Is Keppel DC REIT A Good Buy In 2026? [Fundamental Analysis]

Daniel Lee
2 days ago
8 min read

In this article, we'll conduct a fundamental analysis and review of Keppel DC REIT and its suitability to achieve the following investment objective: To deliver a stable dividend yield of 5% to 6% per year while having high capital preservation ability.

Business Description

Keppel DC REIT is an Industrial REIT that was listed in 2014 and industrial properties across 9 countries that are mainly designed to be data centres.

What I Like About KDC:

Distribution per unit has been rising consistently (CAGR: 5~%) despite the increase in units thereby signifying the manager's capabilities to deliver yield accretive acquisitions
Overall portfolio’s WALE is high & occupancy rate is resilient (Figure 11)
Clear strategic focus on hyperscale assets and strong pipeline from sponsor that could contribute to future portfolio expansion.
Portfolio is well positioned to capture the increasing demand of hyperscalers data centres driven by the ongoing global arms race for compute and data storage

What I Do Not Like About KDC:

62.47% of the portfolio has an underlying land lease of less than 30 years (All the data centers in Singapore). This results in high levels of lease decay which are not priced in by the market.
High levels of tenancy concentration risk (Figure 12). This may result in severe DPU disruption in an event where their top tenants encounter any financial issues.

Updates From Recent Performance (FY 2025)

General Comments:

Secured a 10 year land tenure lease extension for Keppel DC SG 7 & 8 to 15 July 2050 for a total consideration of $350million.
Inclusion in the Straits Times Index from June 2025.
Reported DPU grew by 9.8% at the back of higher contributions from acquisitions (Gross revenue and NPI grew by 42% and 59%), higher income support and government grants, and lower finance cost (-6.91%) offset by the enlarged unit base.
The Guangdong DCs continue to bleed quietly. Carrying value of Guangdong DC 1 fell from S$130.8M to S$109.2M (-16.5% YoY); Guangdong DC 2 from S$130.8M to S$108.9M (-16.7%). Both at original purchase prices of S$141.3M each. The China data centres are now collectively ~S$60M below purchase price, with the Bluesea Intelligence overhang unresolved

Positive Headwinds:

DPU growth in FY2026 is likely going to be driven by the full year contributions of Tokyo DC3.

Negative Headwinds:

Higher energy prices may result in slightly higher operating expenses for those tenancy contracts where energy costs is absorbed by the REIT and not the tenants.

Portfolio Movements

1. Acquisitions

Property	Asset Type	Purchase Price	Effective Stake	NPI Yield /Cap Rate	Lease Structure	Funding Source
Tokyo Data Centre 3 (Tokyo DC 3)	Hyperscale, freehold, fully-fitted (single tenant), Tier-III equivalent	JPY 82.1bn /S$683.0M	98.47%	NPI yield ~4.0%	15-year master lease to a "leading global hyperscaler" with annual rent escalations	S$398.9M net PO proceeds (S$229.8M deployed here) + JPY-denominated debt (TMK bond)
KDC SGP 7 & 8 — remaining 0.51% economic interest	Already-owned colocation campus (now 100%)	S$6.6M (Memphis 1 call option)	Increased from 99.49% → 100%	N/A (incremental stake)	Continuing Keppel lease arrangement	Internal / debt
KDC SGP 7 & 8 — 10-year land tenure lease extension	Land premium paid to JTC for lease extension to 15 July 2050	S$350.0M	n/a	Defensive — protects WALE & terminal value	Extends underlying land tenure	New debt facilities
KDC SGP 3 — remaining 10.0% interest + KDC SGP 4 — remaining 1.0% interest	Singapore colocation, achieving 100% ownership	S$50.5M (aggregate)	SGP 3: 90% → 100%; SGP 4: 99% → 100%	n/a (incremental stake)	Existing Keppel lease structure	Unutilised PO proceeds (S$53.4M)

Acquisition-level commentary

Manager’s stated rationale (Tokyo DC 3) — Hyperscale focus in supply-constrained Tokyo (Inzai); reduces top-client concentration (45.3% → 42.1%); WALE extension (6.9 → 7.2 years); JPY natural hedge via JPY debt.

Independent check & balance:

Yield on cost discipline — questionable. A 4% NPI yield in Tokyo DC 3 sits at the lower end of what the manager could deploy capital into. The 4% NPI yield is below the all-in cost of debt (3.0%) by only ~100 bps — a thin spread once 30 bps of management fees are added. The acquisition is DPU-accretive only because of the unit-issuance arithmetic and JPY low cost-of-debt, not because the asset is cheap.

Strategic verdict (Tokyo DC 3) — defensible but priced for perfection. Inzai is a top-tier sub-market with structural tailwinds (AI workloads, hyperscaler self-build demand) but acquired at a 4% NPI yield on a freehold asset with a 15-year single-tenant lease. Concentration risk is now elevated: single hyperscaler, single asset, single rental escalation curve. Manager’s claim of “diversification” is true at the portfolio level but creates a new dependency at the asset level.

