EV/EBITDA Ratio: Formula, What It Means & Industry Benchmarks

Why this multiple exists

The P/E ratio breaks the moment you try to compare companies with different capital structures. Take two manufacturers with identical plants, identical products, identical operating margins. One funded expansion with equity; the other borrowed heavily. The leveraged company now runs $200M in annual interest expense. That comes straight out of net income, making it look twice as expensive on P/E — not because its business is worse, but because of a financing choice. The EV/EBITDA ratio — enterprise value to EBITDA — was built to strip that distinction out entirely. When capital structures are comparable, P/E is still the faster first read — see our P/E ratio by industry benchmarks for live trailing and forward sector multiples to pair with the EV/EBITDA lens.

The same problem appears across tax jurisdictions. A company domiciled in Ireland faces a 12.5% statutory rate; a comparable U.S. company faces 21% federally plus state taxes. Compare them on P/E and you're measuring tax strategy as much as business quality. Add differing depreciation schedules — one company uses a 20-year useful life on equipment another depreciates over 10 years — and the income statement reflects accounting choices as much as operating performance.

EV/EBITDA collapses all of that to a single question: what does the total business cost, and what does it earn before those noise factors enter the picture? Private equity firms made the multiple standard practice precisely because they operate with debt-heavy structures. When you're planning to refinance a target after acquiring it, the existing capital structure is irrelevant. What matters is the operating cash generation the business can service debt against. EV/EBITDA answers that question — at the cost of some honesty about real cash costs, which we'll get to.

What 'EV' actually includes

Enterprise value is the total cost of acquiring a business outright — not just buying the stock, but absorbing every claim against it. The formula: market capitalization (share price × shares outstanding), plus total debt (short-term and long-term), plus preferred equity, minus cash and short-term investments.

Debt gets added because when you buy a company, you inherit its obligations. Think of buying a house with a mortgage: if you pay $500,000 for a property carrying a $300,000 mortgage, the total consideration is $800,000 — you're now on the hook for that debt. A company with a $3B market cap and $2B in net debt has a $5B enterprise value. A buyer has to make both equity holders and bond holders whole.

Cash gets subtracted because it offsets the purchase price. A buyer can theoretically extract the cash immediately after closing. A company sitting on $1B in cash and trading at a $4B market cap has a $3B enterprise value, before adding debt. The cash belongs to whoever owns the company, so it lowers what you're actually paying for the operating business underneath.

The practical implication: two companies with identical stock prices but different debt loads are not priced the same. A $20 stock at a company with $5B in debt is a more expensive acquisition than a $20 stock at a company with no debt. EV captures that. Market cap alone does not.

What 'EBITDA' actually measures — and what it deliberately ignores

EBITDA is not a GAAP metric. No accounting standard requires it to be reported. But every major financial data provider calculates it the same way: start from operating income (EBIT), add back depreciation and amortization from the cash flow statement. Or work down from net income: add interest expense, income taxes, depreciation, and amortization. What you get approximates cash generated by operations before the cost of financing the business, paying taxes, or writing down long-lived assets.

Each stripped item has a rationale. Interest is a capital structure artifact — two operationally identical businesses financed differently will show different net income purely because of interest expense. Taking it out lets you compare the underlying business without financing noise. Taxes are a jurisdiction and corporate structure artifact — effective rates vary enormously based on where the entity is incorporated, what credits it has accumulated, and what intercompany arrangements it uses. D&A is an accounting artifact — the same physical asset can be depreciated over dramatically different useful-life assumptions, and amortization of intangibles often reflects purchase accounting from past acquisitions rather than current operations.

Strip all three and you get an approximation of operating cash generation that holds up for cross-company comparisons within sectors where capital structure varies but business models are similar. The formula works cleanly there. Where it stops working is when the businesses being compared differ significantly in how much physical capital they must maintain to keep operating. That's the flaw — and it's structural, not accidental.

