Evidence of Insider Trading before September 11th Re-examined

by Paul Zarembka
Department of Economics
State University of New York at Buffalo
September 9, 2011

International Hearings on the Events of September 11, 2001
September 8-11, 2011, Ryerson University, Toronto, Canada

http://torontohearings.org/

This report addresses evidence of insider trading before September 11th, sometimes referred to by a broader phrase, informed trading.   Insider trading refers to using private knowledge of an anticipated event in order to profit financially by engaging in financial market transactions.  In the first weeks after September 11, 2001 a number of financial publications called attention to substantial insider trading in put-options occurring before the attacks.  Some of these early examples have been surveyed in Zarembka (2008, pp. 64-66, 69-71), while this book chapter also commented on certain exaggerations (e.g., an incorrect doubling of the put-option volumes).  Quickly, commentary died out.

Purchasing a put option entitles the owner to sell a stock at a contractually stated price, the “strike price”, any time until the contract expires.  If the market price of the stock goes down below the “strike price”, the owner of the put-option can buy the stock (if not already owned) and simultaneously sell the same stock at that “strike price”, making a profit if the cost of the option itself does not exceed the net revenue.

This report deals with evidence of insider trading only.  It does not deal with speculations.  Nor does it deal with certain open questions about financial issues surrounding September 11th that otherwise deserve investigations: 

• Large increases in the M1 money supply in the United States have been reported for July and August 2001 and explanations have been sought. 
• Huge financial transactions have been reported to have taken place at computers at the World Trade Center minutes before the attacks. 
• Selling short (borrowing a stock and selling it, then returning it later through purchasing). 
• Markets outside the United States. 
• Disappearances of gold and securities from the World Trade Center. 
• The specific financial firms directly hit by planes, and the financial investigations sabotaged by the WTC or Pentagon attacks. 
• Insurance payoffs, particularly to the owner of destroyed buildings, particularly to Larry Silverstein.

This is not a complete list of issues deserving investigation and our contribution can be considered modest at best, compared to these additional questions.  However, some of the above seem to have only one testimonial behind them.  And we are addressing what the 9/11 Commission, however imperfectly, addressed.

In 2004, the 9/11 Commission Report (http://www.9-11commission.gov/report/911Report.pdf, p. 499, fn. 130) stated that the government’s investigations had produced no evidence of insider trading before the attacks.  Yet, it offered little of its evidence to the public.  When a FOIA request asked for the documentary evidence behind that Commission footnote we just mentioned, the SEC even replied on December 23, 2009 that “the potentially responsive records have been destroyed” (http://maxkeiser.com/wp-content/uploads/2010/06/FOIAresponseGIF1.gif).  Such a response is all the more curious, given documents we report below that were made public on January 14, 2009, documents that would have provided at least a partial response to the request made.

• A scholarly article by Poteshman (2006) in the Journal of Business, using econometric modeling, claimed high probability of insider trading for American Airlines and United Airlines put-option purchasing shortly before September 11th.   Were they random, the American purchases had only a 1% probability of occurrence; the United Airlines put-option purchasing was less improbable, but on September 6 had only a 4% probability of occurrence (both measures obtained by comparisons of the airline values reported on p. 1720, Table 4, to the benchmark values reported on p. 1713, Table 1).

Since the government had provided so little evidence of its position, some sharp criticism and reference to Poteshman’s results ensued (e.g., Griffin, 2005, pp. 52-57 and Zarembka, pp. 67-69).

Limited SEC Investigative Evidence Made Public

On January 14, 2009, two post-September 11th SEC investigating memos were finally made publicly available. ¹`
  The simpler one, prepared on May 11, 2004 for the 9/11 Commission, stated that the volume of put-option trades for United on September 6, 2001 (for a $30 strike price with expiration on October 20, 2001) had been erroneously reported in the SEC data: the correct value should have been 1500 – i.e., for 150,000 shares – not 2000.  The memo explained that the SEC had missed the actual cancellation of an intended 500 put sale (included, but not a purchase).  The Option Clearing House (OCC) had the correct number (http://media.nara.gov/9-11/MFR/t-0148-911MFR-00138.pdf ²

 Still, judging by the reported change the next day in open interest (open interest is the amount of the put contract remaining unexercised), a 500 purchase did indeed occur on September 7.   In other words, a volume of 2000 occurred over two days, not one day (1500, then 500).  This would not seem to affect Poteshman’s work since he used the change in open interest for his measure rather than volume data, but it does raise a general concern about the SEC data.  A volume level of 2000 for the first day does appear both in Zarembka (p. 66) and in Chesney, et al. (2010, p. 35, Table 2) and is implicitly retained in the Commission’s own report despite that 2004 memo it had received (Zarembka, p. 68).

Much more important, another memo was simultaneously released that had been prepared back on September 17-18, 2003 and had named the Options Hotline newsletterand its editor Steve Sarnoff as responsible for faxing on September 9, 2001 some 2000 subscribers the recommendation to buy put options on American Airlines (http://media.nara.gov/9-11/MFR/t-0148-911MFR-00139.pdf, p. 14).  The memo further stated that the SEC interviewed 28 people who purchased these options and 26 had said that they had done so because of the newsletter.  This memo reported 27 additional subscribers, not interviewed, as additional purchasers of that put option. 

The same memo went on to report that an unnamed large institutional investor in hedge funds had undertaken the 2000 United put-option purchases – i.e., for 200,000 shares – but was explained away by the fact that the same investor had also purchased 115,000 shares of American stock on September 10.  This information does appear in the Commission’s report, p. 499, fn. 130.

A third memo for the 9/11 Commission, this one dated April 24, 2004, reported an interview with Ken Breen, Deputy Chief, Business and Securities Fraud Section, Department of Justice.  It is also important. It reports that Breen “was not sure about potential trading in index futures (because the volume is so great that analysis proved impossible)” (http://media.nara.gov/9-11/MFR/t-0148-911MFR-00074.pdf).  In other words, the exhaustive governmental investigation was not so exhaustive after all, by its own admission. 



A Contradiction and Its Deepening
Through the above, we have arrived at a contradiction: an econometric result of high probability of insider trading in American Airlines stock and somewhat less for United stock goes against the government’s September 2003 memo (released in 2009).  This contradiction could be resolved by simply claiming that Poteshman himself never claimed certainty and then to assert that an event of low probability had been all that had occurred.  Were such simplicity available! 

Two other econometric studies have since been added to the scientific literature.

• Wong, et al. (2011) examined put options on the S&P 500 and found additional econometric evidence of insider trading before September 11th, now for the index the government had reported in April 2004, “analysis proved impossible”.   Furthermore, the result is significant because some have speculated that option trading was heightened in the period before September 11th because of a falling market.  As we will explain below, Wong obtained their results, even after trying to account for a falling market.

• Chesney, et al. (2010, yet unpublished) have studied about 1.5 million put-option trades for 14 companies: 5 airline companies including American and United, 5 bank stocks, and 4 others, for the period of January 1996 to April 2006.  They report, with high probability, informed trading before September 11th in each of the put options for Boeing, Merrill Lynch, J.P. Morgan, Citigroup, United, American, Bank of America, Delta, and KLM (ordered here from the highest calculated gains downward).

In sum, ten financial instruments, including the S&P 500 option, are each exhibiting, with high statistical probability, evidence of insider trading before September 11th, sometimes more than once.  American and United are identified by separate methodologies, seven additional companies are identified by Chesney, and the S&P 500 is identified by Wong.  The joint probability of all of these being nothing more than random outliners seems astronomically low.

The government, however, had deepened its position.  In that September 17-18, 2003 memo, the SEC refers to investigations of “103 companies and 38 index products and broad-based funds” (p. 3).  It finds no evidence of any insider trading.  It dismissed dramatic comments shortly after September 11th, even by a person as well positioned as the German Central Bank President Ernest Welteke (pp. 10-12).  The report definitively concludes with the SEC’s lead investigator Joseph J. Cella, III, Chief of Market Surveillance, Division of Enforcement, SEC, saying that “he has no questions about any trade and is confident there was no illicit trading pre 9/11 in the United States” (p. 19).

The sharp contradiction between the scientific results and the government’s position is too great to ignore.  Can it be resolved?  On the one hand, are three distinct econometric methodologies implemented with option trading data each erroneous in some manner?  Is the competence of the econometricians, including two articles having been screened through peer review evaluations, to be in question?  On the other hand, if the SEC is accurately reporting the motivating factors about American and United put-option purchases, could the SEC be erroneous about many or all of the other financial instruments for which no evidence has been made public?  Having repeatedly said that the attacks were a complete surprise, has the government been influenced to avoid acknowledging any insider trading before September 11th?  Worse, is it aware of insider trading and is it lying?

Analysis: The SEC Evidence regarding One Named Financial Advisor

The tip that Steve Sarnoff, editor of Options Hotline, offered subscribers on September 9 for placing put options on American Airlines is reported by Mike Williams. ³   Nothing appears unusual with the recommendation itself.  From the government’s memo, somewhat more than 50 of 2000 subscribers seem to have acted upon the recommendation – i.e., about two and one-half percent.  Nothing appears unusual with this outcome.  The 1312 change in open interest on September 10th (Chesney, p. 35, Table 2) represents an average of a bit less than 26 put-options purchases per subscriber who purchased, representing 2600 shares each. Yet, consider the implications of taking this at face value. 

Joe Duarte, another financial advisor, lists ten newsletters dealing with option trading (www.joe-duarte.com/free/directory/options-newsletters.asp).  Options Hotline does not happen to be one of them, perhaps suggesting that Sarnoff has no dominance.  Search the web and get many more.  Recommendations are being made by newsletters daily, weekly, monthly.  If two to three percent of subscribers are following recommendations to buy put options on stocks, we should see many, many examples similar to what occurred for American on September 10, 2001.  Therefore, what happened that day for American would be a rather common event, not a very unlikely one.  That volume on American put options would not have shown up as unlikely, as a statistical matter. 

Absent informed trading, newsletters should be nothing more than instruments, rather than causes, of these market behaviors being analyzed.

It is not only American, but, as we have seen, nine other put options showing spikes before September 11th.  Chesney find only 37 such examples in a decade of some 1.5 million pieces of put-option data on fourteen companies, 13 of which were related to September 11th.   These spikes should have been innocent of ex post shock events because spikes are to be always expected in random statistical outcomes.  Instead, most are centered prior to shock events.

Two Caveats

Let me put one consideration to rest.  Some critics of the 9/11 truth movement, such as Kay (2011), claim that the entire movement is filled with people who go down a rabbit hole, never willing to leave it.  In this case, the suggested claim could be that Sarnoff himself should be added to a conspiracy about 9/11, added as soon as the government released in January 2009 its evidence as to who made what recommendation and with what effect regarding American on September 10.   Such an approach would address the contradiction we have identified.  But it would be at the expense of having no evidence for such an assertion. 

We wish to stay with evidence, evidence from the econometricians, the government, and anywhere else obtainable.  In other words, we wish to push into the contradiction.

Regarding evidence we have to be careful.  For Boeing, Mike Williams, seeking to expose myths among skeptics of the official story of September 11th,, cites a Dutch article of September 11, 2006 placed on the site physics911.net (www.911myths.com/index.php/Put_Options#Boeing, accessed August 4, 2011).  This article had made only a tangential mention to this airline manufacturer, thus representing no more the proverbial “straw man” – a data source is not even provided.  Williams then provides a news report referring to one analyst’s public downgrading of Boeing on September 7th, apparently being unaware that put-options purchases cited by Poteshman were on United occurred on September 6th (as well as in Chesney as we shall see below).⁴

Among known skeptics of the official 9/11 story, discussion of option transactions at a site hosted by Jim Hoffman (http://911research.wtc7.net/sept11/stockputs.html, accessed August 4, 2011) is embarrassingly out of date, mostly refers to 2001 stories, even though claiming an August 2007 update.  Kevin Ryan (2010), followed by Mark Gaffney (2011) whose book Black 9/11 is due for release shortly, each attempt a recent survey, but seem unaware of the Commission documentation on insider trading released in January 2009.

Discerning Evidence of Insider Trading before September 11th

We now proceed to describe each of the econometric studies, the first being single-authored and the following two being triple-authored.  The two first have been peer-reviewed and published in well-established journals, the third is planned for submission shortly, and has been a lengthy work-in-progress. 

The article by Poteshman in the Journal of Business well describes the problem at hand, and is applicable to all three works.  Hinting at the end about his two-pronged approach that we will mention later, Poteshman writes that, in general, option market activity

is motivated by a number of factors such as uninformed speculation (i.e., noise trading), hedging, trading on public information, and trading on private information.  Consequently, when a statistic obtains a value that is extreme relative to its historical distribution, one can infer that there was an unusual amount of activity related to one or more of the option trading motivations.  Although the statistics do not distinguish between trading motivations, if an extreme value is observed just before an important piece of news becomes public, then it is reasonable to infer that there was option market trading based on private information rather than a shock to the trading from one of the other motivations.  Indeed, the fact that the statistic has obtained an extreme value indicates that a shock to trading from another motivation would have to be unusually large to account for the observed option market trading.  Of course, it is possible that the typical option trading from the other motivations varies systematically with changes in the state of the option or underlying security market. This is the reason that conditional as well as unconditional distributions for the statistics will be computed in the next section. (Poteshman, 2006, pp. 1711-12)

Analysis: The Econometric Evidence in Poteshman

Each of the papers we cite has reference to data as quantiles.  Quantiles are defined by the accumulation of the probabilities of occurrences of a random variable.  A quantile at 50% would say that one half of the occurrences of random variable had already occurred over the frequency distribution and one half have yet to occur.  A quantile at 95% would say that 19 out of 20 occurrences of the random variable had already occurred with 1 in 20 yet to occur; a quantile at 99% would report 99 out of 100 had already taken place.  Thus, an event at a quantile of 95% would be rare, while at 99% would be quite rare.

Poteshman’s work examines several measures for the probabilities of insider trading occurring, while addressing market options for American Airlines, United Airlines, the index for airline stocks, and the S&P 500.  I summarize the easiest one of three to understand and the one he seems most comfortable exploring, i.e., the evidence regarding volumes of put-option contracts.  This is measured by the change in a contract’s open interest from one day to the next day (purchases less sales less exercises of options) compared to the average of such change measured by a 126 trading day period, going backwards in time from 22 trading days before the date in question.  This is also normalized for the standard deviations of those 126 trading days.  The statistical results for the four trading days before September 11th are reported in Table 1.

Table 1: Put-Option Market Volume Statistics before September 11th

Volume Statistics	Sept. 5	Sept. 6	Sept. 7	Sept. 10
AMR	-.02	.08	.65	3.83
UAL	-.12	1.45	1.23	.15
Airline Index	-.13	.63	.66	.85
S&P 500	-.07	.25	.54	-.09

                               Source: Poteshman (2006, pp. 1720, Table 4)

Poteshman compares these AMR and UAL statistics to his benchmark data for the 1,000 largest market capitalization firms for the dates from January 2, 1990 through September 4, 2001.  Compared to the historical record of the large companies, the AMR datum for September 10th in the table has only a 1% probability of occurrence and the UAL datum of September 6 has a 4% probability of occurrence.  The airline index datum for September 6th has a 6% probability of occurrence and the S&P datum for September 7th, a 5% probability.

Poteshman also considers a four-trading day interval in addition to the daily values we report in Table 1.  For those who consider this measure to be more appropriate, probabilities are somewhat less unlikely.  In any case, the above results are not conditional upon any underlying factors and so Poteshman also introduces four conditioning factors, “total option volume, the return on the underlying asset, the abnormal trading volume of the underlying asset, and the return on the overall stock market” (p. 1716).  He undertakes quantile regressions for these four factors.  He obtains very similar results.

Analysis: The Econometric Evidence in Wong, et al.

The article by Wong et al. (2011, pp. 7-9) has the most detailed discussion of executions of option trading generally.  It then undertakes a complex statistical investigation regarding S&P 500 option trading before September 11th, centering not only upon put options – whether they were purchased in-the-money (above the market, and thus costing a higher price), at-the-money (at the market level), or out-of-the-money (below the market).  They also consider the type of strategy used, and also call options.  Calls are the contractual right to purchase stocks for a determined “strike price” before an expiration date.  They are a less obvious strategy for anticipating a decline in an asset price.

Wong, et al., first contrast the 2001 period for contracts expiring on September 22, 2001 with the same September expiration in 2000, both being in declining market environments: for the period between January 1 and June 30, 2000, the S&P 500 declined 15 points, while for the period January 1 to June 30, 2001, the S&P 500 declined 96 points.  They consider these as “control periods” (pp. 15-16).  They find that “the trading volume for the SPX index put options during the control periods [i.e., January 1 to June 30] is not significantly different between 2001 and 2000 … the years 2000 and 2001 being similar in regard to option activity in a time period before intense trading began in September index put options” (p. 37).

Furthering the comparison of 2001 to 2000, they examine a short sub-period after the S&P 500 closed at 1134 on August 31 to September 10, 2001 when it closed at 1093, a decline of 39 points in five trading days.  A year earlier, the S&P 500 had closed at 1518 on August 31, 2000, while on September 8 (September 10 was a Sunday) it had closed at 1495, a decline of 23 points, also in five trading days.  They find that “the mean and the standard deviation of the trading volumes for September 2001 contracts were more than double those for September 2000 contracts during sub-period [September 1 to September 10] for both call and put, but not so much during the other sub-periods” (p. 20).  They also find many more extreme volumes in the 2001 period.

They then investigate the various types of puts and calls able to be utilized, and also consider alternative strategies.  Studying statistical results, they conclude that their findings show

a significant abnormal increase in the trading volume in the option market just before 9-11 attacks in contrast with the absence of abnormal trading volume far before the attacks. This only constitutes circumstantial evidence that there were insiders who tried to profit from the options market in anticipation of the 9-11 attacks. More conclusive evidence is needed to prove definitively that insiders were indeed active in the market. Although we have discredited the possibility of abnormal volume due to declining market, such investigative work would still be a very involved exercise in view of the multitude of other confounding factors e.g. coincidence, confusing trading strategies intentionally employed by the insiders, noises from the activities of non-insiders. (p. 44)

They do not claim definitive results, but rather significant statistical evidence of insider trading.  Their procedure attempts to abstract from the fact of a declining overall market, an element that some have attempted as a basis for a counter-claim against evidence of insider trading.  A counter-argument could begin by observing that August was an up-market in 2000 and down-market in 2001.  We have not yet found relevant evidence about option market behaviors in down markets compared to up markets, so we merely state that no presumptions should be made without evidence.

Wong et al. do not attempt to compare their results with Poteshman’s regarding the S&P 500.  Poteshman had noted the fact that “the option volume on SPX options was more than 100 times greater than that on either AMR or UAL options.  Consequently, it would be much more difficult to detect an option market bet …” (p. 1723).  Wong, et al., concur: “any 9-11 insider would not trade directly the airline options in large volume to avoid drawing attention after the 9-11 attacks”.  Since Wong et al. is a much more elaborate analysis for the S&P 500, we are entitled to infer that theirs is a more discriminating analysis.

Analysis: The Econometric Evidence in Chesney, et al.

Chesney, et al., offer the most detailed pieces of evidence.  To understand what they are offering, first consider American Airlines (AMR) for the authors’ time period of January 1996 to April 2006.  It is representative of their general methodology and they provide details for this particular example of so much concern within the 9/11 movement.  Technical details are placed in footnotes as we move toward  our Table 2 presentation of their results.

Chesney, et al., start with the 137,000 AMR put-option contracts, i.e., about 54 possibilities per trading day over the ten-plus years of data they analyze (about 250 trading days per year).  They first identify for each day that put-option contract across strike prices and expiration dates with largest increment in open interest.  These 2560 data points are the more unusual.  After accounting for intraday speculation , they record the prior two years of data for each time t, beginning in January 1998 and ending in April 2006.  This leads to a measure qt, that denotes, for date t, the frequency such a value occurred based upon the prior historical record.  They are referred to as probabilities(p. 9).  For the AMR option on September 10, as one example, qt is reported as 1.2% (p. 35).  This reflects 6 occurrences in the two years of 500 data points on and before that day.  Generally for their study, qt must refer to data that had occurred no more than 5% of the time, i.e., no more than 25 times for the prior two years . 

A second, additional criterion attempts to account for hedging transactions – buying a put option to guard against a fall in an existing stock position, or buying stock to guard against a fall in a put-option’s value as the stock rises.  They offer a rather complicated procedure, not elaborated here.  The two criteria, as they report, reduce the considered spikes for AMR put options down to 141 instances (p. 17), still fairly considerable.

Instead of stopping here, as a third criterion, Chesney, et al., focus upon the most profitable, using ex post information on the behavior of the stock price.  Let rt be the option’s return at time t.   The maximum return over the available contracts after time t is then represented by rtmax.  AMR on September 10th had a put-option contract price of $2.15 for a $30 strike price and October 20, 2001 expiry.  The maximum gain for that contract therefore turned out to be that exercised on September 17 as the stock price fell to a $18 close and the option price rose to $12, a $12 gain per contract on $2.15 invested, or 453%.  That particular rtmax is reported in Chesney (p. 35, Table 2; also, p. 22 with a typo of 458%). 

Now, this third criterion is formulated as a pair of conditions that are presented here in a footnote .  The introduction of this third criterion leads to only 5 incidents for AMR: May 10 and May 11, 2000, August 31 and September 10, 2001, and August 24, 2005, rather than 141 without this third criterion.

For the entire set of fourteen companies studied, only 37 incidents are identified: 13 spikes identifiable before September 11th as reported in Table 2, 14 associated with earnings announcements (all but 4 beforehand, 2 on same day, 2 after), 6 associated with mergers and acquisitions (4 beforehand, 2 same day), and 4 not identified.  In other words, spikes are being shown to relate to real events, most frequently anticipating them. 

The gains from exercising put options, reported in Table 2 for the 13 identified cases of informed trading before September 11th, do not depend upon the econometric procedure, but rather are factually based, close to the maximum possible.  As seen, American purchases on September 10th are by no means the most profitable.  The Merrill Lynch put-option purchase generated almost four times the subsequent gains as that for American.  The extensive put purchases for Boeing were even more profitable.  It could have been background for Sarnoff’s September 9th recommendation to his subscribers regarding American.  That is, an option advisor’s knowledge of prior airline put-option purchases by others may be suggestive for his or her own recommendations.  If the advisor turns out to be wrong, an explanation is at hand: “I was not alone”.  In other words, evidence on American such as Poteshman’s and Chesney, et al.’s may not itself represent of insider trading at all.  If correct, those other put options trades require the particularly careful investigations.  Indeed, if a person actually had prior information about what was to happen on September 11th, why would he or she engage in put-option purchasing for the most obvious of choices, American and United, and subject himself or herself to easy detection?

Table 2: Evidence of Informed Put-Option Purchases before September 11th

Put Option	2001 Date	Change in open interest	Gain from exercising the put options	Proxy for probability as an informed trade
Boeing	29 Aug	2828	$1,972,534	0.998
Boeing	5 Sep	1499	1,805,929	0.998
Boeing	6 Sep	7105	2,704,701	0.998
Merrill Lynch	10 Sep	5615	4,407,171	0.998
J.P. Morgan	30 Aug	3145	1,318,638	0.998
J.P. Morgan	6 Sep	4778	1,415,825	0.998
Citigrroup	30 Aug	4373	2,045,940	0.998
United	6 Sep	1494	1,980,387	0.998
American	31 Aug	473	662,200	0.984
American	10 Sep	1312	1,179,171	0.998
Bank of America	7 Sep	3380	1,774,525	0.994
Delta	29 Aug	202	328,200	0.998
KLM	5 Sep	100	53,976	0.998

                               Source: Chesney, et al. (2010, p. 35, Table 2 and p. 38, Table 4)

What the prior paragraph is suggesting is that spikes in put-option purchases are not independent events, but, in actuality, can be interrelated.  We would thus have to pull back from a conclusion of having ten random pieces of data and then claiming the joint probability of their occurrences to be “astronomically low”.  The joint probability would still be very low, but not “astronomically low”.  In this case, Boeing put-option purchasing moves to the center of attention, not just for the magnitude of profits reported in the table, but as possible background leading to put-option purchasing on American a few days later.

As to the United put-option purchasing, the SEC reports that it was related to a large stock purchase of American stock by the same investor.  Poteshman did not find the option purchase to be highly improbable on a random basis.  Chesney, et al.’s procedure for delimiting hedging transactions would not capture such an example of purchasing American stock while also purchasing put-options on United. 

In sum, we are willing to accept the SEC’s reporting about American and United Airlines and not consider them to represent evidence of insider trading.  Nothing else do we accept, as these have been the only pieces of evidence on the issue of insider trading put forward publicly from SEC investigations.  Specifically, Boeing as well as Merrill Lynch, J.P. Morgan, Citigroup, and Bank of America deserve careful attention as a result of Chesney, et al.’s work.  Nevertheless, we have been warned regarding what can be held back by the government and then released years later (cf. American).  Should the movement for discovering the truth about September 11th jump too quickly with accusations, it can backfire. 

The total gains without United and American reported in Table 2, and then also including other individual stocks not yet analyzed by Chesney, should fall short of $30 million in total.  This level is reported in order to keep in mind the maximum potential of insider put-option trading benefits before September 11th.  We are not claiming that that much actually occurred.  Insider trading could have occurred in individual stocks as Chesney, et al., find, and also served as unsuspecting background to investors and their advisors for United and American put-option purchases. 

Boeing, Merrill Lynch, J.P. Morgan, Citigroup, and Bank of America

For Boeing, Merrill Lynch, J.P. Morgan, Citigroup, and Bank of America we have been unable to find public news motivating large put-option purchases for them before the dates found in Chesney, et al.’s research.  Note that the cited downgrade of Boeing (see http://community.seattletimes.nwsource.com/archive/?date=20010908&slug=boeing08, accessed August 20, 2011) came after the dates cited. 

In any case, research work by Chesney, et al., fails to suggest spikes in put-option trading occurring merely due to rating changes by analysts of corporations.  Indeed, 33 of theirs are associated with September 11th, or earnings announcements, or mergers and acquisitions; only 4 remain unidentified.

Concluding Recommendations

• United and American ought not be recognized as direct insider trading.  Nevertheless, by themselves, they convey little of the larger question. 
• Ken Breen, Department of Justice, reported to the Commission in another interview released in 2009 that, for put-options on indexes, “the volume is so great that analysis proved impossible”.   Therefore, Wong’s result regarding the S&P 500 is not contested in the background reports to the 9/11 Commission, despite what the Commission asserted.  This fact needs to be asserted.
• To the best of my knowledge, none of the three econometric methodologies and results has been contested in the professional literature.  Typically, controversial results generate opposition.  We have attempted to convey enough of three separate methodologies to be convincing that they are solid scientific works. Therefore, accept the econometric research results as meritorious.
• Demand that the SEC publicly report the details of its findings on Boeing, Merrill Lynch, J.P. Morgan, Citigroup, Bank of America, Delta, and S&P 500 index put-option trading before September 11th.  This reporting should be at least as detailed as that already released for American and United.
• Add to that demand of the SEC any additional corporations exhibiting evidence of insider trading before September 11th, e.g., in the expanded material Chesney, et al., plan to release shortly.

        In addition:

• Promote an independent investigation into the events of September 11th, inclusive of subpoena powers, that includes investigations of put-option purchasing.
• Incorporate into that independent investigation the financial issues we have cited, but not examined, in our introduction, most of them having billions of dollars at stake.

References

Arvedlund, Erin E. (2001), “Follow the Money: Terrorist conspirators could have profited more from fall of entire market than single stocks”, Barron’s, October 8.

Chesney, Marc (2010), Remo Crameri, and Loriano Mancini, “Detecting Informed Trading Activities in the Options Markets”, April 15, 2010, at SSRN: http://ssrn.com/abstract=1522157.

Gaffney, Mark H. (2011), “Black 9/11: A Walk on the Dark Side” (Second in a series), Foreign Policy Journal, March 2, at www.foreignpolicyjournal.com/2011/03/02/black-911-a-walk-on-the-dark-side-2/0/, accessed August 5, 2011.

Griffin, David R. (2005), The 9/11 Commission Report: Omissions and Distortions, Northampton, MA: Interlink.

Kay, Jonathan (2011), Among the Truthers, Toronto: HarperCollins.

Poteshman, Allen M. (2006), “Unusual Option Market Activity and the Terrorist Attacks of September 11, 2001”, Journal of Business, Vol. 79, pp. 1703-1726.

Ryan, Kevin (2010), “Evidence for Informed Trading on the Attacks of September 11”, Foreign Policy Journal, November 18 , at www.foreignpolicyjournal.com/2010/11/18/evidence-for-informed-trading-on-the-attacks-of-september-11/, accessed August 5, 2011.

Wong, Wing-Keung, Howard E. Thompson, and Kweehong The (2011), “Was there Abnormal Trading in the S&P 500 Index Options Prior to the September 11 Attacks?”, Multinational Finance Journal, Vol. 15, no. 1/2, pp. 1–46 at http://mfs.rutgers.edu/MFJ/Articles-pdf/V15N12p1.pdf.

Zarembka, Paul (2008), “Initiation of the 9-11 Operation, with Evidence of Insider Trading Beforehand”, The Hidden History of 9-11, P. Zarembka, editor, New York: Seven Stories Press, 2nd edition, pp. 47-74 (1st edition, Amsterdam: Elsevier Press, 2006).

¹ The January 14, 2009 date was reported to this author on July 25, 2011 by Kristen Wilhelm of the Center for Legislative Archives as follows: “The 9/11 Commission's Joseph Cella Memoranda for the Record were scanned and uploaded to the NARA Archival Research Catalog for the opening of the 9/11 Commission records on Jan. 14, 2009.”  See also the NARA prior notice of the general opening at  www.archives.gov/press/press-releases/2009/nr09-41.html.  It may be of some interest that attention was drawn to this release on the very day of January 14 (see the January 15 posting at http://screwloosechange.blogspot.com/2009/01/more-on-911-put-options.html, and its link to postings on January 14 at http://forums.randi.org/showthread.php?t=132904&page=2, both accessed August 12, 2011.  Thanks are offered to “lapman” and “Mike W” – the latter being presumably Mike Williams given his later reference to this release).  Williams receives further attention in the course of this article.

² Another 75 trade is also cited for another contract not in contention.
³ This recommendation was for the put-option contract with a $30 strike price to expire on October 20, 2001.  It read as follows:
September 9, 2001

Vol. 12, No. 28

Stocks Skid On A Jump In The Jobless Rate. This Week, We Take To The Air

This past week, stocks were pressed to the downside – with the highlight being Friday's blue chip decline. Wall Street was surprised by a spike, to a four-year high, in the jobless rate. And the market took its lumps. This week, I see opportunity for you to have fun and profit with an airline play. So, without further ado, here's…

This Week's Option Recommendation

Buy the AMR October $30 put for $170 [100 shares for $1.70 per share, P.Z.], or less, good this week.

Shares of AMR Corp. trade on the New York Stock Exchange under the symbol "AMR". The symbol for this option reco is "AMRVF". American Airlines closed the week at $30.15. The 52-week range for AMR is $27.62-$43.93. My downside price target is $22-$26.

The major airline is under pressure. At $25, each $30 put would have $500 of intrinsic value. If AMR is at or above $30 on the third Friday in October, your option will expire worthless. That is your risk. Set your stop-loss at $100, to preserve capital, in case my expectations go awry.

That's buy the AMR October $30 put for $170, or less, good this week.

– www.911myths.com/index.php/Put_Options#Options_Hotline , accessed 719/2011
⁴ Within the same discussion, Williams cites many reports of put-option volumes without those using accurate data.  Some reported data are about double the actual levels, presumably due to author errors in understanding Optionmetric data which considers the buy and sell sides of one transaction to be distinct.  If one is going to criticize, focusing upon those arguing for insider trading using correct data to make their cases seems preferable.
“The main motivation for considering increments in open interests is the following. Large volumes do not necessarily imply that large buy orders are executed because the same put option could be traded several times during the day.  In contrast large increments in open interest are originated by large buy orders. These increments also imply that other long investors are unwilling to close their positions forcing the market maker to issue new put options.” (Chesney, et al., pp. 8-9)
In order to abstract from intraday speculation, they compare daily changes in open interest to the reported volumes of transactions (the difference between the two should be small).  In other words, purchases are to dominant, with sales or exercises of options small.
This calculation could seem to suggest 103 times in eight and one-quarter years beginning in January 1998.  But a stock like AMR stock price fell considerably from April 2002 to a low of $1.25 within one year thereafter implying much higher volumes then required for similar dollar option positions. 
Actually, AMR closed at $17.90 on both September 21 and 27 before the October 20 option expiration; the $18.00 on September 17 was not quite the lowest.  However, presumably the option price was the highest on September 17.
Let Gt be the cumulative gains achieved through the exercises of the selected option in the shortest time available from the day of the calculated maximum up to ten trading days thereafter. Chesney’s third criterion is then offered as a pair of conditions for the option trade in question, that is,

•  rtmax ≥ q0.90(rtmax)

and

• Gt ≥ q0.98(Gt).

The quantiles at day t for the rtmax  and Gt distributions – q0.90(rtmax ) and q0.98(Gt) – are computed using the preceding two years of data.  These criteria are the quantiles for the top 10% of initial profiting and top 2% of total gains.


"Kindness is the language which the deaf can hear and the blind can see." ~Mark Twain

http://www.ithp.org/articles/septemberinsidertrading.html