Private Credit Is Not What You Fear

The 2008 crisis scared the world so badly that we still look for the next one everywhere. Almost every year since, some corner of finance has been nominated for the role of the-thing-that-breaks-everything, and right now the nomination has gone, nearly unanimously, to private credit.

Before deciding whether the label fits, it is worth asking why we are so certain the next crisis is always out there, waiting.

We humans live at the conjunction of forethinking and optimism bias, so we invented credit before we invented money. In the constant pursuit of excess we long for something as seductive as it is imprecise: we want more, and more is boundless. But whatever we are counting on tomorrow may as well not come, and sometimes it doesn't, or worse, something far darker takes its place, and optimism turns to panic.

When one of us panics loudly enough, the rest of us follow. The panic later gets called a crisis, but, in reality it's just a mass realisation of everything we did in the moments before. That is when credit stops looking like such a good idea.

Then enough time passes, we patch up the harm, and we go looking for a culprit, a single one, one we will always remember, because we hate being wrong. Credit, we notice, has been there all along, at the scene of every boom. So it must be IT, because we want the story we tell ourselves to be that simple, while a financial crisis is anything but.

For a financial crisis to truly detonate, four things have to be true at the same time:

Any one of these alone produces losses. It takes all four, together, to produce a 2008.

We are not wrong that credit is the compounding factor, which is precisely why, after each collapse, we re-engineer it: we make the mechanics more sophisticated and wrap new institutions around them, trying to hold the balance between the gains we want and the risk we fear.

Credit itself is older than money, and older than writing. The oldest financial records we have are clay tokens from Mesopotamia around 8000 BCE, each one standing for grain or livestock owed. By Hammurabi's day there were already interest ceilings and rules that paused debts after a bad harvest, the oldest financial regulation we know of. The story runs on from there through Rome's first recorded credit crunch and bailout in 33 CE, to the great speculative bubbles of the 1700s that set a template every generation since has somehow repeated.

As the economy grew, as industrialisation and then globalisation carried it further, the financial structures that paid for all that development grew right alongside it, and so did their fragility. Every new layer of financing made the next stage of growth possible, and made the whole edifice a little easier to topple. It really only climaxed, though, in the 20th century.

The Crash of 1929 came, and took 9,000 banks with it. Then, the Great Depression followed. Out of the wreckage came Glass-Steagall, deposit insurance, and the SEC, and it worked so well the United States went nearly 50 years without a major banking crisis.

Until, at some point, we started undoing it. In 1971 Nixon cut the dollar's last tie to gold and the ceiling on credit creation slowly vanished. Securitisation let lenders stop holding their own risk. In 1999 Glass-Steagall was repealed, by people who knew exactly what it was for. And derivatives, freed from oversight, swelled past $600 trillion by 2007¹. The guardrails had come off just as the instruments grew too complex to see through.

Which brings us to the one crisis still fresh enough to frighten us. What happened in 2008 is easy to break down in hindsight:

• A group of people inside the credit system made huge bets with borrowed money.
• Those bets went bad and almost broke the entire financial system.
• Afterwards, new rules made those kinds of bets impossible to run inside banks.
• And when the banks pulled back, many companies could no longer get loans.

And yet, for all its scale, 2008 was not a new kind of event, only the largest run of the oldest play: cheap money, an asset everyone believed could only rise, risk buried in instruments too clever to see through, and a trigger to set it alight. Only this time, because of everything we had dismantled, it did not stay contained, it went everywhere.

The aftermath of the GFC is where our conundrum really begins.

Private credit grew, over the following decade, to fill exactly the one gap left by the shock. It lent to the companies the banks had been forced to leave behind, and because it did so privately, it could charge more, which meant it paid more, which meant more and more money flowed toward it.

That privacy is also why it makes people nervous: the deals are bilateral and the terms are hidden, and a thing you cannot see is easy to imagine the worst about. Somewhere along the way private credit picked up its nickname, the cockroach of the financial system, the creature you assume travels in numbers the moment you spot one.

It accumulated, over roughly two years, until a pair of unrelated corporate failures gave it a face and a phrase.

The IMF gave private credit a chapter of its own in 2024³, and through 2024 and 2025 the Fed, the FSB, the ECB and the ESRB each added concerns of their own⁴,⁵,⁶. Then, in September and October 2025, two lenders collapsed within weeks of each other, Tricolor and First Brands, each having hidden far more debt than anyone realised, and that October JPMorgan's Jamie Dimon gave the fear its phrase: when you see one cockroach, there are probably more⁷. That was the line that turned "private credit is the next 2008" into a front-page contagion story.

