0001 1 U.S. SECURITIES AND EXCHANGE COMMISSION 2 3 4 5 6 FIXED INCOME MARKET STRUCTURE 7 ADVISORY COMMITTEE MEETING 8 Amended 2-9-18 9 10 11 12 13 14 15 Thursday, January 11, 2018 16 9:30 a.m. 17 18 19 20 21 22 23 U.S. Securities and Exchange Commission 24 100 F Street, N.E., Washington, D.C. 25 Station Place 1 Multipurpose Room 0002 1 PARTICIPANTS: 2 3 Michael Heaney, Committee Chairman 4 Chairman Jay Clayton 5 Kara Stein 6 Michael Piwowar 7 Robert Jackson 8 Hester Peirce 9 Brett Redfearn 10 Elisse Walter 11 Rachel Wilson 12 Lawrence Harris 13 Mihir Worah 14 Tom Thees 15 Gilbert Garcia 16 Rick McVey 17 Brian Archer 18 Kuman Venkataraman 19 Amar Kuchinad 20 Ananth Madhavan 21 Scott Krohn 22 Carole Brown 23 John Bagley 24 Lynn Martin 25 Amy McGarrity 0003 1 APPERANCES CONT'D: 2 3 Suzanne Shank 4 Horace Carter 5 Daniel Allen 6 Tom Giva 7 Matthew Andresen 8 Kevin McPartland 9 Jeff Meli 10 Sonali Theisen 11 Jim Switzer 12 Richie Prager 13 Drew Mogavero 14 Paul Jakubowski 15 Alexis Rosenblum 16 David Shillman 17 John Roeser 18 Tom Eady 19 Jeff Harris 20 Rebecca Olsen 21 22 23 24 25 0004 1 C O N T E N T S 2 PAGE 3 Call to Order and Opening Remarks 5 4 Review and Consideration of Proposed Bylaws 24 5 Panel 1: Bond Market Liquidity Conditions 25 6 Research 7 Panel 2: Market Participant Perspectives on 69 8 Bond Market Liquidity 9 FIMSAC Members and Panelists Discussion of 121 10 Bond Market Liquidity 11 FIMSAC Members Discussion on Bond Market Liquidity 190 12 Discussion of Committee Next Steps, Future Meeting 214 13 Topics and Subcommittees 14 15 16 17 18 19 20 21 22 23 24 25 0005 1 P R O C E E D I N G S 2 CHAIRMAN HEANEY: Good morning. I believe we 3 have a quorum, so I will call the meeting to order. In 4 addition to the members that we have today in the room, 5 we have Horace Carter attending via teleconference. 6 He'll be on in the morning session, and then I think 7 again rejoining at 3:15, so good morning, Horace. 8 MR. CARTER: Good morning. 9 CHAIRMAN HEANEY: Thank you for joining us for 10 today's inaugural meeting of the SEC Fixed Income Market 11 Structure Advisory Committee. I'm Michael Heaney; I'm 12 delighted to serve as the committee's first Chairman. 13 I'll begin by welcoming Chairman Clayton and asking the 14 Chairman to make his opening remarks. 15 CHAIRMAN CLAYTON: Thank you. Thank you. 16 Thank all of you for being here today. I have to issue 17 our usual disclaimer, which is -- ah, see I knew if I got 18 started -- actually, you three could out-vote me. 19 (Laughter) 20 COMMISSIONER STEIN: What are we voting on? 21 CHAIRMAN CLAYTON: Nothing, nothing. Welcome, 22 welcome. Great, great. Okay. I'm delighted to welcome 23 all of you to the inaugural meeting of the Fixed Income 24 Market Structure Advisory Committee or FIMSAC. This is a 25 significant day for the Commission, and there are a few 0006 1 matters of importance to discuss. I will try to be 2 efficient because I know we're all eager to kickoff 3 today's discussion on bond market liquidity. 4 To start, I'd like to extend a warm welcome to 5 our two new Commissioners, Robert Jackson and Hester 6 Peirce. With Commissioner Stein and Commissioner 7 Piwowar, we have been blessed and benefited from 8 intellect, experience, perspective, energy, as well as an 9 unwavering commitment to our mission. My initial 10 interactions with Rob and Hester have made it clear to me 11 that we will continue to have those characteristics in 12 abundance, so I welcome you both here. I'd also like to 13 extend a sincere thank you to all of you who have agreed 14 to serve on this Committee, and in particular, to you, 15 Michael. 16 I recognize that you all have significant 17 demands on your schedules, and our aim is for this 18 committee to be a valuable use of your time. The 19 insights and recommendations that you will provide will 20 help inform the Commission's policy decisions over the 21 coming months. As many of you know, the Equity Market 22 Structure Advisory Committee or EMSAC has provided the 23 Commission with thoughtful perspectives and 24 recommendations, some of which we have pursued, and 25 others of which we are actively considering as we speak. 0007 1 My hope is that FIMSAC will be similarly valuable to the 2 Commission and to our markets. 3 Speaking of EMSAC, that committee's charter 4 expired this week. Going forward, rather than extending 5 EMSAC, our plan is to organize targeted roundtables on 6 discrete equity market structure issues, which will 7 feature experts on each topic, representative of a broad 8 diversity of viewpoints. We will provide more 9 information on these plans in the future. Having said 10 that, I would like to take this opportunity to thank all 11 of the EMSAC members for their work and many 12 contributions. 13 Turning back to FIMSAC, I'd like to recognize 14 Commissioners Stein and Piwowar for their thoughtful 15 support of this committee and their highly thoughtful and 16 collaborative approaches -- collaborative approaches to 17 its establishment. Thank you, also, to the staff of the 18 Division of Trading and Markets, to the Division of 19 Economic and Risk Analysis, our Office of Municipal 20 Securities and the Office of General Counsel for their 21 work together with Michael in organizing this first 22 meeting. 23 Now, I'm going to take a step back and provide 24 some context to explain why I believe this committee is 25 so important. It is difficult to overstate the 0008 1 significance of the fixed income markets to the American 2 economy and our investing public. First, these markets 3 are massive and growing. The US corporate bond market 4 has experienced significant growth since the early 2000s. 5 Issuance in the corporate bond market has hit record 6 highs five years running. 7 In 2016, almost 1,400 companies issued 1.5 8 trillion of corporate bonds, and there was over 8.5 9 trillion of corporate bonds outstanding. By comparison, 10 in 2006, there was over 4.8 trillion of corporate bonds 11 outstanding. Growth in the US corporate bond markets has 12 outpaced the US equity markets. Between 2006 and 2016, 13 the value of corporate bonds outstanding rose by about 76 14 percent, while the equity market cap rose by 40 percent. 15 The municipal bond market is also large and 16 vital and has experienced significant growth in recent 17 years. By the end of 2016, a total of 31,000 different 18 municipal bond issuers had approximately 3.8 trillion 19 bonds outstanding, up 17 percent from the end of 2006. 20 Second, the fixed income markets significantly impact 21 other markets, both directly and indirectly and vice- 22 versa. For instance, and perhaps the most obvious 23 connection is to the interest rate derivatives market. 24 The notional amount of outstanding OTC interest 25 rate derivatives was approximately 416 trillion in the 0009 1 first half of 2017. The gross market value of OTC 2 interest rate derivatives was approximately 8.5 trillion 3 at the end of June 2017. Obviously, very significant. 4 Third, the corporate and municipal debt 5 markets, which are appropriately the initial focus of 6 this committee, are particularly significant to retail 7 investors. Individual investors are key participants in 8 these markets both directly and indirectly through 9 pension funds and other pooled vehicles. I am very 10 pleased that in recent years, the Commission has focused 11 on retail investor initiatives in the corporate bond and 12 municipal securities markets. I intend for that focus to 13 continue. 14 For the committee members, I urge you to -- I 15 urge you and I -- and I'm going to depart a bit from my 16 prepared remarks, this is what I say to everybody who 17 appears before us for whatever purpose: please put your 18 thoughts and remarks in the perspective of what they mean 19 for the retail investor at the end of the day. The long 20 term interests of the retail investor are what we're all 21 about, and that goes to market efficiency; it goes to 22 capital formation. It goes to -- it goes to fairness and 23 investor protection. The more you cast your views in 24 light of what it means for the retail investor over the 25 long term, the better off we all are. 0010 1 Finally, corporate municipal debt markets are 2 also critical to American companies and our national 3 infrastructure. Companies across the country, both large 4 and small, issue corporate bonds to borrow money to help 5 grow their businesses, fund their operations and create 6 jobs. State and local governments issue bonds to finance 7 a wide range of public projects and provide cash flows 8 for government needs among other things. The size, 9 strength and vibrancy of these markets are part of the 10 reason that our national economy rebounded from the 11 financial crisis more quickly than other countries. 12 I firmly believe this. I believe that the 13 nimbleness of our fixed income markets and the ability to 14 borrow money enabled us to recover more quickly than 15 other similarly situated economies. I will close by 16 noting that I look forward to working with this committee 17 and my fellow commissioners to ensure that our regulatory 18 approach to these markets is sound and continues to meet 19 the needs of retail investors, as well as the American 20 companies and state and local governments. 21 The last comment is I very much welcome your 22 candor and your participation, including if there are 23 things that you don't think we're doing well. That's 24 where, you know, good news is great, but real -- the bad 25 news that's honest and helps us is what we need. So to 0011 1 the extent you have criticisms, don't hesitate to voice 2 them. We welcome them and I look forward to a very 3 productive day and many productive days to come. Thank 4 you. 5 CHAIRMAN HEANEY: Thank you, Chairman Clayton. 6 I'll now ask the commissioners to make their opening 7 statements, starting with Commissioner Stein. 8 COMMISSIONER STEIN: Thank you. I also want to welcome 9 all of the new members of this committee. We're very 10 excited about having you here. I'm going to emphasize or 11 reemphasize what Chair Clayton just said, which is how 12 important it is that you share your candor and your real 13 views. Most of you are out on the edge of the 14 marketplace and you're seeing things that we might not 15 see for quite some time. We won't know -- we won't even 16 know to be thinking about it. So we really do appreciate 17 you bringing those perspectives to us. I also want to 18 welcome our two new commissioners. This is their first 19 public appearance as commissioners. They got sworn in 20 this morning, so that's pretty exciting, Hester Peirce 21 and Rob Jackson, and I think we're excited about, as 22 Chair Clayton was saying, the energy and the perspectives 23 that they're bringing to our commission. 24 I'm not going to go into a lot of detail. I'm 25 going to say some of what Jay said about how important, 0012 1 how vitally important the fixed income markets are to not 2 only our financial system but our economy. As we all 3 know, they allow corporations to raise money, to expand 4 and grow their businesses and they allow savers to 5 invest, so they can pay for things like retirement and 6 children's educations and achieve other financial goals. 7 The fixed income markets have served us well 8 for many years. However, as I think everyone around the 9 table here knows, they're in a period of profound and 10 multifaceted change. Fixed income trading is moving from 11 voice-based to electronic. New players have an 12 increasing role in the fixed income markets, at the same 13 time that the banks have reduced their holdings. The 14 market itself, as Jay said, is much larger. In 2016, 15 there were $8.1 trillion in corporate bonds outstanding, 16 up from just $5.9 trillion in 2010, and these changes 17 have resulted in a multitude of follow-on impacts. 18 Spreads are tighter. Trade sizes are smaller, and 19 liquidity is increasingly concentrated in certain bonds. 20 There's been a great deal of debate about what 21 these changes mean, particularly as they relate to 22 liquidity, the subject of today's discussion. As such, 23 I'm going to reiterate the diversity of viewpoints that 24 you all bring to this discussion is incredibly valuable 25 to us. We, as a commission, need to hear from a variety 0013 1 of stakeholders. Therefore, again, I'm going to say I'd 2 like to thank the committee members and the panelists for 3 your pro bono contributions to this discussion. And we 4 look forward not only to the discussion but any 5 recommendations that you might bring forward in the 6 future, so thank you again, and I look forward to today's 7 meeting. 8 CHAIRMAN HEANEY: Thank you, Commissioner 9 Stein. 10 Commissioner Piwowar? 11 COMMISSIONER PIWOWAR: Thank you. Good morning. 12 Before we begin, I'd like to join Chairman Clayton and 13 Commissioner Stein in welcoming our two newest 14 commissioners, Commissioner Jackson and Commissioner 15 Peirce to the Commission after being sworn in just about 16 an hour ago, I think, by my own account. I'm excited to 17 be working again with my good friend, Commissioner 18 Peirce, and I look forward to working with my new 19 colleague, Commissioner Jackson. It's been approximately 20 832 days since we've had a full commission, but who's 21 counting? (Laughter) 22 The Chairman has set a busy agenda for us in 23 the months ahead, and I look forward to working with my 24 fellow commissioners on advancing the Commission's core 25 mission. It's a great pleasure to welcome you all to 0014 1 this inaugural meeting of the FIMSAC. I'd like to thank 2 Chairman Clayton and Commissioner Stein for their efforts 3 in making this meeting a reality. Since the early days 4 of his chairmanship, Chairman Clayton has recognized the 5 acute need to focus greater attention at the Commission 6 on the fixed income markets and the benefits that the 7 Commission could gain from the insights of an advisory 8 committee in this area. 9 I applaud his decision to form this committee 10 so that we may remain up-to-date with the most current 11 perspectives from industry and academia. It's been a 12 pleasure working collaboratively with both Chairman 13 Clayton and Commissioner Stein to get this committee up, 14 running and populated by an esteemed group of experts 15 representing all relevant constituencies. Thank you also 16 to those committee members who have agreed to serve, as 17 well as today's panelists. When the Commission's Equity 18 Market Structure Advisory Committee began, I challenged 19 its members to take control of the agenda and use their 20 collective expertise to tell us what issues deserved the 21 greatest attention. EMSAC's members rose to that 22 challenge, and their work product has shaped my own 23 thinking, and I think it's safe to say that of my fellow 24 commissioners. 25 I issue the FIMSAC the same challenge. 0015 1 Liquidity is an ideal place to start this committee's 2 conversations today. And while the media may have moved 3 on to other topics, it was not long ago that bond market 4 liquidity, and most especially, the lack thereof was the 5 subject generating the most intense debate and concern in 6 the financial markets. Despite a lower number of recent 7 splashy headlines on this topic, fears of possible 8 liquidity shocks persist and greater depth of analysis is 9 needed to understand what may be appropriate 10 regulatory next steps. 11 I look forward to hearing your perspectives on 12 all dimensions of this debate, and like Chairman Clayton 13 and Commissioner Stein, I hope you will not shy away from 14 rigorous and animated discussions on these issues. Thank 15 you. 16 Thank you, Commissioner. 17 Commissioner Jackson? 18 COMMISSIONER JACKSON:: Well, thank you very much. 19 I'm delighted to be here this morning and wanted to thank my 20 fellow commissioners for their very warm welcome, not 21 only this morning but throughout the last several months 22 of this process. I'll be doing much more listening than 23 speaking today and look forward to learning from the 24 diversity of views, experiences and market perspectives 25 sitting around the table. I want to join my colleagues 0016 1 in thanking you for your service, and I look forward to 2 our work together in the coming months. 3 CHAIRMAN HEANEY: Thank you. 4 Commissioner Peirce? 5 COMMISSIONER PEIRCE:: Thank you, and thank you all for 6 being here and being willing to serve on this committee. 7 It will be a very valuable committee. Too often in the 8 past, it's been easy for the SEC to focus only on equity 9 markets and focus less attention on fixed income, so I'm 10 really supportive of the formation of this committee, and 11 I know that a lot of work has gone into it by the 12 Chairman and by Commissioners Piwowar and Stein, so I 13 appreciate that. And I can guarantee you that the 14 discussions that are held today and throughout this 15 committee's tenure will be extremely valuable for me as I 16 think about fixed income issues, and so I very much look 17 forward to it. And thank you all for the warm welcome 18 today as well. 19 CHAIRMAN HEANEY: Thank you. Next, I'd like to turn 20 it over to Brett Redfearn, Director of Division of 21 Trading and Markets and the Committee's Designated 22 Federal Officer, for his remarks. 23 MR. REDFEARN: Thank you, Michael. On behalf 24 of the Commission, I want to start out also by thanking 25 you and the rest of this committee for your willingness 0017 1 to serve as members. In my former life, I had the 2 privilege of participating as a panelist on three Equity 3 Market Structure Advisory Committee meetings, and I'm sure you'll 4 find, as I did, that the discussions at these meetings 5 are thoughtful and enlightening. I also look forward to 6 working with you on the important work of this committee. 7 Let me briefly introduce my colleague sitting 8 here with me today. From the Division of Trading and 9 Markets, we have right to my right, Dave Shillman and 10 John Roeser, both associate directors in the Office of 11 Market Supervision, and Tom Eady down at the end here, 12 the senior policy adviser in our Office of Analytics and 13 Research. Also joining us, Rebecca Olsen, Acting 14 Director of the Office of Municipal Securities and Jeff 15 Harris, the Chief Economist and Director of the Division 16 of Economic and Risk Analysis on the other side. 17 Before we get started, any views expressed by 18 the staff in this forum are ours alone and cannot be 19 attributed to the Commission or the Commissioners. So as 20 my tenure as director of the Division of Trading and 21 Markets is picking up speed, I'm enthusiastic to devote 22 more resources to the development of thoughtful policies 23 for the fixed income markets, including the corporate 24 bond and municipal securities markets, which will be the 25 initial focus of this committee. 0018 1 In countless public forums, market 2 practitioners have suggested that we place a greater 3 focus on fixed income markets, and this committee 4 demonstrates that this is occurring now in earnest. 5 While the Commission has pursued a wide 6 variety of initiatives to enhance the transparency, 7 fairness and efficiency of equity markets, to date, we 8 have taken a more measured approach with respect to the 9 fixed income markets. But that doesn't mean that fixed 10 income has not been an important part of the regulatory 11 agenda. 12 The Commission has pursued a number of targeted 13 initiatives over the past two decades that have created a 14 solid regulatory foundation for the fixed income markets. 15 Let me mention a few. Following the Commission Staff's 16 1998 review of public debt markets, the Commission began 17 a concerted effort to bring post-trade transparency to 18 fixed income markets. In 2001, the Commission approved 19 the creation of FINRA's TRACE system. Four years later, 20 the MSRB implemented the real-time transaction reporting 21 system for municipal bonds. 22 In 2012, the SEC issued a report on the 23 municipal securities market, making recommendations to 24 improve the transparency, efficiency and competitiveness 25 of that market to better protect investors. The report 0019 1 contained important investor protection recommendations 2 concerning best execution and markup disclosure that lead 3 to recent reforms in the corporate bond and municipal 4 securities markets. In particular, in 2014, the 5 Commission approved an MSRB best execution rule for 6 municipal securities, and in 2015, both FINRA and the 7 MSRB issued detailed and harmonized guidance to assist 8 broker-dealers in achieving best execution in the fixed 9 income markets. 10 Lastly, in November of 2016, the Commission 11 approved FINRA and MSRB rules that would, for the first 12 time, require broker-dealers to provide retail customer 13 confirmation disclosure of their markups on many 14 corporate bond and municipal securities transactions. As 15 we consider the next steps in our policy agenda for the 16 corporate bond and municipal securities markets, it's 17 essential to make a broad-based assessment of the 18 developments that have impacted these markets. The 19 corporate bond and municipal securities markets pose unique 20 challenges, given their substantially different market 21 structures, important distinctions between principal and 22 agency trading models and the extensive number and 23 variety of instruments and lack of product 24 standardization. 25 These markets vary significantly from 0020 1 equities and must be evaluated from a different 2 set of lenses that appreciates these unique 3 distinctions. In some cases, equity market structure may 4 be highly informative, and in other cases, less so. It 5 is critical then in this process, we receive input from 6 market participants so that we can decide whether it 7 makes sense to pursue a particular policy initiative and, 8 if so, whether it should be regulatory, deregulatory or 9 some combination. 10 Today's discussion of bond market liquidity 11 should provide a very good framework for unpacking a 12 range of more specific issues, including the evolving 13 role of intermediation, the increasing prominence of 14 bond funds and the impact of technology. Like all 15 policy, our assessments and recommendations must be data- 16 driven. They must stay focused on improving market 17 quality, efficiency and providing clear benefits for 18 investors, and they must benefit from the expertise and 19 insights of industry participants, like the ones that we 20 are fortunate to have with us today. That's you all. 21 To facilitate that public input, we also have 22 created a comment file on the FIMSAC's new webpage, which 23 can be found under the new spotlight link at SEC.gov. We 24 encourage market participants to actively engage in this 25 process via written comment or otherwise. I look forward 0021 1 to today's discussion. I'm excited to begin working with 2 all of you. And with that, I'll turn it back to you, 3 Michael. 4 CHAIRMAN HEANEY: Thank you, Brett. I want to thank 5 you and your staff for all the hard work done to put this 6 committee together and to host the meeting today. Also, 7 on behalf of all of the committee members, I want to 8 thank Chairman and the Commissioners for establishing the 9 Advisory Committee, focused on the structure of corporate 10 and municipal bond markets. As most of us sitting around 11 the table have spent the vast majority of our careers 12 working in and analyzing these markets, we recognize the 13 importance of corporate and municipal bond markets to 14 both our economy and investors alike, and we welcome the 15 Commission's interest. 16 I'd also like to echo the comments of others in 17 thanking my fellow committee members for their 18 willingness to serve as members of this important 19 committee. Our committee's charter states that our 20 objective is to provide the Commission with diverse 21 perspectives on the structure and operations of US fixed 22 income markets, as well as advice and recommendations on 23 matters related to fixed income market structure. In 24 seeking to meet this objective, I hope we will all draw 25 from our varied experiences to identify and vigorously 0022 1 debate the important issues currently impacting investors 2 and our markets so as to build consensus around 3 actionable recommendations to the SEC regarding the 4 regulatory framework governing corporate and municipal bond 5 markets. 6 Admittedly, this is no small task. Active 7 engagement and the willingness to participate in frank 8 and robust discussions will be critical to successfully 9 achieve our ambitious goals. To that end, we are 10 beginning our efforts today by engaging in what I will 11 hope will be a comprehensive discussion of current 12 liquidity conditions in the US corporate bond market and 13 the drivers and responses to such conditions. Included 14 in all panels and subsequent discussions should be 15 thoughts on future state of fixed income market 16 liquidity, including any tools, products or platforms 17 required as the market continues to evolve post-crisis. 18 Our first panel will focus on the perspectives 19 of industry researchers on liquidity conditions in the US 20 corporate bond market. We'll hear from our panelists on 21 topics such as the methods by which market participants 22 measure liquidity and the inherent limitations on such 23 methods, differences in liquidity resiliency in normal 24 conditions and times of stress, and importantly, our 25 panelists' outlooks on liquidity conditions in both the 0023 1 near and medium terms. 2 Our second panel will focus on the perspectives 3 of buy and sell-side market participants, on the current 4 state of liquidity and the market structure for corporate 5 bonds. These panelists will provide their thoughts on 6 topics including the impact on regulations on market 7 liquidity as well as structural and behavioral responses 8 by market participants to current or expected future 9 liquidity conditions. 10 After breaking for lunch, we'll return to 11 discuss both panels. I appreciate our panelists' 12 willingness to return in this discussion and answer any 13 questions that the members may have. After a brief 14 afternoon break, the members will continue to engage in 15 discussion on liquidity and the potential next steps for 16 committee business. I anticipate that the morning panels 17 and the afternoon discussion on liquidity conditions will 18 touch on a wide range of market issues -- market 19 structure issues that this committee should consider in 20 the future. My intent for the afternoon segments of our 21 meeting is to hear from FIMSAC members regarding their 22 particular issues that have been -- have been raised in 23 the liquidity discussion that may warrant further 24 analysis, but also any other issues that may not have 25 been touched upon that warrant further consideration. 0024 1 And at the conclusion of the meeting, ideally, 2 we will have identified market structure issues around 3 which we may form subcommittees that would further 4 analyze and consider specific topics and ultimately 5 report back to a full committee for discussion at a 6 public meeting. To be successful, we must efficiently 7 deploy our resources and create a results-oriented 8 committee intent on making actionable recommendations to 9 the Commission. 10 Establishing subcommittees to focus our work on 11 particular market structure issues will ensure that the 12 important topics we consider benefit from deep, focused 13 conversations and provide all members with the 14 opportunity to bring their perspectives to bear on the 15 matters we consider important. I look forward to 16 discussing the subcommittee structure and potential 17 market structure topics with you all later this 18 afternoon. So once again, thank you very much for 19 devoting the time and energy to this committee, and with 20 that, let's get to work. 21 Before we get started, we have one quick 22 administrative item to -- which is approval of the bylaws. 23 In advance of today's meetings, we did send out to all 24 members a copy of the proposed bylaws. If there aren't 25 any questions, I would ask that we have a motion from one 0025 1 of the committee members to vote on the bylaws. 2 (Moved.) 3 CHAIRMAN HEANEY: Thank you. 4 All in favor? Calling for a vote, all in 5 favor? 6 COMMITTEE: Aye. 7 CHAIRMAN HEANEY: Any opposed or abstentions? 8 The bylaws are adopted. Thank you very much. 9 Okay. We'll now turn to our first panel 10 focusing on the perspectives of industry researchers on 11 liquidity conditions in the US corporate bond market. 12 Much has been written about corporate bond liquidity 13 conditions since the financial crisis, including academic 14 studies, regulatory reports, industry white papers, and the 15 seemingly unending series of media stories and blog 16 posts. 17 And while there's been much debate, liquidity 18 remains the right starting point while remembering the 19 ultimate goal to discuss is the future state of fixed 20 income markets. Several academic studies and regulatory 21 reports have generally concluded that liquidity 22 conditions, when measured by traditional metrics, such as 23 bid-ask spreads and trading volume, do not show signs of 24 substantial deterioration. In contrast, some observers 25 have focused on other metrics, such as dealer inventory 0026 1 levels and annual turnover as the percentage of par 2 amount outstanding as revealing structural changes in the 3 supply of and demand for liquidity that could lead to 4 extreme volatility in times of market stress. 5 Our first panel will focus on liquidity 6 conditions in the corporate US bond market from the 7 perspective of industry researchers. The topics to be 8 discussed will include the methods used to measure 9 liquidity conditions, differences in liquidity resiliency 10 in normal times and stress times, the data challenges and 11 the limitations of some of the academic research done 12 thus far and the outlook for liquidity in 2018 and 2019. 13 With that, I'd like to briefly introduce the 14 members of our first panel: Kevin McPartland, Head of 15 Market Structure and Technology Research at Greenwich 16 Associates, Jeff Meli, Co-Head of Research at Barclays, 17 and Sonali Theisen, Head of Market Structure and Data 18 Strategy for Global Credit and Securitized Market at 19 Citicorp. Good morning and thank you for joining. 20 Let's start by discussing at a high level some 21 of the findings of academic and regulatory reports on 22 liquidity conditions, many of which have focused on three 23 metrics: trading activity, transaction costs and dealer 24 liquidity supply. While the findings are mixed, as I 25 said before, some of the studies do not show -- do not 0027 1 find significant deterioration of liquidity, particularly 2 those that focus on trading activity and transaction 3 costs. Are these the right metrics to be thinking about? 4 Are there other metrics that should be considered that 5 better inform the analysis of real, true underlying 6 liquidity? 7 Kevin, why don't we start with you? 8 MR. MCPARTLAND: Sure, great. Thanks and 9 thanks, of course, to the SEC for having us all here 10 today and taking this proactive focus on fixed income 11 markets. The liquidity question for fixed income, of 12 course, is more nuanced than just is it good or 13 is it bad. There's a lot more that goes into that, but I 14 think there's no doubt over the last 10 years that it is 15 certainly quite a bit different than it was before the 16 crisis. 17 So if we look specifically at things to 18 measure, I think one of the most notable absences from -- 19 particularly from the academic research are those trades 20 that one, never actually get executed, right? So mostly 21 what we're seeing in the academic research is trades that 22 were done, right? So you're seeing trade records; you're 23 seeing what was done on trades; you're seeing bid-ask 24 spreads. So that's really looking at the active market. 25 So we're missing those trades that just never made their 0028 1 way -- never made their way towards execution. 2 And then secondly, which is even harder to 3 measure, I suspect that there are quite a lot of 4 potential trades that never left the portfolio manager's 5 desk. When doing that analysis and trying to figure out 6 what the next step forward is and what the investment 7 strategy is, where the opportunities are, looking at 8 liquidity metrics, which now there is a lot more data on 9 that for the market participants, which is great, some 10 ideas were probably dismissed before they even got to the 11 trading desk because there was just no way, it was too 12 expensive; we couldn't get it done. The spreads don't 13 fit the strategy, so we never got there. So all of that 14 gets lost in the academic research and really does speak 15 to how things have changed. 16 Also, it's important to understand liquidity by 17 trade type, trade size. So different issuers can be very 18 different in their liquidity profiles, investment grade 19 versus high-yield, of course, is a simple way to look at 20 it. By trade size, which I suspect we'll get into a 21 little bit more throughout the day. Retail size orders 22 are very different than institutional size block orders. 23 And then, again, taking those sizes and applying them 24 back to investment grade or high-yield where different 25 issuers can yield very, very different views into whether 0029 1 liquidity is good or bad. 2 Speed of execution is another important metric, 3 how quickly can we get this done and with what 4 market impact? Will it take weeks or will it take days 5 or hours or minutes? Liquidity in related products also 6 is something interesting to look at. If we look at the 7 rates market, which, of course, has very different 8 dynamics, but there's a very active derivatives market, 9 for instance. There are many, many ways to get rates 10 exposure outside of buying government bonds. The credit 11 market, although it continues to try to create 12 alternatives, has had limited luck there. 13 So that's an interesting -- something 14 interesting to investigate derivatives, ETFs, so I'll 15 follow onto that conversation. And then electronic 16 trading, which is -- which is a tricky one and another 17 one I'm sure we'll spend more time on today. It 18 certainly is a signal, more electronic trading doesn't 19 necessarily guarantee liquidity, but the fact that we've 20 seen such substantial growth in electronic trading, 21 particularly in US corporates over the last three or four 22 years, that does show that the -- I believe there is 23 quite a bit of value there, as they're the primary 24 drivers and the buy-side signaling that they see value 25 would, would lead you to believe that they believe it 0030 1 helps -- it helps their liquidity situation by utilizing 2 technology more than they have in the past. 3 And then last but not least, and we'll get into 4 the details, I think, a little bit later in this panel 5 discussion, but investor sentiment shouldn't be ignored. 6 There's a -- there's a gut feel that comes along with 7 trading in these markets that has always been there, and 8 despite all of the new technology, will always be there. 9 So understanding how it feels to be on the desk and has 10 -- does it just feel that things have changed, even if 11 it's hard to show it in the data. It's certainly very 12 important, but we'll get into that in some more detail 13 before the end of the panel. 14 CHAIRMAN HEANEY: Jeff, as I pass it to you, 15 it's interesting to hear Kevin mention trade type and 16 trade size and, and well, the one piece we sent around to 17 the committee, the behavior modification piece that you 18 talked about certainly incorporates that. I'd be curious 19 as to respond to that question and incorporating that. 20 MR. MELI: Sure, thanks. And I echo Kevin's 21 thanks in having us here. First, I would say I agree 22 that assessing liquidity is very challenging. Liquidity 23 encompasses a large number of different characteristics. 