Strategic verdict (SGP 7 & 8 land extension) — necessary, not optional. Paying S$350M for a 10-year extension to 2050 was a defensive move; without it, the carrying value of those two assets (which together represent 36.7% of investment property fair value — S$1,551.8M of S$4,230M net assets) would have started compressing as the lease ran down toward 2040. The market valuation post-extension (S$1,540M combined) versus purchase + premium (S$1,730M) suggests the manager paid full retail for the extension — fair value uplift was modest.

2. Divestments

Property	Asset Type	Initial Purchase	Exit Price	Exit vs Original	Completion Date
Kelsterbach DC	Shell-and-core, single-tenant triple-net, built 1989	Announced Dec 2019: €81.8M (~S$125.4M); completed May 2020	€50.0M (S$70.6M)	Loss of ~€31.8M (-38.9% in EUR) / ~S$54.8M in SGD terms	24 Mar 2025
Basis Bay DC	Colocation, single-client, 40.2% occupancy	Originally S$42.9M (RM112.4M) — acquired Dec 2014 at IPO	Not yet completed (held for sale at S$17.1M / RM53.9M)	Loss of ~S$25.8M vs original (-60%)	Expected 1Q 2026
NetCo bonds & preference shares (M1 Network Pte Ltd)	Debt securities (NOT a real estate asset)	Subscribed June 2024 as part of Intellicentre Campus sale (Australia)	Held for sale at S$75.4M	Pending completion	Subject to regulatory conditions

Divestment-level commentary

Manager's stated rationale — "Strategically sharpen portfolio mix" toward hyperscale; redeploy ~S$0.2bn into higher-yielding investments.

Independent check & balance:

Kelsterbach was a forced sale, not opportunistic. AR2019 explicitly states Kelsterbach was fully leased on a triple-net basis until end-2025. The asset's master lease expired the year of sale. With a 1989-built shell-and-core building in Kelsterbach (not central Frankfurt) and Germany's punitive new sustainability regulations (PUE ≤1.3 by July 2026, 100% renewable by Jan 2027), re-leasing the asset at decent yields would have required substantial capex retrofits. The "28.2% premium to book" headline disguises a 39% capital loss versus original purchase price in EUR. The manager wrote the asset down 52% from S$125.4M (2019) to €39.0M (Dec 2024) before selling — that's where most of the destruction occurred; the exit merely crystallised it.

Proposed proceed usage — Net divestment proceeds of S$70.6M used to repay bank borrowings. No reinvestment yield uplift here; this is balance-sheet repair.

Basis Bay sale is also distressed. At RM53.7M, exit is ~52% below original purchase (RM112.4M, 2014). 40.2% occupancy as at end-2025 explains why. Eleven years of sub-scale operations in a non-core geography.

Strategic verdict — capital recycling at a loss. Both divestments are necessary sunset-asset exits; neither is "value crystallisation." The framing of these as "rebalancing" is generous — they are losses dressed up as strategy. For an income investor, the more honest read: management took 5–11 years to exit non-core assets, suffering meaningful capital impairment along the way.

Tax tail. Note: FY2025 distributable income included a S$10.83M adjustment for accounting gain on Kelsterbach divestment. The accounting "gain" (book to sale) was added back out of distributable income — it does not flow to DPU. Good capital discipline here.

