Where did private equity returns actually come from?
From three places: cheap leverage, multiple expansion, and operating improvement. For most of the last cycle the first two did the majority of the work. Leverage and multiple expansion together accounted for 59 percent of returns on deals done between 2010 and 2022, on a StepStone Group attribution cited in McKinsey's Global Private Markets Report 2026 (published February 2026, measuring deals struck 2010 to 2022).
That is a striking division of labour. On the published attribution, most of the return on a decade of private equity came from two things that had nothing to do with running the business: borrowing at low rates against an asset whose price was rising anyway, and selling it later at a higher multiple than it was bought at.
The third source, operating improvement, was always present. It was simply not the biggest contributor, and in a rising market it did not need to be. A sponsor could underwrite modest growth, apply leverage the market was giving away, and rely on the exit multiple to close the gap between an acceptable outcome and a good one.
The current composition is the reverse. Revenue and margin growth produced 71 percent of the value created in 2024 exits, up from 64 percent in 2023, and above any prior five-year period on record (Bain and Company data reported in a Moonfare analysis of the Global Private Equity Report 2026, 27 February 2026, covering exits completed in 2024). Read those two numbers together and the shift is not gradual. It is a swap in which contributor leads.
Why have leverage and multiple expansion stopped working?
Because both were priced off conditions that reversed. Multiple expansion requires the exit multiple to exceed the entry multiple, and the spread between the two compressed from roughly four turns across 2010 to 2020 to about 1.6 turns from 2020 onward (iCapital, February 2026). A buyer who needs the multiple to expand is now underwriting a much thinner margin than the last cycle handed out.
Entry multiples are the other half of that problem. Median buyout entry reached 11.8 times EBITDA in 2025, a record that edged past the previous 2022 peak and sits well above the 9.1 times average for 2010 to 2022 (McKinsey Global Private Markets Report 2026, February 2026, measuring 2025 entries and labelled as such). Buying at the top of the recorded range is a difficult place from which to expect expansion.
Leverage has been squeezed from two directions at once. Debt as a share of the entry multiple fell from 44 percent in 2016 to 37 percent in 2025 (McKinsey, February 2026), which means sponsors are writing larger equity cheques for the same asset. And the cost of the debt that remains is not falling: the Federal Reserve held its target range at 3.50 to 3.75 percent on 29 July 2026, with three policymakers dissenting in favour of a hike, the first time since September 2016 that three dissented in the same direction (FOMC statement, 29 July 2026).
The forward curve says the same thing more bluntly than the commentary does. Three-month SOFR is priced at 4.04 percent at the end of 2027, above the 3.76 percent print in early August 2026 and above the Federal Reserve's own 2027 median projection of 3.6 percent (Blue Gamma, 4 August 2026; Federal Reserve Summary of Economic Projections, 17 June 2026). Nobody is underwriting a financing tailwind off that curve.
“For fifteen years a sponsor could be wrong about the business and still be right about the deal, because the market fixed it on the way out. That is no longer available. The whole return now has to be manufactured inside the company, which is why the diligence question has moved from what the business could become to what it has already proved it can do without anyone new in the building.”
How much operating growth does a deal need now?
Roughly twelve percent a year, where five used to be enough. Bain's formulation is that a deal requiring 5 percent annual EBITDA growth a decade ago now requires about 12 percent to reach a 2.5 times return over a five-year hold, because the purchase multiple is higher and the financing costs more (Bain and Company, Private Equity Midyear Report 2026, 8 June 2026). Bain's combined measure of purchase multiples and financing costs is in record territory.
Doubling the required growth rate does not double the difficulty. It changes which businesses are financeable at all. A company compounding at four or five percent was a perfectly good sponsor asset in 2015 on the arithmetic above. On the current arithmetic it produces a return that does not clear a fund's hurdle, no matter how well it is run.
This is the mechanical reason the buyer list for an average business has thinned while the list for a good one has not. Practitioners describe the split directly: multiples for A-grade targets are running very high while lower-grade companies are not attracting bids at all (ACG and GF Data, Q3 2026 Market Pulse Survey, surveyed at the start of Q3 2026, published 15 July 2026). That is not sentiment. It is the growth requirement sorting the population.
It also explains why operating improvement gets described as a return source rather than as good management. When a sponsor says it will create value operationally, it is not making a claim about diligence or governance. It is saying it has identified twelve points of annual growth that the current owner is not capturing, and that it can capture them inside five years while paying a record entry price.
| Return source | What the published attribution shows | Status now |
|---|---|---|
| Leverage | Part of the 59 percent that leverage and multiple expansion produced together on deals done 2010 to 2022, per StepStone analysis in McKinsey, February 2026 | Debt fell from 44 percent of the entry multiple in 2016 to 37 percent in 2025, and the forward curve prices base rates higher at end-2027 than today |
| Multiple expansion | The other part of the same 59 percent; entry-to-exit spread averaged about four turns across 2010 to 2020 | Spread compressed to about 1.6 turns from 2020, against record 11.8 times median entry in 2025 |
| Operating improvement | Revenue and margin growth produced 71 percent of value created in 2024 exits, up from 64 percent in 2023 | The only source still contributing at scale, and the reason a deal that needed 5 percent annual EBITDA growth a decade ago now needs about 12 percent |
What does that change about who wants your business?
