LTM vs NTM EBITDA: Which Earnings Number Sets the PE Entry Multiple

Every LBO starts with a purchase enterprise value. That number is almost always a multiple times some earnings base. The fight is which earnings base. Management wants NTM or "run-rate." The sponsor wants QoE-adjusted LTM unless there is a clean reason to pay for forward earnings. Interviews hammer this because a one-turn mistake on the wrong EBITDA is millions of dollars of overpayment.

The three bases you will hear

Entry EV = Entry EBITDA × Entry Multiple

If Entry EBITDA is wrong, every downstream number (debt capacity, equity check, MoIC) is wrong too.

Why sponsors default to LTM

LTM is audited (or at least historical). You can diligence it. QoE exists to clean it. Paying 10x LTM is a statement about what the business already earned.

Paying 10x NTM is a statement about what you believe it will earn. That is fine when growth is contracted (backlog, price increases already signed). It is not fine when NTM is a hope chart.

Worked gap example

That is a $100M gap for the same "10x" language. Always ask: 10x of what?

Translate NTM marketing into LTM language

If teaser says 9.5x NTM and NTM/LTM = 60/50 = 1.2x:

Implied LTM multiple = 9.5 × 1.2 = 11.4x

Take that number to IC, not the teaser headline alone.

Run-rate: useful and abusable

Run-rate helps when LTM is messy:

• Mid-year acquisition of a chunky add-on
• One-time cost that will not repeat (already backed out in QoE, but timing still skews LTM)
• Clear step-up in pricing that started two months ago

Run-rate abuse looks like:

• Annualizing a peak season quarter for a seasonal business
• Annualizing post-synergy EBITDA before costs to achieve are spent
• Annualizing a month that included a pull-forward of customer orders
• Annualizing a partial year that excludes a known cost reset

If you use run-rate in an interview, say what you annualized and what you excluded.

How QoE sits in the middle

QoE does not automatically mean "use LTM." QoE says: here is normalized historical earnings. Sponsors often underwrite:

Purchase EBITDA = QoE-Adjusted LTM EBITDA

Then build a separate operating case for Year 1 that may be above or below that number. The purchase price is tied to the QoE base. The operating case is the plan. Mixing them (paying for NTM while modeling a soft Year 1) is how models lie.

Add-backs get multiplied

A soft $2M add-back at 10x is $20M of purchase price. That is why QoE arguments are not academic. Challenge every add-back that is about to be capitalized into entry EV.

Multiple conventions in comps and teasers

Trading comps are often shown on LTM and NTM (EV/LTM EBITDA and EV/NTM EBITDA). Precedent transactions are usually LTM at announcement, sometimes with a "next twelve months" footnote if the target disclosed a forecast.

In a paper LBO, if they give you "buy at 10x" and only one EBITDA figure, clarify LTM vs NTM out loud. If they give both, state which one you used for entry EV.

Interview answers that sound clean

"Would you pay for NTM?"
"Only if the step-up is contracted or highly visible: signed price increases, cost-out already executed, backlog converting. Otherwise I pay off QoE LTM and treat NTM upside as equity IRR, not as purchase price."

"Management says LTM is understated because of a warehouse fire."
"That is a QoE add-back conversation, not an automatic jump to NTM. I want the normalized LTM, diligence the add-back, then decide."

"Banker is marketing 9.5x NTM."
"I convert that to an implied LTM multiple using their own bridge. If LTM is $80M and NTM is $100M, 9.5x NTM is 11.9x LTM. That is the number I take to IC."

"What about run-rate EBITDA after a price increase?"
"I will consider it if the new pricing is already in market and churn has not spiked. I still haircut versus full annualization if only two months of evidence exist."

Tie to debt and equity

Senior lenders often size on LTM or a lender-adjusted EBITDA, not management NTM. You can underwrite a purchase at NTM and still find the credit agreement caps leverage on a tighter definition. That creates an equity gap at close.

Sources must fund: Entry EV + fees − debt sized on lender EBITDA

If sponsor EBITDA ≠ lender EBITDA, the equity check moves. This is a common live-deal surprise and a great interview point.

Mini numeric

Purchase at 10x NTM on $60M = $600M EV.
Lender allows 5.0x on $50M LTM = $250M debt.
Equity need ≈ $350M plus fees, not the lower equity implied if you naively levered NTM.

Common mistakes

Practice

On your next LBO case, force yourself to show entry EV three ways: LTM, run-rate, and NTM. Pick one for the base case and put the others in a sensitivity. Then recompute debt capacity on the lender base and show the equity gap. That is how deal teams actually argue about price, and it is exactly what interviews are testing when they ask "10x of what?"