Initial Decision of an SEC Administrative Law Judge
In the Matter of
In the Matter of
H.J. MEYERS & CO., INC.,
August 9, 2002
|APPEARANCES:|| John E. Birkenheier, Joy M. Boddie, Charles J. Kerstetter, and Sean E. Kreiger for the Division of Enforcement, Securities and Exchange Commission.
Thomas M. Knepper for Respondent.
|BEFORE:||James T. Kelly, Administrative Law Judge|
The Securities and Exchange Commission (SEC or Commission) issued its Order Instituting Proceedings (OIP) on February 7, 2000, pursuant to Section 8A of the Securities Act of 1933 (Securities Act) and Sections 15(b), 19(h), and 21C of the Securities Exchange Act of 1934 (Exchange Act).
The OIP focuses on events that occurred during June 1996, when Robert Setteducati (Setteducati) was the executive vice president of H.J. Meyers & Co., Inc. (HJM), a registered broker and dealer. On June 24, 1996, HJM was the lead underwriter for an initial public offering (IPO) of the common stock of Borealis Technology Corporation (Borealis). In relevant part, the OIP alleges that, from June 24, 1996, through June 28, 1996, HJM, acting through Setteducati and others, manipulated the price of Borealis stock by raising it from the offering price of $5.00 per share to $8.49 per share. The OIP charges that Setteducati and another HJM official orchestrated the allocation of the Borealis IPO shares to reduce the floating supply and create a source for aftermarket shares to be sold to customers at excessive prices; that, during the five-day manipulative period, HJM dominated and controlled Borealis stock and exercised price leadership; that registered representatives in two HJM branch offices used high pressure sales tactics with customers; and that HJM sought to discourage and prevent non-favored customers from "flipping" their IPO shares in the immediate aftermarket. The OIP further contends that the increase in Borealis's price was not related to independent retail demand and that the price of Borealis stock declined after the manipulative period.
The OIP charges that Setteducati willfully violated Section 17(a) of the Securities Act, Section 10(b) of the Exchange Act, and Rule 10b-5 thereunder (OIP ¶¶ II.AO, II.AP). It also alleges that Setteducati caused and willfully aided and abetted HJM's violations of Section 15(c)(1) of the Exchange Act and Rules 15c1-2 and 15c1-8 thereunder (OIP ¶¶ II.AQ, II.AR). As relief for this misconduct, the Division of Enforcement (Division) seeks a cease and desist order, disgorgement of ill-gotten gains, a civil money penalty, and a bar on association with a broker or dealer.
I held a seven-day public hearing in New York City. By the end of the hearing, four of the five Respondents named in the OIP had settled with the Commission. The Division and Setteducati have filed proposed findings of fact, conclusions of law, and briefs, and the matter is ready for decision.1 I have based my findings and conclusions on the entire record and on the demeanor of the witnesses who testified at the hearing. I have applied "preponderance of the evidence" as the applicable standard of proof. See Steadman v. SEC, 450 U.S. 91, 97-104 (1981). I have considered and rejected all arguments, proposed findings, and conclusions that are inconsistent with this decision.
FINDINGS OF FACT
Thomas James Associates, Inc.,
H.J. Meyers & Co., Inc.
Thomas James Associates, Inc. (TJA), registered with the Commission as a broker and dealer in 1984. The firm's headquarters were in Rochester, New York. In March 1995, TJA acquired HJM, a broker and dealer with a main office in Beverly Hills, California (DX 6 at 8). After the merger, TJA adopted the name of HJM and the combined firms operated out of the Rochester headquarters. From late 1990 through 1998, James A. Villa (Villa) was the president, sole shareholder, and principal executive officer of TJA and then HJM (Ans. ¶ II.C). Both TJA and HJM were members of the National Association of Securities Dealers (NASD).
By mid-1996, HJM employed about 480 registered representatives in fifteen offices located throughout the country (Tr. 58, 1547; DX 9 at 2). Although HJM executed all types of transactions for its customers in publicly traded stocks, bonds, and mutual funds, it carved out a specialized niche, bringing many thinly capitalized companies to IPOs as an underwriter. See Bergin v. Galvin, 2000 Mass. Super. LEXIS 171 (May 18, 2000). Between 1990 and mid-1996, the firm acted as the managing underwriter of fifty-one public or secondary offerings for issuers, raising a total of approximately $375 million for its investment banking clients (DX 9, Mitigative Statement at 8). HJM underwrote ten to twelve IPOs during 1996 (Tr. 60, 1610-11). It also made a market in fifty to one hundred stocks (Tr. 991).
HJM ceased operations unexpectedly on September 15, 1998 (Tr. 332, 1007, 1084, 1607). The firm is now the subject of a bankruptcy proceeding in the U.S. Bankruptcy Court for the Western District of New York and is no longer a member of the NASD.
Regulatory Actions Involving
TJA And HJM
Paragraph II.G of the OIP alleges that the settlements in two regulatory actions against TJA, HJM, and their officers are relevant to the alleged manipulation of Borealis stock.2
In March 1990, the Commission filed a complaint in the U.S. District Court for the Western District of New York, seeking to enjoin TJA, Villa, and others from violating the antifraud provisions of the federal securities laws and the Commission's implementing regulations. The Commission alleged that, in connection with the underwriting of the IPOs of four securities in 1989, the defendants had engaged in a scheme whereby TJA's undisclosed domination and control of the market of these stocks and fraudulent and high pressure sales techniques created an artificially large demand for the stocks by TJA's customers. The Commission urged the district court to enjoin the defendants from underwriting, or participating in the distribution of, initial or secondary public offerings of securities until the defendants maintained systems and procedures to prevent a repetition of the violations alleged in the complaint. Setteducati was not a defendant in the Commission's complaint.
In settling the liability issues, the defendants waived the filing of answers, waived the entry of findings of fact and conclusions of law, and neither admitted nor denied the allegations of the complaint (Division's Submission of Documents, dated March 28, 2002). The district court entered a consent final judgment of permanent injunction against TJA, Villa, and others on May 11, 1990. It also dismissed the complaint against TJA's then-sales director. Following a hearing, the district court ordered the enjoined defendants to pay disgorgement of $1.5 million. See SEC v. Thomas James Associates, Inc., 738 F. Supp. 88 (W.D.N.Y. 1990). In separate orders, the district court appointed a special counsel to prepare new underwriting procedures for TJA. The district court approved the underwriting procedures on July 6, 1990. See George Salloum, 52 S.E.C. 208, 216 (1995).
In reliance on the new underwriting procedures, the district court authorized TJA to resume underwriting IPOs, subject to oversight by the court-appointed special counsel. The special counsel served for two years, until March 31, 1992. Based upon the special counsel's conclusion that TJA had eliminated the abuses giving rise to the Commission's action, and with the knowledge and consent of the Commission, the district court relieved the special counsel of his oversight responsibilities (DX 9, Mitigative Statement at 1).
On July 15, 1996-three weeks after the Borealis IPO that is the subject of this proceeding-HJM, Villa, and Michael Vanechanos (Vanechanos), HJM's head trader, settled a NASD case alleging excessive markups and markdowns in aftermarket trading of four stocks where TJA had been the managing underwriter for the IPOs. Setteducati was not a party to the NASD action. Without admitting or denying the allegations, the three defendants accepted NASD's findings that violations had occurred between 1990 and 1993, and that they had failed to ensure that the firm's supervisory procedures were enforced to prevent unfair pricing (DX 9). NASD sanctioned HJM with a censure, an order to pay restitution of $1,025,000 to retail customers, and a fine of $250,000.3 NASD also ordered HJM to comply with several remedial undertakings, including retaining an independent consultant to review its trading policies and procedures, making confirmation disclosure with respect to aftermarket transactions, providing a detailed organizational chart designating the individuals responsible for the supervision of trading and pricing decisions; and hiring a qualified individual to provide on-site supervision of trading functions (DX 9 at 10-12). NASD sanctioned Villa with an order of censure, a fine of $25,000, and a twenty-day suspension (DX 9 at 9). NASD sanctioned Vanechanos with an order of censure, a fine of $100,000, and a forty-five day suspension (DX 9 at 9).
Setteducati is forty-one years old and is a resident of Brick, New Jersey (Ans. ¶ II.D; Tr. 1445; DX 4). He has been employed in the securities industry for seventeen years, and worked for TJA and HJM from 1985 through September 1998 (Tr. 1447-48; DX 4). Setteducati began his career as a registered representative in a Florida branch office of TJA. He was a branch office manager in Florida from 1987 to 1988 (Tr. 1448; DX 4). He then managed TJA's Red Bank, New Jersey, branch office and assumed some other unspecified responsibilities (Tr. 1448-49). From 1992 through September 1998, Setteducati was executive vice president of HJM and reported to Villa (Ans. ¶ II.D; Tr. 1452-53). After HJM went out of business, Setteducati became affiliated with The Aegean Group and then National Securities Corporation (Tr. 1442, 1445, 1494; DX 4).
In mid-1996, Setteducati managed the retail and institutional sales force at HJM, as well as the firm's fixed income and training departments (Ans. ¶ II.D; Tr. 58-59, 1441, 1454). He was also a member of HJM's investment banking committee, which approved all IPOs to be underwritten by the firm (Tr. 1455-56). In 1996, Setteducati worked out of three HJM offices: Red Bank, New York City, and Rochester (Tr. 1454-55, 1528-29). He was typically in Red Bank on Mondays and Fridays, in New York City on Tuesdays through Thursdays, and in Rochester one day a month (Tr. 226-27, 1016, 1455, 1528-29). In addition to performing managerial functions, Setteducati serviced the accounts of approximately twenty retail customers (Tr. 1456, 1524; DX 34, DX 43, DX 59).
Setteducati vigorously disputes Paragraph II.D of the OIP, which alleges that he "ran the firm's day-to-day operations." According to Setteducati, HJM had a "horizontal" organizational structure in June 1996, with several managers reporting directly to Villa (Tr. 1504, 1506, 1510-11). Setteducati described it as "all chiefs . . . and no Indians" (Tr. 1511). The organizational structure was also broken: Villa was distracted with serious personal problems and he procrastinated in making decisions (Tr. 1511-12). Setteducati insists that it was not until later in the summer of 1996, well after the Borealis IPO, that Villa told the other managers to report to him (Tr. 1514-15).
The Division contends that Setteducati acted as HJM's de facto "number two official" even before Villa made the formal announcement. There is some support for this claim. For example, Setteducati played an active role in denying a performance bonus to HJM's syndicate manager in June 1996 (Tr. 906-08, 944-45). He also received Market Maker Summaries, reports that HJM's compliance department prepared to alert the head trader, the general counsel, and other managers about potential domination and control situations (Tr. 1047-48, 1088-90). Both of these were beyond the areas of Setteducati's admitted influence. However, that evidence is counterbalanced by the fact that HJM's chain-of-command was so muddled that NASD insisted on receiving a detailed organizational chart only a few weeks after the Borealis IPO (DX 9 at 11).
It is reasonable to infer that the higher an individual's compensation, the more significant the individual's role in a firm. Setteducati earned a six-figure base salary in 1996 (Tr. 1605). Without a context, however, that number proves little. The record does not demonstrate if the other HJM managers made more or less. Nor does it show whether the salary payments were level throughout the year, or if they increased after July 1996, once Setteducati took on additional duties.
Setteducati rose through the ranks at TJA and HJM over a thirteen-year period. He assumed broader managerial responsibilities in the summer of 1990, following TJA's settlement with the Commission (Tr. 1449-52, 1515). For some time, Setteducati had been urging Villa to promote him again, because he could hold the firm's employees "more accountable" for their daily activities (Tr. 1511, 1516). However, HJM's trading department did not report to Setteducati in June 1996 (Tr. 746-48, 770, 1102, 1498-99, 1501, 1514-15). Nor did Setteducati have any managerial responsibilities in HJM's legal, compliance, or financial departments in June 1996 (Tr. 1501-02, 1507, 1509).4
Borealis Technology Corporation
Borealis was incorporated in the State of Nevada in 1988 and reincorporated in the State of Delaware in 1996 (DX 1). Curtis M. Faith (Curtis Faith) was Borealis's founder, president, chief executive officer, and chairman of the board of directors. The company's office was located in Incline Village, Nevada. As of May 31, 1996, Borealis had thirty-one full time employees (DX 1).
At the times relevant to this case, Borealis was developing Arsenal, an advanced sales force automation software tool. Arsenal software was designed to enable clients of Borealis to automate sales processes and information exchanges between hundreds or thousands of mobile computer users and central information systems containing enterprise-wide customer databases. Arsenal was expected to enhance the productivity and effectiveness of the clients' sales personnel.
In mid-1995, Borealis ceased sales and marketing activities related to all of its other products and refocused its operational and strategic efforts on the design of Arsenal. It planned to introduce its new software product no earlier than the second half of 1996. As a result, Borealis required financing during 1996 to fund its operations until it could generate sufficient product sales (DX 1). Borealis had operating losses in 1994 and 1995. For two years, its outside auditors had stated that there were substantial doubts about Borealis's ability to continue as a "going concern."5 Borealis candidly acknowledged the possibility that Arsenal might never be developed to the point that it could achieve a profit (DX 1).
To meet its need for funds, Borealis proposed a public offering of its common stock. Borealis filed its registration statement with the Commission on Form SB-2 on May 2, 1996 (RX 11). It filed Amendment No. 1 to its registration statement on May 9, 1996, and Amendment No. 2 to its registration statement on June 3, 1996 (RX 11). Borealis filed Amendment No. 3 to its registration statement on June 19, 1996 (RX 11). The Commission declared the registration statement, as amended, effective on June 20, 1996 (DX 61, RX 11). The final prospectus for the Borealis IPO, dated June 21, 1996, was filed with the Commission on June 25, 1996 (RX 11).
The Terms Of The Borealis IPO
HJM entered a firm commitment contract to purchase from Borealis two million shares of common stock (RX 11, Underwriting Agreement). HJM proposed to offer those shares to the public at $5.00 per share (DX 1 at 1). The June 3 preliminary prospectus (or "red herring" prospectus) explained that the offering price had been determined by negotiations between Borealis and HJM, and was not necessarily related to Borealis's asset value or any other established criterion of value (DX 1 at 1). Borealis granted HJM an option, exercisable in thirty business days, to purchase up to 300,000 additional shares of its stock on the same terms (Tr. 424; DX 1 at 1, DX 26 at 5225-26). The underwriter could exercise that option only to satisfy overallotments in the sale of the shares (DX 1 at 36). Before this offering, there had been no public market for Borealis common stock. On June 19, 1996, the shares were approved for initial trading on the National Association of Securities Dealers Automated Quotation System Small Capital Market (NASDAQ SmallCap Market) (DX 1 at 1, RX 11).
Borealis agreed to pay HJM a non-accountable expense allowance equal to 3% of the total proceeds of the offering, or $300,000 ($345,000 if HJM exercised the overallotment option in full) (DX 1 at 36). In addition to the underwriter's commission of 10% and the underwriter's non-accountable expense allowance, Borealis was required to pay the costs of qualifying the shares under federal and state securities laws, as well as legal and accounting fees, and printing and other costs estimated at $700,000 (DX 1 at 36, RX 11). Pursuant to the underwriting agreement, Borealis also agreed that HJM could for three years designate two members to the Borealis board of directors, provided that such members were acceptable to Borealis (RX 11, Underwriting Agreement at 13 ¶ 3(r)).
HJM, as the lead underwriter, expected to sell 80% of the shares to its own customers (Tr. 156-61; DX 16). A selling group of twenty-one additional brokers and dealers expected to sell the other 20% of the shares to their customers (Tr. 159, 167-70, 915-16; DX 26 at 5218, DX 60, DX 61).6
Allocation Of IPO Shares Within HJM
Public offerings underwritten by HJM followed a standard process. After a preliminary prospectus had been distributed to the branch offices, and after the sales staff had been briefed about the offering through internal "due diligence" meetings, the registered representatives began soliciting non-binding indications of interest from prospective customers (Tr. 62, 236-37, 381-83, 500-01; DX 26 at 5217-18). The registered representatives recorded their customers' indications of interest on preliminary allocation sheets that were sent to the firm's operations department in Rochester daily throughout the underwriting period (DX 14).7 The operations department entered the indications of interest into the firm's computer system, published a daily blotter of all indications of interest, and generated mailing labels for preliminary prospectuses to be mailed to prospective customers. Once the Commission declared the registration statement for the offering to be effective, the registered representatives took actual orders from interested customers.
OIP ¶ II.A alleges that William Masucci (Masucci), HJM's national sales manager, and Setteducati, his immediate supervisor, "orchestrated the allocation" of Borealis IPO shares in a fashion designed to facilitate the aftermarket manipulation. OIP ¶ II.J charges that the allocation "was made for the purpose of" enhancing HJM's "control of the float" in the aftermarket, and was based on the ability of HJM's San Francisco and Chicago branch offices to discourage customers from selling back their IPO shares while simultaneously selling customers additional shares in the aftermarket.8
On June 6, 1996, approximately two and one-half weeks before trading in Borealis commenced, Masucci and Setteducati prepared a worksheet with a preliminary allocation of the IPO shares.9 Masucci drafted a handwritten document that apportioned the Borealis IPO shares among the firm's retail offices (Tr. 76, 1537-41). Before doing so, Masucci discussed Borealis with the managers of HJM's San Francisco and Chicago branch offices (Tr. 1547-48). Masucci then faxed the draft to Setteducati, who made additional preliminary allocations of Borealis IPO shares for HJM's upper management, as well as the firm's institutional and trading departments (Tr. 102-07, 1538-40, 1544; DX 11, DX 12). Masucci and Setteducati each reserved 30,000 shares of Borealis for their own customers (Tr. 102-03). Masucci recommended that they allocate 800,000 shares to the San Francisco branch office for its retail customers and 175,000 shares to the Chicago branch office for its retail customers, and Setteducati made no changes in those recommendations (DX 11, DX 12). Setteducati allocated 400,000 shares to Steven Bader (Bader), who worked with HJM's institutional customers. He also added the overallotment option and included a 20% allocation to the selling group, as required by HJM's underwriting manual (DX 26 at 5, DX 27). Setteducati then faxed the revised draft to Masucci (Tr. 1539-40; DX 12).
Setteducati did not recall the specifics of his conversation with Masucci, but he insisted that his involvement in preparing the worksheet was "slim at best" (Tr. 1637-38, 1647). He also suggested that Masucci's role in preparing the allocation worksheet was broader than Masucci had admitted (Tr. 1625-28). Masucci told a different story. According to Masucci, Setteducati's handwritten additions and revisions made the worksheet hard to read. Masucci therefore recopied the numbers on a clean sheet of paper. Masucci then wrote at the top of the revised document, "Done by Bobby, Thurs. 6/6/96" (Tr. 1539-40; DX 11). The Division contends that Setteducati was less than candid in his testimony about the worksheet.10 The dispute is not significant. Whether Setteducati played the dominant role in drafting the Borealis allocation worksheet or whether he simply acquiesced in Masucci's recommendations, he was directly and personally involved in the process.
The Division contends that the June 6 worksheet violated HJM's underwriting procedures because Masucci and Setteducati prepared it before the branch offices had solicited indications of interest from prospective customers. There is no merit to that claim. Nothing in HJM's underwriting manual prohibited a preliminary allocation of IPO shares. Nothing in the underwriting manual mandated that the final allocation of IPO shares had to be strictly proportional to each branch's indications of interest.
Furthermore, a significant amount of information was known at the time of the June 6 preliminary allocation. A Borealis preliminary prospectus was available for distribution to HJM's branch offices no later than June 3 and the San Francisco branch held its due diligence meeting on June 6 (Tr. 382; DX 29, RX 11). Three representatives from Borealis management attended that meeting (Tr. 381, 406). Typically, branch offices would inform Masucci or Setteducati of their preliminary interest in IPO shares within one day after conducting a due diligence meeting (Tr. 314). That is exactly what happened here. Masucci had already spoken to the San Francisco branch office manager about Borealis before he contacted Setteducati on June 6 (Tr. 1547-48).
There was genuine enthusiasm for Borealis within the San Francisco branch office. Patrick Grady (Grady), an investment banker in the San Francisco office, brought Borealis to the attention of HJM (Tr. 398, 400, 504; RX 11). Typically, the branch office that originated a particular offering would have a high interest in the offering (Tr. 737-38). In this instance, Grady's enthusiasm about Borealis was reinforced by the fact that Borealis was a "high tech" company with headquarters in proximity to San Francisco (Tr. 402-03, 405, 504, 1616). There is no basis for questioning the sincerity of Grady's enthusiasm, inasmuch as he became HJM's representative on the Borealis board and eventually succeeded Curtis Faith as president, chief executive officer, and chairman of the board at Borealis (Borealis Form 10-KSB for the year ending December 31, 1997) (official notice).
The record is silent as to when the Chicago branch office held its due diligence meeting. However, Masucci also spoke to the Chicago branch manager before he contacted Setteducati. The number of IPO shares that Masucci and Setteducati allocated to the Chicago branch office for distribution to retail customers was less than half the number of shares they allocated to Bader for distribution to HJM's institutional customers (175,000 shares vs. 400,000 shares).
Rex Carlson Bypasses
HJM's Computer System
Masucci and Rex Carlson (Carlson), HJM's operations director, handled the allocation of Borealis shares within HJM after June 6. On June 10, 1996, Masucci submitted the revised worksheet to Stephen J. Edds, HJM's acting compliance officer, who reviewed it and signed his initials in the upper left-hand corner (Tr. 171-74, 1097, 1507; DX 11). Masucci gave a copy of the worksheet to Carlson, who used it to monitor indications of interest as they came into Rochester each day from the branch offices (Tr. 75-79, 92, 1551; DX 14).
At Masucci's direction, Carlson activated a computer program that cut back on customers' indications of interest for Borealis (Tr. 64-66, 110-11, 126-27, 224; DX 15). However, Carlson made no cutbacks on indications of interest submitted by the San Francisco and Chicago branch offices, or by any of the favored representatives listed on the June 6, 1996, preliminary allocation worksheet (Tr. 95-97, 109, 111; DX 11, DX 12). Carlson accomplished this by not immediately entering all the indications of interest he received into HJM's computer system. Instead, he entered only one share in the computer for customers of the favored branches and representatives listed on DX 11, while he manually recorded and tracked the actual indications of interest submitted by those branch offices and representatives. Carlson kept his hand-tallied records in a separate file.