It moves the premium from potential to proof. If the return has to come from operations, the buyer is underwriting an operating plan, and an operating plan is only as good as the evidence that the organisation can execute it. A business that has already professionalised, and can show the effect in its own numbers, removes the largest uncertainty in the buyer's model.
The corollary is uncomfortable for a certain kind of seller. A business whose value rests on what a new owner could do with it is asking the buyer to underwrite the hard part twice: once for the price, and again for the execution. In a market where the exit multiple will not rescue the outcome, buyers are declining that trade, or pricing it with structure rather than cash.
The composition of the buyer pool reflects the same logic. Roughly three-quarters of US sponsor buyouts in the second quarter of 2026 were add-ons to existing platforms rather than new platform investments (PitchBook, Q2 2026 US PE Breakdown, 6 July 2026, as at 30 June 2026). An add-on is the cheapest available form of operating improvement, because the platform already has the management, the systems and the credit facility.
None of this makes the market hostile to owners. It makes it specific. The businesses being bid competitively are the ones where the growth is visible in the historical accounts rather than in the plan, and where the buyer can see who delivers it.
What should an owner do with this?
Decide honestly whether the growth in the plan is already in the numbers. If it is, the market will pay for it, and waiting adds risk without adding price. If it is not, the choice is between selling into a discount that reflects an unproven plan and spending the time to prove some of it first.
The second path has a measurable price. Preparation standards for this market run to at least thirty-six months of clean, normalized monthly financial statements with a quality of earnings report commissioned early rather than in reaction to a buyer's findings; firms that closed successfully were described as having well-prepared packages that reduced the risk of re-trading (Capstone Partners, Capital Markets Update, 4 June 2026). Three years of clean monthlies is also exactly the record that demonstrates a growth rate rather than asserting one.
There is a timing consideration that cuts against waiting indefinitely. For eighteen months the market treated delay as a free option because rate cuts were expected. That has inverted: dealmakers are now discussing the possibility of a hike and, in the same survey, pressure to complete transactions before rates rise (ACG and GF Data, 15 July 2026). Preparation is worth doing. Waiting for conditions is not the same activity.
The broader point is the one the attribution data makes on its own. When 59 percent of the return used to arrive from outside the business and now has to be produced inside it, the work an owner has already done stops being a soft quality and becomes the largest single component of what a buyer is paying for.
As of August 2026
Sources: McKinsey and Company, Global Private Markets Report 2026, February 2026, citing StepStone Group analysis, for the 59 percent share of returns attributable to leverage and multiple expansion on deals done 2010 to 2022, the record 11.8 times median buyout entry multiple in 2025, the 9.1 times average for 2010 to 2022, and the fall in debt as a share of entry multiples from 44 percent in 2016 to 37 percent in 2025; Bain and Company, Global Private Equity Report 2026, 23 February 2026, as reported in Moonfare's analysis of 27 February 2026, for revenue and margin growth producing 71 percent of value created in 2024 exits against 64 percent in 2023; Bain and Company, Private Equity Midyear Report 2026, 8 June 2026, for the deal cost measure and the arithmetic that a deal needing 5 percent annual EBITDA growth a decade ago now needs about 12 percent to reach 2.5 times over five years; iCapital, February 2026, for entry-to-exit multiple spread compressing from about four turns across 2010 to 2020 to about 1.6 turns from 2020; Federal Open Market Committee statement, 29 July 2026, for the 3.50 to 3.75 percent target range and the three dissents in favour of a hike; Federal Reserve Summary of Economic Projections, 17 June 2026, for the 2027 median projection of 3.6 percent; Blue Gamma, 4 August 2026, for three-month SOFR forward pricing of 4.04 percent at end-2027 against 3.76 percent in early August 2026; PitchBook, Q2 2026 US PE Breakdown, 6 July 2026, as at 30 June 2026, for the add-on share of US sponsor buyouts; ACG and GF Data, Q3 2026 Market Pulse Survey, surveyed at the start of Q3 2026 and published 15 July 2026, for the split between A-grade and lower-grade targets and for the rate-direction commentary; Capstone Partners, Capital Markets Update, 4 June 2026, for the thirty-six months of clean normalized monthly financial statements standard and the re-trading observation. Companion articles on this site cover why sponsors now compete on operations rather than leverage, and why three in four sponsor buyouts are add-ons.