Do HFTs increase liquidity or just volume?
Sapient’s David Donovan brought up a question that is being echoed in many places. Are HFT liquidity providers actually providing liquidity or do they just create volume? Here’s a video in which TabbFORUM Editor Greg Crawford talks with Donovan on this subject after the conference.
The high quote-to-cancel rates create a challenging environment for investors looking for real liquidity. The quotes are barely accessible due to their short validity durations. The panel discussed various options that have been under consideration including quote-to-fill ratios, minimum order durations and cancellation surcharges. No option seemed particularly attractive.
The question about liquidity versus volume is an important one. I’ve heard estimates that nearly 70 percent of all volume in the equities market now comes from HFT strategies. If most of the volume is from the HFTs then are they just executing against each other?
Another estimate places volume in the dark pools at 30 percent. Interesting correlation. This raises a question: is most of the “real” liquidity actually hidden away in the dark pools? The correlation of these statistics suggests that a substantial portion of institutional volume is probably being executed in the dark.
Do we need a level playing field?
In the video, Donovan says he’d like to see a level playing field where everyone gets the same data at the same time. But what would this level playing field look like? Do we take away all the technology advancements and slow down data feeds so that no one receives data faster than the slowest buy-side shop located on the west coast subscribing to an aggregated feed?
Or should some benevolent entity grant free technology to all the tech laggards to catch them up with the fastest HFT players? If a firm is able and willing to invest in the technology and infrastructure necessary to process data as fast as possible, then why shouldn’t that firm be able to profit from its investments?
This same buy-side contingent has made extensive investments in portfolio analytics, advanced risk analysis and other technology to help ensure that they get the best returns for their clients. Should we also share that technology with every retail investor to ensure that they can compete with the institutional flow?
Are all the HFTs the same?
The buy-side concern about HFT is real and it’s impacting investor confidence. But in many ways, HFT is getting a bad rap when really, there are many legitimate players, including buy side shops adopting these strategies.
Last year, Bernard Donefer published an outstanding essay about high frequency trading. He made the point that “Referring to HFT as one undifferentiated practice obscures the benefits and risks in each business model. Many press criticisms leveled at HFT are really only relevant to specific (and sometimes already illegal) strategies and only serve to confuse the public.”
This is a hot topic and my opinion is only one of many. I’d love to hear your comments.
Comments | Post a Comment
22 Comments to "Do HFTs Increase Liquidity or Just Volume?":
gbollenbacher
28 June 2011
So, one thing you have to do, to answer this question, is define the term liquidity. Do you mean lots of trading volume, or relatively stable prices, or the ability to execute large trades with little movement in price? It's a market fact of life that short-term traders never provide liquidity for any period of time. The only thing that provides liquidity over the long term is end investors with diverging views on values. Even a lot of end investors who have coincident views on value don't provide liquidity until you reach the value they agree on. To think that trading firms that flash quotes into electronic markets for a nanosecond at a time would supply liquidity is naive in the extreme. What is worse at the moment is that many institutional investors have concluded that HFTs are simply there to front-run their legitimate order flow, so they are migrating to dark venues for their own protection. Whether that's good for the markets is perhaps the most important question of all.
Comments (51)
kurtkujawa
28 June 2011
I am sure whomever came up with the idea to use the term "liquidity" when framing the whole HFT argument, is probably a relative or protégé of the person who worked for the SOES bandits of the 90’s when they were lobbying the SEC. . What is so remarkable is that they can still spin the argument and confuse some of the same people over and over and over. Of course volume is not the same as liquidity; anybody with half a brain should be able to understand that, as gbollenbacher so astutely noted. As for what’s good for the markets? Who cares what is good for the markets, what I care about is what’s good for me and my clients. To me the answer is simple, try and educate every trader I know to stop slicing and dicing every single order and throwing it into some algo, avoid every so called “liquidity” provider, stay upstairs or in select institutional block dark venues and trade in “blocks” and stop worrying about every tick. If the institutions stay upstairs and away from the markets, eventually the HFT’s will cannibalize them selves and go away, so we can get back to doing what the markets were intended for, bringing NATURAL buyers and sellers together.
