A sharp brake hit U.S. and Korean stock markets, which had been on a record-setting run. As U.S. employment data far exceeded expectations, projections spread that the Federal Reserve could pursue aggressive monetary tightening, plunging the stock market into fear.
Behind the market's hypersensitive reaction lies concern that the artificial intelligence (AI) investment cycle, which has driven the recent global stock rally, could turn. The AI investment rally so far was possible thanks to a global low-rate environment and massive liquidity. As ample funds concentrated in the AI ecosystem, expectations for industry growth rose and large sums flowed into equities; accordingly, rate increases from tightening are bound to directly hit stock markets centered on tech shares.
On the 8th, when the KOSPI index plunged more than 8%, Korea's Government Bonds yields rose to their highest level this year. The three-year Government Bonds yield rose 5.8 basis points (1 bp = 0.01 percentage point) from the previous trading day to 3.940% per year. Intraday it climbed to 3.971%, the highest since Nov. 2023. The 10-year Government Bonds yield also rose 9.4 basis points to 4.348%, the highest level this year.
Kim Myeongsil, an iM Securities researcher, said, "The current base rate is around 2.50% per year, but the three-year Treasury is already trading in the high-3% range," and added, "This reflects not just the possibility of three to four rate hikes, but the possibility that the base rate will remain higher than in the past for the next several years."
U.S. Government Bonds yields are also surging. The U.S. 10-year Government Bonds yield climbed back to 4.53% per year. As U.S. employment data released over the weekend showed strength, concerns grew that the Fed will raise rates.
Past cases show that the event that triggers the largest correction in the stock market was a tightening monetary policy. When a rate-hike cycle gets underway, the flow of funds in global asset markets shifts.
Moreover, investors are more worried because the latest global stock rally was driven not only by abundant liquidity but also by massive AI investments by hyperscalers that operate ultra-large data centers and clouds.
According to projections, this year's capital expenditure by five major hyperscalers—Amazon, Alphabet, Meta, Microsoft (MS), and Oracle—will reach $830 billion (about 1,205 trillion won). That is a 79% increase from last year's $462.7 billion, and next year AI-related capital expenditure is expected to reach $1 trillion.
The problem is that as their AI investment scale grows astronomically, funding methods are shifting toward greater external borrowing such as corporate bond issuance. Although these big tech firms have amassed vast cash, intensifying investment competition is pushing them beyond using cash on hand to also increasing borrowing.
According to the Financial Times, Wall Street expects the combined free cash flow of four hyperscalers—Amazon, MS, Alphabet, and Meta—in the third quarter this year to be $4 billion. That is less than 10% of the quarterly average free cash flow ($45 billion) since the COVID-19 pandemic six years ago. Free cash flow is a key indicator of a corporation's debt repayment capacity and dividend capability.
In particular, if hyperscalers' finances in the second half shift from net cash to net debt, their sensitivity to interest rates will inevitably increase.
Lee Seung-jae, an iM Securities researcher, noted, "Previously, big tech's corporate bond funding was limited to securing resources for share buybacks and dividends, and even that could be covered without issuance thanks to massive cash flows. But as vast cash is mobilized for expanded AI-related capital expenditure (CAPEX), they can no longer fund themselves solely with internal cash flow."
As liability-type funding has increased, hyperscalers' credit default swap (CDS) premiums have generally risen since the second half of last year. The researcher added, "The rise in CDS can be interpreted not simply as more bets on default risk but as a point where we must consider the possibility of credit rating changes due to increased debt financing."