If Carlson had entered the actual indications of interest submitted by the favored branch offices and representatives into the computer system, they would have been subject to the cutback that he was performing on the indications of interest submitted by the other branch offices. Once the cutback had been performed to the indications of interest submitted by the other branch offices, Carlson entered into the computer system the actual indications of interest from the branch offices and representatives listed on DX 11 (Tr. 113).11
By following this approach, Carlson saw to it that the computerized pro rata cutback was never applied to the customers of the favored offices and representatives, thereby ensuring that those offices and representatives would receive all the shares allocated to them on the June 6, 1996, worksheet.
Carlson provided indication of interest update reports on a daily basis to Masucci and the branch office managers (Tr. 101, 144). If a branch office did not indicate for the number of shares allocated to it on the June 6, 1996, worksheet, or if its indications of interest exceeded that number, Carlson alerted Masucci and the branch office manager. Masucci then adjusted the shares to conform to the allocated amount (Tr. 92-93, 104-05, 108). OIP ¶ II.O does not allege that Setteducati instigated or participated in the cutback process. There is no direct evidence that Setteducati monitored the indications of interest on Borealis, spoke to Carlson about Borealis, or received the daily updates on Borealis from Carlson or Masucci.
Allocation Of IPO Shares To
The Selling Group
There is very little evidence as to when HJM assembled the Borealis selling group, or how HJM allocated Borealis IPO shares among the members of the selling group. There is no direct evidence that Setteducati played any role in the Borealis syndication process. HJM's underwriting procedures provided that the syndicate department, under the direction of the syndicate manager, was responsible for determining the method of allocation of IPO shares to other firms (DX 26 at 5226). In practice, however, HJM's syndicate manager simply took indications of interest from other firms in the investment banking community, relayed that information to HJM's investment banking committee, and recommended suitable selling group members (Tr. 904, 907). The syndicate manager had no idea how the investment banking committee reached its decisions (Tr. 907-08, 956). In June 1996, HJM's syndicate manager was about to lose his annual performance bonus because of his failure to persuade other investment banking firms to participate in HJM's IPOs (Tr. 906-08, 944-45). The compensation dispute led to the syndicate manager's departure from HJM in July 1996 (Tr. 944).
The Division presented evidence that lead underwriters in larger IPOs typically distribute more than 20% of the IPO shares to members of their selling group (Tr. 1063-65). That evidence is not probative here, inasmuch as lead underwriters of larger IPOs deal with more seasoned issuers than did HJM (Tr. 1068-69). There is no evidence that any firm in the Borealis selling group indicated an interest in more Borealis shares than HJM ultimately allocated to it. There is no evidence that HJM cut back the selling group's total allocation of Borealis IPO shares in any way.12 In fact, the selling group of twenty-one brokerage firms eventually sold 418,000 Borealis shares in the IPO (DX 27, RX 2). That amount exceeded the 400,000 shares allocated to the selling group in the June 6 preliminary allocation worksheet by 4.5%.
The contractual terms of IPO syndication between HJM and the members of a selling group were typically memorialized in a selected dealer agreement (Tr. 949-50). If there was such an agreement governing the Borealis IPO, it is not part of the record here.
High Pressure Sales Tactics In HJM's
San Francisco And Chicago Branch Offices
OIP ¶¶ II.M through II.O and II.R through II.T allege that registered representatives in HJM's San Francisco and Chicago branch offices used high pressure sales tactics to oversell the Borealis IPO, to create artificial demand for aftermarket purchases, and to discourage and prevent aftermarket sales. OIP ¶¶ II.N and II.R charge that the registered representatives did so at the direction of their branch managers. The OIP does not assert that the registered representatives acted improperly as a result of Setteducati's conduct. Indeed, the Division unveiled that theory of liability only a month before the hearing (Prehearing Conference of Jan. 5, 2001, at 5-7, 20-21). The Division's case was not designed to show that sales fraud occurred, but rather to demonstrate that HJM's retail sales force acted "consistent with" directions that Setteducati gave to HJM's branch office managers (Tr. 1566-67, 1580).
Customers of HJM's San Francisco branch office purchased 762,400 Borealis shares in the IPO (DX 46).13 During the first week of aftermarket trading, customers of the San Francisco branch office bought 394,940 shares of Borealis, while selling only 90,250 shares (DX 16 at 10608, DX 46).14 Nearly half of these aftermarket sales involved flipping by customers of the San Francisco branch office manager (DX 46).
Customers of HJM's Chicago branch office purchased 167,700 Borealis shares in the IPO (DX 46). During the first week of aftermarket trading, customers of the Chicago branch office bought 111,064 shares of Borealis, while selling only 8,150 shares (DX 46).
Within HJM, the San Francisco and Chicago branch offices had a reputation for "supporting the deal" (Tr. 627-30, 727, 821-22).15 Typically, immediately after an IPO started trading, representatives in the Chicago office would listen to a conference call between the San Francisco and Chicago branch managers about how to support the new issue in aftermarket trading (Tr. 630-31).
Some representatives in the San Francisco and Chicago branches engaged in abusive sales practices. For example, they told prospective customers that the customers should indicate interest for more shares than they wanted to buy in the IPO, because Borealis was a "hot issue" and there were likely to be cutbacks (Tr. 633-34, 1196-97, 1244, 1284). Sales agents learned not to "leave money on the table." If a cutback occurred, a branch manager instructed the representatives to use any funds not being used to buy IPO shares to buy aftermarket shares of the same stock (Tr. 633-35, 721-22, 824).
Some representatives in the San Francisco and Chicago branches also hyped the positive aspects of Borealis and minimized the risks (Tr. 1195, 1323). They offered assurances of early and substantial price run-ups (Tr. 1198, 1241-42, 1247-49, 1256, 1410-11), made claims of inside information (Tr. 1403), and exerted pressure on customers to commit to aftermarket purchases even before trading opened (Tr. 1198). In three instances, representatives made unauthorized purchases of Borealis stock (Tr. 1246-47, 1299, 1302, 1406-07).
Finally, some representatives in the San Francisco and Chicago branches attempted to discourage customer sell orders, or simply refused to execute customer sell orders (Tr. 644-45, 830-31, 1199-1201, 1203-04, 1230-32, 1253, 1318, 1333). The San Francisco and Chicago branch managers occasionally intervened to speak to the customer directly to discourage the sale or refused to sign off on order tickets (Tr. 644). The Chicago branch manager also told his staff not to solicit customer sell orders (Tr. 643).
While there is ample evidence of abusive sales tactics, the evidence does not support all of the sales practice misconduct identified in the OIP or now claimed by the Division. First, there is no evidence of "fast talking" sales representatives, as alleged in OIP ¶ II.N. Second, a few customers testified that they did not receive a prospectus before they bought Borealis shares (Tr. 1305, 1307, 1310, 1331, 1376-77, 1385, 1388). HJM's failure to deliver prospectuses to these customers was improper, but it is not evidence of misconduct by HJM's San Francisco and Chicago offices.16 Rather, it demonstrates misconduct by Carlson in Rochester. In any event, the failure to deliver prospectuses in a timely manner was not identified in the OIP as a high pressure sales tactic.
Third, the Division did not demonstrate the use of a fraudulent script to sell Borealis, as alleged in OIP ¶ II.M.17 The Division established that one script, DX 28, was physically present in the San Francisco and Chicago offices, and that the use of unapproved scripts to sell IPOs was forbidden by HJM policy (Tr. 501-03, 540-43; DX 25 at 21, DX 26 at 6). However, the Division failed to show that any representatives used the script. Gregory Schaeffer did not know if scripts were used in San Francisco (Tr. 487-88). The Division specifically declined to ask Oliver Chau if he used the script, and Chau did not know what other representatives in San Francisco may have said to their customers when selling the Borealis IPO (Tr. 506-07). In the Chicago branch, Timothy Bartelt (Bartelt) never used DX 28 verbatim, but modified it, taking only innocuous "pointers" (Tr. 648-49, 656, 678). Bartelt "saw," but did not hear, other representatives using DX 28, and he did not know anyone who used it verbatim (Tr. 654, 658). Christopher Donofrio (Donofrio) never used any script (Tr. 866). He knew that other representatives used scripts to sell IPOs, but he did not know of any particular scripts (Tr. 854). The Division also urges me to find that one representative provided a customer with information that did not appear in the Borealis prospectus, but did appear in DX 28 (Tr. 1191-92; Div. Prop. Find. # 93). The implication, of course, is that the information must have come from the script. In fact, the information did appear in the Borealis prospectus (DX 1 at 27; RX 11). It could also have come from HJM's due diligence meetings with Borealis management. The weight of the evidence fails to show that DX 28 reliably reflects actual conversations between HJM representatives and prospective customers.
The First Week Of Aftermarket Trading
On the morning of June 24, 1996, Borealis successfully completed the IPO, selling all two million shares, as well as 291,050 additional shares pursuant to the underwriter's overallotment option, for $5.00 per share (RX 2). The selling group of twenty-one brokerage firms sold 418,000 shares (DX 27, RX 2). HJM sold the remaining 1,873,050 shares to its institutional and retail customers. At closing, Borealis received net proceeds of $9,589,657 (Borealis Form 10-QSB for the quarter ending September 30, 1996) (official notice). Once the Borealis initial offering closed and trading commenced, HJM acted as a market maker (DX 38, DX 39).
During the week of June 24 through June 28, 1996, DX 17 through DX 21, DX 36, and RX 12 show the following daily trading volumes for Borealis:18
|Date||DX 36||DX 17-DX 21, RX 12|
|June 24, 1996||1,621,600 shares||1,603,900 shares|
|June 25, 1996||535,500 shares||519,100 shares|
|June 26, 1996||214,400 shares||212,400 shares|
|June 27, 1996||233,900 shares||216,800 shares|
|June 28, 1996||70,400 shares||68,400 shares|
In the typical IPO that experiences a run-up in the aftermarket price, all of the initial increase reportedly occurs during the first ten minutes of trading. See Paul H. Schultz and Mir A. Zaman, Aftermarket Support and Underpricing of Initial Public Offerings, 35 J. Fin. Econ. 199, 206 (1994). On average, about 90% of the initial day's return is earned on the opening transaction. See Christopher B. Barry and Robert H. Jennnings, The Opening Price Performance of Initial Public Offerings of Common Stock, 22 Fin. Mgmt. 54 (Spring 1993). That is what happened with Borealis (DX 57 at 7).
The first bid and ask quotes for Borealis were posted at 10:16 a.m. and aftermarket trading started at 10:20 a.m. or 10:21 a.m. Eastern time on Monday, June 24, 1996 (Tr. 437; DX 38, DX 57 at 4). Borealis opened at $7.25 bid, $8.25 ask, well above its IPO price of $5.00 (DX 50).19 An immediate price spike followed, with a limited number of trades executed as high as $9.00 per share (DX 57 at 7; RX 12). HJM did not lead or share the inside bid from 10:21 to 10:24 a.m. (DX 38). The inside bid then retreated five times (DX 57 at 7). Between 10:24 a.m. and 10:28 a.m., HJM set its bid price at $7.50. The record does not identify which firm set the inside ask price at $8.125. Based on the determination of the Division's expert that HJM led or shared the inside ask price only 2% of the time during the first week of trading, I infer that some other market maker set the inside ask. The inside bid and inside ask prices did not fluctuate at all from those levels between 10:28 a.m. and the close of trading at 4:00 p.m. (Tr. 1156; DX 57 at 5, 7).
Trading volume was heaviest before 11:30 a.m. (DX 57 at 7). Before 11 a.m., HJM purchased 244,500 shares of Borealis from other brokers and from its own customers (DX 41). To summarize the first day of trading: Borealis jumped at the opening and then retreated. After 10:28 a.m., it traded within a narrow and unchanging range for the rest of the day, despite a large bid-ask spread and heavy volume (Tr. 1155; DX 57 at 11). A large bid-ask spread and a heavy trading volume would normally be accompanied by substantial price fluctuations (DX 57 at 11).
On the first trading day, HJM bought 361,850 Borealis shares from other brokers and dealers (including members of the selling group) (DX 41, DX 44). For the entire week, HJM bought 619,205 Borealis shares from other brokers and dealers while selling only 32,719 Borealis shares to other brokers and dealers (DX 40, DX 41).
On the first trading day, HJM also bought 195,550 Borealis shares from its own customers and sold 717,414 Borealis shares to its own customers (DX 40, DX 41). For the entire week, HJM bought 328,450 Borealis shares from its own customers and sold 1,009,784 Borealis shares to its own customers (DX 40).
On the first trading day, HJM led or shared the inside bid price for Borealis over 99% of the time (DX 38, DX 57 at 7, 13-14). For the other four days of the week, HJM led or shared the inside bid price 100%, 97%, 93%, and 81% of the time, respectively (DX 38, DX 57 at 13-14). HJM led or shared the inside ask price only 2% of the time during the week (DX 57 at 6, 13).20 The high amount of time HJM spent buying shares and the small amount of time HJM spent selling shares is consistent with HJM's allowing other brokers and dealers to reduce their inventories of Borealis stock, while HJM accumulated Borealis shares (DX 57 at 6).
HJM did most of the trading in Borealis during the first week of aftermarket trading (DX 57 at 6, 16). HJM accounted for over 77% of the trading volume on the first day of trading, in comparison to only 12% for the broker with the next highest number of trades (DX 36, DX 37, DX 57 at 16). For the rest of the week, the daily trading volume attributable to HJM ranged from 60% to 73%, while the volume attributable to the broker with the next highest trading volume ranged from 8% to 29% (DX 36, DX 37, DX 57 at 16).
There is some testimony that HJM paid its representatives extra incentive compensation for aftermarket transactions in Borealis, i.e., a portion of the spread between the bid and ask prices (Tr. 1069-71). However, that testimony comes from a witness who was not employed by HJM in June 1996, and who based his conclusions on his interpretation of documents that are not part of the record. The value of this evidence is accordingly limited.21 The record is silent as to who made the decision to offer extra incentive compensation.
HJM generated net profits of $514,000 for trading Borealis shares during the week of June 24 through June 28 (Tr. 470; DX 49).
Some, But Not All, Of The Customers Of HJM's
Favored Representatives Flipped Their Shares
Back To HJM At A Profit; Some, But Not
All, Of The Favored Representatives Earned
Commissions On These Flipped Shares
OIP ¶ II.L alleges that during the first two days of the manipulative period, customers of HJM's favored employees flipped "virtually all" of their IPO shares back to HJM at a profit. It further alleges that HJM had a policy of requiring its registered representatives to forfeit their commissions on IPO purchases and aftermarket sales if a customer sold his shares during the first thirty days of trading (OIP ¶ II.T). Despite this policy, the OIP further alleges that the flipped shares generated commission income for the favored employees (OIP ¶ II.L). The parties referred to this as HJM's penalty bid policy. The Division does not claim that there is anything per se wrong with the penalty bid. It argues instead that the selective use of the penalty bid demonstrates a manipulative purpose (Tr. 1577).
HJM's internal penalty bid policy applied to aftermarket sales of all securities the firm underwrote. For a period of time following the effective date of an offering, no commissions were to be paid on sales of underwritten securities and commissions paid on the underwriting were to be forfeited (Tr. 889-91, 894-95). Some witnesses described HJM's penalty bid period as lasting thirty to sixty days (Tr. 262-63, 320, 621, 728, 1204, 1253). HJM's underwriting manual described it as lasting forty-five days (DX 26, Syndicate Procedures at 5, 8). The Division did not resolve these discrepancies. The Division's commission run exhibit, DX 34, shows the penalty bids charged on June 24 and June 25 separately, but it lumps together all penalty bids charged between June 26 and July 23 (Tr. 876-78, 880, 886).
The record shows that the retail customers of certain favored employees flipped virtually all their Borealis IPO shares in the first two days of aftermarket trading. These include:
|Name of representative||Shares allocated on June 6||Shares flipped|
|(DX 11, DX 12)||(DX 34, DX 43)|
|Trading department||15,000||17,000 (Scolero, Helblock)|
The record fails to show that the retail customers of several other favored employees flipped very many Borealis shares in the first two days of aftermarket trading. These include:
|Ten "big producers"||100,000||none|
It is partially true, but irrelevant, that the institutional customers of two other favored employees flipped their shares in the first two days of aftermarket trading. As a general matter, brokerage firms cannot even be sure that their institutional customers are flipping (Tr. 592-94, 781, 1170-71). The favored employees who dealt with institutional customers were:
|Name of representative||Shares allocated||Shares flipped|
The weight of the evidence does not show that HJM consistently exempted its favored employees from penalty bid assessments. For example, Steve Musielski was identified as a favored employee on DX 11 and DX 12. When his customers flipped 1,300 shares, he was assessed a penalty of $390 (DX 34 at 12098). Joseph Hice was also identified as a favored employee on DX 11 and DX 12. When his customer flipped 3,050 shares, he was assessed a penalty of $915 on the IPO and he earned no commission on the sale (DX 34 at 12048-49). The Division also identifies Keohne and Cruz as favored employees (DX 42, DX 43). When Keohne's customers flipped, he was assessed $5,895 in commission reversals and he received no commission on the sale of 300 shares (DX 34 at 12030, 12033). Cruz, the San Francisco branch office manager, received no commission income on two flips of 3,000 shares during the first two days of aftermarket trading in Borealis (DX 34 at 11989). I find as a fact that HJM's "policy" of exempting its favored employees from the penalty bid during the first two days of aftermarket trading in Borealis was hit-or-miss.
Setteducati's Role In Assessing
And/Or Waiving Penalty Bid
The OIP does not claim that Setteducati had any role in imposing or waiving HJM's in-house penalty bid policy, nor does it challenge HJM's external penalty bid policy (i.e., regarding commission forfeiture by selling group firms whose customers flipped their Borealis shares). In contrast, the Division's More Definite Statement charges Setteducati both with selective enforcement of HJM's internal policy, and with failure to enforce HJM's policy against the selling group firms.
In its posthearing pleadings, the Division alleges that Setteducati imposed the penalty bid against HJM's rank-and-file representatives and waived it for a select few brokers, including himself (Div. Prop. Find. # 84; Div. Br. at 22). The evidence does not support this charge. Setteducati was one of several officials who had the authority to adjust a registered representative's commissions, and he did so in specific cases (Tr. 165, 200-01, 1519-20). However, there is no evidence that he exercised that authority in connection with the Borealis offering. Masucci, not Setteducati, imposed and waived the penalty bid on Borealis (Tr. 262-65; DX 34).
As found above, Setteducati was one of several favored employees of HJM whose customers flipped their Borealis IPO shares in the immediate aftermarket (DX 11, DX 12, DX 43). Twenty-two of his customers bought 30,000 shares of Borealis at the IPO price of $5.00 per share (DX 34 at 12101-03). Several of these customers were Setteducati's friends or neighbors (Tr. 1462-65). All twenty-two then sold their Borealis shares at a profit between June 24 and June 27. On June 24, twenty of Setteducati's customers sold 26,000 shares of Borealis at prices ranging from $7.25 to $7.50 per share (DX 42, DX 59). On June 25, Setteducati's customer # 55411 sold 3,000 shares (DX 34 at 12102). The sales ticket is not part of the record, but the Division's summary exhibit states that the price was also within the $7.25 to $7.50 range (DX 34 at 12102, DX 43). On June 27, Setteducati's customer # 04115 sold 1,000 shares at an unknown price (DX 34 at 12103). There is no evidence that any of Setteducati's customers "supported the deal" by purchasing additional Borealis shares in the immediate aftermarket.
Setteducati retained his commissions on all of these Borealis IPO purchases and on most of the Borealis sales, even though his customers were flipping. However, on two flipped sales, Setteducati forfeited his commission (Tr. 889-91, 894-95; DX 34 at 12101-02; Div. Request for Civil Penalties and Disgorgement, Exhibit C, dated January 15, 2001). Setteducati testified that HJM's penalty bid policy was not enforced against him during the Borealis IPO (Tr. 1456-57). I have considered this testimony, but I place greater weight on the documentary evidence. I find that Setteducati usually, though not always, was exempted from the penalty bid on Borealis.
The Division told its expert witness to assume that HJM had waived the penalty bid as to the members of the Borealis selling group (Tr. 1122; DX 57 at 2). The expert did so, and then testified that he did not know of other IPOs where the penalty bid had been imposed on the lead underwriter's own representatives, but not on the other brokerage firms in the selling group (DX 57 at 18). In its posthearing pleadings, the Division asserts that such a selective imposition of the penalty bid is "unheard of in the securities industry" (Div. Prop. Find. # 79; Div. Br. at 22).
Setteducati was a member of HJM's investment banking committee and it is reasonable to infer that he participated in that committee's selection of selling group members (Tr. 904-05, 1455). However, the Division presented no probative evidence as to what the investment committee's policy may have been as to imposing or waiving the penalty bid for selling group firms, either in the Borealis IPO or on other new issues underwritten during the relevant period.23 I find that the Division has failed to prove that HJM had an external penalty bid policy or waived the penalty bid as to the brokerage firms in the Borealis selling group.
The Price Of Borealis Remained Steady
In The Short Run, But Declined Over The Long Run
The OIP asserts that, on July 24, 1996, thirty days after the IPO, Borealis fell below its $5.00 offering price for the first time (OIP ¶ II.AC). It further alleges that Borealis "continued on a long downward trend" until August 6, 1999, when it closed at $0.06 per share (OIP ¶ II.AC). According to the Division, after HJM ended its upward price manipulation, normal market forces took effect and "inexorably lowered" the price of Borealis (Tr. 9, 34; Div. Br. at 37-38; Div. Reply Br. at 6).24
During the first week of aftermarket trading, DX 50 and RX 12 show that the daily high, low, and closing prices for Borealis were as follows:
|June 24, 1996||$9.00||$7.25||$8.125|
|June 25, 1996||$8.125||$7.50||$8.125|
|June 26, 1996||$8.125||$7.00||$7.375|
|June 27, 1996||$7.375||$6.50||$7.125|
|June 28, 1996||$7.25||$6.75||$6.75|
During the week of July 1 through July 5, 1996, Borealis never traded below $6.375 per share (RX 12). During the week of July 9 through July 12, 1996, Borealis never traded below $5.875 per share, and it closed the week at $6.875 per share (RX 12). During the week of July 15 through July 19, 1996, Borealis never traded below $6.125 per share (RX 12).
