EV/EBITDA

Starter — for Never bought a share

EV/EBITDA compares the full price of a business — shares plus its debts minus its cash — with the profit it makes before interest, tax and accounting for wear and tear.

Why not just use the share price?

Buying all of a company means buying the shares and taking on its debts. If two companies have the same share price but one owes four times as much, the indebted one is the more expensive purchase. Enterprise value fixes this: it adds the debt and subtracts the cash, so you are looking at the price of the whole business.

Why EBITDA as the bottom half

EBITDA is profit before interest, tax, depreciation and amortisation. It is used here because the top half (enterprise value) is also before interest, so the two halves match. It is a rough measure of how much cash the core operations generate before the cost of the equipment and the tax bill are taken out.

What a number means

An EV/EBITDA of 12× means you are paying twelve years of that operating profit for the business. Lower is cheaper only if the businesses are comparable — same industry, same growth, and similar need to spend on equipment. A capital-heavy company deserves a lower multiple than a capital-light one.

What to take away

• Enterprise value = shares + debt − cash: the price of the whole business.
• EBITDA = profit before interest, tax and depreciation. It matches the top half.
• Compare it only within similar businesses, because capital needs differ.

Medium — for Invested before, reads the news

EV/EBITDA compares the price of an entire business with its pre-financing operating profit, which makes it the standard tool for comparing companies with different debt loads.

Building enterprise value properly

Start with market capitalisation at the current share price, using the diluted share count. Add total debt — short and long term, plus capitalised leases and pension deficits where material. Add preferred stock and minority interests, since the acquirer of the whole business buys those too. Subtract cash and equivalents, and subtract any surplus investments that do not contribute to operating profit. The result is what a buyer of the entire company would pay.

What EBITDA includes and leaves out

EBITDA starts from operating profit and adds back depreciation and amortisation. Because depreciation is the accounting spreading of past capital spending, excluding it makes EBITDA flattering for businesses that must constantly spend to maintain capacity: airlines, utilities, telecoms, mining and heavy manufacturing. For those, EV/EBITDAR after lease costs, or EV/(EBITDA − maintenance capex), is the more honest multiple.

Why the metric exists at all

EV/EBITDA became standard in leveraged buyouts and credit analysis: it approximates the cash available to service debt before financing costs, which is exactly what a lender or an acquirer needs. That origin explains both its popularity and its limits. It is a deal metric, not an ownership metric — a shareholder owns what remains after capex, taxes and debt service, which is why a valuation conversation pairs EV/EBITDA with FCF.

Cross-industry comparison is a trap

Multiples cluster by industry because capital intensity, growth and margins differ structurally. Software trades at 20×+ EBITDA with little capex; a steel mill trades at 5× with enormous capex. Judging either against the other says nothing. Even within an industry, check the accounting lives: a company that capitalises development costs spreads them out, while one that expenses them shows a lower EBITDA today and a higher one later.

What to take away

• EV = diluted market cap + debt + leases + preferred + minorities − cash and surplus assets.
• EBITDA ignores capex, working capital and cash taxes — pair it with FCF for capital-heavy businesses.
• The metric came from deal-making and credit; it measures cash available to service debt, not to owners.
• Compare within an industry, and normalize accounting lives and adjustments before ranking.

High — for Works with this number already

EV/EBITDA is the lingua franca of deal and credit work, and precisely because of that, its adjustments are negotiated — the fastest way to tell a serious comparable set from a sales pitch is to read the bridge from reported to adjusted EBITDA.

The adjustment bridge is the analysis

Every "adjusted EBITDA" is reported EBITDA plus add-backs: stock-based compensation, restructuring, acquisition integration, impairments, one-time legal costs, discontinued operations, and pro-forma synergies that do not exist yet. Two rules keep this honest. First, an adjustment must be non-recurring — test it against the last five years of that line item; if it appears every year, it is a cost. Second, the same add-backs must be applied to the peers, because an adjustment granted to one company and denied to another is not a comparison, it is a conclusion.

Capital intensity and the EBITDA/hour mismatch

The gap between EBITDA and operating cash flow is the story for asset-heavy businesses. Two airlines with identical EBITDA can differ by a factor of two in free cash flow because of fleet age and lease structure. Where depreciation materially understates the true maintenance cost — a common situation when assets are old or lives have been extended — the discount is in the cash flow, not in the multiple. A useful discipline is to compute EV/(EBITDA − maintenance capex) for capital-heavy names and check that ranking against the plain multiple.

Multiples, terminals and consistency

In an LBO or DCF, the exit multiple and the discount rate must belong to the same business description. Using a 10× EBITDA exit assumes the buyer at exit prices the asset the same way you did: same adjustment policy, same lease treatment, same growth. Where that assumption cannot be defended, the exit is better modelled on a normalized cash flow or a replacement-cost argument. The most common modelling error is not an arithmetic mistake; it is applying one company's multiple to another company's definition of EBITDA.

When to reach for something else

For financial institutions, EBITDA is meaningless — interest is the revenue. For real estate, use net operating income and cap rates. For pre-profit growth companies, EV/revenue or EV/gross profit carries the analysis, with a path-to-profitability assumption stated explicitly. For deeply cyclical names, normalize both numerator and denominator across the cycle: a trough-EBITDA denominator produces a multiple that looks catastrophic at exactly the moment the asset becomes interesting.

The reading that pays

EV/EBITDA answers "what would this whole business cost, relative to the operating profit it currently declares, compared with similar businesses under the same accounting?" It does not answer whether the profit converts to cash, whether the debt is serviceable, or whether the assets need replacing. Used as the first sort in a peer set and the last check before a leverage conversation, it is efficient. Used as the final word on value, it is the most expensive multiple to trust.

What to take away

• Read the reported-to-adjusted EBITDA bridge and apply the same add-back policy to every peer.
• An adjustment that appears every year is a cost, not an adjustment.
• For capital-heavy businesses, EV/(EBITDA − maintenance capex) is the multiple that ranks the same way the cash flows do.
• Keep the numerator and denominator consistent — and switch metrics entirely for banks, REITs and pre-profit companies.