Highlighting fairness: High-frequency quantitative trading regulation enters the year of "deepening and refinement"

In 2026, new regulatory measures are expected to be rolled out for high-frequency quantitative trading.

During this year’s two sessions (National People’s Congress and Chinese People’s Political Consultative Conference), Wu Qing, Chairman of the CSRC, clearly stated that in 2026, China will “highlight the principle of fairness, further refine and deepen regulation of high-frequency quantitative trading.”

Experts and practitioners interviewed by the “China Business News” (《中国经营报》) said that high-frequency quantitative trading has a distinct “double-edged sword” characteristic. To promote what is beneficial and eliminate what is harmful, differentiated and category-based regulation should be implemented, with upgrades to regulatory technology and penetrative, end-to-end supervision.

On April 3 last year, the Shanghai, Shenzhen, and Beijing exchanges released the “Detailed Rules for the Implementation of Programmed Trading Management” (hereinafter referred to as the “Detailed Rules”), on the same day, making clear provisions for the management of high-frequency trading.

Interviewees said that, since the release of the “Detailed Rules” one year ago, the A-share market’s high-frequency quantitative trading has shown core changes characterized by “rational contraction of scale, significantly standardized behavior, and accelerated industry reshuffling.” For small and medium-sized investors, the trading environment has become more equitable, and the probability of being “snatched orders, intercepted deals, and deceived” has declined.

What has changed for the better in high-frequency quantitative trading?

On April 3, 2026, the Shanghai, Shenzhen, and Beijing exchanges published the “Detailed Rules” upon completion of one year since their issuance.

The “Detailed Rules” set up a dedicated chapter on “High-Frequency Trading Management.” It deems trading to be high-frequency when the maximum number of order submissions and cancellations per second from a single account reaches more than 300, or when the maximum number of daily order submissions and cancellations reaches more than 20k. At the same time, it makes differentiated regulatory arrangements for high-frequency trading relative to other types of programmed trading, including additional reporting requirements, stricter management of abnormal trading behavior, and the implementation of differentiated fee standards.

“Since the ‘Detailed Rules’ were issued one year ago, high-frequency quantitative trading in the A-share market has shown the core changes of ‘rational contraction of scale, significantly standardized behavior, and accelerated industry reshuffling.’” Bound Qi (束其全), General Manager of Shanghai Qianbo Asset Management Co., Ltd., told reporters that the share of high-frequency quantitative trading across the entire market, which was around 30% before the release, has fallen back to around 20% or so.

Hu Conghui, Vice Dean of the School of Economics and Business Administration at Beijing Normal University, also observed that pure high-frequency, intraday round-trip, and millisecond-level order-snatching strategies have clearly cooled off. Institutions have proactively lowered the frequency of submissions/cancellations, staying away from the high-frequency recognition thresholds of 300 orders per second and 20k orders per day. Meanwhile, high-frequency cancellations and “flickering orders” with short lifespans have noticeably decreased, and manipulative behaviors such as fake order placement and deception have been effectively curbed.

“For small and medium-sized investors, the trading environment is fairer, and the probability of being ‘snatched orders, intercepted deals, and deceived’ has decreased. Trading execution quality has improved.” Hu Conghui said.

Luo Ronghua, Dean of the School of Finance at Southwestern University of Finance and Economics and President of the China Finance Research Institute, meanwhile, believes that the “Detailed Rules” have improved the standardization and transparency of trading behavior. The scale of ultra-high-frequency strategies has shrunk and shifted toward low- to medium-frequency and fundamental-data quantitative approaches, but the underlying logic of quantitative trading as a scientific investment method has not changed. “Once the ‘speed premium’ fades, it becomes a necessity to dig deeper into unstructured data through more frontier technology upgrades (such as machine learning and natural language processing) to obtain excess returns relative to low- to medium-frequency strategies.” Luo Ronghua said.

Luo Ronghua also said that in the A-share market, some leading quantitative private funds are also investing huge sums to build multi-agent investment research and decision platforms covering the full process of “data—research—monitoring—attribution,” attempting to establish new “moats” by capturing cognitive alphas such as “managements’ expectation gaps” embedded in massive announcements and market sentiment. However, for small and medium-sized quantitative institutions and day traders, the pressure to survive and transform has increased sharply.

