Charlie McElligott, Nomura’s Cross-Asset Macro Strategist, is flagging a risk that most market participants are probably not thinking about enough: more than $300 billion in autocallable structures, primarily linked to mega-cap tech stocks, sitting in the derivatives market like a loaded mechanism waiting for the right trigger.
That concern arrives alongside a debt binge in the AI and datacenter space that has no modern precedent. Year-to-date issuance from AI companies, hyperscalers, and datacenter operators has reached $269 billion, roughly 12 times the annual average from 2015 to 2024.
The debt machine running at full speed
Broader corporate debt issuance has climbed 61% year-over-year. But the AI and datacenter cohort is driving the bus, absorbing capital markets bandwidth at a pace that dwarfs everything else in the fixed-income universe.
Morgan Stanley had projected $250 billion to $300 billion in hyperscaler issuance for 2026. At $269 billion through mid-August, that range is essentially already met with more than four months left in the year.
