ATN International: The Tower Sale Was the Easy Part

The market had ATN International filed under "slow-growth regional telecom with flat revenue and a heavy balance sheet." Then Q2 earnings printed a headline $10.71 per share and the stock jumped 27.6% to US$30.63 in one day, then another 4 percent the next. In five trading days it was up 34 percent, closing near $32.

The headline EPS grab everyone's attention. But the number that matters is the one nobody is talking about yet: net leverage dropped from 2.36x at the end of 2025 to 0.91x in Q2 2026. That is the kind of balance-sheet compression that changes what a business can actually do with its capital. And the tower sale that drove it — which everyone has already priced in — is only the first leg.

The operating inflection is quieter than the tape

Revenue of $184.5 million was up 2 percent year-over-year. Strip out lower construction work and the loss of a U.S. Virgin Islands government subsidy, and organic revenue grew roughly 3 percent. Adjusted EBITDA — earnings before interest, taxes, depreciation, and amortization, the closest thing this business has to an operating cash proxy — rose 9 percent to $49.7 million. The margin expanded to 27.0 percent, up from 25.3 percent a year ago and from 26.7 percent in Q1.

The margin trajectory is the operating signal. Q1 was 26.7 percent, Q2 was 27.0 percent. The company is generating more cash per dollar of revenue as it shifts from legacy voice and low-margin construction toward higher-margin carrier services and fixed broadband. Carrier services in the U.S. segment alone grew to $32.8 million in Q2, up from the prior-year period.

This is not a dramatic growth story. It's an operating-leverage story. Revenue is flat-to-moderately growing, but earnings power per dollar is climbing. That matters because the company's full-year Adjusted EBITDA guidance of $183 million to $193 million already bakes in the recurring $1 million monthly drag from lost tower rent after the initial closing. If the company hits the midpoint of $188 million, that would be above last year's actual $190 million run-rate despite losing tower income for the remainder of the year.

The balance-sheet mechanics

Here's where the tower sale does the heavy lifting. ATN sold approximately 214 tower sites in the Southwest to an affiliate of Everest Infrastructure Partners. The initial closing in June brought $268 million in cash. Up to another $30 million can come from deferred sites that are subject to construction and operational conditions over the next ten months.

Management deployed $68 million of those proceeds immediately to pay off a revolving credit facility. Total debt fell to $513 million. Cash and restricted cash stood at $332 million. The net-debt ratio — total debt minus cash, divided by annualized EBITDA — collapsed from 2.36x to 0.91x.

The tower transaction multiple far exceeded ATNI's current trading multiple, as CEO Naji Khoury put it on the call. That's the point: infrastructure assets are worth more as standalone towers than as a cost center buried inside a regional telecom operator. Selling them at a premium while the stock still trades at 0.78 times book value is the rerating gap this trade is built on.

What comes after the tower

The tower sale is done. The question is whether there's a second and third leg that keeps the balance sheet moving toward free cash flow.

Management flagged a planned sale of certain U.S. spectrum licenses — those are radio-frequency rights that carriers use to offer wireless service — for up to $41 million, expected to close in 2027. That's a non-core asset in a company that is increasingly focused on broadband and fixed infrastructure, not wireless spectrum.

There's also approximately $150 million in BEAD (Broadband Equity, Access and Deployment) federal broadband funding available within ATNI's footprint in late 2026 and into 2027. These are reimbursable government grants that fund fiber build-out in underserved areas. The reimbursement doesn't flow directly into earnings, but it offsets capital expenditure — the one line item that's been keeping free cash flow negative.

On the capital spending side, the company guided to $105 million to $115 million in capex for 2026, net of reimbursable expenditures. Operating cash flow over the trailing twelve months was $127.8 million, against $174.8 million in gross capex. That produced negative free cash flow of $47 million. If BEAD reimbursements close the gap between gross and net capex, and if EBITDA converts at a similar rate to operating cash flow next year, the path to positive free cash flow is real — not guaranteed, but structurally credible.

The valuation gap

The stock's 34 percent five-day run has made the tape look exciting. But the multiples still look like the old story.

Enterprise value sits at roughly $990 million against trailing EBITDA of about $172 million (the TTM run-rate), implying roughly 5.8 times EV/EBITDA. For a regional telecom with a 3.5 percent dividend yield, 19 consecutive years of dividend payments, and a balance sheet that just moved from 2.36x net leverage to 0.91x, that multiple is below what even cautious mid-cycle telecoms command. The stock trades at 0.78 times book — meaning the market values the entire company below its net asset value, even after a dramatic balance-sheet improvement.

AInvest's aggregate rating signals a Buy, which reflects the structural improvement. But the score doesn't capture the asset-sale pipeline: deferred tower closings worth up to $30 million, spectrum sales worth up to $41 million, and $150 million in reimbursable BEAD funding that will ease the capex burden. Those are the items that turn an interesting telecom story into a free-cash-flow rerating setup.

The target and the tripwire

If the company hits $188 million in 2026 Adjusted EBITDA, adds the $268 million in tower proceeds and up to $70 million from spectrum and deferred tower closings over the next 18 months, and runs capex net of reimbursements at the low end of guidance, the free-cash-flow picture for 2027 starts to look positive. At that point a multiple of 7 to 8 times EV/EBITDA — still conservative for a lightly leveraged telecom generating a 3.5 percent yield — implies an enterprise value of $1.3 to $1.5 billion.

With net debt likely declining further from the current $181 million toward net-neutral territory, equity value would sit in the $1.3 to $1.5 billion range. On roughly 15.5 million shares that works out to approximately $84 to $97 per share over the next 18 months.

That requires the operating leverage to hold, the asset sales to close as planned, and the BEAD reimbursement flow to materialize. If any of those stumbles, the thesis gets dented, not destroyed.

Tripwire: if the Q3 Adjusted EBITDA margin contracts back below 26 percent, the operating leverage story is broken. A margin reversal at this point would signal that carrier services growth is stalling and cost discipline is eroding. That's the condition that makes the old story right again.

Second, if the spectrum sale falls through or is materially reduced, the free-cash-flow timeline extends. That doesn't invalidate the setup — the tower proceeds alone were enough to transform the leverage ratio — but it makes the rerating slower and less compelling at $32.

The real risk isn't the headline

The market reacted to a $10.71 EPS number that was driven by a $230.9 million gain one-time gain. When that headline fades — and it will — the stock needs the organic margin expansion to keep climbing. The Q2 print of 27.0 percent after Q1's 26.7 percent is a good sign, but one quarter of margin growth doesn't lock it in.

The setup is cleanest right now because the balance sheet has already improved while the multiple hasn't fully caught up. The stock is trading below book value after the tower sale closed, which means the market is still pricing the old leverage profile. That's the gap the thesis is built on.

But a 34 percent move in five days is not nothing. The tape is running ahead of the financial proof. If you're entering here, the margin trajectory is the proof that keeps you positioned. Watch the next two quarters for evidence that 27 percent is the floor, not the ceiling.

Discipline over ego. If the margin reverses, the setup is broken.