Tim Bourgaize Murray: Revisiting Circuit Breakers in the Wake of Black Monday

Like any journalist sitting by as global markets quake—as they did on August 24th—one of my favorite pastimes is to wait for the first fintech analysis and commentary to reach my inbox. The “fastest-to-email” gold star this time went to Tabb Group at 2:18 p.m. EST. Analyst Luther Zhao pointed out that the circuit breakers put in place for CME’s equity index futures after the 2010 Flash Crash had been triggered multiple times, causing several trading “time outs” once prices had dropped 5 percent (overnight, pre-open) and 7 percent, respectively.

The same happened for the Dow and Nasdaq under their procedures. The Wall Street Journal reported that more than 1,200 stocks and exchange-traded funds (ETFs) were paused at some point under similar exchange rules. It’s like an entire apartment building had their vacuums running at once—with all of them plugged into the same outlet.

Zhao wrote that authorities would want to revisit these mechanisms once markets stabilize, and that the circuit breakers appeared to do little to staunch a steep decline in prices as liquidity disappeared in these indexes before and during the hour after markets opened.

So far as technology problems go, circuit breakers are pretty easy—on a technical level, I haven’t seen or heard any complaints about their activation in this instance. The debate here, especially for the buy side, is more philosophical: how much intervention is appropriate? And is it even effective?

To start, let’s go to the extreme. Some say China’s aggressive intervention into its markets recently has shaken foreign investor confidence in the country’s willingness to let a free market do its thing. Some say this worsened Monday’s sell-off. Even if the Chinese government’s intervention was improvised, rather than automated, and driven by larger economic issues, it looks and acts a lot like a circuit breaker—lasting longer and enacted more haphazardly, yes, but the same basic theory applies: Don’t let people flee the market as quickly as they’d like to.

It’s like an entire apartment building had their vacuums running at once—with all of them plugged into the same outlet.

Halting trading, whether for five minutes or hours or days, doesn’t always have the desired effect of stability. And we’ve seen that argument regularly made in the past and closer to home, too—after the 2012 Knight Capital meltdown, for example.

Putting in place aggressive triggers that essentially turn off trading in certain names, even briefly, or shut down certain market-makers’ systems can make things worse in instances when a quick reaction could have solved the problem. Leaving the rest of the market in the dark while it gets addressed (or leaving the market-maker out there like a sitting duck) is never ideal.

Trading strategies at hedge funds and elsewhere may well be betting substantial sums, in good faith, that a price will naturally plunge below a certain threshold, by playing in options, for example. With the proliferation of binary options lately, more individual investors might be doing the same, too.

Automatically locked-in pauses at certain percentage loss floors are probably part of their calculus these days (or ought to be), but should they need to be? Does a trading day look different when it’s missing five or 10 minutes? Of course it does. Again, what does a “free market” really look like in 2015?

Further Research
Zhao noted that more research is warranted regarding the breakers’ effectiveness—especially given the high number of listings that were tripped up.

A major market correction in China had been suggested for weeks, but I’m more curious about the short bets that have been lost because of pauses kicking in at various points, or indeed the ways that this could potentially happen more frequently if stronger controls were introduced later on. Perhaps it’s easy to say that on a day when the US markets were only knocked a few percentage points down, despite the greater tumult going on to the east. Perhaps breakers would also begin looking more useful (and possibly effective) next time we get to day three or four of a global sell-off.

Just the same, circuit breakers go to the bigger question of what we really want out of our exchanges—efficiency or stability? What’s a reasonable balance to strike when markets halfway around the world acting rationally—rather than a software glitch—are the cause? I’m not sure we’ve found it.