On July 23, 1996, Borealis traded at a high of $6.50 per share and closed at the daily low of $5.625 per share (RX 12).25 On July 24, 1996, Borealis traded below its offering price for the first time. The lowest trade on that date was at $4.625 per share and the closing trade was at $4.875 per share (DX 50, RX 12).
Borealis traded at or above $5.00 per share on July 25, 1996, and on several days in August, September, and October 1996 (RX 12). It also traded above $5.00 per share on December 2, 1996 (RX 12). One customer's monthly account statements demonstrate that Borealis traded at $5.50 per share on January 31, 1997, at $6.125 per share on April 25, 1997, and at $5.25 per share on September 26, 1997 (Tr. 1356-59).
Over the long term, however, the trading price of Borealis declined gradually. In October 1998, NASDAQ's SmallCap Market delisted Borealis, which was then trading at approximately $0.75 per share (DX 57 at 18-19).
On January 28, 1999, Borealis filed a Form 15 with the Commission to terminate its registration under Section 12(g) of the Exchange Act (official notice). See Exchange Act Rule 12g-4(a)(1)(i), 17 C.F.R. § 240.12g-4(a)(1)(i). Once the Form 15 became effective on April 28, 1999, the company was no longer required to comply with the rules and regulations under the Exchange Act. On August 31, 1999, Borealis was quoted at $0.06 per share on the NASD's electronic bulletin board (DX 50, DX 57 at 19, 22).
Setteducati's Role In The Borealis IPO
And The Immediate Aftermarket
Setteducati's role in preparing the June 6, 1996, preliminary allocation worksheet has been described above. Otherwise, there is little direct evidence of how Setteducati interacted with HJM's branch office managers, registered representatives, and trading desk in the weeks before the Borealis IPO and during the first week of aftermarket trading. In the absence of such direct evidence, the Division attempted to show how Setteducati usually interacted with HJM's sales force and trading desk during other IPOs. It asks me to infer that that he likely acted in conformity with his usual practice during the Borealis IPO (Tr. 304-06; cf. Fed. R. Evid. 406).26
Setteducati's Interaction With HJM's
Sales Force On Subsequent IPOs
The OIP alleges that HJM allocated Borealis IPO shares with intent to ensure that most of those shares would not be traded in the immediate aftermarket (OIP ¶ II.I). The Division's More Definite Statement, liberally construed, contends that Setteducati contributed to this effort by insisting that HJM's branch office managers and registered representatives "support the deal" (Div. Prop. Find. ## 44-51, 59, 60, 90). An additional theory of the prosecution, not evident in the OIP or the More Definite Statement, is that the branch office managers received their marching orders on IPOs from Setteducati in the form of daily conference calls, and that the registered representatives participated in occasional "squawk box" calls on IPOs that enabled Setteducati to communicate with all the firm's representatives at once.
James Battaglia (Battaglia), HJM's Dallas, Texas, office manager, did not pay much attention to the Borealis IPO because his branch was in poor financial shape at the time (Tr. 239, 244). Battaglia instead testified about daily telephone conference calls between HJM's upper management and the firm's branch office managers on other IPOs (Tr. 233-34, 243). During these calls, Setteducati generally wanted the branch managers to commit to taking a specific number of IPO shares (Tr. 249). Setteducati monitored the progress of the branch offices as they took indications of interest from the public (Tr. 251-53). Setteducati told the branch managers that their customers should not only hold IPO shares for the long term, but also should buy more shares in the immediate aftermarket (Tr. 259-61, 1442-45). Setteducati was sometimes abusive, vulgar, and belligerent during these calls. He raised his voice and threatened to fire "underperforming" branch managers (Tr. 250-53, 261).
Jeffrey Briggs (Briggs) was not employed by HJM at the time of the Borealis IPO, but he managed HJM's Rochester branch office from May 1997 to September 1998 (Tr. 303, 332). During that time, HJM underwrote only two IPOs (Tr. 339). Like Battaglia, Briggs participated in telephone conference calls between HJM's upper management and the branch office managers. Briggs recalled that Setteducati personally monitored indications of interest as they were being submitted and would become upset if a branch did not keep pace with its goal (Tr. 322). Briggs described Setteducati's demeanor on such calls as stern, direct, aggressive, and loud (Tr. 313, 319, 327). In these calls, Setteducati emphasized that it was important to hold IPO shares for the long term, to "support the deal" by encouraging customers to buy in the immediate aftermarket, and to avoid flipping (Tr. 323-25). Briggs further testified that Setteducati pressured branch managers to sell a specific number of shares through the use of vulgarity, threats, and reprimands (Tr. 312-13, 316).27 Branch managers who disputed Setteducati were chastised or ignored on subsequent calls (Tr. 316-19).
Two of HJM's registered representatives also described their interaction with Setteducati during IPOs. Bartelt testified about a sales meeting in the autumn of 1996 when Setteducati addressed the Chicago branch representatives (Tr. 685-86). At that meeting (which took place several months after Borealis went public), Setteducati emphasized that HJM specialized in underwriting IPOs. He also used profanity to emphasize to the representatives that, if they did not want to support the firm's IPOs in the immediate aftermarket, the firm did not want them to be employed there (Tr. 625-26, 686, 690). Barry Hayut (Hayut), from HJM's Atlanta office, did not participate in the Borealis IPO, but he recalled a few telephone conference calls with Setteducati on other IPOs (Tr. 744-46, 754). In Hayut's judgment, Setteducati wanted the registered representatives to work hard, but he did not encourage them to generate revenue in an unethical fashion (Tr. 748-49). Hayut blamed Masucci, not Setteducati, for micro-managing HJM's retail sales effort (Tr. 752-53, 764-65).
Two representatives from Chicago said their branch office manager was "getting heat" from "upper management" about IPOs because he made a "commitment" to "corporate" about aftermarket sales (Tr. 642-43, 828-29, 863). The witnesses could not identify who in "upper management" or "corporate" had applied that pressure (Tr. 642, 829). Battaglia was asked if improper conduct by Setteducati or other HJM managers caused him to deviate in any way from an appropriate professional level of care (Tr. 268-74). After fencing with his questioner for several minutes, Battaglia ultimately gave a non-responsive answer. Battaglia acknowledged that he had fired a registered representative for failing to participate in an IPO, but also said that the representative in question had "stiffed" the firm for a $40,000 advance (Tr. 270). In other words, Battaglia thought the discharge was for good cause. Bartelt testified about a "ten or ten" policy-a requirement that representatives in the Chicago branch open ten new accounts or work until 10 p.m. (Tr. 617, 641, 690, 707). However, he emphasized that his own branch office manager instituted that policy (Tr. 617, 709). There is no basis for inferring that Setteducati encouraged that policy, acquiesced in it, or even knew about it.
Setteducati acknowledged his use of profanity, as well as his efforts to "hold people accountable" for sales goals (Tr. 1622-23). He admitted that he was passionate about his work, but denied using high-pressure tactics (Tr. 1622-23). Setteducati could not recall his role in the Borealis IPO and aftermarket (Tr. 1433-36, 1636-38). During the investigation, Setteducati testified that HJM generally recommended that its customers hold IPO shares of thinly capitalized companies for twelve to thirty-six months (Tr. 1444-45). At the hearing, however, Setteducati offered more nuanced testimony: HJM did not have such a philosophy "written in stone," and in any event, the firm's philosophy would apply only if a lengthy holding period made sense for the customer (Tr. 1442-43). I do not give much credence to that aspect of Setteducati's hearing testimony that purported to clarify his investigative testimony.
Setteducati's Interaction With
HJM's Trading Desk
The Division also contends that Setteducati was in frequent contact with Vanechanos, HJM's head trader (Div. Prop. Find. ## 68-72, 108, 111; Div. Br. at 30). Vanechanos worked in HJM's New York City office, and was responsible for trading Borealis in June 1996 (Tr. 1018-19). At the time, Vanechanos reported to Michael Bresner (Bresner), and Bresner reported to Villa (Tr. 746-48, 770, 1102, 1498-99, 1501). Bresner did not begin to report to Setteducati until after the Borealis IPO (Tr. 1514-15).
Setteducati and Vanechanos spoke by telephone daily. Setteducati also visited HJM's trading room when he was in New York City (Tr. 980, 982). Most of these conversations and visits involved social or routine business matters, but Setteducati also mediated disputes between HJM's sales force and the firm's trading desk (Tr. 982-83, 993, 1500-01). HJM authorized its upper management officials and branch office managers to contact the firm's trading desk directly (Tr. 992). Consistent with this firm-wide practice, Setteducati's secretary could place customer orders by telephone (Tr. 992). There is no evidence that any of Setteducati's conversations or visits with Vanechanos involved the Borealis IPO or the specifics of any trade by any broker for any security.28
Late in the summer of 1996, Donnie Casadonte (Casadonte) replaced Vanechanos as HJM's head trader and Theodore Colby (Colby) joined the firm to oversee compliance in the trading department (Tr. 968, 1007, 1068). Casadonte told Colby that he sometimes took trading directions and strategies from Setteducati (Tr. 1025-27). The Division asks me to infer from this fact that Vanechanos also took trading directions from Setteducati on Borealis during the week of June 24 through June 28, 1996. I find this speculative and decline to do so. Setteducati's expanded duties, Casadonte's promotion, Vanechanos's transfer to Red Bank, and Colby's hiring all followed the July 15, 1996, NASD settlement (Tr. 967-68, 1068; DX 9). The Division has not shown, and I decline to infer, that conditions within HJM's trading department during June 1996 were the same as they were after July 1996.29
Colby took his job seriously. He was aggressive in monitoring HJM's trading positions, and he did not hesitate to declare HJM to be in a position of domination and control when he believed it appropriate. Such declarations triggered contemporaneous cost pricing and were unpopular with HJM's sales force (Tr. 797, 807, 1027-31). As the mediator of disputes between sales and trading, Setteducati telephoned Colby to inquire about the reasons for these decisions (Tr. 1500-01). Colby always found Setteducati to be cooperative, cordial, and supportive of his efforts to achieve compliance (Tr. 1030, 1064, 1066). Colby was a very credible witness, and his testimony about Setteducati's demeanor contradicts the testimony of other less credible witnesses who portrayed Setteducati as a bully.
In 1996, HJM's compliance staff prepared and distributed on a daily basis a report, the Market Maker Summary, designed to alert the head trader to situations in which the firm was in a potential domination and control situation (Tr. 1047-48, 1088-89; DX 51). Setteducati received the Market Maker Summary, as did the head trader, the general counsel, and other managers in the firm (Tr. 1090). As a result of inaccuracies and omissions in the Market Maker Summary, that report often failed to indicate situations in which HJM was dominating and controlling trading in stocks (Tr. 1052-53, 1060). The inaccuracies stemmed from tardy reporting of data to HJM by its clearing broker (Tr. 1092, 1095). The omissions arose from the fact that HJM did not include in the report trades that had not yet settled.
Colby opined that HJM's management did not want the Market Maker Summary's methodology corrected, because that would have forced the head trader to begin employing contemporaneous pricing (Tr. 1076). Colby corrected the situation in January 1997 (Tr. 1049-50, 1059-60).
At some point, Setteducati became aware of the deficiencies in the Market Maker Summary (Tr. 1079-80). However, Setteducati insists that he did not keep abreast of the firm's market making positions in June 1996 (Tr. 1633). Colby's testimony cannot be used to impute knowledge of these deficiencies to Setteducati as of that time, because Colby was not then employed by HJM. In any event, this evidence cannot be considered against Setteducati on the issue of Setteducati's liability for markup violations. The Division has twice stated that it would not seek to impose liability on Setteducati for the markup violations that the OIP attributed only to Vanechanos (Division's More Definite Statement, ¶¶ II.AD, II.AE, II.AF; Tr. 5-7, 122-24). The Division is bound by those statements. Setteducati's purported knowledge of the deficiencies is not relevant to the acts and omissions that the OIP does attribute to him.
The Division's Fact Witnesses
Most of the Division's testimonial proof comes from multiple, mutually corroborating witnesses, and I consider it to be reliable. However, there are a few exceptions.
Battaglia, who began his career with First Jersey Securities, had been employed by TJA and HJM even longer than Setteducati (Tr. 230, 275). Battaglia testified that the IPO process at TJA and HJM never really changed from 1984 to 1998 (Tr. 236). In contrast, Briggs, who worked for TJA from 1985 to 1988 and for HJM from May 1997 to September 1998, testified that the IPO process was "significantly changed" with "a lot more procedures in place" in 1997 and 1998 (Tr. 308-09). I credit Briggs and disbelieve Battaglia on this point. Battaglia also asserted that two of his customers flipped 6,500 shares of Borealis on the second day of aftermarket trading, and he believed he lost his commissions on those transactions because of HJM's penalty bid policy (Tr. 266). However, HJM's commission runs showing purchases and sales of Borealis stock during June and July 1996 do not list Battaglia (Tr. 876-78, 880-82; DX 34). It is impossible to say if the fault lies in Battaglia's memory or if DX 34 is incomplete, but neither alternative helps the Division.
I found Battaglia less than forthcoming about his own role in IPOs: he acknowledged that he made commitments in conference calls to obtain customer indications of interest, but insisted that he did so only to "appease management" and did not always keep those commitments (Tr. 249-50). He testified that Setteducati pressured the branch office managers, but he could not or would not identify situations when pressure from Setteducati led him to depart from an appropriate professional level of care (Tr. 268-74). In perhaps his least sincere testimony, he lamented that he "had taken part in a process that I felt was-had been already judged by the [SEC] in 1989 and 1990, as illegal, yet the process continued in much the same fashion that it did before" (Tr. 274). Despite Battaglia's professed discomfort, he continued to work as a branch manager for TJA and HJM for another eight years after the 1990 settlement. I have relied on Battaglia's testimony only to the extent that it is corroborated by evidence from other sources.
Bartelt was not a very believable witness, either. He was quite sure (and quite wrong) about several matters, including the description of the penalty bid in the issuer's prospectus, the job titles of Masucci and Setteducati, the percentage of Borealis IPO shares allocated to the Chicago branch office, and the unique typeface on the Chicago branch manager's facsimile machine (Tr. 621, 625-26, 632, 647-48, 674-75, 706-07). Bartelt presented a series of contradictions. The Division offered Bartelt as one of the few honest men in what was ostensibly a hotbed of fraud, HJM's Chicago branch office. Yet Bartelt came to HJM from L.C. Wegard with a minor disciplinary record of his own (Tr. 608-09). The Division hoped to use Bartelt's testimony to show that Setteducati's profanity was intimidating, yet Bartelt said that he laughed at Setteducati's profane remarks to the Chicago registered representatives (Tr. 686, 690-91). Bartelt considered HJM to be a "chop shop," but he remained with the firm for seventeen months (Tr. 691, 702). Bartelt believed that HJM's Chicago branch office had a good compliance officer. Although that compliance officer had an open door policy, Bartelt never brought any issues about abusive sales practices to the compliance officer's attention (Tr. 659, 672-73). The reason was that he never knew of any such practices (Tr. 716-17). Bartelt portrayed himself as a model of good sales practices: he never told his customers to indicate for more IPO shares than they wanted, never made unsuitable recommendations to his customers, never misrepresented risk, never made an unauthorized purchase, and never refused a customer's sell order (Tr. 633-34, 670-71, 716). He made aftermarket sales commitments only to appease his branch manager, but he never really tried to honor them (Tr. 636-38). I have credited Bartelt's testimony only to the extent that it is corroborated by the testimony of others, or by documentary evidence.
Myers, a customer of HJM's San Francisco branch office, was also an unreliable witness. She was highly educated, earning an A.B. degree from Vassar College, and an M.S. and Ph.D. from the University of California at Berkeley, each with a concentration in mathematics (Tr. 1327).30 Myers was financially sophisticated. At the time of the Borealis IPO, she owned a commercial retail property and had interests in mixed-use development projects. She was also a limited partner and had interests in apartment and offices complexes (Tr. 1370-71, 1373). Myers visited HJM's San Francisco office two or three times (Tr. 1364-65). Borealis was not her first IPO, nor was it her last (Tr. 1335, 1378). In essence, Myers was an IPO veteran, well schooled in the ways underwriting firms "get you to buy these IPOs and they resist your ability to sell them" (Tr. 1363). Myers was the lead plaintiff in litigation that unsuccessfully charged several firms that underwrote IPOs with violations of California state law. See Myers v. Merrill Lynch & Co., 249 F.3d 1087 (9th Cir. 2001), aff'g, 1999 U.S. Dist. LEXIS 22642 (N.D. Cal. Aug. 23, 1999) (alleging that penalty bids and related efforts to combat flipping violated the California Business and Professional Code). She made it clear that she often failed to read mail HJM sent her concerning the status of her account (Tr. 1356-57) ("My life is too busy for that."). On cross-examination, Myers was combative without provocation.31
The Division's Expert Witness
Jay R. Ritter (Ritter), Ph.D., is a professor in the Department of Finance at the University of Florida (DX 57 at 1). He has written and lectured extensively on the subject of IPOs (Curriculum Vitae attached to Division's Designation of Expert Witnesses, dated September 1, 2000). Ritter analyzed the price behavior of Borealis stock in the immediate aftermarket of the June 1996 IPO. He determined that the trading patterns for Borealis on the first day were extremely unusual for an IPO (DX 57 at 11). Ritter concluded that HJM dominated and controlled the market for Borealis and manipulated its price (DX 57 at 2-3).
Certain of Ritter's observations make a great deal of sense and Setteducati has not even attempted to challenge them. For example, Ritter analyzed the trading data and found that many of the IPO shares allocated to brokerage firms in the selling group were flipped to HJM. HJM then sold the appreciated shares to its own customers (Tr. 1142-43; DX 57 at 11-12). In the process, HJM increased its domination and control of the market for Borealis.
Ritter viewed the large amount of time that HJM spent at the inside bid and the small amount of time that HJM spent at the inside ask as consistent with the trading data showing that HJM allowed other brokers and dealers to reduce their holdings of Borealis while HJM accumulated shares of Borealis and sold them to its own customers (DX 57 at 6). In Ritter's view, such behavior was consistent with domination and control of the market and facilitated a manipulation of the stock price (DX 57 at 6).
Ritter also opined that a wide spread between the inside bid and inside ask prices and heavy trading volume on the first day of trading would ordinarily have attracted other market makers (DX 57 at 13). The fact that it did not happen with Borealis was consistent with HJM's domination and control of the market. In Ritter's opinion, the stickiness of Borealis's spread and its high level on the first day of trading were indicative of a manipulation (DX 57 at 9-10). Ritter did not examine the stickiness or level of the spread on the second through fifth day of the alleged manipulative period.
Ritter viewed Borealis as a classic "house stock," with the vast majority of its shares held by the customers of one brokerage firm (DX 57 at 17). He opined that house stocks are attractive to brokerage firms because the absence of independent market makers means that there is no competitive downward pressure on the size of the bid-ask spread. A wide bid-ask spread results in high profits for the dominant market maker (DX 57 at 17).
Notwithstanding Ritter's eminent qualifications, cross-examination brought out several aspects of his analysis that must be read with caution, or discounted in part or in whole. For example, in formulating his opinion, Ritter relied on the Division's summary exhibits and he made several assumptions that the Division asked him to make (Tr. 1121-22; DX 57 at 2, 14, 16). To the extent that the underlying summary exhibits were inaccurate, or the assumptions were unproven, the weight accorded to Ritter's opinions must be reduced.
Ritter repeatedly drew distinctions between underwriters with "high" and "low" reputations (DX 57 at 5, 9-10, 19). He explicitly relied on HJM's poor reputation as an important factor in reaching his conclusions about the manipulation of Borealis (Tr. 1164-65; DX 57 at 5, 9-10, 19). That is exactly the sort of reasoning that Federal Rule of Evidence 404(b) does not countenance. See supra note 2 and infra note 49.
Ritter also expressed the view that Borealis was worthless and doomed to failure from the outset, and that anyone with any common sense should have known that it was worthless and doomed to failure (Tr. 1132, 1168; DX 57 at 3, 4, 21) (opining that "any objective observer" would have placed a low valuation or no valuation on the stock; that Borealis was "almost certain to never achieve profitability;" that Borealis should not have gone public at any price; and that the cash raised in the IPO "was almost certainly going to be squandered"). This hindsight cannot be given much weight, for the reasons discussed below.
Ritter made only passing reference to "the 1996-2000 investment climate" (DX 57 at 4). These were the bubble years for IPOs in the high technology sector of the market. Ritter's opinions about Borealis's dim prospects gave no consideration to the fevered investment climate surrounding IPOs of high technology companies during 1996. In that climate, the public was snapping up IPO shares even though rigorous fundamental analysis might have suggested that there were grounds for skepticism about long-term profitability.
Ritter's own publications observe that IPO investors base their decisions not on a rational analysis of fundamentals, but on fads. As Ritter has previously written:
For IPOs, the prior rapid growth of many of the young companies makes it easy to justify high valuations by investors who want to believe that they have identified the next Microsoft. . . . Investors are betting on longshots. . . . Investors seem to be systematically misevaluating the probability of finding a big winner. It is the triumph of hope over experience.
Tim Loughran and Jay R. Ritter, The New Issues Puzzle, 50 J. Fin. 23, 46-47 (March 1995); see also Jay R. Ritter, The Long-Run Performance of Initial Public Offerings, 46 J. Fin. 3, 18-20, 23-24 (March 1991) (concluding that long-term underperformance of IPOs may be caused by "irrationally optimistic forecasts [by the initial market] or `fads'"); Jay R. Ritter and Ivo Welch, A Review of IPO Activity, Pricing, and Allocations, 57 J. Fin. 1795, 1800 (forthcoming Aug. 2002) (finding that the percentage of firms with negative earnings before going public was 19% in the 1980s, and 37% from 1995 to 1998). Ritter's determination to focus solely on Borealis's fundamentals, without mentioning the public's support of IPOs for reasons having nothing to do with fundamentals, distorts an understanding of the legitimate economic forces at work in June 1996.
