What did Apollo publish about private equity value creation?
A split of realized value by exit year into three components. Apollo's Daily Spark of 6 September 2026, written by Torsten Slok, its chief economist, reports that multiple expansion produced 8 percent of private equity value creation on 2025 exits, against 40 percent on exits completed between 2018 and 2021.
The chart behind those two numbers carries the rest of the series, and the rest is the interesting part. Across five periods, revenue growth runs 46, 52, 63, 73 and 75 percent. Margin expansion runs 14, 15, 17, 11 and 17 percent. Multiple expansion runs 40, 32, 20, 15 and 8 percent. The underlying series is Gain's Private Equity Value Creation Report 2026, which Apollo names as its source.
Read the columns rather than the headline. One component collapsed by 32 points and one component absorbed almost all of it. The third barely moved. The dividing line Apollo draws on the chart is the point at which the Federal Reserve began raising rates, which is why the 2018 to 2021 exits are grouped as a single pre-hike period rather than shown year by year.
One thing this series is not: a full attribution of a sponsor's return. It has no leverage component. A separate attribution already covered on this site, carried in McKinsey's Global Private Markets Report 2026 and sourced to StepStone Group, put leverage and multiple expansion together at 59 percent of returns on deals done between 2010 and 2022, and concluded that both had stopped working. That finding is settled and is not re-argued here. The two decompositions divide different totals and should not be netted line by line. What the Apollo chart adds is the internal split of the part McKinsey left as the remainder.
| Exit year | Revenue growth | Margin expansion | Multiple expansion |
|---|---|---|---|
| 2018 to 2021 (before the Fed raised rates) | 46% | 14% | 40% |
| 2022 | 52% | 15% | 32% |
| 2023 | 63% | 17% | 20% |
| 2024 | 73% | 11% | 15% |
| 2025 | 75% | 17% | 8% |
| Change, 2018 to 2021 against 2025 | +29 points | +3 points | -32 points |
Why does it matter that margin expansion did not rise?
Because it is the component every seller expects the buyer to supply. The received account of what a sponsor does after closing is cost work: consolidate the back office, renegotiate suppliers, tighten the operating model, sell at a better margin. On this series, that work has never been the majority contributor and it has not grown. Margin expansion sat at 14 percent on pre-hike exits and 17 percent on 2025 exits, moving inside an 11 to 17 percent band with no direction across the whole period.
So when the multiple stopped paying, margin did not step in. Revenue did. Revenue growth carried 46 percent of realized value before the hiking cycle and 75 percent on 2025 exits, which is a 29-point transfer landing almost entirely on one line. Whatever a sponsor tells an investment committee it will do to a business, what actually showed up in realized value was that the business got bigger.
This is worth sitting with, because it reverses a common piece of preparation advice. An owner who spends the year before a process squeezing 200 basis points out of the cost base is working on the component that has contributed least and moved least. An owner who spends it demonstrating that revenue compounds without them personally in the room is working on the component that now carries three quarters of the outcome.
It also explains a pattern that reads as inconsistency from the sell side. Buyers will accept a lower current margin than an owner expects, and then refuse to move on a growth assumption that looks conservative. That is not negotiating tactics. On this evidence it is the buyer underwriting the line that has actually produced their returns and declining to underwrite the one that has not.
“Owners prepare for the wrong conversation. They arrive with a cost story and the buyer wants a growth story, and both sides leave the meeting thinking the other one missed the point. The numbers say the buyer is not being difficult. Growth is where their return has come from for four years running, so growth is what they will test hardest and pay for.”
What does a buyer underwrite now?
Durability of the top line, first and hardest. If three quarters of realized value comes from revenue growth, then the diligence that matters is whatever tests whether growth survives the change of ownership. That is cohort retention, pricing power, pipeline that does not depend on the founder, sales capacity that can be added, and a customer base that is not one relationship away from a bad quarter.
