Full text of Financial Crisis Inquiry Commission : Documents Relating to the Financial Crisis of 2007-2009 : Transcript of Interview With Edward J. "Ned" Kelly III, March 3, 2010

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010

United States of America
Financial Crisis Inquiry Commission

INTERVIEW OF
EDWARD J. “NED” KELLY III

Wednesday, March 3, 2010
2:10 p.m.

*** Confidential ***

Financial Crisis Inquiry Commission
Wednesday, March 3, 2010
--o0o--

1

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MR. BONDI: Good afternoon, Mr. Kelly.
It is March 3rd, 2010, at 2:10 p.m.
As I asked you before we got started, do you consent to the tape-recording of this
interview?
MR. KELLY: I do.
MR. BONDI: Okay, Mr. Kelly, my name is Brad Bondi. I’m with the Financial Crisis
Inquiry Commission. I am the assistant director and deputy general counsel.
These are my colleagues who you just met: Jane Poulin, who is an accountant; Donna
Norman; and Karen Dubas, who is a research assistant and paralegal for our group.
As I mentioned, we’re with the Financial Crisis Inquiry Commission. We’re a
commission that was established by Congress as part of the Fraud Enforcement Recovery Act of
2009 to investigate the causes of the financial crisis, both domestically and globally, and to do a
report for the President and Congress and the American public.
The report is due December 15th, 2010. Part of our mandate includes looking at certain
practices, whether it be lending and accounting practices, securitization practices. I believe even
taxes is one of our 22 specific mandates in Congress.
We’re also obligated under statute to look at the financial institutions that collapsed or
would have collapsed but for substantial government assistance.
So we are here as part of our inquiry into the financial crisis.
This inquiry is confidential, so we ask that aside from your lawyers, you don’t talk to
anyone about what we’ve talked about today. But this information that we use today could be
used in a public forum or report to the American public, so this is -- but this is up to our
determination how to use this information.
I am obligated, because we are government agents, to tell you that 18 USC 1001 applies.
And that says, in a nutshell -- and your lawyers have already probably already told you about
that -- but providing willfully or knowingly providing false information to the government is a
crime.
But I have no reason to believe that you’re going to provide us any false information,
Mr. Kelly.
But you’ve been designated, sir, as someone who is knowledgeable about the requirement
for government assistance. In particular – in fact, I think I just read it exactly –- we asked for a
witness to be designated to talk about the requirement of government assistance in 2008. And
we understand that you’re knowledgeable about that topic.
Is that correct?
MR. KELLY: It is.
2

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MR. BONDI: Okay, and because we’re being recorded today, I’d just ask that you
answer questions “yes” or “no” as opposed to nodding your head.
MR. KELLY: Understood.
MR. BONDI: Yes, thank you.
Mr. Kelly, what can you tell me about the events that precipitated the need for
government assistance in 2008?
MR. KELLY: Well, as you know, there were three pieces to that. I’m familiar with the
last two as opposed to the first one.
The first one, as you know, was the TARP injection, which were to the nine largest firms
which we were part. To my recollection, that was in the form of $25 billion worth of
preferred stock. That was, I believe, in October of 2008.
Between October of 2008 and late November -- and I think it was somewhere in
November 23rd, I think, was the agreement with respect to the second injection, and that injection
took two forms. One was an additional $20 billion worth of preferred that we received from the
government; and the second, as you know, was what we described here as a “loss-sharing
agreement,” which I think is sometimes referred to as an Asset Guarantee Program.
I think the reasons for the first one have been well known. Others are more familiar with
it than I. I think the Treasury in particular, when the Administration at the time, along with the
regulators, were very concerned about the systemic issues that became vivid during the fall of
‘08, concluded that it made sense to implement TARP in the way that they did.
With respect to the second one, what I can tell you is what it is that we did and the course
of our conversations with the government.
What I’m less able to tell you is why, because I think the fact is that the reasons for it are,
in one respect, less transparent, or are probably going to be the subject of a lot of discussion and
speculation over the ensuing years. Having lived through it day by day, I have distorted visions
obscured by having been as close to it as I was, and also by not being able to tell you precisely
what the market dynamics were around the creation of the situation that occurred in late
November.
There are two things that strike me as very high level about it. One is that I think in the
wake of the failed Wachovia transaction, there was a great deal of focus on Citi for a couple of
reasons. One is, Citi historically, as you know, had not had the sort of presence as a national
matter that some of our perceived peers did, whether it’s Bank of America, Wells, or
J.P. Morgan. But for years, it had actually done quite well, essentially being more international
and more wholesale focused with a smaller presence in the United States. That’s not to suggest
that it wasn’t substantial, because it was, but it was smaller than our peers.
After we announced the Wachovia deal, ironically, I think there was a perception on the
part of the market that we had now conceded we needed a larger presence in the United States;
that there was a value in class of funding that we had been willing to forgo for some time. But

3

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
having agreed to do the deal was a recognition on our part that we needed it. And if we needed it
and didn’t get it, what did that imply for the strength of the firm going forward.
One thing that’s important to note about that -- and it’s frequently been, I think, either
misperceived or misdescribed -- is that we were approached in connection with that transaction.
We had a very specific idea about what it is that we would do. And we were relatively persistent
in our view that we would do it only under specific circumstances.
Ultimately, I think the regulators and the government -- because it was an open-bank
assistance transaction -- concluded that it was sensible to accept our proposal; and you would
say, we reached agreement, as you know, in the very early morning hours of Monday, whatever
day that was, and went forward from there.
Because the deal was very complicated, it was not an acquisition of the whole firm; it
was an acquisition only of the bank, not of the holding company; and involved insurance, as you
know, around 300 billion of assets of the bank. To be blunt about it, at least as a particularized
documented matter, we weren’t sure what we had bought.
We knew what we had bought conceptually. We thought that it was extremely important
for systemic reasons to protect the debt at the holding company. And what had worried us
generally is that if Wachovia were to fail and were to become -- go into a bank receivership, that
that debt ultimately would, to use a term of art, “splatter,” and have consequences that were, in
my view, dramatic but not necessarily perceivable. We thought that it was important to try to
protect it. So our deal was structured acquisition of the bank for a nominal consideration,
because our understanding was, as a policy matter in connection with open bank assistance, the
government and the regulators would insist that that compensation be nominal. An assumption
of the debt of the holding company.
We had left behind other assets of the holding company, including liability, certain
contingent liabilities. And we agreed, as you know, in a highly structured transaction to
insurance around the 300 billion of refinanced assets at the time for which we agreed to pay, I
think, ultimately $14 billion in preferred stock, and to assume a first loss of 42 billion -- I think
30 billion at the outset and then more than 4 billion over a three-year period, in the aggregate 42.
So when you counted what we were willing to pay for, we essentially provided $56 billion in
consideration against insurance of 300 billion of those assets.
A very complicated deal.
We had to deconstruct the firm and then reconstruct in the process of documenting it.
The only thing that we believed that we could do, was to have an exclusivity agreement with
them to ensure that we would be able to talk to them and have time to structure the agreement.
Very difficult to structure a breakup fee of any description because the truth is, our view was,
that to that extent that you did it, it would be easily invalidated because there was no describable
economic deal at that stage.
And, as you know, ultimately Wells reconsidered its position, and I think now
well-documented in light of the Treasury ruling with respect to taxes, and we concluded that it
made sense for them to buy the whole firm. But they were two very different deals.

4

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
Wells bought the whole company, no insurance. We bought just the bank, assumed the
debt, and had insurance. So there were apples and oranges.
Ultimately, Wells prevailed after three or four days of negotiation initiated by the
regulators between us and Wells as to whether we could reach some agreement with respect to
splitting the firm up.
Thereafter, it was relatively quiet. Although, as you know, developments in the market
were not entirely positive. Just generally. In other words, there was increasing unease, I think,
in the fixed income market, in particular, to some extent reflected in the equity markets. WaMu
had failed. As you know, Lehman had failed. AIG obviously was in distress and had been,
quote, unquote, “rescued” by the Fed.
And as we went into November, there seemed to be almost sequentially focus on various
firms and whether they were weak or whether they had come under a crisis of confidence of
some description in the market. Again, no idea because it’s obviously a huge number of factors
that influence that.
Our stock came under pressure I think that week of November 17th, or whatever it would
have been, which I think is well-known. And you should have probably seen there were a
number of us that bought stock in the firm for our own account the week before. I bought stock
the previous Friday at $10. And by the following Wednesday, the stock was at $3.50.
Part of that, in my own view, was triggered by press speculation that there was some lack
of consensus on the Board and some potential lack of confidence in management, and Vikram
Pandit in particular, that seemed to increase the market’s unease, generating more uncertainty
around the firm; and the stock began to fall.
The feedback would have been the market between the stock price and the comfort of the
liabilities on the debt side had become much closer. So there was concern that to the extent that
the stock continued to fall, some of our liabilities might become insecure; therefore, we’d face
funding issues at the bank.
I was not involved in any direct contacts with the regulators, in other words, any
discussions of what it was that was worrying them. I was vaguely aware that there must be
conversations going on. So the first time I got involved was, I think, that Friday, November 21st,
when I was called and asked to come join a conference call with the New York Fed and Lew
Kaden, who was another vice chairman here, to talk about not why we needed assistance or
whether we needed assistance but how it is that we might structure assistance. In other words,
the presumption was that we were going to get it, the issue was how and in what form. They
asked us to think about it, which we did.
So that’s a very winded way of answering your question, with the short answer being that
I can tell you factually what happened. But, frankly, my speculation is probably no better than
anyone else’s with respect to why.
MR. BONDI: Let me ask a few questions within that. First of all, you said “we were
approached.” And just to be clear, who approached you?

5

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MR. KELLY: Wachovia.
MR. BONDI: Wachovia.
MR. KELLY: Yeah.
MR. BONDI: And who, in particular? Do you know?
MR. KELLY: Bob Steele called Vikram.
MR. BONDI: Okay. And a little housekeeping matter which I should have started at the
beginning.
I understand that I have your background. And rather than go through all the background
and answers, in the interest of time, but if you could briefly just tell us your background in terms
of Citi and when you started in your current role and what your current role is?
MR. KELLY: Yes, I started at -- well, let me go back one step. I can do it briefly.
MR. BONDI: Sure.
MR. KELLY: As you know, I started life as a lawyer at Davis Polk, was a partner there.
Joined the old J.P. Morgan as general counsel in 1994. Was general counsel for a year or
so. Moved to the business side.
Became an investment banker. Ran a financial institutions group around the world, ran
Latin America. We eventually merged with Chase in 2000.
I decided to leave. Went to run a bank in the mid-Atlantic called Mercantile Bankshares.
I was chairman and CEO there from 2003 to 2007, and was CEO from 2001 to 2003.
A reason I go back is because in late 2006 it became apparent to me that credit markets
had gotten a little bit frothy. Mercantile was a very conservative, well-capitalized, well-reserved
bank that had done a very good business in the mid-Atlantic for many years and middle-market
commercial.
We found ourselves in a situation where it was very difficult to compete as a credit
matter, so that we were now losing business or not being able to do business that we had
historically done because price and terms had become so relaxed, that it was difficult for me to
justify it on a risk basis.
We concluded in that light, given the growth challenges we had faced and given how we
thought about the market, the Board and I concluded at that stage that it made sense to sell the
bank. We did. We negotiated a deal with PNC. We reached that agreement in late 2006.
Closed in early 2007.
I was briefly a vice chairman at PNC, but decided that it made sense for me to go do
something else.

