The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
EPR
FEDERAL RESERVE BANK OF NEW YORK
ECONOMIC POLICY REVIEW
The Long and Short
of It: The Post-Crisis
Corporate CDS Market
Nina Boyarchenko, Anna M. Costello,
and Or Shachar
Volume 26, Number 3
June 2020
The Long and Short
of It: The Post-Crisis
Corporate CDS Market
Nina Boyarchenko, Anna M. Costello, and Or Shachar
OVERVIEW
• Regulatory changes altered
the structure of the credit
default swap (CDS) market
following the 2007-09 financial
crisis, with regulatory-imposed
reporting requirements
increasing visibility into this
once opaque market.
• The authors use the resulting
granular supervisory data to
examine the CDS market and
present stylized facts on its
post-crisis evolution across
types of contracts, counterparties, and risk exposures.
They also study institutions’
choices on whether to participate in the four most common
CDS products.
• The study shows that dealers
became net buyers of credit
protection in the second half
of 2014, both by reducing the
amount of protection they sell
in the single-name market and
by switching to buying protection in the index market.
• The authors argue that considering simultaneous positions
in different types of credit derivatives is key to understanding
institutions’ participation in
these markets and how their
decisions affect prices.
Federal Reserve Bank of New York
T
he credit default swap (CDS) market, which became
notorious in the wake of the 2007–09 financial crisis, is
the third biggest over-the-counter (OTC) derivatives market
in the world, with $8 trillion notional value outstanding as
of June 2018 (BIS 2018). Because of the importance of this
market to the world financial system, sweeping regulatory
changes—meant to address fragilities uncovered during the
crisis—were implemented globally in the years following the
crisis. These new regulations changed the market’s structure
and also, through extensive data collection requirements,
allowed greater visibility into the previously opaque
bilateral OTC market. In this article, we exploit granular
supervisory data to study the properties of exposures
taken through the CDS market to corporate reference
entities in the United States and Europe, including which
institutions use these contracts, what kinds of exposures
they take, when they take them, and what factors influence
the prices of these exposures. To examine the CDS market,
we use supervisory position-level data from the CDS trade
repository maintained by the Depository Trust and Clearing
Corporation (DTCC). DTCC provides different data subsets
depending on the relevant supervisory authority’s purview.
Nina Boyarchenko is an officer and Or Shachar an economist at the Federal Reserve Bank
of New York. Anna M. Costello is an assistant professor at the University of Michigan's
Ross School of Business. Email: nina.boyarchenko@ny.frb.org; or.shachar@ny.frb.org;
amcost@umich.edu.
The views expressed in this article are those of the authors and do not necessarily reflect
the position of the Federal Reserve Bank of New York or the Federal Reserve System.
To view the authors’ disclosure statements, visit https://www.newyorkfed.org/research
/epr/2020/epr_2020_post-crisis-cds_boyarchenko.
Economic Policy Review 26, no. 3, June 2020
1
The Long and Short of It: The Post-Crisis Corporate CDS Market
As a prudential supervisor, the Federal Reserve is entitled to view positions and transactions
for which at least one counterparty is an institution it supervises or for which a supervised
institution is the reference entity. Each week, the Federal Reserve receives a weekly snapshot
showing all outstanding CDS positions that meet these criteria.
This data allows us to document properties of both existing and new positions, such as the
credit risk profile of the underlying entity, the maturity of the swap, locations of the party and
counterparty to the trade, and the type of credit derivative used. Thus, we can present stylized
facts on the CDS market’s post-crisis evolution across different types of contracts,
counterparties, and risk exposures. Unlike previous literature, we study jointly institutions’
choices on whether to participate in the four most common CDS products. We show that this
holistic view of exposures is necessary for understanding market changes made in response to
industry- and regulatory-led innovations.
We document four facts about the structure of the CDS market for U.S. and European
corporate credit derivatives. First, while dealers historically were protection sellers in the index
CDS market, they became net protection buyers in the second half of 2014. At the same time,
dealers have continued their historical pattern of selling protection in the single-name market
and buying protection in the index tranche and index options markets. Considering different
types of CDS products simultaneously is thus crucial to understanding institutions’ credit
derivative exposures.
Second, index options have replaced index tranches as the more prevalent levered derivative
product written on index contracts. Historically, institutions used levered products to get
exposure to a particular range—or “tranche”—of losses on a CDS index. The decline of the
collateralized debt obligation (CDO) market and the introduction of options on the index
have, however, led institutions to lever the entire index position.
Third, the maturity at inception of exposures taken through the CDS market has been
declining over time, with index CDS contracts trading almost exclusively in five-year
maturities at the end of our sample. Thus, not only has the gross notional of aggregate
CDS exposure declined since the financial crisis, but so too has exposure duration.
Fourth, most of the decline in single-name CDS gross notional outstanding since the
crisis has been in single-name contracts not eligible for voluntary clearing through a central
counterparty. Thus, the market for plain-vanilla CDS in the United States essentially migrated
wholly to central clearing even without the introduction of mandatory single-name
CDS clearing rules.
The article also serves as a primer on the overall structure of the CDS market in the
post-crisis regulatory environment, providing a summary of the characteristics of the most
commonly traded CDS contracts and the most salient features of the market’s evolution since
its inception in the early 1990s.
The 2010 Dodd-Frank Wall Street Reform and Consumer Protection Act introduced
multiple changes that affected how CDS contracts are traded in the United States, including
registration requirements for market participants, central clearing, and reporting of OTC
derivative positions1. A concern regarding our choice of data set is that the supervisory sample
selection might bias empirical findings. However, we compare coverage of positions data
collected for supervisory purposes in the United States over time with the full universe of
trades maintained by the DTCC.2 We find that the weekly supervisory snapshot of open
positions captures a large fraction of total market activity covered in the DTCC trade
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
2
The Long and Short of It: The Post-Crisis Corporate CDS Market
information warehouse (TIW). In particular, for a median week in our sample, supervisory
data capture over 70 percent of single-name contracts, over 60 percent of index contracts, and
over 85 percent of index tranche contracts3 in the TIW in terms of the number and the gross
notional of contracts outstanding.
The rest of the article is organized as follows. In Section 1, we describe the four credit
derivatives considered contracts in this article—single-name CDS, CDS index, index tranche,
and index options—and how recent regulatory changes have affected trading. Section 2 gives a
short overview of the supervisory version of the DTCC data, discussing the differences and
similarities with other proprietary data sets used in previous literature. We describe the
properties of existing and new positions in Section 3. Section 4 concludes.
1. Overview of the Credit Default Swap Market
A CDS is a bilateral agreement between a protection buyer and a protection seller in which the
buyer agrees to make fixed periodic payments to the seller in exchange for protection against
a credit event of an underlying asset or portfolio of assets. The underlying may be a single
reference entity (single-name CDS), a portfolio of reference entities (CDS index), or a
particular amount of losses in a basket of reference entities (tranche CDS). In this section
we review the definition of these contracts, including how they are priced and traded. We
also review the industry- and regulatory-led changes to the trading mechanisms for these
OTC derivatives.
1.1 Single-Name CDS Contracts
The single-name CDS contract insures the buyer of protection against a default of a single
issuer, such as a corporation, sovereign, or municipality. A credit event triggers a payment
from the protection seller to the protection buyer. To obtain this protection, the buyer makes
quarterly coupon payments to the seller until either default or contract expiry. The reference
obligations are often senior unsecured bonds. The ISDA Master Agreement, published by
the International Swaps and Derivatives Association, specifies contract terms and conditions
including the reference entity, the deliverable obligations, the contract tenor, the notional
principal, and the credit events covered by the contract. Standard credit events include
bankruptcy, failure to pay, obligation default, obligation acceleration, and repudiation or
moratorium. The CDS contract may also insure against debt restructuring, a credit event that
would not necessarily result in losses for the reference obligation holder.
In September 2014, credit event triggers were amended for new transactions on financial
and sovereign reference entities, as well as restructuring and bankruptcy credit events. The
changes included a government-initiated bail-in for CDS contracts on financial reference
entities; a split between senior and subordinated if a government intervention or restructuring
credit event occurs; and an “Asset Package Delivery” provision, under which existing bonds
that were deliverable before the bail-in will be deliverable to a post-bail-in auction to
determine the final auction price.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
3
The Long and Short of It: The Post-Crisis Corporate CDS Market
Before 2005, when a credit event occurred, CDS contracts were physically settled. The
protection buyer delivered the cheapest-to-deliver bond issued by the reference entity and, in
turn, received the bond’s face value. However, with the rapid growth of the CDS market, in
many cases, the volume of CDS outstanding far exceeded the volume of deliverable bonds, and
the market transitioned to cash settlement. An auction mechanism was introduced in 2005 to
determine the fair price of the defaulted reference entity. Creditex and Markit administer these
auctions and publish auction results online.4
In the auction, protection buyers and sellers settle on the net buy or sell CDS position,
reducing the amount of bond trading necessary to settle all contracts.5 The auction mechanism
determines the inside market midpoint for physical CDS contract settlement. The protection
seller then pays the difference between the par value and this auction-identified price per unit
of the contract notional to the protection buyer. Gupta and Sundaram (2015),
Chernov et al. (2013), and Du and Zhu (2017) study theoretically and empirically the auction
mechanism for determining settlement price.
