Transcript summary of an In the Know commentary. Date: 2026-11-07.

Overview

The commentary addresses the recent market decline during the midterm election period, the broader implications of the election, and the outlook for artificial intelligence, inflation, credit markets, labor, and monetary policy. The speaker, Cathie Wood, presents historical market data, sector analysis, and forward projections based on her firm's research.

Tribute to John Nosta

The speaker announced the death of John Nosta, a friend and long-time contributor to the firm's Brainstorm sessions. Nosta had participated in Brainstorm almost weekly since the firm's founding in 2014. He left a wife and three children. Shortly before his death, he published The Borrowed Mind: Reclaiming Human Thought in the Age of AI, a book on the effects of AI on individuals. Nosta's background was in medicine and psychology. The speaker encouraged viewers to purchase the book and support his family.

Historical Midterm Election Performance

• Post-midterm returns: Market performance in the year following a midterm election has generally been positive under both Republican and Democratic administrations, based on data from 1958 to 2022.
• Typical pattern: Pre-election uncertainty is usually followed by a significant turnover of seats held by the incumbent party. The speaker noted that this turnover occurred in the current cycle.
• Precedent: A comparable turnover occurred during the Obama administration. The turnover was smaller in the Senate than in the House.
• 1986 and 1987: The market was down in 1986, the only year among the 1980s and 1990s comparisons whose following year did not end higher. Late 1987 saw a sharp decline, with the market falling by more than portfolio insurance was designed to cover. Investors seeking to exit simultaneously undermined the insurance. The market declined 22% on one day.
• Following year returns: Even after the 1987 decline, the market finished the year up marginally.
• Recent midterms: The market rose roughly 13% to 14% in the year following each midterm election shown.

Election Results and Voter Preferences

• Congressional results: Results appeared to show a Democratic sweep across Congress.
• Ballot initiatives and local measures: Results were more balanced. Voters rejected a wealth tax in California and approved tax limits in multiple states and cities.
• Voting regulations: Measures moved in a more conservative direction.
• Drivers: The Iran war and the resulting inflation, particularly oil and gasoline prices, angered voters across party lines, including independents.
• Market assessment: The speaker stated that the expectation of a quick end to the Iran war was incorrect. The market began pricing a defeat for President Trump and Republicans around May to June. Many AI stocks peaked around that period.
• Defense stocks: Defense stocks declined sharply. The speaker stated that the market expects higher healthcare spending, which would require offsetting cuts, given the current deficit and debt levels.
• Potential revenue sources: Additional taxes on banks and energy companies that have benefited from recent conditions were cited as possibilities.

Data Centers

• Local concerns: The speaker stated that companies building data centers did not adequately account for local political conditions. The working assumption was that data centers create well-paid jobs and increase tax revenue.
• Electricity prices: Electricity prices have risen in areas with data center growth.
• Water usage: Meta and Amazon use effectively recycled water in their data centers, according to the speaker. The speaker stated there is no significant additional water usage.
• Other concerns: Concerns include pollution from generators, community blight, and displacement of residents. Displacement was cited as a concern in Texas.
• Sulphur Springs, Texas: According to the speaker, Elon Musk offered a data center project in Sulphur Springs that would reduce property taxes by 20%, reduce electricity bills by 40%, and employ 5,000 people in a town of 20,000.
• Virginia: Virginia has the largest concentration of data centers in the world. According to a Brainstorm participant, property taxes and possibly sales taxes have declined. Economic development has brought commercial construction, new employment, and new taxpayers. The participant reported that the area is building approximately one high school per year and possibly an elementary school.
• North Carolina: A town in North Carolina reportedly shows a similar pattern to Sulphur Springs and the Virginia communities.
• Recommendation: The speaker stated that companies should consider local concerns and respond to them.

Credit Markets

• High-yield credit default swaps (CDS): Spreads widened sharply during COVID-19. Current levels are elevated but have not reached a new high.
• Hyperscalers: Credit default swaps now exist for hyperscalers. Investors are concerned that these companies have shifted from strong free cash flow generation to negative free cash flow over the past year. Hyperscaler CDS spreads remain well below high-yield levels. They are close to the levels reached for investment-grade CDS during the worst phase of COVID-19.
• Post-election movement: Hyperscaler CDS spreads have improved slightly. Investors expect hyperscalers may slow data center buildout if states or cities restrict data centers. The speaker stated that the worst levels in these spreads have likely been reached.
• CoreWeave: CoreWeave's CDS spreads exceed 1,000 basis points, well above high-yield levels, which were near 300 basis points. CoreWeave's contracts are primarily with hyperscalers and are backed by those contracts. The speaker stated that the market does not believe these contracts provide sufficient security. CoreWeave's business is concentrated in neocloud services, which the speaker stated would be directly affected by a data center slowdown.