Negative EBITDA breaks the multiple entirely. When a company's operating earnings are negative — common for pre-profit growth companies still burning cash — the denominator goes below zero and the ratio becomes mathematically undefined and financially meaningless. You cannot say a stock is "cheap" at −15× EV/EBITDA; a negative multiple tells you nothing about value, only that the tool no longer applies. In that situation, reach for EV/Revenue to screen pre-profit businesses on the size of the opportunity, and fall back to forward EV/EBITDA only when there is a credible, near-term path to positive operating earnings you can actually underwrite. If the profitability timeline is vague, no multiple built on EBITDA will help — the honest read is that the business is not yet valuable on an earnings basis.

The flaw: depreciation is not always fictional

Depreciation is described, correctly, as a non-cash charge. No wire transfer happens when accountants record it. But in capital-intensive businesses, it represents something real: the gradual consumption of physical assets that will eventually need to be replaced. That replacement costs money. When EBITDA strips out D&A, it strips out an accounting entry that, in asset-heavy industries, closely approximates a future cash outflow.

Consider two companies, each reporting $500M in annual EBITDA and each trading at 10x — implying a $5B enterprise value. Company A is a software platform: $30M in annual D&A (servers, office lease improvements), $20M in capex. EBITDA and free cash flow are nearly identical. The 10x multiple is roughly 10x on real cash earnings. Company B is a steel manufacturer: $180M in annual D&A across blast furnaces, rolling mills, and continuous casters — and $220M in actual capex to maintain and replace that equipment. EBITDA is $500M. Free cash flow, after capex and before accounting for it, is closer to $280M. The "same" 10x multiple is actually 18x on real cash. Same number. Very different businesses.

Airlines make this extreme. A major U.S. carrier might trade at 6–8x EV/EBITDA, which sounds almost cheap in most contexts. But aircraft are extraordinarily expensive to maintain and replace. Engine overhauls alone run millions per unit. Heavy maintenance checks on a narrow-body aircraft — the full structural inspection, not the routine between-flight checks — can run $3–5M per airframe. Fleet renewal across hundreds of aircraft requires billions annually. Spread that against the EBITDA and you understand why airline stocks have historically destroyed shareholder capital despite reporting positive EBITDA every year. The multiple was never telling the truth about the underlying economics.

Pipelines illustrate this clearly without the operating drama of airlines. A midstream company running $4B in annual EBITDA, with thousands of miles of steel in the ground, might carry $1.5B in annual depreciation. EV/EBITDA treats that $1.5B as a non-event. But when pipelines need relining, compressor stations need rebuilding, and valves and instrumentation need replacement across 10,000 miles of infrastructure, that cash will come from somewhere. It won't come from the tooth fairy.

Warren Buffett made precisely this point in Berkshire's 1999 letter: "Does management think the tooth fairy pays for capital expenditures?" BNSF, Berkshire's railroad, runs roughly $6–7B in EBITDA annually and also runs roughly $3.5B in annual capex to maintain 32,500 miles of track, bridges, tunnels, and rolling stock. The true earnings power, measured against what actually stays in the business after keeping the infrastructure intact, is materially lower than the EBITDA number implies. The market partially adjusts for this — railroad multiples run at 8–10x, well below the S&P median — but the adjustment is intuitive and approximate. The only way to make it precise is to do the math yourself.

When the multiple is the right tool

EV/EBITDA earns its place in specific situations. Knowing those conditions is as important as knowing the structural flaw.