Ritter next attempted to contrast Borealis with Gensym Corporation (Gensym), a company that also went public in 1996 (DX 57 at 5-6, 8). In Ritter's judgment, Gensym (unlike Borealis) had legitimate expectations and hopes and was brought to market by a "reputable" underwriter (Tr. 1130-31). When Ritter learned that the long-term price performance of Gensym was little better than the long-term price performance of Borealis, he could only shrug and observe: "Many young companies fail to live up to expectations" (Tr. 1132, RX 13).32
Ritter also pointed to the lack of venture capital financing for Borealis as prima facie evidence that both the issuer and the underwriter "knew" that Borealis's prospects were dim (DX 57 at 4). Venture capital firms conduct exhaustive research into the prospects and potential of young companies. They have a level of expertise that is not easily obtained by other market participants. Venture capital backing is certainly evidence of what the "smart money" is doing, and venture capitalists undoubtedly are more adroit at picking winners than is the public at large. Nonetheless, Ritter's point is overstated. First, the Division never presented evidence that Borealis sought venture capital backing and failed to obtain it. Nor was there evidence that Borealis deliberately decided not to approach any venture capitalists because it knew that such overtures would be futile. Second, I reject Ritter's implicit assumption that venture capital firms ought to be the gatekeepers to the public markets and that no young company should go public without the blessing of a venture capitalist. The available data show that only about one in three IPOs had venture capital backing during 1996.33
Ritter did not fully appreciate the extent to which institutional customers with European banking relationships purchased shares of Borealis in the IPO and the immediate aftermarket (Tr. 1159-60). He candidly acknowledged that he should have taken a closer look at the institutional trading in Borealis before he issued his report (Tr. 1172-73).
Ritter focused on the price behavior of Borealis on June 24, 1996-the first day of aftermarket trading. He gave scant attention to the price behavior of Borealis on June 25 through June 28, 1996-the bulk of the manipulative period identified in the OIP. Ritter explained (DX 57 at 5):
The first day of trading receives the most attention in this report partly because [the trading volume of Borealis was heavier then in comparison to subsequent days, as is typical with IPOs], and partly because the academic IPO literature has focused on the first day. . . . If going beyond one day, there is no obvious number of days to choose without being somewhat arbitrary, unless one has a specific purpose in mind. . . . [I]f a stock price can be manipulated up on the first day, there is a tendency for it to stay there with relatively low incremental effort on subsequent days. Thus, whatever level it attains on the first day has disproportionate importance for determining what price it will remain at.
Ritter's focus on the first day's closing price ($8.125), instead of the closing price on the last day of the alleged manipulative period ($6.75), greatly increased the magnitude of his measurement of Borealis's subsequent long-term price decline.34 However, Ritter's methodology failed to acknowledge that the alleged manipulation was still ongoing, and the closing price was dropping, on June 26 through June 28. In my judgment, the closing price on the last day of the alleged manipulative period was the appropriate place to begin this measurement.
Ritter offered very little analysis of the long-run performance of Borealis (DX 57 at 18-19, 22). He made no effort to identify and exclude from consideration any variables other than the cessation of the manipulation (i.e., any adverse business developments) that might have contributed to Borealis's long-term price decline. I have given very little weight to Ritter's opinion that the price of Borealis remained artificially high until August 1999, when Borealis was quoted at $0.06 per share (Tr. 1130). By August 1999, HJM had been out of business for nearly eleven months.
Finally, Ritter's hearing testimony minimized the importance of certain points he raised in his prepared statement. In his written report, Ritter opined that HJM's selective use of the penalty bid provided strong evidence of manipulation (DX 57 at 17-18). He also drew comparisons between the long-run abnormal price performance of Borealis in 1996 and the long-run abnormal price performance of four stocks TJA brought public in 1989 (DX 57 at 20). At the hearing, however, Ritter stated that neither HJM's selective use of the penalty bid nor the Commission's 1990 case against TJA had been an important factor in reaching his conclusions about the manipulation of Borealis (Tr. 1136, 1164). Ritter distanced himself from these two pieces of evidence, but his conclusions about Borealis do not become impossible because he could still rely on other factors. However, his concessions at the hearing detracted from the picture he drew in his prepared statement.
THE APPLICABLE LAW
The Commission has characterized manipulation as "the creation of deceptive value or market activity for a security, accomplished by an intentional interference with the free forces of supply and demand." Howard R. Perles, 77 SEC Docket 896, 905 & n.14 (Apr. 4, 2002); Swartwood, Hesse, Inc., 50 S.E.C. 1301, 1307 (1992).
The basic aim of the antifraud provisions of the securities laws is to "prevent rigging of the market and to permit operation of the natural law of supply and demand." First Jersey Secs., 101 F.3d at 1466. In Section 2 of the Exchange Act, Congress explained that one of the primary objectives of the statute was to "insure the maintenance of fair and honest markets."
The manipulation of securities prices, which "constitutes intentional or willful conduct designed to deceive or defraud investors by controlling or artificially affecting the price of securities," Ernst & Ernst v. Hochfelder, 425 U.S. 185, 199 (1976), runs directly counter to that objective. Accordingly, several provisions of the federal securities laws, including Section 10(b) of the Exchange Act and Rule 10b-5 thereunder, prohibit such conduct. See United States v. Russo, 74 F.3d 1383, 1390-91 (2d Cir. 1996).
Section 10(b) of the Exchange Act
Section 17(a) of the Securities Act
Section 10(b) of the Exchange Act and Rule 10b-5 thereunder prohibit the use of "any manipulative or deceptive device or contrivance" in connection with the purchase or sale of any security. Section 10(b) encompasses "(1) using any deceptive device (2) in connection with the purchase or sale of securities, in contravention of rules prescribed by the Commission." United States v. O'Hagan, 521 U.S. 642, 651 (1997). Rule 10b-5 contains "flat prohibitions of deceitful practices and market manipulations." United States v. Charnay, 537 F.2d 341, 350 (9th Cir. 1976). Conduct falling within the purview of Rule 10b-5 includes "any activities that falsely persuade the public that activity in an over-the-counter security is `the reflection of a genuine demand instead of a mirage.'" SEC v. Resch-Cassin & Co., 362 F. Supp. 964, 975 (S.D.N.Y. 1973); accord, Swartwood, Hesse, 50 S.E.C. at 1307.
Congress did not limit Section 10(b)'s application to the manipulative and deceptive devices or contrivances known in 1934. See Herpich v. Wallace, 430 F.2d 792, 801-02 (2d Cir. 1970). Section 10(b) and Rule 10b-5 are "not intended as a specification of particular acts or practices that constitute `manipulative or deceptive devices or contrivances,' but are instead designed to encompass the infinite variety of devices that are alien to the climate of fair dealing." Id. To effectuate the Congressional intent, "Section 10(b) must be read flexibly, not technically and restrictively." Supt. of Insurance v. Bankers Life & Cas. Co., 404 U.S. 6, 12 (1971).
Section 17(a)(1) of the Securities Act makes it unlawful, in the offer and sale of securities, to employ devices, schemes or artifices to defraud. Section 17(a)(2) of the Securities Act prohibits material misstatements or omissions of material fact, and Section 17(a)(3) of the Securities Act prohibits transactions, practices, or courses of business that operate as a fraud or deceit upon the purchaser.
Recognizing that Section 10(b) outlaws but does not define a "manipulative or deceptive device or contrivance," the Division turns to Section 9(a)(2) of the Exchange Act to determine the elements of the offense of manipulation (Div. Br. at 6, 31-32).
Section 9(a)(2) of the Exchange Act makes it unlawful, with respect to a security listed on a national securities exchange, to effect: (1) a series of security transactions, alone or with one or more persons; (2) which create actual or apparent active trading in such security or which raise or depress the price of such security; (3) for the purpose of inducing others to buy or sell the security. Manipulative activities of the type prohibited by Section 9(a)(2) of the Exchange Act are also violations of Section 17(a) of the Securities Act and Sections 10(b) and 15(c) of the Exchange Act. In addition, Section 17(a) of the Securities Act and Sections 10(b) and 15(c) of the Exchange Act are deemed to prohibit manipulative activities with respect to securities that are not traded on a national securities exchange. See Resch-Cassin, 362 F. Supp. at 975; Halsey, Stuart & Co., 30 S.E.C. 106, 110-11 (1949). When the basis of liability rests on Section 10(b) and Rule 10b-5, however, it is not necessary to show a manipulative purpose in inducing others to trade. Instead, it is sufficient to show that the person engaged in a fraud or deceit as to the nature of the market for the security. See Charnay, 537 F.2d at 350.
The court in Resch-Cassin listed various factors that characterize attempts by manipulators to raise the price of a security. Among them are domination and control of the market, price leadership by the manipulator, restricting the floating supply of the security, and the decline of the market for the security after the manipulator ceases his manipulative activities. Resch-Cassin, 362 F. Supp. at 976-77. The Commission has held that a finding of manipulation is not dependent on the presence or absence of any particular device usually associated with a manipulative scheme. Swartwood, Hesse, 50 S.E.C. at 1307.
Scienter is defined as "a mental state embracing intent to deceive, manipulate, or defraud." Hochfelder, 425 U.S. at 193 n.12. It may be established by a showing of recklessness. David Disner, 52 S.E.C. 1217, 1222 & n.20 (1997) (citing Hollinger v. Titan Capital Corp., 914 F.2d 1564, 1568-69 (9th Cir. 1990) (en banc)). The en banc Ninth Circuit adopted the standard of recklessness articulated by the Seventh Circuit in Sunstrand Corp. v. Sun Chem. Corp., 553 F.2d 1033, 1044-45 (7th Cir. 1977): "[A] highly unreasonable omission, involving not merely simple, or inexcusable negligence, but an extreme departure from the standards of ordinary care, and which presents a danger of misleading buyers and sellers that is either known to the defendant or is so obvious that the actor must have been aware of it."
To prevail under Section 17(a)(1) of the Securities Act, Section 10(b) of the Exchange Act, and Rule 10b-5, the Division must show that a respondent acted with scienter. See Hochfelder, 425 U.S. at 193 n.12; Aaron v. SEC, 446 U.S. 680, 701-02 (1980); Pagel, Inc. v. SEC, 803 F.2d 942, 946 (8th Cir. 1986). No scienter requirements exists for violations of Sections 17(a)(2) or 17(a)(3) of the Securities Act; rather, negligence alone is sufficient. Aaron, Pagel.
Section 15(c)(1) of the Exchange Act
In relevant part, Section 15(c)(1) prohibits brokers and dealers from using "any manipulative. . . device or contrivance" in connection with securities transactions that are not conducted on a national securities exchange. Exchange Act Rule 15c1-2 states that the "manipulative. . . device or contrivance" prohibited in Section 15(c)(1) includes any act, practice, or course of business which operates or would operate as a fraud or deceit upon any person. Exchange Act Rule 15c1-8 states that the prohibitions of Section 15(c)(1) include any representation made to a customer by a broker who is participating or otherwise financially interested in the primary or secondary distribution of any security that is not traded on a national securities exchange that the security is being offered to the customer at a price related to the market price unless the broker knows or has reasonable grounds to believe that a market for the security exists other than the market made, created, or controlled by the broker or dealer.
To establish a primary violation of Section 15(c)(1) by HJM, the Division must show that the broker acted with scienter. See SEC v. Dowdell, Fed. Sec. L. Rep. (CCH) ¶ 91,728 at 98,363 (W.D. Va. 2002); Darvin v. Bache Halsey Stuart Shields, Inc., 479 F. Supp. 460, 464 (S.D.N.Y. 1979); L.C. Wegard & Co., 53 S.E.C. 607, 615 (1998), aff'd, 189 F.3d 461 (2d Cir. 1999). The Division has stipulated that it must also prove scienter to establish the brokerage firm's primary violations of Exchange Act Rules 15c1-2 and 15c1-8 (Div. Br. at 5 n.2).
To establish that one respondent willfully aided and abetted the violation of another, the Division must also show that the aider and abetter acted with scienter. See Adrian C. Havill, 53 S.E.C. 1060, 1070 n.26 (1998). The three elements necessary to find aiding and abetting are: (1) securities law violations by a primary wrongdoer; (2) general awareness or knowledge that the actions of the accused were part of an overall course of conduct that was illegal or improper; and (3) knowing or reckless substantial assistance by the accused in the conduct constituting the primary violations. See Sharon M. Graham, 53 S.E.C. 1072, 1080-81 (1998), aff'd, 222 F.3d 994, 1000 (D.C. Cir. 2000); Russo Secs., Inc., 53 S.E.C. 271, 278 n.16 (1997); Donald T. Sheldon, 51 S.E.C. 59, 66 (1992), aff'd, 45 F.3d 1515 (11th Cir. 1995). Willfulness is shown where a person intends to commit an act that constitutes a violation. There is no requirement that the actor also be aware that he is violating any statutes or regulations. See Wonsover v. SEC, 205 F.3d 408, 414 (D.C. Cir. 2000); Arthur Lipper Corp. v. SEC, 547 F.2d 171, 180 (2d Cir. 1976).
The Commission has held that one who aids and abets a primary violation is necessarily a cause of the violation. See Graham, 53 S.E.C. at 1085 n.35; Havill, 53 S.E.C. at 1070 n.26. However, the case law is not so clear as to whether one who lacks the state of mind to aid and abet a violation can still cause the violation.
Before 1963, the meaning of the word "cause" in the Exchange Act had not been the subject of detailed judicial construction. See Berko v. SEC, 316 F.2d 137, 140-41 (2d Cir. 1963) (holding that conduct which is the "cause" of a violation must consist of more than merely conduct which is "to some degree a factor" in the violation); R.H. Johnson & Co. v. SEC, 198 F.2d 690, 696 (2d Cir. 1952) (rejecting the contention that "cause" must always be interpreted to mean "an immediate and inducing cause").
Subsequent additions to the securities laws, such as Section 21C of the Exchange Act and Section 203(k)(1) of the Investment Advisers Act of 1940, specify that a respondent is a "cause" of another's violation if the respondent "knew or should have known" that his act or omission would contribute to such violation. The Commission has held that negligence is sufficient to establish liability for causing a primary violation that does not require scienter. KPMG Peat Marwick LLP, 74 SEC Docket 384, 421 (Jan. 19, 2001), recon. denied, 74 SEC Docket 1351 (Mar. 8, 2001), petition denied, 289 F.3d 109 (D.C. Cir. 2002).
The Commission has not decided any contested cases in which it explicitly concluded that a respondent was a cause, by negligent acts or omissions, of securities law violations that require scienter to establish the primary violation.35 At least two Administrative Law Judges have held that if the primary violation requires a showing of scienter, there can be no liability for causing that violation without a finding of scienter. See Albert Glenn Yesner, CPA, 75 SEC Docket 220, 255 (Initial Decision) (May 22, 2001), final, 75 SEC Docket 648 (June 19, 2001); Nicholas P. Howard, 69 SEC Docket 1226, 1243 (Initial Decision) (Mar. 24, 1999), review granted. Consistent with Yesner and Howard, and in the absence of any briefing on this issue by the parties, I conclude that the Division must demonstrate scienter to show that Setteducati caused HJM to manipulate the price of Borealis stock in violation of Section 15(c)(1).
Proof of a manipulation almost always depends on inferences drawn from a mass of factual detail, and findings must be gleaned from patterns of behavior, from apparent irregularities, and from trading data. See Wegard, 53 S.E.C. at 613; Pagel, Inc., 48 S.E.C. 223, 226 (1985).36
An adverse inference may be employed to complete a chain of reasoning on a point partially established by direct evidence, but it cannot be used to fill a void where there is otherwise no evidence. See Stanojev v. Ebasco Servs. Inc., 643 F.2d 914, 923-24 n.7 (2d Cir. 1981) (holding that an adverse inference cannot supply the missing element of a prima facie case). Where a plaintiff bears the ultimate burden of persuasion on an element of its case, a rebuttable presumption as to that element shifts the burden to the defendant to provide a persuasive exculpatory explanation for his actions. Once that burden has been met, the fact finder is not bound to find the presumed fact from the basic fact, but he may make that inference. Congress made this clear when it adopted Federal Rule of Evidence 301, which deals with the effect of presumptions in civil cases. See SEC v. Lipson, 278 F.3d 656, 660-62 (7th Cir. 2002).
By analogy to Federal Rule of Evidence 301, if adjudicated facts are subject to competing inferences, the Division, as the party with the burden of proof, must establish that its inferences are more plausible than the respondent's inferences. Cf. Alan J. Rosenberg, [1990-1992 Transfer Binder] Comm. Fut. L. Rep. (CCH) ¶ 24,992 at 37,643 (CFTC Jan. 25, 1991). If the record equally supports both innocent and culpable inferences, the Division fails in its burden of proof. Cf. Louis Abrams, [1994-1996 Transfer Binder] Comm. Fut. L. Rep. (CCH) ¶ 26,479 at 43,136 (CFTC July 31, 1995).
DISCUSSION AND CONCLUSIONS
There is a considerable body of economic and financial literature establishing that IPOs are underpriced in the short run and experience a substantial first-day price run-up. See infra note 42. Flipping, the practice of buying IPO shares at the offering price and quickly reselling them in the aftermarket, takes advantage of the tendency of IPOs to jump in price. Issuers and underwriters "leave money on the table" whenever they systematically price new securities at a lower price than the market will bear. Jonathan A. Shayne and Larry D. Soderquist, Inefficiency in the Market for Initial Public Offerings, 48 Vand. L. Rev. 965, 976 (May 1995). If the relatively sophisticated customers who know about this phenomenon buy shares at the offering and the relatively unsophisticated investors buy from them in the aftermarket, there is a second transfer of wealth, over and above the transfer that takes place at the offering price. Id.
Flippers provide liquidity, but underwriters tend to view them as parasites. See Phillip Zweig and Leah Nathans Spiro, Beware the IPO Market, Business Week, April 4, 1994, at 84. Flipping forces underwriters to sell the same stock a second time, and thus undercuts their legitimate underwriting profits. Underwriters use several methods to combat flipping, including penalty bids, privilege revocation, and overallocation of the offered stock. The Commission has known about these practices for decades. It has elected not to prohibit them and has regulated them with a light touch. See generally Brief of the Commission, amicus curiae, in support of Appellees in Friedman v. Salomon Smith Barney, Inc., No. 01-7207 (2d Cir.).
The OIP alleges that Setteducati, along with HJM, Masucci, and Vanechanos, violated Section 17(a) of the Securities Act, Section 10(b) of the Exchange Act, and Rule 10b-5 thereunder (OIP ¶¶ II.AO, II.AP). It also alleges that Setteducati, Masucci, and Vanechanos willfully aided and abetted and caused HJM's violations of Section 15(c)(1) of the Exchange Act and Rules 15c1-2 and 15c1-8 thereunder (OIP ¶¶ II.AQ, II.AR). Although those paragraphs of the OIP could be read as implying that price manipulation is only "a part of [the unlawful] conduct" on which the charges are brought, the Division has stipulated that no other unlawful conduct, beyond the price manipulation alleged in OIP ¶¶ II.A through II.AN, is at issue (Tr. 1579-80).
That stipulation confirms that there are no negligence issues in this case. Price manipulation requires proof of scienter. I am aware of no Commission opinions holding that an accused manipulated the price of a security through negligent acts or omissions. Accordingly, a showing of negligence sufficient to sustain a charge under Section 17(a)(2) or Section 17(a)(3) of the Securities Act is not enough to prove price manipulation. To the extent that the Division contends otherwise, I reject its argument (Tr. 1586-90, 1654; Div. Br. at 5 n.2). The same analysis applies to Section 15(c)(1) of the Exchange Act. I am aware of no Commission opinion holding that an accused has negligently caused a price manipulation. If the Division is to prevail in its claim that Setteducati caused HJM's violation of Section 15(c)(1), it must prove that he acted with scienter.
Commission opinions considering the offense of price manipulation typically discuss certain "classic factors" or "hallmarks" of manipulation. See, e.g., Castle Secs. Corp., 53 S.E.C. 406, 410 (1998); Fertman, 51 S.E.C. at 948. In general, the Commission's opinions identify three or four such "classic factors" or "hallmarks," although they are rarely the same three or four. As for any absent "classic factors" or "hallmarks," the Commission's opinions also caution that manipulation does not depend on the presence or absence of any particular device usually associated with a manipulative scheme. See, e.g., Swartwood, Hesse, 50 S.E.C. at 1307.
This Initial Decision reviews each one of the "classic factors" or "hallmarks" of manipulation that are alleged in the OIP. It finds that one or two of them have been proven in part, while the others have not. After reviewing the record as a whole, the decision concludes that the weight of the evidence does not support the charge of price manipulation. In the alternative, assuming that a price manipulation had been proven, the decision concludes that the weight of the evidence does not show that Setteducati participated in the manipulation as a principal, willfully aided and abetted it, or caused it.
When the Commission issues an opinion exonerating an accused from a charge of manipulation, it rarely says why. See, e.g., Salloum, 52 S.E.C. at 217 n.44 (declining to hold that the respondent aided and abetted a manipulation because the evidence was "inconclusive"); Peter Martin Toczek, 51 S.E.C. 781, 788 (1993) (finding that the respondent lacked scienter and thus did not manipulate in violation of Section 10(b) of the Exchange Act); Collins Secs. Corp., 46 S.E.C. 20, 31 (1975) (reversing a hearing officer's determination that the respondent aided and abetted a manipulation).
Two aspects of the evidence have been particularly troubling. First, this is a "numbers" case, but I have little confidence in the numbers. Some of the figures presented in the OIP are simply wrong. See supra notes 3, 14, and 25 and infra note 44. In other areas, it has been impossible to pin down such basic numbers as the daily trading volume in Borealis. See supra note 18. The distinction between the "floating supply" and the "public float" has also been lost. See supra note 8 and infra note 38. Finally, certain summary exhibits present data in a fashion that obscures, rather than enlightens (DX 35, DX 36, DX 45, DX 46). See infra note 48. As an illustration, the Division treats allocations to and sales by HJM's Atlanta branch office as a proxy for allocations to and purchases by HJM's institutional customers. In fact, not all Atlanta allocations and sales involved institutional customers, and not all of HJM's institutional representatives were based in Atlanta (Tr. 87-88, 207-08, 993-94; DX 11). Second, the prosecution has missed no opportunity to portray the respondents as regulatory scofflaws. See supra note 2 and infra pp. 60-62. Such evidence often plays a role in sanctioning. It does not belong in the liability phase of the case, and has only a limited function in establishing knowledge and intent. If an otherwise weak case lacks gravitas, arguments that the accused are "bad men" cannot supply it.