Another change that affected single-name CDS contracts during our sample period is the
standard roll frequency. As of December 21, 2015, instead of rolling to a new on-the-run
single-name contract each quarter on the 20th of March, June, September, and December,
single-name contracts now only roll to new contracts in March and September. For example,
under the old convention, on June 2015, there was a move to a new five-year single-name
contract maturing on September 20, 2020. That five-year contract was considered on-the-run
for a three-month period. Under the new roll convention, a five-year single-name contract that
started trading on March 20, 2016, and was set to mature on June 20, 2021, was considered
on-the-run until September 20, 2016, when a new on-the-run five-year single-name contract
started trading. This change aligned single-name contracts with the roll frequency of
CDS index contracts, improving liquidity around the new semiannual roll dates.
1.2 CDS Index Contracts
A CDS index is a portfolio of single-name CDS. A protection buyer is insured against a default
of any constituent in the underlying portfolio. In return, the buyer makes quarterly coupon
payments to the protection seller. As with a single-name CDS, in case of default, the protection
seller pays par less recovery determined in the auction. Today, CDS indexes are the most
common instruments for assuming credit risk exposure. They are more liquid and trade at
smaller bid-ask spreads than baskets of cash bonds or single-name CDS contracts.
The most popular CDS index families are Markit CDX indexes, covering North American
and emerging markets, and International Index Company (IIC) iTraxx indexes, covering
Europe, Australia, Japan, and Asia excluding Japan. The CDX index family includes the
North American Investment Grade CDX index (CDX.NA.IG), the North American
High-Yield CDX Index (CDX.NA.HY), and the CDX Emerging Markets Index (CDX.EM).
The iTraxx index family includes the iTraxx Europe index and the iTraxx Crossover index. In
2018, the combined daily traded volume in the Markit CDX and Markit iTraxx indexes was
approximately $38 billion on average, representing 1,003 daily transactions, $5.6 trillion of
gross notional, and $906 billion of net notional outstanding.6
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
4
The Long and Short of It: The Post-Crisis Corporate CDS Market
For the credit exposures calculated in this article, we consider the three most popular
indexes and their sub-indexes: CDX.NA.IG, CDX.NA.HY, and iTraxx Europe. The
CDX.NA.IG index is a portfolio of 125 North American reference investment-grade
corporations, with $13 billion average traded volume and 226 average daily transactions in
2018. In addition to the aggregate investment grade index, there are also sector-specific
sub-indexes (consumer cyclical; energy; financial; industrial; and telecom, media, and
technology) and the CDX.NA.IG.HVOL sub-index, which includes reference entities with high
volatility. (As of 2018, the latter sub-index is no longer actively traded). The CDX.NA.HY
index comprises 100 North American corporations with a high-yield rating, with $6.2 billion
average traded volume and 276 average daily transactions in 2018. The CDX.NA.HY has been
divided into two rating sub-indexes: CDX.NA.HY.B and CDX.NA.HY.BB. The iTraxx Europe
index comprises 125 equally weighted investment-grade European reference entities. The
iTraxx Europe family includes three sector sub-indexes covering nonfinancial, financial senior,
and financial sub, and a HiVol index, comprising the 30 widest-spread nonfinancial names.
The iTraxx Crossover index consists of up to 75 sub-investment-grade European entities.
Unlike market value–weighted benchmark bond indexes, the CDS index constituents are
equal-weighted by notional, and provide the same default exposure as buying/selling CDS on
each underlying firm.
Importantly, although sectoral representation is taken into consideration in constructing an
index, larger banks and broker-dealers are excluded from CDX indexes. Historically, the
indexes were owned by the International Index Company Limited (iTraxx family) and
CDS IndexCo LLC (CDX family), which were themselves owned by a consortium of large
dealers. Including bank obligors in the indexes would have been a conflict of interest. In the
current market structure, a small set of large dealer participants still dominates transaction
volume. Therefore, as a seller of protection, a dealer would expose the buyer of protection on
an index to “wrong way” risk—that is, the risk that the seller of protection is exposed to the
same risk as the underlying —if that index were to include the dealer as a constituent. As a
buyer of protection, the dealer would be buying protection against its own default, raising
questions about contract legality. Therefore, excluding banks and broker-dealers as
constituents and including other financial firms in the index allows market participants to
gain credit exposure to a diversified basket of firms without introducing conflicts of interest.
The basket’s composition is determined when the index is rolled to the market. That
composition remains unchanged throughout its lifetime, unless a credit event is triggered for
one of the constituents, in which case that constituent is removed without replacement and
settled separately. The protection seller pays the loss on default to the protection buyer based
on trade notional and the weighting of the name. A new version of the index is published,
assigning a zero percent weight on the triggered entity. The contract continues to its full term
at a reduced notional amount, with the defaulted name removed from the portfolio.
Theoretically, the version of the index with the defaulted constituent should not be traded
after the default date. In practice, however, the version including the defaulted entity continues
to be traded until the recovery value is determined in an auction. This approach is in place
because dealers hedge their index derivative positions using the index and only when the
auction results are finalized can the characteristics of a derivative product on the new version
of the index be determined.7
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
5
The Long and Short of It: The Post-Crisis Corporate CDS Market
As time passes, the characteristics of the constituents might deviate from the index’s
desired profile. Therefore, a new index series is introduced twice a year, in March and
September, with extended maturity and updated constituents. This series is considered the
on-the-run series. Although trading continues in previous series, the liquidity of the
off-the-run series is lower than that of the on-the-run series. In the roll, entities that no
longer qualify for index inclusion are removed and new entities are added to keep the
number of reference entities in the index constant. The majority of names remain
unchanged.8 In particular, on average, 4 percent of CDX.NA.IG and 7 percent of CDX.
NA.HY constituents are replaced in each roll.9
The set of rules governing the constituents’ selection has evolved, tracking market
developments. The key change took place in March 2011, when DTCC TIW data were
utilized for the first time to determine the liquidity of potential constituents.10 In
September 2015, since the liquidity of single-name CDS had become a concern, the rules
governing the constituents of the CDX.NA.HY index family were updated to better match
the cash market counterpart. A criterion to avoid excess weighting of certain sectors was
added and criteria to avoid insufficiently liquid single-name CDS were tightened. Entities
that fail to satisfy the rating requirement because of an upgrade or downgrade, or that are
not sufficiently liquid, are replaced by the most-liquid entities with the necessary
credit rating.11
In theory, a CDS index should trade at its intrinsic value, which is approximately equal to
the duration-weighted average of the underlying single-name CDS expressed as a price value
in basis points. In practice, a CDX index’s market value is determined by supply and
demand, often resulting in a spread between its intrinsic and market values. Junge and
Trolle (2014) use this differential to construct a measure of CDS market illiquidity. They find
that CDS contracts with higher liquidity exposures have higher expected excess returns for
sellers of credit protection and trade with wider CDS spreads, with liquidity risk accounting
for 24 percent of CDS spreads on average.
1.3 Index Tranche CDS Contracts
It is also possible to assume a long or short credit exposure to a particular portion of the index
loss distribution by trading a CDS index tranche. An index tranche is defined by its attachment
(minimum level of losses) and detachment (maximum level of losses) points on the loss
distribution. For example, an equity tranche with attachment at 0 percent and detachment at
5 percent will absorb the index’s first 5 percent of losses. When a credit event is triggered, the
appropriate tranche is adjusted for the reduced notional (based on loss-given-default) and a
new detachment point is calculated for the remaining index names.
Exhibit 1 summarizes the relationship between the single-name contracts forming an index
and the tranche contracts on the index, using the CDX.NA.HY as an example. The CDX.NA.HY
includes 100 North American reference entities with high-yield ratings and has the following
tranches: equity, absorbing the first 10 percent of losses; junior mezzanine, absorbing the
next 5 percent; senior mezzanine, absorbing the next 10 percent; junior senior, absorbing the next
10 percent; and super senior, absorbing the last 65 percent. Consider an investor that buys
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
6
The Long and Short of It: The Post-Crisis Corporate CDS Market
Exhibit 1
Relationship between Single-Name CDS Contracts, Index Contracts,
and Index Tranches
CDS1
CDS2
..
.
CDX.NA.HY
Super Senior
35 − 100%
Junior Senior
25 − 35%
Senior
Mezzanine
Junior
Mezzanine
CDS100
Equity
15 − 25%
10 − 15%
0 − 10%
Detachment point – 15%
Attachment point – 10%
Notes: The exhibit illustrates the relationship between contract types using the CDX.NA.HY index as an
example. The CDX.NA.HY includes 100 North American reference entities with a high-yield rating.
protection on the equity tranche with a notional of $10 million. When a name in the index
defaults, with loss-given-default (LGD) set at 35 percent, the payout from the protection seller is
Payout = (Notional × LGD × Weighting) / Tranche Size
= ($10,000,000 × 0.35 × 0.01) / 0.1 = $350, 000.
