The debt Wall Street can't see is now on Wall Street's books

Editor’s note: Every financial crisis has a "we didn't know it was there" chapter. In 2008, it was leverage buried inside CDOs. Today, it is a $140 charge on a checkout button that never touches a credit bureau. And now the largest private credit firms in the world are buying that debt in bulk. Well…what could possibly go wrong?

But before we get to that, let's take a quick look at the markets and what matters…

The week ends where it started, on the AI trade. SK Hynix makes the second-biggest US listing in history today, even as the chips that carried the market all year wobble into the close.

3 Movers in 3 Minutes

Confirmed by satellites 300 miles above the Earth's surface... Elon Musk is rolling out a breakthrough technology that could replace our need for foreign oil and ignite a $10 trillion boom a small group of stocks. Click here to learn how you can invest in this before it becomes mainstream.

And with that out of the way, here's today's big story: the debt Wall Street can't see, and why the biggest names in private credit are now the ones holding it.

On September 18, 1958, 60,000 residents of Fresno, California opened their mailboxes to find an object none of them had asked for.

A small piece of paper. A $300 to $500 line of credit. No application. No signature. No conversation with a banker.

Bank of America had picked Fresno for two reasons. It was big enough to matter and small enough to bury a scandal if things went sideways. And 45% of the town already banked with them.

Within thirteen months, 2 million BankAmericards were in circulation across California. Within a decade, the card became Visa.

But that first year was ugly. Delinquency ran above 22%. Fraud was rampant. Joe Williams, the executive who dreamed it up, quit within two years. He had assumed collections would sort itself out. It didn't.

Because the moment you separate the purchase from the payment, human beings behave differently. And nobody quite knows how much they owe until the envelope arrives.

The BankAmericard didn't just extend credit. It hid the price of things behind a plastic film.

Which brings us to today.

You know that little checkout button. The one that says "4 payments of $34.75." You have probably clicked it. And so have millions of Americans, mostly quietly, mostly without thinking of it as a loan.

The Richmond Fed estimates US buy-now-pay-later transactions hit roughly $70 billion in 2025. That is the polite number. Broader industry counts, including subscription splits and longer installments, put 2025 gross merchandise volume past $560 billion.

Either way, the number itself is not what regulators are worried about.

It's what nobody can see.

Because unlike a credit card, most BNPL loans don't show up on your credit report. The CFPB confirms it. Affirm only began full reporting to Experian in April 2025. Klarna, one of the largest players in the US, still doesn't report transaction data at all. FICO's new BNPL-inclusive score, launched in Fall 2025, is still rolling out.

The result is what the industry now calls…phantom debt.

The average BNPL user takes out nine loans a year. Sixty-three percent of them hold multiple loans at the same time. Sixty-one percent have subprime credit scores. When they apply for a car loan or a mortgage, the underwriter has no idea any of it exists.

The invisible ledger doesn't just hide debt. It hides distress.

Here's the part the retail spending headlines miss.

In 2025, private credit funds bought roughly $136 billion of consumer loans, a category that includes BNPL receivables packaged into securitized pools. That is fourteen times what they bought the year before.

Fourteen times.

KKR has committed €65 billion to buy PayPal's European BNPL receivables through 2028. Warehouse facilities, forward-flow agreements, asset-backed securities. The plumbing that turned subprime mortgages into 2008 is being rebuilt around subprime consumer credit.

And who is on the other side of these private credit funds? Pension funds. Insurance companies. State retirement systems. The ordinary end of Main Street's balance sheet is quietly funding the invisible end of Main Street's spending.

Now stack this up. Late payments on BNPL loans went from 34% of users in 2024, to 41% in 2025, to 47% in 2026. Nearly half of everyone using these products is now paying late. Loss rates on the standard "pay in four" product historically stayed under 1%. But the loans are so short, delinquency can spike before servicers even notice.

Which raises an uncomfortable question. If retail spending has been "resilient" through a period of high rates and slowing wage growth, how much of that resilience is real income and how much is uncounted debt showing up as consumption?

Turns out nobody can quite tell you. And that is the problem.

Joe Williams thought delinquencies on his Fresno drop would stay under 4%. They ran past 20%. He'd built a distribution model without a collections model. He'd invented a way to hand out debt without first inventing a way to track it.

The market eventually caught up. In 1970 banned unsolicited card drops. But not before more than 100 million cards had been mailed into the American population, most of it by BofA and its licensees.

BNPL sits in the same gap now. Distribution is exquisite. Reporting is broken. Credit bureaus scramble to catch up. The CFPB pushes for standardized data while the market grows around it.

And private credit, with no bureau data and no full picture of borrower leverage, is buying the receivables anyway.

Every widely held financial crisis begins the same way. Someone finds a way to move a pile of debt to a place where the usual measuring stick can't reach it. Off-balance-sheet vehicles. Structured products. Reinsurance sidecars. Now, checkout buttons.

The Fresno Drop showed us what happens when credit is easier to give than to see. The BNPL era is showing us what happens when credit is easier to give than to count.

Numbers on an economy only work if the numbers include everything. When the ledger goes dark on even a small slice of the household balance sheet, every macro chart, every "consumer strength" call, every retail sales beat, deserves an asterisk.

Because when debt disappears from the ledger, it doesn't disappear from the economy. It just shows up somewhere else. Usually louder. Usually later. And usually in the hands of someone who thought they were buying safe income.

When a small group of private companies — not the Fed — will perform a major mint of a new kind of money. 

And those who act before this new system fully kicks in could see gains as high as 40X by 2032. 

But those who fail to prepare will be blindsided by this sea change to the U.S. dollar.

Go here now for the details — before the September 8th mint hits the market.

Reply and tell us: if half of BNPL users are already paying late, is the "resilient consumer" story real, or is it running on invisible fuel?

Until then, sip slowly!

The Market Sip Desk