Shear-Thickening Markets: State-Dependent Liquidity Frictions and Endogenous Tail Risk
This paper proposes a state-dependent liquidity framework where transaction costs rise only when aggressive order flow exceeds durable market depth, demonstrating through historical evidence and simulations that such adaptive, convex pricing mechanisms can effectively mitigate endogenous tail risk and transitory price errors compared to flat fees.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the world of modern finance, electronic markets operate like vast, high-speed networks where buyers and sellers meet to trade stocks. Under normal conditions, these systems work with remarkable efficiency: prices adjust quickly to new information, and there is usually enough interest from both sides to absorb trades without wild swings. However, the system relies on a delicate balance between the speed of incoming orders and the availability of "depth," which is simply the amount of stock sitting in the queue ready to be bought or sold. When a sudden wave of orders arrives in one direction—say, a massive rush to sell—the available depth can be consumed faster than it can be replaced. This creates a vacuum where prices can overshoot their true value, not because the news is that bad, but because the market's ability to absorb the shock has temporarily collapsed. This phenomenon is at the heart of a new study that asks a critical question: should the rules of the market change depending on how stressed the system is?
The research, led by Max Fomitchev-Zamilov, investigates whether financial markets have become more prone to these sudden, short-term crashes over time, and if so, how we might fix them without slowing down normal trading. By analyzing nearly a century of daily market data, the author found a striking pattern: while the overall volatility of the stock market has not changed much over the last few decades when looking at yearly returns, the day-to-day swings have become significantly more extreme. Specifically, the study shows that the volatility of daily returns has risen by 72 percent since the mid-1950s, while five-day and twenty-day volatility have also increased, though less sharply. Most importantly, the data reveals that a tiny fraction of days—just 3.33 percent of all trading days—account for nearly half of all the price movement squared. In other words, the market is now dominated by rare, intense bursts of activity that cause prices to spike or crash, only to reverse course shortly after. These are not necessarily signs of bad news, but rather signs of a market structure where liquidity dries up too quickly under pressure.
To understand why this happens, the paper introduces a concept called "durable depth." In a standard view, a market looks deep if there are many orders sitting in the queue. But the study argues that not all orders are equal. Some orders are "durable," meaning they are likely to stay in the queue and actually get filled. Others are "fleeting," likely to be cancelled the moment the market moves slightly against them. When a crisis hits, these fleeting orders vanish instantly, leaving the market with far less real capacity to absorb the shock than it appeared to have. The author suggests that the current market structure allows aggressive traders to consume this fragile depth so quickly that prices are pushed too far, too fast, creating a temporary "overshoot" before the market corrects itself.
The paper proposes a solution that acts like a traffic light for the market, but one that only turns red when traffic is moving dangerously fast in one direction. Instead of a flat tax on every trade, which would slow down normal business, the author suggests a "state-dependent" fee. This fee would be zero when the market is calm and the ratio of incoming orders to available depth is low. However, as the ratio of aggressive orders to reliable depth rises—indicating that the market is being overwhelmed—the fee would increase smoothly and sharply. This extra cost would only apply to the trades that are actually consuming the scarce, fragile liquidity. Crucially, the money collected from these fees would be returned, or "rebated," to the traders who kept their orders in the queue during the stress, rewarding them for providing the stability the market needed.
The study tests this idea using computer simulations that mimic how markets behave under stress. In these simulations, the researchers compared a system with no rules, a system with a flat fee on all trades, and the new adaptive system. The results showed that the adaptive system was far more effective at preventing extreme price swings and keeping the market from running out of liquidity. While a flat fee reduced some trading activity even when the market was calm, the adaptive fee only kicked in when necessary, leaving normal trading untouched. Furthermore, the system with the rebate was particularly good at preventing the market depth from disappearing entirely, which is a key factor in stopping a crash from becoming a disaster. The simulations also identified a "failure region," showing that if the fee is set too high or if traders are too sensitive to the cost, the system could actually make things worse by driving liquidity away. This suggests that the design must be carefully calibrated and tested before being used in the real world.
The author is careful to note that this is a proposal for a new way to think about market rules, not a proven solution that has already been implemented. The historical data used to motivate the idea is descriptive, showing what happened in the past but not proving exactly why. The positive results come from computer models, which are useful for testing logic but cannot perfectly replicate the complex behavior of real human traders. The paper explicitly rules out the idea that a simple, uniform tax on all trades is the right answer, arguing that such a blunt instrument would hurt market quality without solving the specific problem of sudden liquidity shortages. Instead, the research points toward a more nuanced approach where the market's rules adapt to its own condition, adding resistance only when the flow of orders threatens to overwhelm the system's ability to cope.
Ultimately, the paper offers a vision of a market that is resilient rather than rigid. It suggests that we do not need to stop the rapid flow of information or the speed of trading to prevent crashes. Instead, we can design rules that recognize when the market is under stress and apply a gentle, targeted brake to prevent prices from spiraling out of control. By distinguishing between normal trading and the specific conditions that lead to a liquidity crisis, and by rewarding those who provide stability during those times, it may be possible to reduce the frequency and severity of those rare, extreme days that dominate the market's history. The study concludes that while the technology of trading has accelerated, the fundamental need for a market to have enough reliable depth to absorb shocks remains, and our rules should evolve to protect that depth when it is most needed.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.