2. The weight of the evidence fails to show that HJM, acting through Vanechanos, manipulated Borealis during the week of June 24 through June 28, 1996.
In the Borealis IPO, HJM sold almost 1.9 million shares of stock to its customers. During the first week of aftermarket trading, HJM bought 619,205 Borealis shares from other brokers and dealers and sold another 32,719 Borealis shares to other brokers and dealers. Thus, the firm effected a series of transactions and created actual trading in the security. All of the jurisdictional requirements of the antifraud provisions have been satisfied. HJM's activities in Borealis involved the means and instrumentalities of interstate commerce and the use of the mails. In addition, all transactions in Borealis took place on the NASDAQ SmallCap Market, which is not a registered securities exchange. See 2001 SEC Annual Report at 40-41.
a. The weight of the evidence shows that HJM dominated and controlled the market for Borealis after 10:28 a.m. on June 24, 1996.
The OIP alleges that HJM dominated and controlled trading in Borealis during the first week of aftermarket trading. Domination and control of the market for a security by a broker or dealer supports a finding that the broker or dealer manipulated the market for that security. See Resch-Cassin, 362 F. Supp. at 977; Floyd A. Allen & Co., 35 S.E.C. 176, 183 (1953). Domination and control of the market in a security does not necessarily produce a manipulation, but it enables parties to control pricing if they so choose, so as to preclude an independent, competitive market from arising. See Pagel, 48 S.E.C. at 226.
The Commission has found three factors to be probative of whether a firm dominates and controls the market in a particular stock: (1) the firm was an underwriter of the IPO of the stock and sold a substantial percentage of the offering to its own customers; (2) the firm was a market maker in the secondary market for the stock and traded a significant amount of the total aftermarket volume; and (3) there are few other market makers in the stock, who account for only a small percentage of the total trading volume as compared to the firm in question. See G.K. Scott & Co., 51 S.E.C. 961, 964 (1994), petition denied, 56 F.3d 1531 (D.C. Cir. 1995); Meyer Blinder, 50 S.E.C. 1215, 1218 n.14 (1992). Each of these factors is present here.
The fact that HJM sold approximately 80% of the Borealis IPO to its own customers, by itself, establishes neither domination nor control. After all, HJM's underwriting procedures expressly permitted it to sell IPOs at this level to its own customers. Those procedures were adopted under the supervision of a federal district court, and with the knowledge and consent of the Commission. Rather, the crucial factor is that HJM subsequently "dried up" the supply of Borealis shares by acquiring additional shares in the aftermarket from the members of the selling group. Cf. Sanders, 53 S.E.C. at 891-92; Michael Alan Leeds, 51 S.E.C. 500, 503-04 (1993).
For the first week of aftermarket trading, HJM, through Vanechanos, bought nearly nineteen Borealis shares from other brokers and dealers for every Borealis share it sold to other brokers and dealers. HJM also did most of the trading in Borealis during the immediate aftermarket. It accounted for over 77% of the trading volume on the first trading day, and from 60% to 73% of the trading volume for the rest of the week. In comparison, the broker with the next highest number of trades accounted for 12% of the volume on the first trading day and from 8% to 29% for the rest of the week.
Ritter opined that HJM dominated and controlled the market for Borealis "as soon as it went public" (DX 57 at 11). There is no evidence to support such a finding for the period before 10:28 a.m. on June 24, 1996. HJM did not lead or share the inside bid from 10:21 a.m. to 10:24 a.m. The record does not show which market makers were involved in trading, how many shares changed hands, or how many transactions took place during the crucial period before 10:28 a.m. That was the time of the Borealis price spike and an immediate retreat. The OIP attempts to finesse the issue by asserting that Vanechanos entered a bid quote of $7.50 "within minutes after trading began" (OIP ¶ II.Y). The allegation is correct, but the upward price manipulation, if indeed there was one, was over by that time. Without knowing what the other market makers were bidding before HJM entered that bid quote, no negative inference can be drawn from the fact of that bid.37 I agree with Ritter that HJM dominated and controlled the market for Borealis at all other times during the week of June 24 through June 28, 1996.
b. HJM acted to reduce the floating supply of Borealis stock, but it is impossible on this record to quantify the extent of the reduction.
The Commission has defined the floating supply of a stock as "that part of the issue which is outstanding and which is held by dealers and the public with a view to resale for a trading profit, as distinguished from that part of the stock held for investment." Gob Shops of America, Inc., 39 S.E.C. 92, 102 n.23 (1959); Barrett & Co., 9 S.E.C. 319, 327 n.8 (1941). The reduction of the floating supply of a security is one "factor which characterizes attempts by manipulators to raise the price of an over-the-counter security." Resch-Cassin, 362 F. Supp. at 977. A reduced floating supply makes the market more susceptible to manipulation. See Halsey, Stuart, 30 S.E.C. at 128; Harold T. White, 3 S.E.C. 466, 477 (1938). Evidence of a reduction in floating supply is particularly persuasive of manipulation where the amount taken off the market is substantial. See Gob Shops, 39 S.E.C. at 102-03.
In several places, the OIP alleges that Respondents reduced the "floating supply" of Borealis stock (OIP ¶¶ II.A, II.Q, II.U, II.V). In other places, the OIP charges that Respondents reduced, restricted, or controlled "the float" (OIP ¶¶ II.A, II.I, II.J, II.V). While the Commission has not defined the term "the float," the OIP apparently considers that term to be a shorthand reference to the "floating supply." For purposes of this Initial Decision, I have treated the two terms as interchangeable.
The Division's summary witness explained the formula she used to determine the percentage of "the float" controlled by HJM (Tr. 423-24; DX 35). However, she never specified the number of Borealis shares that she considered to constitute "the float." Nor did she apply the Commission's definition of the term "floating supply." A fair reading of Javorski's testimony and Ritter's report persuades me that neither witness addressed the "floating supply." Both focused instead on the "public float" (DX 57 at 11-13, 17).38
I find DX 35, a summary of the percentage of "the float" held or controlled by HJM and others during the first week of trading, to be unreliable for two reasons. First, the exhibit does not use the narrow definition of "floating supply" the Commission announced in Barrett and reiterated in Gob Shops. Under that definition, it is not enough for the Division simply to compute the total number of Borealis shares outstanding and then to subtract the restricted shares held by Borealis management and subject to lock-up agreements. The Commission's definition also requires the Division to subtract the shares of customers who were holding Borealis as an investment (customers such as Mark Faith, Anthony Bergamino, and others similarly situated). Even though none of the Borealis shares in the hands of customers were subject to a lock-up and all were immediately eligible for sale in the public market without restriction, Gob Shops, 39 S.E.C. at 101, teaches:
Securities held for investment are not part of the floating supply and sales by dealers in small lots to retail customers, which carry the likelihood that the purchasers will hold their securities for investment, tend to reduce the number of shares that would come into the market on relatively small price rises.
Second, DX 35 fails to treat all the Borealis shares owned by HJM's institutional customers in a consistent fashion. Such shares may or may not be part of the floating supply, depending upon the intent of the institutional customer owning them. However, none of the shares owned by institutional customers were controlled by HJM. DX 35 properly recognizes that Borealis shares purchased by institutional customers and held at a domestic bank must be excluded when computing the percentage of "the float" held or controlled by HJM. However, DX 35 does not take into account an even larger number of Borealis shares purchased by HJM's institutional customers and held at European nominee banks under the delivery-versus-payment system (Tr. 590, 592, 781, 1160). HJM did not hold or control those shares, either. To the extent that DX 35 assumes otherwise, it significantly overstates the percentage of Borealis shares that HJM held or controlled in the immediate aftermarket.
The record presents little reliable information about the "floating supply" of Borealis during the week of June 24 through June 28, 1996. The "public float" was "tiny" (Tr. 1172). I infer that the floating supply was even tinier. HJM controlled a good bit of the floating supply and was attempting to control more of it. More precise findings are not possible.
c. When HJM dominated and controlled the market, it had the power to control the pricing of Borealis.
The OIP also alleges that HJM exercised price leadership during most of the manipulative period. Price leadership is a factor indicating that a broker or dealer is manipulating the market for a security. Resch-Cassin, 362 F. Supp. at 976. Paying increasingly higher prices for a security, and bidding up a stock or causing it to be bid up, are well-established manipulative techniques. Barrett, 9 S.E.C. at 329; Charles C. Wright, 3 S.E.C. 190, 199 (1938), rev'd and remanded on other grounds, 112 F.2d 89 (2d Cir. 1940). Bidding up prices is particularly effective when carried out in conjunction with a reduction in the floating supply of a security. R.L. Emacio & Co., 35 S.E.C. 191, 199 (1953).
Price leadership can be shown by the insertion of higher bids or by purchases at increasingly higher prices. Halsey, Stuart, 30 S.E.C. at 111-12, 121. The insertion of increasingly higher bids is a device that is used to create a false appearance of activity in the over-the-counter market and to support the prices of securities at inflated levels. Gob Shops, 39 S.E.C. at 101. The Commission has found that purchases at increasing prices stimulate activity at those levels by others in the market. See Halsey, Stuart, 30 S.E.C. at 121, 128.
The Division sought to show price leadership by emphasizing the amount of time that HJM led or shared the inside bid for Borealis during the first week of aftermarket trading (OIP ¶ II.X; DX 38, DX 57 at 7, 14). HJM led or shared the inside bid price 99%, 100%, 97%, 93%, and 81% of the time, respectively, on the first five trading days. The high amount of time HJM spent buying shares and the small amount of time HJM spent selling shares is consistent with HJM's allowing other brokers and dealers to reduce their inventories of Borealis stock, while HJM accumulated Borealis shares.
This is not a price leadership case in the traditional sense. For example, HJM did not lead or share the inside bid from 10:21 a.m. to 10:24 a.m. on June 24, 1996, when the price of Borealis spiked to $9.00 per share and immediately retreated (DX 38, RX 12). The critical period for the alleged manipulation of Borealis was before 10:28 a.m. on June 24, and the absence of pricing information for that interval is striking. Ritter's graph shows that the inside bid for Borealis fell five times before 10:28 a.m. (DX 57 at 7). If the inside bid was falling between 10:21 a.m. and 10:24 a.m. when HJM did not lead or share the inside bid, that is not evidence of an upward price manipulation. If the inside bid continued to fall between 10:25 a.m. and 10:28 a.m., when HJM did lead or share the inside bid, that is evidence that HJM was leading the market down, not up. There is no evidence as to which market makers were responsible for any inside bid upticks from June 26 through June 28, or even if there were any inside bid upticks on those days. Nor is there evidence that HJM raised its inside bid at any time after 10:28 a.m. on June 24. Ritter demonstrates only that HJM was responsible for a stable inside bid after 10:28 a.m. on the first day of aftermarket trading (DX 57 at 5, 7). Based on the daily high, low, and closing trades for Borealis during the remainder of the week, it is probable that HJM lowered its inside bid after the first two trading days (RX 12).
These findings are not dispositive of the issue. The Commission has held that, if an underwriter dominates and controls the supply of a stock, it can control pricing to such an extent as to preclude an independent, competitive market from arising, notwithstanding the presence of other market makers who may be entering bid quotations. See Pagel, 48 S.E.C. at 226. If an integrated dealer dominates and controls the market for a security, it is able to set non-competitive quotations and spreads, as other market makers have no significant competitive influence in pricing the security. Under Commission precedent, the fact that other market makers adopt the dominating and controlling firm's quotations and spreads does not make them independent. Nor is it indicative of the current market. See Goldmen, 75 SEC Docket at 57. Ritter emphasized the same point (DX 57 at 13).
In essence, Goldmen and Pagel establish a presumption that, if a firm dominates and controls the market for a security, it is empowered to determine the price of the security. As to the inside bid, the presumption in this case is unrebutted. Even though the evidence in this case does not show that HJM was engaged in price leadership in the traditional sense (i.e., systematically escalating the inside bid for the entire week, or even for the first few minutes of aftermarket trading), it does demonstrate price maintenance (i.e., shoring up the floor) after 10:28 a.m. on June 24, 1996.
d. The weight of the evidence does not show that HJM, acting through Vanechanos, abused its power to control the pricing of Borealis.
The price leadership in a security resulting from almost exclusive control over the source of supply empowers an underwriter to set prices arbitrarily. Michael J. Markowski, 73 SEC Docket 625, 629 & n.10 (Sept. 7, 2000), aff'd, 274 F.3d 525 (D.C. Cir. 2001). "If that power is abused, the result is a manipulation." Id., quoting Pagel, 48 S.E.C. at 226.
The Division demonstrated that virtually all the aftermarket demand for Borealis came from the retail and institutional customers of HJM. In addition, it presented some evidence that HJM rewarded its registered representatives with higher commissions for Borealis transactions than it did on other transactions. But see supra note 21. The Division seems to believe that this is enough to establish the manipulation: if there is deceptive market activity, there is no need to prove deceptive value, even though the OIP raises the latter allegation. As I understand the Division's argument, HJM violated the antifraud provisions because it misrepresented to the investing public that a market existed for Borealis, when there was only a market made, created, and controlled by HJM itself.
There are two problems with the proof in support of that argument. First, the Division has ignored the central role of Vanechanos, HJM's head trader, in the alleged manipulative scheme. Without proof of trading misconduct by Vanechanos, the Division has abandoned the OIP's manipulation theme in favor of a misrepresentation theory. The Division offers little more than the sales practice case it has insisted it did not bring. In effect, accepting the Division's argument would turn this case into one like Richard D. DeMaio, 54 SEC Docket 1943 (Initial Decision) (Aug. 4, 1993), final, 55 SEC Docket 683 (Oct. 15, 1993), an antifraud case involving sales practice violations in a branch office, which mentions manipulation very sparingly.39 Second, the Division's evidence of trading misconduct relies entirely upon HJM's domination and control of the market for Borealis. As the Commission held in Pagel, 48 S.E.C. at 226, however, domination and control, by itself, does not necessarily produce a manipulation. In essence, the Division's argument takes three of the "classic factors" of prior manipulation cases-domination and control, ability to control pricing, and abuse of the power to control pricing-and attempts to collapse them into a single factor, domination and control.
The most surprising aspect of the record is how little evidence pertains to Vanechanos, and what he actually did during the week of June 24 through June 28, 1996. The OIP alleges that Vanechanos, and no one else, engineered the price manipulation (OIP ¶ II.Y; Tr. 29). The Division argues that HJM's sales force, acting consistent with instructions given by Setteducati and Masucci, created a demand into which Vanechanos could sell Borealis stock at inflated prices (Tr. 1581). The theory hangs together only if Vanechanos can be shown to have done something improper to facilitate the manipulation and to inflate the price. Although the existence of manipulation may be gathered from circumstantial evidence, simply piling inference upon inference is not enough. See Stanojev, 643 F.2d at 923-24 n.7.
Vanechanos was able to follow the market and enter and update quotes for Borealis on his NASDAQ Level 3 Terminal (Tr. 1020-24). He also received daily position updates for Borealis from HJM's clearing broker on the next business day (DX 17-DX 21). There is no evidence that he abused these tools or the information they provided him. Nor can I find such abuse from the fact that HJM established a net short position in Borealis (Tr. 1062-63). The Division argues that Vanechanos established this position so that HJM could buy back the stock without having a downward impact on the price of Borealis. There is nothing per se improper about that. HJM's net short position did not exceed its overallotment option, and no inference of misconduct is warranted.
The OIP's theory that Vanechanos raised the price of Borealis from the offering price of $5.00 to $8.49 per share fails because the Division elected not to explain what happened in the market during the first few minutes of trading on June 24, i.e., before 10:28 a.m. The record shows nothing about the parties and counterparties to transactions during that interval, the volume of shares changing hands in such transactions, and whether retail customers, institutional customers, or interdealer trades were involved. As for quotes, the record shows nothing about the identity of the market makers responsible for upticks and downticks of the inside bid and inside ask prices before 10:28 a.m. It requires a leap of faith to conclude that any overpricing between 10:20 a.m. and 10:28 a.m. was the result of scienter-based fraud, rather than honest misjudgments in the price-discovery process.
The OIP's alternate theory also fails. The Division contends that Vanechanos manipulated Borealis by purchasing from the selling group firms and from the favored HJM accounts at the inside bid and then immediately reselling those shares to HJM's non-favored retail customers at the inside ask (Div. Prop. Find. # 132). It presented no testimony and no exhibits to support that theory.40
Large and sticky spreads do not establish the manipulative trading alleged in the OIP, either. Ritter demonstrated that there was a wide and sticky spread between the inside bid and the inside ask for most of the first trading day (DX 57 at 9-10). He offered no analysis of the spread width and stickiness on June 25 through June 28. I am willing to infer that the wide spread and stickiness continued on June 25, because the range of trades on that date ($7.50 to $8.125) was identical to the inside bid and inside ask quotes at the close on June 24. I am not willing to make any such inferences about large spreads and stickiness for the rest of the week.41 At most, therefore, these phenomena could be indicative of a manipulation on the first two trading days. Domination and control of the market make it possible to dictate substantial spreads between purchases and sales. If two firms act together to control both the inside bid and the inside ask, misconduct may be established. Cf. In re NASDAQ Market-Makers Antitrust Litig., 169 F.R.D. 493, 502-03 (S.D.N.Y. 1996); Sterling Secs. Co., 39 S.E.C. 487, 491 (1959). In this case, unlike in NASDAQ or Sterling, there is no suggestion that HJM colluded with the market makers that established the inside ask price for Borealis. Absent such collusion, it is not possible to hold HJM alone responsible for the large spread on the first two trading days. Any of the other market makers could have reduced the inside ask and thus "broken the spread" at any time, regardless of what HJM was doing at the inside bid. Vanechanos did not manipulate Borealis simply because he bought at an inside ask price that other market makers established.
Since this record does not show that Vanechanos engaged in any acts or omissions that constituted a manipulation, there is no need to consider his state of mind at the time he was alleged to have performed such acts or omissions. Cf. U.S. Environmental, 155 F.3d at 111.
3. The weight of the evidence fails to show two other "classic factors" or "hallmarks" of manipulation that were alleged in the OIP.
a. The weight of the evidence fails to show that Borealis traded at artificial prices during the week of June 24 through June 28, 1996.
"Manipulation is the deceptive movement of a security's price, accomplished by an intentional interference with the forces of supply and demand." Patten Secs. Corp., 51 S.E.C. 568, 572 (1993) (footnotes omitted, emphasis added).
The OIP alleges that HJM sold Borealis at "excessive" and "inflated" prices (OIP ¶¶ II.L, II.Y). It also asserts that the price of Borealis stock "remained high" throughout the manipulative period (OIP ¶ II.Q). The issue for decision is: excessive, inflated, and high, relative to what? The Division has no theory as to the maximum non-manipulated price of Borealis (Tr. 14).
OIP ¶ II.Y does little to clarify the charge. It alleges that HJM manipulated the price of Borealis stock from the IPO price of $5.00 per share to approximately $8.49 per share. It also alleges that HJM bought Borealis IPO shares from its retail customers at prices ranging from $7.19 to $7.25 and then resold those shares to other retail customers at prices up to $8.49 per share. This language may be read either of two ways: first, it could be understood as alleging that price artificiality should be measured as the difference between $5.00 and $8.49 per share; or second, it might be understood as alleging that price artificiality should be measured as the difference between $7.19 and $7.25 per share, on the one hand, and something up to $8.49 per share, on the other.
The Commission has traditionally given its attention to a surge in a security's price if there are no known prospects for the issuer or favorable developments affecting it. See Fertman, 51 S.E.C. at 948; Patten, 51 S.E.C. at 573 (1993). Ritter opined that Borealis's first day percentage return was significantly higher than the average return for an IPO during 1996 (DX 57 at 10). He found that performance difficult to explain, given Borealis's lack of a product and lack of current revenue (DX 57 at 10). I do not reach the question of whether Ritter was right or wrong about Borealis's prospects. As previously shown, however, Ritter has acknowledged that many customers buy IPOs for reasons having little or nothing to do with an issuer's fundamentals. For that reason, a single-minded focus on Borealis's fundamentals is invalid because it ignores legitimate forces that were at work in shaping the demand for Borealis. Rather, I examine Ritter's and the Division's premise that the offering price of a new issue is a valid starting point for measuring a price surge and for comparing one IPO to another. The Commission's opinions provide no guidance on that point.
As a predicate to measuring price artificiality, it is essential to have a reference point known to be a reliable and accurate source of price information. The market does not set IPO prices; there is no public market prior to the offering. In the case of IPOs of small capitalization companies, the offering price cannot be assumed to be a legitimate base line price for measuring artificiality in the immediate aftermarket. The Borealis prospectus explained that the $5.00 per share offering price had been determined by negotiations between Borealis and HJM, and was not necessarily related to Borealis's asset value or any other established criterion of value (DX 1 at 1). Several articles in the economic and financial literature, many of them co-authored by Ritter, document the fact that IPOs are, on average, significantly underpriced.42 The courts have also recognized the phenomenon that IPOs are underpriced in the short run. See, e.g., Friedman v. Salomon/Smith Barney, Inc., 2000 WL 1804719 at *1 (S.D.N.Y. Dec. 8, 2000) ("Typically, the offering price is set below the estimated market value of the stock to make the stock more attractive."), appeal docketed, No. 01-7207 (2d Cir.). The Commission has long recognized that "the correct pricing of new issues is not, and can never become, an exact science." Statement on the Regulation of "Pegging, Fixing and Stabilizing" of Security Prices, at 10, Exch. Act. Rel. No. 34-2446 (Mar. 18, 1940), 11 Fed. Reg. 10971 (Sept. 27, 1946). Recent adjudicatory decisions show just how inexact.43
The Division has presented no evidence to establish that the $5.00 per share offering price accurately represented the "true" value of Borealis on June 26, 1996. I therefore conclude that the Borealis offering price cannot serve as a valid reference point to measure price artificiality in the immediate aftermarket. The size of Borealis's price surge on the first day of trading means little by itself. Instead of serving as a classic indication of a manipulation, it could mean that Borealis should have been offered to the public at $6.00 or $6.50 per share, instead of $5.00. The economic and financial literature has no shortage of articles concluding that a big first day price run-up is a classic indication of an underpriced IPO.
