Crown Castle Inc. (NYSE: CCI) — The Crown, Reclaimed and Repriced: Paying a Premium for the Weakest Tower

Independent fundamental research. Report date: 2026-06-13. The analysis body carries no recommendation and no price target; the sole exception is the clearly-labeled Author’s Take block below.

⚡ Author’s Take

This block is the author’s own subjective opinion. It is general information, not investment advice. The analysis that follows takes no position and names no price target.

Verdict: AVOID at this price / not-a-short / prefer the better peers. Quality oligopoly asset, lowest-quality franchise in it, priced at a premium to its betters. Crown Castle has just finished the most expensive act of corporate self-correction in the tower sector’s history: it sold the fiber/small-cell business it spent a decade and ~$15–20B building, for ~$8.4B net (closed May 1, 2026), cut the dividend ~32%, paid down ~$7B of debt, and emerged as a clean, US-only pure-play tower REIT. The asset is genuinely good — ~40,000 site-monopoly towers, ~98–99% retention, ~65% EBITDA margins, a contractual ~$24B backlog. The problem is the price relative to what you can buy instead. At ~$92 CCI trades at ~21x forward AFFO (~18.6x on the post-close run-rate) and ~22–24x EV/EBITDA — a premium to both American Tower (~16x AFFO, ~19x EV/EBITDA) and SBA Communications (~16x AFFO, ~18x EV/EBITDA), even though CCI is the slowest-growing, most customer-concentrated, most Sprint-churn-exposed, highest-levered, thinnest-covered, worst-margin, and worst-governed of the big three. The widely-repeated “CCI is the cheap tower” claim is simply false on the numbers.

The framing is “great industry, weakest franchise, mispriced expensive.” This is not a value setup; it is a quality-deficit name carrying a recovery multiple. The market is paying up for a clean story — kitchen-sink 2026 trough, ~4.6% covered yield, self-help margin levers, $1B buyback, DISH-litigation optionality — that is plausible but unproven, while ignoring that the same multiple buys a structurally better business in AMT or SBAC. My fair-value accumulation zone is ~$75–84 (~15–17x the ~$4.90 post-close AFFO run-rate, i.e. parity-to-slight-discount vs peers) — roughly the 52-week low ($76) is where the risk/reward turns genuinely interesting. It is not a short: the dividend, the irreplaceable footprint, and a real churn trough in 2026 cushion the downside to the business; the downside here is to the multiple. Conviction: medium. What flips me constructive: a de-rate into the low-$80s/high-$70s, or a 2027 organic-growth guide of 5%+ that proves 2026 was a true trough. What flips me bearish outright: a second dividend reset, a 2028 AT&T/T-Mobile renewal at flat-to-negative terms, or DISH litigation going to zero with organic stuck at 2–3%. Tag: a decade of value destruction, now priced as if the recovery is guaranteed — and cheaper next door.

1. Executive Summary

Crown Castle is, as of mid-2026, a brand-new company wearing a 30-year-old name. On May 1, 2026 it closed the sale of its Fiber and Small Cells segment — Zayo took the fiber-solutions business for ~$4.25B, EQT’s Arium Networks took the small-cell business for ~$4.25B — for ~$8.5B gross / ~$8.4B net, ending a decade-long, ~$15–20B experiment in becoming a converged “towers + fiber” operator. What remains is a focused, US-only communications-tower REIT: ~40,000 towers, ~1,250 employees, ~$4.0B of annual site-rental revenue, ~65% EBITDA margins, and a contractual revenue backlog of ~$23.7B.

The business itself is a good one and sits in a structurally attractive industry. US macro-towers are a three-player oligopoly (American Tower, SBA Communications, Crown Castle) with high barriers to entry, near-zero new-supply risk, ~98–99% tenant retention, and powerful co-location operating leverage — an incremental tenant on an existing tower carries near-100% incremental margin. CCI’s towers are local quasi-monopolies; the moat is real (local economies of scale plus tenant switching costs in Greenwald’s taxonomy).

But within that good industry, CCI is unambiguously the weakest of the three franchises, and on most axes by management’s own admission. (1) Concentration: ~90% of site-rental revenue comes from just three tenants (T-Mobile, AT&T, Verizon), and ~55% of its towers sit under AT&T/T-Mobile master lease agreements where CCI holds only land-buyout options. (2) Churn: CCI is the most exposed of the three to the Sprint/T-Mobile merger consolidation, a ~$200M FY2025 headwind that continues to bleed 1–2% of tower revenue per year through ~2034. (3) A fresh wound: DISH/EchoStar defaulted in January 2026; CCI terminated the lease, is claiming >$3.5B, and is suing — a ~$220M FY2026 churn hit with uncertain recovery. (4) Growth: organic tower revenue growth is decelerating (4.9% FY2025 → 3.3% guided FY2026, which management calls the “low point”). (5) Margin: management concedes a ~200–400bps structural EBITDA-margin gap to AMT/SBAC because CCI owns less of the land under its towers. (6) Leverage: even after the ~$7B fiber-proceeds paydown, net leverage targets ~6.0–6.5x with thin ~2.95x EBITDA/interest coverage. (7) Governance: four CEOs in two years, a two-front activist proxy war (Elliott’s “Reclaiming the Crown” plus co-founder Ted Miller’s competing slate), and a decade of capital destruction that the activist — not the board — forced into correction.

GAAP earnings are noise here, twice over: as a REIT the ~$690M annual depreciation suppresses net income, and the 2024 result was a −$3.9B loss driven by a ~$5.0B fiber impairment. The metric that matters is AFFO per share, which was ~$4.36 in FY2025, is guided roughly flat (~$4.36) for FY2026, and reaches a ~$4.90–4.95 run-rate in the twelve months after the fiber close (~$2.1B AFFO on ~424M post-buyback shares). At ~$92 that is ~21x trailing/forward AFFO and ~18.6x on the run-rate — a premium to AMT and SBAC, both of which trade at ~16x forward AFFO with better growth, better margins, lower leverage, and cleaner governance. The dividend was rebased ~32% (~$6.26 → $4.25/yr, ~4.6% yield) and sits at a ~90% payout near-term, leaving little cushion.

The investment tension is therefore not about the business — it is about price and relative value. CCI is a cleaned-up, de-risked, genuinely-good infrastructure asset that happens to be the lowest-quality member of its oligopoly and trades at a premium to the higher-quality members. The bull case requires believing 2026 is a true trough and that self-help (cost-out, land buyouts, buyback, DISH recovery) drives a re-rating; the bear case observes that you are paying a recovery multiple, today, for the weakest franchise in the group, with no margin of safety and a thin dividend cushion.

2. Business Overview

What Crown Castle is now. Following the May 1, 2026 divestiture, Crown Castle is a pure-play owner and operator of ~40,000 macro communications towers across the United States (including Puerto Rico). It is organized and taxed as a REIT. Its business model is the canonical tower model: CCI owns or leases the land, owns the vertical steel structure, and leases space on that structure to wireless carriers under long-term master lease agreements (MLAs). Each lease typically runs 5–15 years of initial term with multiple 5-year tenant renewal options, contractual annual rent escalators (~3%), and very limited tenant termination rights. The economics are a near-pure annuity: revenue is recurring, retention runs ~98–99% per year, and the contracted future-cash-inflow backlog is ~$23.7B with a weighted-average remaining life of roughly six years (this backlog excludes the now-terminated DISH agreement).

Revenue composition. FY2025 total continuing-operations revenue was ~$4,264M, of which site rental was ~$4,049M (~95%) and network services (installation/colocation application work) ~$215M. Services revenue is small, lumpy, and was deliberately shrunk when CCI exited certain installation activities in a 2023 restructuring (services fell from ~$421M in 2023 to ~$192M in 2024 before a modest ~$215M in 2025). The investable core is the site-rental annuity.

Customer base — the defining feature. CCI’s tenants are overwhelmingly the US national wireless carriers. T-Mobile, AT&T, and Verizon together account for ~90% of site-rental revenue — a concentration materially higher than a casual observer would assume and higher in practical effect than AMT’s (which has international and data-center diversification) or SBAC’s geographic mix. This concentration is the source of both CCI’s annuity stability (these are investment-grade counterparties on multi-year contracts) and its single largest structural vulnerability.

