Two weeks ago, I wrote the following: During the week of September 24–October 1, credit risk premiums on corporate bonds rose. In the high-yield segment, the premium climbed from 280 to 324 basis points, while for the lowest-rated tier—the CCC category—it jumped from 1,112 to 1,215 basis points. The point I highlighted in that piece was that this rise was spreading to companies with stronger credit ratings. CoreWeave was not mentioned that week—nor did it need to be, because stress manifests in a specific market segment before it becomes apparent in the names of individual companies. Here is the sequence of what followed:
On October 13, the yield on CoreWeave’s 2032 bond touched 15%; during the summer, it had been around 9%. A rise in bond yields means that lenders now demand significantly higher interest to provide the same amount of capital to the same company. On October 14, the stock fell by 15%. Nebius, IREN, and Oracle followed suit. That same day, the company announced it would raise capital through a combination of new equity and new debt—meaning it was forced to borrow precisely when its financing costs were at their peak. For me, the core issue here isn’t CoreWeave itself; it’s the sequence of events.
First, the weakest credit ratings—the CCC tier—are affected. Then, the high-yield segment as a whole. Next, the bonds of individual companies. Finally, the stock market. The equity market is the last to grasp this chain of events, because stock prices focus on the "growth story," whereas bond prices focus on whether the debt will actually be repaid.
There is one more detail: throughout this process, the investment-grade bond index barely moved; on October 8, the premium stood at 82 basis points. Someone looking only at that index might have concluded that the credit market was calm. Yet, during that same period, the bond yield for a single company climbed from 9% to 15%. An index is an average; averages mask the extremes.
Therefore, anticipating stress is not a matter of reading charts, but of disaggregation—breaking things down. Move from the index to credit rating tiers, then to sectors, then to individual companies, and finally to the specific maturity and collateral structure. Each layer requires different data, different sources, and different questions. It is a laborious process that often seems to yield no results—until, suddenly, it does.
Anyone can see what is happening after the fact. Identifying it beforehand, however, always keeps you one step ahead. This is not a matter of prophetic foresight, but simply the discipline of looking in the right places, in the right order.
Note: This piece does not suggest that CoreWeave is headed for bankruptcy; the company has a signed order book worth $104 billion. The point is that, within just two weeks, lenders have significantly increased the premium they demand to extend credit to this firm—a trend that, in the upper echelons of the credit market, actually began two weeks prior. This does not constitute investment advice.
