Senator Kaufman Reminds Most HFT Issues Still On Table; Notes Rising Market Structure Concern By Regulators And Market Participants

Yet another much needed reminder that the topic of High Frequency Trading is far from resolved. On Tuesday, Senator Ted Kaufman reminded that increasingly more regulators and market participants remain divided over HFT, even as concern about possible improprieties associated with market structure grows. Kaufman's most recent topic of focus - order cancellations. He said the SEC should address the "burgeoning" number of order cancellations involved in high frequency trading, which, he added, are "clearly excessive" and virtually a "prima facia" case that battles between competing algorithms have become "all too commonplace, overloading the system and regulators alike."

Kaufman's five priorities for the SEC have been as follows:

1. Provide timely guidance on new market practices, like co-location and naked access — before they become too widespread.
2. Use its ‘large trader’ authority to require the tagging and disclosure of high frequency trading, then mask it and release it to the marketplace so there can be independent analysis by academics and others on the effects of high frequency trading on long-term investors.
3. Better define ‘manipulative’ market activity and provide clear guidance for traders to follow, just as the UK has declared “spoofing” to be a deceitful trading strategy.
4. Continue to make the reduction of systemic and operational risk a top priority.
5. Address the growing number of order cancellations.

Lastly, the Senator has called on the high frequency trading industry to “come to the table” and play a meaningful role addressing current market issues.

Congratulations to our friends from Themis Trading for being referenced in the Senator's speech.

Full Kauman speech:

Mr. President, I have spoken on the Senate floor many times about the importance of transparency in our markets.  Without transparency, there is little hope for effective regulation.  And without effective regulation, the very credibility of our markets is threatened.
 
But I am concerned recent changes in our markets have outpaced regulatory understanding and, accordingly, pose a threat to the stability and credibility of our equities markets.  Chief among these is high frequency trading. 
 
Over the past few years, the daily volume of stocks trading in microseconds — the hallmark of high frequency trading — has exploded from 30 to 70 percent of the U.S. market.  Money and talent are surging into a high frequency trading industry that is red hot, expanding daily into other financial markets not just in the United States but in global capital markets as well.
 
High frequency trading strategies are pervasive on today's Wall Street, which is fixated on short-term trading profits.  Thus far, our regulators have been unable to shed much light on these opaque and dark markets, in part because of their limited understanding of the various types of high frequency trading strategies. Needless to say, I’m very worried about that.
 
Last year, I felt a little lonely raising these concerns. 
 
But this year, I’m starting to have plenty of company.
 
On January 13th, the Securities and Exchange Commission issued a 74-page Concept Release to solicit comments on a wide-range of market structure issues.  The document raised a number of important questions about the current state of our equities markets, including “Does implementation of a specific [high frequency trading] strategy benefit or harm market structure performance and the interests of long-term investors?” and, “Do commenters believe that the overall use of harmful strategies by proprietary firms is sufficiently widespread that the Commission should consider a regulatory initiative to address the problem?”
 
Among other potential sources of unfairness, the Commission noted that “short-term price volatility may harm individual investors if they are persistently unable to react to changing prices as fast as high frequency traders.”  
 
Finally, the SEC called attention to trading strategies that are potentially manipulative, including momentum ignition strategies in which “the proprietary firm may initiate a series of orders and trades (along with perhaps spreading false rumors in the marketplace) in an attempt to ignite a rapid price move either up or down.  For example, the trader may intend that the rapid submission and cancellation of many orders, along with the execution of some trades, will ‘spoof’ the algorithms of other traders into action and cause them to buy ([or] sell) more aggressively.” 
 
The SEC went on to ask, “Does…the speed of trading and ability to generate a large amount of orders across multiple trading centers render this type of strategy more of a problem today?”
 
Mr. President, the SEC raised many critical questions in its concept release, and I appreciate that the SEC is trying to undertake a baseline review.
 
As its comment period moves forward, I am pleased to report that other regulators and market participants, both at home and abroad, have taken notice of the global equity markets' recent changes, including the rise in high frequency trading.
 
In the United States, the Federal Reserve Bank of Chicago, in the March 2010 issue of its Chicago Fed Letter, argued that the rise of high frequency trading constitutes a systemic risk, asserting “The high frequency trading environment has the potential to generate errors and losses at a speed and magnitude far greater than that in a floor or screen-based trading environment.”   In other words, high frequency trading firms are currently locked into a technological arms race that may result in some big disasters.
 