The portfolio and the services remnant. CCI’s ~40,000 towers are concentrated in the top US metropolitan markets — a deliberately urban/suburban footprint that skews toward higher-value, higher-tenancy sites (versus a more rural mix), reflecting its origins and its now-divested small-cell ambitions in dense corridors. The network-services business (~$215M revenue) provides site-development, installation, and application/colocation engineering work for carriers; it is lower-margin, project-based, and cyclical with carrier deployment activity, and management has deliberately shrunk it — it is a supporting service to the rental annuity, not a profit engine, and should be valued as such. Puerto Rico is included in the US footprint. The asset base is steel, land rights, and leases; there is essentially no inventory, no technology obsolescence at the structure level (a tower is a tower for decades), and minimal working capital — the balance sheet is dominated by long-lived real assets and the debt that funds them.

The cost structure and operating leverage. The dominant operating cost is ground rent — CCI leases the land under a majority of its towers and owns the land (or long-term easements) under the rest. Management has disclosed it owns or controls in perpetuity roughly 30% of its sites by count and is pursuing land buyouts to lift that toward ~40%, the principal self-help margin lever. Because ground rent and tower maintenance are largely fixed per site, each additional tenant co-located on an existing tower drops through at near-100% incremental margin. This is the engine of the model: revenue growth comes far more cheaply from adding tenants/amendments to existing towers than from building new ones, and sustaining capital expenditure is tiny (under 1% of revenue; total FY2026 net capex is guided to just ~$150–250M, much of which is discretionary land buyouts rather than maintenance).

How the revenue actually grows — the lease mechanics. Site-rental revenue moves on three levers. First, contractual escalators — most US leases carry ~3% fixed annual increases, a near-guaranteed organic tailwind regardless of leasing activity. Second, amendments — when a carrier adds equipment (new spectrum band, more antennas) to an existing lease, it pays more rent on the same tower; this is the highest-margin growth because the cost base is unchanged. Third, new colocations — a second or third carrier mounting equipment on an existing tower, again near-pure incremental margin. Against these positives runs churn — non-renewals and cancellations, currently elevated by Sprint decommissioning and the DISH default. Net organic growth is the sum of escalators (~+3%) plus new leasing/amendments (~+1–2%) minus churn (currently ~−1–2%), which is how a ~3–5% organic number is built. In 2024 CCI signed restructured “holistic” MLAs with its major carriers that traded some near-term billing certainty for longer-dated commitments — a defensive move to lock in the annuity amid the churn wave.

The backlog and recurring-revenue quality. The ~$23.7B contracted future-cash-inflow backlog (~6-year weighted-average life, excluding DISH) underwrites the bulk of forward revenue and is the clearest evidence of the annuity’s durability: ~95% of revenue is recurring site rental, and management has stated ~80% of 2026 organic growth was already contracted entering the year. As a REIT, CCI must distribute at least 90% of taxable income, which (combined with the capital-light maintenance profile) is why nearly all free cash flow is returned via the dividend rather than reinvested — the model is a payout machine, not a compounding-by-reinvestment machine.

Corporate form and why the numbers read oddly. As a REIT with a large historical-cost asset base, CCI’s GAAP depreciation (~$690M/yr on continuing operations) heavily suppresses reported net income, so GAAP EPS and P/E are the wrong lens. The sector reports and is valued on FFO and AFFO (funds from operations / adjusted funds from operations) per share, and on EV/EBITDA. Compounding the distortion, CCI’s recent GAAP history is dominated by the fiber exit: a ~$5.0B impairment drove a −$3.9B GAAP net loss in 2024, and a further −$659M discontinued-operations loss hit 2025. None of that touches the cash economics of the tower business, which is why this memo anchors on AFFO and EBITDA throughout.

Verdict. A simple, durable, high-margin annuity business with exceptional revenue visibility — but one whose revenue is unusually concentrated in three customers and whose reported financials are currently obscured by the accounting wreckage of the fiber exit. The quality of the cash business is high; the reported business looks far messier than the underlying annuity.

3. Industry Dynamics

Structure: a genuine oligopoly. The US macro-tower market is dominated by three public players — American Tower (~global, plus CoreSite data centers), SBA Communications (US plus Brazil/Central America, lean pure-play), and Crown Castle (now US-only) — plus a long tail of private/utility-owned and carrier-retained sites. The competitive structure is about as favorable as real-asset industries get: the customers (national wireless carriers) need ubiquitous vertical real estate to densify their networks, and the supply of suitable, zoned, anchor-tenanted tower sites is constrained. New towers require zoning approval, local permitting, often community (NIMBY) opposition, and an anchor tenant to be economic. The result is that the principal risk in the tower industry is demand, not new entrants — there is no credible scenario in which a flood of new tower supply competes away returns. This is the inverse of most cyclical industries.

The capital cycle (Marathon lens). Through 2020–2023 the combination of 5G mid-band deployment, the Sprint/T-Mobile merger integration, and DISH’s greenfield 4th-carrier build pulled a wave of leasing activity and capital into the sector — exactly the high-return-attracts-capital phase Marathon warns about. That cycle is now normalizing: carrier capex has moderated off the 5G peak, the Sprint integration is producing churn rather than new leasing, and DISH’s build has collapsed into default. The industry is therefore late in its leasing cycle, with growth decelerating across all three players toward a ~3–5% organic range. Importantly, because there is no new-supply response, the down-leg of this cycle shows up as slower growth, not destroyed returns — the towers keep earning their rent; they just add fewer new tenants.

The fourth-carrier question. The 5G era briefly promised a fourth national network: DISH, which acquired Boost and spectrum from the Sprint/T-Mobile divestiture remedy and committed to a greenfield build, leasing tower space from all three operators. DISH’s build stalled under capital constraints, and in January 2026 it defaulted on its CCI lease — removing not only ~$220M of CCI revenue but also the sector’s marginal new-demand source. The practical effect is that the US is now a three-network market (Verizon, AT&T, T-Mobile), with cable MVNOs (Comcast, Charter) riding on T-Mobile/Verizon rather than building. Fewer networks means fewer distinct tenants competing to colocate — a structural negative for tower lease-up that CCI, with its ~90% big-three concentration, feels acutely.

The demand drivers. Long-term tower demand rides on mobile data traffic growth (still compounding at double digits), network densification (more antennas, more spectrum bands, more amendments per tower), fixed-wireless access (FWA) as a genuine new use case, and eventually 6G and newly-auctioned spectrum (e.g. the FCC’s planned 800MHz/AWS auctions expected to seed a 2027+ leasing cycle). Each carrier spectrum deployment typically drives amendments (more equipment on existing leases) and new colocations — the highest-margin growth for the tower owners. These are real, durable secular tailwinds; the question is timing and magnitude, not direction.

The substitution debate. The long-tail bear argument is technological: small cells (which CCI just exited), satellite direct-to-device (Starlink, AST SpaceMobile), and network architecture changes could erode macro-tower demand at the margin over a long horizon. Management characterizes satellite D2D as complementary and “de minimis” today, and that is correct for the foreseeable future — direct-to-device satellite is a coverage-filler for dead zones, not a capacity substitute for dense urban macro networks. But it is a real long-tail risk that belongs in the risk matrix, especially for rural sites.

The capital-cycle read, quantified. Marathon’s framework asks where capital is flowing and what returns it is chasing. In towers, the 2020–2023 period saw carrier network capex surge — the big three collectively spent well over $100B across the 5G mid-band rollout — which drove a wave of amendments and colocations and pulled the tower operators’ organic growth into the high-single digits. That capital has now receded: carrier capex has normalized lower as the initial 5G build completes, T-Mobile is removing equipment (Sprint decommissioning) rather than adding it, and DISH’s build collapsed. The supply side of towers never responded (no new-supply glut, the industry’s saving grace), so the down-leg manifests purely as decelerating lease-up rather than impaired economics. The Marathon signal here is benign-to-cautionary: there is no overbuild to mean-revert, but the demand pulse that drove the last cycle has passed, and the next one (spectrum auctions, 6G) is not yet underway. Tower operators are in the quiet part of the cycle — which is exactly when their growth looks least impressive and when, historically, the better entry points have appeared.

Regulation. The sector is lightly regulated relative to utilities — primarily local zoning, FAA/FCC tower registration, and environmental review. The more relevant policy lever is spectrum policy: FCC spectrum auctions drive carrier deployment cycles and therefore tower leasing. The reinstatement of FCC auction authority and the planned 800MHz/AWS-3 and other auctions are the most concrete forward catalyst — newly-auctioned spectrum must be deployed on towers, seeding the next amendment/colocation cycle, generally expected to build from 2027 onward.

