The financed J-curve.

The wait has a length and a cost. Both can be measured. On Preqin data across 2000 to 2020 vintages, a buyout fund took about four years to reach the bottom of the curve, roughly half the committed capital called net of distributions by then, then close to seven years to turn cash flow positive, with the full range running from four to ten. The trough is not free to sit through. The management fee, commonly 1.5% to 2% a year, is charged on committed capital from the first close, before much of that capital is working. So the investor pays to wait and pays again on capital promised but not yet put to use. That second charge runs through everything that follows.

What has changed is the timing. Cash is not coming back the way it used to. Buyout distributions ran at 14% of NAV in 2025 against a 2014 to 2017 average of 29%, less than half the old pace. Funds are now holding roughly $3.8 trillion of unsold companies, the average position close to seven years old. The curve has stretched. Investors who signed up to a payback schedule that never arrived are pushing managers to do something about it.

The first way to remove the curve is to sell the investor a fund with no end date and a redemption window. Evergreen and perpetual structures take continuous subscriptions, value themselves at regular intervals and let you ask for your money back monthly or quarterly. No capital calls, no ten year lock up, immediate exposure to a diversified book. For a wealth client who never wanted the operational drag of a drawdown fund, the appeal is obvious.

US semi-liquid evergreen funds held roughly $460bn in net assets at year end 2025, according to PitchBook, up more than 30% in a year, across more than 480 funds, over half launched in the past three years. Broader evergreen estimates from Morningstar and PitchBook run above $500bn, depending on definition and universe. Wealth money is about a fifth of the total. Few corners of private markets have grown this fast.

The catch sits in the word liquidity. The fund offers redemptions. The assets underneath are private companies, property and loans that cannot be sold on demand at the marked price. The structure squares that with cash buffers, redemption gates and queues that slow withdrawals when too many arrive at once. In calm markets it works. The promise and the asset only diverge when everyone wants out together, which is exactly when the underlying cannot be sold without a discount. The wait has not gone. It has been moved into a gate that opens in good weather and closes in bad.

The second way to remove the curve works at the fund level. A NAV loan is credit extended to a private equity fund, secured against the net asset value of the whole portfolio rather than any single company. The manager can borrow against the book and use the proceeds to pay a distribution, fund a follow on investment or bridge liquidity, without selling anything into a soft market. Investors see cash come back. Their DPI improves. The fund looks like it is returning capital on schedule.

This is the heart of it. Where it funds a distribution, that distribution is not the proceeds of an exit. The cash is borrowed money, secured against assets that have not been sold, paid out so the return profile looks healthier than the cash flow beneath it. A distribution funded this way is borrowed, not earned. The tool is real and it has real uses. It is also, more and more, a way to make a slow fund look fast.

The market has gone from niche to structural in a few years. Industry estimates put NAV lending at roughly $100bn to $150bn today, with forecasts pointing toward $600bn to $700bn by 2030. Buyout funds are the largest borrowers. Loans now carry investment grade ratings and yield 300 to 600 basis points over the floating rate, which is why insurers are piling in. The growth is not the problem. The structure is.

Where it lands is no mystery. NAV loans cluster on funds in mid to late life, past the investment period, assets maturing, investors pushing for cash. The borrowing shows up exactly where the curve was meant to turn up on its own and has not.

The third cost is the quietest. It is also the one we do most about. A commitment is not an investment. From the moment an investor signs up to the moment money is actually working in an asset, capital sits and waits. At every stage where it waits it should earn a yield rather than a custody fee. There are three places it sits. The industry manages none of them well by default.

The first place is the investor's own reserve. Capital is committed up front but called in tranches over years, so a large part of it sits with the investor, waiting for the call. Held in cash it lags the target return the whole time. Held in a 60/40 blend to do better it takes on market risk, which means it can fall in value just as a call lands and force a sale into a weak market to meet that call. The drag here is well understood. The largest managers now publish on the cost of idle capital, which tells you it is real and being priced rather than a fringe worry.

The second sits inside the fund. When a manager calls capital it does not always go to work that day. Drawn money often sits in a custody or operating account for weeks, sometimes months, before it reaches a deal. The fee clock and the hurdle clock run the whole time. On a large fund that balance can be hundreds of millions at any moment. In a low or zero interest account it earns almost nothing. In short government paper and money market instruments with an ISIN, the same balance can earn around 3.7% to 3.9% with daily access and no lock up. The difference is pure leakage and it lands on the investor.

The third is the redemption sleeve. Evergreen and semi-liquid funds, the structures earlier in this edition, must hold a cash buffer to meet withdrawals. Managers put it at 10% to 20% of the fund and concede it drags returns if it is not run for yield. On a sector approaching $460bn that is tens of billions sitting in cash and treasuries. Here is the part that is left unsaid. The perpetual structure is sold as the cure for idle capital, because it puts the investor's money to work sooner and shrinks the reserve. It does not remove the idle cash. It relocates it, out of the investor's reserve and into the fund's own sleeve. The pool moves down the chain. It does not shrink. The table below sizes the yield at stake on it.

This is where Red Pin Capital Fund Treasury sits. We treat idle capital as a managed position rather than a parking space, the called but undeployed balance and the redemption sleeve alike. The mandate is narrow and deliberately dull. Liquidity generally same day or next day, holdings carry an ISIN, capital preservation first, yield second, full transparency on where every dollar sits. The point is not to take risk with money waiting to be deployed. It is to stop giving up a yield on it for no reason.

Run that against a fund that calls capital in tranches and routinely holds a buffer and the leakage compounds across the life of the vehicle. None of it needs duration risk or credit risk taken with money that is about to be deployed. It needs idle capital treated as a position rather than an afterthought. The structures in this edition move the idle cash down the chain. They do not make it earn. That is the gap we close.

Stack the three. At the investor end, evergreen funds promise liquidity on assets that are not liquid. At the manager end, NAV loans pay borrowed distributions and rank senior to the investor. In between, called capital leaks yield while it waits. Each piece defends itself on its own terms. Together they describe a system that swapped patience for leverage and called it progress.

Then the assets disappoint, or the market reprices. The NAV lender is paid first. The evergreen investor asks to redeem at the same moment as everyone else and finds the gate down. The capital that same investor still has waiting to be called has fallen with the public market it was parked in, so the call lands exactly when there is least cash free to meet it. The smoothed mark, which never moved while public markets ran, has the furthest to fall when it finally catches up. Slow assets are carrying fast liabilities. The claims can arrive together. Nobody has priced that, because nothing like it has been tested in a real drawdown.

Public markets can reprice in a single session. Private marks do not move with them. That gap has to close. The more financing piled on to keep private returns smooth, the worse the close gets.

The other side of the trade follows from all this. If the industry is paying to make slow assets look fast, the cleaner position is the asset that produces cash quickly, under contract, with no engineering at all. We would rather own real cash flow than finance the wait for it.

Power and compute is the clearest case. A data centre with a signed offtake and contracted power does not need a loan to look like it pays. It pays. The cash is real, contracted, arriving on a schedule no lender invented. The capital behind it can often afford to wait for it, with no structure at all. The financing in this edition is trying to manufacture what that asset already has.

None of this says financing is bad, or that evergreen funds and NAV loans should not exist. They have real uses and they are here to stay. The point is where the risk has moved and what that makes attractive by comparison. When much of the industry is paying to make slow assets look fast, the asset that is genuinely fast at producing contracted cash becomes the better trade. That is where we allocate.

Across the five pillars and the treasury work the brief is the same. Back cash flow that is real, contracted and not reliant on a structure to look healthy. Here is how each maps to the thesis.