As a fallback position, OIP ¶ II.Y also asserts that price artificiality may be measured by the difference between the price at which HJM bought Borealis shares from the flipping customers of its favored employees, on the one hand, and the price at which HJM immediately resold those same shares to its non-favored customers.44 The price at which HJM bought Borealis shares from the flipping customers of its favored employees certainly provides one possible reference point for measuring the prevailing market price. However, it is not the base point the Commission prefers for making such calculations. See Sanders, 53 S.E.C. at 894-95 (holding that prices paid in wholesale trades are generally a more reliable indication of the prevailing market price than retail trades because they occur between informed market professionals). Only if there are concerns about the legitimacy of interdealer trade prices has the Commission turned to purchases from retail customers. See Salloum, 52 S.E.C. at 212; Robert A. Amato, 51 S.E.C. 316, 317-18 (1993), aff'd, 18 F.3d 1281 (5th Cir. 1994).
In this case, the number of shares involved in interdealer trades (where HJM bought from other brokers) was more than double the number of shares involved in retail trades (where HJM bought from its own customers) (DX 40). Nothing in the record demonstrates that these interdealer trades were transacted at artificial prices. I conclude that the base line for measuring price artificiality should have been the prevailing market price, as measured by HJM's purchases from other market makers during the week of June 24 through June 28, 1996.
The evidence of price artificiality is inconclusive, but that is not necessarily fatal to the prosecution because the statute demands less of the Division than does the OIP. Section 10(b) of the Exchange Act, unlike Section 9(a), does not require the Division to prove that HJM's manipulation had an effect on the price of Borealis stock. Cf. GFL Advantage Fund, Ltd. v. Colkitt, 272 F.3d 189, 205-06 (3d Cir. 2001), cert. denied, 122 S. Ct. 2588 (2002). "The absence of allegations of market dominance and price movement are not fatal to" a claim of market manipulation, for although "these may be classic attributes of market manipulation, [ ] they are not requisites." In re Blech Securities, Litig., 928 F. Supp. 1279, 1298 (S.D.N.Y. 1996).
b. The weight of the evidence fails to show that the cessation of the alleged manipulation caused a short-term price collapse. Although the price of Borealis declined gradually over the long term, that fact is not probative of the cessation of a manipulation.
OIP ¶¶ II.AB and II.AC allege that the price of Borealis declined after the manipulation ended. The charge is designed to demonstrate that, just as the upward price manipulation had a dramatic effect on the price of Borealis-raising it to an "excessive" and "inflated" level-so too, the cessation of the manipulation triggered a corresponding price reduction.
The collapse of the market price for a security when a manipulator ceases his activity is one of the factors that typically characterize attempts to manipulate upward a security's price. See Resch-Cassin, 362 F. Supp. at 976-77 (finding that, once the manipulators stopped bidding, the price of the security dropped rapidly within a few weeks from $17.50 to $4.00 per share); Trane Co. v. O'Connor Secs., 561 F. Supp. 301, 305 (S.D.N.Y. 1983) (denying a preliminary injunction and dismissing manipulation charges where there was no evidence of a price collapse after the defendants' trading ended), appeal dismissed as moot, 718 F.2d 26 (2d Cir. 1983); Randolph K. Pace, 51 S.E.C. 361, 364-66 & n.26 (1993) (citing Resch-Cassin with favor and concluding that, once the manipulator withdrew its bid, the price of the security fell from $9.00 to $4.50 per share).
The Commission has held that a price collapse is not an indispensable element of an upward price manipulation. See Swartwood, Hesse, 50 S.E.C. at 1307 (stating that the fact that "after the stock reached the $7.00 range it kept climbing [is] not necessarily inconsistent with a manipulation"). Nonetheless, the Commission's recent manipulation opinions have taken care to explain the absence of evidence of a price collapse. See, e.g., Patten, 51 S.E.C. at 575 (concluding that the market price of a security remained relatively high because the manipulator retained a large inventory of the security and because of subsequent favorable developments in the market); Pagel, 48 S.E.C. 223, 227 (1985) (concluding that the stock price remained fairly constant for more than ten months after a manipulation because the manipulator continued to dominate the market). The message of Patten and Pagel is simple: if the OIP alleges a post-manipulation price collapse, and if the evidence does not demonstrate that one occurred, the Division is obliged to provide a plausible explanation for its absence.
In this case, the Division insists that Borealis suffered both short-term and long-term price collapses. The difficulty with the Division's theory is that it confuses correlation with causation and utterly ignores the possibility of intervening events. In essence, the Division engages in the practice of making observations about coincidental and sequential occurrences and, then, from those observations, inviting the fact finder to proceed directly to conclusions about causation. This form of post hoc, ergo propter hoc reasoning (literally, "after this, therefore because of this") is inherently fallacious.
The Purported Thirty-Day Price Drop
Ritter tracked the market price of Borealis for a significant period of time after June 28, 1996, but he found no evidence of an immediate or significant price collapse (Tr. 1127-28). His inability to find such a price collapse is readily confirmed by late-filed RX 12, a summary of Borealis's trading volume, high, low, and closing prices for the period June 24, 1996, through December 31, 1996. Nonetheless, the Division is not willing to give up so easily (Div. Prehearing Br. at 18, 26-27; Div. Prop. Find. ## 83, 134; Div. Br. at 37).
The Division hypothesizes as follows: (1) the penalty bid, which pressured HJM's registered representatives to discourage sales of Borealis by their customers, expired on July 23, 1996; (2) on July 24, 1996, the first day that the penalty bid was not in effect, the price of Borealis fell below the IPO price of $5.00 for the first time, closing at $4.875; (3) the trading volume of Borealis on July 24, 1996 was much higher than the previous day's trading volume and significantly higher than the daily average trading volume over the preceding ten business days; and (4) therefore, the lifting of the penalty bid not only coincided with heavy sales and falling prices for Borealis, but caused it.
The Division's hypothesis fails to acknowledge that the entire market for small capitalization stocks dropped precipitously on July 24, 1996. This phenomenon was widely recognized in the financial press the next day.45 It is also confirmed by a review of the Russell 2000 Index for June and July 1996, which shows that the market hit a bottom on July 24. Ritter used the Russell 2000 Index to measure the decline in Borealis's price against the performance of the small capitalization stock market between June 24, 1996, and October 1998 (DX 57 at 19). For some reason, however, the Division omitted any scrutiny of the Russell 2000 Index for the first thirty days of the Borealis aftermarket. Official notice of the Russell 2000 Index for the relevant dates shows as follows:
|June 24, 1996||347.19||345.16||346.82|
|July 23, 1996||318.30||311.61||311.72|
|July 24, 1996||311.72||305.12||307.77|
The Division never tested its short-term sudden price drop hypothesis by performing a regression analysis on the relevant data.46 The case law requires it to do so. See Sanner v. Board of Trade, 2001 WL 1155277 (N.D. Ill. Sept. 28, 2001) (analyzing the work of an expert witness who performed a series of statistical regression analyses that focused on different variables that were relevant to artificiality of prices in the soybean futures market); Northern Telecom, Ltd., Secs. Litig., 116 F. Supp. 2d 446, 460 (S.D.N.Y. 2000) (finding an expert witness's testimony "fatally deficient" because he did not perform an event study or similar analysis to remove the effects on stock price of market and industry information); cf. SEC v. Sayegh, 906 F. Supp. 939, 944 (S.D.N.Y. 1995) (measuring the price performance of a manipulated security against the performance of the stock market as a whole).
There are too many untested variables to permit any valid conclusions about the purported thirty-day price drop. The Division has not shown that the July 24, 1996, volume and price change figures that it asks me to embrace are statistically significant. It is not even an adjudicated fact that the Borealis penalty bid lasted precisely thirty days. In these circumstances, I give little credence to the Division's hypothesis that lifting the penalty bid on July 23, 1996, caused the price of Borealis to fall on heavy trading volume on July 24, 1996.
Borealis's Long-Term Gradual Price
Decline Is Not Probative Of A Manipulation
Ritter opined that a gradual price decline is just as probative of an upward price manipulation as is a sudden price collapse (DX 57 at 18-19). I do not give much weight to this testimony. When Ritter was pressed to identify other manipulations that did not involve sudden price collapses, he could only point to the four securities that were the subject of the Commission's 1990 injunctive action against TJA (Tr. 1176-77, 1185-86). See Thomas James Associates, 738 F. Supp. at 93 (mitigating a sanction where the four securities continued to trade at or above their IPO prices one to two years later). Because the 1990 litigation was settled without findings of fact and without admitting or denying liability, the record simply does not permit meaningful factual comparisons. Moreover, Ritter conceded that the outcome of the 1990 TJA litigation was not an important factor in reaching his conclusions about the manipulation of Borealis (Tr. 1164).
Expert evidence must be the product of reliable principles and methods. See Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993); Kumho Tire Co. v. Charmichael, 526 U.S. 137 (1999). Daubert provides that a trial judge must perform a gatekeeping function to determine the admissibility of expert testimony. Daubert, 509 U.S. at 589. While not setting out a definitive checklist, Daubert identified four criteria for evaluating the admissibility and reliability of expert testimony: (1) whether the methods upon which the testimony is based are testable; (2) the known or potential rate of error associated with the test; (3) whether the method has been subject to peer review; and (4) whether the method is generally accepted in the relevant scientific community. Daubert, 509 U.S. at 593-94. Although Daubert was limited to scientific testimony, Kumho Tire extended Daubert's reach to include virtually all expert testimony. Kumho Tire, 526 U.S. at 149.
The Commission has not yet considered the applicability of Daubert and Kumho Tire, which are based on Rule 702 of the Federal Rules of Evidence, to adjudicatory proceedings conducted under its Rules of Practice. But see Elliott v. CFTC, 202 F.3d 926, 934 (7th Cir. 2000) ("Daubert and Kumho Tire were decided in the context of admissibility, but the principle for which they stand-that all expert testimony must be reliable-should apply with equal force to the weight a[n agency] factfinder accords expert testimony."); Peabody Coal Co. v. McCandless, 255 F.3d 465, 469 (7th Cir. 2001) (citations omitted):
Daubert does not apply directly in black lung cases, because it is based on Fed. R. Evid. 702, which agencies need not follow. Agencies relax the rules of evidence because they believe that they have the skill needed to handle evidence that might mislead a jury. They have a corresponding obligation to use that skill when evaluating technical evidence. Neither the ALJ nor the [Benefits Review Board of the Department of Labor] did this, however. . . . An agency must act like an expert if it expects the judiciary to treat it as one.
I have admitted Ritter's testimony about the long-term gradual price decline into the record, but that is not the end of the matter. Elliott and Peabody Coal make it clear that Daubert and Kumho Tire are as much about the reliability of expert and technical testimony as they are about its admissibility. Cf. Charles D. Tom, 50 S.E.C. 1142, 1145 (1992) (holding that, in determining when to admit and whether to rely on hearsay evidence, the Commission evaluates its probative value, its reliability, and the fairness of its use). See generally WSF Corp., 77 SEC Docket 1831, 1834-35 (Initial Decision) (May 8, 2002) (applying Daubert and Kumho Tire at the urging of the Division and giving minimal weight to the opinions of a respondent's expert), final, 77 SEC Docket 2336 (May 24, 2002).
A review of Borealis's periodic filings with the Commission demonstrates several developments during 1997 and 1998 that virtually begged for close scrutiny by the Division's expert.47 For example, Curtis Faith resigned as president and chief executive officer of Borealis in January 1997. Arsenal was not sold to the public until April 1997, more than six months after Borealis expected it to be ready for distribution. Borealis continued to raise funds by issuing new stock. There were 3.1 million shares outstanding in July 1996, but over 8.7 million shares outstanding in September 1998. The Commission's investigation of HJM also had an adverse impact on Borealis. In July 1997, in connection with a registration statement for a proposed offering of another 1.4 million shares of Borealis common stock underwritten by HJM, Borealis advised the public about the Commission's then-private investigation of HJM. Borealis stated that any limitation on HJM's ability to make a market in its stock as a result of that investigation could adversely impact the liquidity or trading price of its stock. Finally, in March 1998 and again in August 1998, Borealis advised its shareholders that it might be unable to maintain the standards for continued quotation on the NASDAQ SmallCap Market. It stated that a likely consequence of delisting would be that its stock would become subject to the penny stock disclosure rules. Borealis warned that those rules, in turn, could have an adverse impact on the willingness of brokers to sell or make a market in its stock.
Neither Ritter nor the Division attempted to sort out the many factors that contributed to Borealis's long-term price decline. It was their obligation to do so. The weight of the evidence fails to show that Borealis's long-term price decline was a consequence of the events of June 24 through June 28, 1996, as distinguished from the resignation of the firm's founder, subsequent product disappointments, significant dilution of stock value, liquidity losses from the demise of its principal market maker, liquidity losses from delisting, and the adverse impact of the penny stock disclosure rules. Consistent with Daubert, Kumho Tire, Elliott, Peabody Coal, and WSF, I give this aspect of the Division's proof very little weight.
4. Assuming that HJM, acting through Vanechanos, manipulated Borealis as alleged in the OIP, the weight of the evidence does not show that Setteducati participated in the manipulation as a principal, willfully aided and abetted it, or caused it.
A primary violator is one who, acting with scienter, "participated in [a] fraudulent scheme" or other activity proscribed by the securities laws. First Jersey, 101 F.3d at 1471-72 (finding a firm's president primarily liability because of his hands-on involvement in the pertinent decisions); see U.S. Environmental, 155 F.3d at 111-12 (holding the trader who executed manipulative buy and sell orders primarily liable under Section 10(b) of the Exchange Act). Primary liability may be imposed not only on the mastermind of a fraudulent scheme, "but also on those who had knowledge of the fraud and assisted in its perpetration." First Jersey, 101 F.3d at 1471. No express agreement is necessary to establish the existence of a scheme to defraud; it is enough that each of the individual respondents joined or participated in a common undertaking that he knew or should have known was fraudulent. See Haight & Co., 44 S.E.C. 481, 497-98 (1971).
The Division argues that Setteducati played an important role in the Borealis manipulation by orchestrating the allocation of IPO shares, by selectively imposing HJM's penalty bid policy, and by providing direction to HJM's sales force, both directly and through branch managers. It also alleges that his intent to participate in the Borealis manipulation may be inferred from his knowledge of TJA's previous participation in similar schemes. Each allegation will be considered separately.
a. The weight of the evidence fails to show that Setteducati orchestrated the allocation of Borealis IPO shares to reduce "the float" and/or to create a source of aftermarket shares that HJM could resell to non-favored customers.
The Division asks me to infer that Masucci and Setteducati allocated a large number of IPO shares to HJM's San Francisco branch office to facilitate the manipulation that followed. In support, it showed that the San Francisco office had a reputation within HJM for "supporting the deal." It also showed that the San Francisco branch accounted for 15.9% of HJM's gross revenues over a six month period (January through June 1996) and was allocated 33.9% of the Borealis IPO shares (Supplemental Declaration of Jean M. Javorksi, dated July 24, 2002, at ¶ 4) (Javorski Supplemental Declaration). The inference is strong enough to shift to Setteducati the burden of production, that is, the burden of presenting some rebuttal evidence. Setteducati has done so. He established that there was genuine enthusiasm for Borealis within the San Francisco branch office because of Grady, because of the "high tech" nature of the offering, and because Borealis's Lake Tahoe office was close to San Francisco. He also proved that Borealis's preliminary prospectus was available for distribution no later than June 3, that the San Francisco branch office held its due diligence meeting on June 6, and that Masucci spoke with the San Francisco branch office manager about Borealis before he recommended to Setteducati that they allocate 800,000 shares to the San Francisco branch. The Division remains conspicuously silent about this rebuttal evidence.
Of course, the existence of a legitimate purpose for the allocation does not necessarily negate the possibility of a simultaneous illegitimate purpose. Cf. Lipson, 278 F.3d at 660-62. Masucci and Setteducati could have allocated a large number of IPO shares to the San Francisco branch for two reasons. If so, the question is whether the legitimate purpose sanitizes the illegitimate purpose. I therefore consider the other evidence concerning the June 6 preliminary allocation.
The Division next asks me to infer that Masucci and Setteducati allocated a large number of shares to HJM's Chicago office to facilitate the manipulation. In support, it showed that the Chicago office also had a reputation within HJM for "supporting the deal." However, the Division has not shown that the Chicago office received an allocation of Borealis shares that was disproportionate to its revenue production.48 Chicago accounted for 11.11% of HJM's gross revenues over a six month period (January through June 1996) but it was allocated only 7.41% of the Borealis IPO shares (Javorski Supplemental Declaration at ¶ 5). The allocation of IPO shares to Chicago adds nothing to the Division's requested inference, and in fact, seriously detracts from it.
If that were not enough to sink the Division's theory about orchestrating the allocation to reduce "the float," then the allocation of shares to HJM's institutional customers surely is. Setteducati allocated 400,000 Borealis IPO shares to Bader for distribution to HJM's institutional customers. The allocation to Bader was for more than twice as many shares as were allocated to the Chicago branch office for distribution to retail customers. In aftermarket trading, HJM's institutional customers bought more than triple the number of Borealis shares as the retail customers of HJM's Chicago office (395,000 vs. 111,064 shares) (DX 46; Javorski Declaration at ¶ 5). Because these institutional customers operated under the delivery-versus-payment system and could use their European banking relationships to flip, HJM may well not have known if they were likely to flip (Tr. 425-26, 584, 590, 592-93, 781). The record does not demonstrate that any of HJM's institutional customers were victimized by high pressure sales tactics.
If a group of manipulators intended as a part of their scheme to stop or reduce flipping, then allocating IPO shares in large numbers to institutional customers would be exactly the wrong way to accomplish that goal. The weight of the evidence does not show that Masucci and Setteducati orchestrated the allocation of a disproportionate number of IPO shares to HJM's San Francisco and Chicago branches "for the purpose of" reducing "the float."
Nor does the record support the OIP's second claim: that Setteducati orchestrated the allocation of IPO shares to create a source of aftermarket shares to resell to non-favored customers. As a factual matter, it is simply not true that the customers of HJM's favored employees flipped "virtually all" of their IPO shares. The quantity of Borealis IPO shares allocated to favored employees whose retail customers flipped was outnumbered by the quantity of IPO shares allocated to favored employees whose retail customers did not flip.
The industry's practice with respect to allocations of IPO shares has been common knowledge for several years. See Zweig and Spiro, Beware the IPO Market at 84:
At the retail level, making allocations often amounts to gutter fighting. But predictably, some retail investors and their [registered representatives] come out a lot better than others. In the catbird seat sits the retail coordinator, who decides with the syndicate department how much stock of each deal gets allocated to retail branches, and often within those branches to which [registered representatives]. There is rarely enough to go around. . . .
How much stock an office or [registered representative] gets depends in part on their syndicate rankings, which is a list of [registered representatives] and offices in order of how many new issues, good and bad, each [registered representative] and office has sold.
This was precisely the point that Setteducati raised in his answer to the OIP. Pursuant to a well-established industry custom and practice, certain top producing registered representatives and producing branch managers were allocated shares for sale to their customers, based on their historically demonstrated ability to submit indications of interest that would likely ripen into actual purchases of the offering. Masucci and Setteducati allocated a block of Borealis IPO shares to the Chicago branch office, but the branch manager decided how to subdivide the allocation among the individual representatives in his own office (Tr. 525-27, 548-50).
At the relevant time, neither the Commission nor the NASD had published rules to confine an underwriter's discretion to allocate IPO shares. The Commission has invoked the antifraud provisions of the securities laws to regulate the allocation of IPO shares, but only in the most egregious cases. See SEC v. Credit Suisse First Boston Corp., No. 02-CV-00090 (RWR) D.D.C., Jan. 22, 2002 (consent decree) (enjoining an underwriter from allocating IPO shares to customers in return for extracting excessive brokerage commissions from the same customers in unrelated securities trades); C. James Padgett, 52 S.E.C. 1257, 1273-75 (1997) (sanctioning a respondent for allocating IPO shares to customers only if the customers agreed to resell their IPO shares to the firm at the firm's behest). The Commission has not yet intervened to regulate the process by which an underwriter allocates IPO shares among its own branch offices and its own registered representatives, and the Division has not made a persuasive case for doing so here.
b. Assuming that a manipulation occurred, the weight of the evidence fails to show that Setteducati selectively imposed the penalty bid to facilitate the manipulation.
The Division next argues that Setteducati allowed the selling group brokerage firms to flip their customers' shares with impunity; permitted favored in-house representatives to flip with impunity; and enforced the penalty bid against in-house rank-and-file representatives for thirty days. It also contends that these practices deviated from the industry norm in 1996.
The evidence shows none of this. There is no proof as to what HJM's penalty bid policy may have been as to the selling group firms. If HJM had an external penalty bid policy, there is no proof that Setteducati was responsible for enforcing it or waiving it. Within HJM, Masucci, not Setteducati, enforced the penalty bid policy. To be sure, the favored representatives often benefited from Masucci's lax enforcement. However, Masucci did not uniformly waive the penalty bid policy to benefit the favored representatives. Setteducati, Cruz, Hice, Musielski, and Keohane, all of whom the Division considers favored representatives, were penalized to some degree when their customers flipped Borealis shares in the immediate aftermarket. Finally, Ritter said that HJM's penalty bid policy was not important to reaching his opinion about Borealis (Tr. 1136).
The weight of the evidence does not show that the penalty bid period for HJM offerings typically lasted thirty days, as opposed to forty-five or sixty days. In the case of Borealis, the Division's commission run exhibit, DX 34, lumps together the penalties imposed on HJM's registered representatives between June 26 and July 23, 1996 (DX 34 at 11944, 11946, 11948, 11951). It is impossible to tell from that exhibit if HJM imposed all penalties on June 26, all on July 23, or if it imposed them consistently throughout that period. The omission is significant. Had all such penalties all been imposed on June 26, the Division could not plausibly claim that HJM deviated from the industry-wide custom of waiving penalties after a few days on new issues that were trading above their offering prices. Nor could the Division credibly advance its hypothesis that lifting the penalty bid on July 23 caused the price of Borealis to plunge on July 24.
In any event, evidence of failure to conform to an industry standard is relevant, but not determinative of any issue in this proceeding. Just as conformity to an industry standard is not enough for an underwriter to escape liability, see SEC v. GLT Dain Rauscher, 254 F.3d 852 (9th Cir. 2001), so too, deviation from an industry standard is not dispositive on liability issues, either.