Comments (177)
candyce
28 June 2011
gbollenbacher - thanks for your comments. Your last point - wondering whether it's good for the markets is more complicated than it first seems. HFT and high transaction volume is GREAT for the exchanges and other execution venues that make a small fee on every transaction. Those markets work very hard to attract and retain the HFT "liquidity providers." This is understandable. The exchanges need a predictable revenue stream to survive, and with ever-declining margins, they have to find revenue from every possible source. But in this case, what's good for the exchanges is not necessarily good for the rest of the market participants. However, exchanges serve a vital role in the markets.
Comments (28)
candyce
28 June 2011
kurtkujawa - thanks for your comments also. So what about the role of market makers? Do you think the need for market making is obsolete?
Comments (28)
gbollenbacher
28 June 2011
Candyce, thanks for your response. I wasn't questioning whether exchanges or trading volume were good for the markets. I was pointing out that institutional investors are more and more coming to regard HFTs as frontrunners, pure and simple. The Tabb Forum has featured stories about technology vendors touting order routing software that protects the user from the HFT poachers, and an increasing part of the dark volume is institutions keeping away from the HFTs. If HFTs start snooping on dark venues, and institutions start trying to fake them out (think the Rothschilds after Waterloo), the markets could easily become counterproductive in a big way. That would call for enlightened regulation, something that's been in short supply recently.
Comments (51)
kurtkujawa
28 June 2011
Candice, 1st off you are correct that what is good for the exchanges is not good for the market participants. The whole competitive exchanges argument is just another scam that has been perpetuated on the same fools that bought into the “liquidity” argument. Those same fools, that were so fast to blame specialists and the NYSE rather then themselves for there own ineptitude and lack of trading prowess, are again being lead down the path to ruin by the same charlatans. Talk about tripping over pennies and loosing dollars. The trader that thinks he can save his way to performance is a fool and in the wrong business. Do you not realize for every .25 mills they say they are saving you, they are taking either directly or indirectly pennies and dollars. They are no different then websites selling your personal information, they are just selling your order information. I have always laughed at Liquidnet’s motto of “price improvement” really? Is it not true that on a 2 sided trade, by definition, in order to have price improvement, there must also be the opposite, or a price “disimprovement”? be careful of the wolf in sheep’s clothing. 2ndly, is there a role for market makers. I don’t really think so. At least not in there current form. And please, can we also stop referring to the HFT’s as market makers. A market maker actually takes risk. Simply getting in the way of natural order flow does not constitute being a market maker. Just another lie perpetrated to the unsuspecting fools. One thing I have learned over my 20+ years in the business as a sales trader, market maker and buy side trader. Technology on its own is not destructive. Neither are drugs and alcohol, yet when you start to depend on them it is easy for the pushers to tell you what ever you want to here so you can get the next fix. And believe me when I tell you the FIX is in!
