3 - OVERSTOCKCOM INC a Delaware corporation et al Plaintiffs v MORGAN STANLEY An_Redacted.pdf -

Please enable JavaScript to view the full PDF

OVERSTOCK.COM, INC., a Delaware corporation, et al.,..., 2011 WL 11684655... 2011 WL 11684655 (Cal.Super.) (Expert Report and Affidavit) Superior Court of California. San Francisco County OVERSTOCK.COM, INC., a Delaware corporation, et al., Plaintiffs, v. MORGAN STANLEY & CO., Incorporated, et al., Defendants. No. CGC-07-460147. November 7, 2011. Declaration of Michael A. Manzino in Support of Plaintiffs' Opposition to Defendants' Motion for Summary Judgment or, in the Alternative, Summary Adjudication Name of Expert: Michael A. Manzino Area of Expertise: Accounting & Economics >> Securities & Investments Representing: Plaintiff Jurisdiction: San Francisco County, California Theodore A. Griffmger, Jr. (SBN 66028), Ellen A. Cirangle (SBN 164188), Jonathan Sommer (SBN 209179), Stein & Lubin LLP, Transamerica Pyramid, 600 Montgomery Street, 14th Floor, San Francisco, California 94111, Telephone: (415) 981-0550, Facsimile: (415) 981-4343, [email protected], [email protected], [email protected], [email protected], Attorneys for Plaintiffs, Overstock.Com, Inc., Keith Carpenter, Olivier Cheng, Fern Bailey and Wendy Mather, as Co-Personal Representatives of the Estate of Mary Helburn, Elizabeth Foster, Hugh D. Barron, David Trent, and Mark Montag. Judge: Hon. John E. Munter. Date: January 9, 2012 Time: 9:30 a.m. Dept: 305 Action Filed: February 2, 2007 Trial Date: March 5, 2012 I, Michael A. Manzino, hereby declare as follows: 1. I make this declaration of my own personal knowledge, except to the extent-otherwise stated, and, if called as a witness, I could and would competently testify to the matters contained herein. Professional Background and Expert Qualifications and Experience REDACTED 1 OVERSTOCK.COM, INC., a Delaware corporation, et al.,..., 2011 WL 11684655... 2. I worked on Wall Street for more than twenty years and was employed by Morgan Stanley for almost fourteen years, from 1994-2007. All of my work at Morgan Stanley was on its securities lending desk, and I was promoted over the years from trader, to Vice President, to Executive Director. From 2001-2007 I was the second-in-command of the securities lending desk for the United States and had the title of “Executive Director” at Morgan Stanley. I was involved in all functions of a securities lending desk, including borrowing stock from lenders, lending out stock to clients, dealing with other securities lending desks at other prime brokers, and directly managing our largest hedge fund relationships. In mid-2005, my duties evolved to focus on maintaining relationships with hedge fund clients for securities lending, as well as pitching new potential hedge fund clients, instead of working on the day-to-day borrowing and lending of stock. However, I still had intense direct contact with the entire securities lending desk. Prior to working at Morgan Stanley, I worked at Paine Webber (from 1989-1994), L.F. Rothschild (from 1987-1989) and Drexel Burnham Lambert (1983-1987) on their securities lending desks. 3. At Morgan Stanley, I was responsible for managing a book of more than $100 billion in domestic equity securities to ensure an adequate supply of stock to support short selling by Morgan Stanley clients. Accordingly, I had to be familiar with all aspects of borrowing stock from the various lenders that would lend securities to prime brokers such as Morgan Stanley, as well as all aspects of how Morgan Stanley would in turn lend the stock to prime brokerage clients who sold short, primarily hedge funds. I also was involved in lending securities to Morgan Stanley's internal trading desks. I have experience in all facets of the securities lending business. 4. The securities lending industry is a very close-knit industry. Even though the dollar amounts involved can be enormous, the number of people working on a securities lending desk is not very large. By my estimation, domestic securities lending desks at prime brokers would rarely exceed 50 employees in size and would generally be significantly smaller than that. Persons employed in the securities lending industry frequently interact, both professionally and socially. For example, I would regularly have lunches, dinners and casual drinks with securities lending employees working at various prime brokers, as well as with traders and portfolio managers at hedge funds. In addition, for many years there has always been an annual securities lending conference that lasts for a week, customarily in Boca Raton, Florida. It was my usual practice to attend these conferences, where many securities lending managers from the leading prime brokers would gather and discuss industry issues. These conferences featured both formal presentations on specific topics related to securities lending and informal discussions. It is a unique business in that you must work closely with the very firms with which you are competing fiercely for business. In this way I became knowledgeable with the workings of our competitors' lending desks, and the industry as a whole. 