6

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
So I was at Carlyle for six or seven months. I went to Carlyle, organized a financial
institutions group to focus on private equity with respect to financial firms. For a whole range of
reasons, it was more difficult than we might have guessed at the outset to raise money for that
fund.
I had known Vikram for years. Not well. We were not personally close friends but he
had been an old client of mine when I had been at Davis Polk and he had been at
Morgan-Stanley. We had known each other -- of each other over the years. And John Havens,
as you know, who is another senior person here, we have a very close mutual friend. And I
talked to Vikram and John when they were still with Old Lane about the possibility of joining
them.
They sold to Citi; I went to Carlyle. Six months later, they called and said, “Vikram has
now taken over as CEO. We clearly have some challenges” -- this was in late 2007, early 2008 - “Would you like to come join us?”
I thought of my purpose at Carlyle, in one respect, was to play distress on market. Why
not come to a large distressed opportunity, not having any idea how distressed it would become.
And joined Citi in February of ‘08.
Initially, as president of Citi Alternatives, then as CEO of Citi Alternatives. Then in
September, I became -- retained those titles and became head of Global Banking.
On March of ‘09 I became the CFO. In July of ‘09, I became the vice-chairman.
So during the period we’re talking about, I was actually head of Global Banking, which
at that point included Citi Alternatives. Global Banking, for that purpose, was both the
investment bank and the corporate bank. But the role that I was playing in terms of the fall of
‘08 and the government assistance was as banker to the firm, as an advisor to Vikram, and as the
person who came in and helped think about and structured deals.
MR. BONDI: I appreciate the background. And I apologize for not doing that upfront.
We should have done that first.
MR. KELLY: Not at all.
MR. BONDI: But getting back to the need for government assistance. You’ve
mentioned the Wachovia transaction that failed, failed to go through, failed to consummate.
What other events precipitated the need for government assistance? What other events at
Citi?
MR. KELLY: It’s hard for me to remember. I mean, I think I’ve highlighted those that I
do remember.
I do know that the market was -- it’s important to remember how the whole thing
unfolded, in one respect.
I think the fourth quarter of ‘07, obviously, revealed cracks in the balance sheet of a
number of firms: Citi, Merrill, to my recollection, Morgan-Stanley, too, as I recall.
7

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
The first quarter of ‘08 was pretty horrible. Fixed-income markets froze up. Liquidity
was at a premium. Even the way the industry was funded at the time which, as you know, was
funded short, invested long. To the extent that short liquidity dries up, that generated lots of
asset sales which in turn depressed asset prices, which in turn created a vicious spiral down.
Interestingly, in the second quarter of ‘08 things got a little bit better. You know, I’m
doing this more or less from memory, but my sense is that they got better. A little quieter after
Bear when the government obviously did what it did and JPM bought Bear. I think there was
some sense of relief and relaxation in the market in the second quarter, and, as a result, was a
little better.
The early part of the third quarter, during the summer, which is normally slow, again,
relatively quiet.
And then in September, as you know, we had a sequence of events. Fannie and Freddie.
We had Lehman, we had AIG, we had WaMu. So September, in many respects, was a nightmare
from a market standpoint.
And then there was further disruption, as I’ve described, in terms of market uncertainty in
the Wachovia deal and how that unfolded.
And then I think during October, I don’t recall -- and then that was settled, to some
extent, by the TARP injections. Because, again, you’ll recall Congress first rejected TARP.
Four or five days later they ultimately adopted it. Market uncertainty generated by that. Then
they make the injections which, frankly, nobody had expected because TARP, as you know, was
adopted as an asset-purchase mechanism. Ultimately, it became an equity-injection mechanism,
perfectly consistent with the statute but a little different from what the market expected. But that
settled things out at the end, because the government at that stage was perceived to recognize the
systemic risk, and was willing to support systemically important firms.
And then in October, things were unsettled, but my recollection is not horrible. And
then things began to accelerate again in November, with increasing focus on particular firms.
And the worry in the market about which one was next going to become a target, in part, because
of shorting. And, I telescoped a lot, as you know, there was also an episode around the SEC
banning short-selling for a while, and then that ban was lifted. And there were various efforts
made to stabilize the markets.
Morgan-Stanley and Goldman, as you know, became bank holding companies in the
midst of all of this. There were a whole variety of steps that were taken.
But my point being that it was very choppy. You know, people were very concerned,
high levels of anxiety. But very difficult for me, as I look back on it, to point to anything
specific. And the only thing specific I can point to in the case of Citi really revolved around
publicity and speculation about the stability at Citi, frankly, at that stage, really just on the
management front, had clearly concerns about -- as there was with everyone at that stage -- about
a balance sheet and funding and earnings going forward.
But I don’t remember any specific trigger for it. As a matter of fact, I expressly
remember being bemused by the whole thing. As I said, I bought stock at $10 on Friday. It was
8

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
$3.50 by the next Wednesday, and nothing had changed. And I couldn’t possibly tell you why
that was other than the fact that the market at that stage had adopted, clearly, a more negative
view of the firm and its prospects.
MR. BONDI: Let me ask you, you mentioned short-selling. And there have been some
reports about the effects of short-selling on financial institutions’ stock. And you’ve described
how the stock price dropped in this time period in November.
How much do you attribute the effect of a stock price to short-selling? If you could give
some more details in terms of the observations from Citi’s standpoint concerning any
short-selling on its stock.
MR. KELLY: I wasn’t close to that. I mean, obviously, I know that it was going on, but
I was never focused particularly on the shorts, if you will, as a particularized evil in connection
with Citi. I never really developed my own view about whether -- what the right solution was.
The interaction between short-selling and the CDS market, as you know, was much discussed,
and I know has been much vetted by many. Because the fact is that if you were to -- depending
on what your position was, if you were to buy or sell insurance and then ultimately short the
stock, you found yourself in a position where instability in the firm was a good thing against
which you were hedged. And I think there was a lot of discussion at that time about the
interaction, as I mentioned earlier, between equity market prices and the debt holders’ comfort
with the firm. And that became a very vicious feedback loop.
And I think we were seeing that because our CDS spreads, as you know, were blowing
out at the same time that the equity price was falling. But it was nothing more than my own
impressions, obviously, being a market observer; but no specific knowledge with respect to what
was going on at Citi.
MR. BONDI: You referenced the challenges to the balance sheet. And, obviously, Citi
had its own challenges to the balance sheet, as you’ve referenced, in 2007 -- starting in 2007.
And I understand you’ve mentioned somewhat of the last straw in the concern with respect to the
Wachovia deal.
But how much did the challenges to the balance sheet of Citi really precipitate the need
for financial assistance?
MR. KELLY: Well, that’s a very complicated question. You know, when I think at one
end, if we’d had no losses and the perception had been that our exposure to the market was
limited, we would not have had an issue.
Having said that, I think everybody was perceived to have exposure to the market, and it
had gotten considerably worse than people might have guessed. And, therefore, there was
enhanced focus. And the reason it’s particularly complicated, there was an enhanced focused not
only on capital levels, but the kind of capital. So for Citi in particular, what drove it was that our
“tier 1” ratio, which, as you know, has historically been the ratio that bank regulators have
looked at for purposes of assessing capital adequacy, was actually very strong through the whole
process.
What was curious about it was that our tangible common equity actually got to be quite
9

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
low. And the market was very concerned about what they described as TCE, which as Jane
knows is not even actually a sort of defined term; it’s a market impression. But effectively, it’s
just common equity. They got to be worried about that and began to disregard tier 1 capital.
So notwithstanding the strength of our tier 1 capital ratios, mark was conferred on
tangible common, because that’s actually what’s available to common stockholders. So clearly,
the losses, or at least the perception of possible losses had an impact on the common
stockholders and, therefore, clearly one assumes had an impact on the stock price which for the
reasons that I described, in turn, potentially had an impact on the liability holders. And at the
end of the day, banks are hugely dependent –- both banks and other financial firms are usually
dependent on liquidity in their assets in their fixed-income markets.
MR. BONDI: What do you perceive was the reason to look more at tangible common
equity from an investor standpoint than the capital levels?
MR. KELLY: I think in a very simplistic basis, as I said, that’s actually what’s available
to common stockholders. So they were thinking, if our tangible common equity, notwithstanding
our tier 1 capital, goes negative, there’s nothing left for them. Given the anxiety, there wasn’t the
patience necessary to work through that to restore, if you will, tangible common equity to a
positive value.
I can’t tell you that it was entirely -- I understand it. I’m not sure it was entirely rational,
but I understand it. And as you look back, you’ll find that the regulators for a very long time
continued to hew to the line of tier 1 as being the most important. As you’ve probably seen, and
I’m going to restate it to make a point, there seems to have at least been some implicit
capitulation recently as they focus on tier 1 common as opposed to pure tier 1.
MR. BONDI: Commentators have described what happened with respect to firms as a
loss of confidence. And I’d be interested to get your take on if you perceive there to be loss of
confidence in Citi and the reasons why you believe there might have been a loss of confidence.
Was it the Wachovia transaction? Was it some other things? What did -- was there a loss of
confidence from an investor standpoint?
MR. KELLY: Having lived through that and been an observer of the market in various
guises for many years, I think there was a dramatic loss of confidence across the board and
around the world. I don’t think it was Citi-specific, by any means.
I remember describing it to somebody. You know, we obviously all believed that the
dollar I had in my pocket is worth a dollar and is negotiable. It was as if that willing suspension
of disbelief -- because there’s no particular reason you should believe that -- in fact, evaporated.
People were questioning what everything was worth at the time. As you know, there was a flight
not just to quality and safety, but almost a flight to certainty. You could see that in terms of
where yields were going with respect to Treasury securities and where yields were going with
respect other assets.
There was, I think, in one respect, a classic panic. I think people will be thinking for
years about why that happened. I’m not sure myself. I think part of it was that after an extended
period where liquidity and capital were basically free or cheaper than they had been historically,
the pendulum, not surprisingly, swung even more severely to the other side because there had
10

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
been such a dramatic change in circumstances.
I think Citi in particular, as you know, is global, it was large. It was in both the consumer
and the wholesale businesses.
If you look at it -- and this has been confirmed by the stress tests that the Fed and the
other regulators went through in the fall of ’09 -- we look very much like other banks. Having
said that, if you separate the credit crisis into three pieces -- the markets crisis, the consumer
cycle, and I think what is potentially coming, commercial real estate and commercial industrial
lending -- we had more than our share on the first side. And remembering at least anecdotally by
own impressions, we had a lot of exposures, you know, to superseding securities that were
deemed to be extremely safe. I think in most people’s view, even as things unfolded, were
regarded as being more or less bulletproof.
It turned out for part of the reason they described in terms of liquidity and the need to sell
assets in order to meet those liquidity demands, the marks on the securities were pretty severe. I
think we were perceived at that time, perhaps rightfully, had more exposure on that front than
some of our peers, which arguably made us more vulnerable. But, again, that’s something -remembering my role as head of Global Banking and then ultimately coming in to try to help in
terms of structuring of things -- that I was not ultimately involved in either, while it was evolving
or even after the fact, frankly. But, again, just an impression. Some perceived vulnerability on
that front.
MR. BONDI: How much did valuation play, valuing assets at Citi play in this perceived
lack of confidence by investors?
MR. KELLY: I have no view on that.
I don’t think it was a valuation issue. I think it was a perception of exposures to various
asset classes that were under pressure, and an inability to assess where the bottom might be.
MR. BONDI: Uh-huh.
Citi, through the course of 2000 -- late 2007 and through 2008, had several write-downs.
Can you tell us about the effect of those write-downs and announced write-downs in terms of
adding to the otherwise challenges that Citi faced in the marketplace?
MR. KELLY: Well, I’d be the first to admit my recollection of that is not perfect, by any
means. And, again, going back to it, I wasn’t actually involved with that at the time. I was off
running businesses.
But my recollection -- somebody correct me if I’m wrong, if you’ll allow it -– is that the
biggest write-down we suffered, I think, was in the fourth quarter of ‘07. Actually, I guess it was
toward -- no -- yes, that’s right, the fourth quarter of ‘07. We had a loss in the first quarter of
‘08, partially driven by write-downs but less dramatic than the one in the fourth quarter of ‘07.
I think we actually had a relatively minor loss in the second quarter of ‘08. And in the
grand scheme of things, I think we lost 2.8 billion or something in the third quarter of ‘08. So I
would argue that write-downs, you know, against the backdrop obviously of the scope and skill

11

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
of the firm, didn’t have a lot to do with it.
The really big loss, as you know, actually occurred in the fourth quarter of ‘08, you
know, relatively the first three quarters of ‘08.
So I’m not sure that the write-down in the fourth quarter of ‘07 obviously set the stage for
the perception that Citi had issues. I’m not sure that anything that happened during the first three
quarters was wildly out of line with what was going on in the rest of the industry. So I’m not
sure that there would have been anything there that would have distinguished Citi from anybody
else.
MR. BONDI: With respect to Citi’s subprime exposure in November of 2008, did that
exposure lead to or add to the need for government assistance?
MR. KELLY: Well, the short answer to that is, Brad, I don’t know, because, again,
going back to it and just to stress it, my role in most of these things is, “We have a problem.
Would you please come help us solve it?”
MR. BONDI: Uh-huh.
MR. KELLY: I’m not intimately familiar with the constituent parts of that problem. In
other words, I know it exists. Please help us get out of it. So for all the reasons I described
earlier, it’s hard for me to assess the role that our subprime exposure played quantifying it in
terms of our need for assistance.
As I said earlier, I think our exposures clearly generated a perception that we were
potentially vulnerable. And that, in connection with the market environment that’s characterized
by all the features that we discussed, I think resulted in our needing assistance.
But for me to parse each part of that, in part, because of the role I played, in part because
I’m not sure it’s easy at all, is very difficult.
MR. BONDI: The Wall Street Journal reported, somewhere around November 19th,
2008, that the company, Citi’s announcements on November 17th and 18th stoked investors’ fears
that Citigroup could be swamped by toxic assets flooding back onto its books.
MR. KELLY: Is that November 17th and 18th, 2008?
MR. BONDI: Yes, sir, 2008.
And I’m curious to know if there was this concern with, as The Wall Street Journal
describes, toxic assets flooding back onto Citi’s books in November of 2008?
MR. KELLY: Not that I recall. I recall no discussion of it.
I know that there were a lot of discussion -- misguided in many respects -- about our
off-balance sheet exposures.
MR. BONDI: Uh-huh.

12

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MR. KELLY: The truth is, those off-balance sheet exposures, by and large, I think are
credit-card receivables and certain other securitizations. But I wouldn’t necessarily describe
those as “toxic.”
And, in fact, “toxic” was a term that was thrown around a lot, as you know, during that
period.
When you look at things even like the special asset core, which we now have as part of
holdings, I’m not sure that I would even describe large parts of that as “toxic.”
I think the ring fence that we ultimately agreed to with the government, the loss-sharing
agreement, actually involved assets that you would regard as pretty pedestrian. You know, it
was predominantly mortgages, home mortgages. Some corporate ones. They were basically all
accrual assets. Very few mark-to-market assets. But I remember that vaguely, but I don’t
remember thinking that was an issue particularly.
MR. BONDI: Now, we’ve heard and spoken with Mr. Arnold, Bill Arnold -MR. KELLY: Yeah.
MR. BONDI: -- concerning the special investment vehicles -- the seven special
investment vehicles out of London.
We have a little bit of an understanding about what happened with -MS. BUERGEL: Structured. “Structured investment vehicles,” just to be clear.
MR. BONDI: Excuse me, structured-investment vehicles, sorry.
MS. BUERGEL: As opposed to special-purpose vehicles.
MR. BONDI: Sorry.