The equity tranche is then adjusted for the reduced notional based on the 35 percent LGD,
and 9.65 percent of the notional remains in the tranche. The new detachment point must be
adjusted for the remaining names in the index. Using a factor of 0.99, the equity tranche for
new trades becomes a 0–9.9 percent tranche. The principal of other tranches is unaffected, but
they now have a smaller cushion protecting them against further losses.
A few papers have examined CDX tranche contract pricing. Coval et al. (2009) find that,
from the third quarter of 2004 to the third quarter of 2007, senior CDX tranches offered
too little compensation for their market risk exposure i compared with the compensation
investors were able to earn in the bond and option markets for bearing similar risks.
Collin-Dufresne et al. (2012) argue that senior index tranches provide the risk-neutral
probabilities of catastrophic risks in the economy. Seo and Wachter (2018) incorporate
investor preferences, consumption, and firm cash flows into a rare economic disaster model
to explain spreads of senior CDX tranches before and during the financial crisis.
1.4 Index Options
Credit default options (or credit default swaptions) give the buyer the option of entering into
a CDS contract at a future date. These options, similar in structure to more commonly
referenced interest-rate options, give investors a platform to take positions on volatility in
credit markets or tailor their directional spread views and credit exposure. Two types of
CDS index options trade: a “payer option” gives the holder the right, but not the obligation,
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
7
The Long and Short of It: The Post-Crisis Corporate CDS Market
to buy protection (pay coupons) on the underlying index at the specified strike spread level on
expiry (“European put”); a “receiver option” gives the holder the right, but not the obligation,
to sell protection (receive coupons) at the strike spread level (“European call”).
If a default happens among the index constituents before option expiry, the buyer of a payer
option (seller of a receiver option) can trigger a credit event by exercising the option. Since the
buyer of the payer option receives any losses resulting from default, payer options may be
exercised even if the index spread is below the option strike. The total payoff in a CDS index
option thus has two components: (1) payoff owing to the difference between the spread level
at expiry and the option strike, and (2) payoff stemming from any default losses. Although the
credit default options market has existed since 2003, these derivatives (CDS index options)
only gained widespread traction in 2011. Today, more than 60 percent of the options on the
CDX.NA.IG and the CDX.NA.HY are puts.
Since April 2009, single-name CDS, index CDS, and CDS index options have been traded
with fixed coupons and upfront payments that make the expected present value of the
protection bought equal to the expected present value of protection sold, conditional on the
fixed spread chosen and common assumptions about the recovery rate in a credit event.12 For
both single-name and index CDS, the fixed coupon payments from the protection buyer to the
seller are made quarterly using a 360-day year convention. 13
1.5 The Evolution of the U.S. CDS Market
Although the CDS market has existed since the early 1990s, the contract structure and trading
mechanisms were largely unchanged before the 2007–09 financial crisis. Exhibit 2 provides a
timeline of industry- and regulatory-initiated changes in the U.S. CDS market.
During and after the crisis, industry-led changes focused on revisions to the ISDA Master
Agreement—the document specifying CDS contract terms—aimed at creating greater
standardization and substitutability. Responding to operational inefficiencies and backlogs, the
Big-Bang and the Small-Bang Protocols were introduced in April and July 2009, respectively,
to eliminate redundant offsetting trades and facilitate centralized clearing.
The Big Bang encompassed four main changes: (1) an auction mechanism to determine the
recovery rate following a credit event; (2) Determinations Committees to decide whether a
credit or succession event has occurred; (3) a “looking back” period to determine the effective
protection period; and (4) a fixed coupon (either 100 or 500 basis points) for single-name
North American CDS and an upfront payment at the time of trade. The Small Bang applied
similar changes to European corporate and Western European sovereign CDS, introducing
fixed coupons (25, 100, 500, and 1000 basis points). The Big Bang Protocol also eliminated
restructuring as a credit event in new North American corporate single-name contracts, while
European corporate CDS continue to trade with “Modified Modified Restructuring” as the
standard convention. These steps toward standardization better aligned single-name
CDS contracts with the standard corporate CDS indexes.
Additional contractual change, noted in Section 1.1, came into effect in September 2014
after credit events during the financial crisis exposed flaws in the ISDA’s 2003 Definition.
The key changes ISDA 2014 introduced were (1) a new credit event that covers possible
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
8
The Long and Short of It: The Post-Crisis Corporate CDS Market
Exhibit 2
Timeline of CDS Market Evolution
Mar 2013
Mandatory clearing
CDX.NA
Apr 2009
Big Bang
Protocol
Jun 2003
ISDA 2003
Jul 2009
Small Bang
Protocol
Aug 2013
SEFs
Sep 2014
ISDA 2014
Feb 2014
Sep 2016
Mandatory SEF Mandatory
execution
initial margin
Apr 2013
Mandatory clearing
iTraxx
Notes: The exhibit presents a timeline of major changes affecting the CDS market in the United States. ISDA is
International Swaps and Derivatives Association. SEF is swap execution facilities.
government intervention, (2) deliverables in case of bank bail-ins, and (3) further clarification
on an obligation’s deliverability in a credit event (“Standard Reference Obligation”).
Dodd-Frank regulatory reforms also revamped the U.S. CDS market with the goal of
ameliorating the vulnerability of institutions linked by a complex web of OTC credit
derivatives. Title VII of the Dodd-Frank Act provided a comprehensive framework for
regulating the OTC swap markets, including CDS, to mitigate counterparty risk and to
improve pricing transparency. The bill introduced registration requirements for market
participants, central clearing of certain types of contracts, and reporting of OTC derivatives
transactions to swap data repositories (SDR).
Dodd-Frank divides regulatory authority over swap agreements between the Commodity
Futures Trading Commission (CFTC) and the Securities and Exchange Commission (SEC).
The CFTC has primary authority over swaps, except for “security-based swaps,” defined as
swaps on a single security, which the SEC regulates. The CFTC and the SEC share authority
over “mixed swaps,” which are security-based swaps that also have a commodity component.14
The CFTC has required reporting of swap transactions and pricing for index CDS contracts
since December 31, 2012.
The Dodd-Frank Act also requires mandatory clearing through a regulated central
counterparty (CCP) of all swap trades that the CFTC and the SEC determine should be
cleared. Each party in a CDS contract faces counterparty risk, that is, the possibility the other
party will not fulfill contractual obligations. In a counterparty default, the protection seller
risks the stream of coupon payments for the duration of the contract. The protection buyer
could potentially lose the full notional of the contract, assuming double default and a zero
recovery rate. A CCP reduces this risk by becoming the buyer to every protection seller and
the seller to every protection buyer. The only counterparty risk market participants face in a
cleared transaction is that of the CCP itself.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
9
The Long and Short of It: The Post-Crisis Corporate CDS Market
The clearinghouse is capitalized by its members, which are required to be regulated,
well-capitalized institutions. Each member contributes capital in proportion to its trading
activity. If clearinghouse capital falls below the required minimum level, remaining members
must put up additional capital to compensate for the shortfall. This protects market
participants from a default of an individual counterparty and spreads risk among all members.
CCPs also permit clearing of offsetting trades, since coupon payments, credit event settlement,
and collateral management are carried out through the CCP. These features make cleared
transactions—that is, those transactions that have a CCP as counterparty to the trade—more
attractive to market participants than uncleared transactions.
However, central clearing comes with a requirement to post margin. In the uncleared world,
a bilaterally negotiated ISDA Master Agreement governed collateral posting, which could vary
substantially according to counterparty size and credit rating. Anecdotally, in bilateral contracts,
dealers were rarely required to post initial margin. In contrast, in cleared transactions, the CCP
determines margin and collateral requirements, providing a more standardized approach.
Two forms of margin are required: initial and variation margin. The initial margin is set to
compensate for a scenario in which the counterparty defaults and fails to post the daily
variation margin. Initial margin is calculated at the portfolio level, with netting allowed for
offsetting cleared positions. The variation margin compensates for the trade’s daily
mark-to-market. Duffie et al. (2015) estimate the impact on collateral demand of these clearing
and margin requirements under various scenarios, such as increased novation of CDS to
CCPs, an increase in the number of clearing members, or proliferation of both specialized
and unspecialized CCPs.
Some single-name CDS and CDS indexes were already cleared voluntarily before
Dodd-Frank. As the regulator of the CDS index market, the CFTC called for phased-in
mandatory central clearing of most index trades for different types of market participants
in 2013. The clearing requirement applies to specific tenors and series of the CDX.NA.IG and
the CDX.NA.HY indexes: CDX.NA.IG 5Y, series 11; CDX.NA.IG 7Y, series 8; CDX.NA.IG 10Y,
series 8; CDX.NA.HY 5Y, series 11; and all subsequent series of these four indexes. At the time
of writing, the SEC had yet to finalize rules regarding clearing of single-name CDS, though
some contracts are centrally cleared voluntarily.
In addition, standardized swap trades have to be executed on swap execution facilities (SEFs).
