Semi-liquid funds sell liquid assets to buy illiquid ones

Blackstone Real Estate Income Trust swapped self-storage for data centers. The fund sold the storage assets and put $3.3 billion into data center development. Data centers are the hot asset right now, but they don't produce current income the way self-storage did. A semi-liquid fund already facing redemption queues now holds construction risk that won't pay off for years.

The trade didn't happen in isolation. PWD's tracking shows a $12.7 billion wave of exits hit nontraded BDCs in the first half. The sector's repurchase caps held at 5%. That left a $9.6 billion queue of unmet redemption requests. That queue is now driving what sponsors do with their balance sheets.

HLEND, a nontraded BDC, took a $382.5 million markdown. Cumulative distributable earnings swung to a $342 million deficit. Redemption requests reached 13.3% of shares. A fund that can't cover distributions from earnings and faces redemption requests in the double digits owes more than it earns.

But sponsors aren't responding by raising repurchase caps or offering more liquidity. They're doing the opposite. The liquidity is being created privately, on terms that don't flow through to shareholders. Starwood REIT tapped Apollo for a $1.02 billion affordable-housing joint venture. PCAP, another nontraded BDC, borrowed $292.5 million from Norinchukin Bank. The loan pays SOFR plus 1.58% and matures in 2038. PCAP also expanded its Scotiabank revolver. Neither deal widens the redemption queue.

The BREIT pivot makes the trade visible. Self-storage has been one of the most liquid property types among nontraded REITs because it sells quickly and in smaller parcels. Data centers tie up capital for years. Selling the liquid asset to fund the illiquid one leaves BREIT with fewer ready-to-sell properties if redemptions grow.

The strategy may be right for investors who can stay. Data centers may offer higher long-term returns, and BREIT's managers are placing that bet. But the semi-liquid wrapper promises periodic liquidity. It was built on the idea that part of the portfolio could be sold to meet quarterly redemptions. When the most saleable assets are gone, that promise thins.

The semi-liquid wrapper, which covers interval and tender-offer funds, typically lets investors redeem 5% of shares each quarter at net asset value. The cap protects long-term investors from fire sales. When requests exceed it, the extra rolls into a queue. The first half's $9.6 billion queue is a stress test. Sponsors aren't widening the cap; they're reshaping what the fund holds.

Starwood REIT found a way to raise cash without selling to the secondary market. It brought Apollo into a $1.02 billion affordable-housing joint venture. The structure gives Starwood cash and keeps control of the assets. For shareholders, the deal creates a senior claim on part of the portfolio. That's a private side deal, not a redemption.

PCAP went further. The nontraded BDC borrowed $292.5 million from Norinchukin Bank. The loan pays SOFR plus 1.58% and matures in 2038. PCAP also expanded its Scotiabank revolver. The long-dated loan gives the fund room to hold assets through the queue. But the interest and seniority sit ahead of common shareholders. The sponsor is using private credit to manage what the public redemption mechanism can't.

HLEND's deficit illustrates the trade-off. A $342 million distributable earnings deficit means net investment income doesn't cover what the fund pays out. To keep distributions at the stated rate, the fund may need to return capital or draw on credit. The redemption queue adds another claim on the same balance sheet. So the sponsor turns to the private market, not to shareholders, for the fix.

A sponsor that sells assets to meet redemptions shrinks the fund and locks in losses. A sponsor that finds private capital can keep the fee base and deal flow. The private capital comes with terms—Apollo's JV capital, Norinchukin's 2038 loan—that are senior to shareholders. The sponsor keeps the economics. The investor keeps a capped exit.

Not all funds face the same pressure. PWD's data on NAV BDC repurchase queues shows redemptions hit a record, but demand split sharply. Three large tender-offer funds reported repurchase demand ranging from zero to a sharp jump. One fund's investors wanted out in large numbers; another's wanted to stay. The stress is not uniform across the wrapper.

But for investors who do want out, the options are narrowing. Private credit fund investors face capped quarterly redemptions on one side and steep secondary-market discounts on the other. The secondary market is where investors who cannot wait for a repurchase window go to sell. Those sales clear at discounts. The actual price of liquidity in a semi-liquid fund is not the NAV; it's something lower. The discount is the cost of the queue.

The BREIT pivot can be read the same way. The fund sold self-storage and put the proceeds into data centers, an asset that won't pay off for years. That consumes liquidity rather than adds it. If the redemption queue in nontraded REITs were empty, the trade would be a simple strategic rotation. In a queue, it looks like a bet that the investors who want out are a minority and that the rest will accept less liquidity for higher long-term returns.

The semi-liquid market's pitch was that individual investors could get private assets without giving up all liquidity. What happens when that liquidity is tested is now visible: sponsors do not open the gates; they sell the gates' hinges to buy more building. Investors wait in a queue or sell at a discount in the secondary market. The next test is whether the queue grows faster than the private capital arrives.