Take two private credit funds with the same assets and the same strategy. Under AIMFD 2.0, the revised EU directive that has been applicable to alternative fund managers since April 2026, one of them can run leverage up to 300% of net asset value (NAV) and the other is capped at 175%. This stark difference is not in what assets they hold, rather the difference in whether each is classified as closed-ended or open-ended.
For the semi-liquid or evergreen funds now coming to market, that classification is not always obvious. Open-ended funds must build in liquidity management tools and show that their redemption terms match the liquidity of the underlying portfolio. For a fund holding illiquid positions, landing in the wrong category is an expensive mistake, so it is worth asking whether these funds are open-ended at all.
The legal answer that has emerged is a sensible one. Classification should follow economic substance rather than the literal wording of a redemption clause. The question is not whether an investor can ask for their money back, it is more relevant to ask whether the way they get it back can put the fund under liquidity stress. On that test, a run-off structure, where an investor can request redemption but capital is only returned as the relevant assets are sold down in the ordinary course at the manager’s discretion, is closed-ended. The manager is never forced to raise cash to meet a request, so there is no liquidity mismatch to manage.
This analysis holds, however, questions arise about what happens after the structuring decision has been made.
The run-off classification depends on one thing staying true over time: that the manager actually paces redemptions to the natural sale of assets. The reasoning works because the manager is not obliged to accelerate. But not being obliged to do something is not the same as not doing it.
The commercial incentives run the other way. In a competitive market for evergreen capital, a GP wants to keep its investors comfortable. So, a redemption gets met a little sooner than strict run-off would require. Then another does. Each decision is defensible on its own. Over enough quarters, taken together, they can amount to something closer to an on-demand redemption right rather than the run-off model the offering document describes.
That is where a fund’s actual behaviour and its stated classification begin to part company. Nobody decides it, and no board minute records the fund becoming open-ended. It shows up in the redemption data well before it shows up in a legal review.
This is why classification has been treated too much as a launch-day question, something counsel settles at formation and supports with a risk disclosure. That work is necessary, but it captures a single point in time, while the risk it describes runs for the life of the fund.
The fund administrator is the party best placed to see where the two diverge. It is the service provider that processes the redemption requests, holds the realisation dates, and records the actual time taken to pay each investor out. Where a fund is classified closed-ended on a run-off basis, the gap between the discretion described on paper and the redemption behaviour in the data is something a fund administrator can measure. It can be put in front of the AIFM and the Board while it’s still a trend and not a breach.
This asks something more of fund administrators than simply processing transactions, it’s a judgement call. It applies the same maker-checker discipline the administrator already uses for NAV and capital activity to the classification rationale itself, and treats “is this fund still behaving the way we said it would” as a question with an answer in the data, rather than an assumption based on a classification made when the fund as formed.
GPs need to keep the classification rationale under review rather than simply filing it away. GPs should document the economic basis for it, but also know what their redemption behaviour looks like against the model they are relying on, and be able to show that the two line up. A disclosure from a GP contradicted by their own records creates friction with regulators and investors.
For LPs it’s worth a reminder that the description on the cover does not provide all the information they need. A fund sold as open-ended because that is how its U.S. version is marketed may be closed-ended under EU rules, for example, with different liquidity and leverage. Instead, LPs need to ask how the classification was reached, and how it is kept under review. The second answer tells you more than the first.
For those in fund administration evergreen funds are harder to run than closed-ended drawdown funds. There are more frequent NAVs, continuous subscriptions and rsedemptions, investor bases that increasingly include private wealth, and because of this complexity there are more opportunities for something to go wrong. The firms that deliver the best service will be the ones that can give a GP and its AIFM a clear, evidenced answer to the classification question at any point in the fund’s life, not only on day one.
The definition of open-ended we are all working from is more than a decade old, and AIFMD 2.0 left it largely intact. It was written before semi-liquid evergreen funds existed, and the detailed guidance has not caught up. Until it does, classification rests on economic substance and on whoever can evidence it.
Counsel can set out the rationale at the start. Keeping it honest over the life of the fund is an operational job, and it is one the administrator is best placed to do best. If you’d like to continue the conversation on any of the points raised in this article, please contact us directly.
This article first appeared in the Lux Times.
16 October 2026 Martina Frisch
09 October 2026 Martina Frisch