Tuesday, October 05, 2010

Hot Potatoes

RT Leuchtkafer follows up on his earlier remarks with a comment in the Financial Times:
After a detailed four-month review of the flash crash, looking at market data streams tick-by-tick and down to the millisecond, the SEC concluded that a single order in the e-mini S&P 500 futures market ignited an inferno of panic selling. It was over in about seven minutes, and $1,000bn was up in smoke.
Within hours of the SEC’s report, the CME Group, owner of the Chicago Mercantile Exchange, issued a statement to point out that the suspect e-mini order was entirely legitimate, that it came from an institutional asset manager (that is, the public), and was little more than 1 per cent of the e-mini’s daily volume and less than 9 per cent of e-mini volume during and immediately after the crash.
How did this small bit of total volume cause such a conflagration?
You do it with computers. Specifically, you do it with unregulated computers. You pay rent so your machines sit inside the exchanges, minimising travel time for your electrons. You pay licence fees so your computers eat their fill of super-fast proprietary data feeds, data containing a shocking amount of information on everyone’s orders, not just on your own.
And when your computers spot trouble, such as a larger than expected sell-off, they dump inventory and they shut down – because they can.
No one knows what a “larger than expected sell-off” might be, but on May 6 a single hedge that added just an extra 9 per cent of selling pressure was enough to cause chaos.
When that happened, the SEC’s report says, high-frequency traders “stopped providing liquidity and began to take liquidity”, starting a frenzied race for anyone willing to buy. The report likened the panic to a downward-spiralling game of “hot potato” where, as HFT firms bought beyond their risk limits, they pulled their own bids and frantically sold to anyone they could, which were often just other HFT firms, who themselves quickly reached their risk limits and tried to sell to anyone they could, and so on – into the abyss. Fratricide ruled the day. Firms then fled the market altogether, accelerating the sell-off.
Punch drunk, markets rebounded when other market participants realised what had just happened and jumped into the market to buy.
Fair enough, some might say. Markets do panic, and sometimes for no reason. But the larger HFT firms register as formal marketmakers, receiving a variety of regulatory advantages, including greater leverage. All of this extends their enormous reach and power. In the past, they fulfilled certain obligations and observed certain restraints as a quid pro quo for those advantages, a quid pro quo intended to keep them in the market when markets were under stress and to prevent them from adding to that stress. Over the past few years, however, decades-long obligations and restraints all but disappeared, while many advantages stayed.
Computing power also opened marketmaking to a field of unregistered, or informal, high-frequency marketmakers, what investor and commentator Paul Kedrosky termed the “shadow liquidity system”. Exchanges will pay you to do it, too, just as they pay formal marketmakers, and require little in return.
The result is a loose confederation of unregulated, or lightly regulated, high-frequency marketmakers. They feed on what many consider confidential order information, play hot potato in volatile markets, and then instantly change the game to hide-and-seek if even a single hedge hits an unseen and unknowable tipping point.
The only quibble I have with this analysis is that too many different classes of algorithmic trading strategies are being bundled together under the HFT banner. In particular I would like to see a distinction made between directional strategies that are based on predicted short term price movements, and arbitrage based strategies that exploit price differentials across assets and markets. Both of these can be implemented with algorithms, rely on rapid responses to incoming market data, and involve very short holding periods. But they have completely different implications for asset price volatility. It is the mix of strategies rather than the method of their implementation that is the key determinant of market stability.


Update: Leuchtkafer writes in to say:
I should have been clear in the piece I was talking specifically about market making strategies. 
I appreciate the clarification, and agree with his characterization of the new market makers


  1. Amazing. I've never seen a "mix of strategies" have this much affect on the market in the last ooooohh, I'd say roughly 80 years. Very deep, very deep. So deep I can hardly walk in it.

    Isn't it amazing that up to the last few years a "mix of strategies" has never affected markets like this?? What is so special about this "mix of strategies"!?!?!?!?! Wow, so much insight here.

  2. ONE simple question for you Professor Sethi--take your time to answer it, no rush:

    Can you give me a single example where an algorithm or algorithms created sell-side pressure on the stock market before the existence of HFT??? Just one example. Take your time.

  3. Ted, you are completely mistaken about this. The flash crash was an extreme version of a routine event (as I argued here). The speed with which departures from fundamentals can occur has increased due to HFT but bubbles and crashes have been around as long as speculative asset markets have existed. It seems to me that you have an ideological axe to grind and are not really interested in understanding the phenomenon.

  4. Christian Smith-Socaris has emailed me the following excellent comment:

    "It is the mix of strategies rather than the method of their implementation that is the key determinant of market stability. "

    I think that Leuchtkafer's perscription of making HFTs act as responsible market makers would indirectly change the mix of strategies. The fact that this hot potato phenomena had such a serious effect was because the people engaging in it could just bail out when things got uncomfortable. If they did have liquidity provision responsibility wouldn't they have just two options: build better algorithms that exploited the others or demand better regulation of strategies? Granted they could regulate in ways that don't change the mix, but at least the correcting incentives would be right and things should improve over time.

    In sum: If the HFTs were forced to make markets they would be forced to design more stable systems (even if that meant curbing HFT) in the interest of their own preservation. If the mix of strategies is critical to stability they would have to fix it or perish; ah capitalism.

    Christian Smith-Socaris

  5. I'm ready to believe nine bad things about HFT before breakfast — until Mr. Leuchtkafer adds a tenth, and then I begin to wonder about the first nine.

    Near the end of his FT essay Leuchtkafer writes:

    "the flash crash was a freight train of their [HFT-firms?] own design and manufacture. These firms lobbied for deregulation and lobbied to keep their many advantages over everyone else. So it is ironic the freight train they ran from on May 6 was just another unregulated high-frequency marketmaker...."

