Moody's baseline forecast for U.S. speculative-grade defaults by year-end 2026 is 3.8%, down from recent elevated levels. S&P expects 4.0% by September. Both agencies arrived at these numbers independently, using different methodologies, and both point the same direction: down. That convergence is the single most important signal in the corporate credit market right now, and it tells me the default environment stays contained over the next 12 months.
I understand the anxiety. Bruce Richards at Marathon Asset Management is projecting a 15% default rate for private credit software loans, persisting for 3 or more years. Fitch reported the U.S. private credit default rate hit 5.8% in January 2026, with 11 default events in that month alone, nearly double the 2025 monthly average. Those are ugly numbers. They deserve attention.
They do not deserve extrapolation.
Software Loans Are Not the Economy
The private credit software lending pocket is a specific corner of a $1 trillion-plus market, characterized by loans at 8 to 10 times annual earnings and recovery rates that Marathon estimates at $0 to $0.30 on the dollar. That is genuinely distressed pricing. But conflating this segment with the broader corporate default picture is the analytical equivalent of watching one house burn and calling it a citywide fire.
Consider the wider frame. Neuberger Berman projects emerging market corporate defaults at 2.7% for 2026, below the long-term average of 3.4%. Moody's identifies retail, housing, autos, and consumer goods as the most vulnerable U.S. sectors, but vulnerability is not the same as systemic failure. These are sectors where earnings pressure is visible in quarterly filings, where management teams are already cutting costs and restructuring balance sheets. That is how credit cycles work when they stay orderly.
Richards is right that refinancing conditions for software direct loans have deteriorated badly, with new loan spreads widening 700 basis points. For a mid-market software company carrying $100 million in debt, that spread widening translates to roughly $7 million more in annual interest expense. Some of those borrowers will not survive. But the contagion path from private credit software loans to, say, investment-grade industrials or even the broader leveraged loan market requires several transmission mechanisms that are not yet firing.
The Range Is Wide, but the Center Holds
Moody's own risk range extends from 1.7% to 8.3% through October 2026. That 8.3% tail is the number that generates headlines. Fair enough. But the base case sits at 3.8%, and the agency's own assessment cites stable earnings and gradual deleveraging as supporting factors. When I read an earnings report, I weight the audited results over the scenario analysis. The same logic applies here.
I will grant the bears one point: the leveraged loan market's concentration in B-rated and lower credits means a modest macro shock could accelerate defaults faster than the baseline assumes. Charles Schwab has flagged this, and the concern has merit. But "could" is doing a lot of work in that sentence. The Fed's 2026 stress test scenarios model 10% peak unemployment and a 39% drop in commercial real estate prices. Those are designed to break things. The actual economy is not there.
For a pension fund or insurance company allocating capital today, the defensible institutional view is containment with sector-specific monitoring. Retail and autos will produce defaults. Private credit software loans will produce losses that embarrass some fund managers. Neither of those outcomes constitutes a wave.
The $20 billion in private credit redemption requests is a liquidity event for specific funds, not a solvency event for the corporate sector. Investors who treat every pocket of stress as a harbinger of systemic collapse end up underweight credit at exactly the wrong time. The base rates are declining. The earnings are stable. The 3.8% is the number.
