Private Credit Under the Microscope – Separating Headlines from Fundamentals

Amid heightened scrutiny and evolving market dynamics, we aim to offer a clear-eyed assessment of the state of private credit markets today. What follows is a closer look at the headlines driving recent sentiment, an examination of redemption activity, and our forward view on the asset class.

Beyond the headlines: credit fundamentals stable, but cracks emerging at the margins

Private credit publicly traded sell off amid sentiment headwinds and idiosyncratic cracks, but valuations suggest much of the pain is priced in. Private credit publicly traded BDCs¹ are down ~16% over the past year, with wide dispersion under the surface (ranging from down ~50% to up +10%). We anticipate negative price action to persist in the near term, reflecting a mix of sentiment headwinds and idiosyncratic cracks in fundamentals.

The decline reflects a confluence of pressures: negative sentiment stemming from recent headlines, mounting concerns over redemptions and loan underwriting standards, and the specter of AI-driven disruption—particularly given private credit's outsized exposure to the software sector. Dividend cuts in a handful of managers, as yields normalize, have further weighed on performance. BDCs that have lagged the broader index tend to share common vulnerabilities: subpar risk management and equity exposure, susceptibility to software disruption, or a combination of rising non-accruals and dividend reductions.

That said, at the index level, the selloff does not appear to reflect a broad deterioration in credit fundamentals. For example, non-accruals among publicly traded BDCs that have recently reported remain modest, averaging ~2%². This suggests the pockets of weakness are more isolated than systemic.

From a valuation perspective, the price-to-NAV discount for the public BDC index (CWBDC Index) now stands at ~17%, in line with the prior low in June 2022. On the equity side, multiples for the alternative asset manager cohort³ (Price to Fee-Related Earnings) sit at ~18x. This group has historically troughed in the mid-teens, suggesting that while negative sentiment may persist, a meaningful amount of pain may already be priced in.

Redemptions – what you need to know

Across the largest non-traded, private credit funds, redemptions were ~5% of NAV on average in Q425. We expect greater clarity on first-quarter redemption requests over the coming weeks and will be monitoring developments closely. We anticipate elevated redemption activity to continue through the first half of 2026, at least.

Recent events serve as a timely reminder of how critical it is to understand the liquidity mechanisms embedded in underlying investment vehicles. In private credit, gates and redemption queues are prudent tools, applied in accordance with the terms of an investor's commitment. When portfolios are predominantly illiquid, it is appropriate—and at times necessary—for investor liquidity to align with the underlying asset pool in order to protect long-term value. It is entirely plausible that some non-traded BDCs gate this year—with the appropriate caveats, we would not characterize this as an indicator of stress, but rather a feature of these vehicles given the inherent illiquidity of their underlying investments.

Today, elevated redemption requests appear to be driven more by sentiment than by fundamentals. Several data points support this view: B-rated leveraged loan returns (JLPXB Index) are down only around 2.6% year-to-date⁴; default rates continue to track at or below historical averages across credit markets; and for non-traded BDCs, non-accruals have held at ~1.2% (of cost), below the 10-year average of 1.9%.⁵ Additionally, while yields are still at a premium to public extended credit markets, that premium has been cut in half since elevated levels post the rate hiking cycle in 2022, and we’d argue part of the rise in redemptions is related to taking profits after almost 3-years of meaningful outperformance (both vs. public market equivalents and historical norms). In a normalized rate environment, we acknowledge that risk-adjusted returns in other areas of extended credit (like preferreds, hybrids) are more competitive than they have been in the past.

With parallels being drawn to Real Estate BDCs gating in 2022/2023, it is important to note that private credit is structurally a different asset class (shorter duration asset, 20-30% of the portfolio will turn over per year on average, higher income buffer). Also, importantly with BREIT (as an example), the episode was resolved without permanent impairment of NAV, despite peak negative sentiment, and it was able to fulfill 100% of requests over a 14M period.

With this in mind, we put together an initial ‘rules of the road’ on industry standard liquidity mechanisms across public and private markets – illustrative (not exhaustive) example below.

What is our view on private credit?

• First, we do not believe that growth in private credit poses systemic risk to the economy. Recent headlines have conflated market sentiment with systemic risk. Based on our view, fears of a private credit led crisis are overstated. Why?
  • Private credit AUM grew at over ~14% CAGR over the past decade, but is still only ~9% of total corporate borrowing.⁶
  • Levels of U.S. “risky credit” outstanding has hovered around ~20% of GDP for a decade. There has not been an explosion in U.S. risky credit; private credit has taken share from other risky credit types.⁷
  • The investor base is still primarily institutional (~80% as of the end of 2024). Institutional capital is typically longer-duration and less redemption sensitive, reducing the likelihood of rapid outflows, forced asset sales, or fund gating during periods of market stress.