Artificial Intelligence

Framework

The speaker described AI as the largest catalyst for other innovation. It connects to robotics, energy storage, blockchain, and multi-omics. The framework covers 15 technologies across five platforms.

Enterprise Spending

• Source: Ramp, a fintech company, provided spending data.
• Top 1% of firms: Monthly spending on frontier models is almost $10,000 per employee.
• Top 10% of firms: Monthly spending is slightly above $800 per employee.
• Remaining firms: Monthly spending is approximately $15 per employee.
• Implications: Technology companies and many venture-backed companies are in the top spending group. The speaker identified a risk if venture funding declines. The speaker expects only a brief interruption in venture funding.
• Efficiency: Apple reportedly moved some workloads to lower-cost, lower-performance models where high performance is not required.
• Fast adopters: Early adopters report substantial productivity gains and are expected to pressure competitors to adopt AI.

Inference Costs

• Cost decline: For a fixed level of performance, AI inference costs decline 99% per year. The speaker also cited a 75% to 99% annual decline in AI training and inference costs.

Software and Hardware Investment

• Historical trend: Software and hardware spending both rose as a share of GDP from 1960. Hardware grew faster in the early period, during the mainframe and minicomputer era. Software caught up in the 1990s and has since continued to rise, surpassing hardware as a share of GDP until recently.
• Current position: Spending on data centers and related components is now the main driver. The speaker stated that current levels equal those at the peak of the tech and telecom bubble, and that the buildout has only started.
• Equipment spending: Non-defense capital goods orders as a share of GDP have been largely flat since the early 1980s, with a modest increase in the late 2000s. The speaker expects this share to rise substantially.
• Software disruption: The speaker expects significant changes in software and identified losers. The speaker referred to a "SaaS-calypse." Atlassian and ServiceNow declined sharply. ServiceNow fell 27% in one day, despite its CEO, Bill McDermott, describing the company as the execution layer of AI.
• Frontier labs: OpenAI and Anthropic have achieved substantial success. The speaker expects SpaceX AI to benefit from orbital computing capacity, which may become more important given local data center opposition. Gemini has performed less well than the leaders. The speaker noted leapfrogging among competitors.

Robotics and Autonomous Vehicles

• Humanoid robots: Tasha Keeney and Daniel, who researched the robotics sector, estimate a humanoid robot opportunity of $26 trillion over 10 to 15 years. The speaker stated this equals roughly 20% of current global GDP.
• Robotaxis: The speaker estimated a $10 trillion opportunity.

Personal Assistants

The speaker described AI personal assistants as free tools that reduce the need to open multiple applications. The speaker named Anthropic's Duet, OpenAI's Dots, Meta's Muse, and Instincts, a private company. The speaker stated that applications dependent on transactions face challenges.

Inflation and Measurement

• Equipment inflation: Equipment inflation has diverged from software inflation. The speaker interpreted this as an imbalance between supply and demand and called for more equipment supply, including data centers.
• Software deflation: The speaker stated that official measures understate software deflation. Given AI training and inference cost declines, the speaker estimated the actual deflation is much larger.
• Measurement: The speaker stated that CPI and PPI measures contain significant errors for software and technology.
• Real increases: DRAM prices have risen sharply.

Energy

• Natural gas: Natural gas prices in the United States are rising through increased exports. The speaker noted that U.S. and European prices have diverged.
• Price change: Over the past year, gasoline prices rose roughly 30%. Natural gas prices, the largest feedstock for power generation, fell a little over 35%.
• Competitive position: The speaker stated that lower natural gas prices give the U.S. power grid a competitive advantage over Europe.
• Storage: U.S. natural gas storage is at normal winter levels. European storage is well below normal. The speaker expected stronger demand for U.S. gas in Europe, limited by export infrastructure.
• Oil and electric vehicles: The speaker stated that high oil prices have accelerated the shift to electric vehicles. Transportation accounts for 57% of oil usage. The speaker expects transportation to shift toward natural gas, hydro, nuclear, solar, and wind, and expects oil prices to fall significantly.