Independent Market Review Analysis

Asset-by-asset performance vs market benchmark

Asset / Market	IMR Market Utilisation	REIT Asset Occupancy	Verdict
KDC SGP 1	95.2%	53.3%	Underperforming — 42 ppt gap. Even on a 95% market, this asset sits at half occupancy.
KDC SGP 2	95.2%	98.2%	In-line / slightly outperforming
KDC SGP 3	95.2%	100%	In-line
KDC SGP 4	95.2%	94.5%	In-line
KDC SGP 5	95.2%	100%	In-line; benefits from AEI
KDC SGP 7	95.2%	100%	In-line — but inorganic
KDC SGP 8	95.2%	100%	In-line; future at risk if SGP 8 conversion times poorly
DC1 (Singapore)	95.2%	100%	In-line, but short land tenure (18.6 yrs is the shortest in the SGP portfolio)
Gore Hill DC (Sydney)	92.5%	80.0%	Underperforming — 12 ppt gap in a tight market
Guangdong DC 1, 2, 3	79.0% (Guangzhou)	100% (master-leased)	Reported occupancy is misleading — IMR strips out master tenant; revenue is netted against loss allowances. Material credit exposure to Bluesea unchanged
Tokyo DC 1	92.3%	100%	In-line
Tokyo DC 3	92.3%	100%	In-line — but acquired at a yield that prices in this tightness
Basis Bay DC (Cyberjaya)	82.8%	40.2%	Underperforming in a market with supply CAGR of 26.5% and falling colocation prices. Exit confirmed appropriate
maincubes DC (Frankfurt)	96.1%	100%	In-line, fully-fitted single tenant lease provides protection but counterparty risk concentrated
KDC DUB 1	98.0%	96.5%	In-line
KDC DUB 2	98.0%	98.1%	In-line
Milan DC	94.4%	100%	In-line. But IMR is sponsor-friendly — Milan has lower hyperscale presence and slower permitting
Almere DC (NL)	93.8%	100%	In-line on small base
Amsterdam DC	91.9%	95.1%	In-line — but asset has been written down (€30M → €29.2M); 2035 moratorium should benefit incumbents but hasn't lifted value
Eindhoven DC	90.6% (zero growth)	100%	Marginally outperforming, future at risk — IMR explicitly says no market growth through 2030. Asset already written down (€37.2M → €26.8M)
Cardiff DC	99.7%	100%	Major disconnect — 99.7% market utilisation but Cardiff DC carried at £13.3M vs £34.0M purchase price (-61%). Asset-specific issue not market
GV7 DC (London)	92.2%	100%	Major disconnect — written down from £37.5M to £16.8M (-55%) despite tight London market
London DC	92.2%	100%	Written down from £57.0M to £48.8M (-14%) — milder than GV7

Supply risk map

Market	IMR Forward Supply Signal	REIT AUM Exposure	Risk Level for REIT
Singapore	Supply CAGR 20.1% (2025-29F); but utilisation rising to 98%	62.8%	Low near-term, Moderate by FY2027 when new CFA capacity comes online
Tokyo	Supply CAGR 11.0%; demand 12.2% (demand > supply)	14.4%	Low — power-constrained, defensive
Ireland (Dublin)	Supply CAGR ~7%; CRU constraints	5.6%	Low — extremely defensive; incumbent advantage
Netherlands (combined)	Mixed: Amsterdam 2035 moratorium positive; Eindhoven zero-growth defensive	4.2%	Low (defensive)
China (Guangdong)	Supply CAGR 27.7%, market only 79% utilised	3.7%	High — supply outpacing demand; compounded by Bluesea tenant credit risk
Australia (Sydney)	Supply CAGR 18.8% — but demand matches at 18.8%	3.0%	Moderate
Germany (Frankfurt)	Supply CAGR 17.9% but pre-leased; energy rules pressure older assets	2.9%	Moderate (regulatory risk on older specs)
UK (Cardiff + London + GV7)	Cardiff supply CAGR 24.0% but utilisation 99.7%; London 12.1%	2.2%	Moderate; asset-specific writedowns already incurred
Italy (Milan)	Supply CAGR 14.8%; demand 14.8%	1.0%	Moderate
Malaysia (Cyberjaya)	Supply CAGR 26.5%, prices declining	0.3% (exit)	High — confirmed by manager's divestment

Single asset most exposed to supply unlock during lease rollover:

KDC SGP 8 — the AEI completes Q3 2027, which is exactly when Singapore's new CFA capacity (300 MW + 700 MW Jurong Island park) will be coming online. New supply unlocking just as SGP 8's new data hall seeks tenants. Mistimed entry risk is real.
KDC SGP 7 and KDC SGP 8 — land tenure now extended to 2050, but lease expiries from anchor tenants in 2030-2035 will coincide with deeper Singapore supply unlock.

DPU defensiveness verdict — multi-horizon

Horizon	Verdict	Key Driver
Near-term (FY2026)	Positive	Full-year Tokyo DC 3 + KDC SGP 3/4 consolidation contributions; ~10% sustained Singapore reversion run-rate; KDC SGP 1 occupancy recovery
Medium-term (FY2027)	Mixed	Income support tapers; SGP 8 AEI completes but lease-up risk; new Singapore CFA capacity online; Guangdong assets continue to require provisioning; UK/NL tail assets continue to drag
Long-term (FY2028+)	Mixed-to-Negative	Singapore supply CAGR 20.1% catches up to demand; SGP 7/8 land tenure ends 2050 (24.5 yrs) — beginning of perpetual capex/extension cycle; Tokyo single-tenant lease renewals start in mid-2030s; Dublin defensive position remains the structural bright spot

Download Full Report On Telegram

and continue reading my independent analyst report which will provide you with a detailed look at the fundamentals of the stock and a range of price targets to help you out with your investment decision for Keppel DC REIT:

*Join the channel → click on the channel name → under files → download the report you want!

If you would like to learn about REIT investing, you can find my entire methodology in my eBook: Retire With REITs here:

If you are looking for personalized financial advice, I offer a 1-to-1, fee-only consultation where you will receive personalized strategies to design, implement and manage a profitable REIT portfolio. You can find out more about it here: