�鶹ֱ��

War in the Gulf, an oil spike, rate risk and an AI shock that erased $285 billion of software value in a day have battered sponsors in 2026. Capital is still flowing to firms that return cash — and the law has moved mostly in their favor.

Private equity has taken a beating in 2026. A war in the Persian Gulf, an oil-price spike, a Federal Reserve that may raise rates this month and an artificial-intelligence shock that wiped $285 billion off software stocks in a single session have hit sponsors almost simultaneously.

The industry isn’t broken. Capital is still flowing to firms that return cash to investors. The best assets are still finding buyers. Data centers have turned into a genuine profit center. And on balance, the law has moved in sponsors’ favor this year more often than against them. Here’s where things stand — and where the value is.

THE NUMBERS, YEAR TO DATE

US private equity firms invested $545.1 billion in the first half, on pace to match 2025’s $1.1 trillion — but spread across just 3,926 deals, versus 9,350 for all of last year, according to KPMG. That’s fewer deals with much larger checks.

The second quarter was the soft spot. PitchBook puts deal value at $177.3 billion, down 37.5% from the first quarter, with software deal value down 65.7% from a year earlier.

Exits are the real pressure point. US exit value reached $273.1 billion at midyear across just 508 transactions, well below 2025’s pace of 1,376. Sponsor-to-sponsor sales fell to 94 in the second quarter — the lowest in at least a decade.

Fundraising is thin but not dead. US sponsors raised $159.6 billion across 223 funds through June, roughly on pace with 2025. Experienced managers took $139.3 billion of that; emerging managers took just $20.3 billion. That’s the pattern of the year: concentration, not collapse.

TEN YEARS OF PERSPECTIVE

Two things stand out from the longer record. First, deployment recovered before fundraising did — and it recovered on size, not count. Bain & Co. reports global buyout deal value reached $904 billion in 2025, the second-best year on record, even as fundraising fell 16% to $395 billion, a fourth straight annual decline. The current pace is drawing down dry powder that isn’t being replenished, because fundraising follows distributions, and cash returned to investors has been weak.

Second, the return problem is a cash problem, not a value problem. PitchBook’s global benchmarks show 2014-2017 vintages at a total value multiple near 2.0x. But 2017-2021 vintages show a distributions-to-paid-in ratio, or DPI, under 1.0x, and 2019-2021 vintages are under 0.35x. Investors aren’t questioning whether the underlying assets are good. They’re asking when they’ll get paid.

THE MACRO PICTURE

The US-Iran conflict that began in late February reshaped the year. Leveraged-loan issuance fell 22.5% in the first quarter. Brent crude touched $100 in July, and the two sides were trading strikes again in early September. The Fed held its target range at 3.5% to 3.75% in July on a 9-3 vote, and markets are leaning toward a quarter-point increase this month.

This is a shock, not a regime change. Build a rate cushion into new deals and keep the pipeline warm.

WHERE THE DEALS ARE

Energy and power are the clear winners. KPMG puts private-equity investment in energy and natural resources at $90.2 billion through June, on pace for its best year in more than a decade. Legal-services and health care roll-ups continue apace. Software isn’t closed, but it’s selective: Thoma Bravo’s $4 billion insurtech take-private, and reported talks for Workday Inc., show there’s still demand for regulated, cash-generating vertical software.

AI DATA CENTERS: THE CLEAREST BRIGHT SPOT

Data centers are where sponsors have made real money. S&P Global counted 113 transactions worth more than $69 billion in 2025, a record, led by the $40 billion sale of Aligned Data Centers — a sponsor exit for Macquarie Group Ltd. US data-center deal activity hit a five-year high in 2025 and stayed strong through the first half of 2026. Blackstone Inc.’s digital-infrastructure trust raised $1.75 billion in a May initial public offering. Debt markets are now pricing contracted capacity like a utility: Hut 8 Corp.’s Texas project sold $4.25 billion of notes in June at a 95% loan-to-cost ratio, fully pre-leased for 15 years to an investment-grade tenant, according to J.P. Morgan.

Three things separate a good deal from a bad one: the tenant, the power and the demand assumption. Hyperscaler-leased assets command higher valuations. A site with secured power holds its value; a site waiting on a grid interconnect doesn’t. And S&P’s own analysts warn that hyperscalers are investing as if they’ll all have a business in 10 years — and some may not. The lesson for sponsors: underwrite the lease and the power, not the AI narrative.

THE SAASPOCALYPSE

The biggest shock to technology-focused private equity this year didn’t come from the Gulf. It came from Anthropic. On Jan. 12, the company introduced Claude Cowork, an artificial-intelligence agent built to perform professional work, and on Jan. 30 it released open-source plugins for legal review, compliance, sales preparation and research — the very tasks that sit at the center of high-margin subscription software. On Feb. 3, roughly $285 billion in software, legal-tech and data-provider market value was erased in a single trading session. A Jefferies trader dubbed it the “SaaSpocalypse,” and the name stuck. The S&P Software & Services Select Industry Index lost 25% between the Cowork launch and Feb. 23. Claude Code, Anthropic’s coding agent, had already cut the cost of building software; Cowork convinced the market that the cost of using it would fall too.

