The final showdown of Washington's power struggle will determine the stock market's next move! The U.S. midterm elections will affect trillion-dollar AI computing power investment.

Title context: The final showdown of Washington's power struggle will determine the stock market's next move! The U.S. midterm elections will affect trillion-dollar AI computing power investment. Text: As the U.S. midterm elections approach, the AI trading theme linked to the AI computing infrastructure boom is going through repeated tug-of-war between rebounds and risk repricing. On October 19, the Philadelphia Semiconductor Index rose 1.61%, ending eight consecutive trading days of declines; on October 21 it rose further by about 2% to close at 11,505.6 points, while the Nasdaq 100 Index rose 1.13% to close at 29,822.52 points. However, the Philadelphia Semiconductor Index is still down about 8.89% cumulatively since October and more than 20% from its all-time high, still hovering in technical bear market territory, with the short-term rebound not yet repairing earlier losses; moreover, catalyzed by expectations that the political situation in the Middle East GEO Group Inc seems about to deteriorate, Nasdaq 100 stock index futures fell by about 2% at one point before the U.S. market open on Thursday, which also means AI computing power and even broader semiconductor stocks may once again experience severe volatility. Within the global stock market's AI computing power theme, the repeated swings in the memory chip supply chain are especially evident: Micron and SanDisk rose 1.96% and 0.94% respectively on October 21, after rebounding 2.97% and 5.53% respectively on the 19th; South Korea's KOSPI benchmark index, after rising 3.29% on the 21st, fell 3.52% on the 22nd to close at 5,533.77 points. Shares of Samsung Electronics and SK Hynix, the world's two largest memory chip giants, fell sharply by 4.03% and 5.49% respectively on the 22nd, with cumulative declines since October reaching 20.11% and 31.14%. The coexistence of a short-term rebound and severe volatility in U.S. chip stocks and renewed pressure on South Korean memory chip giants' share prices shows that the repair of risk appetite in the AI computing power and semiconductor supply chain remains unstable. It is understood that as the midterm elections approach, Wall Street still recognizes the earnings expansion brought by AI, but there are clear differences over the timing of adding positions and the concentration of allocations. The CIO of BofA Wealth Management argued in an October 13 report that strong earnings and AI-driven productivity gains may continue to support the stock market through 2027; Citi's own strategist Scott Chronert maintained a target of 8,100 points for the S&P 500 Index while emphasizing that achieving the target requires continued strong AI-related earnings growth, easing concerns about inflation and a hawkish Federal Reserve, and broad participation in the rally by traditional sectors. The strategy team led by Morgan Stanley Chief U.S. Equity Strategist Mike Wilson believes that if large benchmark stock indices further reflect the pressure from the earlier broad correction in individual stocks, opportunities will emerge to add allocations to quality stocks; BofA's derivatives trading strategy team prefers to wait until after the Federal Reserve meeting and the midterm elections before increasing risk assets. Strategists at Jefferies Financial Group Inc. (Jefferies) recently more explicitly proposed diversifying away from AI; although its GREED & fear portfolio still holds Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR, Samsung and SK Hynix, the related allocations are below regional benchmark weights. Being long-term bullish on AI earnings growth prospects while short-term controlling risks from valuations, policy and highly leveraged technology holdings is simultaneously becoming the basis for midterm-election-period trading by Wall Street institutions. The absolute leader and market-cap dominant player in the global AI computing power themethat is, the "AI chip superpower" NVIDIA Corporation (NVDA.US) with a market value exceeding $5 trillionremains an important positive pricing anchor for Wall Street strategists amid this divergence. Using its October 21 closing price of $221.97 as the baseline and assuming total shares of about 24.15 billion remain unchanged: Citi on October 22 maintained "Buy" with a target price of $315, implying upside of about 41.91% and an implied market value of about $7.61 trillion; its bullish logic includes incremental computing demand from continuous agent operation, multi-round model calls and tool execution. Morgan Stanley in early October restored NVIDIA Corporation's status as a "top pick" in semiconductors, maintaining "Overweight" and a target price of $300, implying upside of about 35.15% and an implied market value of about $7.25 trillion, with focus on expansion beyond GPUs into Vera CPU and networking businesses. In addition, BofA on August 18 reiterated "Buy" with a target price as high as $350, arguing that NVIDIA Corporation's strong free cash flow can support its AI ecosystem investments, implying upside of about 57.68% from the above baseline and an implied market value of about $8.45 trillion; Jefferies on August 27 maintained "Buy" with a target price of $300, based on financial results and business outlook continuing to exceed expectations. The average target price of 62 analysts compiled by S&P Global is $328.72, implying upside of about 48.09% and an implied market value of about $7.94 trillion. Whether these valuation expectations can be realized ultimately still depends on whether computing power demand can continue to deliveries, revenue and cash flow, and the midterm elections are adding policy variables to this translation process. Supreme Court reshapes the election battle: how U.S. congressional control enters AI asset pricing Before the November 3 midterm elections, the Supreme Court intervened in election-related cases at an unusual density, and its rulings generally improved Republicans' competitive conditions. Democrats need a net gain of four seats in both the House and Senate to take control; Republicans losing either chamber could obstruct Trump's legislative agenda and open space for Democratic-led congressional investigations. For the market, this