Hold on to this
EBITDA isn't dishonest. It's a metric that tells the truth about some businesses and almost nothing about others. The difference is how much capital a company needs just to stay in place.
Instrument · Drive it
Try it yourself
Load an example
EBITDA ($M)
D&A ($M)
Capex ($M)
Low intensity
EBITDA is a reasonable cash proxy. Capex runs well below D&A — the add-back is defensible here.
What EBITDA Actually Measures
EBITDA is operating profit with two non-cash accounting charges stripped out: depreciation and amortization. That's the whole formula. The question is why stripping them out is ever defensible.
Depreciation spreads the cost of a long-lived asset over its useful life. When a company buys $100 million in equipment with a ten-year useful life, accounting rules charge $10 million to the income statement annually — but all the cash left on day one. In year six, that $10 million depreciation line is a purely notational charge. Adding it back gets closer to what the business actually generates each period.
Amortization works the same way for intangible assets. When Salesforce (CRM) acquires another software company, it records customer relationships and developed technology on the balance sheet and amortizes them over several years. That amortization charge hits the income statement as D&A even though nothing is physically wearing out. Salesforce's acquired customer base doesn't need to be repurchased each year — the charge reflects an accounting rule, not an ongoing cash obligation. For businesses with large acquired- intangible amortization and modest physical capex, adding D&A back is analytically clean. That's the case where EBITDA earns its keep. EBIT adjusted for taxes — called NOPAT — is the numerator in ROIC, the metric that measures how efficiently a company converts invested capital into profit. See the ROIC guide.
Why Companies Love Reporting It
EBITDA almost always looks better than net income. Interest expense, income taxes, and D&A all go in the direction of reducing reported earnings. Strip them out and the number climbs. Management teams have learned this. Many executive compensation packages reference EBITDA targets specifically, which creates a structural preference for leading with it.
None of that is fraud. The cash flow statement shows the full picture plainly. But an investor who reads the press release headline and stops there never sees it. The discipline is to always follow EBITDA with a look at the cash flow statement — specifically, capex and free cash flow.
The Problem: D&A Isn't Free Money
For capital-intensive businesses, depreciation isn't an accounting artifact — it is the income statement delivering an accurate advance warning: you will write that check again. A blast furnace depreciated over 20 years will need to be replaced. Network cables degrade. Aircraft require heavy maintenance every few years and eventual replacement. The annual D&A charge is the accounting system's best estimate of the real ongoing cost of maintaining the asset base. Adding it back assumes the replacement cost goes away. For most physical-asset businesses, it doesn't.
The test is how closely capex tracks D&A. In a steady-state business, the two should converge: you spend roughly what you depreciate, replacing assets at roughly the rate the income statement implies. When capex runs persistently above D&A, the depreciation charge understated the real economic cost. Verizon (VZ) has averaged capex of roughly 150–170% of reported D&A over the past decade — reflecting the ongoing cost of maintaining and extending a national fiber and wireless network. EBITDA at Verizon is a number that requires significant adjustment before it tells you anything about cash generation.
The math is direct. A company with $500M EBITDA, $80M in D&A, and $120M in capex is not generating $500M in operating cash. After capex — a mandatory outflow to keep the network current — it is generating $380M before interest and taxes. That's a 24% gap from the headline number. The Reality Check calculator below makes this concrete for any company you are analyzing. Enter three numbers from the cash flow statement and see the gap.
Businesses that tend to fail the test — where D&A understates real capital needs:
- Airlines (jet maintenance cycles, fleet renewal every 20–25 years)
- Telecom infrastructure operators (fiber, 5G spectrum, switching equipment — continuous build)
- Steel and heavy manufacturing (blast furnaces, rolling mills, chemical reactors)
- Mining (equipment replacement, mine development and reclamation)
- Integrated oil and gas (well drilling and maintenance — depleting reserves require ongoing reinvestment just to hold production flat)
Businesses that tend to pass — where D&A reflects accounting, not real capital needs:
- Enterprise software grown through acquisition (large acquired-intangible amortization, low ongoing physical capex)
- Digital marketplaces and platforms (server infrastructure well below reported D&A)
- Asset-light services — consulting, financial advisory, professional services
For a deeper look at how capex intensity varies by sector, see the capex-intensity benchmarks by industry.
