Bad debts in the US software industry are surging, and a "software-PE" death spiral is unfolding.

The U.S. software industry is facing a credit crisis triggered by artificial intelligence.

On Thursday, according to data compiled by Bloomberg, over the past four weeks, more than $17.7 billion in tech company loans have fallen to distressed levels, pushing the total scale of bad debt in the tech sector to around $46.9 billion, the highest level since October 2022. This crisis, dubbed "SaaS apocalypse" by the market, is quickly spreading from the stock market to the private credit sector.

The core driving force behind this sell-off is market concern that AI is disrupting the traditional software business model. The Software-as-a-Service (SaaS) sector is seen as particularly vulnerable, as AI is replacing traditional software functions such as coding and data analysis.

The impact of this crisis is spreading, with distressed-level debt including loans to medical software company FinThrive and Perforce Software, both backed by private equity firm Clearlake Capital.

Bank of America analysts point out that about 14% of leveraged loan market assets are exposed to the tech sector, while in private credit that proportion is as high as 20%. For collateralized loan obligations (CLOs) that package leveraged loans into bonds, the software sector represents 11% to 16%. The current private credit market is experiencing two shocks: the breakdown of the logic for software company lending, and waning appeal for private credit.

Software industry debt deteriorating rapidly

Bloomberg Intelligence data shows that in the past four weeks, $17.7 billion of tech company loans have fallen to distressed levels, mainly concentrated in the SaaS sector. Distressed-level loans refer to debt with yields more than 10 percentage points above the benchmark Secured Overnight Financing Rate (SOFR).

Beyond the loans already at distressed levels, more software company loans are nearing stress levels. Leveraged loans for HR management software provider Dayforce and call center technology company Calabrio, both held by private equity firm Thoma Bravo, are approaching the brink of distress. Loans for Precisely, a data integrity software company jointly held by Clearlake and TA Associates, dropped 8 cents this week.

Jack Parker, portfolio manager at Brandywine Global Investment Management, describes the current situation as a moment of "sell first, ask questions later." He says: "This is certainly painful for the sector—people are broadly selling all assets in the space with little consideration of how much AI will disrupt these businesses, or how long it will take."

Bank of America analysts found that since January 9, 80% to 90% of issuers in the enterprise software and technology services sector have shown negative price returns. The bank called January 9 the 'turning point' for the sector. Analysts wrote in a report: "As AI advances rapidly, there is growing concern that AI could threaten software and service providers—if not eliminate them entirely, then at least put pressure on their revenue streams."

Private credit faces dual predicament

The software sector’s plight is sending shockwaves into the private credit market. The share prices of alternative lenders such as Blue Owl, Runway Growth Finance, and Golub Capital have started to plunge in tandem with the software industry.

Bank of America analyst Ebrahim Poonawala’s team data shows that using the tech category as a broad proxy, software appears to be one of the largest industry exposures for business development companies (BDCs). Raymond James analyst Robert Dodd points out that actual exposures could be higher because loans to software firms are often classified by end market. For example, a company providing SaaS to healthcare may be categorized under "healthcare" rather than "tech" or "software."

Analysis suggests the private credit market is experiencing two simultaneous "unwinding" processes.

First, the logic for lending to software companies has collapsed. Annual recurring revenue (ARR) was once seen as providing stable, bond-like cash flow. This predictable payment stream supported the argument that lending was reasonable even when free cash flow was negative. But this logic relied on the belief that subscription income would remain stable. When the business model suddenly faces risks of obsolescence, the "stable annuity" turns into a binary bet.

Second, private credit’s own appeal is fading. As yields on public markets catch up, the promised "liquidity premium" is less compelling. The so-called segregation effect—no daily mark-to-market, limited volatility, calm amid the storm—is harder to sell now that default is seen as a real risk, and every headline in the market seems related to sectors where exposures are heavy.

A death spiral is forming

Jeffrey Favuzza of Jefferies' equity trading division describes the current situation as "SaaS apocalypse", and notes that trading style is now "get me out at any cost" panic selling, with no signs yet of stabilization.

Analyses from J.P. Morgan and Goldman Sachs show the market is experiencing unprecedented divergence: on one side are semiconductor companies seen as beneficiaries of the AI supercycle, on the other are software firms viewed as the biggest losers. This divergence is creating a dangerous negative feedback loop.

As software equity valuations plunge, private credit institutions face balance sheet revaluation pressure, which could lead to tighter credit conditions. In turn, this will further squeeze already struggling software companies’ room for growth. Companies whose loans continue to trade at distressed levels will find it difficult to access traditional debt markets, worsening their financing difficulties.

Risk Warning and DisclaimerMarkets have risks, investments should be made cautiously. This article does not constitute personal investment advice, nor does it consider the special investment objectives, financial status or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article are suitable for their particular circumstances. Investment decisions based on this content are at your own risk.