1 1 UNITED STATES SECURITIES AND EXCHANGE COMMISSION 2 3 4 ROUNDTABLE TO DISCUSS SHORT SALE PRICE TESTS 5 AND SHORT SALE CIRCUIT BREAKERS 6 7 Tuesday, May 5, 2009 8 9 Securities and Exchange Commission 10 100 F Street, N.E. 11 Washington, DC 20549 12 13 14 15 16 17 18 19 20 21 22 Diversified Reporting Services, Inc. 23 (202) 467-9200 24 25 Amended 5/26/09 2 1 C O N T E N T S 2 3 Page 4 Opening Remarks 3 5 Panel One - Market Changes and Investor Confidence 5 6 Panel Two - Bid versus Tick versus Circuit Breakers 58 7 Panel Three - Lessons & Insights from Empirical Data 111 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 3 1 P R O C E E D I N G S 2 10:04 a.m. 3 OPENING REMARKS 4 CHAIRMAN SCHAPIRO: Good morning. Welcome to the 5 Securities and Exchange Commission's Roundtable on short 6 selling. The Commission is grateful that so many have agreed 7 to participate in today's meeting. I believe I speak for my 8 colleagues on the Commission in saying that we look forward 9 to the panel's comments, insights and recommendations on this 10 very important subject. 11 As I noted at our recent open meeting where we 12 proposed a variety of short sale restrictions, my brief 13 tenure as chairman has seen the issue of short selling 14 outpace any other in terms of the number of inquiries, 15 suggestions and expressions of concern that we've received. 16 Indeed, well before I arrived at the Commission, I heard from 17 many investors on this issue. 18 We know that the practice of short selling evokes 19 strong opinions from both its supporters and detractors. 20 I've made it a priority to evaluate the issue of short 21 selling regulation and ensure that any future policies in 22 this area are the result of a deliberate and thoughtful 23 process, and that is why we're here today. 24 On April 8th, we unanimously voted to propose two 25 distinct approaches to short selling restrictions. One 4 1 approach would impose a permanent, marketwide short sale 2 price test while the other would impose temporary short 3 selling restrictions upon individual securities during 4 periods of severe declines in the prices of those securities. 5 We proposed these amendments to Reg. SHO with the knowledge 6 that any short selling restrictions must balance the goal of 7 helping to prevent abusive short selling with the view that 8 legitimate short selling can provide tangible benefits such 9 as improved liquidity and pricing efficiency. 10 As we seek public comment on the proposed rules, we 11 believe that this roundtable discussion will help in 12 advancing the debate over short sale price tests and short 13 sale circuit breakers. Throughout the day, we'll hear from 14 three panels. Each panelist will take a few moments to share 15 his or her thoughts on the issues being discussed, and when 16 the opening statements are complete, the floor will be open 17 to questions from the Commissioners. 18 The first panel will discuss the necessity and 19 effectiveness of short sale price tests and short sale 20 circuit breakers. The panelists will also comment on whether 21 they believe a short sale price test or circuit breaker 22 could help restore investor confidence which has been 23 seriously eroded during the current financial crisis. 24 The second panel will take a closer look at the 25 Commission's recent proposed amendments to Reg. SHO and 5 1 evaluate the costs and benefits of the regulatory 2 alternatives outlined in the release. The panelists will 3 discuss the operational implications of these proposed rules 4 and what impact they may have on market function and market 5 quality if the Commission were to adopt them. 6 And the third panel will explore the empirical side 7 of the debate over short sale restrictions. The panelists 8 will discuss academic research on the subject of short 9 selling and evaluate the recent proposals from a quantitative 10 and academic perspective. 11 The panelists we will hear from today are leaders 12 in their respective fields and represent a range of 13 constituencies that includes issuers, financial services 14 firm, self regulatory organizations, investors and the 15 academic community. We are truly privileged to have them 16 here, and we look forward to a spirited and substantive 17 discussion. 18 I'll turn the meeting over now to Jamie 19 Brigagliano, acting co-director of the Division of Trading 20 and Markets who will introduce and moderate our panel. 21 Thank you. 22 Jamie. 23 PANEL ONE - MARKET CHANGES AND INVESTOR CONFIDENCE 24 MR. BRIGAGLIANO: Thank you, Chairman Schapiro. 25 We'll now begin the day's first panel titled "Investor 6 1 Confidence and Market Changes: Are short sale price tests or 2 short sale circuit breakers necessary of effective? 3 Following introductions, the panelists will each make a brief 4 opening statement. Because we have a lot of information to 5 cover in a relatively short time, we ask that panelists limit 6 their opening statements to no more than three minutes. 7 During your prepared remarks, Matt Sparkes over here will 8 hold up a yellow card indicating when there's one minute 9 remaining. Following the opening statements, as Chairman 10 Schapiro said, we will engage in discussion with the 11 Commission. 12 Before we begin, I'd like to welcome our 13 distinguished panel. From left to right, Kevin Cronin is 14 director of global equity trading for Invesco. Brian Conroy 15 is senior vice president and head of global equity trading 16 for Fidelity Research and Management Company, the investment 17 management organizations of Fidelity Investments in Boston. 18 Rick Ketchum is chairman and CEO of the Financial Industry 19 Regulatory Authority, FINRA. He's also the chairman of the 20 World Federation of Exchanges Regulatory Committee. John 21 Kozak is senior vice president and chief financial officer of 22 Park National Bank and Park National Corporation of Newark, 23 Ohio. Dan Mathisson is managing director at Credit Suisse 24 and is head of advanced execution services. Mike McAlevey is 25 vice president of corporate securities and financial counsel 7 1 at General Electric Corporation. He's a former Deputy 2 Director of the Division of Corporation Finance at the SEC. 3 And Justin Schack is vice president for market structure 4 analysis at Rosenblatt Securities, an institutional agency 5 brokerage in New York. 6 Kevin, would you like to start us off with your 7 opening statement? 8 MR. CRONIN: Yes, thank you. Thanks for having me. 9 Thank you, Chairman Schapiro and members of the Commission 10 for the opportunity to speak here today. 11 I'm pleased to participate on behalf of Invesco in 12 this roundtable to examine potential restrictions on short 13 sales. Invesco is a leading independent global asset 14 management firm with operations in 20 countries serving 15 clients in over 100 countries with assets under management of 16 approximately $350 billion. I should also mention that 17 Invesco is a publicly traded company and is listed on the New 18 York Stock Exchange. 19 I'm also pleased to participate in this discussion 20 as the chairman of the Investment Company Institute's equity 21 markets advisory committee to express the ICI's preliminary 22 reviews on the SEC's proposals. 23 An efficient and effective trading environment is 24 crucial to the mutual funds shareholders and to investors at 25 large. We commend the SEC for its continued interest in 8 1 addressing issues that may impact the fair and orderly 2 operations of our interest in the securities markets and 3 investor confidence in those markets. 4 Chairman Schapiro, as you noted yesterday in your 5 speech, investors deserve careful examination of the issues 6 surrounding short sales. To the extent that the additional 7 restrictions on short selling can increase investor 8 confidence in the markets, such restrictions should therefore 9 be carefully considered. We share the view of many, 10 including the SEC, that short selling provides needed 11 benefits to the market, including playing an important role 12 in providing market liquidity and in price discovery. For 13 this reason, we believe that legitimate and lawful short 14 selling must be allowed to continue. 15 It is possible, however, for market participants to 16 use short selling as a vehicle to illegally manipulate the 17 prices of stocks to facilitate or to facilitate other market 18 abuses. Those practices must be stopped. 19 Several of the SEC's actions to date already have 20 made great strides in this area, particularly efforts to 21 reduce fails to deliver to reduce the impact of naked short 22 selling. In considering the proposed restrictions, the SEC 23 faces a difficult dilemma. It is unclear from what available 24 empirical data whether any of the proposals would have had 25 increased investor confidence -- would have increased 9 1 investor confidence or alleviated any of the unprecedented 2 market conditions of the past 18 months. Ideally, market 3 forces would address concerns relating to short selling and 4 there would be no need for further restrictions. 5 We recognize, however, that the impairment of 6 investor confidence may dictate the need for some form of 7 regulatory response to recent market events. While a member 8 of the ICI's including Invesco are still examining the 9 possible impact of the SEC's alternative proposals, we 10 believe that the Commission must proceed deliberately as it 11 considered the consequences of the proposals for investors. 12 As the Commission is aware, we are trading in very, very 13 different market circumstance and conditions than we were 14 just a few years ago. 15 The SEC must also balance the potential benefits to 16 proposed regulation with the costs of the rules. Finally, 17 it's critical that any new short sale regulations have a 18 robust inspection and enforcement regime attached that will 19 provide investors with confidence that violators and abusers 20 of these regulations will be detected and punished. 21 I think it is fair to say that there is no absolute 22 consensus among mutual funds on the optimal course of action 23 relating to the proposals. Invesco, along with many of our 24 peers, believe that immediate SEC action is not warranted at 25 the very least until the impact of the options before us are 10 1 considered further. If the SEC determines that it must move 2 forward, Invesco at this time believes that a circuit breaker 3 imposing the proposed modified uptick rule would be the least 4 damaging to the markets. 5 Our fund group believes that the SEC other -- I 6 should say other fund groups believe that the SEC should 7 adopt a proposed modified uptick rule on a permanent basis or 8 possibly through a circuit breaker that when triggered would 9 impose the modified uptick rule. Either way, the parameters 10 and scope of any SEC action must be further delineated to 11 ensure that it does not negatively impact investors. For 12 example, by reducing market liquidity, you're harming price 13 discovery. 14 I would be remiss if I did not spend a moment on 15 the Commission's interim final temporary rule which requires 16 institutional investment managers to report to the Commission 17 certain information concerning their short sales and short 18 positions. While we strongly support the Commission's need 19 to obtain information to address concerns relating to abusive 20 and manipulative short selling, we strongly urge that the SEC 21 continue to adequately protect the confidentiality of this 22 information to prevent front-running of the fund's security 23 positions. If a public disclosure regime is to be 24 established, we believe it is best achieved by the SEC 25 requiring disclosure of a short position on a periodic but 11 1 sufficiently delayed basis. 2 Finally, speaking from the standpoint of an asset 3 management firm with interconnected trading desks in several 4 locations around the world, I urge the Commission to work 5 closely with foreign regulators to create consistent and 6 sensible cross-border regulations in this area. 7 Invesco and the rest of the members of the ICI look 8 forward to working with the Commission as it continues to 9 examine the options for any additional short selling 10 restrictions and their impact on investors and to expressing 11 our final recommendations in the ICI's comment letter next 12 month. 13 I thank the Commission, again, for organizing this 14 roundtable. I look forward to answering your questions. 15 MR. BRIGAGLIANO: Thank you, Kevin. 16 Brian? 17 MR. CONROY: Good morning. As noted, my name is 18 Brian Conroy, and I am the head of global trading at Fidelity 19 Investments. We're at the end of the first quarter of 2009. 20 We had about 2.5 trillion in assets under management and 21 administration, more than 450 billion of which were managed 22 equities. These assets represent the cumulative investments 23 in over 77 million customer accounts for individuals, 401(k) 24 participants and institutions. 25 As a market leader, we believe it is important to 12 1 engage with the policy makers to ensure that the appropriate 2 balance is struck between regulatory oversight and robust 3 market discipline. As a result, we appreciate the Commission 4 holding this hearing and welcome the opportunity to 5 participate. 6 Less than 1 percent of the assets Fidelity managed 7 are in long-short strategies, so while I worked for 15 years 8 in the broker-dealer community and three years at a large 9 hedge fund, I generally come at this topic with the frame of 10 reference of a long rather than short investor. At its core, 11 our trading desks' fundamental responsibility is to implement 12 the investment ideas of our portfolio managers by trading 13 securities as efficiently as possible. To do that, we seek 14 to participate in and facilitate efficient, liquid and 15 orderly capital markets. 16 Efficient markets benefit our shareholders in the 17 form of lower trading costs which ultimately improve 18 performance. As a stakeholder in our capital markets and as 19 a fiduciary for millions of investors, it is important to 20 Fidelity that the regulatory structure governing the U.S. 21 trading market promotes rather than impedes liquidity, 22 transparency and pricing efficiency. 23 At Fidelity, we believe the vast majority of the 24 time short selling helps create a more liquid, transparent 25 and efficient trading market without creating any harm in the 13 1 integrity of the market. At the time, we recognize the need 2 to curb abusive short selling practices such as naked short 3 selling which we believe can be accomplished with the 4 enforcement of existing regulations, including the rules 5 targeting these practices that were adopted by the Commission 6 in the fall of 2008. 7 As the SEC considers whether the permanent 8 reinstatement of some form of uptick rule or additional short 9 selling restrictions are necessary, we encourage the 10 Commission to consider two important points. First, despite 11 the recent upheaval, the U.S. equity market functions as the 12 most competitive, liquid and efficient market in the world. 13 Daily trading volume is the largest, and spread costs are the 14 lowest of any global equity market. Second, in the last 15 eight months, we've experienced a virtually unprecedented 16 equity market disruption. In reacting to these 17 extreme -- extremely unusual recent market events, we 18 encourage you to act cautiously to ensure no damage is done 19 to the robust equity market that currently exists when this 20 serves the interests of so many investors. 21 In addition, we encourage the Commission to review 22 data relating to the existing and recently repealed rules 23 concerning short selling before determining that any 24 additional regulatory action in this area is necessary. 25 After reviewing the data ourselves, Fidelity believes that 14 1 the best way to protect investors and maintain the benefits 2 of an open, liquid and transparent market is for the 3 Commission to not adopt any of the proposal it has put up for 4 comment. 5 If in the interest of bolstering investor 6 confidence, however, the Commissioner -- the Commission 7 believes some form of regulation is required, we believe that 8 the Commission should implement a remedy that addresses the 9 aberrant markets that occur from time to time rather than 10 posing a potentially costly regulatory regime that could 11 burden the market on a daily basis and ultimately increase 12 trading costs for our investors which is certainly not in 13 their interest. As a result, Fidelity believes that at most, 14 the circuit breaker option should be the only proposals under 15 consideration. 16 Thank you for inviting me. I look forward to 17 participating in the discussion. 18 MR. BRIGAGLIANO: All right. Rick Ketchum. 19 MR. KETCHUM: Chairman Schapiro and members of the 20 Commission, thank you very much for having me here. It's an 21 honor to participate in this critical issue. 22 I'd like to make two points at the beginning. 23 First, certainly, this is an area that FINRA as an 24 organization fundamentally focused on investor protection 25 cares a great deal about as well as its responsibilities to 15 1 help enforce whatever you end up adopting. So we have a 2 great deal of concern also with respect to the structure of 3 any proposal that the Commission does move ahead with. 4 Secondly, let me just give as a disclosure that 5 this is not an issue that our board has considered at this 6 point. So the views I'm expressing are my own and when other 7 people convince me differently, I'll figure out a way to back 8 away from them as well. 9 I'd like to just make a few basic points in working 10 through my thinking with care to note the time and try to 11 avoid repetition. First of which is I have believed and 12 continue to believe that the Commission has been correct in 13 focusing its primary regulatory focus with respect to short 14 sale on the areas that Reg. SHO and the interim rules do 15 apply to with respect to locate and settlement. I think 16 it's remarkable how effectively the industry has been able 17 to comply with the tougher rules that have been in place now 18 for months. It is the area, I think, that provides the 19 greatest appropriate balance from the standpoint of 20 protection. 21 I will make a confession, however, though. I was 22 entirely supportive of the Commission's position with respect 23 to completely rescinding the short sell rule. I have been 24 marked by the trading markets that occurred last fall, and I 25 believe that they demonstrate the appropriateness of the 16 1 Commission at least seriously considering possible actions. 2 But I think in looking at those actions one has to define 3 what you're attempting to address and what facts really, 4 really exist. 5 I do believe that the trading in September and 6 October demonstrated certain truths that we've all known but 7 perhaps more dramatically than we've seen. First, it 8 continues to be a fact that the world is net long and the 9 world reacts to dramatic down movements differently than it 10 reacts to dramatic up movements. Secondly, there are 11 financing issues with respect to dramatic down movements that 12 are different in scope and in many ways more predictable than 13 with respect to upwards movements. Third, thinking of doing 14 anything in here without effective jurisdiction with respect 15 to derivatives in the over-the-counter area, particularly 16 credit swaps is probably futile and should be carefully 17 looked at in anything you consider. 18 Finally, if regulatory action is taken, as I think 19 probably it may well be justified, it should be focused on 20 those areas of concerns from the standpoint of predictability 21 and exposure of the marketplace. In simplest terms, the 22 market is not usually exposed and companies are not usually 23 exposed and people are not usually put into an environment 24 where it's reasonably predictable that there will be a 25 cascading failing as a result of concerns with respect to 17 1 either the company or with respect to intermediaries. 2 So if there is to be regulation, I believe strongly 3 it should be focused in the circuit breaker side of the 4 activity. In addition, quotes are certainly better than 5 trades. Speaking as a regulator, I don't want to be involved 6 in the situation of worrying about who's gaming with respect 7 to creating upticks, and I don't think it makes a great deal 8 of sense. 9 Third, I think that while I would strongly 10 recommend you not have a rule that's in place all the time 11 because I think it creates burdens without any particular 12 benefit, I do think that the present proposal operating 13 circuit breakers with respect to just the rest of the day 14 probably is ineffective just as price limits were ineffective 15 in the past with respect to Asian markets. So if I were 16 doing something, I would, one, focus on circuit breakers. 17 Second, when I speak circuit breakers, I really -- I would 18 avoid prohibitions if at all possible anytime. I would only 19 focus on quote-based tests. Third, I would think of at least 20 giving the flexibility to consider for more than the rest of 21 the day because I don't think otherwise you've changed really 22 the perspective of what's going on. 23 And I believe that if all those things done and the 24 last piece would be that whatever the Commission does, it 25 should set out the tools and require the industry to be in 18 1 place to respond to the various things you require and not 2 have additional tools in your tool bag. You should avoid 3 surprises. Whatever alternatives you have should be built 4 into these rules, should not be part of emergency rules and 5 should operate from that standpoint. 6 And with that, thank you very much for this 7 opportunity to be here. 8 MR. BRIGAGLIANO: Thank you, Rick. 9 John Kozak. 10 MR. KOZAK: Good morning. My name is John Kozak. 11 I'm the chief financial officer for Park National 12 Corporation, and again, I'm delighted to be here and thanks 13 for inviting us. We're headquartered in Newark, Ohio. We're 14 a 7-billion dollar bank holding company that owns and 15 operates two banks. One bank, our lead bank, is $6 billion 16 in size with 12 banking divisions, and it operates 127 17 offices in Ohio. Vision Bank is headquartered in Panama 18 City, Florida, and it's 900 million in assets with 18 offices 19 across the Panhandle of Florida and into Alabama. 20 These banks are traditional community commercial 21 banks. They offer deposits, loans, cash management, 22 retirement and wealth management products and services to 23 their local communities. Park is traded on the New York Stock 24 Exchange, the AMEX part of the New York Stock Exchange. We 25 have close to 14 million shares outstanding and a market 19 1 capitalization of about 925 million. We're largely retail 2 owned. Seventy percent of our shareholders are retail, 30 3 percent institutional. 4 I'm pleased to be here today representing the 5 American Bankers Association. The ABA brings together banks 6 of all sizes and charters into one association. ABA works to 7 enhance the competitiveness of the nation's banking industry 8 and strengthen America's economy and communities. Its 9 members, the majority of which are banks with less than 125 10 million in assets, represent over 95 percent of the 11 industry's 13.6 trillion in assets. Pretty amazing, 95 12 percent all in one trade association. 13 The ABA recognizes that short selling can be a 14 legitimate and important financial tool operating as a 15 mechanism for generating market liquidity, securing price 16 discovery and fostering corporate accountability and 17 responsibility. That said, many ABA members both large and 18 small, believe that some short sellers may be taking 19 advantage of the uptick rule's absence and as a result, 20 banks' stocks may be experiencing excess downward price 21 pressure. The price volatility generated by this short 22 selling activity can be different than the underlying 23 fundamentals of these banks. These disruptions are 24 especially problematic for banks as our customers frequently 25 and incorrectly equate significant drops in bank stock prices 20 1 with safety of bank deposits. 2 Our markets today are much different than July 2007 3 when we were working -- the Commission was working on 4 determining whether to eliminate the uptick rule. Another 5 way to say that is, boy, how can you test that unless you go 6 through a bear market? And so I would be a proponent of 7 re-instating some form of the uptick rule. 8 The Commission has proposed two approaches, two 9 restrictions on short selling: a price test that would apply 10 on a marketwide and permanent basis or a circuit breaker 11 which would apply only to a particular security during severe 12 market declines. The ABA is on record in support of 13 reinstating the uptick rule in some format. 14 Park National supports a price test based on the 15 national best bid. We think that that would be easier to 16 implement and more practical given the changes in today's 17 market and would be fine. Why don't we test that and put 18 that into place? 19 On the subject of today's roundtable, the ABA does 20 believe that more should be done to restore investor 21 confidence in our markets. First, the Commission should make 22 permanent Rule 204(T) with its locate and hard closeout 23 requirements which have had a beneficial impact on short 24 interest volumes. 25 I provided you with a handout that really shows 21 1 that in detail, and really the point of that handout was to 2 show you that on September 17th when you instated the 3 emergency rule, we had an immediate and significant impact on 4 our volume. Our short interest at Park was at 1.6 million 5 shares, and in two weeks, it went down not in orderly fashion 6 by a couple hundred thousand shares. Obviously, folks that, 7 you know, had probably sold the shot naked, sold our stock 8 naked. And their price went up 32 percent. Okay? Things 9 like that bother investor confidence. That's not a good 10 thing. 11 So from being in, you know, the Midwest working at 12 a small bank, that's not a good thing at all. We certainly 13 lose investor confidence when things like that happen. So 14 I've included data for you to kind of look and see what our 15 short interest volume has been over a period of time. But 16 there's times when our stock has gone down in price 17 significantly without us announcing any news. And again, we 18 have a retail shareholder base of 70 percent, and the volume 19 is being dictated by changes in the short positions. We're 20 not for that. So I'd like to say that. 21 Continuing on, the ABA also would say second, more 22 aggressive enforcement of short selling regulations should be 23 undertaken in order to root out manipulative short selling 24 activities. Third, the Commission should consider whether 25 there is some manner in which short sales information could 22 1 be made available to the public on a delayed basis. 2 Overall, again, I'm thrilled to be here 3 representing the ABA and thanks for having me. 4 MR. BRIGAGLIANO: Thank you, John. 5 Dan Mathisson. 6 MR. MATHISSON: All right. Thank you for having 7 me. I run electronic trading at Credit Suisse. To date, 8 we've been responsible for approximately 11 percent of U.S. 9 equity volume. 10 There's a silver lining in the events that we saw 11 in the fall of 2008 in that during the 14-day short sale ban, 12 we had an opportunity to look at what the world would look 13 like if there was no shorting allowed at all. And it wasn't 14 a pretty world. We saw bid-ask spreads dramatically widen. 15 We saw volume dramatically decline, and we saw volatility 16 shoot up. 17 Now, a lot of people don't really get it. You 18 know, out on Main Street, they don't understand why taking 19 away shorting does all this damage to the market. And the 20 fact is that almost no professional traders are net short. 21 Almost all professional traders run long-short strategies. 22 In August of 2008 right before the credit crisis, 23 the Credit Suisse Tremont index showed 0.6 percent of hedge 24 funds were net short. All right. So in other words, almost 25 no funds are net short. They're always long short. For 23 1 every 100 shares they're selling, they're buying 100 shares 2 in something else. What this means is that when you restrict 3 short selling whether by banning it or just by slowing it 4 down, you're also -- you're also taking buys or slowing down 5 buys on the other side, almost in a one to one ratio. For 6 every 100 shares you pull out on the sell side, you pull out 7 100 shares on the buy side. All right? So if you stop the 8 shorting in one bank, you've stopped somebody from buying 100 9 shares of some other bank. And that's the reality of how the 10 market works today. 