What almost nobody said, in the rush to connect them, was that the two failures that lit the fuse had nothing to do with private credit. Tricolor was a subprime auto lender and First Brands an auto-parts maker; both were undone by hidden borrowing, not by the mechanics of private lending.

Despite all that noise, private credit is not built in a way that could cause a 2008-style collapse. It has genuine weak spots too, and we'll come to those honestly. But none of it means much until we know what we're actually dealing with, so it's worth starting with the simplest question there is: how big is this thing, and where does it sit?

Depending on who is counting and what they choose to count, private credit is worth anywhere from about $1.2 trillion to more than $30 trillion. However, the number we are going to hold for the rest of this paper is roughly $1 trillion actually deployed in the United States.

Because there is a lot of anxiety around these numbers, it is worth laying out the rest of the figures in the wild, and why they differ, before setting them aside. Most of the gap comes down to three choices: whether you count money actually lent or also the committed-but-unlent "dry powder"; whether you count only plain corporate direct lending or a much broader set of strategies; and whether you count what exists today or what could one day migrate out of banks.

When the question is risk rather than marketing, only one of these counts, and it is the money actually out on loan, not the dry powder, which, if anything, is a stabilising force. Given all the educated estimates above, the number we will hold to for the rest of this paper is the $1 trillion deployed in the United States.

A number means very little without a proper context, and private credit's natural yardstick is the world it lives in: the non-bank financial system, or NBFI, everything that lends and invests but is not a central bank, a commercial bank, or a public institution. That means insurers, pension funds, investment funds, money-market funds, broker-dealers, hedge funds, finance companies, structured-finance vehicles.

Private credit lives inside one corner of this $256.8 trillion system⁵, the bucket the regulators call Other Financial Intermediaries, itself roughly $169 trillion and the fastest-growing part of the whole system, up 11% in a year⁵. And inside that corner, private credit is about 1%.

Put the two numbers next to each other and the fear starts to look strangely out of proportion.

At $1.5-3.5 trillion however you count it, private credit is about 1% of the $256.8 trillion non-bank system, and roughly 2% of the OFI bucket it sits inside.

Even the slice regulators genuinely lose sleep over, the narrow measure, some $76.3 trillion of entities doing the bank-like things that can threaten stability, is more than 20x its size.

Judging by this, private credit is close to a rounding error in the system it is accused of being able to break.

Because size in the aggregate is not the same as size where it lands. Private credit may be 1% of the non-bank world, but it does most of its lending in a few concentrated places.

There are roughly 33 million private companies in the United States¹³, and together they turn over something like $26 to $29 trillion a year. That averages to a few hundred thousand dollars of revenue apiece, but the average is close to meaningless: the distribution is violently skewed, a handful of very large private firms sitting on top of millions of tiny ones.

By sheer count this private economy dwarfs the public one; by concentration it's the reverse.

Counting them at all is the hard part, and it is what we do at Veridion. We keep a live, company-level map of the global economy, more than 680 million businesses tracked from their real-world footprint on the open web and refreshed as they change, so the private and the opaque become as legible as the listed. That is what lets us do the rest of this paper, sizing private credit from the borrowers up rather than taking the market's word for it.

Let's take the $29 trillion of revenue, and work down. Private firms don't earn public-company margins: let's assume a roughly 7% operating margin, not the ~11% listed companies manage. That leaves something like $1.8 to $2 trillion of operating profit (EBITDA a little higher, $2 to $2.5 trillion).

Debt service comes off first, about $850 billion a year in interest at a blended rate near 6.5%, on $13.2 trillion of private-company debt as gathered by the Fed's latest numbers⁴.

Tax takes another ~$210 billion; say the effective rate sits around 22%, below the ~28% public firms pay. What's left, for all 33 million companies combined, is on the order of $0.74 trillion of net profit.

Three-quarters of a trillion dollars is what the entire American private economy clears in a year after interest and tax.

That is the cushion the whole edifice of private-firm debt ultimately rests on.

That $13.2 trillion comes from six quite different creditors, each with its own risk profile. This is how the debt is split between them:

So private credit is about $1 trillion of $13.2 trillion, roughly 8% of everything private American companies owe. It is also the most expensive money in the table: private credit charges something like 11%, against the ~6.5% blended average⁴,¹¹,¹².

Filter that map down to the firms that actually lend, and Veridion finds roughly 3,000 US companies¹³ engaged in private-credit activity, only about 250 of them also doing private equity. It's a crowded space, but a top-heavy one: the 10 largest managers have raised more than half of all the capital, and the top 100 about 72%⁴,⁸. When people say private credit, they are mostly talking about a few dozen firms.

