Democrats won the House and the Senate on Tuesday. President Trump still has his veto, so neither side can pass much on its own.
For the AI trade, I think that means a slower buildout than companies planned, though it keeps going. The money can keep flowing, but it'll take longer to turn into working data centers. And the companies that already own data centers come out of it a little stronger.
Most of the rest, I expect, will be noise. But Wednesday's market didn't treat every stock the same. The Dow jumped about 1,570 points, its best day since April 2025, and underneath that, a few groups moved hard in opposite directions.
Sectors to watch
AI data centers: infrastructure fell, hyperscalers rose
This was the biggest split I saw.
Companies that build data centers fell hard. Modine and Sterling Infrastructure, which make cooling gear and prepare building sites, were each down more than 20% at midday. Power companies that sell electricity to data centers, like Vistra and Constellation, fell too.
Meanwhile, Microsoft and Amazon were each up 6% to 7%.
The worry is mostly about the states. Texas had already paused new data-center permits in September, under Republican Governor Greg Abbott, while it checks whether the power grid and water supply can keep up. On Tuesday, Abbott lost to Democrat Gina Hinojosa, who campaigned on a pause of her own.
Goldman Sachs expects new Democratic governors in Texas, Georgia and Michigan to pause new data centers while they write tougher rules. It estimates the states likely to add limits hold about 45% of the data centers planned in the US. Jefferies sees it differently: it thinks tighter rules in Texas make building more expensive without cancelling many projects.
A pause still leaves room to build later. So I'd expect slower permits, higher costs and more fights over power and water.
That can help the companies that already have data centers running. The hyperscalers (Microsoft, Amazon and Google's parent Alphabet) said last month they still can't build fast enough to meet demand. If new data centers take longer to approve, the ones already running get more valuable. Investors also bet a slower buildout would rein in Big Tech's huge AI spending. I think that gives those three more pricing power and more time.
Neoclouds like CoreWeave and Nebius are in a harder spot. They rent out AI computing power, and their growth depends on building fast. CoreWeave fell about 10%. Data-center operators like Hut 8 and IREN fell too.
There's a limit to my read. The hyperscalers are building new data centers too, many of them in Texas. And a neocloud with data centers already running and rented out could also benefit. The companies I'd watch most closely are the ones counting on sites that aren't finished or approved yet.
Clean energy
Solar stocks like First Solar and Enphase jumped. A 2025 tax law already cut back many clean-energy tax credits. Cutting what's left would take a vote in Congress, and this Congress is unlikely to give one. That doesn't bring back the credits already cut.
Hospitals and Medicaid insurers
HCA, a hospital chain, was up about 10%. Insurers that run Medicaid plans, like Centene and Molina, rose too. Investors are betting a Democratic Congress blocks another round of coverage cuts by law.
But the cuts already in the 2025 law still happen, including Medicaid work requirements starting in 2027. Undoing them would take a new law, and that still needs Trump's signature.
Oil and defense
Oil fell about 8%. Vice President Vance blamed the Iran war for the Republican losses, and talks to end the war were already under way. I think traders read the election as more pressure to end it, even though the war hasn't ended.
Defense stocks fell on worries that a Democratic Congress gives the Pentagon a smaller budget increase than investors expected.
The old Congress still has until January
The new Congress starts on January 3. Until then, Republicans still run both chambers. The weeks in between are called a "lame-duck" session, and it starts November 9.
Three things to watch:
November 9: the Senate holds a first procedural vote on a bill to speed up federal permits for energy projects and power lines. It's an early step, and it would still need the House. If it moves forward, that would cut against my slower-buildout read.
December 11: government funding runs out unless Congress passes a spending bill.
The debt limit: the government is expected to hit its borrowing limit in the next six months. Hitting it doesn't mean a default right away, because the Treasury can move money around to keep paying bills for a while. Senate Majority Leader John Thune has said Republicans could raise the limit before January. If they don't, raising it needs the new Democratic Congress, which gives Democrats something to bargain with.
What the vote didn't change
The Fed still sets interest rates. Companies still have to earn money. And Congress changing hands doesn't end the war.
On Tuesday I posted J.P. Morgan Asset Management's finding that the S&P 500 had positive returns in the 12 months after each of the 19 midterms from 1950 through 2022. That's history. It doesn't predict next year, and neither does one day of trading.
3 questions I'm asking about my own stocks
Does this company need a new law? If its growth depends on Congress passing something new, like a tax break or a subsidy, that just got harder.
Does it depend on government spending? Big changes to Medicaid or Medicare now need this Congress. Defense contracts and federal grants depend on yearly funding, and the next deadline is December 11.
Does it need to build something new, and where? A company waiting on new sites in Texas or Georgia carries a risk that one with buildings already running doesn't.
If the answer to all three is no, the election probably matters less to that stock than its next earnings report.
Want an AI to do the first pass? Paste this into ChatGPT or Claude:
Act as a skeptical fundamental-research coach for self-directed investors. Help me understand what drives a business, identify dependencies worth investigating, and discover new stocks using explicit criteria.
My holdings: [PASTE TICKERS; ADD PORTFOLIO WEIGHTS IF AVAILABLE]
Research date: [DATE]
Investment horizon: [E.G., 3–5 YEARS]
Political scenario: Assume Democrats control both chambers after the November 2026 midterms, while President Trump retains presidential veto power. The new Congress begins January 3, 2027. Treat this as a hypothetical scenario unless verified through current official sources. Do not assume party control guarantees a particular policy outcome.
