Private credit means lending to companies outside traditional banks. A fund collects money from investors and lends it to businesses that pay interest. The return can be attractive. The risk is harder to see because these loans do not trade on an exchange every day.
In March, we wrote that problems in private credit often appear with a delay. New numbers have since arrived. More investors want to leave major funds, the share of troubled loans has risen, and more private loans are ending up inside life insurers.
The main question is simple.
Who pays the loss if borrowers stop repaying their loans?
In Private Credit: Big Market, Big Risk, we described three weak points in these funds. The loans are difficult to sell, their real value is checked only from time to time, and some companies pay interest with more debt instead of cash.
The second quarter brought another warning sign.
Investors in Apollo Debt Solutions asked to withdraw about 16.8% of the money in the fund. Its rules allowed it to pay out only 5%. Investors in Blackstone's BCRED asked to withdraw about 10%, but this fund also had a 5% limit.
This does not mean that Apollo or Blackstone will fail tomorrow. The limits are written into the fund rules. The reason is simple. A loan to a private company cannot be sold within seconds like an ordinary share.
The numbers do show that some investors are becoming nervous.
If 17 people out of 100 want to leave and the fund can pay only five, a queue forms. Everyone else has to wait.
Fitch reported that the share of troubled US private credit loans reached 6% in the twelve months through April. This was the highest level in its data.
More than half of these cases involved a company that did not pay interest in cash. Instead, the unpaid interest was added to its debt. This is called payment in kind, or PIK.
A simple example helps.
A tenant does not pay the rent. Instead, the tenant signs a note promising to owe you even more next month. Your claim is larger on paper. You still have no money in your wallet.
PIK can help a company through a short period of weakness. If it keeps happening, it may simply hide the fact that the company does not have enough cash.
A 6% default rate looks like a precise number. In reality, the result depends on what is counted as a default.
A company does not have to file for bankruptcy. Its lender may extend the loan, temporarily waive a broken rule, or allow the company to pay interest with more debt. The company may then stay outside the default statistics even though it does not have enough cash.
Fitch, for example, does not count some companies with very weak credit ratings if they have not yet formally defaulted. It also excludes some covenant breaches and PIK interest when the right to defer interest was included in the original loan agreement.
The difference can be large. For December 2025, Fitch reported a 5.6% default rate for its full sample. The rate was 9.2% for a smaller group of companies that it monitored in greater detail. This does not mean that the whole market was at 9.2%. It shows how much the result can depend on the companies included and the method used.
The Financial Stability Board described a similar gap. In one study, ordinary defaults were close to 1%. The figure rose to around 5% after including cases where lenders changed the terms to help a company avoid failure.
Loan values create another problem. Private loans are often revalued only once every quarter. A deterioration may therefore take time to appear on an account statement.
An official rate of 6% does not mean that only six companies out of 100 are in trouble. It means that six have already met the specific rules for being counted as defaults. Others may be surviving through extensions, contract changes and additional debt.
We do not know the exact number. We do have good reasons not to treat the official figure as the complete picture.
A life insurer collects premiums from customers. It invests the money so it can pay insurance claims and other promised benefits in the future.
Insurers have traditionally held bonds, mortgages and similar long term investments. Private credit offers a higher interest rate, which makes it attractive.
Large financial groups may own both an asset manager and an insurer. In other cases, the two companies work closely together. Money belonging to insurance customers can then be invested in private loans selected by a manager from the same group.
The chain looks like this:
When the company repays its loan, the system works. When it stops paying, the problem no longer stays only inside a fund. It may also appear inside the insurer that has promised future payments to its customers.
A new study by Andrew Granato and Pranjal Drall points to a simple problem.
The manager collects fees every year. The insurer may report a better return because of the higher interest. A real loss on a bad loan may not appear for several years.
It is also difficult to know what a private loan is worth today.
We can see the price of a listed share or an ordinary bond every day. The value of a private loan may be estimated for months by the same manager that earns fees from holding it.
A stable number on an account statement does not always mean that an investment is safe. It may only mean that its price is checked less often.
If a US life insurer fails, each state has a safety fund that helps protect the customers of that insurer. Other insurance companies contribute to these funds.
Granato and Drall note that insurers in many states can deduct these contributions from future premium taxes. Part of a very large loss could therefore move indirectly to public budgets.