5. In the 2004-2007 time frame, the securities lending desks at Morgan Stanley and Goldman Sachs were direct competitors, were considered the two leaders of the securities lending industry and shared a substantial number of clients who maintained prime brokerage accounts at both Morgan Stanley and Goldman Sachs, including such large short-selling hedge funds as Maverick Capital and Kingsford Capital. I regularly worked with large hedge fund clients who needed the securities lending desk to borrow shares to support their hedge fund's short sales, including many hedge funds who also had accounts at Goldman Sachs. For a stock that was hard to borrow, like Overstock, the hedge fund would pay a borrow fee to the firm known as a “negative rebate.” Thus, hedge fund clients frequently would want to know whether Morgan Stanley had an available supply of a hard- to-borrow stock like Overstock, and, if so, the cost of the borrow fee (the negative rebate). Availability and rates would also be a topic of discussion directly between brokers such as Morgan Stanley, Goldman Sachs and others. In the ordinary course of business, for a hard-to-borrow security like Overstock, brokers would often call or email each other to see if the other broker had any supply available and, if so, what the rate was. 6. Just as Morgan Stanley and Goldman Sachs (and Merrill Lynch) compete in lending securities to short sellers, they also compete in borrowing securities from the various sources of supply (which in turn will be lent to short-selling clients). Lenders of securities to prime brokers include such large custodian banks as Bank of New York, State Street and Northern Trust. Lenders of securities to prime brokers also include direct lenders, such as a mutual fund like Vanguard. Lenders of securities also include other prime brokers like Morgan Stanley, Goldman Sachs and Merrill Lynch, and these brokers can be important sources of supply when they possess a large pool of customer securities that can be lent. 2 REDACTED OVERSTOCK.COM, INC., a Delaware corporation, et al.,..., 2011 WL 11684655... 7. My opinions are based on my more than twenty years of experience in the securities lending industry, and the specialized knowledge, professional experience and training that I acquired over those years. My opinions do not rely on and do not attempt to analyze all of the deposition testimony and exhibits, though I did review a large number of Goldman and Merrill depositions and their exhibits. My purpose in reviewing the depositions was to gain some general familiarity and background with the issues in the case as well as to assure myself that I agreed with the merits of Plaintiffs' lawsuit as I understand it. Engagement in this Case 8. In genera], I was asked to provide opinions concerning the normal, ordinary workings of a securities lending desk in major clearing firms like Goldman Sachs and Merrill Lynch based on my more than twenty years of experience working on securities lending desks, including fourteen years at Morgan Stanley. As someone who started as a trader and worked his way up to the level of second-in-command on the domestic securities lending desk at Morgan Stanley, I am familiar with all of the workings of a securities lending desk, from both the perspective of a higher-level manager and a lower-level employee. 9. I was asked whether it was typical industry practice for large prime brokers like Goldman Sachs, Merrill Lynch and Morgan.Stanley to have fails-to-deliver in a stock like Overstock that would exceed a million shares. I was also asked whether it was typical industry practice for clearing brokers to maintain a fail-to-deliver position in the Continuous Net Settlement System (“CNS”) for a year or more. I generally understand that Goldman Sachs and Merrill Lynch have multiple accounts within CNS, as did Morgan Stanley. I also generally understand that Goldman Sachs and Merrill Lynch operate through various corporate entities. For purposes of my assignment, the corporate structure of Goldman Sachs and Merrill Lynch had no relevance. I was considering the nature of large, extended fail positions in CNS accounts, not the responsibility of a particular corporate entity for any actions leading to those CNS positions. Moreover, when I spoke with the securities lending desk of another broker, the other broker would answer the telephone “Goldman” or “Merrill,” without referring to the specific name of a corporate entity. When borrowing or lending shares to another prime broker, the identity of the prime broker's CNS account number (also referred to as the DTC number) was relevant, not the corporate name. Ordinarily, Morgan Stanley's transactions with Goldman Sachs would flow in or out of Goldman's CNS 005 account while Morgan Stanley's transactions with Merrill Lynch would flow in or out of Merrill's CNS 161 account. 