Are Regional Banks Sitting on A Private Credit Time Bomb?

If you’ve been skimming headlines lately, it’s easy to assume that credit risk is suddenly everywhere. An auto-parts borrower collapses amid allegations of double-pledged receivables. A subprime auto lender goes under and dents a few banks’ earnings. Several regional banks disclose fraud-linked commercial loans. Regional bank stocks wobble.

It feels like the start of something bigger.

Look under the hood, though, and the picture is mixed. Banks’ exposure to non-deposit financial institutions (NFDIs) has grown quickly, reflective of structural changes to loan origination and the capital markets. Yet this shift has not translated into materially higher default or loss rates. The real questions lie in the details:

• Who are banks lending to?
• How are the loans structured?
• How reliable is the collateral behind them?

In short, today’s credit concerns are real, but we don’t believe they point to systemic stress yet. What we’re seeing looks more like a slow, mild credit cycle than a sudden crash. We may expect more headline noise, but not a replay of 2008.

Headlines vs Data: The Clash Shaping Today’s Private Credit Narrative

The current credit risk debate is being pulled in two different directions.

On one hand, financial conditions are relatively accommodative. Primary credit markets are open, and higher-quality issuers are refinancing without much drama. AI-related capital expenditure is keeping demand for financing alive. Meanwhile, the banking sector is still producing record earnings. Net interest margins have narrowed, but fee income, trading, and wealth management are offsetting the drag.

On the other hand, the credit risk headlines are unsettling. First Brands Group, a large auto-parts company, defaulted with more than $10 billion of liabilities after investors questioned whether certain receivables had been pledged and whether aggressive invoice factoring had inflated reported earnings.[1] Tricolor, a subprime auto lender, entered liquidation, creating losses for several creditors, including Fifth Third Bank.[2] Zions and a few other regional banks also disclosed commercial loans where collateral or borrower information appears to have been misrepresented.[3]

To make sense of this, it helps to separate the broad default cycle from idiosyncratic failures like First Brands. Across credit markets, most defaults today stem from heavy capital structures—especially the 2021 LBO vintage—now facing higher funding costs.[4] The strain is concentrated in non-cyclical sectors where the underlying businesses are still viable, but capital structures designed for a near-zero-rate environment world now carry too much corporate debt for today’s coupons. As refinancing waves hit and floating-rate liabilities reset higher, interest burdens are outpacing cash generation.

The story behind First Brands is different. The core issue was double-pledged collateral and aggressive factoring, which undermined the quality of reported earnings and complicated refinancing efforts. This reflects a company-specific credit failure, not a broad deterioration in underwriting standards or operating performance.

At the bank level, the headline losses are uncomfortable but modest relative to balance-sheet capacity. At Zions, the charge tied to a problematic commercial and industrial (C&I) loan relationship was roughly $60 million—about 5% of annual earnings and around 0.10% of total loans. At Fifth Third, the loss linked to Tricolor was about $178 million, roughly 6% of annual earnings and again close to 0.1% of loans.

System-wide bank credit quality metrics tell a similar story. Across the 25 largest U.S. banks, C&I non-performing loan ratios hover around 0.53%, while C&I net charge-offs sit near 0.33%. Both remain far below levels seen in prior recessions.

Figure 1: C&I NPL Ratios Remain Low But Are Ticking Up

Median C&I NPL ratio among top 25 banks

Source: Regulatory reports and JPM research

Figure 2: C&I NCO Ratios Remain Very Low Versus Recession Levels

Median C&I NPL ratio among top 25 banks

Source: Regulatory reports and JPM research

The gap between scary, idiosyncratic credit risk headlines on one side and surprisingly steady system-level banking data on the other is the backdrop for the rest of the discussion.

How Big Is Bank–NDFI Exposure? A Clear, Data-Backed View of Private Credit Risk

On the surface, the raw numbers on bank lending to non-depository financial institutions (NDFIs) look alarming. Fed H.8 data show about $1.7 trillion in loans to non-bank financials as of November 7, 2025, including nearly $300 billion in loans outstanding to private credit participants. [5] On a reported basis, those balances are up 47% year over year (around $352 billion) and have grown at an 18% compound annual rate since 2018. Even after adjusting for reclassifications, growth still runs at roughly 15% annually, underscoring the rapid growth across the credit markets.

Figure 3: Loans to NDFIs, Large Banks (YoY Change)

Source: Federal Reserve

Loans to NDFIs have become a key driver of loan growth in the U.S. banking sector, outpacing other categories since 2016. Over that same period, NDFI lending has more than doubled its share of commercial and industrial (C&I)—related loans at large banks—from about 14% in 2018 to roughly 34% in 2025. This expansion highlights how NDFIs have become a central force in modern bank credit trends.