Bound Qi also said that leading quantitative institutions rely on comprehensive compliance systems, strong technical reserves, and diversified strategy layouts. By adjusting trading parameters and optimizing strategy logic, they can quickly adapt to regulatory requirements, which further concentrates market share. Compliant high-frequency strategies can still achieve stable returns.

Meanwhile, small and medium-sized high-frequency quantitative institutions, due to insufficient capability for technological iteration, an incomplete risk-control system, and compliance costs under pressure, most choose to scale back high-frequency businesses or transition to low- to medium-frequency quantitative strategies. Some small institutions even directly exit the market.

What are the pros and cons of the “double-edged sword”?

In the view of the interviewees above, high-frequency quantitative trading is a “double-edged sword” for all stakeholders in the capital market, with both advantages and disadvantages that are very clear.

“In normal market conditions, they narrow the bid-ask spread through high-frequency two-sided quoting, making them liquidity providers; but in an environment of extreme volatility or liquidity depletion, highly homogeneous high-frequency strategies may produce ‘resonance’ when triggered by risk-control models—leading to a collective instant cancellation of orders. The so-called ‘fake liquidity’ could instead aggravate the market’s pro-cyclical volatility.” Luo Ronghua said bluntly.

“The share of market-making-type quantitative trading on A-shares is low, and order-snatching-type strategies dominate. A-share algorithmic trading is more likely to raise the bid-ask spread and increase retail investors’ trading costs. This negative effect is more pronounced for small-cap stocks, low-priced stocks, and stocks with a poorer information environment.” Hu Conghui emphasized that speed and information advantages create a “one-dimension-reduction attack,” placing ordinary investors in an unfavorable position.

Bound Qi said that liquidity convenience brought by high-frequency trading can make it easier for small and medium-sized investors to match orders when buying and selling stocks, and can reduce transaction costs for small trades. “But the downsides are even more prominent. The rapid order submission and cancellation of high-frequency trading can create an atmosphere of fake transactions. The intense short-term price fluctuations can cause small and medium-sized investors to form misjudgments. Against a backdrop of information asymmetry and technical gaps, small and medium-sized investors are likely to become the ‘bag-holder’ for high-frequency trading.” Bound Qi said.

Luo Ronghua also said bluntly that small and medium-sized investors are the absolute victims of “information and technology asymmetry.” Retail investors’ manual order placement faces a natural delay disadvantage in front of microsecond-level algorithms. However, high-frequency strategies can detect even slight imbalances in the order book, sensitively capture retail investors’ trading inclinations, and swiftly adjust quotes.

For large institutional investors such as public funds and insurance capital, Luo Ronghua said that large funds themselves can deeply use programmed trading, which helps effectively reduce the instantaneous price impact cost when large capital flows into and out of the market. But large institutions can easily become targets in a game of predatory high-frequency trading.

“Some high-frequency sniffing strategies infer the direction of large institutions’ building or closing positions from short-term order-book characteristics, using their speed advantage to grab positions earlier, and then provide liquidity to institutions at a slightly higher price in the opposite direction.” Luo Ronghua said. “This, in turn, unconsciously raises the friction cost for long-term allocation capital.”

For quantitative institutions themselves, it is not the case that they only get benefits and never bear costs.

Luo Ronghua believes that quantitative institutions can enjoy excess returns driven by technological barriers, but at the same time they may inevitably fall into a “hardware arms race” with increasing marginal investment, such as racing to secure the fastest network lines and developing low-latency FPGA chips.

“As peer competition intensifies, the sunk costs of maintaining a microsecond-level advantage will rise exponentially. At the same time, due to the externalities of their trading behavior, quantitative institutions are facing increasingly severe risks to their social reputation and policy compliance pressure.” Luo Ronghua said.

“Strategy convergence among high-frequency traders and speed-based competition can also intensify market volatility. In extreme market conditions, it can easily trigger liquidity resonance. Therefore, even for high-frequency traders themselves, it is a kind of malignant competition.” Hu Conghui said directly.

