What did the exit market actually do last quarter?
It got smaller and more expensive at the same time. Announced buyout exit value was 245 billion dollars in the second quarter of 2026, down 23 percent from the first quarter, and the 595 companies divested globally were the lowest quarterly count since 2020.
The price went the other way. Median exit multiples of earnings before interest, taxes, depreciation and amortization rose to a range of 14.7x to 17.6x for deals exited in the fourth quarter of 2025 and the first quarter of 2026, against a range of 11.0x to 13.0x over much of 2022 through 2024. That is roughly four turns of improvement.
Note the two readings sit on different clocks. The count and the value are second quarter figures. The multiple is the latest available and covers the two quarters before that, which is normal for exit pricing because a multiple can only be computed once a deal has closed and been reported.
Underneath, the buyout market itself was uneven rather than uniformly slow. US buyout volume fell 45 percent from the first quarter, while volume in Europe, the Middle East and Africa rose 51 percent and exceeded the US by more than 13 billion dollars.
| Measure | Reading | Date or period |
|---|---|---|
| Announced buyout exit value | 245 billion dollars | Quarter to 30 June 2026 |
| Announced buyout exit value, change on prior quarter | -23% | Quarter to 30 June 2026 |
| Companies divested globally | 595 | Quarter to 30 June 2026 |
| Median exit EBITDA multiple | 11.0x to 13.0x | Much of 2022 through 2024 |
| Median exit EBITDA multiple | 14.7x to 17.6x | Q4 2025 and Q1 2026 |
| US buyout volume, change on prior quarter | -45% | Quarter to 30 June 2026 |
| EMEA buyout volume, change on prior quarter | +51% | Quarter to 30 June 2026 |
| Operating company IPOs, US exchanges | 39 | Quarter to 30 June 2026 |
| IPO proceeds, US exchanges | 117 billion dollars | Quarter to 30 June 2026 |
How can prices rise while activity collapses?
Because a median is computed only on the deals that happened. When the number of transactions falls by that much, the surviving sample is not a smaller copy of the market. It is the part of the market that could clear, and the part that can clear in a hard quarter is systematically the better part.
That is our inference and not Carlyle's claim. The filing reports both numbers and does not say the second is caused by the first. But selection is the ordinary explanation for the pattern, and it is worth stating plainly because the alternative reading, that everything got more valuable while almost nothing sold, describes no market anyone has seen.
The mechanism is unglamorous. Sellers with a choice do not run a process into a weak bid, so assets that would have printed at the bottom of the range are withdrawn or postponed. Assets that must transact, or that are good enough to attract competition regardless, go ahead. Removing the low end of a distribution raises its median without any individual company being worth a penny more.
This is why the improvement should not be read as recovery. A recovering exit market shows rising counts. A narrowing one shows falling counts and rising prices, which is exactly the pair on the page.
“An owner sees a headline multiple and hears a quote. It is not a quote. It is the average price of the companies that were good enough to get through a door that most of the market did not get through. The question that number answers is what clearing costs, not what your business is worth, and the two are only the same if you already know you are in that group.”
Did the public route widen while the sponsor route narrowed?
Partly, and it is worth reading carefully. Initial public offerings were the bright spot of the quarter, with 39 operating company listings on US exchanges generating 117 billion dollars in proceeds, an 86 percent increase in transaction count against the first quarter.
That is a genuine improvement, and it is also a very small door. Thirty-nine listings in a quarter is not an alternative route for the great majority of privately held companies, and the proceeds figure is concentrated by definition in the largest of them. The channel reopened at the top of the size range.
The same filing notes what that concentration reflects. Carlyle reads the largest listing of the quarter, a company that came to market twenty-two years after it was founded, as emblematic of private markets capturing an increasing share of value creation before a listing ever happens.
For an owner the practical reading is that both exit routes narrowed to their upper end in the same quarter. The sponsor route cleared fewer companies at better prices, and the public route reopened for a small number of very large ones. Neither is evidence of a market that has become easier to enter.
What decides whether a business is inside the group that clears?
Whatever removes work and doubt from the buyer, which in a thin quarter is worth more than it is in a busy one. When counts are high, a buyer will take on a fixable problem because there is competition for the asset. When counts fall by this much, the buyer has alternatives and the problem is a reason to move on rather than a point to negotiate.
In practice that means the ordinary list, held to a higher standard than usual. Earnings that survive a quality of earnings review without the number moving. Customer concentration low enough that the buyer is not underwriting one relationship. An owner whose departure does not take the revenue with them. Contracted or genuinely repeating revenue rather than a run of good years.
It also means being honest about which side of the line the business sits on before running a process. A withdrawn process is not free. It is visible to the buyer universe, and the next attempt starts from a worse position than the first, so the cost of testing a weak market is paid on the following attempt rather than this one.
The corollary is more useful than the warning. If the multiple range on offer to companies that clear has genuinely improved by about four turns, then work that moves a business from just outside that group to just inside it is being paid for at a better rate than it was in 2022 through 2024. The reward for de-risking widened at the same time the door narrowed.
What would tell you the exit market is genuinely reopening?
The count, not the multiple. A rising number of companies divested is the only reading that cannot be produced by selection, because it is the measure selection suppresses. If counts recover and multiples ease slightly, that is a healthier market than counts falling further while the median improves again.
Watch value and count together rather than either alone. Exit value fell 23 percent quarter on quarter while the count fell to a post-2020 low, so both directions agreed this quarter. If value recovers on a still-falling count, the market has reopened for large assets only, which is what the listing figures already show.
Treat the multiple range as a range. Carlyle reports 14.7x to 17.6x rather than a point estimate, and a spread that wide across a small number of completed deals will move on composition alone. One quarter of exit pricing computed on a shrunken sample is a weak basis for any individual expectation.
The sequence is the thing to hold on to. Prices at the front of the queue improved before volumes did, which means the good news arrived first for the companies that least needed it. An owner reading only the multiple will prepare for a market that is more open than the one they will actually meet.
As of August 2026
Sources: The Carlyle Group Inc., quarterly report on Form 10-Q for the quarterly period ended 30 June 2026, filed with the US Securities and Exchange Commission on 10 August 2026 (accession 0001527166-26-000045), section Trends Affecting Our Business, for the statements that exit activity remained constrained with announced buyout exit value of 245 billion dollars declining 23 percent quarter over quarter, that the 595 companies divested globally represented the lowest quarterly count since 2020, that median exit multiples of earnings before interest, taxes, depreciation and amortization continued to improve, rising to a range of 14.7x to 17.6x for deals exited in the fourth quarter of 2025 and the first quarter of 2026 being the latest data available, compared to a range of 11.0x to 13.0x over much of 2022 through 2024, that US buyout deal volume fell 45 percent from the first quarter while volume in Europe, the Middle East and Africa increased 51 percent to exceed the US by more than 13 billion dollars, that initial public offering activity was a bright spot with 39 operating company listings on US exchanges generating 117 billion dollars in proceeds and an 86 percent increase in transaction count against the first quarter of 2026, and that the largest listing of the quarter, coming twenty-two years after the company was founded, was emblematic of private markets capturing an increasing share of value creation before listing. The reading that a sharply reduced transaction count raises a surviving median through selection rather than through any change in individual company value is our own inference and is not a claim made in the filing. A companion article on this site covers the length of the seller queue and the pressure on sponsors to return capital, another covers the divergence between transaction value and transaction count across the wider merger and acquisition market, a third covers how private valuations are lowered over time in advance of a realization, and a fourth covers why a listing is rarely an exit for an owner.