11 Taking this into account, we believe that 12 restricting shorting is just going to pull both buyers and 13 sellers out of the market as we saw during the ban, resulting 14 in less liquid markets. And for that reason, we think any of 15 the all stocks all the time rules -- you know the up bid or 16 what we call uptick classic are disastrous policy in that 17 what they do is -- is they -- we look at those as essentially 18 chemotherapy for the market in that what you are doing is in 19 order to attack a few cancerous cell within the marketplace, 20 you are sickening the entire market body and you're causing 21 damaging side effects. 22 On the other hand, circuit breakers are highly 23 targeted to just the problem stocks. Circuit breakers go in 24 and essentially they're more like surgery where you go, you 25 remove just the diseased cells while leaving the rest of the 24 1 body untouched and unharmed. Within the circuit breakers, we 2 think that the halt is the best. We think that that would be 3 the most effective in that it gives management an actual 4 breathing space away from shorts, although it is worth 5 pointing out that like in all the restrictions but 6 particularly in the halt model, exemptions for hedge 7 derivatives, ETF and convertibles traders are absolutely 8 critical. 9 Another reason that we like the circuit breaker 10 halt model is it's much, much cheaper and easier to 11 implement, understanding how technology works within Wall 12 Street. We estimate that within the broker-dealer side, any 13 of the rules that require centralization of data and 14 sequencing of data, whether it's sequencing ticks or bids is 15 irrelevant. If you have to sequence the data, it becomes a 16 huge project and it's something Wall Street doesn't currently 17 do. And we estimate it would take approximately 12 months 18 for a bulge bracket broker-dealer like ourselves to implement 19 a tick or a bid rule. On the other hand, we believe a 20 circuit breaker that doesn't require sequencing of bids or 21 ticks would take approximately three months for us to 22 implement. 23 So for all these purposes, we believe that circuit 24 breaker halt with appropriate exemptions is the easiest to 25 implement, is the least damaging to liquidity and would be 25 1 the most effective at giving management breathing space and 2 for these purposes, we support the circuit breaker with halt. 3 MR. BRIGAGLIANO: Thank you, Dan. 4 Mike McAlevey. 5 MR. McALEVEY: Thank you for the opportunity to 6 participate in the panel today. I'm pleased to speak on 7 behalf of General Electric. 8 I should say at the outset that none of my remarks 9 should be understood to oppose short selling generally. 10 While short selling serves functions in the marketplace like 11 price discovery and liquidity, it should not be allowed to be 12 misused by some professional traders to take advantage of the 13 shareholders of public companies by driving the prices of 14 their securities to abnormally low levels. 15 My remarks today are aimed at helping the 16 Commission reach a better balance among market functions and 17 other important considerations like investor confidence in 18 the fairness of financial markets, prevention of manipulative 19 or abusive practices and transparency. Short selling should 20 simply not be allowed to be used solely or in connection with 21 other actions to improperly destroy the financial reputation 22 of companies and create systemic risk in the global financial 23 system. 24 Numerous investors both large and small have 25 brought to our attention concerns about large, well 26 1 capitalized and opaque trading strategies that have a 2 significant effect on stock prices, abusive short selling 3 including naked short selling and false and misleading 4 rumormongering sometimes used in connection with these 5 strategies. 6 The absence of some regulatory response to these 7 types of concerns creates a significant confidence issue, in 8 our view. We support the Commission imposing some form of 9 price test or circuit breaker test to help restore this 10 confidence. Among the tests that the Commission is 11 considering, although it's close, we would support a 12 marketwide permanent short sell test based on the national 13 best bid. And we think that this is a better test for a 14 number of reasons, many of which are identified in the 15 Commission's proposing release: concern about stigmatization 16 of stocks that have hit the breaker, the possibility of the 17 magnet effect which increases downward pressure on stocks if 18 the value approaches the breaker level and in addition, that 19 it may allow the manipulative trading strategies to continue 20 until the breaker limit is actually hit. 21 We think that the bid is better than the sale price 22 because the consolidated bid is updated and reported more 23 frequently than trade data and it is widely regarded as a 24 more accurate reflection of the current price of the 25 security. 27 1 Equally important, I don't think that we can talk 2 about reforms in the area of the uptick rule without also 3 saying a little bit about disclosure. In addition to what 4 I've already talked about, I would urge the Commission to 5 think seriously about some additional disclosures, either 6 through amendments to Form SH or in some other way. First, 7 to consider requiring disclosure about the timing and amount 8 of any long position or an economically short position that a 9 reporting person has established in a reference entity's 10 credit default swaps. Large short sales accompanied by CDS 11 purchases are a powerful tool for manipulative conduct 12 because of the responsiveness of equity prices to changes in 13 CDS spreads. This combined trading strategy should be 14 subject to as rapid reporting as possible in order to ensure 15 that the marketplace is fully informed and participates, can 16 take the short sale CDS relationship into account in their 17 investing decisions. 18 Second, the Commission should consider imposing 19 disclosure requirements about the timing and type of 20 derivative positions that are the equivalent of a short 21 position in the stock. And third, without getting into 22 motive, I think that the Commission should seriously consider 23 imposing disclosure requirements of whether a reporting 24 person directly or indirectly through any contract 25 arrangement or understanding has or shares a short position 28 1 in the stock, a short equivalent position in the stock or a 2 long position in the CDS. If the Commission determines that 3 it doesn't already have the authority to impose any of these 4 disclosure requirements, I would ask the Commission to seek 5 it from Congress. 6 Thanks. 7 MR. BRIGAGLIANO: Thank you, Mike. 8 Now, Justin Schack to wrap up the opening 9 statements. 10 MR. SCHACK: Thank you and good morning. 11 Rosenblatt Securities is pleased and honored to participate 12 in this roundtable. Rosenblatt executes trades for managers 13 of mutual funds, pension funds, hedge funds and other 14 investment firms. Institutions such as these manage the vast 15 majority of U.S. retail investors' equity assets. The 16 Investment Company Institute, for example, estimates that 17 some $3.3 trillion is currently invested in equity mutual 18 funds alone. 19 Changes to market structure arising from regulation 20 often profoundly affect the ability of these institutions to 21 efficiently execute securities transactions. This can in 22 turn impair their investment returns and by extension, damage 23 the capital formation process and the health of our economy. 24 It is therefore vitally important to consider whether any 25 proposed short sale regulations will bring demonstrable 29 1 benefits that will outweigh other potential adverse impact. 2 Rosenblatt also analyzes and advises clients about 3 market structure. In this role, we have gained a unique 4 perspective on how competitive dynamics and regulation affect 5 market quality. Of particular importance to today's 6 discussion, we believe, is the role played by so-called high 7 frequency traders. We estimate that high-frequency firms 8 account for as much as two-thirds of consolidated U.S. equity 9 volume. These firms have become the market's new primary 10 liquidity providers, supplanting NYSE specialists and 11 traditional NASDAQ market makers. 12 High-frequency firms and other market participants 13 regularly sell securities short not to express a fundamental 14 view but rather as part of complex, computer-driven 15 strategies, most of which involve corresponding or related 16 long positions in the same or comparable securities. We 17 believe that the reinstatement of a price test for short 18 sales would cause high-frequency market makers to curtail 19 their trading activity by as much as 20 percent. This would 20 result in wider spreads, less liquidity and higher net 21 transaction costs for the investing public. 22 To be sure, other forms of short selling can be 23 abusive or manipulative. Naked short selling can contribute 24 to artificially large decreases in the prices of securities. 25 However, the Commission's recently adopted rules regarding 30 1 fails to deliver appear to have effectively neutralized naked 2 shorting as a market problem. 3 Most importantly, we must take care to ensure that 4 politics do not get in the way of an intellectually honest 5 assessment of this issue. Congress is pressing for action to 6 do something to boost investor confidence. However, 7 considering the recent equity market rally and the growing 8 sentiment that the worst of the crisis is behind us, the need 9 for an immediate boost of confidence does not appear acute at 10 this time. We believe that imposing regulations that could 11 do more harm than good simply for optics' sake is not the 12 right thing to do for our markets and for our economy, 13 especially during a time of crisis. 14 If, however, the Commission determines that new 15 restrictions are in order, we believe that a price test 16 triggered by circuit breakers would have the fewest adverse 17 effects on market quality. And we strongly urge the 18 Commission to consider exemptions to any new short sale 19 restrictions that would permit high-frequency traders and 20 other market makers to continue to provide the same robust 21 level of liquidity to the market that they do today. 22 Once again, thank you for seeking the opinions of 23 market participants regarding this highly charged and complex 24 issue and for the opportunity to participate today. 25 MR. BRIGAGLIANO: Thank you, Justin. 31 1 And, panelists, we deeply appreciate the very 2 thoughtful statements that you've made. And now, we'll take 3 questions from the Commission. 4 COMMISSIONER CASEY: Good morning. And I want to 5 thank all of you for your very good statements. I'd like to 6 start off with asking you all to give us your best view of 7 the effectiveness of the old uptick rule. I know Rick and 8 others have spoke a little bit about the amount of focused 9 study that the Commission engaged in in looking not just 10 internally at some of the analysis through a pilot program 11 but also I think 13 other empirical studies and data that 12 we've collected that ultimately supported the Commission's 13 decision to remove the uptick rule. 14 But I'm really interested in, one, whether or not 15 you believe the uptick was effective given what we know about 16 the changes in the function and structure of our marketplace 17 over the past several years and then if you don't believe it 18 was effective, then why do you think there were so many calls 19 for us to reinstate the old rule? Begging the question of 20 the fact that we're actually proposing different price tests 21 here that would be effective in this marketplace. And if 22 that is the case that it wasn't effective and folks were 23 calling for us to reinstate it knowing that it wasn't 24 effective, how would that actually increase investor 25 confidence? And any data or analysis you can point to would 32 1 be helpful. 2 MR. CRONIN: I'm sure glad I'm sitting at this part 3 of the table. 4 Let me see, where to start with that one? I think 5 it's very important to understand the context of the uptick 6 rule both in its 1939 to 2007 state and what it would have 7 meant to have been in place today. 8 Clearly, I think any discussion about that has to 9 take into consideration the fact that the market structure 10 has changed dramatically in the last two years. There are 11 now today in the U.S. over 40-something destinations that 12 people like Brian and I can send our order flow to. So I 13 don't think personally that the regime of enforcement was 14 ever particular strong with respect to the uptick rule, but I 15 think it would be nearly impossible given the state of 16 development of the financial markets today. 17 So I think why people would come to you and say 18 let's repeal this, let's be honest. We're all irritated and 19 aggravated that the markets had the negative move that it 20 has. Many of us who do this for a living understand it, but, 21 you know, sort of at the emotional level, it still bothers 22 us. 23 But if there's something that I've learned over the 24 past 20 years in this business, it's this: The best 25 decisions I make on behalf of our clients, shareholders, et 33 1 cetera, is ones that are based on facts, not ones that are 2 based on emotion. And my suspicion is that the emotion of 3 this makes it very easy for anybody to identify either short 4 sellers or the promulgation of a new rule against short 5 sellers as this elixir that cures all the things that are ill 6 in this marketplace. 7 I submit to you that this would not cure all ills, 8 that bringing back in particular the uptick rule based on the 9 last sale would not be effective, could not be, I think, 10 governed and regulated and enforced appropriately to give 11 anybody at any level confidence. So to bring that back, how 12 would this discussion or what's the sort of virtue of 13 discussion about this process, bringing the confidence to the 14 investor? And I hope it's this. People realize that we 15 really care about this. 16 Interest in this issue is going to wax and wane, we 17 understand, based on where the Dow closes on a particular 18 day. To the extent that things have been better for the past 19 couple of weeks -- make it seven or eight now -- probably is 20 not as salient as it was, you know, six, seven months ago, 21 certainly in November of last year and more recently in March 22 of this year. But as their interest wanes, our interest 23 remains. We want to create structure that facilitates the 24 long-term advancement and development of efficient, you know, 25 very sort of open and, you know, transparent market structure 34 1 that by nature of its implementation has regulation in a 2 limited perspective, but the regulation that's there is 3 enforced and it's enforced with vigor. 4 To the extent that 204(T) is made permanent, to the 5 extent that 10b-21 is really enforced and teeth are put on 6 that and it's really sort of held to the level of standard 7 that it should be, I think a lot of the things that need to 8 be in place for this short selling sort of problem that we 9 have today or at least had recently can be addressed. That's 10 my perception of the issue. 11 MR. CONROY: Kevin, I would say you did pretty well 12 for leading off there, by the way. 13 So the first question, Commissioner, I believe was 14 do you think -- do we think the uptick rule is effective. 15 And I would just point towards the pilot program and the data 16 that came out a few years ago during -- I think the quote was 17 "normal market conditions." That was there no evidence that 18 there was an effect. 19 And I think we have to remember in the context of 20 the 1933 market in which the original uptick rule was put in 21 place was a very different market than the one we're in 22 today. I think it was about ten years ago the New York Stock 23 Exchange, volume of New York Stock Exchange listed stocks 24 first dropped below 90 percent. Very little of the volume 25 was done electronically and a large -- I think it was 15 35 1 percent of the volume was done by New York Stock Exchange 2 specialist firms. And if we all recall, that was a day and 3 age where there were two places to buy and sell stocks. 4 There was essentially the NASDAQ marketplace and the New York 5 Stock Exchange marketplace. And with respect to the New York 6 Stock Exchange, stocks traded by human beings who stood 7 around a post. 8 I don't need to remind anybody by looking at CNBC 9 in the visitor's lounge that that rarely takes place today. 10 And to roll back to a time that had regulations around a very 11 different marketplace, I think is a dangerous thing for us as 12 a industry to consider. And as some of my colleagues on the 13 panel pointed out, with 70 or so percent of the market 14 liquidity being provided by high-frequency trading firms, 15 there is a call for the realization that the marketplace of 16 today is, in fact, different than it was ten years ago. 17 I think why would investors be clamoring for a 18 go-back to earlier times, I think it's because of the 19 complexity of the issue. I think it is the topic du jour. 20 It is something that people have heard about. It is 21 something that is potentially not as well understood on Main 22 Street as it is in other places in the industry. And I'm 23 sure that if -- you know, I may have an opinion on the use of 24 nuclear power as a clean energy source, but I couldn't speak 25 intelligently about its safety. 36 1 So I think with respect to what the Commission's 2 doing in holding this roundtable and asking for input from 3 market participants and academics, I applaud you for your 4 research and your time. Thank you. 5 MR. KETCHUM: I think, Commissioner, that's 6 absolutely the right question. I think it's -- and I agree 7 with virtually everything said before with respect to how the 8 market's changed, how much of this perception relates to 9 electronic trading and the rest. 10 But I think actually to answer the question fairly, 11 you have to define what you mean by effective. If you mean 12 by effective did the tick test have an impact in people 13 engaging in substantial short selling, did the tick test have 14 an impact with respect to controlling volatility and the 15 rest? I don't see any academic indications of that. 16 If you define -- if you sort of identify reflecting 17 back on last fall and define quite narrowly, does a tick test 18 have some investor confidence or active trader confidence 19 even with respect to in a very short period of time in 20 exceptional circumstances as to whether shorts can engage in 21 consecutively hitting bids on a continuous basis without 22 break when there -- when essentially most buy interest has 23 withdrawn from the market, I guess my answer to that 24 question, I think that's, to me, the only correct question 25 because I don't think you should regulate beyond that, and I 37 1 don't know. But I think that there -- I think as reflected 2 here to some degree, I think there is some likelihood that 3 there is an encouragement for withdrawal from a buy side 4 because of the concern of the ability for short sale to take 5 the market down. And I can't measure that, and I certainly 6 can't demonstrate it. I do believe in those circumstances 7 activity and the cause and effect of consecutive hitting of 8 the quotes is somewhat more predictable as was made by Mike, 9 a point made very well, I thought, by Michael and it's worth 10 considering. 11 But I that's, to me, the only question, whether 12 there is some social usefulness in providing some comfort to 13 the market that shorts alone can't engage in continuous quote 14 hitting on an uncontrolled basis. Having said all that, the 15 tick test is a lousy way to do that, a particularly lousy way 16 to do that when any tick no matter what the size can allow 17 for any trade. That simply makes no sense, and clearly, your 18 proposals are much better than that. 19 MR. KOZAK: On this panel, I think I kind of 20 represent Main Street. You know, I live in a community that, 21 you know, has 40,000 people, you know, et cetera. Very much 22 in favor of a new uptick rule being reinstated. You know, 23 national best bid, that's absolutely fine. But in more 24 thinly traded stocks, to say that an uptick rule doesn't have 25 an impact is absolutely, in my opinion, ridiculous. 38 1 You know, days to cover the short position in our 2 stock has been high as 40 days. Does that make any sense? 3 Absolutely no. So volume in the short interest position can 4 absolutely influence our price substantially, and that 5 doesn't make any sense. There should be, in our opinion, for 6 stocks that aren't as liquid as, you know, 7 national-market-type stocks, absolutely we need an uptick 8 rule. 9 Then the other thing, my other point would be, yes, 10 everybody says you did a great job on the study -- I applaud 11 you for that -- back in 2007, but, you know, when do you want 12 an uptick rule? You want it in a bear market. Did we study 13 it in a bear market? Well, we didn't have a bear market at 14 the time, but that would be my other comment. Thank you. 15 CHAIRMAN SCHAPIRO: Could I just jump in, Jamie, if 16 you don't mind? 17 I recognize that we don't want to react to every 18 market up and down, but we do want to make sure we're being 19 highly responsive to sort of fundamental changes in our 20 market. And this discussion leads me to wonder whether even 21 if the old uptick rule perhaps was -- had become pretty 22 ineffective because of market structure changes, because of 23 high-frequency traders that you all talked about changes and 24 therefore, the old uptick rule was perhaps not doing what we 25 had hoped it would do, doesn't that suggest that maybe it 39 1 should have been replaced with something else that was more 2 effective in the new market structure? And that that's 3 really what we're really trying to explore here is given all 4 of these other changes, what might replace the old uptick 5 rule or what else might act as a governor on abusive short 6 selling in these new markets with high-frequency traders, 7 with 40 different places to send your orders, with a great 8 amount of fragmentation that we have today in the technology 9 that allows for this tremendous speed of execution. 10 MR. MATHISSON: High-frequency trading, just to be 11 clear about it, typically, the people who are doing that go 12 home flat at night. So for every -- so, you know, so they're 13 not the guys who are causing, you know, John's stock to have 14 40 days of short interest outstanding. They're buying it 15 back the same day that they're shorting it. 16 CHAIRMAN SCHAPIRO: I'm not trying to blame 17 anybody. All I'm trying to say is we've had all these 18 changes since the old uptick rule went away. To me, that 19 doesn't necessarily argue that you don't need something. It 20 just perhaps argues that you don't need the old uptick rule 21 back. That was the only point I was trying to inquire about. 22 MR. MATHISSON: To answer both of those questions, 23 both of that as well Commissioner Casey's earlier question, 24 you know, I believe the uptick rule was completely 25 ineffective, particularly since decimalization. It slowed 40 1 you down slightly to put on your positions. It also allowed 2 you to put out -- you were able to show big orders and then 3 long sellers would jump ahead and hit the bid. It didn't 4 necessarily change the price or lead to any upward price bias 5 which was, I guess, what people are trying to get to. So we 6 believe the rule was ineffective. We think that a circuit 7 breaker with trading halt could potentially be an effective 8 replacement that would fit in nicely with today's market. 9 MR. McALEVEY: If I could just -- I think it's 10 going to be difficult to prove one way or the other whether 11 the reinstitution of something is going to be effective at 12 addressing some of these abuses unless we actually try it 13 either on a pilot basis or some other temporary basis. 14 But it seems to me just as a matter of logic and 15 intuition that imposing some sort of marketwide test based on 16 the consolidated best bid is going to have some slowing 17 effect on very rapid declines in securities, likewise would a 18 circuit breaker test. 19 Addressing this issue of investor confidence, I 20 don't -- I would encourage the Commission to look 21 holistically at this issue because whatever the Commission 22 decides to do with respect to the uptick test, I think that 23 there are a number of other pieces of the puzzle that need to 24 come into place in order to restore investor confidence in 25 the marketplace. None of them alone are probably going to be 41 1 effective. Maybe all of them together may get closer to 2 being effective at addressing concerns. 3 But in terms of academic studies and statistical 4 analyses and economic analyses, when you step back and you 5 just talk to the long investors in these companies -- and I'm 6 not talking about small retail investors. I'm talking about 7 very significant large institutional investors. The 8 observation can replace statistical and academic analysis, 9 and the observation is with respect to some companies that 10 there has been rapid deterioration in stock price accompanied 11 by a number of other things and then suddenly a rapid 12 restoration of value. And that smacks of something going on 13 other than just ordinary operating markets. Thanks. 14 MR. SCHACK: I agree with much of what's been said 15 about whether the uptick rule was effective, why people are 16 asking for it to come back. I won't repeat some of those 17 arguments. 18 I think we need to ask ourselves the question what 19 is the problem we're trying to solve here, when can short 20 selling be abusive and is, you know, shorting on a down tick 21 or a down bid, does that constitute abuse? Does naked short 22 selling constitute abuse? I think it's pretty clear that 23 naked shorting is abusive. You shouldn't be able to sell, 24 you know, more than the shares that actually exist. And the 25 Commission has done a good bit to make sure that that doesn't 42 1 happen anymore. 2 My worry is that when you think about market 3 quality, you think about firms like Brian's and Kevin's being 4 able to put on positions for folks in small-town America who 5 live in those 30,000, 40,000 population towns who are saving 6 for college or retirement through mutual funds or through a 7 pension fund that has money invested with a long-short hedge 8 fund. What is going to be the impact on market quality if we 9 limit short selling and the folks who are providing liquidity 10 in this market now either -- they won't cease to do that, but 11 they will do it in a shrunken form. And that will make it a 12 lot more difficult for firms like Brian's and Kevin's to get 13 good returns for Main Street investors. 14 MR. CRONIN: And I would just add again, I don't 15 think that we can minimize the intersection of all the events 16 that are taking place today. Again, market structure has 17 changed precipitously. We have volatility that's been 18 introduced in the market that we hadn't had for some time. 19 Certainly, we've had problems with financial companies and 20 the economic cycle in general. And all these things are 21 confluencing at one singular time which we really haven't 22 ever seen the planets align like this, certainly in many 23 generations. 24 And so to the extent that that's happening, it 25 makes it very difficult for any of us to say to you, Chairman 43 1 Schapiro, and any of the other Commissioners, ah-ha, if you 2 just did this, we would find this magic answer and we could 3 all go home early today. The fact of the matter is that 4 because we're in the infancy stage of development of a very 5 complicated, complex U.S. financial market today, I can't say 6 for sure what effect any of the rules would have either 7 intended or unintended. And that's my concern. 8 I look at what I believe is the aberrant, you know, 9 bad, egregious behavior in the marketplace, and it continues 10 to seem to me to fall in two dimensions. One is this naked 11 short selling that we've been talking about for some time 12 which I think has been effectively dealt with through 204T 13 which needs to made permanent. The other is this 14 manipulative short selling type which does include, as the 15 gentleman from GE suggests, CDs, CDSs and other equity swaps 16 and different ways to represent similar exposure to short 17 selling. 18 So in order to effectively address that, you do 19 have some rules on your book that give some teeth to 20 enforcement of those particular dimensions of anti-fraud and 21 manipulation. But clearly, there needs to be another step to 22 that which is regulating the other markets. If I had more 23 clarity on that, then I would say to you I would be 24 comfortable really addressing those particular dimensions of 25 the short selling market with those steps. 44 1 I don't view the things that these external 2 liquidity providers and 99 percent of the other people who 3 are laying out shorts in a given day as egregious. 4 Generally, it's hedging behavior. Generally, it's associated 5 with some of our colleagues saying other buys and other 6 things that go on, and those things promote liquidity and 7 effectiveness and transparency. Those are good things for 8 the marketplace. 