Market Volatility Driven by Algorithms, Not Fundamentals, Says Wood

Stock News
Feb 16

Amid rising debates over AI capital expenditures, Cathie Wood, CEO and CIO of ARK Invest, attributes the sharp swings in the U.S. stock market to a chain reaction triggered by algorithmic selling. In her February 14 video segment, Wood stated that recent market turbulence is largely driven by programmatic trading rather than fundamental shifts of similar magnitude. She opened the program by saying, "Most of this volatility is manufactured by algorithmic trading. Algorithms don't conduct research like we do." Wood noted that such fluctuations can "scare people" but also create pricing errors. "During the tariff turmoil last April, many panicked. Those who sold then regretted it for the rest of the year," she recalled. She described the current market as "climbing a wall of worry," adding that such markets often emerge stronger.

Wood explained that the "algorithms" she referred to are not designed to assess corporate cash flows or competitive landscapes but mechanically adjust risk exposures based on predefined rules. She summarized recent trading patterns as "sell first, ask questions later." From a trading mechanism perspective, programmatic strategies are often triggered by factors such as price trends, volatility, correlations, and risk budget allocations. When prices fall or volatility rises, models automatically reduce exposure to risky assets to meet predetermined drawdown or volatility targets. This selling, in turn, further amplifies volatility and correlations, triggering additional algorithmic sell-offs and forming a "feedback loop." In crowded, homogeneously positioned sectors, this chain reaction can drag down both high-quality and low-quality companies, akin to "throwing the baby out with the bathwater."

Wood also highlighted another amplifier: the growing dominance of technically driven trading psychology. "Many people now rely solely on technical analysis," she observed. The more traders focus on the same moving averages or "key levels," the more likely it is to trigger herd-like, unidirectional trading.

Addressing the recent sharp swings in tech stocks, particularly software shares, Wood argued that the market is undergoing a technological transformation from a one-size-fits-all SaaS model toward highly customized AI agent platforms. While traditional SaaS companies face inevitable pressure during this transition, she believes the market has overreacted. "Anyone selling at this moment will regret it," Wood asserted. She elaborated on the mechanism's failure: when the market perceives slowing growth in the SaaS sector, algorithmic trading tends to execute indiscriminate sell orders. Machines cannot distinguish between companies successfully transitioning to AI platforms and those likely to be phased out. This mispricing, stemming from algorithms' lack of deep fundamental research, presents opportunities for active investors. "That's why we concentrate our positions in our highest-conviction stocks. The market is giving us this chance," Wood said.

Regarding concerns over aggressive capital spending by the "Magnificent Seven" tech giants eroding cash flows, Wood holds a contrary view. She recalled the dot-com bubble era, noting that the current environment resembles 1996—the early stages of the internet revolution—rather than the 1999 peak. "If you lived through the tech and telecom bubble, today's conditions are much healthier," she remarked. Wood illustrated the difference with a vivid comparison: at the peak of the dot-com bubble, Jeff Bezos could announce, "We will lose more money as we invest aggressively," and Amazon's stock would surge 10% to 15%. In contrast, today, when the "Magnificent Six" signal increased capital expenditures, the market penalizes them, with shares falling instead of rising. Wood interprets this as a sign that investors are not in a state of irrational exuberance but are instead filled with fear and skepticism. "The market is climbing a wall of worry, which typically forms the strongest foundation for a long-term bull market, not a precursor to a bubble burst," she explained.

Wood also extended her analysis to macroeconomics, suggesting that AI-driven productivity gains could颠覆 the traditional narrative that growth inevitably fuels inflation. She projected that productivity improvements would reduce the fiscal deficit as a percentage of GDP and claimed that the U.S. could achieve a surplus by the end of the current presidential term (around late 2028 or early 2029). She even predicted global real GDP growth of 7% to 8% by the end of the decade, calling the estimate "possibly conservative." Wood repeatedly emphasized that "growth does not equal inflation." In her framework, AI-driven real growth is more likely to suppress inflation through productivity gains than to push it higher.

On the topic of consumer sentiment, Wood acknowledged that consumers "are not happy," primarily due to genuine weakness in the job market and a housing affordability crisis. "Last year's employment figures were revised down by 861,000, equivalent to a reduction of about 75,000 jobs per month," she noted, explaining the disconnect between consumer sentiment and GDP data. However, she saw a silver lining in youth unemployment data. While the jobless rate for 16- to 24-year-olds had spiked, it has recently fallen below 10%. Wood suggested this reflects not just employment recovery but potentially an "entrepreneurial explosion" empowered by AI. "AI has become so powerful that individuals can now go out and start businesses directly," she said, predicting a surge of efficient startups driven by individuals or small teams as AI tools become more accessible.

In conclusion, Wood contrasted the current environment with the dot-com bubble, emphasizing that today's opportunities are genuine. "During the tech bubble, there was extreme speculation. Now, people are terrified," she said. As a portfolio manager in the innovation space, she prefers the current climate of fear and "climbing a wall of worry" over the speculative excesses of the past. While some may argue that AI is experiencing a bubble, Wood believes the market is in a phase similar to 1996—very early in the technology revolution. She pointed out that today's investors carry "scar tissue" from the 2000 crash, making them cautious and fearful, which in turn sustains the wall of worry. Market volatility may be uncomfortable, but like last April, it could present significant opportunities to invest in "the next big thing."

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Most Discussed

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10