Where EV/EBITDA Actually Earns Its Keep
EV/EBITDA became the standard M&A metric for a specific structural reason: when a company acquires another, one of the first things it does is replace the target's capital structure. The acquired debt gets refinanced, the interest burden changes, the tax situation resets. Stripping out interest and taxes to look at the pure operating business is exactly right when the buyer brings their own balance sheet. The multiple prices the business, not the financing.
Cross-company comparisons within the same capital-intensity tier also hold. Two consumer staples companies with similar capex profiles and geography — EV/EBITDA strips out different leverage decisions cleanly and gives a direct comparison of operating business economics. If one company levered up for an acquisition three years ago and another didn't, their P/E ratios will diverge due to interest expense even if the businesses are operationally identical. EV/EBITDA won't. For a full treatment of the denominator, see the EV/EBITDA guide and the sector benchmarks.
Where EV/EBITDA is the right primary tool:
- M&A deal pricing — standard practice in investment banking and private equity
- Screening peers within the same capital-intensity tier
- Cross-jurisdiction comparisons where effective tax rates differ materially
- LBO analysis, where the buyer will recapitalize the business at close
Where it will mislead you:
- Banks and insurance companies — debt is operational raw material, not financing, so enterprise value becomes meaningless
- Capital-intensive businesses where capex persistently exceeds D&A (use EV/EBIT or EV/FCF instead)
- Companies running large adjusted-EBITDA add-backs that recur annually — check the four-year history of "one-time" charges before using the adjusted number
When to Use Free Cash Flow Instead
A working rule: if capex has run above 50% of D&A for three consecutive years, EV/EBITDA comparisons are unreliable for that business. The metric is adding back a charge the company will spend again — probably at a higher level. FCF yield or free cash flow per share gives the more honest picture. EV/EBIT is a middle path — it reintroduces D&A as a real cost, which is appropriate whenever asset replacement is economically necessary at or above the depreciation rate.
FCF yield — free cash flow per share divided by stock price — measures how much of your purchase price you earn back in actual cash each year. A 6% FCF yield on a stable industrial is a more reliable signal than a 7x EV/EBITDA multiple when that company's capex runs at 140% of D&A. The EV/EBITDA multiple is doing optimistic math. The FCF yield is counting real money.
Sectors where FCF or EV/EBIT is the right primary lens:
- Airlines — Spirit, Delta, United all run capex well above D&A in expansion phases
- Telecom carriers — Verizon, AT&T, T-Mobile network build is continuous and capital-hungry
- Mining and metals — equipment replacement costs are real, large, and inescapable
- Integrated oil and gas — well maintenance and drilling programs deplete assets that must be replenished just to hold production flat
The FCF vs. EBITDA comparison guide walks through the reconciliation in detail, including how to find the relevant line items on a cash flow statement in under two minutes.
Questions worth asking
Is EBITDA the same as operating cash flow?
No. EBITDA is an accounting construct — operating profit with D&A added back. Operating cash flow from the cash flow statement also captures working capital changes: inventory builds, receivables, payables. A company can show rising EBITDA while burning cash if it is stuffing its distribution channel or extending payment terms to customers. Always check both.
Why do analysts use EV/EBITDA instead of P/E?
P/E is capital-structure-dependent — a heavily indebted company shows lower earnings than a debt-free peer running the identical business, purely because of interest costs. EV/EBITDA strips out both debt structure and depreciation policy, making peer comparisons cleaner. It is especially common in M&A because the acquirer replaces the target's capital structure anyway.
Which sectors should I be most skeptical of EBITDA in?
Airlines, telecom infrastructure, capital-intensive manufacturing, mining, and oil and gas. In all of these, capex routinely equals or exceeds D&A — meaning EBITDA adds back a charge the business will absolutely have to spend again. EV/EBIT or FCF yield gives a more honest picture of what the business actually generates.
What is adjusted EBITDA and should I trust it?
Adjusted EBITDA adds back items management calls one-time: restructuring charges, stock compensation, legal settlements. The problem is that some of these — stock comp especially — are real economic costs that recur every year. When adjusted EBITDA runs more than 20% above reported EBITDA, treat the gap as a warning sign, not a gift.
What is a normal EV/EBITDA multiple?
It varies widely. Mature industrials typically trade at 6–10x. Fast-growing software companies can trade at 20–40x. The multiple compresses as capital intensity rises because investors discount the gap between EBITDA and actual cash. Comparing multiples across sectors without adjusting for capital intensity is one of the most common valuation mistakes retail investors make.
Keep reading