9 If we got to the point where we were absolutely 10 certain that something had to be done, then I'd say let's 11 take an incremental approach, let's really study the issue 12 just as we did back in 2003 and '4 with Reg. SHO and all the 13 different iterations that we had with that and make sure that 14 we don't come up with something that makes the current 15 condition, the 95 percent of the time that we're 16 participating in the market, beholden to that 2 or 5 percent 17 time of the market that we're really concerned about. I just 18 think that's the bad way to approach this problem. 19 MR. BRIGAGLIANO: Commissioner Walter. You had a 20 question? 21 COMMISSIONER WALTER: Thank you, Jamie. 22 As I listen to you, all of you speak, perhaps this 23 is a Pollyannish view, but I hear a tension between two 24 positive things: a desire to have the price of a particular 25 security be what it should be and really represent the value 45 1 of the underlying company and a desire not to interfere with 2 broader trading strategies which involve but don't center 3 around any particular security. 4 I wondered if you have a comment on if you'll 5 accept that as an assumption for a moment or perhaps want to 6 comment on it and whether that tends to lead towards a less 7 rather than more extreme reaction perhaps to a circuit 8 breaker kind of an approach. And if that's correct, whether 9 the effectiveness of that approach in turn is undermined 10 either by what has been called the magnet effect or by a 11 buildup of short selling interest that builds during the 12 period when whatever follows the circuit breaker is in 13 effect. 14 MR. MATHISSON: There are several academic studies 15 that have been done on the magnet effect. Circuit breakers 16 aren't new to the United States market. We did have Rule 80A 17 which was the -- they call it the program trading collars and 18 Rule 80B which halt the entire market and are still in place 19 and halt the market in the event of a drop of approximately 20 10 percent. So it's not a new concept. 21 There have been a lot of studies around have things 22 been drawn, like do prices get drawn to the circuit breaker 23 price which is, you know, known as the magnet effect in the 24 academic studies. And the overall consensus is there isn't 25 any empirical evidence of this, and there are multiple 46 1 studies that have been done on the futures market and done on 2 foreign stock markets and they have not found evidence of 3 prices getting drawn to this price. And part of this is sort 4 of logical. Like as you're coming into the price, a trader 5 is -- the last thing a trader wants to do is go in and sort 6 of whack the bid, you know, sell the stock down to a level 7 where something is going to trigger, that all of a sudden 8 it's going to pull a whole lot of sellers out of the market. 9 Traders tend to go the other way and actually start pulling 10 back as they approach the price of the circuit breakers, and 11 that's what many of these studies have found which were cited 12 extensively in the comment letter that Credit Suisse wrote. 13 MR. BRIGAGLIANO: Commissioner Paredes. 14 COMMISSIONER PAREDES: Another way of getting at 15 the balancing act here is trying not to be over-inclusive, 16 right? You weed out the, quote, bad stuff. We can debate 17 what's bad. And you let the good stuff in, and we can 18 debate, I guess, at the margin of what's good. 19 One way of addressing that is a circuit breaker 20 type of a concept. Another way of addressing that is by way 21 of making sure you have the right exemptions in place. So I 22 was wondering if you could speak to the exemptions that were 23 offered up as part of the proposal or at least some of them. 24 But if you -- if there's anything in particular that you 25 think we may be missing that would be particularly important 47 1 to add to the mix. 2 MR. SCHACK: I think it would be critical to go 3 beyond what's laid out in the proposal now for things like 4 arbitrage and odd lots and include -- I'm not sure how you 5 would structure it, but include some sort of exemption that 6 would make sure that the folks that I mentioned in my 7 prepared comments, this high-frequency liquidity provider 8 segment of the market is able to continue providing liquidity 9 the way they do today. Some of these firms are registered 10 market makers, but many of them are not so that might pose 11 some difficulties. 12 If you do consider doing something like this, how 13 do you structure it? Is it just registered market makers or 14 is there some sort of cutoff for historical behavior, how 15 much liquidity a firm adds to the market, for instance, on a 16 daily basis. But I would urge you for the sake of market 17 quality to consider something like that in the way of an 18 exemption. 19 MR. MATHISSON: I would echo Justin's remarks. The 20 ability to sell short is absolutely critical to the 21 convertible securities and derivatives as well as the ETF 22 marketplaces. In the rule proposal that you've put out, it 23 does exempt registered market makers, but there is not an 24 exemption for players that are hedge players that are not 25 registered market makers. And we think that this is a flaw. 48 1 For clients -- for companies to be able to raise 2 funding through the convertibles markets, there does need to 3 be exemptions in place. We would suggest that you look at 4 whether or not companies are net long or net short in 5 aggregate, and if they're net long but they're putting on a 6 short to hedge against the convertible or a derivative 7 security, that that be considered the same as a long sale and 8 be short exempt. We think that this would maintain smooth 9 functioning in the convertibles and derivatives and ETF 10 marketplaces which is critical to the functioning of the 11 market as a whole. 12 MR. BRIGAGLIANO: You know, the need for exceptions 13 ties into the power of the restrictions and, you know, I 14 know, Dan, you pointed out the concerns about a ban with 15 respect to liquidity. But several of the panelists have also 16 said that the uptick rule wasn't effective at all, so I guess 17 I'd ask you, Dan, how much of a restriction are -- would the 18 price test proposed by the Commission be? You know, I hear 19 you talk about the need for exceptions, but how long -- how 20 much would short sellers be restricted by a price test at the 21 trading increment? 22 MR. MATHISSON: That's a good question, and you're 23 right. There is a bit of a Catch-22 in the argument if 24 you're saying uptick rule is ineffective and yet exemptions 25 are needed. And that's why the circuit breaker halt which is 49 1 a fundamentally different type of restriction, it's more 2 critical that the circuit breaker halt have exemptions than 3 any of the others. 4 As for the others, we believe it would also be 5 necessary to have exemptions at the end of the day. While 6 the uptick rule is overall ineffective at stopping shorts 7 from trading, at the end of the day it could stop them from 8 sort of getting that last trade-off which could be -- which 9 is critical for guys who are trading converts, ETFs and 10 derivatives so that they can flatten out their book and not 11 take overnight risk. So we would still ask for an exemption 12 in the final half hour of the day. 13 MR. KETCHUM: I would certainly agree that Dan's 14 right, that to the extent that you're dealing with the halt, 15 you need the exceptions from the standpoint of hedge and 16 convert in particular. 17 I would note two things, one from the standpoint of 18 the gravity issue noted earlier. I think you have to be very 19 careful with respect to gravity as to what the impact on the 20 other side. 80A, B with respect to the New York Stock 21 Exchange, basically, it had limited impact with respect to 22 how many participants would be operating and be held out of 23 the market and what they would be restricted from doing. 24 That's very different than, say, a price limit in Japan where 25 I'd suggest there was pretty demonstrably gravity in variety 50 1 of times in the last couple decades. And until economists 2 had looked at that, I would be careful to assume that a 3 circuit breaker halt had no gravity impact at all. 4 The last point I'd just make is a plea that from 5 the enforcement side of this that the Commission got it 6 absolutely right with respect to your broker-dealer provision 7 in providing the flexibility for just as in Reg. NMS for 8 broker-dealers to manage this upstairs at the time either the 9 quote or trade was reflected in their systems with a pattern 10 and practice approach. That's the right approach from a 11 regulatory standpoint to avoid technical violations. That's 12 the right approach from the standpoint of not generating an 13 excessive number of cancellations, and that would be the 14 other request generally would be I would stick to pattern and 15 practice approaches with respect to any of this. No one 16 cares whether one or two things makes a mistake one way or 17 another, and definitely hold on to the broker-dealer 18 provision. 19 MR. BRIGAGLIANO: Thanks for addressing that key 20 area. I think Commissioner Casey wanted to jump in. 21 COMMISSIONER CASEY: Thank you so much, Jamie. 22 I think John mentioned earlier the relationship 23 between extreme market volatility and investors' perception 24 of market quality and integrity and ultimately, their 25 confidence in the markets. And I was hoping that any of you 51 1 could share your experience over the past, you know, eight 2 months on what you believe the key drivers were of the 3 volatility that we've seen in the market, particularly short 4 selling. And then also help me understand whether or not or 5 how much of a role programmatic trading plays in driving 6 volatility. 7 MR. CONROY: Well, Commissioner, I think that we'd 8 all agree that this is a complicated issue and that, you 9 know, short selling is the cough that was -- that first tips 10 one off that they may be getting the flu. And clearly, as 11 pointed out, I believe, in your open comments earlier that 12 month, there were a confluence of events that took place to 13 drive the markets to the state of frenzy. 14 And clearly, as Commissioner Schapiro pointed out 15 in her opening statement, the integrity of the markets were 16 brought into question because of recent market events. And, 17 you know, this is the first time in 71 years that we've seen 18 a precipitous drop in the market in three months. So 19 clearly, it is bringing to the forefront CDS, its influence 20 in the market, highly levered, loosely regulated, very 21 non-transparent, heavily influencing the direction of 22 movements of -- especially financial stocks. You know, 23 clearly, all the issues around the housing bubble, et cetera. 24 So I think, you know, we can sit here and talk 25 about short selling and bans and plus tick rules, et cetera, 52 1 but I think the analogy I would draw is that if you live in a 2 hurricane zone, you put hurricane shutters on your windows 3 but you don't keep them closed and keep the light out for 70 4 years waiting for that hurricane to hit. And so, you know, 5 therefore, because the markets have worked so well and 6 because of the change in market structure, we advocate doing 7 nothing. But if something must be done, putting in a circuit 8 breaker which would help protect and ensure investor 9 confidence during those aberrational events. 10 MR. KOZAK: Again, I feel like I represent Main 11 Street. But I hear the folks on New York talk about all this 12 additional liquidity. Am I worried about that? No, I'm not. 13 So we have hedge funds that are doing shorts and loans and 14 creating this additional volume. On Main Street, do I care 15 about that? You know, it almost seems like the absence of 16 the uptick rule, in my opinion, gives them an opportunity to 17 make more money. And that's what I hear the other -- some of 18 the other panelists argue for. That doesn't play well. 19 MR. CRONIN: I would just add that people like 20 Brian are -- Brian and I do represent Main Street. Our 21 institutional shareholders are retail shareholders, are in 22 401(k) plans, they're in mutual funds, they're saving for 23 college, they're saving for retirement. So we do hear from 24 these people, and our objective is quintessentially to make 25 sure that the market structure supports their best interest. 53 1 As we look at this very complicated issue, it's not 2 hard to for us to figure out why there's more volatility. 3 You know, it's probably not equally too difficult for us to 4 get to a place where we can understand where there was such 5 limited volatility two years prior. In good times, you know, 6 lack of volatility begets a lack of volatility. In bad 7 times, volatility begets more volatility. 8 We had re-leverage or unleveraging of hedge funds 9 which created enormous amounts of selling pressure. We had 10 redemptions coming into mutual funds creating enormous 11 amounts of pressure. This had nothing to do with short 12 selling. These were long positions that we had to sell to 13 raise the cash that people were looking to bring. So it's a 14 vicious cycle, and it continues to sort of grow and 15 perpetuate. 16 Our goal is to say look, these events happen. How 17 do we figure out how we create a market structure which can 18 effectively address that but at the same time doesn't 19 compromise the time, the 90 percent, that 95 percent of the 20 time, whatever it ends up being, the vast preponderance of 21 time where these sort of aberrant market conditions are not 22 in place. 23 So we do care about Main Street. Our thought is 24 that the best way to protect Main Street is to continue to 25 have a high level of transparency, efficiency, effectiveness 54 1 in the capital markets. We just can't get to a place where 2 just closing your eyes and putting a rule in fixes that. We 3 would rather say let's take our time here. Let's have a 4 measured approach. Let's really figure out what the problem 5 here and address that systematically as opposed to just 6 thinking that the politicians or whoever in the world are 7 going to feel good about this because it's an emotional 8 response, not a fact-based response. Let's let the facts 9 figure out where we should go from here. 10 MR. BRIGAGLIANO: Chairman Schapiro had a question. 11 CHAIRMAN SCHAPIRO: Thanks, Jamie. 12 One of the virtues, I think, of the Commission 13 having put out so many different alternatives a couple of 14 weeks ago is that we've seen some fairly creative comment 15 letters come in that even give us some more choices and some 16 combination of choices of the things that the Commission 17 proposed. 18 So, for example, I'd love to hear your thoughts on 19 would a circuit -- if we were to go down a circuit breaker 20 route, when the circuit breaker kicks in, should it kick in 21 to a halt, should it kick in to a price test of some sort or 22 should it kick in to a pre-borrow requirement or should it 23 kick in to something else? And if you have any ideas on 24 that, I'd love to hear them. 25 MR. MATHISSON: If a circuit breaker leads to a 55 1 price test or leads to a tick test, we think on the 2 implementation side, it's no better. Wall Street is still 3 going to have to reprogram all the machines which is 4 significant work. As I mentioned, we estimate it's 5 approximately 12 months of work in that you still have to 6 centralize all the data from all the different exchanges and 7 put them into sequence and that's a big job. So we think a 8 circuit breaker followed by a price test that requires 9 sequencing is overly expensive and too difficult to 10 implement, you know, for a very limited extra benefit. 11 In terms of a circuit breaker leading to a halt or 12 leading to a pre-borrow requirement or leading to some type 13 of a test that does not require sequencing of the quotes, 14 it's -- you know, we think all of those would be -- could be 15 effective. 16 MR. KOZAK: We're in favor of a price test. As I 17 mentioned before, if it fell back to the circuit breaker, 18 then obviously, we would still be in favor of a price test. 19 MR. CONROY: I think, as Kevin said, as someone 20 representing 77 million accounts, many of whom are on Main 21 Street, not Wall Street, we would again advise to do nothing. 22 But in a case as your question asked, if there were a circuit 23 breaker, we would advocate either the bid test or the banning 24 and leave it really to the broker-dealer community to work 25 with the Commission to decide which was the most effective 56 1 and cost effective implementation plan for the markets. But 2 as a buy side representative, we would be agnostic as to 3 which solution was chosen. 4 MR. KETCHUM: I guess my concern with just 5 implementing a circuit breaker is the concern I expressed 6 earlier that I'm not at all convinced that a end of day 7 impact with respect to short selling gets you anywhere and is 8 particularly effective even if defined very narrowly what you 9 want. The difficulty then if you're exactly where you -- the 10 Commission was last fall with no alternatives in your case 11 because the industry won't be programmed to deal with a bid 12 test, won't be programmed to deal with anything and your only 13 choice will be to extend the circuit breaker which I think is 14 an absolute halt which I think is a bad idea. 15 If you're looking for an interim place to operate 16 while the industry makes technical changes, I don't design 17 the systems, but there is a place in between. And also you 18 could consider for a longer -- somewhat longer term halt or 19 action from a circuit breaker which is just require passive 20 orders that don't have them hit bids. I don't know any 21 industry firm that doesn't have a passive order capability 22 now as opposed to worrying about whether the bid is an uptick 23 or a downtick bid. 24 CHAIRMAN SCHAPIRO: When I say "halt," I don't mean 25 halting the stock, I mean halting short selling, just to be 57 1 clear. 2 MR. CRONIN: And I would just add to that, I think 3 again if we were really forced into doing something, our sort 4 of best option would be a circuit breaker based on a bid test 5 rule. 6 Dan, I am sort of curious how earlier you suggest 7 that when the halt was put in the place permanently for those 8 group of financial stocks, how bid-ask spreads widened, how 9 volatility increased and how volume decreased, how it would 10 be a good thing for whatever part of the day to do a halt for 11 the remaining part of the day. It seems to me -- and while 12 giving market makers and others some exemption, it seems to 13 me that if some groups get to do it, everybody should get to 14 do it because we believe that there are valid and good 15 reasons to do that from an investment perspective. 16 So we would present the circuit breaker under the 17 context of a bid-test rule and all the exceptions that have 18 been proposed by the Commission, we would propose to accept 19 as well. 20 MR. MATHISSON: You're right, Kevin, that a halt 21 would clearly widen bid-ask spreads and reduce volume in the 22 main, but it would limited. What's nice about the circuit 23 breaker is it's limited to just a handful of stocks on a 24 handful of days. Like at the proposed 10 percent level in 25 low volatility environment like we saw last spring, you would 58 1 have about one stock a day in the S&P 500 triggering. In a 2 very high volatility environment, you would still have 3 something like 30 or 40. The vast majority of stocks 4 continue to trade as normal. But clearly, when it does 5 trigger, it absolutely damages liquidity. I don't think 6 anyone is arguing against that. 7 Now, a price test that involves -- that's passive 8 in nature and doesn't require sequencing and knowing whether 9 or not the bid is an up or a down is relatively easy and 10 straightforward to implement, similar to a halt. 11 MR. BRIGAGLIANO: Thank you, Dan. 12 The responses to Chairman Schapiro's question marks 13 a great place to pause and end this panel because at 11:30 14 we're going to begin the next panel and get very deep into 15 particular kinds of price tests, operational issues and we'll 16 explore with a new set of panelists. 17 I'd like to thank our panelists for their 18 thoughtful participation, and we'll do a very quick change of 19 panels and begin promptly at 11:30. Thank you. 20 (A brief recess was taken.) 21 PANEL TWO - BID VERSUS TICK VERSUS CIRCUIT BREAKERS 22 MR. BRIGAGLIANO: All right. Today's second panel 23 is entitled "Bid versus Tick versus Circuit Breakers: A 24 Discussion of Short Sale Price Tests and Short Sale Circuit 25 Breakers." And we'll discuss operational implications of the 59 1 Commission's recently proposed approaches to short selling 2 regulation. Again, panelists will present brief opening 3 statement following which the floor will be open to the 4 Commission for questions. 5 And I will introduce our distinguished guests 6 starting from my left with Jeff Brown who is the head of the 7 Washington office of legislative and regulatory affairs for 8 the Charles Schwab Corporation and former senior vice 9 president and general counsel of Schwab Capital Markets. 10 Larry Leibowitz is group executive vice president, head of 11 U.S. markets and global technology for NYSE Euronext. John 12 Nagel is deputy general counsel and head of global compliance 13 of Citadel Investment Group. Jerry O'Connell is chief 14 compliance officer for Susquehanna International Group. Bill 15 O'Brien is the chief executive order [sic] of Direct Edge. 16 And Brett Redfearn is global head of liquidity and 17 algorithmic trading for JPMorgan Securities. 18 Jeff, do you want to start us off? 19 MR. BROWN: Thank you, Jamie. 20 Madam Chairman, Commissioners, I want to thank you 21 on behalf of the Charles Schwab Corporation and in 22 particular, Chuck Schwab, for holding this important 23 roundtable. We believe this is a very important subject. 24 Charles Schwab is -- represents about 7 million individual 25 customer accounts. We have mutual funds. We have -- we 60 1 administer 401(k) plans. So while it seems popular for a lot 2 of people to claim they represent Wall Street -- Main Street, 3 we're -- whatever street we're on, we do have a lot of 4 concern about investors. 5 And we believe that the reintroduction of an uptick 6 rule is necessary for the restoration of investor confidence. 7 Retail investors have been hit hard by the market turmoil of 8 the last 20 months. But in seeking more restrictions on 9 short selling, these investors don't want a guarantee that 10 their trades will win or always go up and they're not asking 11 for a bail-out. They want a market that's fair and that they 12 can trust and that when they put their order in, they feel 13 like -- and the positions they hold, they feel that they will 14 be treated fairly. 15 And what they will avoid is markets where that's 16 not the case, and that's where we are today. In the absence 17 of an uptick rule, our customers believe that the dramatic 18 bear raids of the last year were orchestrated to transfer 19 trillions of dollars of investor assets from customers to 20 manipulative short sellers. As a result, our customers are 21 staying away from equity investing and instead holding cash 22 in extraordinary levels. And I think this is a measure of 23 the concern they have for the marketplace. And in 24 particular, this is, you know, troubling for the future of 25 our capital markets and the capital raising function of our 61 1 markets. 2 So now, when we use the term "uptick rule," we 3 don't necessarily mean returning to former Rule 10a-1. 4 Rather, like the Commission's proposal, we support the 5 modified uptick rule or a bid test. We believe this rule 6 confronts the most serious element of abusive short selling 7 by preventing short sellers from hitting bids at successively 8 lower levels and thereby driving prices down. Limiting short 9 sellers' ability to create downdrafts in a security and the 10 resulting panic that it causes among long investors goes to 11 the heart of the purpose behind Rule 10a-1 as it was created 12 in 1938. At the time -- at the same time, a bid test permits 13 frictionless trading above the bid and eliminates the need to 14 track a tick or set a tick at an increment above the minimum 15 trading increment that we may have today. 16 Finally, a bid test will simplify the enforcement 17 of short sale regulation. Two years ago, I supported the 18 Commission's actions in eliminating uptick rules. At the 19 time, I believed and I think I've been proven wrong that like 20 many that our markets were so liquid that it would be very 21 difficult to manipulate them down. And I guess the problem 22 we faced was that we neglected the extraordinary 23 concentrations of capital that could step in and move 24 markets, move individual securities in a very rapid period of 25 time. And moreover, when combined, those pools of capital 62 1 were combined with unregulated products like CDS, you 2 developed a very toxic cocktail for particular securities and 3 that caused significant problems. 4 Secondly, the Commission noted in its order 5 eliminating the uptick rules that real-time exchange 6 surveillance systems could operate to capture manipulative 7 activity and detect and take action in a swift manner. Well, 8 I think that was incorrect. Detecting manipulative intent 9 from the haystack of trading data is as hard to find as the 10 proverbial needed. It's virtually impossible to determine 11 intent from data. So, you know, that takes a pronounced, you 12 know, examination by enforcement teams. 13 But having in place a prophylactic rule like a bid 14 test allows trading centers to program those rules into the 15 execution systems and thereby ensure compliance and be able 16 to detect violations relatively easily through exception 17 reports. We've done this with respect to Reg. NMS and to 18 other rules, and so even in complicated systems, we can 19 develop the electronic mechanisms to track the rules as they 20 get put in place. 21 So we believe doing so goes a long way to restoring 22 the market conditions that investors can have confidence in. 23 So I want to thank you and I look forward to the questions. 24 MR. BRIGAGLIANO: Thank you, Jeff. 25 Larry. 63 1 MR. LEIBOWITZ: Good morning. NYSE Euronext 2 appreciates the opportunity to appear before the Commission 3 today. NYSE Euronext matches more volume than any other 4 exchange complex or trading venue in the United States, and 5 our constituents include retail and institutional investors, 6 issuers and trading firms of all varieties. 7 We strongly agree with the Commission's decision to 8 reconsider reinstatement of price tests. The global 9 financial crisis has resulted in the loss of public 10 confidence in our financial system. The government has taken 11 a number of dramatic and unprecedented steps to restore that 12 confidence, and we believe that reinstatement of a price test 13 of some form could contribute to that goal. 14 While in a perfect world we'd have clear proof of 15 the effect of these tests, the real world doesn't always lend 16 itself to studying conditions that have not occurred before 17 or which occur rarely. With this in mind, we understand that 18 there's a balance to be struck between measures taken to 19 deter so-called bear raids or panic selling and their 20 consequences on the normal functioning of the most liquid, 21 transparent and efficient market in the world. Indeed, a 22 survey of NYSE issuers show that they too recognize the 23 importance of short sellers in price discovery and providing 24 liquidity. 25 Our general principle is that the stronger the 64 1 medicine, the more selective we should be in applying it to 2 the patient. And no, I didn't hear Dan's analogy before 3 that. This means that the harsher the price test, the 4 greater the need for exemptions and circuit breakers to 5 prevent deterioration of market quality such as when the 6 convertible bond market broke down during last year's short 7 selling ban. 8 The simplest of price tests, the old uptick test, 9 would be ineffective in today's market due to the improper 10 price sequencing caused by permitted reporting delays and the 11 potential for manipulation. The modified uptick rule as 12 defined in the SEC proposal is much more implementable aside 13 from sequencing issues as bids are posted in real time. 14 Firms already use them for compliance with their Reg. NMS 15 trade-through rule. 16 As a reminder, this so-called bid test was in 17 effect in the NASDAQ place in a market structure that was 18 substantially similar to today's, post-decimalization, 19 electronically interconnected markets, high participation by 20 high-frequency traders without detrimental effect. We 21 believe that this change implemented without circuit breakers 22 could be effective in dampening rapid or abusive short 23 selling to underlying market quality. 