means the election results will change the ability to advance policy, the intensity of oversight and the priority of regulatory issues, and AI infrastructure construction therefore enters the of political risk pricing. The Supreme Court currently has a 6-3 conservative majority. In April this year, the Court weakened a key provision of the Voting Rights Act, making it harder for racial minorities to challenge district maps on grounds of racial discrimination, and some Republican-led Southern states promptly redrew House districts with concentrated Black voters. At the same time, Trump last year pushed mid-decade redistricting in the ten-year census cycle, setting off competition between the two parties over House seats; the Supreme Court allowed Texas to implement a new map favorable to Republicans. It is understood that a 2019 ruling held that federal courts have no authority to address partisan gerrymandering, also providing an institutional backdrop for this trend, and liberal Justice Kagan criticized it this week for exacerbating political polarization. On campaign finance, the Court in June eliminated federal limits on coordinated spending between parties and candidates, and in October allowed parties to enjoy preferential television and radio advertising rates originally mainly available to candidates, with Republican committees more actively using the relaxed spending space. But Republicans' judicial advantages coexist with Democrats' electoral advantages. In September the Court blocked Missouri from using a new district map favoring Republicans, also allowed California to implement a map favoring Democrats, and issued rulings on mail voting rules and grace periods that Republicans did not support. Preliminary polling cited by some media still shows Democrats are expected to win the House; as of publication, the most likely scenario reflected by Kalshi and Polymarket exchanges is that Democrats win both chambers. Senior U.S. legal scholar Ilya Somin expects more litigation is still likely after the election, but whether it reaches the Supreme Court is difficult to judge. What the market currently needs to continuously digest is changes in congressional control and policy execution capacity; it should be noted that legal victories on redistricting and campaign finance themselves cannot lock in the final election outcome. From ballots and electricity bills to AI chip orders, trillion-dollar-scale AI computing power investment meets the growing bill figures of ordinary Americans The reason this political uncertainty carries market weight is that AI has become an important earnings pillar of this U.S. stock bull market. Forecasts cited by Reuters show that spending on data centers and other AI infrastructure will exceed $1 trillion in 2027, and around 2030 global capital expenditure closely related to AI computing infrastructure construction may even climb to the super threshold of $3 trillion. Christopher Gannatti, a senior analyst at WisdomTree, pointed out that since ChatGPT was launched in November 2022, one of the signals the market watches most closely is how much money five or six large technology companies are actually investing to accelerate AI data center construction globally. The AI capital expenditure forecasts and outlook from JPMorgan's strategist team show hyperscale cloud providers' capital expenditure at about $950 billion in 2026, about $1.4 trillion in 2027 and about $3 trillion in 2030. As facilities expand, energy bills, environmental pressure and concerns about AI itself are turning the capital market's growth narrative into political issues that community residents can directly feel. A Reuters/Ipsos poll this month showed that 69% of registered Democratic voters and 43% of registered Republican voters support restricting data center construction, and opposition has a bipartisan base. Market participants interviewed regard Democrats controlling both chambers as the scenario among potential election outcomes that would put the most pressure on the AI trade, but generally believe that nationwide legislation to fully block data center expansion is still difficult to enact. Matt Gertken, a strategist at BCA Research, judges that Trump is very likely to block such bills and is more likely to accept legislation strengthening AI regulation rather than "slamming the brakes" on capital expenditure and construction. The importance of data centers to economic growth and concerns about falling behind China may also constrain some Democratic lawmakers. A more direct market shock comes from escalated oversight: Brian Gardner, a portfolio manager at Stifel, expects that after Democrats control Congress, hearings, investigations and negative news may increase, thereby amplifying AI asset volatility. The political power that truly affects project implementation is also distributed at the state and local levels. Who holds governorships, state legislatures and utility regulatory positions may change the rules, restrictions and construction conditions facing data centers. Morgan Stanley is watching gubernatorial elections in seven states: Texas, Ohio, Pennsylvania, Georgia, Nevada, Wisconsin and Michigan; Beacon Policy Advisors emphasizes that investment is highly concentrated in specific regions, so these local elections are especially important. Trump proposed on October 21 that data center developers should pay a "dividend" to communities to win support, also showing that how construction benefits are distributed between companies and residents has entered policy discussion; however, Trump's statement is currently only a proposal. Derived from project economics, policy risk first affects commissioning timing, cost bearing and investment returns. Delays in construction or grid connection may push back some server deployment, equipment acceptance and computing power service billing; if companies need to bear more grid expansion, community compensation or compliance costs, project returns will also come under pressure. For new cloud companies relying on external financing, the mismatch between delayed revenue and ongoing financing expenses is more sensitive. Recent debt concerns surrounding CoreWeave have shown that investors are beginning to scrutinize strictly the funding sources and cash recovery capacity of AI expansion; new policy friction may further affect its credit