EBITDA is the common language of private equity. Deals are priced as a multiple of it, debt is sized against it, covenants are written around it, and management incentive plans usually pay on it. That makes sense as a measure. It is a reasonable proxy for the operating earnings of a business before financing and accounting choices.

The trouble starts when the measure becomes the objective. EBITDA excludes the capital spending that keeps an equipment business running, and it rewards cuts to expenses whose payoff arrives later: maintenance, training, supervision, and sales coverage. When a management team is paid to raise EBITDA and little else, it will find those levers. The result looks like margin expansion. In practice it often works like a loan against the business’s future, taken out at a rate nobody has disclosed.

Why the pressure is rising

The math of private equity returns has become less forgiving. Bain’s 2026 Global Private Equity Report estimated that a typical deal in the 2010s needed about 5 percent annual EBITDA growth to return 2.5 times invested capital over five years, helped by rising multiples. Today, with multiples flat, a typical deal needs 10 to 12 percent annual growth to reach the same result.

That gap has to come from somewhere. Some of it will come from real operating improvement. Some of it will come from decisions that raise reported earnings now at the expense of later ones. Research suggests managers are willing to make that trade. In a survey of 401 financial executives published in 2005, Graham, Harvey, and Rajgopal found that 80 percent would cut discretionary spending on research, advertising, and maintenance to meet an earnings target. More than half would delay a new project to hit a target, even at a small cost in value. That study covered public companies reporting to analysts, but the incentive is the same wherever pay depends on a single number.

What EBITDA leaves out

In asset-heavy businesses, the gap between EBITDA and cash is large and permanent. United Rentals, which describes itself as the largest equipment rental company anywhere, reported 2025 adjusted EBITDA of $7.3 billion on $16.1 billion of revenue, a 45.5 percent margin. Gross rental capital expenditures were $4.2 billion, about 57 percent of adjusted EBITDA, and rental equipment depreciation was $2.7 billion. A fleet does not replace itself, and the cost of replacing it never appears in EBITDA.

That makes several kinds of decisions look better on an EBITDA basis than they are.

Lever Effect on EBITDA now Where the cost shows up later
Cut preventive maintenance on fleet or customer equipment Higher Repair costs, downtime, shorter asset life
Capitalize rebuilds or repairs that were once expensed Higher Capital spending rises, cash is unchanged
Reduce technician training and apprentice hiring Higher Lower first-time fix rates, more overtime, attrition
Thin out field supervision and branch management Higher Drift in quality, safety, and utilization
Freeze sales coverage in new territories Higher Weaker pipeline 12 to 24 months out
Treat recurring costs as one-time adjustments Higher adjusted EBITDA The costs return, and the multiple paid on them does not

The incentive is easy to see in a hypothetical. Consider a rental and service business with $60 million of revenue and $12 million of EBITDA, valued at 8 times. Cutting $600,000 from preventive maintenance and training raises EBITDA by 5 percent and, at the same multiple, adds $4.8 million of equity value on paper. If those cuts raise repair costs, overtime, and downtime by more than $600,000 a year within two or three years, the value was never there. Someone still pays for it, either the next buyer or the same sponsor later in the hold.

None of these decisions is wrong by definition. A business can be overstaffed, overmaintained, or spending on sales in territories that will never pay back. The question is whether the reduction reflects a better way of operating or a deferral. EBITDA cannot tell the difference. Operating data usually can.

The capitalization row deserves particular attention in equipment businesses. The line between a repair, which is expensed, and a rebuild that extends asset life, which is capitalized, involves judgment. A change in that policy can move millions of dollars out of operating expense without changing what the company spends.

The addback problem

Reported EBITDA is often not the number on which a deal is priced. Adjusted EBITDA adds back costs management describes as one-time or nonrecurring, plus savings it expects but has not yet achieved. The evidence suggests those adjustments are optimistic.

A 2018 S&P Global Ratings study, reported by Institutional Investor, tracked companies involved in deals in 2015 and found that EBITDA came in 29 percent below projections in 2016 and 34 percent below in 2017. Addbacks accounted for 47 percent of EBITDA in leveraged buyouts. A later S&P study, reported by Bloomberg in 2023, found that 93 to 97 percent of speculative-grade companies that announced acquisitions from 2017 to 2019 missed their projections in the first year after the deal.

Addbacks are not inherently suspect. Real integration costs and real one-time events exist. But a restructuring charge that appears in three consecutive years is an operating cost. A buyer that pays a multiple on adjusted EBITDA is paying a multiple on management’s forecast, and sponsors who build their own value creation plans on the same adjustments inherit the same risk.

Two tests catch most of the problem. First, line up the addbacks from the last three years and look for categories that repeat. Second, trace every run-rate saving to a specific action, a date, and the cost line it reduces. Savings that cannot be traced are forecasts, and they belong in the model as forecasts, with a probability attached.

Longer holds change who pays

Deferred costs used to be easier to pass on. A sponsor could hold a business for four or five years, sell it on trailing EBITDA, and leave the maintenance backlog to the next owner. That is harder now. Bain found that buyout holding periods at exit have stretched to about seven years, up from five to six years between 2010 and 2021, and that almost 40 percent of portfolio companies have been held for more than five years.

A longer hold means the sponsor is increasingly the owner who pays for the deferral. Fleet repair costs rise, technician turnover climbs, and customers who received less service renew at lower rates, all within the same fund. Buyers’ diligence has also caught up. Fleet age, maintenance backlogs, and addback history are now routine diligence requests.

The better news is that operating improvement under private ownership is real when it comes from operations. A study by Davis, Haltiwanger, and colleagues of about 9,800 US buyouts from 1980 to 2013 found that labor productivity at target firms rose an average of 8 percent relative to comparable firms. Gains were smaller in deals done when credit was easy, which the authors suggest may reflect returns driven more by financial engineering than by operating change.

What to measure alongside EBITDA

The answer is not to abandon EBITDA. It is to pair it with a small number of measures that reveal whether earnings growth is coming from better operations or from deferred costs.

Measure What it protects against
Cash conversion: EBITDA less maintenance capital spending, as a share of EBITDA Deferred fleet replacement and capitalization shifts
Average fleet age and repair cost as a share of rental revenue Deferred fleet maintenance
Preventive maintenance completion and backlog Deferred customer service
Technician headcount, attrition, and training hours Erosion of delivery capacity
First-time fix and repeat visit rates Quality decline
Addbacks as a share of reported EBITDA, and how often “one-time” items recur Inflated adjusted EBITDA
Quoted pipeline and new account wins by territory Future revenue given up for current margin

Three practices make these measures useful. First, set a maintenance spending floor in the annual budget and require board approval to go below it. Second, tie part of management’s incentive to cash conversion and one or two operating health measures, not EBITDA alone. Third, review capitalization policy and addback history at every board meeting where adjusted EBITDA is presented.

There is a trade-off. Too many measures dilute focus, and management teams can game a dashboard as easily as a single number. Four or five well-chosen indicators, reviewed consistently, are usually enough.

EBITDA is a fair measure of what a business earned. It says very little about what the business must spend to keep earning it. Investors who manage to EBITDA alone will usually get the number they asked for, and pay for it later.

Sources