24 Our second preferred choice would be the passive 25 liquidity test, and it's described in a letter sent jointly 65 1 by three exchanges to the SEC shortly before the release of 2 the proposed rule changes. The passive liquidity test is a 3 more stringent requirement than either of the proposed SEC 4 tests requiring that short sales be done in a way that add 5 liquidity to the market and thus should be implemented only 6 in conjunction with a circuit breaker. The passive liquidity 7 test also has the advantage in not requiring the sequencing 8 of ticks or trades, thus greatly easing implementation. 9 Due to the complexity of either of these tests, we 10 recommend that they be implemented in the form of policies 11 and procedures consistent with Reg. NMS trade-through 12 compliance. Circuit breakers in general are confusing, 13 clumsy logistically and will increase the work required for 14 implementation. They introduce such arbitrary considerations 15 as at what level do we set the circuit breakers, whether 16 different priced stocks require different circuit breaker 17 levels, what to do if a stock retraces its losses intra-day, 18 how to handle new situations, how long a circuit breaker 19 should last. We discount the magnet effect previously 20 discussed as not being critical to implementation. 21 Finally, we believe that in no circumstance should 22 there be an outright ban on short selling as the damage done 23 to market quality as seen in increased spreads and volatility 24 is too substantial. Regardless of which approach the 25 Commission adopts, we strongly suggest that the exemptions 66 1 previously in effect during 10a-1 be reinstated in their 2 entirely. It is essential that any rule include a market 3 maker exemption as these participants need to make two-sided 4 markets to enhance market quality. 5 We suggest that the SEC use this opportunity to 6 better define the term market maker under the rule to account 7 for the different types of market participants active today. 8 We also believe it is important that the rule include 9 exemptions for such instruments as ETFs and ETNs which have 10 been provided no action exemption by the Commission in the 11 past and likewise, other exemptions such as options and 12 convertible arbitrage depending on which rule is ultimately 13 adopted. 14 Thank you for the opportunity to express NYSE 15 Euronext's views. We look forward to working with the 16 Commission and the industry to find a balanced solution to 17 this critical and highly emotional issue. 18 MR. BRIGAGLIANO: Thank you. 19 John Nagel. 20 MR. NAGEL: On behalf of Citadel Investment Group, 21 I wanted to thank the Commission and the staff for the 22 opportunity to participate in this important discussion. 23 On an average day, Citadel's funds and market 24 making businesses account for nearly 10 percent of U.S. 25 equity volume and nearly 30 percent of U.S. equity options 67 1 volume. Given this unique vantage point, we believe that the 2 U.S. capital markets can and should play a key role in 3 helping lead the recovery of our economy, but this can only 4 happen if market regulations promote fairness, transparency 5 and capital formation and are based on facts and data 6 carefully analyzed. 7 We fully support regulatory action that attacks 8 fraudulent and manipulative behavior like Rule 10b-21, the 9 Commission's recently adopted short selling anti-fraud rule. 10 We urge the Commission to refrain, however, from imposing 11 new, broad short selling restrictions without persuasive, 12 empirical data to demonstrate that such restrictions are 13 helpful or necessary. While you can always stop reckless 14 driving by prohibiting everyone from driving, such a blanket 15 approach does not achieve the right balance. 16 Last year, the global financial system was on the 17 verge of insolvency and collapse, and financial markets 18 experienced severe market declines and extraordinary 19 volatility. Some have sought to lay the blame on short 20 selling. These critics, however, ignore the direct 21 relationship between last year's extreme market conditions 22 and last year's once-in-a-lifetime collapse in economic 23 fundamentals. Blaming short selling for 2008 market 24 conditions is like blaming a thermometer for starting a fire. 25 As the Commission has recognized, short selling 68 1 provides many benefits to investors and the economy, so I 2 will not belabor the importance of short selling to liquidity 3 and price discovery. But I will discuss a critical benefit 4 of short selling that is not well understood by the public. 5 Most short selling occurs to enable investors to take long 6 positions. As we heard on this morning's panel, over 99 7 percent of hedge funds are net long. Short selling enables 8 investors to take long positions because it is an efficient 9 risk management to hedge some of the risks of making a long 10 investment. 11 We urge the Commission to carefully evaluate calls 12 for a short selling price test or circuit breaker in light of 13 relevant empirical data. Based on the record presented thus 14 far, we believe the outcome of the cost benefit analysis is 15 clear. There is no basis for adopting any new short sell 16 price test or circuit breaker at this time. 17 If the Commission ultimately decides to adopt 18 additional short selling restrictions, we believe they should 19 be narrowly tailored to eliminate collateral damage to 20 investors and the economy. In this regard, a bid test 21 circuit breaker could be such an approach if it includes 22 appropriate exemptions. Thank you. 23 MR. BRIGAGLIANO: Thank you, John. 24 Bill O'Brien. 25 MR. O'BRIEN: Good morning. I'd like to thank both 69 1 the Commission and the staff for the opportunity today to 2 participate on behalf of Direct Edge, the nation's and the 3 world's third largest stock market operator. 4 This debate is both timely and important, and the 5 Commission really merits a lot of praise for taking a 6 leadership position in fostering a constructive dialogue on 7 these issues. 8 Direct Edge believes that the best approach 9 promotes investor confidence without unduly impacting the 10 market liquidity, efficiency and transparency that's 11 consistently been proven to reduce investor costs over time. 12 Narrowly tailored remedies, not broad interference, strikes 13 the appropriate balance of ensuring market integrity without 14 threatening market quality. Applying these principles, the 15 current proposals to reintroduce new versions of bid or tick 16 restrictions on the ability to facilitate short sales would 17 appear to be a significant intervention in the mechanics of 18 trading that does little to combat truly abusive short 19 selling. 20 When effected properly, short selling facilitates 21 the operation of an efficient, liquid market. Many brokers 22 rely on the ability to sell short to facilitate their 23 customer business. Proprietary traders and trading firms, 24 many of which as we've heard a couple of times already, are 25 fundamentally long or market neutral, use short sales in the 70 1 effectuation of their trading algorithms, most of which are 2 completely automated. 3 Current market structure regulation encourages the 4 provision of liquidity from these sources by consistently 5 encouraging the interaction of trading interests and 6 providing minimal friction in the execution of individual 7 transactions. This approach has been validated through 8 consistently high trading volumes and narrow bid-ask spreads, 9 even in volatile market conditions. 10 Broad restrictions on the execution of short sales 11 without efforts to target improper conduct would have a 12 fairly certain and significant negative effect on market 13 operation without accompanying benefits. Studies of mature 14 equity markets evidence that restrictions on short selling 15 reduce trading volumes and increase transaction costs in the 16 affected securities without preventing substantial sharp 17 declines in asset prices. Thus the only likely outcomes of 18 implementing these rules before the Commission is a silent 19 tax on American investors through higher transaction costs 20 and reduced execution flexibility with benefits that are 21 illusory at best. 22 Each variant of the proposed rules have their own 23 unique deficiencies and potential unintended consequences. 24 Tick tests that restrict short selling based on national last 25 sale information ignore the reality that in today's 71 1 competitive market structure where executions occur in a 2 variety of venues, trades are not reported to the tape in 3 sequence. Thus any rule based on the tick would offer hollow 4 comfort given the randomness of such an approach. 5 The effectiveness of bid tests based on national 6 best bid information would be likewise limited given the 7 latencies inherent in the transmission, dissemination of 8 quote information. Circuit breakers that would trigger a 9 rule's application only upon a security declining by a 10 specified percentage could potentially make such declines 11 more frequent stigmatizing issuers and driving certain 12 liquidity out into the market even under normal conditions. 13 While the reform -- need for reform is real, the 14 remedy is not in these proposals. There are concrete steps 15 the Commission can take and in many cases has already taken 16 to address abusive short selling practices without distorting 17 market structure. 18 First, mandate regular and rigorous disclosure by 19 hedge funds and other money managers to appropriate 20 regulators of their short positions, including positions in 21 derivative and exotic securities where the investor would be 22 a direct beneficiary in the decline of the price of the 23 issuer's common stock. In this way, enforcement officials 24 would have a roadmap to locate perpetrators of abuse when 25 necessary, supplementing their ability to enforce rules 72 1 already on the books like 10b-21. 2 Second, continue the stringent enforcement of 3 requirements regarding the location, borrow and delivery of 4 securities sold short which have already served to 5 significantly reduce instances of naked short sales. 6 Supplemented by technological and other improvements in the 7 securities lending market, this can improve inventory 8 management and deprive bad actors of an environment that 9 would facilitate their schemes. 10 By taking this approach to short sale reform, 11 targeting abusive practices without restricting the liberty 12 of legitimate market participants, the Commission can 13 introduce real reform while preserving what already works. 14 Once again, I'd like to thank you all for the 15 opportunity to participate and look forward to any questions 16 you may have for me. 17 MR. BRIGAGLIANO: Thank you, Bill. 18 Jerry O'Connell. 19 MR. O'CONNELL: Thank you, Jamie, Mrs. Chairman, 20 Commissioners. Susquehanna is a large options and ETF market 21 maker, so a lot of my comments will be directed towards that 22 activity. 23 The general problem with abusive short selling was 24 largely fixed when you adopted Rule 204T. And now with few 25 exceptions, stocks settle in a reasonable period of time. 73 1 The fails problem went away mostly with only a small loss of 2 liquidity in the markets. That loss of liquidity was well 3 worth it. This one for these current proposals is not. Each 4 of the proposals would be very costly in this regard. Some, 5 of course, more than others. 6 The tick rule wouldn't work operationally, and I 7 think that that idea should be dismissed. The bid rule is a 8 little closer to working but would create too much needless 9 risk in stocks that are stable. The circuit breaker proposal 10 is the best of the bunch, but I think under the theory that 11 it'll only affect unstable stocks, we have to think about 12 that a little bit. 13 The circuit breaker proposals could create problems 14 for lots of stocks, both stable and unstable. On a down day 15 in the market, you could have more than 100 stocks hitting 16 the 10 percent threshold. Almost all of these stocks belong 17 to one or more indexes, indexes that trade options, ETFs and 18 futures. When we create difficulties in selling these 19 stocks, we also create difficulties in trading all of these 20 indexes at relative prices, the prices that the investment 21 community counts on when they send their orders into those 22 markets. 23 When the indexes become less efficient, so does 24 trading in all the stocks in the index. These proposals add 25 risk to the market which will hurt liquidity, just like when 74 1 you do things to take away risk, liquidity grows. In this 2 respect, I was glad to see market maker relief talked about 3 for riskless principal trades, but you forgot about the 4 options market makers. When an options market maker fills a 5 customer and lays off the risk in the stock market, that is a 6 real form of principal -- of riskless principal trading. 7 Options and ETF market makers are generally well 8 capitalized and very efficient at managing risk. Their 9 liquidity is needed in the marketplace. They serve in a very 10 important role in keeping the markets linked at related 11 prices. Some of these proposals frankly endangers that 12 linking and in a delinked environment, that's where we see 13 the most loss of liquidity. 14 If any of the proposals get adopted, we should make 15 sure the appropriate exemptions are available for options and 16 ETF market makers and for those other liquidity providers 17 where needed. Thank you. 18 MR. BRIGAGLIANO: Thank you, Jerry. 19 Brett Redfearn. 20 MR. REDFEARN: Thank you very much to the 21 Commission for inviting us to speak today, and on behalf of 22 JPMorgan, I really appreciate that. 23 In addition to running JPMorgan's liquidity and 24 algorithmic trading products, I also have been the chair of 25 SIFMA's equity markets and trading committee. I'm not here 75 1 so speak on behalf of SIFMA. However, I will say that in the 2 course of Reg. NMS implementation, I spent numerous hours 3 working with Commission staff and a lot of others to try to 4 work out the details of the rules, so I'm very familiar with 5 a lot of the issues with respect to implementing various 6 types of rules. 7 I'd like to just say that first of all the -- you 8 know, I think the Commission recognizes very well as do most 9 of the panelists, I think all of them, that legal and lawful 10 short selling is extremely important to liquidity and price 11 discovery in the marketplace. As with the others, we commend 12 the Commission with respect to the various actions you've 13 taken, in particular, 204T in combating some of the abuses 14 against naked short selling have been extremely helpful. 15 With respect to these proposals and proposed rules, 16 I think that, you know, even in the proposing release, it's 17 very well recognized what a lot of the benefits are to short 18 selling in the market with respect to liquidity and pricing 19 efficiency and correcting upward stock price manipulation. 20 The key point, I think, that I took out of this was 21 it said it's important that any short sale price test 22 regulation be designed to limit any potential unnecessary 23 impact on legitimate short selling. And ultimately, I think 24 that's what the challenge is that we have, to figure, you 25 know, how can we sort of respond to the investor confidence 76 1 concerns while at the same time make sure that we steer away 2 from some of the potential risks and damages that we're 3 seeing as possibilities here. And obviously, those are both 4 to the issues of liquidity and price discovery as have been 5 mentioned. 6 I'd just like to say that the world is a very 7 different place from when these rules were in place before. 8 We know that the velocity of trading is extremely fast, 9 millisecond trading, microsecond trading, reports come back, 10 bids going up from -- you know, I mean, we're literally 11 taking in direct feeds from over 10 different market centers. 12 We're taking in last sales from over 40 different venues. It 13 really is a different place. 14 The nature of the market participants has changed 15 quite a bit, too. People have mentioned the long-short hedge 16 funds, the quantitative traders in the marketplace, just the 17 activity in the way that people trade is very different. And 18 the nature of the liquidity in the market is very different. 19 So I think those are critical issues that we have to consider 20 in this. 21 I think that JPMorgan supports the -- anything in 22 the interest of restoring investor confidence, and if it's 23 believed by the Commission that one of these proposals will 24 help to do that, then we certainly want to work with you to 25 try to figure out what the best workable solution is. I 77 1 think among the proposals that have been put out there, the 2 general belief is that a circuit breaker with a modified 3 uptick rule proposal is really the best way to go. The issue 4 of making sure there's a circuit breaker so you avoid putting 5 something in place that's very cumbersome and creates 6 frictions when you really don't have a negative market 7 condition, that's very important. So we believe a circuit 8 breaker's important. 9 In addition to that, the modified uptick rule, even 10 though it's actually very operationally complex to put into 11 place, we think that that as opposed to an all-out ban on 12 short selling actually limits the frictions in the market. 13 So we would rather spend the money and do the work to create 14 something that limits the frictions and creates ultimately 15 less negative impact in the market. We think that we should 16 think about 10 percent. It may need to be a higher number 17 for smaller priced stocks, but we can get into those details 18 later. 19 Ultimately, this discussion is extremely helpful. 20 We look forward to working with you and figuring it out and, 21 you know, would certainly urge you to continue to look very 22 closely at this and to apply due diligence and not to rush, 23 so thank you very much. 24 MR. BRIGAGLIANO: Questions from the Commission? 25 COMMISSIONER CASEY: Well, thank you very much for 78 1 all of your opening statements. I wanted to go back, I 2 think, a central point that a lot of us -- without speaking 3 for the rest of the Commission -- but raised in the meeting 4 when we put out a lot of these proposals which was that we 5 would be very keen on receiving any kind of empirical data or 6 analysis that should inform our judgment, not just about the 7 propriety or efficacy, cost to benefits for a particular 8 proposal but also the desirability and basis for supporting 9 taking action on any of these proposals. 10 So with that, I guess I would ask you all to -- if 11 you could point to any data or analysis that you would 12 encourage us to look at in making those judgments, what would 13 those be? 14 I mean, Jeff, you spoke a little bit about what you 15 saw as the dramatic bear raids over the past year. And, 16 John, you had differing views in terms of, you know, what we 17 see in terms of price pressure from short versus long 18 selling. And so I guess I would say if we're looking for 19 data and analysis in trying to make these judgments, which 20 particular studies or analysis or key data points would you 21 focus our attention on? 22 CHAIRMAN SCHAPIRO: If I could just add to 23 Commissioner Casey's question and to the extent -- because 24 one of the things we are wrestling with is this measure of 25 investor confidence, and there are lots of terrific economic 79 1 studies that we can and will look at with respect to the 2 benefits of short selling, the implications of removal of the 3 uptick rule at the time it was done. We're struggling with 4 quantification as the Commission often does when it's doing 5 rule making of the benefits to investors or the benefits to 6 the marketplace of taking a particular action. So anything 7 in that area, most particularly, I think would be helpful, 8 too. 9 MR. O'BRIEN: Well, I would just say on the first 10 part given the impact on micro market structure really, three 11 sources. One would obviously the data analysis done in the 12 wake of the ban last fall which introduced the notion, I 13 think, pretty definitively of wider spreads and lower volumes 14 and higher volatility in affected names. And obviously, a 15 ban is not a circuit breaker on tick, is not a circuit 16 breaker on bid, but I think you can extrapolate a little from 17 that. 18 And moreover, I think a lot of work can be done in 19 highlighting the impact on the individual investors. When I 20 talk about a silent tax in my opening remarks, wider spreads 21 are effectively a silent tax on all investors to get in or 22 out of any individual equity. And that needs to be 23 understood. There is a price to respond to that 24 understandable but visceral concern over the system not being 25 fair due to unlimited short salability. 80 1 Second, I think a lot of good work was done by the 2 FSA analyzing similar restrictions employed in the U.K., and 3 I think, third, some good work is being done by IOSCO in this 4 regard globally as well. And this is an -- you know, 5 investor confidence, that's an international problem, and it 6 needs to be looked at. It's not just a place where only 7 U.S.-centered data points on the confidence and overall 8 market impact work are completely dispositive. 9 In terms of the impact of rules on investor 10 confidence or not, I'm not a psychologist, and I think we 11 could have really interesting panels on that with -- where no 12 one on this panel is really qualified to sit on, but I would 13 just counsel this. Leaders don't give people what they're 14 asking for, they give people what they need and provide the 15 strategic vision of why they're doing what they're doing 16 backed up by very strong, effective, consistent tactical 17 action. And I think that's the best approach over the long 18 term to get investor confidence to where it needs to be. 19 MR. LEIBOWITZ: I think it's interesting to look at 20 all the studies that are out there. There's a number of 21 difficulties with all of them. I think what you'll find is 22 every one on this panel -- and I know most of them 23 personally -- are actually very quantitative and despite the 24 fact that we come out in different positions on the issue. 25 So that itself is kind of interesting. 81 1 First, there's a behavioralist aspect to this. So, 2 for example, when the SEC looked at what would have happened 3 had there been a plus tick rule in effect during this time, 4 well, it's impossible to estimate what the feedback effects 5 of those behavioral changes would have been. So unless the 6 conditions are exactly the way you would have expected them 7 to be then, it's hard to extrapolate that. 8 Second of all, you rarely have a control group 9 where you have one group doing one thing and one group doing 10 another and look at what happened between the two groups. 11 Now, the example that is the pilot study during -- before 12 they removed the uptick the first time, but the issues there 13 were that we were in a very low volatility period. And what 14 we're really looking for is very rare events, right, a panic. 15 Right? We're not looking for things that happen every day, 16 we're not looking for things that have happened for years, so 17 getting quantitative analysis, it's just difficult. It's not 18 that people are just speaking from their gut; it's just that 19 it's very difficult to quantify these things. And I think 20 we're all looking for that in earnest to try to find it one 21 way or the other. 22 MR. NAGEL: I'd like to address both of those 23 questions. As far as data or facts or studies supporting 24 the critics of short sellers and supporting the notion is 25 somehow to blame for the horrible times we all went through 82 1 last year, in my view, the silence is deafening. I haven't 2 seen it. I've heard a lot of unsupported allegations, a lot 3 of suspicions, but I've seen very little or nothing in the 4 way of actual facts and data to support those assertions. 5 On the other hand, I think there have been a lot of 6 studies both around the pilot before the short sale price 7 tests were removed and around last fall's ban supporting the 8 assertions in my opening remarks about all the benefits of 9 short selling that when removed from the market have serious 10 consequences. 11 And turning to the question of investor confidence, 12 I don't have any bright ideas as to an easy way to measure 13 it, but I do think it's important to measure not just the 14 impact on investor confidence of a big market break which can 15 be very damaging to investor confidence. But when you remove 16 short selling or hamper short selling or tax short selling 17 and you remove the many benefits it brings to the market, 18 markets like liquidity, like tight spreads, that erodes 19 investor confidence because I can tell you when you have thin 20 markets trading at wide spreads, that every day, every trade 21 taxes investors. It costs them more money to trade, even 22 whether it's institutional retail investors. Even retail 23 investors are going to be asking why can't I get my trade at 24 the quoted price? Well, the reason you can't get your trade 25 done at the quoted price is because there's only 100 shares 83 1 up, 10 cents wide. 2 So I do think it's important to look at not only 3 the impact on the confidence of the break and the absence of 4 short selling tests but also the -- all those benefits that 5 short selling brings to the markets and what would happen as 6 those are degraded. 7 MR. BROWN: We've talked in meetings about looking 8 for data and at Schwab, we've gone back and tried to find 9 what could we find within our company that could give you 10 some empirical evidence. And we thought about well, let's 11 look at this issue of investor confidence. What are our 12 customers doing? And that seems to us to be the best way, 13 and we've noted -- I mean, obviously, any rational investor 14 has moved out of the market because they just saw tremendous 15 chaos and they have decided to place a high degree of their 16 money in cash and cash-like securities and even when -- in a 17 time when those return minimal amounts. So they're willing 18 to get no return on their money as long as it's safe. 19 And I think that's a demonstration of the concern 20 that they have. So and then the other piece of evidence that 21 we have about what our customers want is we get e-mails from 22 them, and we have a tremendous number of contacts with our 23 customers where they tell us what are you doing about the 24 short sale rule. In fact, we have a customer who FOIA'ed the 25 SEC to ask how many times Schwab had been in to demand that 84 1 the short sale rule be reimplemented. So there are people 2 very concerned about this, and we are trying to respond to 3 those concerns. 4 COMMISSIONER CASEY: And I'd like you all to 5 continue to answer the question, but sort of a follow-on to 6 that as well which is if you don't have the data and analysis 7 which supports what is viewed as the role of short selling in 8 driving the declines that you've seen and in the markets and 9 in financial stocks in particular. But the perception is 10 that shorts are playing this role. And, you know, you get 11 the response that you'd hear it on TV every day, that shorts, 12 you know, you got bear rage. You get folks that are, you 13 know, getting together in shorts or, you know, going to take 14 out a particular company. 15 I mean, how much is that? How much concern should 16 we have over just the fact that it's the perception that's 17 driving this panic because that seems to be the space that 18 we're in. 19 MR. NAGEL: Well, in some instance, I think 20 perception became reality, that lo and behold there 21 were -- you know, people would claim it was the shorts and 22 then the -- as I had mentioned in my opening remarks, I 23 believe the credit default swap market had a tremendous 24 amount of coordination with short selling to create some of 25 the conditions we saw and some of the impacts on securities. 85 1 And so I would agree that we -- there needs to be some 2 regulation brought to that marketplace, but in any case, in 3 some sense, our customer, retail customers do believe that 4 this is a problem and they want to see action to resolve it. 5 COMMISSIONER PAREDES: Can I just jump in real 6 quick on that point? Even if the perception is, if you will, 7 the new reality, if it's not the real reality, is the right 8 response regulation or is the right response some sort of 9 investor education on the part of the Commission and on the 10 part of all of you and anybody else out there? 11 MR. O'BRIEN: Right. I think it's a combination of 12 both, and I think more education needs to be done with 13 respect to what abusive short selling really is. And it's 14 not short selling, per se, that needs to be broadly 15 restricted. I think the Commission has done, as I said in my 16 remarks, a lot already to restrict the abusive conduct. 