What the EV/EBITDA multiple tells you
EV/EBITDA says how many years of current operating earnings the market, or a buyer, is paying for the whole business. At 8.0x, the enterprise costs eight times what it earns before interest, tax, depreciation and amortisation in a year. It is the closest thing corporate finance has to a universal price tag, and it dominates mid-market M&A, leveraged loan pricing and private equity screening.
It earned that position by being consistent on both sides of the fraction. Enterprise value is the claim of every capital provider — equity, debt, preferred and minority holders. EBITDA is the profit available to all of them, measured before interest takes the lenders' share and before tax. Numerator and denominator therefore cover the same set of claimants, which is exactly what the P/E ratio does for equity holders alone. That symmetry is why EV/EBITDA is comparable across companies with different leverage, and why P/E is not.
It also strips out two accounting choices that make raw profits hard to compare: depreciation policy and the amortisation of acquired intangibles. A company that grew by acquisition carries amortisation that an organically grown twin does not, and EBITDA ignores it. That is a feature when you are comparing operating performance and a bug when the assets genuinely need replacing, which is the central trade-off of the whole measure.
Run in reverse, the multiple becomes a valuation. Multiply your EBITDA by the multiple the peer group trades at and you have an implied enterprise value; subtract net debt and the other claims and you have what the equity is worth. That reverse calculation is the whole of comparable company analysis, the convention this calculator follows: it is how a business broker prices a company and how a banker sets a range in a fairness opinion.
The formula, and the bridge back to a share price
The multiple itself is a single division: EV/EBITDA = enterprise value ÷ EBITDA. Everything difficult is in the definitions of the two inputs.
Which EBITDA? Trailing twelve months is the most common because it is a fact rather than a forecast, but forward EBITDA is what a buyer is really paying for. A forward multiple is almost always lower than a trailing one for a growing company, so quoting a multiple without saying which window you used is meaningless. Adjusted EBITDA changes the answer again: if the peer set is quoted on adjusted figures and you use unadjusted for your subject, the comparison is broken before it starts.
Which enterprise value? Market capitalisation on fully diluted shares, plus total debt and capitalised leases, plus preferred stock and noncontrolling interest, less cash and equity-method investments. Leaving out noncontrolling interest while consolidating 100% of a subsidiary's EBITDA is the most frequent mechanical error, and it always makes a company look cheaper than it is.
To turn a multiple into a share price, run three steps. First, implied enterprise value = multiple × EBITDA. Second, implied equity value = implied enterprise value − net debt − preferred − noncontrolling interest. Third, implied price = implied equity value ÷ fully diluted shares. Note what this structure does: net debt does not scale with the multiple, so the entire change in enterprise value from a change in multiple lands on the equity. That is why the equity value of a leveraged business is so much more sensitive to the multiple than the equity value of an unleveraged one, and the reference table below quantifies it.
Worked example: pricing a $890M EBITDA business at 9.5x
Your subject company has enterprise value of $7,325M and trailing EBITDA of $890M. Net debt is $1,420M, noncontrolling interest is $85M, and there are 120 million fully diluted shares. The comparable set trades at 9.5x.
- Current multiple. 7,325 ÷ 890 = 8.23x.
- Implied enterprise value at the peer multiple. 9.5 × 890 = $8,455M.
- Implied equity value. 8,455 − 1,420 − 85 = $6,950M.
- Implied price per share. 6,950 ÷ 120 = $57.92.
- Price implied by your EV input. (7,325 − 1,420 − 85) ÷ 120 = 5,820 ÷ 120 = $48.50.
- Gap. 57.92 ÷ 48.50 − 1 = +19.4%.
Notice the leverage effect in those numbers. Enterprise value rises from $7,325M to $8,455M, a gain of 15.4%, but the equity value rises from $5,820M to $6,950M, a gain of 19.4%. The extra 4 points come from the $1,505M of net debt and other claims that do not move when the multiple moves. Re-rate the enterprise, and the equity gets all of the gain.
How to read the multiple you get
There is no universal fair multiple, so resist any single benchmark number. What a multiple should be is set by three things, and you can reason about each of them from first principles.
Growth. A business whose EBITDA compounds quickly is worth more years of current EBITDA than one that is flat, because the buyer of the fast-growing company is buying a much larger future stream for the same current denominator.