24 It's breadth, depth, immediacy, price impact, along with 25 others, and each one of those characteristics could be 0031 1 measured using a variety of different metrics. And I 2 think it's really no wonder that focusing too heavily on 3 any one individual metric can lead to some ambiguous 4 conclusions. 5 I'll give one example, focusing on a statistic 6 that you cited on the fact that turnover has declined in 7 the corporate bond market. So just to put some numbers 8 on it, investment grade turnover has declined from 2010 9 to this year, to 2017, from, give or take, 100 percent 10 down to about 65 percent over the same time frame, 11 turnover in high-yield markets has declined from almost 12 200 percent to about 140 percent. So those are 13 meaningful declines in turnover. And those could be 14 driven by lower liquidity. 15 However, there are other alternatives and also 16 feasible explanations for a decline in turnover. For 17 example, we could have had changes in the participants in 18 the credit markets. You know, speaking 19 holistically, if you had had a shift of assets, say away 20 from hedge funds, which tend to be high turnover sorts of 21 investors and towards insurance companies, which tend to 22 be lower turnover investors, you would see a decline in 23 turnover, but that wouldn't necessarily translate to less 24 liquidity. It could be that the trades that any 25 individual investor wanted to do were still able to get 0032 1 done. And so I think that you can see that just focusing 2 on something like turnover can be -- can be challenging. 3 You could also look at something like quoted 4 bid offer. A few of the white papers actually quoted a 5 Barclay statistic, which we -- which we developed called 6 a liquidity cost score, which measures a quoted bid offer 7 based on our traders' runs. And you would see using that 8 statistic that a quoted bid offer has largely recovered 9 since the crisis, and it's back closer to levels like we 10 had seen precrisis. Now, I think that raises some 11 questions about what -- for example, what size trades can 12 you get done at those bid offers. 13 I think it's long been understood that larger 14 trades come with higher bid offers, but there's a sense 15 in the market that the slope of that curve, if you will, 16 has increased, that even in some cases, larger-sized 17 trades cannot get done in the market. And so just the 18 quoted bid offer itself could also be potentially 19 difficult to focus on. 20 So I think there's almost two paths to 21 addressing this sort of ambiguity associated with any 22 individual metric. One, which I've seen in a lot of the 23 academic work is to look at almost like a panoply of 24 different liquidity metrics. You can look at turnover, 25 percent of days with zero trades in individual security, 0033 1 various measures of price impact. I've seen academic 2 studies with as many as 10 different liquidity metrics 3 that are being used. And that's not really a measure or 4 really a cogent theory of liquidity, it's more of a 5 shotgun approach to say when doing some empirical work, 6 we want to throw in every sort of metric that we can 7 think of to try to demonstrate that, you know, various, 8 you know, changes in the -- in trading patterns have 9 occurred. 10 It's blunt and doesn't necessarily lead us to 11 understand how market participants are adjusting to these 12 environments, nor does it, I think, really help 13 understand how liquidity metrics may evolve in more 14 stressed situations. 15 Our preferred approach, rather than focusing on 16 an individual metric is actually to look at the behavior 17 of investors. So investors are not static. I think they 18 will react to changes in market conditions, including 19 liquidity. Lower liquidity is not just going to be 20 blindly accepted by the investor community in the form of 21 higher trading costs and higher -- and less flexible 22 portfolios. We would expect a response from investors, 23 and so we look to the change in behavior as a marker of 24 the changes in liquidity and for a guide to the potential 25 impact of lower liquidity, if any, on markets in stress 0034 1 times. 2 So I'll give an example, which we highlight in 3 the piece, Behavior Modification, which was included in 4 the packets. Well, realized bid offer is a very commonly 5 cited statistic. As you mentioned, there are a number of 6 academic studies and regulatory studies that focus on 7 realized bid offer, and they show that realized bid offer 8 is flat to maybe even down in the post-crisis period, 9 which would obviously be a time in which many 10 participants believe liquidity has deteriorated. 11 Our numbers, which we cite in that paper when 12 we look at aggregate bid offer using TRACE data across 13 the corporate bond market confirmed that. We would show 14 that using block trades in investment grade, using TRACE 15 data, our estimate for bid offer from 2010 to 2015 is 16 that it's declined about 10 percent from just over three 17 basis points to just over two-and-three-quarters basis 18 points. 19 So a 10 percent decline in a bid offer spread 20 over that five-year period. However, I think these 21 aggregate bid offer statistics are really driven by two 22 meaningful shifts in behavior, and you need to look 23 deeper at the numbers in order to understand how bid 24 offer is really changing. There were two main shifts 25 that we highlight. One is a shift from trading older 0035 1 vintage bonds towards trading more recently issued bonds, 2 and a second is a shift from trading on a principal 3 basis, meaning that dealers wear the risk on their 4 balance sheets, towards dealers trading more on an agency 5 basis, which means that they've lined up both sides of 6 the trades before they execute, and so never actually, 7 you know, economically wear the risk. 8 We use, again, the investment grade data on 9 block trades to examine these numbers. So, for example, 10 newly issued bonds in 2010 represented 34 percent of the 11 block trades. And in 2015, they represented 41 percent 12 of the block trades. Agent trades, which we define as 13 trades where both sides are executed within 15 minutes of 14 each other, increased by 50 percent over that timeframe, 15 from 16 percent of total block trades to 23 percent. 16 It's a very material increase. And at the same time, 17 trades that were on dealer balance sheets for greater 18 than five days declined from almost half of trades down 19 to just about a third of trades, just over a third of 20 trades. 21 And interestingly, that transition was more 22 severe for older vintage bonds than for new vintage 23 bonds. Now, why is this important? It's important 24 because bid offer is naturally lower for newly issued 25 bonds. They are more liquid and easier to trade, and you 0036 1 can see that in the -- in the -- in the TRACE data that's 2 quoted in the report. It's also true that agent trades 3 have lower bid offers than trades done on principal. So, 4 for example, the bid offer on newly issued bonds is one- 5 third that of more aged bonds. And agent trades 6 typically carry a bid offer of one to two basis points. 7 For the most liquid bonds, it could even be 8 less than one basis point. Whereas principal trades are 9 anywhere from two to seven basis points on average in bid 10 offer, depending on the -- on the age of the bond. So 11 first, I think that just analytically, mathematically, 12 those two changes are fully responsible for the changes 13 in bid offer. Actually, if you look within category, 14 from 2010 to 2015, bid offer widthin category was stable. 15 If you look back precrisis, which is not in the report, 16 but if you look back at the data from '07, you can see 17 that actually within each category, bid offers have 18 increased since the precrisis period. 19 So what's interesting there is you have a -- 20 you have a response, if you will, from the investor 21 community, which effectively needs to have been driven by 22 lower liquidity. These changes, moving away from trading 23 older bonds towards newer bonds, and trading more on an 24 agent basis and less on a principal basis, come with a 25 cost. They come with a cost in terms of immediacy, in 0037 1 terms of portfolio flexibility. As Kevin mentioned, we 2 know for sure that not all trades that are -- that are 3 attempted to be executed on an agented basis 4 actually happen. And I'm sure in the -- in the 5 later panel, we might get some perspectives on just what 6 proportion of trades that are attempted in the agency 7 model are actually executed, but we that it is far less 8 than 100 percent. 9 And so what we can say about liquidity is that 10 it has declined enough to effect meaningful changes in 11 the way investors behave, meaningful enough that we can 12 measure them using, you know, using the 13 available TRACE data. And I think that that behavior 14 itself, I think, can be a guide towards understanding 15 that markets have shifted and adjusted and that is, I 16 think, more meaningful than trying to handpick any one 17 individual metric to measure liquidity. 18 CHAIRMAN HEANEY: Thank you, Jeff. 19 Sonali, you've analyzed and written on this 20 several times in '17, and then the most recent piece this 21 month in '18. Curious your thoughts. 22 MS. THEISEN: Great. Well, thank you very 23 much, and again, thank you, Chairman Clayton for creating 24 this important forum, and to all the Commission for 25 inviting us to participate, to share our comments. 0038 1 I largely agree with my -- you know, the 2 comments that have been made by my panelists already. As 3 a logistical point for the committee members, I will 4 actually reference in your deck, in your packet, I think 5 we're Section F, I will reference some of the charts as 6 we go along with our points, and I welcome you to follow 7 along. And for those in the audience, you know, I'll 8 summarize kind of what's there so that it's in front of 9 you. 10 You know, yes, lots has been written on 11 liquidity metrics. You know, our Slide 6 attempts to, as 12 well, name some of them: volume, depth, you know, bid 13 offer width, resilience of the market. And, you know, 14 fully agree, it's undisputed. There's been a lot of 15 literature about, you know, compressed bid offer widths 16 and higher volumes trading in the market. Rather than, 17 you know, go back to the points that Kevin and Jeff made 18 directly, I want to compliment some of the observations 19 that they made, particularly with respect to bid offer 20 widths and TRACE. 21 One is yes, we absolutely think this is a 22 metric that deserves attention, but we think that looking 23 at bid offer width within the context of -- in relation to 24 break evens of what it actually costs investors to be in 25 and out of trades is very important. So if you flip to 0039 1 Slide 12 of our deck, you know, our view is that 2 investors need twice as much conviction to trade today 3 than they did 10 years ago. Using the exact same bid 4 offer, mid to bid of 2.5 basis points, you can see that 5 if you were to trade the Citi US Broad Investment-Grade 6 Corporate Index in 2007, the breakeven holding period is 7 15 days from the mid to bid, where it's gone up to, in 8 2017, 32 days. 9 So quite simply put, although bid offer widths 10 are, yes, tight, we need -- we think that investors need 11 twice as much conviction to put on a trade. The second 12 limitation of TRACE that I would mention, it's certainly, 13 you know, a very constructive metric for us to, to look 14 at when considering bid offer widths, but I would -- I 15 would highlight that the data is noisy, particularly when 16 looking at, you know, cross-sectionally whether it's 17 retail versus institutional, whether it's looking at 18 high-yield versus investment grade or even the US versus 19 European markets, and the reason for that is, you know, a 20 large part of that is there's not necessarily consistency 21 with the embedding of electronic fees, so I suspect we 22 will, you know, spend a lot of time, again, on the topic 23 electronification, but, you know, prints that hit TRACE 24 may or may not embed within them a fee for trading 25 electronically. 0040 1 And so when looking at, again, trades that 2 happen on different venues or happen voice versus 3 electronically, we need to take that into account, unlike 4 equities, those fees are not explicitly backed out, so 5 the data one can be noisy. 6 The second I would note, limitation with TRACE, 7 which is really a market structure issue is the issue of 8 3:00 p.m. spot. So many investors, you know, because 9 they would like one treasury print with one price on all 10 of the trades that they do in a given day will, you know, 11 look to spot those trades at 3:00 p.m. So a trade -- in 12 risk terms that was executed at 9:00 a.m. may not 13 actually hit the tape until 3:00 p.m. So again, I think 14 when, yes, the bid offer widths, et cetera, looking at it 15 in aggregate makes sense, but when we really try to, you 16 know, pinpoint, you know, what is happening, there is, I 17 think, a significant amount of noise in the data. 18 The other points that I would make in terms of 19 metrics that are not as easily measured or, you know, 20 cited, and I fully agree with -- again, with Kevin and 21 Jeff's comments, one is really a metric for the diversity 22 and concentration of the investor base, and for this, you 23 know, I'd like to refer you to Slide 8 of our deck. And 24 you can see that, you know, on the -- on the left hand 25 chart, you know, ownership may seem diverse, but 80 0041 1 percent of net buying has been with funds and foreigns. 2 And we would also observe that, you know, the foreign 3 investors tend to have, again, less turnover. I think 4 this is the point again that Kevin has alluded as well. 5 So we think that, you know, coupled then, as 6 well, with -- excuse me, I have to find which chart it is 7 of ours, but coupled as well with just the actual 8 concentration of -- excuse me one second, of clients -- 9 I'll find it in a second, but coupled with the 10 concentration of our actual clients in trading with us, 11 in 2017, 50 percent of cities' volumes was with 25 12 clients. We've actually been measuring this metric for a 13 few years now. That number -- that concentration has 14 gone up. 15 In 2015, that number was 36, and on the 16 investor side as well, I think, you know, Kevin has some 17 very good data in some of his literature around, you 18 know, the number of clients that use the top five 19 dealers, it's a few number of participants, so I think 20 creating some type of metric around the concentration and 21 diversity in the investment community would be a helpful 22 metric for us to look at. And then touching, again, on 23 Kevin's points on untraded inquiry, we fully agree that 24 this is a metric that would be helpful to have. I think 25 it's aspirational today to have it, but I do think that 0042 1 as data capture, inquiry capture systems improve in the 2 next couple of years, you know, it's an investment that's 3 being made across the industry in capturing those types 4 of quotes that have not traded, I do think those will be 5 metrics that we will be able to look for going forward. 6 CHAIRMAN HEANEY: Thank you. If we could stick with 7 that theme but maybe just pivot a little bit to times of 8 stress, there's been academic research; there's been 9 regulatory research that talks about some of the factors 10 that may be impacting liquidity, regulation, 11 electronification, some of the things you've touched 12 upon, the role of the dealers, transparency, credit 13 derivative markets, perhaps, and how that's affected or 14 not, and then the general macro-economic factors. But 15 when you think about changes in liquidity and Sonali, you 16 just touched on it a little bit with even the 17 concentration effect or the block trades, how do you 18 think about that as it can change or be exacerbated in 19 times of stress? 20 Jeff, I'll start with you and then move to 21 Kevin. 22 MR. MELI: Sure. Thanks. So first, I'd say 23 you put together a pretty comprehensive list of the 24 catalysts for change in liquidity, I think. I think that 25 it's hard to dissect the individual contribution of each 0043 1 of those, but I think they all have played a role. The 2 way we think about it is that the market really has 3 tipped from a high liquidity equilibrium into a low 4 liquidity equilibrium, and at this point, easing any one 5 individual constraint is very unlikely to result in the 6 market tipping back, particularly not that you've seen 7 material changes in behavior and expectations across the 8 investor and dealer community. 9 If I had to point to one sort of 10 underappreciated catalyst, I would say it's been pressure 11 on the repo market. I think repo is interesting because 12 it affects both the supply and potentially the demand for 13 the liquidity. I noted in one of the white papers, it 14 was the role of repo in the supply of liquidity was 15 noted. Dealers use repo in many ways to source shorts, 16 to fund their investments, to hedge interest rates, and 17 the cost of repo has increased dramatically due to 18 changes in regulations. Notably, the supplemental 19 leverage ratio, which was part of the Dodd Frank, and I 20 would note that the NSFR, the net stable funding ratio, 21 as proposed, would only heighten the pressures on the 22 repo markets. 23 You know, you've seen repo volumes decline by 24 almost 50 percent in response to these increased costs, 25 and I think that's had obvious implications for dealers, 0044 1 sort of piling onto dealers' willingness or ability to 2 provide liquidity. I think less appreciated is the -- is 3 the role that repo may play in the demand for liquidity. 4 You know, repo is really one of the sort of 5 handful of real true cash equivalents that are out there 6 and available to investors, along with things like 7 deposits and treasury bills, it's the short list of 8 investments that are either directly or indirectly backed 9 by US Government liabilities. You know, bills have been 10 flat to down in absolute sense and, obviously, lower as a 11 percentage of sort of financial markets. Deposits are -- 12 remain basically floored at zero. Interesting that 13 despite, you know, several hikes from the fed, you've 14 seen deposit rates be floored at zero. And repo volumes 15 under pressure, I think that in some sense that on the 16 demand for liquidity side, that pressure on the sort of 17 natural cash equivalents may be a factor in some of the 18 inflows we've seen into open-end fund vehicles, where the 19 open-end nature of the funds themselves, the sort of on 20 demand nature of the -- of the investments may be an 21 attractive feature in and of itself. 22 And -- you know -- and in -- you know, in 23 principal, that may -- you know, that may raise some 24 concerns about resiliency if you were to see some of 25 those flows unwind. I want to be very careful about even 0045 1 mentioning that because I think it's highly unlikely to 2 see those kind of pressures evolve in the great majority 3 of fixed income funds, even with a decline in liquidity. 4 Take investment grade as an example. 5 Regardless of the liquidity of investment grade corporate 6 bonds, most investment grade institutional funds are a 7 mix of investment grade corporate bonds, treasuries and 8 agencies. And those latter two remain extraordinarily 9 liquid and more than able to fund any sort of outflows. 10 I think there may be more resiliency concerns in some 11 less liquid, more concentrated parts of the market, such 12 as high-yield or emerging markets. 13 But then I think a second part of your question 14 is periods of stress, and there, I would -- I would -- I 15 would -- I would point again to how investor behavior has 16 changed and ask how has -- how has investor behavior 17 changed to mitigate some of the lower liquidity that we 18 may see in the market, so in particular, we've seen a 19 dramatic increase in the use of portfolio tools to mitigate 20 liquidity risk, particularly in high-yield and emerging 21 markets, which are the less liquid parts of the market. 22 For example, we've seen a dramatic rise in the 23 use of high-yield ETFs by institutional fund managers to 24 manage their inflows and outflows. Now, I think that, 25 you know, we've seen coincident with that a substantial 0046 1 increase in the volume, secondary volumes of high-yield 2 ETFs. To put that in context, the turnover of high-yield 3 ETFs is over 10 times a year, relative to the turnover of 4 the high-yield bond market, which is only 1.4 times a 5 year. So you see major secondary market volumes going 6 through high yield ETFs. 7 And there's, you know, definitively a 8 relationship between those volumes and the levels of 9 inflows and outflows, which supports the more anecdotal 10 evidence that we see institutional investors using those 11 tools. I think that that begs the question of just how 12 effective those tools may be in times of stress. You can 13 look at data, such as -- such as fund level inflows and 14 outflows to get a sense of how correlated inflows and 15 outflows can be, and in normal times, there's -- they're 16 actually surprisingly uncorrelated such that the use of 17 portfolio tools like ETFs should mitigate a large -- 18 could, in principal, litigate a large amount of the need 19 to trade corporate bonds. 20 In fact, we actually believe that high-yield 21 corporate bond trading is likely lower because of the use 22 of some of these portfolio tools to mitigate the need to 23 actually trade the underlying instruments. I think 24 there's less data about how those tools work in periods 25 of market stress, simply by virtue of the fact that we 0047 1 have had very few periods of market stress over the 2 period that these tools have been in use. In principal, 3 using something like an ETF could be no worse than using 4 the underlying bonds since you could always turn to the 5 create-redeem process to end up with the bonds in your 6 portfolio if you needed to sell them. So you really 7 can't be worse off using those tools. 8 I think the question is does the secondary 9 volume in those -- in those sorts of instruments hold up 10 in periods of stress such that there are other buyers of 11 those sorts of securities if you need to liquidate them. 12 Again, we have very few episodes of market stress. 13 Those that we do have would suggest that there was 14 sufficient secondary volume to absorb the selling that 15 you saw in the secondary market for those instruments, 16 which says that they are reasonably effective, but we're 17 really looking at, at most, one or two data points to 18 suggest that. 19 CHAIRMAN HEANEY: Thanks, Jeff. 20 Kevin? 21 MR. MCPARTLAND: Great. Thanks. And I do 22 agree with -- I do agree with most of those drivers. I 23 think put simply, the banks have largely been 24 disincentivized from providing principal liquidity. 25 There's a more regulatory risk. It's become expensive; 0048 1 profit margins are down. So the incentive to be involved 2 in the market has been sort of pulled back quite a bit, 3 and then sort of another outside factor that probably 4 shouldn't be ignored, equity invested in the banks have 5 come to like the I guess you would call it less risky 6 approach, right, where an agency model is more just fees 7 and there's less risk in that. And the equity investors 8 like that and are pushing at that corporate level for 9 that model, which again, further disincentivizes the 10 banks from feeling like they can or should be providing 11 that principal liquidity and taking that risk, even when 12 it is purely to help their clients and investors. 13 So, you know, because of that, those that 14 continue to do well and provide the most liquidity tend 15 to be the largest because they have to scale to overcome 16 a number of those obstacles, so Sonali sort of pointed to 17 it a little bit earlier, the top five dealers in 18 investment grade handle 60 percent of client volume. In 19 high-yield, it's a little bit more, 62 percent. You 20 would think that that number probably would have shifted 21 since before the crisis, and if we look back over time, 22 while there's been some bumps in the chart, we go back to 23 just precrisis and it's almost unchanged. 24 So going through the -- sort of the Dodd Frank 25 process, knowing, of course, that these markets are a 0049 1 little outside of that, there was an expectation that 2 post-crisis, we would see the markets diversify and 3 democratize so we'd have less focus on a small few. That 4 doesn't seem, at least to this point, to be the case. 5 Again, for a lot of these reasons that we just discussed. 6 So that all said, if we start to think about what would 7 happen in a time of stress, of course, investors are 8 thinking about that and preparing. I think there's two 9 major things they're doing as they think about that, 10 based on the research that we've done. 11 So one, the use of more middle market and 12 regional dealers is rising. From a volume percentage, 13 it's still a relatively small piece of the market, maybe 14 about 20 percent, but from our most recent study, 98 15 percent of the investors in these markets that we spoke 16 with are doing some trading with middle market dealers. 17 To a larger extent, that is for -- more for their 18 expertise than it is for their balance sheet, right? 19 Often, those regional banks have particular expertise in 20 particular corporates or maybe they have regional 21 relationships with local asset managers, but to sort of 22 continue on with that point, the average number of 23 primary counterparties that each buy-side firm has -- 24 large or small, has slowly been creeping up over the last 25 four to five years, so which is a sign, despite the fact 0050 1 they're still concentrating so much flow with the top 2 couple, they are lengthening out that list so they have 3 new sources of liquidity or unique sources of liquidity 4 for particular situations if they need those. 5 And then I think we can't ignore electronic 6 trading here, as well, as sort of another source. So 7 even for some of the biggest firms, the biggest asset 8 managers that presumably have the best access to 9 liquidity because of their scale, they are actively -- 10 and obviously a lot of them are here, are actively 11 looking for new electronic channels as well. They are 12 large, so they have scale, but they're large, so they 13 also have quite a bit of trading to do. So the more channels 14 they have to get those trades done, I think the better -- 15 the better they feel, the safer that is for their clients 16 as well. 17 And then one last point that I did notice sort 18 of along that line of thinking about lengthening the list 19 of liquidity providers, there is starting to be a 20 slightly bigger influence of non-bank liquidity providers 21 in this space as well. That picture is very, very 22 different than what you would think of non-bank liquidity 23 in the equity market. These markets don't turn over like 24 that. Those strategies don't work. But if we look 25 specifically at ETF, ETF trading firms who do a lot of 0051 1 those create-redeems, they're playing a bigger and bigger 2 role and contributing more and more volume into the 3 market. 4 Now, again, they're not facing clients 5 directly, but that sort of odd demand liquidity that 6 they're providing as they execute their strategies, I 7 think, ultimately, to some extent, is now but over time 8 could start to benefit investors more as those markets 9 become more liquid. 10 CHAIRMAN HEANEY: Sonali, your thoughts? 11 MS. THEISEN: So, you know, again, I largely 12 agree with everything that our panelists have already 13 cited. I would say that, you know, with respect to 14 dealer balance sheet, one, and again, I'm going to refer 15 to Slide 9 of our -- of our deck, you know, yes, you 16 know, it's undeniable that dealer balance sheets and 17 inventories have, have changed. We don't necessarily 18 know that this is, in and of itself, a cause of a drop in 19 liquidity, but rather, you know, A, we maybe view it more 20 as a correction towards precrisis levels, but also, more 21 than that, we think dealer inventories are largely 22 influenced by volatility, which in turn, influences our 23 ability to earn an appropriate return on capital. 24 So we think that, you know, there will be a 25 mitigating effect in volatility of dealer balance sheets, 0052 1 you know, potentially, you know, changing in size. You 2 know, capital rules may -- we think, really are affecting 3 risk transformation. And this is, again, highlighted for 4 us on Slide 9. You know, the role of a dealer in a 5 principal market is often times not to just trade in and 6 out of the same security, but to really transform that 7 risk, whether that is, you know, trading a funded versus 8 unfunded product, a curve steepener. You know, we hedge 9 with a number of different instruments, not always just 10 trading one bond and then selling that very same bond. 11 So I think it's, you know, fair to say that 12 with, you know, higher costs of capital, again, to 13 highlight on Slide 11, that that risk transformation 14 function has become more expensive. And it particularly 15 becomes true, you know, if you think about sort of the 16 waterfall effect of, you know, the contraction of a 17 structured credit market, which in turn, you know, had 18 implications for the single name market, which also was 19 hurt by, you know, increased capital costs, you know, a 20 hedging tool in the corporate bond space has, you know -- 21 has kind of become a lot more expensive and less liquid 22 for us to have as a tool. So those would be kind of, you 23 know, the two areas with respect to balance sheet and 24 capital rules that we would focus on. 25 Again, coming back to the point of, you know, 0053 1 the evolving role of the investor base as a driver of 2 liquidity, you know, our view is that, you know, with the 3 rise of passive investing, in some ways, it's come at the 4 expense of, you know, a reduction in relative value 5 trading. So, you know, some of these points are 6 illustrated on Slide 13, but if you think back, you know, 7 during periods of less constraint, leverage and repo, 8 investors could more easily turn a small dislocation into 9 a reasonable return on risk, and it became a virtuous 10 cycle of kind of mean reversion. 11 You picked a name that you thought was cheap. 12 You picked a name that you thought was rich. You put on 13 that trade, you levered it. It -- you know, you saw that 14 it'd been reverted, and it became this virtuous cycle. 15 You know, now, you know, in today's world, the fact that, 16 you know, the reach for return has been one of a down in 17 credit quality. There's less dispersion. There's less 18 mean reversion. You know, this has all kind of been a 19 driver as well of the rise in passive investing. 20 And then the other point I'd make here, again, 21 going back to, you know, the comments that we made in 22 Slide 18 about cities' volumes now are 50 percent with 25 23 of our clients; 75 percent with 101 of our clients. You 24 know, this is really, we see it as a function of our 25 client activity. This is not a function of our bank 0054 1 choosing to focus on a smaller number of clients. It's 2 really -- so we talk a lot, I think, about the dealer 3 side of the equation, but the sort of evolving role of 4 the investor in this -- in this market, I think, is one 5 that we also think is very important to discuss. 6 Lastly, with respect to electronification 7 efforts, you know, we're, A, you know, very supportive of 8 all of the advancements that have been made. We don't 9 necessarily believe that in and of itself it's made a 10 tremendous impact on liquidity, particularly in the very 11 long tail of less liquid instruments. So if we think 12 about, you know, again passive investing and the bonds 13 that trade and the sizes that trade in electronic venues, 14 it tends to be, you know, the smaller sizes, the more 15 liquid names where there is a very, very long tail of 16 bonds, I think, you know. close to 40 percent, I've got 17 it somewhere in our deck, of bonds that trade less than 18 five times a day in the corporate bond universe. 19 So we do believe that, you know, 20 electronification is making -- you know, creating a lot 21 of operational efficiencies. We do think it's played a 22 role in, you know, the other sides of the market finding 23 each other when they can. We just don't know that that 24 overall percentage, I think we estimate, you know, that 25 in true client-to-client trading on the institutional 0055 1 side is somewhere around, you know, one percent of 2 volumes of the entire marketplace. 3 We don't necessarily know that 4 electronification in and of itself is sort of, you know, 5 massively changing the dynamic with respect to liquidity. 6 CHAIRMAN HEANEY: Thank you. Insightful answers, 7 covers a lot of territory from electronic trading to the 8 -- to the comments on ETFs, to resiliency. Jeff used the 9 word resiliency. I'm sure that'll come back in the -- in 10 the second panel and kind of the depth of the markets and 11 the potential times of stress. 12 Kevin, I just want to -- I want to turn to you. 13 There's sometimes a disconnect between the data and what 14 clients say. I think around the table, we could all 15 agree to that at certain -- at different points over our 16 careers. Does the research that you're doing and the 17 conversations you're having with clients, does that 18 provide any kind of different points-of-view than what's 19 been expressed so far or other points-of-view, whether 20 it's in times of stress or not in times of stress, but 21 general comments? 22 MR. MCPARTLAND: Yeah, I think the sentiment is 23 important, so just the 10 seconds of background. We, 24 Greenwich Associates, we interview about 4,000 fixed income 25 investors a year. We've been doing so, I think, for the 0056 1 better part of 20 years at this point, so there's a good 2 amount of information that can be sort of gleaned from 3 that quantified, qualitative dataset. The topline number 4 for today, which maybe sounds more negative than it is as 5 we get into the details, 73 percent of investment grade 6 and 68 percent of high-yield investors said they believed 7 that changes in liquidity have impacted their ability to 8 execute their investment strategy, right? 9 So at a top level, I think that goes back to 10 one of the first things that I said was there's no 11 question that things have changed, and they are not like 12 they used to be. With that, one of the other things that 13 we've now been tracking for a couple of years and also 14 goes back to some of my earlier points is how difficult 15 investors feel that it is to execute corporate bond 16 trades of different sizes. So for the past three years, 17 the trend now that we can see is that executing trades, 18 particularly a million and under in size but now looking 19 at the most recent data, which isn't here, we haven't 20 published it yet, but looking under five million, we've 21 seen a pretty notable improvement of folks saying that 22 they find it easy or extremely easy to execute trades 23 under five million, which is a big -- I think a big 24 change. 25 And that sentiment is closing in on what we see 0057 1 for on the run treasuries, which are clearly very, very 2 liquid. Now, to put that in perspective, what an on the 3 run treasury trader thinks of as liquid is probably very 4 different than what an investment grade corporate bond 5 trader thinks is liquid, but nevertheless, I think the 6 sentiment that can be taken from that is important. 7 CHAIRMAN HEANEY: Can I just stop you and ask do 8 they attribute that to electronic trading platforms as a 9 reason or is that not attributable? 