Page 1 
THE UNITED STATES SECURITIES AND EXCHANGE COMMISSION 


MEETING OF THE 
FIXED INCOME MARKET STRUCTURE 
ADVISORY COMMITTEE 

Monday, November 4, 2019
9:00 a.m. 

U.S. Securities and Exchange Commission 
200 Vesey Street 

New York, New York 10281 
Multipurpose Room 


Page 2 Page 4 
1 PARTICIPANTS: 1 PARTICIPANTS(CONT.) 
2 2 
3 JOHN ROESER, Associate Director, Trading & Markets 3 RACHEL WILSON, S.V.P., Finance, Iron Mountain 
4 DAVID SHILLMAN, Associate Director, Trading & Markets 4 MIHIR WORAH, Chief Investment Officer Asset Allocation & 
5 LIZZIE BAIRD, Deputy Director, Trading & Markets 5 Real Return and Managing Director, PIMCO 
6 BRETT REDFEARN, Director, Trading & Markets 6 
7 MICHAEL HEANEY, Committee Chairman 7 
8 JAY CLAYTON, Chairman 8 
9 ROBERT JACKSON, Commissioner 9 
10 ELAD L. ROISMAN, Commissioner 10 
11 REBECCA OLSEN, Director, Office of Municipal Securities 11 
12 ABBY KIM, Financial Economist, DERA 12 
13 JESSICA KANE, Director, Office of Credit Ratings 13 
14 DAN ALLEN, President, Partner, and Senior Portfolio 14 
15 Manager, Anchorage Capital Group, LLC 15 
16 16 
17 MATT ANDRESEN, CEO and Co-founder, Headlands 17 
18 Technologies 18 
19 JOHN BAGLEY, Chief Market Structure Officer, Municipal 19 
20 Securities Rulemaking Board 20 
21 21 
22 GIEDRE BALL, Debt Program Manager, Metropolitan 22 
23 Washington Airports Authority 23 
24 HORACE CARTER, Head of Trading, Fixed-Income Capital 24 
25 Markets, Raymond James 25 
Page 3 Page 5 
1 PARTICIPANTS(CONT): 1 C O N T E N T S 
2 2 PAGE 
3 GILBERT GARCIA, Managing Partner, Garcia Hamilton & 3 Welcome and Opening Remarks 6 
4 Associates 4 
5 TOM GIRA, E.V.P., Market Regulation and Transparency 5 Presentation on Structured Disclosures by 21 
6 Services, FINRA 6 Municipal Issuers 
7 LARRY HARRIS, Fred V. Keenan Chair and Professor of 7 
8 Finance, USC Marshall School of Business 8 Presentation on Alternative Compensation Models 72 
9 SCOTT KROHN, Treasurer, Verizon 9 for Credit Rating Agencies 
10 ANANTH MADHAVEN, Managing Director, Global Head of 10 
11 Research for EIF and Index Investing Blackrock 11 Updates from the Technology and Electronic 146 
12 LYNN MARTIN, President and COO, ICE Data Services 12 Trading Subcommittee and the Corporate Bond 
13 AMY McGARRITY, Chief Investment Officer, Colorado Public 13 Transparency Subcommittee 
14 Employees' Retirement Association 14 
15 RICH McVEY, Chairman and CEO, MarketAxess 15 Presentation on Fixed-Income Construction 151 
16 LEE OLESKY, CEO, Tradeweb Markets 16 
17 SUZANNE SHANK, Chair, CEO and Co-Founder, Siebert 17 Presentation on Interdealer Government 200 
18 Cisneros Shank & Co. LLC 18 Securities Trading Platforms 
19 SONALI THEISEN, Head of Fixed-Income Market Structure, 19 
20 Bank of America, Merrill Lynch 20 LIBOR Transition Update and SOFR Volatility 265 
21 LARRY TABB, Founder and Research Chairman, TABB Group 21 
22 Kuman Venkataraman, Collins Chair and Professor of 22 Adjournment 312 
23 Finance, SMJ Cox School of Business 23 
24 ELISSE WALTER, Former Chair, SEC 24 
25 25 

2 (Pages 2 to 5) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 6 
P R O C E E D I N G S 
CHAIRMAN HEANEY: Good morning, all. 
I believe we have a forum. So, I'll call the 
meeting to order. 
I thank you all for joining us today 
at the SEC fixed-income Market Structured Advisory 
Committee here in the SEC's Regional Office in New 
York. I also want to welcome Giedre Ball from the 
Metropolitan Washington Airports Authority and Lee 
Olesky from Tradeweb Markets for their first FIMSAC 
meeting as new committee members and we look forward 
to working with you in the future. 
I'd also like to say good morning to 
Larry Tabb. He may be joining in a little while. 
I'll begin by welcoming Chairman 
Clayton and ask him to make his opening statement. 
MR. CLAYTON: Thank you, Michael. 
Good morning, everyone. Thank you 
for being here. Welcome to New York. And, Giedre, 
welcome. It's funny that you're from Washington and 
our first meeting we have in New York. 
Today's agenda is full of importance. 
We've assembled expert panels on; One, structured 
disclosures by municipal issuers; Two, rating agency 
compensation models; Three, index construction; 
1 
2 
34 
5 
6 
7 
8 
9 
10 
11 
1213 
14 
15 
16 
1718 
19 
20 
21 
22 
23 
24 
25 
Page 8 
equivalent of roughly 45 percent of the market cap 
of our U.S. public companies. 
Recent growth in these fixed-income 
markets have been significant and reflects a shift 
in our credit markets to debt held by banks, the 
debt held outside of banks, including in funds. Non 
financial corporate debt is at its highest level in 
history. Approximately, 47 percent of new debts of 
U.S. GDP up from about 40 percent in 2010. 
Municipal debt issuance had record years in 2016 and 
2017. 
So, in short, our fixed-income 
capital markets have at least most problems as they 
ever have had in our markets. That's domestic and 
internationally. This is really important stuff and 
it's only getting increasingly so. 
So now, three items on my mind that I 
hope you'll consider with the panelists today. 
First, index construction. Let me ask you, do 
investors and those who advise investors understand 
how indices are constructed. First from an 
technical perspective, weightings, adjustments and 
the like. Second, from a market exposure 
perspective, what opportunities and risks do these 
indices incorporate. And, third, as a subset of 
Page 7 Page 9 
1 Four, government securities trading platforms; and, 1 those risks, and what are the key value judgments 
2 Five, the LIBOR transition. We'll also hear updates 2 the index provider has made, to include or exclude 
3 from our Technology and Electronic Trading 3 certain type of companies and other values. 
4 Subcommittee and the Corporate Bond Transparency 4 I have some concerns in this regard 
5 Subcommittee. 5 and I've had many discussions with investors where 
6 I want to tell you all that I could 6 they expressed their concerns about particular risks 
7 not be more pleased with the work of the committee 7 and choices. Yet, it appears those concerns do not 
8 over the last few years. You've brought diversity 8 apply or are not well understood when they invest 
9 of expertise and perspective which has enhanced our 9 through something. Also, with the proliferation of 
10 understanding of the fixed-income markets and I 10 passive investing of fixations, what happens when 
11 assure you informed policy positions. 11 markets turn downward instead of upward. Two 
12 Now, I want to make a broad comment 12 questions on that. 
13 which again demonstrates the importance of your work 13 Second, with regard to rating agency 
14 and then comment briefly on three of the agenda 14 compensation models, this is not a new issue, but it 
15 items. 15 is an important issue and, as always, the landscape 
16 The fixed-income capital markets are 16 is changing. We need to continually review whether 
17 playing an increasingly important role in our 17 market participants who are substantially influenced 
18 economy on both an absolute and relative basis. 18 that are replied upon by investors are appropriately 
19 This should not surprise us. It is the logical and 19 disclosing, monitoring and managing their conflicts. 
20 predictable result of monetary and regulatory policy 20 While they play different roles and have different 
21 decisions. 21 compensation models and are regulated differently, 
22 A few metrics. In the United States 22 we need to continually monitor the activities of 
23 outstanding non financial corporate debt stands at 23 rating agencies, securities analysts, investment 
24 almost $10 trillion. Municipal debt is almost four 24 advisors, proxy advisory businesses, brokers, 
25 trillion. Together, these markets are the 25 particularly retail brokers, and accountants. None 

3 (Pages 6 to 9) 


1 
2 
3 
4 
56 
7 
8 
9 
1011 
12 
13 
14 
15 
16 
17 
1819 
20 
21 
22 
23 
2425 
Page 10 
of their interests are fully in line with -- fully 
aligned with the interests of investors and our 
regulation and oversight. And, in particular, our 
disclosure needs to recognize this duality. 
With regard to credit agencies a 
broad question, and I recognize that this is not a 
new question, is, are there alternative compensation 
models that will better align the incentives of 
rating agencies to those of investors. 
Finally, LIBOR. As I have noted for 
some time now, depending on the position presents 
risks that market participants working with central 
banks and regulatory authorities need to address. I 
applaud the work to date, including identifying SOFR 
as a potential replacement for LIBOR, as well as, 
efforts to include transition language in loans, 
bonds and products that use LIBOR as a benchmark. 
That said much, much, much, or as my 
son would say, four months, dad, more work needs to 
be done for the transition to avoid substantial 
frictions. Including frictions that will harm our 
investors directly through higher costs and 
indirectly as a result of uncertain agendas. 
With this context I'll ask today's 
panel to comment on a broad issue that has been 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 12 
today's panel to help me with that challenge. 
With those remarks, thank you very 
much. Michael, back to you. 
MR. HEANEY: Thank you, Chairman 
Clayton. 
Also joining us on the phone today is 
Commissioner Jackson. 
Commissioner, would you like to make 
your opening remarks? 
COMMISSIONER JACKSON: I would. 
Thank you very much for the 
opportunity. I'm --I'm terribly sorry to be unable 
to join you today, but I appreciate very much the 
chance to --to share a few opening thoughts. And I 
want to begin by saying how strongly I agree with 
everything the chairman's just said, but in 
particular, his gratitude for all of you for joining 
us at today's meeting and for doing this very 
important work. 
You know, since my very first day on 
the SEC, my first day on the job, my first task on 
the job was to attend my very first FIMSAC meeting, 
I have been consistently impressed with the 
exceptional work of this committee. I'm greatful to 
each one of you for your service to the Commission 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 11 
currently in my mind. Now, I'm going to speak very 
generally. There are people here that are much more 
expert about this than I am, but I want to learn 
more on this issue. And I know it's more complex in 
practice. LIBOR-based securities and products 
generally reflect recent markets; a risk free rate, 
a bank funding spread over the risk free rate and an 
additional fixed spread to or from the lender to the 
borrower or customer. The current LIBOR benchmark 
incorporates both the first and second component, 
the risk free rate and the bank funding spread rate 
over that risk free rate. 
Although each of these components 
fluctuates with market conditions, they are combined 
when we say the word LIBOR. The SOFR benchmark in 
contrast incorporates just the first component, the 
fluctuating risk free rate, but not the second 
component, the fluctuating bank funding spread over 
the risk free rate. As a result a loan, bond or 
product that is comprised of the SOFR rate and an 
additional fixed spread would not be expected to 
fully incorporate that floating bank rate spread, 
that second component. This difference in approach 
appears to me to make like-for-like mapping from a 
LIBOR product to a SOFR product challenging. I ask 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 13 
and the nation's investors. And as we meet for the 
last time this year, I look forward to hearing from 
the committee on today's discussions, especially 
about credit ratings and financial benchmarks. 
I also wanted to share two thoughts 
of my own as the committee continues its important 
work next year and, once again, express my 
appreciation to Chairman Clayton for his leadership 
in extending the committee's work for next year to 
address these, among other, important issues. 
First, I want to highlight your work 
on an issue you raised in your previous meeting, the 
potentially abusive use of last looks or pennying in 
bond auctions. Now, you all know far better than I 
that auctions play an important role in centralizing 
liquidity. And, of course, a last look can be a way 
for dealers to avoid adverse selection, that is, 
losing money on bad trades. 
But we all know there's no free lunch 
in finance. If dealers are systematically abusing 
last looks to obtain free price discovery, well, 
then that comes at the cost of fair competition and 
market efficiency. As in other markets where we've 
worked on similar problems I wanted to call on my 
colleagues at FINRA and the SEC to consider 

4 (Pages 10 to 13) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 14 
collecting data from trading venues to study this 
issue and, if they contend this is warranting, 
consider rule making in the area. 
Second, I want to encourage 
thoughtful discussion about one of today's agenda 
items, alternative compensation arrangements for 
credit rating agencies. We all know that incentives 
play a powerful role in shaping our markets and I 
for one am concerned that credit rating agencies 
lack the necessary incentives to produce reliable, 
unbiased ratings for systematically important 
markets. 
Now, as the chairman explained, I 
thought rather eloquently, these are complicated 
issues, and you know far more about it than I do, 
but it is notable that despite legislative and 
regulatory attempts to introduce transparency, 
oversight and competition to these markets, we're 
now --we now risk repeating the tragic history of 
rating shopping. I very much look forward to 
hearing from the committee on alternative ways we 
can think about this problem, alternative frameworks 
we can bring to the credit rating process and how we 
can produce the transparent and reliable ratings 
that American investors deserve. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 16 
Trading and Markets is Lizzie Baird, one of our -one 
of our deputy directors. To Lizzie's right are 
Dave Shillman and John Roeser, Associate Directors 
in the Office of Market Supervision in Trading and 
Markets. Down on my left are Rebecca Olsen, 
Director of the Office of Municipal Securities, Abby 
Kim from the Division of Economic and Risk Analysis, 
and Jessica Kane, Director of the Office of Credit 
Ratings. 
I'd also like to thank Mark Berger 
and his team here in the New York Regional Office 
for being such gracious hosts for us here at --at 
NYRO. 
Before we get started, I'd like to 
remind you all that the views expressed here are 
just those of the speaker and do not necessarily 
reflect those of the Commission, the staff or any 
other commissioners. 
With that, as we approach this 
two-year anniversary of the committee I want to just 
start by thanking all of the FIMSAC members for the 
hard work and dedication that you've all brought to 
the discussions that we've had. It really has been 
impressive and extremely productive, the number of 
recommendations and the number of thoughtful 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 15 
Thanks again to each one of you for 
your continued engagement throughout this year. 
Look, we know how very busy every professional in 
that room is and we are very grateful for the time 
that you spend to provide us with what I assure you 
is crucial advice on how to move these markets 
forward. 
I very much look forward to today's 
conversations and thank you for the opportunity to 
be apart of them. 
CHAIRMAN HEANEY: Thank you, 
Commissioner Jackson. 
Commissioner Purse and Commissioner 
Elad Roisman will be joining us later in the day and 
we look forward to their participation. 
Next I'll turn it over to Brett 
Redfearn, Director of the Division of Trading and 
Markets and the Committee's designated federal 
officer for opening comments. 
MR. REDFEARN: Thank you, Michael. 
And I'd like to welcome everybody here today to our 
FIMSAC meeting. Thank you. 
Let me briefly introduce my 
colleagues here. 
So, to my right, from the Division of 
1 
23 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
1617 
18 
1920 
21 
22 
23 
24 
25 
Page 17 
meetings that we've had here today. 
Over the past two years we've held 
eight public meetings, made ten recommendations to 
the Commission. These recommendations range from 
improving transparency and promoting liquidity for 
institutional and retail investors in the corporate 
bond market to rethinking the way we regulate 
electronic trading and promoting better investor 
education concerning fixed-income products, among 
other things. And it's clear as evidenced by 
today's agenda that you continue to tackle important 
challenges that have no easy answer. Indeed, 
today's agenda may reflect the most diverse and 
numerous set of issues that this committee has 
considered at any meeting thusfar. 
I thank you all for your continued 
willingness to take on such challenging issues and 
to candidly share your insights with us. 
I also want to take a moment to 
comment on the importance of public engagement and 
to commend this committee on its level of 
inclusiveness. Throughout your tenure you've 
recognized that market participants beyond those 
recognized -- represented on the FIMSAC have 
important insights on a range of topics you access 

5 (Pages 14 to 17) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 18 
and that their views inform your own collective 
thoughts. 
We can see that engagement -engagement 
can inform the many subcommittee meetings 
and we see this in the subcommittee meeting minutes 
reflecting participation by non-FIMSAC members and 
from the publicly accessible panels such as today's 
that you host at your quarterly meetings. And a big 
thank you to our panelists today also for their 
participation. 
As always, interested parties may 
continue to submit comments on the work of the 
committee via the FIMSAC website page and the 
committee's deliberations and topics under 
discussions, preliminary recommendations and all 
recommendations approved by the committee are made 
available to the public on the FIMSAC's web page. 
I look forward to today's discussions 
and thank you again to all the FIMSAC members and 
today's panelists for devoting your time to this 
committee's important work. 
Back to you, Michael. 
CHAIRMAN HEANEY: Thank you, Brett. 
So, we have a full day, as has been 
mentioned, a five-panel discussions. This morning 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 20 
past two years and FIMSAC has made great strides in 
adopting measures which have helped to improve 
market liquidity through our recommendations in 
large part, but also through the outside panel work. 
I look forward to our third and final 
year as we continue down this path of exploring 
potential new recommendations and monitoring current 
ones. And again, I'd like to also thank our 
panelists for their participation today. 
With that, let's dive right into our 
first panel, but before we begin, two housekeeping 
items. I will reiterate, as I always do, please 
with your name tags if you could put it up so we can 
hear from all FIMSAC members. And, secondly, I'm 
being asked to ask the panelists and FIMSAC members 
to leave the volume button unchanged. It will be 
moderated, I guess, in the back with just on and 
off, but no volume adjustments, please. 
So, with that, our first panel will 
be discussing structured formats to provide 
disclosure in municipal markets. Let me introduce 
our moderator, Lynn Martin, who will be moderating 
this panel as Chair of FIMSAC's Municipal Securities 
Transparencies Subcommittee. 
Lynn, I'll turn it over to you. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 19 
the Municipal Transparency Subcommitte will host a 
panel regarding the utility of structured disclosure 
by municipal issuers and the Credit --Credit Rating 
Subcommittee will host a panel focusing on a range 
of alternative compensation models that have been 
discussed in the past concerning credit ratings. 
You'll also hear an update from Rick 
McVey and Mihir Worah on the ongoing work of 
technology in e-trading and corporate bond 
transparency subcommittees respectively. After 
lunch the ETF and Bond Fund Subcommittee will host a 
panel on the role of fixed-income construction and 
Brett will be moderating a panel concerning 
inter-dealer government securities market. 
We'll end the day with a panel 
moderated by Lizzie Baird on the status of the 
current LIBOR transition project as Chairman -excuse 
me, as Chairman Clayton eluded to, and a 
discussion of the recent volatility within the repo 
market. 
Today's agenda is a great example of 
the wide range of issues we have tackled over the 
last two years. I want to join Brett in thanking 
you all for your active engagement on this 
committee. I will say, this has been an exciting 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 21 
MS. MARTIN: Thank you, Michael. 
Chair Clayton, Commissioner Jackson, 
SEC staff and fellow FIMSAC members, thank you for 
giving us the opportunity to talk to you about this 
topic today. 
I'd like to thank my esteemed 
panelists for joining us today. On the panel we 
have Duffy Blackburn, an auditor from Will County, 
Illinois, Lisa Washburn, Managing Director from MMA, 
Campbell Pryde, President and CEO of XBRL US, Mark 
Kim, Executive Vice President and COO of MSRB, and 
Michael Willis, Assistant Director, Office of 
Structured Disclosure from DORA at the SEC. Thank 
you all. 
I'm looking forward to a wide 
discussion on the topic of structured disclosures in 
the muni market. This topic has been raised at 
multiple FIMSAC subcommittees actually. Not just at 
the Municipal Transparency Subcommitte, but also 
more recently as part of the Credit Rating 
Subcommittee work. 
The way we think of this issue is 
really whether the implementation of technology, or 
some form of standardization of reporting, could 
help with the issue of timeliness of disclosures. 

6 (Pages 18 to 21) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 22 
Particularly the financial disclosures which we've 
discussed at various FIMSAC meetings. 
We've seen the positive impact that 
standardization of protocols and common technology 
has had on the corporate bond market with the 
implementation of the XBRL protocol performed easily 
accessible and queriable database within EDGAR for 
the corporate side of the business. And one of the 
things we're wondering is whether a similar standard 
could be applied to muni markets. Again, with the 
idea of standardization and increased transparency. 
With that, I'd like to proceed with 
today's discussion around the topic. 
My first question's going to be 
directed towards Duffy and Lisa. For a typical muni 
bond offering, what is the nature of the basic 
financial disclosures made available to the market, 
particularly on an ongoing basis? 
DIRECTOR REDFEARN: Lynn, might I 
just ask for just one second, I know a lot of the 
microphones were on when we came in here, but I 
would just suggest that if you're not speaking to 
just hit the green light and turn the microphones 
off for those of you not speaking. Thank you. The 
feedback down there can be bad. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 24 
provided is the audited financial statements well 
after the close of the fiscal year. And sometimes 
audited financial statements don't even come out 
every year. 
Sectors that typically have a 
narrower investor base or those that operate in a 
more profit-like environment tend to provide more 
regular financial information. So, for example, 
hospitals as a sector generally provide much more 
regular interim financial information to the 
public --investment public. 
Do you want me to go over to the 
critiques or -MS. 
MARTIN: Let's --let's move 
to --to Duffy for his prospective as well. And 
then --and then maybe the two of you can go back 
and forth between various critiques you may have. 
MR. BLACKBURN: Yeah. You have the 
current system or the current system of this basic 
element of disclosure, has it run its course? And 
at Will County we've issued an XBRL report. And how 
have we done this and how would anybody else do 
this. And the basic question I usually get from -at 
the local level, after I wipe away the --the 
novelty of XBRL or space-age benefits of structured 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 23 
MS. WASHBURN: Thank you all. And 
thank you for having me here today. 
In terms of market disclosure, the 
tolerables, quality, frequency and accessibility 
to --to information varies across sectors and 
issuers and information is typically added to YES at 
the time a bond is issued. The official statement 
typically provides a copy of the most recent audit 
and other updated financial and important 
nonfinancial information. Issuers that have missed 
filings or that are slow to file will typically 
catch up at this point, but after issuance the 
information flow is generally less robust and does 
run the gamut from thorough to lacking. 
There are some issuers that provide 
the full compliment of ongoing financial 
information. So, things like budget comparison, 
cash flows, interim financials, operating 
statistics. For example, the Commonwealth of 
Massachusetts has a website linked to EMMA that 
provides comprehensive monthly financial 
information, but there are many governments, 
particularly those that are infrequent issuers or 
those that are smaller in size for which disclosure 
is only provided --the only disclosure that's 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 25 
data, is usually, aren't you already doing this or 
why aren't we already doing this. 
And when we compile our CAFR, our 
basic financial report, after six months we have a 
30-day extension for the CAFR, the GFOA, 210 days 
from EMMA to put in the --the official statement. 
We do have 120 days to get in our state disclosure, 
the AFR. Most of the timeliness critiques do 
definitely pop up. 
And --so, just being here is -thank 
you for inviting me to be here. Thank you, 
Lynn. Thank you, FIMSAC. 
I've actually followed you and many 
of your organizations through the development and 
work on XBRL. And --and --so, I have had a 
long-term goal of producing an XBRL statement at the 
local level. So, I appreciate your invitation to 
speak. And hopefully we can get into some of the 
critiques, and again, have a lively discussion on if 
this is doable and feasible for the market. 
MS. MARTIN: So, Duffy, I'm going to 
come back to you to go --delve more into the XBRL 
and your experience with the XBRL implementation and 
use and things of that nature, but first I want to 
go to Mark. 

7 (Pages 22 to 25) 


Page 26 Page 28 
1 Mark, how are these disclosures, 1 Then finally, turning to continuing 
2 XBRL, not XBRL, displayed on the EMMA system and 2 disclosures, which is really I think the subject of 
3 distributed to --to customers on the EMMA system. 3 today's conversation, these continuing disclosures 
4 MR. KIM: Sure. And thank you for 4 are made under the Anti-Fraud Provisions of the 
5 the opportunity to participate on the panel. 5 Federal Securities Law. And in particular through 
6 I thought it would be helpful to 6 SEC Rule 15c2-12. At a very high level, this rule 
7 maybe take a step back and talk about the universe 7 requires underwriters to take reasonable diligence 
8 of disclosures that the MSRB makes transparent to 8 to insure that issuers have entered into what's 
9 the market. 9 known as a continuing disclosure agreement, or CDA, 
10 We do serve as the sole designated 10 with the holders of its bonds. The actual 
11 repository of municipal market disclosures. And 11 provisions of the CDA vary from issuer to issuer. 
12 really, at the highest level I would say that there 12 Neither the SEC nor the MSRB has the 
13 are three different types of disclosure information 13 regulatory authority to dictate or to mandate what 
14 that we make transparent to the market. The first 14 the timing of those disclosures or the content of 
15 is primary market disclosures, the second would be 15 those disclosures must be. However, generally 
16 secondary market disclosures and then the third 16 speaking, the CDAs cover two different types of 
17 would be continuing disclosures. And it's really 17 continuing disclosures. The first is material event 
18 that last type of disclosures that's the focus of 18 notices. And then, the second are annual financial 
19 this panel, but I think it would be helpful just as 19 information or operating data. 
20 context and background to just touch very briefly on 20 So, very briefly, the material event 
21 the first two as well. 21 notices touch on credit worthiness of an issuer. 
22 Primary market disclosures are made 22 The SEC has identified 16 material events that must 
23 pursuant to MSRB Rule G32. This rule requires 23 be disclosed. These events cover issues such as the 
24 underwriters in primary offerings to submit a final 24 false delinquencies, bankruptcies and most recently, 
25 official statement, which is really the analogous to 25 any material financial obligation undertaken by the 
Page 27 Page 29 
1 the prospectus in the corporate markets, to the MSRB 1 issuer such as a bank loan. 
2 within seven business days of the underwriting of 2 These material event notices must be 
3 the securities. The MSRB then makes this disclosure 3 filed within ten business days to the MSRB. And the 
4 data available to the public at no cost through its 4 MSRB then makes that --those event notices 
5 EMMA website. This data -- primary market 5 transparent to the public for free at no cost. 
6 disclosure data comes to the MSRB in largely 6 Those event notices come to the MSRB as unstructured 
7 unstructured format. 7 data, typically in PDF format. 
8 Secondary market disclosures are made 8 And then finally, and this is really 
9 pursuant to MSRB Rule G14 which covers trade 9 I think the area where XBRL conversations are having 
10 reporting. Under this rule dealers are required to 10 is the second type of continuing disclosure that 
11 report any trade in a municipal security to the MSRB 11 issuers provide to the market are annual financial 
12 within 15 minutes of the time of trade. The MSRB 12 information or operating --operating data which 
13 then makes that trade data transparent to the 13 really speaks to the financial condition --the 
14 market, again, through EMMA to the public at no 14 ongoing financial condition of the issuer. 
15 cost. 15 I would note that issuers are not 
16 It typically takes the MSRB on a 16 required to audit their financials. Most do, but 
17 normal trading day -- last Friday for instance, it 17 they're not required to do so. And for the issuers 
18 took the MSRB approximately 14 seconds to do 18 that do audit their financials, they're also not 
19 end-to-end processing of that trade information. 19 required to prepare their financial statements in 
20 Most dealers submit the trade information to the 20 accordance with GAAP, or Generally Accepted 
21 MSRB sooner than the 15-minute time requirement. 21 Accounting Principles. So, there are non GAAP 
22 So, in terms of timeliness, that secondary market 22 financials that are presented to the market and 
23 trade information is typically going to the market 23 unaudited financials that are presented to the 
24 within 15 minutes of the time of trade. That data 24 market. 
25 is fully structured. 25 It is importantly up to the 

8 (Pages 26 to 29) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
1011 
1213 
14 
15 
16 
1718 
1920 
21 
22 
23 
24 
25 
Page 30 
individual issuer to establish the time frame in 
which it will present its annual financial 
information or operating data to the market. Again, 
neither the SEC nor the MSRB has the regulatory 
authority to determine what that time frame or what 
the content of those financial disclosures are. 
When we receive the annual financial information or 
operating data it comes to the MSRB in unstructured 
format. 
So, hopefully that gives an overview 
of -MS. 
MARTIN: Yeah, and I think it 
very eloquently points out some of the challenges, 
particularly when it comes to financial disclosures, 
which is what the -- the market appears to be most 
acutely aware of. 
I'm going to pivot now to Campbell 
and Michael. 
As I said, in my opening remarks, and 
as you've heard from the panelists so far, EDGAR and 
the corporate bond market has been sort of a model 
that has been used leveraging technology, allowing 
customers to have the data in queriable format, 
improving disclosure information. Could either of 
you comment on what the process was to implement 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 32 
number in the paper financial it changes it in the 
underlying financial format. 
So, that --that --when that data 
hits the SEC there are a number of people who are, 
basically, reading an SEC feed. A large number of 
hedge funds are pulling that data in and --and, 
basically, as they're pulling that data in -typically, 
today most of that is being done by large 
institutional investors. It probably represents 
we've estimated about 70 percent of assets under 
management --under management of the funds are 
pulling that data in. 
We, for example, we take that data -we 
make that --pull that data into databases that 
we then distribute to people who are interested in 
looking at the data, make the data available for 
free for people as well. And so, we get --you 
know, we're roughly getting a million hits a month I 
think it is on the data. And that's just people who 
are just generally interested in understanding that 
information. Most of those companies --most of 
those people either do it for their own personal 
investment choices or then you've got mid market 
players who want to start then distributing data as 
well. Sort of, small start ups. So, the cost for 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 31 
XBRL in that market. And, more specifically, talk 
about how customers actually use the data today, 
what you've observed since the transition of the 
corporate bond market to this format. 
MR. PRYDE: I'll let Mike cover the 
process of implementing it. And I'll --and I'll 
just talk about just in terms of how people using 
that data today. 
So, just if you're not familiar with 
how the process works, basically, all public 
companies report to the Commission in a electronic 
format and the report in an XBRL format. That 
includes today the financial statement information, 
also footnote information. It does not include NDA 
disclosures. That information covers both quarterly 
and annual financial statement information. 
That --that information --once the 
company files with the SEC, they've got to file both 
their XBRL filing and their internal financial 
statement filing. However, in the last year that 
has changed and that's a phase in process at the 
moment. They are filing something called in line 
XBRL which basically means, it's just one document. 
You just have a --a filing that comes in, the 
number that's reported is the same. You change the 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 33 
someone into that marketplace is very, very small 
and makes it much, much easier. 
And most of the benefit there is, you 
know, traditionally people would read --you know, 
get a paper financial, they've got to read through 
that data. They've got to pull it. Most of it -as 
you know, most of these documents are similar to 
the CAFR and to the extent they're very large 
documents. They over hundreds of pages long. It's 
very hard for people to read through those. 
So, one of the things that's 
happening in the marketplace is people can just 
extract the data they want from those reports. They 
can bring it instantly into practical 
representations. And so, you know, one of the 
things that, you know, we're trying to promote and a 
lot of other people in the marketplace are promoting 
is really making that data accessible to investors 
through seeing it looking on a smart phone, looking 
on a website, see it in a graphic representation. 
I think that it's unrealistic today 
to expect anyone who's going to be investing in the 
future, certainly under 40, to be looking at pulling 
these documents down and reading them. That they 
can get that information on a smart phone just 

9 (Pages 30 to 33) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 34 
makes --it just makes this more where it gets that 
information out. And having that data in an 
electronic format, you have to do that. I don't 
think, you know, distributing these massive 
documents to people is not a smart way to distribute 
information to investors. 
So, I'll leave it to Mike, you can 
talk about the process. 
MR. WILLIS: Sure. I'll --I'll 
cover two processes. One we see from reporting and 
also how we implemented it initially. 
So, the evidence sort of is sitting 
right in front of us with this machine readable tag 
here. So, I use the bar code as the analogy in that 
it helps information be delivered in a much more 
timely manner, at a lower cost, it gives the 
consumers much more diversity in products and they 
have a lower cost. So, it actually works for the 
entire supply chain of the standardization play. 
So, there's no winner or losers. There's actually a 
way to make the entire supply chain work more 
efficiently. Any of us that have been in grocery 
store, we see that today in spades, but it didn't 
happen overnight. 
So, the way the SEC program was 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 36 
decision making process. 
A large Fortune 500 company did that. 
They wrote an article on it called, Driving Fashion 
Decisions which sort of makes the point. That 
appeared in the Journal of Accountancy. 
In terms of how the data gets used 
from an analyst, it is more timely depending on how 
the process worked, but it also allows them to 
identify company-specific disclosures. So, if it's 
not in the taxonomy, they create an extension which 
is a company-specific idea which actually turns out 
to be very useful for an analyst to be able to 
delineate what's unique about this company. That's 
very useful. 
It also can identify disclosures by 
topic. So, rather than looking for something by 
words, we can now look by concept which is a very 
powerful idea, almost a magic trick. The words may 
not appear in the report, but the topic is there. 
So now we can find them. 
Additionally, it allows you to --to 
really dig into the report and find things 
regardless of where they're located. It's --on an 
individual report you can do that, but you can also 
do that in the aggregate. Now, a lot of 
1 
2 
3 
4 
5 
6 
7 
8 
9 
1011 
12 
13 
14 
15 
16 
17 
1819 
20 
21 
22 
23 
24 
25 
Page 35 
implemented is there was a voluntary program for a 
number of years, three or four years, and then, the 
taxonomy was actually developed by the expert, Mr. 
Campbell, and his team. That got transitioned over 
to FASB. They maintain that taxonomy today. It 
goes through a public review process and it's 
approved by the Commission on an annual basis. And 
there's other taxonomies that are used by 
registrants for publishing. 
If we go back to the bar code for a 
second. Back when I had dark hair, that was 
implemented and it was implemented by inventory 
clerks in the grocery stores putting the bar codes 
on the products when they stocked the shelves. So, 
we quickly see that added cost, but they figured it 
out, they pushed it back to the producers, the 
manufacturers. 
What we see in the corporate space is 
sort of the same thing. They finish their financial 
statements and then they put the tags on it. So, 
it's sort of like how the bar code was first 
implemented. There are some companies who are 
beginning to push that back against their operating 
systems which really generates a lot of internal 
transparency which kind of flows into their own 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 37 
municipalities are very unique in how they operate. 
And so, that's where you kind of want to dig in. 
This doesn't change the model. Just 
like the bar code didn't change my grandmother's 
recipe on pecan pie, but it did definitely lower the 
cost for getting those ingredients. 
The --the recipe is equivalent to 
the analyst's model. So, this just allows them to 
be delivered and the analyst still does their thing. 
And I think that's the key process as a way to make 
the information more efficient, lower cost and more 
specific and granular so that the analyst can 
actually determine what's standardized and what 
might be entity specific. 
MS. MARTIN: One of the --one of the 
challenges that folks have pointed out because this 
standard has been proposed in the muni markets is 
administrative burden and additional costs that 
could be imposed on the industry. Do you have a 
sense as to how that was received, at least on the 
corporate side, or at least the magnitude? Either 
Mike or Campbell. 
And then, Lisa, I'm going to pivot to 
you to ask, you know, what are the drawbacks of 
using some form of standard. 