Capital intensity. EBITDA is measured before capital expenditure, so two companies with identical EBITDA but very different reinvestment needs are not worth the same. Where maintenance capex is large relative to EBITDA, less of that EBITDA reaches the owner, and the multiple should be lower. The gap between the EV/EBITDA and EV/EBIT multiples is a quick read on this: EBIT is after depreciation, so the wider the gap, the more of the EBITDA is being consumed by the ageing of the asset base.
Risk and cyclicality. A cyclical business at the top of its cycle has an inflated EBITDA and will therefore look cheap on a trailing multiple exactly when it is most expensive. This is the single most dangerous failure mode of multiple-based valuation, and it is why credit analysts normalise EBITDA across a cycle before pricing anything.
When you compare your subject with a peer set, do it on the same window, the same adjustment convention and the same bridge. A difference of two turns between two companies is far more likely to be a definitional inconsistency than a genuine valuation gap, so reconcile the definitions before you draw a conclusion. And where the implied equity value comes out negative, the arithmetic is telling you that the multiple does not cover the debt: the value has passed to the creditors, and no per-share figure derived from it is meaningful.
One turn of multiple, four balance sheets
| Net debt | Equity value at 8.0x | Price per share | Equity value at 9.0x | Price at 9.0x | Change in price |
|---|---|---|---|---|---|
| $0M | $800M | $16.00 | $900M | $18.00 | +12.5% |
| $200M | $600M | $12.00 | $700M | $14.00 | +16.7% |
| $400M | $400M | $8.00 | $500M | $10.00 | +25.0% |
| $600M | $200M | $4.00 | $300M | $6.00 | +50.0% |
Every row is 8.0 x 100 = 800 less net debt, divided by 50 million shares; the change is $100M of extra enterprise value divided by the equity value in that row. The enterprise re-rates by 12.5% in every case, and the equity captures all of it.
Pitfalls that produce a wrong multiple
- Mismatched windows. A trailing EV against a forward EBITDA, or a stale enterprise value against a fresh EBITDA, produces a number that means nothing. Date both inputs.
- Adjusted against unadjusted. If the peer multiples come from a screen using adjusted EBITDA, adjust your subject the same way or the gap you find is the adjustment, not the valuation.
- Omitting noncontrolling interest. Consolidated EBITDA includes all of a partly owned subsidiary; enterprise value must include the outside shareholders' claim or the multiple is understated.
- Applying a public trading multiple to a private company unchanged. Public multiples embed liquidity and, in a takeover, a control premium; precedent transaction multiples embed the premium but not the liquidity. Know which one your comparables are.
- Valuing a cyclical at the peak. Trailing EBITDA at a cycle high makes the multiple look low precisely when the risk is highest. Normalise across the cycle before applying any multiple.
- Ignoring lease accounting. An IFRS 16 reporter capitalises leases into debt and keeps them out of EBITDA; a US GAAP peer with operating leases does neither. Compare them without adjustment and you are comparing two different measures.
- Reading the multiple as a return. The inverse of EV/EBITDA is not a yield to shareholders. It is before tax, before interest and before the capital expenditure needed to keep the EBITDA where it is.
When to use a different multiple
EV/EBITDA is the default, not the answer to every question. Reach for something else in four situations.
When EBITDA is negative or near zero, the multiple is undefined. Use EV/revenue or EV/gross profit as a placeholder, and be explicit that you are valuing a business that does not yet make money.
When capital intensity differs sharply across the peer set, use EV/EBIT instead. Because EBIT is after depreciation, it charges each company for the ageing of its own asset base, which is precisely the difference EV/EBITDA hides. For a fuller answer, value the free cash flow directly.
When the capital structure is the point, as in a leveraged buyout or a covenant test, look at debt/EBITDA alongside the valuation multiple. Lenders size facilities in turns of EBITDA, and the multiple you can pay is constrained by the leverage the credit market will fund.
When you need a value rather than a comparison, build a discounted cash flow. A multiple is a shorthand for the assumptions a DCF makes explicit: growth, margin, reinvestment and the cost of capital. The two belong together — run the DCF for the intrinsic answer, then back out the implied exit multiple and ask whether anyone has ever paid it for a business like this one. If the answer is no, the model is wrong somewhere, and the multiple has just told you where to look.