Bound Qi believed that leading institutions can achieve stable spread/price-difference returns in high-frequency trading by leveraging technical, capital, and compliance advantages, and can also improve market pricing through完善ing their trading behavior. But homogeneous high-frequency competition among small and medium-sized institutions can trigger “involution” within the industry. Some institutions, in pursuit of returns, may touch regulatory red lines, not only facing their own punishment risks, but also further destabilizing the industry.

In Bound Qi’s view, for discretionary and low- to medium-frequency strategy private funds, reasonable high-frequency trading can enhance market liquidity, making it easier for them to rebalance holdings and switch stocks in response to changes in fundamentals, and to reduce strategy execution costs. But excessive high-frequency opportunities can magnify short-term market volatility, disrupt the pace of restoring stock valuations, and increase the difficulty of assessing the market’s short-term sentiment for strategies. It can even lead to high-quality targets being mistakenly sold off due to short-term price distortions, affecting the long-term stability of strategy returns.

How to highlight the principle of fairness?

“Highlight the principle of fairness, and further refine and deepen regulation of high-frequency quantitative trading.” On March 6, during the two sessions, Wu Qing said that in 2026, efforts should focus on strengthening regulation of new business models. The overall considerations are to pursue beneficial outcomes while avoiding harm, regulate development in a standardized way, ensure effective supervision, and strictly control risks.

“An approach could be to explore establishing differentiated fee mechanisms. For liquidity-supplying orders such as long-limit orders, stable market-making, and orders with low cancellation rates, lower fees and incentives should be offered. For liquidity-consuming behaviors such as high-frequency cancellations, short lifespans, and flickering quotes, higher fees should be imposed, and limits on cancellation rates and minimum resting time should be set to suppress fake liquidity.” Hu Conghui said.

Hu Conghui also suggested that for stocks with a high degree of information asymmetry and low levels of institutional participation, monitoring and standardizing algorithmic trading behavior could be strengthened to enhance protection for small and medium-sized investors. For blue-chip stocks with abundant liquidity and a high degree of institutionalization, more market-oriented space could be granted under the premise that risks are controllable, so that algorithmic trading can play a more positive role in improving market pricing efficiency.

Regarding upgrading regulatory technology to enhance penetrative regulatory capabilities, Luo Ronghua believes that in the face of increasingly complex trading algorithms, regulatory authorities need to build real-time, per-order, whole-network monitoring systems that are superior to the hardware level of quantitative institutions. By introducing AI large-model and graph network analysis technologies, regulators should strengthen their ability to penetrate and identify abnormal trading behaviors, accurately locking onto new market manipulation behaviors such as high-frequency coordinated “hitting the limit” orders, fake submissions, and more—hidden behind links among multiple accounts and multiple products.

Luo Ronghua also said that regulators should improve the level of interconnection and information sharing of data, and strictly prevent risks from external leverage from spilling over. For the capital side of quantitative trading, it is necessary to focus on conducting penetrative reviews of business models that effectively amplify high-frequency trading leverage through off-exchange derivatives.

To better highlight the principle of fairness, Luo Ronghua believes that besides standardizing the allocation of trading resources and eliminating hardware and channel privileges, a stricter mechanism for compensation and a reversal of the burden of proof should also be established. For market abnormal fluctuations caused by major defects or “mishaps” in programmed trading systems, a more stringent accountability and remedy system must be in place. At the same time, for small and medium-sized investors who are damaged due to follow-the-herd trading or sudden liquidity exhaustion, civil compensation channels should be further broadened and made more accessible to truly protect the lawful rights and interests of vulnerable groups.

“High-frequency quantitative trading itself is not a ‘flood of monsters and beasts.’ Its positive value to market liquidity and pricing efficiency is worth affirming. But it must be regulated and developed under a strict and fine-grained regulatory framework.” Bound Qi emphasized that only by highlighting the principle of fairness, deepening and refining regulation of high-frequency quantitative trading, can the regulatory goal of “promoting what is beneficial while avoiding what is harmful, and ensuring standardized development” be achieved—so that high-frequency quantitative trading becomes a tool for improving the efficiency of the capital market, rather than a factor that harms small and medium-sized investors’ interests or disrupts market order.

（Source: China Business Network (中国经营网)）