Verdict: a structurally good industry, late in its leasing cycle. High barriers, no new-supply threat, sticky investment-grade customers, secular data tailwinds, and pricing power via escalators. The near-term reality is decelerating organic growth as the 5G/Sprint/DISH cycle rolls off, but the long-run structure is among the best in real assets. The industry is not the problem with CCI; the company’s position within it is.

4. Competitive Position

The moat, named. In Greenwald’s taxonomy, the tower moat is local economies of scale combined with customer captivity (switching costs). Each individual tower is effectively a local monopoly: once a carrier’s antennas are installed, mounted, integrated, and propagation-tested at a given location, moving to a competing structure nearby is costly, slow, and operationally risky (coverage gaps during transition), and a suitable alternative structure frequently does not exist within the required radius. The captivity is reinforced by long contracts with escalators and renewal options. The scale economics are local, not global — what matters is owning the right structure at the right location, and a national footprint of ~40,000 such local monopolies is extremely hard to replicate. The moat is real and it is visible in the financial outcomes: ~74% site-rental gross margin, ~65% EBITDA margin, ~98–99% retention, and contractual escalators that deliver positive organic growth even in a soft leasing year.

But CCI is the weakest of the three, and concedes it. The moat is shared by all three players; the question for an investor is relative quality, and on that axis CCI consistently loses:

1. Structural margin deficit. Management openly acknowledges a ~200–400bps EBITDA-margin gap versus AMT and SBAC, driven primarily by the fact that CCI leases more of the land under its towers (peers own more outright). Ground rent is the largest cost; owning more land would lift margins. CCI’s land-buyout program (30% → 40% ownership target) is explicitly an attempt to close a gap it admits exists — i.e. it is starting behind.
2. No diversification. Post-fiber, CCI is ~100% US wireless. AMT has international towers and the CoreSite data-center platform; SBAC has Brazil/Central America. CCI’s concentration is a double-edged feature: simpler and higher-quality assets, but zero offset when the US carriers pull back.
3. Worst Sprint exposure. Of the three, CCI carries the largest relative exposure to the Sprint/T-Mobile consolidation churn — the single largest drag on its organic growth through ~2034.
4. Less control of its own base. ~55% of CCI’s towers sit under AT&T and T-Mobile master lease agreements on which CCI holds buyout options rather than freehold ownership — a structurally weaker position than owning the asset outright.
5. Worst recent organic growth. Management itself characterized CCI’s 5G-cycle growth as roughly in line with one peer and lagging the other.

Co-location economics — the one unambiguous strength, worked through. Consider the canonical tower. A single-tenant tower might generate ~$25–30k/yr of rent against ground lease + maintenance costs that are largely fixed; its standalone return on the build cost is modest (high single digits). Add a second tenant at a similar rent and almost the entire incremental rent drops to the bottom line — the ground lease and steel are already paid for — roughly doubling site revenue while site costs barely move, lifting the site-level return into the high teens/low twenties. A third tenant pushes returns into the 20s%+. This is why tenancy ratio (tenants per tower) is the single most important operating statistic in the industry, and why escalators-plus-amendments on an existing tenant base compound so cheaply. CCI’s ~65% EBITDA margin is the financial fingerprint of this co-location leverage. The corollary, though, is symmetric: when a tenant churns off a multi-tenant tower (as Sprint and DISH are doing), the lost rent comes off at near-100% margin too — which is precisely why CCI’s churn wave bites EBITDA so hard. Co-location leverage cuts both ways, and CCI is currently on the wrong side of it.

The head-to-head, quantified. On the metrics that matter, CCI trails: organic tower growth ~3.5% (ex-DISH) vs ~4–5% at both AMT and SBAC; EBITDA margin ~65.1% vs ~65.5% (AMT) and ~65.6%+ (SBAC, historically as high as ~68%); ROIC ~7% vs ~7.8% (AMT) and ~12.4% (SBAC, the clear leader); AFFO/share growth roughly flat near-term vs mid-single-digit-plus at peers. SBAC, the lean pure-play, consistently posts the best margins, the highest returns, and the lowest payout (~41–45% vs CCI’s ~90%) — the textbook demonstration that the same tower moat, run with more owned land and tighter cost discipline, throws off materially better economics than CCI extracts from a near-identical asset base. CCI’s land-buyout program is an explicit attempt to migrate toward that model, but it is years and billions of capital away from closing the gap.

The Greenwald tests. Greenwald’s diagnostics for a genuine moat are market-share stability and persistently high returns on capital. Towers pass the first emphatically — share among the three operators is extremely stable because the assets are fixed in place and rarely change hands, and a carrier almost never relocates from one operator’s tower to another’s. They pass the second only partially: site-level returns are high, but corporate ROIC is dragged down by the price paid to assemble the portfolios and (for CCI) by the leverage and the fiber detour — CCI’s ~7% ROIC is respectable but not the hallmark of an unassailable franchise, and it trails SBAC’s ~12%. The honest synthesis is that the asset has a strong local moat (stable share, high site returns) but the equity captures a diluted version of it after the capital structure and the price of entry. For CCI specifically, the moat is real at the tower level and second-tier at the enterprise level.

Verdict: a real but second-tier moat in a commoditizing competitive set. All three players own near-identical national footprints of local-monopoly towers and compete for the incremental dollars of the same three customers; differentiation is increasingly about execution, cost structure, and cycle-time rather than structural advantage. CCI has the moat but the thinnest version of it — the lowest margins, the highest concentration, the worst churn exposure, and the least control of its own asset base. The land-buyout program is genuine self-help, but it is closing a gap, not building an edge.

5. Growth History and Forward Opportunities

Headline revenue is declining — but the reasons matter. Continuing-operations (tower) revenue fell from ~$4,734M (2023) to ~$4,460M (2024) to ~$4,264M (2025). Two effects dominate, and only one is a quality problem. First, the deliberate shrinkage of low-margin network-services revenue after the 2023 restructuring (services down ~$230M). Second, and more important, Sprint/T-Mobile merger churn: as T-Mobile decommissions redundant Sprint sites, CCI loses the associated leases — ~$200M of non-renewals in FY2025 alone.

Underlying organic growth is positive but decelerating. Stripping the noise, organic site-rental growth (escalators + new leasing/amendments − churn) was ~4.9% in FY2025 (~3.8% excluding DISH), slowed to ~3.1% in Q1-2026 (~3.3% ex-DISH), and is guided to ~3.3% for FY2026 (~3.5% ex-DISH). Management explicitly calls 2026 the “low point” and notes ~80% of 2026 organic growth is already contracted. The shape of the story is therefore a decelerating-into-trough year, with the bull thesis resting on re-acceleration thereafter.

The DISH/EchoStar blow. DISH had been a contractual (not activity-based) growth contributor of ~$50M/yr. In January 2026 DISH defaulted; CCI delivered notice of default and termination, asserts DISH/EchoStar owe >$3.5B, and has sued for declaratory judgment and breach. The practical effect is ~$220M of FY2026 churn (the loss of the contracted DISH revenue), with any litigation recovery a separate, low-visibility, multi-year option that the base case should exclude.

The forward leasing drivers. The re-acceleration case rests on: (1) the next spectrum-deployment cycle — C-band continuation, plus the FCC’s planned 800MHz/AWS spectrum auctions seeding a 2027+ build; (2) fixed-wireless access, where carrier data growth compounds 30%+; (3) carrier convergence deals (Verizon/Frontier, AT&T/Lumen) potentially driving fiber-to-the-tower and densification; (4) selective new builds; and (5) early-stage edge-compute/data-center colocation trials (not material today). These are real but they are call options, not underwritten growth — management has pointedly declined to give a multi-year organic-growth re-acceleration guide.

The 2028 renewal cliff — the unguided variable that matters most. A large slug of CCI’s master lease agreements with its anchor carriers come up for renewal/renegotiation around 2028. Because ~55% of CCI’s towers sit under AT&T/T-Mobile MLAs, the terms struck at that renewal — escalator rates, holistic-commitment volumes, churn allowances — will substantially shape organic growth into the back half of the decade. This is the single largest source of forward uncertainty that management has not guided, and it cuts both ways: a constructive renewal (the carriers value the footprint and lock in long-dated commitments with healthy escalators) would underwrite the re-acceleration thesis; a hard-nosed renewal (carriers using their concentration leverage to extract concessions, having just demonstrated with Sprint/DISH that they will decommission) would entrench low-single-digit growth. The ~90% revenue concentration that makes the annuity stable also hands the three customers enormous negotiating leverage at renewal — a tension that has no clean resolution until the terms are signed.

