US midterm elections are one week away, and for markets the question is not who wins but whether the result leaves the White House and Congress divided — with the president’s party losing control of at least one chamber. Divided government is nothing new in Washington: spending fights and intermittent federal shutdowns have played out repeatedly. This year, though, the risk carries a new market meaning, because it collides directly with the AI infrastructure buildout the US is betting heavily on.
That matters because current US equity valuations rest largely on expectations of expanding AI compute capacity. If political gridlock slows the AI buildout, the damage would go beyond policy — it would hit the entire stock-pricing logic built around AI.
The parties’ AI infrastructure split is real
The two parties’ positions on AI infrastructure differ sharply, and Republicans are split internally as well. One camp wants to build fast, pushing data center construction ahead at maximum speed and deferring cost and public backlash. The other wants legislation requiring AI infrastructure projects to resolve water and power supply for nearby residents before construction can begin. If that second approach becomes law, it would significantly slow data center construction and raise US compute costs.
In other words, the election result will directly determine which policy path prevails. The two paths mean very different things for AI-linked assets: rapid expansion favors the compute supply chain, while strict preconditions mean longer timelines, higher costs and lower earnings expectations.
Two transmission channels under divided government
If divided government materializes, uncertainty would reach markets mainly through two channels. The first is a fiscal and budget standoff. If Congress and the White House refuse to budge on spending bills, shutdown risk rises, rattling market sentiment in the short term and affecting the pace of federal contracts and projects. The second is AI infrastructure policy uncertainty. Mandated water and power requirements, longer environmental reviews and slower permitting would make data center timelines unpredictable, in turn affecting how investors assess returns on AI companies’ capital spending.
Together, these two channels mean parts of the market previously priced for “continuous AI infrastructure acceleration” may need recalibration.
BofA strategist’s scenario forecasts
Bank of America chief equity strategist Michael Hartnett offered rough directional calls on two extreme scenarios in a note to clients: if Democrats win both chambers, US stocks could fall more than 10%; if the president’s party unexpectedly holds both, stocks could rise more than 10%. The forecast is not a precise model but a directional read on political outcomes and market risk appetite — yet it shows Wall Street already treats the midterm result as a major short-term variable for US stocks.
Note that markets do not read divided government as purely negative. Historically, a split Congress has sometimes been seen as a check on aggressive policy, reducing the risk of sharp swings. But on AI infrastructure specifically, divided government more likely means slower approvals and more legislation, not a policy vacuum.
A measured view
With one week to go, polls and betting markets offer no certain answer. For investors, the thing to track is not the election-night result itself but where AI infrastructure legislation and budget negotiations actually go afterward. Whether divided government really slows compute expansion depends on how hard the new Congress fights over specific provisions such as water and power requirements and permitting authority. Until the outcome is clear, market pricing of the AI trade is likely to stay sensitive and fragile.
Source: Zhihu https://www.zhihu.com/question/2098307695795245562/answer/2099088083484407574?utm_medium=openapi_platform&utm_source=b38f6831b176
0 条评论
还没有评论，写下第一条。