1. Establish the business baseline
For each company, briefly explain:
• What it sells, who pays, and its main revenue and profit drivers.
• What management identifies as the main sources of future growth.
• Whether that growth is supported by current operations, signed contracts, or projects and assumptions that have not yet materialized.
Separate established facts, management expectations, and your own inferences.
2. Investigate three dependencies
A. Legislation
Does the existing business—or a material part of expected growth—depend on federal legislation that has not become law?
Identify the specific legislation and its status. Distinguish:
• A law that has not passed.
• An existing law whose funding, implementation, expiration, or renewal remains uncertain.
• An agency rule or executive action.
• A possible policy benefit that is not necessary for the business to succeed.
Explain the mechanism connecting the policy to revenue, costs, cash flow, or investment spending. Do not equate a potential benefit with business dependence.
B. Government funding
How much revenue depends directly or indirectly on government spending?
Distinguish direct government contracts, healthcare reimbursements, subsidies or tax credits, and sales to customers supported by government funding.
Report the disclosed dollar amount or percentage, fiscal period, and denominator. If the company does not quantify the exposure, say so. Do not substitute a segment's total revenue for its government-funded revenue or confuse contract backlog with recognized revenue.
Explain what could change the economics: reimbursement rates, appropriations, contract renewals, eligibility rules, or payment timing.
C. Facilities and permits
Does expected growth require new or expanded physical facilities with outstanding permits or approvals?
Identify material projects, locations and states, expected timing, and disclosed investment amounts. Distinguish federal, state, and local approvals from utility interconnection, financing, and construction constraints.
State whether each project is proposed, awaiting approval, approved, under construction, or operating. Distinguish company-owned projects from customer projects that affect demand.
Do not infer permitting difficulty from a state's political affiliation.
3. Use evidence carefully
Start with the latest available 10-K, or 20-F for a foreign issuer. Check subsequent 10-Qs, 8-Ks, and official company disclosures for material updates. Use official legislative, agency, and permitting sources to verify current policy and project status.
For each material finding:
• Include a short supporting quotation where available, the document date, section name, and source link.
• Explain what the evidence establishes and what remains uncertain.
• Label the conclusion documented, inference, or not found.
• If support is missing, write "not found in the sources reviewed" and name those sources. Absence of disclosure is not proof of no exposure.
• Keep quotations brief and within applicable quotation limits; use cited paraphrases for additional evidence.
If browsing or filings are unavailable, explain the limitation rather than inventing evidence.
4. Connect exposure to fundamentals
For each holding, explain whether the identified dependency affects current earnings, future growth, or both.
Then assess its importance alongside:
• Revenue growth and operating margins.
• Operating cash flow, capital spending, and free cash flow.
• Debt, liquidity, and funding needs.
• Customer concentration and competitive position.
Use dated, sourced figures where relevant. Avoid unsupported precision or a composite score that hides uncertainty.
Give me:
1. The most important dependency to investigate.
2. One fact that would strengthen the investment thesis.
3. One fact that would weaken it.
4. Two specific questions for my next filing or earnings-call review.
5. Handle funds separately
For ETFs and mutual funds, use the latest official holdings and fund documents. Do not analyze a fund as if it were an operating company.
Review its ten largest holdings and any additional material concentrations relevant to this screen. State the holdings date and percentage of fund assets covered. Label this a partial look-through analysis, and do not imply precise exposure for the unreviewed holdings.
6. Turn the findings into new-stock research
First derive 3–5 screening criteria from the analysis, before naming companies. Focus on business characteristics such as internally funded growth, durable customer demand, manageable debt, and limited dependence on unresolved approvals.
Then identify up to five publicly traded companies outside my list that appear to meet those criteria. For each:
• Explain which criteria it meets, supported by primary sources.
• Identify its main tradeoff or disqualifying risk.
• State what still needs verification.
• Include one relevant, dated valuation measure, explaining its limitations.
• Explain why a strong business might still be a poor investment at the price examined.
Present these as research candidates, not buy recommendations. Do not force five names if the evidence supports fewer.
Output
Begin with a concise comparison table:
Holding | Business driver | Pending-law dependence | Government funding exposure | Facility/permit dependence | Evidence gaps
Follow with short company notes, a separate fund section, and the new-stock research shortlist.
Finish with the three most useful next research steps and briefly explain what each would teach me.
Use plain English and define necessary financial terms. Do not predict stock prices, give price targets, or make automatic buy/sell judgments based on election outcomes.
Then look up two of those quotes in the actual report. If one is wrong, check the rest yourself.
Tuesday's Insider issue
On Tuesday, Insiders got my full read on Microsoft, Amazon and Alphabet after earnings: how much capacity they have, how they're paying for it, and how I rank them. It's for paid subscribers only, on email and on the web.
Stay disciplined,
Koh
Market moves are November 4, 2026 closing figures unless marked midday.
Sources: CNBC, Reuters, Stocktitan, AP, Texas Tribune, Goldman Sachs via Morningstar, Jefferies via Bloomberg, the US Senate schedule, Harbinger Strategies, the Committee for a Responsible Federal Budget and J.P. Morgan Asset Management.
Positions as of publication: I may hold positions in some of these companies.
Disclaimer: Nothing in this email, the Reset and Invest newsletter, the Make Your Own Alpha book, or any course or digital product from Starshine Media LLC constitutes investment advice or a recommendation to buy or sell any security. Numbers and observations are as of publication. I may hold positions in companies discussed. Always do your own research and consult a licensed financial advisor before making investment decisions.