10. I was also asked whether a securities lending desk would have an awareness of which stocks were the subject of high interest among securities lending clients and why some stocks would develop a very high interest among securities lending clients. My practical experience includes frequent, daily communications with short-selling clients, primarily hedge funds, which occupied a large part of my time in the 2004-2007 time frame. Virtually every day I would be continuously on the phone with lenders on behalf of our hedge fund clients seeking to locate a very limited number (usually 15-20 stocks that stayed relatively constant) of very hard-to-borrow securities like Overstock. Based on my years of experience in the securities lending industry, including my regular interactions with short-selling clients, I developed an understanding of why a limited number of stocks became very active, very hard-to-borrow securities. That short list of stocks was the focus of a securities lending desk. It was our job to know everything about these stocks as it related to our firm's needs, including the supply and rates available in the marketplace, and also to know about any fails, recalls, buy-ins, etc. related to the stocks that may affect our firm or any corporate entity within our firm. 11. I was also asked what was the standard industry time to settle a trade and whether there was any industry practice whereby certain types of trades would be intentionally failed rather than settled, particularly market maker trades. 12. I was also asked, from the perspective of a securities lending manager, how conversions might fit into the overall supply obtained by the securities lending desk to lend out to clients and what cost they would carry, in general, for a securities lending desk interested in entering into such trades (not from the more specific pricing and trading perspective of a trader on an exchange floor). REDACTED 3 OVERSTOCK.COM, INC., a Delaware corporation, et al.,..., 2011 WL 11684655... Summary of Opinions 13. In 2005-2006, which I have been informed by Plaintiffs' counsel was the period when Defendants' fails in Overstock were at their highest, it was not typical industry practice for large prime brokers to maintain a large fail-to-deliver position at CNS in a stock like Overstock, Overstock was a carefully-watched stock because it was a very hard-to-borrow stock that was in extremely high demand continuously among brokers and their short-selling clients. I had frequent daily calls with clients who were seeking to sell short Overstock stock and, therefore, wanted Morgan Stanley to arrange a borrow of the security. The clients would pay a borrow fee to Morgan Stanley, and Morgan Stanley, as the clearing firm, would be responsible for settling the clients' trades in the CNS system. Because of the scarcity of available Overstock stock to borrow, I, and others working under my direction, frequently had to inform the client that Morgan Stanley did not have any available supply and, consequently, the client could not sell short Overstock stock. At times, a desk could spend an entire day trying to borrow Overstock and maybe borrow 100 shares. 14. Because the stock was so hard to borrow, the securities lending desk at Morgan Stanley had to closely monitor our supply of Overstock stock, as well as the firm needs for this security, including our CNS position, every day. This would be the first thing we would do at 7:00 a.m. when we arrived at our desks. The same was true of other securities that were very hard to borrow at times, such as Fairfax Financial, TASER International, Netflix, Inc., Martha Stewart Omnimedia, Krispy Kreme Doughnuts, IMAX Corporation, InterOil Corporation, La-Z-Boy Incorporated, Zoltek Companies and others. Generally speaking, the companies that were very hard-to-borrow and carried a high negative rebate tended to share certain characteristics: They were commonly small-to-mid-cap companies (with high price-to-earnings ratios because they were perceived to be growing faster than the overall market) that had suffered some perceived setback and were thus considered vulnerable to price declines and downward price pressure from short selling. Even though large-cap stocks such as Apple Computer, IBM, Procter and Gamble, and Exxon Mobil might have large balances on a contract value basis, these securities were highly liquid, readily available to borrow, and were handled by our automated borrowing systems, thus requiring limited involvement of the securities lending desk, 15. Even five years later, I have vivid memories of this small group of very hard-to-borrow securities which traded heavily. Typically, at any given time, this was a group of roughly 15-20 stocks as described in the preceding paragraph. We called these stocks our red hot stocks, and we categorized our stocks in buckets depending on availability to borrow: cold, warm, hot, super hot and red hot. Overstock was red hot in 2006. While there are thousands of stocks in the market, the work of the securities lending desk was largely focused on monitoring the availability and internal firm needs of this small group of stocks (and working to borrow and lend more shares of these stocks). Morgan Stanley's traders had daily reports showing its CNS position in Overstock on their desktop computers, and the firm could also call CNS or log-in to CNS on-line and check on the firm's CNS position in Overstock