Comments (177)
Anonymous
29 June 2011
Gbollenbacher suggests “Even a lot of end investors who have coincident views on value don't provide liquidity until you reach the value they agree on.” It might be semantics but liquidity occurs, a trade occurs, when two investors agree on price but disagree on value; hence a seller and a buyer exist. What we do need to agree on is the value of creating fair and transparent equity markets. Given the debt crisis equities activity should be booming – why isn’t it? Where as a low tide exposes, the draw from a Tsunami exposes much more. The debt crisis , our financial Tsunami, with its consequences for trading volumes has exposed fundamental flaws in the way that equity trading is being conducted. A tightening of commissions (brokers’ not notably fund managers’) coupled with falling volumes has exacerbated dark trading as banks and brokers with scale look to maintain a standard of living by extracting as much profit as possible from every trade irrespective of the wider or longer terms connotations for the market. The emergence of off exchange trading is to some extent due to HFT but to a much greater extent the product of highly intelligent honed machines developing self-serving arguments to persuade the buyside that in the light they are easy game for these highly adept predators. It is this vested interest that has led to the banking fraternity’s opinions being ignored by the EU in their MIFID II deliberations. Even the instutions are tainted in the eyes of the EU who see them as doing anything the banks want in exchange for a continuous flow of analysis. The main argument put forward for the perils of the lit exchange is the statistical argument that most of the flow on exchange is HFT flow. One is aghast at this revelation until you consider that all market-making flow is low latency dependent and because we live in a world of fast moving news and information prices adjusted frequently. Does this mean its HFT flow; absolutely not. The market making function whether obligatory or incentivized is done by Principal Trading Group Firms (PTG) who do in fact take risks by quoting on a broad base of securities. This is real liquidity that appears on all lit venues and exists contrary to a shot in the dark – it might be adjusted frequently for price but it’s there for lifting! The reality is that across the EU, except the UK market, everyone from retail investor to buyside institutional trader has the opportunity to interact with this liquidity. Brokers on the other hand are incentivized to push it elsewhere! 50% of the flow on lit pools, assuming that a market maker or liquidity provider is one side of all transactions, is then reasonably derived from Principal Trading companies with liquidity provision mandates; this should be of no concern but rather welcomed. What’s left of the stipulated 70%, is that HFT flow and real money currently trades in roughly equal measure on exchange. The reality is that if the broker directed real money flow was moved to the lit pools it would dwarf the HFT flow. The article correctly alludes to the fact that the majority, possibly >80%, of banks’ real money flow is internalized. Driving real money or natural liquidity away from lit exchanges exacerbates the ratio between HFT and real money volumes and the perception that it’s a den of card sharps. This despite the fact that there is minimal price improvement in off exchange trading as the practitioners of internalizes ensure that they give the worst possible price they can get away with. When did it become acceptable for the retail investor to be such fair game that their flow is actually worth paying for – payment for dumb order flow! Best Execution is obviously nonexistent for these suckers! This is not best execution its self annihilation. Have the exposures of the credit crisis taught us nothing of the propensity of individuals and corporations to be self-serving. Candyce is right the exchanges are the fulcrum of our markets, of indeed, capitalism. We must reinstate the value of exchanges, their dominance as the forum for trade and in so doing help reinstate a belief in the integrity of equities markets. Empower the exchanges and MTFs with sufficient business to seek out and move against predatory HFT firms that continuously aggress a handful of stock making life harder for the liquidity providers and denying opportunity to the brokers. I heard a trader comment that the more complex and stacked against them the individual investor thinks the market is, the more he is going to entrust is flow to the professional manager. Quite the contrary, coupled to the duds offloaded by the Private Equity firms and other over valued IPOs trading in the dark is a shady business running contrary to fair access and unacceptable in the eyes of the investor. While the buyside trader may feel they’ve performed a good execution it’s killing the industry. For all their folly, the EU smells a rat even if it can’t see it.
elucidate
29 June 2011
Just reading the new One Size Fits None document from Credit Suisse. I noticed that when the bank puts out arguments for dark trading it fails to include internalisation in its examples just dark pools. How does this skewe the stats ?
crammond1964
29 June 2011
the argument about volume and liquidity is missed as open interest is really the key to success . a thought will be when dark pools take over from the exchanges ; as HFT deter folks from trading . = exactly what gsf and csfb and jpm want .
Comments (257)
Anonymous
29 June 2011
Mr. Kujawa states the following:"Who cares what is good for the markets, what I care about is what’s good for me and my clients." Isn't that the EXACT same attitude he and many others attack HFT for? I won't even get into what a non-sequitor the following statement is; "If the institutions stay upstairs and away from the markets, eventually the HFT’s will cannibalize them selves and go away, so we can get back to doing what the markets were intended for, bringing NATURAL buyers and sellers together."
kurtkujawa
29 June 2011
Anonymous, since when is consummating a trade between a natural buyer and seller non-sequitor? An exchange should be nothing more then a clearing house. It is the for profit model that has lead to all the fragmentation and for the rise in HFT. Investors are not here to serve the HFT industry or the exchanges, they are here to serve themselves and there clients. As for your comment about minimal price improvement. The whole idea that there should be price improvement is just another marketing ploy you’ve been sold on. Have you ever heard of a limit order? The whole term “price improvement” is absurd. As is “best execution”, do you know why I am buying or selling something, do you know how much I am buying or selling, there are many factors involved in best execution, LEAST OF ALL IS PRICE BASED ON 100 SHARES! Until you leave your odd lot world, I can not begin to explain to you how the real world operates.