10 MR. MCPARTLAND: Perfect lead in. So it's an 11 interesting point, so I -- there's no question that some 12 of that goes back to e-trading and to auction, not just 13 electronic trading, but the -- all of the new technology 14 and the tools that they have on their desk they didn't 15 before, which can be a whole host of things: data and 16 analytics, even just that help the process or even help 17 them figure out who to call more effectively than they 18 had in the past. That's not the whole story. There's 19 certainly been a lot of change at all of the dealers, the 20 large and medium-sized dealers because, obviously, as 21 Jeff and Sonali have been saying, they've been adapting 22 to this new environment and thinking about how do they 23 serve clients better. 24 So e-trading certainly is a big part of this 25 story, but the way that the dealers are servicing 0058 1 clients, the type of people that are on those trading 2 desks, the skillsets have changed. We've seen quite a 3 bit of that as well, right? We're moving more towards a 4 group of folks who understand technology and market 5 structure and economics and understand electronic trading 6 and can have those conversations with clients to help 7 them figure out should they put it in the system, should 8 they call, when should they call, who's the right person 9 to talk to, all of those sort of more -- I don't know if 10 you call them workflow changes and personnel changes in 11 terms of skillsets have made a big, big difference, I 12 think, for market participants as well, so I think the 13 improvements there are twofold. 14 And if we flip to the dealer side for a minute, 15 we've talked a lot about investor sentiment, when we ask 16 dealers sort of how they're -- how their -- how their 17 work with clients has changed over time, not 18 surprisingly, both bulge bracket and the regional dealers 19 said they're much more selective with which clients they 20 provide liquidity to. I think that goes back to some 21 earlier points, thinking about it's another metric, 22 right? It's what -- the liquidity for some clients is 23 different than liquidity for others. So it's something 24 definitely to think about. I don't think there's 25 anything wrong with that, right? Large scale orders and 0059 1 profitability, these are still private -- or for-profit 2 companies, but nevertheless, dealers, of course, are 3 going to make us think a little bit harder about what 4 they do with that capital that they have a little bit 5 less of today. 6 And then lastly, which I thought this was 7 interesting is we asked the sell-side trading desk what 8 they thought their challenges to success were in the 9 coming year. For the bigger dealers, it was more about a 10 reduction in budgets and balance sheet. For the regional 11 dealers, it was overwhelmingly the burden of compliance, 12 so regulatory compliance and regulatory uncertainty for 13 the regional dealers were a much, much bigger deal. They 14 felt they didn't have the staff and the lawyers and the 15 presence -- the presence down here to really be able to 16 understand what's coming and know what to deal with. 17 We had a number of conversations with decently 18 sized regional dealers, and we said, you know, two, three 19 months ago, MiFID II, and they looked at us like what's 20 that, right? Which, again, not a US regulation, but 21 certainly something they should be aware of as that will 22 bring changes around the world. 23 CHAIRMAN HEANEY: Thank you. And then just our 24 final question, and we'll ask all three of you, just 25 really your thoughts, best guess in the research that 0060 1 you've done on your outlook for liquidity conditions in 2 2018 and '19, and then I'll just throw one little second 3 part of that curveball, if you can implement one change 4 based on your research, what would it be that would 5 improve it? 6 Jeff, why don't we start with you? 7 MR. MELI: Sure, I'll start with that. So 8 first, I think that liquidity conditions are highly 9 correlated with market conditions. And so if the market 10 remains robust, then I think that liquidity will remain 11 manageable, which with spreads, you know, near, you know 12 -- at or near post-crisis heights, I think the market 13 conditions are strong. And so you really see liquidity 14 deteriorate when market conditions become more stressed 15 and, you know, there is little sign of that, at least in 16 -- you know, in the -- in the immediate term. So I think 17 the forecast for liquidity would be that it would sort 18 of, you know, remain manageable over that -- over that 19 time. 20 In the meantime, what I think between now and 21 whenever the next inevitable market disruption occurs, I 22 expect we'll see more use of portfolio products and more 23 fine-tuned use of portfolio products. For example, we're 24 seeing portfolio trades increase. That's where investors 25 use portfolio products as a way of accessing portfolios 0061 1 of bonds. So for example, you could buy and ETF and then 2 redeem it for the -- for the underlying cash bonds or 3 maybe create an ETF with a custom basket of bonds if you 4 were looking to try to sell a specific basket. So we're 5 seeing more innovative use of portfolio trades, as I 6 think investors figure out how to use some of the new 7 tools that they've started to incorporate in their 8 portfolio. 9 I think that the question is how does the -- 10 you know, the market or investors respond to a period of 11 market stress or market shock when we see it, and I'd 12 expect to see liquidity risk spike as well, and I think 13 that would be an accelerant to any spread widening that 14 occurred, based purely on fundamentals. I would point to 15 the experience in the investment grade market around the 16 route in oil prices as informative of the potential 17 increase in volatility that could occur. When oil was 18 deteriorating, we saw investment grade spreads peak at 19 over 200 basis points, which would normally be sort of 20 recessionary levels, fully pricing the risk of a 21 recession. 22 Now, clearly, energy companies themselves, 23 which are a sensibly large part of the market, were 24 experiencing material stress, but what we saw was that 25 even non-energy credits widened materially and 0062 1 contributed to that overall aggregate spread level 2 hitting such wide levels. And I think that it's possible 3 that poor liquidity conditions, which accelerate during 4 those periods of stress, add to the level of market 5 volatility and sort of increase the level of market 6 stress that you would feel based purely on fundamentals. 7 CHAIRMAN HEANEY: Sonali? 8 MS. THEISEN: Thanks very much. So I think, 9 you know, just as a starting point before we talk about 10 liquidity, I would like to just make an observation that, 11 you know, we believe that the credit markets remain 12 predominantly a principal market. Again, this is 13 something that we cover in Slides 3 and 4 of our 14 presentation, but, you know, we think that, you know, the 15 credit markets will continue to rely on the warehousing 16 of risk and we don't believe, you know, again, the 17 electronification efforts, the rise of passing investing 18 materially changes this dynamic for the very long tail of 19 less liquid trades and block trades that happen in the 20 marketplace. 21 So, you know, with that as a -- as a backdrop, 22 we think that, you know, if the trend continues towards, 23 you know, low volatility, more passive investing, we 24 expect to see, you know, relatively low and potentially 25 further decreases in turnover and, again, this topic of, 0063 1 you know, less homogenating in the market and diversity 2 of sort of investment strategies. And then consequently, 3 I think, you know, the metric that is most impacted is 4 just the depth of liquidity that can get done. 5 You know, we actually believe that bouts of 6 volatility within credit, particularly if coupled with 7 the ability for investors to utilize more leverage, would 8 actually, again, go back to improving the diversity of 9 investment strategies. And then, you know, with respect 10 to a macro event, we think that would lead to a repricing 11 of risk, but in and of itself would not materially 12 impact, you know, liquidity metrics that are -- that we 13 have been talking about traditionally measuring. 14 In terms of looking, you know, ahead and what 15 kind of, you know, changes potentially could be made, one 16 point that I think is foundational, and not necessarily, 17 again, directly related to liquidity today, but you know, 18 the theme of electronification has been widely covered in 19 every single paper and, you know, from all of our 20 panelists today. I think it's beholden upon us as an 21 industry to ensure that we've built a strong foundation 22 as the migration to digitization continues. To that end, 23 you know, technology standards, rulebooks, fee 24 transparency, we think, are a very high priority. Again, 25 we cover some of this in Slide 20, but if you look at the 0064 1 US fixed income markets versus, you know, the US equity 2 markets and again, now with the implementation of MiFID 3 in Europe, you know, I think we're further behind in 4 having clear standards that are given to the market for 5 this. 6 You know, despite all of our collective best 7 efforts, we view it as a when, not an if, that there will 8 be some sort of technology issue that happens as a -- to 9 the marketplace, and we think that, you know, to avoid, 10 you know, a material event, it's very incumbent upon us 11 as an industry to, you know, tackle, you know, building 12 kind of robust electronic trading practices. 13 You know, the role of data, again, we've kind 14 of touched on it in a lot of our discussions. Again, we 15 all think that the role of data will continue to grow, 16 and it'll be very interesting to see, you know, whether 17 that data kind of feeds into supporting diversity in the 18 market or making it more homogenous. You know, in other 19 words, we want to make sure that we're not necessarily 20 all listening to the exact same signals, all getting the 21 exact same data and all making the exact same, you know, 22 investment decisions based on the exact same information. 23 And I think, again, it's important to recognize here 24 that, you know, instruments may have the same -- the very 25 same instrument may have, you know, a different price 0065 1 based on size, likelihood of execution, counterparty 2 behavior, transparency features, you know, so we think if 3 constructed properly, we believe advances in the search 4 feature and data will allow market participants to become 5 more targeted in their engagement, which again is more -- 6 is helpful for improving heterogeneity, rather than 7 having the entire marketplace engage everyone, you know, 8 on again this very long tail of less liquid instruments. 9 You know, the last point that we would make 10 that we think is certainly just worth having a discussion 11 going forward is, you know, the less liquid and block 12 trades and whether the kind of one-size-fits all approach 13 of TRACE is accommodating towards liquidity in those 14 types of executions. 15 CHAIRMAN HEANEY: Thank you. You know, you touched 16 upon concentration earlier and looking at single 17 signaling type events, which kind of leads to the herding 18 potential or possibility and, again, perhaps that's a 19 second panel topic as well, but Kevin? 20 MR. MCPARTLAND: Yeah, I think the one thing 21 that liquidity providers want the most is the one we have 22 the least control over, which is just a return of some 23 good volatility and volume, which will certainly bolster 24 the whole market, but again, that's not something we have 25 much control over. If there was a single change, if we 0066 1 could sort of reincentivize liquidity providers to commit 2 that capital to this market to making the market liquid 3 for investors, that would have the biggest impact. Of 4 course, you know, it's much -- that is much easier said 5 than done, and it is a multifaceted thing to do, but I 6 think that really is front and center. 7 As for -- as for new sources of liquidity, when 8 we asked the buy-side what they thought would be the top 9 source of new liquidity going forward, their answer was 10 the buy-side, right, which talks to growth in all-to-all 11 trading. Again, I don't think that's something that will 12 take over for sure. I think the importance of principal 13 liquidity in this market and large dealer balance sheets, 14 the market couldn't function without those. However, as 15 we mentioned earlier, providing new tools and new 16 channels for access to liquidity based on needs and 17 investment strategies and whether we're in times of 18 stress or not is critical. 19 We're at about 20 percent of institutional 20 investment grade volume in the US traded electronically 21 today. If we look at e-trading adoption in Europe and in 22 other products that have gone through this cycle already, 23 in the next few years, we could very easily get to about 24 a third. There's some market dynamics and market 25 structure issues that, you know, will make sort of 0067 1 punching through that maybe a bit harder, but, you know, 2 over 80 percent of investors are trading at least some 3 volume electronically today, which is as high as some of 4 the most electronic markets, like equities and FX. So 5 there's been a tremendous amount of change in that 6 regard. 7 Things certainly won't get back to the way they 8 were. We probably don't want them to, but at this point, 9 based on all the conversations, we've had a good portion 10 of the market has sort of accepted where we are today and 11 has worked really hard to adopt their business models and 12 approaches to fit this market structure. 13 CHAIRMAN HEANEY: I'd just ask one follow-up, when 14 you say a third of the market, would you say that as a 15 consequence of volume and not necessarily the amount of 16 issues being traded or is it -- 17 MR. MCPARTLAND: Yeah, it's a -- right, of the 18 total volume traded in any given year, right. 19 CHAIRMAN HEANEY: Well, I want to thank Kevin, Jeff, 20 Sonali, this was a -- this was a great discussion, very 21 insightful, a great way for us to kick off the meeting. 22 I'd remind the committee members that all of the 23 panelists will be here after lunch so you can engage in 24 either questions and a discussion with them as you have 25 it. 0068 1 Right now, we'll take a 15-minute break and 2 we'll come back -- 3 CHAIRMAN CLAYTON: Could I just say -- 4 CHAIRMAN HEANEY: Yeah. 5 MS. WALTER: -- I want to thank each of you, 6 these materials were terrific. They were accessible. 7 They talked about the principal shifts that have 8 occurred, the drivers for those shifts. You know, it's a 9 multifaceted problem, but this discussion, you know, I 10 didn't spend my life trading fixed income securities, but 11 you pitched at a level where I get it and I -- and I very 12 much appreciate that, so thank you. 13 CHAIRMAN HEANEY: And we're going to reconvene here 14 at 11:05, please. Thank you. 15 (Whereupon, at 10:55 a.m., a brief recess was 16 taken.) 17 MR. REDFEARN: We will get started in just a 18 second. Well, we might as well get going. I know 19 Michael will be back here in just a second, and so -- but 20 he said to go ahead, so on my remarks here, I want to 21 thank him. Maybe I'll thank him afterwards. Let me just 22 start by saying that I am excited to pick up where we 23 left off on the last panel. It was a great discussion. 24 Thank you very much. I thought there were a lot of good 25 data points. For those of you who haven't seen a lot of 0069 1 the research on this, it's well worth looking at. 2 There's just a phenomenal amount of data and insight that 3 comes from those reports, so I would encourage folks to 4 take a look at that. 5 We have our second panel now ready to get 6 underway, and we are going to be discussing a number of 7 different topics, but some of them will include dealer 8 intermediation and the evolving role of dealers and 9 institutional investors. Electronic trading, we're going 10 to pick up on some of the themes that were raised earlier 11 and look at the changing role of technology and trading 12 venues. 13 Of course, regulation and its impact on market 14 liquidity, and also, some potential policy considerations 15 that could foster more and efficient markets. So just to 16 briefly recap some of the themes from the first panel 17 that are worth noting and that will continue to come up, 18 one, of course, is the shift from principal trading to 19 more agency trading by dealers. The point was raised 20 about banks being disincentivized from providing primary 21 principal liquidity. Another issue was the change in the 22 diversity of investor base and how that diversity has 23 decreased and, in particular, there was a point made 24 about foreign funds and less turnover. The increased use 25 of portfolio products like ETFs and the more innovative 0070 1 use of these products and the impacts that that's having 2 on the marketplace, including on the liquidity of the 3 market, and new e-trading opportunities, but still 4 uncertainty regarding the evenness of the electronic 5 trading impact across the marketplace when you compare 6 the more liquid segment of the market to I think what was 7 described as the long tail of less liquid products in the 8 market. 9 And also, of course, there was a recognition of 10 MiFID II. As all of you know, this went into effect just 11 earlier this month, a huge regulation in the European 12 market where there have been a number of changes that 13 were put in place that will affect the fixed income 14 market and bond trading and pretrade transparency and a 15 whole set of issues that we will certainly be watching as 16 time goes forward. So building on that discussion, let 17 me focus first on introducing our panelists. I want to 18 start with, let's see, we have Jim Switzer from Alliance 19 Bernstein right over here. And Jim is the global head of 20 credit trading. 21 Next to him, to his right is Richie Prager, 22 head of trading, liquidity and investment platform at 23 Blackrock. Just next to him is Drew Mogavero, who's the 24 head of US flow credit trading at Barclay's. He's our 25 sell-side guy. And then we have, lastly, Paul 0071 1 Jakubowski, global head of credit at Vanguard. So thank 2 you all very much for being here and participating in 3 another panel -- excuse me, we're looking forward to 4 kicking off on some of the things where we just left off. 5 So let me start with just a general question to 6 all of the panelists, I'd like to ask based on your 7 personal experience from all of the themes that you just 8 heard, let's just sort of have a little bit of a 9 transition here and tell me what are the themes that you 10 just heard that resonated most for you, and I guess we 11 can start at the end there maybe with you, Jim. 12 MR. SWITZER: Thank you very much for having 13 me. So let me start -- let's start with a story. A 14 couple years ago, I was sitting around a table, dinner 15 with a bunch of buy-side and sell-side participants, and 16 we were debating market structure and liquidity and 17 what's going on. And there was a person that's on the 18 FIMSAC Committee that was there that asked me a question. 19 He said, "Jim, are we talking about liquidity or are we 20 talking about efficiency?" And that was my ah-ha moment 21 where I went, "Wait a second, we're looking at this 22 wrong. We're trying to solve for liquidity, but we 23 should be solving for efficiency first because we can't 24 get more liquid until we figure out a way to get -- to 25 get more connected." 0072 1 And I started to look around Wall Street, and I 2 started to look at the changes, and I started to look at, 3 you know, more people from the equity side moving over to 4 the fixed income side and running that equity playbook, 5 and I -- and I started to figure this out that how do we 6 do this. So for electronic trading to start to really 7 move, how do we become more efficient? How do we process 8 trades? How do we -- how do we move smaller trades off 9 our desk so that we can focus on the larger things? 10 So, you know, a lot of what we talked about 11 over -- you know, over here was liquidity and so on and 12 so forth, but for me, it's -- for me, it's about -- it's 13 about efficiency. And it's about behavior modification. 14 Blackrock wrote a white paper many years ago with a 15 number of suggestions, and the -- I think the fourth 16 thing on that list was behavior modification. And I 17 think that is actually the most important thing, and 18 that's what we've seen change. I mean you look around 19 this room here and the people that we have in this room, 20 four or five years ago, I don't know that we all knew 21 each other, but I think now, we all know each other very 22 well because we're constantly involved in these 23 discussions about what do we need to do? What do we need 24 to change? You know, we don't want a MiFID like solution 25 here, we want a solution that all of us around the table 0073 1 can come up with it that's really going to solve our -- 2 solve our problems. I'll just leave it at that. 3 MR. REDFEARN: Thank you, Jim. 4 Richie? 5 MR. PRAGER: Thank you, Brett, and also, I'd 6 like to thank our chairmans and the Commission for 7 hosting this discussion. And so I -- just a few 8 comments, and you asked what resonated, a lot of it 9 resonated. I think, though, we should always step back. 10 It's good to really set context when we ask ourselves 11 about liquidity. You know, liquidity's not a God-given 12 right, and as a former colleague of ours at Blackrock, 13 Peter Fisher said, you know, liquidity is really a 14 function of confidence, and if there's confidence in the 15 market and there's confidence in, you know, what you're 16 doing, you're -- it's going to be liquid, and that's a 17 good thing. 18 But, you know, we talk about the tail, there's 19 nothing that says that every bond that's ever issued is 20 meant to be traded. So I do think we always have to 21 have, you know, context and ask ourselves what are we 22 solving for. And we also have to ask ourselves which 23 benchmark we're using. We keep talking about since the 24 crisis. I don't know anyone who really wants to go back 25 then to that. Is that the right benchmark? You know, 0074 1 the first long bond I traded matured five years ago, so 2 all that says is I'm kind of old, but so I think you 3 really -- you know, to go back to, you know, the crisis, 4 as the benchmark and say that's when things should be 5 compared to, you know, I think we've got to ask ourselves 6 is that -- is that appropriate. You know, there was a 7 little sidebar over here, and I think Drew, you said, you 8 know, everything keeps changing and it's evolving. 9 I think that's like such an important point, 10 and we are in constant evolution. And the behavior, you 11 know, it's -- you know, the buy-side, we've had to make 12 tremendous behavior changes in terms of investment 13 strategies, what's appropriate instruments, what's not, 14 portfolio construction, business models on the sell-side 15 have changed. I mean so that's natural. That's 16 evolution. The different participants, you know, the 17 electronic market-makers, the buy-side becoming a price- 18 maker, you know, these are all, you know, natural 19 evolutions, different technology, different protocols, 20 different products, new products, innovation is certainly 21 a big part of this, ETFs we'll talk about; you know, all 22 of the different venues, all to all. 23 This is all -- you know, this is all very 24 healthy. Bid-ask, we talk about it a lot. I think we -- 25 you know, it sounds like a fruit salad, though, we're 0075 1 comparing apples and oranges. To go back and compare the 2 bid-ask spread of a purely principal market to a bid-ask 3 spread of a hybrid market structure, which is part 4 agency, part principal, it's top of book. I mean I 5 really think we should, you know, unpack some of these 6 things and just not -- you know, I think that's some of 7 the research I've seen that I really worry about, what is 8 the meaning when they say bid-ask, are we actually saying 9 the same thing. 10 You know, and then things like what's the 11 definition of best ask, that's evolving, you know? So 12 this evolution we have to look at. I really think it's 13 important that everyone needs to pause and make sure 14 they're educated on the issues before we, you know -- you 15 know, run too quickly, and whether that's the education 16 of the role of an asset manager versus the role of, you 17 know, an asset owner, I heard some of the concentration 18 points. I agree with some; I disagree with others. We 19 have 100 -- over 120 different investment teams at our 20 firm, all doing different things for different clients 21 with different mandates, and they all trade through one 22 desk. That doesn't mean that one desk has any authority 23 to do anything outside of what the client's mandate is 24 and what the fiduciary who was the portfolio manager says 25 to do. 0076 1 So we really, again, let's unpack these things 2 and make sure we understand, we understand what ETFs do, 3 the elegance of ETF technology versus an open-end fund. 4 So I think the education, still more work to be done 5 before we're all, you know, saying the same things there. 6 And then I think it's -- the last comment is this, we'd 7 really have to look at things in the round, you know? So 8 when literally everything that was said here by the end, 9 you know, Michael, you took everyone through it, you 10 really started to understand how these things are 11 interconnected. You know, how the portfolio trade is 12 touching the ETF, which is touching this, and if we just 13 look too narrow, I think some of the research, and I 14 think this panel did a really nice job. But some of the 15 stuff I've read and some of the academic stuff I've read 16 is just way to narrow. 17 We've just got to take -- you know, look at 18 this in the round and just make sure we make informed 19 decisions, so that would be my opening comments. 20 MR. REDFEARN: Thank you. 21 Drew? 22 MR. MOGAVERO: Thank you, everybody for having 23 me here and I'm excited to have a nice, robust dialogue. 24 Before I get to the points that resonate from the first 25 panel, just two things that have been touched on already 0077 1 by Jim and Richie, which really resonate with me. First 2 is I couldn't agree more with the comparison to the pre- 3 crisis era. To me, it's like comparing it to the, you 4 know, steroid era in baseball when guys who looked like 5 me had 50 homeruns, you know? It's just like -- it's not 6 the right comp. And then I would add on a second point 7 around the evolution and how fast it's happening, I 8 always like the movie Jurassic Park, and there's a line 9 in Jurassic Park where the crazy scientist guy goes, he's 10 like, "How do two female dinosaurs mate?" And then he 11 just says, "Life will find a way." And it's like if you 12 look at what's going on in the market right now and you 13 go through a lot of the data and you look at the stuff 14 and you say like, "Geez, how's the liquidity going to be 15 the same? How are we going to handle this massive growth 16 in the market? How are we going to handle that things 17 are different?" 18 And I think that literally evolving by the -- 19 at least from my perspective as a dealer and how we're 20 engaging with our clients and how we're using technology 21 and how we're using e-trading and just how we're thinking 22 about our business, it's changing by the month. And so 23 we have got to keep that in perspective, that things are 24 evolving quickly and that you have to let life find a 25 way, to a degree. That doesn't mean we can't have a 0078 1 good, smart dialogue and discussion, but I think it's 2 important to keep that in mind. 3 Getting back to just the thing that resonated 4 the most with me from the prior panel, the word has been 5 said a lot today, you know, Commissioner Stein spoke 6 about it straight away, it's the word concentration. I 7 think as I think about what -- as we move through the 8 discussion around what has changed, how is the sell-side 9 role differing, how is e-trade and ETFs, any of these 10 things, but I'm sure we're going to spend the next hour 11 talking about the concentration issue is one that is 12 important from in terms of how we compete as a sell-side 13 dealer and how we engage with the clients that, you know, 14 hold a large amount of assets and are a big important 15 part of the ecosystem. And so I think that's -- it's 16 been mentioned a lot of times. I wish I had kept count, 17 but I'm guessing it was probably around 15 to 20, and I 18 think -- so it's an important -- it's an important theme 19 to kind of just override as we advance the discussion. 20 MR. REDFEARN: Fantastic, thanks. 21 Paul, you've got Jurassic Park. You've got 22 baseball. You've got -- 23 MR. JAKUBOWSKI: I don't have any good 24 analogies, but thank you again for having us. We applaud 25 the committee for forming this. It's very important to 0079 1 the markets. And the one thing, and I'll take a 2 different -- I agree with everything that the previous 3 people said. The one thing I'll take a different angle 4 on, and I think Kevin said this, was there needs to be a 5 focus on the trades that weren't executed and how can we 6 stimulate those trades to get done, whether it's through 7 better data, better technology, better ways to transact 8 essentially market structure, right? 9 I think we spend a lot of time focusing on the 10 trades that were done, but to really improve liquidity, 11 how do we get those trades that someone couldn't find a 12 level or didn't know what the level was but wanted to 13 trade, someone didn't have a deep enough pool for 14 liquidity but wanted to transact; someone didn't have the 15 efficiency, and I think that word was used a couple of 16 times, that they can transact, whether you're a large 17 institutional client or whether you're a retail client. 18 And I think that's really got to be the key question that 19 we ask ourselves is what trades didn't get done and why, 20 and how can we focus on solutions and moving things 21 forward and evolving so that a larger percentage of those 22 trades do get done, and they get done with a low cost and 23 with very best execution to the benefit of, again, both 24 institutional and retail clients. 25 MR. REDFEARN: Fantastic. Okay. So now we 0080 1 want to move on and talk about dealer intermediation. 2 And we heard themes about this sort of increasing shift 3 for principal trading to agency trading and some other 4 dynamics in terms of keeping inventories and so on, and I 5 think, Drew, we want to start with you as the sell-side 6 representative, if you can expand on Barclay's 7 intermediation of bonds and how has that changed, I hate 8 to say since the financial crisis, but how has that 9 changed lately and perhaps, you know, give an example, 10 too, of what do you see changing from that perspective? 11 MR. MOGAVERO: So I think to answer that, you 12 know, question, I'd break it down into two parts. You 13 know, what has changed and also, what hasn't changed. So 14 I think, you know, what has changed, and again, bringing 15 it back to that -- to that concentration theme, you know, 16 for us, from my perspective on the sell-side, it does 17 seem like whether it's due to regulation, cost of balance 18 sheet, cost of compliance, there are less dealers that 19 have the full suite of global trading products, you know, 20 with that I mean it's just they're strong across all 21 areas in terms of the ability to provide that broad suite 22 of products. 23 I'd say, you know, back in -- it doesn't have 24 to be the financial crisis, but back in the day, for lack 25 of a better word, you had a lot of niche players, and 0081 1 they definitely still do exist, but I think as technology 2 expands as to be good at trading ETFs, to be -- have the 3 technology infrastructure, to be good at electronic 4 trading, to be able to handle the, you know, costs of 5 regulation, you need a deep pool of resources that you 6 can provide to your clients to handle that, so I feel 7 like put all of those things together and it does lead to 8 more of a concentration where, for me as a dealer, I feel 9 back in the day, I could maybe get by by having -- being 10 good at a few things and pick my spots a little more; 11 there was plenty to go around, plenty of activity, plenty 12 of trading. Now, I just feel it is imperative to be in 13 that top three position, you know, with the core clients 14 and to have the full suite of products. 15 So I feel like that is something, you know, 16 that has changed from concentration levels so that we can 17 best service our clients. What has not changed in my 18 mind, and I think this is important, is at the end of the 19 day, we trade credit. You have to know your companies. 20 You have to know we're seen as experts who know how to 21 respond to volatility. All bonds are certainly not 22 created equal. What are the -- our clients want to know 23 what are the good bonds to be buying when things are 24 dislocated? What are the, you know, bonds that have some 25 stress in the credits that we should sell when the market 0082 1 is frothy? And that expertise in bond-by-bond, sector- 2 by-sector, product-by-product, region-by-region, I don't 3 feel like that is going to change no matter we've seen it 4 over, you know, the 18 years I've been doing this and 5 certainly the five plus -- the matured bond, long bond 6 that Richie's been doing. It is that you're going to -- 7 it's been that way, and I don't see that changing. 8 And I think that that just -- that that leads 9 to trust, transparency and just having an honest dialogue 10 with your -- with your clients and internally, like that 11 part of the market, if anything, it's certainly not 12 changing, and I would actually argue that because of the 13 concentration issues, it's actually increasing, like the 14 need for when I talk to, you know, traders or PMs or 15 whoever at our clients, the color and information, not 16 just a level or a price transparency or anything like 17 that, but just the color and information before a trade, 18 during a trade, after a trade because there's only -- 19 there's less and less dealers who do it well, and there 20 are more and more clients with a lot of assets that need 21 that service that that trusted transparency part 22 certainly hasn't changed, and I would argue just 23 continues to become more and more of a differentiator for 24 sell-side dealers that want to do it well. 25 MR. REDFEARN: Thank you. So Richie, you're 0083 1 dealing with a lot of different dealers in the 2 marketplace, and so you're seeing sort of a broad array 3 of responses to the current market environment. What can 4 you tell us about sort of your view from the buy-side in 5 terms of how you're experiencing changes in dealer 6 participation? 7 MR. PRAGER: Okay. Thank you. So again, I 8 think I want to just parse it into what we see, and then 9 the what do we do. So every single firm out there is -- 10 they're solving for their own business needs, so we have 11 Drew, Barclay's, they're going to look very different 12 than the next bank; the U.S. bank's going to have a larger 13 corporate footprint. They're going to have a bigger 14 calendar. They're just going to be -- it's just 15 different. So when we look at everything, actually, we 16 don't see the same concentration issue because we have -- 17 as a fiduciary for our clients, we have to solve for our 18 client's liquidity need, which means then we need to be 19 resourceful, so we, you know, think that the -- you know, 20 the market is evolving, it is becoming a hybrid market, 21 the principal market is here, and it's here to stay. 22 Banks are some of our most important partners, 23 but they can't do everything for us and nor should they 24 do everything for us, so, you know, we see the 25 alternative liquidity providers come in; we see more 0084 1 liquidity coming from the buy-side itself, and we see 2 liquidity coming from new financial instruments such as 3 ETFs. So, you know, there's no -- you know, the 4 intermediation model is still alive, it just depends on 5 which bank you're talking about. So while I do agree 6 there's still a very top tier of commercial banks or 7 universal banks that are proficient in the principal 8 market-making game, and that number may be fewer, the -- 9 you know, the access to the liquidity is actually quite 10 broad, so that breadth forces us to respond. 