For sponsors, the damage is direct. Software is the largest sector in most buyout portfolios, and the Bank for International Settlements reports that private credit’s exposure to software loans has grown to more than $500 billion, or 19% of direct loans. Companies underwritten between 2018 and 2020 — before generative AI — are the vintages most at risk. The lesson isn’t that software is dead. It’s that per-seat pricing is now negotiable, and proprietary data, regulatory moats and embedded transaction flows are what hold value. PwC finds that firms using AI to lift portfolio margins are outperforming peers in both fundraising and exit processes. Multiple expansion is gone; margin is the lever that remains.

Private credit had a scare, not a crisis. Three Blue Owl Capital Inc. business-development companies sold $1.4 billion of loans at 99.7% of par after the smallest of the trio suspended redemptions. Note the price: the problem was a liquidity mismatch, not a credit loss. Morgan Stanley sees defaults possibly reaching 8%, well above the historical norm of 2% to 2.5%, but calls that significant rather than systemic. For sponsors with clean credits, terms remain available.

EXITS AND FUNDRAISING

The IPO window is real. PwC counts roughly $114 billion raised in traditional IPOs through June — seven times the prior year’s pace. But the line to get through is long: CBIZ estimates more than 13,500 US companies sit in sponsor portfolios, nearly a third of them older than six years, and clearing that backlog at the current pace would take about nine years. The takeaway for sponsors holding a genuine gem: sell it. Waiting costs more than the discount.

DPI has replaced IRR as the metric that matters most to limited partners. Through early 2026, firms with strong distributions raised new funds quickly; others didn’t. About 70% of fund series expected to come back to market have done so, and Bain expects the long-run failure rate to resemble the financial crisis — roughly 20%. LPs have consolidated capital behind managers with strong distribution records and scale, and fees are compressing: the average buyout fee was 1.6% in 2025. Scale and specialization are the two ways out.

THE LAW: WHAT CHANGED, AND WHAT IT MEANS

The legal story of 2026 will outlast the oil price. Seven developments matter, roughly in order of importance to a sponsor.

• Retirement capital. On March 30, the Labor Department proposed a rule giving 401(k) fiduciaries broad discretion to include private equity, provided they follow a prudent process. The Supreme Court will hear Anderson v. Intel this fall, on whether participants suing over private assets in target-date funds must plead a meaningful benchmark; the Labor Department has sided with Intel. A favorable rule and a favorable decision by mid-2027 both look likely. The price of entry is a defined-contribution-ready product — frequent NAV, no capital calls, real liquidity, ERISA-grade fee disclosure and independent valuation. Whichever firms finish that build first will own a distribution channel their competitors don’t.
• The SEC. The 2023 private-fund adviser rules were vacated by the Fifth Circuit in 2024 and haven’t been revived. The commission’s agenda now favors retail access through registered funds and looser limits on performance fees; that proposal reached the White House budget office on Aug. 31. Sponsors should still draft as if the vacated rules applied — LPs continue to demand the fairness opinions and disclosures those rules would have required.
• Antitrust. A federal court in Texas threw out the FTC’s expanded merger-filing form in February, and the Fifth Circuit refused to stay that ruling in March, so filers are back on the pre-2025 form. States are moving the other way: California will require a copy of every federal filing starting in 2027, and more than a dozen states now review health care deals below the federal threshold, several naming private equity specifically. Build 90 to 150 days into any health care roll-up in a notice state.
• Delaware. In Rutledge v. Clearway, decided Feb. 27, the Delaware Supreme Court upheld the 2025 statute giving controlling stockholders a safe harbor — committee approval or a majority-of-the-minority vote — for most controller transactions, including going-private deals. For a sponsor with a control stake in a public company, it’s the most useful decision in a decade. Build a Section 144 committee into the playbook now.
• Credit documents. The Serta and Mitel rulings — one against an uptier exchange, one for it — turned on the exact wording of each credit agreement. Lenders are now negotiating express blockers and signing cooperation agreements before a borrower can divide them up. The room for coercive exchanges has narrowed.
• Tax. Last year’s US tax law left carried interest untouched, restored the EBITDA-based interest deduction and made bonus depreciation permanent. The UK went the other way: from April 6, carry is taxed as income, at an effective rate of about 34.1% for qualifying carry with a 40-month average holding period, up from 32% before. Fund design in London now turns on holding period.
• Europe. AIFMD II took effect in April with new rules for loan-originating funds and new liquidity requirements for open-ended funds — both of which reach the semi-liquid vehicles sponsors are building for private wealth.

BOTTOM LINE

Bain calls the new deal math “12 is the new 5.” A 2015-vintage buyout needed about 5% annual Ebitda growth to return 2.5 times capital; today’s leverage and multiples require closer to 10% to 12%. That’s harder. It’s also honest — the easy-money era was the anomaly.

The sponsors that do well from here will sell good assets rather than hold them out for a better mark, prove they can grow earnings rather than rely on multiple expansion, know their software exposure by vintage, own contracted data-center capacity with secured power, and have a DC-ready retirement product already in design.

Watch five things this fall: the Fed’s rate path after the September FOMC meeting, any settlement over the Strait of Hormuz, the text of the SEC’s retail-access proposal, oral argument in Anderson v. Intel, and whether third-quarter exit data shows the IPO window finally reaching sponsor-backed companies. My bet: at least three of the five break the right way.

Louis Lehot is a Silicon Valley growth and private-equity lawyer at �鶹ֱ�� & Lardner LLP, where he co-leads the firm’s Growth practice. He advises technology, life-sciences and clean-energy companies, their boards and their sponsors on growth-equity investments, buyouts, take-privates and exits.

Special thanks to Bella Tappin for editing assistance.