JPMorgan’s $250B Private-Credit Test Is a Liquidity Trap
A 9% yield is not income if you cannot get your money out. JPMorgan says roughly $250 billion of private credit now sits in semi-liquid funds largely owned by individual investors.
The yield is real. So is the trap.
A 9% yield is not income if you cannot get your money out when the world turns ugly.
That is the uncomfortable truth sitting underneath the private-credit sales pitch. JPMorgan Asset Management estimates roughly $250 billion of private credit now sits in semi-liquid funds owned predominantly by individual investors. “Semi-liquid” is finance-speak for: you may be allowed to ask for your money back, but the manager may not have to give it to you on your timetable.
That distinction matters a hell of a lot more than the glossy brochure’s yield chart.
Private credit has been marketed to ordinary wealthy investors as the grown-up alternative to shares: higher income, less volatility, access to deals banks supposedly left behind. Then 2026 handed the industry a nasty stress test. Fears that AI could weaken parts of the software sector hit the very companies many private-credit funds had lent to. Investors began requesting redemptions. Some funds limited withdrawals.
Nobody needs to panic. But anyone treating private credit as a cash substitute needs their head read.
Why JPMorgan is looking under the bonnet
The central issue is software.
JPMorgan’s wealth team estimates private credit has roughly 21% exposure to software. Add broader technology and business-services borrowers, and the exposure rises to around 40%. That is not a trivial side bet. It is a meaningful chunk of a market that sold itself on diversification.
The concern is not that every software company disappears next Tuesday. It is that AI has made investors reassess which software businesses have durable pricing power, sticky customers and actual cash flow — and which ones are expensive digital paperwork with a subscription button.
That matters because private-credit lenders often financed companies that were bought at punchy valuations, with meaningful debt, during the software boom. If a borrower’s revenue growth slows, customer retention slips or its product becomes easier to replace with AI-enabled tools, the lender’s downside changes quickly.
Public markets mark that pain every day. Private credit does not. Its valuations are generally updated less frequently, using models and manager judgments rather than a live market price. That can make the ride look calmer than it really is.
Calm is lovely. Fake calm is dangerous.
The Wall Street Journal reported in March that JPMorgan CEO Jamie Dimon had ordered a review of the bank’s own software exposure, while the bank tightened lending to some funds based on their exposure. The same report said banks were developing ways for sophisticated clients to hedge or bet against companies exposed to private credit.
Read that again. The people providing the plumbing are checking the pipes while retail investors are still being sold the view from the penthouse.
The product’s promise was always conditional liquidity
Private credit itself is not rubbish. Good lending is a terrific business. I have built businesses and invested in them long enough to know that lending against sensible cash flow, with proper covenants and real security, can be a far better risk than chasing fashionable equity stories.
But the product structure matters more than the asset-class label.
A listed bond fund owns assets that can normally be sold quickly. Its price moves every day, sometimes brutally. A private-credit fund may own loans that take months to sell, renegotiate or restructure. Yet many newer vehicles offered wealthy individuals periodic redemption windows — monthly or quarterly — to make an illiquid asset feel more convenient.
That is the mismatch: investors receive a suggestion of liquidity while the manager owns assets that are genuinely illiquid.
Gates are not necessarily evidence of fraud or impending collapse. In fact, limits on withdrawals can protect remaining investors from a fire sale. If a fund receives redemption requests from 12% of investors but permits only 5% of assets to be redeemed in a quarter, that cap can stop the manager from selling the easiest loans first and leaving everyone else with the dregs.
But it also proves the point: the liquidity is conditional.
You do not own cash. You own a place in a queue.
The overlooked risk is not just default — it is behaviour
Most coverage focuses on whether private-credit defaults will spike. Fair enough. Defaults matter.
JPMorgan’s stress work put some useful numbers around the downside: if software defaults reached 15% and recoveries were only 40%, losses in private credit could reach around 2% before fund leverage and roughly 4% including leverage, against a starting yield of about 9%.
That is a scenario, not a forecast. It is also more nuanced than the hysterics suggest. JPMorgan’s own analysis said stresses appeared concentrated in smaller borrowers and selected sectors, and it did not see an imminent broad-based default cycle. Its April update noted non-accruals in non-traded business-development companies were around 1.2% of cost, below the 10-year average of 1.9%.
So, no, I am not saying sell everything and hide under the doona.
I am saying the more important risk may be human behaviour. Investors who thought they owned a low-volatility income product can react very differently when they learn the exit door is narrow. The worst time to discover you dislike illiquidity is when everyone else has discovered it too.
That is why the 9% yield needs to be judged against three things: credit risk, fee drag and access to capital. Most people only look at the first one.
The contrarian view: gates can be a feature, if you price them honestly
Here is the bit most people miss: a redemption gate is not automatically bad. For a genuinely long-term investor, it can be rational.
If you have capital you will not need for seven to 10 years, understand the fund’s lending standards, accept that values may lag reality and are being paid properly for giving up liquidity, private credit can have a place in a diversified portfolio.
The problem is that plenty of investors have been sold a different mental model. They hear “monthly liquidity” and translate it as “my money is available monthly.” Not the same thing. One is a request mechanism. The other is a guarantee.
Private credit should be treated more like a small holding in commercial property or private equity than a high-interest savings account. It may produce income. It may compound well. But it should never be money earmarked for a tax bill, a house deposit, a business payroll gap, university fees or the emergency fund that keeps you from making desperate decisions.
And founders should pay attention too. This is not only an investor story. If your company relies on private lenders, AI is now part of your credit committee whether you like it or not. Lenders will ask harder questions about churn, customer concentration, pricing power, automation risk, debt-service coverage and whether your software is essential or merely convenient.
“AI strategy” has become an annoying phrase. But if you run a software business, showing precisely why your customers cannot replace you is no longer branding. It is financing.
What this means for you
Do this tomorrow, not after the next wobble:
1. Find every illiquid investment you own. Look through your superannuation, adviser portfolio, family trust and investment platform. Search for private credit, non-traded BDCs, interval funds, private debt and alternatives.
2. Read the redemption terms. Do not read the marketing sheet. Read how often you may request redemption, how much the fund may repurchase, whether requests can be prorated, and whether the manager can suspend withdrawals.
3. Match the asset to the job. Keep emergency money and any money needed in the next three years in genuinely liquid assets: cash, term deposits, short-duration government securities or equivalent boring stuff. Boring is underrated when life gets expensive.
4. Ask what you are being paid for. A yield premium is compensation for risk. If a fund offers 9% while liquid high-quality bonds offer less, the difference is not a gift. It is payment for credit risk, illiquidity, leverage, fees or all four.
5. Make managers explain the software book. Ask for sector exposure, the largest borrower exposures, proportion of payment-in-kind income, non-accruals, loan-to-value assumptions and historic redemption fulfilment. If the answer arrives wrapped in corporate fog, that is your answer.
6. Do not confuse a smoother statement with a safer investment. Private assets can be excellent. But a quarterly valuation is not a force field. It is just a slower mirror.
The sensible conclusion is not “private credit is dead.” That is lazy.
The sensible conclusion is that liquidity is an asset. Treat it that way. If you give it up, demand to be paid properly, understand exactly when you can get your capital back, and never put essential money behind a gate.
That is how you keep a nice yield from becoming an expensive lesson.