Four banks. $128 billion in private credit exposure. Executives call it "comfortable."
I call it a time bomb with a short fuse.
Here's the raw data from S&P Global and Reuters: 53 Business Development Companies (BDCs) analyzed. 23 posted net losses in Q1 2026. That's 43% of the sample bleeding red ink. Net income across the group dropped 36% year-over-year.
Meanwhile, payment-in-kind (PIK) loans — the crypto equivalent of rug pull tokens that pay in more promises — doubled as a share of portfolios. Off-balance-sheet leverage surged.
Audit passed. Trust failed.
Context: The Private Credit Machine
Private credit sits in the shadows between retail and institutional finance. BDCs lend to mid-sized companies that can't access bank loans. Banks then lend to BDCs through various vehicles: NAV loans, warehouse facilities, even derivatives. The chain is long.
The borrower needs cash. The BDC provides it at high interest. The bank provides leverage to the BDC. Everyone takes a fee. Nobody checks the underlying math.
In crypto, we call this liquidity mining. You deposit token A, get token B as yield. The yield is 50% APY. But token B is printed by the protocol. Real value? Zero.
Private credit is the same. PIK loans pay interest in more debt, not cash. The borrower's health doesn't improve. The BDC's income looks fine on paper. Until it doesn't.
I audited the Ethereum 2.0 beacon chain specs in 2017. Found a slashing condition bug others missed. That experience taught me one thing: complexity hides risk. The private credit market is complex. The risk is hidden.
Core: The Data Doesn't Lie
Let's break down the numbers.
1. BDC Losses - Q1 2026: 23 of 53 BDCs unprofitable. - Net income down 36% from same quarter last year. - Losses primarily from loan impairments and higher funding costs.
In DeFi, when protocols start posting negative yields, the TVL drops. Here, BDCs can't print new loans fast enough to mask the red ink.
2. PIK Loan Explosion - PIK loans as a share of BDC portfolios doubled from 2024 to 2026. - These are loans where interest is paid in-kind (more debt) instead of cash. - Essentially, the borrower is kicking the can down the road.
In crypto terms, it's like a stablecoin that pays yield in its own governance token. The yield looks real until the token price collapses.
3. Off-Balance-Sheet Leverage - BDCs increasingly use off-balance-sheet vehicles to raise capital. - These structures hide leverage from regulators and investors. - Financial Stability Board (FSB) warned about this in 2023. The warning is still relevant.
Remember the 2008 crisis? Off-balance-sheet vehicles (SIVs, CDOs) were the silent killers. Private credit is building the same infrastructure.
4. Bank Exposure - JPMorgan, Citigroup, Bank of America, and Wells Fargo collectively hold $128 billion in private credit exposure. - This includes direct loans, NAV loans to BDCs, warehouse facilities, and derivatives. - Bank executives say they are "comfortable."
Comfortable? I've heard that before. FTX executives were "comfortable" two weeks before bankruptcy.
The Transmission Chain
Borrower struggles → BDC takes impairment → BDC net income drops → BDC borrows more from banks → Bank exposure rises → Regulator discovers hidden leverage → Crisis.
It's a chain of nine blocks. Each block is a fragile smart contract. One oracle failure and the whole structure collapses.
Contrarian: The Narrative vs. The Truth
The mainstream narrative: Private credit is safer than public debt because it's illiquid and managed by professionals. Illiquidity reduces volatility. Professionals manage risk.
Bullshit.
Illiquidity doesn't reduce risk. It delays recognition. In crypto, we learned this the hard way with Three Arrows Capital. Their assets were illiquid. Their leverage was hidden. The collapse was sudden.
The same pattern exists here. BDC net income is down 36%. PIK loans double. Off-balance-sheet leverage grows. Yet bank executives say "comfortable."
Why? Because they don't want to mark down these assets. Marking them down would trigger margin calls, redemption waves, and a liquidity crisis. So they smile and say everything is fine.
I call this "audit theater." The numbers pass an auditor's review but fail the reality test.
Another blind spot: The circularity of exposure. Banks lend to BDCs. BDCs lend to companies. Some of those companies are suppliers to the banks. Some are hedge funds that trade bank stocks. The web is tangled. When one node fails, the whole network feels it.
In crypto, we called this "contagion." It killed Terra, then Three Arrows, then BlockFi, then FTX. The same physics apply to TradFi.
Takeaway: The Next Financial Crisis
Watch the banks. Watch the BDCs. Watch the PIK ratio.
If private credit implodes, it won't happen overnight. It will start with a single BDC defaulting on a bank loan. The bank will write down $100 million. The market will panic. Other BDCs will face margin calls. The bank will need to raise capital. The stock will drop. The Fed will step in.
Sound familiar?
In 2008, it was subprime mortgages. In 2022, it was crypto leverage. In 2026, it's private credit. The names change. The pattern doesn't.
BDC stable. Fragility remains.
Based on my audit experience across both DeFi and TradFi, I see the same signatures: high leverage, opaque structures, and a belief that "this time is different."
It isn't.
Yield fiction. Reality fact.
Private credit's yields look attractive in a bull market. When rates stay high, the fiction breaks. The yield becomes a liability.
What to Watch
- Next BDC earnings season: Track net income and PIK share. If the trend continues, sell bank stocks.
- Fed Financial Stability Report: Look for specific private credit warnings.
- Off-balance-sheet disclosure: If regulators force transparency, leverage will spike and trigger a selloff.
Final Thought
The $128 billion on bank balance sheets is just the visible tip. Below the surface lies a mountain of hidden leverage, impaired loans, and circular exposure. The executives are comfortable. The data is not.
In crypto, we learned to trust the code, not the narrative. In private credit, trust the data, not the executive. The data is screaming. Are you listening?