U.S. rules governing these trades were finalized on May 16, 2013, and went into effect in
August 2013. For CDS, these include all index transactions in the CDX.NA.IG, CDX.NA.HY,
iTraxx Europe, and iTraxx Europe Crossover families. The rules also define the types of trading
platforms that must register as SEFs, the core principles by which they must operate, and the
execution method required to trade swaps. With the introduction of made-available-to-trade
(MAT) on SEFs in January 2014, the current on-the-run and first off-the-run series of the
five-year CDX.NA.IG, CDX.NA.HY, iTraxx Europe, and iTraxx Europe Crossover have been
required to trade on SEFs since February 2014.
Most recently, mandatory initial margins for noncleared positions were introduced
in the U.S. swap market in September 2016 and in the rest of the world in March 2017.15
These reforms increase bilateral trading costs by requiring dealers as well as customers
to post margin and by setting higher initial margin levels than in comparable cleared
contracts. The changes incentivize market participants to migrate to cleared trades for
clearing-eligible instruments.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
10
The Long and Short of It: The Post-Crisis Corporate CDS Market
2. Supervisory DTCC Data
Though CDS have been traded since 1994, the lack of detailed data on transactions and
positions before the 2007–09 financial crisis limited our ability to study the decisions made
by the wide range of participants that traded credit risk through these instruments. 16 Since the
crisis, detailed trade-level information has become available. In this section, we describe the
available data sets and what we can learn from them.
2.1 Review of Data Used in the Literature
Despite the CDS market’s inherent decentralized nature, the DTCC has been collecting
transaction information through its widely used lifecycle event processing service Deriv/SERV.
The DTCC estimates this service covers approximately 98 percent of all standard credit
derivatives contracts.17 Following the financial crisis reforms, the DTCC began using these
data in two ways. First, it publishes weekly statistics on CDS volume and activity through the
TIW. Since November 2008, these statistics have included notional outstanding by participant
type (dealer, nondealer, central counterparty), product type (single-name, indexes, and index
tranches), term, and currency. Oehmke and Zawadowski (2016) exploit a subset of these
data—total net notional amount of CDS protection written on the top 1,000 single-name
reference entities—to investigate participant trading objectives.
Second, the DTCC provides global regulators with transaction- and position-level data,
giving supervisory authorities a more granular view of the market. The transaction-level data
include new trades, assignments (novations), and terminations. For each record, the DTCC
data contain the names of the protection buyer and seller, submitter of the transaction to the
DTCC, reference entity, trade date, termination date, notional amount, and currency. The
DTCC distributes different subsets of its worldwide data set in different jurisdictions,
supporting relevant authorities in regulating and supervising OTC derivatives markets.18
Chen et al. (2011) examine a three-month sample of global single-name (corporate,
sovereign, municipal, asset-backed, and loan) and index CDS transactions to evaluate the
market’s size and composition, trading frequency, and the level of standardization of
CDS products before post-trade public reporting began in the CDS market. Their sample
comprises all CDS transactions occurring globally between May 1 and July 31, 2010, in which
at least one of the fourteen major over-the-counter derivatives dealers was a counterparty to
the trade. Shachar (2013) analyzes transactions in single-name CDS contracts on thirty-five
financial firms, as well as transactions in CDX Index contracts. The sample includes all
CDS transactions occurring between February 2007 and June 2009 regardless of counterparty
region. Counterparty identity is masked, but counterparty type is shown.
Using these data, Shachar (2013) shows that bilateral exposures in the interdealer market
are empirically relevant in determining counterparty risk, dealer intermediation capacity, and
market resilience in times of stress. Applying a similar methodology, Gehde-Trapp et al. (2015) use
single-name CDS with German firms as reference entities from January 2009 to June 2011.
They show that CDS premiums reflect market frictions rather than the credit risk of the
underlying reference entity. Du et al. (2015) observe CDS transactions from January 2010
through December 2013 in which at least one of the dealer banks regulated by the Federal
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
11
The Long and Short of It: The Post-Crisis Corporate CDS Market
Reserve Board (FRB) is a counterparty to the trade or the reference entity. The FRB-regulated
institutions are Bank of America Corporation, Citigroup Inc., Goldman Sachs Group, Inc.,
JP Morgan Chase & Co., and Morgan Stanley. The authors focus on how market participants
price and manage counterparty risk. Siriwardane (forthcoming) uses a granular DTCC data set
that identifies counterparties, terms of trade, and covers nearly all outstanding CDS exposures
for transactions that reference North American entities and/or U.S. participants starting in
2010. Using the same data set, Eisfeldt et al. (2018) study the extent to which dealers exert
pricing power in the index CDS market. They find that credit spreads in dealer-to-dealer
trades are 6 percent lower than credit spreads in dealer-to-nondealer trades.
A few papers exploit nonsupervisory versions of the DTCC data or other data sources.
Duffie et al. (2015) obtain a version of the DTCC data that encompasses gross and net bilateral
exposures between any two counterparties for 184 single-name CDS (9 G20 sovereigns,
20 European sovereigns, and 155 global financial entities), with no restriction on counterparty
origin. Their data set does not identify counterparties, trade dates, or position maturity.
Loon and Zhong (2016) use publicly disseminated Index CDS transactions. As noted in
Section 1, since December 31, 2012, index CDS transactions have had to be reported to an
SDR, which in turn publicly disseminates transaction details, including price, size, and time.
Loon and Zhong collected CDS Index transactions executed between December 31, 2012, and
December 31, 2013, from the DTCC data repository. They merge these transactions with
intraday and end-of-day quotes to calculate the transaction-level relative effective spread and
other liquidity measures. Tang and Yan (2017) use transactions data from the GFI Group from
January 1, 2002, to April 30, 2009. They argue that CDS spreads not only change in response to
fundamentals, but also in response to supply–demand imbalances and market liquidity.
Arora et al. (2012) use a proprietary data set from one of the largest fixed-income asset
management firms, which contains both actual CDS transaction prices for contracts entered
into by the firm as well as actionable quotations obtained from a variety of CDS dealers. Their
data extend from March 2008 to January 2009.
More recently, SEF transaction data have become available.Collin-Dufresne et al. (2017) collect
transaction data for multiple dealer-to-dealer and dealer-to-nondealer SEFs from
October 2, 2013, to October 16, 2015, and find that average transaction costs are higher for
dealer-to-client trades. Using May 2016 message-level data for two dealer-to-nondealer SEFs,
Riggs et al. (2017) find that customers contact fewer dealers if the trade size is larger or
nonstandard, while dealers are more likely to respond to customer inquiries if fewer dealers
are competing, if the notional size is larger, or if more dealers are making markets.
Although these papers examine different parts of the network of CDS exposures, a
consistent picture emerges. First, credit exposures fluctuate over time. Net credit protection
sellers in one period can become net protection buyers in another. Second, the pricing of the
exposure traded depends on the counterparties to the trade—though the literature has yet to
arrive at a consensus on whether and to what extent dealers are able to exert market power—as
well as the effect of mandatory clearing on the pricing strategies of other market participants.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
12
The Long and Short of It: The Post-Crisis Corporate CDS Market
2.2 Our Data
Our version of the DTCC data is obtained through the Federal Reserve System’s supervisory
authority. Each weekly snapshot reports all outstanding CDS positions in which at least one
FRB-regulated dealer bank is a counterparty to the trade or the reference entity itself. We refer
to this subset of all CDS trades collected by the DTCC as “supervisory DTCC.”
For the positions the DTCC reports to the Federal Reserve, we observe detailed contractual
terms, including the identities of the counterparties; pricing terms, including the fixed spread
and the upfront payment; notional amount of the contract; trade date; maturity; and the
restructuring clause. The sample period is January 2010 through June 2019, representing
497 observation weeks, 12,971,160 unique contracts, 3,427 unique reference entities,19
13,474 unique protection buyers, and 12,555 unique protection sellers. In a median week,
22,212 new positions are opened, corresponding to 829 unique reference entities, with
exposures exchanged between 752 unique buyers and 766 unique sellers. Although these
data cover only a subset of all CDS transactions and thus have inherent limitations, the
six institutions for which we observe all open positions on a report date are major market
participants and their trades cover a large fraction of overall activity.
Overall activity
Summary statistics on the number of contracts and gross notional amounts for different
subsamples of the supervisory DTCC data are presented in Charts 1 and 2. The top two panels
of Chart 1 present overall activity in the market for single-name, index, and index tranche
CDS instruments—specifically, the distribution of gross notional amounts outstanding in
U.S. dollar billion equivalents and the number of contracts in thousands for the DTCC TIW.
The bottom two panels show the fraction of the total reported in the supervisory DTCC data.
Comparing the positions observed in our data to total market activity captured by the DTCC
TIW, the positions of the six supervised institutions account on average for 70 percent of total
activity in single-name derivatives, 58 percent in index products, and 91 percent in index tranche
products, as measured by the number of contracts and gross notionals. Aggregating the three
types of contracts, in an average reporting week, the supervisory data capture 60 percent of the
gross notional outstanding and 62 percent of the number of contracts.