Figure 4: Loans to NDFIs as Share of C&I-related Loans, Large Banks

Source: Federal Reserve

So why has bank lending to NDFIs risen so sharply?

After the financial crisis—and especially following the 2013 interagency leveraged-lending guidance—banks pulled back from riskier direct leveraged loans but continued to lend to the intermediaries that finance those borrowers. This pivot allowed banks to diversify revenue streams by funding business-credit platforms and providing subscription lines to private-equity funds rather than holding every middle-market loan on their own balance sheets.

Favorable market conditions in recent years also accelerated this growth: floating-rate assets repriced higher, term funding remained accessible with low-term premia on a flat or inverted curve, and liquidity was plentiful. But rapid growth often signals higher embedded credit risk. As the curve steepens and liquidity tightens, those tailwinds are fading, making asset-liability management, stronger disclosure, and collateral quality more critical. It’s not hard to see why investors are becoming more cautious.

But the details matter.

Although NDFI loans are trending higher, the rise does not reflect an across-the-board surge in bank lending. A large share of the “growth” is due to regulatory reclassification. The FDIC’s expanded NDFI definition, introduced in late 2024, prompted banks to shift loans out of traditional commercial & industrial (C&I) lending into NDFI categories to comply. As this transition stabilizes, reported NDFI growth is expected to moderate.

Part of the current “nonbank” activity has long existed inside banking groups. Over the past several decades, bank holding companies (BHCs) have added broker-dealers, specialty lenders, insurers, and investment funds, increasing nonbank subsidiaries from roughly 10% of consolidated assets in the mid-1990s to over 30% pre-GFC, and stabilizing around 20% in recent years[6]. This broad-based expansion across large BHCs means that BHC-affiliated NDFIs represent a meaningful share of the overall NDFI sector. With this context, it’s easier to see why the data series “jumps” without implying brand-new credit risk.

Reporting differences further complicate interpretation. Bank call reports and Y-9C filings often disagree on NDFI balances because Y-9C data includes loans booked in non-bank subsidiaries, while call reports capture only the bank entity. Meanwhile, the Fed’s recent update to the H.8 NDFI definition boosted reported balances, inflating growth rates.

Yes, the banking system does have more NDFI exposure than it did seven years ago—but the recent spike in reported balances overstates how much of that increase reflects genuinely new credit risk.

What’s in the NDFI Book? Breaking Down the Real Private Credit Exposure

“NDFI” can sound like a single, homogeneous credit risk category, but in practice, it is a catch-all label for a wide range of non-bank borrowers. For regional banks, the piece that matters most in this discussion is the private-credit slice, which sits mainly in two categories:

• Business credit intermediaries: explicitly identified in the data as private credit, including loans to direct-lending platforms and specialty finance companies.
• Private-equity funds: mostly subscription lines secured by LP commitments, which are senior, collateralized, and have historically exhibited very low loss rates.

Across these categories, combined private-credit exposure generally represents the low- to mid-teens as a percentage of total loans at the most active lenders—making up less than half of the overall balance sheet.

Regional Banks: Stronger Fundamentals Behind the Private-Credit Noise

While GSIBs also hold sizeable private credit exposure, investors are less concerned because these exposures are smaller relative to their larger, more diversified balance sheets, skew toward secured fund-finance, and benefit from stronger capital, liquidity, and regulatory oversight. Meanwhile, recent blow-ups have pulled regional banks back into the spotlight.

Tricolor, First Brands, and certain Cantor-linked real-estate loans all involved one or more regional banks as lenders. Fifth Third and Zions have faced the most heat, but both used those episodes as catalysts to dig deeper: Fifth Third re-examined its broader consumer finance exposure, while Zions reviewed its auto book and engaged a third party to verify nearly all VINs. So far, neither bank has reported evidence that similar issues are emerging across their franchises.

System-level data reinforces this perspective. Among super-regional and regional banks, median changes in net charge-offs in 3Q25 were largely flat, with notable changes clustered in a small group of banks that had already flagged specific problem credits.

Figure 5: Credit Continues to Hold Up. Regional Bank Net Charge Offs Remain Stable.[7]

Source: Company filings

In addition, regional banks are absorbing these idiosyncratic hits from a stronger position than they were in early 2023. Since the liquidity shock, they have rebuilt capital, lengthened and diversified funding sources, and increased liquidity buffers. The rate cycle is easing, which should gradually lower funding costs, but consumer credit has softened compared with twelve months ago, and net interest margins remain under pressure.

Overall, fundamentals are sturdier than they were at the height of the 2023 stress. For the active lenders, loans to business credit intermediaries and private-equity funds tend to sit in the low- to mid-teens of total loans, with a large portion of that exposure in senior, collateralized structures that have historically demonstrated strong performance. The high-profile problem credits seen so far—about 0.1% of loans at the affected banks—are too small to destabilize these institutions on their own.