The exit multiple has become close to a rounding error in the plan. At 8 percent of realized value, an assumption of any re-rating at all is now a small part of the case, which cuts both ways for a seller. It means the buyer is not relying on the market being kind in year five, so a soft market is less of an objection than it was. It also means the buyer has no room to make up a disappointing top line with a better exit print, so they will not underwrite one.
It changes the shape of the price too. A buyer whose return depends on growth is buying a rate, not a snapshot. That is why deferred structures keep appearing in offers on businesses whose growth is real but not yet proven to a third party: an earnout or a rollover is the mechanism by which a buyer pays for a growth rate they cannot verify at signing. An owner who reads those structures as distrust is misreading them. They are the price of the component that now decides the outcome.
And it raises the value of anything that makes growth legible. Two businesses with the same revenue line are not the same asset if one can show where the growth came from by cohort, channel and product and the other cannot. The first one is underwriting evidence. The second one is a management assertion, and a buyer who has to discount an assertion will discount it in the price or move it into contingent consideration.
What should an owner or a sponsor do with this?
Decide honestly whether you are selling a growth business, because that decision sets everything else. If revenue is compounding and can be shown to compound without you, the market is currently paying for exactly that and the case for going now is strong. If revenue is flat and the plan was to sell on the multiple, the component you were relying on has fallen to 8 percent of what buyers realized last year, and no amount of process design recovers it.
Build the growth record before the data room, not inside it. Cohort retention by year of acquisition, revenue by customer with the concentration visible, pricing history with the increases that stuck marked separately from the ones that were discounted back, and pipeline conversion measured over enough quarters to be a rate rather than an anecdote. All of that is ordinary management information. It is also the evidence base for three quarters of the buyer's return, which is why its absence costs more than it used to.
Do not over-correct into a cost story. The point of this series is not that margin is unimportant, it is that margin work has never been the majority contributor and has not increased its share. A business with structurally poor margins still has a problem. A business with reasonable margins and a good growth record does not improve its position by shaving another point off the cost base in the six months before a process, and may weaken it if the saving comes out of the sales capacity that produced the growth.
Then ask what the buyer in front of you actually needs. A sponsor whose fund is underwriting revenue growth wants a platform that can absorb capital and add capacity. A strategic buyer is usually buying a specific capability or a customer base and may be indifferent to your standalone growth rate. Those are different arguments and different prices, and the same set of numbers supports one much better than the other. Knowing which conversation you are in before the first meeting is most of the preparation.
As of September 2026
Sources: Apollo, The Daily Spark, "Since the Fed Started Hiking, Multiple Expansion Has Nearly Vanished as a Source of PE Value", published 6 September 2026 and written by Torsten Slok, Partner and Chief Economist, read first-hand: for multiple expansion falling to 8 percent of private equity value creation on 2025 exits from 40 percent before the Federal Reserve began raising rates, and for the statement that value now has to be built rather than repriced. The full series by exit year is read from the chart published with that note, titled "PE value creation drivers by exit year": revenue growth of 46, 52, 63, 73 and 75 percent, margin expansion of 14, 15, 17, 11 and 17 percent, and multiple expansion of 40, 32, 20, 15 and 8 percent, for exits completed in 2018 to 2021, 2022, 2023, 2024 and 2025 respectively. Apollo attributes the underlying data to Gain, The Private Equity Value Creation Report 2026, together with its own chief economist's analysis; Gain's report was not read first-hand for this article, so no claim is made here about its sample, geography or definitions beyond what Apollo published. McKinsey and Company, Global Private Markets Report 2026, February 2026, citing StepStone Group analysis, for the 59 percent of returns on deals done between 2010 and 2022 attributable to leverage and multiple expansion together, which is covered in a companion article on this site and is cited here only to mark that the two decompositions divide different totals and are not netted against each other. The reading of margin expansion as the component that never rose, of revenue growth as the line a buyer underwrites, of deferred consideration as the mechanism for paying for an unverified growth rate, and the guidance to an owner or a sponsor are ours.
This position sits within our sell-side M&A advisory practice.