Locklin on science

Da plunge

Posted in finance journalism, microstructure, systematic trading by Scott Locklin on May 14, 2010

So, I don’t claim to know what’s going on here any more than anyone else does. Thus far, we have the Fat Finger theory. The fat finger theory at least makes a little bit of sense. When people dump enormous blocks of stock … things happen. Microstructure 101. I was going to try to work through the microstructure of how this could happen using a little of Hasbrauck’s book, but unfortunately, the SEC says it didn’t happen that way.

The usual band of numskulls is out blaming HFT in their typically nebulous manner. Nobody seems to want to go out on a limb and show us how such a thing could happen, other than, “things we don’t understand are the source of all evil!” It boggles my mind that people who don’t know a limit order from a fill or kill feel free to weigh in on this subject. Hey man, I know about iphones; it must be computers!

Other seemingly less likely possibilities include “cyber attack” or market manipulations. My favorite unlikely speculation is that Taleb did it. Probably the real answer is complicated.

Rather than pointing fingers around like a bunch of hysterics, or attempting to clean the Augean stables of figuring out exactly, precisely what happened, why don’t people think about what market rules would have prevented this? I mean, that’s what everyone seems to be agitating for: changing the rules. Rather than changing the rules to please self interested parties who want trading to go back to the days of ticker tape and the horse buggy whip, why not change the exchange rules to prevent this sort of thing and make everybody happy?

Right now, people are talking about circuit breakers, stop price logic and so on… The standard circuit breaker in the NYSE never engaged, as it was never hit (it’s linked to the DJIA). Stop price logic seems to be circuit breakers for individual instruments: stop trading for a few seconds, then go back to normal. This is apparently implemented in US futures exchanges. Since the futures did a little better than equities, we’re supposed to believe this would have solved the problem. Well, maybe, but maybe not. According to the WSJ, one of the proximate causes of the plunge was … well, an effective halt in trading in PG&E.

Assuming this is true, can someone explain to me why halting the trade for a few more seconds would be anything remotely resembling a good thing? As far as I can tell, that would be a silly thing which wouldn’t help anyone. While I can’t claim this is an original idea: it seems like doing what happens during other market dislocations would make more sense. You know, like when you start trading at the beginning and end of the day. Dudes with market orders would be fucked, but they were fucked anyway, and at least you’d restore some semblance of order.

Maybe dropping to a call auction model when something goes pear shaped is what they’re actually proposing, but it doesn’t look like that reading the funny papers. The Europeans do it this way. The Asians do it this way. Why not do it this way? You need liquidity? Drop to a call auction. I’m not really optimistic that anybody is thinking of this; a google on NYSE and vol interruption auctions yields the following link …which rather indicates the NYSE never heard of the idea.

Seems like the right thing to do to me. Beats stockbrokers jumping out the window.

Edit add:
Apparently this is how it works in the NYSE.
http://nyserules.nyse.com/NYSETools/TOCChapter.asp?manual=/nyse/rules/nyse-rules/chp_1_13/default.asp&selectedNode=chp_1_13
Don’t look like an auction to me.
CME debug:
http://cmegroup.mediaroom.com/index.php?s=114&item=159

And … the mystery seller is … Waddell & Reed! Who? Some financial advisor who got the world’s worst execution from Barclays. Who dumps 4 billion of anything on the market?
http://www.reuters.com/article/idUSTRE64D42W20100514

There you have it: as usual, the biggest fuck ups come from the “respectable” bozos in the investment business, rather than the eeeevil high frequency traders. I, for one, eagerly await the apologies from the numskulls who blamed HFT.

16 Responses

Subscribe to comments with RSS.

  1. L said, on May 14, 2010 at 2:09 am

    “… can someone explain to me why halting the trade for a few more seconds would be anything remotely resembling a good thing?”

    The theory is that ‘doing something’ — anything — is a good thing.

  2. Tom said, on May 14, 2010 at 2:57 am

    Check out Max Keiser’s theory YouTube title “Keiser Report № 42: Markets! Finance! Scandal!” Over dramatic, but perhaps true. Gaming psychology applied en masse so day traders. They follow the lead as provided. Too much? Stick with it for Damon Vrabel.

  3. maggette said, on May 15, 2010 at 11:38 am

    “I, for one, eagerly await the apologies from the numskulls who blamed HFT.”

    We both know there will be no appologies. The publich memory is to short. When the smoke clears everybody who told that crap on national TV and the web won’t be asked about there former statements..