Sizing the optionality. Fixed-wireless access is the most tangible new-demand vector: carrier FWA subscriber bases are growing rapidly and the underlying data consumption compounds 30%+, which over time pressures carriers to densify — favorable for towers. Edge-compute/data-center colocation at tower sites is a genuine but early and immaterial trial. Net new builds are selective. None of these is large enough, soon enough, to move CCI’s ~$4B revenue base meaningfully in 2026–2027; they are the call options that could turn a 3% grower into a 5% grower in 2028+ if they land — but they are options, not a plan.

Per-share growth is currently a self-help story, not a top-line story. With AFFO/share roughly flat (~$4.36 FY2025 → ~$4.36 FY2026E), near-term per-share progress depends more on the ~$65M cost-out program, land buyouts (margin), ~$120M/yr of interest savings from the debt paydown, and the $1B buyback than on the revenue line. That is a lower-quality growth profile than a business compounding on organic demand.

Verdict: low-to-mediocre-quality growth near term, with genuine but unproven re-acceleration optionality. The honest read is that CCI is in a trough year driven by churn it cannot control (Sprint, DISH), with positive but decelerating organic growth, and a per-share story currently carried by financial engineering and cost-out rather than demand. The 2027+ spectrum cycle could change this — but the company itself won’t underwrite it yet, and an investor shouldn’t either.

6. Financial Quality

Read the cash, not the GAAP. CCI’s reported earnings are doubly distorted: REIT depreciation suppresses net income structurally, and the fiber exit produced a ~$5.0B impairment (−$3.9B GAAP net loss in 2024) plus a −$659M discontinued-ops loss in 2025. Continuing-operations diluted EPS was ~$2.52 in 2025; the headline GAAP net loss figures are accounting artifacts of the divestiture and should be set aside.

The AFFO bridge. The metric that matters is AFFO per share. The bridge: continuing-operations net income + real-estate depreciation/amortization (~$690M) + stock-based comp (~$73M) + financing amortization ± straight-line rent/amortization of prepaid rent − sustaining capex (tiny) → AFFO. On this basis:

• FY2025 AFFO ≈ $1.9B = ~$4.36/diluted share (down ~4% YoY), confirmed in the FY2025 release and the 2026 proxy pay-versus-performance disclosure (which shows $4.36 for 2025 versus a recast ~$6.98 for 2024).
• FY2026E AFFO ≈ $1.9B (~$4.36/share) — roughly flat: the ~$65M cost-out (~$55M in-year) and ~$120M interest savings approximately offset the ~$220M DISH churn and residual Sprint churn.
• Post-close 12-month run-rate AFFO ≈ $2.1B (~$4.90–4.95/share) on ~424M post-buyback shares — the figure bulls anchor on (it was revised down from ~$2.34B after stripping the ~$280M DISH contribution).

A quality-of-earnings flag in the payout metric. CCI targets a dividend of 75–80% of “AFFO excluding amortization of prepaid rent” — a self-defined sub-metric that is more generous than headline AFFO. Investors should track the dividend against reported AFFO (~90% near-term payout) rather than the adjusted sub-metric.

Two more QoE notes worth flagging. First, straight-line accounting: under lease accounting, CCI recognizes rent (and ground-lease expense) on a straight-line basis over the lease term, which creates non-cash revenue/expense and a growing straight-line-rent receivable that can run ahead of cash — a standard REIT feature, but one that means reported revenue slightly leads cash collection. AFFO adjusts for this, which is part of why AFFO (not GAAP revenue or net income) is the honest cash proxy. Second, the negative book equity (−$1.9B) is not a distress signal but an accounting artifact: years of dividends and buybacks in excess of GAAP retained earnings, compounded by the $5B fiber impairment, have driven cumulative distributions/losses below paid-in capital. It renders P/B and book-based return metrics meaningless (hence the null own-history P/B percentile), which is normal for a mature, high-payout, asset-heavy REIT — but it also means there is no equity-book cushion absorbing a downside, only the market value of the towers themselves.

Margins and operating leverage. Adjusted EBITDA margin was ~65.1% in FY2025 ($2,776M on $4,264M); site-rental gross margin ~74%. Incremental co-location margins approach 100%. The catch, again, is the conceded ~200–400bps structural margin deficit to peers — CCI’s 65% is the lowest of the three despite a near-identical business, because of the land-lease cost structure.

Leverage and coverage — the balance-sheet constraint. This is where CCI is genuinely stretched. Net debt/EBITDA was ~8.7x at year-end 2025 on towers-only EBITDA, before the ~$7B fiber-proceeds paydown; pro-forma, management targets ~6.0–6.5x. Total debt including ~$5.2B of finance/capital leases (ground leases) was ~$29.9B at Q1-2026; cash was a thin ~$55M; book equity is negative (−$1.9B), a combined buyback-and-impairment artifact that renders P/B meaningless (and returns a null P/B percentile in own-history screens). Interest coverage is thin for an investment-grade REIT at ~2.95x EBITDA/interest (~2.22x on operating income). Maintaining the IG rating is the explicit gating constraint on capital allocation — it is precisely why the company skewed the fiber proceeds heavily toward debt paydown (~$7B) over buyback (~$1B). The ~$120M/yr of interest savings from that paydown is the single largest tailwind to FY2026 AFFO.

The refinancing wall and rate sensitivity. With ~$22–25B of debt outstanding after the paydown and a weighted-average coupon built largely in the low-rate era, CCI faces a multi-year ladder of maturities that must be refinanced at materially higher prevailing rates. Each tranche that rolls from (say) a ~3% legacy coupon to a ~5–6% current coupon is a direct, permanent hit to AFFO — and with coverage already thin at ~2.95x, the company has limited room to absorb a sustained higher-rate environment. This is the structural reason management prioritized debt paydown over buyback: not because paydown is more accretive at today’s price (it arguably isn’t), but because protecting the investment-grade rating and the cost of ~$22B+ of debt is existential to the equity. Rate sensitivity, not operational risk, is the dominant financial vulnerability.

Dividend coverage — thin, by design. The rebased $4.25 dividend against ~$4.36 of FY2026E AFFO is a ~97–98% payout on a strict basis (~90% on the company’s preferred “AFFO excluding amortization of prepaid rent” sub-metric). That leaves almost no retained cash to fund land buyouts or buybacks from operations — both are being funded from the one-time fiber proceeds, not recurring FCF. The dividend is “covered” only in the narrow sense that AFFO exceeds it; there is no cushion for a churn or rate surprise, and management has already demonstrated (with the ~32% cut) that it will reset the distribution rather than over-distribute. An investor buying for the ~4.6% yield should underwrite the possibility of a second reset in a downside scenario.

Capital intensity and FCF. Sustaining capex is under 1% of revenue; total FY2026 net capex is guided to ~$150–250M (much of it discretionary land buyouts). The tower model is genuinely capital-light to maintain — the heavy capital was the fiber build, now gone. FCF conversion is strong on the continuing business: the ~$2.7–3.0B of operating cash flow converts to ~$1.9B of AFFO after interest and the modest maintenance capex, essentially all of which is distributed.

Verdict: genuinely good unit economics inside a stretched, low-equity, thinly-covered balance sheet, with AFFO/share flat-to-down near term. The cash economics of the tower annuity are high quality. But CCI carries the highest leverage and thinnest coverage of the big three, negative book equity, a ~90% AFFO payout, and a structural margin deficit. Economics improve with scale — but CCI captures less of that scale benefit than AMT or SBAC. The financial profile is that of a good business on a tight balance sheet, not a fortress.

7. Capital Allocation

This is the heart of the bear case, and it is damning on the record. Over roughly a decade, Crown Castle pursued a “converged” strategy, building and buying a fiber/small-cell business it believed would ride 5G densification. It acquired Sunesys (~$1.0B, 2015), Wilcon (~$600M), FiberNet/Quanta assets, and — the centerpiece — Lightower for ~$7.1B in 2017, roughly ~$10.7B of fiber/small-cell M&A in three years, plus ~$1–2B/yr of subsequent capex. Conservatively, $15–20B+ of capital was sunk into the fiber/small-cell platform.

