Trading halts and circuit breakers: when exchanges pause trading and why

When the S&P 500 falls more than 7% in a single day, trading on US exchanges automatically pauses for 15 minutes. When a single stock moves more than a defined percentage in a 5-minute window, the same kind of pause applies to that individual security. Trading halts and circuit breakers are the mechanisms exchanges use to prevent disorderly trading—and their design and history reveal the trade-offs between continuous price discovery and orderly market function.

What trading halts and circuit breakers are

A trading halt is a temporary pause in trading on an exchange or in a specific security. Halts can be triggered by defined volatility thresholds (the circuit-breaker case), by pending news announcements (the regulatory-halt case), or by exchange-specific operational issues (the technical-halt case). The duration of the halt can range from seconds (in the case of single-stock circuit breakers) to days (in the case of regulatory halts pending news).

Circuit breakers specifically refer to the volatility-triggered halts, which were introduced in the US after the October 1987 stock market crash. The original design—paused trading at defined market-wide drawdown levels—has evolved through several iterations and now operates at three levels: 7%, 13%, and 20% drawdowns from the prior day's close, with progressively longer halts at each level.

Single-stock circuit breakers (the Limit-Up Limit-Down framework in the US, introduced after the May 2010 flash crash) operate at the individual security level. A stock that moves more than its defined percentage band in a 5-minute window enters a 5-minute pause, with the band sized to the stock's typical volatility. The pause prevents the disorderly cascades that produced the flash-crash event and similar episodes.

How they work

Market-wide circuit breakers in the US work as follows. A 7% drop in the S&P 500 from the prior day's close triggers Level 1—a 15-minute pause in trading on all major US exchanges. A 13% drop triggers Level 2—another 15-minute pause. A 20% drop triggers Level 3—trading is suspended for the rest of the trading day. The mechanisms are coordinated across exchanges so that the pause applies uniformly.

Single-stock circuit breakers (Limit-Up Limit-Down, LULD) work differently. Each stock has a defined percentage band—typically 5% for liquid large-cap stocks, 10% for mid-cap, 20% for less-liquid names. If the stock's price moves outside the band in a 5-minute window, trading enters a Limit State: orders that would push the price further outside the band are not executed, and the stock can only trade back within the band. If the stock remains in Limit State for 15 seconds, a 5-minute pause is triggered.

Regulatory halts are different again. They are triggered by exchanges or regulators in response to pending material news—earnings announcements, merger announcements, regulatory enforcement actions, or other events that would produce disorderly trading if the market had not yet absorbed the information. The halt typically lasts until the news is released and the market has had time to digest it.

Other halt types include news halts (paused pending news the company has not yet released), volatility halts (above the LULD framework, paused for extreme volatility episodes), and operational halts (paused due to exchange-side technical issues).

What the evidence shows

Market-wide circuit breakers have been triggered very rarely. The 7% Level 1 was triggered for the first time in the modern framework on March 9, 2020, March 12, 2020, and March 16, 2020—three times in two weeks during the COVID drawdown. Before that, no Level 1 halt had occurred in the modern framework. The mechanism is therefore designed for genuinely extreme events rather than for routine volatility.

Single-stock circuit breakers have been triggered far more frequently. The LULD framework triggers thousands of pauses per year across the US equity market, mostly in less-liquid names that experience large price moves on small volume. The vast majority of pauses are 5 seconds in the Limit State and resolve without escalating to a full 5-minute pause.

The empirical literature on circuit-breaker effectiveness is mixed. Studies of post-pause trading typically find that volatility resumes at high levels for a period after the pause ends, suggesting that the pause may delay rather than prevent the underlying price discovery. The case for circuit breakers is therefore less about price-discovery improvement and more about preventing operational chaos—giving the exchanges and the participants time to assess what is happening and rebalance their positions in an orderly way.

Limitations and trade-offs

Circuit breakers can produce unintended consequences. The threat of a halt can cause disorderly trading just before the trigger as participants rush to execute before the pause; the resumption after the pause can produce the same disorderly trading in the opposite direction. The May 2010 flash crash itself was partly attributable to liquidity-providers withdrawing as they anticipated halt triggers.

For investors with stop-loss orders, circuit breakers complicate execution. A stop-loss triggered just below the halt threshold can execute at materially worse prices than the trigger level if the post-halt resumption opens with a large gap. The 2010 flash crash produced exactly this pattern for many retail stop-loss orders that triggered at unrealistic prices during the disorderly trading.

The trade-offs in circuit-breaker design are not fully resolved. The thresholds are calibrated for typical conditions and may be inappropriate in genuinely extreme regimes; the duration of the pauses is similarly calibrated and may be too short or too long depending on the specific event. Each major market disruption produces a debate about whether the existing framework needs adjustment, and the framework has been adjusted multiple times since the original 1988 introduction.

Trading halts in pfolio

Trading halts and circuit breakers are exchange-side mechanisms that affect trade execution, not portfolio analytics. pfolio's price series reflect the actual traded prices on each market day; days with halted trading produce gaps in the data that the platform handles via standard time-series interpolation.

Related articles

• Stock exchanges and trading venues: NYSE, NASDAQ, LSE, and how exchanges differ
• ETF liquidity explained: bid-ask spreads, premium, and discount to NAV
• Market makers and liquidity providers: who stands between the buyer and the seller
• Order types in investing: market, limit, stop, and how each affects execution