We count a little over 200,000 mid-market companies in our data¹³. The National Center for the Middle Market counts roughly the same cohort¹⁴. These companies are earning $10 million to $1 billion in revenue, across every sector, but private credit touches only a small portion of them. KBRA's direct-lending database¹⁵ tracks somewhere between about 1,972 and 2,416 private-credit borrowers, carrying more than $1 trillion in debt between them. Most of that lending rides on private equity: around 70% of private-credit deals are PE-sponsored³, and private-debt funds supplied 77% of global leveraged-buyout debt in 2024¹⁶.

Because private credit lends against cash flow and enterprise value rather than hard collateral, it wants a particular kind of borrower: asset-light, with recurring, predictable revenue, usually PE-sponsored, and, at the core of the market, throwing off EBITDA in the $25 to $100 million range¹⁷. That profile funnels the money into a few industries; the top three sectors account for roughly 60% of all private-credit portfolios.

Our data puts hard numbers on that headroom: only about 95,000 to 125,000¹³ US companies are "scaled" enough and sit in the right sectors (healthcare, software, business services) to plausibly make a private-credit target list. Against that, the penetrated market is small: a typical middle-market loan runs $100 to $300 million, so $1 trillion deployed implies somewhere between roughly 3,000 and 10,000 actual borrowers.

Even against the smaller end of the addressable pool, that's penetration in the low double digits at most, which describes a market still expanding, not a saturated one.

Knowing its size and shape and the fact that quite a lot more borrowing can be done, we can now ask the unsettling question: what happens when some of it goes bad?

Despite the very human instinct to reach for the worst scenario first, it's worth saying plainly that a 100% default rate across an entire asset class isn't even a remote possibility, it would mean the end of an economy. Something far worse than defaults would be happening in that world, and we'd be a good deal more concerned with our own survival than with the survival of a few credit institutions.

So the honest test is: what if far more defaults than ever has, and then some? The debate today puts the headline default rate around 2-3%²¹. Fitch calls that conservative, it excludes several kinds of distress, and puts it at 5.8% on its broad tracker and 9.4% on its privately-monitored slice²⁰. Morgan Stanley bets on 8%, while UBS inflates it to 13%⁹,²².

Let's say it's far worse than all of that and assume a 20% default rate, nearly ten times the historical norm, which would mean $200 billion of loans going bad. But a defaulted loan isn't vaporised: lenders recover, through collateral, enterprise value, restructuring, and private credit, lending senior and secured against going concerns, recovers well. So on that supposed $200 billion of defaults, the actual losses land somewhere around $80 to $120 billion. That is still a lot of money, yet it lands below a single year of JPMorgan's revenue, and at roughly a tenth of the ~$1.8 trillion American financial institutions lost in 2008²⁶.

The securitised products at the heart of the last crisis weren't enormous either, and one is roughly the size private credit is now. So why would this time be different?

Well, it is so, by design, and to see why, you have to remember what those products actually did.

The mid-2000s were awash in cheap money, a Fed that cut to 1% and stayed, a savings glut from China and the oil exporters pinning down long-term rates, and a housing market everyone had reason to keep rising. The trouble was never only that subprime borrowers took on debt, but what was built on top of that debt.

Mortgages were pooled into bonds (MBS), the riskier leftovers were repooled into CDOs and stamped AAA on the assumption that mortgages don't all go bad at once, and then the multipliers arrived: CDO-squared built from the slices of other CDOs, synthetic CDOs that held no loans at all but let the same mortgage be bet on over and over, and a single insurer, AIG, standing behind much of it. So when house prices turned in 2006 and adjustable-rate borrowers could neither refinance nor sell, the damage didn't stop at the loan, it multiplied through everything stacked above it. That leverage-on-leverage, not the mortgages themselves, is what turned a housing downturn into a global crisis.

This is where every skeptic raises an eyebrow at once: CDOs and CLOs rhyme, so surely private credit is just the same trick under a new name. A CLO is a securitisation: it pools loans, tranches them, sells the tranches. The resemblance is real, but it ends at the surface, and for two reasons.

First, the CLO market is overwhelmingly built from broadly-syndicated loans arranged by banks, roughly $600B+¹⁶,¹, versus about $150B of private-credit CLOs¹⁶ (2025): about 80/20. Private credit is the minority of it, even as its share of new issuance climbs (~17% in 2025).