The outcome: it was exited for ~$8.4B net — roughly half (or less) of the capital invested. The 2024 financials crystallized the destruction: a ~$5.0B goodwill impairment plus write-downs drove a ~$5.07B discontinued-operations loss and a −$3.9B total GAAP net loss, with a further −$659M in 2025. This is one of the larger large-cap-REIT capital-misallocation episodes of the decade, and — critically — the correction was forced by an activist, not initiated by the board.

Why the fiber strategy failed — the lesson. The strategic logic in 2017 was superficially coherent: 5G would require dense small-cell networks in urban corridors, small cells need fiber backhaul, and owning metro fiber would let CCI capture that build the way it captured the macro-tower build. The economics never materialized. Small cells lacked the co-location leverage that makes macro towers so profitable — each small-cell node served essentially one carrier, the deployment was slower and more capital-intensive than promised, municipal permitting was a quagmire, and the returns came in far below the macro-tower business they were compared against. Crucially, CCI was paying tower-like acquisition multiples and reinvestment rates for fiber-like (lower, more competitive) returns, and funding it with debt and a stretched dividend. The episode is a textbook Marathon/Greenwald cautionary tale: management extrapolated the economics of its genuinely-advantaged business (macro towers) into an adjacent business that lacked the same barriers and co-location leverage, and the market’s eventual verdict — a halving of invested capital — was brutal but correct. The most charitable read is that CCI bought an option on a densification wave that arrived more slowly and less profitably than underwritten; the less charitable read is empire-building dressed as strategy.

The dividend cut as an admission. Concurrent with the fiber sale, CCI rebased its dividend ~32%, from ~$6.26/yr (~$1.565/quarter) to $4.25/yr (~$1.0625/quarter). After years of paying ~$6/share against insufficient pure-play cash generation, the reset is an implicit admission that the prior distribution was propped up by the (capital-destroying) fiber strategy and by leverage. The new dividend sits at a ~90% payout of current AFFO; policy holds it flat at $4.25 until the payout naturally falls to the 75–80% target, then grows with AFFO. Notably, when an analyst (KeyBanc) argued a deeper cut plus more buyback would be more accretive, management declined.

Use of the $8.4B proceeds. Locked: ~$7.0B to debt paydown (to protect the IG rating and cut ~$120M/yr of interest) and ~$1.0B to a share buyback — CCI’s first-ever meaningful repurchase (historical “buybacks” of ~$23–33M/yr were merely RSU tax-withholding). This is a rational, conservative deployment given the balance sheet, even if a more aggressive investor might have wanted more buyback at a depressed price.

Compensation — a yellow flag. The 2026 proxy reveals that the 2025 annual incentive plan replaced AFFO/share with Adjusted EBITDA (weighted 70%, up from 50%) plus organic revenue growth (30%) — i.e. the bonus now keys off EBITDA/scale rather than per-share value, in the very year the company is shrinking itself by design. The 2025 plan paid above target. The long-term incentive provides some offset (three-year cumulative AFFO/share with a relative-TSR modifier, downgraded to a ±15% modifier), but the shift of the annual bonus toward an absolute scale metric, mid-transformation, is not reassuring on alignment.

The peer contrast makes it worse. Over the same decade that CCI sank ~$15–20B into fiber and exited at a loss, American Tower built an international tower platform and bought CoreSite to add data centers (a debatable but at least non-destructive diversification), while SBA Communications ran a disciplined, focused, low-payout, high-return pure-play and compounded AFFO/share at the best rate of the three. CCI’s strategic detour did not just fail in absolute terms — it represented a decade of relative underperformance versus peers who allocated capital more sensibly within the same industry. An investor evaluating management quality has three real-world counterfactuals sitting right next door, and CCI’s prior regime looks worst against all of them. The current team inherits that hole; protecting the dividend and delevering is the right first move, but it is digging out, not building.

Verdict: a poor historical record, with rational current cleanup that does not yet constitute a good track record. Be direct: the fiber saga destroyed an enormous amount of shareholder capital, and it took an activist to force the correction. The new Hillabrant/Patel regime is doing the sensible things — delever, modest buyback, focus, cost-out, dividend discipline — but that is repair, not evidence of skilled allocation, and the comp-metric shift toward EBITDA/scale is a caution. Capital allocation is the single most important negative in the thesis.

8. Changes and Headwinds — Last Two Years

The last two years at Crown Castle have been among the most turbulent for any large-cap REIT, defined by an activist campaign, a governance crisis, a strategic dismemberment, and a major customer default.

• The Elliott activist campaign (“Reclaiming the Crown”). Elliott Investment Management engaged from 2020 and escalated sharply in late November 2023, demanding leadership change, a fiber strategic review, and capital discipline. On Dec 7, 2023, CEO Jay Brown retired (director Anthony Melone interim) and Elliott signed a cooperation agreement securing board seats and a strategy/operations review committee.
• A two-front proxy war. In February 2024, co-founder Ted Miller (via Boots Capital) launched a competing proxy fight challenging the Elliott settlement — the source of the wall of contested proxy filings (DEFC14A, DEFN14A, PRRN14A, and 26 DFAN14A) in the public record. The dissident slate was defeated at the May 22, 2024 annual meeting.
• CEO carousel — four in two years. Jay Brown (out Dec 2023) → Anthony Melone (interim) → Steven Moskowitz (permanent, Apr 2024) → Moskowitz terminated after under a year (Mar 2025) → Daniel Schlanger (CFO, interim) → Christian (Chris) Hillabrant (permanent, ex-Vantage Towers CEO, Sept 15, 2025). CFO Sunit Patel joined April 2025. This degree of leadership instability is a genuine governance red flag.
• The fiber sale + dividend reset (announced Mar 13, 2025; closed May 1, 2026). The strategic-review endgame: ~$8.5B sale of Fiber + Small Cells and the ~32% dividend rebase — the single largest change to the thesis, converting CCI into a clean US-only tower pure-play.
• The DISH/EchoStar default (Jan 2026). DISH defaulted; CCI terminated the MLA, asserts >$3.5B owed, and sued DISH (and added a claim against EchoStar in Q1-2026). ~$220M of FY2026 churn; recovery uncertain and excluded from base case.
• Operational reset. A ~20% workforce reduction (to ~1,250 FTEs), a ~$65M cost-out program, and a renewed land-buyout push to attack the peer margin gap.

Verdict: the changes are net stabilizing for the go-forward business but reflect a company emerging from crisis, not strength. The fiber exit and delevering genuinely de-risk the equity and simplify the story. But they were forced corrections to self-inflicted damage, the leadership remains freshly installed and unproven, and a fresh wound (DISH) opened just as the old ones (Sprint, fiber) were being dressed. The trajectory is improving; the starting point was a crisis.

9. Risk Analysis

The dominant risks are customer concentration (a structural, low-probability/high-severity tail), continued churn (high-probability/medium-severity, the live drag), balance-sheet/rate sensitivity (the leverage and thin coverage amplify any cash-flow disappointment), and — distinct from all business risk — multiple de-rate, since CCI trades richer than better peers.

How the risks interact — the transmission mechanism. The danger in CCI is not any single risk but their correlation. The same ~90% customer concentration that stabilizes the annuity in good times hands the three carriers pricing leverage at the 2028 renewals; a hard renewal would deepen churn and slow organic growth; slower growth against a ~90% payout and thin ~2.95x coverage would pressure the dividend; and a dividend reset or growth disappointment is precisely the trigger for the multiple to de-rate from its current premium toward (or below) peers. In other words, the customer, growth, balance-sheet, dividend, and valuation risks are not independent — they are a single chain that a carrier-concentration or rate shock could pull through. The mitigant is equally structural: the assets are irreplaceable, retention is ~98–99%, and the carriers genuinely need the footprint, so even a bad renewal is a concession, not an exit — the towers keep earning. This is why the downside is best characterized as a de-rate-plus-dividend-trim scenario (painful but recoverable) rather than an impairment of the franchise (which is remote). The catastrophic-loss probability is low; the disappointing-total-return probability is not.

The interest-rate point deserves emphasis because it is the least-discussed and most mechanical risk: CCI is, in effect, a leveraged spread business — it earns rent on towers and pays interest on ~$22B+ of debt. With coverage at ~2.95x and a refinancing ladder rolling low-coupon legacy debt into a higher-rate market, a sustained high-rate environment is a slow, compounding drag on AFFO that no operational improvement can fully offset. The equity is long real assets and short a large fixed-income book; rates are the hidden second factor in the stock alongside leasing fundamentals.