Over our sample period, both the gross notional and the number of contracts outstanding
for single-name contracts have declined steadily, driving an overall decrease in gross notional
and contracts outstanding for the market. This significant decline is partly attributable to
“compression,” when redundant contracts on the same reference entity are terminated and
replaced with new ones with the same net exposure. The gross notional outstanding for index
and tranche index contracts have also declined somewhat since 2010, though not to the same
extent as those of single-name contracts. Interestingly, while the number of single-name
contracts traded is much larger than the number of index trades, the gross notionals are
comparable. That is, while contracts using single-name reference entities are more frequent,
the notional amounts of contracts written on single-name reference entities tend to be much
smaller than those of index contracts.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
13
The Long and Short of It: The Post-Crisis Corporate CDS Market
Chart 1
Worldwide CDS Positions and Coverage in Supervisory DTCC Data
All products
Gross notional, (billions of U.S. dollars)
30,000
Gross Notional in TIW
Index
Index tranche
Contracts (in thousands)
2,500
Single-name
Contracts in TIW
Contracts in TIW
2,000
20,000
1,500
1,000
10,000
500
0
0
Percent of reported in TIW
100
Percent of reported in TIW
Gross Notional Coverage
100
80
80
60
60
40
40
20
20
0
2010
2012
2014
2016
2018
0
2010
Contracts Coverage
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: Number of contracts is reported in thousands. Notionals are given in billions of U.S. dollars. Coverage
is reported as the ratio between the corresponding quantities in the supervisory DTCC data and the TIW data
(in percentage terms).
It is not surprising that the supervisory DTCC sample has the lowest coverage for index
trades because the index provides more-diversified credit risk exposure. In addition, index
product trading was standardized before the start of our sample and, therefore, the identity
of the protection seller is less detrimental to the value of the contract. Moreover, index
CDS trades frequently occur with a CCP as a party.
Trading by type of counterparty
In our sample, 7,493,715 contracts are exchanged between dealers, 1,968,379 between dealers
and their customers, and 2,858,924 between dealers and a CCP. In a median week,
9,892 contracts are exchanged between dealers, 3,657 between dealers and their customers,
and 5,417 between dealers and a CCP. Chart 2 shows the average number of contracts and
gross notionals exchanged between different types of market participants by product category.
We use the DTCC classification to designate institutions as dealers, nondealers (customers),
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
14
The Long and Short of It: The Post-Crisis Corporate CDS Market
Chart 2
Comparison of Supervisory DTCC and TIW Data by Participant Type
Dealer-to-dealer
Percent of reported in TIW
100
Single Name
80
Dealer-to-nondealer
Dealer-to-CCP
Percent of reported in TIW
100
Single Name
80
60
60
40
40
20
20
0
Percent of reported in TIW
100
Index
80
Percent of reported in TIW
100
Index
80
60
60
40
40
20
20
0
0
Percent of reported in TIW
100
Index Tranche
80
Percent of reported in TIW
100
Index Tranche
80
60
60
40
40
20
20
0
2010
2012
2014
2016
2018
0
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: Coverage is reported as the ratio (in percentage terms) between gross notional in the supervisory DTCC data
and gross notional in the TIW data (left column), and the ratio between the number of contracts in the supervisory
DTCC data and the number of contracts in the TIW data (right column).
and CCPs. Overall, the supervisory DTCC data cover 83 percent of dealer-to-dealer and
58 percent of dealer-to-customer contracts, and 82 percent and 55 percent of gross notional,
respectively. For dealer-to-CCP trades, the supervisory data cover approximately 48 percent
of contracts and 55 percent of the gross notional.20
This relatively low coverage of trades between CCPs and dealers explains the relatively
low coverage of index trades and the relatively high coverage of index tranche trades, since
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
15
The Long and Short of It: The Post-Crisis Corporate CDS Market
index tranche positions cannot be traded with a CCP. The middle row of Chart 2 shows
that, while the supervisory data capture around 75 percent of the number of
dealer-to-dealer contracts, they capture only slightly more than a quarter of index
contracts exchanged between a dealer and a CCP.21 Combined with the relatively good
coverage of transactions between dealers and nondealers in the single-name market, this
suggests that nondealer participants prefer index contracts for taking credit risk exposure.
3. Aggregate Market Activity
3.1 Metrics of Activity
We examine weekly financial institution activity in each CDS market segment by computing
each participant’s buy- and sell-side positions in single-name, index, tranche, and option
products, as well as the corresponding net positions. The net position is equal to the difference
between the buy and sell positions for each underlying. A positive position indicates that an
institution is, on net, buying protection. Formally, participant p’s position in contracts on
reference entity i with maturity τ at snapshot date h is the sum of notionals in contracts in
which p buys protection less the sum of notionals in contracts in which p sells protection:
Net Positionp,i,τ,t = Notional boughtp,i,τ,t − Notional soldp,i,τ,t .
We construct two measures of market activity: gross and net notional. Gross notional sums
the par amount of credit protection bought (or, equivalently, sold) in all the contracts. Net
notional sums the net positions of all the participants that are, on net, buying or selling
protection. These two concepts of notional for reference entity i on date t are defined as
Gross notionalit =
boughtp,i,τ,t
∑ Notional
it
p,τ
Net notionalit = ∑
p
(∑ Net position τ (1 (∑ Net position
τ
p,i, ,t
τ
p,i,τ,t > 0
(.
Net notional positions represent the maximum possible net transfer between protection sellers
and protection buyers in a reference entity credit event. When the recovery rate in the credit
event is above 0, the funds transferred are a fraction (equal to one less the recovery rate) of the
net notional. Gross notional measures total transaction volume in the CDS market. An
important caveat concerning these measures is that they are based on contract face value, and
do not reflect the market value or the duration of the contracts. Thus, neither gross nor net
notional captures market participants’ exposure to credit events. In Appendix 2, we show that,
overall, the qualitative properties of the duration-risk-adjusted positions are similar to those of
the raw positions. We thus focus on the unadjusted gross and net notionals in the main body
of this article.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
16
The Long and Short of It: The Post-Crisis Corporate CDS Market
Table 1
Sample Summary Statistics
Sample
RedCodes
Contracts
Buyers
Sellers
Original
4,049
14,641,667
13,997
12,934
...and 2010-2019 (June)
3,427
12,971,160
13,474
12,555
...and SN, Index Product Corporate
2,801
12,945,839
13,465
12,543
...and US/EUR Corporate
1,773
10,353,448
11,129
10,584
...and 5Y Maturity
1,351
3,673,703
10,264
9,521
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information
Warehouse (TIW); Markit.
Notes: The first row of the table reports the characteristics of the full sample; subsequent rows
describe how the data set changes with the specified filter applied. The “RedCodes” column
presents the number of unique Markit RedIDs in the sample; “Buyers” (“Sellers”) columns are
the number of unique firm organization ID numbers in the sample that bought (sold) protection.
3.2 Sample Selection
We focus on single name, index, index tranche, or index option contracts entered into between
January 2010 and June 2019 covering U.S. and advanced European corporate reference entities.
Table 1 summarizes how the original sample changes with each filter applied. Appendix 1
describes the details of the overall sample construction and splits by characteristics used below.
3.3 Characteristics of Outstanding Positions
We examine the characteristics of all positions outstanding on a given snapshot date. The
outstanding positions provide an overview of total exposures at a point in time and illustrate
longer-term trends by capturing gradual exposure changes. Charts 3–7 plot monthly average
gross and net notional of outstanding positions by counterparty type, counterparty location,
on-the-run index membership, clearing eligibility, and master agreement type.
Positions by counterparty type
The gross and net notional exposures in single-name, index, index tranche, and index option
contracts of dealers, nondealer customers (buyside counterparties), and CCPs are plotted in
Chart 3. The bottom two rows show that, while index options are a relatively new contract
type, they have grown more prominent, replacing index tranches as the preferred levered
position on the index. In both index tranches and options, dealers have historically been
protection buyers from buyside counterparties. The only exception is from January 2012 to
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
17
The Long and Short of It: The Post-Crisis Corporate CDS Market
Chart 3
Monthly Average Gross and Net Positions by Participant Type
Dealer
Gross notional (billions of U.S. dollars)
20,000
Single Name
50
10,000
0
5,000
-50
0
-100
8,000
Index
CCP
Net notional (billions of U.S. dollars)
100
Single Name
15,000
Gross notional (billions of U.S. dollars)
10,000
Buyside
Net notional (billions of U.S. dollars)
100
Index
50
6,000
0
4,000
-50
2,000
0
-100
Gross notional (billions of U.S. dollars)
5,000
Index Tranches
4,000
3,000
Net notional (billions of U.S. dollars)
50
Index Tranches
0
2,000
1,000
-50
0
Gross notional (billions of U.S. dollars)
1,500
Index Options
Net notional (in billions of U.S. dollars)
100
Index Options
50
1,000
0
500
-50
-100
0
2010
2012
2014
2016
2018
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the monthly average gross notional of outstanding positions in the four contract types
by participant type. The right column shows the monthly average net notional of positions in the four contract types
by participant type. Notionals are measured in U.S. dollar billion equivalents; positive net notional indicates net
buying of protection. Participant classification is provided by DTCC.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
18
The Long and Short of It: The Post-Crisis Corporate CDS Market
mid-2013 when dealers were selling protection to buyside counterparties in the index tranche
market. Comparing gross and net exposures shows that dealers also exchange a large fraction
of the total index tranche gross notional with one another.
