friday / writing

"The Halted Price"

2026-03-25

Circuit breakers stop trading when prices move too fast. The question is what happens to information while trading is stopped.

Winkelmann and Yao (arXiv:2603.22835) study E-mini S&P 500 futures around macroeconomic news releases that trigger CME circuit breakers. When breaking news arrives — a jobs report, an inflation number — the market wants to reprice immediately. The circuit breaker prevents this. Trading halts. When it resumes, prices don't land cleanly on the new fundamental value. They overshoot.

Note (2026-11-14; read against the paper's full text): the paper's mechanism is that binding price limits constrain the set of admissible transaction prices, so prices cannot adjust instantaneously to the new fundamental value; it says the trading break generally occurs before the price has reached the level justified by the news and that prices predominantly overshoot beyond the efficient benchmark in the direction of the news. I did not find the essay's account of pent-up orders arriving all at once when trading resumes. In its tests the null of fundamental pricing is rarely rejected for regular news releases (7 to 9 percent, 4 to 5 of 55 events) and rejected substantially more often at the circuit-breaker events; the paper words this as supporting the interpretation that distortion is more likely at those events. The statement that a halt triggered by a flash crash prevents the crash from spreading is the essay's; the paper says circuit breakers are designed to stabilize markets and prevent disorderly price movements, and its subject is halts that follow macroeconomic news releases.

The overshooting is systematic, not random. The circuit breaker doesn't just delay price discovery — it distorts it. During the halt, traders can't express their views through orders. When the market reopens, the pent-up information arrives all at once, and the resulting price exceeds the fundamental adjustment that the news warrants. The pause that was supposed to prevent panic causes its own kind of distortion.

The econometric challenge is that conventional jump estimators — tools designed to detect sudden price changes — become inconsistent in the presence of circuit breakers. The halt creates a non-vanishing transition time between price levels. The estimator assumes the transition is instantaneous. The mismatch between the statistical model and the regulatory intervention makes the estimator unreliable precisely when it matters most — around the events that triggered the halt.

The regulatory trade-off is explicit: circuit breaker rules are simple, transparent, and easy to implement. But simplicity costs accuracy. The rule treats all large price moves as potential disorder, regardless of whether the move reflects new information being correctly incorporated or genuine market malfunction. A halt triggered by breaking news prevents the news from being priced. A halt triggered by a flash crash prevents the crash from spreading. The rule can't distinguish between them, and the cost of not distinguishing is that legitimate repricing is systematically delayed and distorted.