Tag Archives: HFT

Ceresney Presents Unconvincing Defense of Increased SEC Administrative Prosecutions

The SEC has been besieged by criticism of the Division of Enforcement’s new determination to bring more antifraud civil prosecutions in the SEC’s administrative court, and was stung by a speech from Judge Jed Rakoff criticizing that decision in no uncertain terms.  The day after Judge Rakoff’s talk, Andrew Ceresney, the current Director of the Enforcement Division, tried to defend the decision.  He plainly needs more practice; or perhaps he could start with a better understanding of his own administrative proceedings.  The arguments he made were so off-base that they can be explained in only two ways: he either doesn’t know very much about defending SEC enforcement actions (which I hope is not true), or he is just presenting makeweight arguments in the hope that by mere repetition they might be believed.

In a PLI conference panel discussion the day after Judge Rakoff’s speech at the same conference, Ceresney attempted a rejoinder to the Rakoff criticism.  Essentially, he argued that the SEC has no comparative advantage when it litigates in front of its own administrative law judges.  You can see a description of his comments here.  I don’t mean to be dismissive, but that’s a bunch of hooey.

He argued that respondents in these proceedings are not disadvantaged by the fact that they get no discovery, have limited subpoena power controlled by the ALJ, have very little time in which to prepare a case, are unprotected by the rules of evidence, and have no independent (that is, independent of the SEC, the prosecutor in these cases) review of the sufficiency of the “evidence” that gets presented in the administrative court.  His arguments are, to be kind, unconvincing.

Discovery.  Ceresney says the lack of discovery doesn’t matter because the SEC delivers all of its files to the other side.  That’s bunch of malarkey; a fiction; a mere pretense.  The SEC delivers to respondents a part of the official enforcement “case file,” but that may not include vast amounts of important information the SEC has bearing on the claims asserted.  For example, say the SEC were to bring an action alleging a company violated the antifraud provisions of the securities laws because its release of material information used a process that allowed certain investors to get a slightly faster data feed.  The enforcement division is investigating such potential charges in connection with high frequency and algorithmic trading right now.  The documents the SEC would deliver to respondents would be only the SEC’s own formal case file.  It would not include any information about the SEC’s own distribution of information, and how it may be doing the same thing (releasing information in a way that benefits certain traders) that it is charging as a fraud.  Such information would likely be obtainable in discovery in a federal court, but would not be delivered by the SEC in an administrative case and almost certainly would not be ordered to be produced by an administrative law judge (ALJ).  The SEC gains from the choice of an administrative court; the respondent loses.

Another example.  The SEC brings fraud charges against a public company alleging an improper accounting treatment to account for certain expenses.  The accounting approach, however, is in a gray area and is common in the industry.  The SEC’s enforcement file will include only its inquiry into the respondent’s accounting determinations.  What it will not include is information the SEC gathered about other industry participants, which would be relevant to determining the “generally accepted” approach in that industry.  Because that information is plainly relevant to the issue of scienter — whether the respondent knowingly or recklessly chose an incorrect accounting treatment — it would likely be obtainable in discovery in a federal district court, although the SEC would surely contest it.  But the chance of obtaining such information through an order by an ALJ is close to zero.  Once again, the SEC gains from being in the administrative court; the respondent loses.

Apart from the narrower scope of documents the SEC must deliver to respondents in an administrative proceeding, Ceresney ignores the SEC’s inevitable, overbroad assertions of work-product and “deliberative-process” privilege.  The SEC does not deliver to opponents documents in the “case file” over which they claim work-product or deliberative-process protections.  Often, these are the most important documents in the file.  In courts, the refusal to provide those materials in discovery can be (and has been) overruled, but the ALJs don’t do this.  For example, the SEC may have gathered 5 million pages of documents in electronic form.  They examine those materials for months, or more likely years, and identify the ones they think best support their case as well as ones that are harmful to their case.  When the “case file” is delivered, all the respondent gets is the 5 million pages (other information being withheld under claim of work product protection), and a short time before trial to try make head or tail out of them.  The SEC identifies as its exhibits the documents it found to be helpful and hopes the documents harmful to its case remain buried in the 5 million pages.  Federal courts can, and have, ordered that the SEC provide documents in the form they are kept in the SEC’s files (that’s what the Federal Rules of Civil Procedure require) — including files identifying what documents may be useful or harmful in certain areas.  ALJs are highly unlikely to do this.  The SEC gains; the respondent loses.

Likewise, the SEC will not deliver key interview material if there is no official transcript, claiming work-product privilege.  SEC lawyers often call potential witnesses during their investigation to learn what they might have to say on relevant issues.  They take notes of those interviews.  Some of those witnesses are subpoenaed for investigative testimony, which is transcribed.  But many of those potential witnesses are not subpoenaed for testimony, often because what they told the SEC lawyers is not helpful to prosecuting a case.  (The SEC investigative lawyers’ bias in creating an investigative record is another huge problem, but not the subject of this discussion.)  The SEC delivers the investigative transcripts as part of the case file, but keeps for itself all non-transcribed witness interviews.  As a result, evidence helpful to a respondent remains hidden.  In a federal court, discovery will likely allow the defendant to get copies of these untranscribed interviews, but an ALJ will never issue such an order.  The SEC gains; the respondent loses.

Ceresney says: “There is no discovery in criminal prosecutions, and I don’t think anyone claims due process violations there.”  That is wrong in many respects.  First, one justification for this is the considerably higher burden of proof the government faces in criminal cases.  Second, there are forms of discovery in criminal cases and judges to determine whether requested discovery should be permitted.  Third, criminal procedure rules mandate that materials obtained by prosecutors be delivered to defendants if they are exculpatory or can be used to impeach witnesses.  That is what is known as so-called “Brady material” (after the Supreme Court case Brady v. Maryland).  But the SEC, in its civil prosecutions, refuses to comply with Brady v. Maryland.  For administrative proceedings, there is a strange provision in the SEC Rules of Practice that lists documents the SEC must disclose, and those it may withhold, and then says the SEC may not “withhold, contrary to the doctrine of Brady v. Maryland, 373 U.S. 83, 87 (1963), documents that contain material exculpatory evidence.”  SEC Rule of Practice 230.  But the staff takes a narrow view of what is “exculpatory,” and does not look for exculpatory material outside of its own investigative file.  That, combined with the fact that the SEC administrative law judges are highly restrictive on what they consider “exculpatory,” means that much evidence that could be used by the defense to disprove SEC claims is never made available.  See, e.g., In the Matter of Thomas C. Bridge et al.,  File No. 3-12626 (Aug. 27, 2007).  When that happens in a federal court, a defendant at least has the chance to gather that same information in the civil discovery process.  But in the administrative court the SEC gets to try to bury evidence that a respondent can use to defend himself or herself.  The result?  You guessed it: the SEC gains; the respondent loses.

So much for Ceresney’s claim that the lack of discovery in administrative proceedings doesn’t harm respondents because the SEC delivers its case file.

ALJs as securities law experts.  Ceresney says the SEC administrative courts are better for these cases because they involve highly technical securities law issues that judges and juries just can’t understand as well as ALJs with securities law expertise.  To begin, there is no requirement that an SEC ALJ have securities law experience when he or she gets appointed.  Many of them learn on the job.  Beyond that, however, there is a kernel of truth in Ceresney’s argument, but not much more.  Over time, SEC ALJs may develop expertise in areas involving technical regulatory rules and industry practices.  But that expertise applies only to enforcement actions against SEC-regulated entities like broker-dealers or investment advisers.  The rules governing those businesses are detailed and non-intuitive, so previous experience with these cases can be helpful.  That is why for many, many years the only cases brought by the SEC in its administrative courts were those involving its own regulated entities.  But the issue raised by Judge Rakoff involves not those cases, but the broader, higher impact fraud cases like, for example, insider trading cases.  As Judge Rakoff noted, trying those cases in administrative courts is a relatively new phenomenon.  And because the SEC has resisted any clear statement of what are and are not fraudulent practices, and fraud cases that get tried (and not settled) often explore the outer edges of the fraud concept, ALJs have no meaningful technical edge on federal courts.  To the contrary, judges and juries are far more capable of applying the concept of fraud than ALJs are likely to be because in such case, technical skills are far less important than more intuitive understanding and application of common sense and community standards.

Timing advantages.  Ceresney says it is no great advantage that the SEC has been able to take months or years to prepare its case and the defense has no such luxury.  He argues that the defense has probably been doing its own investigative work during that period.  But defense counsel have no power to get people to talk to them or deliver documents.  When the SEC lawyers call people subject to SEC regulatory power, say they represent the U.S. government, and ask them to chat off the record, you can bet they get cooperation that defense counsel are unlikely to get.  That is especially true if the people contacted are subject to SEC investigation themselves.  SEC lawyers are not shy about playing on the fear of prosecution to encourage people to talk freely with them.  In fact, the Enforcement Division touts its policy of giving credit to cooperators.  Defense lawyers have no such leverage.  They get their leverage only through the use of subpoena power in a federal civil action, where ample time is allowed for discovery before trial.  Theoretically, defense lawyers can get subpoenas in administrative cases, but the time-frame is much shorter, making use of this process much more difficult.  And ALJs are reluctant to approve the issuance of subpoenas, each of which requires their approval, which almost always must come over SEC objection.  Finally, as noted above, defense lawyers also get hit with large volumes of materials a short time before trial.  The SEC has had months and years to examine and choose the few documents that favor their case.  The defense lawyers are left trying to scour those documents for the ones harmful to the SEC case, which the SEC has intentionally left in the haystack.  Try examining 5 million pages of electronic materials to prepare a case in a month or two.

Due process.  Cereseny says the respondents get great due process with multiple levels of protection. But the only meaningful levels of review of the evidence come from the SEC itself.  The ALJ is an SEC employee and deals every day with the SEC litigators but only rarely with each set of defense lawyers.  The ALJ’s ability to consider even complex cases is limited by the SEC, which sets time limits on the completion of decisions.  Once the ALJ has ruled, the only avenue of appeal is to the SEC itself.  Yes, that’s the same folks who decided to bring the case in the first place.  The Commissioners are supposed to take off their prosecution hats and don judicial hats.  Perhaps in some bizarre utopia this is viewed as due process, but in the real world, prosecutors don’t do so well at deciding they were wrong to bring a case (and incur the major expense in doing so).  And beyond that, the Commissioners will typically defer to the fact-finding of ALJs because they weren’t there to see and evaluate witnesses (like an independent judge or jury is in a federal court case).  After the SEC approves its employee’s decision, the respondent finally gets a chance to have an independent review  — only, of course, if he or she has the financial resources to continue this quest against the government, and many do not.  Even then the court of appeals that reviews the SEC decision is required to defer to SEC judgments unless it can find them unreasonable.  So the value of independent review is diminished by imposing a high bar before the SEC decision can be rejected.

All of those layers of protection touted by Ceresney are more like layers of inquisition.  The expense is enormous.  The review is limited.  The likelihood of reversal is small.  And the only independent review is done under standards that shackle the reviewers.  Compare that to being before an independent judge for the pretrial process and then having an independent judge and jury hear and consider the evidence.  No contest.  Once again, in the administrative process, the SEC wins and the respondent loses.

Despite his statements to the contrary, Ceresney knows the truth here.  He himself admitted it when he announced the new SEC policy of administrative prosecutions.  He said that the rationale for the SEC’s use of its captive courts is that it has a higher success rate in prosecutions brought there.  See here.  The SEC’s improved success rate is not because the administrative courts are fairer, it’s because the playing field is tilted in the SEC’s favor.

Straight Arrow

November 11, 2014

Contact Straight Arrow privately here, or leave a public comment below:

How the SEC Helps High Frequency Traders (Redux)

The Wall Street Journal reported previously on a new study revealing that the SEC was providing a special advantage to high frequency traders (HFT) by allowing them to get information filed with the SEC more quickly than most investors.  The Journal recently updated its earlier report here, and tells us that although the SEC appears to have reduced that advantage following the publication of this information, it still retains a 2-3 second timing advantage for early recipients of its data.

For those of you interested in the study, which reveals SEC conduct similar to possible similar arrangements in the private sector that the SEC Enforcement Division is investigating as potentially fraudulent securities trading, a copy is available here: How the SEC Helps Speedy Traders, a Columbia Law School Millstein Center for Global Markets and Corporate Ownership paper written by Robert J. Jackson, Jr. and Joshua R. Mitts of the Columbia Law School.

Straight Arrow

November 10, 2014

Contact Straight Arrow privately here, or leave a public comment below:

SEC Found To Be Assisting High Frequency Traders

As they say, you couldn’t make this stuff up!

The Wall Street Journal reported today (October 29, 2014) that a new study confirms that the Securities and Exchange Commission (SEC) is feeding information to high frequency traders (HFT) before it goes to the general investment public.  What?  Could you say that again?

Yes, it’s true.  Your SEC — the U.S. government agency tasked with assuring a fair and efficient securities marketplace — provides special information access to favored investors, including the much ballyhooed high frequency traders.  Read about it here in this article: Fast Traders Are Getting Data From SEC Seconds Early.

The Journal reports that:

Hedge funds and other rapid-fire investors can get access to market-moving documents ahead of other users of the Securities and Exchange Commission’s system for distributing company filings, giving them a potential edge on the rest of the market.  Two separate groups of academic researchers have documented a lag time between the moment paying subscribers, including trading firms, newswires and others, receive the filings via a direct feed from an SEC contractor and when the documents are publicly available on the agency’s website….

When a company submits a document, the contractor forwards it to the Edgar subscribers and to the SEC website “at the same time,” according to the SEC. But the studies suggest that the SEC website can take anywhere from 10 seconds to more than a minute to post the documents, giving an advantage to the Edgar subscribers or their customers, who are often professional investors. Mom-and-pop investors can download the documents from the SEC website, but the information may already be known to others in the market, the studies indicate….

[A] forthcoming paper will document that investors could make about several cents a share, on average, on market-moving filings by receiving it in advance of those who rely on grabbing the document from the SEC website.

Straight Arrow

October 29, 2014

Contact Straight Arrow privately here, or leave a public comment below:

SEC HFT Enforcement Action Looks Nothing Like “Flash Boys”

On October 16, 2014, the SEC entered into an administrative settlement of an enforcement proceeding involving a high frequency algorithmic trader — Athena Capital Research, LLC.  But if you review the Administrative Order, which is available here (In re Athena Capital Research), don’t expect to see anything like the fanciful musings of Michael Lewis in “Flash Boys.”  In fact, there is no suggestion that high frequency trading might be unlawful outside of the complex, labyrinthine alleged manipulation scheme described in the SEC’s charges.

I won’t bother to try to describe the charges here because they are so complex.  But they involve an algorithm of multiple high speed buy and sell transactions at the close of the trading day designed to create a favorable price increase for Athena Capital in the after-closing auction.  The important thing is that the SEC did not assert that the problem was high frequency trading (HFT), but that HFT was a vehicle for pursuing “a familiar, manipulative scheme: marking the closing price of publicly-traded securities.”

SEC Chair Mary Jo White made it clear that it was the manipulative conduct that was key: “When high-frequency traders cross the line and engage in fraud we will pursue them as we do with anyone who manipulates the markets.”  SEC officials in the division of the SEC tasked with examining trading systems and practices previously expressed skepticism about Michael Lewis’s accusations and follow-on proceedings by New York Attorney General Eric Schneiderman.  See SEC’s Berman Says Critics Like Schneiderman Misjudge Regulator and SEC official suggests order cancellations not currently a problem.

Another interesting aspect of this proceeding is that the settlement occurred based on the respondent neither admitting nor denying the charges — meaning the SEC did not have the leverage to insist on an admission of liability — and, although there was a $1 million penalty imposed, the SEC did not seek any “disgorgement” of alleged unlawful profits — probably meaning they had a hard time proving significant profits actually occurred.

Straight Arrow

October 21, 2014

Contact Straight Arrow privately here, or leave a public comment below:

Does High Frequency Trading (HFT) “Rig” the Market?

Michael Lewis’s “Flash Boys” purports to reveal a system of “market-rigging” by the intervention of high frequency traders into executions of market orders.  Using some of the same sources that he used to source his highly misleading book “The Big Short,” he created a “sky is falling” atmosphere by promoting his book as the revelation of rigging the market.  He portrays a supposed fraud on ordinary traders by high frequency traders because of their ability to step in front of a trade and execute it for themselves at that price and then resell to the trader with a miniscule mark-up, thereby gaining huge profits when these miniscule mark-ups are repeated millions of times.

This has been labeled “insider trading” by many uninformed people, but it almost surely is not, because the high frequency traders do not appear to possess material non-public information when they trade.  I have not yet seen evidence suggesting they were engaged in committing any form of fraud, or breaching any fiduciary or fiduciary-like duty of confidentiality through the process by which they gain access to trade order information before it gets from the exchanges to market-makers responsible for trade execution.  They certainly do not have material non-public information about the public company whose stock is being traded, which is the standard fare of insider trading violations.