10. Valuation Discussion (Embedded Expectations)

Where it trades (2026-06-13, ~$92.16). Market cap ~$40.2B; enterprise value ~$68–70B as reported (pro-forma ~$61–63B after the ~$7B fiber-proceeds paydown). On REIT metrics: ~21.1x trailing/forward AFFO (~$4.36) and ~18.6x on the post-close run-rate AFFO (~$4.90–4.95); ~22–24x EV/EBITDA; ~4.6% dividend yield. On its own 10-year history, EV/EBITDA (~22x pro-forma) sits mid-range (the historical band is roughly 20x–33x), and the own-history P/S percentile (~86th) is contaminated by the fiber-divestiture revenue shrink (a smaller revenue base mechanically lifts P/S versus history), while the P/E percentile is REIT-noise and P/B is null on negative equity. EV/EBITDA is the cleaner own-history lens, and it says “mid-range,” not “cheap.”

The peer comparison is the crux — and it inverts the popular narrative.

GAAP P/E is meaningless for all three; AFFO multiples are the right lens. The table delivers a blunt conclusion: CCI does not trade at a discount to its peers — it trades at a premium. It is ~5 turns richer on forward P/AFFO (21x vs ~16x), still a premium on the run-rate (~18.6x), and the richest on EV/EBITDA (~22x PF vs ~18–19x). Yet on every fundamental axis — growth, margin, concentration, leverage, coverage, governance, capital-allocation record — CCI is the weakest of the three. The “CCI is the cheap tower” thesis is false on current numbers. Whatever narrative discount exists (the optically depressed FY2026 AFFO from DISH churn and pre-buyback share count) is deserved, not an opportunity: every quality axis argues for a multiple below AMT/SBAC, not above.

Embedded expectations / reverse read. At ~$40B of equity with a 4.6% yield, a ~90% payout, and only ~3.5% decelerating organic growth, justifying ~18.6x run-rate AFFO requires the market to underwrite some combination of: (a) 2026 as a genuine churn trough with re-acceleration toward 5%+ on the C-band/spectrum/6G cycle; (b) successful execution of the 200–400bps self-help margin program; © DISH litigation recovery as a free option; and/or (d) a re-rating toward a “scarce US-infrastructure” premium or an outright takeout. That is the bull operating stack being priced as the base case. If any leg slips, compression toward the peer multiple (~16x) is ~15% downside before any AFFO miss.

The total-return decomposition. Strip the narrative and the math is sobering. A buyer at ~$92 receives a ~4.6% dividend yield. On top of that, AFFO/share is guided flat in 2026 and grows only as fast as escalators-plus-modest-leasing-minus-churn plus the one-time benefit of the buyback and interest savings — call it low-to-mid-single digits in a good outcome. So even holding the multiple constant, the expected annualized total return is roughly the ~4.6% yield + ~3–5% AFFO/share growth ≈ ~8–10% — and that already assumes 2026 is the trough and the self-help levers work. For that same ~8–10% (or better), an investor can own AMT or SBAC — faster-growing, better-margin, lower-levered, better-governed — at ~16x AFFO, i.e. with re-rating upside rather than de-rating risk. The CCI bull is therefore implicitly betting on either multiple expansion (hard to justify from a premium starting point) or a discrete catalyst (DISH win, takeout) that the base case cannot count on.

Asset value / SOTP sanity check. Tower portfolios change hands in private markets at high-teens-to-low-20s EV/EBITDA multiples, so CCI’s ~$61–63B pro-forma EV on ~$2.7–2.8B of tower EBITDA is not detached from private-market reality — a strategic or infrastructure-fund buyer could rationalize a similar or modestly higher figure for a clean, financeable, irreplaceable US footprint, which is the kernel of the takeout-optionality argument. But note the implication: the public multiple already embeds close to a private-market valuation, so a take-private would need to pay a premium over an already-full public price — limiting the margin of safety that the takeout thesis is supposed to provide. The asset is worth a lot; it is also already priced like it.

Scenarios (return drivers; no price target).

• Bear: a de-rate toward or below peers (~14–16x AFFO) as a second churn leg emerges, the 2028 AT&T/T-Mobile renewals come in flat-to-negative, DISH recovery goes to zero, the margin program stalls, and organic settles at 2–3%. The ~90% payout leaves no buyback firepower and raises the specter of a second dividend reset. Negative total return despite the yield — and the damage is to the multiple, not the franchise.
• Base: the multiple roughly holds (~18–20x run-rate); 2026 proves the trough; organic re-accelerates to ~4–4.5%; cost-out plus land buyouts add ~100–200bps of margin; the $1B buyback is modestly accretive; the dividend is held then grown. Total return ~= the ~4.6% yield + low-to-mid-single-digit AFFO growth ≈ ~8–10% annualized, with DISH/edge as unpriced upside.
• Bull: a re-rating to 22–24x on faster growth — C-band 2027 + 6G/800MHz spectrum drive a new leasing cycle to 5%+, the full 300–400bps margin gap closes, DISH delivers a meaningful recovery, edge/new-build becomes real, or a strategic/private takeout materializes at a premium. Double-digit total return.

Verdict: full price for the weakest franchise. The distribution is roughly symmetric-to-negative: the moat and dividend cushion the downside to the business, but the premium entry multiple caps the upside and creates real de-rate risk. You are not being paid to take CCI’s incremental risks versus AMT or SBAC — you are paying more.

11. Variant Perception

Consensus. The prevailing Street narrative is constructive: a cleaned-up, de-risked pure-play; 2026 as a “kitchen-sink” trough; a ~4.6% covered dividend to be paid while you wait; and a self-help-driven re-rating toward peers as churn troughs and margins recover. In this telling, CCI is the contrarian recovery play in a great industry — buy the simplification, collect the yield, and wait for the leasing cycle and margin program to lift AFFO/share. The implicit anchor is that CCI “should” trade in line with AMT/SBAC and currently offers a cleaner US-only story. The flaw in that anchor, as the valuation section shows, is that CCI does not trade at a discount to be closed — it trades at a premium that would have to widen for the consensus return to materialize from multiple expansion, leaving the thesis dependent on fundamental re-acceleration that management itself declines to guide.

The strongest bull case. An irreplaceable, US-only oligopoly infrastructure asset with no FX/EM risk and the simplest story of the three; a sustainable post-rebase dividend; churn troughing in 2026 with ~80% of organic already contracted; concrete self-help levers (cost-out, land buyouts, $1B buyback, ~$120M interest savings); DISH litigation as a free ~$3.5B+ option; a 2027+ spectrum-auction leasing cycle; and genuine takeover/re-rating optionality (a focused, financeable US tower pure-play is an attractive target for private infrastructure capital).

The strongest bear case (the variant view this memo finds more persuasive). CCI is the weakest of the big three on essentially every axis — concentration, Sprint churn through 2034, structural margin deficit, leverage, coverage, organic growth, and governance/capital-allocation track record — and yet it trades at a premium to its better peers. AFFO/share is flat-to-down; the ~90% payout leaves no cushion; the comp plan now rewards EBITDA/scale rather than per-share value mid-shrinkage; and a decade of value destruction was corrected only under activist duress by a leadership team installed months ago. The stock is mispriced expensive, not cheap, and the return math is dominated by multiple risk.

The 3–5 assumptions that matter most. (1) Is 2026 a genuine trough, or the first of multiple churn legs (Sprint runs to 2034; 2028 AT&T/T-Mo renewals loom)? (2) Does organic growth re-accelerate to 4–5%, or settle at 2–3%? (3) Does the 200–400bps self-help margin program actually close the peer gap? (4) Is the dividend durable at a ~90% payout? (5) Does the market re-rate CCI down toward its (better) peers?

What would falsify each side. Bull falsified by: a 2027 organic guide still below 4%; a second dividend reset; flat-to-negative 2028 AT&T/T-Mobile renewal terms; a DISH suit dismissed or settled near zero; the margin program slipping. Bear falsified by: a 2027 organic guide of 5%+; on-schedule margin-gap closure; a meaningful DISH cash recovery; a credible premium takeout bid; or the buyback scaled up materially as the payout normalizes.