10 (Pages 34 to 37) 


12 
3 
4 
5 
6 
7 
8 
9 
1011 
12 
13 
14 
15 
16 
17 
1819 
20 
21 
22 
2324 
25 
Page 38 
MR. WILLIS: So, in the initial 
movement, the estimate, it was around $25,000.00 as 
the estimate. Recent surveys of the smaller 
companies have indicated that number is closer to 
2,500. So, that's a ten-fold decrease. As I 
mentioned earlier, we're still seeing many 
companies, just like the grocery stores, it's sort 
of an after fact. I think as the supply chain 
matures we may see the flip side of that. 
So, just as an example, in my prior 
life I was able to architect a -- an audit platform 
that actually used XBRL from the ledgers and sub 
ledgers across an entire sector. Within two years 
we applied that to 5,000 clients and we were able to 
automate over 40 percent of the audit procedures. 
That's not about cost increase. That's about cost 
decrease. 
I think it's more than just report 
preparation. That actually begins, just like the 
bar code, have a very positive cost impact on 
everybody, but it depends on the implementation. 
That's the key point. 
MR. PRYDE: And Mike and I can talk 
about this a lot. This initiation with the SEC 
program we saw an approach of what we called a bulk 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 40 
In other markets such as --if you 
look in the markets like the UK, every single 
company has to follow an XBRL format. If you're a 
doctor or you're an accountant and you've got a 
small business you have to report on an XBRL --your 
XBRL financials to both the tax department, in the 
UK Customs House. They make that voluntary. You 
can file either in paper or an XBRL format. 
However, if you file paper-based format you've got 
to mail in your financial statements. And I think, 
like, 90 percent of participants choose to file it 
in XBRL format. And the reason -'
cause it's 
cheaper than the postage. The reason for that is 
because all of the accounting systems and everything 
else has been changed to, basically, support XBRL so 
that that data can be filed. So, really it's an 
infrastructure issue and infrastructure changes. 
And then you can get much, much more efficiency over 
the long term. 
I think where we're at today, most of 
this financial reporting got to a point where it's 
kind of a dead end. You --you can't get that cost 
down. It's a lot of work. You've got to take all 
of the data from the financial systems. You've got 
to pull that data down, get it in, put it in, format 
Page 39 Page 41 
1 point. So, what this means is, I've got all of my 1 it, get it into a paper format where it looks good 
2 data in an electronic format sitting in my sub 2 and then get it out. That's not necessarily what 
3 systems. I put it into paper and then I put it back 3 the market wants. And at the same time, this is 
4 into an electronic format again. I send it --then 4 kind of a dead end in terms of getting efficiency in 
5 I send it out to the financial markets, all right. 5 process. 
6 So, today we have the same thing. 6 MR. BLACKBURN: If I can add -7 
The data is in electronic format, putting it into a 7 MS. MARTIN: Yeah. I actually was 
8 paper like CAFR, send that paper based format out to 8 going to go to you Duffy before I go to Lisa. 
9 the financial markets. They take it, they put it 9 So, Duffy, you --you did this 
10 back into the electronic format. So, they put it 10 without a mandate. You thought it was the right 
11 down, back up. 11 thing to do, as you mentioned in your --in your -12 
So, this bulk point approach where 12 in your opening remarks. Has how has the experience 
13 you're taking the paper and putting it into an 13 been moving your financial statements at least to 
14 electronic format, it's inherently not an efficient 14 this format? Have there been costs? What lead you 
15 thing to do. You should be taking the data straight 15 to make that decision? 
16 out of your financial systems. And there is a 16 MR. BLACKBURN: Yeah. Yeah. As we 
17 transitional period to do that. 17 believe, Will County is the --as we believe, is the 
18 So, you will see now with most market 18 first CAFR to be produced in a smaller format. Just 
19 participants and the SEC program they've moved 19 a few data points along the larger CAFR that we 
20 towards that. There's still people doing bulk point 20 produce. And as Campbell and Mike said, we actually 
21 approaches. A lot of providers in the marketplace 21 take our --as a side note to technology and local 
22 provide that very cheaply. I think the cheapest 22 government, we take Advantis 400 system, IBM, put 
23 I've seen is like $250.00 to do it, but that's for a 23 it --take it out with Sequel, put it into Excel. 
24 very, very small company and wouldn't apply to most 24 Then, manipulate that through the functions of -25 
large --large municipal security issuers. 25 just database functions with Excel. And then, take 

11 (Pages 38 to 41) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 42 
that and ship it into a 346-page document that is 
just --it's not being used as we understand from 
the market, from the people that it should impact, 
from the users of our financial data. 
And then, we then send it in PDF 
format to EMMA, to the --the investors. And they 
then also apparently don't appreciate that format as 
well. And it feels like it just seems glaring that 
local governments, local units, are doing a lot of 
work and wasting a lot of cost and burden to do so. 
If we're going to take just the 850 
different data points that we have tagged in our 
CAFR and put it into XBRL it seems that's a much 
better use of local units' time and effort. 
We went through a survey and a cost 
estimate with the Center for Governmental Studies at 
Northern Illinois University. So, they helped us 
just define and just, sort of, review what we had 
done. And --and we put together a CAFR that's 40 
to 80 hours worth of prep time and review time. And 
it's about five to $10,000.00 worth of cost to us. 
Where if we had been, sort of, you know, a smaller 
subset of data that has more impact and use at the 
beginning, and this is assuming that in lieu of an 
AFR that goes to our state and to GFOA and to the 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 44 
make available? Was that something that -MR. 
BLACKBURN: It was based on --it 
was based on right now --yes. This is then --this 
is to determine what's being tagged. Ultimately is 
with the people tasked with this for the national 
CAFR taxonomy. 
So, we started with just four basic 
financials. With GASBE there's two different 
financials. There's modified accrual. So, we did 
basic four financials, the statement of net 
position, revenues, the balance sheet and then, also 
pensions. And then --but if there is an issue, if 
analysis and data shows issues or reveals issues 
with assets, tagging information could be applied to 
some capital assets that we're investing based on 
type or age of those assets. But again, it's going 
back to whoever is tasked with putting together this 
taxonomy. 
MS. MARTIN: Lisa? 
MS. WASHBURN: Sure. So, nothing 
against structured data as a concept, right, but in 
a municipal market we have a lot of hurdles that -some 
could be maybe surmountable, but others that 
maybe not be that could limit the utility of 
structured data reporting for the investor community 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 43 
official statement that Mark Kim had brought up, 
it's 100 --it's 1,000 to $1,200.00. The savings 
is --it's clear that we would save money. If we 
are producing at a local --at a municipal unit 
something that actually has more impact that would 
be much better. 
As far as the three different --the 
three different reports that we actually have to 
make, each one of them have costs. So, the CAFR has 
costs. That's the prescribed by --just by -prescribed 
by GFOA. Then we have the state. And 
then we have the EMMA that is also then rendered by 
our financial advisor on retainer that they send to 
the EMMA. So, all together it's about $20,000.00 we 
believe. Where if we just had a smaller subset. 
There's also the assumption that with 
all of this, the 346 pages of --of financial 
disclosure, we wouldn't attempt or we should not 
attempt to at first tag every point. The --the 
Iris --the Iris Technologies who is one of the 
producers of XBRL for the SEC public companies, they 
had said that it would cost maybe five to $7,000.00, 
but that's considering 21,000 points of data. 
MS. MARTIN: How did you decide what 
points of data you thought would be appropriate to 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 45 
beyond just really making it an easier way to 
populate our spreadsheets. 
U.S. local governments are difficult 
to standardized. So, governments that are called 
the same things don't necessarily have the same 
responsibilities, revenues or reporting conventions. 
So, for example, you've got New York City, Boston, 
they operate their school districts. Chicago has a 
separate entity that does that. In Maryland 
counties do the same thing that cities do elsewhere, 
right. So, they have a lot more of the 
responsibilities and revenues. 
Governments account for things --the 
same items differently, right. So, Hugh did some 
research on cities and pointed out that in Florida 
different cities reported the same item differently. 
So, the states share --I'm sorry, the local --the 
local government sale of --share of state sales 
taxes, some reported that as own source revenues. 
Whereas, others reported it as having received it 
from the state. 
It's been mentioned before that 
governments use different accounting standards that 
range from cash accounting up through full accrual. 
So, you've got very different reporting structures. 

12 (Pages 42 to 45) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 46 
And then also, the self --the 
ability to self-report and create these unique 
taggings, unique fields, is also likely to inhibit 
how reliable the structured data can be. Without 
oversight and quality --the quality and 
comparability of the data is likely to deteriorate 
from whatever point it starts at. 
Even in the regulated corporate 
market there have been some reported concerns about 
the accuracy and consistent use of the developed 
taxonomy. And, you know, many analysts are going to 
want to review the source documents anyway. They'll 
want to appreciate what's in the notes and other 
elements that can't be captured in just providing a 
subset of data points. 
You know, in --in thinking about how 
structured data may be able to be used more readily, 
it may be better suited for a state by state 
enactment of structured data where you can have 
fully customized data for the particular state based 
on what their governments do and how their 
governments account for things if you can get the 
state to agree to standardize that, but across 
states it gets very, very difficult. 
MS. MARTIN: Thanks. So, clearly 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 48 
it --it delays getting financial information out 
there, that would be problematic for investors. 
And if we're talking about, you know, 
structured data being --improving the --the 
availability of data by, like, five days, I'm not 
sure that getting a financial, you know, on day 205 
after the end of the fiscal year is any better than 
getting it 210 days after the fiscal year. 
And some of the issues that have been 
discussed about why it's difficult to get financial 
information have to do with costs, resource 
constraints, the availability of auditors that do 
governmental accounting. You know, they're just, 
there are many more issues that are out there 
that --oh, and the complexity of component units of 
government having to do their audits first and then 
report in to other governments in order for audits 
to be finished. 
So, if we can't get at kind of the 
whole, I don't know whether the right word is supply 
chain here, and figure out how to get more frequent 
or more --more timely information. I think really 
the issue for investors is getting more frequent 
information. And that's really our first cause of 
action is to try to get interim financial 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 47 
there's a lot of challenges with implementing any 
sort of standard in the market. 
MS. WASHBURN: Can I add one more 
thing? 
MS. MARTIN: Absolutely. 
MS. WASHBURN: Okay. You know, I 
think that one of the biggest issues for investors 
is timeliness of data. 
MS. MARTIN: Yeah. And that's really 
what has lead us to --to assume this panel is --is 
the challenge around timeliness of financial 
disclosures that --that you've all pointed out. 
MS. WASHBURN: Right. And, you know, 
it --it doesn't seem to me that the --there can be 
efficiencies with structured data, but the --but 
structured data doesn't get at the heart of why it 
takes so long to get a government financial audit 
out. And so, I think what we --you know, in 
looking at structured data, if it didn't delay from 
the time period and you could --you could overcome 
all of the hurdles I've talked about with 
governments, if it didn't delay the time getting 
information out to the public then I think that 
investors would be --you know, would either be 
indifferent to happy, right. But to the extent that 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 49 
disclosures versus just getting what we get today in 
a structured format. 
MS. MARTIN: Okay. So, I'm going to 
turn it to Mark and Mike just on the back of some of 
Lisa's comments. 
Do you think there is applicability 
for a standard like XBRL in the muni market or 
should the focus really be on what can be done to 
improve the timeliness of financial disclosures in 
any event? 
MR. KIM: So, I think Lisa raises a 
really good point that the XBRL question doesn't 
really get to the heart of the timeliness question. 
I think what XBRL does do is get to the content 
question of trying to make financial disclosures 
more comparable and more standardized and more 
uniform. And --and that is a value to this market. 
The timeliness question is, I think, a separate one 
that isn't addressed directly by the XBRL questions. 
MR. WILLIS: That may be more of 
what's required by the different people who are 
dictating the timeline, but certainly the examples I 
gave earlier and just the idea of the bar code as 
the analogy. 
Now, with the bar code, fulfillment 

13 (Pages 46 to 49) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 50 
is now an automated basis. So, that's a whole 
different game than before we had the bar code. So, 
that's what the supply chain actually standard does, 
it enables that, but it also depends on again where 
it's implemented. If it's bolt on problem it's 
not --you have a much more efficient idea. 
So, I think something that Duffy said 
earlier was that he's actually mapping this against 
his operating system. That's not his financial 
system. That's a different idea. So, I think 
there's something there to think about. It's more 
granular data not summarized like we see it today in 
a report. There's something there to maybe 
investigate and to really think about what 
information is useful to the analyst community, how 
can they get it in a more timely manner, maybe even 
a more frequent manner. But certainly back to the 
bar code, one of the things it enabled was more 
frequent fulfillment. 
MS. WASHBURN: Just to kind of chime 
in. I think one of the --the key differences is 
that, you know, when we're talking about 
corporations, right, they're motivated by 
different --different elements that maybe a 
government --than a government is, right. So, 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 52 
not seem that it's satisfying the market's wish. 
And that's much different than the SEC. And as far 
as, like, the incentives at the --the local 
government level or state or local, there is a 
different --a different goal. 
And there's sort of a tertiary goal 
in this and it's just transparency in the, you know, 
democracy and transparency for citizens and more 
granular level of details within operations of how 
many roads and how many curbs and how many law 
enforcement and all of that that also can be 
included within a taxonomy that can offer some 
benefits. 
But as far as the --the goals and 
the objectives of a local --local government, there 
is a different reward system. And it's funny that 
the GFOA has always been giving awards to people. 
The CAFR award which is also, you know, synonymous 
with CAFR itself, but they give local governments 
awards. I think politicians or local units of 
government they like recognition. They like --they 
like these things that they get. And if you're 
going to incentivize local governments to do such, I 
think it has to be --I don't mean to be crude about 
this. You have to sort of, like, give them an 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 51 
that's not to say there isn't value in getting them 
to prepare their financials more quickly and all of 
that, but governments are doing a whole host of 
different --different things, right. They're -they're 
running water systems. They're providing 
just garbage pick up. They're --they're doing a 
whole host of basic services and providing public 
safety within a community --wellness and public 
safety in a community. 
You know, I don't know that they're 
necessarily staffed with full financial departments 
across the board. Maybe Duffy's area is, but you 
know, there are many, many governments that have 
part-time financial areas that, you know, trying to 
get the efficiencies and dedicating more resources 
to it just might be ahead of left. 
MR. BLACKBURN: Yeah. There's a 
large spectrum of well-staffed units and 
well-skilled units and well-sized units. The 
sophisticated issuer is much different than somebody 
in a rural setting. Most of the metropolitan units 
of government are able to take on this and 
definitely solve the content issue. 
As far as timeliness though, it's -the 
210 days as a deadline is --is not --it does 
1 
2 
3 
4 
5 
6 
7 
8 
910 
1112 
13 
14 
15 
16 
17 
18 
19 
20 
21 
2223 
24 
25 
Page 53 
award, right. They put it up on their wall and they 
take pictures and -- and they show them -- they get 
to be in the paper. And if you could maybe --you 
know, since EMMA is the -- the repository, if they 
can offer just voluntary, like, hey, I got the 
early -- the early filer award or if you're going to 
satisfy XBRL, give them the XBRL award. It sounds 
crude, but I think that's a -- a good, good way. 
MR. PRYDE: Just a general comment 
on -- on standardization. 
I think, you know, all these issues 
are completely and aptly valid. However, to --to 
solve a lot of those problems, it's not going to go 
away if you do something. And if you look at it -and 
Mike talks about the supply chain. If you 
looked at New York in the '70s, there were a lot of 
wharfs here, that was totally revolutionized by 
standardization, right, through something like the 
shipping container. You know, now we can get goods 
from China or wherever, you know, in we're talking 
days or weeks versus months. 
So, the standardization of data or 
standardization of anything gives a lot of incentive 
for people to move towards that cause it's going to 
drop costs over time. So, you know, this is a 

14 (Pages 50 to 53) 


1 
23 
4 
5 
6 
7 
8 
910 
11 
12 
13 
14 
15 
16 
17 
18 
19 
2021 
22 
23 
24 
25 
Page 54 
complicated problem. 
EMMA, you know, the MSRB council has 
a problem, FIMSAC council has a problem, individual 
states have to get on board. It's a --it's a more 
of -- but you need to have a stake in the ground 
which people can start to move towards. Without 
having anything to move towards you're not going to 
get any change. 
Now -- and that change is going to 
take --it's not going to be a week, it's not going 
to be a year, it's not going to be two years, it's 
not going to be ten years probably, but to solve a 
lot of these problems you can achieve a lot of 
things through automating it. But to be able to 
automate any of this stuff, all of this stuff I'm 
hearing, it's very manually created it. 
Everyone's -- a very manual process. So -- but 
starting to set the framework for standardization is 
going to pay off benefits years from now. 
And I don't think everyone should be 
looking, hey, we're going to move to XBRL and we're 
going to have all these financial reports in years 
times and everything is going to be great. That's 
not the fact. States have to be convinced, everyone 
else. So -- but that process needs to start 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 56 
MS. WALTER: Excuse me, my short arm 
doesn't reach the button. 
I agree with much of what has been 
said, but --and I leave it to David in particular 
to correct me if I'm wrong, but my recollection is 
that there isn't anything in the securities laws 
that prohibits the Commission from setting 
disclosure standards with respect to municipal 
issuers. The power of members is widely 
misunderstood to prohibit that. It doesn't. It 
prohibits the Commission and the MSRB from requiring 
filing. 
So, it is --now, maybe it would get 
struck down by the courts, but the Commission I 
think could step up and do rule some making. And 
I'm not suggesting a Reg SK, but something at a 
higher level to sort of walk people towards 
standardization, which is incredibly helpful for 
investors, and at the same time push for legislation 
that would provide explicit authority. 
And I think one of the keys will be, 
I think the stake in the ground, even if it's some 
time out, is good. You could stage this. So, you 
first put in place requirements for the regulatory 
or legislative which is perhaps at the state level 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 55 
somewhere. And I think it's, you know, just make 
sure everyone realizes that it's going to be --you 
know, it's got to be --you've got to do something. 
MS. MARTIN: So, Mark and Mike, from 
your perspective, and I think you eloquently pointed 
out that there isn't a filing deadline for --for 
states associated with their financial disclosures 
every year. What --whose mandate is it ultimately 
to compel the states, aside from Duffy's idea of 
giving a best in class trophy, what else can be done 
to get municipalities to file on a more timely 
basis? 
MR. KIM: So, this is probably not 
going to be a very popular answer, but Congress has 
the authority. 
MS. MARTIN: Okay. 
MR. KIM: But absent congressional 
action, the second best solution is voluntary 
adoption by the municipality. 
MS. MARTIN: Okay. Michael, maybe we 
can open it up for broader FIMSAC committee 
questions. 
MR. HEANEY: I'm happy to at this 
point. 
Elisse. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 57 
and then sort of move down in terms of what 
resources different municipal issuers have. 
I think one thing that is key 
particularly with respect to moving forward without 
legislation, but both is that investors really need 
to speak up very loudly about the fact that this 
really needs to happen because when we sit around 
tables like this, and I've done it for many years 
now, everybody kind of nods their head, yes, that's 
a really good idea, but in --in broader terms 
people aren't really pushing for it. So, I think 
that's something we should consider sort of on dual 
fronts. 
MR. HEANEY: So, let me if, I can, go 
to Rebecca to respond to that. And, Elisse, I will 
agree, there's a lot of head nodding, with myself 
included, going on with this. And if retail doesn't 
stand up and --and ask for it, I would hope that 
this group stands up for them. That's the one 
comment I would make before I pass to Rebecca. 
MS. OLSEN: So, I'll just say I agree 
with Elisse, but I --I would note even though 
the --you know, the tower amendment doesn't 
preclude the Commission from promulgating disclosure 
standards, there's also no expressed authority for 

15 (Pages 54 to 57) 


Page 58 Page 60 
1 us to do so. 1 MS. WASHBURN: I think what Elisse 
2 Where we have done rule making in the 2 says is absolutely correct in terms of access to 
3 past, which is primarily Exchange Act Rule 15c2-12, 3 additional information. Since it --it may be 
4 is based very much on our anti-fraud authority. So, 4 difficult to get to the timeliness issue for 
5 in thinking about this and --and things we can -5 
structured data, but the thing that we really need 
6 we can do, we definitely have to ground it in 6 is more frequent information in order to make sound 
7 anti-fraud which is why I think, you know, many of 7 investment decisions. And if information is 
8 the recommendations the Commission made in its 2012 8 being --being compiled at municipality level, 
9 report, the municipal securities going to the 9 whether that be quarterly financials, 
10 disclosure side did ask Congress for more authority. 10 budget-to-actual statements, cash flows, making that 
11 Otherwise, we are very much left with relying on the 11 information available to analysts will go a long way 
12 anti-fraud. 12 to bridge the gap in terms of the --the time lag in 
13 In terms of kind of format of 13 getting the audits. 
14 submitting financial disclosures, you know, the 14 MR. HEANEY: Amy. 
15 architecture set up by Rule 15c2-12 does a couple of 15 MS. McGARRITY: Thank you. Thanks to 
16 things. First of all, it designates the MSRB as the 16 all of you for participating in today's council. 
17 single and central repository for where you have to 17 It's been very helpful. 
18 submit your disclosure. And it also gives to the 18 My question is directed at Duffy, and 
19 MSRB authority for prescribing the electronic 19 maybe at Lisa as well. And I think we've talked 
20 format, as well as, any identifying information that 20 about this on the subcommittee a lot, and maybe even 
21 has to be included in that filing. So, back when 21 at this FIMSAC meeting in the past, but, Duffy, have 
22 that rule making was done, the MSRB did a rule 22 you had any feedback from investors on your 
23 making and they --you know, they put down that it 23 implementation of XBRL? And, is there any change at 
24 has to be sent in in Word searchable PDF. 24 all in your cost of capital? Are there market-based 
25 So, you know, could something in that 25 incentives aside from the best-in-class trophy that 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 59 
way be done in the future. You know, it would have 
to, you know, very much ground it in the authority 
we have or --or the MSRB and its authority as set 
up under the architecture or the rule, but it's -you 
know, it's certainly something we could think 
about. 
MS. WALTER: I don't want to hog the 
microphone, but let me just make one final point, 
and that is that I think the municipal issuers need 
to key in more closely to two facts. One, their 
constituents are frequently their investors. So, to 
say they're looking after their constituents more 
than their investors can be a funny game, but 
secondly, also many municipal issuers put out 
information more frequently than annually with 
respect to budget and the like. And in terms of 
anti-fraud authority the things that you're not 
saying may get you into a box. 
So, I think municipal issuers should 
be encourged to realize that this can be a win for 
them as well because they are vulnerable given the 
fact that they do make disclosures and that people 
want those disclosures. 
MR. HEANEY: Thank you, Elisse. 
Would you like to respond? 
1 
2 
3 
45 
6 
7 
8 
9 
10 
11 
12 
13 
1415 
16 
1718 
19 
20 
21 
22 
23 
24 
25 
Page 61 
we could potentially look to encourage or 
incentivize other municipalities to adopt this type 
of disclosure.? 
And then, secondarily, I think either 
of you, or both of you, or maybe all of you are 
active in GFOA. And to the extent that this cannot 
or there is some resistence to mandating it at a 
regularly level, is the GFOA and their standards, do 
they have the ability to have, like, extra credit 
GFOA standards wherein there could be some sort of 
award? Because that does motivate. I mean, we have 
on our wall lots of plaques for our CAFR and they 
are -- we are proud of it. 
So, I'm just curious as to your 
thoughts of other ways to implement some positive 
change here. 
MR. BLACKBURN: Yeah. And other than 
Will County I think the -- the XBRL issuers are 
vacant. So, we actually have not heard anything 
back from any investors. We would love to hear -get 
some at-away for that. But we look towards a 
long time horizon. So, if this is 2019, we plan to 
retire our current bonds right now in 2050. So, we 
do see schedules that show that timeline. So, we're 
aware of a longer timeline. And if it takes us that 

16 (Pages 58 to 61) 


1 
2 
34 
5 
6 
7 
89 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
2324 
25 
Page 62 
long or if it's 2100, we're planning to, you know, 
move into the future going forward. 
So, I would like to see -- I would 
like to see if we are pushing better data, more 
quickly, maybe unaudited and more timely, that that 
would give us some sort of reduction in cost of 
capital. That would be a -- a goal. 
MS. WASHBURN: So, I would say this, 
it's very difficult in today's environment to kind 
of say that there would be a cost of capital 
advantage for just about anything, right. I mean, 
supply and demand is just completely imbalanced, but 
that said, I think that where there could be a 
benefit down the road is in the providing of this 
interim financial data more quickly. I think that 
that's where you'll have investors make 
distinctions. If they're getting regular updated 
information from an investor --I'm sorry, from an 
issuer, that will -- and they're not from another 
issuer, they'll make the decision to go with the 
issuer that is providing the updated information. 
And that should increase the demand for those bonds. 
So, I think it really has to do with 
the providing of the information. Not necessarily 
the method in which you get the structured 
1 
2 
3 
4 
567 
8 
9 
10 
11 
12 
1314 
15 
1617 
18 
19 
20 
21 
22 
23 
2425 
Page 64 
would expect after voluntary issuance they -- they 
would be able to supply more data. And more data is 
probably, and rightfully so, will be -- will be 
requested. 
MR. HEANEY: Larry. 
MR. TABB: Might I suggest that we 
think creatively about the -- that we think 
creatively about the leverage that the government 
can exercise to -- to get the type of disclosure 
that would be beneficial, not only to the markets, 
but most especially to the citizens of all of these 
different entities. 
I have two suggestions, but I also 
want to ask you guys to think of other places where 
we might look. 
So, the strongest leverage that the 
federal government can exercise in municipal finance 
is held by IRS. Is it possible that IRS might 
determine within its authority that it needs to 
review the regular financials to insure that -- that 
issues remain tax exempt. If that's the case, 
perhaps a cooperative gesture or a inquiry to IRS 
might improve things. 
Another place where we might look is 
if the municipalities need anything from the SEC, do 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 63 
disclosures or the method you get the disclosures. 
The other thing too that I think is somewhat 
inhibiting is that --I think, Duffy, you said 
you --you do a portion of your CAFR gets tagged. 
And from an investor's standpoint, right, we have to 
look at --at all of the notes and everything else 
that's in the CAFR to really fully appreciate what's 
going on in the government. 
So, if we're not getting information 
about a large contingent obligation, about a bank 
loan that's got a bullet maturity or covenants that, 
if breached, are going to result in the 
acceleration. Those are really critical elements to 
the credit quality that can't be found in just a 
subset of data. 
MR. BLACKBURN: Yeah. Currently we 
have the --the process that we take our --our 
annual financial figures, those are actually tagged 
into a Word document. Again, just a side note on 
our technology, we just --we do have those numbers 
tagged so that all those figures that then end up in 
the paper document, the Word document, that then is 
exported to PDF are tagged. And once we feel more 
comfortable and others feel more versed and 
comfortable with the thought of data tagging, I 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 65 
we --do we have to review the original filings? If 
so --before they --before they can move forward. 
If so, why don't we create two lines, one line for 
those --for filings coming from states that are 
fast in their reporting and another line from the 
states that are slow. 
Now, I focus on the states and not on 
the entities because ultimately we could get a lot 
of power by pushing the states to do what we want 
them to get --to have the states push upon the -the 
different entities. This is especially 
important because many of these smaller entities 
mainly come to the market every 20 years and they 
figure, well, why bother with any of this until we 
have to come to the market. But if one of those 
entities is slowing down the rest of the state then 
it's very likely that the state will wisen up to it 
and start acting in --in the common interest. 
So, those are two suggestions, but 
perhaps all of you and others at the table can think 
of other levers besides perhaps those that we see in 
the powers that --and --and as a practical matter 
move forward without further congressional input. 
MR. WILLIS: I have a thought on the 
other agencies, specifically the IRS. 

17 (Pages 62 to 65) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 66 
The corporate filers, the structured 
data that comes in, it comes in as a filing. Every 
quarter we take those filings and strip them all 
down just to the data and we publish it as a data 
set. And those data sets are among the most popular 
downloads on the SEC's website every quarter, but 
they're also of interest to other agencies. 
Specifically, the IRS, census, Bureau of Economic 
Affairs, fed, et cetera. And the reason is because 
they're very granular. More granular than what's in 
the tax returns. 
So, they're using them for fraud 
assessments, looking at accruals, all that kind of 
thing, which sounds a lot like some of the things 
that Lisa is interested in in terms of that granular 
data on a specific debt instrument or a contingent 
liability. That's the kind of thing that's in those 
structured reports. 
MR. TABB: So, that would be another 
example, and the SEC's resources are limited and 
certainly we should be allocating those resources to 
maximize your agenda. Perhaps it's the case that -that 
those disclosures which ought to be timely 
will --the SEC first devotes resources to those 
people who are making it easy for us to do it. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 68 
MR. BLACKBURN: Yes. We produce a 
GFOA CAFR in that expectation currently. 
MS. BALL: Okay. Speaking of costs, 
you spoke about implementation costs. What are the 
ongoing costs every year? Like, once you build 
taxonomy, do you --is that a one time thing or do 
you constantly need to do something every year? 
MR. BLACKBURN: Yeah. The current 
tagging and formatting would have a cost, but it 
would be greatly reduced afterwards. Just like any 
formatting of a large financial document. 
So, we're hoping that once we have in 
place now the only thing we have to do is just 
update the figures as --and then, just like any 
chart of accounts within an organization, maybe 
once, twice, three times a year new accounts are 
added in in a block. So, if --if we're tagging 
less numbers and less figures than that, we're 
expecting not much, you know, ongoing continuing 
costs. 
MS. BALL: Is this done internally or 
via some software or externally? 
MR. BLACKBURN: It would be --right 
now we are using just Excel. And some of it is, 
sort of, the spoke work that we have to do because 
Page 67 Page 69 
1 MR. HEANEY: Thank you. Let me go to 1 we are the --the only person doing this with just 
2 Giedre because we have about four or five minutes 2 Excel, but other applications are --and products 
3 left. 3 and solutions are on --out there in the 
4 MS. BALL: Okay. So, I want to talk 4 marketplace. 
5 a bit about costs. 5 We've just signed a contract for a 
6 What I'm hearing is that the costs 6 new ERP situation with Microsoft, the Dynamics 365, 
7 are going to be bourne by the issuer, but the 7 which has and offers --throughout Europe they've 
8 benefits are going to be reaped by investors because 8 implemented XBRL taxonomy and schema and everything 
9 if you are not a large frequent issuer, what is the 9 that would just automatically come from our chart of 
10 benefit really to you, right? 10 accounts and be placed right into the instance 
11 Second thing, timeliness. That does 11 document. 
12 not help with timeliness. That actually does the 12 MS. BALL: Mr. Willis, is that what 
13 opposite to timeliness. Issuers are dealing with 13 you were talking about, the operating system, right? 
14 multiple complex components, right. They are also 14 MR. WILLIS: Yes. 
15 dealing with different GASBE amendments and -- like 15 MS. BALL: Okay, thank you. 
16 latest one, GASBE 87. And then you add another 16 MR. HEANEY: Thank you. I'll go to 
17 requirement that's going to require their resources 17 Suzanne and then John, please. 
18 and time, tagging, creating taxonomy, that's going 18 MS. SHANK: Yes. Most of my comments 
19 to affect the timeliness even more. 19 and questions have been already asked or stated, but 
20 So, Duffy, question to you. Are you 20 I wanted to just mention that we have approximately 
21 a frequent large issuer? 21 50,000 municipal issuers and this is not going to be 
22 MR. BLACKBURN: Yes. 22 helpful to investors if the vast majority of them or 
23 MS. BALL: Okay. So, you expect the 23 all of them do not, you know, participate. And, you 
24 benefits of XBRL reporting to be -- come through the 24 know, with all the things that Lisa and others have 
25 bonds transactions, right? 25 stated, that is just a mammoth, mammoth undertaking. 

18 (Pages 66 to 69) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 70 
We're either just going to be 
providing outside consultants, you know, new 
business lines so they can make more money on the 
heels of taxpayers. And so, I think we should just, 
you know, be cautioned that we are accomplishing 
something that is industry-wide, issuer-wide, that 
would have low impact for investors. 
MR. HEANEY: Thank you, Suzanne. 
John. 
MR. BAGLEY: Lisa, I wanted to ask 
you, do you think the market's ever really 
differentiated for issuers that provide interim 
financials? Understand, supply and demand right 
now, but if you look back over the last 10 or 15 
years, the ones that were better, were they rewarded 
by a cheaper cost of funding for similar credit -MS. 
WASHBURN: It's really hard to 
tell, right, because you have different tax 
structures. You've got different --different times 
that people are coming to market. So, it's really 
hard to tease that out, but I do know that, at least 
anecdotally, there are people that say --investors 
that will say, given the --the choice between two 
issuers, the one that provides the regular, more 
timely information, and one that hasn't reported, 
1 
23 
4 
5 
6 
7 
89 
1011 
121314 
15 
16 
17 
18 
192021 
22 
23 
24 
25Page 72 
governmental institutions. 
MR. HEANEY: Let me take this 
opportunity now to thank the panelists for your time 
for participating with us. This was incredibly 
insightful. So, you've given a lot to the 
subcommittee to think about and FIMSAC as a whole 
for sure. So, we appreciate your time. 
Lynn, thank you for moderating such a 
great panel. 
We'll take a quick, five-minute break 
next and get ready for the next panel. 
(Brief recess taken.) 
MR. HEANEY: So, our second panel 
will focus on alternative models for compensating 
credit rating agencies that were outlined in the GAO 
report in 2012. Amy McGarrity, Chair of the Credit 
Rating Subcommittee will be moderating this panel. 
And I will now turn it over to Amy. 
MS. McGARRITY: Thanks, Michael. 
Since the last FIMASC meeting on 
July 29th, the Credit Rating Subcommittee hosted 
numerous calls related to the topics of conflicts of 
interest and our SRO competition, use of ratings by 
investors and alternative payment models. 
A topic that has come up on various 
Page 71 Page 73 
1 they won't --you know, if the audit is, you know, a 1 calls is the use of credit ratings by index 
2 year-and-a-half late, they won't buy that particular 2 providers and other investor guidelines that might 
3 bond and they'll buy the other bond. So, they're 3 produce the level of competition amongst NRSROs. 
4 making distinctions particularly in the secondary 4 The subcommittee discussed this topic and believe it 
5 market. 5 is a relevant one worthy of further discussion. 
6 MR. BAGLEY: And one other question 6 All of the subcommittee members of 
7 related to Suzanne's. With this marketplace, right, 7 the ETF and bond fund subcommittee are also on this 
8 it's tough to say that a small school district would 8 credit rating subcommittee. And, as such, we 
9 benefit from this. Is there any value to looking at 9 recognize that we also heard that topic in some of 
10 it from, how much bond you issue and, you know, 10 their work. In fact, that subcommittee will be 
11 should over a certain size matter? So, if you are 11 hosting a panel discussion on that topic later 
12 this kind issuer and you're coming to market five 12 today. 
13 times a year and you issue billions of dollars, that 13 We have learned a lot over the months 
14 would cover most of the --at issuance the par 14 and have honed our research on the alternate payment 
15 amount. Maybe it wouldn't --it would only cover a 15 models. While the subcommittee has not coalesced 
16 small fraction of the issuers, but we know that if 16 around a particular model, or even whether a 
17 you get through, what, two or 300 issuers, in terms 17 recommendation should be considered, we feel the 
18 of what's issued, that's going to be a huge chunk of 18 FIMASC members will benefit from a base setting on 
19 it. 19 work that has already been done in this area. To 
20 MS. WASHBURN: And the problem with 20 that end, today we will have the opportunity to hear 
21 disclosure really is with not the two to 300, right. 21 from people who have either proposed alternate 
22 It's more prevalent among the smaller. So, really 22 models or who are actively involved in SEC's work 
23 the --the issue with timeliness and accessibility 23 studying alternate models detailed in their 2012 
24 to information exists much more --it's much more of 24 report to Congress on assigned credit ratings which 
25 a problem below the states and the large 25 Michael forwarded to you in our pre-reading. 

19 (Pages 70 to 73) 


Page 74 Page 76 
1 While today's panelists do not 1 number of investment banks in the ranges of premium 
2 represent all of the seven models outlined, you will 2 deal flow to the rating agencies. Also, the 
3 get a sense of some of the ideas that were discussed 3 arranger itself could be a rated entity. So, it 
4 and how their thinking has evolved or not since that 4 could seek to curry favor with the rating agency 
5 time. The objective of today's discussion is to 5 that's issuing ratings on it. 
6 present some of these ideas to the FIMSAC. 6 The report also describes the other 
7 With that, let me introduce our 7 conflict which is the other model, the subscriber 
8 panel. 8 paid and investor paid model. And here it was noted 
9 Today we will hear from Neil Barren, 9 that investors may want a particular type of rating. 
10 Consultant, Mahesh Kotecha, Structured Credit 10 So, if they need a particular type of rating to -11 
International Corp., Jeffrey Manns, George 11 to get -- meet an investment guideline or they might 
12 Washington University Law School, David Raboy, 12 want a high rating or they might want a low rating 
13 Professor and Randall Roy with the SEC. 13 based on whatever their investment position was. 
14 I have asked each of the panelists to 14 Perhaps shorting the company or they might have a 
15 limit their introductory remarks to five minutes, 15 CDS. 
16 with the exception of Mr. Roy. 16 Also, with the investor pay or 
17 Mr. Roy, I'll turn it over to you to 17 subscriber pay model, there's no public scrutiny of 
18 kick us off. 18 the ratings. So, you have to pay to get the 
19 MR. ROY: Thank you, Amy. 19 ratings. Where on the issuer pay side, the ratings 
20 So, among many things that the 20 are made publicly available to the extent that 
21 Dodd-Frank Act asked the Commission to do one was to 21 public scrutiny is viewed as a check on confidence. 
22 do a study and report to Congress on the feasibility 22 So, turning to the CRA board. What 
23 of what I'm going to refer to as the CRA board 23 this was was, there was an amendment to the 
24 model, which I'll describe in a minute, and other 24 Dodd-Frank Act which I believe passed in the Senate, 
25 alternative means of compensating NRSROs for 25 but ultimately didn't make it into the -- the final 
Page 75 Page 77 
1 structured finance rating. So, the focus was 1 statute and it would have set up what I'm calling 
2 obviously coming out of the 2008 crisis structured 2 the CRA board. 
3 finance. 3 The SEC was asked to view that 
4 Around that time the GAO also did two 4 amendment and opine on the --or give, you know, 
5 reports that were helpful to the SEC's work, one in 5 some views on the --the feasibility of it. So, 
6 2010 and one in 2012. Among other things those 6 what that amendment would have required is that the 
7 reports identified other alternative compensation 7 SEC establish a credit rating agency board which 
8 models, some of which we'll hear from the other 8 would have been an SRO. In terms of the composition 
9 panelists, and also it set forth the framework for 9 of the board, a majority of the members would have 
10 evaluating a model, an alternate compensation model. 10 to represent investors. At least one member would 
11 The SEC requested comments on these 11 have to represent an issuer, one member a credit 
12 issue and received 32 comment letters from NRSROs, 12 rating agency, and there needed to be an independent 
13 academics, market participants. SEC staff 13 member as well. 
14 interviewed some of the model authors and also 14 The function of the board would be to 
15 looked at relevant literature. The staff issued a 15 assign structured finance ratings to qualified 
16 report in 2012 that the Commission approved for 16 NRSROs. So, from the pool of NRSROs, those entities 
17 release to Congress. The report dealt with a number 17 that sought to get ratings assignments from the 
18 of issues in structured finance, but I am going to 18 board would need to apply to the board and the court 
19 focus on obviously the conflicts and then, the 19 would determine whether they were qualified in one 
20 alternative models. 20 or more ratings of structured finance to get 
21 So, the conflicts identified in the 21 assignments from the board. Issuers would be 
22 report appears to be the issuer paid conflict. I 22 prohibited from getting a rating from an NRSRO or a 
23 think that's well understood by everybody. It was 23 credit rating agency without first going to the 
24 noted in the report it might be more acute in the 24 board and having the initial rating assigned through 
25 structured finance area where you have a limited 25 the board process. 