11 And so, you know, the response for us really 12 breaks down into what we said are sort of three Ts. 13 There's technology, tactics and TCA. And the technology 14 is it's forced us to invest a lot of money to be able to 15 reach all that breadth of liquidity where all of those 16 liquidity pools may be, you know, regional dealers, large 17 firms, alternative market-makers, exchanges, et cetera. 18 And tactics is a fancy word for judgement and people, it 19 just happens to start with T, so I can say three Ts. 20 And, you know, it's -- you know, there's a lot of new 21 skills that are required in terms of, you know, when to 22 use technology, when not; when to be a price-maker, when 23 not, and that's really important now. 24 And then the last is TCA, I mean without the 25 tools of transaction cost analytics to really test the 0085 1 quality of execution, to test the quality of the venue, 2 the judgement of the people who were deciding what to do, 3 you know, without robust tools, it's really hard to see 4 what we're doing. 5 MR. REDFEARN: Thank you. 6 So Jim, to you in terms of the changing 7 experience you're having with dealer participation in 8 intermediation? 9 MR. SWITZER: Sure. So I guess to start off 10 with the one premise is that myself, Richie and Paul, you 11 know, we provide daily liquidity to our clients, and the 12 question we have to ask ourselves is does the market 13 provide daily liquidity to us. So if you go back 14 precrisis and you -- and you look at the difference 15 between then and now, and I think -- I think the big 16 change, as we've touched on, is dealer balance sheets 17 look different. Depending on who you ask, they're much 18 lower. Some people argue that they're not, but I will 19 tell you the composition of those balance sheets is very 20 different. 21 You don't have the prop desks anymore, and you 22 certainly have the leverage and the big balance sheets 23 that the hedge funds were running in the credit space are 24 curtailed. And, you know, I worked for Steve Cohen, and I 25 was at UBS Principle Finance, I saw those balance sheets, 0086 1 and they were -- they were robust. Why is that 2 important? You know, that was the other side of the 3 trade. That was the uncorrelated capital to daily fund 4 flows, and I think that's a big change. So that, to me, 5 is the buffer capital that is kind of missing in the 6 market. 7 And that's forced us to change the way we've 8 managed portfolios and the way we manage liquidity in the 9 portfolios and just it's forced us to change our 10 investment process, you know, entirely, if you will. The 11 concept of the buy-side becoming the new price-maker or 12 market-maker, some people -- we're never going to be a 13 market-maker, that's not us. The price-maker, yeah, when 14 it's convenient for us. So if you're relying on the buy- 15 side to really fill those shoes, I don't -- I don't think 16 that's going to happen. 17 And then this just brings us back to the 18 efficiency side of it, so okay, what do we do? The Holy 19 Grail in the marketplace, in my opinion, and what I would 20 recommend to the Commission to think about is pretrade 21 transparency. It's about gathering information and 22 there's so much information out there, and bringing it 23 all together in one place. We've invested a lot of 24 money, much like Blackrock and Vanguard, in terms of 25 aggregation tools to bring all of that information to the 0087 1 front. 2 Now, that being said, the dealers play a very, 3 very vital role. There's no disintermediation that's 4 going to happen here. We can't function without that. 5 And I know ETFs are going to come up at this, so I'm not 6 going to touch on it. I'll let these guys start and then 7 we'll talk about the ecosystem that that's actually 8 created and how we're using that and how we're using the 9 dealers to actually facilitate and look for new pools of 10 liquidity. So I'll leave it at that. 11 MR. REDFEARN: So back to you, Paul. Why don't 12 you tell us your thoughts on this issue? 13 MR. JAKUBOWSKI: Yeah, I want to talk a little 14 bit about cyclicality. You know, I think Drew brought up 15 some points about dealer balance sheets shrinking, but, 16 you know, historically, and Drew may disagree with this, 17 that, you know, dealers have been procyclical. When 18 liquidity has been poor, like they have tried to shed 19 risk, too, amplifying the problem. And to the extent 20 that banks have been shored up and balance sheets are 21 smaller, right, that risk has somewhat been mitigated. 22 They still play a key role in the marketplace. The 23 second thing I want to talk about is just target date 24 funds. The growth of target date funds, again can be 25 counter-cyclical to the market, right? As you see big 0088 1 moves or big volatility in the market, they may actually 2 step in and buy as they rebalance at the end of the month 3 and provide another source of liquidity to the market. 4 So I think those have been a couple of things that we've 5 seen may have actually enhanced liquidity and actually 6 made the market safer to transact in. 7 MR. REDFEARN: Excuse me, I just want to ask 8 one follow-up question. So Richie, you had mentioned the 9 alternative liquidity providers and since we're talking 10 about sort of provision of liquidity, I'll just throw 11 this out generally to anybody on the panel, just what are 12 you seeing in terms of the emergence of alternative 13 liquidity providers, and a little bit of a side-bar 14 question there in terms of whether or not there are 15 impediments to the emergence of that or not? 16 MR. PRAGER: I'll just pick up to start. And I 17 think someone in the research panel mentioned earlier, I 18 mean they're not really a counterparty to clients, to us, 19 okay? So it's not as if we see them directly, but they 20 live on the exchanges. And what they are doing at this 21 point, which is different, and again, I think it was 22 Kevin who said it, I agree with Kevin, they're playing 23 the fixed income market very differently than they're 24 playing the equity market. What they're doing in the -- 25 in the fixed income market is actually -- they're 0089 1 actually connecting the dots. So if you think about, you 2 know, my comment before about, you know, well maybe you 3 see this concentration among banks. Yeah, we see a few 4 banks over here, and then we'd see the rest of the world, 5 which is, you know, all of the different venues. And 6 what they can often do is marry up the fragmented 7 liquidity and bring it to you. 8 So that's the role in many cases, and that 9 means whether that's -- you know, they're addressing the 10 fragmentation across just marrying two venues. They 11 could be addressing the fragmentation by bringing the 12 exchange market, which is ETF markets, to the OTC market. 13 So actually, they are providing an important function. 14 That said, it's -- they're not necessarily solving for 15 the block liquidity, but they are -- it's another reason 16 you're seeing a lot of, you know, top of book type bid- 17 ask spreads because they're actually in there tightening 18 up all of that fragmented liquidity. 19 MR. REDFEARN: Anybody else on that one or 20 shall we move on? Okay. So the next topic we have is 21 we're going to switch over to electronic trading. 22 Obviously, this is a big development and an important one 23 in fixed income markets. I do want to get a better sense 24 of how electronic trading is evolving in corporate bonds 25 and a read on how this is perceived both, you know, in 0090 1 contrasting the sell-side view and the buy-side view. So 2 let me throw it out there. I'll let whoever wants to 3 jump in first on this, but just some examples that you 4 can give about how you're seeing electronic trading come 5 into the business and what trading protocols you're 6 seeing emerging and so on. Thank you. 7 MR. MOGAVERO: I can start it off from the -- 8 from the sell-side things. Electronic trading has been - 9 - it's been a major growth area and something that we 10 really embrace. I think it plays to your advantages that 11 we talked about before as a -- as a top three dealer 12 where if you're able to potentially just be able to trade 13 the bonds for your clients and know where the prices are 14 and be able to really take advantage of your position to 15 price a lot of different things, you know, we've created 16 investments in technology for us to have grids in place 17 that, you know, populate all of our levels where we can 18 quickly and automatically, whether it's a human or an 19 algo, respond to inquiries just faster, you know? 20 We want to do -- generally, the sell-side, we 21 want to do more trades with more of our clients, and we want 22 to do them faster and as efficiently, to Jim's point, as 23 possible. And electronic trading allows us to do that, 24 and it allows us to just kind of get more reps and to 25 know -- to know where -- you know, where the prices are, 0091 1 and that helps us with trades, both large and small, for 2 clients both large and small. And I think it's 3 something that we've really embraced. I think there's -- 4 you know, we'll talk about it more with ETFs when we get 5 there, and I think -- and Richie brought up some great 6 points about some of the alternative market-makers that 7 trade frequently electronically, and we trade with them. 8 So it is an ecosystem that is -- that is -- 9 provides some nice cyclicality to the market, where 10 people are -- the price discovery process on where bonds 11 are trading based on at the second conditions in the 12 market, whether it's through an ETF or through an 13 exchange, through an inflow and outflow that an asset 14 manager had, those are trading real-time, and the levels 15 are getting, you know, published to the marketplace real- 16 time, and so we're -- we at Barclay's are big proponents 17 of that. You know, we've put a lot of effort into it in 18 terms of really from the top down from a cultural 19 perspective of making sure that our traders know that 20 this is important, this is growing, if you're going to be 21 a good trader, you've got to be good at this. 22 MR. SWITZER: So look, I would -- I would say 23 our goal by the end of the year is to trade -- every 24 trade we do is going to be done electronically. That 25 doesn't mean the voice trade is over. It just means 0092 1 it'll be processed electronically. So I think you need 2 to kind of break this up into compartments. You take 3 credit, something as simple as, you know, auto x-ed where 4 we -- where we price a corporate bond at a spread over 5 treasury. We can now process that on electronic 6 platforms, auto spot, create a lot of efficiencies. Auto 7 x-ed where we can take low-hanging fruit, and we take, 8 you know, one million or two million or five million 9 under, depending on a liquidity score that we put on it, 10 and we can just auto route it on an electronic platform 11 and execute it. 12 Now, for that to work, again, it comes back to 13 transparency and understanding where the markets are 14 trading and that's ultimately -- there's many people 15 sitting around this table that are trying to come up with 16 composite pricing or constantly evaluating prices, and I 17 think that's going to be a big part of this solution down 18 the road. You know, when you -- when you get away from 19 that and then you start to talk about the platforms where 20 there's all-to-all trading, again, I think all-to-all 21 trading is here to stay, and it will only grow as the 22 market gains more and more confidence around, you know, 23 what the correct price of a security is. 24 And look, lastly, and, you know, I keep 25 alluding to the -- to the -- to the ETFs, is lastly as we 0093 1 -- as we get more and more involved in the ETF ecosystem, 2 I think electronic trading is going to -- and electronic 3 platforms are going to play a big role in the 4 transmission of information and the execution and the 5 negotiation, which is really going to help us because, 6 look, you look at the markets today and volatility's low, 7 you have a lot of time to make a decision to do a trade. 8 But as volatility picks up down the road and we get into 9 -- into those more volatile times, its speed is of the 10 most -- it's the most important priority that we have is 11 to be able to gather as much information as we can, as 12 quickly as we can, make a decision, process the 13 information and get out there and do it. And that's 14 where electronic trading is going to help us to do it is 15 the speed side. 16 MR. JAKUBOWSKI: Yeah, I agree with a lot of 17 what was said already. I think it's crucial that you 18 connect the data to the electronic system to your back 19 office and have it flow smoothly, like that just speeds 20 things up. It reduces risk. It's really essential. I 21 would say something else that, you know, needs to evolve, 22 and I think electronic trading has evolved and we've 23 obviously used it a lot more than we have in the past is 24 just the fragmentation. There are so many systems out 25 there, and it causes that you have individual pools of 0094 1 liquidity that you either need to aggregate or 2 eventually, you know, there has to be, you know, a 3 handful of systems that we now -- like there are just too 4 many systems, I think, you know, I don't know for Jim and 5 Richie, but, you know, we probably have 20, 30 people 6 come in every year showing us a new system. And, you 7 know, with those systems, what they need to focus on is 8 how can the pool liquidity; how can they get better 9 execution, and less so on some of the technology bells 10 and whistles that come along with some of these things. 11 I think that's what they're trying to 12 differentiate themselves on, and I think they need to 13 focus on how do they change market structure and how do 14 they benefit all liquidity providers and all liquidity 15 consumers, but there are just a lot of fragmentation, a 16 lot of providers out there. 17 MR. PRAGER: And just a few more points, I mean 18 I agree with most of the comments here, and like Jim, our 19 goal is to be 100 percent electronic in, just as you 20 described it, some part of it, even if it's voice and 21 then, you know, we record electronically so we can STP 22 down. So I talked about earlier the three Ts that we 23 have. We also slice everything we do into high-touch, 24 low-touch and no-touch, so depending on the asset class 25 and how standardized and how available liquidity, 0095 1 something just might be auto x-ed out of the gate, you 2 know, futures and treasuries, they just are auto x-ed. 3 Today, you know, even corporate credit, about two-thirds 4 of what we do, by transaction number, I know you were 5 asking yourselves before in terms of what is the right 6 metric in terms of transaction number, we already do 7 about two-thirds electronically. 8 However, by volume, okay, it's very different. 9 It's only about 30 percent. So there's still a lot of 10 the volume gets done in -- on a high-touch voice basis. 11 It is here to stay. You know, if you look at some of the 12 innovations out there, look at market access, open 13 trading, all-to-all, I follow their stats every month, 14 you can see the adoption that's going on. When I heard 15 Kevin's stat before that 73 percent of the clients say 16 they're worried, those are the -- those are the dinosaurs 17 that don't want to change. That's just, you know, you 18 could ask the question another way, how many dinosaurs 19 are there? Seventy-three percent. 20 You're just going to have to, you know -- 21 you're just going to have to evolve and use these tools. 22 And the more protocols, the more innovations that come 23 from the platforms, and whether it's open trading or 24 auctions, and specifically ones that can help with the 25 blocks because that is where, you know, you can look at 0096 1 how the market has naturally evolved and electronic 2 trading has really helped in so many ways, but we still 3 have, you know, in terms of the larger block size, I'd 4 say there still is a challenge, and a lot of that's just 5 breaking it up and trading it in smaller mounts. But I 6 think, you know, that's probably where there's work to be 7 done. 8 MR. REDFEARN: And just to follow up on that 9 for anybody in terms of the types of products where 10 electronic trading is more suited versus others, I mean 11 obviously we do have very liquid products, and we have 12 others that are extremely illiquid, you know, how do you 13 see electronic trading as sort of making its span across 14 the field? 15 MR. PRAGER: Again, I think -- sorry to jump 16 in, I think this was the point that Jim was making and we 17 look at it exactly the same way. There's all -- 18 electronic -- the front end process lends itself, you 19 know, electronically somewhere along that. And even if 20 it's a voice trade, the sooner you can capture that trade 21 capture, that electronically to make sure your STP's 22 going well, you're confirming that the trade's matching, 23 so, you know, there is the, the very standardized 24 products you can just auto x right through, you know, hands 25 don't need to touch it if you're integrated between your 0097 1 OMS and your EMS. 2 So it really depends on the product, the more 3 standardized, the less humans need to intervene, the more 4 the machines can do, the more illiquid voice you have to 5 have, but you want to capture that as soon as possible 6 electronically. 7 MR. SWITZER: I would just also add away from 8 the electronic platforms, there's electronic trading 9 platforms. There are other -- there are other electronic 10 platforms that are delivering information. You take a 11 Symphony, a Project Neptune or even Bloomberg, right, and 12 that information's coming through and, you know, we have 13 technology and we've spent a lot of time building it to 14 basically aggregate this information together, which 15 gives us a picture pretrade of everything that's out 16 there. We've aggregated everything. We know everything 17 that we have internally. 18 Everything that manually a trader would have to 19 do, which then gives us the opportunity to decide which 20 electronic platform or which dealer or which, you know, 21 phone we're going to pick up. That's the most crucial 22 part. And that's the speed part because if we can 23 aggregate the information and we have it, ultimately, you 24 know, all we're doing here is copying the equity markets. 25 The equity markets have aggregators and EMSs, you know, 0098 1 plenty of them, right? So you think about what we're 2 trying to do here, is just -- is just to bring that 3 efficiency forward to be able to do that. So there's 4 electronic platforms, there's data firms, and there's 5 just utilities, conduits that are pushing information to 6 us. 7 Somebody told me, I believe, that we get three 8 million messages a day, unique messages that come into 9 the firm. How do you -- how do you have a trading desk, 10 you know, an IG desk of four guys with eight eyes that 11 can look at three million messages and actually make any 12 sense in rapidly changing and dynamic markets? You know, 13 that's where -- that's where we're headed, and that's 14 where the market, I think, has to go. 15 MR. REDFEARN: Very good. Okay. I want to 16 move on to the next topic, which is on ETFs and bond 17 funds. So several panelists had mentioned earlier 18 talking a little bit about the ETFs and mutual funds, 19 including the increased usage by both retail investors 20 and institutional and portfolio managers, and in the case 21 of ETFs, in cases as a substitute for less liquid cash 22 bond holdings and high-yield credit portfolios. So the 23 first question I wanted to ask was do ETFs pose any 24 special risks to investors and the markets in times of 25 stress? And maybe, Paul, I'll start with you on this. 0099 1 MR. JAKUBOWSKI: Yeah, we would say no, and I 2 think the history would show that, and I think a good 3 example is there was recently a high-yield fund that was 4 highly concentrated, highly illiquid that got into 5 trouble, and the high-yield ETF market functioned 6 beautifully and served the purpose. So ETFs are just a 7 better way of price discovery on an intraday basis, and 8 they're much more aligned or correlate to regular mutual 9 funds than they are to any other product that's out 10 there. And they function in very much a similar way, 11 through price discovery. And, you know, I think a lot of 12 the data that, you know, Jeff threw out earlier supports 13 the fact that they do not pose any risks to the system. 14 MR. REDFEARN: Drew? 15 MR. MOGAVERO: Yeah, I would add, I definitely 16 do not think so. I think on one hand, you know, the 17 prices sometimes when you're trading with ETFs are 18 happening quickly, they may be happening outside of a, 19 you know, "pretrade level," but maybe I'm -- maybe I'm an 20 old school trader, but the -- you know, to me, I've 21 always grown up getting -- get bonds to the level where 22 they trade fast, and then you'll get the most liquidity, 23 and you'll do the -- you'll give an opportunity for 24 buyers and sellers to meet. And I feel like the ETF 25 process really does -- it does that exceptionally well. 0100 1 And it allows for alpha opportunities for our clients, 2 where sometimes the baby gets thrown out with the 3 bathwater and you can find those good bonds, and this is 4 where, you know -- where the sell-side steps in and says, 5 "We really like the security, and the ETF's selling it 6 just because it's one of 450 bonds, so let's take 7 advantage of that dislocation." So we've found where 8 that bond trades, what's the next bond? Let's find where 9 that one trades. 10 And it happens real-time, quickly. It's -- I'm 11 a very big proponent. The other thing I would add, too, 12 is, you know, spoke about it earlier is, you know, 13 Commissioner Clayton around the perspective of the retail 14 investor. And I think one thing that's really great for 15 ETFs is when you're doing redeems, you know, I'll do a 16 redeem basket, and my traditional, you know, trader would 17 be like, "I don't want 450 YMs with 20 of each one of 18 them, this is crazy." And it's just the wrong way to 19 look at it. You instead, by having a whole portfolio of 20 securities, then even if it's just 100 bonds on an 21 electronic exchange, when someone comes in, is like, 22 "Hey, I need 50 of those," back in the day, that price 23 would've been marked up or something, and now, it's just 24 like I don't differentiate, you know? I've got -- I 25 picked up 200 bonds from a redeem; someone wants 50 of 0101 1 them, here's the context. This is where I see the mid 2 is. I know that the ETF will sell to me below the mid, 3 so I'll happily sell them to you just a little at the mid 4 or above the mid, whatever. 5 It's just because that ETF process is creating 6 not just liquidity for large inflows and outflows that, 7 you know, Jeff spoke about for the large desk managers, 8 but I think it's going to naturally lend itself to 9 smaller trading for smaller investors electronically that 10 can be through Barclay's; that can be through an 11 alternative liquidity provider. It could be from a lot 12 of different sources to some of the points that Richie 13 was making. And I think that's something that as we get 14 more involved and grow our scales in trading ETFs, I 15 personally noticed that, and I think that's something 16 that's definitely worth highlighting. 17 MR. PRAGER: I think that was incredibly well 18 said. I mean I think that when you look at the way the 19 question was framed, I also say no, but I think that the 20 risk is just the lack of education in terms of 21 understanding the product, the technology, the -- both 22 the primary and the secondary nature of the -- you know, 23 that we talk about terms create and redeem, which is the 24 primary how they're created and how they're -- you know, 25 how we unzip them and access the OTC market. You know, 0102 1 they are a financial instrument. They're used as a 2 financial instrument. There's many use cases for them, 3 and, you know, I think if you do look at the data, and I 4 know Jeff, you said there's only a couple of data points, 5 you can actually look at many data points in terms of 6 points of stress taper, going back taper tantrums, you 7 know, August 24th, and look at what happened with fixed 8 income ETFs. They behaved beautifully. 9 In fact, if you look at the liquidity spikes 10 around all of those market traumas, you know, I think you 11 also mentioned the energy sell off, and there was a high- 12 yield and credit phenomenon at that time. If you look at 13 the intraday volumes of HYG and JNK, you saw, you know, 14 just tremendous spikes of volume showing you that is -- 15 that was the market. That's where the liquidity went to. 16 That's where there was the transparency. And, again, to 17 your point from the retail perspective, complete 18 democracy retail institution is the same price. 19 So, you know, it's -- we don't see those risks, 20 and we just see it would, you know, encourage more 21 education and more adoption for those reasons. 22 MR. SWITZER: So I'm definitely the uneducated 23 ETF guy that Richie's talking about. So -- but let me -- 24 I'll take a crack at this. So it seems to me that we've 25 been trading ETFs for a really long time because we've 0103 1 been trading in on the derivative side, you know, CDX 2 with the underlying CDS's that are involved in the index 3 arm and so on and so forth. So, to me, it seems like 4 that's very similar. The difference is you're talking 5 many, many more underlying securities, which again, not 6 to beat a dead horse, but leads back to the need for 7 greater pretrade transparency. 8 And just to be clear, I don't mean pretrade 9 transparency the way MiFID has tried to come up with 10 pretrade -- I just mean good old-fashioned know what's 11 going on out there and price discovery. That being said, 12 with the ETF market, I think one interesting thing that 13 has happened, and this certainly happened at Alliance 14 Bernstein, is looking at the ETF market and looking at 15 that ecosystem that is developing and trying to identify, 16 okay, what can we take from that. And what we've 17 actually been doing a lot of is what we refer to as, you 18 know -- you know, diversified portfolio, you know, risk 19 transfers or portfolio trades. These guys would probably 20 just call it a create redeem. 21 But there's actually a market developing out 22 there where this is -- this is not flowing into an ETF. 23 It's actually, there -- you know, dealers are actually 24 crossing large blocks of risk, you know, just blocks of 25 beta that we're trying to shed and somebody's trying to 0104 1 source in a -- in a portfolio trade, so it might hit the 2 tape and you see a whole bunch of one million or two 3 million or three million bond trades, but in reality, 4 300, 500, 800 million just crossed in one trade. And 5 that's what it's forcing us to do. It's forcing us to 6 think about different ways. Because in times of 7 dislocation and we want to be a liquidity provider in the 8 market, that's a way we can do it. 9 What are the positives? Think about it. If we 10 have an inflow or we have an outflow and we -- and we go 11 to sell individual securities, by definition, we're 12 risking that portfolio, but if we can sell or buy a risk, 13 a slice of that risk in the portfolio, that's more 14 optimal for us. So it's really this just behavior, you 15 know, Drew talked about it, about things changing every 16 other week. This is happening at such an epic speed 17 right now, more and more dealers are embracing it. 18 They're setting up separate desks where they can hedge 19 it. It's quite interesting, so just throw that out 20 there. 21 MR. REDFEARN: So let's pick up on a similar 22 question about mutual funds and mutual fund holdings and 23 just asking the question if there's any risk in investors 24 in this trend that we've seen in the marketplace, in 25 particular, in times of market stress? 0105 1 Paul? 2 MR. JAKUBOWSKI: Yeah, I would think, you know, 3 you look at the mutual fund universe and it's still a 4 very small portion of the overall universe, even in 5 corporate bonds and, you know, Sonali had some graphs 6 that showed that. So that's one point, you know, 7 unleveraged diversified assets where a Vanguard or a 8 Blackrock is acting as an agent on behalf of client 9 assets poses little risk to the market. There are 10 clients in there that have a variety of different time 11 horizons, risk tolerances, objectives with that money, 12 and, you know, there's a long history of the market that 13 just shows in time of stress that they don't hurt; that 14 there aren't massive outflows. You know, we look at 15 different time periods, whether it's the GFC or the taper 16 tantrum, and we've seen, you know, less than five percent 17 of redemptions across mutual funds. 18 So we have some very concrete data points that 19 just show they don't pose a systematic risk, and again, 20 back to my point on, you know, balance and target date 21 funds, they may actually be counter cyclical, and they 22 may actually provide some liquidity to the market when 23 things are in times of stress, just through rebalancing 24 of those portfolios. 25 MR. REDFEARN: Anyone else want to comment on 0106 1 that? 2 MR. PRAGER: We would agree with those 3 comments. I mean I think it's really important and 4 actually, also, the work that the Commission has done in 5 terms of the liquidity risk management, it has been -- 6 you know, there has been improvement in terms of, you 7 know, TRN portfolios and, again, change of behavior that 8 we've all been talking about. Portfolio managers are 9 much more liquidity aware, but, you know, Paul's points 10 there are all well-taken in terms of just look at the 11 data over the last period of time when there's been the 12 most concern. You all have them in your IRA. I don't 13 know if you trade them. I mean it's the type of thing 14 where it's just you don't see those type of redemptions. 15 And, again, the point you made is very 16 important. In an unlevered way, I mean the risk is with 17 that investor who is -- that's appropriate. It's not 18 with the bank and the bank balance sheet in putting 19 government money at risk. It is with the -- with the 20 asset owner, and that's the appropriate place to hold 21 that risk. 22 MR. SWITZER: I mean the only thing I would add 23 is, you know, in these funds, you know, over the years, 24 the way we manage them is different, and we manage them 25 with more diversified liquidity or certainty of 0107 1 liquidity. And certainly, you know, you could look at a 2 time like today where, you know, spreads are tight; 3 yields are low, dispersion has been squeezed out of the 4 marketplace, and we tend to put a little bit -- a little 5 more certainty of liquidity in there. That could be 6 derivatives, it could be ETFs. It could just be cash and 7 government bonds. So, you know, we manage -- we manage 8 for that. 9 And when you really look at our performance 10 versus our peers, that liquidity drag doesn't show 11 through, so I think it's pretty constant across the -- 12 across the market that people are managing portfolios 13 differently. 14 MR. REDFEARN: Great. Thanks. So I want to 15 move on now to the topic of transparency and talking a 16 little bit about pretrade and post-trade transparency. 17 Everybody's aware of the longstanding pretrade, post- 18 trade transparency regime for corporate and municipal 19 bond markets with FINRA's TRACE and MSRB's EMMA. 20 Certainly it was mentioned earlier, the sort of MiFID II 21 regime coming in and different protocols for pre and 22 post-trade transparency, in particular, pretrade changing 23 pretty significantly. That's early, so it's, I think, 24 hard to see what the overall impacts of that will be, but 25 ultimately, this is a very important issue as we think 0108 1 about how this marketplace evolves with respect to 2 electronic trading and so on. 3 So I guess the first question, and I don't 4 know, I guess we'll start down with you, Paul, is, you 5 know, should the US post-trade transparency model, this 6 time post-trade, be reviewed to better reflect the needs 7 of market participants in current market conditions? And 8 specifically, we're wondering if there are any particular 9 aspects of the post-trade transparency model that are 10 right for consideration right now. 11 MR. JAKUBOWSKI: Yeah, I think any model that 12 gets introduced always has to be reviewed to make sure 13 it's accomplishing what it needs and should it be scaled 14 back or should it go even further. I think we should 15 argue, you know, at Vanguard that we're for more 16 transparency. And we should look to find other sectors 17 that can be included in the TRACE regime to give both, 18 you know, institutional and retail clients a necessary 19 look into the market that they need to make, one, more 20 accurate decisions, but also, one, faster decisions. 21 And, again, I think we've talked a lot about 22 using technology and data to make accurate decisions, 23 but, you know, that requires actually having levels and 24 knowing what trades happened. So in short, you know, I 25 think we should review it; it should always be looked at, 0109 1 but we would argue that we're for more transparency 2 versus less. 3 MR. REDFEARN: So Drew, this is obviously a 4 slightly different question from the perspective of a 5 dealer where with such, you know, I mean we've already 6 talked about risk in inventory and that and certainly 7 these transparency models do have some impact in the cost 8 there, and so what are your thoughts about this? 9 MR. MOGAVERO: I think more transparency helps 10 everyone. I think that -- and there's when we, you know, 11 do that electronic trading example, when I -- when I have 12 a good feel of where that bond is, it helps me price it; 13 it helps me price the portfolios that Jim was talking 14 about. I do feel that dealers competing over time -- 15 over time, more and more people are going to know where 16 that top half of the liquidity stack is, and there's not 17 going to be a ton of edge in that. And the edge will 18 come in at that -- at that bottom tranche, that 80 to 100 19 percent tranche where because you know those companies, 20 you know where the bones are buried; you just -- you know 21 what's -- because not all bonds are created equal, that's 22 really the bottom line. 23 And so I do feel that when you think about 24 post-trade transparency or pretrade transparency, while 25 you want that transparency, the goal should be to get 0110 1 more information to the market. My thing that I think 2 should be considered or at least talked about is the 3 timeliness of when the information gets to the market 4 because of the fact that not all bonds are created equal. 5 And so while I want to encourage more trading and more 6 liquidity for Barclay's, for our clients, for everybody, 7 and so we have to ask ourselves how do we encourage more 8 trading in that 80 to 100 percent of the market? 