12. Fact vs. Interpretation

13. Open Questions

1. Is 2026 truly the trough? Sprint churn runs to ~2034 and the 2028 AT&T/T-Mobile MLA renewals are a major unguided variable — does organic actually re-accelerate, or grind sideways at 3%?
2. What is the realistic DISH/EchoStar recovery? CCI claims >$3.5B; the form (cash, equipment, settlement), timing, and probability are all unknown. Base case excludes it — but it is the single largest swing option.
3. Can the 200–400bps margin gap actually close, and over what horizon? Land buyouts are slow and capital-consuming; the program’s pace and ROI are unproven.
4. Will the dividend hold at $4.25 at a ~90% payout if 2026 disappoints? Management resisted a deeper cut, but a churn surprise could force a second reset.
5. Will the new Hillabrant/Patel leadership allocate capital well after a decade of destruction — and does the EBITDA-keyed comp plan undermine per-share discipline?
6. Is there a strategic endgame (take-private by infrastructure capital, or a merger), given how clean and financeable the pure-play now is?

14. What Must Be True

Bull case — what must be true:

• 2026 is a genuine churn trough and organic growth re-accelerates toward 4–5% on the 2027+ spectrum/6G cycle.
• The self-help program (cost-out, land buyouts) closes a meaningful chunk of the peer margin gap, lifting AFFO/share growth above the flat near-term trajectory.
• The dividend holds and eventually grows; the $1B buyback proves accretive; DISH delivers a non-trivial recovery.
• The market continues to award CCI a peer-or-premium multiple (re-rating optionality, possibly a takeout).
• Falsification test: a 2027 organic-growth guide below 4%, a second dividend reset, or flat-to-negative 2028 AT&T/T-Mobile renewal terms would break the bull case.

Bear case — what must be true:

• CCI’s premium-to-peers multiple compresses toward (or below) AMT/SBAC’s ~16x AFFO as the market recognizes the relative-quality deficit.
• Organic growth stays stuck at 2–3% as Sprint churn persists and DISH recovery disappoints; AFFO/share stays flat-to-down.
• The ~90% payout and thin coverage leave no cushion; a churn or rate surprise pressures the dividend.
• Falsification test: a 2027 organic guide of 5%+, on-schedule margin-gap closure, a meaningful DISH cash recovery, or a credible premium takeout bid would break the bear case.

15. Source Appendix

See CCI_source_appendix.md for the full source list. Primary sources: Crown Castle FY2025 Form 10-K and Q1-2026 Form 10-Q (SEC EDGAR, CIK 0001051470); Crown Castle 2026 DEF 14A proxy; the May 1, 2026 fiber-close press release and updated FY2026 outlook; Q1-2026, Q4-2025, and Q2-2025 earnings-call transcripts; the contested-proxy filing record (DEFC14A/DFAN14A/PRRN14A, 2024 Elliott/Boots Capital campaign); Form 4 insider-transaction corpus (CIK 0001051470); American Tower and SBA Communications Q1-2026 results and outlooks (for peer comps); and quantitative cross-checks via third-party financial-data aggregators. All figures reconciled to filings where possible; management commentary is treated as hypothesis and validated against filings and external evidence.

APPENDIX A — Standard Diligence Questionnaire

Crown Castle Inc. (NYSE: CCI) — supplemental to the research memo. Report date: 2026-06-13.

Answers are grounded in the underlying analysis; Fact/Interpretation/Assumption labels applied where it matters.

General

What thoughtful questions have other investors asked about this company? The dominant questions: (1) Is 2026 genuinely the organic-growth trough, or the first of several churn legs (Sprint to ~2034, DISH now, 2028 AT&T/T-Mobile renewals)? (2) Is the rebased $4.25 dividend safe at a ~90% payout? (3) Will the new leadership allocate capital well after the fiber disaster? (4) Does CCI deserve to trade at a premium to AMT/SBAC, or should it converge down? (5) What is DISH/EchoStar litigation worth? (6) Is CCI a take-private candidate now that it is a clean US pure-play?

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: Organic growth is at a cyclical low — management explicitly calls 2026 the “low point” as the 5G/Sprint/DISH leasing-and-churn cycle rolls off. AFFO/share is flat-to-down (~$4.36). So the company is closer to a trough than a peak on growth, though the absolute multiple is not depressed.

Driven by the external environment or internal actions? Both: external (carrier capex normalization, Sprint/DISH churn) and internal (the deliberate fiber exit, services shrinkage, dividend reset, cost-out). Near-term per-share progress is driven more by internal self-help (cost-out, interest savings, buyback) than by demand.

How stable are revenues? Very stable at the contract level — ~98–99% retention, multi-year MLAs, ~3% escalators, ~$23.7B contracted backlog. Instability comes from large discrete churn events (Sprint decommissioning, DISH default), not from volatile underlying demand.

Outlook for products/services? Site rental is a durable annuity with secular data-traffic tailwinds; the FY2026 guide is ~3.3% organic (3.5% ex-DISH), with potential re-acceleration on the 2027+ spectrum cycle.

How big is this market — growing, shrinking, domestic/international? US macro towers: a mature, slow-growing (~3–5% organic), domestic-only market for CCI post-fiber. The total demand pool grows with mobile data, but CCI’s share is fixed within a 3-player oligopoly.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Structurally stable (3-player oligopoly, no new-supply threat), but the competition for the same three customers’ incremental dollars is intensifying as the leasing cycle slows — a demand-side, not supply-side, squeeze.

How profitable is the business (ROIC, ROE)? Fact: Low-single-digit ROIC (~7% on a heavily-levered, goodwill-laden base; ROE is not meaningful on negative equity). Cash margins are high (65% EBITDA, 74% gross), but returns on total invested capital are modest and below AMT (~7.8%) and SBAC (~12.4%).

How profitable is the industry — competitors, barriers? Highly profitable at the asset level (65%+ EBITDA margins across all three); high barriers (zoning, site scarcity, anchor tenant). CCI is the lowest-margin of the three by a conceded 200–400bps.

Can the business be easily understood? Yes — post-fiber it is a clean, single-segment tower-leasing annuity.

Can it be undermined by foreign low-cost labor? No — the assets are physical US real estate; labor is a minor cost.

Do brands matter? No — towers are bought on location, price, and reliability, not brand.

Nature of competition? Location, cost-to-lease, speed of deployment, and existing footprint — execution rather than structural differentiation.

Customers’ switching costs? High — relocating installed, integrated, propagation-tested antennas is costly, slow, and risks coverage gaps; a suitable alternative structure often does not exist nearby. This is the core of the moat.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The franchise value of ~40,000 site-monopoly locations and ~$23.7B contracted backlog far exceeds book value (which is negative). Conversely, goodwill (~$5.1B) remains on the books post-fiber.

Off-balance-sheet liabilities? Ground-lease commitments are substantially capitalized (~$5.2B of finance/operating lease obligations); long-dated operating-lease and purchase commitments exist (standard for the sector). The DISH litigation is a contingent asset (claim), not a liability.

How conservative is the accounting? Mixed. The REIT/AFFO framework is standard, but CCI’s dividend-policy AFFO sub-metric (“AFFO excluding amortization of prepaid rent”) is more generous than headline AFFO — track the payout against reported AFFO. The 2024 fiber impairment was taken decisively (not drawn out).

How CapEx-hungry is the business? Post-fiber, very capital-light to maintain (sustaining capex <1% of revenue; total net capex ~$150–250M, mostly discretionary land buyouts). The capital hunger was the fiber build — now divested.

Capital Allocation & Management

How much FCF, and how is it used? Strong FCF conversion on the tower annuity (~$2.7B+ OCF, tiny capex). Uses: dividend (~$1.85B, the priority), ~$7B one-time debt paydown from fiber proceeds, ~$1B buyback, modest land buyouts.

Philosophy? Newly reset under Hillabrant/Patel: delever to protect IG, focus on US towers, discipline on the dividend (hold $4.25 until payout normalizes), modest buyback, cost-out. Interpretation: rational repair, but no track record yet.

Significant acquisitions recently? No — the story is divestiture (fiber, May 2026), not acquisition. The prior decade’s fiber M&A (Lightower $7.1B, Sunesys, Wilcon) is the capital-destruction history.

Buying back shares? Yes, for the first time meaningfully — a ~$1B program from fiber proceeds. Historical “repurchases” were RSU tax-withholding only (~$23–33M/yr).