• M&A transaction comps. Investment banks building fairness opinions and pitch books use EV/EBITDA almost universally for precedent transaction analysis. The buyer in an acquisition is going to refinance the target's debt anyway — the existing capital structure doesn't matter. What matters is what EBITDA they're acquiring and what multiple recent comparable deals paid. For this purpose, the multiple is exactly right.
• Cross-border sector comparisons. Statutory corporate tax rates range from under 10% to over 30% across developed markets. A German industrial at 30x P/E and an equivalent U.S. industrial at 22x P/E might be priced identically on an operating basis — the spread is tax rate, not business quality. EV/EBITDA neutralizes most of that and lets you compare underlying economics directly.
• Sectors with uniform capex intensity. If every company in the comparison set has similar capex-to-revenue ratios — U.S. wireless carriers, large-cap quick-service restaurant chains, specialty chemicals — EV/EBITDA differences across the group more likely reflect real valuation differences than capex adjustment issues. The condition matters: verify the capex profiles are actually similar before assuming they are.
• Infrastructure build-out phases. A data center REIT or cell tower company in aggressive expansion will show suppressed free cash flow because capex massively exceeds D&A on the assets already running. EBITDA captures the economics of the operating asset base without penalizing the company for building future capacity. In this narrow context, EV/EBITDA is a more useful lens than EV/FCF because FCF is being temporarily distorted by deliberate growth investment.
• Quick screens for outliers. When ranking 30 companies in a sector by relative cheapness before doing deeper work on the cheapest ones, EV/EBITDA is fast and directionally useful. The danger is stopping there. Use it to identify which companies deserve closer examination — not to close the investment argument.

Worked example: How to calculate EV/EBITDA for Middleby (MIDD)

The four inputs you need — market cap, total debt, cash, and EBITDA — all come from public filings. Here is the full path from balance sheet to multiple using Middleby Corporation (MIDD) FY 2024 figures.

Step 1 — Build Enterprise Value

EV = Market Cap + Total Debt − Cash = $5.02B + $2.1B − $0.42B = $6.7B

Step 2 — Compute EBITDA

EBITDA = Operating Income + D&A = $0.72B + $0.12B = $0.84B

Operating income comes from the income statement; D&A from the cash flow statement (or income statement footnotes). Adding D&A back strips the non-cash depreciation charge to get the operating cash proxy.

Step 3 — Divide

EV/EBITDA = $6.7B ÷ $0.84B = 7.98×

Step 4 — Compare to the sector median

Industrial Machinery companies typically trade at 10–12×. At 7.98×, MIDD trades at a discount to the sector median — reflecting either a value opportunity or slower expected organic growth relative to peers.

EV/EBITDA by sector: what counts as cheap?

There is no universal "cheap" multiple. A number that screams bargain in software would be a warning sign in energy, and one that looks expensive for a pipeline might be fair for a drug franchise. What counts as cheap is set by how much of each sector's EBITDA survives the trip to free cash flow — capital-light businesses keep almost all of it and command high multiples; capital-heavy ones surrender a large slice to capex and trade low. The table below anchors the ranges you should expect before you decide a given stock is mispriced.

Ranges reflect S&P 500 constituent medians circa 2024; they compress in high-rate environments and expand in low-rate ones. Source: Damodaran (2024 dataset), cross-checked against FactSet sector medians. Read the ranges as a starting reference, not a verdict: a company at the low end of its band may be genuinely cheap, or it may be signalling a broken thesis the market has already priced in. The band tells you which question to ask — it does not answer it. See live sector data in our EV/EBITDA by Industry benchmark table.

EV/EBIT vs. EV/EBITDA: When Depreciation Changes the Picture

EV/EBIT includes D&A as an expense; EV/EBITDA adds it back. The gap is largest for capital-intensive businesses where D&A represents real economic cost — telecom towers, oil rigs, manufacturing plants that will eventually need full replacement. For asset-light software, D&A is minor so both multiples converge and the distinction rarely changes the investment conclusion. Use EV/EBIT when capex consistently exceeds depreciation — a sign D&A understates replacement cost and EV/EBITDA is doing flattering work on the headline number.

P/E, EV/EBITDA, EV/EBIT, EV/FCF: which to reach for?

The four multiples investors reach for most differ in exactly one way: how much real capital cost they let through to the ratio. P/E leaves everything in and breaks on debt differences; EV/EBITDA strips out the most and flatters capital-heavy businesses; EV/EBIT and EV/FCF add the asset cost back, one via accounting depreciation and one via actual cash spent. Reading them side by side shows why a single multiple never settles a valuation on its own.