9 Sometimes, it may not just be there to trade 10 because the bonds may be just very well held or whatnot, 11 but I do feel, at times, the treating that 80 to 100 12 percent bond in terms of timeliness of transparency the 13 same way we treat, you know, the graph that Sonali had 14 where there's a 1.5 percent of the bonds that trade every 15 single day, I feel like that may be something, you know, 16 to Paul's point as all policies should be reviewed, and 17 my opinion is that if there is maybe some form of delay 18 on those less liquid bonds, such that it allows the 19 marketplace to find that level and find where a bunch of 20 bonds can trade, as opposed to just when a print happens 21 in a less liquid bond, right away. Sometimes, it's 22 actually used to the disadvantage of the client, the 23 dealer and that real-time transparency the same way as, 24 you know, the JP Morgan on the run 30 years may make 25 sense to think about in a different way. I just think 0111 1 that's something to consider. 2 MALE SPEAKER: Yeah, absolutely. 3 MR. PRAGER: I mean I'll pick up on that theme, 4 slight twist to it. So let's start backwards. As you 5 said, we'll start post and then we'll go pre. And so no 6 one's -- I don't think anyone's going to argue against 7 transparency. So transparency is obviously good. But, 8 you know, as Paul said, every now and then, it's time to 9 revisit something and see if there's some process 10 improvement. So, you know, we would like to see some 11 sort of pilot that addressed, you know, better 12 calibration on -- and you say liquidity, that's a hard 13 thing to measure, as we were discussing with the 14 research, but I would prefer to look at it as maybe some 15 calibration around trade size and that we do consider 16 some. You know, where's the right calibration of 17 reporting trade size, which will actually encourage more 18 transaction? 19 So you know, I think we should look for that 20 calibration. We have sort of one in five now. Maybe if 21 everything was, you know -- this is not based on -- you 22 know, on data, again, this would be -- this would be the 23 pilot. We should look to find what is that number and 24 what is the appropriate delay. Fifteen minutes is 25 actually, I think in some ways, constraining, as I 0112 1 mentioned early. We sell for a lot in the market whether 2 it is efficiency or it is liquidity, certainly around the 3 smaller trade size. I think the market still has some 4 challenges with blocks, and we should -- the Commission 5 consider some sort of pilot to look at the right 6 calibration and the right delay. 7 So I agree with Drew, and I don't think we're 8 going to compromise the concept of transparency at all. 9 If there was some end-of-day reporting for, you know, 10 whatever, wherever we decided to draw that line. So I 11 think there -- you know, that your committee should 12 really give that some thought. 13 In terms of the pretrade, you know, I do agree 14 with Jim that, you know, it would be wonderful to see a 15 lot of pretrade. It is challenging, and I also agree on 16 some of the MiFID style stuff. We need to -- let's just 17 see what happens. It's way too early. But, you know, I 18 do think any pretrade should be anchored on transactable 19 pricing because if we just invite -- you have the, you 20 know, your three million of messages coming in, we have a 21 similar number sort of eight digits' worth of information 22 coming in every day. You know, it's subject to abuse if 23 it's not transactable pricing. So it needs to be 24 anchored in real transactable pricing, otherwise, I'm not 25 sure it's usable or healthy information when you think 0113 1 about it. 2 MR. SWITZER: I mean the only thing I would say 3 about that is, you know, the equity markets, NASDAQ 4 started out as just a streaming quote, right? It wasn't 5 an executable. We have to start somewhere. So to me, to 6 be able to see depth of thought around securities would 7 certainly help in terms of the post-trade transparency, 8 look, you know, we're not asking for a delay in 9 reporting. We'd be -- you know, it would be a delay in 10 dissemination to the market, I guess, right? So that 11 seems reasonable to me. The only thing is as pretrade 12 gets better, and as everybody has their algos that are 13 running, that are pricing every bond in the market and 14 ICE and MarketAxess and Trumid are out there streaming 15 their CEPs. You know, I think as confidence creeps into 16 the market, the market understands that there should be 17 off trades for size and off trades for efficiency. 18 I mean certainly the equity markets understand 19 that, so I think that ultimately, it will have less of an 20 impact, but look, the fact of the matter is in the bottom 21 30 percent of the market, the trades once a month or once 22 every six months, you get a print. It reprices the 23 markets, so certainly, that's a problem. The pretrade, 24 the only thing I wanted to say is, you know, when you 25 think about like why don't bonds trade and we talked 0114 1 about part of it is if you think about best ex policies 2 and a buy-side, you know, a buy-side trader, and this is 3 part of the behavioral modification and the way the firm 4 is looking at best ex, but a buy-side trader, very often, 5 is just used to be shepherded through the process by the 6 sell-side. 7 So the concept of being a price-maker is a 8 difficult thing, especially in a market where you might 9 not be confident or confident enough because -- in what 10 something should trade and that -- and that's why I think 11 pretrade transparency will actually -- will actually 12 help. 13 I wrote one other thing down here. And really, 14 it's just really just the lack of -- the lack of 15 confident around the market. Now, not to harp on this 16 ETF and the whole ecosystem thing, but to be able to 17 price these underlying blocks of risk is really going to 18 require if you -- if you imagine trying to trade $500 or 19 $1 billion, which is very, very reasonable to -- you 20 know, expectation in times of volatility, you have to be 21 able to do that fast, and that's where being able to take 22 these feeds -- so at the very least, what we're doing is 23 we're taking all of these CEP feeds or composite pricing, 24 and we're feeding it into aggregate, our aggregator, so 25 we're at least looking at it pretrade, and I just think 0115 1 that's something that maybe the Commission is supposed to 2 think about, you know, something more broadly. 3 CHAIRMAN CLAYTON: So I -- thank you, a question 4 that I think I should ask that has occurred to me 5 listening to this and I want to -- I want to thank you 6 all very much, I think the transparency discussion is 7 great because transparency affects liquidity, right? And 8 the right type of transparency, whether it's actionable 9 transparency or calibrated transparency affects 10 liquidity, but for our Main Street investors, we're spend 11 -- a lot of really talented people giving up their time, 12 I just -- why does liquidity matter? 13 Why do we focus on it and why does it matter? 14 I think I want -- I'm going to say somethings you guys 15 know better, but Main Street investors want liquidity. 16 They really do. They're willing to give their capital. 17 You provide daily liquidity to them, but they're willing 18 to provide their capital because they know if times 19 change, they can get their money back. It matters to 20 them, also, because the better liquidity is, probably the 21 easier they get in and out. The costs of being an 22 investor go down. And lastly, it is a sign of stability 23 and resiliency, it's not a guarantee, but it's a sign is 24 -- I mean that's why we focus on it. It's a focus of the 25 Commission, but if you guys could just elaborate as to 0116 1 why it's important for us to facilitate a ruleset that 2 provides liquidity, I would appreciate that. 3 MR. PRAGER: I mean I'll start. I mean you 4 might've missed my earlier comment when I -- when I was 5 sort of giving my context. So first of all, I don't know 6 that liquidity is a God-given right. It is a -- you 7 know, we see liquidity as a function of confidence, and 8 the more confident that people are in the market, you 9 will see liquidity. And I also don't think for the, you 10 know, myriad of CUSIPs out there, is there meant to be 11 perfect liquidity on everyone and nor should every retail 12 investor own a single CUSIP. 13 So I do think there's some appropriateness 14 there, and so, you know, that's why a lot of the retail 15 owns -- you know, whether they own mutual funds and they 16 own ETFs and, you know, that's where we do have, you 17 know, a lot of pretrade transparency in ETFs where there 18 is on exchange. They can see it. Same price for 19 institutional or retail, so I think that we just -- 20 again, the context of that question is really important 21 because if we're -- you know, if there has to be 22 compulsory liquidity on every single CUSIP issued, I 23 don't know that if that's a good thing for a retail 24 investor in terms of the execution cost to get in and -- 25 in and out of something could actually be quite 0117 1 expensive. So we have to put that context in place where 2 what is the appropriate, you know, vehicle for the -- for 3 the retail, what is that, you know -- what is the 4 appropriate liquidity conditions? 5 I think a lot of the markup rules have been 6 really good in that -- in that nature, and, you know, so 7 there's been great progress made, but I don't know -- you 8 know, again, what are we solving for? We can't -- if 9 we're actually saying it has to have perfect liquidity in 10 every single CUSIP, we probably wouldn't issue a lot of 11 CUSIPs, so I think we're just trying to get to the most 12 market confidence so that we can have, you know, adequate 13 liquidity through, you know, different market positions. 14 CHAIRMAN CLAYTON: Liquidity means different things 15 for different instruments. 16 MR. PRAGER: Yeah. 17 CHAIRMAN CLAYTON: And our retail investors should 18 understand that. 19 MR. PRAGER: Yeah. 20 CHAIRMAN CLAYTON: And our ruleset should reflect 21 that. 22 MR. MOGAVERO: Yeah, and I think it's important 23 that last point, the ruleset should reflect that, because 24 I think what I would caution against is because every 25 bond is not created equal and is so different that you 0118 1 would want to make sure that if some form of pretrade 2 transparency is -- my fear would be that if that's, you 3 know, imposed, for a lack of a better word, on all 4 securities in a one-size-fits-all type of thing, that 5 that's going to be misleading to the Main Street 6 investor, where they feel that -- 7 MR. PRAGER: And costly. 8 MR. MOGAVERO: Yes, exactly. And they think 9 that there's liquidity at a pretrade price, and there may 10 not be. And I think it's just really, really important 11 to recognize that and to understand that not all bonds 12 are created equal and to advocate for more appropriate 13 solutions, whether it's portfolio products, ETFs, liquid 14 bonds, like, you know, you can create a framework. These 15 are things that are all good for discussion. 16 MR. REDFEARN: So we are getting very close to 17 being out of time, but I wanted to just close and maybe 18 try to keep our answers limited to, you know, maybe one 19 minute or less, given the time here, but it has to do 20 with the policy response, so I think part of -- part of 21 the thrust behind Jay's question and the discussion is 22 what we're looking at and what are the policy things that 23 need to be done to help liquidity or help create a more 24 efficient market. 25 So if each of you could just opine briefly on, 0119 1 you know, if we had to focus on, you know, a regulatory 2 or a regulatory initiative, whether it has to do with 3 transparency or something else, what would be the -- you 4 know, the one thing that you would suggest that should be 5 on our radar right now? 6 MR. JAKUBOWSKI: I guess I'll start. You know, 7 I'll go back to my original comments on what trades and, 8 again, Kevin mentioned this, what trades aren't getting 9 done and, you know, the fragmentation in the market and 10 why aren't buyers and sellers connecting more frequently 11 and in bigger size and at the right level? And I think 12 that's the ultimate question to ask, especially as we 13 move into using more electronic trading technology and 14 data. You know, how do you really get at that crux 15 question right there? 16 MR. MOGAVERO: Sorry about that. I would add 17 from -- to Paul's exact point and, you know, we touched 18 on it earlier, to me, it's -- I think Richie picked the 19 perfect word, it's the calibration of the -- of the post- 20 trade transparency, such that you can -- you know, we 21 want more transparency, more bonds to trade and then that 22 leads to more liquidity. And I think the thought process 23 around making sure that while those bonds can get to the 24 levels or especially in block size, where buyers and 25 sellers can meet and can do it in a way that is -- 0120 1 creates more trading and is advantageous to allow the -- 2 those participants to find each other, whether it's 3 through a dealer, whether it's through all-to-all, 4 whatever it ends up being, you know, that we're able to 5 have a way for those folks to find each other and to 6 consider that those situations may be different than the 7 on the run JP Morgan 30-year bond. I think that's a 8 primary for consideration that can lead to more liquidity 9 and transparency in the marketplace. 10 MR. PRAGER: Yeah, I made the recommendation 11 before on the would recommend a pilot around that 12 calibration, so that's the short crisp answer, and I 13 think as a secondary one, to pick up, I think some of the 14 research panelists mentioned before, you know, the 15 treasury market paper talked about some of the easing of 16 bank regulation around some of repo and financing, I 17 think to the extent that there were some more financing 18 availability in the marketplace, you would see an 19 increase in liquidity around a lot of instruments. 20 MR. SWITZER: Surprise, pretrade transparency. 21 So to me, that solves a lot of problems. That solves 22 trapped or dead balance sheet bonds that we're afraid to 23 try to trade. It incents new capital into the market. 24 There's more confidence. It will -- it will aid the buy- 25 side to be more of a price-maker. I think it will -- it 0121 1 will increase the velocity of trading on the electronic 2 platforms. I think it'll increase the velocity of 3 trading styles or techniques, if you will, that are -- 4 that are starting to develop. And so that's my thinking. 5 MR. REDFEARN: Fantastic, we are -- we are done 6 with our time here. I want to thank the panelists for a 7 very engaging and informative discussion. Thank you very 8 much. And just a brief announcement, at this point in 9 time, the committee and the panelists, we're all going to 10 be heading directly to the 9th floor for our lunch, and 11 so we'll do that in just a second, and just so everybody 12 else knows that after the -- after the lunch break, we 13 will, indeed, be back here at, I guess, 1:45. 14 At that point, there will be a broader 15 discussion with all of our panelists, from both of the 16 panels with the broader committee, and then, of course, 17 after that, there will be a broader discussion solely 18 among the committee. 19 So thank you all very much, and we 20 will be back here at 1:45. Let's now all head up to the 21 9th floor. Thank you very much. 22 (Whereupon, at 12:18 p.m., a lunch recess 23 was taken.) 24 A F T E R N O O N S E S S I O N 25 CHAIRMAN HEANEY: Okay. I think we're ready to get 0122 1 started. I just wanted to thank, again, the panelists 2 for taking the time this morning. Two very insightful 3 sessions, I think kind of really set the table for the 4 discussion going forward this afternoon. As we said we 5 would start at this point opening it up to questions or 6 discussion or comments, but really more questions from 7 FIMSAC members to our panelists before we lose them in a 8 little over an hour. So with that, if I can just ask as 9 we get to know each other a little better, that you say 10 your name and your firm when you're asking a question and 11 perhaps if it gets where there's several people jumping 12 in, perhaps just raising the nametag so we can just keep 13 it a little bit orderly as well. 14 Elise? 15 MS. WALTER: In the interest of moving things 16 along, I will agree to go first. I actually have two 17 things I wanted to ask about, and why don't I lay them 18 both out and give you people choices as to which you want 19 to address, two rather different. First, from the point- 20 of-view of transparency, I think looking at this from the 21 small investor, from the retail investor perspective, 22 people who used to be called my Aunt Millie but are now 23 called Mr. and Mrs. 401(k). Although, I would say Mr. 24 and Ms. or Mr. and Mr. 401(k). I think one thing that 25 didn't come across this morning, which I think is an 0123 1 important factor to take into account is the value of 2 transparency either as a direct link to greater investor 3 confidence in the marketplace, which I think is the 4 foundation for everything, or the value itself to people 5 of having the sense that they can find out whatever they 6 need to find out, and I harken back to the field hearings 7 we did before we did the MUNI report, which came out in 8 2012, where basically, small investors, and this is 9 particularly pointed in the MUNI market where there are 10 more people who hold directly, said, "If I want to sell 11 my bond, I have no idea what it's worth. I just 12 basically have to trust, and I have no way to verify." 13 So I wanted your reaction to what is the value 14 of transparency? In a sense, it's a value in and of 15 itself, rather than as a road to other values. And the 16 second thing I wanted to ask about is with respect to 17 intermediation. Should we -- is it important, regardless 18 of whether liquidity is decreased or it's bad that it's 19 decreased or people have adjusted and it's fine, do we 20 want to -- do you favor taking affirmative steps to 21 increase intermediation? Re, perhaps developing a more 22 definitive concept of a market-maker with some 23 incentives, "payments," not necessarily payments, but 24 some incentives for people to play that role and also 25 having them have marketplace responsibilities. And I'm 0124 1 Elise Walter, unaffiliated for this purpose. 2 CHAIRMAN HEANEY: Who'd like to take the first shot 3 at -- stab at that? 4 MR. SWITZER: I'll take a crack at that. Bond 5 trading is hard, and it's hard to figure out values for 6 lots of bonds. And so when you think about a retail 7 investor and not understanding what something's worth, 8 sometimes we have to do a lot of work to figure out what 9 something's worth ourselves. So it's difficult to figure 10 out how to pass that along. We have very robust best 11 execution policies and a process, let's call it more of a 12 process than even policies, that are assuring that our 13 clients are getting the best price possible. And 14 remember, the best outcome for a client isn't always best 15 price. There are other outcomes that could be better. 16 And so that's kind of how -- that's kind of how I think 17 we look at it. 18 MR. MOGAVERO: I would add from our 19 perspective, I think the -- again, getting back to the 20 all bonds not being created equal part of it, it's the -- 21 I think the value of transparency, in and amongst itself, 22 is clearly important. It's clearly helpful. It's 23 clearly helpful for, you know, if and when you can -- you 24 can talk about will the market evolve to something where 25 liquidity is getting pushed out to whether it's 0125 1 alternative liquidity providers, whether it's certainly 2 the RFQ phenomenon is there, and you have a different -- 3 the market structure evolves to look at alternative ways 4 of trading. There'll be certain bonds that I imagine 5 will gravitate well towards that and then certain bonds 6 that will not. 7 And I think the value of transparency is going 8 to be a lot of a function of that, and maybe even in 9 reverse, whereas for some of the more liquid bonds, the 10 value -- you know, Jim probably has as much pretrade 11 transparency as you could possibly ever want for some of 12 those things and maybe, arguably, too much. 13 And the retail investor has a lot of post-trade 14 transparency through TRACE because there's a lot of 15 prints, large and small, allotted various different 16 times. So your -- that transparency is well-defined and 17 maybe it's good, but the value of it is maybe less than 18 the value on those harder bonds, and I think that's where 19 you have to start to think about, you know, I think the 20 ETFs have done a good job of helping because it's maybe 21 getting some of those trade more often, electronic 22 trading gets them to trade more often. That's really the 23 transparency I think Jim put it well, you know, it's like 24 you've got to kind of -- you've got to start somewhere, 25 right? You've got to start with some of the information; 0126 1 start by having some levels of pretrade transparency. 2 Maybe we can talk about recalibrating the post-trade 3 transparency and then building off of that and building 4 off of -- you know, because that's where, to me, that -- 5 the transparency value is enhanced on the bonds that are 6 harder. 7 MS. ROSENBLUM: And I might just add -- hi, I'm 8 Alexis Rosenblum, I'm sitting in for Richie Prager, he 9 apologizes for not being able to stay for this panel. 10 You know, I think one thing just to add there is that I 11 think transparency is helpful when it promotes fair and 12 efficient markets and when it reduces costs for 13 investors. I think one of the challenges in the fixed 14 income market is that it's a different market structure 15 than the equity markets, and so some of the concepts of 16 transparency don't necessarily translate as well in the 17 fixed income markets and in some cases, could increase 18 costs for investors. 19 So I think it's really a question about 20 properly calibrating the transparency, and that's what 21 Richie tried to raise in the earlier panel. 22 MS. THEISEN: I would -- sorry, I would add I 23 think addressing your second question, if I understood it 24 correctly, and I want to ensure that I did, if the 25 question was around sort of designated market-makers in 0127 1 sort of a -- 2 MS. WALTER: That would be one wish to it, I'm 3 basically asking do you think that there's a need to 4 encourage further intermediation than there is today? 5 MS. THEISEN: I see. So I would just kind of 6 highlight in the context of designated market-makers in 7 kind of offering, you know, firm orders, I would -- I 8 would mentioned that I think that those initiatives, 9 those incentives have been out there. I don't 10 necessarily know that regulation is required or -- so I 11 think that, you know, the market has organically tried 12 that. Our view is that, you know, it has not moved en 13 masse towards that type of a model because of the -- 14 again, the inherent differences between, you know, credit 15 and equities for an -- as an example of just the market 16 structure being one. In many ways, it's almost 17 diametrically opposed in terms of number of instruments 18 traded to number of participants, if you think about, you 19 know, the credit markets, there's a large number of 20 instruments to a relatively small number of participants. 21 And it's the opposite triangle in equities. 22 And I think, again, that leads to a market structure. 23 Again, we talked about it a bit in our -- in our 24 presentation on Slides 3 and 4, it leads to a market 25 structure where, you know, knowing your counterparty is 0128 1 often, you know, a lot more important. There's not 2 necessarily as much order-driven business, and I think, 3 you know, the kind of DMM structure of providing 4 incentives, you know, has worked historically in more of 5 an exchange market, which I just don't know that our 6 market structure is there today. 7 MR. MELI: I'd add to that, I think implicit in 8 your question is a sense of is there -- is any decline in 9 liquidity that can be measured or documented sufficient 10 to cause problems in the market such that you need to 11 sort of -- you need some way to increase intermediation, 12 I think. And I think maybe that equity channels, like 13 designated market maker is one of the ways that you could 14 do that. I mean I think that what -- Richie made some 15 comments in the second panel about, you know, trying to 16 benchmark what an appropriate level of liquidity is. You 17 know, I think that in a -- in an ideal world, you know, 18 more is better than less, but the question about whether 19 there's too little or to allow the market to function 20 smoothly, I think, is the key question. And, you know, I 21 think that if investors and dealers were all behaving the 22 way they had been behaving in, you know -- in earlier 23 times, precrisis era or earlier, then you -- I think 24 given the growth of the market, you might -- you might be 25 able to raise some questions about the ability for it to 0129 1 smoothly function. 2 But I think when you -- when you combine, you 3 know -- you know, you have to combine both the fact that 4 maybe single-name liquidity has changed, along with the 5 fact that the market has responded to that. And then I 6 think, you know, it's a harder question then to say that 7 or a harder position to take, I think, to say that the 8 market is not functioning smoothly, despite some of the 9 changes in liquidity. 10 MS. WILSON: Hi, Rachel Wilson, Iron Mountain. 11 Just to kind of follow up -- thank you, just to follow 12 up on that question, we're using liquidity, I mean that's 13 probably the word du jour today so far as the most 14 commonly used word. And yet, I hear it in very different 15 context. We talk about it as where buyers meet sellers, 16 right? That's how you were using it. And so we're 17 having more transactions happen. We've used it to say 18 ease of execution. We've used it to say bond frequency 19 of trading of specific bonds of frequency of trading. 20 We've used it as a market participant access to 21 funding availability, and we've used it to enable, as you 22 just talked about, the market to smoothly function. And, 23 you know, these are -- these are pretty different 24 concepts. We're using the same word, but these are 25 pretty different concepts. So as you think about the 0130 1 prioritization or where this is an obstacle, okay, or 2 creating an issue, I think that would be -- I'd like to 3 refine your thinking on what you really mean by liquidity 4 and where it is a problem. Thanks. 5 MR. JAKUBOWSKI: I guess I'll start. And I 6 think everyone would agree with this, it's you really 7 need to connect the buyers and the sellers, right, and 8 how you do that may look different, but, you know, I 9 think anyone, you know, on the buy-side or the sell-side 10 would love to find people who want to transact the same 11 bonds. And I think they exist out there, but the methods 12 of connecting them are disjointed, are not clean. You 13 know, even in, you know, Drew's business, like he 14 probably owns bonds or is in touch with bonds. He would 15 love to reach out to everyone and just survey who wants 16 to buy those. But there's no streamlined method for 17 doing that. There's no way of capturing a variety of 18 different investors and just saying, "Here is the pool of 19 liquidity, and hey, I can easily make those connections." 20 And I think that's the top priority. And 21 everything else that you listed either helps to 22 facilitate that or could make it easier for that stuff to 23 happen. 24 MS. THEISEN: I would weigh in, while I agree 25 with those comments, and I do think that the search 0131 1 function in the marketplace of connecting buyers and 2 sellers has drastically improved since the crisis, so I 3 think that that technology, while it's still being, you 4 know, developed, I think we've gone through a good period 5 of development already. When we think about liquidity, 6 and I think it was underscored in some of our comments 7 earlier of kind of where the problem is, we really do 8 focus on the markets becoming more homogenous from the 9 varieties of factors that we discussed earlier today. 10 We think that creating more diversity of 11 strategies in the marketplace is one, again, I keep 12 focusing on that long tail of illiquids and, in some 13 sense, the market has the danger, and I think we have 14 seen, you know, some evidence, it's already becoming more 15 barbelled, the liquid -- you know, the liquid small 16 sizes, the bonds that go into the ETFs become more and 17 more liquid, but you still have the call it, you know, 18 25,000 other CUSIPs that as they become seasoned, if 19 they're from smaller issue and size, et cetera, you know, 20 the holder concentration becomes, you know, a lot higher, 21 and when there is someone that needs to buy or sell those 22 bonds, there's not necessarily -- I don't necessarily 23 think the problem is always finding the other side, it's 24 that A -- you know, going back to the points that we made 25 earlier, A, there's not as much conviction. There might 0132 1 be someone who thinks those bonds look rich or cheap, but 2 because they can't achieve the same level of return when 3 there was, you know, more leverage in the system, they 4 may not trade them, and B, you know, the strategies have 5 become, you know, more homogenized. So I think those are 6 the points. I do agree that, you know, connectivity is 7 something that we should continue to focus on, but I also 8 feel that we've made a good bit of progress there. 9 MR. MOGAVERO: And maybe I'm being too 10 idealistic, but I feel like it's some combination of D, a 11 lot of the above. Like if I picture A is, you know, a 12 state of increased liquidity, I picture things of where 13 you've got, you know, small trades that are occurring 14 that are defined from flows, whether it's from ETFs or 15 smaller retail flows. I picture that leading to levels 16 in transparency where you -- where you've got a decent 17 idea if stuff is going to trade. 18 Then there's going to be larger trades that 19 occur around that context, give or take, but when you 20 think about what creates that liquidity for the larger 21 trades, it's a combination of some element of risk- 22 taking, where a dealer or a hedge fund or a client says, 23 "Okay, these bonds look cheap, I'm going to provide a bid 24 to the marketplace." And then -- or an idea of, you 25 know, certain RFQs that traded and set a price, and I can 0133 1 see a -- to me, it's like a combination of however we can 2 connect, to Paul's point, the dots between all of those 3 various different forms of, you know, to Sonali's point, 4 of the various different actors and participants in the 5 marketplace and figure out how that can be best 6 disseminated to the marketplace, either before a trade, 7 during a trade, after a trade, like that -- all that 8 combination together, to me, seems is where you could 9 potentially see future state liquidity evolve. It sounds 10 kind of Utopian, but I think that's -- 11 MR. MELI: I think a follow-on to that is that 12 like I think you could think about this as -- think about 13 it in two regimes. If you think about it as a normal day 14 and then a stress day, and in a normal day, I think I 100 15 percent agree with Drew that it's -- that the different 16 kind of descriptors that we use that you cited are all 17 sort of saying -- they're all speaking around, I think, 18 the same general concept. And I actually would agree 19 with Sonali that I think there's been a lot of progress 20 made on building systems and behaviors that will allow 21 participants to find the other side on normal times and 22 be able to close that gap. And I think there is more to 23 be done there, for sure. 24 But the amount of progress is encouraging and 25 some of the new innovations, like some of these portfolio 0134 1 trades, are great -- you know, are great new forms to be 2 able to do that. I think there's another set of concerns 3 that sits around, you know, how does the market respond 4 to periods of stress, and I think Paul said, I think in 5 the second panel that -- you know, that over -- being 6 over reliant on the dealer community in times of stress 7 was probably never a great strategy for unwinding your 8 positions in the first place. And I would agree with 9 that. I think that there are open questions about how a 10 less liquid market might respond and how quickly it might 11 reprice and to what levels that -- you know, even that 12 being said, those open questions still exist, and I think 13 those, the sort of connection systems, I think, are less 14 likely to solve, given, I think, the homogenization that 15 Sonali is talking about. 16 MR. MCPARTLAND: I mean I think there needs to 17 be an appreciation how diverse this market is. I mean 18 even really almost not barely 10 years ago, there was one 19 way to do corporate bond trading, you called a few 20 people, right? And now, here we are 10 years later, and 21 there's a couple of ways, but there probably should be a 22 whole -- a whole bunch more based on who you are, the 23 size of the trade, the type of the bond, how much of it 24 you need to trade at any given time, should it be an RFQ; 25 should it be an auction? Should it be with the buy-side? 0135 1 Should it be with -- should it be anonymous? Should it 2 be name disclosed? It's just the list goes on and on and 3 on. And, you know, equities, to some extent, figured 4 that out. Although, equity has since, some ways, looked 5 simple. 6 We figured out that there should be dark pools 7 to do large trades. Okay, now there's a whole variety of 8 those, and there's a whole variety of exchanges. I'm 9 preaching to the choir on that. So we don't need to go 10 into that, but I think there needs to be an appreciation 11 that electronic trading or execution efficiency in the 12 bond market isn't one thing, it's many, many, many, many 13 things. And the solution and liquidity is definitely in 14 the eye of the beholder in this case. 15 MR. SWITZER: And one other thing I would say 16 is liquidity, the market or liquidity in general is very 17 bifurcated. There's the -- there's the on the wire 18 liquidity or that immediate liquidity you can get, and 19 then there's the liquidity we talked a lot about moving 20 to the riskless principal model, where you're working 21 orders, and really, it depends on what kind of market 22 we're in, whether we're in a volatile market, then you're 23 going to see less immediate liquidity and more kind of 24 you've got to work it off, and in times like this, it's - 25 - there's a little bit more immediate liquidity, so it's 0136 1 going to fluctuate back and forth. So it's very hard to 2 define, you know, what this is. 3 MS. ROSENBLUM: Yeah, and just to echo that 4 point, you know, from our perspective, we manage money 5 for thousands of different clients, and each of those 6 clients have different investment objectives, so 7 different aspects of what does liquidity mean or what is 8 important to me about liquidity can be very different, 9 depending on the investor, their objectives, their time 10 horizon, so I think it's hard to say that there's one 11 problem because the market has many, many diverse 12 participants, and I think you have to look at it from the 13 investor's perspective, from their individual situation. 14 CHAIRMAN HEANEY: I think Larry had a question, and 15 if I could just again for the benefit of the public who's 16 here, just the name and where you're from. 17 MR. HARRIS: I'm Larry Harris. I'm from the 18 USC Marshall School of Business. My question's about 19 post-trade transparency. Richie proposed that perhaps we 20 need some pilot on reporting trade size, and in 21 particular, that while we presently limit the reports to 22 one million and five million, depending on the type of 23 bond in 15 minutes, he suggested perhaps that 15 minutes 24 is too short and that blocks are challenging. And both 25 Jim and Drew echoed that concern, suggesting that perhaps 0137 1 we should visit the question of how quickly should we 2 report the trades. So here's the question, if we require 3 that the trades be reported quickly, then the dealers are 4 not able to run with the position that they've acquired, 5 and they would claim that -- presumably that this ties 6 them up and makes it more difficult for them to trade. 7 But the other perspective on that is if we 8 require the positions to be reported, the dealers run 9 before they do the first trade, and so the question is 10 should the Commission, or any other entity, providing 11 this slack, shouldn't the -- if the deal has to be done 12 between natural buyer to natural seller, it can be 13 arranged beforehand, before you actually print the trade, 14 if it's really basically a broker trade, it's a broker 15 trade. 16 So the question is this, as we consider the two 17 alternatives, one is to maybe maintain the status quo or 18 even tighten it up to make it look more equity-like, 19 versus the proposal that Richie offered, and was seconded 20 by two, to relax these standards. The question is what 21 do we gain if we -- if we relax the standards? What is 22 the benefit of allowing the trade to -- with the first 23 party to take place and give the extra time to find the 24 second party instead of essentially requiring that both 25 parties be found upfront? 0138 1 And as you answer the question, I would just 2 like to make one final observation. In the equity 3 markets where we're dealing with far more credit risk, 4 and therefore, far more overall risk, small stocks that 5 look a lot like bonds have significant block trades 6 reported on a regular basis, and they have to report it 7 as they occur, so given that, what's the benefit that 8 we're looking for here? 9 CHAIRMAN HEANEY: Jim, why don't you tackle it? 10 MS. THEISEN: I'll start it. I'd love to -- 11 CHAIRMAN HEANEY: Go on, Sonali. We're going to 12 definitely get Jim to answer it as well. 13 MS. THEISEN: Sure. One of the points -- so I 14 don't necessarily view the two frameworks as one 15 tightening or loosening or put another way, I think, 16 again, there could be increased transparency in certain 17 sizes. Maybe that's those sizes go up, those thresholds 18 go up from five million and one million to something 19 higher, and I think that was maybe something that Richie 20 was alluding to earlier, but then having accommodations 21 for larger blocks and another point is that also just 22 uncapped transparency today happens after six months. 23 That could potentially be moved up. I think there's 24 value in that data that could be moved up. 25 Like if you look at MiFID, for example, there's 0139 1 full uncapped transparency after a month, right? So we 2 think that there's ways to modify both increasing 3 transparency and making accommodations for blocks. But 4 to answer your -- what I think is the second part of your 5 question around what do we gain by making accommodations 6 for blocks or less liquid trades, I think it goes back to 7 a point that we made earlier around the role of a dealer 8 and risk transformation. It's not just finding, you 9 know, okay, I'm buying this 20 million bonds from you, 10 and I'm going to find four, you know, other people who 11 want to buy those back from me. You know, often times, we 12 are hedging our risk with other instruments, and we need 13 to go out and engage in those marketplaces as well, and 14 that is an important component of how these -- how these 15 principal markets function. 16 So it's -- I think having a accommodation for 17 risk transfer, and that risk transfer does not always 18 have to be a one-for-one, is an important flexibility for 19 us to discuss. 20 MR. HARRIS: If I may very quickly, those 21 hedges would be done in interest rate markets and in the 22 credit markets, the ultimate credit markets, which are 23 equities, both which trade very liquid and very quickly, 24 so I'm not quite understanding -- 25 MS. THEISEN: There's single name CDS, there's 0140 1 loans. I mean there's other -- you know, other products 2 with which there may be -- like, in other words, though, 3 it's not a one-for-one of the same marketplace with which 4 we would be hedging, and I think you mentioned like why 5 wouldn't you just ensure that buyer A, you know, finds 6 seller B and then also, you know, B, C, D. My only point 7 there was that, you know, it may be in a different 8 marketplace that that risk is hedged immediately and the 9 participants in the first marketplace might not be 10 accessing all of those markets at the same time. 11 MR. SWITZER: I mean look there's -- when you 12 think of those small stocks, I think you're referring to, 13 there's still at least pretty good transparency in those 14 markets before you do a trade. When you get into the 15 bottom echelon of the -- of the credit space, you can 16 have bonds that don't trade for a week, two weeks, a 17 month, three months. So when you're asking a dealer to 18 put -- you know, put capital forward and take risks, 19 there's probably a price for that media and a price for 20 that size. And all you're doing is -- I'm not -- I think 21 a trade should be reported immediately or 15 minutes, 22 it's just when you disseminate that information to the 23 marketplace so that the regulators can see that 24 everybody's playing above board and doing the right 25 thing, but, you know, we get predatory behaviors in these 0141 1 markets when you get into the illiquid space and somebody 2 takes a block of bonds and it's used against them by, you 3 know, other people in the market. So I think all we're 4 trying to do is create some kind of incentive for the 5 dealers to commit capital when they think they have a 6 fundamental view or a quantitative view, they think they 7 can line up the buyer or the seller that's necessary on 8 the other side. 9 And I need the liquidity today because I'm 10 providing daily liquidity to my clients, so we're just 11 trying to create some kind of easing, so to speak, just 12 to allow that. 13 CHAIRMAN HEANEY: Mihir, please. 14 MR. WORAH: Mihir Worah, PIMCO. So just to -- 15 just to add on this post-trade transparency, I think -- I 16 think I agree with a lot of the statements made by the 17 panelists that yes, you should report right away to the 18 regulators, but talking about market dissemination, what 19 happens is you really -- you, you know, one of the other 20 unintended consequences you're having with quick 21 decimation to the market is it compressing, you know, the 22 -- you're making investors less diversified. So there's 23 a whole ecosystem of retail investors who want to invest, 24 some just care about lowest transaction costs and that's 25 passive investing, and some care about thinking that the 0142 1 markets are not fully efficient, and you can try and beat 2 the markets, that's active investing. And it hurts the 3 individual investor who's opting, for example, for active 4 investment. You spend a lot of time, you spend months 5 doing research on a company, and maybe you need a couple 6 of days because of the size of the trade to do the trade, 7 I mean months of research are gone because the minute you 8 do the trade, the whole world of guys who don't do 9 research, don't spend the money and the resources trying 10 to beat the market know if you trade, and they can 11 piggyback off it. 12 So absolutely, there should be transparency to 13 regulators, but some decent gap between market 14 dissemination, given the size of the block, I think, 15 should be considered because it compresses everyone 16 towards passive investing. And there's a role for 17 passive. There's a role for active, those kind of plays. 18 CHAIRMAN HEANEY: Tom, please. 19 MR. THEES: Yeah, I just want to -- Tom Thees, 20 Castle Oak. I just want to make the point that everyone 21 should acknowledge that in today's marketplace, the 22 structure is such that there is actually a disincentive 23 to provide that liquidity for the subset of bonds that 24 we're talking about. I agree that the word incentive is 25 the proper one, but today's behavior is disincented from 0143 1 providing that liquidity to that investor that needs that 2 retail or institutional on-demand because the risk factor 3 of that dissemination is too high, and then, therefore, 4 that falls into the bucket of trades Paul was talking 5 about before, it never gets done. The cost of 6 transaction and the willingness of a dealer to step in 7 and provide that, there is too much disincentive to do 8 that behavior today. 9 Whereas with a slight adjustment, with a 10 calibration, with a pilot program, we might be able to 11 find that proper balance. 12 CHAIRMAN HEANEY: I just want to make the 13 distinction for everybody on the two different themes 14 that have been brought up, one of which is supplying 15 liquidity to the sub clients, the one that Jim discussed 16 that Alliance Bernstein is offering liquidity to their 17 client and, therefore, may need to do the trade that 18 minute, that time, in a block trade, and then the one 19 really Mihir talked about, which is now they've amassed 20 an opinion on a credit, could be an equity for that 21 matter, but an opinion on a credit, although, it'll take 22 time to get that equally in every portfolio they have, 23 they probably can't do that in a day's time, and so it 24 make take over two or three days. 25 Two very different trades, but come back to the 0144 1 same issue that was raised, at least on the block 2 trading. 3 Please, Gilbert. 4 MR. GARCIA: I'm Gilbert Garcia with Garcia 5 Hamilton. We're a bond firm in Houston, and I'm also a 6 trustee on the Dallas Fire and Police Pension Fund. I'd 7 like to go back to the concept of trades that aren't 8 getting done because I'm in the view that a significant 9 percentage aren't getting done. You know, in the old 10 days, we used to do a lot of work, and that would drive 11 the portfolio composition. Now, it's what trades can get 12 done are what's driving our portfolio composition. So 13 for any of the panelists, what can we do to get some 14 sense for how many trades aren't getting done, what types 15 of those trades, I mean what can we do to get that type 16 of information so when we look at these studies, we 17 really get a good full view of what's happening? 18 MR. JAKUBOWSKI: I mean I think it's -- right, 19 I mean it's hard to do that, but, you know, one area may 20 be just looking at cash flows in the certain products and 21 mandates and, you know, I think passive investing is a 22 good area to look at where money comes into an index 23 fund, you know, that they have to replicate that index 24 and what trades go through and what trades don't go 25 through and why is that? Is it they don't have 0145 1 transparency on pricing? They don't know where the bonds 2 are or no one can make the phone calls. That might just 3 be one suggestion because that's an easy mandate to say, 4 hey, money comes in, you know, you have to buy the index, 5 you know, why did you transact in X, Y, Z securities, why 6 didn't -- and there may be other reasons. 7 There may be liquidity. There may be cost, so 8 it may be tough to tease out, but that would probably be 9 the cleanest way to get at what trades didn't get done, 10 but I think it really gets back to this, you know, you 11 have to call so many people or you have to use so many 12 systems that everyone may not be getting connected. You 13 may have a seller on this system and a buyer on this 14 system, and they -- they're not getting aggregated, and 15 that trade never happens, for some reason, but there may 16 be other factors that may be tough to tease out as to why 17 that happens. 18 MR. MELI: You know, evidence for something 19 that didn't happen is, by its nature, very difficult to 20 assemble, right? So I think we should -- we should 21 temper our expectations probably to a certain extent, but 22 a few -- a few thoughts on that. First of all, Sonali 23 made mention earlier about the fact that like order 24 tracking is still someone nascent on the sell-side, and I 25 think over time, you know -- actually, over a reasonably 0146 1 short period of time, we'd have a better sense of order 2 flow that is successful versus unsuccessful, and there's 3 various reasons why we'd need to track orders, both 4 regulatory and internal that are -- that are -- that are 5 driving that. 6 So I think we'll have -- we'll have a sense of 7 say one category of trades that didn't happen, which is 8 orders we got that we couldn't find the other side of, 9 right? Then there's another category of trades, which is 10 orders we never got because nobody bothered to submit it 11 because they -- you know, they sort of like presumed that 12 it wouldn't happen, and those are -- you know, I think 13 it's very difficult to lay out what those trades might 14 be, but what you can -- what you can look for for 15 evidence that it, in principal, exists is to look at some 16 of the sort of natural relationships and look at the 17 volatility around some of those natural relationships and 18 how that's changed over time. So things like curves, 19 basis versus CDS, relative value across names. There's a 20 lot of evidence that there's a wider range that things 21 trade in before trades get executed, which says that, you 22 know, in more liquid times, those relationships might've 23 compressed more rapidly. 24 In other markets, you could look at things like 25 negative swap spreads or cross-currency basis as evidence 0147 1 where there are transactions that are -- that would have 2 compressed those sorts of like pseudo arbitrage 3 relationships that are no longer happening. It's very 4 hard to put a number on that, whereas the order flow, at 5 least, you could, you know, in theory put a number on. 6 But at the very least, you could say that there are -- 7 there are some modest implications on the market in terms 8 of the tightness of some of these relative value 9 relationships. 10 MR. MOGAVERO: I think that to an extent, 11 Lawrence's question and Gilbert's question are somewhat 12 related and maybe, you know, I might call on Jim to help 13 me out with this one, but there are -- regarding trades 14 that don't happen, I think there is, you know, anecdotal 15 behavior in the market where on the less liquid bonds, 16 you know, you might hear clients say, "I've got one phone 17 call to make, and if I make the wrong one, I'm in -- I'm 18 in trouble." Because -- and I do think that that 19 transparency element of it, that the instant print can 20 matter, you know, you brought up -- Lawrence brought the 21 example of, you know, when you line up the buyers and 22 sellers, you know, you're good to go, right? Like that's 23 no problem, everyone gets their trades done. 24 But to Jim's point of when you need the 25 liquidity, there's plenty of examples where, oh, yeah, it 0148 1 actually fit that person to buy it, no problem. It goes 2 on the -- it goes on the tape, and that's where -- that's 3 the new -- that's the new price. There's also an example 4 sometimes of if that -- if that doesn't fit that person 5 to buy and then Jim twists my arm and tells me I should 6 buy it, then if there's multiple other sellers in the 7 market, immediately and whether it's an interdealer 8 broker market or something, if I buy those bonds at 65, 9 low and behold, people are just offering them at 63, 62, 10 whatever, just to -- just because they want to get the 11 price lower because we're trying to find out the level 12 where something trades. 13 So there's numerous examples, and there's a lot 14 of different examples that -- of type of trades that can 15 or cannot occur. I do feel that in the instances what we 16 would really -- whatever we, you know, think about or try 17 to recalibrate or consider, my biggest ask would be to 18 think about how do we avoid the scenario where if a 19 client is playing a little bit of a guessing game on who 20 to call and if they call the wrong person, yes, that 21 trade then never happened, and, B, that bond's not going 22 to trade for a few more weeks. That's to no one's 23 advantage. So I feel like trying to figure out ways to 24 minimize that, which I anecdotally hear from folks 25 frequently is -- I think would be a good thing for the 0149 1 market. 2 MR. SWITZER: I mean you're basically talking 3 about just information flow, right? It's -- for us, we 4 have to have all of the information at our fingertips to 5 make a decision if we're going to try to sell the bond, 6 where is the best place to go. And sometimes, that's 7 more art than science, and what we're trying to do is 8 bring more science into it and bring technology into it 9 to get more pretrade transparency. And it's not just 10 prices or access, but it's inquiry, it's thought. It's 11 what have you traded over the past three to six months? 12 How do we get that information? And I think what the 13 market's getting good at is it's getting good at working 14 together to share that information. 15 I think we're trying -- we're showing you our 16 road blocks and our hurdles, and you're doing the same 17 thing with us, and I think that's where we're getting. 18 And also, you know, it comes and it goes. Like if you 19 want to sell a bond over the last six months, there's no 20 bond pretty much you couldn't sell. And it's harder to 21 buy because we're in a one-way market going that 22 direction. So these things ebb and flow. But when you 23 just drill down to that bottom 40 percent of the market, 24 that's the tricky part. It's about trust. It's about 25 information, and really, we just have to kind of start to 0150 1 open up and share what we're doing because that's 2 ultimately -- it's making that decision at that moment in 3 time. 4 MR. MCVEY: Rick McVey, MarketAxess. Following 5 on in this topic of improving liquidity in the fatter 6 tail of less liquid bonds, one thing we didn't hear at 7 all this morning was the impact of the Volcker Rule, and 8 I'm just wondering, especially from the bank 9 representatives that are trying to make markets in 10 securities that trade very infrequently, whether Volcker, 11 in your opinion has had any impact on your ability to 12 make markets in those bonds? 13 CHAIRMAN HEANEY: I'm going to -- I'm going to ask 14 Brian to kick it off and then we can -- we can move 15 around. 16 MR. ARCHER: Hi, I'm Brian Archer from 17 Citigroup. Before I answer your question, Rick, I think 18 there's a very important distinction we need to make, and 19 Drew brought this up a number of times, that every bond 20 is not created equal. And when you think about fixed 21 income investing, in particular corporate bond investing, 22 there's more to it than just trading that credit, and one 23 of the reasons I think things don't always trade when you 24 think they will is because you -- when you're looking for 25 a buyer for a particular bond, you might be looking for a 0151 1 buyer of a bond on a particular part of a curve. The 2 curve in corporate credit goes out 30 years, so whereas 3 somebody might have a seven-year bond for sale, we might 4 not have a buyer in the seven-year bucket for that 5 particular security or that person's portfolio doesn't 6 require a seven-year risk at that point in time. So it's 7 a lot more nuanced than just thinking about the way the 8 equity market works in terms of there's one security and 9 it's a homogenous market. 10 Our market is very different, and we need to 11 think about that. And that's why I think a lot of times 12 a trade doesn't happen is because that -- it's a very 13 unique type of market. Now, I do think some of the 14 electronification of the market has been very helpful in 15 disseminating information and having more of those trades 16 happen. I think the networks that are being created in a 17 marketplace by numerous vendors, by numerous banks and 18 clients alike are really helping that process along. If 19 that process doesn't help, then it goes back to incenting 20 the dealer to take upon that risk, and for that -- and as 21 Sonali is saying, then to find a way to hedge that risk 22 or warehouse that risk for a period of time. 23 And what's the incentive to do that? To me, as 24 a dealer, the biggest incentive for me is to find a 25 sufficient return on my capital. All right. So over the 0152 1 course of time -- I mean we're not going to get 2 sufficient return of capital on every bond trade we do, 3 but over a whole series of bond trades we do, we're 4 looking to get a return on capital, right? So how do we 5 incent dealers to get that return on capital, which I 6 think is one of the things that we need to think about. 7 And that goes back to some of the TRACE reporting 8 questions that came up. 9 In terms of Rick's question, I do think Volcker 10 Rule has had an impact on liquidity, but not in a way I 11 would -- I would think most people would think about it. 12 I think, you know, as we -- as dealers operate, you 13 know, our method of operations has changed somewhat, but 14 I think the more nuanced part of it is prop desks don't 15 exist. And you think about it, we're liquidity providers 16 to our clients. Going back pre Volcker, we had a lot of 17 liquidity providers to us that used a lot of leverage, 18 right? And I think somebody on the panel said before 19 that, you know, it's uncorrelated buyers in the 20 marketplace. And I think that's a very important point 21 in the sense that because prop desks don't exist, because 22 hedge funds don't use as much leverage or is not as an 23 important part of the market today as they were in the 24 past, our outlets for risk have changed. And I think 25 Volcker has -- Volcker has reduced our outlets for risk 0153 1 in terms of prop desks. 2 CHAIRMAN HEANEY: Drew, do you want to add to that 3 at all, different perspective or the same? 4 MR. MOGAVERO: I would -- I would -- I think 5 when we think Volcker on the trading desk, we think 6 reasonably expected near term demand. You know, that's 7 the one that we kind of drill into the traders as part of 8 the law, and I think for -- again, it gets back to the 9 not all bonds are created equal thing. It's like the 10 bonds that are trading actively, you know, if somebody -- 11 if Paul calls me and needs to sell 100 and it's an active 12 trading bond, it's a big chunk, it's a lot of risk, but I 13 can easily justify that I can reasonably expect near term 14 demand, whether it's in a bunch of fives and tens or a 15 twenty-five or if somebody takes fifty of them, I can 16 expect to be able to trade those in a near term scenario 17 in most market environments. 18 Where it gets trickier is in some of the less 19 liquid parts of the market where, you know, if I -- if I 20 buy some bonds and I'm trying to make an orderly market 21 and keep trying to trade them and they're less liquid, if 22 all of the sudden, I'm just accumulating a bunch of them 23 and I haven't been able to find another side because they 24 just don't trade as much, all of the sudden, am I, you 25 know, is there really reasonable near term expected 0154 1 demand there? Like I'm trying to find it, but I'm not 2 really sure, you know? So the uncertainty in some of 3 those less liquid parts of the market is where I would 4 say it is a more likely place you would see an impact of 5 Volcker. I think in the regular liquid go-go stuff, you 6 know, at Barclay's, we always run a pretty turnover-based 7 client-centric model, and so that's been pretty 8 consistent. We haven't felt any real impact in that part 9 of Barclay's. 10 MR. MELI: I'd add to that that there was a -- 11 there was a question in one of the first panels, I think, 12 where they listed a whole bunch of like potential changes 13 that might've been a spark for lower liquidity, and 14 Volcker may or may not have been on the list, but it 15 should be on the list. It was one of the long list of 16 things that I think tipped us, just to reiterate this 17 point, from a -- sort of a high liquidity equilibrium to 18 a low liquidity equilibrium. 19 But it's -- I think it's highly unlikely that 20 you could pull individual levers that contributed to all 21 of that, pull one individual lever and you flip -- and 22 you flip back. So at this point, to Brian -- to echo 23 Brian's comments, prop desks are gone, with or without 24 Volcker. I think it's less likely that like the equity 25 investors in the big, large banks would be very enthused 0155 1 about having huge prop desks spring up again, you know? 2 I think that, you know, the hedge funds are using less 3 leverage. It would be very unlikely that just switching 4 Volcker off, you know, causes the hedge fund community to 5 reassess their use of leverage and to see, you know, 6 leverage metrics go back up to where -- you know, where 7 they were before these rules came into place. 8 So I think that while it may very well have 9 been one of the sort of set of factors that contributed 10 on the one hand, I don't think anyone of those factors 11 individually would be enough to actually undo the changes 12 that have happened. 13 CHAIRMAN HEANEY: Larry? 14 MR. HARRIS: Just a quick question about this 15 last point. My understanding, and I'd like to be 16 corrected if I'm wrong, is that as the prop desks left 17 the banks, those traders just formed their own 18 proprietary trading firms, collected capital and 19 continued to do business, to some extent. So the 20 question is to what extent is that correct? 21 MR. MELI: Okay. I'll give one answer, but I 22 think some of the other bank people serving those clients 23 may have other perspectives, but I think that that was 24 very true in the -- in the moment, you know, this is now 25 whatever, seven, eight years ago, so it was very -- it 0156 1 was very true in the moment, but over the ensuing period, 2 the sort of -- the proportion of the business being done 3 with those sorts of accounts has shrunk. Those accounts 4 have shrunk. Many of them have been after having entered 5 the credit markets, have subsequently exited the credit 6 markets. And I think now if you survey the landscape, 7 you would say that that capital is not there; those 8 people are, you know, either doing something else or 9 trading in other asset classes that they've subsequently 10 then exited. 11 So there was a rush of people setting up their 12 own funds after having left prop desk, but I think that 13 was a very short-lived phenomenon, and their business 14 models were premised on a high liquidity equilibrium as 15 became apparent we were no longer in that equilibrium, 16 they had to turn to other -- to other asset classes. 17 MR. SWITZER: And that's definitely the case 18 that I see. You know you come out of the financial 19 crisis, the large banks were damaged. They were trying 20 to delever, derisk, and they weren't providing, you know, 21 the liquidity, the market-making or whatever it was that 22 were precrisis. And there was a void, and that's where, 23 you know, 30 or 40 or 50 or 60 regional dealers or bucket 24 shops stepped up and really actually acted as a 25 transmission mechanism. And there were a lot of those 0157 1 people that you talk about that started those firms and 2 it was very lucrative for the first couple of -- first 3 couple of years. I -- you know, a lot of guys used to 4 call it making boats and making cars in a day. 5 But that has gone away, and now, when you look 6 at Alliance Bernstein, I think from 2011 when I got 7 there, we were looking to increase as many counterparties 8 as we could possibly find because we were looking for 9 liquidity anywhere we could. And now, here we are in 10 2018, and that number's going down because now we have 11 more trusted counterparties, and we're doing business 12 with, you know -- much to what the numbers suggest that 13 we're -- that we're more focused with our -- with our top 14 echelon dealers. 15 MR. MOGAVERO: I mean I would maybe provide 16 like some examples of -- to the point of how that 17 liquidity that was there and whether from a bank prop 18 desk, for example, and yeah, maybe they gave it a go, but 19 it's to -- I second Jeff's comments that that, for a lot 20 of intents and purposes, is gone. You know, there were 21 examples of basis training, you know, for example. There 22 was a lot of derivative contracts floating around, and 23 there were a lot of accounts who would trade basis by 24 owning bonds and owning protection and capturing that 25 difference and then levering it up and earning returns on 0158 1 that. And there were a lot of those players. Now, there 2 are less of them. There's still a few, but I can count 3 them on one, maybe two hands. 4 And there was an example, I remember it well, 5 trading IDARC, it was a -- it was a directories company. 6 It was a real garbage credit, and it was going down. 7 And it was very obvious it was going down. And -- but 8 there was a lot of protection floating around trading at 9 80 points upfront or $.20 on the dollar and I had -- 10 you'd have all of these basis clients who would be like, 11 "All right. Well, if I can capture three points and buy 12 those bonds at 17, like I'll do that trade. I'll lever 13 it up. I'll earn a return. Give me as much as you can." 14 And the reality -- and those bonds that were 15 trading at 17, I knew were worth nothing. You knew this 16 company was going down, and it didn't have any real 17 assets. It was a subordinated part of the capital 18 structure. It was going to be bad news. And you would 19 call the clients and be like, "I've got a deep bid at 17 20 for these bonds, like you should -- you should sell them 21 to me." And people would be like, "You know what? 22 You're right, Drew. Thanks for the advice. Really 23 appreciate it. How many do you want?" 24 And that protection and those leverage basis 25 players created this triangle of extra liquidity that 0159 1 that part of it doesn't exist anymore. So I don't think 2 it would be fair to characterize that as just picked up 3 shop and went somewhere else. 4 MR. HARRIS: Another quick closely related 5 question. So Kevin talked about the growth of e-trading 6 in these markets, and we're all aware of it. Interactive 7 Brokers finds a two-sided quote in a huge number of 8 bonds. And so the question is in these electronic 9 venues, who's supplying these quotes? Most of them, of 10 course, are indications, but they're pretty firm, at 11 least for small size. And so who is it who's now 12 providing these quotes? Most of it seems to be automated 13 trading. Is it coming from you guys? 14 MR. MOGAVERO: I think it's coming from a lot 15 of different areas. You know, we'll provide prices to 16 ECNs, but I think, you know, ETF -- we mentioned before, 17 I can't definitively speak for them since I'm, you know, 18 not one of these other authorized participants, but it 19 wouldn't surprise me if they're providing small little 20 lots of liquidity to ECNs and trying to get smaller sizes 21 trading because that's a lot of the business model of it. 22 So I think the growth that the ETF product has had some 23 element of that. There's also algorithmic trading shops. 24 Again, I can't necessarily speak to their 25 strategies, just because I'm not one of them. But 0160 1 there's -- you know -- and maybe Rick can comment, 2 there's plenty of folks who are more -- they're not your 3 big bulge bracket dealers, but they're very active in 4 algo flows on some of the ECNs and trading venues. So a 5 lot of these alternative providers are all adding 6 liquidity to the ecosystem. And it's definitely helpful, 7 you know? It leads to trades and prices and it's 8 something that, you know, I imagine is going to continue 9 to grow and be part of the market. 10 MR. MCPARTLAND: And to large extent, the 11 liquidity that an Interactive Brokers or similar are 12 tapping into are very much retail-focused platforms. 13 They're very different from where, you know, Jim and Drew 14 will trade with each other, right, like through 15 MarketAxess. And yeah, there's certainly some automated 16 liquidity in there, but it's a very, very small size, and 17 it's only in those names that turnover enough where they 18 can do that often right after it's like a new issue or 19 something that's trading a lot for a couple of weeks and 20 then it'll die down. 21 Those markets, to some extent, are truly all- 22 to-all because there is so much retail in there, so there 23 certainly can be investors trading with investors. But 24 again, it's such a small size that at least it seems like 25 over time, you should have an aggregate of retail 0161 1 liquidity that can come together and interact with 2 institutional liquidity, and I think there's some 3 attempts to do that, but so far -- at least as far as we 4 can tell, that's not really happening yet. 5 MR. VENKATARAMAN: Kumar Venkataraman now, Cox 6 School of Business, SMU. There's been a record level of 7 primary market issuance activity. Can you comment on the 8 interaction between primary market activity and secondary 9 market liquidity and what might happen if primary market, 10 you know, activity drops off and how that might impact 11 the liquidity management of fund managers? 12 MR. JAKUBOWSKI: I guess I'll start. You know, 13 with the primary market growing so much, right, there's 14 less demand for secondary liquidity because you can know 15 that there'll be an issue, maybe multiple issues the next 16 day, the following day to put cash to work. It brings 17 out the concept of substitute-ability in the market, 18 that, you know, kind of to Drew's point, that not all 19 bonds are created equal, but some of them are close 20 enough that you can replicate the risk and get the 21 exposure that you need and you can use the primary 22 market. 23 And I think that's a reason why you see 24 turnover is down, just people are using the primary 25 market as a source of issuance, right? To the extent the 0162 1 primary market declines, right, that's essentially saying 2 either that there's less investor demand for fixed income, but 3 our assumption would be that that would spill over into 4 the secondary market, that people would demand more as 5 they have to trade either outflows or inflows and don't 6 have that access to readily available new issuance. 7 MR. SWITZER: I think, you know, when you -- 8 when you think of the new issue market, you know, the 9 buffer cap that I was talking about in the market that's 10 going down and you were looking for that shock absorber 11 in the market, the primary market is actually acting as a 12 buffer or, you know, buffer capital on a market that's 13 going up. 14 There's, you know, global coordinated central 15 bank easing going on. There's demand for credit 16 everywhere, and if we didn't have that primary market, I 17 think we'd even be tighter than we are right now, so it's 18 -- so it's actually -- and what would happen, you know, 19 to either, you know, the primary market or the secondary 20 market if one were to -- you know, if the primary market 21 stopped, it would depend on why the primary markets 22 stopped. 23 If it's stopping because rates are 24 skyrocketing, unhinged rates. Is it stopping because, 25 you know, the economy, whatever that might be. It's 0163 1 really -- it's really going to depend. It's hard to -- 2 but right now, they're acting in concert, and they're 3 actually supporting one and other. 4 MR. MOGAVERO: I generally view a healthy 5 primary market, while it certainly can substitute for a 6 secondary trading activity to the point that Paul just 7 described, for sure, a healthy primary market is usually 8 good for the overall market because you can see you get a 9 lot of level validation, like it helps. The points all 10 help. It's like a big block trade that just happened. 11 You're like, "Well, if that -- if a, you know, billion 12 bonds just issued there, then I can provide relative 13 value why this, you know, piece should be worth this and 14 this piece should be worth that." And all of those extra 15 data points provide more transparency and data points. 16 So I think that that's an important thing 17 that's usually a good part of a functioning market to the 18 extent that goes away, I think Jim's got it right. It's 19 going to depend on why it -- why it goes -- why it goes 20 away. Is it because, you know, the market demand is so 21 strong and companies can't issue anymore and like the 22 bid's still there? In which case, you'll have together 23 secondary spreads? Or is it because, you know, you've 24 had a shutdown in the marketplace because of macro stress 25 or global stress? And if it's the latter, that's 0164 1 certainly not great because then you lose that 2 transparency and liquidity and pricing reinforcement 3 mechanism. And then it becomes trickier and trickier. 4 MR. KUCHINAD: That's maybe the -- that brings 5 me to my question about pretrade transparency because -- 6 sorry, Amar Kuchinad, Trumid Financial. We're a 7 corporate bond trading platform. That might be the sixth 8 comment in this thread that we've been discussing where 9 some amount of transparency fosters volume and liquidity. 10 How do we as a marketplace or the SEC as a rule-setting 11 organization incentivize participants to provide pretrade 12 transparency? And maybe we can start with the dealer -- 13 the dealer groups because they're the ones that 14 traditionally have provided a lot of that. 15 MS. THEISEN: Yeah. So I think that's a space 16 that is rapidly evolving and one that we're very 17 supportive of, of providing targeted pretrade 18 transparency, excuse me, to our clients. Back to, you 19 know, some points made earlier, though, I think over 20 time, you know, there's benefits, and there's also 21 dangers of everyone getting the same information 22 pretrade, especially when it's quote-driven, just that 23 being broadly disseminated because, again, going back to 24 like there might -- if I know that you're a buyer of a 25 certain sector or a certain part of the company, et 0165 1 cetera, it might be beneficial for me to show you, you 2 know, larger size, et cetera that like I think that those 3 feeds, over time, will become a virtuous cycle of 4 becoming more targeted, both from, you know, what we know 5 of our investors and what investors will expect from the 6 dealers. 7 But I think like in terms of setting up and 8 establishing the pipes and plumbing, I think that we've - 9 - again, there's a lot of work left to do, but there's a 10 lot that's already, you know, foundational work that's 11 already, you know, being done by the marketplace. And I 12 think if you look forward, whether it's -- you know, I 13 don't want to set a timeframe, but I think if you look 14 forward, not only, you know, providing greater pretrade 15 transparencies to our clients, but then having direct 16 execution back, you know, through those same pipes is 17 going to be an evolution that our marketplace goes 18 through. 19 MR. REDFEARN: I just want to throw out there 20 that it's interesting, we've had the conversation about, 21 you know, how one -- all bonds are not alike and one size 22 doesn't fit all. I would argue that when we look at the 23 US equity market or other equity markets, to a certain 24 extent, there is a great similarity in terms of how very, 25 very liquid stocks trade and very, very illiquid stocks trade, 0166 1 at least in terms of the market structure and the 2 platform. So we have, I don't know, literally hundreds, 3 if not thousands of stocks that trade less than 50,000 4 shares a day. They don't trade a lot on exchange. They 5 tend to trade in an upstairs market. 6 They trade very differently, but the market 7 structure is effectively the same, so in light of the 8 last question, you know, I thought there was a slide in your 9 deck, Sonali, Page 23, since you seem to have them 10 memorized generally on the different ones, but so when we 11 talk about electronic trading and we talk about marrying 12 the appropriate market structure with the appropriate 13 characteristics of a given security, I wanted to just 14 throw out there, you know, how do we start to define this 15 challenge of, you know, we have electronic trading? It 16 could be, you know, one set of tools for one group and 17 another set of tools for another group. You know, and 18 how do you -- how were you looking at this and how do you 19 think we should create a framework so that we can marry 20 the market structure to the characteristics of the 21 securities appropriately? 22 MS. THEISEN: Our view on this is that figuring 23 out this sort of progression, this evolution of the 24 market, both by size and liquidity, et cetera, is one 25 that the market organically does as it has more 0167 1 efficiencies, as it has more technological efficiencies. 2 So our view is actually that you should allow the 3 markets, based on conditions, based on, you know, where 4 they are in the process of electronifying, et cetera, to 5 figure that out. It's certainly a discussion that we 6 have. I know, you know, Jim and I have had this 7 discussion before. It's a question that we always ask 8 our clients, like someone else was making the point 9 earlier, you know, which trades will you blast to 10 everyone? Which trades is it really important for you to 11 know my full size of what I can offer you? Which trades 12 would you put in RFQ to 3 or 4? 13 I believe it firmly that we're not supposed to 14 be prescriptive about how that evolves, rather, we're 15 supposed to focus on building kind of robust, safe, sound 16 markets, efficient infrastructure to let that evolution 17 happen. It will happen when there is confidence. And, 18 again, Jim has been mentioning this all afternoon, when 19 there is like speed, efficiency of being able to try 20 every protocol, go down the waterfall of how you might 21 want to execute. And I think it's our job as an industry 22 to make the ecosystem robust but not to prescribe the 23 protocols. 24 MR. MCPARTLAND: I agree. And if we -- if we 25 take the -- if we take the SEF process over the last 0168 1 couple of years as an example, it was very, very 2 prescriptive as to what you could and couldn't do. And I 3 think at this point, most everybody in the market will 4 agree that it was overly prescriptive. It didn't really 5 -- I think the clearing, the reporting and even the 6 requirement to trade on SEF, I think ultimately was 7 positive to the market, and most people agree with that, 8 but telling people how many quotes you need to get and it 9 was just -- it was too much, and it didn't ultimately 10 benefit investors because there are cases, right, 11 depending on security, where you really do only want to 12 call one person because -- I think Drew, did you say 13 that, right? If you make the wrong call, then you -- 14 then you only get one shot, right? 15 So this market has evolved -- as the corporate 16 bond market has evolved, and those e-trading tools have 17 evolved, it has evolved organically based on real market 18 demand, which I think is very, very telling, right? The 19 fact that all-to-all trading and open trading, right, and 20 MarketAxess has picked up and the way that it has picked 21 up is because the market found it useful and it worked in 22 those cases. And in other cases, it hasn't. And they've 23 -- and they've used other methods. So I think that's 24 very telling, and we're only going to see more of that 25 through the new venues and the new protocols that have 0169 1 come to the market. 2 CHAIRMAN HEANEY: So let me just ask a follow-up 3 then to Brett's question. Tom, you brought it up when 4 you talked about calibration of a pilot program, and 5 slightly to take the other side on this, and not trying 6 to sound cynical, but it's been 10 or 12 years that this 7 migration of systems, electronic trading platforms is in 8 the evolutionary phase of getting to the right place. 9 And it may be another 10 years if you allow that amount 10 of time for the gravitational pull to get there. So if 11 you took the other side and said, "Let's try a 12 calibration of a pilot program," if you got a little 13 prescriptive, and since before we're going to lose you 14 guys at 3:00, to pick your brain because I'm guessing 15 this will come up in our discussion, could you get 16 prescriptive? 17 Think of it of a prescription of -- and again, 18 Chair Clayton brought it up about kind of what is 19 liquidity? And there was the liquid and then the 20 illiquid sector. So thinking about the liquid sector, 21 not the tail, for the time being, could you come up with 22 a prescriptive way that you would actually -- to Brett's, 23 again, question, more equity-like create bonds that would 24 have to be on a platform, that would have to trade on the 25 platform? Could you come up with a prescription, do it? 0170 1 If so, could you -- how often would you change the 2 characteristics of those bonds that went out on there? 3 And if you can't, then tell me how long you really think 4 this evolution takes on its own. 5 MS. THEISEN: Sorry, I would just start by 6 saying I think we're running that experiment currently 7 with MiFID in Europe, right? So I think, you know, it 8 remains to be seen, but, you know, there is a framework 9 for defining what's liquid, and then there is strong 10 incentive to move those trades onto an electronic trading 11 platform. Now, there is not the prescription of you must 12 trade on an exchange; you must, you know, request for 13 quotes from so many people. Again, I think that will 14 evolve organically on those smaller tickets that are 15 deemed liquid. 16 You know, there is always a balance. No 17 transparency framework, there's always going to be, you 18 know, advantages and there's always going to be 19 challenges around them. You know, the challenge is 20 exactly what you said, the transient nature of liquidity, 21 and there's so much has been written about it. We've, 22 you know, published a lot on this, but the transient 23 nature of liquidity in the new issues, the larger sizes 24 and it tapering off, so, you know, there's the balancing 25 act of then how often do you recalibrate? What are the 0171 1 sizes, et cetera? And how much fatigue is the 2 marketplace going to have keeping up with that and what 3 are you -- what is the cost benefit analysis? And so my 4 only view would be that, you know, we are running that 5 experiment and we'll have some good data points in the 6 near future on that in Europe. 7 MR. SWITZER: So I would disagree that it's 10 8 years, you know, changing. I think you just need to look 9 back two years, and the amount that we've accomplished in 10 the last two years and how this market is changing. In 11 the previous eight, nothing happened. I mean when I 12 started in this business, you know, it was like starting 13 fires by banging two rocks together. And, you know, 14 eight years ago, we were still banging rocks together. 15 And now we're talking about utilities and transmission 16 mechanisms and Project Neptunes and all-to-all trading 17 and everything's changing. We're talking about portfolio 18 trades. And you actually see a willingness of the buy- 19 side and the sell-side to get together and try to solve 20 these problems. 21 So I think that's a really good sign, and I 22 think if we just look back those two years, we've come a 23 long way. I don't think it's going to take another ten. 24 I think we might be evolving for 10 years, but I think 25 the market is getting there very quickly. 0172 1 MR. MOGAVERO: I would even give just some 2 specific examples of what's just happened in the last 3 year, especially if you're talking about prescribing on 4 the more liquid parts of the market. The liquid parts of 5 the market, the transparency is happening. I mean you 6 can go to your website and there's 22,000 bonds and you 7 can click and see -- you can pick it up on your phone 8 right now. You can look and see this is where some 9 pretrade transparency is on these 22,000 corporate bonds. 10 That's awesome. You -- we have dealers who a year ago 11 when I grew up trading, you would -- you would send a 12 morning run and then some one-liners throughout the day, 13 and maybe if you weren't going to go out with the client, 14 a closing run. And now, you know, we started this year 15 as we've built technology and we've invested in 16 technology, our runs used to go out every two hours. 17 They were automated, so all we had to do is update 18 prices, boom, off they went. And then, of course, you 19 know, organic competition, all of the sudden, another 20 dealer's like, "You know what? I'm going to do an hour." 21 You're like, "Oh, yeah? Well, I'm going to do a half-an- 22 hour." "Oh, yeah, I'm going to go out every 15-minutes." 23 And that's been happening throughout the course of this 24 year. This is happening where I'm sure Jim sitting here 25 is like, "Oh my God. Do I have to see another run every 0173 1 10 minutes from this person? Like I -- like we know 2 where the bond is. I have the transparency, like I've 3 got it. And so we have to ask ourselves is what are we - 4 - you know, that almost -- does that harm the efficiency 5 that Jim's talking about getting? You know, like those 6 are -- we have a lot of those things and they're 7 happening quickly. 8 You know, a year ago, we didn't have the -- I 9 call it the runs war. You know, we didn't have the runs 10 war, and so the natural progression, of course, is going 11 to be like do we even need runs? You know, is there a 12 level streaming of the system? There's a lot of -- these 13 are things that I'm going to -- we're all evolving and 14 thinking about. Ritchie mentioned that the market's 15 evolving every month. And these are things that are 16 happening real-time. And I think, you know, it's 17 important to realize that a lot of this stuff, it's 18 happening. It's happening right now. 19 MR. SWITZER: You know, it's interesting, our 20 software that we've built, which is just the aggregator, 21 we've actually had to create suppression software because 22 the dealers have figured out if they send it enough, 23 their information is always at the top of the screen. So 24 we actually suppress it if doesn't change so that we kind 25 of get a better view of it, but that's how much 0174 1 information is pushing through. 2 MR. MCPARTLAND: And just to support the things 3 have changed quickly argument, two more numbers and then 4 I won't do any more numbers today, but in 2013, we had 24 5 percent of investment grade investors in the US trading 6 high-yield electronically. In 2017, it's 73 percent, in 7 only five years. That's pretty huge. That's a -- that's 8 a great looking chart. 9 CHAIRMAN HEANEY: Rick, 22,000, I think was the 10 CUSIP number, should it be 100,000? I mean, again, to be 11 a little provocative, the question is is it as much as it 12 can be? 13 MR. MCVEY: Well, I would agree with the 14 comments we've come a long way in the last three or four 15 years. A little known fact, judging by the -- some of 16 the comments is that there are -- we have a larger share 17 of less liquid bonds trading on MarketAxess than we do of 18 liquid bonds. And I think what that tells you is that 19 investors are using electronic marketplaces where they 20 need help with liquidity the most. And the liquid bonds 21 are newly issued and the dealer liquidity is ample and 22 they're trading in blocks and that's not where investors 23 today are going for electronic liquidity across a broad 24 network. 25 So I think that the market is solving these 0175 1 issues on their own. We're all for seeing it get a lot 2 bigger sooner, but I'm very happy with the trends that we 3 see in the automation of the corporate bond market in 4 particular and the beginning signs in the municipal bond 5 market. And the transparency has been a key ingredient 6 of making the electronic market successful because we 7 have created a lot of new market-makers through having an 8 efficient and transparent US credit market. So in these 9 illiquid areas now, the pool of potential market-makers 10 is significantly broader today than it was even three or 11 four years ago. 12 And the all-to-all trading is clearly taking 13 hold. Four years ago, it was about two percent of our 14 volume. In the year we just finished, it was about 16.5 15 percent of our volume. So you think about that added 16 layer of liquidity, it's as if in the aggregate, it's 17 created two new dealer markets. So we're very encouraged 18 to see how quickly the market is embracing all-to-all 19 trading. And it's clearly not just client-to-client 20 trading. Dealers are using the open trading liquidity 21 pool for their own liquidity, which is allowing them to 22 run their businesses more efficiently and keep their 23 balance sheets down. So by the end of last year, 30 24 percent of our all-to-all activity was initiated by 25 dealers. So yet, another tool for the dealer community 0176 1 to manage their liquidity needs and to manage their 2 balance sheets through all-to-all trading. 3 So my overall sense is that the market is 4 finding solutions on their own and the growth rates are 5 very high, and I'm optimistic that as we keep investing 6 and others do as well, we will have a much bigger 7 electronic market in the years ahead. 8 MR. HARRIS: This one's a real quick one. In a 9 world where your, by some means, aware that there is an 10 offer to sell say 80 bonds at $100 that is a locked in 11 offer, it's a firm quote, and you would like to trade say 12 1,000 bonds at $100.25 for a customer who wants to buy, 13 will you be able to trade through that 80 bonds at $100? 14 And just obviously, you could be prohibited by the SEC, 15 but given your client relationships, the common law or 16 whatever you think, would you be able to trade through 17 that? 18 MR. MOGAVERO: Is the question can I buy bonds 19 at $100? 20 MR. HARRIS: Okay. So you want to -- so you 21 want to print 1,000 for at $100 and say $.25, and there's 22 a locked in quote to sell at $100 for 80 bonds. Do you - 23 - would you feel comfortable printing the 1,000 without 24 picking up the 80 for the client? 25 MR. MOGAVERO: In most -- in most of our stuff 0177 1 in credit, we'll usually ask the client to make sure that 2 they're comfortable to transact, in most cases. Like 3 maybe it's just the nuance of the credit market, 4 especially in high-yield, but if there's -- if we feel 5 like we have an indication of interest or an order for 6 interest for a client and we're able to source liquidity, 7 we usually call that client and just be like, "Hey, just 8 want to make sure you still care to purchase these 9 securities." If that's the question. I may not 10 understand it. 11 MR. SWITZER: I mean in this day and age, you 12 know, if you -- if you traded at a quarter and you left a 13 guy behind at $101 or whatever it was, the guy's going to 14 see the print at $100.25 and wonder why he's not done, 15 right? So I mean I don't -- I don't think that's going 16 to happen. I mean typically, if these guys have a seller 17 at $101 and I want to buy bonds at $101.25, they're going 18 to tell me, "Well, you can buy bonds at $101." 19 MR. HARRIS: Or the more likely story is that 20 when you print the 1,000, you're selling 1,000 to your 21 client. 22 MR. SWITZER: Right. 23 MR. HARRIS: You're going to turn around and 24 immediately pick up the 80 for yourself. 25 MR. SWITZER: Well, I'm going to see the print 0178 1 and I'm going to go back and go, "What the hell just 2 happened?" 3 MALE SPEAKER: Exactly. 4 MR. SWITZER: "What did you just do?" 5 MR. HARRIS: So that's the question, I mean 6 where's the client going to be in this? Will you do this 7 or will you -- 8 MR. SWITZER: This is -- it's -- this is a 9 trust business, right? So if Drew does that to me, I'm 10 going to go back and go, "What just happened?" You know, 11 we -- you know, I just bought bonds. I paid this for it, 12 now you had that in your pocket? That is a problem. 13 14 MR. MOGAVERO: And we talked about it earlier 15 in the very beginning, the whole trust -- what hasn't 16 changed is the trust and transparency elements of the 17 marketplace, right? If I -- especially now with the 18 concentration in the market, it's just the -- if you do 19 anything to fray the trust and transparency relationship 20 with a client, that's just -- that's a -- that's a -- 21 that's a no go. It starts -- 22 MR. SWITZER: This market -- you know, this 23 market, the riskless principal model, the order business, 24 the riskless business, which is going up. I think you've 25 said the numbers before, or somebody said the numbers. 0179 1 You know, that's the golden goose. You don't want to 2 kill the golden goose, right? So there's a -- there's a 3 -- there's a protocol here that we're all very aware of 4 how this -- how this works. 5 MR. HARRIS: Right. And if the client doesn't 6 know about the $100 offer, then are you better off or 7 worse off? 8 MR. SWITZER: If the client doesn't -- 9 MR. HARRIS: So there's an offer for $100 for 10 80 bonds -- 11 MR. SWITZER: Again, you're -- okay, so you're 12 MR. ANDRESEN: Sorry, do you mind? I might be 13 able to take a crack at this. 14 MR. SWITZER: Go ahead. 15 MR. ANDRESEN: So the critical distinction 16 here, I think, in the question, Larry, is that there is 17 an offer for 80 bonds at one price, and then a slightly 18 worse price, but they're making it for 1,000 bonds. So 19 there's different ways that this could be treated by 20 regulation. The end story is that the seller at $100 at 21 80 will get filled, either by the dealer picking it up as 22 part of the basket of bonds they need to fill the full 23 1,000, or they will just fill it and hand it as riskless 24 principal or as agent to the client. But if you -- in 25 your example, if you went out and tried to fill 1,000 0180 1 bonds as agent, you would be getting 80 at $100, and then 2 you'd be buying dribs and drabs up for like 1.5 points 3 until you got completely filled. So it's an interesting 4 question about block trades versus retail sides. The 5 retail market should absolutely be in a firm and 6 transparent framework. But as you get up to the larger 7 size, then the question is how to balance that against 8 the retail, which I think we've heard today. 9 MR. SWITZER: And remember, we -- in credit, we 10 don't trade as agent. Agent, I think, would imply a 11 Commission. We trade principal. It's either principal 12 where they're taking the risk, the risk is transferring, 13 you know, onto or off their books, or there's riskless 14 principal. That's the order business, but there is still 15 an implied spread built into that. 16 CHAIRMAN HEANEY: All right. As it is just past 17 3:00, it's time to -- for us to go on break, and for us 18 to, again, thank all the panelists for your time, your 19 participation, unbelievable thought leadership the 20 morning and through the afternoon, so thank you very 21 much. And we'll be back at 3:15. 22 (Whereupon, at 3:00 p.m., a brief recess was 23 taken.) 24 CHAIRMAN HEANEY: Let me just start off by saying 25 we've worked through this agenda pretty well. For those 0181 1 that are traveling and for those that are not, right now, 2 it looks like we're going to try to conclude this by 3 4:00, so I think that's going to help a bunch who are 4 looking to travel or need to leave around that time. So 5 perhaps just another productive 30 minutes and we'll be 6 in great shape. 7 So I think, as Brett and I thought about this, 8 the next steps maybe is just to get the temperature of 9 people in this room, any further thoughts on the topics? 10 You know, we obviously have been in a fairly, I'll use 11 the word, robust discussion, as was planned, but further 12 thoughts before we break into what we think proposed 13 subtopics could look like? So either thoughts on -- 14 let's go broadly with thoughts on this discussion and 15 then as part of that, if you'd like to integrate what you 16 think subtopics may be. 17 MR. WORAH: Michael, I had some thoughts to 18 add. 19 CHAIRMAN HEANEY: I'm sorry, Mihir. Let me -- let 20 me go to Mihir. 21 MR. WORAH: No, go ahead, Lawrence. 22 MR. HARRIS: Just a real quick comment about 23 liquidity. So one of the debates is has liquidity 24 dropped or not, and the evidence is on two sides, one is 25 that we see smaller spreads and more volume, that 0182 1 suggests more liquidity, and then we also see less 2 turnover and essentially endless complaints by buy-side. 3 And so I wanted to address that question. To the extent 4 that spreads are smaller because the buy-side is 5 providing liquidity to the buy-side, you know, that's not 6 a problem. That's a good thing. And so we should be 7 aware of that. Now, as to turnover, we're only counting 8 turnover in the bonds, and we're not counting turnover in 9 the ETFs. So we're just told like the high-yield ETF is 10 turning over ten times a year or something like that. 11 Those are all trades largely between retail and 12 retail, but also institutional and institutional, and 13 also the cross-relationship as well. Those are trades 14 that would otherwise be taking place in the underlying 15 market if it weren't that there was just a cheaper way to 16 do it. And so there's a lot of liquidity there, and when 17 you ask like what would happen if there was serious 18 volatility or something like that, then historically, 19 dealers always disappear when -- in those circumstances 20 anyway, so it's just the buy-side meeting the buy-side. 21 The hope is that you won't have as much volatility 22 because they can meet each other in the ETF already. And 23 if they can't, well then, you know, it's sort of the same 24 situation. We'll just end up in the underlying, and 25 that's the way we would've been in the first place. 0183 1 MR. GARCIA: Can I follow-up on that? 2 CHAIRMAN HEANEY: Sure. Can I just go to Mihir and 3 then we'll come back to you if I can, Gilbert? 4 MR. GARCIA: Sure. 5 MR. WORAH: Yeah, so two points, and one of it 6 relates to that. So one is I think -- I think the 7 panelists did a great job in talking about how bid offer 8 spreads aren't the best measure of liquidity because it's 9 all top of the stack, it's all small size; it's for -- 10 it's for new issues. So along those lines, something I - 11 - you know, I worry about and I think people should be 12 concerned about, so that's a fact for sure, but we've 13 always also been an environment which has been and I've 14 even talked about the inflows. 2017 was a record year 15 for investors investing in the corporate bond market. 16 And when that's happening, you know, new issues 17 are the way to get liquidity. So buying is easy when 18 there's a lot of -- a lot of issuance. When that turns, 19 and we don't know when investor sentiment will turn, but 20 we know one of the things that's turning, which is the 21 Fed and the ECB. The Fed is buying zero bonds in 2018, 22 and the ECB is buying half of what it used to in 2018. 23 So new issues, just like in the treasury 24 market, auctions are a great way to get liquidity, but 25 you don't have a -- you don't have a way out. And I 0184 1 wanted to touch upon -- so let's -- you know, so all of 2 these measures, besides the fact that bid offer only 3 measures the top of the stack in small volumes, we've 4 also been in a bull market for corporate bonds for the 5 last seven or eight years where new issuance has supplied 6 a lot of liquidity needs. 7 Something to talk about in terms of ETFs, I 8 don't think ETFs are a danger to the market, forget that. 9 They're good for the market. They're not -- they're not 10 a danger, but I think a lot of the points that were made, 11 including the one that Lawrence just made about ETFs and 12 portfolio trading, at the end of the day, that -- again, 13 I think it's a function of the bull market and the 14 demand. At the end of the day, an ETF is sold and 15 there's a create and redeem process. At the end of the day, 16 the underlying bonds are changing hands. So in a 17 stressed market, again, ETFs are not a danger to the 18 market, but in a stressed market, we're going to go back 19 to the same old days and ETFs won't work. 20 And then finally, there is a third portfolio 21 trade, and, you know, we use it a lot, and again, it's 22 not a panacea, which is -- which is synthetic exposure. 23 The -- you know, the credit -- you know, the credit 24 derivative indexes. I think that's a big source of 25 liquidity in markets. And for an isolated stress test, 0185 1 which is what PIMCO went through in 2014 when, you know, 2 Bill Gross left and we had tremendous outflows, the use 3 of synthetics was really what kept the market liquid and 4 kept things going. So it's something to think about as 5 well. 6 MR. GARCIA: The comment I had to make, again, 7 I'm Gilbert Garcia, is that, you know, I understand a lot 8 of the research we receive, and I understand the concept 9 of the bid offer and so forth, but anecdotally, I can 10 just say -- and we trade every single day, is the 11 liquidity is vastly different. You know, in the old 12 days, you could do a $50 million corporate bond trade and 13 you could get that done. Now, even a $20 million trade 14 is hard, and you've got to do it on order; you've got to 15 work it. Even something that used to be very easy has 16 now become very difficult. 17 And it's happening in an environment when, as 18 you mentioned, this has been a great environment for 19 corporate bonds. It's almost as good as it gets, right? 20 So I think what we need to do is really think through, 21 you know, what can we do to improve things to be ready 22 for the next crisis because if this is as good as it gets 23 and liquidity is what it is, imagine what it would be 24 like the next time we have some sort of stress. So I 25 think that's what we should start thinking about. 0186 1 MR. THEES: Can I just ask a follow-up question 2 to you -- 3 MR. GARCIA: Sure. 4 MR. THEES: -- based on that comment -- 5 MR. GARCIA: Sure. 6 MR. THEES: -- 50 or 20, would you make the 7 same comment five or two or one? Or do you feel like 8 that part of the market is not the part of the market 9 you're referring to because that's gotten -- that's 10 gotten better? 11 MR. GARCIA: Sure. That's gotten much better. 12 In fact, what electronic trading has done, it's been 13 wonderful for odd lots. Where in the old days, you know, 14 your sales coverage or your trader that you're speaking 15 to wouldn't even want to -- ah, don't even want to bother 16 with that. The electronic trading takes the human out 17 and so now, it's been fantastic for trades under two 18 million. 19 MR. THEES: I'd like to point out it only takes 20 the salesperson out, not the human -- 21 MR. GARCIA: I understand. 22 MR. THEES: -- as a former bond -- as a former 23 bond trader. 24 MR. GARCIA: Point well taken, thank you. 25 MR. MADHAVAN: Hey, Ananth Madhavan from 0187 1 Blackrock. I just want to comment a little bit about the 2 ETF flow implication. So I think this has come out in 3 bits and pieces in some of the previous discussion from 4 the panel, but, you know, the ETF flows, obviously, can 5 come both ways. You know, there can be inflows and 6 outflows and ultimately, I think as Mihir pointed out, 7 those flows have to be manifested in primary market 8 activity at the end of the day. 9 There's a lot of settling out. There's a lot 10 of, you know, secondary market trading for every dollar 11 of primary market trading. We have seen ETF stress -- 12 tested in stress situations like the taper tantrum, like 13 the financial crisis, and we've actually seen that 2A 14 mechanism doing just fine. 15 The second point I just want to make, which was 16 really in comment to the notion that somehow the ETFs 17 will gravitate towards the most liquid bonds. If you 18 actually look at the large ETFs, they track their 19 constituent benchmarks very closely. So these 20 constituents may be 1,100, 1,200 bonds, yet, the tracking 21 error is typically, depending on the ETF, quite low, 22 meaning the fund managers are doing a good job of that 23 being able to access some of those less liquid bonds and, 24 in fact, providing liquidity into some of the bonds, to 25 Gilbert's point, in smaller sizes, in some of these areas 0188 1 that are less traded. 2 CHAIRMAN HEANEY: Kumar, please. 3 MR. VENKATARAMAN: I would like to share some 4 research coming out of academic studies on post-trade 5 transparency, the impact of transparency. When it was 6 rolled out between 2002 and 2005 for corporate bonds, it 7 was done in a very structured way, such that one can 8 actually design very clean tests of the impact of 9 transparency and, in fact, Commissioner Piwowar and Larry 10 Harris and myself and others have looked at this very 11 carefully. 12 And I've also looked at the recent 144A 13 transparency event as part of my research. And the 14 evidence is, at least based on the data, it suggests that 15 transactions cause of decline, turnovers have not been 16 hurt. Block trading activity has not been affected in a 17 significant way. Dealer concentration has declined, 18 implying that smaller dealers can compete more 19 effectively with larger dealers, and clients of these 20 dealers actually benefit in the form of lower 21 transactions cost. So at least based on the evidence 22 that we have, based on the best data that we have, 23 evidence seems to support that post-trade transparency 24 has benefited market participants. 25 MR. REDFEARN: Again, what was the -- for the - 0189 1 - for the study, what was the example of the transparency 2 that you were looking at specifically? 3 MR. VENKATARAMAN: So this would be the 4 initiation of transaction reporting in July of 2002, as 5 well as the sequence of bonds that were -- became 6 transparent over time, so 500 bonds were introduced in 7 July of 2002, and subsequently, bonds were introduced in 8 stages, so you had a very nice sort of classic even study 9 type approach where you can compare bonds that were 10 treated with control samples which were not. And this 11 was thanks to the design by the SEC at that time on how 12 transparency should be ruled out. So there's a lot of 13 academic evidence on that. 14 MR. ANDRESEN: So just a point I would like to 15 make in the interest of clarity. There are, as we've 16 noted today and all of the panelists noted, there are -- 17 not all bonds are created equal. There's lots of 18 different types of fixed income instruments. Likewise, 19 there are many different types of fixed income consumers 20 in terms of what -- the size of the bonds, the types of 21 credit, risks, et cetera that they're willing to take. 22 And I just wanted to maybe take a shot at trying to say 23 what those large buckets are so that when we're speaking 24 about liquidity or talking about incentivizing dealer 25 behavior, et cetera, that we're always speaking to one of 0190 1 the particular quadrants of those things. 2 So if you think about fixed -- today, we heard 3 a lot about institutional size trading in corporate 4 credit, and that was the focus from the majority of our 5 panelists. But the fixed income world is very large and 6 very diverse, and if you look at things that are the most 7 liquid to the least liquid, you'd have -- on the most 8 liquid end, you'd have US treasuries, which the on the 9 run US treasury trades as continuously and as order- 10 driven a way as the most liquid US equities. So that is 11 one extreme. Then that would be followed by off the run 12 US treasuries and US corporates. And then at the extreme 13 end would probably be municipal securities where there is 14 more CUSIPs and more complexity. 15 So when interest is concentrated in a fewer 16 number of instruments, sort of six buckets for on the run 17 treasuries down to a million municipal CUSIPs, that's 18 sort of the gamut. And so when we think about 19 suggestions for the Commission about what regulation can 20 or should do, we should be mindful of which of those 21 buckets maybe we're talking about and what the 22 anticipated benefits might be. 23 Likewise, it's important to recognize the 24 difference in the type of consumer, so the retail sizes, 25 which are the odd lot as typically you think of maybe 250 0191 1 bonds or less, all the way up to blocks and then mega 2 blocks, we -- you know, PIMCO, obviously, and Blackrock 3 being at the one extreme end of that. But it is worth 4 nothing that the demand for transparency is probably most 5 -- the benefit is most strongly felt down at the retail 6 end of the spectrum, and when a retail customer has no 7 idea when they're selling a bond, no idea within five 8 points what price they're going to get, we can agree that 9 perhaps some degree of pretrade transparency might be 10 useful. 11 Fortunately, really, that does exist already in 12 the market structure. You have ATS platforms like 13 MarketAxess and others that are out there, and the RFQ 14 process where you can generate an auction. If there's no 15 prices in the marketplace, you can just ask the 16 marketplace, "Hey, does anybody have any interest?" And 17 those auctions are actually quite robust. We were very 18 surprised when we got into this business three years ago 19 to find that we were not alone and that actually, the 20 average number of bidders, even in municipals is over 21 seven bidders per auction. 22 So if you had seven bidders on your house, 23 you'd be thrilled; seven bidders on your bond, you'd also 24 be thrilled. So that mechanism does work, but we have -- 25 and I think a thing that we can examine as a committee 0192 1 and the Commission can look at is that being made maximal 2 use of by the market because that is a proven market 3 structure now that is benefiting retail, what can we do 4 to extend that? 5 And on the institutional side, I'm mindful of 6 well, can we take transparency too far? But I think I 7 would always err on the side of starting with, you know, 8 let's have every trade printed, and then we can have a 9 reasoned debate about where to draw the line to something 10 that is market-moving, either because of the particular 11 underlying credit or from the relative size of the print, 12 and you can always figure out what might be the right 13 block trade exemption for either delay or suppression of 14 that. But I think we should -- my view is to err on the 15 side of -- for -- especially for the retail side is on 16 full transparency. 17 CHAIRMAN HEANEY: Thank you, Matt. 18 Rick, let me go to Rick and then we'll go back 19 to -- 20 MR. MCVEY: Just want to follow-up, you know, I 21 -- because I really did not think it got enough air time 22 in the panels, but just to follow-up on some of the 23 previous comments, I think we should recognize how 24 extraordinary the times have been the last six or seven 25 years. With 12 trillion in new liquidity from 0193 1 quantitative easing from central banks going into bond 2 markets around the world, sending 10 trillion in 3 government bonds to negative yields. And so we have had 4 this flood of money coming into the US credit markets, 5 leading to unprecedented demand from asset managers for 6 more bonds and corporate issuers accommodating that 7 demand by record levels of issuance. 8 And my point on that is, as we heard earlier, 9 we all know that the corporate market is now 75 percent 10 larger than it was the last time we had a significant 11 increase in credit spreads in a very different market 12 environment, the timing of this work for the committee 13 and the Commission could not be any better because we're 14 starting to see signs now that quantitative easing is 15 going the other way, and central banks are starting to 16 raise rates as economic growth around the world gets a 17 lot better. 18 And so what I think we need to focus on is what 19 are the steps that we can do to strengthen our secondary 20 markets and make them as robust and vibrant as possible 21 because the liquidity needs in a scenario where rates go 22 higher have never been greater. And against that 23 backdrop, we heard a lot today about the bank balance 24 sheet constraints, and they are real, and they are not 25 going away. So in our opinion, it's really important to 0194 1 create vibrant secondary markets and encourage the 2 broadest possible participation in US fixed income 3 markets and open access in all ways to create that 4 breadth of participation in secondary trading that we 5 will certainly need if we're starting to enter into a 6 rising rate environment with all of this new debt in our 7 market. 8 The last thing I'd like to say is that I don't 9 think there was enough commentary today recognizing that 10 we have the broadest and deepest, most transparent and 11 most efficient fixed income markets in the world. We 12 operate globally in markets across Europe and Asia and 13 even Latin America, and we are the envy of the world in 14 terms of the way that our markets operate. It's great 15 that we're all here to talk about ways that our markets 16 can get even better, but my first rule of thumb would be 17 do no harm. These markets are much better and much 18 deeper and the capital markets are wide open. 19 The level of issuance is unprecedented, and 20 it's moving through our markets very smoothly. So I 21 think -- I think there's a lot to be proud of in terms of 22 the way that the US fixed income markets are operating 23 today. 24 CHAIRMAN HEANEY: If I could, and then I'll go to 25 the questions, just draw from what you initially said 0195 1 with Mihir, what you said, which is it has been a bull 2 market. There is a functioning primary market, and 3 obviously, we have two issuers with us who can chime in, 4 but it is a robust inflow oriented market. And Gilbert, 5 to your point, and going back to what I said initially, 6 our job is to come up with recommendations to the 7 Commission on market structure, forward-looking, future 8 state. And I think we'd all agree, maybe not, that we 9 are in a higher interest rate environment, as you're 10 talking about, and at some point, there's going to be 11 trip wires. And this is a fascinating time to have the 12 start of this committee for all those reasons. 13 And I agree it's an incredibly functioning 14 market. The question is how do you make it better? And 15 how do you cut those trip wires, knowing what we know 16 about the last seven years? And maybe, to Rich's point, 17 the crisis isn't the right starting point, but take today 18 going forward as the starting point, what do we do to cut 19 the trip wires knowing that there is certain control in 20 the hands of this committee, hopefully, with the 21 Commission to do that? So I'd just -- I throw that out 22 there. I know I cut somebody off. 23 Larry? 24 MR. HARRIS: Since equity markets are so much 25 more volatile than credit markets, we'd expect 0196 1 transaction costs to be much higher there, but they're 2 much lower, and they're much lower for two reasons. One 3 reason has been brought to our attention repeatedly 4 today, that there are just so many bonds, and the search 5 for bonds makes things costly. 6 But the other reason we have to pay close 7 attention to as well. In the equity markets, the buy- 8 side very easily meets the buy-side, and this happens 9 regularly. And so much so that they've largely pushed 10 most traditional dealers out, and the only dealers left 11 are the high-frequency traders, who are extraordinarily 12 efficient at what they do because they have information 13 systems that know how to play one stock against another 14 and so forth. And play it against another, I mean by 15 hedging. 16 So Matt is right about we have these systems in 17 the bond markets now that allow the buy-side to meet the 18 buy-side. The trouble is that many of the brokers are 19 not allowing their clients to participate in these 20 systems. And so my expectation is that all we have to do 21 is figure out how to make it easier for the buy-side to 22 meet the buy-side, initially starting, say, at the retail 23 level, and that will suck in as much institutional 24 interest as the institutions are interested in. 25 And if they want to do it, that's great. If 0197 1 they don't, they don't. But my guess is that they'll end 2 up splitting up their orders as they've done in the 3 equity markets, and they'll participate and further 4 strengthen those markets and that's what'll eventually 5 merge. 6 But the real challenge is that there are -- the 7 most, perhaps the vast majority of retail traders cannot 8 get to the fixed income markets that have these 9 facilities that allow them to trade easily at prices that 10 are readily available to the -- to the brokers and 11 dealers who represent their orders. 12 MR. REDFEARN: Larry, when you say the buy- 13 side, meaning the buy-side directly and equities, are you 14 talking about like the Liquidnet Luminex example or are you 15 talking about the fact that brokers are providing agency- 16 based algorithms into the marketplace where whether or 17 not they're engaging with intermediaries in the process 18 or not, they -- it's just easier to find the other side? 19 MR. HARRIS: Okay. So when I say the buy-side, 20 I mean the retail institutions, and anybody, who's the 21 natural holder of the security, meaning somebody who 22 wants to -- somebody's buying; somebody's selling. The 23 seller, of course, is on the buy-side as well, but so 24 what I mean by buy-side is the people who actually 25 ultimately hold securities, as opposed to the dealers who 0198 1 intermediate them. And so when we say the buy-side meets 2 the buy-side it's because in an exchange, my order to buy 3 and your order to sell can match, could match with an 4 institutional order, could happen at an exchange, in an 5 ECN, a dark pool or who the hell knows what, but it 6 doesn't have to go through a dealer. 7 And it's not that the dealers don't provide 8 service, they provide a lot of service, but when they're 9 not needed to provide the service, then probably we 10 shouldn't be paying for them. 11 MR. MADHAVAN: I think the historical element 12 is important here. So historically when search costs 13 were high, dealers were important because you knew the 14 dealer would ultimately find the buyer if you wanted to 15 sell. Might take a few days, might take a few weeks, but 16 whatever. You'd find the dealer -- you'd go through a 17 dealer. But now in the -- in the modern era with search 18 costs being driven down by technology, the technology 19 that we have built enables, as Larry said, the buy-side 20 to meet the buy-side. And you're using more general 21 description of the buy-side than I would, but you're 22 including all of the retail -- the ultimate holders. 23 And they can meet each other now much easier 24 through technology, which didn't even exist five or eight 25 years ago. 0199 1 MR. THEES: Michael? So just to follow-up on 2 Rick's point and to your point, Larry, whether it's the 3 buy-side meeting the buy-side or the sell-side meeting 4 liquidity, the fact -- the fact of the matter is the 5 markets are evolving. We have been in a tremendous 6 evolution over the course of the last five years. Open 7 trading enunciated to the marketplace that the all-to-all 8 market could work, and it provided the technology. It's 9 not like conceptually we just found out about all-to-all 10 trading three or four years ago. We knew about it 20 11 years ago, it just didn't happen, and there wasn't any 12 technology to be able to affect that. 13 That technology is advancing rapidly. There 14 are a number of competitors in the space; that evolution 15 is going to continue to evolve. But the fact of the 16 matter is the dealers are taking just as much advantage 17 of that technology as the buy-side because liquidity is 18 the main -- and again, it's not a God-given right, as 19 Richie said, but it is the main objective in the course 20 of -- the course of transferring risk on a given day, 21 whether it's the sell-side or buy-side, we're getting 22 into an equalization mode, number one, the buy-side is 23 tremendously larger than the sell-side. They need the 24 liquidity more, but there's no less need for the 25 liquidity in the intermediary on the sell-side as well. 0200 1 And I -- and I think we have to give credit to 2 the markets for that rapid two, three-year evolution 3 that's occurred on already, and I expect that that, you 4 know, acceleration is only going to continue with the 5 advancement in technology going forward. And I do think 6 technology is a big part of the answer. 7 MR. HARRIS: One more -- 8 CHAIRMAN HEANEY: Please. 9 MR. KROHN: Just to jump in as an issuer, and 10 maybe to connect this a bit more to Main Street from I 11 think the financial markets that we've heard a lot from 12 today. You know, I think the discussions on the panels 13 this morning, you know, from a Verizon perspective, we 14 have $50 billion of pension assets that represent 15 401(k)s and DB employees of 200,000. So this is real to 16 them, whether it's their life path funds, whether it's 17 our ability to continue to derisk and give out more LDI 18 mandates to many of the people in this room. 19 And those LDI mandates are chunky. And so you 20 can't just buy in the primary market. Efficient 21 secondary markets are important so that our employees get 22 the benefit of the execution from our managers to achieve 23 those outcomes. So that would be my comment on the 24 pension side. 25 On the corporate debt side, so Verizon is one 0201 1 of the largest corporate issuers out there. We have 2 about 120 billion of debt. And we have a very high class 3 problem, and I learned today that maybe we've earned some 4 of this liquidity. That was nice to hear. But what 5 we've noticed in the market and Commissioner Stein 6 mentioned this this morning in terms of the 7 concentrations, which we really haven't talked about 8 today, I'd say Verizon is a very high concentration of 9 the corporate market liquidity that's offered. 10 Now, that's a huge benefit for us, but it 11 doesn't come without the drawback. So, for example, the 12 oil shock was mentioned from a few years ago. What's the 13 first thing that a trader does to express their view in 14 the corporate bond market when something's going wrong in 15 a sector away from our 115,000 mobile phone customers, so 16 like when the oil shock happened? There's nothing about 17 our 115,000 mobile phone customers that changed; nothing 18 about our cash flow that changed, but our spreads blew 19 out like 50 basis points because that's where the 20 liquidity is. 21 And so, you know, there's certainly an 22 advantage to having a very liquid bond complex, but there 23 are also some drawbacks. The procyclicality that I think 24 has come into the market since the Volcker Rule and since 25 these banks have shrunk their balance sheets is very 0202 1 concerning to me as a corporate treasurer as we think 2 about the next downturn. 3 So, you know, I'm fearful our spreads will blow 4 out even more in the next downturn. Now, arguably, we 5 would be able to issue a lot easier than other names, but 6 I think part of what we should be looking at here, not 7 just the concentrations on dealers, not just the 8 concentrations on which investors are consolidating the 9 market, but if we're really thinking about triggers and 10 downturn planning for that next crisis, you know, in 11 terms of which bonds are trading, are we looking at those 12 concentrations? 13 And as we think about not having to do another 14 TALF, not having to do another CPFF, not having to do any 15 kind of bailout program to the capital markets, are we 16 really studying the sectors of the economy? Yes, you 17 earn liquidity, but there -- you know, there are some 18 parts of the market where maybe, you know, there should 19 be some focus to make sure that those sectors of the 20 economy are getting liquidity. 21 CHAIRMAN HEANEY: Thank you. 22 MR. HARRIS: One real quick one. As long as 23 we're talking history, let's go way back. Bruno Biais 24 and Rick Green did a study of bond trading at the New 25 York Stock Exchange in the 1920s. So then in the New 0203 1 York Exchange basically pretty much had an exclusive fund 2 bond trading. And they found that the bonds then traded 3 at lower transaction costs than we presently see. Okay. 4 So it's not technology, it's rules and systems and 5 access and perhaps agency problems as well. 6 So what was their database system? The 7 database system was what -- was the bond cabinet. The 8 bond cabinet was a wall of drawers, which held filing 9 cards, upon which limit orders were written. Can you 10 imagine how much better we could do that now? That's 11 what computers are really good at. So I don't think it's 12 so much the technology. I think it's ultimately access. 13 The concentration of the order flow into a single place, 14 obviously, is advantageous. And the rules at the New 15 York Stock Exchange facilitated pretty tight spreads. 16 Now, I'm not suggesting that we compel 17 everybody to trade in one place, but it's pretty sobering 18 to think that at a time when the advanced technology was 19 the telephone, they had better transaction costs for 20 similar issues than what we presently have. 21 CHAIRMAN HEANEY: Rachel? 22 MS. WILSON: And just as another theme, as an - 23 - as an issuer to kind of echo some of what Scott's 24 saying as well, we also with the changes in the tax regs 25 that's come down, and we've been taking a lot just 0204 1 looking internally at our US market, but guess what? 2 With these regs, we're looking externally. So we've 3 already placed 25 percent of our debt overseas. We've 4 got, you know, a very global business, so 25 percent of my EBITDA 5 naturally matches, but you're going to see more and more 6 of us kind of issuing overseas. 7 And we said what is the concentration in the 8 buyers when we were listening to Citi. You know, a lot 9 of it was foreign money. So what's going to happen as we 10 -- you know, as things change? So just a -- you know, 11 another thought of what are the external pressures beyond 12 just the US market? 13 CHAIRMAN HEANEY: So if I can, perhaps let's pivot 14 just a little bit to potential subtopics, which again, 15 feeds into the subcommittees to come back in April. 16 Brett and I have kind of put our heads together at 17 different points through the day. It seems like there's 18 common ground, but I'll throw them out here for the 19 group. If there's agreement that perhaps these are the 20 topics to consider, we'll take a step forward and then -- 21 and go to the next step of that. 22 If there's things that aren't mentioned that 23 you think should be, please, let's have that discussion. 24 But broadly, it seems like one topic could be let's call 25 it modernization, which is technology and electronic 0205 1 trading platforms. It's going to sound broad, but just 2 giving it a shot at these. 3 The second could be fixed income, ETFs, still 4 the discussion including open-ended mutual funds. It'll 5 have concentration effects as part of that, in some way, 6 shape or form. Jeff has talked about it. Sonali talked 7 about it. Now, Scott's talked about it in different 8 ways. 9 And then the third being transparency. I know 10 that's going to sound very broadly defined, but thinking 11 pretrade, post-trade, certainly the TRACE discussion came 12 up. I think there's ways to talk about primary issuance 13 embedded in the transparency portion of that. So I'll 14 stop there and just kind of drop those in the middle of 15 the room and see what people think as broad themes. I 16 wouldn't get -- because Brett and I will go away with the 17 working committee with the group from the SEC, if those 18 are the broad themes, and kind of drill down a little 19 more granularly on that, that's really on the fly, but 20 just the broad themes as concepts of three subgroups -- 21 three subtopics that then we'd form the subcommittees 22 around and come back and after the analysis between now 23 and April. 24 MR. KUCHINAD: Michael, I was thinking one -- 25 this might fit into the electronification and platforms, 0206 1 but one topic that came up sort of on the periphery of 2 Volcker and some of the regulation talk and capital 3 flight is new participants. I mean I was surprised that 4 in this entire conversation of liquidity, we really 5 didn't talk about new -- trying to incentivize or create 6 -- find new pockets of liquidity, as opposed to really, 7 we talked a lot about data and transparency, so maybe 8 it's not important enough, but I don't know if we want to 9 have that as a topic of discussion of how do we encourage 10 or how does the Commission incentivize the competition or 11 the formation of capital to provide new liquidity in the 12 marketplace? Maybe that's electronic trading. 13 MR. REDFEARN: It's a good suggestion, and I 14 certainly think that with technology, electronic trading 15 and transparency, those are probably some of the key 16 underpinnings there, but it's a good point. Thank you. 17 CHAIRMAN HEANEY: You know, the interesting thing of 18 this discussion, at least my own personal opinion 19 takeaway, is this is the ultimate Venn diagram. Each one 20 of these three overlap in some way, shape or form through 21 the discussion from minute one to the last conversation 22 that -- the last point that Scott made or that you're 23 making, and they're all intertwined, so the tricky part 24 may be separating these as to three, where you go off and 25 do separate analysis, and again, I'll get into that in a 0207 1 minute, because they do have so many overlapping 2 characteristics. The positive of that would be if you 3 can actually come up with ways to think about these 4 independently, maybe when you put it back together in 5 April, there is a consensus of one or two things that 6 could actually touch all three, just throwing that out 7 there. 8 MS. BROWN: Michael the one -- 9 CHAIRMAN HEANEY: Sorry, Carole. 10 MS. BROWN: -- I'm sorry, the one point I 11 wanted to make on transparency, when you -- if you go 12 back to the spectrum that Matthew talked about from most 13 illiquid to most liquid, I think that that question of 14 transparency issue gets toward the end, and it's the end 15 I'm most familiar with right now with the most liquid in 16 the muni market, that transparency question and the price 17 discovery and all of those things are going to -- it's 18 going to have a much different paradigm than if you're 19 talking about the corporate bond market, and so I agree 20 with you, but I just wanted to make that point. 21 CHAIRMAN HEANEY: And I agree, I'll go back to 22 Matt's point and yours, but interestingly, it really is 23 these subcommittees, I don't want to overstate or 24 understate, but this is where the -- a lot of the work 25 will really, truly get done, and I think the 0208 1 differentiation across the spectrum that Matt laid out 2 will have to be part of that subcommittee work, for sure. 3 MR. GIRA: I had maybe two areas, maybe they're 4 subsets of what you just laid out. And I would say -- I 5 don't know that this came up today, but I would call it 6 gaps. And that is there are -- there are some 7 instruments that people are trading that actually -- and 8 I'm thinking of sovereign debt here, and I know this 9 could raise a lot of issues, but do we -- or do we want 10 to make sure that within our whatever transparency regime 11 we sort of look at in modernization, are there some 12 instruments that are not being say, for example, reported 13 through TRACE today that maybe should be? 14 And similarly, this is coming up with 15 treasuries, we -- and hopefully these gaps are filled, 16 but when you look at sort of holistically, it seems like 17 if people are trading the same instrument, they should be 18 subject to similar requirements. And so for exempt 19 securities today, there are some entities that are 20 subject to TRACE requirements and some that aren't that 21 are trading the same instrument. So I would just throw 22 that out a potential gap issue to look at. 23 And then another one, and this could be 24 modernization and maybe to your point when we started, 25 Chairman Clayton, is best execution and there were -- and 0209 1 we even had some examples about, you know, trade through 2 them. And while we don't want to bring all of the equity 3 notions into the fixed income market, should we be 4 thinking, though, as the markets do start to tilt that 5 way, are there -- are there some customer protection 6 issues or rules on the equity side that, in some form, 7 might also apply to the fixed income market? 8 CHAIRMAN HEANEY: Thank you. I want to -- 9 MR. ANDRESEN: To that -- sorry -- 10 CHAIRMAN HEANEY: Go ahead. 11 MR. ANDRESEN: -- to that point about the -- 12 those seem perfectly reasonable cuts at subcommittees. 13 One thing just to question is these are obviously, by 14 definition, as you noted, very broad topics. And as 15 we've seen today, there is a dichotomy between sort of 16 the institutional problem and the retail problem. So a 17 thing that may be worth considering as you guys think 18 about how to divide these up is perhaps more of a retail 19 focus on one versus more of an institutional focus on 20 another. 21 So in that retail, you'd have, you know, best 22 execution and more straightforward transparency issues. 23 Whereas for some of the institutional sides, you're 24 dealing with more -- you know, more prospective solutions 25 that might be very different. 0210 1 CHAIRMAN HEANEY: It's a great -- it's a great 2 point. And you could think about -- and I'm just looking 3 at two of these three where you have two trees under each 4 one, which is going to be -- not to put all the work on 5 the subcommittees, but all of the work's on the 6 subcommittees, you have two trees on each one of these, 7 it's going to have to approach them slightly differently. 8 MR. HARRIS: I was going to suggest the same 9 thing that Matt and Tom are driving at, that retail is 10 different, and in particular, the -- there ought to be a 11 -- perhaps another subcommittee, not necessarily named 12 retail but perhaps named access. So we've been talking 13 about transparency, but access is a different issue. And 14 so being able to ensure that people can access a fixed 15 quote anywhere or something like that, so in a potential 16 interest of the Commission. We've certainly done it in 17 equities. And even if -- even if we don't require that 18 people access it, we might easily imagine creating an 19 environment where people will choose to access it if they 20 can, but we have to make sure that they can. 21 CHAIRMAN HEANEY: Thank you. All right. So let me 22 just quickly -- just for -- to kind of fill in some of 23 the gaps on the subcommittees and how this will work or 24 could work, and again, Brett and I will come back to this 25 group, but there's Commission staff on-hand to help in 0211 1 some of the work and some of the analysis from -- with 2 the subcommittees. 3 The subcommittees can go out and invite and 4 solicit experts to come in, so it doesn't just have to 5 be, for simplicity's sake, if there's a subcommittee of 6 10 people, the 10 people getting around a table and 7 trying to do this on their own. You can bring other 8 experts in to do it, similar to how we did the panel 9 today. 10 The idea is to keep the subcommittees around 11 the same numbers. We're certainly not going to force 12 everybody on FIMSAC initially to join one. Although, 13 with three topics, it would seem that there's probably 14 one that is applicable. Given everybody's different 15 backgrounds here, it would be better to do it that way. 16 You know, again, I don't want to overstate, but the work 17 of the subcommittees, and I'm -- we're not pushing this 18 to that group and we're sitting back. Brett and I will 19 be involved in every one of the subcommittees. So that - 20 - this is where a lot of the kind of roll your sleeves up 21 recommendations should come from. 22 So with that out there, we'll take into account 23 a lot of what has been said as to kind of the granularity 24 of each of one of these as topics. We'll go back out and 25 solicit -- either Tom or David will solicit where you 0212 1 have interest to join on which subcommittee, which 2 obviously, ideally, we're going to try to do our best to 3 match that because that'll be the most productive way 4 going forward with this. 5 And then the idea is to come back in April, 6 after these meetings have occurred; you've got the work 7 of the subcommittees, and again, we're kind of flushing 8 this out, but you could see that it's a third, a third, a 9 third the way the day would be split with the 10 subcommittee findings into this public meeting. So 11 you've got the findings of A; you've got the findings of 12 Subcommittee B; you've got the findings of Subcommittee C 13 with some time at the end of us to then kind of debate 14 and discuss it. 15 So just to kind of put a little color on the -- 16 on the painting as how this should -- this could and this 17 should and this probably does come together in April. So 18 let me just stop to see if there's any questions. 19 Anything, Brett, you want to add? 20 MR. REDFEARN: You know, I would just echo 21 those sentiments and say that, you know, I think starting 22 with Rick's point, you know, we all -- we all get it, do 23 no harm. We understand the efficiency and the depth of 24 our market, and certainly, we're not looking to do things 25 that don't need to be done, but to the extent that there 0213 1 are concrete suggestions that emerge from these 2 subcommittees where we can be action-oriented and come in 3 and help improve our market for -- our markets for retail 4 investors and make it more efficient, then, you know, 5 it's -- we're ready, you know, to come up with concrete 6 ideas and to see if we can move forward. 7 So we want to make this well worth everybody's 8 time for doing this because we appreciate all of your 9 time very much, and we're enthusiastic about some great 10 ideas coming out of these groups. 11 MR. GARCIA: Does that mean you're going to let 12 us know what committee we're being assigned to and things 13 of that nature, or do you want to solicit interest, or how 14 do you want to proceed? 15 CHAIRMAN HEANEY: I think the first thing, and I'll 16 -- and I'll put the words in David's mouth, David will 17 reach out to people individually to see where the 18 interests lie. What we first have to do is huddle on the 19 subtopics, kind of really get a little bit more granular 20 on that. That'll be step one. Those will be sent out to 21 committee members. Step two will then be for David will 22 reach out and see where your interest may lie, one-on- 23 one, on email. And then we'll come back and that has to 24 be done pretty quickly because this will be time 25 intensive, and April will come pretty quickly. 0214 1 CHAIRMAN CLAYTON: Well, no, I'll let you guys 2 close. I don't know if Kara, Hester, do you have 3 anything? 4 COMMISSIONER PEIRCE:: I want to thank you all for 5 coming today and just wanted to underscore a point that 6 Commissioner Piwowar made this morning, which is that 7 it's really important for you to tell us what we need to 8 be thinking about. And so if there are issues that are 9 not being talked about at the SEC, I really encourage you 10 to bring those issues to our attention so that we can 11 think about them as soon as possible. Thanks. 12 COMMISSIONER STEIN: I'm just going to reiterate thank 13 you. You are going to be put to work, I think, during 14 this next quarter. And we, again, really appreciate what 15 Hester's saying, as you bring in your best thoughts to us 16 about what we should be focused on, right? I think even 17 what our Chair and Brett are saying is they're trying to 18 solicit feedback on what these three subcommittees should 19 be about. If you think after this meeting about things 20 you didn't think about during the meeting, get back to 21 them, you know, as they sort of try to think through how 22 to do this. I think you guys are also bringing up 23 repeatedly how different parts of the market might need a 24 different type of treatment. And I don't know if that 25 means different subcommittees or this idea of two 0215 1 subcommittees within each (laughter) subcommittee, but 2 thinking through how to do that, I think is very helpful. 3 One of the things I thought about, which is not 4 volunteering, but can people be on more than one subcommittee 5 if they were so inclined? 6 CHAIRMAN HEANEY: That's a lot of work. 7 MR. REDFEARN: I mean in -- if they're willing 8 to contribute at that level, we are, I think, probably 9 thrilled. 10 COMMISSIONER STEIN: Okay. 11 MR. REDFEARN: I know that I think in EMSAC, we 12 had each person could be on two of the committees that 13 were there, so -- 14 COMMISSIONER STEIN: I just don't want people to feel 15 limited. 16 MR. REDFEARN: Right. 17 COMMISSIONER STEIN: Okay, great. 18 MR. REDFEARN: I'm inclined to say they could 19 be on all of them, but (laughter) I thought we were the 20 only crazy ones doing that. 21 CHAIRMAN CLAYTON: Right. That's -- thank you. I 22 want to say something that Brett said reminded me that I 23 should say this, and one of the lucky things that we have 24 in this job is we interact with a lot of our counterparts 25 around the world, central bankers, markets regulators. 0216 1 And I've been doing a lot of that lately as a result of 2 MiFID and other assignments. And it gives me some 3 perspective in that the US markets -- and I mentioned 4 this at the beginning, the US markets, particularly the 5 US, you know, public capital markets and the debt 6 markets, are the envy of the world. 7 We finance more things through the capital 8 markets than anybody else, using whatever metric you 9 want. I really do believe having had a number of 10 conversations about how different economies came out of, 11 you know, the financial crisis and the recession that 12 followed that the nimble nature of our capital markets 13 contributed to a much quicker and hopefully stronger 14 recovery. And we shouldn't forget that. That's a 15 longwinded way of saying we don't want to do any harm. 16 We want -- we want to make it better, but we recognize 17 that, you know, doing harm is -- you know, I'd rather do 18 nothing than do something that harmed these markets. 19 So we really do need your good ideas. They're 20 going to evolve. The electronification of the markets is 21 going to change them. We have to react to that. It's an 22 opportunity to make them better, but, you know, like I 23 said, they're pretty good. We want to make them better, 24 but let's not do anything that we feel is harming them. 25 Thank you. 0217 1 CHAIRMAN HEANEY: Thank you all for your 2 participation, and this was the robust discussion we were 3 hoping for, so we really appreciate it. With that, I'll 4 entertain a motion to adjourn. All in favor? 5 COMMITTEE: Aye. 6 CHAIRMAN HEANEY: Have a great day and safe 7 travels. 8 (Whereupon, at 4:10 p.m., the meeting was 9 adjourned.) 10 * * * * * 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 0218 1 PROOFREADER'S CERTIFICATE 2 3 In the Matter of: FIXED INCOME MARKET STRUCTURE 4 ADVISORY COMMITTEE MEETING 5 File Number: OS-0118 6 Date: Thursday, January 11, 2018 7 Location: Washington, D.C. 8 9 This is to certify that I, Christine Boyce, (the 10 undersigned) do hereby swear and affirm that the attached 11 proceedings before the U.S. Securities and Exchange 12 Commission were held according to the record, and that 13 this is the original, complete, true and accurate 14 transcript, which has been compared with the reporting or 15 recording accomplished at the hearing. 16 17 18 ________________________________ __________________ 19 (Proofreader's Name) (Date) 20 21 22 23 24 25