20 (Pages 74 to 77) 


Page 78 Page 80 
1 The CRA board would select a 1 to fund the -- the board and its activities. 
2 qualified NRSRO to do a rating. There was a number 2 Also, in terms of the selection 
3 of selection methods that it could use. It could 3 process, the random selection process was -- was -4 
use a lottery. It could use a random selection 4 was thought it could potentially be unworkable in 
5 process or a --a rotating process or it could make 5 the sense that you might be assigning ratings to 
6 selections based on performance. Qualified NRSROs 6 entities that aren't -- don't have the capacity or 
7 would be --would set the fees for the assigned 7 the expertise to do the work. 
8 ratings that they got, but the board could step in 8 Switching to a performance based 
9 and set fees by rules. For example, if it thought 9 selection process, there were concerns raised that 
10 that the NRSROs were setting up reasonable fees. 10 that could cause the rating agencies to be overly 
11 The CRA board would have to evaluate 11 conservative and to avoid downgrades and cause them 
12 each qualified NRSRO annually. That evaluation 12 to hesitate to use downgrades. The complexity of 
13 would be based on an exam of the qualified NRSRO and 13 developing a performance metric also was met. 
14 the performance of the qualified NRSRO's rating. 14 Particularly when you think of the different types 
15 The CRA board finally would be funded by fees on 15 of ratings out there. Some -- some rating agencies 
16 qualified NRSROs and by credit rating agencies that 16 may seek to have ratings that are stable through 
17 applied to become qualified NRSROs. 17 economic cycles. Others may want ratings that are 
18 Once an issuer got a rating through 18 more sensitive to market factors and, therefore, 
19 the board process, it was free to go and get a 19 more volatile. 
20 second or third rating from the NRSRO of its choice. 20 Also, some rating agencies take into 
21 The report went on to identify potential benefits, 21 account loss given default and others don't. So, 
22 as well as, potential concerns that commenters and 22 having a performance metric that would take into 
23 others raised with respect to this model. 23 account all those different methodologies has made 
24 First, in terms of benefits. 24 this quite a difficult process. 
25 Obviously it could mitigate the issuer pay conflict 25 Also, there was a concern that a 
Page 79 Page 81 
1 by removing the issuer from selecting the NRSRO. It 1 performance based metric could cause the rating 
2 could reward qualified NRSROs for good performance 2 agencies to alter their methodologies to sort of fit 
3 by being assigned more ratings if performance was 3 within that metric leading to homogenous ratings or 
4 the metric in which assignments were made. It could 4 speculative results. 
5 provide smaller NRSROs with an opportunity to 5 Also, finally with performance was a 
6 compete and develop track records in the structured 6 view that it could be contrary to the statute which 
7 finance area. Also, it was thought that having a 7 prohibits the Commission from regulating the 
8 majority of investors on the board would focus it on 8 substance of ratings or the methodologies by which 
9 accurate ratings. 9 NRSROs do ratings. They sort of cater their rating 
10 Potential concerns raised were that 10 methodologies to this performance spectrum. 
11 it may ultimately not mitigate the issuer paid 11 Other concern was that it would 
12 conflict because after getting the initial rating 12 replace the issuer pay conflict with a different 
13 through the board process they could go out and get 13 conflict. So, the CRA board members potentially 
14 the rating from the NRSRO of their choice. Also, 14 could desire specific rating outcomes or the credit 
15 there was a concern that NRSROs may not even 15 rating agencies themselves, given the government 
16 participate. So, if nobody applied to become a 16 involvement in this board, might give favorable 
17 qualified NRSRO the board would have nobody to 17 ratings to the government. 
18 assign to or, you know, you would have a couple of 18 Finally, it was noted that the CRA 
19 rating agencies that applied and they might not be 19 board and the government involvement could be 
20 able to handle the volume of deals that come through 20 countered to the goal of reducing reliance on credit 
21 and create a bottleneck. 21 ratings. In other words, if investors and others 
22 It also was viewed as potentially 22 thought that the activities of the board and the 
23 costly. And there was a question of --of, would 23 government involvement in the board created a 
24 there be enough qualified NRSROs. And even if every 24 premature of these ratings that could cause more 
25 NRSRO applied, would they be enough in terms of fees 25 over reliance than undue reliance when there was a 

21 (Pages 78 to 81) 


1 
23 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17181920 
21 
22 
2324 
25 
Page 82 
shift at that point of reducing investor reliance. 
Quickly, the report also looked at 
the other models and identified potential benefits 
and potential concerns. On the benefits side, 
obviously mitigating the issuer pay conflict, the 
potential to improve ratings quality, promote 
competition and increase investor choice. Some of 
the models you hear are designed to do all of that, 
but there was also concerns that, again, they may 
not ultimately mitigate the issuer pay conflict. 
They may replace that conflict with other conflicts. 
Some of the models have a presumption that investors 
are less conflicted than issuers. That may be the 
case, but certainly investors may desire them, as 
noted early, specific types of rating. Feasibility 
and complexity were also raised as issues. 
And with that, I'll turn it over. 
MS. McGARRITY: Thank you, Mr. Raboy. 
So, now let's go through some of the 
models themselves. I said earlier, we don't have 
all seven here, but we have a solid representation 
of those that were presented at that time. 
Mr. Barron, let's start with you. 
And if you could, please limit your initial comments 
to under five minutes, I would appreciate it. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 84 
because they felt, the ones that I talked to --and 
let me just say parenthetically right now that 
there's much less of a need for it now than there 
was when I was working on it which was back in 2009, 
2010. So, it will be harder to coral fixed-income 
investors to get involved. I have some suggestions 
though, but all of them felt it would be good for 
their relationships at that point with their 
constituents, whether retirees or whether mutual 
fund shareholders. Their investments would be 
limited to seven to $8 million each and their 
returns would also be limited. So, their profit 
motive would be --would be actually gone. 
The idea would be for them to --to 
serve on the board and control the --the internal 
governments, including the three, in my opinion, 
governments failures that contributed so heavily to 
the financial crisis. One was internal compensation 
structures that were skewed towards the raters and 
away from those whose job it was to challenge the 
ratings. The allowance of rating shopping, 
competing for issuer fees --for --for issuers by 
lowering rating standards. 
And I want to say, I started 
representing Standard & Poor's in 1968 and --and 
1234 
5 
6 
7 
8 
910 
11 
12 
13 
14 
15 
1617 
18 
19 
20 
21 
22 
23 
2425 
Page 83 
MR. BARRON: I'll try hard. 
MS. McGARRITY: Thank you. 
MR. BARRON: I start off with just 
the proposition that you can -- you can legislate 
behavior that is inconsistent with human incentives 
and motivations, but it's more effective to have the 
right motivation and the right incentive to start 
with. So, that's the basis of what I'm suggesting. 
The model that -- that I'm suggesting 
preserves the issuer pay model, but in my view, it 
minimizes and maybe even eliminates the conflict. 
The --it's also a private sector solution that 
doesn't require any legislation, regulation or 
government involvement in it once it's up and 
running, okay. 
The idea would be that -- that you'd 
have an investor run rating agency. I call it an 
IROR, which would be owned and the board would be 
populated by six to eight of the largest income 
investors that we have that are invested in the 
structured markets. I have met with seven or eight 
of them earlier and I'll tell you what the outcome 
of that was. 
These fixed-income investors would be 
motivated by accuracy over revenue production 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 85 
then went to Fitch and left there in 1998. So, you 
can't blame me for anything, but in my view, one of 
the things that happened was the culture evolved and 
it lost track of the importance of --and the impact 
of ratings on our financial systems, on our economy 
and also on the ratings business. 
The --the fixed-income investors on 
the board would control internal governance that 
would have nothing to do with rating methodologies. 
They would not be privy to those. Nothing to do 
with rating outcomes. It would not be privy to any 
nonpublic information that is being received by the 
analysts. They would be insulated. Unlike some of 
the other --or at least one of the other models, 
their --their --they would be very limited to the 
information that they got and what their job was. 
Compensation I think is a big one. 
For example, and just an example, bonuses could be 
paid over a five-year period. And if a rating --if 
a --let's just say a triple A rating went into 
default, a committee would take a look at the 
analysis and determine whether there was a problem 
with the analysis, whether it was deficient. And if 
it were, part of the bonus would be lost. And also, 
salary increases would --would not be made. But 

22 (Pages 82 to 85) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 86 
paying bonuses over five years, by the way, also 
happens to be a --a pretty good retention policy. 
And the analysts that were responsible for 
challenging rating methodology would be paid at 
least as much as the analysts that generate the 
rating. 
The one way it would minimize rating 
shopping is that because --this is the market. 
Seven or eight of the largest institutional 
investors, at least then, and probably now, 
constitute a significant part of the structured 
finance market. So, they are saying, right, to 
issuers and bankers, you know, get our ratings. And 
if they got IORA's rating, and I've spoken to 
bankers --investment bankers about this, they would 
perceive a basis point benefit and that would be a 
very strong motivation. They would --the ratings 
would enjoy a very high degree of credibility 
because of who's managing all the governance 
internally. And I think they would reinstate a lot 
of those culture --you know, things that the 
culture lost over the years that allowed the 
behavior witnessed the Department of Justice 
complain against S&P. And just take a look at what 
happened there. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 88 
Tennessee Valley of Florida. 
From there, investment banking, MBIA, 
which you've heard of, and then financial advisory 
firm which I thought I would be out of business in 
two years. We are in number 21-year this year. So, 
we advise on ratings. We indulge in rating shopping 
of course. That is permitted currently, but we do 
it with some degree of integrity we think. 
So, the rating model we proposed, it 
was written up in four --three or four articles 
between 2010 and 2012. And it was summarized quite 
well by --by the SEC, their report. Ronald Roy 
sent me a memo in 2012 which summarizes it pretty 
well. I'm going to read a couple of paragraphs on 
that to summarize on it and then comment on it 
briefly. 
The basic model is --the problem is 
that you get paid as a rating agency by the issuer 
and the issuer can say, I'm not going to you, I'm 
going to somebody else or the banker can say the 
same. So, there is this issue of whether it skews 
the ratings. 
We break that by having a rating 
fund --U.S. rating fund which will pay the rating 
agencies, not the issuer, not the banker. That is 
Page 87 Page 89 
1 MS. McGARRITY: Mr. Barron, if you 1 simply put, our model. 
2 could just wrap. 2 So, how does the fund get the money? 
3 MR. BARRON: Pardon me? 3 MSRB, which you know, collects money with statutory 
4 MS. McGARRITY: Can you wrap up your 4 authority from a secondary market trade in municipal 
5 comments. 5 bond markets and from primary issuance. A small or 
6 MR. BARRON: I'll wrap it up right 6 larger trade --term. We would propose that the 
7 now, right. 7 same thing be done in the public markets, including 
8 I think --I think it has the 8 144As and commercial paper to the extent it's asset 
9 advantage of not having government involvement and 9 backed. And this collection of money could be 
10 risking moral hazard. And I would say it --and 10 placed in a fund, a U.S. rating fund. And then that 
11 also I would just repeat that it requires no new 11 would pay the ratings, the first two or three, 
12 regulation or legislation or government involvement 12 that's two. I'll come back to how many ratings. 
13 in the selection of ratings. 13 And --and, therefore, you wouldn't need for the 
14 MS. McGARRITY: Great. Thanks very 14 first couple of ratings to go to the rating agency 
15 much for that. 15 and say, what's your fee. Instead, the fund would 
16 Mr. Kotecha, could you please 16 pay for it. 
17 elaborate on your proposal as well. 17 So, you break completely the link 
18 MR. KOTECHA: Thank you very much. 18 between the issuer and the --the rating agency or 
19 My background, I've held five 19 the banker and the rating agency with regard to the 
20 corporate positions. The most important and 20 conflict. Conflict over for those two things. 
21 interesting, at least to me, was Standard & Poor's. 21 How many ratings? First, we propose 
22 I was there for nearly a decade running 22 there should be a queue of qualified rating agencies 
23 international ratings, public sector, all ratings 23 which would rotate randomly --not randomly, but on 
24 outside the U.S. And actually, I also got some U.S. 24 a rotating queue. And when the name comes up of a 
25 agencies and things that didn't quite fit like 25 rating agency fora new issue it would be assigned to 

23 (Pages 86 to 89) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 90 
the first two that come up. They would not have the 
option to say no. Otherwise, I mean, if they were 
not qualified they would not be on the queue. So, 
if they're qualified they must do it. They must 
accept the fees they've agreed to with the fund. 
The fund will pay them. So, there 
will be no link between the commercial side of the 
rating agency which would talk to the fund and the 
markets which are getting the ratings and a little 
side of the rating agency. That breaks the 
conflict. 
Now, there are lots of complexities 
and we can talk about the complexities, but this is 
fundamental. It is surgically eliminating the 
conflict. It means that the funding for the fund 
would come from the secondary market operations and 
the primary market issuance, meaning, both the 
issuer and the investor are paying. The investment 
banker knows --I mean, MSRB knows what the trades 
are, what the issue is. They charge this already. 
It's not --it's not a miracle. They do it today. 
So, it would simply be using --using that mechanism 
and just leveraging it up. 
So, mechanically in terms of 
implementation it's been proven. It exists. It's 
1 
2 
3 
45 
6 
78 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
2122 
23 
24 
25 
Page 92 
move on, one aspect of your model I don't know that 
you talked about. Isn't that the assignment would 
eventually be based on ratings performance. 
MR. KOTECHA: It could, but ratings 
performance measurement is not -- it sounds simple, 
but it's not. And it takes a long time. 
So, yes, it should be adjusted 
initially based on performance, but if you look 
at -- I think what was said earlier, some rating 
agencies prefer to have ratings through the cycle, 
stable ratings. Others might try to reflect ratings 
that change fairly regularly based on sort of 
cyclical changes that take place in the market, in 
the performance of the collapse or what have you. 
And so, you have to find a way that is empirically 
correct, but also based on consensus, methodological 
view and measured by an independent party. We have 
allowed for that to happen, but this should be done 
five- to ten-year cycles. Not everyday because 
performance can sway. 
I think more importantly, market does 
feed back to the -- to general -- to participants 
what's going on. You can -- there have been three 
articles in the Wall Street Journal about -- about 
high ratings by one party or the other. You can 
Page 91 Page 93 
1 just scaling it up. 1 quickly pick up and then check it out. So, if there 
2 Second, in terms of governance, there 2 is something going on like that you can put a stop 
3 would have to be a board for the fund because now it 3 to it. You can put corrective measures in place. 
4 has money. It would be a board which would have 4 You can call people in and ask them to explain. 
5 representatives of the industry, a banker, an 5 So, I think that kind of thing should 
6 issuer, a rating agency, but the rating agency would 6 be done, but performance measurement, while it 
7 not chair it. Never. It would be chaired by some 7 sounds wonderful, it is not so easy. 
8 third-party government officials. 8 MS. McGARRITY: Thank you. 
9 The priority of the fund could 9 Mr. Manns. 
10 include reviewing the fees. They'll see everybody's 10 MR. MANNS: Great. Thank you very 
11 fees. They might see that one is charging X, 11 much. 
12 another is 10 X. That would --that would cause a 12 I want to start with a global point 
13 flag to be raised and they could discussed the mater 13 which is really two things. One, the issuer pay 
14 and potentially issue guidance on what to do, but 14 model. The issuer pay model we have is 
15 this could be dynamically done. 15 fundamentally broken. Empirically what many 
16 And --and finally, this should be 16 academics have shown, we have a systematically 
17 done in coordination with consultation with the 17 inflated ratings system that exists. And the second 
18 international regulators because the markets, if you 18 point is, we have an oligopoly, it's the elephant in 
19 protect the markets here in the U.S., they'll have 19 the room, of three leading rating agencies that 
20 global impact as you've seen already with the 5G. 20 dominated the market ten years ago and they almost 
21 So, that's basically the proposal. 21 consistently dominate the market now. Also, we have 
22 Lots of implementation issues, but it surgically 22 seven other nominal participants, but they compose a 
23 removes the conflict. That's the --that's the 23 very small segment of the market. 
24 conclusion I have. Thank you. 24 The real question is, how do you 
25 MS. McGARRITY: Mr. Kotecha, before I 25 change the oligopolistic domination? How do you 

24 (Pages 90 to 93) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 94 
change the conflict of interest that comes from the 
issuer pay system? 
The second point would simply be 
this, there's a mandate from Congress to go about 
and have a pilot program in structured finance. 
That mandate was out nine years ago. It's shocking 
to me that nine years after the Dodd-Frank Act 
nothing really has happened in regards to that. 
That would be --my recommendation forward would be, 
every single proposal entails trade offs. I propose 
the trade offs. It has limitations. Every proposal 
Randall indicated has it, but in the face of a 
massive oligopoly and in the face of a systematic 
conflict of interest created by the do not bite the 
hand that feeds system of issuer pays, there's a 
case that's being made for a pilot program to do 
something different. 
My proposal is straightforward. It 
consists of either having a user fee imposed on 
investors or having a ratings tax that would go 
about and fund a CRA board or an independent 
commission to go and select and compensate rating 
agencies. That part is simple. What's the 
challenge? How do you select the rating agencies? 
How do you compensate them? And that's the other 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 96 
based government contract model would effectively 
allow a board to choose higher cost raters to go 
about and participate in at least parts of the 
structured finance rating market. 
The second aspect of it is the 
challenge that Mahesh and Randall both referred to, 
which is, you need to have some performance based 
dimension either in the selection or in the 
compensation of rating agencies. That is another 
very big problem. I think there's a problem with 
virtually every model. It's, what metric do we 
choose? If we choose a given metric, do we have 
herding affects of rating agencies simply gaming to 
test of whether they'll be compensated or however 
they're going to be selected, but that being said, 
any serious proposal has to go about and try to 
address that particular issue. 
Last issue to talk about would be the 
issue of expert liability. One of the other most 
shocking aspects to Dodd-Frank is how the leading 
rating agencies in blatant defiance of the law 
refuse to go about and will be subjected to Section 
11 liability of experts. And the SEC effectively 
caved in for fear of going about and disrupting the 
market. I think that's something that has to be 
1 
23 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
1920 
21 
22 
23 
24 
25 
Page 95 
big elephant. 
My proposal would call for having a 
cost based government contract model. And the basic 
logic of it is this, is if we have a competition 
based off of price we'll end up with ratings that 
have even less informational value than the ratings 
that we have because obviously folks would go about 
and bid and try to pledge to have transparent 
methodologies that are very high on qualitative or 
quantitive dimensions, but very thin on actual 
scrutiny and due diligence. And so, from that 
perspective the argument would be, have rating 
agencies go about and lay out in greater detail what 
their methodologies are, what due diligence steps 
they're going to be doing about it, taking what type 
and extent of exposures they're going to go about 
and scrutinize in order to go about and justify the 
rating. 
What's the thinking? It may be, 
frankly, ratings are too cheap and that helps to 
explain why their informational value is so thin. 
It may be better to go about and have a spectrum of 
competition where smaller players, such as Egan 
Jones, can compete even if their cost structure is 
higher. And so, going about and allowing a cost 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 97 
remedied. 
My proposal is to go about and have a 
more narrow scope of expert liability. Namely, 
having a due diligence certification duty imposed on 
rating agencies where they go about and have to 
certify that they have gone about and engaged in due 
diligence and they're exposed to gross negligence or 
recklessness liability to --either to the SEC or to 
private litigants, namely, the creditors committee 
representing investors with a cap damages system. 
So, the big elephant in the room in 
terms of going about and having full blown Section 
11 expert liability is the amount of liability 
exposure the rating agencies might face. If you 
encapsulate damages tied to the fees for the ratings 
that rating agencies are issuing you can go about 
and have a much more manageable burden that could be 
addressed through an SEC adjudicative process. 
So, in that sense, you don't open up 
the door to a massive blow of plaintiff lawyers 
going about trying to cash in and holding rating 
agencies extensively accountable. Instead, you can 
focus on giving the SEC, the Office of Credit Rating 
Agencies or going about empowering investors to go 
and scrutinize what rating agencies are doing to the 

25 (Pages 94 to 97) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 98 
extent to which they're grossly negligent, have some 
real financial consequences as opposed to an annual 
report that goes about and documents what rating 
agencies have done wrong. 
Money matters, and in that sense, 
having financial accountability should be part of 
any meaningful model. Thank you. 
MS. McGARRITY: Thank you, Mr. Manns. 
Mr. Raboy. 
MR. RABOY: Thank you. 
The model that I came up with in 2009 
was toxic, very umbratic and since that time I've 
really been thinking about the component and the 
problems with implementation. The components are, 
basically, the same thing that everybody's talking 
about, eliminating conflicts of interest and 
incentivizing CRAs to give better ratings. 
Some of the proposals I've seen out 
there I don't think do the job. And I'd be happy to 
get you to that in the question and answer, but I 
completely agree with Jeffrey, the elephant in the 
room is oligopoly. We have a small number of rating 
agencies. Indeed, probably a small number of 
structured finance issuers. What that means is that 
the game is really a game theory model where people 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 100 
has arisen is in the legislation there's the right 
of refusal by CRAs. And I understand why that's out 
there. It's to allow new entrants. It's to allow 
people to build up. The problems are that during 
the period where there is no --where there is a 
right of refusal that's incredibly easy to gain the 
system. So, if you want to institute a model where 
you have new entrants then you have to think back to 
the huge barriers of entry, the huge intellectual 
and physical capital requirement. 
So, if you look at a new entry trying 
to take on those costs, they won't do that unless 
there's some type of guaranteed revenue stream. And 
who's going to guarantee that. Otherwise, those sum 
costs will never be recovered. 
Finally is the --the fee structure 
itself. Whether we have a government body or an SRO 
administering fees, how is that done. Do we really 
want issuer --or CRAs to have to come up with some 
functional model like we do for transfer pricing 
where they list all of the components of their costs 
in order to justify them. I mean, that has a 
potential for dating. It also is incredibly 
complex. 
Finally, I think there needs to be 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 99 
interact, where there's informational symmetries, 
there's problems between who's to benefit and who's 
to suffer. And everything kind of has to be 
analyzed within that structure because there's any 
number of ways that CRAs could still take advantage 
of the system. 
The body itself in my model was an 
SRO basically. If you have too much government 
participation you do have a moral hazard finding. 
You know, just like flood insurance. You get flood 
insurance, people build by the river. It's the same 
type of thing where there would be the implementor 
of government participation which would almost give 
a sanctity or guarantee to ratings which could cause 
some problems. 
I still believe that some type of 
random type of assignment is --is necessary. If 
you have a queuing system that's very easy to gain. 
The --you can --you can figure out what's going 
on. If you have a rotation, that's even worse. An 
issuer can just wait 'til their best bud comes on 
and then get the issuer there. They can sort of 
predict it. So, I think some type of random system 
is necessary, but that is all in the details. 
I think one of the big problems that 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
1112 
13 
14 
15 
1617 
18 
19 
20 
21 
22 
2324 
25 
Page 101 
signaling mechanisms. And I don't know the 
legalities of these things. I mean, one of the 
proposals that I've seen in the research is 
disclosure of the resources committed to ratings 
which signals investors as to the quality of the 
ratings and also the reputation of the CRA. The 
other is disclosure is fees. So you can see what's 
going on, but essentially, we have a very long way 
to go even to implement the concept that I came up 
with and it's --it's documented. 
MS. McGARRITY: Thank you. And I 
sure can appreciate the daunting task in front of us 
all, but thank you for -- for that sentiment. And 
again, thanks for all of your views on these very 
interesting topics. 
Mr. Barron, I'd like to direct a 
question to you. Based on what you talked about 
earlier, one of the challenges I think of your model 
is that six to eight large fixed-income investors 
would need to be willing to invest money to -- to 
start the IORA. This would also cost investors 
time, money and other resources. 
Do you have any thoughts about how 
they could become interested, why they would want to 
participate in -- in this model? 

26 (Pages 98 to 101) 


12 
3 
4 
5 
6 
7 
8 
9 
10 
1112 
13 
14 
15 
16 
1718 
19 
20 
2122 
23 
24 
25 
Page 102 
MR. BARRON: Well, as I said before, 
they would want to participate much less now than in 
the teeth of the crisis. However if they did feel 
that it would send a good signal to their 
constituents, number one, and back then I had the 
chairman, chairwoman at the time's, support for 
this. And the SEC was at that point considering 
talking to a number of fixed-income investors and 
maybe exercising some moral suasion and trying to 
encourage them to get on board. 
They -- I met or had calls with seven 
to eight of them. None of them said yes. None of 
them said no. And I never got to senior management 
because the chairperson of the commission was -- was 
then replaced and lost that support. What would 
have happened, I can't tell you. 
I did go to Congress. I did go to 
Treasury. And the response was good. Especially 
the fact that it didn't require more legislation or 
regulation. 
So, my answer -- my answer to your 
question is, they would have to feel strongly enough 
that they're sending the right message to their 
constituents, sending a good message to their 
constituents, about them taking action which would 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 104 
responsibility to make any investment decisions 
there. And this is what we had at Fitch. 
The --the --the --the people who 
were working for and were on the board for the major 
owner of Fitch agreed in writing that they had no 
responsibility to purvey or use information from 
Fitch for the benefit of the employer. And the 
employer itself would agree in writing that -that 
--to the same thing, that they would not try 
to obtain any information or any benefit from the 
fact that their employee --that his employee was on 
the board. It worked. 
I was on the board of Fitch and I 
think --I think could work. It minimized the 
conflict which, by the way, is a little bit 
different from the concept where you have --I think 
it was IOCRA was one --one of the suggestions where 
the board has institutional investors on it, but at 
the same time has a lot more to do with the 
governance of the organization, including 
involvement in --in --in ratings criteria, et 
cetera, so. 
MS. McGARRITY: Okay, thank you. 
Mr. Manns, my next question is for 
you. You talked a little bit about how your model 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 103 
prevent 2007, 2008 from ever happening again. If 
you believe that it will never happen again it's 
less necessary, but, you know, over --over time 
greed evolves. So, that's my --my answer. 
MS. McGARRITY: Okay. Do you think 
that there --do you think there might be a 
potential for a different conflict of interest with 
your model. For example, could there be an instance 
where investors could seek to delay or avoid 
downgrading a troubled security to avoid its sale on 
a mark to market on their books? 
MR. BARRON: No question. Their -their 
--you know, investors, institutions with 
capital requirements or who want to sell bonds would 
want higher ratings. Those who want to buy bonds 
would want lower ratings, but there's significant 
conflict mitigation which we --I was vice chairman 
and general counsel at Fitch. I set up in 1998 
which we instituted at Fitch, right, 'cause the two 
thirds owner of Fitch was heavily involved in the 
bond markets. Bob Van Kampen at the time. 
The fixed-income investors would have 
no participation in criteria development or rating 
outcomes. And the --the IORA would --would choose 
employees of --of fixed-income investors who had no 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 105 
would create additional accountability and then you 
specifically mentioned there spoke liability. Are 
there other accountability mechanisms associated 
with your model that we should learn about? 
MR. MANNS: Sure. So, I think that 
the one --as I mentioned, one of the glaring 
failures of the SEC, and again, I apologize to our 
hosts because the SEC is our host, one of the most 
glaring failures was the failure of an expert 
liability under Section 11. So, that opened the 
door to meaningful financial accountability at the 
rating agencies. 
What's the big picture issue. When 
you step back we had a couple of extraordinary 
settlements involving S&P and Moody's. And so, what 
has happened is extraordinary circumstances there 
will be a financial settlement. What doesn't exist 
are settlements in ordinary circumstances when 
rating agencies are falling short. Instead, for the 
most part they get chastised at the annual report 
from the Office of Credit Rating Agencies. It would 
be valuable to have some meaningful mechanism for 
ongoing financial liability. 
And so, that's why my suggestion of 
going about having a due diligence certification 

27 (Pages 102 to 105) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 106 
duty and having gross negligence liability to 
investors is one mechanism for that. 
Another mechanism would be opening up 
a broader swap that a financial accountability to 
the Office of Credit Rating Agencies. In other 
words, having more explicit accountability naming 
the individual rating agencies that have been 
falling --have fallen short, naming the amount 
financial accountability that comes from them having 
falling short. 
MS. McGARRITY: Understood. Thank 
you. Has your thinking evolved at all from your 
original pitch in 2009? 
MR. MANNS: Sure. So, I think, if 
anything, the --the big lesson of rating agency 
reform was how hard reform actually is in practice. 
And so, to be quite honest, I think it's --when we 
look at what's happened with the rating agency 
accountability, there was a flurry of activity 
during the wake of the financial crisis and the 
implementation of the Dodd-Frank Act. And then, 
since then there hasn't been a lot of action in 
terms of rating agency reform or accountability. 
Frankly, I'm waiting for the next recession or 
financial crisis to hit because what ultimately -1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 108 
cost of securitization for issuers, investors and 
underlying borrowers? 
MR. KOTECHA: Not materially. They 
would be free to --they would be assigned a couple 
of rating agencies from the --the U.S. rating fund, 
but they could also, as they currently do, ask for a 
third rating or a fourth that they themselves 
select. 
Rating fees tend to be in the basis 
points ranging from three to five, sometimes more. 
So, if you get two or three they add up to not -not 
too much. Maybe 15 basis points for three at 
worst. 
Some of the more complicated ratings, 
commercial real estate securitizations, for example, 
they are more expensive. But, generally speaking, 
the cost is not a huge factor in the --in the cost 
of issuance. It's higher when interest rates are 
lower, as they are now, as a --as a portion of the 
total cost of issuance, but you save it in the 
spread that you pay. 
So, the cost is easy to measure, but 
the benefit --I heard the earlier panel, the 
benefits are not always precisely measurable, but 
they are there. If you look historically at --at 
Page 107 Page 109 
1 when you do look back in time, recessions are 1 ratings based pricing there are distinct reports you 
2 cyclical. We don't know when the next one's going 2 can see that suggest 25 basis points pick up for a 
3 to happen. We can predict with a fair amount of 3 notch or something that like. The more, the lower 
4 certainty that rating agencies would have 4 the ratings. 
5 underestimated the risk in terms of debt issuances. 5 So, you would make it back up. It's 
6 And at least part of the blame is going to rest on 6 easy to measure the cost, but it is actually quite 
7 their shoulders. 7 significant to have better ratings. And, therefore, 
8 And so, I think the dialogue today 8 tighter spreads in fact. And then, if you did 
9 and the dialogue going forward really is looking to 9 choose to have an additional rating it's because you 
10 the next recession or financial crisis that happens. 10 think that it's going to save you money and you can 
11 And at that time some of the particular failures, 11 actually choose not to take an extra rating. So, I 
12 such as the failure to implement a pilot problem for 12 think that the cost factor is --is exaggerated. 
13 the selection of structured finance ratings, or the 13 I would also say one other thing is 
14 failure to implement expert liability under Section 14 exaggerated is the notion that every model has a 
15 11, or the failure to hold rating agencies 15 conflict. I hear that from rating agencies all the 
16 accountable under the -- under Regulation FD. All 16 time. Oh, everything is a conflict. Not so. In 
17 of those issues will be back on the radar screen 17 our --in our rating proposal we have taken the 
18 again. And so, in that sense it's valuable to talk 18 conflict out, period. Because the issuer will not 
19 about those issues now with an eye to two or three 19 be the party paying and the investment banker will 
20 years from now when reckoning eventually will come. 20 not be the party paying. That's a complete break of 
21 MS. McGARRITY: Thank you. 21 the conflict. 
22 Mr. Kotecha, I have a couple of 22 So --and I think on the oligopoly 
23 questions for you. If, per your model, issuers want 23 issue, the industry is an oligopoly. It will remain 
24 to or have to get a rating from a second or third 24 an oligopoly because how many doctors do you want to 
25 rating agency, wouldn't this increase the overall 25 see for --for cancer? You want to see one, maybe 

28 (Pages 106 to 109) 


1 
2 
34 
5 
6 
7 
8 
910 
11 
12 
13 
14 
15 
16 
17 
18 
19 
2021 
2223 
24 
25 
Page 110 
two, maybe three. After that, the value addition 
kind of tails off. 
So, I think that in the hamburger 
business if you have Blimpe's and -- and McDonald's 
and Burker King, quality improves and prices go 
down. In the ratings business, quality goes down 
and pricing may or may not go up or down because of 
competitive edge sometimes is easy -- easy criteria. 
MS. McGARRITY: So, one of the 
benefits I think of the continuous queue model or 
even the random selection model is potentially 
increased competition wherein new entrants can get 
into the queue based on their own expertise, et 
cetera. But I wonder if -- if there could be an 
instance where investors, based on their investment 
guidelines or their benchmarks, would be required to 
have ratings from one of the top NRSROs and -- and, 
thus, that would impede their ability to purchase 
these types of securities. 
Have you had any push back on your 
model regarding that topic? 
MR. KOTECHA: We discussed that and 
that's why I've maintained our system will not 
replace the existing issuer pay model in totally. 
It will replace it for the most part. It will 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 112 
ratings of mortgage-backed securities --let me go 
back. 
I represented the Reagan 
Administration in drafting and enabling legislation 
for the former S&P up until 1989 in all their deals 
and then went to Fitch. And I said before, I left 
there in '98. And for --for many years, 20 years, 
ratings of mortgage-backed securities, or ratings 
generally I would say, performed well. The outcomes 
were those outcomes predicted by the ratings almost 
in all circumstances, right. 
I --I --I agree with Mahesh when he 
says that the result of competition can be more 
rating shopping. Putting words in your mouth, I 
understand that. And --and --and the reason I say 
that was --is, of course, a very blatant example -one 
very blatant example. 
I --if Fitch had not been 
successful, okay, Fitch passed S&P and Moody's in 
mortgage-backed securities in 1993, okay. Had they 
not been successful, and this is a big statement, I 
don't think the crisis would have ever happened. 
And the reason for that is, they started taking 
market share away from Moody's and S&P. And how did 
Moody's and S&P compete? By lowering enhancement 
Page 111 Page 113 
1 maintain pressure from an independent, un-conflicted 1 levels. 
2 model, but it will allow those who wish to, to pay 2 MR. KOTECHA: So, it was your fault 
3 for the big three if they wish because investors saw 3 after all. 
4 it wild. 4 MR. BARRON: No, I said I left in 
5 So, it is a force to retain pressure 5 '98. I don't think an oligopoly or a monopoly is 
6 on the --on the rating fund based allocation of 6 good under general circumstances in any sense, but 
7 ratings and to allow freedom to --to --to have the 7 the increase in competition I think over time, you 
8 dominant market participants, in some respect, 8 know, the taste of the 2008 crisis is still in our 
9 continue to some extent for some period of time that 9 mouths. And bad mortgages, I think the option arms 
10 dominance would be the independent choice outside 10 and negative amortization loans and no-doc loans are 
11 the rating model because you're not going to change 11 substantially gone. So --but over time things are 
12 this overnight. The big ones have 80 percent, 12 going to start to --human nature is that things are 
13 90 percent market shares. The little guys are 13 going to start to creep back in --back into the 
14 struggling in their own issues. And for them to get 14 system. 
15 higher you will see that there will be incentives 15 So, I don't have a solution to the 
16 for them to go to easy ratings. You don't want more 16 fact that the rating business is dominated by a few, 
17 competition because more competition goes to easy 17 but I will say that opening it up to competition, 
18 ratings. That is the fact of life. 18 ultimately, I think poorer ratings are going to 
19 MS. McGARRITY: Mr. Barron, did you 19 return. 
20 have a comment? 20 MS. McGARRITY: Mr. Raboy? 
21 MR. BARRON: Yeah. I think I've 21 MR. RABOY: I don't think we can 
22 known you, what, for 30, 40 years now and this is 22 dismiss the oligopoly aspect that easily. Yes, it's 
23 the first time I've agreed with him about something. 23 true that when Fitch entered it created a certain 
24 MR. KOTECHA: We're making progress. 24 situation, but the gaming is in the oligopoly. I'm 
25 MR. BARRON: Right. And that is that 25 not sure new entrants is not a bad thing. 

29 (Pages 110 to 113) 


12 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
1819 
20 
212223 
24 
25Page 114 
You have a problem, when a new 
entrant comes in, how do they gain business? How do 
they get market share? Probably through 
reputation -- building reputation and associated 
costs of rating. If -- if you don't have reputation 
you can conflate a rating as high as you want the 
market will discount you. But then you get into a 
classical prisoners dilemma situation where a new 
entrant comes in, competes on the basis of quality. 
Others try to follow suit, but eventually they get 
to the point of passive collusion where implicitly, 
not explicitly, signals are being sent and ratings 
quality can go down and inflation can go up over 
time to get to a new equillibrum. The only way to 
guard against that is somehow to encourage new 
entrants so you get beyond three and there's an 
incentive for others. 
MS. McGARRITY: Speaking of quality 
and competing on the basis of quality, I wanted to 
explore that a little bit with you some more. 
Is it possible that some ratings -MR. 
TABB: Hey, this is Larry Tabb. 
I can't hear people, but I have a comment. So, let 
me know when I can jump in. 
MS. McGARRITY: Okay. I just have 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 116 
you. And it's not really based on alternate models. 
It's rather, I think, to inform us. 
Is there any more information that 
NRSROs should be required to disclose that could be 
helpful for investors to assess NRSRO performance? 
MR. KOTECHA: Yes, I think so. I 
think that, you know, the issue of expected losses 
of the LGD --the LGD was mentioned and Moody's was 
also expected loss, S&P would be for the first 
dollar loss in the definition, but in very practical 
terms, you can guess the basics of both. And 
Moody's publishes these. And I think all agencies 
should publish statistics against a standard that 
should be set by the SEC hiring somebody to set 
those standards. 
I think that should be done because 
there is inconsistency in the data. There's 
inconsistency in reporting that makes it difficult 
to compare rating agency statistics partly because 
the --the grades themselves are slightly different 
at the bottom end. So, I think somebody who 
independently looks at this could say, look, if you 
were to do these three things, then --then all of 
the data coming in could be compared through and to 
a part and you can get a general rating default 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 115 
one more question and then --and then I'll try 
to -MR. 
HEANEY: We have plenty of time 
for questions after. 
MS. McGARRITY: Okay, thank you. 
Would it be difficult for a board or 
other assigning entity to determine the metrics for 
past performance ratings and ratings accuracy in 
order to determine future assignments do you think? 
MR. RABOY: Yes. It would be very 
difficult. And that's kind of above my pay grade, 
but I do know that there's a lot of academic 
research that's been done by statisticians, decision 
theorists, that sort of thing, where they have done 
a lot of work on establishing benchmarks, various 
staffing techniques, where agreement on metrics 
would certainly have to occur. And then, through 
some sampling technique which would be necessary 
because you can't review everything, compare it to 
those benchmarks. So, I do think that there is 
progress on that front. Probably more progress than 
on the other areas. 
MS. McGARRITY: Okay. Before I turn 
it over for the broader FIMASC and SEC questions, I 
just wanted to ask one --one question to all of 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 117 
matrix that goes back, aggregates everybody's data 
one by one, agency by agency. 
There is a system called Gems which 
is a system for multilateral develop banks. I do a 
lot of multilateral development bank advisory work. 
They have collected this under the leadership of BIB 
or IFC, I forget, or both. There are about dozen of 
them. They collect data from all of the 
multilateral development banks that lend to third 
world countries and they --they have project 
finance ratings, both defaults and --and LGDs. And 
they have a standardized method for doing so. 
So, I think if you could provide a 
standardized method that --that the agencies must 
use. And then, have independent party collect the 
data and collate it and put it together so there's 
an aggregate and --and it is aggregated default 
study that comes with a vetted, consistent and -and 
accepted method for measuring details, the 
defaults settings, that would be very valuable. 
MS. McGARRITY: Thank you. 
Does anyone else have any comments 
that you'd like to share. Mr. Raboy? 
MR. RABOY: Again, what I'm 
interested in is the signaling elements of that 

30 (Pages 114 to 117) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 118 
investors themselves have information because all 
they can view currently is default rates. So, they 
don't have information to judge. If somehow rating 
agencies were required to disclose the resources 
they put into a different rating or the fees 
themselves, or maybe both, then that information 
could be digested by the investors in addition to 
rates and provide more information to investors. 
MR. BARRON: I would say broad 
disclosure of the compensation structures within the 
rating agency itself and how different --not 
individual analysts, but classes of analysts, 
whether they're measured by how much revenue they 
bring in or by the performance of their rating, how 
the --their compensation of those classes of 
analysts are set. 
MS. McGARRITY: Thank you. 
Mr. Manns. 
MR. MANNS: And I --I agree with my 
fellow commentators comments. The one additional 
element I might add is this, there's an awful lot of 
opaqueness when it comes to the qualitative factors 
that go into the rating agency methodologies. And 
so, from that perspective having greater, more 
detailed disclosures as to how the qualitative 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 120 
points in a matrix to find a rating and you can go 
through like a mouse to find where the rating is and 
it's impossible to find it. So, transparency 
pressures from the regulators have lead to lack of 
transparency in fact in many cases. It should be 
changed. 
MS. McGARRITY: Thank you. 
Michael, should we open it up to the 
broader FIMSAC? 
MR. HEANEY: Sure. 
Amy, thank you. And I want to thank 
the panelists. This has been incredibly 
interesting. 
I'm going to kick off with just two 
quick questions. Then, I'll go to Larry and then 
we'll go --on the phone and then we'll go around 
the table. 
Altering comp schemes, many mentioned 
it, the banking industry obviously did it after '08, 
'09, there were a lot of complaints and agitations 
and objections, but the fact of the matter is the 
different pay structure probably enhanced the safety 
and security of the industry. 
Are there any negatives as you see it 
changing the compensation models within the rating 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 119 
dimensions are indeed measured factored in would be 
very valuable to have. Especially when you're 
making comparisons in the industry or across 
industries. 
MR. KOTECHA: May I just add to that. 
I think that Dodd --the SEC is the host, but --and 
I think you've done a great job, but --and I don't 
mean this to be disrespectful at all, but Dodd-Frank 
which required I think in some provisions for 
methodologies to be independently verified, if I 
recall right, has made life more opaque frankly. I 
do advisory work day in and day out. The criteria 
have become mechanically transparent, but in fact, 
more opaque. 
So, the greater transparency that is 
sought in certain segments has made it more like 
what I call a useful tranche in Boston Museum of 
Science there is a machine in the atrium --there 
used to be a machine in the entrance which had 1,000 
moving parts that just ran and did all kinds of 
stuff. And nothing came off of it. You could see 
everything moving, but no --no --no sense came out 
of it. 
I think it's become --you --you 
have mechanical rules, matrices, 140 different 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 121 
agencies understanding at least all of us and 
yourselves included, just what an important part 
they play in efficient markets now. Any negatives 
that you see that we should be aware of when you 
think about changing the compensation structure? 
MR. KOTECHA: Yes. 
MR. RABOY: Not necessarily 
negatives, but be very aware of the incentive 
natures that are going on here. Economic referred 
to is a principal agent problem. So, the agent is 
the pension fund. They're the ones who rely on 
this. If you have a compensation system, say, 
that's based on volume of ratings, these two don't 
align. So, whatever you do with the compensation 
system it has to be with an eye towards aligning the 
wishes and desires of both parties. 
MR. KOTECHA: Just quick, three 
points. One, the industry is global. When you 
preserve it here it effects what happens outside the 
U.S. You've seen that with '17 G5. So, whatever 
you do, please consult with your European 
regulators, Japanese regulators and others. Bring 
them on board so there's a buy in and --and input 
two ways. That's one. 
Two, if you start with structured 

