Palo Alto is a great town.
While there sponsoring IR Magazine’s West Coast Think Tank last week we feasted at Evvia and Fuki Sushi. Denver’s got fine sushi. Our Sushi Den on South Pearl Street flat demoralizes Bryant Park’s Koi. Proprietors Toshi and Yasu Kizaki each day fly in hand-picked selections from the Tokyo fish market. You gotta get up pretty early to beat our fish. Fuki Sushi apparently rises at dawn. We ate to dullness.
Speaking of lulled, exchanges began introducing new SEC-approved Limit-Up/Limit-Down (LULD) single-stock circuit breakers Monday, smartly easing the program into effect. More will be added until the largest 2,000 are covered by late May and the rest of the market through August.
“It sounds simple, but for firms managing thousands of customer orders, you have to program how you’ll manage them, how you’ll deal with quotes and trades across 50 destinations, routing decisions and execution quality,” Chris Concannon, partner at high-frequency trading firm Virtu Financial, told Bloomberg reporter Nina Mehta.
Under LULD, stocks won’t be permitted to trade more than a certain percentage from their rolling five-minute average prices. The SEC mandated these changes after the Flash Crash of May 6, 2010, sent the S&P 500 plunging over a hundred points and the Dow Industrials a thousand points, before both rebounded, all in roughly twenty minutes.Also getting facelifts are market-wide halts in place since the 1987 crash, when markets dropped 22%. These weren’t triggered in 2010 and indeed have only been tripped once, Oct 27, 1997, suggesting the bands were too elastic. Now, all U.S. equity-trading will stop for 15 minutes if the S&P 500 falls 7% in a particular span, or 13% before 3:25p ET. Down 20%, and trading closes for the day.
I’m unclear how that helps. But anyway, what’s LULD mean to IROs tracking market structure for company shares? More uniformity in behavior. There are 4,300 brokers, but only 185 (4%) are executing trades at any given time, our data indicate. In the past 20 days a sample of executions shows a dozen large firms handling 60% of all volume.
The National Market System hosts 3,600 issuers (more brokers than issuers!).Yet there are hundreds of thousands of index products – 850,000 from S&P/Dow Jones alone, along with 175,000 more from Russell, MSCI and the NASDAQ – some portion of which tie directly or indirectly to US equities and which calculate continuously in increments as small as every second. So every equity we track – even one that traded just 174 shares the other day – relies in large part on algorithmic executions, generally for baskets.
These rules are well intentioned. If you don’t know what load a structure can bear, strap on guy wires and meter traffic through it. But that doesn’t exactly inspire confidence. What’s more, layering on requirements for algorithms running vast seas of disparate executions tied to baskets, indexes and derivatives means but a handful can comply and still meet best-execution standards. The entire market will be more dependent on how well algorithms work at Too-Big-To-Fail banks behind 60% of equity trading.
It’s ironic. And how does one square potentially larger tick-sizes for small-caps with a market that won’t let stocks trade outside a tiny standard deviation? No wonder money prefers private equity where years-long lockups abolish arbitrage, and function is free.
Lest ye be disheartened, it’s more reason to know your market structure. How these rules affect investors and traders will be evident in the evolution of behavior. I can’t help but think of Hudson River Trading: “Hudson River Trading (HRT) develops automated trading algorithms using advanced mathematical and statistical modeling techniques and an extremely high-performance computing environment.”
HRT will be studying change with statistical models to trade more effectively. And we’ll be studying them with statistical models so you can understand how Speculators adapt. You, public companies, are equal market constituents with HRT, entitled to know what drives your share-price (something exchanges seem to have forgotten). In fact public companies are more equal as there would be nothing to model without them.
So take heart. This adds to the thrill of the IR job. In my IR-chair days when it was uphill both directions and we were barefoot, all we could do was go tell the story to investors. That’s so yesterday.