MR. KELLY: SIVs.
MR. BONDI: SIVs, structured investment vehicles. I apologize.
Ms. Buergel is keeping me honest with my acronyms, and I’m a slow learner when it
comes to acronyms. But the SIVS. The seven SIVS out of London.
We’ve spoken to Mr. Arnold about those SIVs. But the timing of some
announcements concerning those SIVs and the government, ultimately the announcement of the
government assistance, is curious to me. And by that, I mean, the following is, there was an
announcement on November 19th, 2008, that’s of a 1.1 billion write-down on the SIVs. And I’m
going off of my own notes. So to the extent that these numbers are incorrect, please correct me.
But I’m not meaning for you to attest to numbers that you don’t know, so let me just say that
right at the offset.

13

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MR. KELLY: Sure.
MR. BONDI: But according to my notes, Citi announced a $1.1 billion write-off on the
SIVs on around November 19th, 2008. And Citi also announced that it would pay somewhere
around 17 billion to unwind, as announced here, unwind and compensate the holders of the SIVs
commercial paper. Then the government assistance package was announced not too long after,
on around November 23rd, 2008.
I’m just curious, sir, if there was a connection there between what happened with respect
to the SIVs and the need for government assistance?
MR. KELLY: I don’t think so. You know, again, I can’t say conclusively. But my own
sense of it was, that that was, by and large, at that stage, old news. And my recollection is that
the SIVs were consolidated at the end of 2007, I think, based on the initial three and a half billion
that we had injected at the mezzanine level. It had been roughly a year before, when I was at Citi
Alternatives, I was actually involved in the wind-down of the SIVs.
But SIVs, as you know, were basically banks, they were non-deposit funded, that were
short-funded. So they were funded very short in the market and invested very long.
The issue we had in the SIVs was not credit quality. Credit quality was actually almost
pristine, remarkably good.
When I was there, my recollection is, the only default in the entire portfolio was WaMu.
Otherwise, the credits were all very solid. The marks were driven by the market dynamic I
described, where short-funding dries up, you’ve got to meet demands; therefore, you sell assets,
selling assets into a falling market, you begin to lose sight of fundamental values and ultimately
take big haircuts.
I can remember looking at that, and I used to cite it frequently, where there were assets
that I was absolutely confident in my bones were “money good,” and I’m sure had proven to be
money good. They were trading at 65 or 70 in the market, principally because there was a
liquidity issue. You know, there was another classic example of that, which was Carlyle Capital
in early ‘08, as you know, was basically invested in implicitly government-guaranteed securities
because it was funded short because the haircuts changed.
The whole thing ultimately imploded because the fact is that they had to meet
short-funding demands, had to sell assets, and they ultimately had to wind it up.
The reason that we made the decision that we did, as I recall in November of ‘08, was
that we had injected the three and a half billion.
I think at one point, we may have injected -- and I can’t remember, but I think at one
point we may have injected another billion. In other words, in order to support them. And at
some point, we concluded, “This is crazy,” right? You know, the fact is, we just ought to take
these back onto our balance sheet.”
To your point, it was only 17 billion at the time. There were marks associated with them
because of that market dynamic that I described. But in the grand scheme of things, I remember

14

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
thinking at the time it was unexceptional but a perfectly rational decision; they were already
effectively on the balance sheet, anyway.
MS. BUERGEL: Could I ask a clarifying question?
MR. BONDI: Please.
MS. BUERGEL: On the SIVs, I think you referred to them as they were basically banks
that were short-funded.
MR. KELLY: Uh-huh.
MS. BUERGEL: I don’t recall whether I -- I understood what the assets were. And I’m
struck by your reference to them being banks. So maybe that’s something -MR. KELLY: I was saying analytically. Analytically.
MS. BUERGEL: Oh, okay.
MR. KELLY: I mean, banks basically incur liabilities and aggregate assets and manage
them into merging. They just incurred non-deposit liabilities that were short in the wholesale
markets and invested those in long assets.
MS. BUERGEL: Then the assets were long?
MR. KELLY: Precisely. And there’s just an asset liability mismatch.
MS. BUERGEL: Okay. Was there any similarity to the types of assets that they held?
MR. KELLY: No. They’re pretty broadly based across the board. Lots of financial
assets, some mortgage exposures. You know, I can’t remember. But pretty broad-based. But by
and large, as I said, the unifying principle around them: good credits.
MS. BUERGEL: The Qs and Ks have a breakdown of the underlying asset pool, it’s
about 60 percent financial institution debt, 20 percent European residential housing. Very, very
little subprime U.S. housing, almost de minimis.
MR. KELLY: A lot of financial institution paper.
So unless you believed the world was coming to an end -MS. BUERGEL: Which is what was causing the stress.
MR. KELLY: -- it was going to be money-good.
But I apologize for describing it. I was thinking of it -MS. BUERGEL: No, you were using the term generically, I understand.
MR. KELLY: Yes.

15

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MR. BONDI: What -- with respect to the SIVs, we were talking about, really, the
payment on November 19, 2008. But taking us back to 2007, what was your role, Mr. Kelly,
with respect to the SIVs and this issue?
MR. KELLY: I wasn’t here.
MR. BONDI: Okay.
MR. KELLY: I didn’t show up until February of ‘08. You know, that was done when I
got here.
MR. BONDI: I thought so. I wanted to make sure that there wasn’t a role there that I
didn’t -MR. KELLY: No.
MR. BONDI: -- had not been mentioned.
You’ve mentioned and described, in effect, that we’ve seen across several institutions,
and that is the illiquid market affecting the valuation of assets.
MR. KELLY: Yep.
MR. BONDI: And I was just wondering if you could provide some further detail in
terms of the challenges specifically that Citigroup has faced with respect to marking assets and
the results in terms of investor perception and the like?
MR. KELLY: Well, certainly in the run-up to all of this, but I wasn’t involved in it at all.
So I have no view. No clue.
As a more general market matter, as you know, I think it’s fair to say that there was
skepticism in the market generally about the marks, not just that Citi was taking, but that all
firms were taking. In other words, given the illiquidity in the markets and given the lack of price
transparency around certain assets, there were questions that emerged about, are they marked
properly, which goes back to the point I tried to make earlier about an inability to perceive the
bottom. In other words, where are these things going. But I don’t remember and was not
involved in anything that was Citi-specific.
MR. BONDI: And part of our mandate is to look at mark-to-market accounting.
MR. KELLY: Yes.
MR. BONDI: I was just wondering what your view was in terms of mark-to-market
accounting and how it functioned during the course of the crisis and the aftermath?
MR. KELLY: Well, I have a variety of views on that, but that’s -- I think mark-to-market
accounting is fine, all right. And I don’t have any -- I mean, obviously, given the impact of it, it
clearly has the effect of exaggerating market movements in the context of equity constraints.
And it becomes difficult to separate, as I mentioned earlier, the fundamental value of a credit
from its value in a context where it is very difficult to finance.
16

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
So you essentially take an asset that people have assumed had value because they were
able to finance it at certain levels. You remove that financing, it becomes a question of what it is
that somebody will pay for it. And at that stage, clearly not surprisingly, they’re willing to pay
less because the returns implicit in it are lower by virtue of the fact that they can’t borrow against
it.
Having said that, I have no particular issue with the fact that it existed. But one broader
issue that I’ve had with mark-to-market accounting is, as you know, it deals with the asset side of
the balance sheet but not the liability side. You know, and the fact is that the liability sometimes
have values that offset those potential marks on the asset side.
But I think it’s fair to say for reasons that I perfectly understand that accounting has not
evolved to the point where they are marking both sides of the balance sheet. In certain isolated
cases they are but not generally.
And then, as you know, on the banking side -- and I mean by that, the commercial
banking side, specifically with respect to accrual assets -- there is no mark-to-market.
And there’s a reason for that. Because if there were mark-to-market and there were no
liability marks, I think virtually every bank in the company would be insolvent.
I can tell you, when I went to run a mid-cap bank in the mid-Atlantic after having been
on Wall Street for a while, Mercantile, as I said, one of the best capitalized, most highly reserved
banks in the country with probably some of the best credit experience, if I tried to sell our loan
portfolio, markdown, right? And that has to do with the information on efficiencies in the
middle market in particular. Because the fact is that we knew those were money-good. We
knew the borrowers, we knew the credits, we had security in the form of their first-born, their
wife, their house, and everything else they own. We knew we would be repaid. But if you were
to try to market that, we wouldn’t get par.
MR. BONDI: What’s Citi’s exposure to Lehman, both directly or indirectly?
MR. KELLY: No idea, unfortunately.
That was a little bit prior to my time in terms of actually getting involved.
I know we had exposure and I think we managed through it.
MR. BONDI: And my colleague, Jane, is well-versed in accounting. I am not, so I’m
going to shortly turn things over to Jane to ask.
MR. KELLY: I’m sure who will test me as well.
MR. BONDI: I confess, my colleagues know the story, but I was an accounting major
for a short period of time, and then I realized, I said, “I don’t like accounting.” And I went into
law because accounting wasn’t my forté, so…
MR. KELLY: I’ll do my best.
MR. BONDI: Some may argue that law is not my forté, too. But I’ll reserve judgment
17

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
on that.
But let me ask a few sort of final questions on this.
There have been a lot of experts and a lot of academics talk about the interconnectiveness
of the markets. There are some who say that the markets aren’t really interconnected; there are
others who say they’re very interconnected; and as a result, this is why we had our financial
crisis.
And I was interested to hear your views on whether the interconnectiveness of the
marketplace caused or contributed to challenges at Citigroup? And if you could explain what
those were?
MR. KELLY: Well, there are two, I think, sort of very broad questions embedded in that.
Let me just take the last one first. I mean, I think there was a perception that Citi was, of
course, not surprisingly an extremely important financial services firm, not only in the U.S. but
globally. That the impact of its being, for lack of a better description, disabled, would likely be
fairly severe, both as a confidence matter, both in terms of the market thinking, “My goodness,
Citi is vulnerable and could fail,” and in terms of the actual impact on other banks in the world.
In the U.S. economy, because you’re obviously removing a large provider of liquidity and credit
to the economy.
So I think its interconnectedness was part of what -- and I’m assuming this, I don’t know
for sure but I think it’s a fair assumption -- was part of what moved people to think it’s necessary
to provide Citi with assistance as described.
I think the other element of interconnectiveness, of course, we touched on, which was
this interaction which has become much more acute than it had ever been before in my
experience between certain derivatives markets and the cash markets in terms of the CDS
spreads and the impact that that had on debt holders and the stock price, and the fact that they
seem to move, as you know, in correlated but completely opposite directions which, of course,
had a particularly damaging effect not just on us but on others.
And then to take your very broad question away from Citi, I think markets are
interconnected. And I think if you were to ask a number of us, we would tell you that the
biggest -– and the biggest and most troubling interconnection, which I know policymakers are
thinking about, is how to deal with the derivatives markets. Because you know the
over-the-counter derivatives markets are not hugely transparent. There are a number of
interconnections that are not as visible to people as I think one would hope.
My sense is that there are efforts underway essentially to try to force them onto
exchanges or into clearinghouses or to standardize them on a basis which makes it easier for
people to see what the exposures are and to see where they sit, and accordingly, possibly make it
easier to resolve firms if they do get into trouble. Because the biggest problem you have, I think,
in terms of resolutions is sorting through, you know, the various counterparty exposures. And as
you know, for a whole host of reasons, I’m not sure that the current structure that we have is
well-suited to that.