The top row of Chart 3 plots single-name exposures, showing that dealers again exchange a
large volume of gross notional with one another. In contrast with levered products, dealers are
net protection sellers throughout our sample while buyside counterparties are net protection
buyers. Despite an increase over time in single-name contracts eligible for clearing, CCPs do not
appear to have raised their gross exposure to single-name contracts and they oscillate between
net protection seller and buyer positions. Overall, both gross and net exposures in the
single-name market have fallen over time, consistent with the global trend illustrated in Chart 1.
In contrast, while we see some decreases in index contract gross notional before
January 2014 (plotted in the second row in Chart 3), the net notional traded has remained
mostly constant throughout our sample period. While dealers were primarily net protection
sellers in the index CDS market until mid-2014, they have become net protection buyers,
primarily from buyside counterparties and secondarily from CCPs.
Positions by counterparty location
Credit exposures of the four CDS products by jurisdiction are plotted in Chart 4. The largest
fraction of gross notional is traded between counterparties in the same jurisdiction, while
exposure to counterparties outside the United States and advanced economy European
countries is negligible. U.S. counterparties are primarily net protection buyers in the index
tranche market, net protection sellers in the index and index option markets, and oscillate
between net buying and net selling in the single-name market. In the index and index tranche
markets, their primary counterparties are in Europe. While European counterparties also
represented a sizable fraction of net notional traded in the single-name market before
January 2016, more recently, institutions outside the United States, Europe, and Japan have
been net protection buyers from U.S. and Europe-based counterparties. In the index option
market, these “rest-of-the-world” institutions have been the principal protection buyers since
January 2013, with European counterparties switching between net protection buying and
selling in that market.
On-the-run versus off-the-run positions
Membership of outstanding positions in an on-the-run index series is plotted in Chart 5. A
series is considered on-the-run if it is the latest series and version of an index on a given date.
For single-name reference entities, we consider four cases: the entity belongs only to an index’s
on-the-run series, only to an off-the-run series, to both an index’s on-the-run and off-the-run
series, or it does not belong to an index. In both gross and net notional terms, the majority of
single-name exposure is traded in contracts that belong to both on-the-run and off-the-run
series. Little exposure is in the on-the-run-only category, reflecting the slow pace of index
membership replacement. Single-names that do not belong to an index also represent a sizable
fraction of gross and net notional traded, though their prominence has declined.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
19
The Long and Short of It: The Post-Crisis Corporate CDS Market
Chart 4
Monthly Average Gross and Net Positions by Location of Counterparty
Home
Gross notional (billions of U.S. dollars)
20,000
United States
Single Name
Advanced European
20
10,000
0
5,000
-20
0
-40
Index
8,000
Rest of world
Net notional (billions of U.S. dollars)
40
Single Name
15,000
Gross notional (billions of U.S. dollars)
10,000
Japan
Net notional (billions of U.S. dollars)
200
Index
100
6,000
0
4,000
-100
2,000
0
-200
Gross notional (billions of U.S. dollars)
5,000
Index Tranches
4,000
3,000
Net notional (billions of U.S. dollars)
100
Index Tranches
50
0
2,000
-50
1,000
-100
0
Gross notional (billions of U.S. dollars)
1,500
Index Options
Net notional (billions of U.S. dollars)
40
Index Options
20
1,000
0
500
0
2010
-20
2012
2014
2016
2018
-40
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the monthly average gross notional of outstanding positions in the four contract types
by location of the counterparty to the contract. The right column shows the monthly average net notional of new
positions in the four contract types by location of the counterparty to the contract. “Home” refers to counterparties
domiciled in the same jurisdiction. Notionals are measured in U.S. dollar billion equivalents; positive net notional
indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
20
The Long and Short of It: The Post-Crisis Corporate CDS Market
Chart 5
Monthly Average Gross and Net Positions by Type of Index
On the run
Gross notional (billions of U.S. dollars)
10,000
8,000
Single Name
Neither
Net notional (billions of U.S. dollars)
800
Single Name
400
Singlename
Singlename
200
2,000
0
2010
Both
600
6,000
4,000
Off the run
2012
2014
2016
Gross notional (billions of U.S. dollars)
5,000
2018
Index
4,000
0
2010
2012
2014
2016
2018
Net notional (billions of U.S. dollars)
600
Index
400
3,000
2,000
200
1,000
0
2010
2012
2014
2016
2018
Gross notional (billions of U.S. dollars)
2,000
Index Tranches
1,500
0
2010
2012
2014
2016
2018
Net notional (billions of U.S. dollars)
300
Index Tranches
200
1,000
100
500
0
2010
2012
2014
2016
2018
Gross notional (billions of U.S. dollars)
800
Index Options
0
2010
600
150
400
100
200
50
0
2010
2012
2014
2016
2018
2012
2014
2016
2018
Net notional (billions of U.S. dollars)
200
Index Options
0
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the monthly average gross notional of outstanding positions in the four contract types
by type of index. The right column shows the monthly average net notional of new positions by clearing eligibility. For
index tranches and index options, “on-the-run” corresponds to those written on on-the-run indexes. Notionals are
measured in U.S. dollar billion equivalents; positive net notional indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
21
The Long and Short of It: The Post-Crisis Corporate CDS Market
In the index market, the on-the-run indexes represent about a fifth of the gross notional
outstanding but nearly half the net notional bought. In contrast, index tranche exposures are almost
exclusively written on off-the-run series, while index option exposures are primarily written on
on-the-run series. This is because index options have maturities of less than one year so that the
option market can adapt quickly to the introduction of new index series (see Chart 9, p. 27).
Positions by clearing eligibility
Outstanding positions have shifted in response to two market changes: the introduction of
central clearing and the adoption of the ISDA 2014 Master Agreement. Chart 6 plots the gross
and net notional traded according to clearing eligibility—that is, whether a contract is eligible
for clearing on a CCP, not whether the contract was actually cleared. For index tranches and
options, this means that the contract is written on an index eligible for clearing.
In the single-name market, as more reference entities have become eligible for clearing,
both the gross and net notional of clearing-eligible contracts have increased. The drop in the
overall gross notional of single-name contracts traded has been driven by the segment of the
market not eligible for clearing. With the introduction of mandatory central clearing for index
contracts in the United States in March 2013, both gross and net notional outstanding in
index, index tranche, and index option contracts migrated quickly to clearing-eligible series.
Overall, Chart 6 suggests that, as more single-name reference entities are accepted for CCP
clearing, single-name activity will increasingly be conducted in clearing-eligible contracts,
even without mandatory clearing rules.
Positions by master agreement type
The gross and net notional outstanding by ISDA Master Agreement are plotted in Chart 7.
As with the introduction of mandatory clearing of index products, outstanding positions in
single-name, index, and index option products transitioned quickly to the ISDA 2014 Master
Agreement. Contracts with the new Master Agreement represented most gross and net
notional outstanding within six months of the agreement going live in October 2014. However,
the index tranche market responded more sluggishly, with ISDA 2003 contracts representing at
least 50 percent of gross notional outstanding as late as January 2017.
3.4 Characteristics of New Positions
New positions provide an overview of the risks market participants trade each week,
capturing faster-changing features of the market than outstanding positions do. To examine
the characteristics of new positions, we identify the first occurrence of a contract for which
the effective date is reported to be after the trade date. We determine net and gross notional
of new positions according to the credit rating of the underlying reference entity, initial
contract maturity, reference entity sector, and contract currency.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
22
The Long and Short of It: The Post-Crisis Corporate CDS Market
Chart 6
Monthly Average Gross and Net Positions by Clearing Eligibility
Eligible
Gross notional (billions of U.S. dollars)
10,000
Single Name
8,000
Net notional (in billions of U.S. dollars)
1,000
Single Name
800
6,000
600
4,000
400
2,000
200
0
0
Gross notional (billions of U.S. dollars)
5,000
4,000
Index
Net notional (billions of U.S. dollars)
800
Index
600
3,000
400
2,000
200
1,000
0
0
Gross notional (billions of U.S. dollars)
2,500
Index Tranches
2,000
1,500
Net notional (billions of U.S. dollars)
400
Index Tranches
300
200
1,000
100
500
0
0
Gross notional (billions of U.S. dollars)
800
Index Options
Net notional (billions of U.S. dollars)
200
600
150
400
100
200
50
0
2010
Non-eligible
2012
2014
2016
2018
0
2010
2012
2014
Index Options
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the monthly average gross notional of outstanding positions in the four contract types
by clearing eligibility, determined as described in Appendix 1. The right column shows the monthly average net
notional of new positions by clearing eligibility. For index tranches and index options, “eligible” corresponds to those
written on clearing-eligible indexes. Notionals are measured in U.S. dollar billion equivalents; positive net notional
indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
23
The Long and Short of It: The Post-Crisis Corporate CDS Market
Chart 7
Monthly Average Gross and Net Positions by ISDA Master Agreement
ISDA 2003
Gross notional (billions of U.S. dollars)
10,000
Single Name
8,000
6,000
ISDA 2014
Net notional (billions of U.S. dollars)
800
Single Name
600
400
4,000
200
2,000
0
0
Gross notional (billions of U.S. dollars)
5,000
Index
4,000
Net notional (billions of U.S. dollars)
600
Index
400
3,000
2,000
200
1,000
0
0
Gross notional (billions of U.S. dollars)
2,500
Index Tranches
2,000
Net notional (billions of U.S. dollars)
300
Index Tranches
200
1,500
1,000
100
500
0
0
Gross notional (billions of U.S. dollars)
800
Index Options
Net notional (billions of U.S. dollars)
200
Index Options
600
150
400
100
200
50
0
2010
2012
2014
2016
2018
0
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the monthly average gross notional of outstanding positions in the four contract types
by ISDA master agreement. The right column shows the monthly average net notional of new positions in the four
contract types by clearing eligibility. Notionals are measured in U.S. dollar billion equivalents; positive net notional
indicates net buying of protection. ISDA is International Swaps and Derivatives Association.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
24
The Long and Short of It: The Post-Crisis Corporate CDS Market
Trading by credit rating
The distribution of credit ratings of the underlying reference entities is plotted in Chart 8.