17 I think there's more information and what's really 18 changed most in the last couple of years is the promulgation 19 of these exotic instruments that people can use in concert 20 with short sale activity to really leverage abusive 21 strategies and by having a greater information network and 22 enforcement ability to pursue perpetrators of truly abusive 23 conduct that's -- and educating the average investor about 24 the steps that you're taking because it is a relatively 25 intellectually lazy approach to rely on a rule that people 86 1 have some familiarity with as a way to give them comfort when 2 you know full well it's not really giving them the come back 3 comfort and it is leaving wide very large loopholes for 4 abusive perpetrators or conduct or continue operating in. 5 And over time, that might have a short boost to 6 investor confidence, but it's going to degrade faith in the 7 ability of the regulator to improve our markets in the long 8 term. And that's the challenge because you can't change 9 media perceptions overnight, but if you're consistently 10 dedicated to that backing up articulation of a clear vision 11 with concrete action along the way, I think we can get there. 12 COMMISSIONER WALTER: I guess I could follow that up 13 by asking a question. Whether it's based on perception or 14 the actual economic impact of the rule, I wouldn't ask of you 15 to predict if we were to take action what the impact really 16 would be. I mean, we can speculate about that. 17 There was some support on the last panel for doing 18 something in the nature of a pilot and I think there were 19 also some countervailing concerns about that, particularly on 20 the cost side of the analysis because assuming that the pilot 21 is imperfect, you could end up having to make vast system 22 changes and then make then again. So I wonder if you could 23 feed that into the approach and give me a little bit of a 24 point of view on how you think we figure out what will change 25 if we do take action and whether a pilot approach is viable 87 1 or not. 2 MR. O'CONNELL: I think one thing we can count on, 3 most of the empirical data on the subject over the years has 4 one common strain, and that is that when you create risk, you 5 kill liquidity. So we have this dilemma where there is an 6 issue that we saw last fall and there was some selling that 7 really couldn't be explained by any of the current studies. 8 And I know people are out there still trying to get their 9 finger on it. 10 But in the meantime, we have proposals in front of 11 us which seem like they're going to worsen the situation not 12 help it because if it were to occur again, what we'd need 13 most in those instances is tighter and more liquid quotes. 14 And I think what we do know is that when you tell somebody he 15 can't sell something on the bid or you tell an options market 16 maker he can't buy a call at that price because he may not be 17 able to lay off on the stock bid, you've created risk and 18 you've widened the quotes and they're not going to be there 19 when you need them most if it does happen again. 20 MR. REDFEARN: The thing I would add on that is if 21 we're looking at how, let's say, the modified tick test or 22 the bid test would work in practicality in today's market, if 23 you were to just simply look at the number of bids and ticks 24 that we have in a given second with all the different quoting 25 venues that are out there, you'll see hundreds literally 88 1 within the course of a second in an active stock, in a 2 volatile circumstance. In several cases, you'll see actually 3 directional changes in the bid occurring in different 4 directions also in the same exact second. 5 So one of the interesting questions here when we 6 look at again the velocity of the market and the way the 7 market data works is you'll see significant differences 8 between the SIP feeds versus direct feeds that firms get. So 9 I think we need to understand that, A, you'll have different 10 people seeing completely different tick directions depending 11 on which feeds that they're using; B, you'll be creating a 12 situation where, in fact, there may very well be a down 13 market where there's actually a momentary up bid. It might 14 only last for three milliseconds, but there may very well be 15 a momentary up bid in which case the shorts would potentially 16 be able to shoot off other short positions. 17 So effectiveness is something that I think we 18 really need to look at, and that goes in addition to, I 19 think, the other comments that have been made with respect to 20 just really let's just get behind the question of was it bear 21 raiders or was it long sellers and try to provide the public 22 with real information because I think there is still somewhat 23 of a gap between what the research has shown and what the 24 public believes. And I think that's something that we really 25 need to get out there. We may have done a poor job of doing 89 1 that so far. 2 MR. LEIBOWITZ: I think it's hard to find the 3 smoking gun that says a stock has been manipulated, but just 4 like I think we would probably all agree that most people who 5 say oh, short sellers are the demon, just like we're hearing 6 from a lot of other industry people, I think just as much 7 exaggeration to the cost of such a role is going on. 8 I think short of an outright ban which clearly has 9 detrimental effects, the Reg. SHO pilot study didn't show 10 significant help to the uptick rule but it also didn't show 11 any harm, right? So I think that one that does allow 12 continuous trading, that isn't overly onerous, that has the 13 right exemptions can actually not cause harm to the market 14 and then if it does build investor confidence, then we're all 15 to the good. 16 To answer the question about the pilot, I think it 17 is potentially a good idea to have it so that we could have a 18 controlled study. And to your concern that well, we've spent 19 a bunch of money to discover a problem and then we have to 20 spend it again. Well, yes, it means that we made a rule that 21 didn't work the way we intended and now we're going to make 22 it better. So that's exactly the right approach, and then 23 either we'll discover well, we shouldn't do anything and we 24 should pull it back out or we should modify it in a way that 25 allows us to have more efficacy in the market. 90 1 I think the problem here is we don't have a 2 controlled experiment. It's hard to imagine putting 3 something out there, then studying it and then changing it 4 again unless we have a good way to compare. 5 MR. BROWN: To address your pilot idea, I think 6 also it's a good idea, and, in fact, you may get an 7 indication of how issuers feel about it when they all call 8 you to sign up for the pilot because they might want to be 9 included in that protection, so. 10 MR. LEIBOWITZ: I can tell you what happened the 11 day after the ban was announced, we got blizzarded with 12 issuers saying I'm a financial stock. 13 MR. BRIGAGLIANO: Chairman Schapiro. 14 CHAIRMAN SCHAPIRO: I want to change tracks for 15 just a moment. Are you all aware of short sale restrictions 16 or tests in other countries or in other kinds of markets that 17 we ought to be taking a look at to look and to see what the 18 experience was that may have some parallel use for us? 19 MR. O'BRIEN: I think they should all be evaluated, 20 but they need to be done in context. It can be done on a 21 variety of issues, the efficacy of the rule generally, the 22 impacts of certain variants of implementation of rule. 23 For example, with circuit breakers, we heard a lot 24 about the magnet effect. I'm personally concerned about the 25 potential of a magnet effect should we be in a circuit 91 1 breaker driven rule application. Some studies have been 2 proffered to say that that magnet effect doesn't exist. A 3 lot of that's based off of the Kuala Lumpur stock exchange. 4 And I'm a big proponent of the development of Malaysian 5 securities markets, but don't think it's dispositive, an 6 interesting data point but one that needs to be held in 7 context. 8 So I would advocate casting a fairly broad net. 9 We've seen a variety of rule implementations, outright bans. 10 Australia would be an example of a very broad net cast. 11 Others have taken more limited approaches. I think IOSCO has 12 done a good job again, as I said, synthesizing a lot of that 13 data. But it does need to acknowledge the fact that we have 14 a unique and probably unique in a good way level of scale, 15 efficiency and automation in our market that may not be 16 replicated elsewhere and may distort the data point. 17 CHAIRMAN SCHAPIRO: But we also probably have a 18 different scale of individual investor participation where 19 the confidence issue is that much more important for us to 20 try to address. 21 MR. O'BRIEN: Absolutely. 22 MR. REDFEARN: One thing on that point is it's safe 23 to say that this issue being debated is not uniquely a United 24 States phenomenon. We have a e-mail that comes out weekly 25 with a list of I don't know how many countries, but I bet I 92 1 could go with 25 to 30 different countries that's giving us 2 updates on the various short selling provisions or rules that 3 exist in various markets. I'd be happy to forward along 4 the -- I can't say that we've studied the details of what's 5 happened in all of the different cases, and many of the 6 markets are very different than our markets. They're not as 7 liquid and don't have a lot of the same dynamics, but 8 certainly, there is a lot of activity out there and you've 9 got the halts in Japan. And you could go on and on about 10 different. 11 MR. O'CONNELL: I would also add to that that 12 we -- our liquidity levels in our markets are the envy of the 13 world. A lot of the catching up that foreign countries are 14 doing right now is largely because of U.S. trading houses 15 incorporating their trading strategies overseas and providing 16 liquidity in those markets as well. 17 So I think it's something worth protecting, but I 18 have heard that some of the countries that kept the 19 restrictions on after the September problems didn't meet with 20 much success. Those that withdrew them had sort of the same 21 impact that they did, but I can't think of where I saw that. 22 But I think that that was the outcome for those that left 23 them on. 24 MR. BRIGAGLIANO: Commissioner Casey. 25 COMMISSIONER CASEY: Just a sort of follow-on to 93 1 your point, Larry, which is we've heard before the fact that 2 we understand what the costs might be with some particular 3 proposals that have been offered up and that you can 4 ultimately make a judgment of whether or not it actually 5 causes harm, right? And I guess my question is: You know, 6 what is the appropriate standard for the Commission? Is it 7 one of saying we don't think it actually causes that much 8 harm, therefore it's worth taking the chance that it'll 9 improve investor confidence in the absence of something more 10 formative? I mean, I understand. I completely get the 11 behavioral issues here and the perception. 12 But I'm just trying to say as far as the 13 rigorousness with which we engage in rulemaking, I mean I'm 14 just trying to grapple with the fact that it makes it quite 15 challenging if that's where the proposition that we're faced 16 with. 17 MR. LEIBOWITZ: You have a tough job. I mean in 18 all seriousness, it is very difficult because the gain is 19 somewhat nebulous. It's Jeff's investors being willing to go 20 back into the marketplace. Well, how do we measure that and 21 what's the cost? Well, we're going to have to look at what's 22 the cost in changing spreads and changing behaviors, 23 particularly if we can do it in a controlled situation, at 24 least we'll be able to get some numbers out of it. 25 I think maybe that is really in the end the best 94 1 way to do this because otherwise, it's going to be very 2 difficult to know because even when we put it in, it's going 3 to be very different conditions than it was in the fourth 4 quarter. If you look at the difference between the fourth 5 quarter and the first quarter, a 10 percent circuit breaker 6 on an average day in the fourth quarter would have triggered 7 400 stocks every day, right? On average in the first 8 quarter, that was down to 225, right? And if you look 9 historically backwards, it goes down to as low as 50 at 10 10 percent. 11 So the conditions under which we put this in are 12 going to be very different. That's why having a pilot 13 actually might be interesting, but this is going to be a very 14 tough choice because it will be very difficult for you to 15 figure out the cost/benefit. And the only sure cost will be 16 what it costs all of us to implement it, right? So of 17 course, all of us are going to be saying wow, this is really 18 too expensive. 19 MR. O'BRIEN: Let me be a little stronger than 20 that. The American stock market is not petri dish to be 21 experimented with. I'm a big believer in markets as 22 ecosystems that are a function of the participants, the 23 economics and the rules. And when you're dealing with 24 abusive conduct as opposed to an abusive market structure 25 generally, you want to avoid structural remedies to 95 1 conduct-based concerns. No one likes being stung by a bee, 2 but you don't kill all the bees and then become surprised why 3 all the flowers have all died. 4 We really need to look at very, very targeted 5 remedies. The fact that oh, we don't see any -- we haven't 6 seen any evidence before that it may not hurt, so let's try 7 it and maybe it'll help is not the standard to be made. That 8 does not mean the Commission cannot take controlled steps to 9 see what impacts may be on a market structure. Pilots would 10 be an example of that. The Commission did take a very 11 thoughtful approach in this regard when deciding to repeal 12 the short sale rule, but I would strongly urge not to give in 13 to the temptation of trying things just to see what happens 14 on a broad basis because I think your first obligation is to 15 do no harm. 16 COMMISSIONER CASEY: I have -- oh, sorry, 17 Jamie -- just one other question which was, Jeff, when you 18 talked about the notion of a pilot and what kind of a line 19 would be formed in terms of companies that would want to be 20 part of that. Do you really think there'd be a tremendous 21 amount of demand? I mean this is sort of the same question 22 that you had around the questions associated with the ban and 23 who would seek to be, you know, to consider themselves within 24 the scope of the ban. I mean is it a sign of strength or 25 weakness on the part of companies ultimately judging whether 96 1 or not they'd want to be part of any kind of pilot and 2 especially in light of what we believe might be some of the 3 adverse consequences of having such a restriction? 4 MR. BROWN: I think issuers, yes, would want to be 5 a part of the pilot. In one sense they -- if in adopting it, 6 the Commission would determine that the rule has merit, that 7 it will provide benefits to investors and shareholders, an 8 issuer would say well, if it's going to benefit the 9 shareholders, I want my shareholders benefitting. So why 10 wouldn't I participate in the -- want to participate in the 11 pilot? It would seem very -- you know, you have a duty to 12 help your shareholders as much as you can and to inject them 13 into a rule that the Commission would determine has benefit 14 would be a very important part of that duty. 15 MR. LEIBOWITZ: I think it's actually a very 16 interesting question because initially the day it was 17 announced, our phones were flooded with everybody and their 18 brother, even the auto companies, trying to say they were 19 finance companies. You know GM has a finance arm or they 20 did. 21 And yet as it went on, some of the firms wanted to 22 get pulled off of it because they thought there was a stigma 23 attached, just like some firms don't want to take TARP money 24 because they don't want to be viewed as being in trouble. On 25 the other hand, some firms say well, if I don't have TARP 97 1 money, people are going to question whether I'm backed by 2 somebody. 3 I think that that's going to happen much more with 4 a black-and-white rule like a ban. I think if you end up 5 with something in between like a tick test or a bid test, 6 then I think you'll have a much more balanced approach to it, 7 and I also think we shouldn't take suggestions because to be 8 truly a control group, we need this to be randomly selected, 9 distributed among industries so that half the finance groups, 10 half are in each pot, half high cap, half low cap and really 11 do this in a scientific -- if we're really going to do this 12 as a control test, we need to do it that way. 13 MR. BRIGAGLIANO: I'd like to follow-on to a 14 question, an area raised by Commissioner Casey. I don't want 15 to let this group go without asking about the execution 16 impact or bite of the Commission's proposals. The Commission 17 proposed price tests at the trading increment. 18 To what extent -- and maybe we could start with 19 John and Jerry and Brett -- would such an increment be a 20 silent tax, a systems cost, something you have to just put up 21 with, spend money, make capital contributions to deal with or 22 to what extent would a price test impact executions in your 23 view? 24 And if a penny increment would not have an 25 execution impact, at what point, at what level would a price 98 1 test become a ban? 2 John, you want to start? 3 MR. NAGEL: I'll take a shot at that. Let's start 4 with at a penny increment. A price test would have a 5 material impact on execution. It would impact some trading 6 strategies more than others, particularly those trading 7 strategies that need to be active and a lot of algorithmic 8 trading strategies need to be active and maybe taken out of 9 the market for various amounts of time depending on how 10 frequently the test is triggered and which direction it's 11 moving. 12 I do think it is something the markets could bear, 13 and I've made clear, I think, we believe it is not helpful 14 and is actually counterproductive. But we did have price 15 tests for many years. The markets functioned generally quite 16 well and were quite successful, and so we do think it's 17 something that the markets could bear with appropriate 18 exemptions, I should add because I don't want to lose sight 19 of the fact that under the old NASD bid test, equity market 20 makers had a general exemption from the bid test. Option 21 market makers had a general exemption from the bid test. 22 There were exemptions for domestic arbitrage which 23 were included in your proposal, although I would add that 24 it's very important -- the way they had historically been 25 interpreted required the hedges to be put on roughly 99 1 contemporaneously with the position. And when you're hedging 2 a derivative instrument like a convertible bond or an option, 3 as the stock price moves, the stock equivalent exposure of 4 that long position moves. And it's important to allow 5 whoever's hedging the position to be able to adjust. Delta 6 hedge is what people call it in the market, but basically, if 7 the stock price goes up, your -- you'd be further out of the 8 money and you'll need less of a hedge, and as the stock price 9 moves down, it's the opposite direction. And you'll buy it 10 to cover. 11 So we do think those are important exemptions. 12 Also, ETFs, index products, if those are not exempt, we 13 believe there's going to be a tremendous, completely 14 unnecessary harm on the markets because those are really 15 important hedging tools and there's really not a material 16 risk of a bear raid on an index product. They've been exempt 17 previously, and so we think that it's important that those be 18 exempt. 19 As far as increments, when you go beyond a penny, 20 you don't have to get very far. It quickly turns into a de 21 facto ban on short selling. Many stocks trade. It's not 22 uncommon to see a penny or two wide. As soon as you go 23 anywhere near the actual spread or certainly over the spread, 24 there's an actual ban because you've always got somebody in 25 line ahead of you. And so it doesn't take very far to go. 100 1 There's no -- the right or wrong answer is depending where 2 you're trying to set, how restrictive the test is. But just 3 keep in mind that a penny, we've done it before. We've 4 seen -- had some idea how it works and once you go very far 5 beyond that, it's basically a de facto ban. 6 MR. BRIGAGLIANO: Jerry, would a penny increment 7 impact your trading? 8 MR. O'CONNELL: It would, Jamie, because it 9 introduces an element of risk that you'd account for in all 10 your models. It would be something that you would have to 11 keep an eye on every time you had an offer and you couldn't 12 hit the bid and maybe you're not going to offer them at 12 13 anymore if you can't hit the 11 bid. Maybe you're going to 14 be a 13 offer. Those decisions aren't going to be made. 15 Maybe on every other trade or every fifth trade or tenth 16 trade, but somewhere along the line, it's going to be made 17 enough times that it's going to impact best execution and 18 perhaps significantly. 19 And the one great thing I think of getting rid of 20 the tick rules of the past was that old cat and mouse game 21 people used to play with respect to if somebody was short and 22 they didn't want to show their offer because they thought 23 people were going to jump ahead of them and that I think that 24 also had an effect with respect to the quality of the markets 25 at the time. And I think we've enjoyed the period since that 101 1 time. I'd hate to see it come back, but I think it would. 2 MR. BRIGAGLIANO: Brett. 3 MR. REDFEARN: I mean specifically to your 4 question, I would generally agree with what the others have 5 said before me, just adding that certainly, as you know, for 6 stocks under $1.00, the quoting increment can go below a 7 penny. So for any of those stocks, a one cent increment 8 would too high. It would have to be in line with the 9 quotable stocks under a dollar. 10 And in addition, for the rest of the stocks, if you 11 went above a penny, that would be very harmful. And I think 12 it would in many cases become a de facto ban. So I don't 13 think for the rest of the universe going above a penny would 14 be advisable at all. 15 MR. BRIGAGLIANO: Jeff. 16 MR. BROWN: Jamie, if I might, just you referred a 17 lot this morning a lot about how whatever rule the Commission 18 would adopt may harm liquidity. I think if you go back and 19 look at rules the Commission's adopted or proposed over the 20 last 25 years, there's always been a claim that we're going 21 to harm liquidity. And yet liquidity continues to grow. We 22 now, as Brett was saying, make hundreds of thousands of 23 trades a second. I'm not so sure that that impact -- the 24 market -- the trading community will learn to trade in a new 25 environment that -- and we'll adjust. 102 1 MR. BRIGAGLIANO: Chairman Schapiro -- oh. Well, 2 my next question would be we understand that there are 3 systems costs, there would be systems costs in building any 4 kind of price test capacity and the Commission explored that 5 in its release. Do those costs differ if you're talking 6 about a marketwide test that's on all the time versus a 7 circuit breaker or does it cost as much to put in the 8 system's capacity to have a circuit breaker price test as it 9 would for a price test that's on all the time? 10 Larry, you want to start? 11 MR. LEIBOWITZ: I think it's actually incremental. 12 You're just additive because you can think of these as 13 independent system changes. You're going to put in the 14 ability to do a bid test, and then you're going to add a 15 circuit breaker on top of it. So in terms of the programming 16 work up-front and the testing work up-front, they're purely 17 additive. 18 I think going forward there's probably some 19 incremental capacity differences based on whether I have to 20 have enough capacity for it to be on all the time or not, but 21 that's a different issue than the up-front work. 22 MR. O'BRIEN: Again, I would agree with Larry. I 23 think it's more in the structure of the underlying rule as 24 opposed to whether or not it's triggered by a circuit breaker 25 or on all the time. There is some incremental work to react 103 1 in response to a dynamic event in real time to trigger 2 application of the rule, but rules that would require storage 3 of -- and integration of data marketwide as opposed to the 4 modified uptick rule that would be effectively be post only, 5 that would be very easy for any one individual trading center 6 to implement without storage of any third-party data. That 7 drives implementation costs and time lines, I think, more 8 than anything else, circuit breaker notwithstanding. 9 MR. REDFEARN: I would add to that a circuit 10 breaker with an all-out ban, as I mentioned earlier, is 11 certainly the cheapest way to go, but because of the 12 frictions it caused and some of the issues of liquidity, we 13 don't think that that's the right way to go. 14 If you go beyond that and then you add in the 15 modified bid tests, there certainly are additional costs. In 16 fact, it looks like there will be even more costs to that 17 than there would be just dealing with all-out bid test. 18 Again, it's interesting because we've sort of looked at it 19 and thought that each one looks -- each proposal looks more 20 and more desirable, gets more and more expensive. 21 Again, the expense is, I think, in this case not 22 the issue, but it does speak to the question of piloting. 23 And I would beg to differ with Larry with respect to the 24 costs relative to what's involved here. I don't think 25 they're overstated. I think for larger broker-dealers, if 104 1 you're looking at over a dozen different trading desks with 2 four or five different front-end systems and middle office 3 and back office, if you go through the details, it really 4 does add up and become expensive. 5 MR. LEIBOWITZ: Just to clarify, Brett, I actually 6 wasn't talking about the costs in terms of modification of 7 systems. I was talking about the costs of the Chicken 8 Littles, about oh, spreads are going to widen and it's going 9 to cost so much to trade here more than systems. 10 MR. REDFEARN: Oh, okay. 11 MR. O'BRIEN: One thing about implementation that I 12 would also add and I think it needs to be a function of any 13 rule is you need to give, I guess, downstream trading centers 14 the ability to rely on compliance work done upstream. So 15 JPMorgan sends an order to Direct Edge which is then routed 16 to the floor of the New York Stock Exchange which would then 17 be rerouted to another trading center for ultimate execution. 18 If you're requiring a compliance analysis, an event, an 19 action be taken by each of those trading centers along the 20 way, not only does that create incremental implementation 21 risk, it creates operational risk as well because you have 22 orders ping ponging back and forth between trading centers. 23 "It was good when you sent it to me. It wasn't good when it 24 was received." 25 So very similar to the approach taken with Reg. NMS 105 1 where the inter-market sweep order exemption was a way to 2 give downstream comfort and reliance for trading centers 3 executing orders. You'd need to apply a similar approach. 4 COMMISSIONER WALTER: While we're in this space, 5 can you comment on the -- whether there's a cost differential 6 of following a policies and procedures kind of approach as 7 opposed to a prohibition? 8 MR. REDFEARN: I would just point out that we -- I 9 mean assuming that I understand clearly what the prohibition 10 means versus the policies and procedures approach. I'm 11 assuming that the prohibition would have it such that the 12 exchanges would, in fact, just block any short sales that 13 were in perceived violation as opposed to a policy and 14 procedures approach being on where as with Reg. NMS, 15 broker-dealers, we all have different quote feeds. We would 16 have to do our sort of best efforts with the data feeds that 17 we would to at that point in time make a determination of 18 whether or not it's an up bid or not. 19 In that particular case, I think the key point 20 there is would it be absolutely necessary for our trading 21 strategies and our trading algorithms and our business 22 generally to have a policies and procedures approach because 23 of the differential in looking at the market data quotes. 24 It's very likely that they would actually be seeing something 25 completely different even if it's only 10 milliseconds later 106 1 than we're seeing. Hence we wouldn't be able to sort of time 2 our trades right. 3 So I think that the ban approach would be -- would 4 just wouldn't work anyway. Similarly speaking, again, the 5 approach is more expensive for us. 6 MR. NAGEL: I wanted to add that I think the 7 policies and procedures approach has one big advantage that 8 has really proved valuable in the implementation of 9 Regulation NMS which takes a similar approach. And that's it 10 allows market participants to allocate among each other. 