• Second, are there historical parallels we can draw from as it relates to bank interlinkages? How should we think about AI disruption risk?⁸
  • Bank Lending. In the mid‑2000s, real estate loans were ~53% of overall bank lending; today, loans to BDCs account for about 12.5% of total bank lending. Additionally, unlike the 2000s credit boom, bank lending has been subdued relative to the broader economy, reducing the likelihood of a bank‑led amplification loop.
  • Credit Stress.
    • To make a historical parallel when it comes to credit stress related to technological disruption, consider the micro default cycle of brick & mortar retail (which was disrupted by e-commerce) back in 2016-2019. Utilizing leveraged loan data, we find that credit episode was contained and didn’t spill over to the broader loan market. The leveraged loan default rate for retail spiked up to 6.3% in 2018, while the broader market default rate actually fell by 10bps to 1.45% over the same period.
    • Notably, the perceived driver of stress in credit markets today, software, is bigger than retail was. AI disruption of software is likely to be more consequential to the credit markets than retail was in 2016-2019, given higher relative exposure. Default rates are likely to go up for software, but we view this as more of a 3-5 year future development (vs. immediate) and contained as the disruption will not be uniform. One of the important risks we are monitoring in the software sector is refinancing risk.
  • AI Disruption Risk. Not all software or services are equally exposed, and understanding the nuances is critical.
    • AI-related disruption is creating cracks in pockets of software. Software exposure has steadily increased in the leveraged loan market over the past 15 years. Private credit has ~21% exposure to software, driven by the attractiveness of SaaS recurring revenue models and the fact that ~96% of software companies are privately held. Exposure rises to ~40% when including broader tech/business services—the highest among extended credit markets. We see software volatility as a sector led reset rather than the start of a macro default cycle.
    • For services, the risk is at the task level. Standardized, repeatable work is most vulnerable; tasks requiring trust, accountability, or physical execution are more defensible.
    • For SaaS, the question is whether AI erodes the product’s moat. Deeply embedded, data-rich, mission-critical software is safer. Generic, labour-scaled, or easily replaced tools are at risk. Seat-based revenue models are more at risk than usage-based.

• Third, at a macro level, fundamentals remain stable. Metrics are less robust than public markets, but investors have historically been compensated for the incremental risk.
  • Steady corporate fundamentals. Interest coverage ratios for direct lending have stabilized and are holding steady after past-rate shocks, hovering around ~2.0x (less than ~4.0x for public borrowers which investors have historically been compensated for). Additionally, all size cohorts showed positive year-over-year trailing 12-month EBITDA growth, except for the smallest borrowers generating less than $25M of EBITDA.⁹
  • Defaults have remained contained. High yield defaults are around 2% vs. ~3.3% historical; leveraged loans are around 2.8% vs. ~3.1% historical; and private credit is ~2.5%, broadly in line with historical norms.¹⁰
    • Micro-level stress is concentrated. Default dispersion persists across two key dimensions:
      • Borrower size: Defaults thus far are concentrated among companies with $25–$50M EBITDA.
      • Industry exposure: Automotive leveraged loan default rate (last 12M) is 10.6%—5x the historical average. Retail leveraged loan default rate is 3.7% (in line with the historical average, elevated vs. market). While tech/software defaults remain relatively muted now, it is likely to tick up over the medium term.

• The bottom line:
  • Private Market Portfolio Perspective. We believe senior secured direct lending still can serve as a critical component in a core private market portfolio as it serves an alternative source of potential income in a client portfolio. In terms of sizing, we think private credit can be ~15% of a core private market allocation (anywhere from ~1.5-4.5% of overall portfolios), balancing direct lending with asset-backed finance and opportunistic/distressed to enhance durability and diversify risk. We think it is an opportune time to assess position sizing, whether exposure has become outsized given direct lending’s outperformance over the past three years, and that risk taken is intentional.
  • Where do direct lending yields go from here? In the U.S., we expect yields to normalize to historical ranges in our base case (high single digits). We continue to monitor defaults/credit losses closely, but note that current yields are healthy and provide some buffer to this risk. Selectivity is key as we expect performance dispersion to widen in 2026 (a break from historical precedent)—focus on portfolios with diversified sector exposure, seniority in the capital structure, and lean into larger companies (EBITDA > $50M).
  • ‘Credit complements’ will be critical in 2026 as direct lending yields normalize and cracks in credit emerge. Consider diversifying through asset-backed finance which can offer key portfolio benefits including low correlation to corporate credit and direct lending, diversification through varied collateral pools, self-amortizing structures versus bullet maturities, multiple layers of structural credit enhancement, and access to a growing opportunity set created by structural bank retrenchment—though the asset class's complexity makes manager selection and expertise critical. Find opportunity in "micro" credit cycles through opportunistic/distressed credit managers that can lean into dislocations.