Telecom stocks have a reputation as the boring uncle of the market. Slow, steady, pays a dividend, buy it and forget about it. That reputation is doing a lot of unpaid labor for the people selling you the shares. Under the hood, telecom is one of the most capital-hungry, debt-dependent, accounting-flexible industries on any exchange, and the numbers these companies lead with are almost always the ones that make them look best.
None of this means the sector is a scam. It means that evaluating a telecom stock properly takes about twenty extra minutes of work, and most retail investors never do it. Here’s the version nobody spells out.
Why Telecom Looks Simpler Than It Actually Is
On paper, the pitch is great. Recurring subscription revenue. An essential service people won’t cancel during a bad economy. Enormous barriers to entry because building a network costs a fortune. What’s not to like?
The catch is that the same barrier to entry that protects the business also traps the money inside it. Telecom is a fixed-cost industry. Once the network exists, adding one more customer costs almost nothing, which sounds amazing until you realize the inverse is also true: every dollar of price competition drops straight to the bottom line.
So the business model becomes a treadmill. You spend heavily to build and upgrade, you depreciate that spend over a decade, you borrow to cover the difference, and you pay a dividend to keep shareholders from noticing. Understanding that loop is 80% of evaluating the stock.
The Metrics They Want You To Watch
These get the press releases. They’re also the easiest to bend.
Subscriber counts
“Connections” is not the same thing as customers. A single household can produce several connections once you count tablets, smartwatches, connected devices, wholesale lines sold to resellers, and industrial sensors. Headline net-add figures often blend all of it together. When you see a big number, go find the specific line item for the highest-value category — usually phones on contracts with actual people paying actual bills — and compare growth there instead.
ARPU
Average revenue per user sounds neutral. It isn’t. ARPU moves when a company bundles in device financing, raises admin fees, changes promotional timing, or shifts the mix toward cheaper plans. A rising ARPU can mean customers are paying more, or it can mean the company shoved hardware sales into the same bucket. Look for service revenue per user, separate from equipment.
Churn
Churn is the percentage of customers leaving. It’s also one of the most conveniently defined numbers in the industry.
- Blended churn mixes cheap prepaid customers with expensive contract customers and hides weakness in the high-value base.
- Voluntary versus involuntary churn — cutting off non-payers counts as churn too, but it’s a very different signal than losing customers to a competitor.
- Promotional cycles distort everything. Churn can look fantastic right up until a discount expires and a wave of cancellations lands.
If churn is improving while ARPU is falling, that’s not a win. That’s customers downgrading.
The Accounting Flex Nobody Talks About
This is where telecom gets spicy, because the industry is genuinely capital-intensive and accounting rules give management real room to choose how that capital shows up.
EBITDA is close to useless here
EBITDA strips out depreciation and amortization. For a software company, that’s reasonable. For a telecom company, depreciation is the cost of doing business. The network wears out. Fiber gets buried and eventually needs replacing. A company can post beautiful EBITDA numbers and still burn cash every single year once you subtract what it actually spends to keep the lights on.
Depreciation schedules
Management picks the assumed useful life of equipment. Stretch it out and annual depreciation falls, which raises reported earnings — without a single extra dollar of cash coming in. It’s legal, it’s disclosed in the notes, and it’s very easy to miss if you only read the earnings summary.
Capitalized costs
Some labor, software development, and overhead can be capitalized and spread over future years instead of being expensed today. More capitalization means smoother, prettier earnings in the short term and a bigger depreciation bill later.
Adjusted numbers
Every “adjusted” metric was defined by the company. When the adjustment list is longer than the actual income statement, stop and ask what’s being removed and why it keeps recurring. Costs that show up every year are not one-time costs, no matter what the label says.
Leases and off-balance-sheet history
Telecom runs on leases — towers, rooftops, land, fiber capacity. Lease obligations are debt in everything but name, and sale-leaseback deals can generate a nice-looking cash inflow today in exchange for permanent rent payments tomorrow. Read the commitments section, not just the headline debt figure.
Free Cash Flow: Read Every Line
Free cash flow is the honest number in telecom, but only if you build it yourself.
- Start with operating cash flow.
- Subtract capital expenditures.
- Subtract spectrum and license purchases — these are real, recurring, and enormous. Many companies report free cash flow before spectrum, which is not free cash flow at all.
- Check for working capital games. Pulling receivables forward or stretching supplier payments boosts cash flow once and can’t be repeated forever.
- Remember that device financing makes a telecom partly a consumer lender. Money owed on handsets sits on the balance sheet and carries default risk.
Debt Is The Actual Business Model
Telecom companies borrow because building networks costs more than operations generate in any given year. That’s not automatically bad — it’s how the industry has always worked. What matters is whether the debt is manageable.
- Leverage ratio: total debt against cash earnings. How does it compare to peers, and is it trending up or down?
- Maturity wall: how much debt comes due in the next few years, and can it be refinanced if rates are higher?
- Fixed versus floating: floating-rate debt means rising rates hit earnings directly.
- Dividend coverage: is the payout covered by free cash flow, or is it being funded by new borrowing? A dividend paid with debt is a countdown timer, not income.
Red Flags Checklist
- Subscriber growth driven almost entirely by low-value connections or wholesale lines
- Adjusted earnings diverging further from reported earnings each year
- Capital expenditure falling as a percentage of revenue while management promises a major network upgrade
- Useful-life assumptions extended without explanation
- Dividends consistently exceeding free cash flow
- Debt rising while buybacks continue
- Churn definition changing between quarters
- Heavy promotional activity described as “disciplined”
The Competitive Reality Missing From The Slide Deck
The classic bull case assumes three or four national networks politely splitting the market. In practice, capacity gets resold. Smaller brands rent the big networks and undercut them on price. Cable and fiber operators offer bundled service. Fixed wireless eats into the low end. Satellite providers take the customers nobody else can reach.
That’s fine — the pie is huge — but it means pricing power is weaker than the marketing suggests. When two major players run the same promotion at the same time, margins compress and nobody wins except the customer.
How To Compare Two Telecom Stocks Without Fooling Yourself
- Normalize on service revenue growth, excluding equipment sales.
- Build free cash flow after capex and spectrum.
- Check leverage and interest coverage.
- Look at ARPU and churn specifically in the highest-value customer tier.
- Compare capital intensity across a full upgrade cycle, not a single year.
- Verify the dividend is covered by free cash flow.
- Read the risk factors and the non-standard metric definitions in the annual filings. That’s where the uncomfortable stuff hides.
The Bottom Line
Telecom is a legitimate industry with real, durable demand. It’s also a business where the story is always smoother than the cash flow, and where the headline metrics are picked because they look good, not because they’re informative.
So treat it like any other investment. Ignore the subscriber press release, build your own free cash flow number, subtract the debt, and see what’s left. If the dividend survives that math and the leverage isn’t creeping upward, you’ve got something real. If it only works when you use the company’s own adjusted figures, you don’t own an income stock — you own a story with a coupon attached.