11 There's many -- as we were discussing, there's many steps in 12 the chain of execution of an order. And it allows market 13 participants or market centers, market makers, brokers to 14 figure out who's in the best position to ensure that the rule 15 is complied with. 16 You could end up at roughly the same place whether 17 you have a prohibition with appropriate exemptions or 18 affirmative defenses for race conditions or I did the right 19 thing and the market moved just as my order got there versus 20 saying you have to have policies and procedures to do the 21 right thing in the first place. But allowing the various 22 market centers and market participants to allocate 23 responsibility, I think is really -- makes it much more 24 efficient when it is implemented. 25 MR. LEIBOWITZ: First, I wanted to thank Bill for 107 1 routing his orders to the floor of the New York Stock 2 Exchange. 3 MR. O'BRIEN: We're the number one router of flow 4 to the floor of the New York Stock Exchange on many days. 5 MR. LEIBOWITZ: But I think what you'll find out is 6 when first Reg. NMS was proposed, it was meant to be a hard 7 ban on the trade-through and the more you dig into the 8 details, the more you realize it really is impossible. And 9 so I think that we can say anything we want, but the reality 10 is it's almost going to have to be a policies and procedures 11 by the time we're all said and done given how much volume is 12 done in the TRF, given how much is done in other places, et 13 cetera. It's the only way it's going to work. 14 MR. O'BRIEN: Well, the one final thing I'd add to 15 echo Larry's comments is an outright prohibition, I think, 16 would be an improper allocation of enforcement resources. It 17 wouldn't be targeting truly egregious conduct. It 18 would be -- Rick Ketchum's organization or others saying you 19 had 17 instances, explain -- out of 10 million, explain why. 20 As opposed to allocating to those resources and ferreting out 21 and prosecuting truly abusive conduct. 22 MR. BROWN: I'd just would like to echo that I 23 think the only way it would work is under a policies and 24 procedures approach. The market data differences alone would 25 cause you to -- you were absolutely -- Brett's absolutely 108 1 right. You'll see different things at different times within 2 milliseconds, and that alone allows you -- a policies and 3 procedures approach allows you to capture the information 4 that you see and be able to record that you've met your 5 obligations and not have that then the short exempt -- being 6 able to send the orders short exempt and allows it to be 7 executed knowing that you've met your obligations. I think 8 that's a very important part. 9 MR. O'CONNELL: And also with policies and 10 procedures, you're going to have to do it if you're a large 11 trading house because as much as it may cost, what, millions 12 of dollars to implement, the cost of missing trades by 13 leaving it up to latency to get to the exchange in time is 14 going to gobble up that money in no time flat. So everyone 15 is going to have to remodel and develop policies and 16 procedures if they're active traders. 17 MR. BRIGAGLIANO: Other questions? 18 CHAIRMAN SCHAPIRO: Jamie, is there time for one 19 last question? 20 MR. BRIGAGLIANO: Absolutely. 21 CHAIRMAN SCHAPIRO: In a lot of areas, the 22 Commission works very hard to try to be coordinated with or 23 take a similar approach to an issue as our regulatory 24 counterparts around the world do. Is this one of those areas 25 where it's important for us to try to achieve some kind of 109 1 convergence or does it -- is it not really very important 2 here? 3 MR. NAGEL: Speaking on behalf of Citadel, I would 4 say getting the right result would outweigh convergence. If 5 there are other jurisdictions that adopt approaches that 6 impose substantial frictions in their market or harm the 7 quality of their markets, I think in our view that would be a 8 race to the bottom that I'd rather not be a part of and I'd 9 rather take that advantage to our capital markets and to help 10 employ people in the United States when our markets win out 11 in the end. 12 On the other hand, if there is convergence around 13 an end result that is workable, that could allow for some 14 efficiencies in implementation. But I think those 15 efficiencies would be very quickly drowned by the real world, 16 everyday trading costs of a rule that is counterproductive. 17 MR. LEIBOWITZ: I think sharing information is 18 really important so that -- because all the regulators must 19 be looking at this. So what have you seen, what do you hear, 20 what are you thinking, but this is really different than not 21 coordinating on CDS regulation where there could be 22 regulatory arbitrage, right? If you don't do it here, it's 23 going to pop up in London if you ban it here. 24 We're not going to see short sellers arbitraging us 25 in IBM stock by trading in London or something like that as a 110 1 result of this, so I don't think you have that danger. 2 MR. BROWN: I would say that I think it's important 3 that we do what's right here, and the world will watch what 4 we do. 5 Also, I think you'll hear during this comment 6 process that well, one of the problems with instituting a 7 price test or some sort of short sale restriction is that 8 even within our different markets, there's ways to evade it, 9 to go around, to do some -- to establish the same type of 10 position. And I would say to the Commission that 11 without -- that may be true. You still have to do what's 12 right for the market that you regulate, and if it's right to 13 protect investors in a way you determine, then you should do 14 it regardless of somewhere else they could trade. 15 MR. O'CONNELL: Larry, I hear what you're saying, 16 but I also remember the overseas market and a lot of our U.S. 17 listed stocks would attract a lot of volume in down days. 18 And it was a place that people would go to, so I wouldn't be 19 so sure that if we adopt it and they don't, that you wouldn't 20 see some migration of our order flow overseas, some. 21 MR. LEIBOWITZ: You had a point there except that 22 in general the amount of liquidity in those places is so low 23 compared to here that the trading friction even 24 with -- unless we do a ban. I think that the trading 25 friction here will still be better than over there, but I 111 1 understand what you're saying. 2 MR. O'CONNELL: Or you could be crossing over 3 there, not necessarily hitting bids. I think that was more 4 the pattern in the past. 5 MR. O'BRIEN: I think the key in whether it's in 6 terms of the standard or the application of the rule here is 7 to avoid opportunities for arbitrage. That could be U.S. 8 versus offshore, that could be in the promulgation of 9 exemptions which could give certain trading centers inherent 10 competitive advantages over others. You need to apply an 11 approach that's going to be able to be upheld consistently 12 without degrading other aspects of market quality. 13 MR. BRIGAGLIANO: All right. I'd like to thank the 14 panelists very much. We're on time and under budget. At 15 1:45 promptly, we will begin our third panel which is 16 "Lessons and Insights from Empirical Data: Short Sale Price 17 Tests and Short Sale Circuit Breakers by the Numbers." We 18 have a distinguished panel of academics, and we look forward 19 to seeing everyone back here at 1:45. Thank you. 20 (Whereupon, a luncheon recess was taken.) 21 A F T E R N O O N S E S S I O N 22 PANEL THREE - LESSONS AND INSIGHTS FROM EMPIRICAL DATA 23 MR. BRIGAGLIANO: Welcome back, everyone. Our 24 third and final panel of the day is entitled "Lessons and 25 Insights from Empirical Data: Short Sale Price Tests and 112 1 Short Sale Circuit Breakers by the Numbers." Our panelists 2 will be discussing costs and benefits of short sale price 3 tests and short sale circuit breakers from a quantitative 4 perspective. Similar to the morning sessions, each panelist 5 has an opportunity to present a brief opening statement and 6 subsequently, we'll open it up to questions from Chairman 7 Schapiro and the Commissioners. 8 Again, we'll begin with a very brief introduction 9 of our distinguished panelists. Dr. James Angel is an 10 associate professor at the McDonough School of Business at 11 Georgetown University. Dr. Frank Hatheway is chief economist 12 of the Nasdaq OMS (sic) Group and former professor of finance 13 at Penn State. Dr. Charles Jones is the Robert W. Lear 14 professor of finance and economics at the Columbia Business 15 School of Columbia University. Dr. Robert Shapiro is the 16 co-founder and chairman of Sonecon, LLC. He's a senior 17 fellow at the Georgetown University School of Business and a 18 former undersecretary of commerce for economic affairs. 19 Dr. Ingrid Werner is the Martin and Andrew Murrer professor 20 of finance at the Fisher School of Business of the Ohio State 21 University. 22 Dr. Angel, why don't you start us off? 23 DR. ANGEL: Thank you. I'd like to thank the 24 Commission for the honor of the invitation to be here. We're 25 here because of the uproar over short selling in the 113 1 financial debacle of last year, and we're here to talk about 2 whether some form of the uptick rule should be reinstated. 3 I'd like to point out that the Commission got rid 4 of the old uptick rule after a very careful study. I thought 5 the original study was very well done. Your staff did an 6 excellent job of designing and analyzing the study, and I 7 think the Commission did the right thing. It did a 8 scientifically controlled experiment. It discovered that the 9 old uptick rule really did absolutely nothing and got rid of 10 it. 11 It was a mere coincidence that the market fell 12 apart after the repeal of the old uptick rule. However, many 13 people, as you well know, are calling for a return of some 14 form of an uptick rule. And some of them are responding to 15 emotion, some of them are responding to an inherent gut 16 feeling that there is something wrong with the price 17 discovery mechanism in our market, that there are some 18 reasons to be concerned about short selling. 19 Short selling has a lot of good legitimate uses 20 which I defend on numerous occasions, but there are also some 21 problems there. And yet there is this instinctive feel there 22 is something wrong with the price discovery mechanism. With 23 our transition to all electronic markets which trade at light 24 speed, inter-day volatility appears to be excessive. 25 Last week's incident with Dendreon is a smoking 114 1 gun. The stock dropped by almost half within a few minutes. 2 Nasdaq stopped that stock as fast as humanly possible, but 3 still tremendous damage was done. Yesterday there was a 4 similar incident on the upside with Better Online Systems 5 where the stock jumped for no apparent reason and then 6 settled back down. 7 So I'd like to point out that these kind of moments 8 of excessive inter-day volatility can be caused by short 9 selling. They can be caused by a lot of other things as 10 well. It can be caused by long selling. It could be caused 11 by errors in trading. A lot of things can cause these kind 12 of glitches that damage the reputation of the markets, that 13 damage investors' confidence because when people see a stock 14 drop by a factor of two within seconds, they start to wonder 15 is there some kind of manipulation. There may be. There may 16 not be. 17 But we need to look beyond just short selling. I 18 think it would be a mistake for the Commission just to focus 19 narrowly on short selling bid test uptick rule and step back 20 and say hey, wait a minute. What people are really looking 21 for is a shock absorber. What people really are saying is 22 hey, there's something wrong with our price formation 23 mechanism. It's misfiring and whether it misfires once out 24 of 1,000 times or once out of 10,000, it only takes a few of 25 these events to damage the reputation of the markets. 115 1 And so what people are saying is hey, we need a 2 shock absorber. And whether it's the old uptick rule, a 3 modified uptick, a big (sic) test, I think we need to look 4 beyond mere short selling. If we look at other computerized 5 exchanges around the world, almost all of them have some kind 6 of circuit breaker, price halt, special quote system, special 7 restart mechanism to deal with these situations. 8 So this is what I would urge the Commission to do 9 is to think beyond the narrow technical box of short selling 10 and move into the broader and say hey, what, if anything, 11 should we do to deal with this excessive intra-day 12 volatility? 13 And finally, even though I think the Commission did 14 the right job getting rid of the old and useless uptick rule, 15 I think the Commission made a serious mistake in dropping the 16 transparency we had with respect to tick by tick short sales. 17 One of the great things about the pilot is that we had tick 18 by tick data telling us which trades were short. We don't 19 have that anymore. So whenever anything bad happens in the 20 market, there's a natural tendency to blame the short 21 sellers. People have been doing that for 400 years, will 22 probably do it for the next 400 years. But if we have better 23 transparency about the amount of short interest, better 24 transparency about which trades are short, better 25 transparency about settlement failures, then people can see 116 1 for themselves whether or not there's a problem. 2 So once again, I want to thank you for the 3 invitation to be here, and I'll turn it over to Dr. Hatheway. 4 DR. HATHEWAY: Thank you, Jim. Good afternoon, 5 Madam Chairman and Commissioners. I want to thank you for 6 the opportunity to participate in this panel. 7 I also want to applaud the Commission for its 8 actions to reduce naked short selling. Through these 9 efforts, the Commission reduced the number of listed 10 companies on the threshold securities list by over 98 percent 11 in the eight months following adoption of Interim Final Rule 12 204T. 13 Throughout my 25-year career in finance as an 14 options trader on the Philadelphia Stock Exchange, as a 15 professor of finance at Pennsylvania State University and as 16 Nasdaq's chief economist, I have closely observed the 17 evolution of short selling on domestic and global capital 18 markets and the inter-related asset classes that trade the 19 equivalent short positions at an ever increasing pace. 20 Regulating short selling is about balancing 21 liquidity, transparency, price discovery to best benefit all 22 market participants. Investors, member firms, listed 23 companies agree that short selling provides valuable 24 liquidity and price discovery that contributes to the 25 world-class efficiency of our capital markets. Participants 117 1 also agree that abusive short selling harms investors and 2 companies listed on our exchanges and also erodes confidence 3 in the U.S. markets. 4 And by abusive short selling, I mean attempts by 5 speculators to artificially push down stock prices through 6 selling short. Although the SEC has anti-manipulation tools 7 at its disposal, including newly adopted Rule 10b-21, many 8 market participants believe that real-time restrictions are 9 required. In setting the restrictions, regulators are aware 10 that a link exists between domestic and global markets and 11 also between the asset classes. 12 Short selling in equities and options are closely 13 linked, and there are additional links to index products, 14 futures and other derivatives. As a result, regulation of 15 one asset class impacts linked asset classes and domestic 16 regulation can impact global trading patterns. 17 Recognizing the need to balance liquidity, 18 transparency and price discovery and the link between asset 19 classes and global markets, Nasdaq's view is that the 20 Commission should adopt two measures. First, the circuit 21 breaker that the Commission described in its proposing 22 release that would be triggered when an individual stock 23 experiences a decline of 10 percent or more from the previous 24 day's closing price. Second, when the circuit breaker is 25 triggered, short sales would be subject for the remainder of 118 1 the trading day to the exchanges' modified bid tests that we 2 described to the Commission in a joint letter dated March 3 24th, 2009. These measures are in Nasdaq's view not 4 separable. 5 As discussed in the morning panels, circuit 6 breakers are preferable to price tests because price tests 7 are significantly over-inclusive in application. By 8 contrast, circuit breakers impose such costs only when a 9 potential problem may exist which could be in over 10 percent 10 of the S&P 500 on a stress day. While these costs will prove 11 unwarranted in some percentage of circuit breaker events, the 12 circuit breaker eliminates unnecessary compliance costs in 13 the up to 99 percent of trading where trades can proceed 14 safely without order by order short sale restrictions. 15 After a circuit breaker is triggered, the follow-on 16 restrictions imposed by the exchanges' modified bid tests 17 will be superior to the traditional tick test and the 18 modified tick test that the Commission proposed. Under our 19 modified bid test, short selling can only occur on a passive 20 basis at a price materially above one tick, the highest 21 prevailing national bid. This type of passive test was 22 mentioned earlier by Dan Mathisson and Rick Ketchum. 23 As such, short sales would only execute at a higher 24 price than the prevailing market. In today's automated 25 markets in which a bear raid can be completed within 60 119 1 seconds as Jim just described, modified bid tests would be 2 more restrictive than either the traditional or modified tick 3 tests due to the vast increase in the number and speed of 4 quotation changes. You now don't have to follow the 5 sequencing under the exchanges' proposal. 6 Consequently, the Commission's traditional tick 7 tests would be less effective now than their antecedents were 8 in 2007. It's important to note that the original tick test 9 of the NYSE was derived for an auction market. Nasdaq bid 10 test was derived for a dealer market. We now operate order 11 driven markets with electronic books. Whatever price test 12 should be implemented needs to take into account the fact 13 that the markets work different in a very fundamental way 14 than they did in the 1930s or in 1994. 15 Again, thank you for this opportunity to present 16 Nasdaq's position on short sale proposals currently under 17 discussion. I anticipate the opportunity to expand on this 18 summary and provide additional empirical evidence to assist 19 the Commission in adopting measures that strike the right 20 balance for today's marketplace. 21 MR. BRIGAGLIANO: Dr. Jones. 22 DR. JONES: I'd like to thank the Commission and 23 the staff for conducting this important information gathering 24 effort and thank you for the opportunity to appear before you 25 today. 120 1 Over the years, many people have studied the data 2 on short selling, and the empirical evidence is remarkably 3 uniform. Research has shown time after time after time that 4 short sellers ferret out overvalued companies and help keep 5 stock prices from getting too high. For example, short 6 sellers were the ones who uncovered the Enron fraud. 7 However, when stock prices go down, short sellers are always 8 blamed. It happened in the Great Depression here. It 9 happens the world over every time stock prices fall sharply. 10 In the current financial crisis, short sellers have 11 been blamed for bringing our financial system to the 12 precipice. But never mind that the fundamentals at firms 13 like Citigroup, Merrill Lynch, Lehman Brothers were 14 absolutely horrible. Short sellers should have received 15 accolades for seeing this ahead of other people. Instead, 16 we're demonizing them. 17 There will always be bad apples, and short sellers 18 are no exception. Based on the record, it appears that 19 manipulative or abusive short sales are quite rare, and I 20 think they can be deterred by aggressive enforcement action. 21 The data clearly say that short sellers in 22 aggregate perform a valuable watchdog function. If you want 23 to further restrict short sellers, I think you need to argue 24 that those data are somehow now relevant, that something is 25 different now than what it was before when all these data 121 1 were collected. From my vantage point, I don't see anything 2 in this particular downturn that renders all that previous 3 data moot. 4 Now, I understand you may have access to some 5 additional nonpublic information on recent actions of traders 6 and investors, and by all means, you should incorporate that 7 knowledge into your decision making. But if you do impose 8 new restrictions on short sellers, I hope you will explain 9 what is different this time around and why it is that we 10 should discard all the evidence to date on the valuable role 11 of shorting in making stock prices as informative as 12 possible. 13 Now, let me talk a little bit about some of the 14 empirical evidence that we have in place on the likely 15 effects of some of the proposed rules. Along with Ekkehart 16 Boehmer and Xiaoyan Zhang, I've studied the 2007 repeal of 17 the uptick rule. It turns out stock price levels and market 18 quality were essentially unaffected by the repeal of the 19 uptick rule. There does not seem to be a change in short 20 interest, either, associated with the uptick rule. 21 This implies that actually reinstating short sale 22 price tests -- and I expect that a bid test would have 23 similar effects to the old uptick test so that those results 24 from 2007 could be extrapolated. I don't think they'll be 25 much effect on stock price levels or liquidity of the stock 122 1 market, and it seems like it would have little effect based 2 on the short interest data on short sellers' ability to amass 3 a short position over the longer term. 4 However, it's important to remember that price 5 tests will impeded shorting considerably. In July and August 6 2007 just after the repeal of the uptick rule, shorting was 7 about 40 percent of trading volume, and based on the evidence 8 around the repeal, about 7 to 15 depending on how you count 9 of those percentage points would have disappeared if we had 10 gone back to having price tests. So that's about one-fifth 11 of the shorting activity that would disappear if we had 12 the -- if we reinstated these price test rules. 13 Now, since short interest doesn't change much, this 14 suggests that the short sale price tests are mainly impacting 15 short sellers with a shorter horizon. Now, we talked a 16 little bit about some of these high-frequency algorithmic 17 traders and liquidity suppliers this morning, and certainly, 18 they're in that category. And so this would certainly affect 19 some of their trades, but also an investor, for instance, who 20 believes that IBM's earnings are going to be short tomorrow, 21 that they're going to fall short of expectations may have 22 difficult achieving her desired short position in the 23 presence of an uptick rule or another price test. These 24 effects will actually -- these will affect real traders with 25 real fundamental views on the prospects for a stock. 123 1 In a second paper, the same co-authors and I have 2 studied the effects of the temporary shorting ban in the fall 3 of 2008. The main thing we find, of course -- and this has 4 been alluded to several times in today's panels -- is that 5 market quality was severely degraded in the stocks that were 6 subject to the ban. Spreads were much, much wider. 7 Volatility was much, much worse. Price impacts and other 8 measures of market quality, severely degraded. 9 So we believe that the shorting ban sharply 10 restricted the activities and informal market markers, and so 11 the proprietary trading desks and hedge funds that do this 12 informal market making were simply unable to provide 13 liquidity. And so market quality was simply horrible as a 14 result. 15 So based on what we saw during the shorting 16 ban -- and again, this was mentioned by a couple of panelists 17 this morning -- I would strongly counsel against any sort of 18 circuit breakers that includes an outright ban on shorting. 19 We are likely to get poorly functioning markets exactly at 20 the time when it's important that markets work well. In sum, 21 I think the Commission should be aware that new restrictions 22 on short sellers are likely to reduce the amount of 23 information incorporated into stock prices. Thus the less 24 often the new restrictions come into play, the better in my 25 view. And in terms of the type of restrictions being 124 1 considered, price tests are vastly superior to any sort of 2 ban on shorting activity. 3 Thank you very much. 4 MR. BRIGAGLIANO: Dr. Shapiro. 5 DR. SHAPIRO: Thank you very much for this 6 opportunity to be here and address these important issues. 7 I'm going to step back and discuss the economic 8 context for some of these issues with regard to one aspect of 9 short sales that has raised calls for more regulation, and 10 that's fails to deliver. Since 1989, investors, scholars, 11 market analysts have urged the SEC to address the problem of 12 fails to deliver of shares sold short which principally 13 consist of naked short sales, and since 2004, the SEC has 14 taken a number of steps to discourage new fails and encourage 15 or require investors and broker-dealers to resolve their 16 outstanding fails, principally through Reg. SHO and its 17 amendments. 18 And these efforts have reduced fails during certain 19 periods, but the number of fails also have periodically risen 20 to record levels. There's also evidence that very 21 large-scale naked short sales and the fails they produced 22 played a role in the sudden and unmanaged collapse of Bear 23 Stearns and Lehman Brothers, not that they caused the 24 collapse but that they affected the manner in which those 25 companies collapsed. 125 1 Despite the measures taken by the SEC, fails 2 continue to persist at levels greater than those which 3 occurred before Reg. SHO. These findings come from a new 4 study I've completed on short sales and fails. Let me review 5 a few of the findings. Current regulation has not stemmed 6 fails in a meaningful, consistent and permanent way. In the 7 first three months of 2008 before the collapse of Bear 8 Stearns, fails on any given day affected almost 4,000 9 companies, averaged more than 1.1 billion shares 10 cumulatively, more than twice the average level of both the 11 previous year and the first 15 months of Reg. SHO. Those 12 fails also were highly concentrated sufficiently to affect 13 the share prices of many stocks under certain conditions. 14 One hundred companies or less than 3 percent of those with 15 fails of more than 10,000 shares accounted for 70 percent of 16 total fails. 17 By March 2008, the 100 stocks with the largest 18 outstanding fails averaged 9.3 million fails each. As the 19 financial and economic crisis unfolded fails increased 20 sharply, reaching more than 2 billion shares in July 2008 21 with an estimated value of $30 billion based on share prices 22 one month before fail soared and helped depress prices. 23 These fails were linked to sharp increases in short sales, 24 including those affecting Bear Stearns and Lehman. 25 From the first quarter 2007 to March 2008, short 126 1 sales of Bear Stearns increased fourfold to 23 million shares 2 while fails to deliver those shares increased 145-fold to 14 3 million shares or 59 percent of the stock's short interest. 4 Despite the emergency measures by the Commission to stem 5 naked short sells, the same dynamics unfolded at Lehman 6 Brothers. From third quarter 2007 to September 2008, short 7 sales of Lehman increased fourfold to more than 100 million 8 shares, and failures to deliver those shares increased 9 151-fold to 50 million shares or 46 percent of the stock 10 short interest. 11 Fails fell in the fourth quarter compared to the 12 historic highs of the crisis. SEC regulations certainly 13 played a role here, especially in ending the option makers' 14 exception to Reg. SHO which reduced fails in optionable 15 securities, threshold securities by over 77 percent. 16 But naked shorts and their attendant fails still 17 remained at troubling levels. Average monthly fails of more 18 than 525 million shares during the fourth quarter were 19 greater than the average quarterly fails in the first year 20 after Reg. SHO was implemented and comparable to levels in 21 mid 2006. 22 While we expect a decline in short interest in 23 fails following a sharp decline in equity prices regardless 24 of regulation, measures other than average monthly levels of 25 fails on a quarterly basis show fails remaining at even 127 1 higher levels in 2008. Using -- in late 2008. 2 Using SEC data to track maximum fails in December 3 2008 for all companies with at least 10,000 fails, we found 4 that despite the new close-out rules and the end of the 5 option makers' exception, these maximum measure of fails in 6 December 2008 reached 885 million shares and maximum monthly 7 fails over the whole fourth quarter averaged nearly 1 billion 8 shares. 