    Let me begin my rebuttal by clarifying some things.

    1) First of all, Waddell & Reed precipitated the market's plunge by attempting to sell 75,000 e-minis too quickly into a thin, already weak market. Such a plunge could happen again tomorrow if Fidelity or Vanguard unwisely decided to sell such a large quantity into a weak market.

    2) Second, and more importantly, Waddell & Reed is most definitely NOT, I repeat, not a HFT market-maker.

    Leuchtkafer reveals his ignorance when he says that Waddell & Reed is "just another unregulated high-frequency marketmaker....")

    Leuchtkafer is totally wrong.

    Aspiring critics of HFT must not confuse or conflate Waddell & Reed (or Fidelity, Vanguard, Putnam, etc.) with the likes of Getco, Citadel, Tradebot, and too many other HFT market-making firms (that Leuchtkafer probably still hasn't heard of).

    Anyone who likens a firm such as Waddell & Reed with a HFT market-making firm does NOT know what he is talking about -- and should be ignored.

    Waddell & Reed (Fido, Vanguard, et. al.) is "paper"; they are representative of the retail customer-order flow that HFT market-making "locals" live off.

    It's quite true that "paper" and "locals" are both traders, but the two entities trade with each other seeking entirely different ends.

    It has ALWAYS been true that when "locals" got hit or lifted by "paper" that they would quickly scramble in hot-potato-like fashion to cover their unhedged positions, hopefully at a profit.

    And Mr. Leuchtkafer thinks that that ought to be changed?

    He is clueless.

    (And the font Mr. Sethi uses at this blog is hard to read.)

  6. Thanks for your comment and sorry about the font - I just used one of the few default options available when I started this. The RSS feed may be more readable.

    You're absolutely right that Waddell & Reed is not a HFT market maker and should not be confused with the likes of Getco and Tradebot.

    But I think that you're misreading the quote from RTL. He is not referring to the Waddell & Reed Sell Algorithm in that quote, but rather to the response by HFT marketmakers to the changes in price and volume that the order gave rise to.

    This will be clear if you just look earlier in his comment, where he refers to W&R as an "institutional asset manager" and "the public". Here's the relevant para:

    "Within hours of the SEC’s report, the CME Group, owner of the Chicago Mercantile Exchange, issued a statement to point out that the suspect e-mini order was entirely legitimate, that it came from an institutional asset manager (that is, the public), and was little more than 1 per cent of the e-mini’s daily volume and less than 9 per cent of e-mini volume during and immediately after the crash."

    RTL does know what he's talking about, though his rhetoric is sometimes over the top (that's why I didn't quite the freight train part of his comment).

  7. Professor Sethi,
    Interesting you still didn't/can't give a single example of what I asked you for above. I only asked for ONE. But I am the one with an "ideological axe to grind"???? Fascinating....

  8. Ted, there are dozens of examples, but you can start with the October 1987 crash in which program trading was implicated. Also, I would appreciate it if you take your sarcastic comments elsewhere, they are not welcome on this blog.

  9. I'm very sorry if Nanex doesn't support your theory on "mix of trading strategies" or your theory of "algorithmic trading strategies" being the cause of the May 6th Flash Crash. Their analysis shows it was a few players that caused the crash. Those using HFT---High Frequency Trading. I encourage all your blog's readers to read Nanex's Final Conclusion themselves here.
    I don't blame you for disliking my comments though. No more after this one. I'll let you continue on as you like.

    Good luck with that 1980 paper with Beja and Goldman.

  10. Here's what the Nanex report says:

    "First of all, the Waddell & Reed trades were not the cause, nor the trigger. The algorithm was very well behaved; it was careful not to impact the market by selling at the bid, for example. And when prices moved down sharply, it would stop completely.

    The buyer of those contracts, however, was not so careful when it came to selling what they had accumulated. Rather than making sure the sale would not impact the market, they did quite the opposite: they slammed the market with 2,000 or more contracts as fast as they could. The sale was so furious, it would often clear out the entire 10 levels of depth before the offer price could adjust downward. As time passed, the aggressiveness only increased, with these violent selling events occurring more often, until finally the e-Mini circuit breaker kicked in and paused trading for 5 seconds, ending the market slide.

    Because of arbitration, when the e-Mini changes price with high volume, many ETFs are repriced (quotes updated, trades executed). The component stocks of ETFs are also repriced, along with many indexes. And finally, all the option chains for the ETFs, their components and indexes are also repriced. The entire system simply cannot absorb the impact of a sudden move in the e-Mini on high volume. A sale (or purchase) of 2,000+ contracts which rips through one-side of the depth of book in 50-100 milliseconds, will immediately overload many systems. The impact reverberates for a much longer period of time than the sell (or buy) event itself.

    The first large e-Mini sale slammed the market at approximately 14:42:44.075, which caused an explosion of quotes and trades in ETFs, equities, indexes and options -- all occurring about 20 milliseconds later (about the time it takes information to travel from Chicago to New York). This surge in activity almost immediately saturated or slowed down every system that processes this information; some more than others. Two more sell events began just 4 seconds later (14:42:48:250 and 14:42:50:475), which was not enough time for many systems to recover from the shock of the first event. This was the beginning of the freak sell-off which became known as the flash crash.

    In summary, the buyers of the Waddell & Reed e-Mini contracts, transformed a passive, low impact event, into a series of large, intense bursts of market impacting events which overloaded the system."

    The buyer described here is using a particular strategy, implemented through an algorithm. Alongside this are a variety of other arbitrage based strategies, also implemented through algorithms, that maintain parity across prices of equivalent bundles of securities across different markets. The former class of strategies are destabilizing while the latter are non-directional. There are lots of different strategies that operate under the HFT banner and not all of them are destabilizing. That was my point.