neither here nor there.
What caused the bounce on 5 November? I could not find a company-specific reason. The most recent filing was a routine issuance of 869,429 new units to the manager to settle base fees, which is not a price driver. The units also went ex-distribution on 3 November for the 2.97 cent second-quarter payout due on 9 December, and buying after an ex-date is common when a stock is sitting near a 52-week low. Treat the move as trading noise rather than news.
What the trust owns and how it earns
MIT is an industrial REIT sponsored by Mapletree Investments. Its Singapore portfolio covers flatted factories, stack-up and ramp-up buildings, business parks and hi-tech space, including Mapletree Hi-Tech Park @ Kallang Way. Beyond Singapore it owns data centres in the United States, a portfolio of 29 facilities picked up in 2021, plus a data centre in Osaka and other assets in Japan.
The money comes from rent. Gross revenue feeds net property income after property expenses, then the trust deducts borrowing costs, manager’s fees and trust expenses to arrive at distributable income, which is paid out quarterly. Two things matter most for unit holders: how much of that rental income stays in the door, and how much of it goes to lenders.
The latest numbers point one way
Results for the quarter ended 30 September 2026, released on 27 October, showed gross revenue of S$167.0 million, down 1.9% year on year, and net property income of S$119.8 million, down 3.5%. Distribution per unit came in at 2.97 cents, down 6.6%. For the half year, revenue fell 4.8% to S$329.3 million, net property income fell 6.1% to S$242.0 million and DPU fell 5.7% to 6.08 cents.
There are three drivers, and only one is about operations.
First, occupancy. Portfolio occupancy slipped to 89.7% from 90.7% the previous quarter. The data centre segment fell to 82.1% from 85%, and the North American data centre portfolio dropped to 79.1%. The trigger was the non-renewal at 7337 Trade Street in San Diego, which shows up in full this quarter.
Second, costs. Borrowing costs rose 9.1% quarter on quarter to S$20.2 million from S$18.5 million, as interest rate swaps locked in at lower rates matured and repriced higher. Management’s average cost of debt was 3.3% for the quarter and it guides to 3.5% to 3.6% for FY26/27, then just under 4% in FY27/28 if swaps reprice at current market rates.
Third, recycling. Divestments shrink the income base, at least in the short term. On 29 September the trust agreed to sell 3255 Neil Armstrong Boulevard in Eagan, Minnesota to its tenant DataBank for US$9.81 million, triggered by the tenant’s purchase option. That is 9% above the 2021 purchase price but 10% below the March 2026 valuation of US$10.9 million, and the asset contributed about 0.2% of gross revenue in FY25/26.
Is the payout covered?
This is where the debate lives, so let’s use only verified figures.
| Metric | Figure | | Price (5 Nov 2026) | S$1.77 | | Distribution yield | 7.13% | | P/B | 1.07 | | Aggregate leverage (30 Sep 2026) | 37.7% | | Portfolio occupancy | 89.7% | | 2QFY26/27 DPU | 2.97 cents (-6.6% yoy) | | 1HFY26/27 DPU | 6.08 cents (-5.7% yoy) |
Gearing is the first check, and 37.7% is comfortable. It is up from 34.0% at 31 March 2026, but still well below the 50% regulatory ceiling that applies to Singapore REITs. The trust also says about 90.8% of the amount available for distribution over the next 12 months is hedged or derived in Singapore dollars, so currency is not an unmanaged risk.
Interest cover is the second check. For FY25/26, MIT reported interest coverage of about 4.0 times. That is a reasonable buffer, and it is why the DPU decline has been gradual rather than abrupt. But the direction of travel is clear: refinancing costs are rising, and the guidance of just under 4% for FY27/28 means the drag gets worse before it gets better.
The third check is the shape of the payout history. FY25/26 DPU was 12.71 cents, down 6.3%, and the first half of FY26/27 is down another 5.7%. Two consecutive years of decline is not a distressed REIT, but it is a REIT whose distribution is being squeezed by forces outside its control.
Note also that FY25/26 distributions of 12.71 cents were well above earnings per unit of 7.40 cents. That gap came largely from a S$131.5 million fair value loss on investment properties, mostly on North American assets. A non-cash write-down does not stop a distribution, so this is not a red flag by itself, but it explains why the headline P/E of 0.29 from our data source is meaningless. Fast-growing P/E figures on REITs rarely are; use P/B of 1.07 and the yield instead. Annualising the first-half DPU gives about 12.2 cents, roughly 6.9% at S$1.77, a bit below the 7.13% trailing figure, which reflects the FY25/26 payout.
The bull case
Singapore is doing the heavy lifting. Rental reversions on renewals hit 6.9% in the quarter, up from 5.3%, and that is positive across every Singapore segment. Kallang Way is filling up, with committed occupancy expected to exceed 75%. That means leases rolling over are being signed at higher rents even as the trust shrinks elsewhere.
The data centre void is being filled with long leases. After the quarter closed, MIT signed new leases at Alpharetta (15.5 years) and San Jose (5.3 years), starting October 2026, with annual escalations. The North American portfolio’s weighted average lease term is 6.7 years, which reduces near-term vacancy risk.
And the balance sheet has room. Gearing of 37.7% gives debt headroom, and management is targeting S$500 million to S$600 million of divestments by the end of FY26/27, according to coverage of the results call. Proceeds go to paying down debt and funding the reallocation towards data centres, which is the stated strategy.
At 1.07 times book, investors are paying close to the appraised value of the assets while collecting a 7% plus yield.
The bear case
The North American data centre book is the soft spot. Occupancy there is 79.1%, and the new Alpharetta and San Jose leases were signed at lower starting rents than the ones they replace, according to coverage of the results call. Coverage of the reits, and be honest about the more you lease, the lower the revenue base.
Cost is the second problem. Swap repricing is contractual and predictable, which is worse in a way, because management has already told you the cost of debt goes to just under 4% next financial year. Every 25 basis point rise in the cost of debt eats into a distribution that already yields more than 7%.
Third, divesting at 10% below valuation, as with the Minnesota asset, tells you the bid for older North American data centres sits under carrying values. More such sales mean more write-downs and more earnings dilution before the money is redeployed.
Units are also issued to the manager quarterly, which is a slow dilution that unit holders bear.
The bottom line
MIT is not a broken REIT, but it is a shrinking one, and the 7.13% yield is the market’s price for that. The income stream is real: Singapore rents are rising, leases are long, gearing is modest and just under 91% of near-term income is hedged or in Singapore dollars. The problem is that the two forces cutting into DPU, North American vacancy and swap repricing, are both still at work.
Rather than asking whether the yield is high or low, watch three things over the next two to three quarters. First, whether North American data centre occupancy climbs from 79.1% now that the new leases have begun. Second, whether DPU stops falling in FY27/28 once the swap repricing peaks, or whether falling rents offset the higher Singapore reversions. Third, whether divestments keep coming in below book value and how quickly the proceeds reduce borrowing costs. If the first two improve, the 7% plus yield at 1.07 times book would be better supported by a stabilising distribution. If they do not, the yield is doing exactly what it should: compensating for a distribution that is still drifting lower.