9 Moreover, fails have remained highly concentrated 10 even as their total numbers decline. The number of companies 11 with at least 10,000 fails which reached 3400 in four firms 12 in July 2008 fell to 1275 companies by December 2008. 13 However, the top 3 percent of those companies in July 14 accounted for about 74 percent of nearly 1.6 million 15 outstanding fails, an average of 11.6 million fails each for 16 the top 3 percent. In December, the top 3 percent of 17 companies with at least 10,000 outstanding fails accounted 18 for 79 percent of 501 million total fails or an average of 19 9.9 million fails each. 20 We also established that these fails are not 21 particularly concentrated in any sector. They occur across 22 the economy. They are not particularly concentrated on any 23 exchange. They are in the New York Stock Exchange, the 24 Nasdaq and the over-the-counter. They are not particularly 25 concentrated in small, very small companies, in micro cap 128 1 companies. They occur in large cap, medium cap, small cap 2 and micro cap companies. Nor are they concentrated in 3 companies with greater than average insider ownership. They 4 are again distributed as you would expect. 5 Finally -- and we created a database of all 5,500 6 companies that the SEC reported had fails of at lease 10,000 7 shares from January 2007 to December 2008, and then we ran a 8 series of regression analyses. The regression analyses on 9 these data did establish a close relationship between short 10 sales and fails, again across sectors, exchanges, market caps 11 and insider ownership. However, the analysis did not find a 12 close, or significant relationship between short sales and 13 trading volume in these companies, suggesting that short 14 sales are not a critical factor for these stocks' liquidity 15 and that other factors drive the liquidity of individual 16 stocks in the overall market. 17 This relationship was strong and significant only 18 for companies in one sector, the consumer goods sector. 19 There were no significant connections between short sales and 20 trading volume in the other seven economic sectors for stocks 21 listed on the NYSE or traded over the counter for small and 22 medium cap companies. 23 Finally, we tested the incidence and significance 24 of short sale abuse. To do that, we used fails as a share of 25 short interest, as a proxy for short sale abuse. We found 129 1 that overall financial sector companies were 2 generally -- and this is over 2007 and 2008 -- less 3 vulnerable to short sale abuse than companies in other 4 sectors, that this phenomenon is not generally concentrated 5 in finance. 6 Again, using fails as a share of short interest, as 7 a proxy, we found that fails represented 2 percent to 4 8 percent of short interest among financial companies compared 9 to between 4 percent and 11 percent for companies in the 10 other seven non-economic, non-financial sectors. 11 MR. BRIGAGLIANO: Dr. Shapiro, I think we should 12 move along to Dr. Werner. Thank you. 13 DR. SHAPIRO: Okay. Sorry. 14 DR. WERNER: I guess I need to press the button. 15 First, I would like to thank Chairman Schapiro and the 16 Commission for inviting me to participate at this roundtable. 17 I think it's an important event and an opportunity to clearly 18 evaluate the Commission's recent proposed amendments to Reg. 19 SHO. 20 I last participated in an SEC roundtable in 21 September of 2006 as we discussed the empirical evidence from 22 the Reg. SHO pilot. The empirical evidence presented by 23 academic researchers as well as by Commission economists at 24 the time showed that temporary suspension of price tests was 25 associated with a slight increase in short sales but did not 130 1 cause stock prices to fall or volatility to increase. 2 Importantly, there was no evidence of an increase in downside 3 volatility or other return patterns associated with bear 4 raids and reversals. Further, the effects on inter-day 5 market quality measures were limited and mostly associated 6 with asymmetries in order flow created by the specific form 7 of the price test in force. 8 So after evaluating the empirical evidence and 9 considering the public comments as well as the feedback from 10 industry participants, the Commission voted unanimously to 11 permanently suspend the uptick rule and the bid tests in July 12 of 2007. The SEC also prohibited the use of any short sale 13 price tests by SROs in the future. 14 And now in an interesting turn of events, we are 15 back here again, and this discussion is seeking public 16 comment on introducing either a price test that would apply 17 on a marketwide basis and permanently or one that would apply 18 only to particular security during severe market declines in 19 that security. 20 I must say that is unclear to me what the SEC hopes 21 to achieve by these proposed amendments. As I already 22 mentioned, in 2006 we all agreed that price tests did not 23 significantly affect short sellers' ability to execute their 24 short sales, had no effect on prices and volatility and did 25 not cause a decrease in return patterns consistent with bear 131 1 raids and had very limited effects on market quality. 2 More recently, the Commission staff economists and 3 other academics have studied short sales in declining markets 4 and confirmed that price tests do not impair short sellers' 5 ability to execute their orders even in rapid declining 6 markets. Moreover, as I read it, the experimental and 7 existing empirical evidence suggests that circuit breaker, 8 particularly if a company is by a short sale ban, may cause 9 an increase in short-term volatility due to the so-called 10 magnet effect. 11 Even though price tests aren't likely to have a 12 significant effect on markets, in my mind, I am opposed to 13 reintroducing constraints on short sales. I echo Dr. Jones' 14 comments. Academic research shows that short sellers are 15 important contributors to liquidity and market efficiency. 16 In fact, I can think of no academic study that has found 17 short sellers to have a negative effect on markets. 18 Nevertheless, investors, issuers, media 19 representatives and politicians argue that short sellers are 20 responsible for the recent precipitous decline in financial 21 stocks and other stocks. Yet I have found no empirical 22 evidence showing that short sale caused the price declines of 23 financial stocks. Incidentally, I have not found any 24 evidence that short sellers were engaging in fraudulent 25 market manipulation during the recent financial market 132 1 crisis, either, but obviously, I do not have access to such 2 data. 3 Since prices tests and circuit breakers are likely 4 to be relatively innocuous, what's the problem with adopting 5 the proposed amendments you may ask? The problem, as I see 6 it, is that the SEC cannot afford to get sidetracked. 7 Instead, I believe that the SEC should focus its scarce 8 resources on enforcing its existing rules against delivery 9 failures and market manipulation and on continuing its 10 important work towards greater market transparency. 11 I applaud the Commission's regulatory action to 12 eliminate delivery failures by adopting tighter rules on 13 locates and closeouts both in Reg. SHO and subsequent 14 Regulation 204T and also the elimination of the option market 15 maker exemption from the Reg. SHO closeout rules. As shown 16 by the Commission's own staff and academic studies, these 17 regulatory initiatives have dramatically reduced the number 18 of stocks with significant failures to deliver. 19 I'm also pleased to see that the Commission intends 20 to get serious about prosecuting cases against market 21 manipulation through the adoption of the anti-fraud exchange 22 Rule 10b-21. I believe that together with the existing 23 securities law, this should provide ample ammunition to go 24 after market manipulation. 25 Finally, I would urge the Commission to push for 133 1 more transparency in all aspects of short sales. While stock 2 level data on shorting activity is regularly gathered and 3 reported, investor level data on short positions have until 4 recently not been collected. Such data would permit the 5 Commission to regularly monitor short sellers to more easily 6 detect possible manipulative short selling activity. The 7 Commission's move to expand the reporting obligations of 8 13(F) institutions to include reporting of short positions on 9 Form SH is an important step in the right direction. 10 I would also encourage the Commission to require 11 more pre-trade as well as post-trade transparency in the 12 stock-lending market. I believe that this would go a long 13 way towards demystifying the mechanics of short sale 14 transactions. I would also facilitate -- excuse me. It 15 would also facilitate the determination of a fair-market 16 price for stock loans that is based on market rights of 17 lendable shares. Finally, it may encourage brokers, 18 custodians and other stock lenders to pass through a greater 19 proportion of the stock-lending fees to long investors. 20 In closing, I urge the Commission under its new 21 Chairman Schapiro to refrain from adopting short sale price 22 tests and circuit breakers and instead focus on adopting 23 regulation that enhances market efficiency, liquidity and 24 transparency. Thank you. 25 MR. BRIGAGLIANO: Well, thank you, Dr. Werner and 134 1 thank all the panelists for their thoughtful statements. 2 Questions from the Commission? Chairman Schapiro. 3 CHAIRMAN SCHAPIRO: Thanks, Jamie. 4 Let me thank you all very much for being here. 5 It's enormously helpful to us to hear your views, and you 6 will absolutely be informing our process as we go forward. I 7 would just say in response to Dr. Werner that what has 8 changed is that we have just gone through the worst economic 9 crisis since the Great Depression. And while that doesn't 10 dictate a particular outcome, I think it does dictate that we 11 consider whether changes in market structure and in how the 12 markets are operating could be appropriate in the interests 13 of helping to restore investor confidence. 14 And that's really where my question goes. Those of 15 you who were here this morning heard our struggle with trying 16 to assemble some data or some studies or information about 17 investor confidence and what has been the impact of the 18 removal of the uptick rule on investor confidence or what 19 impact reinstatement of some kind of governors whether it's 20 an uptick rule, a bid test, other methodologies, disclosure 21 or other processes could help to restore confidence of 22 investors who might be then willing to come back into the 23 marketplace. 24 And is there data, are there studies that you're 25 aware of, are there things that we could be directing our 135 1 economists to look at that might help us get at that 2 ephemeral investor confidence question that we're so deeply 3 trying to understand here? 4 DR. JONES: I'll take stab at it if nobody else 5 wants to. 6 I think that data is very hard to come by, of 7 course, and I don't know of any studies that are out there. 8 I thought we did have somebody from Charles Schwab this 9 morning, and I think some information about his customer base 10 and how many of them have withdrawn, for instance, completely 11 from the stock market, that might be sort of instructive in 12 telling us what fraction of the populace might have lost 13 confidence in the markets. 14 We could get -- it's possible perhaps to find a 15 Vanguard or some other large asset manager who collects a lot 16 of retirement accounts and things like that and ask them 17 certain questions about what fraction of their people have 18 gone to all cash. I think that's the kind of thing that is 19 knowable, although it's proprietary data. So I think that 20 academics are unlikely to ever have access to it. 21 CHAIRMAN SCHAPIRO: Of course, it doesn't give us 22 the absolute connection to the short sale restrictions. 23 People could have pulled out of the market for a variety of 24 different reasons, so I guess we'll keep searching. 25 DR. ANGEL: I don't know of any particular studies 136 1 with good data on investor confidence, but there are numerous 2 studies on consumer confidence that are based on survey data. 3 So the Commission may want to consider starting a regular 4 data collection process that would actually measure investor 5 confidence. It would require quite a bit of resources, quite 6 a bit of planning, but it could be extremely useful going 7 forward for collecting the data because as you rightly 8 realize, investor confidence is extremely important and yet 9 we don't exactly know but we can feel that there was a big 10 loss of investor confidence. 11 COMMISSIONER WALTER: Professor Angel, do those 12 studies also follow what factors create either increases or 13 decreases in consumer confidence? Because it seems to me 14 that the question we're really focused on here -- I think we 15 can probably all pretty much agree there's been an incredible 16 decrease in investor confidence. But what we want to figure 17 out is what levers will increase investor confidence, and 18 since there are so many variables, I wondered whether we'd be 19 able to track even long-term. Do you have a point of view on 20 that? 21 DR. ANGEL: I don't have a view on the consumer 22 confidence studies. I know the data are collected, but 23 exactly what variables have been found to drive consumer 24 confidence, I'd have to pass on that one. 25 MR. BRIGAGLIANO: Commissioner Paredes? 137 1 DR. SHAPIRO: While I don't know of any American 2 studies, there may very well be -- there certainly are 3 studies going on right now in a number of countries of the 4 impact of new rules which have been placed on short sale 5 transactions throughout this crisis. And I know the 6 Australian stock exchange has been studying the impact of 7 their new rules. There have been a number of studies done by 8 the Hong Kong exchange of the impact of their rules. We have 9 new regulation of short sales in about six or seven European 10 countries. 11 So there are -- while we may not be able to get 12 directly at consumer confidence, we can probably get at the 13 result of that confidence; that is, in actual transactions. 14 So I'd recommend a review of those and conversations with 15 your fellow regulators in Australia and Europe. 16 DR. HATHEWAY: If I may, another dimension of 17 confidence we can look at is our own history. 1987, we had 18 short sale constraints, we had a market crash. That was the 19 previous once-in-a-lifetime experience I've been through. 20 And again in the 1970s, we had a substantial bear market in 21 the presence of short sale constraints with a loss of 22 consumer confidence or investor confidence in both cases. 23 Whether it's a difference of degree or not, it would be very 24 hard to tell. 25 I'd like to differentiate between macro consumer 138 1 confidence or investor confidence which is what we're talking 2 about for the last couple minutes and a micro level which is 3 one of the things that Mike McAlevey talked to the morning 4 panel and Jim Angel brought it up a moment ago. And that's 5 short-term trading events that can damage investor 6 perceptions about the market as a fair game. 7 And again, it's hard to measure the confidence 8 impact of a stock like Dendreon having the trading event that 9 happened to it or Better Online Systems. But that's an area 10 we can focus on. My team does look at these trading events 11 and refer to see information we find over to FINRA and to the 12 SEC as appropriate. And at that level of detail -- and not 13 to speak to any particular event and any particular stock 14 because of the potential enforcement actions that may be 15 involved -- you do see behaviors that one would ex ante 16 expect to be part of a bear raid strategy in terms of the mix 17 of short selling and long selling, what occurs when, these 18 very short time horizons. 19 COMMISSIONER PAREDES: On the question of how much 20 has changed or the extent to which we've just gone through is 21 unique in terms of the amount of stresses and strains and all 22 the rest, the financial system and the economy, it certainly 23 calls for reconsideration of questions along these lines, but 24 it doesn't necessarily mean that the prior economic studies 25 don't continue to obtain by way of their results. 139 1 And so the question I have is, is not withstanding 2 that we can identify this as a unique period, are there 3 reasons that we think the prior studies and their results 4 were not obtained or have obtained during a period like that; 5 that is to say that in a period like this an uptick has more 6 bite than it might in other periods or that it has different 7 consequences in terms of adverse consequences for market 8 quality? 9 DR. ANGEL: The pilot study occurred on during a 10 period that Federal Reserve Chairman Ben Bernanke referred to 11 it once as the great moderation, when volatility in the 12 equity market fell to abnormally low levels. So one might 13 say oh, it was an abnormally quiet time, so it doesn't apply. 14 I disagree because even though it was during the great 15 moderation, there were numerous times during that sample 16 period when individual stocks experienced periods of great 17 volatility and we got the results we did. 18 So I still believe that those results are basically 19 valid in that the old uptick rule really did not provide any 20 benefit to the market. So I think it's pretty clear that 21 going back to the old rule would be a big mistake, but I 22 still think that we need to think about what kind of shock 23 absorber we need now that we have transitioned to all 24 electronic markets. 25 DR. HATHEWAY: I think one of the weaknesses in the 140 1 studies that were done during the pilot -- and this is no 2 reflection on the people who did them. This is a weakness 3 that was acknowledged in most of those studies -- is the bid 4 test rule was not a SEC rule. It was not a national rule. 5 It was an NASD rule, and there was some uncertainty in at 6 least in the minds of certain market participants whether the 7 bid test constraints on shorting and Nasdaq listed stock 8 applied to trading on ARCA or on INET. 9 So there was ample reason to believe ex ante that 10 particular bid test may not have been effective, and so 11 studies of the bid test during the pilot period don't 12 necessarily reflect how a bid test could constrain short 13 selling or what its costs or benefits might be at a national 14 level. 15 DR. WERNER: If I may chime in, I have looked at 16 some of the studies that have more recent data to try to 17 evaluate whether an uptick rule would work as a brake against 18 market decline in severe stress. And my reading, although 19 these are not my own studies, is that the evidence suggests 20 that it's not effective. In fact, I believe that some of the 21 Commission's own economists have found that the uptick rule 22 was less effective when needed most during panics that drive 23 prices down and volatility up. And even with the delays that 24 may be involved in the execution strategy; that is, waiting 25 for an uptick, there's enough volatility to execute the short 141 1 sales if the short seller so desires. 2 And this type of evidence has been found by 3 academics independent of the Commission as well in looking at 4 negative earnings announcements and looking at the 5 participation of short sellers in control versus pilot stock. 6 That's back to the Reg. SHO, the period during the pilot, but 7 also found that the uptick rule did not work as an effective 8 brake. 9 DR. HATHEWAY: If I could respond to that for a 10 second -- go ahead. I'm sorry. I already had my shot. 11 DR. SHAPIRO: No. It's okay. But this may not be 12 very helpful, but the truth is we don't know. We think 13 that -- we assume that the markets will behave essentially 14 after this crisis the way they did before. We don't know 15 that. This is a very large discontinuity. We'll have to 16 see. 17 I think that the prudent course at this time is to 18 identify those factors which played some material role in the 19 unfolding of this financial crisis and address those and then 20 observe how those changes in the regulatory topography affect 21 the behavior of markets which arise out of this crisis. I 22 think it is a mistake to assume either that markets will 23 behave just as they did before or that markets will behave 24 significantly differently. We're going to have to see. 25 DR. HATHEWAY: If I may, the point I wanted to make 142 1 in response to Ingrid is I wanted to draw a distinction 2 between the ability to execute a short sale which we think 3 there's nothing wrong and the ability to artificially push 4 down prices which we think there is something wrong with it. 5 And one of the distinctions between the exchanges' modified 6 bid test and the one in the proposing release is the ability 7 in the proposing release bid test to hit bids when it's an up 8 bid. In the exchanges' proposal, you cannot hit the bid, so 9 you have to be on the passive side of the trade. 10 In addition to making compliance potentially easier 11 because you don't have to follow the sequencing as was 12 discussed this morning, you do force short selling into a 13 more passive role if the goal is price manipulation, not so 14 much the execution of a short sale. 15 One other point on that, the Commission staff can 16 see who's aggressive and passive. I can see it. My 17 colleagues on this panel cannot. So as you talk about 18 requesting empirical results and research making available to 19 the academic community the information on aggressive and 20 passive selling at the level known by the exchanges and the 21 Commission has some advantages as opposed to using something 22 called a Lee and Ready algorithm which is basically a 23 scientific guess. 24 DR. WERNER: Thank you, Frank. 25 CHAIRMAN SCHAPIRO: Your request is noted. 143 1 Let me ask a question. We talked about this a 2 little bit this morning but -- and I said one of the benefits 3 of the Commission putting out so many different ideas was 4 that people have almost taken it as a menu of things and 5 combined different components of the Commission's several 6 proposals. 7 And one of the things that's been interesting to me 8 is to see the idea that the circuit breaker latched onto by a 9 number of people but with the circuit breaker triggering 10 different potential outcomes. One would be a halt on short 11 selling for a period of time, maybe till the end of the 12 trading day, maybe for several days thereafter. The other 13 would be that it triggers a price test or some 14 other -- either a tick test or a bid test. And the other is 15 that it might trigger a hard borrow requirement in that stock 16 for some period of time. 17 And I wondered if you all had any thoughts on the 18 relative merits or demerits of any of those approaches, 19 assuming -- understanding that some of you don't like the 20 idea of anything at all in this case. 21 DR. HATHEWAY: May I? One of the things that 22 struck me about this morning in discussion about the circuit 23 breaker, particularly for those commentators that didn't want 24 one, so this is a chance for my colleagues to think about 25 what I'm going to say after I've said it rather than before. 144 1 If you don't have it, what else in terms of how the trading 2 community is going to behave? We have seen an emergency 3 order with very, very little notice. So if there is no 4 constraint on short selling in place and we have a market 5 break, what goes through the minds of market participants as 6 they plan their trading strategies? Is it not perhaps the 7 same thing or maybe even more severe than the losses of 8 liquidity we heard about this morning in the event of a 9 circuit breaker going off? 10 So I think that the analysis this morning was 11 missing that part of the game theoretic analysis -- yes, I 12 was at Eric's going-away last night -- that needs to be done 13 when you think about this process. 14 CHAIRMAN SCHAPIRO: Just so I understand, are you 15 talking about just predictability, people knowing what the 16 rules are -- 17 DR. HATHEWAY: Exactly. 18 CHAIRMAN SCHAPIRO: -- before the event hits and 19 then everybody's scrambling instead of understanding at this 20 level, this will happen? 21 DR. HATHEWAY: Right. And Rick spoke to also this 22 morning the need for the Commission to have some flexibility 23 within this approach. One nice thing about a circuit 24 breaker, 10 isn't magic. If you get on -- the rule is 25 written in such a way that 10 can be changed during times of 145 1 stress in terms of circuit breaker level. We code to it. We 2 know it. It's a possibility, and we can take that into 3 account. Hard rules, you don't necessarily have that 4 flexibility to that. 5 DR. JONES: I want to reiterate that I think a 6 complete ban on shorting after a certain level of decline is 7 sort of the worst thing we can do to a stock that is down. I 8 think that stock really needs liquidity at that point. It 9 needs maximal participation, and if you put these rules 10 in -- if you put a total shorting ban in effect on a stock 11 that declines by a certain amount, you're going to have 12 withdrawal of participation in that stock. And so I think 13 it's going to have exactly the reverse effect than what you 14 hope to have happen there. 15 I guess I think most people here are pretty 16 comfortable -- maybe I won't speak for everybody, but it 17 seems like the pre-borrow and locate requirements that are in 18 place right now are working well. I'm not sure I see a need 19 for making those contingent on a certain amount of a decline 20 or not. It seems to me like we could impose some of those 21 requirements all the time, and that would work just fine 22 which I think leaves us with your last proposal which is some 23 sort of price test in the event of a certain amount of a 24 decline would I think be the least damaging of any 25 restriction that you could put into place. 146 1 COMMISSIONER WALTER: There was conversation 2 earlier today about a variety of proposals, but one thing 3 that we have not yet talked about which I have heard some 4 people discuss is the notion of at least in less liquid 5 stocks actually mandating that they only be available to be 6 borrowed once. Do you believe that that has any merit in 7 terms of tightening up the system? I throw that open to 8 anybody. 9 DR. ANGEL: I think it would be very hard to 10 monitor and implement that in that if somebody owns a 11 stock -- I buy a stock, part of my ownership right is I can 12 lend it out. Now, the fact that I happened to have bought 13 that stock from a short seller, I don't know that. Our 14 markets are anonymous. So how can you tell me that I bought 15 a stock and I can't lend it out when I don't even know who I 16 bought it from? Especially given the netting that occurs in 17 the CNS settlement system, I don't see how a system like that 18 could be easily monitored and enforced. 19 What you could do, though, is you could have a 20 limit on the total amount of stock lending. But again, I 21 would be very hesitant to put such a system into place. 22 I think Rule 204T has done an excellent job of 23 removing most of the settlement failures. If you look at 24 what's left on the threshold list -- I looked yesterday. 25 There were two companies on New York, one operating company 147 1 on Nasdaq. But there were about 40 approximately exchange 2 traded funds, and this, I think, is the big unsolved problem. 3 And that's one of the reasons why Dr. Shapiro's getting such 4 high numbers for the fails because you do see some very large 5 fails in the exchange traded funds, and that's an issue that 6 the Commission really hasn't addressed. 7 Now, the ability to short exchange traded funds is 8 absolutely essential for their functioning. And my broker 9 was telling me yesterday I could not short a bunch of those 10 ETFs which happened to be on the threshold list. So there is 11 a market malfunction going on here. Does it have to do with 12 the pre-borrow requirement, the locate requirement or is 13 there some other friction in the ETF creation process? I 14 tend to think it's the latter, but I do think the Commission 15 needs to look at the ETF issue separately from the rest of 16 the stocks. 17 DR. SHAPIRO: I do think that Dr. Angel is correct 18 that a rule of lending only once would interfere with the 19 ability to lend stock which has been bought for the purchaser 20 of a short sale. However, a hard restriction in the DTCC 21 stock borrow program, that the same shares can only be lent 22 once, I think would make a great deal of sense since there is 23 evidence of churning of shares which are hard to locate in 24 small companies through the stock borrow program. 