HJM's underwriting manual emphasized that customers were never to be prevented from selling IPO shares in the aftermarket (DX 26 at 5237, 5240). While the record shows that violations of that policy occurred, the policy worked as it was supposed to work in other instances. John Berquist bought 500 IPO shares and 850 aftermarket shares and flipped them all on June 27, 1996 (Tr. 1309-10). David Mosher bought 200 IPO shares and 300 aftermarket shares and flipped at least 300 shares on July 2, 1996 (Tr. 1405, 1408, 1421). Three of the registered representatives who testified for the Division stated that they would never refuse a customer's sell order, even if it meant the loss of commission income (Tr. 266, 288, 716, 846, 867-68, 916; DX 34 at 11972-73).
The Commission is well aware that a primary objective of a penalty bid is to encourage sales of securities to those persons who intend to hold them rather than to engage in short-term profit taking, i.e., to combat flipping. See Trading Practice Rules Concerning Securities Offerings, 61 SEC Docket 2021, 2037-38 (Apr. 18, 1996) (Proposed Rules). As an illustration, an HJM representative sought customer Anthony Bergamino's promise that he would not flip his IPO shares on the first trading day, and Bergamino gave it (Tr. 1284). No one at HJM ever refused Bergamino's sell order (Tr. 1290-91). There was no misconduct in using such a procedure.
Penalty bid policies undoubtedly place registered representatives in conflict with their customers. Such policies also arguably put retail customers who wish to flip IPO shares at a disadvantage to institutional customers who want to flip IPO shares. The Commission has known about anti-flipping devices, such as penalty bid and privilege revocation, for a long time. As of June 1996, it had not elected to regulate the practices. In the Friedman case, a district court rejected the plaintiffs' claim that several investment banks engaged in price fixing in violation of the Sherman Antitrust Act when they discouraged investors from selling IPO shares during the penalty bid period. Plaintiffs' appeal is now pending before the U.S. Court of Appeals for the Second Circuit. The Commission has filed an amicus curiae brief, urging the Second Circuit to affirm the district court. The Division nonetheless encourages me to conclude that the particular anti-flipping policy imposed here was unlawful. The Division has not addressed the market factors that create the incentives to flip. Because there were no clear rules in June 1996 about how far an underwriter could go to resist flipping pressure, I decline to do so.
c. Assuming that a manipulation occurred, the weight of the evidence fails to show that Setteducati's interaction with HJM's branch office managers and registered representatives facilitated the manipulation.
The OIP alleged that the registered representatives in HJM's San Francisco and Chicago offices engaged in impermissible sales tactics at the direction of their branch managers. It did not charge that Setteducati played any role in this misconduct.
The Division's More Definite Statement identified an additional theory of liability: that Setteducati directly or indirectly implemented a firm-wide policy which discouraged sales of IPO shares and encouraged the purchase of additional aftermarket shares by the firm's customers. I understand the More Definite Statement to relate only to the allegation that Setteducati urged the firm's registered representatives to "support the deal" and not to "leave money on the table."
The record demonstrates that Setteducati exhorted the firm's representatives to "support the deal." The term did not necessarily have a sinister meaning (Tr. 412-13, 486). Certainly, stabilizing a new issue so that it does not fall below the offering price is permissible and, according to Ritter, expected. Nothing in the record warrants an inference that when Setteducati said "support the deal" what he really meant was for the registered representatives to sell shares of new issues in the aftermarket at artificially high prices or without regard to the antifraud laws.
The weight of the credible evidence does not establish that Setteducati personally taught or encouraged the sales force not to leave customers' money on the table (Tr. 721-22 ("possibly, maybe Bobby Setteducati" taught it), 824-25 (Setteducati did not teach it)). Rather, the record shows that the San Francisco and Chicago branch managers were responsible for originating that instruction (Tr. 486-87, 633-35). I find no basis to infer that Setteducati knew of such tactics or that he intentionally and knowingly acquiesced in them.
One month before the hearing, the Division announced its intent to pursue an even broader theory of liability: that Setteducati was also responsible for misconduct that OIP ¶¶ II.N and II.R attributed only to the San Francisco and Chicago branch office managers (Prehearing Conference of Jan. 5, 2001, at 5-7, 20-21). In its posthearing pleadings, the Division has finally made this theory explicit. It now alleges that the "boiler room" tactics used by HJM's retail sales force to sell Borealis to the public was "a direct and predictable result" of the pressure that Setteducati applied to the firm's branch managers (Div. Prop. Find. # 107; Div. Br. at 25).
This theory and language belonged in the OIP. If the Division wanted to claim that Setteducati pressured the branch office managers and that the branch managers responded by pressuring the representatives in their branches, or if the Division wanted to claim that Setteducati directly pressured the registered representatives, it could have easily sought to amend the OIP to so allege. Instead, it resisted Setteducati's request for a More Definite Statement.
While one month's notice prior to the hearing may be (barely) sufficient to blunt Setteducati's claim of unfair surprise, this theory still remains well beyond the scope of the OIP. Nor is this expanded theory of liability fairly within the scope of the More Definite Statement. I conclude that the Division is not free to argue this theory of liability under the circumstances. However, because the Commission may have an opposite view, I examine the merits of the Division's expanded charges.
I have analyzed Setteducati's dealings with branch managers and registered representatives in light of the Commission's opinions in three contested adjudicatory cases.
In Haight, 44 S.E.C. at 486-87, 497-98, the Commission held that if a firm conducts training programs and staff meetings where instruction is given in sales techniques and if the sales techniques are designed and operated to defraud clients, then the individual respondents who taught the sales techniques engage in a scheme to defraud investors. Because the president of a brokerage firm participated in the training program, attended the staff meetings, and selected the underwritings, the Commission determined that he was fully cognizant of and directed the fraudulent scheme.
In Universal Heritage Invest. Corp., 47 S.E.C. 839, 844-45 (1982), the Commission found that only supervisory liability, but not primary liability, should be imposed on the executive vice president of a brokerage firm. One branch office of the firm engaged in manipulation and other fraud, but the firm's other nineteen branch offices did not. The managers of the one offending branch office encouraged high pressure sales tactics. The Commission held that there should be no primary liability on the executive vice president for participating in the firm's manipulation and fraud, because there was insufficient evidence that he was actually aware of the misconduct of the branch office. The Commission also declined to find scienter, because the record did not show that the executive vice president intentionally and knowingly acquiesced in the illegal activity of the branch office.
In Padgett, 52 S.E.C. at 1264-66, the Commission considered the situation of the president and chairman of a brokerage firm, who oversaw the creation of a training manual and discussed it at a meeting of regional vice presidents. The two officials then ordered a consultant to collect sample scripts used throughout the firm and add them to the manual. The two officials never saw the scripts the consultant added until a year later. At that time, they determined that some scripts were inappropriate and they ordered the offending scripts removed from the manual. The Commission determined that the two officers had acted negligently, but not recklessly, with respect to the distribution of the scripts. The Commission's opinion analyzed the substantive violation of Sections 17(a)(2) and 17(a)(3) of the Securities Act in failure-to-supervise language.
Setteducati's behavior was not comparable to the conduct the Commission found reckless in Haight and negligent in Padgett. It was much closer to the conduct the Commission found not to involve primary liability in Universal. The impermissible sales practices, as proven on this record, were confined to two branch offices out of HJM's nationwide total of fifteen offices. There is evidence that the branch office managers in San Francisco and Chicago personally encouraged high pressure sales tactics. However, the record does not support an inference that Setteducati knew of such misconduct, or that he intentionally or knowingly acquiesced in it. Quite the opposite is true. Hayut, a credible witness, testified that Setteducati did not encourage unethical sales practices. Hayut also blamed Masucci, not Setteducati, for micro-managing HJM's retail sales effort.
The Division presented evidence as to how Setteducati interacted with HJM's sales force on other IPOs, and it asks me to infer that he likely acted the same way during the Borealis IPO. That inference depends upon the unspoken assumption that the Borealis IPO required the same strenuous selling effort as HJM's other underwritings. In other words, it requires a determination that the Borealis IPO was sold, not bought. There is evidence on both sides of that issue. Ordinarily, the payment of extra incentive compensation to representatives would warrant an inference that Borealis needed some sort of a boost in the aftermarket. However, the evidence concerning the extra incentive compensation is quite weak, and there is little reason to assume that such compensation was only available on Borealis transactions, as distinguished from transactions in all of HJM's "proprietary products." See supra p. 17 & note 21. On the other hand, the Borealis IPO was quite popular with HJM's representatives (Tr. 405, 504, 851-52). It was also quite popular with the public. Customers willingly assisted HJM's sales effort by providing registered representatives with the names of their old school mates, co-workers, and siblings (Tr. 1229-30, 1264, 1305, 1308, 1320, 1382, 1385-86, 1392-93). I decline to draw the requested inference.
There should be no need to decide this issue on the basis of inferences. The Division could have asked Carlson or Masucci if Setteducati actively monitored the indications of interest in Borealis, if Borealis was a hard sell or an easy sell, and if Setteducati cracked the whip to motivate the sales force. Only if the witnesses had no knowledge of these subjects would the requested inference be necessary. When the prosecution disregards the most obvious source of direct evidence and relies solely upon inferences, it is difficult to accept its proof as meeting the weight of the evidence standard.
Setteducati resorted to profanity and threats to prod branch managers into making and then keeping commitments to sell some of HJM's other offerings. He also intimidated HJM's branch managers by ignoring them or by singling them out for harsh criticism in front of their peers. Such managerial conduct is deplorable. Cf. Stephen B. Carlson, 53 S.E.C. 1017, 1020-23 (1998) (finding that implied threats of unpleasant consequences are unacceptable). However, the record does not warrant an inference that Setteducati engaged in such tactics in connection with the Borealis offering.
d. Assuming that a manipulation occurred, prior regulatory actions against Setteducati and others do not demonstrate Setteducati's knowledge of the Borealis manipulation or his intent to participate in a scheme to manipulate Borealis.
A respondent's prior disciplinary action is often relevant to sanctioning; however, it may not be used to demonstrate the respondent's propensity to violate the federal securities laws. See Perles, 77 SEC Docket at 916 & n.44; cf. Fed. R. Evid. 404(b). It may be used for more limited purposes, as previously discussed. See supra note 2.
Setteducati's prior disciplinary history. In March 1990, at the same time that the Commission brought an injunctive action against TJA in federal district court, the Division of Securities of the Florida Department of Banking and Finance charged TJA, Setteducati, and others with violating the Florida Securities and Investor Protection Act (DX 4). Setteducati requested a formal hearing on the charges, but none was ever held. After a year of legal skirmishing, Setteducati and the State of Florida reached a stipulation and consent agreement. See Thomas James Associates, Inc., 1991 Fla. Sec. LEXIS 82 (April 19, 1991). The State of Florida did not find that Setteducati had committed any violations of law. Without admitting or denying the allegations in the Florida complaint, Setteducati agreed not to reapply for registration in the State of Florida for eighteen months. He also reimbursed the State of Florida $3,000 for its expenses. See id.
The Division has repeatedly cited from the Florida complaint, as if that document contained adjudicated facts (Div. Br. at 12, 16-18, 21, 24, 27, 37, 41-42, 44, 49, 52; Div. Reply Br. at 12). In reality, the Florida complaint was simply a collection of allegations that were never proven. In light of the wording of the April 19, 1991, stipulation and consent agreement, I have given the allegations in the Florida complaint no weight. The Florida stipulation and consent agreement cannot be used to show that Setteducati is liable for the manipulation charged in the OIP. Cf. Fed. R. Evid. 404(b), 408. It may be used for the limited purpose of showing Setteducati's knowledge and intent. See Wegerer v. First Commodity Corp., 744 F.2d 719, 723-24 (10th Cir. 1984); United States v. Gilbert, 668 F.2d 94, 97 (2d Cir. 1981). However, because the stipulation and consent agreement found no violations by Setteducati, its probative value here is nil.
In August 1994, NASD's District Business Conduct Committee for District No. 8 issued a complaint against TJA, Setteducati, and others, alleging violations of NASD's Rules of Fair Practice. The NASD complaint charged that Setteducati failed to supervise an individual in TJA's Chicago branch office who was subject to his supervision, and failed to prevent unauthorized trading by that individual in a customer's account (DX 4). In September 1995, NASD accepted Setteducati's settlement offer. Without admitting or denying the NASD's allegations, Setteducati agreed to accept a censure and a fine of $5,000 (DX 4). The NASD settlement letter specifically found that Setteducati did not have a prior disciplinary history (DX 4, Decision and Order dated Sept. 14, 1995, at 8).
Like the 1990 Florida complaint, the 1994 NASD complaint is not entitled to any weight here. The NASD acceptance, waiver, and consent may be used to show Setteducati's knowledge and intent. However, the Division is obliged to explain how a failure to supervise finding in the NASD proceeding is probative here, where Setteducati has not been charged with any supervisory violations. The Division has not done so.
The prior disciplinary history of others. The Division makes repeated references to TJA's previous participation in similar schemes (Div. Prehearing Br. at 5 n.6; Div. Br. at 2, 23, 37, 48). It urges me to infer that Setteducati knew of the Borealis manipulation because he "was familiar with" the charges in the Commission's 1990 enforcement action against TJA (Div. Prop. Find. # 153). According to the Division, Setteducati simply "utilized several of his firm's tried and true manipulative devices throughout 1996" (Div. Br. at 42). The Division also seeks an inference that Setteducati "knew" that conduct "like his" had been "found fraudulent" twice (Div. Br. at 6 n.2 and 43).
There are several problems with the Division's attempt to use settlements by others to prove knowledge and scienter by Setteducati. As previously shown, the federal district court's May 9, 1990, consent decree contained no factual findings about TJA. The same is true of TJA's stipulation and consent agreement with the State of Florida. See Thomas James Associates, Inc., 1991 Fla. Sec. LEXIS 379 (Apr. 19, 1991). Evidence of TJA's May 1990 settlement with the Commission and its April 1991 settlement with the State of Florida would not be entitled to any weight against HJM in this proceeding, if the Division had offered them as proof of HJM's liability for the Borealis manipulation. Cf. Fed. R. Evid. 408.49 I conclude that HJM's compromise agreements with the Commission and the State of Florida are not entitled to any weight against Setteducati for any purpose.50 The same conclusion applies to HJM's 1995 settlement with the NASD.
Even if the Division were able to establish that Setteducati knew of the precise metes and bounds of TJA's prior regulatory problems with the Commission, that showing would not provide much help to it here. TJA installed appropriate procedures to prevent recurrence of the violations identified in the Commission's 1990 complaint. Once the Commission approved those procedures, TJA resumed its underwriting programs, subject to monitoring by a court-appointed receiver. If Setteducati knew anything about the 1990 injunctive action, he knew that the Commission considered the problems it raised to be solved. If he had read Judge Tellesca's opinion, he also knew that TJA had a "relatively clean record with securities regulators." See Thomas James Associates, 738 F. Supp. at 92-93. Alternatively, if the Division's purpose is to show that Setteducati should have been paying closer attention to his colleagues as the Borealis IPO was undertaken, that is essentially a failure to supervise argument, and beyond the scope of the charges raised.
As a final matter, the Division also seeks an adverse inference from the fact that Setteducati knew that Joseph Gianni (Gianni), TJA's sales director, left the firm a few months after the Commission's 1990 injunctive action (Tr. 1450-51; Div. Prop. Find. # 153). The argument is frivolous. The Division neglects to mention that the Commission dismissed with prejudice all its charges against Gianni (Exhibit E, Division's Submission of Documents, dated March 28, 2002). The Division has failed to demonstrate that Gianni's departure from TJA carried any stigma.
Assuming arguendo that Vanechanos manipulated the price of Borealis as alleged in the OIP, the weight of the evidence does not establish that Setteducati, acting with scienter, participated in the manipulation as a principal, willfully aided and abetted it, or caused it.
Pursuant to Rule 351(b) of the Commission's Rules of Practice, 17 C.F.R. § 201.351(b), I certify that the record includes the items set forth in the record index issued by the Secretary of the Commission on November 6, 2001, as amended on August 8, 2002.
Based on the findings and conclusions set forth above, IT IS ORDERED THAT the proceeding against Respondent Robert Setteducati is dismissed.
This Order shall become effective in accordance with and subject to the provisions of Rule 360 of the Commission's Rules of Practice, 17 C.F.R. § 201.360. Pursuant to that rule, a petition for review of this Initial Decision may be filed within twenty-one days after service of the decision. It shall become the final decision of the Commission as to each party who has not filed a petition for review pursuant to Rule 360(d)(1) within twenty-one days after service of the Initial Decision upon such party, unless the Commission, pursuant to Rule 360(b)(1), determines on its own initiative to review this Initial Decision as to that party. If a party timely files a petition for review, or the Commission acts to review as to a party, the Initial Decision shall not become final as to that party.
James T. Kelly
Administrative Law Judge
|1|| Setteducati's answer to the OIP, filed April 6, 2000, will be cited as "Ans. ___." The hearing transcript, as amended by my Order of April 5, 2001, will be cited as "Tr. ___." The hearing exhibits offered by the Division and by Respondent will be cited as "DX ___" and "RX ___," respectively. The Division's Proposed Findings of Fact and the Division's Proposed Conclusions of Law and Post-Hearing Brief, both dated May 4, 2001, will be cited as "Div. Prop. Find. ___" and "Div. Br. ___," respectively. Setteducati's Proposed Findings of Fact, Conclusions of Law, and Post-Hearing Brief, dated July 27, 2001, will be cited as "S. Br. ___." The Division's Reply Brief, dated August 22, 2001, will be cited as "Div. Reply Br. ___."
Pursuant to Rule 323 of the Commission's Rules of Practice, 17 C.F.R. § 201.323, the parties have been afforded an opportunity "to show the contrary" on all matters as to which official notice has been taken. See Order of July 16, 2002.
As a result of the four settlements, the Division has stipulated that OIP ¶¶ II.C, II.F, II.AD through II.AN, II.AS through II.AU, and II.AV through II.AW are no longer at issue (Tr. 5-7, 122-24).
|2||There are well-recognized limits to the proper use of such "prior bad acts" evidence, and those limits are addressed below. See infra notes 26 and 49. In general, the Division may not attempt to prove that HJM manipulated Borealis by showing a propensity for wrongdoing on the part of TJA or HJM or their officers. It may use such evidence for the limited purpose of proving motive, intent, knowledge, or the like. Cf. Fed. R. Evid. 404(b). In addition, special care must be taken in evaluating settlements that were negotiated without admitting or denying liability. Cf. Fed. R. Evid. 408.|
|3||Paragraph II.G of the OIP alleges that NASD fined the firm $500,000. That claim is inaccurate (DX 9 at 9).|
|4|| The Division's witness list included several of HJM's senior managers who might have clarified the scope of Setteducati's duties, but the Division elected not to call them. While I do not speculate what the absent witnesses might have said, I do give the strongest weight on this issue to the evidence provided by Setteducati. Some witnesses who testified on this subject were lower level employees and their testimony lacked a reliable foundation. They were confused about who held what title and about who reported to whom within HJM at the relevant time (Tr. 59, 346-47, 746, 939-40). Two witnesses stated that Setteducati got things done and made all the important decisions (Tr. 368, 732). One was referring only to the years 1997 and 1998 (Tr. 331). It is unclear if the other was referring to the period before July 1996, the period after July 1996, or to both periods.
The OIP also overreached when it alleged that another respondent, William Masucci, was responsible for "all sales" by HJM's offices (OIP ¶ II.E). In fact, Masucci was responsible for retail sales, but not institutional sales (Tr. 58-59).
Finally, the OIP alleges that Villa failed to supervise two subordinates, but it does not allege that Setteducati failed to supervise anyone (OIP ¶¶ II.AG-II.AN). The Commission's settlement with Villa found that Villa supervised the head trader and did not properly delegate his duty to supervise the head trader to anyone else. See H.J. Meyers & Co., 73 SEC Docket 2594, 2597 at ¶¶ II.F, II.H (Nov. 17, 2000).
|5||Continuation of an entity as a "going concern" is assumed in financial reporting in the absence of significant information to the contrary. Information that significantly contradicts the "going concern" assumption relates to the entity's inability to continue to meet its obligations as they become due without substantial disposition of assets outside the ordinary course of business, restructuring of debt, externally forced revisions of its operations, or similar actions. See Codification of Statements on Auditing Standards § 341.01 (1996).|
|6||HJM's underwriting procedures required the firm to syndicate at least 20% of each offering to a selling group of other brokers and dealers who were willing to participate in the IPO (DX 26 at 5218, 5225, DX 57 at 11 n.3).|
|7||Typically, registered representatives called prospective customers and asked how many IPO shares they would be interested in obtaining. A positive response with a specific number of shares was considered a non-binding indication of interest (Tr. 237-38, 612-14). Collective lists of prospective customers were then compiled showing how many total shares were "indicated for" (DX 14). Actual orders were taken from the interested customers later, after the registration statement for the IPO had been declared effective (Tr. 238; DX 26 at 5217-18, 5226). Customers could reduce or cancel their indications of interest at the last minute (Tr. 160-64; DX 46, lines 1-2).|
|8||The OIP sometimes refers to "the float" (OIP ¶¶ II.A, II.I, II.J). At other times, it references "the floating supply" of Borealis stock (OIP ¶¶ II.A, II.Q, II.U, II.V). The Division has used the terms interchangeably. As discussed below, the Commission has defined the terms "floating supply" and "public float" differently, and there is no commonly understood meaning for the term "the float." See infra note 38 and associated text.|
|9||Masucci was in California at the time, and Setteducati was in New Jersey (Tr. 1538; DX 11, DX 12 (facsimile number showing Area Code 908)).|
|10|| In his investigative testimony, Setteducati claimed that allocations of IPO shares were based on indications of interest "in the system" (Tr. 1638-39, 1641). At the hearing, Setteducati testified that allocations were "sometimes" based on indications of interest "in the system," but that "we obviously allocated" Borealis "a little bit differently" (Tr. 1638, 1643-44). I credit Setteducati's hearing testimony on this point.