In other words, these mostly one-off credit issues may generate more headlines and volatility as boards push for deeper reviews. But when you weigh the size and structure of these exposures against still-low loss ratios and stronger balance sheets, what appears to emerge is a noisy, idiosyncratic credit cycle, not a private-credit bomb buried inside regional banks.

Beyond the Headlines: Key NDFI Signals for Investors

Earnings commentary this season has been unusually consistent. JP Morgan describes the “vast majority” of its non-bank financial institution (NDFI) lending as highly secured, structured, or securitized, emphasizing that it is not in the business of bulk, low-rated, high-beta NDFI loans. In their view, the real risk lies in collateral issues at a handful of borrowers, not in a sea of weak underwriting.

Wells Fargo makes a similar point, noting that most of its NDFI book consists of plain-vanilla capital-call facilities to large, established private-equity firms, deliberately focused on big sponsors to reduce tail risk. Regional banks echo this sentiment from different angles: PNC reports that all recent C&I growth came outside NDFI, U.S. Bancorp now includes an “NDFI Transparency” slide and highlights that NDFI credit quality is better than its core C&I portfolio, with NDFI representing 12% of loans, and KeyCorp stressing avoiding the “esoteric” corners of NDFI, focusing on exposures such as REITs and CLOs that fit within its risk appetite.

While this commentary provides useful context, it is not proof. Management teams may have incentive to frame these books in the best light, so investors should verify claims against hard evidence:

• Depth of NDFI disclosure by subsector
• Collateral and advance-rate details
• Top-counterparty concentrations
• Borrowing-base and audit practices
• Trends in nonaccruals, charge-offs, and reserves at banks with large NDFI books

Where these checks align with stable credit metrics and a clear tilt toward senior, sponsor-backed fund finance, confidence is warranted; where disclosure is thin, “other” buckets are large, or early stress appears in migration data, the benefit of the doubt should be much smaller.

Over the next few quarters, the most useful questions for investors regarding regional banks and NDFI lending are less about total exposure and more about loan composition and disclosure quality:

• How clearly do banks separate private-credit exposure from the rest of NDFI, and how much detail is provided on collateral, maturities, and counterparties?
• Are C&I and NDFI non-performers and charge-offs drifting gradually higher, or starting to accelerate at banks with heavier NDFI concentrations?
• Are regional bank spreads widening in line with fundamentals, or overshooting market expectations?

The takeaway for investors is straightforward: this is not a stress-free credit cycle, but it is not a broken one either. The goal is to scrutinize the composition and disclosure of regional bank NDFI books, rather than fleeing bank credit exposure altogether.

Key Takeaways

• Private credit exposure at U.S. banks is real, but not uniformly risky. Loans to NDFIs have grown quickly, but much of that growth stems from definitional changes and strategic shifts since 2013, rather than a wholesale increase in bank risk appetite.
• Recent defaults are high-profile but largely idiosyncratic. Losses at Fifth Third and Zions, tied to Tricolor and First Brands, were isolated events, rather than evidence of broader deterioration in credit quality or underwriting standards across regional banks.
• The structure and concentration of exposures matter more than headline size. Most regional bank NDFI lending is directed to business credit platforms or private equity funds, often in secured, senior, and collateralized structures. Problem credits to date are small relative to capital and earnings.
• Regional banks remain fundamentally healthier than in early 2023, but face a mixed outlook. While capital and liquidity positions are stronger, rising funding costs and consumer credit are notable headwinds. Overall, this appears to be a more selective, slower-moving credit cycle, not systemic stress.
• Management commentary emphasizes sound underwriting and robust collateral, but investor scrutiny is still warranted. Transparency on subsectors, borrower types, and collateral structures varies widely, so claims should be tested against disclosure and performance data, not narrative alone.

---

[1] https://www.reuters.com/markets/us/auto-parts-maker-first-brands-files-bankruptcy-protection-2025-09-29/

[2] https://www.bloomberg.com/news/articles/2025-10-17/fifth-third-house-to-house-search-finds-just-two-bad-car-vins

[3] https://www.reuters.com/business/finance/investor-behind-zions-western-alliance-bad-loans-is-tied-270-million-troubled-2025-10-20/

[4] https://www.franklintempleton.lu/articles/2025/alternatives/public-insights-on-private-credit-the-art-of-vintage-selection

[5] https://fred.stlouisfed.org/series/LNFACBW027SBOG

[6] https://libertystreeteconomics.newyorkfed.org/2025/11/u-s-banks-have-developed-a-significant-nonbank-footprint/

[7] Note: FITB and ZION losses were driven by alleged fraud from Tricolor and Cantor group exposure.