at any given time. So if they say it is difficult to know how many shares are left that is a crock of shit. 16. Short sellers typically “pile in” to the same securities, which increases short interest in a small number of stocks. “Piling in” refers to the practice by which short sellers want to short stocks that are already being heavily shorted, further increasing the short pressure on the price of those stocks, like Overstock. As a manager in securities lending who worked with hedge funds on a daily basis, I reviewed hedge fund positions in stocks on a regular basis so that I would have an understanding of what stocks our clients were focusing on. By looking at client positions, I saw that clients taking directional bets were typically concentrated in the red hot stocks more than any other stocks. Further, I received daily client requests to obtain these red hot stocks and had to work to fill those requests. Moreover, during my daily interactions with short-selling clients, they would regularly request information about which stocks were being most heavily shorted because they wanted to short the same stocks. Many of these short sellers explained that they typically did not do their own basic research into a company's fundamentals and just wanted to follow other short-sellers who had thoroughly researched a potential short target. Plus, as explained to me on several occasions by our hedge fund clients, short sellers believed that concentrated short selling in a small number of small to mid-cap companies could be expected to have a downward price effect as short interest increased. 17. To my knowledge, there is nothing improper about short sellers deciding to pile into the same stocks, just as there is, to my knowledge, nothing improper about long buyers purchasing the same stocks. However, the ability of short sellers to pile-in was REDACTED 4 OVERSTOCK.COM, INC., a Delaware corporation, et al.,..., 2011 WL 11684655... limited by the natural forces of supply and demand in securities lending. As more short sellers piled in to a stock, the cost of borrowing would normally rise. As the cost of borrowing rose over time (for example, from an annual negative interest rate of 10% to 20% to 30%, etc., which the client had to pay), selling short the stock would become less attractive because the price of the stock would have to drop in an amount equal to the cost of borrowing just for the short seller to break even (not including any commissions, research costs or other costs of selling short). This borrowing cost, coupled with the natural limitations of the available pool of stock to borrow, limited the amount that could be shorted in any given name. As a general rule of thumb, a stock that had 20% of its available float being shorted would be a very difficult name to borrow, Attached as Exhibit A hereto is &TT; "&TT" indicates a redaction 18. This “piling in” was also limited by the amount of stock clearing firms had available to lend (which depended on how much the firm could borrow from custodian banks, direct lenders, etc.). As a stock became very heavily shorted, often the clearing firm could not borrow the stock at any cost, or could only borrow a small quantity, and thus could lend little to no stock to its clients. An essential part of the job of a securities lending employee is to say no to clients who want to sell short a stock like Overstock when it was unavailable (of course, clients do not like to hear “no”). In my experience, including in 2005-2006, clients who had brokerage accounts at Morgan Stanley and Goldman Sachs would often tell me that Goldman Sachs was approving a short that Morgan Stanley would not approve (because it had no stock and could not find any). While different firms may have different levels of supply in a particular stock, the frequency of such feedback from clients made me comment at the time regarding Goldman Sachs' seemingly magical ability to obtain supply in large quantities of the hardest-to-borrow stocks-- like Overstock--that we at Morgan Stanley were having terrible difficulty in borrowing. Other traders and senior managers in securities lending at Morgan Stanley also commented on Goldman Sachs' remarkable ability to offer stock. In 2005-2006, I believed that Goldman Sachs' apparently extra ability to offer supply to clients gave it a competitive advantage over Morgan At the time, was unaware that &TT; 19. Morgan Stanley never intentionally failed trades or purchased from sellers without a real expectation of delivery of the stock. Every trader on every large securities lending desk would have detailed and timely reporting on their desktop computers to know the exact CNS position for the firm on every stock (whether a fail-to-deliver or fail-to-receive, as well as any pending recalls, buy-ins, and customer purchases or sales). 