The practical workflow: start with EV/EBITDA for a fast, structure-neutral screen, then walk toward EV/FCF to see how much of that headline cheapness survives real capital spending. The wider the gap between the two, the more the EBITDA multiple was doing to make the business look inexpensive.

How to check whether a low multiple is real

The translation takes three steps and requires two inputs you can pull from any financial statement: D&A as a percentage of revenue (from the cash flow statement or income statement footnotes) and capital expenditures as a percentage of revenue (from the investing section of the cash flow statement).

Step one: start with the EV/EBITDA multiple as given. Step two: compute EV/EBIT. Because EBIT = EBITDA minus D&A, the multiple expands by a factor of EBITDA margin divided by EBIT margin. A business with a 20% EBITDA margin and 8% D&A runs a 12% EBIT margin — the EV/EBIT multiple is 20÷12 = 1.67× larger than EV/EBITDA. A 9x EV/EBITDA becomes roughly 15x EV/EBIT. Step three: substitute actual capex for D&A using the same structure. FCF margin equals EBITDA margin minus capex as a percent of revenue (a simplified pre-tax estimate). That ratio between EBITDA margin and FCF margin is how much larger EV/FCF is than EV/EBITDA.

The gap between EV/EBITDA and EV/FCF tells you how much work the multiple is doing to make the company look cheap. A tight gap — 2–3x — means capex closely tracks depreciation and EBITDA is a reasonable proxy for earnings. A wide gap — 10–15x — means the business eats a large fraction of its EBITDA in capital spending before it can return anything to investors. The calculator below makes this concrete. Enter a real company's numbers and see where the multiple actually lands once capital costs enter the picture.

For a live example, see how a capital-heavy steelmaker like Cleveland-Cliffs (CLF) — where depreciation and capex run large — carries its EV/EBITDA in our stock analysis.

A cheap multiple only pays off if the capital already in the business earns its keep. Pair EV/EBITDA with ROIC to assess whether that capital is deployed efficiently — see our Return on Invested Capital guide.

Translating the multiple: Middleby Corporation (MIDD)

Middleby Corporation (MIDD) makes commercial cooking, food-processing, and residential kitchen equipment — a real industrial, but a comparatively capital-light one. That makes it a useful case for the translation, because unlike a steel mill its capex actually runs below its depreciation, so the multiple stays roughly honest as you walk it down. Using Middleby's most recent trailing figures — an enterprise value near $8.0B, EBITDA around $700M, depreciation and amortization of roughly $105M, and capital expenditures of about $72M — here is the three-step translation.

The lesson is in the tight spread. Middleby's EV/FCF (~12.7×) sits close to its EV/EBITDA (~11.4×) — barely more than a turn apart — because it does not need to spend heavily to keep its plants running. Run the same three steps on a steelmaker or an airline and the EV/FCF blows out to two or three times the EBITDA multiple. That gap, not the headline number, is what separates a genuinely cheap operating business from one that only looks cheap until the capex bill arrives. See the live figures on our Middleby (MIDD) stock analysis.

Try it: From EV/EBITDA to EV/FCF

Enter any company's EV/EBITDA multiple and its cost structure below. The translator shows what that multiple actually implies once D&A and capex enter the picture. Toggle between the SaaS and manufacturer presets to see an identical starting multiple land in very different places after the adjustment — that's the lesson in one interaction.

Questions worth asking

Keep reading

• EV/EBITDA Screener — Filter Stocks by Multiple
• DCF Calculator — Build a Full Intrinsic Value Model
• WACC Calculator — Set the Discount Rate for Your DCF
• What Is Enterprise Value?
• How to Read an Income Statement
• Free Cash Flow: What It Is and Why It Matters
• P/E Ratio Explained
• How to Compare Stocks in the Same Sector