US midterm elections could widen AI data center gap

The US midterm elections could widen the AI ​​divide: Tighter data center rules could reward tech giants with spare capacity and constrain debt-financed companies trying to catch up.

Democratic victories in key gubernatorial elections have raised the prospect of stricter scrutiny of data center projects, heightening investors’ concerns about how quickly new capacity can be built and financed.

In Texas, Democratic Gov.-elect Gina Hinojosa proposed a statewide moratorium on new data center permits, although projects already under construction would be exempt.

As markets digested the results on Wednesday, Microsoft And Amazon each jumped more than 6%, while AI infrastructure provider CoreWeave plunged about 10%. Chipmakers Micron and Nvidia fell 7% and 3.17%, respectively.

Differences in borrowing costs are playing a key role in the split, James Sullivan, managing director and co-head of global fundamental research at JPMorgan, told CNBC’s “Squawk Box Asia” on Thursday.

Sullivan said borrowing spreads have widened significantly for neocloud companies, leaving their incremental cost of capital significantly higher than hyperscalers who benefit from more diverse customers and revenue streams.

“The fact that you’re seeing this bifurcation in stock performance is largely driven by the idea that existing capacity is now much more valuable because of increasing supply constraints,” he said.

The midterm results could reinforce this view if they make new projects more vulnerable to permitting delays.

BMI warned that a sharp slowdown in data center construction and AI investment could lead to a larger-than-expected decline in overall investment, the Business Times reported, citing the company’s post-midterm analysis.

The potential impact is uneven across the construction pipeline. Much of the data center capacity expected to come online in 2027 is already under construction, limiting the short-term impact of the new restrictions.

Projects planned for 2028 are more exposed to permitting delays and regulatory uncertainty, investment bank Nomura said in a mid-term results note on Wednesday.

Stricter regulations on data center construction could become a bigger drag on AI investments starting in 2028, the company said.

Shares of Nvidia-backed neocloud companies CoreWeave, Nebius and Iren fell at least 30% in October as tightening credit markets raised questions about how AI infrastructure projects would be financed. The spread for CoreWeave’s five-year credit default swaps rose above 1,100 basis points, reflecting rising costs for investors seeking to protect themselves against a potential default.

The distinction may lie in the ability of companies to absorb delays and finance the capabilities they develop. Hyperscalers have broader cloud businesses and more internally generated cash flow, while lower-rated neoclouds are more exposed to borrowing costs, energy prices and refinancing terms, JPMorgan Asset Management said in an Oct. 20 research note.

The asset manager said financially stronger hyperscalers could also obtain cheaper capacity or acquire assets if weaker operators had to scale back operations.