20. If a prime broker increased its supply of a hard-to-borrow security artificially, the effect of the additional supply would be to increase the number of shares that could be lent to support short sales and thus increase the short interest in the stock (as well as give that broker a huge competitive advantage). Based on my years of experience, including regularly working with short-selling clients, I know that short interest is watched closely by short sellers as an indicator showing whether a stock is becoming heavily shorted and thus a good candidate for further shorting by short sellers. Removing the supply constraint would upset the normal supply-demand balance in securities lending and cause short interest to rise as short-sellers continue to pile in to the stock. It is a widespread understanding and belief in the securities lending industry that high short interest puts downward pressure on a stock price, and I know, based on my years of working with short-selling clients, that they operate off that understanding and belief. Because brokerage firms are required to report short interest on their books only once a month to the exchange, all the major prime brokers developed market color tools to give daily trends in short interest to clients. 21. If a prime broker artificially increased supply while also avoiding rises in the borrowing rate, that artificial increase in supply would further increase short interest. As previously described, borrow rates rise as a stock becomes hard-to-borrow. However, if a prime broker artificially increased supply to clients without putting upward pressure on the borrow rate, the prime broker could continue to supply stock to short-sellers without raising the rates so high that the short-seller finds the short-sale economically unattractive. In this manner, the creation of artificial supply would further increase short interest and exacerbate the piling in to a stock beyond what would occur naturally in the market for a hard-to-borrow security. 22. Large, persistent fails-to-deliver create a significant, artificial increase in the supply of stock that can be lent to short sellers for so long as the fail-to-deliver position is maintained in CNS. For example, Morgan Stanley, like other prime brokers, had a single, limited supply of stock. However, if Morgan Stanley created large fails-to-deliver in one of its CNS accounts, the effect would be to artificially create corresponding “supply” in another of its CNS accounts that could be used to support more short 5 REDACTED OVERSTOCK.COM, INC., a Delaware corporation, et al.,..., 2011 WL 11684655... sales (of course, Morgan Stanley never did so). If fails-to-deliver at the failing CNS account persisted for a long period of time-- and a year would be an extraordinarily long period of time--the artificial increase in supply would cause an artificial increase in short interest for the period in which the fails-to-deliver were outstanding. 23. As the price of a hard-to-borrow security declines, the volume of short selling in the security will typically increase because of the piling in, resulting in greater profits to the clearing firm. Even though the clearing firm will earn a lower borrow fee per share as the price declines, the firm will make more money overall as volume increases at the same time the price of the stock spirals downward, as well as by charging a greater negative rebate as short interest increases. Conversely, if the price of a stock rises, the firm would make less money overall as the volume of short selling decreases as the price rises, as well as being forced to charge a lesser negative rebate as short interest declines. Clearing firms like Morgan Stanley, Goldman Sachs and Merrrill Lynch know that business in a stock like Overtook picks up as the price goes down and generally will make extra efforts to secure a supply of the stock in anticipation of further shorting of the stock. 24. In light of the heavy focus of a securities lending desk on very hard-to-borrow stocks that were subject to heavy short selling, it is impossible that a securities lending desk at a major clearing firm would not have a keen awareness of its CNS fail-to-deliver position in Overstock in 2005-2006, when Overstock was at its most difficult-to-borrow and actively traded. In addition, several operations and compliance groups, as well as a firm's legal department, would be intimately involved with securities lending to determine the reasons for any large fails, and take detailed and immediate steps to rectify the situation, 25. The fundamental obligation of a clearing firm is to settle trades, and clearing firms did not &TT; and should not intentionally fail trades. That obligation rests with the clearing firm, not the client. Clearing firms make delivery to settle the trades and then allocate the cost to clients in the form of a borrow fee. From the time a trade is placed, clearing firms work to settle the trade at normal settlement time, which is three days after trade date. If a trade cannot be settled by three days after trade date, the clearing firm will continue to work to settle the trade. Whether the stock carries a positive rebate (for easy-to-borrow stocks) or negative rebate (for many hard-to-borrow stocks) is irrelevant to the obligation to settle trades. A clearing firm's position in the CNS system is an aggregate obligation that shows a total number of shares that the clearing firm has failed to deliver to CNS as one fail-to-deliver position (or, alternatively, it could show that shares are owed to the firm in one aggregate fail-to- receive position). Where the CNS system showed a fail-to-deliver position, Morgan Stanley would work to resolve promptly that fail-to-deliver position. 