Citing a number of instances in which trading errors have occurred, the Chicago Fed stated that “a major issue for regulators and policymakers is the extent to which high frequency trading, unfiltered sponsored access and co-location amplify risks, including systemic risk, by increasing the speed at which trading errors or fraudulent trades can occur.”
 
Moreover, the letter cautions us about the potential for future high frequency trading errors, arguing, “Although algorithmic trading errors have occurred, we likely have not yet seen the full breadth, magnitude, and speed with which they can be generated.  Furthermore, many such errors may be hidden from public view because a large number of high frequency trading firms are privately held, rely on proprietary technology, and have no customers.”
 
There is action internationally as well.  On February 4th, Great Britain's Financial Services Secretary, Paul Myners, announced that British regulators were also conducting an ongoing examination of high frequency trading practices, stating, “People are coming to me, both market users and intermediaries, saying that they have concerns about high frequency trading…High frequency trading is a very new development.  Does that have consequences that regulators need to be focusing on?”
 
This development comes on the heels of another British effort targeting so-called “spoofing” or “layering” strategies in which traders feign interest in buying or selling a stock in order to manipulate its price. In order to deter such trading practices, the Financial Services Authority (FSA) announced that it would fine or suspend participants who engage in market manipulation.  The revamped enforcement effort is partly intended to more clearly define manipulative activity.  Noting that some market participants may not be sure that spoofing or layering is wrong, an FSA spokeswoman said, "This is to clarify that it is.”   
 
In Australia, market participants are also requesting clearer definitions of market manipulation, particularly with regard to momentum strategies like spoofing.  In a review of algorithmic trading published February 8th, the Australian Securities Exchange called on the Australian Securities and Investments Commission to, “Ensure that…market manipulation provisions…are adequately drafted to capture contemporary forms of trading and provide a more granular definition of market manipulation.” 
 
Mr. President, high frequency trading poses two risks:  systemic risk to the market and the risk of manipulation.  As the Chicago Fed stated, it is high time that we weigh the risk of major high frequency trading errors against the supposed “efficiency” benefits of cutting down trading speeds by several milliseconds. 
 
The risk of an algorithmic trading error wreaking havoc on our equities markets is only magnified by so-called “naked,” or unfiltered sponsored access arrangements, which allow traders to interact on markets directly — without being subject to standard pre-trade filters or risk controls. 
 
Robert Colby, the former deputy director of the SEC’s Division of Trading and Markets, warned last September that naked access leaves the marketplace vulnerable to faulty algorithms.  In a speech given at a forum on the future of high frequency trading, which was cited by the Chicago Federal Reserve’s recent letter, Mr. Colby stated that hundreds of thousands of trades representing billions of dollars could occur in the two minutes it could take for a broker-dealer to cancel an erroneous order executed through naked access.
 
According to a report released December 14th by the research firm Aite Group, naked access now accounts for a staggering 38% of the market’s average daily volume compared to 9% only four years ago. 
 
Let me reiterate that: almost FORTY percent of the market’s volume is executed by high frequency traders interacting directly on exchanges without being subject to any pre-trade risk monitoring.  
 
In January, the SEC acted to address this ominous trend by proposing mandatory pre-trade risk checks for those participating in sponsored access arrangements.  This move would essentially eliminate naked access, and I applaud the SEC for its proposal.   
 
While I am pleased that the SEC has taken on naked access and has issued a concept release on market structure issues, there is much work that still needs to be done in order to gain a better understanding of high frequency trading strategies and the risks of frontrunning and manipulation they may create.  In the last few months, several industry studies aimed at defining the benefits and drawbacks of high frequency trading have emerged.  While these studies may not be the equivalent of peer-reviewed academic studies, they do have the credibility of real-world market experts.  And they begin to shed light on the opaque and largely unregulated high frequency trading strategies that dominate today’s marketplace.  
 
In addition to the Aite Group study, reports by the research firm, Quantitative Services Group (QSG), the investment banking firm, Jefferies Company, the dark pool operator, Investment Technology Group (ITG), and the institutional brokerage firm, Themis Trading, all raise troubling concerns about the costs of high frequency trading to investors and reinforce the need for enhanced regulatory oversight of these trading practices.
 