31 (Pages 118 to 121) 


Page 122 Page 124 
1 finance, be aware that it will interrupt the non 1 institutional investors or some of the largest 
2 structured finance market. So, you need to have a 2 institutional investors on the board, you know, 
3 migration thought process that is clear cut and that 3 saying to issuers, come and get our rating. 
4 is --that is implementable. And, third, I would 4 Remember they're --they're buyers. It's --they're 
5 add a principal. GAO did seven principles for 5 buyers. And so, that's going to be a real pull. 
6 evaluation. I would add one more. Tested against 6 And not only that, the fact that 
7 minimal invasiveness because the industry, as you 7 their incentives are not profit motive incentives, 
8 said, has performed generally well. It screwed up 8 but more accuracy given, you know, dealing with the 
9 in some ways big time. 9 conflicts. Like I said before, right, I think that 
10 MR. BARRON: Big time. 10 will be a base point for other rating agencies or a 
11 MR. KOTECHA: Big time, thanks to 11 reference point for other rating agencies and make 
12 this man here in part. 12 them more careful with their internal compensation 
13 So, we need to fix the problem, 13 structures. 
14 but --but --but we need to allow the industry to 14 So, I don't --I don't think you -15 
function. I think it is a natural oligopoly. 15 you're going to need any --they're going to need 
16 MR. HEANEY: Thank you. 16 any help once they get started. 
17 Second really quick question. We've 17 MR. HEANEY: Thank you. 
18 talked a lot about the structured finance market 18 Larry Tabb, we'll turn to you for a 
19 generally as to how this is applicable. Any 19 question on the phone. 
20 negatives, just to roll it out broadly? I mean, the 20 MR. TABB: Yeah, thanks. 
21 plain vanilla ratings market has many of the same 21 First of all, when --when folks 
22 underlying issues that structured finance does. It 22 speak, can they speak in the microphone because some 
23 seems to me it would be fairly simple to make this 23 people come through clearly and other people -24 
change more broadly. 24 like, other people I can barely here. 
25 MR. MANNS: So --so, I think there's 25 This --this comment may be a little 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 123 
a virtue to having a pilot program in order to 
understand all of the complexities that come with it 
and the potential spill over effects. And 
especially --and so, I think I respectfully 
disagree with my colleagues about the rating agency 
industry being a natural oligopoly, but instead, the 
logic it would be with the pilot program, you 
provide an outlet for smaller rating agencies to 
show that they can viably compete with high quality 
ratings and establish a track record. Without a 
track record, there's no way that they can compete 
with the three leading rating agencies that dominate 
95 percent of the market. 
So, in that sense having it start off 
as a pilot program it helps to give some oxygen to 
the smaller rating agencies. It would also provide 
a context to understand, are there secondary effects 
that we aren't anticipating that could be 
significant. 
MR. RABOY: I agree with that. 
MR. BARRON: I think that if you 
started a brand new rating agency that met the 
requirements of IORA, of an investor owned rating 
agency, I don't think that we need any help. I 
think that the fact that you have the largest 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
1314 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25Page 125 
bit mute and I'm not sure depending upon some of the 
comments, but a couple of people ago there was 
someone talking about the issue of competition 
creating, in effect, a pressure to reduce ratings 
around, you know -- you know, reduce the quality of 
ratings. And when I think about competition in -in 
the rating -- in any type of market base, that 
somewhat tells me that folks don't necessarily need 
accurate ratings in the investment process. Because 
if people really needed an accurate rating in the 
investment process to get competition to increase 
ratings rather than decrease the quality of ratings. 
So, I'm not sure whether this is a 
comment or a question or whatever, but how do we 
get, you know, the need, you know, for better 
quality ratings into the investment process so the 
competition comes more from, you know, raising the 
quality of -- of ratings rather than issuer -- you 
know, or what seems to be the -- the business model 
associated with getting more raters or more issuers 
rating products with an incentive to in affect, you 
know, entice them with lower quality ratings. I'm 
not sure what the answer to that is, but at least 
that's from my external two cents. 
MR. MANNS: So, I think one way to 

32 (Pages 122 to 125) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 126 
try to tackle this is a theme that's come up in many 
of the models which is having performance based 
selection or compensation with a three-, five-or 
ten-year look back. And the logic of that being, 
what's missing in the current system is any 
incentive to necessarily go and elevate the quality 
of ratings over time. And that's just a critique of 
the fact that in an oligopoly there's going to be 
stickiness of reputation. There will be some 
accountability, but limited accountability. 
And under most of the models we've 
talked about, there at least is the consideration of 
trying to go and factor in performance. It's a 
contestable issue as to what metrics to use, but 
certainly would go about and advance your underlying 
concern. 
MR. KOTECHA: Yeah. I have to say as 
an ex-rating agency analyst, I remember somebody 
working for me who said, Mahesh, I love working here 
because I can say what is the truth. This is I 
think the creed of the rating analyst. Analysts 
like to know what is real. They like to call a shot 
as they see it and they get supported largely for 
doing so. 
So, I think that --that --that what 
1 
2 
3 
4 
56 
7 
8 
9 
10 
11 
12 
1314 
15 
16 
17 
1819 
20 
21 
2223 
24 
25 
Page 128 
whole idea that people don't want to pay for 
accurate information, you know, there's -- there 
must be some other issue here. It just doesn't make 
a whole lot of sense, at least to me. 
MR. HEANEY: I just want to make one 
comment on the back of Mahesh's comment. And the 
subcommittee would have heard this, but we had a 
representative from one of the large rating agencies 
who worked there and made, basically, the same 
question, and Amy correct me if I'm wrong, that the 
business side of it basically choked off the 
independence of the analysts themselves. 
And so, I just throw that out there 
anecdotally for the rest of, you know, FIMSAC to 
hear it. He was unable to make it here at this 
meeting, but perhaps he'll be at a future one, but 
he reiterated that exact point. 
MR. RABOY: That's not just 
theoretical or anecdotal. There's a lot of research 
that establishes that that the analytical side was 
basically cut out of the system. 
MR. BARRON: You know, when we first 
got to Fitch we had periodic lectures from the 
analysts that explained the impact of ratings on our 
financial system, on our economy, on investors. And 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 127 
happened or that the business side got hold of 
the --the golden goose. The analytical side 
remains very committed to quality. And I think some 
of the reforms that you're talking about regarding 
compensation, disclosure, some of the things that 
have been done, Chinese walls between the business 
side and the analytical side, they've helped, but I 
think overall, the --the agencies have done 
remarkable job because the default studies shows 
great correlation between defaults and ratings -low 
ratings. So, I would say that the industry is 
such that you cannot promote higher quality when 
somebody has to pay for it. 
You go to me and say, I want a 
rating. And I say, I'm going to give you a low 
rating because you're --I'm tough, but I want you 
to pay for me. The guy will go to the next guy. I 
mean, it's a natural situation. Nobody is going to 
pay for it. 
MR. TABB: Let me jump in here real 
quick. You know, Lynn's whole business in terms of 
her market data business really revolves around 
providing quick accurate data. So --so, you 
know --and her data is integral to valuation and 
integral to the pricing of everything. So, this 
1 
2 
3 
45 
67 
8 
9 
10 
11 
12 
13 
1415 
1617 
18 
19 
20 
21 
22 
23 
24 
25 
Page 129 
they took it seriously. And I -- and I -- this is 
amazing. This is the second time I'm agreeing with 
him in 40 years. 
MR. KOTECHA: It's beginning to be 
dangerous. 
MR. BARRON: But the business side as 
you said did take over. And I think it might be a 
good idea to require rating agencies to make their 
analysts aware using 2008 as an example of what can 
happen in our country for bad ratings. There should 
be another obligation to lecture analysts so that 
rating agencies can avoid the systemic risk that 
they did not avoid in 2006, 7, 8, whenever it was. 
MR. HEANEY: Let me go to Rachel, 
please. 
MS. WILSON: So, a lot of discussion 
around moving from the oligopoly essentially to 
assigning readings and getting a more broader group 
involved in that process and talking a lot about 
quality though and the need -- there's been kind of 
some conflicting thoughts from the panel around 
fewer is better and quality is better with fewer. 
And then, this idea of competition. We did have one 
of our panelists say that a pilot would be a way to 
check if you're compromising quality relative to 

33 (Pages 126 to 129) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 130 
having that develop over time. However, as an 
issuer, do you want to be the one in that pilot? 
You know, no, okay. That's their answer. No, I do 
not want to be the one in that pilot. 
And, you know, the other thought of, 
well, another approach is you can more gradually get 
there by saying, you know, it's almost like a bo/go 
model, you get one assigned to you, you can go buy 
your other one. One free, the other, go buy. 
Does that though then really break 
the conflict issue or have you just created more 
complexity. 
So, I just wanted to throw that back 
because I think the concept is very appealing 
intellectually and you all did start your talks, all 
of you said, pragmatically and practically 
implementing though is another matter. So, I want 
to get us back to that with some final thoughts. 
Thank you. 
MR. MANNS: So, if I may just kind of 
address your second point. 
I think the --the model of allowing 
issuers to go and get whatever ratings they want, 
there could be no limitation on that. And so, in 
the sense that's implicit in any approach we're 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 132 
Standard & Poor's to go about and provide ratings in 
addition to whatever the pilot program produces. 
MR. RABOY: I agree with all of that. 
I don't think it's a natural oligopoly. I think it 
is an oligopoly, but that isn't necessarily there 
for future. Reputation matters. If any of these 
systems are going to work it has to be the case that 
reputation affords profit. And the pilot program 
has an incentive for people to participate because 
that does enhance reputation for trading. 
And unless you --if you have 
successfully broken the conflict, then reputation 
matters. And that matters very much to new entrants 
because that's the way competition occurs. 
MR. KOTECHA: May I suggest a couple 
of quick ways to --to start our little experiment. 
You could say, pick an asset class which is not the 
biggest, not the smaller, and say, we're going to 
start there. It could be CRE, CMBS or VWAP, RMBS, 
just pick some class and you could try. So, that's 
one possibility. 
Another, the issue is oligopoly. If 
you have five, that's still an oligopoly. And I 
think you might end up with five. KBRA is making 
progress. Morningstar is making progress. If you 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 131 
going to have. And in that sense, it would raise 
the cost for whoever in is the pilot pool. They 
might not be overly excited about that for sure, but 
we then have an attractive problem otherwise, which 
is, assuming my --the folks on my left, though they 
may be on my right depending on how you put them on 
the political spectrum, but the folks on the left 
basically say it's a natural oligopoly and it 
becomes a talkology, which is, three rating agencies 
control 95 percent of the market 10 years ago, they 
still do, they still will. And it's --the logic 
is, well, they've made some mistakes along the way. 
This is the simply the natural order of things. 
Without some type of pilot program 
initiative, we can't really have any kind of real 
world market test. We won't be able to have enough 
data out there to go about and show whether, given 
the proper incentives, abolishing the issuer pay 
conflict of interest, could we have viable 
competition that elevates the quality of ratings. 
And I think that's kind of the catch 22. 
And so, in that sense you're right, 
it effectively would be a tax on the folks that are 
going to be subject of the pilot problem, but they 
can still go about and get Moody's, Fitch or 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 133 
five, is it still not an oligopoly? It is still an 
oligopoly. So, I think the --the point is that how 
many opinions do you need that you're about to die 
of cancer? You need maybe two or three. After that 
you say, hey, I've had enough. I go for the 
treatment. 
So, I think it's a natural oligopoly 
because the opinions that are credible are going to 
be limited and their value diminishes with more 
opinions. 
MR. BARRON: You know, one thing, 
again drawing off the experience of Fitch that will 
help new rating agencies, when we first took over 
Fitch, Standard & Poors and Moody's had spent hardly 
any time going out to investors and talking to them. 
We went out --I personally probably went out and 
met with 60, 70, maybe 80 different institutional 
investors. And what we did was we explained our 
model. We explained our analysis. We explained 
compensation structures internally, how they worked. 
And many of the institutional investors at that 
time --this was, again, in '89, you know, didn't 
really understand how mortgage-backed securities 
worked. And we --we explained it and showed our 
criteria and why we think our ratings are going to 

34 (Pages 130 to 133) 


Page 134 Page 136 
1 be accurate. 1 Model' applied from equities into debt. Basically, 
2 Having done that, they started 2 you say that if you have, you know, equity prices go 
3 calling up the bankers and saying, we like Fitch. 3 up then your balance sheet is more able to pay for 
4 We want to get some Fitch ratings. So, even though 4 debt. 
5 I have said that, you know, competition begets lower 5 And then from there you -- you do 
6 standards, generally speaking, I think that if 6 mathematics, fancy mathematics, and take the market 
7 you're going to have a pilot program, any new and 7 prices of your stocks and come up with a default -8 
upcoming rating agency has to have a lot of contact. 8 default rate. What happens with those models is 
9 And it's not like panelists. It's not like, you 9 that they are much more timely. Every time the 
10 know, lecturing. It's really one-on-one with 10 market jumps up and down, the default rate goes up 
11 investors to educate them about what they're doing. 11 and down. So, they -- the apply two errors. 
12 MR. KOTECHA: Let me adjust one 12 They've cried wolf and the wolf doesn't come. So, 
13 footnote and I'll shut up. 30 seconds. 13 type one they predict better, but they also predict 
14 There was an agency called Duff and 14 things that don't happen. So, that's the problem 
15 Phelps. I called them easy ratings. They made a 15 with rapid changes in ratings. It's more timely, 
16 reputation as a small potatoes entity in utilities. 16 more accurate, but frankly, it's noise. 
17 They become very, very successful by going, like you 17 MR. HEANEY: Tom Gira. 
18 did, Neil, talking to investors, doing excellent 18 MR. GIRA: I just want to go back to 
19 reports and their reputation in utilities lead them 19 the oligopoly issue for a second. 
20 to other sectors. 20 This is more for Professor Manns, but 
21 The moment they begin a cafeteria 21 I have issues here that others might have on this. 
22 rating agency, not a uni --not a mono-line rating 22 You made the good point I think about raising the 
23 agency, they lost it because they then --they 23 accountability standard -- the reasonable 
24 became easy ratings because they couldn't get the 24 accountability standards. And I think that makes a 
25 business. And then, NAIC said, one NRSRO rating 25 lot of sense, but do you have any concerns though 
Page 135 Page 137 
1 and, boom, if it's investment grade, you can buy. 1 that by doing that it could entrench the oligopoly 
2 So, they --that become their model to a 200 million 2 or --or make it --make the white elephant even 
3 IPO and a life on the beach. 3 bigger and how would --how would you sort or 
4 MR. HEANEY: Thank you. Lynn. 4 respond to that. 
5 MS. MARTIN: Yeah. As Larry 5 MR. MANNS: Sure. No, I think 
6 mentioned, I --I do run a database and my database 6 that --that is a significant issue. I think the 
7 is an operated rating agency. 7 idea of trying to cap the damages to a multiple of 
8 I guess one thing that I heard from 8 fees would be the design that would go about making 
9 the different proposals --or one thing that I think 9 a potential liability exposure manageable for even 
10 was missing from the different proposal is, a way to 10 smaller entities because I think the underlying 
11 really democratize this is to leverage observable 11 concern about expert liability under Section 11 was 
12 data. And that's typically what we do in --in my 12 just, as we know from Rule 10b 5 and Section 11 
13 business and we don't really have a subjective 13 assumes the expert liability to be monstrous and 
14 nature. 14 that --that truly could be debilitating to smaller 
15 Is there observable information that 15 rating agencies in terms of competition. 
16 perhaps could democratize this process and level the 16 If you have an SEC adjudicative 
17 playing field as opposed to the subjectivity? 17 process where you try to go and limit the expense of 
18 Because I think that's what you all are really 18 the adjudicative process itself and cap the damages, 
19 getting at, the issue is the conflict between who is 19 then I think it would be something that would be 
20 paying for subjectivity. 20 more manageable for smaller entities to do. 
21 MR. KOTECHA: There is a rating 21 MR. HEANEY: Sonali. 
22 agency kind of a model, KMV, comma, Corp. Those are 22 MS. THEISEN: I have a quick question 
23 the two that are market driven data base ratings. 23 around the pilot --the proposed pilot program. 
24 There is a mathematical, you know, kind of 24 So, Rachel's questions centered I 
25 discipline that can lead from the Black-scholes 25 think a little bit around, you know, incentivizing 

35 (Pages 134 to 137) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
1314 
15 
16 
17 
18 
19 
20 
21 
22 
23 
2425 
Page 138 
issuers to participate, etc. What about on the 
rating agency side. So, let's just say that you 
could construct a pilot. I would assume that there 
will be, you know, many newcomers that want to earn 
their stripes through this process. So -- so, how 
would you go about selecting -- I mean, if you have 
too many -- obviously, if you have too many people 
in that pool, you're not going to get enough kind of 
information to compare them. That's -- that's 
apples to apples. So, how would you propose going 
about constructing the selection process for the -for 
the newcomer rating agencies themselves? 
MR. MANNS: I think as a starting 
point for all of the models, as I would understand 
it is, being an NRSRO. In other words, going 
through it being certified with that status to be 
the starting point you then would be able to go 
participate in the pilot program. Currently I 
believe there are 10 NRSROs that exist. Obviously, 
if the pilot program existed and were -- had a 
longer time horizon, more entities would go and try 
to get status, but it's not an easy --it's not a 
rubber stamped process to become an NRSRO. 
So, effectively what that is is the 
SEC setting minimal standards. You have the 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 140 
MR. HEANEY: Larry. 
MR. HARRIS: I'm going to make some 
general observations about this area and then 
conclude with a suggestion for yet another model 
that we haven't discussed before. 
So, we're meeting today on this issue 
and have met before over the years because there's a 
need for reliable, unbiased, informative research 
that's easy to digest and doesn't facilitate 
manipulative behaviors. I think that's the 
definition of our objective here. 
And we know with respect to credit 
rating agencies that there are two big economic 
problems that make this very difficult. The first 
one is the free writer problem. Everybody would 
like somebody else to pay for it. This one actually 
has a simple, well-known solution. There's a free 
writer problem, basically, you compel everybody to 
pay for it one way or the other. And then the 
question is, how do you manage that money. But the 
much more difficult problem is the fact that the 
product is extremely difficult to evaluate. 
Defaults just don't happen very often. And in 
evaluating the product by whether your ratings 
change when everybody else's ratings change is not 
Page 139 Page 141 
1 credibility to go about issuing credit rating 1 the right way to do it. That's self-referential. 
2 agency. So, I think in that sense, that would be a 2 And this is a fundamental problem. It's hard to get 
3 meaningful barrier and would be a meaningful 3 around. 
4 preliminary step for anyone who would take the 4 So, we could imagine -- we could 
5 ratings seriously. 5 imagine that one solution to the problem, and it 
6 MS. THEISEN: So, would you believe 6 actually already exists, one solution to the problem 
7 that all 10 of those NRSROs if a pilot conducted is 7 is to create an information market where people can 
8 that --is that too many or do you think that's 8 bet on -- essentially on ratings or on quality of 
9 manageable? 9 information. So, if you think that the rating is 
10 MR. MANNS: I think part of it would 10 too low, you can buy it and sort of push it up. If 
11 really turn on what segment in the marketplace 11 you think the rating is too high then you can sell 
12 you're going to have the pilot program in. And not 12 it. That's actually a description of the credit 
13 all 10 of those NRSROs are participating all 13 default swap market, okay. 
14 segments in the market, at least currently. They 14 So, with that observation, let me 
15 may have reasons to go about and shift their focus 15 note that there are three places where we discovered 
16 to be able to participate in the pilot study, but I 16 information about the value of fixed-income 
17 think the underlying issue simply is not a matter of 17 securities. We have the credit rating agencies, and 
18 three, five or ten. It's a matter of trying to 18 we're all quite familiar with it, we've spent a lot 
19 provide enough oxygen in the air for folks who are 19 of time talking about that. We have the markets for 
20 offering credible ratings to be able to go and 20 those securities where people who have opinions who 
21 participate in the process. And the only way you 21 will buy or sell based on the price. And then, of 
22 build credibility is by having enough business that 22 course, the credit default swap market where it's 
23 you can empirically test in retrospect how credible 23 essentially the same thing, but where there's less 
24 your ratings were. 24 capital required. 
25 MS. THEISEN: Thank you. 25 So, the suggestion is is that -


36 (Pages 138 to 141) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
1415 
16 
17 
18 
19 
20 
21 
22 
23 
2425Page 142 
well, first, the next observation is that the latter 
two markets, the markets for information, where 
people are trading the securities or they're trading 
the swaps, those are places where people put their 
money on the line when they're forming an opinion, 
which is really disappointing. It -- it keeps them 
from lying. You lie and you take a position in 
your -- in your -- in your statement, if you're 
wrong, you're going to lose a lot. So -- and of 
course the academic evidence suggests that those 
markets actually produce better information. Though 
the information tends to be a little bit more 
volatile. 
Here's the suggestion, we presently 
have a regulatory framework, the NRSRO framework, 
that's designed to deal with the problems as we 
recognize them in the existing credit rating agency 
world. And that world is a world in which you have 
a bunch of analysts who collect fundamental 
information. They go talk to managers. They get 
paid this way or that way. They process it and 
produce ratings. As Lynn mentioned and 
Mr. Kotecha -- did I pronounce it correctly? 
MR. KOTECHA: Correct. 
MR. HARRIS: Yeah, elaborated, we 
1 
2 
3 
4 
5 
6 
7 
8 
910 
11 
12 
13 
1415 
16 
17 
18 
19 
20 
21 
22 
23 
2425 
Page 144 
derivative fourth actually, source of information 
that might be -- might be pretty useful. And we 
solve the volatility problem by recognizing that to 
the extent that there's negative zero correlation in 
ratings you can smooth that out. And if there's not 
than you just accept that the world changes and 
changes are what you want to see. And so, you have 
the volatility. 
So, I guess the question that comes 
out of this is, is there a strong reason to suspect 
that such a proposal like this would generate -- it 
certainly would generate its own problems. The 
question is, what would those problems be. 
MR. KOTECHA: Yeah, I'm not sure I 
would support a rating agency that's just based on 
CDS spreads or other market spread. Because I think 
the issue is, you pointed out, the volatility is too 
high. I actually advised on a transaction where CDS 
spreads that would be used in order to impute 
credit -- credit risk in a CDO and it just -- the 
CDS spreads not only have credit risk, they have 
liquidity risk and they have other factors that come 
in. 
So, they are rather volatile because 
many other facts are reflected in there than simply 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 143 
know that it's possible to transform price 
information into easy-to-digest statements about -about 
the future default, but --but we don't have a 
way of sort of overseeing people who make those 
statements. And it might be sensible to have such a 
thing. 
So, the suggestion is is that perhaps 
we should have a --another designation that the SEC 
recognizes which is parallel to the NS --the NRSRO 
designation that applies only to those entities that 
are taking price information and translating it 
to --into easily to --easy-to-digest information. 
The translation is not easy because it may involve a 
Black-scholes Model' or other models. It's going to 
involve financial engineering or the understanding 
of the characteristics, but it's based fundamentally 
only on --on the price information and the quality 
of the translation. And the regulator, whoever it 
may be, maybe the SEC or some board or something 
like that, simply says that you're qualified to do 
this if we give you this designation because we know 
that you understand the --the financial engineering 
well enough to create a --a reasonable statement. 
And then we would allow those 
entities to compete as a third or sort of a 
1 
2 
3 
4 
5 
6 
7 
8 
910 
11 
12 
13 
14 
15 
16 
17 
18 
1920 
21 
22 
23 
24 
25 
Page 145 
credit. And separating them out is a complicated 
job and the volatility is a big, big job. So -- and 
you don't need a license for that. The market is 
quite broad. People can use it and they can -- they 
can short a security that they think is, you know -you 
know, just going to go down in the future. So, 
I don't think you need that today to -- to have a 
newly designated rating agency for that. 
MR. HARRIS: So, volatility is 
something that I've studied a lot. There are two 
types of volatility. There is volatility that comes 
from fundamentals and though we don't like, we're 
probably best off seeing it. And the other 
volatility comes from bouncing back and forth. 
Things go up and then they bounce back down and so 
forth. That volatility is called negative zero 
correlation or associated with negative zero 
correlation. 
Financial engineers understand how to 
separate the two types of volatility. And the way 
to deal with the negative zero correlation is simply 
to smooth it out. And so, there are financial 
engineering techniques that could deal with that 
issue. And so again, the question is, what other 
problems might we see from -- I don't want to 

37 (Pages 142 to 145) 


Page 146 Page 148 
1 dominate this -1 
The first --the first is normal bonds are trading 
2 CHAIRMAN HEANEY: Unfortunately, I'm 2 in terms of portfolios as a basket. So, often 
3 going to have to just cut this off. We're ten 3 these --these trades are won or lost in terms of 
4 minutes over as it is, but I appreciate it. 4 the best price in the basket, but it may or may not 
5 A big thanks to the panelists. 5 have the best price on the individual components in 
6 Fantastic. This is a heavy topic obviously for the 6 the basket, the winning trade. So, to put a flag to 
7 subcommittee and for FIMSAC as a whole, but this has 7 identify the trade before. 
8 been eye opening. I appreciate it. 8 The other one of course is corporate 
9 And thank you, Amy, for moderating. 9 bonds often trade in terms of spreads and the 
10 (Applause.) 10 discrepancy as spreadable treasuries to be marked at 
11 CHAIRMAN HEANEY: So, we are going to 11 a certain time. And between the time of the 
12 move right away into the --the two updates. Rick 12 trade --between the time the trade was negotiated 
13 McVey first who chairs the Technology and Electronic 13 and the time the trade was executed spreads moved, 
14 Trading Subcommittee will give a brief update on 14 prices moved as a better way to better capture 
15 what that subcommittee is thinking and working on. 15 spread trades that are negotiated one time and 
16 And then, we'll move to Mihir to do the same. 16 executed at another time. 
17 MR. McVEY: Thanks, Michael. 17 And the third one is to try and 
18 The Technology and E-Trading 18 capture and trace riskless trades and mark ups, 
19 Subcommittee continued its ongoing support for 19 whether it be the Commission or another flag talking 
20 previous FIMASC recommendations on pennying at a new 20 about when a broker-dealer buys a bond and turns 
21 issue corporate bond reference database. On the 21 around --turns around and sells it instantly 
22 corporate bond new issue reference database market 22 without taking risks. So, that's --that's in terms 
23 participants have had ample opportunity to provide 23 of what we're looking at for modifying TRACE 
24 comments and FINRA continued to refine their 24 reporting. 
25 proposal. 25 The second issue we worked on was 
Page 147 Page 149 
1 In addition, the committee has 1 revisiting the --the block trade proposal. As many 
2 entertained a number of new topics without any firm 2 of you might recall, one of the first proposals that 
3 conclusions to date. Including, critical market 3 came out of FIMASC or our subcommittee is modifying 
4 infrastructure, accessibility and pricing models, 4 the dissemination rules for large block trades and 
5 cyber security education and best practices, 5 to see if that improves liquidity and execution 
6 especially targeted for municipal bond issuers, the 6 costs for large blocks. 
7 use of composite pricing services for portfolio and 7 So, FINRA put out a proposal. It's 
8 package trades and potential improvements to TRACE 8 been out for comment. And the reason we decided as 
9 reporting for these trades when conducted on 9 a subcommittee to revisit this is we've had the 
10 electronic trading venues. 10 comments out there. We've got the FINRA proposal. 
11 And, finally, we have had recent 11 We've had input from other market participants and 
12 interest from asset managers to consider a 12 we've seen the comments that people have officially 
13 recommendation to modernize SEC regulations for 13 put forth to FINRA. And we thought as a 
14 internal cross trades utilizing independent pricing 14 subcommittee maybe we should revisit this and see if 
15 services on electronic trading venues. The 15 we can come up with a consensus to try and move this 
16 committee hopes to reach consensus on one or more of 16 issue forward. 
17 these topics in the coming quarters. 17 Both of these in hindsight, maybe if 
18 MR. WORAH: So, an update from the 18 we focused on one rather than two issues, we'd have 
19 Corporate Bond Transparency Subcommitte. 19 a recommendation today, but where we are, we've made 
20 So, over the last seven months we've 20 progress on both of these issues, but no 
21 been working on two parallel topics. The first one 21 recommendation today. 
22 relates to, is there a way to modernize the data 22 So, that's an update from us. 
23 collected by TRACE either in terms of more fields or 23 MR. REDFEARN: Okay, thank you very 
24 flags to better reflect the way bonds trade today. 24 much. 
25 And this -- this breaks down into three categories. 25 I think we are going to break for 

38 (Pages 146 to 149) 


Page 150 Page 152 
1 lunch. For those of you who are visitors here, you 1 remarks, you know, on each person. 
2 should know that there is a cafeteria one flight 2 So, starting alphabetically, we have 
3 down from this. I think it's the -- the American 3 Bill Ahmuty from State Street Global Advisors, Steve 
4 Express cafeteria, whatever, or you can obviously go 4 Berkley from Bloomberg Index Services, Phil Galdi 
5 out to Brookfield Place. 5 from ICE Data Indices, Anish Karyat from Citadel 
6 And for those of you on the panel I 6 Securities and Dan Li from the Federal Reserve 
7 think we have a -- or the committee we have a -7 
Board. 
8 right down the hall we have a room. 8 So, with that, let me turn it over to 
9 CHAIRMAN HEANEY: 409 they're saying. 9 our panel for remarks. Maybe we'll begin with Steve 
10 MR. REDFEARN: And then we will -- we 10 and -- and Phil and -- from the index provider 
11 will assemble back here at 12 -- I think we're on 11 viewpoint. And then, proceed to hear from Bill 
12 for 12:40. 12 about the, you know, indices from an issuer 
13 (Lunch recess taken.) 13 perspective. And then to Dan Li from the Federal 
14 A F T E R N O O N S E S S I O N 14 Reserve on -- on kind of her perspectives from the 
15 CHAIRMAN HEANEY: Okay. Welcome 15 market side and then close out with Anish from 
16 back. Before we move back to the next subcommittee 16 Citadel about the trading of -- of bonds around bond 
17 I'd like to introduce Ronald Tidal from the Division 17 and reconstitution events. 
18 of Economic Risk Analysis. He will be joining us 18 Okay. So, with that, let me turn it 
19 for the afternoon session. Welcome. 19 over to our distinguished panel. Thank you. 
20 Our first panel this afternoon will 20 MR. BERKLEY: All right. Thank you. 
21 address matters related to index construction and 21 Well, I'll be the canary here. 
22 reputation methods and disclosures and the daily 22 Good morning. I'm Steve Berkley, CEO 
23 pricing of induces. 23 for BISSELL which is Bloomberg's Index Service 
24 With that, let me turn it over to 24 business and I'm happy to be here. 
25 Ananth the chair of the ETF and the Bond Fund 25 I look at today's meeting as a way to 
Page 151 Page 153 
1 Subcommittee who will be moderating the panel. 1 educate people on how fixed-income indices are 
2 MR. MADHAVEN: Great. Thank you very 2 constructed and the differences between bond and 
3 much, Michael. And thank you for our panelists 3 equity benchmarks. 
4 for -- for coming today to inform us about index 4 Bloomberg Indices have been published 
5 construction within fixed-income. 5 since 1973, albeit, under the organizational names 
6 Before I introduce the panel, just I 6 of Kuhn Loeb, Lehman Brothers and Barclays. And if 
7 wanted to go back to Chairman Clayton's remarks this 7 you leave here with one memory I hope it is the 
8 morning when we kicked off the agenda. In terms of 8 knowledge that the fixed-income indices are rules 
9 index construction, Chairman Clayton highlighted 9 based --thank you. Thank you. 
10 three areas for focus. One had to do with technical 10 If you leave here with one memory I 
11 issues such as the weighting schemes that are 11 hope it is the knowledge that fixed indices are 
12 applied within index construction. The second had 12 rules based and transparent. A little feedback. 
13 to do with market exposures, some of the risks and 13 That means there are no subjective 
14 opportunities provided by indexation within 14 decisions made as to which bonds are added or 
15 fixed-income where it's still MASINT. And then, the 15 dropped from the index. There are no committee 
16 third, which I thought was a very interesting remark 16 decisions upon which the world waits. Just like the 
17 was the question about value judgments on what 17 boardgames we all played as kids, the rules are on 
18 should be included and what should be excluded out 18 the back of the box. 
19 of bond indices. 19 So, what are these rules. Like most 
20 So, these are very large questions. 20 fixed-income indices, including those of other index 
21 I think we have a number of very technical questions 21 providers, they are a set of criteria that must be 
22 as well about this. And we have a -- a all star 22 met. A key example includes a size constraint. For 
23 panel here to answer some of those questions. Let 23 example, the U.S. aggregate index, all corporate 
24 me introduce them briefly and then I'm going to turn 24 bonds must have at least 300 million paramount 
25 it over to our panel to make some introductory 25 outstanding. All securities for this index must 

39 (Pages 150 to 153) 


1 
2 
34 
5 
6 
78 
9 
10 
11 
12 
13 
14 
15 
16 
1718 
19 
20 
21 
22 
23 
24 
25 
Page 154 
also be rated investment grade. We use Moody's, S&P 
and Fitch in this algorithm. 
In the case of split ratings we'll 
use the middle of the three as the determining 
factor. If only two agencies rate the security, we 
use the lower of the two. 
We also have exclusion rules. For 
example, private placements are not eligible to be 
included in the U.S. Act, neither are floating rate 
securities. These rules can and have changed over 
the years. We've addressed -- we've added asset 
classes such as U.S. MSBS, ABS and CMBS. We've 
added a rating agency as Fitch was not initially 
part of the ratings prior to. We've raised the 
amount outstanding constraint as well over the years 
to reflect changing issuance patterns. 
The point being that markets evolve 
and the indices evolve a long with them. 
Philosophically, the indices are designed to 
represent and measure the marketplace. There are a 
number of key differences between equity and 
fixed-income indices. People are always questioning 
the higher turnover in bond benchmarks. There's a 
really simple answer. Bonds are born and bonds die 
at issuance and maturity respectively. It is the 
1 
2 
3 
4 
5 
6 
78 
9 
1011 
1213 
1415 
16 
17 
18 
1920 
21 
22 
23 
24 
25Page 156 
all the 11,000 bonds in the U.S. So, managers 
typically use a sampling approach where they match 
the risk characteristics of their chosen benchmarks. 
These risks factors include things such as duration 
and exposures along the curve, sector weights, 
quality distribution, among others. 
I think I'm out of time. So, I'm 
going to stop there and I'll turn it over to my 
colleague Phil. 
MR. GALDI: Thanks. I'm Phil Galdi 
at ICE Data Indices which covers fixed-income -CHAIRMAN 
HEANEY: I'm sorry. Can the 
panelists just all put their green lights on. 
MR. GALDI: Fixed-income, equity and 
commodity indices. Prior to that I was running the 
index business at Bank of America/Merrill Lynch for 
over 30 years which was then acquired by ICE two 
years ago. 
With the significant growth in 
fixed-income UTFs that we've seen over the past few 
years it's important that the indices that support 
these products be well understood. Over the years 
many changes have taken place in index construction, 
but basic principles are the same. 
First, as Steve explained, the 
Page 155 Page 157 
1 circle of fixed-income life, so to speak, but --but 1 universe of outstanding bonds is filtered using an 
2 equities live forever. 2 objective set of criteria relating to terms and 
3 Most bonds are issued in the two-to 3 conditions such as maturity and issue size and other 
4 30-year range, but some stocks have existed for 4 factors such as credit rating and sector 
5 decades longer than that. So, it's easy to see why 5 classifications. Second, there's a process for 
6 equity index constituents appear more stable. 6 determining the weight of each security in the 
7 A second big difference between the 7 index. 
8 two asset classes is how they're traded. Stocks 8 Together these two process comprise 
9 trade on exchanges while bonds are largely OTC. 9 the index skewed rules. We make these rules 
10 This results in a complication regarding pricing. 10 publicly available on an unrestricted basis in 
11 Investors may disagree on the value of Apple stock, 11 accordance with the IOSCO Principles for Financial 
12 but there is no uncertainty as to its price. You 12 Benchmarks. Having determined the index 
13 can find it throughout the day on TV and the 13 composition, constituents are priced daily and index 
14 internet. 14 results are compiled in accordance with their 
15 Bond pricing is a bit more difficult 15 published methodologies. 
16 because there are many inputs and observations that 16 Finally, the indices are broadly 
17 go into establishing an end-of-day price. These 17 disseminated with index values and statistics 
18 include TRACE feeds, bid and offer lists, issue of 18 available on an unrestricted basis. And in 
19 curves, models, et cetera. 19 addition, for indices used by the ETFs, constituent 
20 The last thing I want to comment on 20 holdings are also made available to the general 
21 in my opening statement are the differences in 21 public. 
22 techniques used by passive equity and fixed-income 22 Most bonds don't trade every day and 
23 investors. Equity managers can actually buy all the 23 as a result in most cases we are using what's 
24 constituents of their benchmarks to minimize or 24 referred to as evaluated prices that we license from 
25 eliminate tracking error, but it's impossible to buy 25 our affiliate ICE Data pricing reference data. They 

40 (Pages 154 to 157) 


Page 158 Page 160 
1 also use these same prices and license them to the 1 Securities cease to exist and need to be replaced by 
2 funds to start their NAVs. 2 new issuance, but it does raise a question regarding 
3 The rules for the earliest 3 the removal of bonds that have less than a year to 
4 fixed-income indices were developed back in the '70s 4 maturity. That rule may be fine for active managers 
5 and, basically, tried to emulate the investment 5 who use the indices, but it's less than ideal for 
6 arena for the typical ERISA portfolio. The thinking 6 passive fixed-income managers who would generally 
7 at the time was that the investments were allocated 7 prefer to avoid selling bonds before they are 
8 to three main asset classes; stocks, bonds and cash. 8 redeemed. 
9 To draw a line in demarcation between bonds and 9 As a result, the rules of many of the 
10 cash, most bond indices adopted a one-year minimum 10 indices that we license to ETF issuers do not 
11 maturity requirement and bonds were removed once 11 include a one-year minimum maturity requirement and 
12 they fell within a year of their final maturity. 12 they do hold their bonds to maturity. Some indices 
13 I'll come back to that point in a minute. 13 used by ETFs would naturally have higher turnover 
14 Most ERISA portfolios have 14 because they are trying to capture specific 
15 restrictions on credit ratings. So, the indices 15 investment opportunities such as our fallen angle 
16 were limited to investment grade rated bonds. High 16 index. That index will have higher turnover of 
17 yield indices didn't enter the scene until the '80s. 17 about ten percent of the year relative to the high 
18 Rather than relying on a single 18 grade index that I described, but that's due in part 
19 rating agency, ICE Indices used an average of 19 to the migration of issuers below and then back 
20 Moody's, S&P and Fitch as explained in our published 20 above the investment grade line. 
21 rules. Our method may differ slightly from others 21 Some believe that this segment of the 
22 for determining whether a bond is investment grade 22 market is attractive from a relative value 
23 or not, but the resulting differences are relatively 23 perspective and they are willing to accept more 
24 minor. 24 turnover in order to capture that investment 
25 The indices have a minimum issue size 25 opportunity. 
Page 159 Page 161 
1 criteria to avoid selection of less liquid bonds 1 Another consideration for indices 
2 while still capturing the majority of the market. 2 used by ETFs is diversification. I initially stated 
3 The earliest bond indices were cap 3 that most of the well-known fixed-income indices are 
4 weighted which means that each constituent was 4 market cap weighted, but that can result in issuer 
5 weighted relative to its outstanding issue size. 5 concentrations that do not meet continuous listing 
6 That holds true today for many, but not all indices. 6 requirements for fixed-income ETFs and their 
7 For most bond indices, the selection of weighting is 7 underlying indices. 
8 securities, which is referred to as the index 8 As a result, we construct many 
9 rebalancing, occurs monthly. That's different than 9 indices that are constrained to meet or exceed the 
10 equity indices which occur much less frequently. 10 continuous listing requirements rather than simply 
11 You might conclude that bond indices 11 defer to market value. 
12 therefore have a necessarily high turnover, but the 12 To sum up, we have thousands of bond 
13 more frequent rebalancing is necessary due to the 13 indices and they are used for many purposes from 
14 fact that the bonds mature or get called and 14 performance benchmarking for active funds to the 
15 mortgages have principal payments. 15 underlying for passive ETFs. Not every index is 
16 Over the last ten years our U.S. 16 ideal for every one of these purposes, but when it 
17 broad hybrid index has seen on average about 20 17 comes to passive ETF community, we and others have 
18 years annual turnover. Over half of that is 18 the market data methodology and technology to 
19 removable bonds that had been redeemed and no longer 19 construct indices that meet their needs. And we 
20 exist. And most of the remainder are bonds that 20 provide full transparency to investors in those ETFs 
21 fall under a year to maturity. Only about five 21 as to the rules and methodologies used to compile 
22 percent of the turnover is attributed to other 22 the underlying index and its owners. Thank you. 
23 factors such as rating changes. 23 MR. MADHAVEN: Thank you very much. 
24 So, as you can see, the majority of 24 We'll continue with Bill from State Street. 
25 the turnover in bond index is --is unavoidable. 25 MR. AHMUTY: Thank you very much. 

41 (Pages 158 to 161) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 162 
Bill Ahmuty from State Street Global 
Advisors. I'm here representing the SPDR 
fixed-income ETF business. This is an overall 
business for the SPDRs is $670 billion in assets 
under management and the fixed-income business in 
the U.S. is over $72 billion in fixed-income ETF in 
either assets we're managing as a direct manager or 
by employing sub-advisors. 
For the purposes of this session my 
comments are going to be focused on fixed-income 
ETFs managed in-house at State Street Global 
Advisors which are primarily index strategies and 
represent roughly $57 billion in fixed-income 
assets. 
It's important to note that the 
portfolio managers who oversee the SPDR fixed-income 
ETFs are part of the broader SSGA fixed-income beta 
team which managers over $400 billion in 
fixed-income assets. This allows the team to 
leverage the tools and techniques across various 
portfolios regardless of the wrapper. 
The purpose of providing the panel 
with some of these details is to provide context in 
terms of our capabilities in fixed-income 
indexation. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 164 
the opportunity cost of not owning a position versus 
owning a position. 
As it relates to fund disclosures and 
transparency, all of our ETFs provide daily holdings 
on our website and other third-party sites along 
with the index holdings of the said benchmark index 
on our website. As markets continue to evolve we 
look for consistency in index and pricing 
methodology and increased transparency in the 
marketplace. Thank you. 
MR. MEDHAVEN: Thank you. 
Let's turn to Dan Li. 
MS. LI: Hi, I'm Dan Li from the 
Federal Reserve Board. I want to thank the FIMSAC 
committee and to Ananth and Kumar for inviting me to 
this panel. 
Before I start, I'd like to make a 
disclaimer that the views here are my personal ones 
and do not --does not represent that of the Federal 
Reserve Board. 
Some of my recent research have been 
focusing on the role of institutional investors in 
the OTC bond market. In one of my published paper, 
I and co-authors found that corporate bond funds 
tend to buy and sell bonds in a correlated fashion 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 163 
Our goal as an index manager is to 
provide the returns of the investors of the 
benchmarked index in a manner that is consistent 
with the overall risk profile and stated benchmark. 
Broadly speaking, our approach is to replicate the 
risk profile of this index which, if managed 
effectively, will ultimately deliver these return 
streams. 
Due to the OTC nature of fixed-income 
markets, full replication on a CUSIP by CUSIP basis 
as a technique is neither practical nor prudent. 
Replicating and index strategy from the bottom up on 
a CUSIP by CUSIP beta level. Typically it results 
in the manager being a price taker in the market and 
can result in high costs to the fund. 
In addition, in certain indices where 
the CUSIP count may be in the thousands, full 
replication by CUSIP count would become too onerous 
and costly. We aim to replicate the risk profile of 
the fund using stratified sampling not full 
replication on a bottom up CUSIP by CUSIP basis. 
The key risks we look to identify in 
managing the portfolios are credit spreads, ratings 
and sector exposures. In measuring costs we look at 
the implicit cost of the bid/ask spread, as well as, 
1 
2 
3 
45 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
1718 
19 
20 
2122 
23 
24 
25 
Page 165 
which we call herding. We found that when they herd 
to sell corporate bonds price get temporarily 
depressed and may take a few quarters to revert. 
In two other working papers of mine 
we find that when mutual funds are forced to 
liquidate corporate bonds due to investor 
redemptions, so-called fire sales, they exert price 
pressures and may even cause second round affect 
where -- where the returns of the funds owning the 
same bonds will suffer triggering outflows from them 
as well. And, in a nutshell, bond funds investment 
strategy have important financial stability 
implications. And this could be particularly true 
when more funds own similar bonds. Something that 
the growing prevalence of passive investment 
strategy could only exacerbate. 
Research about benchmarks or indexing 
has been active in the equity space, but relatively 
sparse for fixed-income. And this might be a result 
of data availability, which I will get to later. 
Generally speaking, it is widely 
documented that when stocks go in and out of an 
index, especially the widely followed index such as 
the S&P 500, there are large price movement and 
trading activities associated with such events. 

42 (Pages 162 to 165) 


Page 166 Page 168 
1 In the fixed-income space, one recent 1 U.S. bond index fund universe around 100 different 
2 paper by Dick Nelson and Rossy documented that when 2 indices are used, which sounds pretty competitive. 
3 bonds drop out of the Barclays corporate bond index 3 However, 98 percent of the assets under management 
4 due to either rating changes or nearing maturity, 4 of tech to indices under the Bloomberg markets 
5 there is elevated demand for immediate execution by 5 franchise. 
6 mutual funds which requires dealer to use their 6 In the bond ETF universe over 300 
7 balance sheet to accommodate. Increasing share of 7 different indices are being referenced, but the top 
8 bonds held by mutual funds and the post crisis 8 three franchise, the Bloomberg Barclays, ICE and 
9 regulatory regime may have contributed to an 9 J.P. Morgan indices account for more than 90 percent 
10 increase in the price of immediacy charged by 10 of the assets under management. 
11 dealers. Since mutual fund behavior may interact 11 In the non index bond mutual fund 
12 with the post crisis regulatory environment for 12 space over half of the assets under management has a 
13 banks understanding the impact of indexing is 13 reference benchmark listed. There are over 1,000 
14 particularly important now. 14 different indices or sub indices referenced with 
15 In reviewing the literature and 15 many of the indices referenced by only one fund. 
16 research on this topic, a few unique features caught 16 This echos the finding by Robertson at Toronto 
17 of the bond indices caught my attention. Relative 17 University who looks at the stock mutual fund 
18 to the most popular indices in the stock market, 18 universe and finds similar kind of diversity in the 
19 bond indices tend to have high turnover as the other 19 universe of indices. Some of these single user 
20 panelists have mentioned and as bond get issued and 20 indices seem to be created by the fund or the fund 
21 new bonds draw closer to maturity they will have to 21 family themselves. Such practice of 
22 turn over. Bond index constitution typically 22 self-referencing might be more cost effective for 
23 happens at month end. One potential problem with 23 investors, but could lead to comparability issues or 
24 this is that when month end coincides with quarter 24 governance issues. 
25 end or year end it is precisely when major dealers 25 I want to finish by making some 
Page 167 Page 169 
1 tend to view elevated balance sheet pressure and 1 policy recommendations to the FIMSAC committee, 
2 cost of trade could be higher. 2 again, from the point of view of a researcher. 
3 Unlike some indices bond indices have 3 Research on the topic of bond indices 
4 mutual rating created and maintained by dealer firms 4 are relative sparse likely due to the lack of the 
5 and the old ones are the Lehmans and Merrills and 5 quality of data available to researchers. In 
6 the JP Morgans were the big ones. This makes sense 6 particular, data on index constituent history -7 
as dealer have a competitive advantage in providing 7 constituent price history, these are hard for 
8 pricing for the index constituents particularly for 8 researchers to access. Understandably, index data 
9 OTC traded illiquid bonds. 9 is a large revenue source for the index creators and 
10 This trend is slowly changing though. 10 their continued incentive to provide such a valuable 
11 In the past decade we saw part of --we saw the 11 service should be considered. However, to the 
12 Lehan Barclay franchise being purchased by Bloomberg 12 extent that historical lacked data can be shared 
13 and the Merrill indices being purchased by ICE. As 13 with the research community, it will be a bit 
14 trade reporting become more ubiquitous we're seeing 14 impetus for innovation and intellectual development 
15 large data vendors or trading platform owners 15 in this area. 
16 becoming more competitive in this space. This could 16 Another related point is about 
17 be potentially beneficial as the index prices will 17 transparency for investor protection. It would be 
18 rely less on one dealer's quote. However, this 18 great to have a one-stop-shop where investors can 
19 could make index further away from regulators reach. 19 search for the index information for each fund and 
20 Uncompetitive sales practices of bumping could also 20 where they can easily find out for each index what's 
21 potentially make the cost of requiring information 21 the criteria for index inclusion and exclusion and 
22 on these indices prohibited. 22 how many other funds reference the same index. 
23 The bond index landscape is 23 Which firm maintains or owns the index that the fund 
24 surprisingly diverse, yet, highly concentrated. 24 or ETF is tracking? Any potential conflict of 
25 According to data pulled from Morningstar in the 25 interest? And, furthermore, what are the 

43 (Pages 166 to 169) 


1 
23 
4 
5 
6 
7 
8 
910 
11 
12131415 
16 
17 
18 
1920 
21 
22 
23 
24 
25 
Page 170 
constituents for these index? 
Why is this important? Because as I 
was doing the research I found out that, to my 
surprise, the Bloomberg Barclays green bond index 
constituents are issued primarily by banks. And 
this -- this is easy of you to see a list, but it 
might not be that obvious to the investor just 
looking at the name. 
So, that's just some of my 
recommendations. And I look forward to further 
discussions. 
MR. MADHAVEN: Thank you, Dan. 
Anish, over to you. 
MR. KARYAT: Hi. Anish Karyat from 
Citadel Securities. I joined the firm about a year 
ago. Prior to that I was at Jane Street. My role 
is to run our fixed-income ETF business. And the 
context here is along the lines of ETF market. 
As a brief overview, the ETFs are, 
basically, a basket or portfolio of bonds. They 
have features of both closed end funds and open end 
funds. Close end funds, they trade on exchange 
throughout the day, there's plenty of transparency 
and like close and open end funds, as investor 
demand for a given ETF changes over time, funds can 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 172 
price the ETF, the other big factor that we 
constantly calculate in terms of providing liquidity 
to is the actual creation of redemption mechanism. 
Given the new ETF rule most retails are now on a 
level playing field. So, being able to do custom 
creation redemptions is an extremely critical part 
of it. While an ETF might track an index of 
thousands of bonds, the actual redemption -creation 
redemption can be achieved using a much 
smaller subset of bonds which significantly reduces 
the cost of providing liquidity on ETFs. 
With regards to this panel and some 
of the concerns that are being addressed, while I 
think a clear understanding of the index methodology 
and constituents is important, as the ETF market may 
burst, being able to accurately calculate the ETF 
fund holdings on a daily basis is critical. And by 
in large, most ETF issuers do provide a sufficient 
details data to help us supply funds accurately, but 
there has been a lot of push from various issuers to 
give us a lot more clarity on the holdings to reduce 
the noise when it comes to striking it down which I 
think will help the market a lot. 
With regards to the index rebalancing 
discussion we're having, my priority will be that at 
Page 171 Page 173 
1 issue or redeem shares to meet those demands. 1 least given ETFs creation redemption mechanism, as 
2 Unlike, open end funds though, the 2 funds managers they're able to add the index into 
3 creation redemption process for almost all of the 3 ETF creations and at least in the redemptions and 
4 fixed-income ETFs there's an incline transfer of 4 using the primary market creation redemption 
5 bonds in exchange for ETF shares. We as market 5 activity get those transactions done. So, 
6 makers to make realtime markets, whether it's 6 therefore, minimizing the cost of --transaction 
7 systematically on exchange, or on our view 7 cost of the bond itself. So, I'll be curious to see 
8 platforms, we put a lot of effort into building 8 in these studies we do if there is a difference 
9 realtime pricing models for the underlying 9 between mutual funds and ETF as far as tracking goes 
10 securities. And to that extent we use various data 10 from rebalances. 
11 sources including broad-based market predictors like 11 And, finally, and as a last comment, 
12 futures, equities, EDX, bond quotes, but by far the 12 in terms of exposures to investors, some of the 
13 most valuable data source we've come across is 13 biggest ETFs that hold corporate bonds have a lot 
14 actual transaction details on --on the underlying 14 more liquidity than the online bonds would imply, 
15 bonds. 15 but I think investors should be made aware of the 
16 As you know, only a small percentage 16 fact that in fact they are trading --they are 
17 of the corporate bond universe trades on any given 17 investing in securities that do have low liquidity. 
18 day, but those retails are extremely important in 18 So, while during times of stress and consistent 
19 being able to calibrate the prices of ETFs. And 19 outflows in the market, the cost of getting out of 
20 particularly given that less than five percent of 20 those ETF --ETF positions in fact could approach 
21 the corporate bond market actually trades on any 21 the cost of trading the underlying passive ETFs. 
22 given day, being able to model various bonds 22 MR. MADHAVEN: Okay, terrific. 
23 extremely critical based on the trade details we get 23 So, I have a few questions for the 
24 on other bonds. 24 panel before we turn it over to the broader FIMASC. 
25 Besides being able to accurately 25 And my fellow subcommittee members, you know, feel 

44 (Pages 170 to 173) 


Page 174 Page 176 
1 free to join in. And it's Kumar and others. I know 1 what they pay attention to. And that will include 
2 you have some questions, so. 2 all the new issuance, bonds that are dropping out, 
3 But let me --let me just begin 3 et cetera. So, come the end of the month it becomes 
4 actually, you know, maybe with --going back to the 4 the established returns universe for the following 
5 start with --with Phil and Steve. You know, in Dan 5 year. 
6 Li's remarks she made some --several comments about 6 MR. MADHAVEN: Okay. 
7 herding, about the possibility that concentration 7 MR. BERKLEY: So, that's -- that's, 
8 among indices means that, you know, passive 8 basically, how it works. And -- and managers will 
9 investors or their managers are reacting in the same 9 use this projected universe. They'll look for 
10 way in the same bonds. I mean, I'd love your 10 characteristics in their portfolios that need to be 
11 comments on --on that line of, you know. 11 adjusted. As I mentioned earlier, the different 
12 MR. GALDI: To the degree that there 12 risk factors, do they need to extend duration, do 
13 is herding triggered by a rebalancing, bear in mind 13 they need to purchase more corporate-specific with 
14 that the people who are going to specifically trade 14 the space of corporate issuance, things like that. 
15 those securities on the date of the rebalancing in 15 MR. MADHAVEN: Great, thanks. Okay, 
16 the amount of the rebalancing are just the passive 16 Bill. 
17 investors. Active managers will trade around those 17 MR. AHMUTY: Sure just a comment on 
18 events. Some of them have already sold out those 18 this. 
19 bonds well in advance of --of the downgrade or the 19 Phil made a comment, with respect to 
20 upgrade that may have caused the move in or out of 20 rebalancing and the passive funds, as an ETF manager 
21 an index. 21 we -- we don't always -- we don't necessarily 
22 Considering that then take account of 22 rebalance our funds the way an equity fund would be 
23 the fact that the fixed-income markets passive 23 managed, meaning, at a certain date and time looking 
24 investing is a drop in the bucket. It's well under 24 to execute the buys and sells that we need to do. 
25 two percent of total assets. So, to the degree that 25 Again, we're looking to track the overall risk 
Page 175 Page 177 
1 the rebalancing is going to affect the overall 1 characteristics in the fund. So, we do look at the 
2 market pricing, that's not likely given that 2 overall equity of bonds that are entering or exiting 
3 context. 3 the index. And there may be times where based on 
4 MR. MADHAVEN: Great. Steve. 4 our sense of what may be moving certain --certain 
5 MR. BERKLEY: Yeah. I agree with -5 
securities, we may enter that rebalanced position a 
6 with Phil's comments. We don't have the situation 6 few days before the month end, right at month end or 
7 like we do in equity space. When folks do their 7 after month ends. So, we do maintain some 
8 re-balancings they typically have cash that's come 8 flexibility within the guidelines of managing the 
9 in from --from payments and principal payments. 9 ETF. 
10 They start formulating their rebalancing trades as 10 So, I just want to make that point 
11 the month goes on. When you look at the 11 clear that we are not just going in there at month 
12 construction of fixed-income indices there's really 12 end and looking to rebalance those at a specific 
13 two indices that people have to look at. There's 13 time. 
14 the returns universe, which is a static set of 14 MR. MADHAVEN: Great, thanks. 
15 securities that's established at the end of a month. 15 Important qualification. 
16 It's held constant throughout the month. And then 16 Anish, quick question for you about 
17 you have a new index from a returns perspective the 17 this rebalancing activity. From your perspective of 
18 following month, but you also have --investors are 18 market maker, do you see unusual patterns in 
19 also interested in what we call the projected 19 liquidity of pricing of constituent bonds that are 
20 universe, right, which is really deciding --so, for 20 being added to or deleted from indices. And how do 
21 example, we --we just published the end of October 21 you, as a market maker, position around those kinds 
22 index which will be used for November. 22 of events. 
23 Now that we're in November people are 23 MR. KARYAT: Sure. So --so, we're 
24 really starting to formulate their rebalancing for 24 focused obviously on the ETFs space and a lot of the 
25 December, right. And this projected universe is 25 times what we do see is some of the bonds that are 

45 (Pages 174 to 177) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 178 
potentially overweight are more concentrated on the 
redemption baskets and vice versa on the creation 
side. So, we're actively trading those bonds on the 
back of creation redemptions and helping translate 
that into the rebalancing of the system. 
So, you know, as market makers we're 
cognizant of potential rebalances. And to the 
extent the margin will have positions in bonds that 
could go into creations, we are optimistic about how 
we source those bonds and go into those details. 
MR. MADHAVEN: Okay, great. I don't 
know, Michael, if you want to allow time for robust 
discussions. So should we open it up to the broader 
FIMSAC. 
MR. HEANEY: Happy to do so. 
Kumar. 
MR. VENKATARAMAN: Thanks everyone 
for your very thoughtful comments. 
I had a question on the daily pricing 
of the bond indices. And in this context of --I 
want to just read a short summary from 2003 report 
by the International Organization of Securities 
Commission which listed up some challenges in 
pricing benchmarks in the financial markets. And 
these are relevant for the bond market indexes 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 180 
of the sources for all of the different indices that 
we produce. And they produce the majority of our 
fixed-income pricing. They in turn have published 
disclosures on their methodologies and procedures 
for coming up with their evaluating prices. 
They are receiving input information 
from many sources. Not just actual trades, but also 
dealer quotes and other information take accountable 
of that in developing the pricing for a security on 
a given date. Also, as a security that has not been 
trading recently does trade, that gives them 
feedback into how the model works so that they can 
continually fine tune and calibrate the models. 
MR. BERKLEY: What I also want to 
point out to folks is that it's hard to manipulate a 
bond index because of pricing. And the reason is 
that there's thousands and thousands of 
constituents, right. And so even if one bond was 
off, that bond only has a very small weight in terms 
of the overall index, right. I think that's an 
important point. 
The second thing I want to raise is 
that the prices that go into the indices are 
available to thousands of eyes, right. So --so, 
people can see, for example, in the Bloomberg 
Page 179 Page 181 
1 because many bonds, as has been pointed out, do not 1 terminal what the constituents to the indices are 
2 trade on daily basis. 2 and what the prices are that are being used, right. 
3 So, the data that can be used for 3 And so, if there are any anomalies it's open for 
4 price and index in the benchmark comes from many 4 everybody to --to see. 
5 sources. Observable transactions, as somebody just 5 And --and there are in bond space 
6 pointed out, only about five percent of the bonds 6 as --as our colleagues at ICE would know, there are 
7 trade. So, you don't observe these transactions for 7 challenges to prices that come across, you know, the 
8 a very high number of the bonds. Then you have 8 desks from time to time, right. And so, this is an 
9 executable bids and offers. And then you have some 9 open --you know, it's an open system where people 
10 sort of imputed price based on expert judgment of a 10 can actually see the valuations and question them, 
11 panel of participants or some, sort of, model based 11 right. And it's --it's not an exchange. As I 
12 price. 12 mentioned earlier, it's --you know, there's a 
13 So my question is, when --when it 13 little bit of calculations and perspective that go 
14 comes to the daily pricing of the bond index, is 14 into establishing a price for an individual 
15 that sufficient these disclosures that are being 15 security. 
16 provided on what these different sources are for the 16 MR. VENKATARAMAN: I have a follow-up 
17 different types of bond indexes that are being 17 question. 
18 offered to the market and how does this change over 18 So, on --on a daily basis in terms 
19 time for a bond index. So, for example, as you hit 19 of the disclosures, do you --do you provide 
20 periods of market stress, the quality of information 20 information on what the composition is of these 
21 may change quite a bit. So, could you --could you 21 sources. You say five percent of it came from 
22 comment on that. 22 transactions and, you know, the remaining 25 percent 
23 MR. GALDI: In our case as we said, 23 came from codes and remaining came from some sort of 
24 we obtain our pricing from our affiliate and ICE 24 model based pricing. Is there disclosure on that. 
25 Data Services and we make transparent what are all 25 MR. AHMUTY: Again, we're the index 

46 (Pages 178 to 181) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 182 
providers, right. I think the important thing here 
is that if there are questions around the specific 
techniques that the pricing businesses run, I think 
those questions should really be directed toward 
them. 
MR. GALDI: And I would add to that 
that's an important point to raise that there is a 
wall of separation between pricing services at our 
respective organizations and the index businesses. 
And indices are not the only users of these prices. 
When Steve referred to thousands of eyes, those 
aren't just the people who are looking at the 
indices, but all of the many users of pricing 
information, be it for portfolio valuation and 
accounting, risk management departments, trading 
desks who are putting this information into their 
models. So, all of those end users are leveraging 
the same price that, yes, also gets used by the 
index. And they are communicating back to the 
pricing sources when they see something that doesn't 
look right. 
MR. HEANEY: Rick. 
MR. McVEY: Thank you. This is 
probably for Anish and Bill primarily, but there 
have been some important changes in the index and 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 184 
in the expected creations or redemptions or the 
inventory of bonds that we carry in being proactive 
in responding to our queues that sort of match our 
positions. And so, we would love to see a lot more 
transparency and a lot more proliferation of wall to 
wall trading. 
On the data side, again, knowing 
about actual --not just trading volume, but also 
activity --potential activity in bonds is extremely 
useful in helping us calibrate overall liquidity 
metrics for trading bonds. 
MR. AHMUTY: And so, from the ETF 
issuer we certainly have seen broker-dealers using 
ETFs and ETF to help manager positions in their 
balance sheets. That is something that really is 
continuing to help. It does two things I think. 
One, it allows the dealers to make sure that the 
ETFs are staying at a tight spread to their NAV. 
So, we've seen that reduce over time. That's the 
price relative to the net asset value. 
In addition that, that process also 
allows these dealers to use a create redeem 
mechanism --to use a create redeem mechanism within 
the ETF to help provide liquidity in underlying 
bonds. So, we see this as something that does help 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 183 
ETF markets over the last several years. Major 
dealers are using ETF shares interchangeably with 
underlying bonds for risk transfer, but the market 
has embraced all-to-all trading which allow relative 
value our players to participate accurately 
throughout the day on both sides of the market. And 
both Bloomberg and ICE are --are now conducting 
analytical work to improve the create/redeem process 
in an attempt to standardize that. 
My question is, collectively do you 
think that's improved the risk profile of the ETF 
and index space in a way that will minimize basis 
risk and is there a better liquidity model now to 
avoid the herd mentality when funds are getting out 
of the same positions. 
MR. KARYAT: I can start. 
So, from an ETF market maker 
perspective having access to liquidity on bond 
trading is essential. And with the proliferation of 
all the electronic trading menus, we are able to 
source liquidity in much more effective ways than 
just bilaterally with dealers. Dealers are a very 
important source of liquidity for us. 
At the same time, we can be a lot 
more opportunistic about sourcing bonds whether it's 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 185 
the ETF market become more efficient. 
MR.KARYAT: And along those lines one 
more comment. 
I think the creation redemption 
mechanism being customized there's a lot of effort 
being put in, quote, by Bloomberg and ICE and they 
show this directly into making that work flow a lot 
more standardized which helps reduce some of the 
friction and sort of operational issues that could 
come about from the creation redemption process. I 
think that's a good step forward in terms of making 
the process more effective. 
MR. HEANEY: Amy. 
MS. McGARRITY: Thank you. 
This question is directed I think at 
Phil, but if anyone else has any thoughts on it, 
please feel free to chime in. 
So, we debated on whose subcommittee 
should host this panel. I chair the Credit Ratings 
Committee and so this question is more credit 
ratings oriented, but you indicated that you take an 
average credit rating to determine inclusion in your 
indexes. 
A lot of work has gone into post 
Dodd-Frank reform, removing references to credit 

47 (Pages 182 to 185) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 186 
ratings in federal regulation. And so, some of the 
perspective we've gotten from a lot of our calls as 
a subcommittee over the past months is that there 
remains references to credit ratings in indexes as 
well as investor guidelines. And to the extent that 
exists, I'm just kind of wondering what your 
thoughts are on new players as NRSROs or smaller 
NRSROs and their ability to be used --their ratings 
being able to be used in your indexes going forward 
or is there like a period of time in which they need 
to be covering a name, et cetera. Just curious as 
to how you determine that. 
And in addition, if you could talk 
about when an issuer is downgraded and potentially 
falls into the fallen angel indices how that works. 
MR. GALDI: Sure. And, first of all, 
there is a cause and effect relationship between the 
indices and the investment policy guidelines. It's 
the policy guidelines that typically drive the index 
rules. You know, I described at the beginning the 
fact that the very first indices were investment 
grade rated. Why? Because they were trying to 
support ERISA funds that typically had investment 
grade rating criteria for the portfolio managers. 
And you want the index and the portfolio manager 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 188 
through. 
So, the important thing is when a 
downgrade occurs that hits that projected universe 
that I talked about earlier right away. So, folks 
know that, okay, this is security is downgraded, no 
longer meets the index rules and will be dropping 
out of the index come the end of the month. So, 
people don't have to wait until the end of the month 
to get this information. They'll see it as soon 
as --as soon as it's being captured by --by the 
benchmarks. 
As far as the smaller rating agencies 
go, we have included, for example, in our Canadian 
index DBRS, right. That's an important provider up 
in --up in Canada, but I think what --what really 
is important is that the index shouldn't be driving 
the business for these smaller rating agencies. 
If --if they're doing their job and they have 
enough coverage, our investors will tell us, you 
know, please add them to --you know, to the 
calculus. That's how Fitch evolved to be included 
into the benchmarks, you know, a long time ago, 
right. 
So --so, the process is really open 
to feedback from --from investors. We take all of 
Page 187 Page 189 
1 guidelines to be similar so that they have a --a 1 that feedback when we change these index rules, as I 
2 fair benchmark. 2 mentioned earlier, and then we --you know, we make 
3 So, that is the reason why we have so 3 announcements and we make these announcements as to 
4 many different cuts. And it's not just investment 4 the index rules changes with an unscheduled press 
5 grade high yield. There --there is many different 5 release. So, that all market participants have 
6 slices of the --the index world that we create 6 access to the information at the same time. That's 
7 based off ratings. 7 typically how it works. 
8 As far as how the process works, we 8 MS. McGARRITY: Thank you. 
9 look at the average rating once a month and current 9 MR. GALDI: If I could just add to 
10 with the rebalancing cycle. We look at the average 10 Steve's point, that's an important part of this 
11 rating for the security and compare it to the rules 11 whole process is rule changes. There is a process 
12 for all of the potential indexes that it might be a 12 and a process that's spelled out in the IOSTA 
13 part of. If it was in an index, but no longer meets 13 principles for what one does before changing the 
14 the rating criteria for that index then it gets 14 rules. You go through a consultation process. 
15 removed at that next rebalancing. If it was not 15 So, we have annually a public survey 
16 previously in an index, for example, securities 16 where we spell out all the different potential rule 
17 rated below investment grade, but it gets upgraded 17 changes under consideration. It's an open survey 
18 during the middle of the month, at the end of the 18 for the entire public to respond to. We collect 
19 month if it now carries investment grade rating we 19 that feedback and we consider changes in index rules 
20 take it out of the high yield index and put it into 20 based on that feedback from the public. And then, 
21 the higher grade index. 21 having considered that we then make the 
22 Does that answer your question? 22 announcements as --as Steve said as to which 
23 MS. McGARRITY: Yes, thank you. 23 changes, if any, will get effected for a goodwill 
24 MR. BERKLEY: And what Phil describes 24 counsel. 
25 is --is the same process that --that we go 25 MS. McGARRITY: Thank you. 

48 (Pages 186 to 189) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 190 
MR. HEANEY: Other questions? Sorry. 
MR. VENKATARAMAN: This is a question 
for --for Bill. 
So, my understanding is that even if 
we take a bond index that is widely followed and you 
look at the range of strategies that different funds 
use, there's quite a bit of variation. You have 
some funds that hold maybe 20 percent of the bonds 
and largely use --have a heavy reliance on a model 
and a sampling strategy that is of other funds which 
have a very high number of bonds and which are 
primarily holding index. 
So, my question is, are the fund 
disclosures that we have today with respect to how 
some of these passive funds are tracking the index, 
is that sufficient? To the extent that there is 
quite a bit of variation in the reliance on a model, 
should there be more information that is provided? 
Are there some metrics that can be provided to bond 
investors so that they have a better idea of the 
extent of model risk that may be implicit in --for 
the particular mutual fund or the ETF? Thank you. 
MR. AHMUTY: Sure. Thank you. 
So, I'm going to address this 
question from the perspective of an ETF issuer. And 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 192 
again, there are some significant guidelines where 
we're not going to deviate from that, you know, 
too --too widely. And again, 20 percent that you 
stated, is not something that we'd ever do. 
I don't think many, and I don't want 
to speak for other people, but it seems like 
unlikely that other index managers would be 
representing index with 20 percent. 
MS. LI: I thought what Kumar said is 
that 20 percent of the constituents of the indexes 
in your portfolio. I guess you're referring to 
40 percent of your portfolio is in the index. Is 
there a difference between these two ratios. 
MR. AHMUTY: Well, let's take a step 
back here. 
So, if we have our fixed-income ETFs 
we're tracking a Bloomberg Barclays investment 
created index, right. At any one point in time 
we're required to have 80 percent of our fund 
constituents that are of the index. 
MS. LI: 80 percent? 
MR. AHMUTY: Yeah, but --but 
practically speaking, it's probably going to be 
closer to 90 to 95 percent of the constituents in 
our fund that are in the index. 
Page 191 Page 193 
1 these are '40 Act funds and the guidelines of these 1 MR. HEANEY: Any other questions? 
2 funds state that the index --the constituents of 2 I'm sorry. 
3 the fund --80 percent of the constituents need to 3 MS. LI: Can I ask a question to the 
4 represent that of the index. 4 panel? 
5 So, you made a comment early in your 5 So, for -- for a mutual fund or ETF 
6 statement here saying that the funds were holding 6 which is represented by Bill, how do you decide on 
7 maybe 20 percent of the bonds in the index. So, 7 which index you will be benchmarking against as you 
8 that's not --that would not qualify in our process, 8 start a new fund or ETF? Or -- and when do you 
9 right. So, we're going to look to hold --by the -9 
decide you want to switch or does it ever change. 
10 by the guidelines of the '40 Act a minimum of 80 10 MR. AHMUTY: Sure. 
11 percent. 11 Well, a lot of what we decide to do 
12 Now, philosophically at State Street, 12 on the ETF side is client driven. We also evaluate 
13 for example, if you look at our ETFs on our index 13 what's out there in the market in terms of the 
14 funds and what we're holding, at any one point in 14 products that everyone sees that are already out 
15 time we may have 90 to 95 percent of the index 15 there. Do we want to be the third to market or the 
16 constituents in our funds. So, while we don't look 16 fifth to market in the same ETF that's tracking a 
17 at it from a bottom up CUSIP by CUSIP level, we do 17 similar index. So, there's a lot of things that go 
18 look to represent the overall characteristics of the 18 into the consideration. 
19 fund. So, it's really a question of how we get to 19 Primarily we're talking to clients to 
20 that the 90 to 95 percent. 20 see what they need and what they want. We want to 
21 And so, the bond that's falling in or 21 make sure that when we do choose the overall 
22 coming out we may not trade it that day, but that is 22 strategy we're using indices that have guidelines 
23 ultimately we want to make sure we're delivering 23 that provide for liquidity. So, we're not going to 
24 those returns. So, every ETF issuer does have a 24 have an index where the -- the liquidity threshold 
25 slightly different process on how to do this, but 25 is maybe 50 -- you know, 50 million in par 

49 (Pages 190 to 193) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 194 
outstanding. We're going to look to set those 
levels at a --at a level where we think we can have 
an index that can be created and redeemed, provide 
enough liquidity. 
So, there are a number of 
considerations that go into. You know, it's changed 
quite a bit over the past couple of years. And if 
you look at the statistics from ETF issuers, there's 
probably a lot of less issuance in the fixed-income 
space than there has been. And a lot of it is, 
we're looking to see what our clients want. 
MS. LI: Does the cost of the -getting 
the index ever come into play in deciding 
what index to track? 
MR. AHMUTY: Does the cost of --of 
the index, like our licensing. 
MS. LI: Right, licensing revenue. 
MR. AHMUTY: That would be an input, 
but it's not going to be driver. 
MS. LI: Okay. 
MR. BERKLEY: I'm just going to pile 
on here and talk about customizations because even 
though there's --there is a number of indices that 
the world uses, we publish over 100,000 of them in 
the fixed-income space everyday, right. People -1 
2 
3 
4 
5 
6 
7 
8 
9 
1011 
12 
13 
14 
15 
16 
17 
18 
19 
2021 
222324 
25 
Page 196 
described, if you have a large issuer that's falling 
out of the investment grade index moving into the 
high yield index, that's like a bear showing up on 
your doorstep. And so, issuers --I'm sorry, 
investors may say, okay, I want to cap the weight of 
any particular issuer in my index to two percent or 
some number that -- that they think is appropriate 
so as not to create exceptional risk due to -- you 
know, due to large issuers being involved. 
So, that's -- that's an important 
point when -- when it comes to indexing. And I know 
my -- you know, our -- my colleague Phil puts 
together custom indices as well, right. This is -this 
is kind of how we work. And to Bill's point, 
we can -- we can take these custom indices and 
create products around them for -- for investors. 
Sometimes the custom indices are a type of strategy 
that -- that people believe will outperform the 
standard benchmarks. I'll leave it at that. 
MR. HEANEY: All right. There's a 
quick question then we're right on schedule. 
MR. HARRIS: It is. 
Bill, can you describe how the 
deposit redemption portfolios differ from the 
investment portfolio and whether you might expect 
Page 195 Page 197 
1 we're seeing a trend toward customized benchmarks 1 variation in the event of a substantial market 
2 where people create their own criteria. We then 2 crisis. 
3 take that criteria and put together a benchmark and 3 MR. AHMUTY: Sure. 
4 back test it for them and then publish it moving 4 So, I'll walk through the creation 
5 forward as if it was one of our standard indices. 5 redemption and how it can be impacted by substantial 
6 When you look at customized indices 6 volatility in the market. 
7 there's really three categories, if I --if I may, 7 MR. HARRIS: So, the creation 
8 that we kind of look at. One is a constraint index. 8 redemption portfolios that the authorized 
9 So, for example, let's say you like the U.S. Agg you 9 participants deliver or -- or receive different from 
10 don't really want to own triple B securities in your 10 the actual portfolio if only for the discreteness of 
11 index. We can constrain the index further and say, 11 bonds, all right. So, are you asking for the entire 
12 okay, U.S. Agg triple Bs, that's one example. 12 portfolio -- the entire list or is a fraction of the 
13 A second is --is what we call 13 list. If it's a fraction of the list, how might 
14 composites and this is where we allow investors to 14 that vary in volatile periods. 
15 establish a benchmark with the weights that they 15 MR. AHMUTY: Sure. So, a couple of 
16 determined for their needs, right. So, for example, 16 things here. And we -- we have a couple of options 
17 you may want to overweight the corporate sector in a 17 when participants are creating and redeeming and 
18 benchmark and underweight the others. So, you may 18 that has evolved over time as well. 
19 say, 50 percent corporate bonds, 50 percent U.S. Agg 19 I would say that, you know, if we 
20 X corporate bonds, right. So, you can --you can 20 think about a plain vanilla IQ fund or one of our 
21 play with the weights a little --excuse me, a 21 high yield funds, we will -- there are two real 
22 little bit. 22 methods; one is that we will submit a -- we'll 
23 And the third major type of custom 23 publish a list out to the street that will be a 
24 index is what we call capped indices, right. So in 24 subset of the overall portfolio. So, let's say that 
25 the case of fallen angles, which --which Phil 25 the portfolio has 900 CUSIPs in it, we may publish 

50 (Pages 194 to 197) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 198 
to the authorized participants a list of three to 
400, right, that represent the overall risk 
characteristics of the fund and ask the participants 
to deliver in either that fullest or a subset of 
that list. Again, everything that comes into our 
fund would represent the overall risk 
characteristics of the fund. And everything that 
goes out would be similar. 
That process has evolved and while we 
still do that, what we're seeing now more and more, 
are what I would call the bespoke created redeems 
where folks such as Anish and firms like that would 
come to us and say, here's what we can deliver into, 
here's what we don't. So, this puts the optionality 
I guess you would call it on the ETF issuer and say 
yes or no to certain individual securities. 
At the end of the day though we're 
going to make sure that whatever we take in from the 
street is going to represent that fund and if you -you 
know, if you're looking for a number of 
securities that we would accept, that's going to 
change. It's going to deviate depending on the size 
of the creation unit, the volatility in the 
marketplace. There are a number of factors that we 
consider when thinking about the number of actual 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 200 
number of units. Passed that they are going to 
reissue new baskets that other markets can use. 
MR. HEANEY: Great, thank you. Thank 
you for moderating this thoughtful panel. And to 
the panelists, thank you. Incredibly informative 
and we appreciate you sharing your perspectives. 
Thank you very much. 
(Applause.) 
MR. HEANEY: Okay. We'll take a 
short, five-minute break and start again at 1:45. 
(Brief recess taken.) 
MR. HEANEY: Okay. Good afternoon, 
everybody. And thank you --thank you for your 
continued participation and engagement. We have 
another panel getting ready to start on the 
Interest-dealer Government Securities Trading 
Platforms panel. 
MR. REDFEARN: Okay. Good afternoon, 
everybody. And thank you --thank you for your 
continued participation and engagement. We have 
another panel getting ready to start on the 
Interest-dealer Government Securities Trading 
Platforms panel. 
So, this committee's focus today has 
been largely on corporate bond and municipal 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 199 
securities that we take in. 
When it comes to, you know, volatile 
markets we also want to consider the overall flow 
directionally, what we're seeing in the markets, 
that will also be an input. So, if one market maker 
is coming in and saying we will --can deliver you 
these bonds for creation, we may take that in part 
and parcel with what other folks are looking to 
deliver to us, but overall each person needs to 
represent the risk characteristics to the fund. So 
we wouldn't say, okay, from this bank we'll take 
energy bonds. From this bank we'll take financials. 
So, every bank or dealer is going to have to give us 
that representation. 
MR. KARYAT: Only one other comment I 
would make is all the different issuers have 
slightly different variations along the same lines, 
but to the extent it is one of those days when there 
is excessive redemptions, the issuers are also 
managing to the fact as one participant remains, to 
the extent it makes some issuers --some issues 
underweight, those bonds are no longer available for 
further redemptions from other market participants. 
So, generally speaking, most of these 
baskets or custom baskets are good to a certain 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 201 
securities market, but this is also a very good 
forum for us to have a public discussion of 
government securities markets issues including the 
U.S. treasury market structure. 
So, since October 15th, 2014 flash 
rally in the treasury market, the SEC, along with 
the U.S. Treasury Department, Board of Governors, 
the Federal Reserve Bank of New York and the CFTC 
have participated in a collaborative process, 
including the official and private sectors, to 
assess and monitor the current U.S. treasury market 
structure. 
The most recent event in that 
collaborative process was the fifth annual U.S. 
Treasury market conference which was held on 
September 23rd at the Federal Reserve Bank of New 
York. At that conference, Commissioner Elad Roisman 
stressed the need to keep pace with changes in this 
market. And in so doing, he recognized some of the 
parallels between segments of the treasury market 
and our equities markets and emphasized the 
importance of investor protection and market 
stability. 
Today's panel will have an 
opportunity to focus on the inter-dealer market for 

51 (Pages 198 to 201) 


Page 202 Page 204 
1 treasury securities. The market is comprised of 1 So, among our panelists we have Ted 
2 institutional market participants trading through 2 Bragg, head of U.S. fixed-income at Nasdaq Fixed 
3 electronic inter-dealer platforms and voice or 3 Income, Dan Cleaves, CEO of BrokerTec North America, 
4 manual inter-dealer brokers with the electronic 4 Shawn Bernardo, senior managing director for rates 
5 platform participating primarily trading on-the-run 5 at TP ICAP Americas, Deirdre Dunn, co-head of global 
6 treasury securities and manual or voice IBDs, more 6 rates at Citigroup, Nate Kalich at Ronin Capital and 
7 often, specializing in off-the-run treasury 7 Ryan Sheftel head of FIC at GTS. 
8 securities. As was recognized in the joint staff 8 So, to begin I'd like to turn it over 
9 report on the events of October 15th, 2014, the 9 to our panelists for a brief introduction and ask 
10 electronic inter-dealer platform segment of this 10 each of them to quickly describe or briefly describe 
11 market has become highly automated. 11 your firms, your roles at your firms and how your 
12 Based on data released by the 12 firms engage in the inter-dealer market for treasury 
13 Treasury Department and discussed by Deputy Treasury 13 securities. 
14 Secretary Justin Muzinich at the recent conference, 14 So, Ted, why don't we starred with 
15 between September 2018 and August 2019 volume in 15 you. 
16 nominal coupons in inter-dealer venues, which 16 MR. BRAGG: Hi. Thanks, Brett. And 
17 includes both electronic platform and manual or vice 17 good afternoon, everybody. 
18 IBDs, represent between 55 to 60 percent of the 18 My name is Ted Bragg and as 
19 total volume on average of nominal treasury coupons. 19 mentioned, I run fixed-income for Nasdaq. Nasdaq as 
20 And according to data published by the staff of the 20 you know is known for our exchanges and clearing 
21 Federal Reserve Bank of New York late last year 21 houses. Twenty that Nasdaq run and --and one of 
22 electronic inter-dealer platforms represented 22 those, Nasdaq Fixed Income, is part of that. 
23 approximately 40 percent of the total volume of 23 Another 70 that we run for other brokers, markets 
24 treasury coupons traded between August 1st, 2017 and 24 and regulators around the global. 
25 July 2018. 25 I'm also the CEO of Execution Access 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 203 
In addition, principal trading firms, 
or PTFs, represent a significant portion of the 
electronic platform volume. The same treasury 
department analysis showed that between April and 
August 2019 PTF share hoovered around 60 percent of 
the volume on electronic inter-dealer broker 
platforms. Note to this day that it is limited to 
electronic platforms where the majority of principal 
trading firms trading occurs. 
So, there's a few key areas that will 
be in focus for our conversation today. We're going 
to look generally at the U.S. treasury market 
structure. We're going to be looking at operational 
transparency. And also, risk controls and system 
integrity as an issue. We've assembled a panel of 
market participants ranging with a range of 
perspectives on the inter-dealer market for 
governments securities. And we'll have an 
opportunity to hear from both inter-dealer trading 
platforms and users of those platforms. 
So, before I introduce the panel -the 
panelists, I also just wanted to welcome 
Commissioner Pierce who has joined us for this 
afternoon's conversation as well. So, welcome 
Commissioner. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 205 
which is a FINRA registered broker-dealer and 
affiliate of Nasdaq which runs fixed-income ATS, 
Nasdaq Fixed Income. 
The platform itself offers electronic 
trading for U.S. treasuries, predominantly 
on-the-runs, both the capabilities to trade 
off-the-runs and U.S. T-bills. The protocol 
represents essential in an order book, matching 
orders on a price time priority basis 23 hours a 
day, five days a week. 
As a representative of this panel and 
Nasdaq I look forward to discussing the benefits of 
possible modernization, greater transparency and 
efficiencies which could be accomplished in the 
treasury market provided it's done on a level 
playing field and doesn't overlay rules or 
procedures that might apply to U.S. treasuries. 
MR. REDFEARN: Dan. 
MR. CLEAVES: Thank you. 
Dan Cleaves, CEO of BrokerTec. I've 
been with the firm since 2001. One of the original 
hires there and did a lot of work getting the 
platform up and going, taking 13 years of voice 
brokering experience and applying that to building 
out electronic trading platform. So, I did a lot of 

52 (Pages 202 to 205) 


Page 206 Page 208 
1 design there. 1 a treasury department --department which is more of 
2 BrokerTec was acquire by ICAP back in 2 an investor in treasuries as well. 
3 2003 where we were brokering at the time the entire 3 MR. REDFEARN: Thank you. 
4 career of benchmarks and all the off-the-runs. Then 4 Nate. 
5 we joined forces with Shawn's company at ICAP and 5 MR. KALICH: Nate Kalich, Ronin 
6 created a hybrid model where a lot of the active 6 Capital. I appreciate the opportunity to speak here 
7 trading benchmarks traded on --on the BrokerTec 7 before the Commission and --and obviously FIMSAC. 
8 system and what was the OM and then evolved into the 8 I have been at Ronin Capital for 20 
9 Nasdaq trading system which is a very robust 9 years and been electronically trading for 20 years. 
10 platform. 10 The E-Speed BrokerTec and obviously at various other 
11 And then we had a --a hybrid 11 inter-dealer brokers over that period of time. 
12 offering where you can pick up the phone and talk to 12 Ronin's a proprietary trading firm 
13 the broker for trading off-the-runs or you could 13 and we're based in Chicago. And what that means is 
14 trade them on a point-and-click via a screen into 14 that we don't have any customers. So, we --we 
15 the same pool of liquidity. Pretty powerful 15 would be classified as being, you know, a PTF, 
16 combination there. 16 principal trading firm, but I'd say that we're, you 
17 Recently in --things have gone on we 17 know, different than a lot of principal trading 
18 have split from the voice broken group and created a 18 firms. And there's a lot of variety and broadness 
19 company called NEX. NEX operated autonomously for a 19 with principal trading firms. 
20 year-and-a-half and then subsequently was purchased 20 We're a registered broker-dealer. 
21 and acquired by the CME. So, as of November 2018 we 21 We've been members of FICC since 2005 and we engage 
22 are now part of the CME. And we have plans to move 22 in --in relative value and basis trading and also 
23 our technology off of the Nasdaq, which is the new 23 participate in treasury options. And we're here, of 
24 system right now the Nasdaq GENI-Ant system which 24 course, because we actively participate in the 
25 has been excellent for us, and we're moving on to 25 inter-dealer market. 
1 
2 
3 
4 
5 
678 
9 
10 
11 
12 
13 
14 
15 
16171819 
2021 
22 
23 
24 
25 
Page 207 
the CME Globex platform which is the same platform 
for the futures and options. A separate instance, 
but that will be taking place in a large migration 
process project that will culminate with the 
movement over on Q4 of 2020. 
MR. REDFEARN: Shawn. 
MR. BERNARDO: Shawn Bernardo from TP 
ICAP. TP ICAP is one of the world's leading IBDs. 
We broker products across the entire globe in rates, 
FX, equities, commodities and credit. We also have 
an industry-leading data and analytics group. We 
currently operate two sets globally for all of the 
regulated products of the company. And the reason 
I'm here today is we run one of the largest 
liquidity pools for off-the-run traders. 
MR. REDFEARN: Thank you. 
Deirdre. 
MS. DUNN: Yes. So, Deirdre Dunn 
with Citi. Thank you so much for having us. 
So, Citi is, as probably all of you 
are familiar, is your traditional broker-dealer 
activities. It was my background and what I am 
responsible for. And so, we engage in market making 
for clients globally across the board in treasuries 
and many other products. Citi of course as well has 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 209 
MR. REDFEARN: Thank you. 
Ryan. 
MR. SHEFTEL: Hello. My name is Ryan 
Sheftel. I'm with GTS. I'm a partner and head of 
our fixed-income business there. First, thank you 
very much for the opportunity to speak today. 
A little bit about GTS. So, we are a 
quantitative and technology driven capital markets 
trading firm. We would also qualify as a PTF based 
on the description of that. And, you know, our 
business is, you know, a variety of different 
business lines, but one of which in our core is 
we're a committed liquidity providing market maker 
across a variety of different products and 
geographies. And that gets us engaged in a large 
number of different venues, platforms, ATSs, et 
cetera, across the broad spectrum of security types. 
Obviously, including U.S. treasury markets, which is 
why we're here today. 
And, lastly, I think one of the 
unique things about us as well is we are one of the 
largest DMN, formerly specialists firms down on the 
New York Stock Exchange. So, we have a lot of 
experience with operating in different trading 
platforms that have a variety of different 

53 (Pages 206 to 209) 


Page 210 Page 212 
1 regulatory constructs. 1 wasn't the fastest system out there. So, we 
2 MR. REDFEARN: Great, thank you very 2 evolved. In 2007, 2008 we introduced a colo 
3 much. 3 facility so that people could house their strategies 
4 So, I think we're going to start with 4 closer to our matching edge and --and provided 
5 Dan and Ted talking just a little bit about the 5 cross connects into that. 
6 evolution of sort of the electronic platforms and 6 In 2012, we --we moved to what was 
7 how they've evolved. And, you know, so just some 7 called the GENI-Ant system where we took round trips 
8 thoughts in terms of the technology and, you know, 8 times that were at about three milliseconds down to 
9 how you look at the interaction of those -- those 9 close to 100 microseconds. Followed that up with 
10 markets and market participants these days. 10 market data distribution using --using the ITCH 
11 So, Dan, do you want to start. 11 protocol which was mark standard, but very quick. 
12 MR. CLEAVES: Sure. Thank you. 12 And then followed that up with OUCH which was the 
13 So, as I referred to earlier, you 13 order entry protocol which gave round trip times 
14 know, a number of years on the -- on the broker 14 significant lowering of latency. 
15 floor, right, screaming and yelling making sure that 15 Bottom line is --is what you have 
16 everything matched up at the end of the day. I 16 to --what you have to think about is, how do you 
17 started in 1987, had the privilege of writing hand 17 provide, whether it's a manual trader who's still 
18 tickets in October of '87, for those of you that 18 pointing and clicking who's providing a very 
19 remember, it was a pretty typical day. Things are a 19 valuable natural interest in the marketplace, or 
20 lot more efficient now. What we did is we automated 20 your most sophisticated low latency trader to 
21 a lot of the matching that was done by the voice 21 deterministic trading environment and a level 
22 brokers at the shop that I was a broker at. And, 22 playing field. 
23 essentially what evolved is they moved that 23 The manual trader who's pointing and 
24 technology out from the brokers and put it on the -24 
clicking trying to buy ten-year notes isn't 
25 on the desks of the -- of the traders. 25 necessarily going to compete against somebody 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 211 
The --the benchmarks were actually 
transacted in the open outcry as well. Now, if you 
tried to --to broker benchmarks it would be very 
difficult, right. There's active fast moving. So, 
that technology moved out to the traders. 
Eventually it evolved and I think a lot of it had to 
do with the migration out of the pits in Chicago 
where people traded more automated fashion both cash 
and futures. It created more automated trading. 
When BrokerTec was founded a lot of 
it was around the API and trading on the API as 
opposed to just a point and click. And about 2004, 
2005 I'd say the percentage of --and really saw 
climate change in the percentage of activity that 
was done, not only in market making, but also in the 
volume grew substantially as you had a lot of the 
sophisticated participants join the market, not with 
customers, but coming in making markets or running 
strategies against other asset classes or other 
products and trading aggressively or as in the 
market making capacity. It really changed the 
complex. 
It made us think about the systems 
that we had. As I said, we started off with the -the 
Nasdaq click system, which is very robust, but 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 213 
that's --that's co-located and --and running round 
tripe times in the micro seconds, but they're 
trading against those market makers who are 
providing very important liquidity who, when the 
market's busy, and we saw this on October 15th, 
2014, I know it's been a big discussion for many 
years now, when we traded all the price points up 
and we traded all the price points down in that 
point-and-a-half move we never gapped. I've got the 
data to show. I've met with treasury and --and fed 
and CFTC and even the Bank of England wanted to know 
about what happened on that day and we went through. 
The system performed as expected, but 
bottom line is bringing everybody together and 
providing deterministic trading for everybody is 
important. 
Getting back to benchmark treasuries. 
They have evolved and the way they trade now is very 
different. Shawn will talk about off-the-runs. As 
I said, we assist in their market in terms of 
providing distribution and an electronic means for 
trading it, but it is a very different market. 
Knowing --you know, being on the phones it's -it's 
institutional driven and it's --it's supply 
and demand. It's not liquid and --at times it's 

54 (Pages 210 to 213) 


1 
2 
34 
5 
6 
78 
9 
10 
11 
12 
13 
14 
15 
16 
17 
1819 
20 
21 
22 
23 
24 
25 
Page 214 
not liquid and it doesn't trade by the second or the 
microsecond. 
MR. REDFEARN: So, Ted, could you 
pick up on that. And, in particular, this sort of 
changing nature of the market participant that you 
see in the marketplace as well. 
MR. BRAGG: Sure. Absolutely. 
And -- and to compliment what Dan said, you know, it 
was probably 20 years ago that eSpeed, so that's 
what Nasdaq called it, Nasdaq Fixed Income six years 
ago and -- and BrokerTec were created. And it was 
really to move the market from a tranche model, a 
hybrid model, the voice into electronic. And now 
where we've gotten to is probably 95 percent of our 
clients, as Dan mentioned, are connected via colo or 
via an internet service provider. So, an ISP. Only 
five percent of our clients are actually using GUI. 
Then there's the -- a clear 
demarcation line between where the liquidity 
providers to end clients, the D to C, the dealer to 
client model, internalizes their order flow, trades 
with their client. Then, possibly now trades with 
single streams and bilateral streams. And once they 
can't accomplish all of liquidity limit, the reason 
they convince our limit order books is purely that, 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 216 
trade on a daily basis. 
So, given that it doesn't necessarily 
lend itself to that API driven high frequency type 
trading that happens in the on-the-run market. So, 
it is a slightly different market. 
I would say, currently 65 percent of 
the business that we do is still voice. Our clients 
do have the ability to trade in a hybrid fashion. 
They have on the front end in front of them. And 
that we provide to them to disseminate prices and it 
allows them to execute if they so choose or they 
could come in via a broker and we can execute the 
orders on their behalf, but we have --we do provide 
all of the execution protocols that you would 
expect, meaning, auction platforms and executing in 
the --in the swap box. 
MR. REDFEARN: Okay. Thanks. 
So, Deirdre, as a primary dealer and 
a significant participant in --in these markets, 
can you just describe how Citigroup trades in the 
inter-dealer market, including the types of trading 
venues or streaming or other technologies that 
you're engaging with. 
MS. DUNN: Sure. 
So, as mentioned, Citi is a primary 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 215 
for liquidity. And that's really a dealer to dealer 
model. It's the banks and now it's the PTFs. So, 
it's who's the market maker. 
PTF has become in some ways a large 
market maker for the bank who's getting the 
additional liquidity that they couldn't get from an 
internalization or direct connects. And all of that 
is done through, in our sense, again, it's Nasdaq 
market data coming out, which is called ITCH, order 
entry OUCH, then FICC. So, it's very much a 
technological solution to --to service the whole 
rates and fixed-income complex. 
MR. REDFEARN: So, Shawn, could you 
just provide a little contrast with how this works 
in the off-the-run market and some of the 
distinctions you would see there. 
MR. BERNARDO: Sure. In on-the-runs 
the liquidity's consolidated to six or seven issues. 
And in off-the-runs it trades slightly differently. 
They're not trading --one, they're not trading 
outright, they're trading in a swap box. So, each 
position is hedged when it trades, but the liquidity 
itself, as Dan said, is episodic. It's not constant 
and you're not talking about six instruments. You 
could be talking about hundreds of instruments that 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 217 
dealer and primarily is an intermediary to clients. 
And so, we tend to participate in the inter-dealer 
market as a hedge to our principal risk position 
which tends to be driven by either one of those 
activities. 
I would say that our participation in 
the inter-dealer market can be driven both by 
products --product risk specific to treasuries, but 
could be other products, could be on-the-run, could 
be similar things. We will always look to use the 
various product suite that is available in order to 
minimize footprint because we have found that 
enhanced technology that we're all here today 
discussing that has enabled everyone to transmit and 
process data faster has lead to much broader 
information dissemination and tends to impact -trade 
impact tends to be felt across a broader suite 
of products and it's spread to others much faster 
than it used to. 
In terms of how we engage, I would 
say that can be through both passive and aggressive 
orders through auctions and through sweeps. It can 
be in any of the platforms that we've talked about 
so far today. Also, we have --we trade with 
clients via voice, via single dealer platforms which 

55 (Pages 214 to 217) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 218 
at Citi is called Velocity which is pretty standard 
across the street. And they're, you know, the 
multi-dealer platforms. So, I would say we 
participate in both from a streaming and an RFU 
capacity responding to clients. 
MR. REDFEARN: Thank you. 
So, Ryan, I'm going to turn to you 
for a second. Can you just describe, do you see 
yourselves primarily as a liquidity provider or, you 
know, a liquidity taker in the marketplace? And can 
you tell us just a little bit about your trading 
activities and the sorts of platforms that you 
engaged with when you're trading. Is it --is it 
mostly these platforms or how do you see the --sort 
of the spectrum there. 
MR. SHEFTEL: Sure. Thanks. 
So, to answer your first question, 
you know, our activity in the treasury market is 
primarily as a liquidity providing market maker. In 
terms of the platforms that we transact on and the 
nature of our --our trading obviously in the 
treasury market activity is 100 percent automated 
and technology driven. And for that we participate 
on a --on a wide variety of platforms, both ones 
that are represented here, as well as some of the 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 220 
electronic platforms. And we participate in 
treasury auctions. So, we essentially do everything 
that you could possibly do, whether it's with a 
person directly or electronically. 
I think I'd like to make a point that 
the inter-dealer platforms in general are great for 
price discovery, as well as, providing liquidity. 
And that's an important concept. And I'd say in 
particular, the central limit order book model is 
great because it does that and it allows you to 
access anonymously. And it provides fair access 
because it is anonymous. So, you know, I think that 
in general that's --that's a great model. But, you 
know, the electronic trading in general has brought, 
you know, tremendous sufficiencies to the trading 
market. 
In terms of risks I think that, you 
know, one of the risks is fragmentation. So, in 
other words, price discovery is more complete 
clearly when you can go to one venue. When you have 
to go to multiple venues it's more complicated. If 
you have to go to many, many different venues and, 
in fact, access bilateral skew streams or a few 
models, etc., you can see how it gets a little bit 
more complicated. And, you know, in the treasury 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 219 
newer ones that aren't here today. Probably the one 
style of platform that we are not actively 
participating are, you know, the traditionally 
colloquially termed RFQ platforms. Outside of that, 
generally the platforms that allowed for fully 
automated pricing transacting between parties like 
ourselves, dealers, banks other PTFs, those are the 
platforms that we're actively involved in. 
MR. REDFEARN: And, Nate, same 
question. Just, you know, if there's any way you 
would differentiate sort of what you're doing. 
MR. KALICH: Yeah. I'd say that 
we --we participate on all the platform in addition 
to non platforms. So, electronic ranging from 
BrokerTec, you know, Nasdaq, to, you know, Shawn, 
ICAP. 
And in addition, we participate in 
our queue markets where I'd say we'd be 
characterized more as a liquidity taker because 
we're asking for quotes and --and quotes come back, 
but we --we provide liquidity obviously on --on a 
lot of the electronic platforms where you're able to 
do both, provide liquidity and take liquidity. We 
trade over the phone. We --we trade with ICAP as 
an example over the phone, as well as, on their 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 221 
market I would say it is --it is important to 
determine what the price of the ten-year is. 
I would say in general though that 
it's probably not important to determine the price 
of the tenure to the nanosecond. So, I think 
that's --that's kind of where we've gotten a lot of 
efficiencies out of electronic trading over the 
years, but I think now it's gotten to the point 
where it's somewhat marginal. So, anyway, I think 
at the end of day what we're --what we're most 
focused on is liquidity, price discovery and fair 
access. 
MR. REDFEARN: And so, Deirdre, just 
a quick follow-up, I want to follow-up with you and 
both Ryan. Simply on the issue of the amount of 
electronic trading that is now occurring in the 
treasury marketplace and the on-the-runs, are there 
any other risks that you see that have emerged 
that --that come with this evolution. 
MS. DUNN: Yes. That's a very big 
question. I would say there --yeah. I think it 
has introduced a lot more technological dependence. 
I think concentration and venue risk I think can be 
something that the market has been concerned about. 
I think transparency --some of this we may get to 

56 (Pages 218 to 221) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 222 
later, but I think transparency around how orders 
are routed, how matching logic works, how, right, 
there isn't necessarily transparency across all of 
those different concepts, or at least not to the 
same level to every market participant. 
There are times where we have only 
found out --out about a change to matching logic 
because it changed the way that we need to, like, 
physically connect or send the messages, right. In 
other words, we would have no idea. So, I think 
there's some concerns around that. 
Also, you know, as part of the 
treasury market practices group we spend a lot of 
time on clearing and settlement over the past number 
of years. And Ryan was involved in this as well, 
published a 60-odd page white paper on it for anyone 
interested in some light reading on the way home. 
But I would say there we've identified as well a 
number of different models whereby market 
participants are connecting without as robust an 
understanding of what happens after the fact to get 
from trade time to settlement time which introduced 
a whole host of other risks for the market. 
MR. SHEFTEL: Yeah. I think what it 
comes to risk --so, I've been involved in the 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 224 
You know, it's very different when 
you can, you know, manage risk generally on the end 
of day cycle or you can sort of trust that a human 
is part of every trade, you know. The kind that you 
would want to have involved in that trade is very 
different than if you go more automated and now 
trades are occurring without human intervention at 
every level. It sort of shifts the nature of the -you 
know, what capabilities you have to have to 
understand that risk. 
So, to sum it up, I think that for 
any market going through the transition it's a 
question of, you know, what are the capabilities of 
the firms that are involved and what are the 
capabilities of the individuals in that line of 
command to truly understand the risks and manage 
them properly. 
MR. REDFEARN: So, I want to switch 
for a second to the topic of just systems operation 
and operational transparency and go back to Dan and 
Ted for a second and just talk about --so, in terms 
of how your systems operate, in terms of, you know, 
the information that's made available. So, what is 
provided to customers? Are they operations manual? 
Is it on your website? Is it public? Like, how 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 223 
fixed-income markets for almost 25 years now. I 
started in '95. So, I've had the luxury of seeing 
whether it was futures or FX or equities, you know, 
a variety of different products and markets go 
through this transition. 
I think the one common theme I kind 
of think of through all of those is that really 
the --the sort of nature and capabilities of both 
firms and individuals that it takes to properly risk 
manage a voice business and an electronic business 
are very, very difference. Rather they're very 
different and often they're not embodied in the same 
people or the same firms. And so, I think that 
often times the risks are --during these 
transitional periods are, do the firms respect that, 
do they understand that and are the right people in 
charge of the right points of the process, you know. 
What I mean by that is, you know, 
it's very easy --you know, a lot of vendors create 
a lot of tools to make electronic trading capable, 
which is great, but the question is is, you know, 
whose hand do they land in. Do the people utilizing 
these tools truly have the understanding, the 
internal risk management processes to understand 
what that's all about. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 225 
would you describe the way that the --the way the 
systems operates is sort of made clear to your 
customers and then more broadly. 
MR. CLEAVES: So, from a --from a 
BrokerTec perspective, any new trader that comes on, 
whether it's a manual trader or an API trader, it 
goes through a rigorous on-boarding process where we 
review the worthiness of them to be on the platform 
and the --and the credit worthiness and any number 
of other factors. We find out how they want to 
integrate with us, whether they want to be a manual 
trader, whether they want to trade via an ISD, 
whether they want to trade in --in the colo, etc. 
Any new manual trader will go through 
rigorous training with our salespeople who will be 
on the floor who work with them in the front end 
configuring all the --the checks and balances and 
the limits on the system. They'll find out what 
type of trader they are, what their risk tolerance 
is, et cetera. And they'll set up their screen, 
show them the pages that they want to see, the 
instruments that they want to see and make sure that 
they are ready to be able to transact in the live 
marketplace. 
Similarly, the same thing goes for 

57 (Pages 222 to 225) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 226 
any new type of API trader. They will get 
on-boarding docks. They'll get sample code. They 
will have to pass a conformance test to make sure 
that their algorithm or strategy works against our 
marketplace in a test environment as they would 
expect it and, quite frankly, as we would expect it. 
You want to make sure that there's no disruption to 
the marketplace. 
So, we're pretty clear in terms of 
how we operate the market. We're pretty 
straightforward. It's FIFO. We do still have the 
workup on the benchmark treasuries which we'll be 
retiring when we move to Globex, but that's really 
from a visual. So, really from the most part, it's 
simple logic. The order types are limited. It's 
pretty straightforward. So, there's not --from a 
benchmark treasury perspective there's not a lot of 
sophisticated order routing or order matching. 
It's --it's as simple and straight forward as we 
would like it to be. 
MR. REDFEARN: Ted. 
MR. BRAGG: So, not dissimilar from 
what Dan said. It's --it's a simple professional 
market. To your question about how we provide that 
to our client, it's behind the log in, right. So, 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 228 
with our clients. The testing will be done months 
in advance. As --as Dan said, it's all done in a 
testing environment after hours. It's proven and 
then rolled out to the customer base, but they're 
well aware of what the enhancements are, how it's 
going to operate and it is fair and open access. 
It's the same --same exact protocols for everybody. 
MR. REDFEARN: So, we'll get to this 
in a minute. And we talked before about how, you 
know, in the U.S. equity markets we've gone through 
a lot of different issues in terms of transparency. 
So, for example, we have our Reg ATS, our Reg ATSN 
now for --for alternative trading systems where 
there are significant disclosure requirements that 
are put in there. 
And so, there are these questions as 
out markets become --as markets evolve and they 
become more electronic, just in terms of the level 
of detail that is provided out to market 
participants, it is a question. 
So, Deirdre, you mentioned a minute 
ago that there are some things that might be 
meaningful to you. Do you have any additional 
thoughts as to, you know, the sorts of things that 
you think might be very useful to have, you know, 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 227 
there's say --there's a secure portal for our 
customer base, but then behind that log in whether 
they get access to --they get an operation or --or 
subscriber manual. It includes our own NSJ policy. 
It includes access rules. It includes our risk 
controls, order types, displays, trading suspensions 
if we were to have them. 
There's an electronic trading 
agreement with our client and that's just basic 
on-boarding of broker-dealer KYC. And again, an 
explanation of those rules and then walk through the 
connectivity. And the rules I suggestion are --are 
fairly simple because the product is fairly simple. 
There's not 50 different order types. There's not a 
whole bunch of different order routing requirements. 
Nor is there multiple APIs. It's --we kind of lay 
that out in a straightforward question. 
MR. REDFEARN: Shawn, same question. 
MR. BERNARDO: It's --it's 
essentially the same for us as well. There are 
protocols for trading offering treasuries. They are 
not publicly available, but they are available to 
all of our clients. Protocols are straightforward. 
From a tech perspective any new release or 
enhancements that do go out we work hand in hand 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 229 
in --either in a public document or some other 
format that makes very clear the nature of how 
orders interact or how disputes are resolved or, you 
know, how fees are working or anything like that in 
terms of just another level of transparency. 
MS. DUNN: So, I think, you know, the 
treasury market is obviously quite unique in what 
it's servicing and what the purpose is. And so, I 
think it's important when considering any additional 
regulation to make sure that it's not burdensome 
without some counter balancing material benefit. 
What I would say in reference to my 
comments earlier around the information disclosure 
that as a user of various platforms, no one 
specifically, I think one of the things that we're 
always trying to evaluate is, what is going to be 
the footprint or the impact to the market of action. 
And so, understanding at times, you know, 
methodologies around ties, you know, whether there 
is any benefit for speed to the microsecond as we're 
talking about or not and different order types and 
how those do get routed in the matching engine and 
when there are changes to that. All of that 
information helps to evaluate that and I think also 
as the end user how or where you would want to 

58 (Pages 226 to 229) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 230 
engage in order to minimize your footprint on the 
market, you know, for the benefit of your client. 
So, I think that's sort of the 
primary focus I would say from my end. And I think 
that it --the state that we're in right now is 
satisfactory, but sometimes you have to do a little 
bit of extra homework to get the answers to all of 
those questions. 
MR. REDFEARN: Thank you. 
Nate, same --same question. 
MR. KALICH: Yeah. I mean, I'd say 
the first of all, the market is very different from 
the equity market. I think they hinted to the fact 
that the order types are much simpler. It's much 
simpler. It's not as complex. 
I'd say historically, again, we've 
been trading electronically on some of these 
platforms for over 20 years, we've had very few 
problems. They're very open. They do a good job of 
transmitting information. I'd say if I was to 
mention one area that I think, you know, more 
transparency would be helpful would be where fees 
are concerned. A lot of the fees are discussed, you 
know, in a bespoke manner and you don't actually 
know where you stand. 
1 
2 
3 
45 
6 
7 
8 
9 
10 
11 
1213 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25Page 232 
funding the nation's debt. So, having the most 
confidence around that market is sort of some ways 
more important than any other market. 
Inside of that conversation, I would 
agree. I think when it comes to the -- you know, 
the platforms have a very strong incentive to have 
us be safe and prudent participants on their market 
and do do an excellent job of providing us with all 
the information required whether it's testing, 
notification, et cetera, so that we could be, you 
know, very -- a safe participant. 
You know, I think as Nate mentioned 
though, when it comes to knowing whether or not we 
are competitive or how our business model looks, you 
know, where there maybe there isn't that necessary, 
you know, obvious line of interest, you know, there 
would be useful to have some additional information 
because I think, you know, to answer your question, 
part of is that we don't know what we don't know. 
We don't know what might be happening and it's not 
been disclosed to us. So, it's difficult to know 
whether it's important or not. And I think just in 
some ways knowing that information and putting it to 
bed one way or the other would be helpful. 
Secondly, I'm very sympathetic --I 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 231 
And so --and that I think kind of 
plays in a little bit --it's not the same thing as 
Fair Acts, but you can imagine if you're paying 
significantly higher fees than others that that 
obviously affects you negatively. But I would just 
say, you know, in --in general, you know, I think 
the markets function, you know, pretty well and 
they've done a pretty good job of --of 
communicating over the years. 
MR. REDFEARN: Any thoughts on that 
one, Ryan? 
MR. SHEFTEL: Sure. I mean, just to 
add, obviously, you know, we are always supportive 
of, you know, a very thoughtful and prudent 
regulation. We think they've consistently shown to 
improve markets and really boost investor confidence 
which I think at the end of the day is one of the 
most important things for any market to have, you 
know, investors that are confident and know what 
they're getting into when they transact with the 
market. 
When we talk about the treasury 
market obviously, you know, as the lynchpin --you 
know, it is the lynchpin of the entire fixed-income 
market. In addition to it's important to obviously 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 233 
mean, we're all here very active, professional 
participants on these markets. You know, I'm 
sympathetic that there are lots and lots of other 
participants in the treasury market that cannot 
dedicate, you know, individuals to be solely focused 
on this. And for them to know, you know, what is 
the access criteria, where can they transact, where 
are they not allowed to transact, are they getting a 
fair deal, you know, to those participants I think 
it's probably almost in some ways even more 
important than us so that they feel confident 
they're getting a fair shake. 
MR. REDFEARN: Right. That is -that 
is an issue that did come with --you know, 
we've learned a lot of things from the evolution of 
the equities market into a much more electronic 
space. And one thing that we did realize at times 
were that those people who, to your user term, did 
the digging and made the calls and talked to the 
operations people and so on and so forth, were 
indeed in cases getting information that was --was 
above and beyond what others might have gotten. And 
that that had some implication in terms of their 
engagement in the marketplace. And so, that was one 
of the reasons why some of these rules evolved the 

59 (Pages 230 to 233) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 234 
way they did. 
So, basically, in this context --so, 
Reg ATS, basically, these amendments were adopted in 
July 2017 for the equities markets, for MMS 
securities, to enhance operational transparency at 
ATS as trade exchange listed stock. 
So, MMS stock and ATS is now must 
publicly disclose detailed information about how 
they operate, including order types, priority and 
execution rules, market day-to-day use, connectivity 
fees, as well as, the nature of their subscribers 
and their relationship to the ATS with its 
affiliates. These disclosures are designed to help 
subscribers to understand how their orders will be 
handled and executed on the ATS and allow them to 
evaluate the ATS as a potential designation for 
their orders. 
So, look, we understand that, you 
know, all regulation needs to be thoughtful and 
prudent and there --there are pros and cons on all 
of this to the extent we've been down this road or 
we've contemplated this, but I guess the question 
that I have is, given the importance of electronic 
trading venues now, is there some baseline level of 
transparency or disclosure that might be prudent at 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 236 
doesn't need to be shown to everyone. 
MR. CLEAVES: I would just echo what 
Ted is saying. I mean, we're not --we don't have 
interconnected markets, right. We're run 
independently and --and actually, compete against 
each other and try to be as fair and open with our 
customers as possible. And --and we provide all of 
the information whether you're a manual trader or 
whether you're an API trader. 
And there's --you know, it's very 
clear when we --when we on-board you we tell you, 
this is --this is the basic market data product, 
here's your --here's your premium market data 
product. And everybody has access to it and there's 
really --there's nothing that we're doing --you 
know, we are a broker-dealer, but we're not trading. 
We don't have affiliates that we give extra priority 
to everybody. It's --it's a single threading 
matching engine. When it hits, it matches and then 
data comes out. And everybody has access to the 
same amount of data. 
We're also the --the pricing source 
of T-500. So, there's on a daily basis about 25,000 
people that watch our prices and I think there's 
close to 50,000 subscribers that watch that page on 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 235 
this point in time for certain venues in --in the 
marketplace. And, if so, what type of information 
did you think that might be. So, I'm just going to 
open that one up and maybe if you guys on this side 
want to chime in and let me know your thoughts 
first. 
MR. BRAGG: Just to start, you know, 
I would suggest, it's a bit different than the 
equity market to what you described because it's a 
professional choice to come and --and trade 
treasuries at Nasdaq Fixed Income as an access point 
to get liquidity. There is no broker-dealer at 
Nasdaq that trades. We don't make any routing 
decision. It's a simple limit order book. 
And, as I mentioned earlier, there's 
no multiple ways to connect and have a different 
response on the platform. There's no NMS, there's 
no NBBO to consider. It's really a --a smaller 
institutional space as opposed to a retail 
marketplace with some investor protection components 
to that. 
So, we believe we're fairly 
transparent to all of our clients or we are very 
transparent to all of our clients on a limit order 
book from what we produce for them, but not --that 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 237 
a semi-regular basis. So, I would agree with Ted. 
MR. REDFEARN: Anybody else want to 
chime in on that one. 
MR. BERNARDO: It would be the same 
in the off-the-run markets as well. The guidelines 
is posted. Everything is posted. It is an open 
marketplace. Everyone is clear and transparent on 
how we operate, the rules of engagement. So, it's 
very, very clear to all of our participants. 
MR. SHEFTEL: I'll chime in. 
You know, I think our view is that, 
you know, thoughtful regulations that have really 
gone through a lot of cycles and other markets make 
a lot of sense to bring over to the treasury market. 
You know, I think the reason for that is that, you 
know, even though the treasury market as we've 
discussed in these, quote, unquote, inter-dealer 
platforms might only be professionals who can spend 
the time and effort digging. They still provide the 
baseboard as we've discussed. I mean, they are the 
liquidity source. Kind of a go-to price formation 
liquidity platforms for the treasury market and, 
thus, by extension pretty much the entire 
fixed-income market. 
So, even I think participants that 

60 (Pages 234 to 237) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 238 
aren't actively involved participants on these 
exchanges have vested interests that they are --you 
know, that they are fair and that they are being 
operated efficiently. And that there's proper 
disclosures so that they can feel confident that if 
they are, you know, trading in corporate bond 
someone is telling them the reference price of the 
treasury for that corporate bond that they can have 
complete confidence in that reference price. 
So, I think anything that creates, 
you know, greater investor confidence throughout the 
entire sphere of participants, not just in 
treasuries, but in the fixed-income market at large 
is a sort of positive good for the entire 
marketplace we need to take into account. 
And, you know, we are respectful that 
no regulation is free. They all bring some 
incremental costs, but I think the key is really is 
that incremental costs provide a benefit to the 
entire fixed-income market, not necessarily to the 
direct participants doing each trade. So from that 
perspective, you know, we are supportive of taking 
lessons learned from other markets. 
You know, one of the benefits of the 
treasury market being sort of late to this 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 240 
market or things like that kind of I think can be 
individual and not --you know, not as well known. 
MR. KALICH: I think I'll just say 
real quick that I think the market is actually 
pretty simple and it's different from the --the 
equity market that way. And I do --I do not think 
that there's a large number of potential market 
participants that are not participating in the 
market because they don't feel they function as 
well. So, I worry about, you know, kind of changes 
in regulation maybe going the other way and 
potentially diminishing liquidity. 
I actually think the market functions 
pretty well right now. And of course there are 
some, you know, corner areas, but I don't think that 
it needs to be significantly changed. It's 
competitive. It's simple. I think it's been 
functioning really well for --you know, for 20 
something years I've been involved trading 
electronically. 
MR. REDFEARN: Yeah. But it's always 
interesting when you ask a bunch of market 
participants if they want more regulation. So, it's 
not something that everybody always just jumps at 
the chance to chime in on that. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 239 
conversation is that where there's a lot of lessons 
learned elsewhere. And if, you know, thoughtful 
regulation has been implemented elsewhere and it 
showed to provide the benefit of increased 
participation or just increased confidence in the 
market, we should really take that into account and 
see how we can bring that over to the treasury 
market. 
MS. DUNN: And so, I would just --I 
would say as well, I'm not --I think the level of 
information is inconsistent and maybe not as fully 
transparent. And while I think many of these 
platforms are not directly interconnected, the fact 
is that people trade across them electronically as 
if they are. 
So, I think where you can access 
liquidity is built into a lot of the algo 
development that exists currently. So, it's hard to 
think of them as not somewhat interconnected even if 
it's the second order affect, but I would say things 
like how ties are broken, to the extent that they 
are, you know, whether there are penalties around 
misbehavior or we have found sometimes that the 
ability to find spot errors may be different or what 
is the protocol for breaking trades that happen off 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 241 
So, there are these concerns, right, 
that --the digging and the asymmetries, right. To 
the extent that there were market participants who 
don't have the people or the staff conceivably to be 
making those calls, to find out the details that 
come from those direct engagements. And it is the 
kind of thing that we're just, we are sensitive to. 
So, I want to switch onto the other 
issue of sort of operational risk, right. So, 
again, you know, there is such a significant --it's 
such an important market and there's such a 
significant amount of liquidity that's trading on, 
you know, electronic platforms as well as, you know, 
off-the-run market work that, you know, we have 
concerns about sort of what are the risks, what are 
the operational risks in the marketplace. 
And so, this comes back to the --the 
other issue that we've dealt with in the U.S. equity 
markets with our --so, we have our Reg SCI, system 
and compliance and integrity. It's a rule that was 
put in place after there was a period of time in the 
equities markets where we saw a lot of --you know, 
we saw a lot of glitches there for a while, right. 
This were, this was down for three hours and that 
was down for two hours and this was, you know. And 

61 (Pages 238 to 241) 


1 
2 
3 
4 
56 
7 
8 
9 
10 
11 
12 
13 
14 
1516 
17 
1819 
20 
21 
22 
23 
24 
25 
Page 242 
there were operational issues on very important 
markets. And so, obviously there's some similar 
concern if we have such an -- such an important set 
of venues in the -- in the U.S. treasury market. 
So, we -- we try to think about the 
way in which operational risk is also managed. And 
so, maybe -- maybe I'll --I'll start with, you 
know, with you guys as well, Shawn, Ted and Dan. 
Just your thoughts in terms of, you know, what do 
you have in place to sort of insure against system 
issues and operational risks in the marketplace. If 
there's a problem, are there checks in place where 
it might stop the trader if they're -- if something 
is going on. 
Please give me a few thoughts for 
what's in place for managing some of these 
operational risks. 
MR. BERNARDO: So, from an 
off-the-run perspective we're not under the same 
rules of engagement as Dan and Ted because we don't 
have that API type trading. We don't have that high 
frequency type trading in -- in the overall trading. 
So, a fair amount of the business, as I said, about 
65 percent of it is done via voice. So, we wouldn't 
run into an issue as far as algo going crazy doing a 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 244 
might want to have from when they might want to 
trade again. And then, to evaluate when we have 
those changes in increments and changes in trades, 
is it something we need to look at from an execution 
quality perspective. 
Those are set by Nasdaq though. So, 
we now have to work with our clients to make sure 
that they know what the settings are just because 
you do get a lot of requests from a lot of our 
broker-dealers and banks to audit those settings and 
possibly test those settings as well. 
I don't think you're asking about the 
kind of technological SCI -MR. 
REDFEARN: So just to follow-up 
on that last point. So, if somebody breaches one of 
those limits then what happens? 
MR. BRAGG: So, if it's credit, hard 
stop. Risk limits, hard stop. If it's increment it 
can be --depending on how they wanted it set a hard 
stop or a speed bump. Speed bump is the wrong word. 
An, Are you sure message. Yeah, are you sure 
message, would you like to trade there from -from 
--from where the market has moved to, but 
we --we don't normally dictate where that --where 
that price formation would be. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 243 
million trades and you accumulating a massive 
position. That's not going to happen in the 
off-the-run space knock on wood. 
But you do have --if --if we were 
to go down or the systems were to go down, we 
obviously have disaster recovery sights where it's 
a --a hot standby and it rolls over. Fortunately, 
at TP ICAP we also have two brands. So, we have a 
Tullett Prebon brand that does off-the-run 
treasuries and we have an ICAP brand. 
So, it's a bit interchangeable. It's 
the same product, same group, doing essentially the 
same thing. So, we're in a --in a relatively good 
position from that perspective if there were to be 
an issue. 
MR. REDFEARN: Got it. 
Ted. 
MR. BRAGG: So, operationally I think 
you mean more like the 15c3-5 controls, risk and 
credit limits and --and clearly we work with our 
clients and set those pre-trade risk limits, credit 
limits, that we review with clients as well. We 
look at minimum size, maximum sizes, changes in 
increments, meaning, how far the markets move from 
the last price and what type of buffer that client 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 245 
MR. BERNARDO: That also goes down to 
the trader level, correct? 
MR. BRAGG: Correct. To each 
individual --so, the firm will set an aggregate 
level --thanks, Shawn. We'll have an aggregate 
level and roll it down. 
MR. BERNARDO: So, you'll have an 
individual trader that sits there that has access to 
his front end. He has his own limits just as long 
as they're not going to exceed the firm's limits 
because Teddy probably works with somebody else. Or 
the desk manager or somebody in risk says, set the 
firm level at this and no one can exceed that. 
So there are, without answering for 
you, limits in place, checks and balances. 
MR. CLEAVES: Similarly, we've got 
controls on each and every user in terms of 
long/short, deviation, duplicative orders. So, if 
somebody's it hitting a button too many times or 
somebody's hitting the keyboard by mistake. And 
similar to an API that might be entering the same 
order over and over again there's this there's a 
threshold quote to fill ratios, et cetera. You 
know, and all --everything that we do is around 
protecting the system, the integrity of the system. 

62 (Pages 242 to 245) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 246 
We've got a market control, market 
support team that will monitor, both from a manual 
perspective and from an automated prospective. 
Alarms will go off if there's an off market price. 
Alarm will go off if there's an off market trade. 
We do have an error trade policy that 
everybody knows and it's pretty clear what will 
happen. If it's within a certain band it would be a 
good trade and we'll do our best --best efforts to 
try to rectify that trade. If it's outside the 
trade will be automatically broken. We'll make that 
decision and communicate that as quick as possible, 
but it does need to be brought to our attention 
within a time frame as well. 
So, that's in terms of protecting 
the --the traders. Then you're talking a little 
bit about the systems as well, right. So, as --as 
I said before, we used the Nasdaq GENI-Ant system 
which is a, you know, really super robust and is 
being used globally to run major exchanges. I think 
the Australian exchange and the Singapore exchange 
and it's a great system. 
It goes through serious, serious QA 
before it ever reaches us. Then it goes through a 
UAT process with our team. Then it actually hits 
12 
3 
4 
5 
6 
7 
8 
9 
1011 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25Page 248 
And if you do have an issue, I think 
we've all learned, we've all been in the market for 
a long time, it's how you respond to that, right. 
It's, how do you communicate it to your customers? 
How do you make sure that there's no nobody at risk, 
there's no trade at risk, there's no open position 
that -- that somebody doesn't know about that 
carries over to the next day or -- or the next hour 
even. So, that's something that you can do. 
And people often when they go down 
they will want to bring a system back quickly, 
right. People will come in and say, we need to get 
back on line, but that's -- that's the worst mistake 
you can make. Trying to come back too fast and not 
making sure that you fully understand the problem. 
Bringing it back on with the integrity, making sure 
that, you know, your sequence number on your trades 
all matched up because it could throw off somebody's 
STP. It's very important that you go through all of 
that -- all of that process to make sure you bring 
it back and the system behaves as possible. Because 
if you go down a second time in that same day you 
lose the confidence of your confidence and the 
industry as a whole. 
MR. REDFEARN: So, I mean, obviously 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 247 
our salespeople, if you can believe that. They're 
actually some of the toughest testers out there in 
terms of making sure that it works exactly the way 
we want. 
We are in the Equinix Status Center. 
We have two matching engines. And then we have a -we 
have a fail over site as well. When we move to 
Globex we'll actually have a fail over site in 
Aurora out in Chicago. So, we'll have two robust 
environments for a total of, you know, I believe 
four matching engines total. 
So, we take uptime very seriously. I 
know that it's something that was on the agenda 
here. As I said, we monitor our system. We also 
have the back up of the Nasdaq team that are on call 
24 hours a day to handle any type of issue that may 
come up and help us rectify. Our track record is -is 
very solid because we don't take our position in 
the marketplace, which is fairly large, lightly. 
Not only from a transactional perspective, but as 
you guys referenced earlier, you know, we are the 
market data and the --and the benchmark that 
everybody uses. Not only to trade U.S. treasuries, 
but to trade other products. So uptime is --uptime 
is critical. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 249 
U.S. treasury market, very big market, very 
important market. Significantly less regulated, I 
would say, than what we have right now in the --in 
the U.S. equity markets on a couple of these fronts. 
The --the context as well here. So, in the U.S. 
equities markets we have Reg SCI. As I had 
mentioned, it was adopted in 2014 to help insure the 
integrity and resilience of the technology of key 
markets and market utilities including significant 
ATS. 
So, I think ATS is one. Once it's 
over one percent notional they then become SCI 
entities. In general, Reg SCI requires the covered 
entities establish written policies and procedures 
designed to insure their systems have levels of 
capacity, integrity, resiliency, availability and 
security to adequately maintain operational 
capability; and, two, then to notify the Commission 
and market participants of system outages, 
intrusions and compliance issues and take 
appropriate corrective action. That was again, put 
in place after there were a number of issues and 
anecdotally it looks like it's been helped. 
So, I guess maybe I'll turn it to the 
folks on the other end now. Maybe starting with 

63 (Pages 246 to 249) 


1 
2 
3 
4 
5 
67 
8 
9 
10 
11 
12 
13 
14 
15 
1617 
18 
19 
20 
21 
22 
23 
24 
25 
Page 250 
you, Deirdre. Just in terms of, you know, what are 
the concerns -- so, sometimes systems do go down and 
when they go down, how does that affect the price 
formation in the marketplace or how you see your 
engagement in the treasury market. 
MS. DUNN: Yeah. So, I think it is 
important to note that, you know, it is expected to 
have occasional disruption due to technology. 
That's not, you know, a totally unexpected thing. 
And I think the market for short periods of time and 
especially with multiple venues can figure out how 
to accommodate and adjust. And, frankly, the 
outages that we've seen this year, you have seen 
some resilience from the market in terms of still 
being able to transact and settle without an issue. 
I think market participants should 
have contingency plans in place, you know, as users 
of the technology to, you know, just prevent 
dependence. I think it gets a little trickier when 
you have a heavy concentration in one specific venue 
or heavy dependence of the market in one specific 
place. That said I think that you'd rather have 
multiple places via compensation, not via 
regulation, right, because you just want to have a 
competitive and a strong -- strong platform not 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 252 
MR. KALICH: I mean, I'll say that, 
you know, in general I think the --the market is 
pretty robust. You know, I expect --I fully expect 
systems to go down, trading venues to --to have 
hiccups once and a while and --but I think the 
treasury market is pretty resilient. I mean, at the 
end of the day there's lots of different ways that 
you can trade in a treasury bond. 
I'll contrast that with, you know, if 
the CME were to go down, for example, you could only 
trade treasury futures at the CME. So, that's kind 
of a different level of problem. So, you know, I -I 
think --I think in general that --you know, I'm 
not necessarily encouraging fragmentation because 
there's problems with price discovery with markets 
being fragmented, but it does have sort of a side 
benefit of kind of creating some resiliency. So, if 
BrokerTec were to go down you could still trade at a 
number of venues. And so, it's not as if the market 
goes dark. 
You know, and I will say that, you 
know, as a person that was trading on September 11th 
and saw, you know, the east wing, basically, go down 
and go dark, the market is a lot more resilient 
today just in terms of --of what they're actually 
Page 251 Page 253 
1 something that's necessarily mandated just given the 1 doing to ensure that they don't go down and there's 
2 level of service and everything else that needs to 2 healthy competition. 
3 be provided. 3 And I think ultimately at the end of 
4 Overall I would say, you know, from 4 the day the treasury market is important. And if 
5 our perspective we have many ways to make markets 5 you can establish, you know, the price, some way, 
6 for our customers and don't depend on any one 6 somehow I think that, again, the market is helpful, 
7 individual platform, nor do we depend on any one 7 so. 
8 individual platform for our price formation. It is 8 MR. SHEFTEL: Again, you know, just 
9 easier and more transparent when all of the price 9 echoing a lot of the other comments. Obviously 
10 formation platforms are working and visible to 10 anti-fragile market structure is the key. I think 
11 everyone who wants access to them. 11 we do have that in a lot of ways which is great. 
12 When there is a hiccup, whether it's 12 You know, to answer your direct 
13 using, you know, futures or other platforms or 13 question of, has there been, you know, Reg SCI as 
14 different ways of communicating with clients, we are 14 proven --again, if it's proven effective and 
15 able to adjust. Obviously, the longer that it goes 15 beneficial in other markets, and I believe this is 
16 on I think the more disruptive overall to the 16 one instance where there's complete alignment of 
17 market. 17 interest across the table here where everybody wants 
18 MR. REDFEARN: So, I guess, you know, 18 a robust, workable anti-fragile system, obviously 
19 for Ryan and maybe Nate as well the question is, 19 within the construct of understanding that there is 
20 again, should significant electronic trading 20 no such thing 100 percent uptime. 
21 platforms be subject to any of these baseline 21 And to the extent there are, you 
22 standards of operational integrity. 22 know, regulations in other markets that maybe, you 
23 Do you have any thoughts on that, 23 know, we encourage competition, but I think one of 
24 whether or not that would be helpful here or --or 24 the keys will be --or one of the benefits will 
25 not. 25 potentially bring in a regulation like Reg SCI is 

64 (Pages 250 to 253) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 254 
making sure that new participants who look to get 
into this market do not use regulatory arbitrage or 
a race to the bottom as their means to be, say, a 
lower cost provider or something. 
So, yes, the participants on this 
table, probably not, but for the ability to make 
sure that new entrants are also held to the same 
high standards I think would be important and 
beneficial. 
MR. HEANEY: Why don't we open to 
FIMSAC questions right now. We'll go a few minutes. 
I'll start with Larry. 
MR. HARRIS: Thank you. 
We've talked a lot about regulatory 
issues. One that we missed though is the discussion 
about surveillance for an identification of bad 
actors. So, I want to ask first, Ryan, are you 
finding that you're occasionally losing, just before 
large events that affect the treasury market, 
whether they're domestic events or foreign events. 
And then more generally, I'd like to ask the 
managers of these various entities that are matching 
things together, what are the capacity for doing the 
forensic research that's necessary to identify when, 
say, perhaps foreigners are intervening in our 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 256 
80 percent of our volume. We have another total of 
115 clients that are again, professionals. We know 
who they are and all those goes through that SMARTS 
application. There's no other orders that get 
routed in in an exchange model. 
MR. BRAGG: We do use a SMART system 
and it's proven to be very effective looking for 
spoofing, layering, any type of behavior that's 
potentially disruptive to the marketplace. In 
addition to that, as I mentioned, we have market 
support, market control who are individuals using 
automated tools or even manual tools. They will 
interject themselves into the marketplace. They'll 
suspend a user, they'll suspend a firm, they'll 
suspend an instrument if they think there's a 
problem, whether something going on in order to 
maintain integrity. 
MR. HEANEY: Let me go to Rick. 
MR. McVEY: Thanks, Michael. 
So, I'm going to reframe a question 
perhaps in a little bit more palpable way and 
redirect it to the reality that all of you are 
regulated. You're just regulated today primarily as 
broker-dealers by FINRA. 
And you may be aware that one of the 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 255 
markets in ways that are inappropriate. Perhaps 
even simply identifying spoofing issues. 
MR. SHEFTEL: So you asked me. I 
guess I'll ask answer. 
No. We've seen nothing in particular 
that makes us lose any confidence in these markets 
or believe that there is, you know, an individual or 
group of bad actors. We certainly have not seen 
that. We don't expect our models to stand through 
time. We're constantly evaluating them, exploring 
and looking at data. It's what we do, but I can't 
say today that we've seen anything today that raises 
concern beyond just the typical, this is a highly 
competitive marketplace with a wide dispersed 
variety of participants and markets change all the 
time and you have to, you know, adapt and keep up 
with it. 
MR. KALICH: And then from a platform 
perspective, and not to speak for Dan, but Nasdaq 
has a product called SMARTS. And Dan's platform 
uses it as well. It's actually how we surveil the 
market to look for spoofing and many other bad actor 
characteristics. And it's an over-the-counter 
market. So, from a bad actor, foreign interactions 
to market, we've --we have 20 clients that are over 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 257 
recommendations we passed at the FIMSAC committee to 
the SEC was that we strongly encouraged them to 
create a working group with FINRA and MSRB to create 
a consistent set of regulations that was efficiently 
applied across the entire industry. So, we've got 
single dealer platforms, we've got inter-dealer 
platforms, we've have multi-dealer RFQ platforms 
generally regulated by FINRA. We have retail cross 
patching systems that are active in government bond 
trading that are primarily regulated by SATS. 
In the case of munis, which we went 
through, MSRB is involved as well. And we also 
through that journey identified systems that are 
operating in U.S. fixed-income markets 
electronically without any regulation at all. So, 
I'm just curious if you --if you think about that 
question a little bit differently and you view FINRA 
as the logical ongoing regulator, would you too 
encourage the SEC and ATS staff in particular to 
coordinate their efforts with FINRA so that we can 
get an effective and consistently applied regulatory 
framework for government bond trading. 
MR. CLEAVES: BrokerTec in addition 
to being a --a broker-dealer is also a non exempt 
agency ATS. So, we go through, we file the Form 

65 (Pages 254 to 257) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 258 
ATS, we comply with the ATS exam rule, we comply 
with the ATS record keeping rule, the ATS 
confidentiality rule. So, different than a 
treasury-only platform due to the fact that we trade 
government Canada bonds as well. So, we are not in 
that exempt ATS. So, we're already adhering to a 
slightly higher standard already. 
MR. BRAGG: And --and, Rick, to your 
question, Nasdaq NFI is actually a registered ATS. 
So, we are complying with the rules and filing as we 
make changes. But to your point, I don't think that 
it's a level playing field. And I think there might 
be other platforms that are not. And we'd be if the 
regulations were to continue to be placed on our 
platforms and the alternative liquidity sources 
could --could gain business around that which is 
what I think I said earlier. The regulation, 
including SBCI, comprised as well the answer would 
be yes, but as long as it didn't create some type of 
unfair advantage or unlevel playing field which 
would negatively affect our business. 
MR. SHEFTEL: I have one point which 
is, I think, to your point. You know, we use a lot 
of terms, like, IDER and those are really more just 
colloquialisms about how these platforms started, 
12 
3 
4 
5 
6 
7 
89 
10 
11 
12 
1314 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25Page 260 
And, while there is an enormous 
amount of trading in on-the-run securities, that is 
a fraction in terms of the number of CUSIPs that are 
based off of this, both whether it's the entire rest 
of the treasury market, you know, which helps to 
fund obviously the country or credit markets or 
anything else. 
So, I think, you know, one thing is 
these products as a reference point and the scope of 
impact of any market structure change is that much 
more material. So, I think that should just be, you 
know, in consideration as working groups are formed. 
MR. KALICH: And I would just say, 
you know, just to add on the flip side it's 
important to just make sure that -- that there 
aren't various entry created. I think at the end of 
the day we want the treasury market to be very 
healthy and liquidity is what makes the market 
healthy. And so, just, you know, to be careful to 
not, you know, basically, entrench existing 
participants and create potent barriers for entry. 
I think that's particularly important in the 
treasury market particularly because again, it is a 
professional market. 
MR. HEANEY: So we've run over a bit. 
Page 259 Page 261 
1 you know, 29 years ago. And the RFQ platforms offer 1 Let's just take one -2 
streaming and a variety of different protocols. So, 2 MR. TABB: This is Larry Tabb. 
3 I think, you know, kind of echo that, I think the 3 MR. HEANEY: Go ahead, Larry. 
4 key is that, you know, any new regulatory regime 4 MR. TABB: Now, talking about open 
5 should cast a wide net to make sure that there's not 5 access that historically there are indices in 
6 that kind of, you know, regulatory arbitrage 6 over-the-counter markets where you know if IADs open 
7 occurring where a platform could just hang up its 7 up access to various participants, the more 
8 shingle with a different name and have an unlevel 8 participants they get put in the penalty box from -9 
playing field. So, I think the key is to, you know, 9 from some other bigger dealers; A, does that still 
10 not get too hung up on the titles of these platforms 10 go on; and, B, should this committee be thinking 
11 and just say, you know, are you a venue that allows 11 about some sort of open access rules around that. 
12 electronic trading of treasuries or other 12 And that could also mean clearing issues as well. 
13 fixed-income products. If yes, then, you know, you 13 Getting access to the FICC and GFCC or something 
14 should be in the scope. 14 like that. 
15 MS. DUNN: So, I would add as well, I 15 MR. BRAGG: I'll start and then I'll 
16 mean, the way people are phrasing these questions, 16 turn over to you, Shawn. 
17 they're very hard to say no too, right. I think we 17 The days of putting people in the box 
18 tend to agree we want high integrity in the 18 really are over. Everybody needs access to the 
19 marketplace. 19 platforms, right. You've got need for access to 
20 I would just -- any exploration that 20 liquidity and pricing, et cetera. And --and for 
21 does happen on a working committee or whatever else, 21 somebody to turn themselves off to a platform right 
22 I think it's very important to focus on the fact 22 now would probably do more damage to their franchise 
23 that the U.S. treasury market is very different than 23 than --than help. You know, back in the days where 
24 many of the other markets where these things already 24 it was, you know, trader/broker relationship putting 
25 exist. 25 them in the box, it was his livelihood that was at 

66 (Pages 258 to 261) 


Page 262 Page 264 
1 risk. 1 it by accident to stop somebody from trading because 
2 We -- we at BrokerTec, I don't want 2 they've gone over their limit. Not because it was 
3 to speak for Shawn and Ted, but we make the decision 3 some functionality to how our matching engine worked 
4 on who we have on our platform autonomously. 4 and the price con priority. And --and our view or 
5 There's nobody that impacts that. What we do is we 5 Nasdaq's view of our current matching engine in the 
6 look for the best ecology that we can -- that we can 6 marketplace is --is a very transparent one. Not to 
7 generate. And it's a combination of banks, 7 advantage any participant in that regard. 
8 professional traders with a couple systematic hedge 8 MR. HEANEY: I want to --I want to 
9 funds and -- and your PTFs on the platform. 9 thank Brett and I want to thank the panelists for 
10 We're not in the business to, you 10 taking the time to come and present in front of us. 
11 know, disenfranchise our major liquidity providers 11 It was very insightful and we appreciate it. Thank 
12 and providers of natural interest into our streams. 12 you very much. 
13 It's not our objective, but ultimately we're making 13 (Applause.) 
14 that decision and we're making that decision because 14 MR. HEANEY: We've got a short 
15 we want to keep the balance in the marketplace. 15 five-minute break and the we have our last panel for 
16 The B to C market is -- is very 16 the day. 
17 strong right now. It's providing great liquidity to 17 (Brief recess taken.) 
18 their customers and I don't think there's really any 18 MR. HEANEY: Okay. Today's final 
19 need to nix that right now. It's not our objective. 19 panel will focus on the update on the status of 
20 MR. BERNARDO: I would reiterate what 20 LIBOR transition efforts and the recent events in 
21 Dan just said. I do -- if they're speaking about 21 the repo markets and the volatility of SOFR. 
22 certain firms, asset managers or different types of 22 Lizzie Baird, Deputy Director in the 
23 firms that have access to our -- to our marketplace, 23 Division of Trading and Markets at the SEC, will 
24 I think those clients are served very, very well by 24 serve as the moderator and I'll pass it over to 
25 Deirdre and the other banks out there, but if they 25 Lizzie. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 263 
wanted access, you know, it would absolutely be 
considered on an open access basis. 
MR. BRAGG: Just --just to 
compliment the one part of Larry's question though. 
I know TMPG did a lot of work on what can we do with 
FICC, the risk that the platforms are underwriting 
so that are a non broker-dealer, professional 
trading firm, as large a market maker as they are, 
FICC made changes two years ago on those 
recommendations and nothing happened. And I still 
think that's as --if I have to evaluate a client 
and I --and if Nasdaq dials down the risk of that 
client, does it --do I just lose market share while 
not everyone else takes the same view at that time. 
MR. HEANEY: Sonali for one final 
question. Sorry, we're running well over. 
MS. THEISEN: Thank you. 
I'm just curious, I think one of the 
panelists mentioned a word in passing, and also just 
because it's a topic that's been contemplated in 
other asset classes, what is the view of any of the 
panelists around speed bumps and whether they have a 
place in the treasury market and whether they would 
and what are the pros and cons. 
MR. BRAGG: So, I said it and I said 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 265 
MS. BAIRD: Thanks, Michael. 
So, as many of you may remember at 
the FIMSAC's April meeting we heard from a panel 
consisting of issuers, investors and a bank dealer 
on the ongoing work to prepare for the transition 
from U.S. dollar LIBOR to SOFR. As with today's 
panel, many of our April panelists serve as members 
of the Alternative Reference Rates Committee, or the 
ARRC, which is the group tasked by the Federal 
Reserve with, among other things, developing an 
implementation plan for the transition away from 
U.S. dollar LIBOR to a more robust reference rate. 
Our final panel today will provide us 
with an update on those transition efforts. And -and 
if you remember back in July of this year the 
staffs from the SEC's Division of Corporate Finance, 
Investment Management, Trading and Markets and the 
Office of Chief Accountant issued a statement 
encouraging market participants to identify and 
manage their risks associated with the impending end 
of LIBOR which I think is now 790 days away, do I 
have that right? The statement also included 
guidance to assist in this effort. 
The LIBOR transition efforts have 
recently received a great deal of attention from the 

67 (Pages 262 to 265) 


1 
2 
3 
4 
5 
6 
78 
9 
10 
11 
1213 
14 
15161718 
19 
20 
21 
22 
23 
2425 
Page 266 
financial trusts due to the volatility experienced 
by the repo markets in September and the resulting 
impact on SOFR. That volatility was so remarkable 
that it precipitated action by the fed brining 
liquidity to the repo market each day that week and 
on an ongoing basis. 
In updating us on the market's 
progress in transitioning away from LIBOR to SOFR 
our panel will also discuss this recent volatility 
in repo and how it's impacted the transition 
efforts, if at all. 
To begin with, I'll ask each of the 
panelists to briefly introduce themselves and 
describe their role at their respective firms. 
Debbie, let's start with you. 
MS. CUNNINGHAM: Sure. 
I'm Debbie Cunningham, Chief 
Investment Officer of the Global Liquidity Markets 
at Federated Investors. Federated is a -- an asset 
management firm based in Pittsburgh, Pennsylvania. 
About 520 billion in assets under management. About 
370 billion of that is in the liquidity markets. 
So, much very much involved in the LIBOR aspects. 
270 billion or so of our assets in 
liquidity products are in 2a-7 products. So, ones 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 268 
the impacts that we have with respect to our models, 
with respect to our applications and other systems, 
as well as, exposure that we have through, excuse 
me, our contracts. 
And so, my job is to manage a small 
team that insures to the greatest extent possible 
that we have an informed group of people in the 
lines of business and other functions that are 
executing consistently and with the correct facts on 
LIBOR transition. 
MS. BAIRD: And you're also with the 
ARRC, right? 
MR. GRABENSTEIN: I am. I represent 
Wells on the ARRC. 
MS. BAIRD: Thank you. 
Tom Pluta. 
MR. PLUTA: Tom Pluta, J.P. Morgan. 
I've been at the firm 24 years in a variety of 
trading and trading management roles across rates, 
effects and emerging markets. I'm currently the 
global head of linear rates trading for developed 
markets. And I'm also the business lead for the 
corporate investment bank for the LIBOR transition 
program and on our firm-wide committee as well. 
Similar to what Brian described, we 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 267 
that are overseen from an SEC perspective and then 
the other 100 billion or so is non 2a-7 actuary or 
separate accounts, other types of short-term 
fixed-income securities markets. 
We're large users of LIBOR-based 
floating rate notes in all of our products. And, as 
such, this is a central aspect to, basically, 
everyday life with the issuers that we're -MS. 
BAIRD: Thank you. 
Brian. 
MR. GRABENSTEIN: Yeah. I'm Brian 
Grabenstein. I'm head of LIBOR transition at Wells 
Fargo. I've with the firm for 17 years in a variety 
of roles that weren't so much project or program or 
transition management, more front office, investing 
roles primarily. 
We set up our LIBOR transition office 
back in February of 2018. Since that we've 
inventoried our exposure across the firm which 
entails something along the order of half a trillion 
of on-balance assets, in excess of 100 billion of 
liabilities. We have significant off-balance sheet 
exposure in terms of servicing trustee activities, 
as well as, contingent exposure related to advisory 
activities. That doesn't begin to describe some of 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 269 
set up a very broad and all-encompassing LIBOR 
reform effort across all parts of the business back 
in 2018. 
MS. BAIRD: Thank you. 
Sam. 
MR. PRIYADARSHI: Good afternoon and 
thank you for the opportunity to speak to the SEC 
and to the FIMSAC. 
My name is Sam Priyadarshi and I'm 
global head of portfolio risk management and --for 
those that don't know Vanguard, it's a large asset 
management globally managing 5.6 trillion and we 
offer 300 funds to 30 million investors globally. 
I'm on Vanguard's LIBOR steering 
committee which is responsible for the transition 
from LIBOR. I also serve on the CFPC's market 
advisory committee and on MRACs interest rate 
benchmark subcommittee and its market structure 
subcommittee. 
MS. BAIRD: Tom Wipf. 
MR. WIPF: Tom Wipf, I'm the Vice 
Chairman of Institutional Securities at Morgan 
Stanley and --and co-lead of our LIBOR transition 
efforts at the firm which is a firm-wide imitative 
to insure across all of our businesses that we've 

68 (Pages 266 to 269) 


1 
2 
3 
4 
5 
678 
9 
10 
11 
121314 
15 
16 
17 
18 
19 
20 
212223 
24 
25Page 270 
got readiness with a key focus from my side of that 
on the client outreach and education. Away from 
that, I'm -- since May I've been chairing the ARRC. 
I am on the board of ISDA and I'm chairing the 
interest rate subcommittee of the MRAC at CFCT. 
MS. BAIRD: Great thank you. 
Tom Pluta, let's go right to the 
recent repo market events. As head of a rates 
business that includes an active repo trading desk 
could you share with us your thoughts on the drivers 
of these recent events. 
MR. PLUTA: Sure. 
So, the backdrop in September was 
that coming out of the QE era the fed had been 
actively and predictably reducing its balance sheet, 
and with that, excess reserves in the system were 
also declining. And we always knew that there would 
be a point with the decline in excess reserves that 
the supply and demand to the funding markets would 
come out of balance. 
The unfortunate -MR. 
TABB: I'm sorry to interrupt, 
but could you please speak into the microphone, I 
really want to hear this. 
MR. PLUTA: So, coming out of the 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 272 
The --the GC markets are trading where you would 
expect back in the normal range. SOFR is setting 
back in the range that you would expect and it's all 
good. 
Additionally, on October 15th the fed 
began purchasing 60 billion a month of T-bills which 
will gradually raise the reserves in the system back 
up again. So, certainly, the measure --this 
combination of actions has certainly had the desired 
results and brought the system back into balance. 
MS. BAIRD: Tom Wipf, do you have any 
thoughts about the potential effects of this 
volatility on SOFR's suitability as a replacement 
for U.S. dollar LIBOR. 
MR. WIPF: Well, certainly it wasn't 
the best day we had on the SOFR PR department, but I 
would say what it did really do, you know, when you 
think about this, certainly it had brought people to 
the table really understanding the differences 
between SOFR and LIBOR. So, there was a lot of work 
heading into this, but really if you think about the 
ARRC's role here, the ARRC was tasked with creating 
a referenced rate that was based on market 
transactions. 
And so, when we think about the 
Page 271 Page 273 
1 fed's QE era they'd been reducing their balance 1 underlying repo market, we've got well over $1 
2 sheet and excess reserves in the system had been 2 trillion a day that's calculated against SOFR. It 
3 declining. And we knew that we would hit a point 3 certainly -- you know, it was -- when we think about 
4 where supply and demand would for the funding 4 what took place, it certainly has focused people on 
5 markets would be out of balance. 5 the negatives, but the fact remains that looking for 
6 The problem that occurred on 6 a market rate that has underlying transactions was 
7 September 16th was it sort of happened all at once 7 our goal. We've achieved that. And I think this -8 
and we had a perfect storm where -- it was a 8 this sort of activity that we've seen in the past 
9 corporate tax day which generally removes about 100 9 few weeks is not going to define SOFR. 
10 billion from the system. And there was also a --a 10 SOFR is a strong rate. It's a repo 
11 settlement of 54 billion of treasury coupon 11 rate based on again, over $1 trillion in 
12 securities. So, that reduction took us out of 12 transactions. So, it certainly brought attention to 
13 balance all at once and repo rates spiked that day. 13 it. I think the one point that we tried to make, I 
14 The -- the fed did respond quickly 14 think pretty clearly, once we sort of, you know, 
15 though. Within a couple of days they came in with 15 unpacked the whole thing was that the fact remains 
16 the TOMOs, temporary open market operations, where 16 that people are not going to use one-day SOFR as 
17 initially they were offering overnight lending into 17 they set their coupons. They're going to use an 
18 the market to put the system back into balance and 18 averaging of these rates. And when on a comparative 
19 then additionally added the term rates of the 19 basis, including the five-and-a-quarter rate that we 
20 two-week repos. 20 saw, you're -- basically, the change in three-month 
21 So, now we have about 70 billion of 21 SOFR on a compounded basis was two basis points. 
22 demand for overnight repos every day and there's 22 The change in three-month labor was four basis 
23 about 150 billion of the two-week funding in the 23 points. 
24 market from the fed. So, that's about 220 billion 24 So, when we really kind of put it all 
25 in the system has brought it back into balance. 25 together and think about how people will use SOFR it 

69 (Pages 270 to 273) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 274 
seemed to be just something that, you know, that was 
going --it wasn't great for in terms of getting 
people on it. Again, people who didn't like SOFR to 
begin with aren't going to like it anymore. 
Nonetheless, the fact remains that we have --we've 
sort of put together a rate that people can use. 
And, in fact, today the fed sent out 
a consultation which will be a way to create SOFR 
averaging and indexes that people can use as well. 
So, when we think about how --how people will use 
this rate, for the most part they're going to use 
averages, compounded in advance or arrears. So, 
when you really look through all the information, 
again, it was a --a bump in the road, but certainly 
not anything that we see as --as an impactful in 
the long run. 
MS. BAIRD: Tom, sticking with you 
for a minute. 
At the April FIMASC meeting you 
provided an overview of the work that the ARRC had 
been doing and the implementation plans from 
transitioning away from LIBOR. Can you provide us 
an update on some of the recent progress made 
referred to the transition. 
MR. WIPF: Sure. And in fact, away 
1 
2 
3 
4 
5 
67 
8 
9 
10 
1112 
13 
14 
15 
16 
17 
1819 
20 
21 
22 
23 
24 
25 
Page 276 
regulatory accounting implications and now having 
answers to those questions really opens a key door. 
So, there's been serious hurdles removed I think, 
you know, by the official sector and beyond that 
really open up the door for people to transition. 
Additionally, obviously the SEC's 
guidance back in July was extremely helpful and we 
found that that created a lot of energy from more 
buy side clients who really wanted to know more 
about this topic on the back of that. 
Away from that, we also see, you 
know, potentially some further guidance from IOSCO 
on, basically, advising market participants that the 
best way out of a hole is to stop digging. So, 
start using SOFR when and where you can. We've got 
over 300 billion in floating rate note issuance 
using SOFR over the last year. 
We have heard from the treasury bond 
committee that the treasury -- that they've advised 
the treasury to take a look at potentially issuing a 
SOFR floater from treasury. And I think really 
where we're getting to right now is, with the 
barriers being removed and a little bit more 
liquidity building, and the last piece of the 
puzzle, which we'll see in the second half of next 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 275 
from what the AARC has been doing which, you know, 
we've got our consultation out on adjustable rate 
mortgages, we've closed off most of our --most of 
the fallback language that we had to put in place, 
new cash products. So, those are all available to 
the market. 
But away from that really over the 
last six months there's been, you know, I think very 
significant developments even outside that, the 
official sector or the quasi official sector. We've 
seen --when we sort of lay out what's taken place, 
over the last several months we've seen regulatory 
accounting and tax news that is very positive. 
So, in terms of the --what the IRS 
has come out with, their first round of guidance, 
which is all in draft form, says that you can 
convert a trade from LIBOR to SOFR and not have a 
tax event. From --from a FASB perspective, same 
outcome. And I think from a regulatory and margin 
perspective we've seen some drafts that sort of get 
us there. 
So, when we were speaking to clients 
say even six months ago, the question we would have 
had is, well, if we're going to want to convert 
trades from LIBOR to SOFR, what are the tax 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 277 
year, is where the CCPs are going to change the 
discounting from fed funds to SOFR. And that's 
going to have a big impact on sending SOFR liquidity 
into the system and could be a real catalyst for 
getting people to get moving. 
So, if there's enough liquidity and 
there's no barriers to conversion and that really 
speaks to the derivatives market, which is an 
enormous piece of the puzzle, and clients have the 
ability to actually transact, you know, you can 
almost see forward to a point where it's just easier 
to sort of restructure than it is to actually put 
all of these fallbacks in place and deal with the 
operational risk that we're going to have at the end 
and actually really implementing those fallbacks. 
So, I's say some, you know, really 
positive developments. You know, the ARRC is again, 
has --has all of our language out. This 
consultation today on the --on how to use SOFR 
averaging and how it can be helpful to people who 
can use this either in advance or arrears. The fed 
being able to hopefully the first half of the next 
year publish those averages on a screen people. It 
will be IOSCO compliant that people can use. And we 
have heard much, much more about that topic from non 

70 (Pages 274 to 277) 


Page 278 Page 280 
1 financial corporates and others who don't 1 come in two different manners, monthly contracts for 
2 necessarily, you know, need a rate that's, you know, 2 the nearest seven months and quarterly contracts for 
3 minute-by-minute. They could just take a rate 3 the nearest 20 quarterly months. And volume in SOFR 
4 that's an average of the last month or two. Also, 4 futures has been growing expedientially. 
5 have a mortgage market plans to approach this as 5 As of August month end the average 
6 well. 6 daily volume in SOFR futures on both these exchanges 
7 So, I think from what we've seen, 7 was over $132 billion. And the total outstanding 
8 it's actually been just a series of events that, you 8 open interest was a little over $1 trillion. 
9 know, on --on their face may seem small, but when 9 Similarly, SOFR interest rates cleared swaps have 
10 you add them all up, there's a lot that you've built 10 reached over 586 billion in outstanding notional 
11 up over the last six months. And I think it's 11 created and over 210 billion outstanding notional 
12 shifting us as we head to the two-year point on 12 cleared across NCH and CME. 
13 this, you know, I think it will focus people on 13 However, average daily volume and 
14 execution and I think we're firmly in at three out 14 open interest in SOFR futures and outstanding 
15 of three right now. 15 notional SOFR swaps are still several magnitudes 
16 MS. BAIRD: When did you see the CCPs 16 smaller than their LIBOR counterparts. Average 
17 are going to switch from fed funds to SOFR. 17 daily volume in Euro dollar futures is over $1 
18 MR. WIPF: The most recent one, 18 trillion and open interest is over 34 trillion. 
19 they're both out in draft forms, is in October -19 
The outstanding notional in interest 
20 October 2020. There was a couple of debates about 20 rates swaps index to LIBOR is over $81 trillion. So 
21 different dates and different methodologies which 21 my expectation is that the big bank for SOFR is 
22 still haven't been resolved. And we did on our part 22 starting in the third quarter of 2020, October as 
23 committee at CFTC invite them to present their 23 Tom mentioned when CME and LCH will switch the 
24 models. They have both have kind of different 24 discounting of swaps --cleared swaps from effective 
25 models, but they have agreed on a similar date. 25 ten points rate to SOFR discount rate. It will 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 279 
Obviously, there's not so much they can do together 
with any trust concerns. 
So, I think we're trying to see if 
there's a role for the CFTC to play to create space 
where they can actually in fact maybe --maybe 
synchronize their models. On having to present the 
risk back, one has an auction method and one just 
hands you the risk back. 
So, there's a few different ways to 
get there. We think they will get there, but we 
think they will get there. And that will actually 
be really, really critical to the --to the 
liquidity into the SOFR market. 
MS. BAIRD: Thank you. 
Sam, one of the issues that market 
participants have asked us about a lot is the 
development of a SOFR turn structure. 
Based on your experience in the 
derivatives markets, we'd be interested in your 
views on the development of the SOFR like futures 
markets and how these markets are developing and how 
they're going to help with the development of term 
rates for SOFR. 
MR. PRIYADARSHI: Yes. So, SOFR 
features straight on the ICE and the CME and they 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 281 
necessitate swap dealers to hedge their discounting 
risk using SOFR futures and SOFR swaps. 
And this may be the inflection point 
as Tom mentioned in the build up of liquidity in 
SOFR futures and SOFR swaps and the potential 
decline in LIBOR swaps. 
On the second part of the question, 
the development of return SOFR is important for 
the --our fallback provisions for cash products 
like floating rate notes, business and consumer 
loans, adjustable rate mortgages and securitization 
that is currently tied to LIBOR all for new issuance 
of SOFR. 
So, SOFR is an overnight rate based 
on volume weighted median of transactions in the 
repo market. Unfortunately, term lending volume on 
transactions in the repo market are not significant. 
To have an IOSCO compliant benchmark turns SOFR rate 
has to be computed from silver futures and SOFR IOS 
funds. 
So, fed's staff has proposed a 
methodology to compute a forward path for overnight 
SOFR from futures pricing allowing the average rate 
to jump down from the CME meeting dates. This 
methodology assumes that the overnight SOFR rate is 

71 (Pages 278 to 281) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 282 
constant until the MC meeting dates. And a forward 
looking compounded SOFR rate can be in for --once 
the forward SOFR rates are computed. 
So, two questions arise. One is the 
volatility of the term SOFR and the other is the 
correlation of the term rate with real SOFR. On 
both these metrics the proposed methodology does 
well. As Tom mentioned, the volatility in the SOFR 
market is actually dampened if you take the 
compounding both MX over a period of three months. 
So, that dampens just one single day volatility. 
And indeed the --the term SOFR does has to realize 
the SOFR historically speaking. 
Governor Rand Paul, from the Federal 
Reserve Band and currently the chairman of the 
Financial Stability Board, has recently proposed 
that the fed consider abolishing the compounded SOFR 
that the market participants can use. And he called 
this safer, sharpe for secured average financing 
rate. One can safely assume that pun was intended. 
MS. BAIRD: Thank you. 
Debbie, recently market participants 
have questioned using SOFR as a reference rate for 
certain products because it lacks a credit 
component. Can you provide us your views of the 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 284 
free securities versus the securities with a credit 
component associated with them. 
And, for that reason when we owned 
securities at that time that had LIBOR as their 
basis, our securities as they were resetting off of 
this very much higher index at that point was --was 
resetting higher as well. So, we felt less 
compelled from a needs perspective to want to sell 
them into the marketplace. 
We felt we were maybe not being as 
compensated as we would be if they were new issued 
securities, but, you know, there were 12-month 
security and they were six months into it, what we 
thought was adequate compensation was coming simply 
from the rise in the --in the benchmark or the 
reference rate at that point. So, SOFR will act in 
exactly the opposite -MS. 
BAIRD: Right. 
MS. CUNNINGHAM: --direction. And 
based on the fact that it's overnight, doesn't have 
a term, it's collateralized not, you know, unsecured 
and it's collateralized with treasury securities, 
the risk free rate, we expect that SOFR will act 
much --much the opposite as I said during that 
periods of credit stress. 
Page 283 Page 285 
1 adoption of a reference rate that is secured rather 1 And as risk free securities the 
2 than one that has a credit component. 2 yields that we'll see on this bench market 
3 MS. CUNNINGHAM: Sure. It's 3 essentially are probably likely to decline during 
4 definitely less than ideal and it has a lot to do 4 that time period because their flight to quality 
5 with what we've gotten used to in the context of 5 securities. They're going to be more in demand, in 
6 having LIBOR as our reference rate for many decades 6 desire, in a higher risk marketplace. 
7 at this point. LIBOR as we all know is based on 7 So, as owners of the floating rate 
8 banks and those banks definitely have a credit 8 issues that have that as a benchmark, we're now 
9 component to them. 9 being compensated less during periods of stress 
10 And as such, during periods of stress 10 based on what was historically being compensated for 
11 in the marketplace, the interest rate that we would 11 more. 
12 receive from LIBOR-based floating rate securities 12 And -- and so, as a liquidity fund 
13 was higher. And it, thus, compensated us for owning 13 manager, I would be more compelled to -- to try to 
14 and continuing to owning those securities and --and 14 likely sell these securities in periods of stress 
15 trying to make up for the additional risks that were 15 which, you know, maybe I'd get a higher price for 
16 inherent in the marketplace at that particular point 16 them because they are more desired at that point in 
17 in time. 17 time, but quite frankly, I'm probably less -- less 
18 It was --it was probably most noted 18 compelled to actually own them in the first place 
19 in the 2007, 2008 time period during the financial 19 without some sort of additional spread involved all 
20 crisis when one-and three-month LIBOR which are the 20 the time based on, you know, basically, what might 
21 most popular for benchmarks for index resets in 21 be instead of what has been. 
22 floating rate securities in the --the marketplace, 22 And -- and, you know, therefore, I -23 
rose anywhere from 150 to 300 basis points during 23 even if I do buy them, I'm probably more compelled 
24 that time frame above what would be basically the 24 to buy shorter maturities where there's less risk of 
25 similar tenor treasury securities. So, the risk 25 immediate change in the market condition. So, less 

72 (Pages 282 to 285) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 286 
than ideal without a credit component associated 
with it. 
MS. BAIRD: That makes sense. 
Brian, would you share with us some 
of the steps that the ARRC has taken with respect to 
fallback language for floating rate notes and --and 
tell us how successful in your view the industry has 
been in assessing the risks to cash products that 
reference LIBOR now. 
MR. GRABENSTEIN: Sure. Well, with 
respect to fallback language maybe I'll take a step 
back just to make sure we're all on the same page 
with --with respect to what we mean by fallbacks. 
So, fall back language is the 
contractural provisions that are available in most 
LIBOR based contracts that say what is to be done if 
LIBOR doesn't exist, right. 
Now, historically, fallback language 
was written with the assumption that maybe you 
wouldn't have LIBOR for a day or a couple of days. 
It wasn't written again historically with the 
possibility that LIBOR might go away entirely, 
right. So, if you will, legacy fallback language 
that's in trillions of contracts would be very 
problematic if LIBOR were to go away tomorrow, 
1 
2 
3 
4 
5 
6 
7 
89 
10 
11 
12 
1314 
15 
16 
17 
18 
19 
20 
2122 
23 
24 
25 
Page 288 
publicly issued securities. And this is a little 
bit dated, but it looks like certainly a plurality. 
Something on the order of three quarters or more of 
the newly issued floating rate notes are using 
either almost verbatim ARRC language or something 
very close to it. We don't see the same level of 
uptake in some of the other asset classes. 
MS. BAIRD: Debbie, we've seen 
different countries pursue different approaches to 
transition away from their currency based LIBOR. 
What are your thoughts on the global approach to 
managing the -- the transitions from LIBOR. 
MS. CUNNINGHAM: Well, currently the 
LIBOR panel banks provide indices for five different 
currencies across seven different maturities. And 
all five of these indices are, basically, the same 
with maybe a few different banks, but the components 
are, basically, representing credit spreads for what 
are high quality bank liquidity in those 
marketplaces. 
So, whether I'm investing my dollar 
denominated products here in the United States, our 
U.K. denominated products -- or our sterling 
denominated products in the U.K., euro denominated 
Luxenberg, yen denominated in Japan, essentially 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 287 
right. 
So, one of the major bodies of work 
that the ARRC undertook beginning in 2018 and it 
took the better part of 15 months was for cash 
products to look at developing and endorsing and 
issuing to the market better, more robust fallbacks. 
And they did that across floating rate notes, 
syndicated loans, bilateral business loans and 
securitizations including CLOs. And as Tom eluded 
to earlier, we are almost through the process -finished 
with the process to endorse fallback 
language for residential adjustable rate mortgages. 
I should finally add that ISDA has 
been undertaking a similar work stream for the 
better part of three years now with respect to 
derivatives. 
So, with respect to floating rate 
notes specifically, we endorsed the language in 
April of this year --the ARRC endorsed the language 
in April of this year. It was the product of a -of 
a consultation process and it aligns at a very 
high level with all of the other cash products as 
well as what we expect ISDA to land on. 
Covenant review has done some --some 
research on adoption of the fallback language in 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 289 
we've got the same thing when we're buying an IBOR 
based index. It's high quality bank lending spreads 
in the marketplace. 
As we move forward, what the U.S. has 
chosen from a SOFR perspective as we mentioned is an 
overnight rate that is secured by treasury 
collateral. So it's, basically, risk free in the 
context of, you know, the --the reference process. 
The U.K. has actually chosen SONIA which is Sterling 
Overnight --the Sterling Overnight Index Average 
which, basically, tracks the rate of actual 
overnight funding on all the deals that are offered 
in the wholesale market in the U.K. So, it is not 
collateralized and it has a credit component. 
Although that credit component is only on an 
overnight basis. 
The ECP has recommended ESTR which is 
the euro short-term rate and that's its 
recommendation for its replacement for euro IBOR. 
ESTR is based on transaction data collected as part 
of their daily money market statistical reporting 
from the 52 largest euro area banks. And as such, 
it also includes some sort of a credit component for 
spread, as well as, some term associated with it. 
And again, no collateralization. 

73 (Pages 286 to 289) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 290 
So, what you have is a mixed bag. 
And because of these different choices there's, 
basically, going forward what we would consider to 
be a lack of continuity across these major global 
financial regions in the world. And, therefore, the 
financing that happens out of Europe, the financing 
that happens out of Tokyo, the financing that 
happens out of New York and how they are funded and 
what the indices look like are going to be likely 
different. And --and, therefore, structured 
differently which ultimately, in our minds, leads to 
what we think is probably less efficiency from a 
structuring perspective, less efficiency for global 
issuers who may in fact have securities issued in 
all of those markets and what essentially probably 
leads to less liquidity for market participants in 
these markets. 
MS. BAIRD: Okay. Let's close with a 
discussion of market preparedness. I'd like to ask 
each one of you to share your views on --on the 
progress made by the industry in transitioning from 
LIBOR and talk about what further steps you think 
remain both in the short-term and the long-term. 
So, Tom Wipf, if you can start us 
off. 
1 
2 
34 
5 
6 
7 
8 
9 
1011 
12 
13 
14 
15 
16 
17 
18 
1920 
21 
22 
23 
24 
25 
Page 292 
fallbacks we would have had significant disputes 
litigation and worse. 
So, the fact is that when we think 
about this is the 19 panel banks that actually 
agreed to stay on to the end of 2021, that the FCA 
as the regulator of LIBOR, and these are Andrew 
Bailey's comments from the FCA, which is, when a 
bank or banks leave he's required by law to declare 
this rate representative or not representative. 
If he were to declare it not 
representative that triggers all sorts of things in 
Europe and U.K. and other places in the world. So, 
the idea I think that people have really come around 
to with those explanations which is, wait a minute, 
this isn't something where someone just waves a wand 
and extends the deadline. This in fact has a lot of 
different component pieces that are not necessarily 
under anyone's complete control. 
So, the idea would be that, you know, 
preparing for this becomes -- you know, it becomes 
really important. And I think as we talked about 
the fallbacks and protocols and the things that 
people need to do, that's actually happening. We're 
seeing it in new issue. I think the use of 
fallbacks is going very, very well. I think people 
Page 291 Page 293 
1 MR. WIPF: Sure. Thank you. 1 going straight to SOFR. We'll be seeing some 
2 I think when we think about the sort 2 developments. 
3 of the end of LIBOR and the preparations that have 3 And I think what has happened is 
4 to be in place and the tools available to smooth 4 there's been a change in sort of tone in terms of 
5 those outcomes, I think the important consideration 5 what the deadline is and things like term and credit 
6 I have to say over the last six to twelve months 6 are not necessarily going to be there when we reach 
7 people are in the market have really come to accept 7 this deadline. And people in the market are now 
8 that the end of LIBOR is at the end of 2020. And 8 beginning to think about what they can do. 
9 that really came as a series of statements from the 9 For instance, regional banks struggle 
10 official sector. Certainly, most recently, 10 with the lack of return piece in terms of just a 
11 President John Williams of the Euro fed talked about 11 lending versus a cost of funds. Many are talking 
12 death, taxes and the end of LIBOR are the three 12 about putting floors in and creating different ways 
13 certainties in life. 13 from keeping that spread from collapsing. They can 
14 And I think there's also a lot of -14 
at least attach floors and things like that. And I 
15 it took a long time because of just the long-dated 15 think what we've seen at the ARRC is, you know, we 
16 nature as we approach two years, people will get 16 set out to create something that was based on the 
17 there, but the fact is that, you know, the idea 17 criteria durable and it was --and it was defined by 
18 isn't that the FCA is trying to shoot LIBOR. Nor is 18 transactions. 
19 the official sector. 19 Neither the ARRC, nor any --any 
20 The fact is that the 19 panel banks 20 private institution out there, has been able to 
21 back in 2014 many were being to leave the panel in a 21 actually find enough underlying credit transactions 
22 less than orderly way. So, the end of LIBOR could 22 to meet the standard scope guidelines. So, the fact 
23 have come with five or ten big banks leaving and 23 remains is that without that we look at something 
24 creating an enormous amount of chaos. And as Brian 24 that would be reliant on submissions which feels a 
25 mentions, on the fallbacks, you know, without those 25 whole like LIBOR and if we think about what was 

74 (Pages 290 to 293) 


1 
2 
3 
45 
6 
7 
8 
9 
10 
11 
12 
13 
1415 
16 
17 
18 
19 
20 
21 
22 
2324 
25 
Page 294 
really at the heart of the LIBOR scandals, it was 
about that bank credit piece -- that sort of shared 
bank credit piece. 
Within three pieces we have, we have 
issuer credit and we have this little slice which 
is -- which is this assumed bank credit rate. So, 
without that I think what people are beginning to 
see in the market is they have to actually think 
about different ways to approach this. And knowing 
that, you know, certainly in terms of term I think 
if we get the big bang, we start seeing liquidity, 
there may be enough transactions to create an IOSCO 
compliant term and that could happen. 
That's obviously contingent and the 
market has to kind of put the liquidity there for 
that to happen, but if that doesn't happen then 
people will be very reliant on averaging of SOFR 
both compounded in advance and arrears. If we think 
about the markets, the derivatives markets, the 
floating rate note market, that's all workable 
because that's fairly consistent with the 
conventions that we have today. 
So, I think where we are right now in 
terms of progress is with -- with the things that we 
laid out earlier in terms of the help we're getting 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 296 
and advance and average in arrears. 
So, there's been a little bit more of 
a reckoning. I certainly think that again, we've 
heard from the official sector, the guidance we've 
received from the SEC has brought people to the 
table. And I think in terms of awareness, plans, 
I'd say, every large organization has a LIBOR 
transition program in place. And I think medium 
firms as well. Everyone we visit has people 
responsible for this. And every client that we see 
is concerned about it. 
So, I'd say where we are right now 
there's a hug focus, a lot of programs in place, a 
lot to do and certainly a lot more to do as we 
approach the next two years. 
MS. BAIRD: Sam, what you have you 
seen. 
MR. PRIYADARSHI: I think the 
industry is making decent progress along the 
transition plan. A lot of market participants are 
coming --still coming to the realization that LIBOR 
fixation is a given and a certainty. A critical 
step in the transition will be the big bang for SOFR 
discounting in third quarter of 2010 and will cause 
the markets to proactively manage the new value 
1 
2 
3 
4 
5 
6 
78 
9 
10 
11 
12 
13 
14 
1516 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 295 
in terms of the transition that we have people in 
the market who I think really believe in this 
deadline as they begin to understand exactly how 
LIBOR ends. Now I think we just have to understand 
what tools will be available by then and what tools 
aren't going to be available by then. 
And by no means is the ARRC trying to 
dispute, you know, any -- if anyone can figure out a 
credit spread, we haven't and we have a pretty big 
group of people, about 800 people in the market are 
involved with the ARRC. The current administrator 
of LIBOR has not been able to find one. 
Nonetheless, if there's a way to create something 
like that that meets the standard, that's great. 
I think what has to happen between 
now and then is people in the market have to really 
start thinking about -- a little more creatively 
about how they actually continue the activities. 
And what we're seeing a bit more of is that there's 
a competitive dynamic here. You know, people want 
to offer these products, and if they have to offer 
these products without LIBOR, then people are much 
more thinking about, how do I actually continue my 
lines of business in a world without LIBOR with the 
tools on the table of being SOFR, overnight average 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 297 
transfer during before and after this event. 
Market liquidity for SOFR should 
double up and needs to reach a critical mass for 
market participants to transition voluntarily. 
Jurisdictional differences in terms of the 
alternative reference rates and their fallback 
provisions need to be harmonized and many 
differences in the timing of the fallback could lead 
to mismatches between cash products and their 
hedging with realtors. 
Now, ISDA also needs to finalize the 
protocol for the big bank adhering to the fallback 
provisions for derivative products. And that's a 
much simpler lead than the fallback for cash 
products. Most of the LIBOR index products are in 
place. Almost 35 percent of it in U.S. dollar 
LIBOR. So, once it's finalized and once the market 
participants adhere to the protocol then the 
transition will be well on it's way. 
MR. BAIRD: Thank you. 
Tom Pluta. 
MR. PLUTA: I guess I would just 
highlight a couple of things that I think are going 
well and a couple of things that are still hurdles 
or uncertainties. 

75 (Pages 294 to 297) 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 298 
So, clearly development of tradeable 
SOFR products that are functioning well is great. 
The market has come a long way. It's growing from a 
low base, but growing rapidly. It's all quite 
positive. And development of the fallback language 
and offering the various products that Brian 
described I think is also another good thing. 
Some of the uncertainties that I see 
this issue of having a forward-looking term rate is 
huge. There's certain products where we had an 
IOSCO compliant forward-looking term rate sooner the 
market would transition sooner. Particularly cash 
markets, loan markets, things like that. Which, 
just the way the market operates, the systems, the 
mentality of the market is very much need a rate in 
advance, I know my payment three months forward. To 
the extent that we had a forward-looking term rate 
come sooner would be a massive accelerant to the 
transitioning. 
The second thing I would highlight is 
on this topic of credit risky rate. I think there 
is a bit of confusion around whether a second credit 
risky rate will develop alongside SOFR. There's a 
couple of rates that are being suggested or 
developed, but as Tom highlighted, the challenge has 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 300 
were implemented, again, just for legacy portfolio, 
that would reduce the litigation significantly. 
MS. BAIRD: Thank you. 
Brian. 
MR. GRABENSTEIN: So, I'll try not to 
repeat anything that we've heard already. Although, 
I broadly agree with it. Maybe just a couple 
additional points. 
So, a tremendous amount has been done 
in the last year, in the last, you know, 24 months. 
I think back a year ago and the levels of 
mobilization, the levels of awareness of 
understanding of acceptance that SOFR was --that 
SOFR was going to be the replacement, that LIBOR was 
actually going to go away. They've all increased 
dramatically. 
That said, there are still 
significant parts of the market that have not 
accepted SOFR as the LIBOR successor. And with, you 
know, 790 days left that's kind of a precarious 
position to be in, right. You know, I --I think we 
mentioned --we talked a little bit about fallbacks. 
They're available for almost all products at this 
point. ISDA is almost done its work. It's going to 
come out with a protocol and facilitated option, but 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 299 
been the underlying transactions to support them 
are --are generally fairly low level. So, we're 
not sure we'll get there in a way that will be 
deemed IOSCO compliant and robust enough for usage. 
The third thing I would highlight, 
it's a huge concern for us and a lot of others in 
the industry is litigation risks. So, no matter how 
well the fallback language is designed, they're not 
perfect. And no matter what we do collectively as 
an industry through the ARRC and other --and other 
bodies and ISDA, there will be groups of people who 
say, this fallback was created and I've been harmed 
by it because it's higher or lower than it would 
have been otherwise. So, that's a huge concern for 
us. 
It's unclear if there's going to be 
any relief or anything that can be done to avoid 
that, but there's certainly a scenario where you 
could actually get to the end of 2021, and while 
banks wouldn't be compelled to submit LIBOR beyond 
that point, you could see a situation where banks 
could choose to submit for some period of time LIBOR 
that would apply to the legacy portfolio. And I 
know that that suggestion has come and gone at 
different times, but it's certainly something if it 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
Page 301 
people aren't using them, you know, six months or so 
after they've been endorsed to the extent we would 
have hoped. We being the ARRC. So, I think we need 
increased uptake of the ARRC fallbacks. 
One thing that hasn't been touched on 
is --is the technology and --and the associated 
vendor dependencies to get this. So, we've only 
seen two SOFR-based loans issue to date, right. 
Well, you know, my take on that is it's not --it's 
partially because of lack of SOFR liquidity, right. 
It's partially because clients maybe aren't 
demanding it yet, but it's largely because the 
infrastructure to support those loans isn't yet up 
and going. I don't mean to pick on anyone, but they 
are used by many banks for syndicated loans. So 
until, Loan IQ and other vendors can support SOFR 
it's very difficult for markets to develop. 
And the ARRC's engaging with vendors 
and --and other infrastructure providers, but 
that's --that's a long, you know, work stream --a 
long and very complicated work stream where a lot 
more progress will need to be made over the next 
year. 
MS. BAIRD: That's a good point. 
Debbie. 

76 (Pages 298 to 301) 


Page 302 Page 304 
1 MS. CUNNINGHAM: I'll just summarize 1 standardized fallback language. 
2 by saying that it seems that a lot of the progress 2 MR. WIPF: Yeah. I think the ARRC 
3 that has been made has been focused on choosing the 3 has pursued that and are legal working group is 
4 index that has the sufficient volume. So never, 4 approaching that from really what we call a tough 
5 ever to go down, you know, the scandalous, 5 legacy. Things that have no fallbacks things that 
6 manipulative sort of path again when -- when setting 6 have. Fallbacks that reference some version of 
7 these types of rates durable as -- as Tom has put 7 LIBOR. 
8 it. 8 So, there's been some work done in 
9 And that's step one, but I still 9 terms of approaching this at a New York level, you 
10 think, you know, steps two through something more 10 know, New York Law level. There are particular 
11 than that are going to include all the things that 11 hurdles to that, but nonetheless our legal group 
12 we identified that I'm not going to go back over 12 sort of put it together, is still in progress and 
13 here today with my answers, credit spread term 13 I'll pass to Brian where he sees this, but I think 
14 curve, global continuity, market participation in 14 the --the view right now is, there could be a 
15 all aspects has many challenges, but you know, we 15 legislative path for things that are most 
16 are all challenged I think with coming up with 16 challenging and can't be repaired through protocol 
17 creativity. 17 or existing fallbacks. 
18 I as an investor in this marketplace 18 That work is going to happen --that 
19 have to figure out a way to go forward using what 19 work is underway. We've sort of identified that we 
20 indices available to me that makes sense for the 20 have something that would meet a constitutional 
21 various products that I manage. And, you know, in 21 challenge. We have some things that we think could 
22 today's environment I don't buy just LIBOR based 22 apply, but it's a really, really long road and it's 
23 floaters. And, you know, we -- we buy fed funds 23 a really difficult path. We are pursuing it, we 
24 based floaters. We buy SOFR based floaters. We buy 24 think, at a reasonably aggressive pace, but there's 
25 treasury based floaters. 25 a lot we have to do in terms of actually how do you 
Page 303 Page 305 
1 So, there are ways of getting around 1 take something like that up. There's implications 
2 where you don't have to buy the reference benchmark 2 from the trust venture act. So, there's a lot on 
3 rate as your floater. So, it's a matter of, you 3 this. Brian, kind of. 
4 know, coming up with those types of product 4 MR. GRABENSTEIN: Yeah. No. I think 
5 enhancements or creativity again, as Tom has called 5 you did a good job of describing what the ARRC's 
6 it, in order to fit the type of --of arrangements 6 done to date on legislative relief. 
7 that you need for what your aspect of the market is. 7 I will add that it's not help that 
8 MS. BAIRD: Questions. 8 2020 is an election year and, therefore, a shortened 
9 MR. HEANEY: Why don't we open it up 9 legislative calendar. It's going to make it even 
10 to FIMSAC member. Horace. 10 more challenging to get anything, you know, even 
11 MR. CARTER: Thank you all for being 11 considered in the near term. So, I don't think 
12 here today. 12 we're going to have -- I think it's unlikely we'll 
13 Just a question for you, Tom. You 13 have certainty on this for quite some time which 
14 mentioned the possibility of litigation risk and 14 means we have to proceed assuming that we won't get 
15 continuation of LIBOR after it's expiration. I've 15 it. 
16 heard it called zombie LIBOR and all of that. 16 The only other thing I would add is 
17 That's been a concern. In the last meeting, the 17 this would be -- what the ARRC is currently 
18 last time you all were here, we discussed the 18 evaluating in terms of legislative relief would not 
19 possibility of a legislation to prevent that. Is 19 apply to all LIBOR based contracts. And it is under 
20 that still on the table or is that --is that 20 New York State law, right. So, there's a lot of 
21 something that you discussed at all with New York 21 LIBOR based contracts that are subject to other 
22 and Delaware. 22 jurisdictional laws that would not be covered under 
23 MR. PLUTA: Legislation to prevent a 23 it with the ARRC's pursuing right now. 
24 zombie LIBOR situation? 24 So, it's --it's uncertain whether we 
25 MR. CARTER: Legislation to --to 25 would get it. It's uncertain exactly what it would 

77 (Pages 302 to 305) 


Page 306 Page 308 
1 look like. Although we have sketched out what we 1 I mean, the numbers we're talking 
2 would pursue and it's uncertain the timing. 2 about are in the United States the dollar LIBOR over 
3 MR. HEANEY: Larry. 3 200 trillion in financial contracts currently 
4 MR. HARRIS: I understand that the 4 resting on half a billion in transactions. So, when 
5 search for a benchmark rate that includes credit 5 we really think about it part of this is, A, what 
6 risk is challenging. I want to suggest an 6 the underlying transactions are; and, B, what 
7 alternative and wondered what the challenges are 7 they're supporting, right. 
8 associated with the alternative. 8 So, I think when you think about 
9 Instead of looking for a benchmark 9 how --how big you want that inverse pyramid, there 
10 rate, you'll find a benchmark spread so that people 10 are states of the world where you can see --where 
11 who construct contracts can then benchmark to the 11 you could definitely see for particular products, 
12 SOFR and then add something else based on some 12 cash products, consumer, there's room for all of 
13 fraction of --of a spread. So, for a spread then 13 that, those ideas. 
14 one might imagine could get something like that, 14 I think when we think about the vast 
15 say, from short-term corporates or something like 15 majority of the market, the reason we have this 
16 that. 16 problem with LIBOR is that because it became 
17 So, I'm sure you guys have thought 17 extremely popular across many, many, many products. 
18 about this. The question is, what are the 18 It was designed for an oil loan in 1969. So, it's 
19 challenges of moving in that direction. 19 just --the fact is we have too much resting on this 
20 MR. WIPF: I think that's the work 20 pilar and the goal at the ARRC is to say, we've got 
21 that's been done. I think the current administrator 21 overnight SOFR, we've got compounded SOFR. Those 
22 is going down that road and pulling the data, but 22 seem to be the pillars that can support a vast 
23 the data is really thin. And I think what we've 23 majority of the market, but when we get under that, 
24 noticed even during this SOFR spike litigation is 24 there's nothing to say that there are things that 
25 that there wasn't much paper issuance, if any. So 25 could happen in smaller cash products. 
Page 307 Page 309 
1 what you find in this sort of theoretical time of 1 We're not ruling anything out, but we 
2 stress that the fact is that you're not going to see 2 know that if we make meaningful progress in the 
3 an issue. People are typically not going to issue 3 derivatives market and large institutional markets 
4 when spreads are at their widest. 4 and mortgage market, there are real opportunities 
5 So, the fact is we will find 5 here for a lot of ideas to develop. What we don't 
6 ourselves back again looking at submissions or 6 want people to think is that our deadline and good 
7 guesstimates by whomever wants to continue to 7 deadline is predicated on those answers. 
8 participate. And this idea of zombie LIBOR, we did 8 MR. HEANEY: Other questions? 
9 not have a single bank on the panel that said they 9 Elisse. 
10 wanted to continue to submit one day beyond 2021. 10 MS. WALTER: Just one quick follow-up 
11 So, it's very hard for us to speculate. 11 question I think, Tom, for you again. 
12 The fact remains is that there are 12 You mentioned using an average and 
13 clearly -- there just has not been a big enough data 13 are there going to be -- is it is under 
14 set to do anything that would even look like it and 14 consideration, I would assume it is, to develop 
15 it would be reliant in many cases on some submitter 15 standardization with respect to how the averages 
16 actually putting that guesstimate continuing to 16 would work. 
17 short that litigation risk which is sort of how we 17 MR. WIPF: Yeah, the goal -- and this 
18 got here. 18 is something that's been long awaited. The fed 
19 MR. HARRIS: How important is it to 19 talked about this thing, SAFR. I'm not sure that's 
20 have a benchmark based that's based on new issuances 20 going to be the final name. There's more work to be 
21 as opposed to secondary trades. 21 done in the marking department. The idea that the 
22 MR. WIPF: I think the current -- the 22 New York fed will produce averages of SOFR and an 
23 current administrator has looked at this under this 23 index that people can look at the convention or the 
24 heading of bank yield index. There's just not 24 consultation that was sent out today is really 
25 enough data to support it. 25 asking for responses on how that convention would be 

78 (Pages 306 to 309) 


Page 310 
1 applied. 
2 So you --you'd be surprised at how 
3 many people in the market that we speak to, you 
4 know, who are infrequent users or, like, if I could 
5 just --is there a screen I can look at this? Can I 
6 see a rate? Can I use that rate for my next 
7 payment? The mortgage market that --the design 
8 that Fannie and Freddie came out with is it's, 
9 basically, going to take the average of the last 30 
10 days of SOFR and that will be your next payment. 
11 So, the idea that these averages are 
12 probably going to have a lot of utility across the 
13 market and the fact that it will come from an 
14 unsaleable source from the New York fed, that's all 
15 going to work out. 
16 So, the conventions will be 
17 standardized. We're hopefully that that's the one 
18 people will use because they're putting a lot of 
19 work out. The consultation was just released today 
20 on the fed website, the ARRC website. And we're 
21 hopeful that there's going to be a reasonable take 
22 of that particular products who aren't that 
23 concerned about the last month of the curve. 
24 MS. WALTER: Thank you. 
25 MR. HEANEY: That's --any other 
Page 311 
1 questions. 
2 Okay. I want to thank Lizzie and the 
3 panelists. This was incredibly informative. And if 
4 I had to guess, I'm going to guess we'll see you 
5 again at least one more time in the next year if you 
6 have the time to join us. 
7 If you just sit tight, we'll be ready 
8 to go in one minute. 
9 Thank you all again. It's been a 
10 very successful day, insightful, long day. With one 
11 year to go we had a lot of answers to some of our 
12 questions. I would say certainly in the afternoon 
13 panels. And we have some more questions that remain 
14 from the morning panels. I think both municipal 
15 disclosure issues and the affect on retail questions 
16 still seem to be outstanding. 
17 And, similarly, the credit rating 
18 agency construct and the real questions regarding 
19 the clear conflicts of interest that we've heard. 
20 Again, we had a blank sheet of paper we would 
21 reconstruct the model the way it's currently 
22 constructed. So, plenty more work to do for this 
23 committee over the next year. 
24 Again, I thank everybody. I'll 
25 Chairman Clayton and Brett from this morning, 

Page 312 
1 everybody's enthusiasm and hard work as we are 
2 finishing up our second year and headed into the 
3 third year there's plenty of important things to do 
4 and thank you for your time. 
5 As always, if there's futures topics 
6 for consideration, please send them through. On 
7 that -- in that light we have a task force within 
8 the pre-trade -- within municipal transparency that 
9 will focus on pre-trade transparency. It came out 
10 of FIMSAC membership as a topic to delve into in the 
11 next year. So, that's going to be on the docket. 
12 So, with that having been said, I'll 
13 entertain a motion to adjourn. All in favor. Thank 
14 you very much. 
15 (Whereupon, at 4:10 p.m., the meeting was 
16 adjourned.) 
17 * * * * * 
18 
19 
20 
21 
22 
23 
24 
25 
Page 313 
1 PROOFREADER'S CERTIFICATE 
2 
3 In the Matter of: FIMSAC 
4 File Number: OS-1104 
5 Date: Monday, November 4, 2019 
6 Location: New York, New York 
7 
8 This is to certify that I, Christine Boyce 
9 (the undersigned), do hereby certify that the foregoing 
10 transcript is a complete, true and accurate 
11 transcription of all matters contained on the recorded 
12 proceedings of the meeting. 
13 
14 _______________________ _______________________ 
15 Proofreader's Name) (Date) 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 

79 (Pages 310 to 313) 


Page 314 
1 C E R T I F I C A T E 
2 
3 I, SHAUNNA H. MORAN, a Certified Shorthand 
4 Reporter and Registered Professional Reporter in the 
5 States of New Jersey, New York and The District of 
6 Columbia, and Notary Public of the State of New 
7 Jersey, do hereby certify that the foregoing is a 
8 true and accurate transcript of the testimony as 
9 taken stenographically by and before me at the time, 
10 place and on the date hereinbefore set forth. 
11 I DO FURTHER CERTIFY that I am neither a 
12 relative nor employee nor attorney nor counsel of 
13 any of the parties to this action, and that I am 
14 neither a relative nor employee of such attorney or 
15 counsel, and that I am not financially interested in 
16 the action. 
17 
18 SHAUNNA H. MORAN, CSR, RPR 
19 Shorthand Reporter 
20 
21 
22 
23 
24 
25 

80 (Page 314) 


Page 315