Summary of Key Points
Since July, the A-share technology sector has experienced a rollercoaster market pattern of "limit-downs → rebounds → further declines → more rebounds." The market initially attributed these fluctuations to quantitative trading. However, the real cause was a chain reaction resulting from the overcrowding of stocks in the global AI industry chain at high prices. The upward trend was driven by AI capital expenditures from overseas giants, a supercycle in memory chips, and profit-making effects, while the downward trend was triggered by factors such as Meta selling off idle computing power (shattering the narrative of a shortage of computing resources) and rising macroeconomic interest rates. Quantitative trading is not the "engine" of these fluctuations; rather, it acts as an "amplifier." Under normal circumstances, it provides liquidity by helping buyers and sellers find each other. But in extreme market conditions, due to the similarity of algorithm models and risk control rules, it can cause simultaneous reductions in positions, compressing what would otherwise be several weeks of adjustment into just a few days and intensifying the volatility.
Detailed Analysis
1. Global Tech Stock Decline: Is Quantitative Trading to Blame? Actually, It's Due to Overcrowding at High Prices
The recent decline in tech stocks was not unique to the A-share market; stocks in the AI industry chain on the Korean and US stock markets also saw similar adjustments. The root cause lies in the fact that prices had risen too sharply and the distribution of shares became too concentrated:
- Upward Trend: Overseas giants (such as Meta and Microsoft) invested heavily in AI infrastructure, leading to a shortage of memory chips (Samsung's profits increased by 18 times), attracting more funds into the market (the A-share memory chip index rose by over 165%).
- Downward Trigger: Meta announced the sale of idle computing power, suggesting that AI infrastructure might no longer be in such high demand; SK Hynix' profits fell short of expectations; rising US interest rates hit tech stocks with high valuations even harder.
- The A-share Market Was More Affected: The technology sector accounts for nearly half of trading volume, and shares are highly concentrated. Even minor disruptions prompted investors to seek profit-taking opportunities, and quantitative algorithms exacerbated this by simultaneously reducing positions, further compressing the adjustment period.
In simple terms, it wasn't quantitative trading that caused the decline; rather, many investors wanted to sell at high prices, and quantitative trading simply made these sales more concentrated.
2. What Role Does Quantitative Trading Play in the Decline? Not the "Criminal," But Definitely an "Accelerator"
There are claims that quantitative trading drove the market down, but data and logic refute this:
- The Trigger Wasn't Quantitative Trading: The simultaneous decline in global tech stocks indicates a sector-wide adjustment, not the action of a particular type of trader.
- Large Sales Orders Came from Institutional Position Adjustments: On July 7, the electronics sector saw a net outflow of 10.3 billion yuan, with individual chip leaders losing between 2 and 2.8 billion yuan. Such volumes are more consistent with large institutional investors (such as public funds) selling their positions, not quantitative trading (which accounts for only 2%-3% of market trades).
- Quantitative Trading as an "Accelerator": Many quantitative institutions use similar algorithms (e.g., analyzing stock prices and market sentiment) and risk control rules (reducing positions when losses reach a certain level). When stock prices fall, these algorithms trigger simultaneous sales, compressing weeks-long adjustments into just a few days, creating the impression of widespread limit-downs.
For example, on July 13, 187 stocks hit limit-downs because quantitative algorithms simultaneously reduced positions, intensifying the decline. Conversely, on July 9, the Sci-Tech 50 index rose by 8.41%, which was also partly due to quantitative trading programs buying into the market.
3. What's the Debate About Quantitative Trading? Is It a Stabilizer or a Factor That Fuels Volatility?
Critics argue that quantitative trading amplifies volatility:
- Although its share of market trades is low (2%-3%), it accounts for 30%-40% of trading volume, especially in smaller stocks, giving it significant influence on pricing.
- Similar algorithms and risk control rules cause a "resonance" effect: When prices fall, all traders selling at the same time depletes liquidity, dragging down even well-performing stocks.
Defenders argue that quantitative trading is a stabilizer:
- Quantitative trading doesn't have emotional biases and thus doesn't contribute to panic-driven sell-offs.
- It has "reversal strategies" (automatically buying when prices fall), which can slow down the decline.
- High turnover provides liquidity: When no buyers are available, quantitative trading programs act as buyers, allowing sellers to execute their orders.
Intermediate Conclusion: Under normal circumstances, quantitative trading stabilizes the market by providing liquidity and correcting price distortions. However, in extreme conditions, its risk control rules can suppress buying signals, leading to simultaneous reductions in positions and amplifying volatility. This is not a flaw of quantitative trading itself but a "phase transition" of the system (similar to water freezing under extreme temperatures).
4. How Should Regulation Address This Issue? Learn from International Practices
Regulation should not aim to ban quantitative trading outright but to address the problem of amplified volatility in extreme situations. International approaches include:
- Regulating Behavior, Not Identity: For example, identifying and punishing deceptive practices (such as quickly canceling orders to create fake demand or flooding the market with large numbers of cancelations). The US uses "large-trader reports" and order tracking systems to monitor traders' activities; the EU requires algorithmic market makers to maintain liquidity during market downturns.
- Regulating Externalities, Not Just Returns: Quantitative institutions often invest heavily in upgrading their systems (e.g., reducing transaction delays from 1 millisecond to 0.1 millisecond) to gain a competitive advantage. However, when everyone does the same, the advantage disappears, leading to a "arms race" that wastes resources. A "cancelation fee" (higher fees for more cancellations) can encourage institutions to reduce unnecessary transactions and allocate funds to research.
- Using "Breakers" in Extreme Situations: For instance, implementing a brief pause (e.g., a few minutes) when a stock drops by 7%-8% to allow for re-pricing and prevent further panic.
The A-share market could benefit from adopting these measures: identifying related accounts to prevent manipulation, implementing tiered cancelation fees, and setting different cooling periods for different levels of price changes.
Final Thoughts
Quantitative trading itself is neither good nor bad; it's a tool. Under normal circumstances, it makes markets more efficient. However, in extreme situations, its homogenization can amplify volatility. The goal of regulation should be to ensure that quantitative trading provides liquidity while keeping risks within acceptable limits—just like a car needs brakes and speed limits. For investors, understanding the underlying causes of market fluctuations (such as overcrowding at high prices and global synchronization) is more important than focusing on identifying the "culprit."