Last November, QSG analyzed the degree to which orders placed by institutional investors are vulnerable to high frequency trading strategies.  Large investors typically seek to break up their orders into a series of smaller ones in order to hide their trading interest and avoid price swings. 
 
Despite this tactic, QSG found that institutional order flow is very susceptible to high frequency predators who sniff out such order flow and trade ahead of it. 
 
Specifically, the study concluded that splitting large orders into several smaller ones not only enhances the risk of unfavorable changes in price, but also increases, “the chances of leaving a statistical footprint that can be exploited by the ‘tape reading’ HFT algorithms.”   While traders have long tried to trade ahead of large institutional orders, they now have the technology and models to make an exact science out of it.   
 
A second QSG study released on February 11th analyzed changes in price that occur minutes after a large order is filled.  According to the study, institutional investors routinely experience unfavorable price movements while attempting to make a transaction only to see prices move 25 percent in a more favorable direction once the sale is completed.  This outcome is evidence of high frequency strategies that act "as a motivated competitor for liquidity, not a supplier," the study’s press release says. 
 
In a study put forth on November 3rd, the Jefferies Company examined the advantages high frequency traders gain by co-locating their computer servers next to exchanges and subscribing directly to market data feeds.
 
Jefferies estimates that these advantages afford high frequency traders a 100 to 200 millisecond advantage over those relying on standard data providers.
 
Under such conditions, Jefferies concludes, high frequency traders enjoy “(almost) risk-free arbitrage opportunities.”
 
A Themis Trading white paper released in December elaborated on Jefferies’ conclusion, noting that the combination of speed and informational advantages allow high frequency traders to “know with near certainty what the market will be milliseconds ahead of everybody else.”
 
Themis estimates this ability to essentially predict the future and trade ahead of unsuspecting investors translates into $6 to $12 million a day for high frequency traders employing these predatory strategies, or roughly $1.5 to $3 billion a year in “profit generated from traditional institutional and retail investor assets under management.”
 
ITG, which runs the block trading dark pool POSIT, also conducted a study last November on predatory high frequency trading in dark pools.
 
ITG’s study warns that, “due to the overwhelming participation level of high frequency trading firms in dark pools, adverse selection is occurring much more frequently to the detriment of buyside participants,” who represent institutional investors, including pension and mutual funds. 
 
Regulators need to determine whether institutional investors’ pockets are being picked in the dark pool, pennies at a time.  The evidence shows that high frequency traders have models that effectively allow them to trade at the most optimal time, exploiting institutional investors with less sophisticated trade execution strategies.  In the past, these dark pools were never accessible to high frequency traders, but now they appear to have virtually risk-free money-making opportunities there for the taking.  
 
Equally troubling, ITG notes that the current market structure, conflicts of interest among some broker-dealers, and the lack of adequate quantitative methods and relevant data mean that poor executions are “often invisible” to buyside traders.
 
These findings demonstrate that some institutional investors, including those managing the pension and mutual funds of average Americans, may be failing not only to obtain best executions for their clients when trading in dark pools with high frequency traders, but are also unaware of this reality.
 
While concerns expressed by regulators and members of the industry have mounted, high frequency trading firms continue to enter and expand their presence in the marketplace.  Recently, GETCO, a high frequency trading firm that already serves as a registered market-maker for NASDAQ and BATS Exchange, and represents as much as 10 to 20% of average daily trading volume according to various estimates, announced that it would become one of the New York Stock Exchange’s “designated market makers.” 
 
 As high frequency traders carve out an increasingly dominant role in the market, the lack of transparency into their trading practices looms as a reminder of the troubling gaps in our regulatory framework. 
 
Mr. President, the studies and papers I have mentioned underscore the need for the SEC to implement stricter reporting and disclosure requirements for high frequency traders under its “large trader” authority, as Chairman Mary Schapiro promised she would in a letter to me on December 3rd. We need tagging of high frequency trading orders and next day disclosure to the regulators, and we need it now.
 
Mr. President, for investors to have confidence in the credibility of our markets, regulators must vigorously pursue a robust framework that maintains strong, fair and transparent markets.  The regulators owe it to investors and the American public to actively address these problems.  I would make five points.
 
First, the regulators must get back in the business of providing guidance to market participants on acceptable trading practices and strategies.  While the formal rule-making process is a critical component of any robust regulatory framework, so too are timely guidelines that bring clarity and stability to the marketplace.  Co-location, flash orders and naked access are just a few practices that seem to have entered the market and become fairly widespread before being subject to proper regulatory scrutiny.  For our markets to be credible, it is vital that regulators be pro-active, rather than reactive, when future developments arise.
 
