The argument, stated up front. For most of the last decade, deciding to hold a portfolio company another year was a view — you formed it, you said it out loud at the quarterly, and the reasoning stayed in the room. That is changing. Enough assets have now aged past their natural exit window that the hold itself has become the thing under examination: by limited partners who have waited four years for distributions, by auditors testing whether the mark is real, and in the tail case by counsel constructing an argument about why a fund kept charging fees on an asset it could not sell. The decision has not become harder. It has become discoverable.
The sponsors who come out of this cleanly will be the ones who can produce a file showing what they knew, what they tested, and why holding was the better of the available options.
That last point is where and why a firm like Stratis Group LLC gets hired to help. I will come back to it at the end.
What the numbers actually say
The first half of 2026 was worse for liquidity than the headlines suggested. There were 872 private equity–backed exits in the US against 1,210 in the first half of 2025 — a decline of nearly 30% year over year. Deal count fell in parallel: 3,999 transactions in H1 2026 versus 4,619 in the second half of 2025, with roughly $314 billion deployed. Median entry multiples compressed a full turn, to 13.4x EV/EBITDA from 14.6x across 2025.
The inventory number matters more.
Read those together. A quarter of the portfolio is already past the exit window. Another third gets there inside twenty-four months. This is not a cyclical air pocket that resolves when the IPO market reopens — it is a structural backlog, and the sponsors sitting on it have to decide, asset by asset, what to do about each one.
Dry powder has fallen to roughly $880 billion from a $1.3 trillion peak in December 2024. That cuts the other way from how it is usually reported. Less undeployed capital means less pressure to transact for its own sake, which removes the forcing function that used to break the tie. The default now is to hold.
What changed underneath the decision
Here is the part the trade press has mostly not caught up to. Foley & Lardner, writing in July, framed the aging-asset problem as a governance question rather than a market one, and their formulation is the sharpest I have seen on it:
A GP has to be able to show the decision was reasoned and in the fund's interest, not just a way to keep charging management fees.
Sit with the implication. The standard is not whether the hold turned out well. It is whether the decision to hold was made — deliberately, on evidence, with the alternatives actually examined — as opposed to arrived at by default because no buyer appeared and nobody forced the question.
Three things follow, and each lands on a different desk.
The mark has to be defensible on its own terms. If an asset is carried at a value the sponsor would not itself pay today, someone will eventually ask why. Valuation and disclosure move from a quarterly exercise to a documented position.
The fund documents stop being boilerplate. Extension provisions, LP consent rights and term limits sit unread for years, then become the binding constraint the moment a fund approaches the end of its life. What the LPA permits starts driving what the sponsor can choose.
Every alternative carries its own governance load. A continuation vehicle requires a conflicts process and a fairness opinion. A minority recapitalization introduces a new governance layer with its own consent rights. A secondary sale requires diligence and consent management. None of these are lighter than a clean exit — they are heavier, and they land on a portfolio company board assembled to oversee an operating business.
The National Association of Corporate Directors' 2026 governance outlook shows the same pressure from the board side: portfolio company boards standing up audit committees, formalizing documentation, and moving toward public-company process well ahead of any exit.
The exposure is the undocumented hold
Most holds are defensible. That is worth saying plainly, because the risk here is not that sponsors are behaving badly.
The exposure is narrower and far more common. An asset gets held for a fourth or fifth or eighth year. The reasoning is real and gets discussed at every board meeting. And none of it is written down in a form anyone outside the room could evaluate. Two years later the fund is raising, an LP asks why Company G is still carried at 2021 marks, and the honest answer — we believed the operating plan and the buyer universe was thin — has no file behind it.
That is not misconduct. It is a documentation gap — and it is the specific thing that turns a reasonable decision into an indefensible-looking one.
The same gap shows up in a second place. When an aging asset finally does go to market — trade sale, sponsor-to-sponsor, or into a continuation vehicle — the buyer's first question is what happened in years five through eight. A company that cannot narrate its own recent history with numbers gets discounted for the ambiguity, whatever the underlying performance actually was.
What a defensible hold actually contains
Not a memo. Four things — and they are the same four things that make the asset saleable when the window finally opens.
An operating case with a number attached. Not "the business is improving" but a specific bridge from current EBITDA to exit EBITDA, with the initiatives named, sized and dated. If the plan is real it can be written down. If it cannot be written down, it is a hope.
A forecast that has been stress-tested rather than presented. The board has seen the base case. The file needs the downside — what happens to covenant headroom and cash if revenue lands 15% short — and evidence that someone ran it before it was needed.
A capital structure with a stated path. When the facility matures, what the refinancing options look like at today's terms, and what the company has to demonstrate to a lender to get there. An asset with an unaddressed maturity inside eighteen months is not a hold decision; it is a deferred problem.
A clear-eyed read of the alternatives. What the asset would fetch today, who the realistic buyers are, and what a continuation vehicle would actually price at. A hold is only reasoned if the sale was genuinely evaluated.
None of this is legal work. It is finance work — the kind a sitting CFO would do if they had the bandwidth, which at an aging portfolio company they generally do not, because the finance function was built for a business two-thirds this size and never grew.
Where Stratis Group fits
This is the gap I said I would come back to.
The deal partner has the judgment but not the hours. The portfolio company CFO has the hours but is closing the month. The auditor tests the mark but does not build the case. Counsel reviews the process but does not underwrite the business. Each of them is doing their job, and none of those jobs is this one.
Stratis Group takes the operating CFO seat inside a portfolio company through exactly this work — building the operating case, stress-testing the forecast, mapping the financing path, and putting the whole thing in a form a sponsor can hand to an LP, a lender or a buyer without adding a word.
It is the same work whether the answer turns out to be hold or sell. That is rather the point: you cannot know which one is defensible until you have done it.
This is the gap Stratis Group works in. I take the finance seat inside a portfolio company through a capital event — a refinancing, a covenant reset, an acquisition financing, or the preparation that precedes a continuation vehicle or recapitalization. Build the case the counterparty will actually test, surface the uncomfortable parts before diligence finds them, and carry the process through to close, so the CEO and CFO stay on the operating plan the buyer is underwriting.
If a portfolio company is heading into one of these, twenty minutes is usually enough to tell whether it is useful.
jon@stratisgroupllc.com