Comments (177)
billcowie
29 June 2011
The comment by gbollenbacher makes a good point about HFTs being perceived as front-running. If the HFT crowd is really making the large sums that they are widely reported to be making, (and I have no reason *not* to believe that they are), the question is: where is that money coming from? To channel Milton Friedman, it’s not the Tooth Fairy. Much of it comes from the trader who is adding visible liquidity to the market by displaying a limit order. If some other trader continually stays 1/100 in front, the liquidity-adding trader must pay an extra penny per share when they move to the other side of the spread to get their order executed. Or perhaps they “only” pay 99/100 of a penny if someone “price improves” their order by 1/100 – and steps in front of the resting limit order on the other side. I don’t see any difference between a market maker illegally stepping in front of an order in a price/time priority market (i.e. buying at the bid or selling at the offer, but stepping in front of resting limit orders), and a trader legally improving an order by 1/100th of a cent. (Actually I do see the difference – it’s an insignificant 1 penny per hundred shares). One possible solution is to require larger price improvement. I did not attend the Tabb Forum, so I don’t know if this was discussed. The minimum improvement might be set at the midpoint of the current spread, or even the full Minimum Price Variation. Those of us who were involved in this industry last century remember that this problem didn’t exist when we traded 1/8ths. I’m not advocating a return to that big a spread, just suggesting that perhaps the pendulum has swung a bit too far. As a side note, I find it highly unlikely that the HFTs are trading only with each other. That environment is a zero-sum game. Every penny of profit comes at the expense of some other trader, and again, the reports are that the group as a whole is making large profits. On the issue of liquidity vs. volume, the Exchanges and ECNs have maker/taker pricing models that do not charge traders anything to submit an order, or to cancel it. As we all learned in Economics 101, if the price of something does not fully reflect its cost, too much of it will be demanded. The problem for the Exchanges/ECNs is that the customers who pay the bills with their revenue-generating executions are the same ones who send in (and cancel) ten or twenty free orders per trade. Other businesses have a similar problem. Airlines charge you for a ticket, and promise to transport your luggage as well. Recently, baggage charges have become a price and marketing differentiator. On the other hand, no consumer is charged for sending email messages, regardless of how many they send, and thus we have spam. A possible solution here would be for the Exchanges and ECNs to institute a charge for orders submitted. Candyce reports that forum participants were not enthusiastic about charging for cancellations, but I didn't see any mention of whether an order fee was discussed. There are many issues around an order fee, and they all need to be considered carefully. A flat fee might not be the best solution, but creative minds should be able to craft something that addresses most reasonable objections. Brokerage firms would have to decide how and when to pass that cost back to the customer, and it might become a competitive differentiator, like the airline baggage fee. To protect the small investor, perhaps the first 5 or 10 or some number of orders per month would be free. Or the fee could be credited against the execution of that order. Or perhaps the fee could be tied back to the number of executions… And it appears as though this is something that might have to be regulated into existence, as distasteful as that is to some. Free market forces have not been sufficient to do so, and probably won’t be any time soon.
