Chapter 3

Figures converted from Indonesian rupiah (IDR) at historical FX rates — see data/company.json.fx_rates. Ratios, margins, multiples, tenancy ratios, and percentages are unitless and unchanged.

Carrier Consolidation

TOWR's revenue now comes 87% from three carriers, and after the 2025 XL–Smartfren merger its single largest customer, XLSmart, supplies 42% while Indosat supplies another 34% — two merged operators covering three-quarters of the top of the book [1]. The same consolidation that concentrated the base is the demand-side reason tower-leasing revenue barely grows: merged carriers deduplicate overlapping sites and keep one tenant where two once sat. Real secular tailwinds — double-digit data-traffic growth, 5G, fibre — reach TOWR through a narrowing set of buyers.

A market that keeps collapsing into fewer buyers

Indonesia's mobile market has consolidated twice in three years. Indosat Ooredoo merged with Tri (Hutchison) into PT Indosat Ooredoo Hutchison in 2022, and in 2025 XL Axiata combined with Smartfren to form PT XLSmart Telecom Sejahtera [2]. What was a field of five or six wireless operators is now effectively three national buyers of tower space — Telkomsel, Indosat, and XLSmart. Management's own framing is telling: it reports growth "despite ongoing consolidation in the telecommunications industry," a phrase that appears where a landlord would normally describe demand [3].

For an independent tower company, fewer carriers is a structural fact, not a passing one. Every tenant a tower can hold must come from three counterparties, and each of those counterparties now has more sites of its own and more bargaining power at renewal. This is the mechanism sitting underneath the near-flat tower-leasing line documented in Financials and Estimates.

The concentration, from the audited note

The revenue-concentration disclosure in the financial statements makes the shift precise. In 2023 the top three customers were Indosat (38%), XL Axiata (31%), and Telkomsel (12%) — 81% combined. In 2024 they were 79%. In 2025, with XL and Smartfren folded into one line, the top three reached 87% [4] [5].

Source: FY2025 Annual Report, revenue-concentration note [6].

Source: FY2024 and FY2025 Annual Reports, revenue-concentration notes [7] [8].

Two things stand out. First, the single largest customer share rose from 35% (Indosat in 2024) to 42% (XLSmart in 2025), because the merger fused what had been two separate tenants — XL at 32% and Smartfren-related billings — into one relationship [9]. Part of the 79%-to-87% step is therefore mechanical relabelling rather than fresh dependence — a former "other" customer now sits inside a named top-three line. But the economic point is unchanged: the number of independent decision-makers who set TOWR's tower demand fell from four to three, and two of them now write three-quarters of the cheque.

Source: FY2025 Annual Report, revenue-concentration note; total revenue about $800M [10] [11].

Concentration this high is not unique to TOWR — it is the shape of the industry. Peer Mitratel reports that 89% of its revenue comes from Indonesia's largest mobile operators, anchored on Telkomsel [12]. A tower landlord's customers are, by definition, the handful of carriers licensed to run a national network. The question is not whether concentration exists but whether it is rising and whether the counterparties are sound.

One fact cuts the other way on soundness. As the base concentrated, credit exposure actually improved: gross third-party trade receivables fell to about $120M at end-2025 from about $209M a year earlier, and the allowance for expected credit losses dropped to about $2.1M from about $5.8M [13]. The three counterparties are large, investment-grade-adjacent carriers that pay; concentration here is concentration among strong names, not weak ones.

Why consolidation holds the tower line flat

The demand-side damage from a carrier merger is specific and well understood in this industry. Smaller peer Gihon described it plainly a year ahead of the event: after a merger the two operators "consolidate their tower leases, potentially retaining only one tenant per site that was previously occupied by two entities," mirroring what followed the Indosat–Hutchison combination, "which caused a decline in tenancy ratio" at tower providers [14]. A year later Gihon confirmed it had happened: the completed XL–Smartfren merger "led to network adjustments by the operators, which temporarily affected the demand for tower leasing in the second half of the year" [15].

TOWR's own operating detail carries the same fingerprints, and it complicates the headline tenancy figure. The company reports a tenancy ratio of 1.67x on 60,540 tenants, up modestly from 58,035 in 2024 [16]. But management explained on its year-end call that the uptick came in part from reclassification, not new demand: it "restructured some reseller contracts to become direct lease," so tenants that were previously not counted "as part of the XL, Smartfren merger" now are [17]. On the fibre side, where the accounting is cleaner, the merger's drag is visible directly: fibre-to-the-tower utilisation slipped to 1.79x from 1.84x, which management attributed to the "impact of mergers" [18].

The Indosat–Hutchison rationalisation, meanwhile, is still not finished four years on. Of the 847 towers TOWR added over 2025, a "couple of hundred towers" of relocation work for Indosat and Hutchison remained to be concluded — much of the year's build activity was relocations, not net new demand [19].

The tailwinds are real — and they run through a narrower funnel

None of this means demand for connectivity is weak. The secular pull behind the business is genuine and worth stating on its own terms.

Data traffic has grown at a double-digit CAGR for years and management expects that to continue, driven by heavier mobile and fixed internet use, the early adoption of cloud and AI, and the eventual rollout of 5G once spectrum is released [20]. Indonesia's internet penetration has reached roughly 80% with rising ARPU as usage deepens [21], and the decade-long shift from wired to wireless communication continues to feed demand for tower and fibre infrastructure [22]. Management's 2026 outlook rests on the same pillars: sustained demand for high-speed internet, continued 4G build-out, potential 5G, cloud and AI adoption, and — importantly for a landlord — carriers' shift toward "asset-light" strategies that push infrastructure ownership onto tower companies [23].

The tension is that these two forces point in opposite directions at the tower line. Rising data traffic pulls demand for capacity up; carrier consolidation pushes the number of tenants per site — and the number of buyers who can pay for that capacity — down. The visible outcome so far is that data-driven demand shows up mostly in the non-tower businesses (fibre, connectivity, FTTH), while the tower-tenancy line stays close to flat because the buyer count keeps shrinking. The asset-light tailwind is real, but it is a tailwind delivered by three counterparties with growing leverage over price.

The read, and what would change it

On balance, carrier consolidation is a durable structural headwind to tower-tenancy growth and a genuine rise in customer concentration — the demand-side reason forward growth is modelled at only low-single digits. It is the strongest fact on the permanent-derating side of the report's central question.

The counter-case is not trivial. Lease contracts run 10–13 years and are non-cancellable, so churn from deduplication is a slow grind of non-renewals and relocations, not a cliff — the towers already built keep paying while the merged carriers work through their overlaps. The counterparties are financially strong and their receivables are shrinking, not stretching. And each merged carrier still has to densify its network to carry double-digit traffic growth, which over time argues for more equipment on towers, not less. What would move the read toward the mispricing side: evidence that a merged carrier's net site count has stopped falling and turned up, or a return to a fourth national buyer. What would harden the bear read: a step-down in tenancy or tower-leasing revenue as XLSmart works through its site overlap the way Indosat–Hutchison still is.