Imagine a world where the AI revolution stalls, leaving tech giants drowning in debt. Sounds far-fetched? Think again. As companies race to invest billions in artificial intelligence, a silent alarm is ringing in the financial world: traders are scrambling for protection against a potential AI debt implosion. And this could impact everything from your retirement fund to the stability of global markets.
Tech companies, particularly the 'hyperscalers' (the giants of cloud computing and AI), are preparing to borrow massive amounts of money – hundreds of billions of dollars, to be precise – to fuel their AI ambitions. But what happens if these investments don't pay off as expected? Lenders and investors are increasingly worried about this scenario, and they're taking action to shield themselves from potential losses.
One of the key strategies they're using is trading derivatives, specifically credit default swaps (CDS). Think of CDS as insurance policies against a company defaulting on its debt. If a company fails to make its debt payments, the CDS pays out to the holder, compensating them for their losses. The demand for this kind of protection is soaring. For example, the cost of credit derivatives on Oracle Corp’s bonds has more than doubled since September, signaling increased anxiety about their debt obligations. And this is the part most people miss: it's not necessarily because Oracle is in trouble now, but because the potential for things to go wrong down the line is becoming more apparent. According to Barclays Plc credit strategist Jigar Patel, trading volume for credit default swaps tied to Oracle jumped to approximately $4.2 billion in the six weeks leading up to November 7th, a massive increase from less than $200 million during the same period last year. This stark comparison underscores the growing unease surrounding AI-related debt.
John Servidea, global co-head of investment-grade finance at JPMorgan Chase & Co., confirms this trend. He notes a "renewed interest from clients in single-name CDS discussions," a trend that had faded in recent years. While hyperscalers typically have strong credit ratings, their increasing borrowing has led to greater exposure for investors, prompting them to seek hedging strategies. It's like owning a house in an area prone to earthquakes; even if the house is well-built, you'd still consider earthquake insurance.
While the trading volume of these derivatives is still relatively small compared to the overall amount of debt expected to flood the market, it's a clear indication of the growing influence of tech companies in capital markets. Investment-grade companies could potentially sell around $1.5 trillion of bonds in the coming years, according to JPMorgan strategists. Recent major bond sales linked to AI include Meta Platforms Inc's $30 billion offering in late October (the largest corporate issue of the year in the US) and Oracle's $18 billion offering in September. These colossal figures highlight the scale of investment flowing into AI and the potential risks involved.
JPMorgan reports that tech companies, utilities, and other borrowers linked to AI now dominate the investment-grade market, surpassing banks, which were long the largest segment. This shift indicates a fundamental change in the composition of the debt market. Furthermore, junk bonds and other major debt markets are also expected to experience a surge in borrowing as companies build data centers worldwide – the physical infrastructure necessary to power the AI revolution.
So, who's buying these credit default swaps? Banks, for one, have seen their exposure to tech companies increase significantly in recent months, making them keen to protect their investments. But here's where it gets controversial... Equity investors are also jumping into the CDS market, seeing it as a relatively inexpensive way to hedge against potential stock price declines. Why? Because if a company defaults on its debt, its stock price is likely to plummet. An example: on Friday, protecting $10 million of Oracle bond principal against default for five years cost about $103,000 annually. In contrast, buying a put option (which gives you the right to sell shares at a specific price) to protect against a 20% drop in Oracle's stock price by the end of next year would cost approximately $2,196 per 100 shares, or about 9.9% of the share value. This comparison illustrates the relative affordability of CDS as a hedging tool.
And there's a valid reason for money managers and lenders to consider reducing their exposure. A report released this year by an MIT initiative revealed that a staggering 95% of organizations are seeing zero return on their generative AI projects. While some of the largest borrowers currently have strong cash flow, the technology industry is notoriously volatile. Companies that were once dominant players, such as Digital Equipment Corp, can quickly become obsolete. Therefore, bonds that appear safe today could become considerably riskier over time, or even default, if profits from data centers, for instance, fall short of current company expectations. This is a critical point: the future is uncertain, and past success is no guarantee of future performance.
Even Meta Platforms Inc., despite its vast resources, saw its credit default swaps begin actively trading for the first time late last month, following its jumbo bond sale. Derivatives tied to CoreWeave, a provider of AI computing power, have also become more actively traded. CoreWeave's shares even tumbled recently after the company lowered its annual revenue forecast due to a delay in fulfilling a customer contract. This highlights the immediate impact that setbacks in the AI sector can have on investor confidence.
Interestingly, the single-name credit derivatives market saw even higher volume in the years leading up to the financial crisis. Proprietary traders at banks, hedge funds, and bank loan book managers used these products to manage their risk exposure. However, following the collapse of Lehman Brothers, trading volume in single-name credit derivatives declined sharply. Market participants believe it's unlikely to return to pre-financial crisis levels, partly because there are now more hedging instruments available, such as corporate bond exchange-traded funds, and partly because credit markets themselves have become more liquid due to electronic trading.
Sal Naro, chief investment officer of Coherence Credit Strategies, believes the recent surge in single-name CDS trading is temporary, driven by the current data center build-out. He expresses a desire to see the CDS market truly revived, but for now, he views the activity as a short-term phenomenon.
Despite Naro's perspective, traders and strategists at banks report that activity is indeed on the rise. According to Barclays' Patel, the overall volume for credit derivatives tied to individual companies increased by approximately 6% in the six weeks leading up to November 7th, compared to the same period last year, reaching about $93 billion. Dominique Toublan, head of US credit strategy at Barclays, confirms this trend, stating, "Activity has picked up. There’s definitely more interest."
So, the million-dollar question: Is this increased demand for credit protection a prudent measure to mitigate risk, or an overreaction fueled by fear? Are we on the cusp of an AI debt crisis? Or is this simply a temporary blip in the market? What are your thoughts? Do you agree with Sal Naro that this is a temporary situation, or do you believe that the increased hedging activity is a sign of deeper underlying concerns about the sustainability of AI investments? Share your perspectives in the comments below!