Comments (4)
Anonymous
29 June 2011
Kujawa, you'd make a lot more sense if you stick to the topic at hand without stooping to boorish insults, so ignoring your attempts (Leave my odd lot world?). The point is that Institutions alone cannot satisfy liquidity demands on both sides of the market, thus even they need short term liquidity suppliers to provide that liquidity. You may want to ask the operators of one of the dark pools you trade in just what the match rate is, it is usually well below 20%. Thus if all Institutions did was trade "upstairs" over 80% of their orders would never be executed. Most HFTs are in effect Short Term liquidity providers which bridge that liquidity gap.
kurtkujawa
29 June 2011
Anonymous, while I agree that not everything can get done "upstairs" or in the dark, I do not agree that HFT's are short term liquidity providers, getting in between natural orders is not providing liquidity. Selling me 100 shares at $20, then taking 200 at 20.01 and 100 at 20.02 and selling me 200 at 20.04 is NOT providing liquidity, do the math, 600 shares traded, (volume) but only 300 actually traded hands (liquidity). You can always find liquidity; it’s just a matter of time and price, I don’t need someone getting in-between. If I want in or out and time is the priority, I can either move the stock until I find liquidity, or have a broker (PTG) make a bid/offer. If price is the priority, I will have to wait until liquidity shows at the level I want. Either way, I don’t need HFT. If the PTG wants to use a HFT strategy to reduce his exposure AFTER he takes down my piece, I don’t care. I think we can all agree, that there are various HFT strategies, like all things some good and some bad. What myself and the majority of the street refers to when discussing HFT are the predatory strategies and the firms who use them.
Comments (177)
Anonymous
29 June 2011
Kujawa, Agreed. There are HFTs that take advantage of the market and are predatory. I do not think regulation will solve that issue, predators will always be part of any market structure. The best we could hope for is tiered market structure that advantages size and natural liquidity, unfortunately that may result in even more fragmented markets.
kurtkujawa
29 June 2011
I agree that regulation will not solve it, if anything it will make it worse, it always has! We need to get back to the basics, which are that the exchange should act as a clearing house, not a profit center. I really think the only way to get less fragmentation and larger markets is with some type of CLOB. I know the NYSE might not have been perfect, but at least the markets were deeper and you at least had a market based on size not 100 shares.
Comments (177)
crammond1964
01 July 2011
what happens when "dark pools" have a larger percentage of the market than the exchanges ? Will only then we will realise what damage HFT and dark pools have brought to the market ! and at this current rate it be by end of the year !
Comments (257)
Anonymous
01 July 2011
Dark Pools will never have a larger percentage of the exchange market for two reasons. First regulation is already in proposal stage that will ensure much of what is now dark will move to lit markets. And second, the amount of real actionable latent liquidity is much smaller than people realize, of that amount a very small fraction gets executed. I should add it's executed in only 3 or 4 venues for the most part.
crammond1964
01 July 2011
dear sir , currently bunds on eurex are trading over 30% in dark pools and rumored to be higher . i know because i trade it daily . retail rather trade in dark pools than be ripped by co location thieving that exchanges allow HFT to do . I repeat once the exchange loses its retail customer it will quickly lose its market share .. its happening right now .
Comments (257)
Anonymous
01 July 2011
crammond, I was strictly referring to cash equities. As you probably know retail orders are mostly gone from the US listed exchanges as well.
crammond1964
06 July 2011
There has to be some mistake here: according to the just released June CBOT volume for futures and options across the 4 key product categories: interest rate, equity index, energy, and commodities, plummeted by 92.9% Year over Year for the month of June. Although apparently not really per Reuters: "Trading volume at the Chicago Board of Trade was down 92.9 percent in June 2011 at about 5.3 million contracts versus about 74 million contracts traded in June 2010, CME Group said in its monthly volume report. The year-to-date volume through June 2011 was about 442 million contracts, compared with 443 million contracts for the same period in 2010, down by 0.3 percent." Some of the more jarring observations: $25DJ index futures: 2 contracts in June 2011, Mini Dow futures: 154K versus 3.7 million, and a complete collapse in IR futures and options: 5 and 10 Year Note futs down from 24MM and 10MM respectively to... 1.9MM and 934K! We can only assume this is due to some recalendarization of trading as otherwise this implies an epic collapse in any investor participation.
Comments (257)
crammond1964
22 July 2011
if the retail trader turns his back on the markets then we will have to impose speed limits as with co location the HFT and algos are making volatility and losing liquidity .
once you start losing open interest your volumes dropped very fast ....... the warning signs have been there for a while .
fill or kill orders have never helped a market .
Comments (257)