Behind the video
How Midterms will F**K Up the Stock Market
Overall Thesis
The host argues that midterm election outcomes will influence Treasury yields, government shutdown risk, and approvals for substantial AI data center investment. He favors modest, low-debt equity exposure after midterms and expects established hyperscalers to benefit if restrictions constrain new compute capacity, while warning that divided congressional control could delay investment and hurt growth.
Narratives
The host views historical post-midterm equity performance as supportive of remaining invested with limited leverage. He also warns that election-driven shutdowns, data center restrictions, and elevated yields could undermine growth and market performance.
Key Arguments
- He cites Bank of America research showing consistently positive six- to twelve-month post-midterm periods since World War II.
- A Democratic sweep could restrain deficit expansion and Treasury supply.
- A Republican sweep could support spending and earnings but keep yields elevated.
- Split congressional control could create shutdowns and delay data center investment.
Risks acknowledged
- Historical performance does not justify an aggressive all-in position.
- The market may be underestimating elevated yields.
- Inflation is exceeding wage growth and pressuring consumers.
Predictions (1)
NBISNebius GroupThe host expects Nebius to break out toward 336 while arguing that neocloud growth depends on reduced opposition to data centers. Financing access and the operational complexity of running data centers remain important constraints.
Key Arguments
- The host states his own near-term breakout expectation.
- Reduced political opposition would allow neoclouds to finance additional capacity and grow.
- AI usage and demand for compute are expanding.
Risks acknowledged
- Data center financing becomes harder when political restrictions intensify.
- Buying chips alone is insufficient; facilities require substantial engineering and operational support.
- Established hyperscalers may gain an advantage if new capacity is constrained.
MSFTMicrosoftThe host explicitly agrees that restrictions on new data center capacity could benefit Microsoft and other established hyperscalers. Existing infrastructure would become more valuable as compute shortages increase pricing power and reduced capital spending limits debt issuance.
Key Arguments
- Incumbents already possess substantial data center infrastructure.
- Slower competing capacity growth could prolong compute shortages.
- Lower capital expenditure could reduce debt issuance and credit risk.
Risks acknowledged
- Restrictions would delay the broader AI buildout.
- Compute shortages could increase AI inflation.
- The benefit depends on the extent of financing and permitting constraints.
Predictions (1)
METAMeta PlatformsThe host includes Meta among established hyperscalers that could benefit when new data center financing becomes harder. He expects existing infrastructure and scarce compute capacity to support incumbent pricing power.
Key Arguments
- Existing data centers confer an advantage over new entrants.
- Capacity shortages could support premium pricing.
Risks acknowledged
- The thesis depends on restrictions slowing competing capacity.
- The broader AI buildout would be delayed.
Predictions (1)
AMZNAmazonThe host expects Amazon's established infrastructure to provide an advantage if restrictions impede new data center entrants. Scarce compute capacity could allow incumbents to charge a premium.
Key Arguments
- Existing infrastructure reduces dependence on new entrant financing.
- Slower capacity additions could reinforce pricing power.
Risks acknowledged
- The benefit depends on financing restrictions.
- Compute shortages would delay deployment and increase AI inflation.
Predictions (1)
GOOGLAlphabetThe host names Google as an incumbent that could benefit if financing constraints limit new compute suppliers. Existing data centers could command premium pricing during a prolonged capacity shortage.
Key Arguments
- Established infrastructure creates an advantage.
- Constrained new supply could increase pricing power.
Risks acknowledged
- The outcome depends on restrictions affecting new capacity.
- The broader buildout could be delayed.
Predictions (1)
CRWDCrowdStrikeThe host reports Morgan Stanley's favorable cybersecurity view and expresses his own enthusiasm for the sector. He acknowledges expensive valuations and does not give a company-specific forecast.
Key Arguments
- Morgan Stanley maintains a favorable view of CrowdStrike as a cybersecurity investment.
- The host expresses general support for cybersecurity.
Risks acknowledged
- Cybersecurity stocks are very expensive.
- The host mentions less expensive alternatives without identifying them.
The host highlights Morgan Stanley's favorable assessment of Palo Alto Networks alongside his general enthusiasm for cybersecurity. Valuation concerns temper that positive sector view.
Key Arguments
- Named as a favorable cybersecurity play by Morgan Stanley.
- The host likes the cybersecurity sector.
Risks acknowledged
- The sector is expensive.
Okta is mentioned among Morgan Stanley's favored cybersecurity names. The host supports cybersecurity generally but supplies no specific Okta prediction.
Key Arguments
- Included in Morgan Stanley's favorable cybersecurity list.
- The host expresses sector enthusiasm.
Risks acknowledged
- Cybersecurity valuations are expensive.
BEBloom EnergyThe host reports Morgan Stanley's interest in Bloom Energy as a behind-the-meter power investment associated with data centers. He does not clearly make or endorse a Bloom-specific forward claim.
Key Arguments
- Highlighted in Morgan Stanley's behind-the-meter power ideas.
- Discussed in the context of growing data center infrastructure demand.
Risks acknowledged
- Data center investment faces permitting and political obstacles.
NVDANvidiaNvidia is discussed as a supplier of leading-edge chips used in costly data center deployments. It also appears in Morgan Stanley's list of potential open-weight model beneficiaries, without a clear host-endorsed Nvidia forecast.
Key Arguments
- Leading-edge chips are a major component of data center spending.
- AI usage and compute demand are expanding.
Risks acknowledged
- Political opposition could delay substantial data center spending.
- Operational and financing requirements extend beyond buying chips.
Hedges & Caveats
- A narrow Democratic House victory with a Republican Senate could produce damaging shutdowns and prolonged data center opposition.
- Higher yields, inflation exceeding wage growth, and weak consumer sentiment remain risks.
- AI employment effects are uncertain, and some highly exposed occupations have experienced rising unemployment.
- Positive AI breakthroughs could improve public acceptance, but that outcome is uncertain.
- Cybersecurity stocks are expensive.
- The host recommends low-debt, low-risk exposure rather than aggressive positioning.
- An unexpectedly bad CPI report could change his expectation that the Fed will avoid a rate hike.