Productivity and Labor

• Productivity growth: Productivity grew 3.6% at an annualized quarterly rate.
• Unit labor costs: Unit labor costs fell 0.7%.
• Compensation per hour: The gap between productivity and unit labor costs implies compensation per hour grew slightly below 3% at an annual rate.
• Four-year average: The four-year moving average of productivity is rising. The speaker stated that compensation growth should also rise if productivity continues to accelerate.
• AI-intensive sectors: Real output per payroll job in AI-intensive sectors is 11% above trend, based on Ramp data. The speaker stated this gain has occurred within the last few years. Output in the rest of the economy shows no comparable improvement.
• Equity market expectations: The speaker stated that investors will press companies that fail to show productivity gains.
• Employment data: Nonfarm payroll employment was stronger than expected. Household survey employment was broadly in line with weak expectations. Average hourly earnings rose 0.4%, above expectations. Unemployment is approaching 3%. The speaker noted the labor force has been reduced by retirements of baby boomers and reduced immigration.
• Youth unemployment: The unemployment rate for ages 16 to 24 fell from 9.9% to 9.3%.
• Hiring: Ramp data shows AI-intensive companies are hiring faster than non-AI-intensive companies.

Real GDP and Long-Term Growth

• Historical average: Real GDP growth has averaged 3% since the Industrial Revolution.
• Projection: The speaker projected growth of 7% through 2030 and approximately 9% through 2035, assuming the technology revolution proceeds.
• Uber analogy: The speaker compared AI's expected adoption to Uber's entry into New York City, which proceeded before regulatory approval was obtained.

Monetary Policy, Yield Curve, and Fiscal Outlook

• Yield curve: The yield curve remains flat. During the Industrial Revolution, the yield curve inverted as short rates rose with nominal GDP growth, while long rates were held down by deflationary effects of new technology. The speaker stated long rates were held back by 100 to 500 basis points.
• Deficit: The speaker expects the deficit to reach 2.8% of GDP by the end of 2028, citing expected gridlock and vetoes.
• Negotiation: The speaker suggested that President Trump may negotiate with Democrats.
• Space policy: The speaker stated there is bipartisan support for space programs for national security and competition with China. The speaker stated the United States is 10 years ahead of China in reusable rocket technology.
• Real yields: Interest rates have risen more because of real yields after the election.
• 5-year, 5-year inflation expectations: This measure remains stable. The speaker stated that expected government spending after the election is not reflected as inflationary in this measure.
• Money supply: M2 growth is 5.7%. Inflation has generally remained below money growth since 1997, when velocity began declining.
• Diffusion index: The index tracked by Federal Reserve Chairman Powell is moving in the right direction.
• Gold: Gold fell below $4,000 from roughly$5,600. The speaker expects further declines as inflation risks ease under Chairman Warsh.
• Bitcoin: Bitcoin has shown no correlation with gold since 2019, with a correlation of 0.1. The speaker expects Bitcoin to rise relative to gold.

Housing, Credit, and Consumers

• New home prices: New home sale prices are falling faster than reported because builders are buying down mortgage rates, absorbing approximately one percentage point of the rate buyers would otherwise pay.
• Builder inventory: Builder inventories are high. The speaker expects further price declines.
• Housing cycle: Manufacturing has exited the rolling recession. Housing has not. The speaker expects housing to recover.
• Savings rate: The savings rate has declined. The speaker attributed this to hand-to-mouth affordability pressures and to wealthier households feeling less need for buffers while stock prices rise.
• Bank CDS: Bank credit default swap spreads show a slight increase but no serious stress compared with 2008–2009, the European sovereign debt crisis, or COVID-19.
• Consumer and K-shaped economy: The speaker described a K-shaped pattern in which asset holders and non-holders experience the economy differently.

Trade and Corporate Factors

• Trade deficit: The speaker stated that the trade deficit increase reflects strong U.S. growth. Most recent import growth is related to data centers.
• Corporate tax: The effective corporate tax rate is 8%, the lowest in the world, according to the speaker. The speaker stated this has increased foreign direct investment. Accelerated depreciation also supports investment.
• Profit margins: Margins are near historic highs. The speaker expects some productivity gains to be shared with workers.

Growth Outlook

• Third quarter GDP: Growth was 3.4%.
• Outlook: The speaker expected growth to accelerate. A housing recovery would have a large multiplier effect, almost as large as capital spending.
• Inflation: The speaker expects inflation to be lower than expected because of productivity, lower oil prices, and AI cost reductions.
• Interest rates: The speaker expects nominal GDP growth to raise short-term rates and long-term rates, with short-term rates rising more. The speaker stated the increase would not be large and would not constitute a crisis. Investors overweight bonds would be most affected.