Second, the SEC must gain a better understanding of current trading strategies by using its “large trader” authority to gather data on high frequency trading activity.  Just as importantly, this data – once masked – should be made available to the public for others to analyze. 
 
I am concerned that academics and other independent market analysts do not have access to the data they need to conduct empirical studies on the questions raised by the SEC in its concept release.  Absent such data, the ongoing market structure review predictably will receive mainly self-serving comments from high frequency traders themselves and from other market participants who compete for high frequency volume and market share.
 
Evidence-based rule-making should not be a one-way ratchet because all the "evidence" is provided by those whom the SEC is charged with regulating. We need the SEC to require tagging and disclosure of high frequency trades so that objective and independent analysts — at FINRA, in academia or elsewhere — are given the opportunity to study and discern what effects high frequency trading strategies have on long-term investors; they can also help determine which strategies should be considered manipulative.  
 
Third, regulators must better define manipulative activity and provide clear guidance for traders to follow, just as Britain’s regulators have done in the area of spoofing.  By providing “rules of the road,” regulators can create a system better able to prevent and prosecute manipulative activity.
 
Fourth, the SEC must continue to make reducing systemic and operational risk a top regulatory priority.  The SEC’s proposal on naked access is a good first step, but exchanges must also be directed to impose universal pre-trade risk checks.  If left solely in the hands of individual broker-dealers, a race to the bottom might ensue.  We simply must have a level playing field when it comes to risk management that protects our equities markets from fat fingers or faulty algorithms.  Regulators must therefore ensure that firms have appropriate operational risk controls to minimize the incidence and magnitude of such errors while also preventing a tidal wave of copycat strategies from potentially wreaking havoc in our equities markets.
 
Fifth, the SEC should act to address the burgeoning number of order cancellations in the equities markets.  While cancellations are not inherently bad – potentially enhancing liquidity by affording automated traders greater flexibility when posting quotes – their use in today’s marketplace is clearly excessive and virtually a prima facie case that battles between competing algorithms, which use cancelled orders as feints and indications of misdirection, have become all-too-commonplace, overloading the system and regulators alike.
 
According to the high frequency trading firm T3Live, on a recent trading day only 1.247 billion of the 89.704 billion orders on Nasdaq’s book were executed – meaning a whopping 98.6% of the total bids and offers were not filled. Cancellations by high frequency traders, according to T3Live, were responsible for the bulk of these unfilled orders. 
 
The high frequency traders that create such massive cancellation rates might cause market data costs for investors to rise, make the price discovery process less efficient and complicate the regulators’ understanding of continuously evolving trading strategies.  What’s more, some manipulative strategies, including layering, rely on the ability to rapidly cancel orders in order to profit from changes in price.  
 
Perhaps excessive cancellation rates should carry a charge.  If traders exceed a specified ratio of cancellations to orders, it’s only fair that they pay a fee.  The ratio could be set high enough so that it would not affect long-term investors (even day traders), and should apply to all trading platforms, including dark pools and ATSs as well as exchanges.  
 
The high-frequency traders who rely on massive cancellations are using up more bandwidth and putting more stress on the data centers.  Attempts to reign in cancellations or impose charges are not without precedent. In fact they have already been implemented in derivatives markets where overall volume is a small fraction of the volume in cash market for stocks.  The Chicago Mercantile Exchange’s volume ratio test and the London International Financial Futures and Options Exchange’s bandwidth usage policy both represent attempts to reign in excessive cancellations and might provide a helpful model for regulators wishing to do the same.   
 
Finally, the high frequency trading industry must come to the table and play a constructive role in resolving current issues in the marketplace, including preventing manipulation and managing risk.  In order to maintain fair and transparent markets and avoid unintended consequences, market participants from across the industry must contribute to the regulatory process.  I am pleased that a number of responsible firms are stepping forward in a constructive way, both in educating the SEC and me and my staff.  I look forward to continue to working with these industry players.
 
Mr. President, we all must work together, in the interests of liquidity, efficiency, transparency and fairness to ensure our markets are the strongest and best-regulated in the world.  But we cannot have one without the other – for markets to be strong, they must be well-regulated.  So with this reality in mind, I look forward to working with my colleagues, regulatory agencies, and people from across the financial industry to ensure our markets are free, credible and the envy of the world.
 