Second, and decisively, the multipliers are gone. No CDO-squared or synthetic CLOs or AIG-style insurer behind the market are possible today. Post-crisis capital rules make re-securitisation so expensive it isn't worth doing. Everything that's left is the plain-vanilla version of securitisation, pooling and tranching, without the leverage-on-leverage that was the actual detonator in 2008.

To further understand why this cannot go systemic, let's look at the structure of a private credit institution: a manager (the GP) raises money, 98-99% of it from LPs (pensions, insurers, sovereign funds, increasingly wealthy individuals) and lends it to companies, usually PE-owned, that either can't or won't borrow from a bank.

Say a PE-owned software firm with ~$50M EBITDA wants to buy a competitor and needs $300M. A private credit manager originates a single senior secured loan, cushioned by the ~45% equity the sponsor put in, and funds it by allocating slices across the vehicles it runs: a closed-end drawdown fund, a perpetual BDC, an insurer's managed account, an interval fund. If the company thrives, the loan is repaid with interest. If it can't cover cash interest, it flips to PIK. And if it defaults outright, the manager, lending senior and secured, recovers around 60 cents, leaving perhaps $120M of loss, split among those funds in proportion to their slices.

Importantly, the loss stops there. It lands on long-lock LP capital that signed up for exactly this risk, not on a chain of bets held by banks that owe each other.

Blue Owl's tech BDC saw investors ask for 40.7% of shares back in Q1, its credit BDC 21.9%, some $5.4B, and it capped withdrawals at 5%; Blackstone's $82.5B BCRED drew $3.8B of requests, and Blackstone put in $400M of its own to meet them.

Sounds alarming, but the model IS illiquidity: investors trade access to their cash for higher yield, and most of private credit simply cannot be "run." BDCs are the one exception, because they let retail in, and even there the design holds: a publicly-traded BDC just sees its share price fall while it keeps the loans, and a non-traded one meters exits through those ~5% quarterly gates. Redemptions surged, valuations dipped, and the thing that would make it systemic, a bank run that forces a fire-sale, cannot happen by design.

The beauty of BDCs is that they file with the SEC and report quarterly, so using the Veridion infrastructure, we pulled the loan-level books of the 30 largest BDCs¹³: 139,000 individual loans across ten quarters, end-2023 to end-2025, recomputed from the filings:

It doesn't look terrible. But it isn't nothing, either: we can see a real, concentrated drift that we'll keep watching the following quarters.

Rising defaults and too much leverage alone did not blow the housing downturn out of proportion. It was that all of it was sitting inside the banks. Banks were, and still are, at the heart of the global economy: they fund themselves overnight, they run the payments system, they hold the savings of people who assume that money is simply there. One of the more recent, and more serious, worries is exactly that: that the banks are getting tangled up with private credit.

In practice, this entanglement is far more tangled than we'd like it to be, and not necessarily as dangerous as it's made to sound.

The banks have got into private credit from two directions at once.

The first is that they've started joining it, teaming up with private-credit firms, or building vehicles of their own. Citigroup and Apollo have set up a $25 billion direct-lending programme, one of the largest partnerships in the space. JPMorgan has raised its own direct-lending commitment to $50 billion, with another $15 billion from co-lenders. Wells Fargo and Centerbridge built a $5 billion BDC, Overland Advantage, run through Wells's commercial-banking network. Bank of America launched a $25 billion private-credit initiative in February 2026. And it isn't only the giants: regionals like Fifth Third and Webster have formed their own private-credit partnerships.

The second direction is that they lend to it. The six largest US banks now have somewhere around $300 to $322 billion of committed lending²⁴ to private-equity and private-credit sponsors, through subscription lines, NAV loans, warehouse facilities and BDC financing. That is up from roughly $10 billion in 2013, a thirty-fold jump in a decade²³. Moody's frames it as about $300 billion of private-credit exposure sitting inside more than $1.2 trillion of total bank lending to non-bank financial firms, with estimates running anywhere from $270 to $500 billion depending on whether you count what is committed or only what is actually drawn.

Banks extend subscription lines, short-term credit against the money investors have promised a fund but not yet paid in. They make NAV loans against a fund's existing book. They provide the leverage facilities behind the roughly 1.2x leverage a BDC runs, borrowing at around 5% to lend out at around 10%. And they extend warehouse lines that hold loans while a fund or a CLO is assembled. Beyond funding it, banks feed it deals, sometimes buy the AAA tranches of private-credit CLOs, and sell the whole asset class on to their own clients.