18

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
So I subscribe to the notion that the markets are interconnected.
I think as it’s currently structured, that interconnectiveness does pose obstacles to orderly
resolutions of firms that are systemically important. But my own sense is that people are
beginning to think through how they might reduce the size of that obstacle at least in terms of
addressing the lack of transparency and lack of standardization in the derivatives markets in
particular.
MR. BONDI: Did the interconnectiveness of the marketplace contribute to the financial
challenges of Citi specifically?
MR. KELLY: Other than -- as I said, other than in this interaction between the equity
and the CDS markets, in particular, not that I can think of. Not any differently than it would
have anyone else.
As I said, my own view, for what it’s worth -- and I don’t know for sure -- but I’m
inferring that it was part of what informed the judgment on the part of the regulators and the
government authorities in providing assistance that Citi made sense because of that
interconnectiveness.
I’m not sure that the interconnectiveness itself had anything to do with Citi’s distress.
MR. BONDI: What was the CDS exposure to Citi leading up to the government
assistance? And did that cause or contribute to the need for government assistance?
MR. KELLY: Well, I don’t think it was our exposure. I think it was the fact that CDS
spreads had moved where they were. I can’t remember, but I think they peaked at six or seven
hundred over.
You know, at that stage -- or at six or seven hundred. At that stage, the impact is on other
market participants, because they look at those spreads and say, “Is this some place I really want
to put my money?” And that’s not just in terms of wholesale funding, that’s people who also
have deposits with us at various points. Some governmental institutions or other
quasi-governmental entities who are obviously subject to some -- well, they’re all subject to
fiduciary duty, which may argue with special obligations on the part of governmental entities.
To the extent that they see a firm that’s got CDS spreads at 600 versus those who have them at
200, it’s a relatively easy choice. And to the extent that that funding gets displaced, that
aggravates liquidity issues that may persist.
So the impact for us was not in terms of our participation in the market or in terms of our
connection to it; it had more to do with the market perception around where our CDS spreads
were at that stage.
MR. BONDI: Understood.
It’s obviously difficult to ask the “what if” type questions, the hypothetical-world
questions. Those would, as you know, from your days at Davis Polk, those would usually cause
an objection of “calls for speculation” and the like from counsel.
Since we’re not in any litigation posture and we are fact-finding in trying to test
19

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
hypotheses, in your view, what would have happened to Citi had they not received government
assistance, either in this first wave or in the subsequent two waves?
MR. KELLY: I have not the foggiest idea, right, in the sense that there are so many
co-dependent variables. It’s very difficult to say.
I think with the benefit of hindsight, looking at how it did evolve, you know, it would be
silly to say that if we hadn’t gotten the first wave, we would have been able to make it. Having
said that, it’s not clear to me what alternatives we could have pursued either as a policy matter or
otherwise, or how the market might have evolved in terms of what our fate might have been. So
I just don’t know.
And again, not being evasive, it’s just I have no clue, really. It’s a very difficult question.
MR. BONDI: And, Mr. Kelly, we had a hearing in January with the CEOs of
J.P. Morgan.
MR. KELLY: I watched.
MR. BONDI: You watched?
They were posed the question, which is a complicated question, but I will pose it to you
as well.
There’s a concept of being “too big to fail.”
MR. KELLY: Yes.
MR. BONDI: Was Citi too big to fail?
MR. KELLY: Well, I think the judgment that the government made was that there
were -- whether you use “too big to fail” as a moniker or not, I think there were some firms that
the government believed too important to fail.
I actually subscribe to that view. I think part of the reason that they came to that
conclusion is because they had limited tools at their disposal to deal with it otherwise. You had a
Fed, which I think it’s fair to say had 19th century tools in the context of a 21st century crisis,
limited ability to deal with it. You’ve got that combined with resolution authorities that, as you
know, are inadequate, and a bankruptcy code which is not necessarily friendly to an orderly
resolution.
So if you were to ask me for my personal view, not a Citi view -MR. BONDI: Sure.
MR. KELLY: -- I do think that “too big to fail” as a policy is not a good one, but I think
it is equally bad to allow firms that are big and interconnected and important to fail in a
disorderly fashion.
And I think people -- what people are working towards is trying to come up with a
20

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
resolution mechanism so that firms, even if they are important systemically, can, in fact, be
resolved in an orderly way.
MR. BONDI: There was some suggestion that having the moniker “too big to fail,”
either explicitly or implicitly, creates an advantage in the marketplace. There was a perception
to some extent in the marketplace in 2008 that some firms may have been too big to fail.
MR. KELLY: It didn’t help.
MR. BONDI: I was going to ask you about that.
What was the -- what was, in your view -- and I appreciate some of these views are your
own views.
MR. KELLY: Absolutely.
MR. BONDI: But what was your perception of the effect of having that view of firms
having the view in the marketplace that they were too big to fail -- or excuse me, the perception
by others in the marketplace that certain firms, financial services firms, were too big to fail?
MR. KELLY: Well, what was interesting about that, Brad, is -- a couple different things,
right. “Too big to fail” has been around for a very, very long time. And I think the Fed’s policy,
for at least as long as I’ve been a professional, is one of constructive ambiguity. “We’re not
going to tell you whether you’re too big to fail or not, but we’re going to let the market guess,
and if it makes sense, we’ll open bank assistance and take care of it, but it’s perfectly possible we
may not.”
My recollection is, the only time that they did so expressly -- although I can also tell you
as a market participant in my own sense is they’ve done it quietly from time to time -- was in
Continental Illinois -- where they concluded that for whatever reasons -- and, as you know, the
scale of Continental Illinois relative to where we are today is just remarkably different. But they
did intervene and they did resolve Continental Illinois, in a way that was not a classic resolution
but, in fact, was open bank assistance. And they resolved it in an orderly way because their
concern was that it would have a systemic impact if they didn’t.
We then went through a period where basically the failures, if you will, were of relatively
smaller firms. We went through the thrift crisis. No reason really to call it into question,
although there were some banks that were troubled in the early nineties that were large that were
perceived to be important, but I think the Fed, through other mechanisms, whether it was rates or
other provisional liquidity to the market or just a fortuitous turn in the economy, were able to
work our way through that.
We’ve had shocks over the years. ‘98, long-term capital management, the fact is that the
market shut pretty tightly for a couple weeks around that failure, but reopened quickly enough so
that nobody was really threatened.
I think there has been discussion about various firms being in trouble as a liquidity matter
during the nineties but the regulators being able to handle it quietly.
So the first time ultimately that we’ve gotten into a real test was in 2008.
21

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
I think part of what happened was that after Bear, there was some confidence in the
market that, in fact, the government would, from the market standpoint, intervene and ultimately
ensure that there was some kind of orderly resolution. Then I think as things became more
severe, Fannie and Freddie, the decision to let Lehman go, that certainty subsided. In other
words, people said, “Hmm, my goodness. Not entirely clear they are going to do that.” So there
was a guessing game about who was and who wasn’t. But, you know, as you know, the way
markets work, if there’s a live possibility that you are not or that nobody is, they’re clearly going
to test that. And you’ve found circumstances, as I’m sure you’ve seen, where they were sort of
testing it firm by firm, to see whether, in fact, they were going to be taken care of or not.
Merrill, as you know, was acquired by BofA, which made that question moot. And
Goldman and Morgan-Stanley became bank holding companies, which helped them. Part of the
process was, obviously, doing the capital injections, which seemed to confirm it. But then the
question was -- which always happens and we’ve seen this in the currency crisis in the nineties,
for example -- the market will always test to find out where the limits are and how far it is that
the government is willing to go.
So part of what happened during ‘08 was that there was some perception that there were
firms too big to fail but the market wasn’t sure, tried to find out, and, obviously, it drove the
government limits, in some respects, to test the proposition that there were.
I think that, in part, explains why my flip comment about why it didn’t help. I don’t
think it helped because nobody was sure.
So in a curious way, constructive ambiguity persisted, and was obviously resolved on
almost a case-by-case basis. But it was that case-by-case resolution that effectively led to the
same constructive ambiguity that the regulators were working from for years. It was much more
suspended in this case.
MR. BONDI: And the failure of Lehman, what effect did that have then on the
perception of -MR. KELLY: Panic. I mean, the fact is that I think if you look back -- and, like you,
there’s a cottage industry in books about the period -- I’ve read lots of them, not all of them -but I think if there was one common theme that comes through from virtually everybody, is that
Lehman was a mistake. Whether it was a mistake in the broadest sense from a policy standpoint
I think is open to question.
Whether it was a mistake as an immediate matter in terms of the market impact, I think is
pretty clear, that unsettled the market.
It certainly, for a while, made things worse. And that generated, as I said, you know, the
greater anxiety, which ultimately led some more disquiet in the markets.
The truth is, as I said, the regulators, I think, in the government were handicapped by a
variety of deficits as a policy matter. The Treasury doesn’t have guarantee authority. The fed
doesn’t have guarantee authority. The only entity with guarantee authority is the FDIC. You
know, clearly it gets around the Treasury’s ability, which is why they needed the statutes -- or
the Feds, for that matter -- to make essentially capital investments in firms. So their resolution
22

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
authority that was inadequate.
So we confronted a crisis which really was genuinely unprecedented. In my own view,
probably the worst in history, without having been able to think through at the outset what it is
that we would need to cope with that, I think, in part, because it was just so unanticipated.
Look, I told you, I had a window onto what I thought might happen. What happened was
a thousand times worse.
MR. BONDI: To what extent did Citigroup’s super senior positions and CDOs
precipitate or add to the need for government assistance?
MR. KELLY: Well, insofar as it was sort of the triggering point, in terms of those
exposures we had and in terms of, as I said, the creation of the initial perception of vulnerability,
it was obviously a part of it.
I think in the event -- very difficult to quantify, other than the fact that, as I said, it
generated this perception of our vulnerability to additional losses. And part of the reason is you
know we structured the deal that we did with the government -- this is an important point -- just
so you know, the 20 billion that they provided us, we never asked for. It showed up on Sunday.
We were happy to have it, but we never asked for it.
What we had originally proposed was something very similar to Wachovia, because that
was fresh in our minds and, frankly, we had developed it. And that, as you know, was around -or at least our initial conception of it -- was around cutting off the perception of tail risk in the
market. In other words, we will get a guarantee against this 300 billion of assets. That will give
the market comfort that the catastrophic risk has been taken off the table.
So to the extent that the CDO -– super senior CDO exposure had generated the
perception of further marks, we were trying basically to put a limit on those marks, and that’s
why we structured the way we did.
Interestingly, none of those securities ended up in the asset -- in the ring fence. So for
other reasons.
MR. BONDI: And there were several mark-downs that Citi did with respect to its
positions. And we were working through the documents, and we’ve asked for documents
concerning where things are, and so I’m sure we’ll get to the precise numbers.
But can you tell me what ultimately the effect of the super-senior positions were on Citi
and Citi’s balance sheet?
MR. KELLY: I can’t. I don’t know. And, as I said, it was not through this process; I
wasn’t involved in that.
I mean, I could go back and look just as sure as everybody else is, but I just wasn’t
involved in it. I know it was substantial, but I don’t know the particulars of it.
MR. BONDI: And some of the exposure that Citi’s had with respect to subprime, in
particular the CDOs, I understand may be marked –- may be marked up -- or excuse me, I’m
23

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
probably not using the correct term.
MR. KELLY: Yes, you are.
I mean, the fact is that I think in the third quarter of this year in particular, I think there
was some -MR. MODE: Fourth quarter.
MR. KELLY: Fourth quarter, I’m sorry. Precisely. I’m sorry.
There were some markups and specialized pools. Obviously, that just reflects the fact
that markets are recovering; that liquidity is, by degrees, coming back into the market, which
allows people to focus more, as we discussed earlier, on sort of underlying values as opposed to
distressed value of these assets. But, again, I’m familiar with it but not immersed in the details
of it.
MR. BONDI: One of our many mandates is rating agencies, and to look at the effect of
rating agencies on the financial crisis.
MR. KELLY: Sure.
MR. BONDI: And I do appreciate, Mr. Kelly, your views in terms of the general
marketplace, as well as Citi.
But my question, at least initially here is, what effect did ratings have on the challenges
that Citi faced throughout the financial crisis?
MR. KELLY: I don’t know that it had any specific effect. And, again, understanding the
role that I played and where I was, I wouldn’t know particularly what our interactions were with
the agencies or what specific impact it had with us. I think it’s fair to say, and everybody knows
this, that, clearly, there were securities that were perceived to be super-senior AAA+ that
ultimately didn’t work out.
Rating agencies, obviously, reached those conclusions based on their own analysis. I
think it’s fair to say if you look back ex post, there were clearly some mistakes that were made.
Having said that, I’m not sure that those mistakes were wildly different than mistakes that market
participants made in terms of how they thought about the future. In other words, things that were
completely unanticipated actually crystallized and that had an impact.
The rating agencies, as you know, are something that are embedded in a variety of
statutes and are an inherent part of the market. I think it’s clear to say that they had some
stumbles, as we’ve gone through this. But that doesn’t make them any different from any other
market participant in that respect. But in connection with its specific impact on Citi, I can’t
address it. I’m really addressing that more or less as a market observer.
I’ve also joked that I understand it’s a problem. And I have thought about it a fair
amount, but I have no creative or ingenious view about how you might fix it. It’s a problem.
MR. BONDI: With respect to rating agencies specifically, did the subscriber-based
24

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
model have any impact with respect to Citi?
You mentioned that you didn’t believe that rating agencies had an effect on -MR. KELLY: This is Citi-specific.
MR. BONDI: -- Citi specific.
MR. KELLY: Uh-huh.
MR. BONDI: This question is, again, Citi-specific. But did -- with respect to the type of
model that was used, I assume that that same answer would be -MR. KELLY: It would be no. Precisely, it would be industry -MR. BONDI: Industry.
MR. KELLY: -- wide as opposed to Citi-specific. And I wouldn’t know if it did with
respect to Citi, anyway. But as I said, I think these were industry-wide phenomena.
MR. BONDI: Risk management has been a topic of discussion with respect to the
financial crisis as a whole.
MR. KELLY: Yes.
MR. BONDI: The question is, did concerns or -- concerns over risk management at Citi
cause or contribute to the need for Citi to have government assistance?
MR. KELLY: Not specifically -- you know, that I’m aware of.
As I said, it’s hard to separate risk management from the fact that we had the exposures
which, in fact, generated the perception of vulnerability. So I’m not suggesting that.
But having only arrived in February of ‘08, I have no idea how risk was managed on sort
of a particularized level at Citi. But I have my more general view, which you’ve heard me
suggest, which is that there were a lot of events in the market that were completely unanticipated
outside the models, outside people’s reasonable expectation, that clearly did violence to
expectations, not only here but also across the street.
I will say, you know, that as a unifying principle, in terms of thinking about what
happened here, I think in many respects, it was not just here but across the street, there was a
good, old-fashioned failure of credit judgment. And I think the fact is that there were some
decisions that were made, you know, where the benefit of hindsight and frankly even with some
respects of the benefit of foresight across the industry where people made some credit decisions
that clearly they came to regret.
MR. BONDI: Mr. Kelly, you occupy a unique position in the sense that you came into
this situation somewhat after the fact.
But have you, since joining Citi, developed any views concerning any lessons learned,
25

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
Citi-specific, that could have avoided the challenges that Citi faced in ‘07 and ‘08?
MR. KELLY: Yes. Again, it’s important -- I think, yes, they are Citi-specific. But to the
point, I think they’re also more broadly applicable.
We went through an extended period in the U.S. where, as I said, liquidity and capital
were, by and large, free. I think in some respects, that lulled people to sleep a little bit. And,
obviously, during a period of attractive low rates, people began to reach for yields since
short-funding was readily available, and you felt you could invest that on a mismatch basis with
relatively little anxiety. I think that that was a very common practice.
I had mentioned, or analogized the SIVs, the banks. But I think there were a lot of
financial firms that were SIV-like in the sense that they were short-funded and invested long in
the context of a liquidity crunch, ultimately you find yourself in very dire straits.
I think if there is a lesson learned, it is one that was taught to me when I was a very
young lawyer at Davis Polk doing bank due diligence, which is you die short. You know, if
there’s any lesson from this crisis, it is that, that you’ve got to be sure that your liability structure
is solid enough and long enough to give you the staying power necessary to weather whatever it
is the market might have to offer.
MR. BONDI: Mr. Kelly, I would be remiss if I didn’t ask you the question that we’ve
been asked to investigate by Congress, the central question, and that is the causes of the financial
crisis. That’s not a Citi-specific question, but certainly Citi was a market participant during the
financial crisis.
In your view, what were the causes of the financial crisis?
And I know that there will be treatises written on this for years, and we certainly are
investigating those. But, centrally, what do you view as the causes of the financial crisis?
MR. KELLY: Well, you know what’s interesting about that -- I had mentioned this to
P.J. and Brad at one point -- when I was a very young lawyer, one of the things that I was sent
off to do was to look at the origins and legislative history of Glass-Steagall. And, as you know,
part of what drove Glass-Steagall was The Depression of the banking crisis in the late 20’s and
the early 30’s. Having spent more or less a year, a year and a half in the library, I came up with a
list of the stated causes of The Depression, which was two pages, single-spaced bullet points.
I suspect that what’s going to happen is that we’ll have a similar analysis of what
happened here in terms of the contributing factors. It’s very difficult to focus on any one in
particular because markets are complicated things, and there were a number of things that
converged to create this.
But for me, at least, what it highlighted -- again, and this was true in The Depression as
well, and I think Chairman Bernanke’s made this clear -- is that liquidity is absolutely crucial to
the functioning of the world economic system and the system in the United States.
To the extent liquidity disappears for whatever reason –- as liquidity is usually dependent
on confidence -- the impact is dire. So you have the absence of the deposit insurance in the late

26

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
20’s, early 30’s, lots of small banks, they all failed. Huge contraction of the money supply. The
Fed didn’t act. It made matters worse. Because to the extent there was less money and less
liquidity out there, asset prices were going to fall, ended up aggravating the situation.
In some respects, even though it’s 80 years later and it’s a much more sophisticated
economy, ultimately the immediate cause of this was liquidity. People basically lost faith.
To your point, there was a crisis of confidence. They were no longer willing to believe.
They were no longer willing to take risk. And as a result, given that broad uncertainty, they were
unwilling to make decisions that, in the past, people had made in the normal course of every
single day, which had a huge impact, given the fact that it all happened simultaneously on the
system as a whole.
In terms of what led to that loss of confidence, which is really the broader question, I
really don’t know, and I’m not sure that anybody does.
But I could probably list, you know, six or seven factors that were contributors.
I think a protracted era of low rates -- you know, not to be flip about it, but when rates are
1 percent, everybody is a genius. You know, it’s impossible not to make money by levering.
I think the fact that there was a bubble that evolved and the impression generally was that
housing prices in the U.S. were going to continue to rise, and because fixed-income investors, in
light of those low rates had an appetite for yield, you were able to provide funding to people that
would have been deemed to be less creditworthy in prior periods.
The fact that even though the mortgages that would have been less creditworthy ended up
being dispersed by virtue of the sophistication of the financial system into a variety of
instruments, and there was no transparency as to where they resided or who held them, that
generated anxiety, lack of confidence with respect to who your counterparties could be.
So not surprisingly -- and I’m only, you know, a third of the way through the list if you
pressed me -- but it’s very difficult to pinpoint.
But at the end of the day, it seems to me that the inquiry has to be what factors
contributed to that, to such an erosion of confidence that it ultimately led to a seizure of the
credit lines. And that’s a very complicated question and one that I wish I were competent to
answer, but I’m not.
MR. BONDI: Given your reference to
Glass-Steagall, I have to ask you, did GrahamLeach-Bliley cause or contribute to the financial crisis, would that be on your list?
MR. KELLY: No, sir. Because in the absences of it –- in other words, assume GrahamLeach-Bliley never happened, this all could have happened. So I think it’s, by and large, in my
own view, sort of unrelated.
MR. BONDI: Thank you.
My colleague, Jane, has several more specific and fine-tuned questions than my inartful
questions. And so I’d defer to Jane now to ask you about some specifics.
27

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MS. POULIN: First, I’d like to go back and clarify a few things that you discussed with
Brad just to educate myself a little bit better.
One of the things -- I’m paraphrasing, if I’m not stating this correctly, please correct
me -MR. KELLY: Sure.
MS. POULIN: -- you said that when there was pressure on Citi’s stock price, that your
debt holders would -- had gotten nervous?
MR. KELLY: Yes.
MS. POULIN: And that that would cause your debt-funding to be less secure.
MR. KELLY: Sure.
MS. POULIN: So could you elaborate on that?
MR. KELLY: Sure.
MS. POULIN: I don’t understand it.
MR. KELLY: If depositors had the perception that the firm is troubled as reflected in
liquidity price, they might very well choose to withdraw funding. So they take their deposits out,
they don’t roll their credit that they had with us. It magnifies the liquidity crisis that was going
on in the market generally.
MS. POULIN: So you’re referring to customer deposits –MR. KELLY: Sure.
MS. POULIN: -- which I would understand.
MR. KELLY: Yes.
MS. POULIN: And short-term funding?
MR. KELLY: Absolutely.
MS. POULIN: There’s not -- would there be any long-term funding or medium-term
funding that would have any provisions?
MR. KELLY: Well, you know, they don’t have exit. It’s just that the price of it goes up,
so your ability to replace, if you will, the short-funding given the fact that it’s dried up is much
less, because the long-funding markets and the short-funding markets are closed, they’re closed
as well.
MS. POULIN: Okay. One of the other things that you mentioned was on the capital
ratios.

28

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MR. KELLY: Yes.
MS. POULIN: That you said that the tier 1 capital, that you were fine and healthy on
that, people were focusing on -MR. KELLY: TCE.
MS. POULIN: -- TCE.
I’m curious if you have a view on whether Citigroup and Citi Bank, each individually,
would have been able to clear the tier 1 capital requirements without the government funding?
MR. KELLY: Absolutely. Remember, the
recollection, were 4 and 8 at the time.

tier 1 -- the capital funds, to my

My further recollection -- I may be wrong, Jane, if you could go back and check this -- I
think our tier 1 ratio in November of ‘08 was like 11, if I’m not mistaken.
MR. BONDI: It was 10.8.
MR. KELLY: Yeah, so it was well in excess of the minimum.
But the tension that you see -- the tension is this, right, in other words, it depends on
one’s point of view, right? Are you running a bank for the bondholders and the depositors, or are
you running a bank for the equity?
I think what the regulators have concluded and what they thought about historically was,
they didn’t care about whether you were running a bank for the equity; they cared about whether
you were running a bank to generate security for the depositors and the bondholders, because
that was -- their funding was crucial for all the reasons that we’ve described.
What happened was that the feedback loop, which I don’t think was true historically
between the equity markets and the fixed-income markets, became much tighter. And what that
was reflected in is the relationship as we discussed between the CDS markets and what people
saw in our spreads, and the equity-market price began to unsettle people in the fixed-income
markets.
And the only way to stop this stock price from falling -- because, clearly, shorts don’t go
in for no reason -- was by virtue of having something that the equity believed would be left for it.
So preferred stock, for example, is there to absorb losses and to protect bondholders and
depositors, but it does the common stock no good at all.
So the regulators, historically, have been focused on the entire capital structure, you
know, including preferred stocks and subordinated debt, for example, so long as it was
loss-absorbing.
What happened was, the market, the common-stock market said, “That’s interesting, but
we don’t care. You know, what’s left for us?”

29

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
So if you actually look at our ratios pre and post, you’ll find that our tier 1 ratio,
notwithstanding the fact that we now have a tier 1 common ratio, which is I think is 8.2
pro forma for the first quarter, given some other issues, has increased dramatically, our tier 1
ratio is basically flat.
So what we did, was to take some of what people -- what we knew to be and what the
regulators thought to be loss-absorbing capital, and we converted it to common stock. So the
preferred exchange we did last summer, 58 billion, that was essentially preferred stock in
trust-preferred securities that became common, and then the common equity offering we did this
past December and, you know, the conversion of the government’s preferred common back in
early 2009 generated more common equity which, in turn, stabilized the stock price. But it
didn’t change the capital ratios from the standpoint of the regulators.
Now, in fairness, clearly, there were increases associated with those objections. And my
basic point is that if you look at tier 1, it’s been pretty constant. It’s the equity measures, the
TCE and the tier 1 common that have varied. But the great news is that we were able to use that
tier 1 capital that we had and convert that to equity which, in turn, helped to stabilize the stock
price -- a huge dilution, obviously -- but it gave the common equity holders more hope, if you
will, that they would have a stake in the firm. And that, in turn, I think, quiets the debt markets
because you cut that feedback loop.
Sorry, that was so long-winded and complicated.
MS. POULIN: When you -- when the government assistance was originally negotiated,
the loss-sharing arrangement, or the ring-fence arrangement, that was -- I think you said that that
was struck on a prototype of what you had thought about for Wachovia; is that correct?
MR. KELLY: Yes, uh-huh.
MS. POULIN: And could you just tell us what transpired? How is it that the deal was
struck? You spoke to me -A short story.
MR. KELLY: Well, you know, in one respect, from our standpoint, it is. Because I can
tell you that I had the one conversation. I think, as I’ve told others, New York Fed was on the
phone -– now-Secretary Geithner was on the phone very briefly because his nomination was
announced that afternoon, so we didn’t talk to him anymore and he didn’t participate. So it was
essentially first the New York Fed. I can’t remember whether anybody from Washington was on
the phone, and then the New York Fed evolved into New York Fed, Washington Fed, OCC,
FDIC, Treasury.
They said, “Okay, what do you want to do?” more or less. “We are in a position where
we believe we need to do something. What would you suggest?”
We went back to them, as I recall, that Friday night, or if not Saturday morning, with a
proposal that was very similar to the Wachovia proposal in terms of the ring-fence and the
loss-sharing agreement.

30

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
I remember having a conference call with them on Saturday about it. It was essentially
what we had served up. I think it was, by and large, one page. It had expected losses associated
with it, had the assets described, the amount of assets, suggested of what the loss-sharing
arrangement would be, and what our first loss would be in light of those expected losses.
There was some conversation back and forth, and then there were a number of
conversations during the course of the weekend that were sort of one-offs. And we didn’t really
hear back conclusively, again, to my recollection until late on Sunday. And at that stage, they
came back to us, if you will, with a revised term sheet with some variations in the ring-fence or
loss-sharing agreement, and with a proposal that they inject a further $20 billion in preferred
stock.
MS. POULIN: And, again, just to confirm, this, in your view, was to stabilize people’s
views of -MR. KELLY: Of Citi.
MS. POULIN: Of Citi?
MR. KELLY: Right.
MS. POULIN: To stabilize the stock price?

MR. KELLY: And the ring fence had the effect of cutting off that perceived tail risk.
But the other important thing to remember about it was that it also increased our capital
ratios in the following sense: We got the 20, but we also got asset-weighting relief with respect
to the assets were covered by the ring fence. In other words, irrespective of what their risk
weights were for regulatory purposes, they were deemed to be weighted at 20 percent. So my
recollection at the time was that we got $16 billion of capital relief for that, combined with the
$20 billion that we got as an injection. That was $36 billion. And then I think we were able to
count the 7 billion that we paid, in other words, that we issued to them in connection with that
loss-sharing. We were able to count, on a market basis, about 4 of that as an insurance premium.
So it increased our capital by about $40 billion.
MR. BONDI: Tier 1?
MR. KELLY: Tier 1. And that’s why we were willing -- that’s why we did it.
We didn’t actually think. And in the event, you know, we never got close to reaching the
first loss, which was ultimately 31 billion -- or 30 billion and 9 billion in reserves.
I think the ultimate losses we had before we unwound it were sort of 9 billion over five
quarters.
We did it because of the capital relief issue of, and the notion that it would enhance the
view of our capital issues in the markets and, therefore, help us stabilize the firm, in conjunction
with eliminating the tail risk.
31

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
But remember, it was a $39 billion first loss. So it wasn’t -- we still had lots of exposure,
potentially.
MS. POULIN: So the losses ultimately on the first loss -MR. KELLY: -- for that nine -MS. POULIN: -- was only nine?
And so let me go back to my question on the capital. I guess maybe I don’t understand.
If you got 40 billion -MR. KELLY: -- in capital.
MS. POULIN: -- in capital and it increased your tier 1, you would still think that through
the end of 2008 -- you used December 31, ’08, as the marker -- that you would still have satisfied
the minimum requirements.
Would you have satisfied the well-capitalized requirements?
MR. KELLY: I don’t know, I’d have to go back and look. But I think our tier 1 would
have still been adequate. I’m pretty sure that’s true. I don’t know for sure, but that would be my
recollection.
MS. POULIN: At the bank and at Citigroup?
MR. KELLY: Oh, yes. The bank was, frankly, always fine. You know, the issue was at
the group level.
That’s commonly true. Bank holding company structures, the banks are always more
adequately capitalized than all the companies, to my knowledge.
MS. POULIN: So then the -MR. KELLY: But, again, don’t hold me to that. I want to go back and verify it.
And my basic point was that we had lots of tier 1 all through this process, obviously
enhanced by what it is we did with the government. But the market didn’t care, right? The
market at that stage tended to be more focused on intangible combination.
MS. POULIN: Also at the time you were negotiating the exchange agreement and the
ring-fence arrangement, did Citi project -- this was in late November, right -- did Citi project
what losses you would expect to incur through the end of that year, of ‘08?
MR. KELLY: With respect to those assets?
MS. POULIN: With respect to those assets.
MR. KELLY: Yes, we did.

32

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MS. POULIN: And more broadly?
MR. KELLY: No.
MS. POULIN: Would you have -MR. KELLY: At least I didn’t. In other words, in terms of what I was working with, we
had a $300 billion pool of assets, which changed markedly, you know, during the course of the
negotiations.
But we did have an expected loss, we had reserves against it. And I can’t remember right
off the top of my head, I think the expected lifetime loss was 18 billion, and we may have had
9 billion in reserves against it. I can’t -- don’t hold me to that, but we did. We showed them
what we believed the losses to be in those assets for purposes of calculating what a first loss
might be. And then the view was, we’d have loss sharing beyond that.
MS. POULIN: Okay. I think I mentioned earlier the sup. senior -- the CDOs where you
had sup. senior exposure.
Were those in the ring-fence assets? Did you expect them to be originally and they were
never in?
MR. KELLY: Well, what happened -- and, again, it’s something that, unfortunately, is
not simple -- what happened was that the way the ring fence ended up being structured, it ended
up being very unfriendly in mark-to-market assets. So as a result, the ring fence was almost
exclusively accrual assets.
MR. BONDI: What do you mean by “the ring fence became unfriendly from mark to
market assets”? I don’t follow that.
MR. KELLY: Well, I’ll try to explain. It’s going to be difficult.
But, in any event, we have a $39 billion first loss, right, and we have -- the Treasury, I
think -- and P.J. or Brad will correct me -- the Treasury, I think, has the next five.
MR. MODE: Right, uh-huh.
MR. KELLY: The FDIC has the next ten, and the Fed then has the rest of it, you know,
in terms of the losses. So it’s 39, 5, 10, 90, 10. You’ve got to do the math. But essentially that
gets you to somewhere like $56 billion. And then beyond $56 billion, the Fed has it.
The Fed is not in the form of a guarantee; the Fed is in the form of a non-recourse loan.
And what became apparent is that we had thought about this -- we lose money, you pay
us. The Fed was, “We’re not going to lose any money, so whatever gains or losses are
embedded in these assets, we’re going to share in those. So if there’s a gain in an asset, that’s
going to offset whatever loss there is that we’ve incurred on any other asset.”
So if you put mark-to-market securities in, which we believed had higher fundamental
value, if you will, than was currently reflected in the market, we were going to be deprived of the
33

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
gain in terms of the erosion of a liquidity premium over time.
So that would go to the Fed rather than to us. And that made no economic sense to us.
So what we did was to go in with accrual assets, right, that had no marks associated with
it and they only had reserves associated with them. But they were easier to deal with and made
more sense economically.
And you may find that I think the reason that BofA ultimately concluded that they would
not execute their ring-fence agreement is because they had lots of mark-to-market assets in there,
and they concluded that it didn’t make any sense.
MS. POULIN: Let me just restate what I think you said -- and then help my colleagues.
So you have the accrual assets and you have mark-to-market assets. If the mark-tomarket assets went in the ring fence, anything -- any losses you share according to the
arrangement, but also any gains?
MR. KELLY: Any gains because -MS. POULIN: Whether they’re -MR. KELLY: -- it was going to be on a net basis.
MS. POULIN: Right, it was on a net basis.
And your perception was that the mark-to-market assets had more -- I’m using the
term “intrinsic value” but not in the classic sense -- more intrinsic value than they were -MR. KELLY: Yes.
MS. POULIN: -- the fair value they were stated as -MR. KELLY: I think –MS. POULIN: -- would go in.
MR. KELLY: -- we estimated the liquidity premium on the assets that we put in there
originally -- and, again, don’t hold me to this, but it’s a [inaudible].
I think we thought it was like $23 billion worth of liquidity premium that was embedded
in those assets that we put in there. So given the way the deal was ultimately structured, that’s
23 that we would be getting back.
MS. POULIN: Right.
MR. KELLY: Do you see what I’m saying?
MS. POULIN: So you saw an upside to those assets?
MR. KELLY: Absolutely.
34

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MS. POULIN: Don’t put them in?
MR. KELLY: Precisely.
MS. POULIN: And so that begs the question, what’s happened with the assets that might
have gone in, the ones we’re referring to, mark-to-market assets?
MR. KELLY: Well, I think by and large, those marks have held up pretty well. I mean, I
think our subsequent marks in the mark-to-market securities had been pretty modest. And, in
fact, as we just pointed out, there have been some markups associated with those assets, hence
the markets have recovered.
But the reason we put, going back to it –- just not to put too fine a point on it -- is accrual
assets, as you know, are held at par, net of reserves. Downside is downside but there is no
upside, if you see what I’m saying?
MS. POULIN: Right. Okay.
MR. MODE: Is it fair -- this subject, by the way, is incredibly complex, so that what
you’re getting, even though Ned has got a better grasp of it than anybody else -- what you’re
getting is about 2 percent of it.
But is it fair -- and I really don’t know the answer to this -- is it fair to say that the
government also liked this because, in general, it was getting -- the accrual assets that we’re
getting had a less tail risk than the -MR. KELLY: I think that’s fair. Absolutely.
MR. MODE: So then what was happening is, the government was also pressing us at the
time, as I recall, to put these accrual assets into the ring fence, because it was minimizing -- they
were taking the tail risk, and it was minimizing the tail risk.
MS. POULIN: Oh, they were taking the longer term -- the further-out tail risk.
MR. MODE: Yes, we had the first –-

whatever –-

MS. POULIN: It was all tail risk.
MR. KELLY: We had the -– we basically had the first 39, plus 10 percent of the rest.
MS. POULIN: Okay.
MR. KELLY: So you had 90 percent of the risk.
MR. MODE: So the government was trying to get the safest possible assets into -MR. KELLY: Yes.
MR. MODE: We announced the deal, and then we had to go negotiate it. And part of the
negotiation was, what goes into the pool? And the government wanted the safest possible assets
35

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
to protect the taxpayers because they were taking the tail risk.
It turned out that the economics were such that we wanted the accrual-based assets. And
it turned out to be -MR. KELLY: Consistent with what they wanted as well.
MR. MODE: -- consistent with what the government wanted.
MR. BONDI: While we’re on that topic of the negotiation, who from Citi was involved
in negotiating that?
MR. KELLY: Principally, myself and Brian Leach.
MR. BONDI: Anyone else?
MR. KELLY: Yeah, Gary Crittenden was around. You know, Vikram certainly had
conversations. Michael Helfer, you know, was our general counsel, was having conversations,
but -- and then Brian had a team obviously working with him who may have had conversations
at various points with the New York Fed. But that was the team.
MR. BONDI: Was Mr. Rubin involved, either directly or indirectly in these
negotiations?
MR. KELLY: Not that I’m aware of.
MS. POULIN: Could you tell us, broad-brush -- you commented that the assets that you
originally had scoped, that you thought would go into the ring fence, ended up being somewhat
quite different at the end.
MR. KELLY: Yeah.
MS. POULIN: We talked broadly accrual assets versus mark-to-market assets.
But is there more to it than that?
MR. KELLY: Sure. There were TARP limitations, Fed prudential limitations. There
weren’t just prudential but in some cases, statutory. So there was a whole series of filters, right,
that were imposed as we went through the negotiation. Because, remember, we did this on a
two-page term sheet over two days, had a high-level agreement, and then we had to implement
it -- I don’t think the agreement was executed -- P.J., correct me if I’m wrong -- until
mid-January, maybe, six weeks later date? With the documents. And then the asset selection, I
think, was going through the summer.
MR. MODE: Yes, that’s what I was thinking. The asset selection took forever.
MR. KELLY: Right, because of these filters.
So no asset that was originated past, I can’t remember, March 26th -- it was the date in
TARP -- March 26th, 2008, or whatever the date was.
36

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MR. MODE: That was the date, they were foreign versus domestic.
MR. KELLY: And then the Fed basically would not take security in foreign assets.
MS. POULIN: The list that’s included in the agreement?
MR. KELLY: Yes, precisely.
MS. POULIN: The not-very-long list, and then you mentioned the first –MR. KELLY: No equities.
MS. POULIN: -- three of the first four that are relatively straightforward. But it’s
everything that’s in that list?
MR. KELLY: Exactly.
MS. POULIN: Which would require you to go down asset by asset to make sure
everything cleared all those fences?
MR. KELLY: I think we ended up throwing out, I don’t know, north of 80 billion in
assets, as I recall, you know, just in terms of various substitutions.
MS. POULIN: Because when you did the initial -MR. KELLY: They didn’t pass the filters.
MS. POULIN: Yes, when you did the initial pass, it would be based on an aggregate -MR. KELLY: Right.
MS. POULIN: -- pile, yes.
MR. KELLY: And, see, we weren’t anticipating because we didn’t know -- we didn’t
understand how the Fed was going to structure the back end.
Remember, it’s a non-recourse loan. The Fed has limitations on its authority over what
they can lend against. So that drove some of the filters.
The statute, TARP drove some of the other filters. So what we were doing is operating
on the assumption, which turned out to be wrong, that effectively the Treasury and the FDIC
would be providing the guarantees.
The Fed showed up on the back end, when we didn’t know they were, if you see what
I’m saying?
Remember, the Fed wasn’t involved in Wachovia. That was an FDIC guarantee for
which we paid.
MS. POULIN: So let me understand a little bit better. We have a list of the hurdles that
the assets have to clear in the arrangement.
37

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MR. KELLY: Yes.
MS. POULIN: But the term sheet was struck, and it was a couple of pages.
And the date that the term sheet was struck was what –MR. KELLY: November -MS. POULIN: -- versus the date that the agreement finally was signed?
MR. KELLY: I think November 23rd or the 24th was the date of the term sheet. And the
agreement -- and I may be wrong, is the agreement was signed -- it may have been signed -- I
think the capital injection, I know, was December 31st. Maybe we signed the agreement
December 31st. I just can’t -- but it was later. I -- for some reason, I thought it was January 15th,
but I may be wrong.
MR. MODE: No, I think -MR. KELLY: December 31st.
MR. MODE: December 31st.
MS. POULIN: So that’s why you had one group of assets in mind and another group of
assets in -MR. KELLY: And remember, we signed the agreement without having determined the
assets.
I think the agreement itself suggested that there would be a review of the assets and a
final determination of the list, which I think actually, ultimately, slipped into August.
MR. MODE: It was certainly the summer.
MR. KELLY: Yes.
MR. MODE: It was a huge task. You can’t -- there were dozens of people involved. I
mean, I wasn’t personally involved, but I remember the magnitude of the task. It was
consuming.
MR. BONDI: Was Mr. Karp involved or -MR. KARP: I was standing around.

MR. KELLY: That’s right, you were the basket --

MR. KARP: Right.

38

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010

MR. BONDI: Well, you were nodding your head, Mr. Karp, and I was wondering if you
were involved.
MR. KELLY: It was an immense task, is the point. Because as you might imagine,
300 billion in assets and an awful lot of individual loans. And you’ve got to go through and
figure out whether they pass the filters or not.
So we had discussions, for example, about whether a nominal foreign obligor guaranteed
by a U.S. entity would count, all right. So Ford Motor Company has got a subsidiary in the UK.
If it’s a subsidiary, a loan guaranteed by Ford Motor Company, does it get in or does it not?
MS. POULIN: Okay.
MR. KELLY: My recollection is, it got thrown out. I can’t remember. Even though it
seemed to us it’s obviously a U.S. credit, we’re lending on the strength of the U.S. parent. But
because it was to a foreign obligor, no good.
MS. POULIN: I understand.
MR. KELLY: Yes.
MS. POULIN: But when the ring-fence arrangement was sized, did Citi suggest the size?
MR. KELLY: Yes.
MS. POULIN: And you had some view on which assets broadly would be included?
MR. KELLY: Well, we had two.
One is, we had -- frankly, again, you have to keep in mind that we’re doing this very
quickly, all right. So the size actually was very similar to Wachovia. I think Wachovia was 300;
this was 306.
We went through asset categories that we thought might fit, backed into that number and,
in part, were driven by how much would be enough to give the market comfort that we have, in
fact, eliminated that tail risk.
There was not a huge amount of science in coming to that number.
MS. POULIN: Okay.
MR. BONDI: Why was the Wachovia -- Wachovia was 300, 306.
I don’t think of Wachovia as being an equivalent to Citigroup. Why then use the
ring-fence equivalent for Wachovia as the model?
MR. KELLY: Well, as we looked through it -- and, again, thinking about it -- we were
thinking about our exposures, how much would we need to persuade the market that we had
39

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
eliminated tail risk. And it was coincident in one respect, but it triangulated into about 300.
And what Wachovia had actually done, to the extent we could over the course three days,
a pretty detailed analysis of Wachovia’s portfolio. Remember, they had 120 billion in option
ARMs. They had an awful lot of commercial real estate. They had some other structured
securities. And we actually had come to the conclusion there that it was 300 billion.
But remember, that was a much larger percentage of Wachovia’s balance sheet than 300
was of ours. So 300 in Wachovia, I think, on a balance sheet -- you know, correct me if I’m
wrong -- but it’s probably a trillion at the time. So that was 30 percent of the balance sheet. We
were 300 on 2 trillion, so it was 15 percent. But we were trying to get to a number that was -and, again, not a lot of science -- but a number that was large enough to persuade the market that
we had eliminated tail risk but not so large as to be impractical.
MS. POULIN: With regards to the SIVs, should I stay away from questions that relate to
them coming on the balance sheet?
MR. KELLY: Yes, because I wasn’t -- exactly. I wasn’t here.
As I said, I showed up in February of ‘08. I was here, but in a slightly different role, but
still responsible for it in November.
But at that point, essentially the die had been cast. And at that stage, it was just an
economic decision about what made more sense. Continued to chase it as a mezzanine matter or
actually take the assets into our own -– in other words, actually buy the assets.
MR. BONDI: Were you involved -- you said you got here in November of -MR. KELLY: February ’08.
MR. BONDI: February ’08? Excuse me, I’m sorry. I’ll withdraw that question then.
MS. POULIN: Some of these questions, I may have a view on. I want to just confirm
with you.
MR. KELLY: Sure.
MS. POULIN: Let’s see. There were $3.3 billion of losses on the SIV assets in 2008.
Were there losses on the SIV assets in any other period in addition to that? Or was that
the only losses you recognized on the SIVs?
MR. KELLY: I don’t know. But I think -- remember, those losses were driven, by and
large, by marks. It would have been most acute in ‘08. I think they probably -- I just don’t
know.
And I can’t remember -- I can’t remember. Some of those SIV assets may have been
held to maturity. In other words, once we took them onto the balance sheet, I’m not positive, but
my recollection is that they may have been; which, of course, would have eliminated the issue
around the marks. But I may be wrong but I just don’t know the answer.
40

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
Mr. KARP: Actually, this is publicly disclosed; and you’re going to be talking to other
witnesses who are familiar with this. We can clear this up with you and -MS. POULIN: It may have been held to maturity?
MR. KELLY: Yes, precisely.
MS. POULIN: Okay. So back on the loss-sharing arrangements. Were the SIV assets
ultimately included in the ring fence?
MR. KELLY: I don’t know.
MS. POULIN: I’m not recalling the wind-down because there’s a wind-down.
MR. KELLY: I don’t think there were any SIV assets in the ring fence. I’m not positive,
but I don’t think there were.
MS. POULIN: Could we clarify whether there is or not?
MR. KELLY: Sure.
MR. BONDI: Perhaps -- yeah, Brad, you can get back to us separately, and that would
be helpful. Thank you.
MS. POULIN: And the public filings refer to the SIVs being wound down.
Can you explain to us what you mean by “winding the SIVs down”? The SIV assets
come on the books and there was some liability still standing that also came on the books of Citi?
MR. KELLY: Well, no, we would have paid off the liability.
MS. POULIN: You paid off all the liability?
MR. KELLY: Right. So in other words, we owned the assets. And at that stage, there
are two ways to wind it down. One is through sales. The second is obviously just through
organic runoff, because they had an average life associated with them. I think it was all term
debt.
I think the fact is that it was relatively short, even though it was much longer than the
funding we had. But a lot of financial institutions’ papers we discussed.
So I think that it’s -- the wind-down essentially is, we’re obviously not reinvesting any
proceeds; we are going to let this portfolio run off naturally and hopefully facilitate it through
sales.
And that’s why I’m hesitating around the health and maturity, because I’m not sure we
did put it maturity, but I think we probably -- we can confirm that.
MR. BONDI: And am I correct to say that there was no Lehman debt in those SIVs?

41

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MS. POULIN: There was no Lehman debt in SIVs.
MR. KELLY: WaMu, as I said, is the only default that I was aware of.
MS. POULIN: And it remains, I think, to this day the only actual default.
MR. KELLY: You see, the funny part about the SIVs is that the credit judgment was
fabulous.
MR. BONDI: Right, right.
MR. KELLY: It was the funding structure that was the flaw.
MS. POULIN: This is to clarify my understanding. When -- perhaps you can tell me
how -- when you look at Citi’s off-balance-sheet business, how would you view that in terms of
different buckets? For example, there’s CDOs and CLOs, there’s mortgage and student loan,
credit-card securitizations -MR. KELLY: Just ordinary course. And I think that was the -- again, I’m focusing sort
of on my time as opposed to what happened before I got here. But I think those standard
financial institution securitizations are the vast majority of what we had off-balance sheet.
I think with 166, 167, what have we brought back on, 180 billion or so, basically, which
is all sort of consumer securitizations.
MS. POULIN: Okay, CDOs, CLOs, then I have the asset-backed -MR. KELLY: Commercial paper.
MS. POULIN: -- commercial-paper conduits that you administer?
MR. KELLY: And I think there are about 19 or 20 -- don’t hold me to that, but my
recollection is there are 19 or 20 or so.
MS. POULIN: Okay.
MR. KELLY: I don’t know. You know, but that’s something we can check.
MS. POULIN: And then there’s third-party commercial paper conduits, single and
multi-seller.
This is the kind of -- as I look -- if I looked across all the off-balance-sheet stuff, I’m
trying to understand, are these the types of categories, if I were to try to categorize the
off-balance-sheet structures -MR. KELLY: Yes.
MS. POULIN: -- vehicles, these are the categories?
MR. KELLY: And I think we’ve got fairly extensive disclosure around that. I think

42

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
what you’ve just described, if I’m not mistaken, would be in that 180 we just brought back on
balance sheet as a result 166, 167.
SIVs would have been separate from that because they came back on a while ago. But
166, 167 would have covered, I think, most of those synthetics.
MS. POULIN: And then the other category being the SIVs?
MR. KELLY: Yes. And they were already on.
MS. POULIN: I think there is a footnote in the K’s, it actually breaks it down in Citi’s
sense, kind of the Enron days, has had a market meeting in the footnote that actually breaks it
down by category and gives a lot of information about it.
I wanted to get your take on what I see in the public filings is, in fact, how you would
view it? If you look at the off-balance-sheet stuff, and it falls into those basic categories?
MR. KELLY: Absolutely.
MS. POULIN: Okay. And -MR. KELLY: There was -- I think the point that Brad made -- look, there was always a
lot of -- I can remember reading blogs and other things talking about how, “Oh, my God, Citi’s
got, you know, gazillions in off-balance sheet.”
I remember reacting at the time, silly; right? At the end of the day, it’s pretty standard,
plain, vanilla stuff with certain exceptions. But predominantly, it’s precisely what it is you
would expect to see.
MS. POULIN: And you mentioned 166, 167, in your profile, and you’ve got the impact
on the Citigroup. But is there any markedly different impact if you were looking at just the
bank?
MR. KELLY: I think most of those assets are in the bank; are they not?
I’m pretty sure they’re coming back.
I don’t know, actually. I take that back. I haven’t looked at it in that light. But I take
that back. But I wouldn’t think so.
MS. POULIN: That would be just interesting for us to know.
MR. KELLY: Sure.
MS. POULIN: To give us a better sense of what assets are in the bank and what the
assets aren’t in the bank.
MR. KELLY: Just to give you rough numbers, I think we have 1.8 trillion -- now, with 2
trillion in assets, my suspicion is there are two-thirds of them in the bank – roughly, two-thirds
are in the bank, and the balance is in the holding company, in the broker deal, in particular.
43

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MS. POULIN: And then in 2009, there’s what I understand to be the three major credit
card -- or three major card trusts: The Master, Omni, and Broadway.
Can you just elaborate what occurred in 2009? I’ve read it in the filing, but I’d like -- if
you could explain that to us.
MR. KELLY: I don’t know. I don’t know enough to answer. But I know enough to be
dangerous, right, which I think had to do with the OCC at one point. And I think it was pursuant
to a regulatory ruling that was -- I take it all back.
What I’m remembering is that there was either a regulatory rule which caused us to
consolidate those trusts. What I can’t remember is whether there’s anything we did specifically
to support those trusts which caused us to have to consolidate them. But there was about
80 billion, as I remember.
MS. POULIN: What I believe I read was there was approximately 80 billion -MR. KELLY: Yes. And I just can’t remember what the trigger was.
MS. POULIN: -- coordinated support provided by the Citi.
MR. KELLY: Right. Well, that’s probably what the trigger was. In other words, we had
to step in, we supported the trust and, therefore, we were forced to consolidate them.
MS. POULIN: And it says it impacted your -MR. KELLY: It would have impacted the balance sheet.
MS. POULIN: Yes, no, I thought what I read was, it impacted your capital requirements.
But it wasn’t clear as I read it whether they came on balance sheets.
Ultimately, that’s my question is, did they actually come on balance sheets?
MR. KELLY: Well, that’s the interesting point.
They did for regulatory purposes.
MS. POULIN: Oh, okay.
MR. KELLY: In other words, they did for regulatory purposes, but they did not, for
GAAP purposes, I think, until 166, 167.
MS. POULIN: And these would have been -- let’s make it -- this is some accounting,
more detail. I apologize, I’m trying to keep away from the accounting stuff. But these would
have been QSBs?
MR. KELLY: I don’t know.
MS. POULIN: So my question is ultimately if you provided the subordinated funding,
why wouldn’t -- if these are SPEs, why wouldn’t this have been a trigger event, a reconsideration
44

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
of that under 1046R?
MR. KELLY: In terms of bringing it back up for tax purposes?
MS. POULIN: There are QSPEs -- maybe because they’re QSBEs, and then -- and even
then, I’d like to better understand, if you -- the QSPE would have been set up in a certain
fashion -MR. KELLY: I think the accounting guys can walk you through that.
MS. POULIN: Yes.
MR. KELLY: I know, again, based on my recollection -- and I may be entirely wrong -that they were brought back on for regulatory capital purposes but did not show up on a GAAP
balance sheet. That’s my recollection. And, in fact, we had to make a capital injection into one
of the banks. And I can’t remember whether it was Citi South Dakota or Citibank N.A., in order
to compensate for their coming back online.
MR. MODE: You were relying on his recollection of accounting issues?
MR. KELLY: Right.
MR. MODE: Which is -MR. KELLY: Right. And I’m just saying that I just don’t -- I’m just -- again, my
recollection is back on for regulatory, not for GAAP. I don’t remember specifically what the
triggering event was, nor do I know why they would have come back on it for regulatory and not
for GAAP.
MS. POULIN: Okay. And this is a little bit of a knit. In your disclosure in the 2009
10-K, I believe it’s in the footnote where you discuss the sup. senior CDOs and CLOs, in the
2009 filing, you refer to 24 out of 39 of the CDO structures coming on balance sheet between
2007 and 2009. In your prior year filing, 10-K, is the exact same paragraphs but refers to 34 out
of 46 coming on between 2007 and 2008.
So in two years, I took 34 out of 46. In three years, I took 24 out of 39 on some -- so I’m
confused by the disclosure.
MR. MODE: I’m sorry, I didn’t follow exactly. But we have sold some of these things.
MS. POULIN: Yes, that’s what I was wondering, if some of them have been sold and not
included in the current numbers.
MR. KELLY: Well, they could have sold, or presumably some of them could have been
matured, less likely.
MR. MODE: Or sold or finished.
MR. KELLY: That’s right.

45

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010

MS. POULIN: Actually, no, some of them. Some of the ‘03 ones had a six-year tenure,
so, yes…
MR. KELLY: That’s right.
Theoretically, these things have 20-year lives. But, in fact, they have -MS. POULIN: Actual tenure is between five and ten years, yeah.
MR. KELLY: But we have -- you’re going to talk to Brian, right, tomorrow? Brian will
be able -- is more likely on SIV. But we, opportunistically, when there’s a good price for these
things –MR. MODE: We sold them.
MR. KELLY: -- we’re selling them, so…
MS. POULIN: I think it might be helpful for us to understand comparative numbers. So
through 2009, on the same basis as the numbers that were presented in the 2008 10-K, out of the
total 46, how many came on? And then we know some of those either were matured or sold.
MR. KELLY: -- sold, uh-huh.
MS. POULIN: And the distinction -- I mean, the super-senior positions, the company
retained them back when the deals were done, were already on balance sheets. So this is a
reconsideration of that potentially, depending on what happened.
MR. KELLY: Right.
FEMALE VOICE: Structure.
MS. POULIN: Yes, to bring the whole structure on, I understand.
MR. BONDI: And just for housekeeping purposes, I know Mr. Leach has been
designated to talk about the same topic, and I probably should have asked at the outset again, and
I apologize, what areas is he designated to talk about that may be different from Mr. Kelly? Or
is there -MR. MODE: Well, as you know, he’s a chief risk officer.
MR. BONDI: Correct, yes.
MR. MODE: And Brian was involved in a bunch of these negotiations; but more as
opposed to negotiating the deal, as it were, he was focused on the assets, the risk analysis, what
we could put in, what made sense to put in. And his team was involved in the ultimate selection
of the assets that went into the ring fence. And then Brian took it over as chief risk officer, I
think, about the same time I arrived with the firm, February of ‘08.

46

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
MR. KELLY: That’s right.
MR. KELLY: So he clearly inherited lots of this, and then has worked through it.
So is very knowledgeable in marking valuation of these incidents.
MR. KELLY: Right.
MR. BONDI: All right, thank you.
MS. POULIN: Could you -- to the best of your ability, could you describe for me the
differences between your role and the chief first officer’s role? So we can understand -MR. MODE: They’re entirely different.
MR. KELLY: I think the simplest way to think about it is Brian is the chief risk officer;
I’m the deal guy.
I’m your classic investment banker.
MR. MODE: He’s had other roles in this life, but -MR. BONDI: Formerly a lawyer.
MR. KELLY: Precisely. And formerly, I ran a commercial bank. But that’s the
difference.
In other words, would I -- and I think you know this -- what I do is deal with a set of
facts, trying to figure out what they are, and then we come up with a structure that satisfies both
parties.
MS. POULIN: Okay. And how would you describe the risk officer’s?
MR. KELLY: Well, I mean, to P.J.’s point, he’s clearly involved in setting standards
around the assumption of risk, risk limits, how it is that we think about the valuation of assets,
how they are being valued on the desk, whether it makes sense optimally to sell this asset or that
asset. Works very closely with the business people in assessing all of those things.
MS. POULIN: Okay.
MR. BONDI: I have a couple more, and that’s, you alluded, Mr. Kelly, earlier to reading
books on topics. And I do confess that I have read a couple on the topic of the financial crisis,
and one of which was Mr. Paulson’s book, Secretary Paulson’s book. And he describes in his
book a telephone call that he received from Mr. Robert Rubin on or about November 18th, which
then prompted a conversation between the Secretary and the President, President Bush, on
November 19th. Then, of course, on November 23rd was when the government assistance with
respect to Citi was announced.
My question is, did you have a role in preparing or briefing Mr. Rubin for this
November 18th call to Secretary Paulson to the extent that it took place? I’m not attesting to the
47

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
veracity or the date of that call. It’s only what I’ve read.
But did you have a role in briefing Mr. Rubin for a call or conversation with Secretary
Paulson?
MR. KELLY: No. And I have -- to your point, I have no idea whether it occurred.
The first time I got involved in the run-up -- I mean, intimately involved in what we were
talking about was that Friday afternoon, as I described, when I was asked to come down and
participate in that conference call.
MR. BONDI: Mr. Kelly, there’s been some news reports of some comments that you’ve
made about regulators. I don’t want to ask you about those comments specifically, but I would
like your views in terms of the role of regulators in leading up to the need for government
assistance. I had asked you several questions about events at Citi that prompted the need or
requirement for government assistance.
And I want to ask specifically this question with respect to the regulators: Were there
any actions by any regulators that caused or contributed to Citi’s need for financial assistance
from the government?
MR. KELLY: No, no. And what I mean by that is that regulators include players in the
market in the sense that they are participants and have various authorities and respect the firms
that they supervise.
But there was nothing that the regulators did that, in my view, for what it’s worth, led
directly or indirectly to Citi’s need for government assistance.
Now, if you wanted to take a jaundiced view -- which I don’t, right, for all the reasons
that I’ve described -- you could argue that insofar as they were there and that this happened, that
they somehow must be complicit. But that’s a far different issue.
MR. BONDI: We had tangentially talked about liquidity puts, not directly but we talked
about super-seniors and questions with respect to liquidity puts.
What was the effect of these liquidity puts on Citigroup and its financial condition?
MR. KELLY: Don’t know.
But I do know -- I don’t know specifically. Again, I wasn’t involved.
I can tell you that anecdotally I heard that obviously, that the impact was, in fact, adverse
because they were there. And I think in many cases, they were exercised.
MR. BONDI: Right. And taking that question just one step further, do you know if the
fact of the liquidity puts and their exercise caused or contributed to Citigroup’s need for
government assistance?
MR. KELLY: For all the reasons that I’ve described earlier, just in terms of the
perception of Citi and market reaction, they were at peace with that.
48

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
But I don’t know that they had any direct, quantifiable impact, if you see what I’m
saying. But there were a variety of a factors.
MR. BONDI: Was Mr. Prince at all involved in the negotiations for government
assistance?
MR. KELLY: Not to my knowledge.
MR. MODE: He was gone.

MR. KELLY: I think he had been gone for a

year.

MR. MODE: He was long gone.
MR. BONDI: Since November 2008 and not 2007. My apologies. You’re correct.
MR. KELLY: Yeah.
MR. BONDI: Because there are so many questions that I probably have forgot to ask
you, Mr. Kelly, I’m going to defer to my colleagues -- the remaining colleague, colleagues
Ms. Norman and back to Ms. Poulin, to see if they have any additional questions that I may have
overlooked in the course of talking to you.
MR. KELLY: Sure.
MR. BONDI: Ms. Norman?
MS. NORMAN: I’m sure that this is in the public realm.
MR. KELLY: I’m sorry?
MS. NORMAN: I’m sure that this is in the public realm and any agreements. But if you
could explain to me, what is the natural end of the government assistance agreement that Citi has
right now?
MR. KELLY: Well, as you know, we’ve repaid the 20. And we -MS. NORMAN: So you’re out of the TARP house?
MR. KELLY: Exactly, and we unwound the ring-fence agreement which, as I
mentioned, we paid, in rough numbers, 7 billion for, and then ultimately paid the government for
the one year of coverage, where we had had losses of about 9 billion, we paid them 5 billion for
that, which was part of the unwind. So there was a very healthy fee that was earned by the
49

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
government for the provision of that protection.
What we have remaining is, as you know, 25 billion was the initial TARP injection that
was converted to common stock, or agreed to be converted into common stock in February or
March; and then ultimately, when a preferred exchange was done in July or August, it was
converted to common at 7.7 billion shares.
The basis -- cost basis is $3.25.
Now, the government has announced that it plans to exit that position by the end of this
year in connection with the equity offering that we did in December. There was a 90-day lockup
which expires March 16th. The government is considering how it is it might dispose of those
shares. And there a variety of ways to do it.
You know, they can do it through an underwritten offering, they can do it through a
dribble-out offering, they can do it through a continuous program in the market, they can do it
through some combination of the two.
But I know that the Treasury and the TARP officer, Allison and his people are thinking
throughout, that they want to effect those sales. But they’ve made it clear that they’d like to be
out sometime over the course of the next year, if not by the end of the year.
MS. NORMAN: And this probably won’t be clear until the beginning of the following
year, but what do you think net-net will be, the cost to the U.S. government, for the cost to Citi,
of that combined package? If you could net that out.
MR. KELLY: Well, I think what I can do -- well, we could get for you the amount that
was paid to the government in dividends, you know, with respect to the preferred interest.
As I mentioned, we paid a combination of the Treasury and the FDIC -- we paid the
FDIC three, Treasury four. But I think we ultimately solved that, leaving 5 behind. I don’t
know how that was allocated. I think the FDIC may have ended up with 3 billion and the
Treasury with 2 billion and change, but I’m not positive.
MS. NORMAN: Is this something that Citi has internally documented, what it expects
the net loss to be?
MR. KELLY: I think what I’m struggling with is trying to figure out how much the
government made.
I don’t think that, you know, based on what it is that we see so far -- look, I don’t have
perfect foresight, I have no idea where the stock price is going to end up; but as it stands, it made
a lot.
MS. NORMAN: Unlike Jane or even Brad, I have no accounting experience which
makes things very complicated, and it does seem to be going up and down. But if you could -MR. KELLY: It’s not, though. See, think about it this way, right -- in other words, they
earn dividends in the 20 billion, 8 percent is outstanding for roughly a year. So they would have
earned 1.6 billion on that.
50

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
We paid them 5.2 billion, including the coupon that we paid on that 8 percent trust
preferred during the period of time 2009, while the insurance was in effect. So they got -- they
got $5.2 billion, plus the interest that they earned on that. They got repaid the 20, but they got a
1.6 billion coupon, rough numbers. It’s probably more than that because there are stub periods
in there in terms of the interest that they collected. And they have 25 billion of common stock,
which probably today is worth 26 billion.
To the extent the stock stays above 3¼, they have a gain, 7.7 billion times the difference
between 3¼ and the market price. To the extent that it is below 3¼, they have a loss -- same
math -- 7.7 billion times the amount of that is below 3¼.
So as it stands now, as I said, they’ve got a $5 billion fee for a risk that never crystallized,
that was insurance, which is important and valuable but they got 5 billion for that. They got the
interest associated with that, and they got 1.6 billion on the 20 billion. And they got some
interest payment even before on the 25, before they converted to common stock.
MR. MODE: And these -- with regard to the common stock, it depends on the stock
price.
With regard to everything else -MR. KELLY: It’s fixed, it’s fixed.
MR. MODE: -- these are -- we can give you the exact numbers from there, it’s just
arithmetic.
MR. KELLY: Yes.
MS. NORMAN: I think that helps for me.
MR. KELLY: Sure, sure.
MS. NORMAN: It probably is already in the public domain. Thank you.
MR. MODE: We’ll just do the calculations.
MR. KELLY: But they’ve made money so far.
MS. POULIN: I have a question on the off-balance sheets.
So the SIVs and the CDOs and the CLOs, the Citi has recognized some losses on those
arrangements, those off-balance-sheet vehicles. Some of them come on balance sheets and that’s
what caused the losses. But the Citi-administered aspect, commercial paper conduits, I don’t see
the same thing going on with that group of off-balance-sheet structures.
Can you explain to me what it is about those structures that we’re not seeing losses? And
if –- and is there any liquidity issues in them now? Has there been?
MR. KELLY: You know, I don’t think so. But I’m going to defer to others. And we
could find the right person to answer that question.
51

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010
I think my impressionistic view is in all the assets that you’ve cited, there have been mark
losses. I suspect there just haven’t been any. In other words, there have been mark losses or
credit losses.
In what you’ve described, I think there may not -- just not have been any losses.
MS. POULIN: So maybe the composition of the assets?
MR. KELLY: Precisely.
MS. POULIN: And so I’m curious, is it the composition of the assets, is it the structure
of the entities? Is it long versus short assets and liabilities mismatch, in some structures and not
in others?
MR. KELLY: My suspicion is it’s probably -- well, you’re right because composition of
assets and length of funding are related because the assets can be money-good. But if there’s
short liquidity, you are obviously going to have a problem because you’re selling on a fire sale,
so I just don’t know. But somebody can figure that out.
MS. POULIN: Well, my recollection is with the SIVs and the CDO structures, it was
credit and liquidity.
MR. KELLY: Yes, but in the SIVs, remember, I keep going back to that, really it wasn’t
credit.
MS. POULIN: It was liquidity? Okay.
MR. KELLY: I’m not saying that it was absolutely not credit. But as we discussed, the
credit judgments were actually quite good. It was the fact that liquidity dried up and they had to
be sold into distressed markets.
MS. POULIN: So that would be the open question. Just so we can understand those
structures versus the others, it will help us better understand our balance sheet.
MR. KELLY: We can track that down.
MR. BONDI: Mr. Kelly, I always like to ask someone at the end of an interview if
there’s anything that we didn’t cover that you expected to talk about. Or is there any question
that we didn’t ask you that you were expecting?
MR. KELLY: Not at all. I think you were very comprehensive. I tried to give my best.
And I appreciate it.
MR. BONDI: Mr. Kelly, I think -- do we have any -- Ms. Dubas, do you have any -MS. DUBAS: No, I don’t. Thank you.
MR. BONDI: Anyone else?
[No response]
52

FCIC Interview of Edward J. “Ned” Kelly III – March 3, 2010

MR. BONDI: We certainly appreciate your time and that of Citigroup in cooperating
with our inquiry.
As I mentioned at the outset, our inquiry is confidential, so we ask that you don’t discuss
what we talked about outside of your attorneys, of course. We certainly appreciate your time.
Mr. Karp, did you have anything?
MR. KARP: I don’t -– I do not. Thank you.
MR. BONDI: We did have -- I’m reminded by Jane that we had a few accounting
questions to break down some losses that we can do, I think, off-line with your accounting staff.
MR. KELLY: Sure.
MR. BONDI: It is just mainly subclassification of what’s already been disclosed in your
public filings -MR. KELLY: Sure.
MR. BONDI: -- breaking down some of those losses. So we’ll deal with that off the
record.
MR. KELLY: And, Jane, I’m sure they can get the answers. In fact, if I had it in front of
me, I could probably figure out some of the answers with you. But they could easily go through
and give you the precise answers that you need.
MR. BONDI: Terrific.
Thank you very much. We appreciate your time.
(End of interview with Edward J. “Ned” Kelly III)
--o0o--

4850-9824-2309, v. 1

53