spreads and equity valuation. Investment analysis of the AI supply chain needs to focus further on power supply guarantees, construction permits, customer contracts and financing capacity for downstream data center projects, as well as post-commissioning computing power utilization and revenue realization progressthat is, who has secured power supply and underlying energy guarantees, obtained permits for accelerated AI data center construction, who has reliable customers, higher utilization and sufficient financing capacity, and when these conditions revenue. The accelerated expansion of AI agent applications led by Meta Muse and OpenAI Astra still provides long-term demand support for NVIDIA Corporation and memory suppliers, but progress in AI data center construction and operation across different customers, regions and large infrastructure projects may change the pace of equipment procurement, delivery and revenue recognition. The latest judgment by Kevin Shea, a portfolio manager at BNY Wealth, reveals this risk: policy uncertainty may not necessarily push the AI investment theme off track, but it may delay project implementation and raise construction and operating costs. Therefore, the substantive impact of the midterm elections on the AI trade mainly centers on whether policy changes hinder massive capital expenditure from being converted on schedule into operable computing power and further into sustained revenue and cash returns. Three potential outcomes of the U.S. midterm elections are shaping the next stage of global stock market pricing The U.S. midterm elections are pushing the pricing focus of the AI bull market from the record scale of AI capital expenditure around the AI computing infrastructure boom further toward policy constraints, commissioning progress and financing costs. The Supreme Court's rulings have changed the election's competitive conditions, and congressional control will affect legislation, investigations and the regulatory agenda. Some media cited Wells Fargo & Company statistics reporting that in the 12 months after every midterm election since World War II, the S&P 500 Index has recorded gains; but as of this moment this year, the S&P has already risen 13% cumulatively and has not experienced a typical pre-election pullback or even a pre-election bear market trend. Therefore, whether the post-election rally can continue depends more on corporate earnings, 2027 economic growth and consumer resilience, while bipartisan opposition to data center construction gives this election cycle special weight for the AI trade. Divided Congress is the scenario Citi senior strategist Stuart Kaiser considers most favorable for the current stock market. If Democrats retake only one chamber, divided chambers plus Trump's veto power will reduce the possibility of major tax and regulatory policy changes, leading investors to price more around earnings and economic growth. Carson statistics show that since 1950, during Republican presidents, divided Congress has corresponded to an average annual gain of 13.7% for U.S. stocks, higher than 8.3% and 4.9% when Republicans and Democrats respectively controlled both chambers. Limited room for fiscal expansion may also ease pressure from new deficits and Treasury supply, support bond prices, and improve corporate investment returns through lower financing costs. However, Trump can still influence the market through tariffs and foreign policy, and congressional gridlock does not mean administrative policy volatility disappears at the same time. If Democrats retake both chambers, the first shock may be to AI asset valuations and risk appetite, and only afterward will it depend on whether policy can substantively affect operations. Stricter data center construction rules or model regulation may delay purchases of chips, servers and supporting equipment and postpone realization of AI service revenue. But Trump tends to support AI, and it is extremely difficult for Democrats to obtain the supermajority needed to override a presidential veto, so nationwide legislation severely restricting AI expansion still faces a high threshold. More realistic pressure comes from hearings, investigations and investors' reassessment of future policy direction; combined with previous Reuters reports, state and local permitting, power supply and community coordination are the important channels through which construction timing and costs change. For leaders such as NVIDIA Corporation and Microsoft Corporation, what needs attention is how these changes transmit to customer commissioning and order fulfillment. A Democratic sweep of both chambers could also drive repricing across industries and assets: if Democrats use political influence to ease spending cuts to Medicaid, some healthcare service companies may benefit; stricter scrutiny of military spending may suppress defense contractors' expectations. The direction of bonds and the dollar depends more on the subsequent policy mix: if Trump turns to more aggressive tariffs or foreign actions, rising inflation and risk premiums may pressure bonds; if slower AI investment weakens U.S. growth expectations, it may weigh on the dollar, but high energy prices and Federal Reserve rate hike expectations may again provide support. This is also why the same election outcome can simultaneously produce different effects on fiscal, inflation and growth directions and cannot be directly equated with a single conclusion of stock, bond and currency gains or losses. Republicans holding both chambers may extend a policy environment favorable to AI expansion and drive trading in economically sensitive sectors such as financials and industrials on growth expectations. But if subsequent fiscal stimulus pushes up demand, deficits and bond yields, corporate financing costs and equity discount rates may also rise, offsetting part of the growth benefit. From this perspective, the core investment judgment corresponding to the three election outcomes is whether AI capital expenditure can, under tolerable funding costs, smoothly complete approval, construction and grid connection and orders, revenue and cash flow. For mega AI infrastructure projects with clear power supply arrangements, reliable customer contracts and sound financing capacity, their ability to deliver will more easily demonstrate value amid policy divergence.