NAV loan, preferred equity or CFO: what each costs and when an LP should use which

Last reviewed November 2026.

Not every investor who needs cash from a private portfolio wants to sell it. The LP-led secondaries market did $58bn in the first half of 2026 by PJT Partners’ count, but a growing slice of that activity is what the market now calls partial liquidity, where the investor keeps the assets and borrows or raises structured capital against them instead.

A recent episode of Secondaries Investor’s Second Thoughts podcast, hosted by Madeleine Farman with Adrian Millan of PJT Park Hill and John Ketnick of StepStone, is a useful tour of the three tools in that category and how a large LP chooses between them. You can find the episode on the Secondaries Investor podcast page. Below is a summary of what they said on cost, leverage, tenor and fit, and then the numbers they did not give, which are the ones most of our readers actually want.

The three tools

Millan’s framing is simple. There are three ways to take liquidity from a fund portfolio without selling it. A NAV loan, historically a bank product, where a lender advances against the portfolio and is repaid from distributions. Preferred equity, provided by specialist funds, where the investor takes a senior slice of future distributions up to an agreed return and the LP keeps everything above it. And a collateralised fund obligation, or CFO, where the portfolio is placed in a vehicle that issues rated debt in tranches to insurers and other buyers of rated paper, with the LP keeping the equity.

All three rest on the same premise. The LP believes the portfolio will return more than the market will pay for it today, so it would rather pay a cost of capital for a few years than take a discount on sale. Together they were around 10% of LP-led volume last year on PJT’s numbers, up from roughly 1% five to ten years ago. Most LPs now run a sale process and a partial liquidity process in parallel and pick once they see both prices.

What the episode said on cost of capital

This was the clearest set of numbers in the conversation. Millan put the implied cost of a traditional LP portfolio sale at around 15%, meaning the return the buyer needs and therefore the discount embedded in the price. Preferred equity sits inside that at roughly 10%, which he described as high single digits to low double digits. NAV loans and CFO debt sit inside that again, all in, with the level depending on loan to value. He treated NAV and CFO debt as broadly similar in cost.

Ketnick added the leverage point. On the same portfolio a CFO will typically support something like twice the LTV of a conventional NAV loan, which is why an LP that wants to pull out a meaningful amount of capital while keeping the equity is increasingly steered towards a CFO. NAV loans were described as the most constrained of the three, both on how much they lend and on which assets count as collateral.

Tenor and process

NAV loans run around five years, six at most. CFOs can be structured for longer and in tranches, with the senior paper rated and placed mostly with insurance companies, where the NAIC capital treatment has made the asset class attractive. StepStone’s own credit team said it concentrates on the B tranche of CFOs and on preferred equity rather than the senior, lower returning debt.

A CFO takes four to five months against two to three for a straight sale. The rating agency work alone is six to eight weeks, though it can run alongside the debt raise, and repeat issuers get quicker. CFO issuance passed $40bn in 2025, helped by GPs using the structure to tap insurance capital for fundraising as well as LPs using it for liquidity.

Where it does not work

Both guests were direct about this. An aged portfolio with low single digit forward returns and most of its cash coming out in the next three years should be sold, because any debt will cost more than the assets earn. Concentrated portfolios do not work, since all three structures need diversity by manager and by asset. Venture, emerging markets and China were called poor CFO candidates because distribution timing cannot be predicted, which is what a debt stack needs, though venture may suit preferred equity where there is no requirement for current cash flow. A portfolio heavy in public look-through positions can work well in a CFO on the way up and badly on the way down, since a sharp fall ramps LTV with little cushion for the lenders.

The ideal collateral is a diversified, mature book that throws off distributions most weeks, with seasoned secondary fund interests the textbook example.

The numbers they did not give

The episode is pitched at pension funds, insurers and sovereigns running processes with PJT. The cost figures were deliberately broad. For family offices, GPs and smaller institutions looking at the same three tools on portfolios of $30m to $300m, this is roughly where the market actually sits.

NAV loans start at around 500 basis points over the reference rate for a well diversified portfolio of institutional fund interests at a conservative LTV, and move up from there with concentration, asset type and loan to value. Most of what we see and write closes between 550 and 650 over, with one to two points of OID, on a two to four year PIK structure. Banks will go tighter for a large relationship client, but they rarely lend against this kind of collateral at all below institutional scale. Our piece on pricing grids covers how the margin moves with LTV.

Preferred equity providers are generally underwriting to high teens IRRs or better, taken as a combination of a preferred return on their capital and a share of the upside above it. The headline preferred return is often quoted in the low teens, which is what the podcast’s “roughly 10%” cost reflects, but the upside participation is where the provider makes its return and where the LP gives up the most. It is the right tool when the portfolio cannot support current cash flow, when LTV needs to go well beyond what a lender will do, or when the assets are too long dated or too venture heavy for debt. It is the wrong tool if a NAV loan would have got you there, because you will have paid equity pricing for debt sized risk.

CFOs are harder to generalise. The all-in cost depends on how the tranches are cut, what rating the senior paper achieves, how much equity the LP retains and what the market for rated private fund debt is doing that quarter. Senior tranches can price very tightly. Blended cost across a full stack can land anywhere from NAV loan territory to something closer to preferred equity, and the fixed costs of rating and structuring mean the product only makes sense at scale, in practice several hundred million of NAV and up.

How to think about the choice

For most of the investors we speak to, the decision is simpler than the podcast makes it sound, because scale rules out the CFO. The question is whether a NAV loan or preferred equity fits.

If the portfolio is diversified and distributing, and the amount needed is 20% to 30% of value, a NAV loan is almost always the cheaper answer and the quicker one, closing in six to eight weeks rather than months. If the portfolio is venture heavy, long dated or concentrated, or the amount needed is 40% to 50% of value, a lender will either decline or price so wide that preferred equity becomes competitive, and at that point you are choosing between paying high teens with a capped downside or taking the discount on a sale.

One thing the episode gets exactly right and worth repeating. The right answer depends on the forward return you believe in, not on the structure you like. If you do not think the portfolio will beat the cost of capital of whichever tool you pick, sell it.

If you would like a view on which of these fits a specific portfolio, send us the schedule of positions and what repays the facility. We will tell you quickly whether a NAV loan works and roughly what it would cost, and if it does not, which direction to look instead.

Nodem Capital is an FCA-authorised AIFM (fund manager) providing facilities of $10m to $100m and above to family offices, GPs, LPs and holding companies. Contact Alex Branton at abranton@nodem.com.

General information, not advice. Nodem Capital is an FCA-authorised AIFM (fund manager). Nodem Ltd is authorised and regulated by the Financial Conduct Authority, FRN 1017481.