I am troubled by Massucci's testimony about DX 11 and DX 12. The former document is three pages long, while the latter is two pages long. Masucci acknowledged that one page common to both exhibits was not prepared until the effective date of the Borealis IPO (Tr. 1555, 1557). Clearly, that page was not "done by Bobby" on June 6, 1996.
|11|| Among other things, HJM's underwriting procedures were designed to ensure that every prospective investor who indicated interest in an IPO would promptly receive a copy of the issuer's preliminary prospectus (DX 26 at 5). By entering the prospective investor's indication of interest into the computer, Carlson generated a mailing label (Tr. 98). Carlson was responsible for mailing the red herring prospectuses to prospective investors each day during the indication of interest period (Tr. 166, 185, 188, 309-10; DX 26 at 5). The mailing labels were printed, and the prospectuses were mailed, whether the prospective customer indicated for one share or one thousand shares.
The Division does not allege that every underwriter's allocation of IPO shares must be "fair" or made on a "first-come, first-served" basis (Tr. 12-13). Rather, it asserts that the allocation method used here constituted a manipulative device in violation of HJM's underwriting procedures (OIP ¶¶ II.I, II.J).
|12||The present case is distinguishable from Salloum, 52 S.E.C. at 209 n.5, in which selling group firms got only a small fraction of the shares they had sought. In addition, HJM's syndicate manager found himself in hot water in June 1996 for failing to persuade other investment banking firms to participate in HJM's IPOs. This supports an inference that HJM was not attempting to limit selling group participation in its IPOs to the "bare minimum" of 20%.|
|13||Three representatives sold 337,300 shares in the IPO, or 44% of the San Francisco branch's total of 762,000 shares. Dennis Keohane (Keohane) alone sold 230,700 shares, or 30% of the branch's total (DX 16, DX 29). Angel Cruz (Cruz), the branch office manager, sold 54,000 shares, and Paul Garvey sold 52,600 shares (DX 16, DX 29).|
|14||The evidence does not support the charge in OIP ¶ II.P that the San Francisco office sold 417,000 Borealis shares in the first week of aftermarket trading.|
|15||"Supporting the deal" and "supporting the market" were the terms used throughout HJM to mean soliciting customer buy orders in the immediate aftermarket (Tr. 498, 623-25, 630-31, 724-26, 820-21). The terms did not necessarily have a sinister meaning (Tr. 412-13, 486). The Division's expert witness testified that, to get a role as lead underwriter in an IPO, the underwriter must typically pledge the issuer that it will support the price (Tr. 1169). The Division's expert also observed that an underwriter that allows its IPOs to drop in price is unlikely to receive future underwriting nods (Tr. 1169). As explained by one HJM registered representative: "[I]n all of our IPOs, we were the sole underwriter and if we didn't buy any of the stock, no one else was going to buy any of the stock. And if there was no buying and there was only selling, the price would dramatically go down" (Tr. 623-25).|
|16||On June 17, 1996, one week before the IPO, the San Francisco branch submitted indications of interest to Carlson in HJM's Rochester headquarters for prospective customers John Berquist (Berquist) and Joan Myers (Myers) (DX 14 at 22820, 22825). There was ample time for Carlson to mail these customers the red herring prospectus in advance of the effective date. The third customer, Mark Faith, was the brother of Borealis's president and chief executive officer. He decided to buy only at the last minute (Tr. 1383) ("[I]t was a bit of a hurry to get the account opened for [ ] the first day of trading."). As a result, he missed the IPO and purchased only aftermarket shares. Mark Faith was home in Sussex, England, when trading started (Tr. 1383). His name does not appear on the daily indication of interest list (DX 14). The parties have not addressed the issue of whether delays in the international mail system should be considered a mitigating factor as to Mark Faith's tardy prospectus.|
|17||The script evidence here is quite weak in comparison to the script evidence in other proceedings. See SEC v. First Jersey Secs., Inc., 101 F.3d 1450, 1473-74 (2d Cir. 1996) (finding that a branch manager would come out of his office, tell the representatives to clear their telephones so they could hear, instruct them to write down the script verbatim, and repeat the message several times until the representatives had it correct in their notes); Richard J. Puccio, 52 S.E.C. 1041, 1042 (1996) (describing a "classic boiler room" firm as one that held mandatory sales meetings at which sales techniques were demonstrated, scripts for the firm's "house stocks" were distributed, and representatives were expected to follow the scripts and only give customers the information they contained); Jay Michael Fertman, 51 S.E.C. 943, 947 (1994) (finding that the respondent provided scripts, told the sales force to use them, and said that deviations from the texts would be permitted only if he approved the changes); M. Rimson & Co., 63 SEC Docket 2910, 2918-19 (Initial Decision) (Feb. 25, 1997), final, 64 SEC Docket 757 (Apr. 9, 1997) (crediting a witness's testimony that he heard salesmen using a script verbatim with customers).|
|18||I offered the Division's summary witness an opportunity to reconcile the discrepancies between the daily trading volume figures in DX 36, on the one hand, and those in DX 17 through DX 21, on the other hand. See Order of March 15, 2002. The witness was unable to do so, but opined that the differences were not material (Declaration of Jean M. Javorski, dated March 22, 2002) (Javorski Declaration).|
|19||In the NASDAQ SmallCap Market where Borealis traded, market makers post a bid price at which they are willing to buy and an ask price at which they are willing to sell. The highest bid price is known as the inside bid, and the lowest ask price is known as the inside ask (DX 57 at 5).|
|20||The percentages are based upon the amount of time and are not weighted by the number or size of transactions (DX 57 at 6).|
|21|| In A.S. Goldmen & Co., 75 SEC Docket 49, 62-63 (May 21, 2001), the evidence showed what percentage of total commissions the brokerage firm paid to its registered representatives, what the firm ordinarily used as its cost basis for calculating commissions, and how the extra incentive compensation paid on the relevant transactions differed from the norm. The present record offers no such detail.
In Daniel Richard Howard, Exchange Act Rel. No. 46269 at 3 n.4 (July 26, 2002), the Commission recently found that HJM paid its registered representatives extra compensation for transactions in six other securities (not including Borealis). The question arises as to whether the extra compensation offered on Borealis differed in any way from the extra compensation offered on these other securities, in which HJM was also the primary market maker. If not, then the proper inference is that HJM encouraged transactions in its proprietary products, rather than in Borealis alone.
|22||DX 34 at 12064 shows 20,000 shares flipped, while DX 43 shows 17,000 shares flipped. Some of McAuliffe's customers were institutions and others were retail (Tr. 993-94).|
|23|| As noted above, the selected dealer agreement is not part of the record. I infer that if HJM and the members of the Borealis selling group had any understanding about the penalty bid, it would have been memorialized in that document. J. Stewart Schultz (Schultz), HJM's syndicate manager, testified that, if HJM had imposed the penalty bid against the members of the Borealis selling group, the effective letters and retention letters that HJM transmitted to the selling group members on the eve of the IPO were one of the places where such a policy might have been memorialized in writing (Tr. 933-35, 953-56; DX 61). The Division suggests that the absence of penalty bid instructions from DX 61 "proves" that HJM waived the penalty bid on the Borealis IPO as to the selling group members. That contention is rejected. The fact that the letters in DX 61 do not mention the penalty bid does not establish that HJM's investment banking committee, in general, or Setteducati, in particular, made a conscious choice to waive the penalty bid against the brokerage firms in the Borealis selling group.
At the start of the hearing, the Division identified its expert witness as its authority for the proposition that the penalty bid did not apply to the members of the selling group (Tr. 11, 35). That assertion is frivolous: the expert simply assumed that the penalty bid did not apply to selling group members because the Division told him to assume it (Tr. 1122; DX 57 at 2).
|24|| The Division is vague as to when the alleged manipulation ended. As stated in Div. Br. at 38 n.13:
|25||Paragraph II.AC of the OIP alleges that Borealis closed at $5.48 on July 23, 1996. That assertion is incorrect.|
|26|| Courts have generally proceeded cautiously in considering a pattern of conduct as habit, "because it necessarily engenders the very real possibility that such evidence will be used to establish a party's propensity to act in conformity with its general character, thereby thwarting [Federal Rule of Evidence] 404's purposes." Simplex, Inc. v. Diversified Energy Sys., 847 F.2d 1290, 1293 (7th Cir. 1988). In other words, courts are concerned that Federal Rule of Evidence 406, admitting evidence of habit, will swallow Federal Rule of Evidence 404(b), which excludes character evidence. See United States v. Mascio, 774 F.2d 219 (7th Cir. 1985). "To offer evidence of a habit, a party must at least demonstrate a regular practice of meeting a particular kind of situation with a specific type of conduct." Jones v. Southern Pac. R.R., 962 F.2d 447, 449 (5th Cir. 1992).
The best proof of a habit is that someone acted in a particular way at specific and frequent instances, and also that the person did not act otherwise at other similar instances. See Simplex, 847 F.2d at 1294. Whatever the nature of the proof, the touchstone of reliability is to provide an "adequacy of sampling and uniformity of response" in specific circumstances. See McWhorter v. City of Birmingham, 906 F.2d 674, 679 (11th Cir. 1990). I have applied these principles here.
|27||The Division also relies on Briggs's testimony to show that Setteducati demanded that branch managers find different customers accounts in which to place aftermarket shares when customers insisted on selling new issues immediately (Div. Prop. Find. # 76; Div. Br. at 23). Briggs offered no such testimony. He stated that there was "only one instance" of such a demand. The call in question "came from [Setteducati's] office. It could have been Mike Bergin. . . . But I can't say that it was specifically Bobby, I don't recall" (Tr. 324-25).|
|28||The evidence in this case is quite weak in comparison to the evidence in Steven P. Sanders, 68 SEC Docket 982, 998-99 (Oct. 26, 1998), where the Commission found that a firm's president was typically in the trading room during the initial aftermarket for the firm's new offerings and that his compensation was tied to profits from trading activity. It is also deficient in comparison to the evidence in Edward J. Mawod & Co., 46 S.E.C. 865, 875 n.46 (1977), where the Commission found that the manipulator "stationed himself" in Mawod's trading room and learned about orders for the manipulated stock "at once." In affirming the Commission's Mawod opinion, the U.S. Court of Appeals for the Tenth Circuit noted that the manipulator was spending three to four hours a day in the trading room. See Edward J. Mawod & Co. v. SEC, 591 F.2d 588, 595 (10th Cir. 1979). Here, the record does not establish that Setteducati was even in New York City on June 24, 1996. The evidence shows that he was usually in Red Bank on Mondays.|
|29||In March 1997, Setteducati threatened to fire Peter Baxter, a representative who dealt with institutional customers, for complaining about HJM's aftermarket trading practices and its slow delivery of stock certificates (Tr. 797-98; DX 33). The evidence demonstrates the extent of Setteducati's authority in March 1997, but it is not probative of Setteducati's managerial control of the trading department in June 1996.|
|30||Before the hearing, the Division described Myers as an "engineer" (Div. List of Fact Witnesses, dated September 27, 2000, at 12). After the hearing, the Division described Myers as a "housewife" (Div. Prop. Find. # 194).|
|31||The Commission has rejected arguments that customers were not defrauded because they were sophisticated or experienced. See Steven D. Goodman, 74 SEC Docket 707, 714 n.20 (Jan. 26, 2001); Joseph J. Barbato, 53 S.E.C. 1259, 1275 n.17 (1999). My evaluation of the witness does not clash with this precedent because sophistication in investing presents a different issue than credibility in testifying. I find that this highly-educated mathematician with prior experience in IPOs is not entitled to much credibility after she admitted that she failed to read mail from her brokerage firm, professed confusion about understanding her account statements, was unable to demonstrate the source of the confusion, and bristled with unwarranted hostility to legitimate cross-examination.|
|32||Gensym was delisted from the NASDAQ National Market System in August 2001. It is now posted on the electronic bulletin board, where it is subject to regulation as penny stock (Gensym Form 10-Q for period ending September 30, 2001) (official notice).|
|33||Scott DeCarlo, Michael Schubach, and Vladimir Naumovski, A Decade of New Issues, Forbes, March 5, 2001, at 180-81 (reporting that 233 of 707 IPOs during 1996 were backed by venture capital) (official notice); Venture-Backed IPOs Begin to Rebound in 4th Quarter According to Thomson Financial/Venture Economics and the National Venture Capital Association, Business Wire, January 7, 2002 (reporting that 268 of 771 IPOs during 1996 were backed by venture capital) (official notice).|
|34||The size of Borealis's first day's run-up also depends on the fortuity that the last reported transaction of the day was a buy at the inside ask ($8.125), rather than a sale at the inside bid ($7.50). Had the last reported transaction been a sale at the inside bid, then Borealis's first day run-up would have been 50%, rather than 62.5%.|
|35||Erik W. Chan, 77 SEC Docket 851, 867 n.47 (Apr. 4, 2002) involved a respondent who caused a violation of the antifraud provisions, but the Commission never specified whether Chan acted negligently or recklessly. As I read the Commission's opinion, the respondent acted at least recklessly.|
|36|| Just because the Division requests an inference, it does not always get one. On several occasions, the Commission has refused to draw inferences the Division urged it to draw. See, e.g., Russo, 53 S.E.C. at 282 (refusing to infer substantial assistance on the part of an accused aider and abetter); William R. Carter, 47 S.E.C. 471, 505 (1981) (refusing to infer knowledge of materiality); Fox Secs. Co., 45 S.E.C. 377, 383 n.17 (1973) (refusing to infer that a respondent misappropriated assets).
Inferences and presumptions are related but distinct concepts. In its April 1994 manipulation concept release, the Commission proposed to adopt a presumption that a distribution participant or its affiliated purchaser who engaged in proscribed activity during a distribution had done so with manipulative intent. The presumption would have placed the burden on that person to prove that the conduct was not done with manipulative intent. See Review of Antimanipulation Regulation of Securities Offerings, 56 SEC Docket 1442, 1454 n.121 (Apr. 19, 1994). The proposed presumption, which would have significantly reduced the Division's burden in price manipulation cases, was never adopted.
|37||For example, HJM led or shared the inside bid from 10:20 a.m. to 10:21 a.m. (DX 38). If no trades were then taking place, the Division must explain its theory of how HJM was adding to its market share. Conversely, if the inside bid was falling between 10:21 a.m. and 10:24 a.m., when HJM did not lead or share it, and if the inside bid was falling further between 10:25 a.m. and 10:28 a.m., when HJM did lead or share it, the Division is obliged to explain how a series of downward movements of the inside bid is probative of the upward price manipulation alleged in the OIP. Ritter's graph persuades me that the inside bid was moving down in just this fashion before HJM set its inside bid at $7.50 (DX 57 at 7). The Division has offered no proof that HJM was adding to its 80% share of the Borealis market before 10:28 a.m. Bader flipped 30,000 shares for an institutional customer at 10:24 a.m. (DX 42), but there is no evidence that the transaction added to HJM's market share.|
|38||HJM calculated the "public float" of Borealis at two million shares (DX 51). Javorski used a larger number of shares, because she included HJM's overallotment option in her computations (Tr. 423-24). The Commission's definition of "public float" is set forth in 17 C.F.R. § 228.10(a)(1)(Note).|
|39||"Liability for manipulation wholly independent of fictitious transactions [such as wash sales or matched orders] . . . raises interesting questions. Without such transactions, the core of the offense can become obscure." Markowski v. SEC, 274 F.3d 525, 528 (D.C. Cir. 2001), petition for cert. filed, No. 01-1749 (May 28, 2002). Commission opinions finding that proof of deceptive value is unnecessary if there is proof of deceptive market activity have involved wash sales and matched orders. The present case does not.|
|40||OIP ¶¶ II.AD-II.AF allege that Vanechanos charged retail customers excessive and undisclosed markups on June 24 and June 25, 1996. If the Division is attempting to equate excessive markups with manipulated prices, its argument fails. First, the Division has made clear that the allegations of excessive and undisclosed markups were in addition to the allegations of price manipulation (Div. Prehearing Br. at 1). A firm can charge excessive markups without manipulating the price of a security and it can manipulate the price of a security without charging excessive markups. The Division did not show how the two allegations were related, or if they were related. Second, the excessive and undisclosed markups were alleged to have occurred only on the first two days of the manipulative period. There must be other proof to support the charge of price manipulation by Vanechanos on June 26 through June 28, 1996, the last three days of the manipulative period.|
|41||The range of trades on June 26 through June 28 supports an inference that there were changes in the inside bid and inside ask prices: downward movements and a narrowing of the spread that are entirely inconsistent with the theory of the prosecution.|
|42||See, e.g., Jay R. Ritter and Ivo Welch, A Review of IPO Activity, Pricing, and Allocations, 57 J. Fin. 1795, 1802 (forthcoming Aug. 2002) ("We know of no exceptions to the rule that the IPOs of operating companies are underpriced, on average, in all countries."); Roger G. Ibbotson, Jody L. Sindelar, and Jay R. Ritter, The Market's Problems with the Pricing of Initial Public Offerings, 7 J. App. Corp. Fin. 66, 68, 74 (Spring 1994) (finding that smaller offerings are underpriced by more, on average, than larger offerings, and concluding that "the pricing of young growth companies going public is difficult, and there is evidence that the market fails to get it right."); Roger G. Ibbotson, Jody L. Sindelar, and Jay R. Ritter, Initial Public Offerings, 1 J. App. Corp. Fin. 37 (Summer 1988) ("Numerous empirical studies have shown that unseasoned new issues are significantly underpriced, on average."). "[P]art of the problem [is] that the new breed of profitless IPO candidates, and the unprecedented bull market, [make] it a challenge for Wall Street to value stocks accurately." Shawn Tully, Betrayal On Wall Street, Fortune, May 14, 2001, at 81. As Ritter and Welch have shown, 37% of the IPOs between 1995 and 1998 involved issuers with negative earnings in the year before going public. See 57 J. Fin. at 1800.|
|43||The preliminary prospectus for the IPO of Netscape Communications Corporation (Netscape) estimated an offering price of $12-$14 per share. Shortly before the Netscape IPO in August 1995, the issuer and underwriters raised the offering price to $28 per share. On the first day of aftermarket trading, Netscape opened at $71 per share, rose to $74 per share, and closed at $56 per share. See Fu-Sung Peter Wu, 77 SEC Docket 922, 924-28, 931 n.25 (Apr. 4, 2002) (rejecting the respondent's claim that the high opening price in aftermarket trading for Netscape surprised the industry as a whole). In Monetta Fin. Servs., Inc., 72 SEC Docket 72, 77 (Initial Decision) (Mar. 27, 2000), review granted, an Administrative Law Judge found that a firm commitment IPO only occurs where the underwriters judge that the offering is comfortably oversubscribed based on realistic indications of interest that are well in excess of the number of shares that they expect to sell. The ALJ further credited expert testimony that IPO shares are priced, on average, to appreciate immediately on the first day, and that those accounts receiving the shares at the offering price garner instant first-day profits. Id. at 78.|
|44||OIP ¶ II.Y alleges that HJM purchased Borealis shares from the flipping customers of its favored employees at prices ranging from $7.19 to $7.25 per share on June 24. The record shows that the price HJM paid to its flipping customers on June 24 ranged from $7.16 to $8.25 per share (DX 17).|
|45||See Edward Wyatt, NASDAQ Sinks as Investors Turn to the Blue-Chips, New York Times, July 25, 1996, at D1 ("Small-company stocks. . . continued to erode yesterday, as investors turned further away from that increasingly volatile sector of the market in favor of blue chips. . . . Most analysis call the current downturn in NASDAQ stocks `a classic correction,' similar to declines in 1992 and 1994, when the NASDAQ index fell 10% to 20%."); NASDAQ Drops Despite Boost from Compaq, Atlanta Journal and Constitution, July 25, 1996, at 1B ("The stock market, stuck in the throes of the most severe correction in nearly two years, is expected to revisit, and probably sink beneath, the lows reached at the depths of last week's sell-off. . . . Analysts say investors are selling into the good news from any liquid instruments that have held their value relative to the market."); Patricia Lamiell, Associated Press, Stock Market Rout, Federal Settlement, Has So Far Not Deterred NASDAQ, AP Worldstream, July 25, 1996 ("Prices for many NASDAQ stocks have plummeted from record highs set last month. . . . NASDAQ has been hit with a one-two punch in the past week. On July 17, two dozen Wall Street firms settled a nearly two-year federal investigation of NASDAQ price fixing and fee padding. . . . The settlement was reached in the middle of a significant market retreat.").|
|46||A regression analysis looks at the relationship between two variables to determine whether the relationship between the variables is statistically meaningful. The methodology is well established and reliable. See Askew v. City of Rome, 127 F.3d 1355, 1365 n.2 (11th Cir. 1997).|
|47||Consistent with Rule 323 of the Commission's Rules of Practice, I advised the parties that I would take official notice of these specific matters from Borealis's quarterly and annual reports to the Commission. See Order of July 16, 2002. Neither party objected.|
|48||Masucci testified that the San Francisco and Chicago branches ranked first and second, respectively, in terms of their contributions to HJM's gross revenue production (Tr. 1547). The Division's proof at the hearing lumped the two offices together in an exhibit comparing the percentage of Borealis IPO shares allocated with the percentage of revenue generated (DX 45). The Division has acknowledged that office production is a legitimate basis for an allocation (Div. Prehearing Br. at 7). Nonetheless, the data were presented in a way that made it impossible to consider the Chicago office by itself, to determine if its allocation of 175,000 Borealis IPO shares was disproportionate to the revenue it generated. See Order of July 16, 2002.|
|49||Under Federal Rule of Evidence 404(b), evidence of other bad acts is not admissible to prove character as the basis for suggesting the inference that conduct on a particular occasion was in conformity with it. Character evidence is of slight probative value and may be very prejudicial. It tends to distract the finder of fact from the main question of what actually happened on the particular occasion. It subtly permits the trier of fact to reward the good man and punish the bad man because of their respective characters, despite what the evidence in the case shows actually happened.|
|50||As to the 1990 federal district court consent injunction, the Division might have sidestepped this problem. Federal Rule of Civil Procedure 65(d) provides that an injunctive order is binding not only on the parties, but also on their employees who receive actual notice of the injunction. Setteducati was an employee of TJA in 1990. The Division might have asked Setteducati if he had actual notice of the 1990 injunction against TJA. If he said "yes," the problem of proving his knowledge would be solved. If he said "no," or if he said he had forgotten the TJA injunction during the intervening six years, the Division could have asked for an adverse credibility finding. The Division elected not to ask Setteducati anything on this subject, and chose to proceed via the "inference" route.|
|Home | Previous Page||