Investor Michael Burry highlighted this possibility on 25 July.
It is important to say what this means and what it does not mean.
It does not mean that the US government is currently rescuing a specific private credit fund. It also does not mean that a well known insurer will fail tomorrow.
The study shows that private companies may receive profits and fees for years, while other insurers and public budgets might eventually carry part of a very large loss.
This is a warning about poorly designed rules. It is not a prediction of an immediate collapse.
Private credit is increasingly financing data centres, software and other projects linked to artificial intelligence.
The Bank of England cited an OECD estimate showing that private credit's share of AI financing rose from 9% in 2024 to 34% in 2025.
The money may move through the system like this:
There is no problem if the data centre makes enough money. If demand for computing power is weaker than expected, the company may struggle to repay the loan. The loss can then move from the borrower to the fund and from the fund to the insurer.
This is why it is not enough to watch Nvidia's share price or the results of large technology companies. Debt linked to the AI story may be sitting somewhere else entirely.
No.
The Alternative Credit Council examined more than 70,000 private loan valuations. In total, 87% were valued above 97% of the original amount lent. These numbers suggest that the problems are still concentrated in only part of the market.
Redemption limits also have a purpose. They stop a fund from selling loans quickly at a bad price simply because investors have become frightened.
A 6% share of troubled loans is a warning. It is not proof that every fund is collapsing.
The Granato and Drall study is also still a working paper. It may start an important debate, but it is not the final word on the market.
There is no reason to panic. There is a reason to demand clearer information.
Four things matter most to an ordinary reader:
A higher return is never free. The important questions are who valued the loan, who receives a fee for managing it, and who carries the loss if the borrower stops paying.
Private credit did not turn from a safe investment into a disaster within one quarter. The new data do show that earlier warnings are no longer only theoretical.
More people are asking to withdraw money. The share of troubled loans has risen. Some companies are paying interest with more debt. Risky loans are appearing more often inside insurers.
What could this mean for US markets?
Shares of companies directly linked to private credit would probably react first. These include asset managers, listed credit vehicles, some insurers and banks that lend money to private credit funds. Software companies and AI projects that depend on private loans could also come under pressure.
If investors began to doubt loan values across the market, the pressure could spread to other shares. Investors would demand a greater reward for taking risk. Less profitable and heavily indebted companies would probably fall fastest. Strong companies with plenty of cash should hold up better, but a large selloff rarely affects only the weakest shares.
The effect on corporate bonds would be more direct. Prices of risky bonds would fall and their yields would rise. Companies that need to replace old debt with new borrowing would face higher costs. Some might not be able to obtain new financing at all. This could create another round of distress and defaults.
High quality corporate bonds would probably perform better than risky debt. Their spreads over US government bonds could still widen if investors began to demand more compensation for every form of corporate risk.
US government bonds may act as a safe haven during a pure credit panic. Investors buy them, their prices rise and their yields fall. This is not guaranteed. If the same shock also involved expensive oil and higher inflation, equities and long term government bonds could fall at the same time.
The Federal Reserve could provide liquidity or cut interest rates if the financial system were genuinely threatened. It would not necessarily intervene simply because several funds limited withdrawals. High inflation would also reduce its room to act.
It is important to keep the scale in perspective. The Federal Reserve estimated that private credit represented about $1.4 trillion, or roughly 10% of the debt of US nonfinancial companies. At the same time, it described the redemption requests seen so far as manageable. Private credit does not currently make a new financial crisis inevitable.
The main danger is a chain of events. Borrowers stop paying. Funds reduce the value of their loans. Investors request withdrawals. Funds draw on bank credit lines or sell the assets that can be sold quickly. Prices of risky assets fall, and financing becomes more expensive even for healthier companies.
This is how a problem in a group of private loans could become a problem for the wider market.
The most important questions are very ordinary:
Who receives the fee?
Who decides what the loan is worth?
Who pays when the borrower stops repaying the loan?
If these questions do not have clear answers, a calm number on an account statement may not mean a calm investment.
Note: This text is for information only and does not constitute investment advice. Investing in financial markets involves risk, and readers should conduct their own analysis before making any decision. The text reflects publicly available information as of 3 August 2026.