For both single-name and index products, the majority of new gross notional is investment
grade. iTraxx Europe Crossover, the unrated category for index products, represents
around 10 percent of new gross notional traded, and, interestingly, only includes index
options, not index tranches. More generally, index tranche net notional is almost exclusively
traded in the investment grade category, though tranche contracts written on high-yield
indexes represent around 20 percent of new gross notional traded in the second half of our
sample period.
For single-name reference entities, we break down the investment grade rating categories
into AAA, (AA,A) and BBB. The top row of Chart 8 shows that around 40 percent of new gross
notional is BBB, around 20 percent (AA,A), and a negligible amount is AAA. Contracts on
BBB-rated reference entities also represent most new net notional exchanged. This ranking of
CDS traded exposures according to reference entity credit rating corresponds to the relative
amounts of debt outstanding in these categories. Most U.S. debt issuance since the financial
crisis has been BBB, with only two AAA issuers remaining.22
Trading by maturity
The distribution of new gross and net notional traded by initial contract maturity is plotted in
Chart 9. Three striking features emerge. First, as noted earlier, index options have exclusively
short maturities, probably reflecting the index contract’s semiannual roll. In unreported results,
we find that an overwhelming majority of index option contracts have initial maturities of less
than six months. Second, while index and index tranche contracts are increasingly traded in
five-year maturities, the distribution of initial maturities in the single-name market is less
concentrated: 20–30 percent of gross notional is traded in the five-year category, around
60 percent in the one-to-five-year category, and a further 10–20 percent in the
one-year-or-less maturity. Thus, a substantial fraction of single-name transactions have
maturities of less than five years. Third, although index and index tranche contracts historically
had as much as 15 percent of index-contract and 50 percent of index-tranche-contract gross
notional traded in maturities of ten years and greater, such longer maturities were never
prevalent in the single-name market. Thus, while market participants were willing to trade
long-term exposures in index contracts, the single-name market has always been concentrated
in intermediate maturities.
Trading by reference entity sector
Exposure by reference entity industry is plotted in Chart 10. Around 25 percent of new gross
notional traded is written on manufacturers and around 20 percent on financial institutions,
the two most prevalent industries. Contracts on energy companies have increased in share
since January 2015, but still represent a negligible fraction of net notional traded.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
25
The Long and Short of It: The Post-Crisis Corporate CDS Market
Chart 8
Monthly Average Gross and Net Positions by Credit Rating
Single-name:
Index:
Both:
Percent new gross notional
100
80
AAA
Investment grade
Unrated
Single Name
(AA,A)
High-yield
BBB
BB
Net notional (billions of U.S. dollars)
500
400
60
300
40
200
20
100
Below BB
Single Name
0
Percent new gross notional
100
Index
80
Net notional (billions of U.S. dollars)
1,500
Index
1,000
60
40
500
20
0
0
Percent new gross notional
100
80
Index Tranches
Net notional (billions of U.S. dollars)
150
Index Tranches
100
60
40
50
20
0
0
Percent new gross notional
100
80
Index Options
Net notional (billions of U.S. dollars)
250
200
60
150
40
100
20
50
0
2010
2012
2014
2016
2018
0
2010
2012
2014
Index Options
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW);
Markit (credit rating information).
Notes: The left column shows the percentage of monthly average gross notional of new positions in the four contract
types by credit rating category. The right column shows the monthly average net notional of new positions in the four
contract types by credit rating category. Notionals are measured in U.S. dollar billion equivalents; positive net notional
indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
26
The Long and Short of It: The Post-Crisis Corporate CDS Market
Chart 9
Monthly Average Gross and Net Positions by Initial Maturity
(0,1 years]
Percent new gross notional
100
80
(1-5 years)
Single Name
5 years
(5-10 years)
>10 years
Net notional (billions of U.S. dollars)
600
Single Name
400
60
40
200
20
0
0
Percent new gross notional
100
Index
80
Net notional (billions of U.S. dollars)
2,000
Index
1,500
60
1,000
40
500
20
0
0
Percent new gross notional
100
80
Index Tranches
Net notional (billions of U.S. dollars)
150
Index Tranches
100
60
40
50
20
0
0
Percent new gross notional
100
80
Index Options
Net notional (billions of U.S. dollars)
250
Index Options
200
60
150
40
100
20
50
0
2010
2012
2014
2016
2018
0
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the percentage of monthly average gross notional of new positions in the four contract
types across initial maturity buckets. The right column shows the monthly average net notional of new positions
across initial maturity buckets. For index options, maturity is the maturity of the option, not the underlying index.
Notionals are measured in U.S. dollar billion equivalents; positive net notional indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
27
The Long and Short of It: The Post-Crisis Corporate CDS Market
Chart 10
Monthly Average Gross and Net Positions by Sector
Manufacturing
Percent new gross notional
100
Energy
Utilities/Telecommunications
Gross Notional
80
Financial services
Net notional (billions of U.S. dollars)
400
Other
Net Notional
300
60
200
40
100
20
0
2010
2012
2014
2016
2018
0
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW);
Markit (reference entity sector information).
Notes: The left column shows the percentage of monthly average gross notional of new positions in single-name
contracts by reference entity sector. The right column shows the monthly average net notional of new positions in
single-name contracts by reference entity sector. Notionals are measured in U.S. dollar billion equivalents; positive
net notional indicates net buying of protection.
Trading by currency
Both new gross and net notionals in contracts on U.S. and European reference entities
are split between contracts denominated in U.S. dollars and contracts denominated in
euros, with very little notional exchanged in other currencies, as shown in Chart 11.
Euro-denominated index contracts have slowly become more common, rising from
around 30 percent of new gross notional traded in January 2010 to around 50 percent
in June 2019. In contrast, the single-name market remains primarily denominated
in U.S. dollars, creating a potential currency mismatch between the single-name and
index markets.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
28
The Long and Short of It: The Post-Crisis Corporate CDS Market
Chart 11
Monthly Average Gross and Net Positions by Contract Currency
U.S dollar
Percent new gross notional
100
Single Name
80
Net notional (billions of U.S. dollars)
500
400
60
300
40
200
20
100
0
0
Percent new gross notional
100
Index
80
Euro
Other
Single Name
Net notional (billions of U.S. dollars)
1,500
Index
1,000
60
40
500
20
0
0
Percent new gross notional
100
80
Index Tranches
Net notional (billions of U.S. dollars)
150
Index Tranches
100
60
40
50
20
0
Percent new gross notional
100
80
Index Options
Net notional (billions of U.S. dollars)
250
200
60
150
40
100
20
50
Index Options
0
0
2010
2012
2014
2016
2018
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the percentage of monthly average gross notional of new positions in the four contract
types by contract currency. The right column shows the monthly average net notional of new positions in the four
contract types by contract currency. Notionals are measured in U.S. dollar billion equivalents; positive net notional
indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
29
The Long and Short of It: The Post-Crisis Corporate CDS Market
4. Conclusion
In the wake of the 2007–09 financial crisis, OTC markets have been a focus of regulatory
reform. This article has examined the evolution of the CDS market in the current
regulatory environment. We find that market participants have reacted quickly to
changing circumstances. For example, the ISDA 2014 Master Agreement was adopted
widely within six months of its introduction. Although central clearing of single-name
CDS contracts is not mandatory in the United States, activity has been moving to
products that are eligible for clearing. Reductions in the sectors of the market not
eligible for clearing are almost exclusively driving the overall decrease in outstanding
single-name contract gross notional.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
30
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix (Continued)
1: The DTCC Data
Market participants submit their trade records to the DTCC’s Trade Reporting Repository (TIW)
for operational purposes. The TIW holds the most current contract details on the official,
or “gold,” record for both cleared and bilateral CDS transactions. The DTCC estimates that it
captures approximately 98 percent of all credit derivative transactions in the global marketplace.