Mr. President, I ask that links to some of the studies and reports I have mentioned be included in the record.

• www.qsg.com
  • “Liquidity Charge® & Price Reversals: Is High Frequency Trading Adding Insult to Injury?” February 11, 2010
  • “Beware of the VWAP Trap,” November 11, 2009
• http://www.themistrading.com/article_files/0000/0519/THEMIS_TRADING_White_Paper_--_Latency_Arbitrage_--_December_4__2009.pdf
• http://www.itg.com/news_events/papers/AdverseSelectionDarkPools_113009F.pdf
• http://www.aitegroup.com/reports/200912141.php
• http://www.asx.com.au/about/pdf/20100211_review_algorithmic_trading_and_market_access.pdf
• http://www.chicagofed.org/digital_assets/publications/chicago_fed_letter/2010/cflmarch2010_272.pdf

Mr. President, I yield the floor.
 

Comments

[AnonymousMonetarist]

Ode to high frequency trading:
nihoncassandra.blogspot.com

In simpler times, when Atari elicited a "wow" instead of a "huh?", this might have been achieved by leaving sizable but plausible limit, stop or MOC orders, then pulling them to see who and how much was leaning on them. Now more cunning is required. Torquemada-like fear and surprise. Spoof them back. Double-spoof. Triple spoof. In size. BOOOO!! Randomize. Maybe relax or eliminate all restrictions upon transaction etiquette, thereby allowing trade with oneself to paint the tape as required shake the parasites, perhaps leveling the playing field. Or merely to make it more like quick-sand. Declare all-out war so anything goes. Let - no, encourage the 'bots to fight and predate each other. "Greetings, Mr Anderson". Is this real or are we in the Matrix??! Hunt the hunter. "Kill the bear" as Anthony Hopkins rallied in The Edge. "What one man can do - another man can do!!" Heck, it's not about investing anymore (if ever it was) - its about winning the game, and as this time of The Quickening approaches, an algorithmic battle to the death, an epic battle cannot be far behind, leaving in its wake, RAIDs, bandwidth, over-educated Russians with no scrupples, recursive bloodshed and, then yes, SkyNet...

[Fritz]

Good read.

However, once the finance lobbyist crowd works their magic, it will be right back to business as usual.

[Bob]

"Burgeoning" is an interesting term.  Makes me wonder just what percent of trades get cancelled, by whom, and under what conditions. 

[Anonymous]

Wall Street: Powered By Bullshit.

One way maybe the best way to argue HFT is to shut them all off for a week. Then extrapolate the data....what data?
The market will be real humans involved in real transactions.

[mynhair]

Funny, no mention of GS or their profits or the inability of 'reversion to the mean' to apply in their case.

[peterpeter]

> According to the high frequency trading firm T3Live, on a recent trading day only 1.247 billion of the 89.704 billion orders on Nasdaq’s book were executed – meaning a whopping 98.6% of the total bids and offers were not filled. Cancellations by high frequency traders, according to T3Live, were responsible for the bulk of these unfilled orders. 
 

Given that these orders are priced in penny increments, a 1.4% hit rate on orders adding liquidity to the books is not bad at all.  The HFTs would of course love for more order flow to hit their standing orders.  I suspect however that isn't what the folks at Themis and Senator Kaufman have in mind.

The high cancellation rates are a result of changes in internal calculations of fair value (i.e. a movement in a related asset), or in response to another order for the same equity (i.e. another market maker has placed a higher bid than you, so you then place a yet higher bid).

How Senator Kaufman believes this fails to aid in price discovery is a mystery to me.  Any order - whether it is hit or cancelled helps in price discovery.

There's nothing wrong with trying to put a tax on each order cancellation (albeit a very very small one - perhaps under 0.0001 per share), but the net result will be wider spreads, as the costs of those cancellations get factored into the same algorithm that is calculating fair value.

So, one would expect a decrease of in order cancellation rates, and an increase in spreads (the sizes of which would be determined by each strategies average order cancellation rate per equity and the order cancellation cost).

That in turn would somewhat hurt price discovery, as a zone of uncertainty in terms of the market's view of value bound by a standing bid and ask would widen.