When people panic about a downturn in private credit, what they're really afraid of is those losses spilling out of the funds and into the banking system, and from there into the wider economy. It's nearly impossible to say for certain how real or how imminent that risk is. But some of it is visible, if you're willing to do the reading.

Veridion went through the disclosures of the 41 biggest banks in the world²⁴ and pulled out everything each one reveals about its private-credit ties, straight from the filings and the earnings calls. We sorted every figure into the channel it belongs to and tagged each one with exactly what it covers: genuinely private-credit-specific, or a broader "non-bank" or private-equity-inclusive bucket.

The banks have lent about $166 billion to private credit²⁴, the sum of that first group, the genuinely private-credit-specific, senior, secured lending. They sit senior, secured and over-collateralised, with the funds' own money set to take the losses first. Much of that $166 billion hasn't even been handed over; a good part of it is credit lines the banks have promised the funds can draw when they need to, rather than cash already out the door.

The genuine exposure is far smaller than $166 billion, but how protected the banks really are comes down to the one thing we can't independently check: the marks on the collateral, which is itself private-credit paper, valued by the people who hold it. And the rest of the doors into private credit, the CLO slices, the equity stakes, the in-house arms, aren't credit risk in the classic sense at all. They are different bets, or simply very profitable businesses.

This is why you have to resist the urge to add the bold numbers up. Stack every figure in that table together and you get something like $1.5 trillion, but it's apples, oranges and the orchard they grow in. BNP's $352 billion and Goldman's $188 billion are their own asset-management businesses, not exposure to anyone. Bank of America's $242 billion and UBS's $307 billion are private-equity-inclusive buckets many times broader than private credit. The only figure you can honestly sum is the first group, and it comes to $166 billion.

So while many say flatly that "banks are $300 billion-plus exposed to private credit," what's actually visible is both less catastrophic and a good deal murkier than we'd like. The unsettling part has nothing to do with an actual number, and everything to do with the fact that we can't fully tell how big the real exposure is, or how safe it truly is, because both come down to the one thing nobody can fully verify.

The problem with loud headlines is that they risk becoming a self-fulfilling prophecy. It's true that there's no transparent market price for this asset class, so you're just left with the manager's word for it, and the managers have every incentive to keep the marks looking fine, which is quite concerning. The one thing that offsets this is who the money belongs to: most of it comes from institutional investors, pensions, insurers, sovereign funds, who, in principle, understand exactly what they bought.

A second concern is that about 70% of private-credit borrowers are owned by private-equity firms, and private equity is in a distribution drought. Exits are down, holding periods have stretched, and sponsors who can't sell or refinance their companies may struggle to repay their private-credit loans on time.

Then there's the wider weather, inflation, a possible recession, and the concentration problem: too much of the book still sits in software, the exact sector an AI shock could hit all at once, though the funds do seem to be diversifying away from it. None of this has ever been stress-tested at the size and complexity the market now has. The 2028 maturity wall, when a large share of these loans comes due at once, will be the real test, and probably the moment we find out whether there's contingent leverage in here that we can't see today.

If contagion is going to form, two channels are the ones starting to take shape, and the striking thing is that neither runs through the banking system.

The first is insurance. Private-equity firms have been buying insurers, Apollo and Athene, KKR and Global Atlantic, to get hold of a captive pool of long-term money, and private credit now makes up around 10% of life-insurer assets³. The danger here is reflexive: if the marks fall, the insurer's capital drops, regulators step in, policyholders start pulling their money, the insurer is forced to sell illiquid loans, and those sales push the marks down further still. And unlike a pension, policyholders can withdraw more or less at will, no lockup, no smoothing, no government backstop.

The second is retail. The 2025 executive order and the 2026 Department of Labor rule²⁵ are opening more than $10 trillion of retirement money to private credit. Push ordinary savers' money into illiquid, gated funds at scale and you set up the same liquidity mismatch, only now it's ordinary people's money, which makes it political as well as financial.

To break the system the way 2008 did, private credit would need leverage stacked on leverage, funding that can vanish overnight, and enough of the wreckage lodged inside the banks to carry it into everyone else's life, all at the same time.

What we find instead is capital locked up for years, losses that stop with the investors who signed up for them, senior and secured claims wherever the banks are involved at all, and a book of loans that, stressed ten times past anything on record, would still cost a fraction of what the last crisis did. The regulators have also started asking more questions: the Fed is now collecting data on how much the banks actually lend to private credit, and the FSB and the OFR have published their own reports, which is a sign that this asset class is here to stay, and that it will be watched a great deal more closely from here on.

What private credit cannot do, built the way it is today, is trigger a systemic crisis.