25 COMMISSIONER CASEY: Professor Angel, I want to 148 1 pick up on something that you mentioned earlier which was 2 your belief that price discovery mechanism was broken. And I 3 was hoping you all could comment on your view as to we 4 understand the value of short selling in ordinary times in 5 terms of obviously, the correcting function it plays in 6 overvalued stocks and the efficiency it brings to the market. 7 Is there something unique or different about short 8 selling behavior in extreme market conditions that we should 9 be concerned with? Does behavior change? 10 DR. ANGEL: Well, our markets have changed. We've 11 gone from markets in which humans traded with humans to one 12 in which computers trade with computers, and things happen so 13 much faster now that the damage can occur literally within 14 seconds before any human can possibly stop the stock. And so 15 I think that's one of the reasons why we do need some kind of 16 shock absorber that kicks in. Now, exactly what form it 17 should look at -- or what form it should look like remains to 18 be determined. But I think we need a broad public debate on 19 how to deal with it. 20 But I'd like to state that the price mechanism 21 works very well most of the time. But we do observe these 22 situations like Denedron where it broke, and it doesn't take 23 many of those situations to destroy investor confidence. 24 COMMISSIONER CASEY: And you're saying that may or 25 may not be attributable to short selling. It could be some 149 1 other operational issue or something like that that would 2 have the same consequence. 3 DR. ANGEL: Exactly. I mean I don't have access to 4 the data, but, for example, a large investor who sets up the 5 algorithm to get rid of a large block thinking that they're 6 going to dribble it out 100 shares at a time over the next 7 week may put the wrong time frame into the algorithm and 8 suddenly dump a million shares instantly. Something like 9 that could also trigger a stock to just fall out of bed on 10 short notice. 11 COMMISSIONER WALTER: Following on that, in our 12 first panel this morning with two issuer representatives, 13 they expressed concern which I read and I'm extrapolating 14 from their remarks, so maybe I've misinterpreted them. But 15 they expressed concern about what I would call the micro 16 distortion on their particular stock which follows on in that 17 idea. 18 And one of the things that occurred to me in 19 listening to them was whether we've sort of overlooked that. 20 We seem to in this analysis be talking a lot about the health 21 of the market mechanism, what keeps the market going as a 22 whole, but there is also an important value to be served by 23 trying to make sure that there aren't things that are 24 inappropriately disrupting the price discovery mechanism on a 25 company by company basis. 150 1 And I wondered if you could react to how we would 2 go about taking that into account in our analysis here in 3 determining what to do. 4 DR. HATHEWAY: I'll take a jump at that. Maybe 5 pick up Jim's question as well. 6 In times of market stress, at a macro level, micro 7 level, the opportunities to short under the two proposals, 8 uptick or up bid, don't change much. They decline a little 9 bit but not significantly. What does change is the 10 aggressiveness of short sale, and we're going to do something 11 that's much more detailed than I have with me today which at 12 this point only covers about 12 days. But effectively, the 13 aggressiveness of short selling picks up about 5 percentage 14 points of total volume as the market goes down. Now, what we 15 have differs from OEA's study that they did about the market 16 break so that's one of the reasons I want to be a little 17 cautious here. 18 Now, to your question, Commissioner Walter, about 19 the micro level. We have in place an approved rule to put in 20 an electronic circuit breaker for trading on Nasdaq. That's 21 a little bit of a problem because it's Nasdaq, and 22 we're -- it's not a national market system. So I think if we 23 can engage on a national solution to the question of sudden 24 price swings in the security that could only be addressed in 25 automated context because, as Jim said, the events of last 151 1 week, we acted as fast as we can act with human speed, but 2 the damage was done in 45 seconds. You need something the 3 machine can deal with, and as I like to joke with my friends 4 at general counsel, that means we all have to cede authority 5 to the little black box and the programmers which people 6 aren't necessarily comfortable with. 7 COMMISSIONER CASEY: When you talk about price 8 swings and this was a question about whether we should be 9 concerned about not just extreme price swings on the downside 10 but on the upside. 11 DR. HATHEWAY: That's definitely. In both cases, 12 you want to tell a story or have a theory about how someone 13 makes money out of it, and I'm not going to sit here and spin 14 my own manipulation stories because we are being recorded and 15 I don't want to give anybody an idea that doesn't already 16 have one. But, yes. 17 COMMISSIONER CASEY: And one other follow-up on the 18 issue of bench-marking volatility if that were the goal which 19 was to be concerned about creating a shock absorber for 20 excessive inter-day volatility. Obviously, there were 21 restrictions around the world which I think have been 22 mentioned. How much inference or conclusions should we take 23 that there's been significant volatility worldwide and we saw 24 significant stock price declines across the world even with 25 jurisdictions that had short sale price restrictions or not 152 1 even price restrictions but just restrictions generally? How 2 much inference or judgment should we take about -- 3 DR. SHAPIRO: Well, let me just say one thing and 4 that is that Hong Kong which has one of the strictest short 5 sale regulatory regimes, including a hard pre-borrow and an 6 uptick rule, very broad disclosure and auditing requirements, 7 they experienced much less volatility than other markets. 8 That's only one example, but it is probably the strictest 9 regime in the world. 10 DR. HATHEWAY: In terms of volatility halts, the 11 Nordic markets which the OMX part of Nasdaq OMX, have 12 volatility halts. They were hit often during the crisis, and 13 these are trading halts. These are not short selling halts 14 because there's no short selling constraints in the Nordic 15 markets. And during the crisis, the exchange and the 16 exchange community and the regulator widened those thresholds 17 rather than narrowing them. So that's one of the downsides 18 of thresholds is they're conditional on what's happening in 19 the market at that time, and they can actually become an 20 impediment if they're too tight. 21 In terms of the overall decline of the market, they 22 really didn't make much difference. The Nordics had a tough 23 time. The Icelandic, the Swedish banks had a large exposure 24 to the Icelandic banks. A lot of stress on the financial 25 sector securities, so the market certainly went down. But 153 1 that's a macro event. 2 Circuit breakers are really designed for micro 3 events, and so as such, they didn't perform as desired and 4 the exchange moved to eliminate them because they were 5 effectively getting in the way of trading. 6 I think one of the things our markets did 7 exceptionally well, again harking back to '87, they kept 8 going. The morning of the 20th, the second day of the crash, 9 our markets basically shut down. They were open, but they 10 weren't trading. It was the highest stress, boring period 11 I'd ever had in my life, just sort of waiting for the next 12 trade on the floor of the Philly. That did not happen this 13 time. For better or worse, we kept going, and I think it 14 provided an important signal to the public that the markets 15 were still open. If you wanted to raise capital, you could 16 sell equity at the price you saw on the screen. And if you 17 thought this was the greatest buying opportunity you'd ever 18 see, you could do that as well. 19 DR. JONES: I think it's actually important to 20 point out that most of these circuit breakers are really 21 trading halts in most of these other markets, and so in some 22 sense, they're not really apropos as to what we're 23 considering here. And so I don't think, for instance, that 24 we can look to any of the evidence anywhere else on the 25 presence of a magnet effect to tell us whether there might be 154 1 a magnet effect here because I think the events -- a trading 2 halt is very different than what we're considering here as a 3 change in who might be able to short sell and when. 4 And so I personally don't think that we are likely 5 to see any sort of magnet effect if we impose a circuit 6 breaker just because I don't think any of the restrictions 7 that we are contemplating are all that severe other than a 8 ban. 9 DR. ANGEL: And I'd like to echo that and get back 10 to Commissioner Walter's previous question about the issuer 11 concerned about the short interest in his company. There are 12 really two legitimate concerns that issuers have about short 13 selling. The first is the short term, the micro concern 14 we've talked about that our whole discussion of the uptick 15 rule is all about. It's, yeah, short selling is a good 16 thing, but too much too fast can overwhelm the liquidity in 17 the market. So the first concern is the liquidity exhaustion 18 argument that if we get too much too fast, it'll muck up the 19 market mechanism. 20 The second concern is a longer term concern in 21 which you have these companies that have had very high levels 22 of short interest and they're engaged in long-term battles 23 with the shorts and they -- you know, accusations fly left, 24 right and sideways about who's manipulating, who's violating 25 what rules. And there what we're discussing here with 155 1 respect to short selling and the uptick and circuit breakers 2 won't do anything for the short and distort problem that 3 causes many issuers to file many complaints with the SEC. 4 One of the issues there is that because of the tax 5 treatment of short selling, there's actually a strong 6 disincentive for short sellers to cover their positions. 7 They've already gotten their collateral back after the stock 8 has gone down, but they don't have to realize a gain for tax 9 purposes until they actually cover. So many of them have a 10 financial incentive to stay short forever if they can. And I 11 think that's one of the things that leads to these long-term 12 battles, but I don't think the Commission can really do 13 anything about that. 14 COMMISSIONER PAREDES: One of the phrases that was 15 used earlier was the kind of micro side of this and you're 16 talking about damage occurring. And I guess my question, 17 though, is if you have a circumstance where the, quote, 18 damage occurs but then it bounces back, right, and so at some 19 point people realize the stock is undervalued for some reason 20 or another. People come in, and so ultimately, in that 21 scenario, that scenario is the one that obtains. You could 22 say fundamentals win the day, although it may take a day or 23 whatever it happens to take. 24 The other possibility, of course, is that you have, 25 quote, the damage occurring. You have the precipitous drop, 156 1 and it stays flat which is to suggest perhaps that's actually 2 also reflecting the fundamentals. And so the fundamentals 3 again are perhaps winning the day. 4 I'm just curious about depending upon what happens 5 in the hours or days that follow, are we talking about a 6 damage occurs, a damage persists? How do we dissect whether 7 that's because of an abuse, whether that's something an 8 uptick would have done something about, whether it ultimately 9 is a reflection of the fundamentals? And another way of 10 asking that in some part is, is to what extent the market 11 itself is, in fact, self-corrective at least over the longer 12 term, if not in a moment-by-moment respect? 13 DR. HATHEWAY: I think the market again -- now with 14 the ex-trader hat on, I think the market is self-correcting. 15 Not atypically for an economist, I think markets are 16 efficient. Perhaps a little atypically, I think they're 17 efficient because people make them efficient. It doesn't 18 just happen automatically because people assume it can't be 19 inefficient, therefore it says efficient. No, it gets out of 20 line, and somebody pushes it back. 21 The way I would normally go about considering a 22 aberration in the market is where you have a sharp price 23 change and an equally rapid -- close to equally rapid price 24 recovery, particularly in the absence of news. And I think 25 the harm in those situations comes from the different 157 1 constituencies that may be taking part in these events as 2 they unfold. 3 DR. JONES: I think they're actually -- there's 4 another kind of manipulative activity that we should be 5 concerned about. It's not just these short-term reversals. 6 It's a sharp decline followed by some other -- and that sharp 7 decline forcing some sort of corporate act that is taking 8 place at these wrong prices. So for some reason, forcing an 9 equity issue at these abnormally low prices. That, I think 10 is actually -- in some sense, these reversals are just about 11 transfers between investors, but if we're starting to -- you 12 know, we're starting to liquidate a company that should be 13 liquidated or a company is raising equity at the wrong value 14 or making investment decisions based on a stock price that's 15 at the wrong level, that's the kind of manipulation I think 16 that we should be fundamentally, first order concerned about. 17 COMMISSIONER PAREDES: Can I just ask a quick 18 follow-up on that, though? Which is if I'm in a boardroom, 19 I'm not going to make that decision over a two-minute period 20 of time, right? And so if we actually make that decision 21 because we've seen the stock price stay flat for some 22 extended period of time, how do we separate out whether 23 that's, in fact, what's the company's worth versus some 24 persistent reflection of some manipulation that actually 25 distorts the market and gets us away from the fundamental 158 1 value? 2 DR. JONES: I totally agree with you. It's often 3 hard to tell those two things apart, but I guess what I have 4 in mind is something where the stock price falls rapidly and 5 then the board is forced to do something either because they 6 issued one of these death-spiral convertibles or there's some 7 other -- there's some contract out there that forces 8 something bad to happen to this firm. So that's sort of what 9 I have in mind here, where somebody is trading really with an 10 intent to impose this externality on the firm in this bad 11 way. 12 DR. SHAPIRO: Let me say there certainly is a lot 13 of evidence that the kind of manipulation that Dr. Jones just 14 referred to and described, it may not be common, but it's not 15 rare, either. 16 Let me also say and maybe this is -- you know, this 17 is - we talked before about whether our approach to this 18 should be the same after the crisis as it should be before. 19 I have to say as an economist that it may be that on average 20 markets are efficient and it may be on average markets are 21 functional, but our markets are clearly, regularly very 22 inefficient and regularly very dysfunctional. 23 The housing market over the last five years was 24 profoundly dysfunctional. The market in mortgage backed 25 securities was profoundly dysfunctional. The market in 159 1 credit default swaps was profoundly dysfunctional. We need 2 to approach these issues not in an abstract way, that the 3 neoclassical model of markets tells us that on average in the 4 long run, all the results will be optimal. We need to 5 instead look at each market, look for the dysfunctions or the 6 efficiencies, address those dysfunctions and not be satisfied 7 with a kind of fundamentalist approach that says that in the 8 end the results will be optimal because the results have not 9 been optimal. The results have been fairly catastrophic for 10 tens of millions of people. 11 And this is a very good point historically to 12 recognize that and figure out in what respects we should 13 follow the course we've been following and in what respects 14 in order to protect the economy and the markets we need to 15 follow a different course. 16 COMMISSIONER WALTER: Let me follow up on that and 17 come back to something that Professor Angel said because it 18 seems to me that there may be -- and I don't know whether 19 this is right or wrong, but I'll throw this out as an idea, 20 something different today. You talked about the short-term 21 and long-term concerns of issuers. And I'd like to challenge 22 that just a little bit in terms of whether the long-term 23 concern or the what you call the short and distort -- it has 24 a nice ring to it -- isn't also -- hasn't also turned into a 25 short-term concern because you can see an issuer sitting out 160 1 there. It has as a result of benign or malevolent market 2 pressure been hit extraordinarily badly in a Dendreon kind of 3 way or perhaps not quite that badly, but there is more of 4 a -- as soon as it goes out of sync with the marketplace 5 today because people are so attuned to coming disaster, you 6 could see it taking a long time for an issuer like that to 7 recover the confidence of its shareholder body and its 8 potential shareholder body. And to me, that becomes also 9 a -- perhaps a short-term concern and it may be that some of 10 the things that we're talking about today with limited 11 intrusion into trading strategies and the overall good flow 12 of the marketplace could perhaps address some of those things 13 in some sort of a circuit breaker kind of way unless the 14 intent, the taint associated with the circuit breaker is 15 enough to create the same sort of problem again. 16 Does anybody have a reaction to that? 17 DR. HATHEWAY: I don't really think there's a taint 18 any more than there is around any other sort of halt. I mean 19 it's something that happens probably to every company at some 20 point in its life. So in that sense, whether it's a circuit 21 breaker regarding the introduction of short selling or it's a 22 circuit breaker about stock level volatility that halts 23 trading in that stock, I don't think there's a taint. 24 I think part of the challenge operationally from 25 looking into this ourselves, different stocks have different 161 1 thresholds, and that's a bit of a problem to communicate into 2 the investing community simply because it's complicated or 3 you take a simple heuristic like 10 percent knowing that 4 certain types of stocks. Low-priced stocks, for example, 5 it's not going to work very well for you. You have to come 6 up with something different. 7 DR. JONES: I think it's also worth pointing out 8 that everything we're talking about here in terms of 9 manipulation is really -- we can tell a story. We can talk 10 about short and distort as short selling being the form of 11 the manipulation, but almost every story we tell, we can turn 12 it around and there's a long version of that same 13 manipulation story. And so we really should be concerned 14 about both of those equally, I think. And I think it's wrong 15 to really single out the short side of this for special 16 consideration. 17 COMMISSIONER WALTER: That may be right, but I 18 think on the upside, we've at least been somewhat successful 19 in going after the people who have done this, finding them, 20 determining that it's happened, not always. But there's 21 enough of it that it self serves as a deterrent for those 22 people who are at least somewhat attuned to obeying the rules 23 not to do it. And on the short side, it seems much more 24 complicated, at least to me. 25 DR. JONES: Do you think you have a harder time 162 1 identifying -- this is probably a question for the 2 enforcement staff, I think. But it sounds like it might be 3 harder for you all to identify this kind of manipulative 4 activity on the short side. 5 COMMISSIONER WALTER: Let me say I'd have a hard 6 time identifying either and leave it at that. But yes, it's 7 a question for the enforcement staff. But I do think there 8 is certainly a difference. Maybe we're more attuned to 9 upside manipulation, but if you look historically, look at 10 the number of pump-and-dump cases that have been brought. 11 And all of these things are complicated in an evidentiary 12 sort of way, and maybe it's just because of the greater 13 sophistication in the marketplace. But that's my perception, 14 at least at the moment. 15 DR. JONES: I guess another possibility is that the 16 short and distort stuff just doesn't happen as often. 17 CHAIRMAN SCHAPIRO: I wonder if you would agree 18 that one of the other differences between the long side and 19 the short side is that there are many more almost Draconian 20 actions that a corporate issuer will take based on their 21 stock price being depressed for a long period of time. So 22 they might be subject to a takeover. They might feel they 23 have to do a partnership with someone. They might close 24 factories or lay-off employees. I mean, there are a lot of 25 issues I think that have more profound impacts throughout the 163 1 economy from the short side than there are from the long 2 side, at least that's been my experience. 3 DR. JONES: I agree that there are these 4 existential things tend to happen more on the short side than 5 they do on the long side, but I think there are plenty of 6 value reducing strategies that happen on the long side. You 7 just have to think back to the Internet bubble and Pets.com 8 and all the Super Bowl ads it bought with its silly sock 9 puppets, right? If we had had -- if we'd basically had a 10 more active, I think, short selling brigade to sort of rein 11 in those kinds of firm, I think we might not have had those 12 value reducing strategies. But your point's well taken, I 13 think. 14 DR. WERNER: Also, you can think about in a 15 pump-and-dump scheme, a corporation might choose to issue 16 stock at an inflated value, thus destroying investor value as 17 they purchase the stock. So you can think about losses 18 either way. So I agree with Charles that symmetric treatment 19 would be really something to really consider here. 20 DR. ANGEL: And, however, I'd like to point out and 21 make an argument for asymmetric treatment for the following 22 reason: that with a -- if you look at the real economy, if a 23 stock is undervalued, then the -- as you point out, the 24 company may be forced to take actions. They may be unable to 25 raise financing on affordable terms. Employees who are 164 1 motivated by stock options may leave the company. You have 2 customers may walk away. And the short sellers have a 3 financial incentive to interfere with the operations of an 4 operating company, and this is one of the reasons why issuers 5 have such a visceral loathing of short sellers. 6 Now, I'm a great defender of short sellers. That, 7 personally, I don't think they caused our debacle. I 8 actually bought Lehman Brothers a week before they failed as 9 a speculation they would turn around. With retrospect -- in 10 retrospect, I wish there'd been more short selling. If 11 there'd been more short sellers who pushed the price down 12 further faster, I would have lost less money. 13 So I'm a great defender of the short sellers, but 14 if a stock is truly selling below its true value, it can 15 cause damage to the real economy. And that's why we need to 16 pay attention on the short side. 17 DR. HATHEWAY: I'll pick up on the symmetry idea 18 just for a second. As was said this morning and by 19 Dr. Shapiro on this panel, disclosure of short positions, I 20 think, is well warranted. 21 COMMISSIONER CASEY: That was going to be my -- I 22 mean just how much do we buy and increase transparency in 23 terms of addressing the concerns that have been raised today? 24 I know Dr. Werner has suggested that -- and we've heard from 25 other panelists as well that enhanced reporting and 165 1 disclosure would be valuable. I guess I'm trying to 2 appreciate how much do you think that actually contributes to 3 addressing some of the perceptions and perceptions about the 4 role that short selling is playing? 5 DR. SHAPIRO: Well, it depends on what the data 6 show. The data could undermine confidence as well as support 7 it, but certainly, as a general proposition, the more 8 disclosure, I think, generally, the more efficient the market 9 is likely to be. There are enormous significant distortions 10 that arise from asymmetries of information. Some people have 11 access to it; others don't. And I think this would reduce 12 those as well increase the general level of information. And 13 it would make the price discovery which occurs under short 14 sales more valuable, frankly. 15 DR. ANGEL: Whenever you have a lack of 16 transparency, you create room for the conspiracy mongers to 17 think that something bad is going on. So with additional 18 disclosure on how much short selling there is, how much 19 interest there is and what level of settlement failures we're 20 experiencing, I think that transparency will do a lot to 21 restore investor confidence in the market. 22 And if there is a problem, then the fact that the 23 data are out there and we can all look at the data, I think 24 it will really help the SEC in its enforcement activities to 25 essentially deputize the general public to troll through the 166 1 data and find the problems. 2 DR. HATHEWAY: That's an interesting distinction 3 perhaps between real time and exposed transparency. So the 4 fail data was terrific when the Commission began to release 5 that with a one quarter lag. Still having it provides firms 6 and people such as myself and the others on this panel the 7 opportunity to really look at some trading events and try and 8 figure out what's going on as well as to identify strategies. 9 Real time, you do run the risk of information 10 overload. So there's disclosure and there's informing the 11 investing public, and I think that balance needs to be struck 12 because as Jim and I were talking about at lunch, I could see 13 certain types of real disclosure -- real-time disclosure 14 being used as part -- yet a different form of a manipulation 15 strategy to try and create an appearance that may not 16 necessarily be so. 17 MR. BRIGAGLIANO: That brings us to the end of the 18 roundtable unless the Commission has further questions? 19 On behalf of the staff, we'd like to thank all the 20 panelists for their insights, candor and really the enormous 21 effort that it took to be here. Your participation is 22 crucial. 23 I'd like to turn the program over now to Chairman 24 Schapiro for any closing remarks she may have. 25 CHAIRMAN SCHAPIRO: Thanks very much, Jamie. 167 1 Before closing today's roundtable on short selling, 2 I really want to extend my sincere thanks and those of my 3 colleagues on the Commission to all of you and our panelists 4 this morning. We so appreciate your taking the time, some of 5 you to travel to Washington from other places to be here and 6 really help inform our decision making in this really 7 critical area. We have a lot to think about based on today's 8 conversations. 9 And we also want to thank my commissioners for your 10 insightful questioning throughout the day. 11 We are charged with ensuring that the securities 12 markets operate within a regulatory framework that promotes 13 efficiency while protecting investors from the potential 14 abuses and the manipulation that can cause them to lose faith 15 in our very well proven system of capital formation, and we 16 take that charge very seriously across the entire spectrum of 17 securities regulation. And the current debate over short 18 selling policy is certainly no exception. 19 So for that reason, we are committed to closely 20 reviewing the potential benefits and the costs of our 21 recently proposed amendments to Reg. SHO and their overall 22 impact on investor confidence. I do think that today's 23 candid discussion of highly informed and differing viewpoints 24 will go a very long way in assisting the current rulemaking 25 process. 168 1 So in that regard, I would note that the public 2 comment period for the proposed amendments to Reg. SHO will 3 run until June 19th. And I would encourage those here in 4 attendance but also those of you viewing on our webcast to 5 contribute to this ongoing policy discussion and share your 6 viewpoints on the proposed approaches to short selling 7 regulation. 8 Before we conclude, I want to thank those people at 9 the SEC who also made today's event possible. I'll start 10 with none other than our intrepid moderator Jamie 11 Brigagliano. Thank you, Jamie, for getting us through the 12 day. Also, our short selling roundtable team of JoAnne 13 Swindler, Josephine Tao, Tory Crane, Steward Mayhew, Amy 14 Edwards, Tim McCormick and most especially, Matt Sparkes. 15 And again, thank you very much to all our panelists 16 for all your help today. 17 (Whereupon, at 3:10 p.m., the roundtable was 18 concluded.) 19 20 21 22 23 24 25