Issuing large amounts of stock to insiders? No — SBC is modest (~$73M/yr).

Compensation policy? Yellow flag: the 2025 annual incentive was re-weighted toward Adjusted EBITDA (70%) + organic revenue (30%), rewarding scale rather than per-share value during a deliberate shrinkage. LTI uses 3-year cumulative AFFO/share with a ±15% relative-TSR modifier.

Motivations of management? New team (CEO since Sept 2025, CFO since Apr 2025) installed post-activist-campaign; mandate is cleanup and credibility-rebuilding. No officer open-market share purchases to signal personal conviction.

Valuation & Market Data

ADR, MLP, or K-1 issuer? None — CCI is a US C-corp taxed as a REIT; it issues a 1099-DIV, not a K-1.

Dividend policy? Rebased ~32% to $4.25/yr (~4.6% yield); held flat until the payout falls to a 75–80%-of-AFFO target, then grows with AFFO. Currently ~90% of AFFO.

How profitable is the business? High cash margins (65% EBITDA), modest returns on capital (~7% ROIC), lowest of the three peers.

Is net income diverging from cash from operations? Yes, massively and expectedly — GAAP net income is distorted by REIT depreciation and the fiber impairment (−$3.9B GAAP in 2024) while OCF remained strongly positive. Use AFFO, not net income.

Risks & Downside

What would cause the stock to decline? A multiple de-rate toward peers (the largest risk — CCI trades at a premium it arguably doesn’t deserve); a second churn leg or weak 2028 renewals; a dividend reset; rising rates against a levered, thinly-covered balance sheet; DISH recovery going to zero.

Risk of catastrophic loss? Low — the asset base is irreplaceable physical infrastructure with investment-grade tenants and ~98–99% retention. The downside is to the multiple and the dividend, not to the existence of the franchise.

Chance of total loss? Very low. Even in a severe scenario, the towers retain substantial value; the equity is cushioned by the asset base, though the high leverage means equity holders bear amplified cash-flow risk.

Recent News & Events

Has the business environment changed recently? Profoundly: the fiber/small-cell sale closed May 1, 2026, converting CCI into a US-only tower pure-play; the dividend was rebased ~32%; DISH defaulted (Jan 2026) triggering ~$220M of churn and a >$3.5B claim; and new leadership (CEO Hillabrant Sept 2025, CFO Patel Apr 2025) took over after a two-year activist-driven governance upheaval.

Significant acquisitions? No acquisitions — a major divestiture (fiber).

Change in accounting policies? No material change; fiber reclassified to discontinued operations.

Recent changes — new markets, facilities, management? Exited fiber and the small-cell/installation businesses; ~20% workforce reduction to ~1,250 FTEs; ~$65M cost-out program; renewed land-buyout push; four CEOs in two years culminating in the current Hillabrant/Patel team.

APPENDIX B — Source Appendix

Crown Castle Inc. (NYSE: CCI). Report date: 2026-06-13. Primary sources prioritized over secondary; management commentary treated as hypothesis and validated against filings and external evidence.

Primary — SEC filings (EDGAR, CIK 0001051470)

• Crown Castle Inc. Form 10-K, FY2025 — business description, customer concentration (~90% T-Mobile/AT&T/Verizon), ground-lease/land-ownership detail, churn disclosure, discontinued-operations (fiber) treatment, segment and revenue detail. https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001051470&type=10-K
• Crown Castle Inc. Form 10-Q, Q1-2026 — Q1 organic growth (~3.1%), DISH default/termination disclosure and litigation, balance-sheet detail (total debt ~$29.9B incl. ~$5.2B leases, cash ~$55M, negative equity −$1.9B). https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001051470&type=10-Q
• Crown Castle Inc. 2026 DEF 14A (proxy statement) — compensation structure (2025 AIP re-weighted to Adj EBITDA 70% / organic revenue 30%; LTI cumulative AFFO/share + ±15% rTSR modifier), pay-vs-performance AFFO/share ($4.36 2025; recast ~$6.98 2024), board/governance.
• Contested-proxy record (2024 campaign): DEFC14A, DEFN14A, PRRN14A, PREC14A, and DFAN14A filings — Elliott “Reclaiming the Crown” cooperation agreement and the competing Boots Capital (Ted Miller) slate; outcome (dissidents defeated, May 22, 2024 annual meeting).
• Form 4 insider-transaction corpus (CIK 0001051470) — transaction-code tally since Jan 2024 (grants/exercises/withholding dominate; 7 de minimis director open-market purchases; no officer/CEO/CFO open-market buys).
• 8-K material-event record — CEO transitions (Brown retirement Dec 2023; Moskowitz appointment Apr 2024 and termination Mar 2025; Hillabrant appointment Sept 2025; Patel CFO Apr 2025); fiber-sale announcement (Mar 13, 2025) and close (May 1, 2026); dividend reset; DISH default.

Primary — Company disclosures

• Crown Castle press release, “Closing of Sale of Fiber and Small Cell Businesses and Updated Full-Year 2026 Outlook,” 2026-05-01 — ~$8.5B gross / ~$8.4B net proceeds; Zayo (fiber solutions) and EQT/Arium Networks (small cells); updated FY2026 outlook (site rental, Adj EBITDA, AFFO); use of proceeds. https://investor.crowncastle.com/news-releases/news-release-details/crown-castle-announces-closing-sale-fiber-and-small-cell — and https://www.globenewswire.com/news-release/2026/05/01/3285983/7657/en/Crown-Castle-Announces-Closing-of-Sale-of-Fiber-and-Small-Cell-Businesses-and-Updates-Full-Year-2026-Outlook.html
• Earnings-call transcripts (public): Q1-2026 (2026-04-22), Q4-2025 (2026-02-04), Q2-2025 (2025-07-23) — management framing of organic-growth trajectory (“2026 low point”), dividend policy and payout target, leverage target (6.0–6.5x), $1B buyback, cost-out (~$65M), DISH, and the margin-gap/land-buyout program.

Primary — Peer comparables

• American Tower (AMT) Q1-2026 8-K / press release and earnings call — AFFO/share (~$10.99), organic growth, EBITDA margin, leverage (~5x), dividend (~$7.16). https://www.sec.gov/Archives/edgar/data/0001053507/000105350726000094/pressreleaseq12026.htm
• SBA Communications (SBAC) Q1-2026 8-K outlook and earnings call — AFFO/share (~$12.1–12.3), organic growth, margins, leverage, dividend (~$5.00). https://www.sec.gov/Archives/edgar/data/0001034054/000119312526191545/d49075dex991.htm

Secondary — Trade press and financial media

• CNBC, “Crown Castle CEO to retire, handing win to activist investor Elliott,” 2023-12-07. https://www.cnbc.com/2023/12/07/crown-castle-ceo-to-retire-handing-win-to-activist-investor-elliott.html
• CNBC, coverage of the Elliott/Boots Capital proxy contest, 2024-05-22. https://www.cnbc.com/2024/05/22/crown-castle-cci-activist-elliott-boots-capital.html
• Light Reading, “Crown Castle sticks to fiber diet with $7.1B Lightower deal,” 2017 — the centerpiece fiber acquisition. https://www.lightreading.com/fttx/crown-castle-sticks-to-fiber-diet-with-7-1b-lightower-deal
• Fierce Network, “Zayo closes its $4.25B purchase of Crown Castle’s fiber business,” 2026. https://www.fierce-network.com/broadband/zayo-closes-its-425b-purchase-crown-castles-fiber-business
• MarketScreener — dividend-rebase coverage and consensus framing.

Quantitative cross-checks

• Third-party financial-data aggregators — income statement, balance sheet, cash flow, profitability/credit ratios, enterprise value, and valuation multiples for CCI, AMT, and SBAC (reconciled to filings; SEC filings remain primary).
• Public market-data sources — live prices, market cap, share count, and quick comps for CCI/AMT/SBAC.
• Own-history valuation percentiles (third-party) — P/E 27.8th (REIT-noise); P/B null on negative equity; P/S ~86th (contaminated by the fiber-divestiture revenue shrink); composite 56.9th.

Note on EV reconciliation: ROIC’s headline AMT enterprise value understates AMT debt; AMT EV was taken as ~$131–137B (diluted-EV basis) for the peer table, giving AMT EV/EBITDA ~19x. CCI’s pro-forma EV (~$61–63B) reflects the ~$7B fiber-proceeds debt paydown.