Whether this is a form of “front-running” is another issue.  Front-running is a description of trading that occurs when someone learns about an impending trade order, knows that this order will “move the market price,” and steps in front of the order to make a trade before the order hits the market (and moves the price), then usually pockets the profit by selling at the new market price.  This typically would be seen only when people have access to confidential order information involving large block orders of a stock that could conceivably move the market price enough to allow for a front-running profit after costs of order execution are taken into account.  What Lewis describes as high frequency trades are not this type of front-running because they do not involve large block, market-moving, trades.  And it is not at all clear, as compared to front-runners who misuse confidential access to trading information for their own benefit, whether the high frequency traders are engaging in an impropriety at all, since, once again, there is no apparent source of a duty of confidentiality they would be breaching.

Finally, the “market rigging” aspect of HFT would seem to be a hyperbolic, publicity-grabbing description, at best.  The focus appears to be on cumulative profits allegedly gained by the high frequency traders, but the actual impact on specific trades is so small as to be immaterial to the person who entered the order.  How many of us, or even institutional traders, care very much whether our trade is executed at $1 per share or $1.001 per share?  To be sure, the SEC should be making  efforts to try to assure best execution of trade orders, but the real question to ask is what costs we are willing to incur to prevent  that $.001 mark-up from occurring?  In some instances, analysts appear to be showing that there could be an issue of phantom liquidity that may be worth pursuing, not because it is a fraud, but because it may be allowing incorrect liquidity expectations to impact market executions.  I’ve included one interesting such analysis in the “securities markets” links to right (Nanex.net analysis of HFT trade cancellations), which appears to show that HFT sometimes results in unreliable reports by exchanges of trade offers because of high frequency traders cancelling offers before executions can occur.  Whether spending, and causing others to spend, hundreds of millions of dollars or more to address such an issue in investigations and “settlements” of enforcement actions addressing what appear to be lawful practices may make little sense.  Moreover, if this is an issue, it would appear to be a regulatory one, not an enforcement one, and the SEC’s experts on market practices, not the enforcement division, should lead the way.  The politically-oriented state Attorneys-General with no market expertise and no authority to regulate markets should be excluded from the process altogether.

The main problem here is a common one that people like Michael Lewis, many commentators, and, unfortunately often the SEC itself, get in a huff when financial market participants find even lawful ways to make a lot of money.  They seem to forget that the entire market process is, and must be, driven by the ability to garner lawful profits from trading, because that is what creates market efficiency, which is the driver of the market system.  The huge ruckus created by “The Big Short” was just such a mistake.  The fact that some people made a lot of money by betting on the housing market to decline, and huge institutions lost a lot of money by failing to protect themselves against a massive decline in housing values, is what markets are all about.  The short trades Lewis described typically were not unlawful (the contrary verdict in the Fab Tourre case was, in my view, a travesty, because it is plain that the counter-party was not defrauded in any meaningful respect).  They were either very good bets based on informed judgments about over-exuberance in housing markets, or, in many cases, very good luck when short positions put on as a hedge became wildly, and unexpectedly, profitable.  The SEC wasted millions of taxpayer dollars investigating these matters for more than five years, and caused billions of dollars of expenses and “settlements” for no good reason other than to punish financial institutions for allowing a marketplace driven by huge, well-informed, institutional investors to function as intended.

Whether the same is true for the now-blossoming HFT investigations will remain to be seen, but the hype and hyperbole attendant to the investigations and commentating on this issue suggests we are moving down a similar path.

Straight Arrow

July 20, 2014

Contact Straight Arrow privately here, or leave a public comment below:

High Frequency Trading, and Proof that the SEC Approach to Insider Trading is Completely Wrong

Some interesting thoughts by Mark Cuban on the SEC’s approach to high frequency trading and insider trading.

blog maverick

Got to love Mary Jo White, the Chairwoman of the SEC.  While Michael Lewis’s book Flash Boys was getting all the headlines and was the topic of some of the best television  on CNBC, ever, Ms White used the firestorm to ask for more money for the SEC.

Shocking ? The only shock would be if she didn’t use any occasion the SEC was in the public eye to ask for more money. It is unfortunate because there is no greater waste of money than what the SEC spends trying to enforce  insider trading laws.

Let me give you some examples of just how poorly the SEC manages our tax dollars when it comes to insider trading:

1. Did you hear the one about Gary and Clif of the Florida East Coast Railroad ? Gary and Clif noticed that there were a lot of tours of the company…

View original post 768 more words