    • Scott Locklin said, on May 15, 2010 at 6:27 pm

      Maybe someone should compile an “enemies list.”

      • maggette said, on May 17, 2010 at 10:00 am

        Maybe I will write a couple of hate mails…but I don’t think anybody will answer them;)

        The thing that scares the hell out of me: I wouldn’t consider myself an expert of capital markets and finance, but I can see, read and listen to an incredible large bunch of media clowns that obviously have NO clue what they are talking about! Still they form (or at least influence ) the public opinion.

        What if their research about other topics (health care, war overseas, energy policy..) is as bad as their knowledge and journalism on finance related topics?

        Scary…

        • L said, on May 17, 2010 at 11:47 am

          “What if their research about other topics (health care, war overseas, energy policy..) is as bad as their knowledge and journalism on finance related topics?”

          At the risk of being branded a ‘conspiracy theorist’, there’s always the possibility that our betters are not incomptent dunces.

        • Scott Locklin said, on May 17, 2010 at 6:28 pm

          Oh, I’m certain this is true. Reading the physics funny papers and comparing to what I know about that subject … if I extrapolate to everything else, reading any sort of media should be considered harmful for your knowledge of what goes on in the world.

          “Oh noes, the fourth estate might be out a job;” can’t happen quickly enough, IMO.

  4. William O. B'Livion said, on May 17, 2010 at 4:13 pm

    “What if their research about other topics (health care, war overseas, energy policy..) is as bad as their knowledge and journalism on finance related topics?”

    Do you have any evidence that it is better? I mean consistently better?

    “At the risk of being branded a ‘conspiracy theorist’, there’s always the possibility that our betters are not incomptent dunces.”

    Betters? You some sort of European boy?

    Betters?

    Fuck you “betters”.

    The chance that Claire McCaskill, Barbara Boxer, and the rest of those power hungry fucks in the Senate are NOT my betters. They have some domain-specific knowledge I do not posses (Senate rules mostly), but they are NOT my betters, and in the case of the two mentioned specifically my daughter (the 19 year old, maybe the 3 year old) could do a better fucking job.

    I submit, with no respect at all, that NO ONE is competent to “run”, “manage” or “lead”, much less “control” a country the size and complexity of ENGLAND, much less a nation-state the size and economy of the United States.

    You CANNOT write rules for something like a capital market that prevent downside. At best you can rate-limit bad decisions, but frankly making sure that the people who make the mistakes take it in the shorts is probably the best answer.

    Shame we can’t make my “betters” take it in the shorts when they fuck up.

    Betters. Bah.

  5. Lew Burton said, on May 19, 2010 at 3:16 pm

    Halting or slowing a single security that is a member of an index is an illusion–it can be traded synthetically. Often, two or more halts can be traded synthetically. But illusions serve a purpose. If a trade doesn’t make it into TAQ, then it doesn’t enter the risk models. It disappears from history.

    The game playing between ECNs is a little funky. Instead of ‘self help’ it might better to mandate a temporary central limit order book (CLOB) under times of stress.

    The aftermath of this story speaks volumes about the lack of money available to pay people with real experience.

    • Scott Locklin said, on May 22, 2010 at 12:08 am

      I’m of course very curious about how the flow gets routed around the ECN’s in conditions like this. Should I route? Sweep to fill? Do the network effects cause discontinuities? Was it a dislocation between the E-mini and the equity market? I dunno.

      FDAXHunter is saying it couldn’t have been these Wadell dudes. Probably not a good idea to bet against him.

  6. Mitchell Porter said, on June 8, 2010 at 5:23 am

    A lot of people seem to interpret this as “Rise of the Machines”. One month later, it suddenly occurs to me – is the sequence of events evidence that the system has some stability built in? Because it crashed, but then it went back up just as fast!

    • John Flanagan said, on June 9, 2010 at 8:37 am

      It’s evidence of quantum tunneling in the markets!

      Ford help us if we ever tunnel through to a false vacuum.

  7. John said, on July 6, 2010 at 7:20 pm

    I was there watching my Quotron during the 87 crash when the S&P futures cascaded down with huge gaps between trades. I also was there duing Fat Finger Thursday when the Dow plunged 1,000 points in minutes. I conclude that no ‘regulations’ can stop a full-blown selling panic – the emotion of the herd has to play out until exhausted.
    Incidentally, theer are many more fat fingers waiting in the wings before this bear market is done.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: