Repos: A Deep Dive in the

Collateral Pool
Macro Credit Research August 1, 2012
www.ftchratings.com
Analysts
Macro Credit Research
Martin Hansen
+1 212 908-9190
martin.hansen@fitchratings.com
Robert Grossman
+1 212 908-0535
robert.grossman@fitchratings.com
Kevin DAlbert
+1 212 908-0823
kevin.dalbert@fitchratings.com
Fund and Asset Manager Group
Viktoria Baklanova
+1 212 908-9162
viktoria.baklanova@fitchratings.com
U.S. Money Fund Exposure and European
Banks: Eurozone Hits Fresh Low,
July 26, 2012
Repo Emerges from the Shadow,
Feb. 3, 2012
Related Research
Research Highlights
Sample based on 10-largest prime money
market funds.
Repo collateral by asset class (% share of
total collateral)
Treasurys and Agencies ........... 65%
Structured Finance ................. 18%
(see table at right)
Corporate Debt and Equities .... 15%
Corporate and equity issuers (% of
corporate debt and equity collateral pool)
SPDR Gold ........................... 7.0%
Westpac ............................... 1.4%
Verisign ................................ 1.4%
Repos in the Spotlight: Repurchase agreements (repos), a core part of the shadow banking
system, are increasingly in the spotlight, given both their importance as a funding mechanism
and their role in past episodes of market distress. This study updates Fitch Ratings earlier
report, Repo Emerges from the Shadow, dated Feb. 3, 2012, which highlighted the post-
financial crisis resurgence in the use of structured finance collateral within triparty repo markets.
As revealed through Fitchs analysis of the 10 largest U.S. prime money market funds (MMFs)
disclosures, structured finance repos are typically collateralized by pools of securities that are
of lower credit quality (e.g. CCC and below), deeply discounted, and small in size. Additionally,
while Treasurys and agencies represent a significant majority of collateral, repos are also used to
finance corporate debt, gold, and equity securities. Funding relatively less liquid, more volatile
assets through repos (which are effectively short-term loans) creates potential liquidity risks for
both repo borrowers and the underlying assets.
Focus on Structured Finance: Triparty markets are used to finance roughly $90 billion of structured
finance securities, based on estimates from Federal Reserve Bank of New York (FRBNY) data (see
table, Structured Finance Repo Collateral: A Drill-Down). As context, average daily trading volumes
for non-agency residential mortgage-backed securities (RMBS) and asset-backed securities (ABS)
combined is about $6 billion, according to SIFMA. Fitchs sample, which captures $21.2 billion
or almost one quarter of the structured finance securities funded through triparty repo, reveals the
potential liquidity risks inherent in much of this collateral. These liquidity risks stem from both
the small size of many of these securities, whose median value is about $800,000, and their
distressed nature. For example, roughly 50% of this sample consists of legacy CDOs and subprime
and Alt-A RMBS, much of which was originated by financial institutions that experienced severe
distress related to their securitization and mortgage-related exposures during the U.S. credit crisis.
Approximately 60% of structured
finance repos within Fitchs
sample were conducted with
three institutions: Credit Suisse,
Royal Bank of Scotland, and
Bank of America (Merrill Lynch).
Interestingly, each of these
institutions has been active in the
FRBNYs Maiden Lane auctions
of crisis-era RMBS and CDOs.
Funding Risks: Structured
finance repo yields averaged
72 bps as of end-February,
roughly four times greater than
the 18 bps yields on repos
backed by Treasury and agency
securities. Structured finance
repos provide a potentially
attractive return opportunity for
some money funds in the current
low-yield environment. However,
U.S. prime money market funds,
which act as repo lenders within
the U.S. triparty repo market, are
Structured Finance Repo Collateral:
A Drill-Down
(As of End-February 2012)
Fitch Sample Size (Based on Top 10 U.S. Prime MMFs) ($ Bil.) 21.2
N.Y. Fed Sample Size (Based on Entire Triparty Market)
($ Bil., Estimated) 90
Value of Individual Collateral Securities (Median, Rounded) ($) 800,000
Number of Securities Within Collateral Pool (Rounded) 4,200
Repo yield (Structured Finance Collateral) (bps) 72
Repo yield (Treasury and Agency Collateral) (bps) 18
Haircut for Structured Finance Repo (Median) (%) 7.6
Repo Leverage (Implied from Median Haircut) 14 to 1
Subsectors of Interest (% of Structured Finance Collateral Pool)
RMBS (Alt-A) 26
RMBS (Subprime) 17
CDO 6
Top 3 Largest Issuers (% of Structured Finance Collateral Pool)
1. Countrywide 7.2
2. Credit Suisse 6.6
3. Royal Bank of Scotland 3.8
Top 3 Largest Counterparties Structured Finance Repos
(% of Structured Finance Collateral Pool)
1. Credit Suisse 30
2. Royal Bank of Scotland 19
3. Merrill Lynch (Bank of America) 13
Source: Fitch Ratings, Form N-MFP, Federal Reserve Bank of New York.
2 August 1, 2012
Repos: A Deep Dive in the Collateral Pool
short-term and highly risk-averse investors, as demonstrated by the
severe decline in structured finance repo during the height of the
U.S. financial crisis. Therefore, any disruption in repo funding for
structured finance could impair the liquidity and valuation of these
assets, affecting not only repo market participants but also cash
investors that take long positions without the use of leverage. Such
disruptions might also affect financial institution counterparties
that use repo markets to fund inventory of structured finance and
other riskier securities. As noted in the Financial Stability Oversight
Councils recently published 2012 Annual Report: Some dealers
remain very dependent on short-term repo funding and are heavily
exposed to rollover risk. Of particular concern is the use of short-
term borrowing to finance less-liquid collateral, such as asset-
backed securities or corporate bonds.
Triparty Repo: Leveraging the
Available Data
Despite the extensive interest in repo markets, data is relatively
scarce. This study helps to fill that gap by analyzing disclosures of
the 10 largest U.S. prime money funds. MMF disclosures provide
the most detailed publicly available historical information on repo
haircuts, pricing, collateral, and counterparties. Based on these
disclosures, Fitch has been able to construct a time series of repo
attributes back to end-2006, capturing trends before, during, and
after the U.S. credit crisis (see pages 5 and 6 for Fitch time series
data for both collateral mix and haircuts).
The recently developed SEC Form N-MFP is an unparalleled source
of granular, security-level repo information, including issuers and
valuations, enabling Fitch to take a more in-depth view of the
collateral pool. This detailed analysis complements the aggregated,
high-level data provided by the FRBNY, which offers a comprehensive
summary of the triparty repo market as a whole (see table, U.S.
Money Fund Disclosures and N.Y. Fed Data: Complementary
Perspectives). Taken together, the greater transparency afforded by
both FRBNY and MMF data sheds light on risk trends in the triparty
market and offers a glimpse into the potentially larger, but more
opaque, bilateral repo market.
This study is based on repo transaction information sourced from
a sample of the 10 largest U.S. prime MMFs public disclosures.
Fitchs sample encompasses about $116 billion in repo transactions
as of end-February 2012, which represents roughly 7% of the
$1.76 trillion U.S. triparty repo market (see table, Triparty
Repo: Comparing the Fitch and N.Y. Fed Samples). In terms
of nontraditional or riskier collateral (i.e. structured finance,
corporate debt securities, and equities), Fitchs sample captures
$40 billion in collateral, or 16% of the $248 billion financed in
the triparty market. Much of the triparty collateral not reflected in
Fitchs sample consists of relatively smaller prime MMFs, MMFs that
transact strictly in U.S. government securities as collateral (and,
therefore, do not accept riskier forms of collateral), and securities
lenders reinvesting cash proceeds, for which there is not publicly
available data of comparable granularity.
Fitch constructed the repo dataset based on a variety of MMF disclosures.
Fitch derived the historical time series (year-end 2006 through mid-
year 2010) from quarterly, semiannual, and annual reports, but used
the more recently developed Form N-MFP for subsequent periods.
Importantly, there appear to be differences, in some cases significant,
in how shareholder reports and Form N-MFP categorize the form of
collateral. As a result, some of the changes between midyear 2010 and
year-end 2010 observations could result from differences in reporting
standards. Several methodological assumptions underlie this study and
are detailed in the Appendix on page 6.
Structured Finance Repos: Thirst for Yield,
Less Liquid Collateral
Fitchs analysis of Form N-MFP, which captures roughly $21 billion
in structured finance repo collateral, provides a detailed glimpse into
the risk attributes of the roughly $90 billion of structured finance
securities funded in the triparty repo market. First, the generally
small size of these positions, which are typically pooled within a
single repo, raises potential concerns about their liquidity during
more stressed market conditions. As an indication of this high
granularity, the median value of the more than 4,200 structured
finance securities in Fitchs sample is about $800,000, with almost
a quarter less than $50,000 (see table, Structured Finance Repo
Collateral: Smaller Positions Potentially Less Liquid). Smaller-sized
tranches can make it difficult for potential investors to research and
price these securities and, in turn, for existing holders to either fund
or liquidate their positions.
U.S. Money Fund Disclosures and N.Y. Fed Data:
Complementary Perspectives
Fitch Sample of
10 Largest Prime
Money Funds
Federal Reserve
Bank of New York
Coverage of Triparty Market
(as of End-February 2012) 7% 100%
Beginning of Time Series End 2006 May 2010
Frequency of Available Data Monthly
a
Monthly
Collateral (by Broad Sector) Yes Yes
Collateral (Security Level/Issuer) Yes No
Collateral (by Subsector/Industry) Yes No
Collateral Security Value Yes No
Haircuts (by Collateral Type) Yes Yes
Repo pricing (i.e. Transaction Level) Yes No
Repo Maturity Yes No
Counterparty (by Institution) Yes No
a
Form N-MFP has been available on a monthly basis (with a two month lag) since
November 2010. Prior to that period, detailed information was available on a quarterly
basis for each fund. Fitchs time series is based on select periods and does not include
each month.
Source: Fitch Ratings, Form N-MFP, MMF nancial statements, Federal Reserve Bank
of New York.
Triparty Repo: Comparing the Fitch and
N.Y. Fed Samples
($ Bil., As of End-February 2012)
Fitch Sample of
10 Largest Prime
Money Funds
Federal Reserve
Bank of New York
Structured Finance 21 91
Corporate (Debt) 12 78
Equities 6 79
Total Risky Collateral 40 248
Treasury 20 579
Agency 56 907
Total Treasury and Agency 75 1,485
Other 1 26
Total Collateral 116 1,759
Source: Fitch Ratings, Federal Reserve Bank of New York, Form N-MFP.
3 August 1, 2012
Repos: A Deep Dive in the Collateral Pool
Additionally, structured finance collateral tends to be of low credit
quality. Many of these securities are lowly rated, have long tenors,
and are deeply discounted as illustrated by three securities that are
broadly representative of the structured finance collateral within Fitchs
repo sample (see table, Structured Finance Repo Collateral Tends to
Be Deeply Discounted, Low Credit Quality). For example, as a rough
measure of their valuation, the ratio of the security value to the principal
amount for these three securities ranges from 0.15 (essentially, 15
cents on the dollar) to 0.71 (essentially, 71 cents on the dollar). These
securities are all rated CCC (or an equivalent rating) or below.
Another illustration of the distressed nature of much of this collateral
is that more than three-quarters of the structured finance pool
consists of RMBS, much of which is backed by legacy subprime and
Alt-A loans (see table, Structured Finance Repo Collateral Half of
Pool is Alt-A, Subprime, and CDO). Additionally, several of the largest
issuers of the structured finance collateral in Fitchs sample were
financial institutions that experienced severe distress related to their
securitization and mortgage-related exposures (see table, Structured
Finance Repo Collateral Some Legacy Names Among Top Issuers).
Since Fitchs sample represents about one-quarter of all structured
finance collateral within triparty repo, it is unclear whether this
analysis also accurately describes the risk profile of the remaining
75% of structured finance securities funded through this market.
FRBNY data for ABS and private label CMO combined both investment
and non-investment-grade securities into single categories from mid-
2011, making it difficult to determine whether these attributes apply
to the broader pool of structured finance repos.
For MMFs (which act as lenders in the triparty repo market), repos
backed by structured finance securities generate markedly higher
returns than repos on traditional forms of collateral, a potentially
significant motivation in the current low yield environment. Median
yields on repos backed by structured finance collateral (72 bps)
were four-times higher than Treasury and agency repo yields (18 bps)
as of end-February (see table, Structured Finance Repos Generate
Higher Yields). While part of this yield differential likely stems
from the relatively longer maturities of structured finance repos,
it is likely also a reflection of the incremental risks associated with
this lower quality form of collateral. Fitchs understanding is that
MMFs generally make a binary yes/no decision about whether to
transact with a specific counterparty and, therefore, focus their repo
activities on financial institutions that they would otherwise lend
to on an unsecured basis. From this perspective, MMFs view repo
collateral, even if of lower credit quality, as an additional measure
of security in their counterparty management.
Structured Finance Repo Collateral:
Smaller Positions Potentially Less Liquid
Value of Security (Range) No. of Securities % of Total Pool
<$50,000 928 22
$50,001 to $1,000,000 1,322 31
$1,000,001 to $10,000,000 1,646 39
>$10,000,000 329 8
Total Securities 4,225 100
Total Value of Pool $21.2 Billion
Total Value of Pool (Security-Level Details) $14.3 Billion
Mean Value $3.4 Million
Median Value (Rounded) $800,000
Note: Above analysis is based on the portion of the roughly $14.3 billion in structured
nance collateral that is identied at a security-level in the N-MFPs. This analysis does
not consider the roughly $7 billion in structured nance collateral for which security-
level details are not provided in the N-MFP.
Source: Fitch Ratings, Form N-MFP.
Structured Finance Repo Collateral Tends to
Be Deeply-Discounted, Low Credit Quality
Security Maturity
Value
($)
Principal
($)
Ratio of
Value/
Principal
Rated
CCC
(or
Equivalent)
or Below?
WAMU 2006-AR10 1A1 09/25/36 6,801,157 9,554,349 0.71 Yes
SASCO 2004-21XS 1M1 12/25/34 799,524 3,945,084 0.20 Yes
JPMAC 2007-CH5 M1 05/25/37 1,203,966 7,800,000 0.15 Yes
Source: Fitch Ratings, Form N-MFP, Bloomberg.
Structured Finance Repo Collateral
Half of Pool is Alt-A, Subprime, and CDO
Sector % of Structured Finance Repo Collateral Pool
RMBS 76
RMBS Alt-A 26
RMBS SUBPRIME 17
RMBS PRIME 7
RMBS Various 27
ABS 5
CMBS 12
CDO 6
CDO SF 1
CDO Various 5
Source: Fitch Ratings, Form N-MFP, Bloomberg.
Structured Finance Repo Collateral
Some Legacy Names Among Top Issuers
Issuer (Structured Finance) % of Structured Finance Repo Collateral
Countrywide 7.2
Credit Suisse 6.6
Royal Bank of Scotland 3.8
Bear Stearns 3.4
Bank of America 3.3
JPMorgan Chase 3.0
GMAC 2.9
Morgan Stanley 2.3
Washington Mutual 2.3
Lehman 2.3
Deutsche Bank 1.7
Citigroup 1.6
Goldman Sachs 1.4
Indymac 1.2
Merrill Lynch 1.2
Santander UK 0.9
UBS 0.9
Wachovia 0.8
CIT 0.8
Wells Fargo 0.7
Source: Fitch Ratings, Form N-MFP.
Structured Finance Repos Generate
Higher Yields
Yield (Median)
Haircut
(Median)
(%)
Implied
Leverage
(x)
Structured Finance 0.72 7.6 14
Agencies (U.S.) 0.18 2.3 44
Treasury (U.S.) 0.18 2.0 51
Source: Fitch Ratings, Form N-MFP.
4 August 1, 2012
Repos: A Deep Dive in the Collateral Pool
For securities dealers (which act as borrowers in the triparty repo
market), repos provide a source of leverage and cost-effective funding
for their inventory of structured finance securities. While median
repo haircuts for structured finance collateral have increased from
about 5% at end-August 2011 to 7.6% as of end-February 2012,
this higher haircut still implies roughly 14:1 leverage, meaning that
a dealer can finance $14 of securities with $1 of invested capital.
Additionally, repos provide financial institutions with a relatively
low-cost form of financing for their structured finance holdings,
particularly in light of the low credit quality, small size, and longer-
tenor of many of these securities.
As of end-February, five financial institution counterparties Credit
Suisse, Royal Bank of Scotland, Bank of America (Merrill Lynch),
Citibank, and Barclays together accounted for almost 80% of all
structured finance repo borrowing within Fitchs sample (see table,
Structured Finance Repo Largest Counterparties). Credit Suisse,
which was not among the five largest counterparties as of end-
August 2011, accounted for roughly 30% of all structured finance
repos as of end-February. This emergence at the top of the list seems
consistent with Credit Suisses activity in the Maiden Lane auctions
of structured finance securities that the FRBNY has acquired as part
of its crisis-era interventions with Bear Stearns and AIG. Interestingly,
each of the top five counterparties was at some stage this year one of
the winning bidders of a Maiden Lane auction, an occurrence that
might indicate repos are used to fund these securities.
Equity and Corporate Debt Repos: Gold,
Tech, and Financials at the Top
The equity and corporate debt repo collateral in Fitchs study
includes a wide mix of names, with more than 1,500 total unique
issuers within the combined pool as of end-February 2012. Much of
this collateral consists of securities issued by large, well-established
companies, whose obligations trade actively in secondary markets.
However, as illustrated by the sheer number of names represented
within this sample, the securities of smaller and medium-sized
companies are also financed through triparty repo.
Notably, the repo collateral sample includes some lower quality
names, including issuers that have either defaulted or experienced
severe credit distress during the past few years. While not generally
representative of the credit quality of the sample as a whole,
examples of names appearing within the corporate debt pool are
AMBAC, Lehman Brothers, and Sino Forest (see table, Corporate
Debt and Equity Collateral: Mix of Names, Some Low Quality).
A notable change since Fitchs prior study is the presence of several
technology-related companies among the largest names within the
corporate debt and equity collateral pool. Nine of the 20 largest
issuers are tech companies, an indication of more active trading and
investor interest in the sector (see table, Corporate and Equity Repo
Collateral Gold, Tech, and Financials). The largest single name
is gold exchange traded funds, an asset class that has also been a
focal point for many investors.
Five of the 20 largest issuers are financial services names, creating
potential wrong-way risks in which the same risk factors affect the
financial performance of both the collateral issuer and the financial
institution counterparty. This correlation between the creditworthiness
of the counterparty and the value of the collateral could reduce
recoveries in the event of a counterparty failure. The presence of
structured finance repo collateral likely exacerbates this risk given its
correlation with the performance of financial institutions.
The presence of gold, tech companies, and financials sectors
which have been the subject of ongoing market interest reflects
the important role that repo markets play in providing liquidity to
trading-oriented investors. For example, hedge fund positions that
are financed through bilateral repos might, in turn, be financed in the
triparty market by their dealer and prime brokerage counterparties.
Collateral, Counterparty, and Haircut Trends
Over the past several observation periods, the collateral mix within
Fitchs sample has remained relatively stable with a few notable
exceptions. In the six-month period between end-August 2011 and
end-February 2012, Treasurys increased from 5% of the overall
Structured Finance Repo
Largest Counterparties (Repo Borrowers)
(% of Total Structured Finance Collateral)
End-August 2011 End-February 2012
Bank of America/MerrillLynch 27 Credit Suisse 30
Royal Bank of Scotland 23 Royal Bank of Scotland 19
Barclays 22 Bank of America/Merrill Lynch 13
Citibank 9 Citibank 7
JPMorgan Chase 7 Barclays 7
Top Five (Aggregate) 87 Top Five (Aggregate) 77
Source: Fitch Ratings, Form N-MFP.
Corporate Debt and Equity Collateral:
Mix of Names, Some Low Quality
Security Maturity
Coupon
(%)
Value
($) Principal($)
Ratio of
Value/
Principal
AMBAC Financial Group 12/05/35 5.95 1,951,756 13,472,000 0.14
Lehman Brothers Holdings 01/24/13 5.63 6,869,424 25,093,788 0.27
Sino Forest 10/21/17 6.25 1,052,333 3,075,000 0.34
Source: Fitch Ratings, Form N-MFP.
Corporate and Equity Repo Collateral
Gold, Tech, and Financials
Issuer % of Corporate and Equity Repo Collateral Pool
SPDR Gold 7.0
Westpac 1.4
Verisign 1.4
General Electric 1.4
Bank of America 1.3
Micron Technology 1.3
Virgin Media 1.3
Netapp 1.3
Lehman 1.2
Hartford 1.2
General Motors 1.2
Citigroup 1.1
Cadence Design 1.0
Microchip Technology 0.9
Novellus 0.9
Hologic 0.8
Xilinx 0.8
Biomarin 0.8
BorgWarner 0.8
Ford 0.8
Note: Above analysis is based on the portion of the roughly $11 billion in corporate
debt and equity collateral that is identied at a security-level in the N-MFPs.
Source: Fitch Ratings, Form N-MFP.
5 August 1, 2012
Repos: A Deep Dive in the Collateral Pool
repo collateral mix to 17%. On a combined basis, Treasurys and
agencies, which represented roughly half of the repo collateral pool
in end-August, increased to roughly two-thirds of the pool as of end-
February. Over this same period, the share of equities and corporates
both declined, while structured finance remained relatively stable at
18% of the pool. Interestingly, on a dollar basis, the amount of
structured finance repo collateral within Fitchs sample increased
by about 10% over this period (from $19 billion as of end-August
to $21 billion as of end-February) given that Fitchs total repo
collateral pool increased from roughly $93 billion to $116 billion
over this period.
A relatively small number of financial institutions tend to
dominate repo trading, although there has been a mild decrease
in counterparty concentrations of late (see table, Concentrations in
Largest Counterparties Declines from Crisis Peak). As of the end
of 2008, the five largest counterparties accounted for more than
80% of all repo trading in Fitchs sample. Since then, the top fives
share of repo activity has declined steadily and, as of end-February,
accounted for roughly 50% of all trading. In terms of the largest
individual counterparties, seven institutions were among the top 10
as of both end-August 2011 and end-February 2012 (see table,
Largest Counterparties Relatively Stable, with Some Shifts). Over
this period, notable changes include the rise of Credit Suisse as the
largest repo counterparty; a moderate decline in the shares of activity
by Bank of America/Merrill Lynch, Barclays, JPMorgan Chase, and
Deutsche Bank; and the entry of Goldman Sachs, Mizuho, and Royal
Bank of Canada within the top 10. Interestingly, as of end February,
only one euro zone institution, Deutsche Bank, remains among the
10 largest counterparties.
Another notable trend is that median haircuts on repos backed by
riskier forms of collateral have increased moderately since end-
August 2011 (see table, Moderate Increase in Haircuts on Riskier
Collateral). Over the six-month period between end-August 2011
and end-February 2012, median haircuts for repos collateralized by
corporate debt securities increased from 5% (or 21:1 leverage) to
7% (or 15:1 leverage). Over this same period, median haircuts on
structured finance repos increased from 5% (or 21:1 leverage) to
7.6% (or 14:1 leverage).
This increase in haircuts could indicate greater conservatism by repo
lenders, since higher haircuts imply a larger buffer against potential
price volatility and a deleveraging for repo borrowers. However, it
is important to note that the price declines already experienced by
deeply discounted legacy structured finance collateral significantly
exceed haircut levels. As a hypothetical example, for a pool of
structured finance securities that on average is valued at 45 cents
on the dollar, the realized 55% decline in value is several multiples
Repo Collateral Mix
(% of Fitch Sample)
Period Treasury Agency
Treasury
and Agency Corporate
Structured
Finance Equity
Unallocated
Riskier
Collateral CP
Mixed
Various
Riskier
collateral Total
2H 2006 0 38 38 36 18 2 0 1 6 62 100
1H 2007 0 32 33 23 33 5 0 6 0 67 100
2H 2007 17 32 49 20 24 2 0 2 3 51 100
1H 2008 0 41 41 27 11 14 0 1 6 59 100
2H 2008 9 44 53 8 5 31 3 1 0 47 100
1H 2009 13 43 57 9 0 19 12 3 0 43 100
2H 2009 13 40 54 13 4 14 8 3 4 46 100
1H 2010 10 45 56 22 4 10 8 1 0 44 100
2H 2010 16 28 45 11 13 21 7 2 1 55 100
Aug. 2011 5 47 52 17 21 10 N.A. 1 0 48 100
Dec. 2011 23 37 60 14 19 6 N.A. 0 0 40 100
Jan. 2012 18 44 63 12 19 5 N.A. 1 0 37 100
Feb. 2012 17 48 65 10 18 5 N.A. 1 0 35 100
N.A. Not applicable, as all nontraditional collateral is allocable due to form N-MFP.
Source: Fitch Ratings, Form N-MFP, and MMF nancial statements.
Largest Counterparties Relatively Stable,
with Some Shifts
(% of Total Repo Collateral)
End-August 2011 End-February 2012
Bank of America/Merrill Lynch 17 Credit Suisse 12
Barclays 15 Bank of America/Merrill Lynch 11
JPMorgan Chase 13 Barclays 11
Deutsche Bank 10 Royal Bank of Scotland 9
Royal Bank of Scotland 7 JPMorgan Chase 8
Credit Suisse 5 Deutsche Bank 7
BNP Paribas 4 Goldman Sachs 5
Wells Fargo 4 Mizuho 5
Citibank 4 Citibank 5
ING 4 Royal Bank of Canada 4
Top Five (Aggregate) 61 Top Five (Aggregate) 52
Top 10 (Aggregate) 82 Top 10 (Aggregate) 78
Source: Fitch Ratings, Form N-MFP.
0
20
40
60
80
100
10 Largest Repo Counterparties
Five Largest Repo Counterparties
Source: Fitch Ratings, Form N-MFP, and MMF financial statements.
(% of Total Repo Activity)
Concentration in Largest Counterparties Declines
from Crisis Peak
6 August 1, 2012
Repos: A Deep Dive in the Collateral Pool
higher than the median haircut of 7.6%. Put differently, a 7.6%
buffer on a security valued at 45 cents on the dollar would cover
a valuation decline of roughly 3 cents on the dollar (i.e. with the
security now pricing at 42 cents on the dollar). Given the relatively
lower liquidity of much of this collateral, it is unclear whether these
haircut levels would fully cover the bid-ask spreads for structured
finance collateral, particularly under stressed market conditions.
Moderate Increase in Haircuts on Riskier
Collateral
Period
Treasury
(Median)
Agency
(Median)
Corporate
(Median)
Structured
Finance
(Median)
Equity
(Median)
2H 2006 N.A. 2.0 3.0 4.5 5.0
1H 2007 2.0 3.0 3.0 4.2 5.0
2H 2007 3.0 3.0 3.2 5.0 5.0
1H 2008 2.0 2.8 5.0 5.0 5.0
2H 2008 2.0 2.0 5.0 3.5 10.0
1H 2009 2.0 2.5 5.0 N.A. 10.0
2H 2009 2.0 2.4 6.8 6.9 8.1
1H 2010 2.0 3.0 7.7 8.0 8.0
2H 2010 2.0 2.5 6.2 8.0 8.0
Aug. 2011 2.0 2.3 5.0 5.0 8.0
Feb. 2012 2.0 2.3 7.0 7.6 8.0
N.A. Not applicable (i.e. no collateral of that type within Fitchs sample).
Source: Fitch Ratings, Form N-MFP, and MMF nancial statements.
Background on Fitch Study
This research study is intended to provide market participants with information on triparty repo markets. Although it is based on
MMF financial statements, the study does not comment specifically on Fitch-rated funds. This report does not have any ratings
implications at present.
The sample set is based on public filings from the 10 largest prime institutional and retail MMFs, as measured by assets under
management, as of each observation period. Therefore, in some cases, the MMFs sampled differ slightly from period to period.
Because this analysis is based on aggregated data for the 10 MMFs sampled, it does not capture potential differences in repo
activity across individual funds. To identify the 10 largest funds that make up Fitchs sample, Fitch uses rankings of fund assets
under management sourced from Crane Data LLC. Fitch sources all of the underlying repo data directly from MMF disclosures, not
through a third-party data source.
Repo collateral is classified into one of the following categories: Treasurys, agencies, corporates, equities, structured finance,
commercial paper, and mixed pool/various. For certain periods, the disclosures of a fund in the sample did not provide information
on the asset class of the collateral underlying its nongovernment repos.
Agencies encompass both debt instruments issued directly by the U.S. government-sponsored enterprises (GSE) and mortgage-
backed securities guaranteed by GSEs. Structured finance includes whole loans and collateral classified as mortgage.
Commercial paper is primarily issued by corporate entities.
Data for the more recent periods (i.e. end-2010 and after) is based on MMFs Form N-MFP, which generally provides security-
level details for repo collateral. Historical data (i.e. end-2006 through mid-2010) is based on MMFs quarterly, semiannual, and
annual reports, which in most cases provide sufficient data to both classify repo collateral by broad asset class and to calculate the
corresponding haircuts.
In some cases, however, the asset class information provided in the quarterly, semiannual, and annual reports is ambiguous.
In particular, it is sometimes unclear whether collateral identified as government securities is Treasury or GSE securities.
In these cases, Fitch has made a classification determination based on inference from other data points. For this reason, as
compared with Fitchs prior study, Repo Emerges from the Shadow, the relative mix of Treasury versus GSE securities has for
some historical periods been revised in this study. However, the overall combined share of Treasurys and GSE securities together
is largely unchanged.
Additionally, analysis of the N-MFP shows the prevalence of mixed pools of collateral (i.e. in which securities from multiple asset
classes collateralize a single repo transaction). Therefore, classifications provided in the quarterly, semiannual, and annual reports
might in some instances reflect pools comprised of multiple asset classes.
The period of observation covers nine distinct semiannual periods and month-end August 2011, December 2011, January 2012,
and February 2012. Note that prior to 2011, financial reporting dates often varied across MMFs. Therefore, Fitch has applied a
degree of judgment in categorizing individual MMF filings into the appropriate semiannual bucket within its historical time series.
For the last observation period (Feb. 29, 2012), the repos in Fitchs sample represent roughly $116 billion.
Mixed pool/various includes both collateral instruments that do not fit into the other categories and pools of collateral that include
multiple categories. A collateral pool is allocated to one of the categories whenever possible given the quality of disclosure, even if
including some mixing across categories. For example, a pool consisting primarily of agencies, but also including some Treasurys,
is typically designated as agency collateral.
7 August 1, 2012
Repos: A Deep Dive in the Collateral Pool
ALL FITCH CREDIT RATINGS ARE SUBJECT TO CERTAIN LIMITATIONS AND DISCLAIMERS. PLEASE READ THESE LIMITATIONS AND DISCLAIMERS BY FOLLOWING
THIS LINK: HTTP://FITCHRATINGS.COM/UNDERSTANDINGCREDITRATINGS. IN ADDITION, RATING DEFINITIONS AND THE TERMS OF USE OF SUCH RATINGS ARE
AVAILABLE ON THE AGENCYS PUBLIC WEB SITE AT WWW.FITCHRATINGS.COM. PUBLISHED RATINGS, CRITERIA, AND METHODOLOGIES ARE AVAILABLE FROM THIS
SITE AT ALL TIMES. FITCHS CODE OF CONDUCT, CONFIDENTIALITY, CONFLICTS OF INTEREST, AFFILIATE FIREWALL, COMPLIANCE, AND OTHER RELEVANT POLICIES
AND PROCEDURES ARE ALSO AVAILABLE FROM THE CODE OF CONDUCT SECTION OF THIS SITE.
Copyright 2012 by Fitch, Inc., Fitch Ratings Ltd. and its subsidiaries. One State Street Plaza, NY, NY 10004.Telephone: 1-800-753-4824, (212) 908-0500. Fax:
(212) 480-4435. Reproduction or retransmission in whole or in part is prohibited except by permission. All rights reserved. In issuing and maintaining its ratings, Fitch
relies on factual information it receives from issuers and underwriters and from other sources Fitch believes to be credible. Fitch conducts a reasonable investigation of the
factual information relied upon by it in accordance with its ratings methodology, and obtains reasonable verication of that information from independent sources, to the extent
such sources are available for a given security or in a given jurisdiction. The manner of Fitchs factual investigation and the scope of the third-party verication it obtains will
vary depending on the nature of the rated security and its issuer, the requirements and practices in the jurisdiction in which the rated security is offered and sold and/or the
issuer is located, the availability and nature of relevant public information, access to the management of the issuer and its advisers, the availability of pre-existing third-party
verications such as audit reports, agreed-upon procedures letters, appraisals, actuarial reports, engineering reports, legal opinions and other reports provided by third parties,
the availability of independent and competent third-party verication sources with respect to the particular security or in the particular jurisdiction of the issuer, and a variety
of other factors. Users of Fitchs ratings should understand that neither an enhanced factual investigation nor any third-party verication can ensure that all of the informa-
tion Fitch relies on in connection with a rating will be accurate and complete. Ultimately, the issuer and its advisers are responsible for the accuracy of the information they
provide to Fitch and to the market in offering documents and other reports. In issuing its ratings Fitch must rely on the work of experts, including independent auditors with
respect to nancial statements and attorneys with respect to legal and tax matters. Further, ratings are inherently forward-looking and embody assumptions and predictions
about future events that by their nature cannot be veried as facts. As a result, despite any verication of current facts, ratings can be affected by future events or conditions
that were not anticipated at the time a rating was issued or afrmed.
The information in this report is provided as is without any representation or warranty of any kind. A Fitch rating is an opinion as to the creditworthiness of a security. This
opinion is based on established criteria and methodologies that Fitch is continuously evaluating and updating. Therefore, ratings are the collective work product of Fitch
and no individual, or group of individuals, is solely responsible for a rating. The rating does not address the risk of loss due to risks other than credit risk, unless such risk is
specically mentioned. Fitch is not engaged in the offer or sale of any security. All Fitch reports have shared authorship. Individuals identied in a Fitch report were involved
in, but are not solely responsible for, the opinions stated therein. The individuals are named for contact purposes only. A report providing a Fitch rating is neither a prospec-
tus nor a substitute for the information assembled, veried and presented to investors by the issuer and its agents in connection with the sale of the securities. Ratings may
be changed or withdrawn at anytime for any reason in the sole discretion of Fitch. Fitch does not provide investment advice of any sort. Ratings are not a recommendation
to buy, sell, or hold any security. Ratings do not comment on the adequacy of market price, the suitability of any security for a particular investor, or the tax-exempt nature
or taxability of payments made in respect to any security. Fitch receives fees from issuers, insurers, guarantors, other obligors, and underwriters for rating securities. Such
fees generally vary from US$1,000 to US$750,000 (or the applicable currency equivalent) per issue. In certain cases, Fitch will rate all or a number of issues issued by a
particular issuer, or insured or guaranteed by a particular insurer or guarantor, for a single annual fee. Such fees are expected to vary from US$10,000 to US$1,500,000 (or
the applicable currency equivalent). The assignment, publication, or dissemination of a rating by Fitch shall not constitute a consent by Fitch to use its name as an expert
in connection with any registration statement led under the United States securities laws, the Financial Services and Markets Act of 2000 of the United Kingdom, or the
securities laws of any particular jurisdiction. Due to the relative efciency of electronic publishing and distribution, Fitch research may be available to electronic subscribers
up to three days earlier than to print subscribers.