There has never been a single tier of access to markets and there never will be....  Unlike the past however when premium access was restricted to a select group of exchange members, today's markets are a meritocracy, and anyone with $20K to spend on a fast server and the programming expertise to do something with it can play.  The costs of co-locating a 1U server in NASDAQ is probably cheaper than the costs per month to operate this blog.

[TimeToChange]

I'd describe at least some HFT practices as more of a kleptocracy than a meritocracy.  It's true there has never been a single access tier and there never will be.  In the old days there were privileged intermediaries with a lot of advantages over other market participants, and today there are advantaged participants, though it's an advantage of a different kind.  Participants today pay for their advantage through capital investment in technology, but they're advantaged nonetheless. 

The difference between the old days and today is regulation.  In the old days, privilege came with at least some responsibilities, including meaningful quoting requirements and negative and affirmative obligations.  Today, anything goes.  If an HFT firm needs to rebalance, it jams the stock up or down until its risk model comes back into equilibrium; regulators don't know what those risk models are, don't study their effects, and may not even know HFT firms deploy them in the first place.  In exchange for the transient liquidity HFT firms provide, and provide only when they are close to or in equilibrium, we pay the price of the volatility they induce and the illiquidity they cause while they're rebalancing. 

Meritocracy?  Hardly.  They rip through the market with impunity.

[Bob]

The argument seemed to rest on "anyone with $20K to spend on a fast server and the programming expertise to do something with it can play." 

Like, in another universe, maybe!  Not the most desireable one, that "Play Universe," but surely a different one than ours. 

[Anonymous]

Ok...humor me...a better than 98% cancellation rate is an aid to price discovery and liquidity ?

[peterpeter]

Why not?

If XYZ last traded some time ago at 100.0 and more recently had a bid of 99.97 and an ask of 101.03

Moments later,

HFT-1 places a bid at 99.98 making a new national best bid
HFT-2 places a ask at 101.02 making a new national best ask

If neither of those orders execute, the best that you can say is that XYZ is worth somewhere between roughly 99.98 and 101.02.

However without those offers, the zone of uncertainty is 2 pennies wider.

Having orders not hit is itself valuable information towards price discovery.

The cancel rate for the quotes at the NBBO means nothing in terms of price discovery... but the closer both sides of the quote get (i.e. the narrower the bid/ask spread), the clearer the price becomes - regardless of whether any of those orders ever execute.

[Anonymous]

HFT's do not blindly improve on the NBBO. Doing so involves risk. Their strategies all revolve around performing (almost) risk free arbitrage. In your example the typical real buyer or seller might have experienced price improvement in times before HFT. Nowadays the order is scalped as the HFT mechanically routes the order to someone else. Massive cancellations are necessary because the liquidity provided is an illusion. Taxing very high cancellation rates seems like an elegant regulatory approach.

[Anonymous]

In a vacuum you may be on to something. In the real world, your so-called bids and offers are micro second illusions. There is no depth, no substance and thus no market. Let us also take the viewpoint of the non HFT market participant. Why are they being forced to use the tools of HFT to execute ? A cancellation rate of 98% hurts everyone. Have you ever had to execute large orders in this environment ? Real orders, not some bullshit model-driven nonsense ?

[Anonymous]

Solution: eliminate information.

[Anonymous]

T3 is a prop trading firm, they have no high frequency businesses. I am impressed by this mans knowledge of the HFT space, usually left to the prop traders and quants, its refreshing to see that normal people are beginning to understand that the primary goal of HFT, is to A) Collect exchange rebates B) Cost traders and institutions pennies, nickels and dimes on order execution and to C) add up all those pennies, nickels dimes and rebates into a very profitable business.
There is something to be said about being the smartest in trading, but most of the edge these HFT firms gain is through special access or co-location, so its hard to say that they actually are any smarter than any other participants, if the market has any volatility, they suddenly are the dumbest participants.
I believe it is a business that creates little if any overall value in markets, and I think if you have ever seen a move like SWM a few weeks ago you understand fully what it means when these alogs they are running don't understand a move. Simply put, they go completely haywire and actually cause total liquidity loss almost immediately. It is literally a lie to call them liquidity providers because when there is a move of real substance or one side of a trade cannot be taken, they completely freak and cause moves that are so far out of whack it's almost unbelievable.
I remember when these things got shut off at points in late 2008, you were seeing real liquidity spreads, where real people were willing to take real positions, not trying to make a penny on a 100 or 10 millisecond advantage and I'm sure many of you remember how different the market was when WAL-MART had a 2 dollar spread or STT was trading with a 50 cent spread and ripping in a 80 point range intraday. Needless to say, it was a much much different market, but it was a real market, and unfortunately we trade in a largely artificial market today.
It is this artificiality that poses the greatest threat to public investment in the markets. It is simply guaranteed that the public will be holding the bag when this game of musical hairs ends.

