Before borrowing against a private markets portfolio: a checklist

Before borrowing against a private markets portfolio, a family office should be able to answer four questions in writing: whether the need is temporary and defined, whether the portfolio is diversified and one the family means to keep, whether the cash sweep and covenants still leave enough cash for known capital calls in a downside case, and whether repayment rests on an assumption the investment committee would accept in any other context. That is the decision test set out in a paper published by The Family Office Association in October 2026, What a family office should establish before borrowing against a private markets portfolio, which I co-wrote with Marc Sharpe. This article condenses it.

The four-part test

Each test is independent. A diversified portfolio is still the wrong collateral if the need is permanent, and a temporary need is still badly served if repayment depends on marks recovering.

Eligible NAV is not reported NAV

The advance rate in a NAV facility applies to eligible net asset value, which is a defined term in the loan document, rather than to the GPs' reported marks. The paper's checklist covers excluded positions, concentration caps by fund, manager and underlying asset, valuation and currency adjustments, the treatment of unfunded commitments and existing debt, whether capitalised interest counts as debt, and whether the lender can challenge valuations between reporting dates.

The gap between the two numbers can be wide, and it is the smaller number that the covenants test. It also moves. The paper's illustration is that a facility drawn at 35% LTV rises to roughly 44% after a 20% markdown if the debt is held flat. Advance limits, maintenance thresholds and cash trap levels are usually set at different points, so a borrower needs all three figures. Lenders publish their own parameters in outline, for example Nodem Capital's investment parameters page, but the document that governs is the facility agreement. The NAV loans explainer covers the structure in more detail.

Model the sweep before the breach

Most borrowers stress-test the LTV covenant. Fewer model the cash sweep, which often starts at closing while the loan is performing. Distributions go to the lender first, interest keeps accruing, and capital calls keep arriving. The paper recommends modelling the sweep alone before modelling any breach, and asking three things of the term sheet: what share of each distribution is captured, whether there are carve-outs for capital calls, taxes and running costs, and when the sweep switches off.

A sweep that captures most distributions from day one turns a portfolio that was funding its own calls into one that needs external cash for them. That is the mechanism by which a facility intended to solve a liquidity problem can create a larger one. The article on cash sweeps, LTV triggers and covenants works through the arithmetic.

When a secondary sale is better

The paper does not offer a single rule, and it warns against comparing a loan to a published secondary price. Jefferies' July 2026 review, cited in the paper, put average LP portfolio pricing near 87% of NAV, with buyout at about 91% and real estate at about 68%. Those are market observations. Buyers select the strongest interests, deferrals are common so the headline price may not settle as cash, and a sale transfers unfunded commitments while a loan does not.

The fair comparison raises the same cash on the same date by each route. The worked example is a $20m requirement: at a 91% bid, raising it means selling about $22m of reported value, a discount of roughly $2m before costs and taxes, against the interest, fees and sweep of a loan over its expected life. The secondary sale versus portfolio financing article sets out that comparison in full.

The paper lists the situations in which a facility is the wrong answer: a recurring shortfall, concentration in one or two positions, repayment that depends on a recovery or refinancing, an unsettled governance process for curing breaches, and cases where the family would exit the assets anyway. In that last case, a discount to a stale mark may simply be what the assets are worth.

What to establish, in order

Taken together, the paper's checklist runs as follows. Write down the amount and date of the need and test whether it recurs. List the portfolio by position with the latest mark, the date of that mark, the unfunded commitment and any transfer restrictions. Obtain the lender's definition of eligible value and recompute LTV on that basis. Model the sweep on the base case and the downside case, including known calls, taxes and running costs. Identify who in the family's governance can approve a cure, add collateral or authorise a sale, and how quickly. Price the alternative of a sale on the same cash and dates. Only then decide.

None of this is specific to one lender. It is the diligence a committee would apply to any other leveraged position, applied to the family's own balance sheet.

The author is Chief Investment Officer of Nodem Capital, an FCA-authorised lender. Family Office Primer is published by Nodem Capital; see our editorial standards.