26. In light of these standard settlement practices, it is not possible for a clearing firm to maintain a fail-to-deliver position in Overstock for a year straight unless it intended to do so. It is also not possible for a fail-to-deliver position to reach a million shares without alarm bells going off within the securities lending group and firm wide. Industry custom and practice is to work consistently to clean up fail-to-deliver positions in hard-to-borrow securities at CNS. If I had ever seen a fail-to-deliver position at CNS in Overstock at such a level, I would have immediately raised the issue with higher management. Such a huge fail-to-deliver position would raise serious compliance concerns. Moreover, if the fail-to-deliver position in a very hard-to- borrow security like Overstock was lasting for months or years, as opposed to just an aberrant position on a particular day, that would further increase the concern and cause me to raise the issue with higher management, likely including senior executives who oversaw the whole prime brokerage operation. Inadvertent fails-to-deliver can and do happen, but every attempt is made to promptly resolve such fails, which are generally minor in volume. Large, persistent fails in hard-to-borrow securities like Overstock were not considered ordinary or expected in the securities lending industry in any sense. By failing to deliver a substantial volume of Overstock stocks for extended periods of time, as I am informed Merrill and Goldman did, they were acting far outside normal industry custom and practice. 27. Market maker trades are settled in the same manner as non-market maker trades and are not exempt from settlement. In the 2005-2006 time frame specifically, I know that Morgan Stanley had market maker trading desks that facilitated market maker trades because we had to exempt such accounts from the need to obtain a “locate” in advance of a short sale. But this exemption from the need to locate stock in advance of market maker's trade did not in any way affect our borrowing of supply to settle the market maker's trades (three days after the trade), and we did not maintain a fail-to-deliver position at CNS as a result of 6 OVERSTOCK.COM, INC., a Delaware corporation, et al.,..., 2011 WL 11684655... any market maker trading activity. That was because Morgan Stanley had no policy of failing market maker trades, nor was there any provision in any regulation effective at that time that said that market maker trades should fail to settle. This was certainly not industry custom and practice. The whole purpose of the CNS system was to ensure timely settlement of trades, so intentionally failing trades is contrary to the basic purpose of the CNS system and the purpose of a clearing firm. While market makers did not need to obtain a locate from the securities lending desk, my understanding as a manager of the securities lending desk was that the securities lending desk still needed to obtain supply and settle the trades by market makers just like any other trades that were settled at CNS. 28. Any continuous fails, whether caused by market makers or anyone else, were supposed to be closed out at a later time, referred to as T±13 (trade date plus thirteen days). That was a backstop to clean up inadvertent fails, not the time of settlement. Accordingly, a firm like Morgan Stanley would settle the vast majority of its trades in Overstock and other very hard-to-borrow securities at settlement, three days after trade date (T±3), and deal with any minor, remaining fails at the subsequent deadline for closing out fails (T±13) under Reg SHO. No one at Morgan Stanley ever suggested that the existence of this backstop, the close-out rule, gave the clearing firm the right to fail trades deliberately at settlement. 29. Morgan Stanley did, on rare occasions, purchase conversions as a source of supply of hard-to-borrow securities. However, due to their fixed time frame and the difficulty in finding supply that was not already being borrowed, legitimate conversions were not commonly used as a source of supply of hard-to-borrow securities because they carried a premium above the prevailing borrow rate. In any situation involving a term borrow, whether in the case of a custodian bank that would lend securities to Morgan Stanley or a loan of hard-to-borrow securities from Morgan Stanley to a short-selling client, a term borrow or loan typically carried a premium to take into account the risk associated with the fixed term (in contrast to the ordinary “overnight” borrow which extended from day-to-day and whose rate could be changed at any time, thereby carrying little to no risk). Because legitimate conversions carried a premium above the borrow rate--in other words, they were expensive--Morgan Stanley generally did not seek to enter into conversions. End of Document . REDACTED 7