[Anonymous]

It's clear you have no idea what you're talking about when it comes to how HFT desks make money - but there's really no reason to correct you.

On your other point, it's true HFT makes money from traders and institutions, but institutions have always paid to execute - and used to pay more than they do now. Previously, you had specialists, more broker-dealers and pit traders - specialists alone made more money than HFT.

Also, according to Elkins McSherry, which specializes in trade cost analysis for institutions and publishes an annual survey in Institutional Investor found that transaction costs (commissions, fees, and market impact) have plummeted in the HFT era - over the last 8-10 years. Overall, they have dropped 35% and dropped the most in markets that have the highest HFT prevalence (50-60% in us equities) and the least in markets that have the least HFT prevalence (~10% in asian equities, where HFT only represents 5-10% of volume).

[Anonymous]

Since things are so "clear" to you, why don't you factor in the associated IT and consulting costs with todays environment. Thanks to all of our "progress" we now need more consultants, computers, and software to tell us how we are doing. Specialists made money because they took risks , provided real liquidity, were held to standards , and helped to quarterback trades.

[Anonymous]

You're reference to Elkins McSherry's study is all over the place on zerohedge. Let me shine some like for you on this subject, since, I'm taking it you are not a specialist in measuring implicit costs of trading. It all really boils down to a single question here: (1) how is Elkins 'measuring' these costs of trading? While I do not know of the specific study you're referencing, I can tell you one thing with certainty: Elkins is using 'average price order' data and comparing these prices to benchmark prices. Which benchmark is Elkins using in the study you're referencing? That makes a huge difference. 'Costs' of trading when measured by an average price-benchmark differential can be extremely different when using different benchmarks. Are they using Arrival Price? If so, do they have the timestamp data on when the decision on the orders were made? Are they using the Open as the benchmark or the Previous Night's Close? Are they using, god forbid, some type of VWAP or participation-weighted benchmark? Using any type of volume-weighted price derived benchmark to measure 'costs' of trading is like determining your 'costs' of goods sold by making average comparable estimates of your goods to the costs of goods sold of other firms in the same industry during the same time-frame. Not sure how in-line that methodology is with US GAAP or IFRS accounting standards.

The point I'm trying to make is that, while the trading industry has moved miles per millisecond in innovation over the last decade, most TCA (transaction cost analysis) methodologies being used today are methodologies that were also being used in the 1990s.

You should check out recent reports from Quantitative Services Group (QSG). They have innovated in the measurement of trading costs with a tick-based methodology in which every individual fill executed in an order is matched to the trade and quote data, whereby each fill's true 'cost' of obtaining liquidity is measured and accumulated throughout the order. This cumulative liquidity charge value is thereby separated from the competing order flow's price changes in the name, making it very apparent what the "cost" of transacting was. This methodology can identify situations where an institutional-sized order in a participation algo has found liquidity on the other side to have dried up suddenly and thus 'liquidity charge costs' to continue to participate increase drastically. Their most recent report pairs these exact situations with short-term price reversals of 10bps+, bringing into question the quality of liquidity these participation algos are interacting with. Please post the link to the Elkins study when you see a chance to do so.

[Anonymous]

So now "high frequency trading" is an industry! People still think that they are socially useful and give great contribution to economic growth and they do not need to be taxed heavily because otherwise it is the consumer to pay and markets would become less liquid and efficient. Gosh!
Can somebody show simply that high frequency trading is a zero-sum game and create no wealth or improve any economic condition apart from the one of the successful trader and its company?
http://mgiannini.blogspot.com/2010/03/make-finance-industry-to-pay.html

[mark456]

ucvhost is a leading web site hosting service provider that is known to provide reliable and affordable hosting packages to customers. The company believes in providing absolute and superior control to the customer as well as complete security and flexibility through its many packages. cheap vps Moreover, the company provides technical support as well as customer service 24x7, in order to enable its customers to easily upgrade their software, install it or even solve their problems. ucvhost offers the following different packages to its customers.