The DTCC’s customer base includes all major global derivatives dealers and more than
2,500 buyside firms and other market participants located in over seventy countries.
Trade records are submitted by participants to the system. The quality of reporting varies
across participants and time. We detail below the filters and data adjustments applied to the
positions data to avoid biases in our empirical analyses.
Positions Data
We combine weekly snapshots of position data into a single data set. We then apply the
following filters:
1. A single position is recorded from the point of view of each party. We remove duplicate
records based on the transaction ID (dtcc_reference_id), which is assigned by the
DTCC upon submission, and the reporting date (rpt_date_key).
There are some records for which rpt_date_key is missing. When that is the case,
we use the date in the file name, which corresponds to the date on which the data
were captured.
2. We backfill RED code (red_id). RED codes are alphanumeric codes assigned to
reference entities and reference obligations, and are used to confirm trades on trade
matching and clearing platforms.
3. We use the RED code to backfill key variables related to the reference entity: reference
entity location (reference_entity_jurisdiction) and reference entity sector
(reference_entity_sector).
4. We use RED codes to backfill locations of the counterparties
(counterparty_settlement_location).
5. We keep records where transaction_status = Certain.
6. We keep records where reference_entity_type ∈ {CORP, INDEX, missing}
and exclude records where reference_entity_type ∈ {CMBS, ECMBS, ELCDS,
ERMBS, MUNI, Other, RMBS, SOV, STATBODY, STATE, SUPRA}.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
31
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix (Continued)
1: The DTCC Data (Continued)
7. We create a variable called prod_type based on the values of two variables,
product_type and subproduct_type, as follows:
prod_type
product_type
subproduct_type
Index
CreditDefaultSwapIndex
(missing)
Index option
CreditDefaultSwapIndex
(‘OPTION’, ‘SWAPTION’)
Index tranche
CreditDefaultSwapIndexTranche
(missing)
Single name
CreditDefaultSwapShort
(missing)
8. The notional values in the positions data set as delivered by the DTCC are scaled
down by a factor of 100 from January 2011 through April 2016. We multiply these
values by 100.23
After the initial cleaning of the data, we apply the following conditions to construct the
samples of outstanding and new positions used for Charts 3-11 in the main text of this article:
1. Exclude reference entities located outside of U.S. and advanced European economies.24
2. Exclude the following indexes: MBX, MCDX, LCDX, IOS, PO, ABX, IBOXX COCO,
and IBOXX LOANS.
3. The location of a party is determined based on the reported settlement location rather
than registered office, because the registered office data are far sparser; however, both
contain almost the same information when both exist. The settlement location field
(as well as the registered office field) is at the account level. Therefore, there are cases
where a party with multiple accounts is associated with multiple locations in the
same week. For such cases, we keep the location of the party as the location of the
account through which the greatest notional amount is traded in that week.
4. The rating of a reference entity is assigned based on the rating at the reporting date
rather than the rating at origination.
a. Ratings for single-name positions are obtained from Markit’s composites by
convention data set. While the positions are at a weekly frequency, the rating data
are at a daily frequency. We take the latest available rating from Markit in a
given week as the rating for the reference entity for that week. If the rating in
Markit is missing for the week of the trade date, we use the latest available rating
within the month prior to the trade date.
b. Ratings for index, index option, and index tranche products are assigned based
on the name of the product. CDX.NA.IG is considered investment grade and
CDX.NA.HY is considered high yield. Among regional iTraxx indexes, the
Europe Series, Senior Financial, and Asia Ex-Japan are considered investment
grade. The European Crossover is considered high yield.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
32
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix 1: The DTCC Data (Continued)
5. The clearing indicator is constructed from an amalgamation of three sources: ICE, CCP
trades in the DTCC, and the mandatory clearing dates for indexes. For each RED code,
the reference entity is considered eligible for clearing starting on the first trade date
identified across the three sources. The first source is the list of clearable instruments
published by ICE. For single-name contracts not required for clearing, ICE reports
the date included for clearing for each reference entity (red6). ICE clears
single-name contracts conditioned on a specific ISDA definition, tenor, and
tier. ICE also reports clearing eligibility for indexes and sovereigns. CDX and iTraxx
indexes for which we do not find information through ICE are given a clearing date of
April 26, 2013, the date of mandatory clearing for Category 1 indexes.25 The iBoxx index
family is never cleared. Lastly, we consider the earliest trade date at which one of the
transacting parties is a CCP as a possible candidate for the clearing date dummy for a
specific reference entity.
6. The splits by location of counterparty, documentation ID, participant type, clearing
indicator, and reference entity location are performed on existing positions. The
remaining splits are performed on new trades only. The new trade data set is generated
by taking the cleaned positions data set and removing duplicate transaction identifiers
based on trade date.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
33
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix (Continued)
2: Risk-Adjusted Positions
All splits are filtered based on the split sample variable and exclude missing split
variables. The exceptions are for the split by rating, where we include missing as its own
category (Unrated category), and for the split by reference entity sector (Other category).
In this appendix, we study the properties of risky duration-adjusted positions. More specifically,
for each position of participant p in reference entity i at date t with time to maturity τ, we follow
He et al. (2017) and compute the approximate risky duration of the position as
4τ
j
p,i,tτ,t
j (λit + rt /4)
,
RDp,i,τ,t = 14 ∑ e 4
j=1
where r the risk-free yield on a Treasury maturing in j/4 years, and λit is the default hazard
rate implied by the quoted (Markit) five-year spread on reference entity i, date t
j/4
t
(
λit = 4 log 1 +
(
Markit spreadit
,
4Lit
with Lit the priced loss-given-default for reference entity i on date t. The risk-adjusted gross
and net notional for reference entity i on date t are then given by
Risk-adjusted gross notionalit =
Notional boughtp,i,τ,t
∑ RD
it p,i,τ,t
p,τ
(
((
(
Risk-adjusted net notionalit = ∑ ∑ RDp,i,τ,tNet positionp,i,τt 1 ∑ Net positionp,i,τ,t > 0 .
p
τ
τ
One caveat for this procedure is that it requires a match to a Markit quote on a five-year
CDS contract on reference entity i on date t. We can compute the risky duration of the
majority of positions in single-name and index contracts throughout our sample.
Appendix Charts 2A–2E plot monthly average duration-weighted gross and net notional of
outstanding positions by counterparty type, counterparty location, on-the-run index
membership, clearing eligibility, and master agreement type. Qualitatively, the results in
Appnedix Charts 2A–2E are similar to the unweighted gross and net notional of
outstanding positions plotted in Charts 1–5 of this article, with the only exceptions the net
notional traded by participant type in the index market and the net notional traded by
counterparty location in the single-name and index markets. On a duration-adjusted basis,
dealers become buyers of protection in the index market in the second half of 2012, about
two years earlier than on an unadjusted basis. Appendix Chart 2B further shows that, on a
duration-adjusted basis, U.S. counterparties are buyers of protection from the rest of the
world in both the single-name and the index market starting in the second half of 2015,
while, on an unadjusted basis, U.S. counterparties have close to zero net positions
throughout this period.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
34
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix (Continued)
2: Risk-Adjusted Positions (Continued)
Chart 2A
Duration-Adjusted Monthly Average Gross and Net Positions by Participant Type
Dealer
Gross notional (billions of U.S. dollars)
500
Single Name
400
300
Net notional (billions of U.S. dollars)
4
CCP
Single Name
2
0
200
-2
100
-4
0
Gross notional (billions of U.S. dollars)
300
Index
Net notional (billions of U.S. dollars)
4
Index
2
200
0
100
-2
0
-4
2010
Buyside
2012
2014
2016
2018
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the duration-adjusted monthly average gross notional of outstanding positions
in single-name and index contracts of different participant types. The right column shows the duration-adjusted
monthly average net notional of positions in single-name and index contracts of different participant types.
Notionals are measured in U.S. dollar billion equivalents; positive net notional indicates net buying of
protection. CCP is central counterparty.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
35
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix (Continued)
2: Risk-Adjusted Positions (Continued)
Chart 2B
Duration-Adjusted Monthly Average Gross and Net Positions
by Location of Counterparty
Dealer
Gross notional (billions of U.S. dollars)
500
Single Name
400
300
Net notional (billions of U.S. dollars)
4
CCP
Single Name
2
0
200
-2
100
-4
0
Gross notional (billions of U.S. dollars)
300
Index
Net notional (billions of U.S. dollars)
4
Index
2
200
0
100
-2
-4
0
2010
Buyside
2012
2014
2016
2018
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the duration-adjusted monthly average gross notional of outstanding positions
in single-name and index contracts by location of the counterparty to the contract. The right column shows
the duration-adjusted monthly average net notional of new positions in single-name and index contracts
by counterparty location. “Home” refers to counterparties domiciled in the same jurisdiction. Notionals are
measured in U.S. dollar billion equivalents; positive net notional indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
36
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix (Continued)
2: Risk-Adjusted Positions (Continued)
Chart 2C
Duration-Adjusted Monthly Average Gross and Net Positions by Type of Index
On the run
Off the run
Both
Gross notional (billions of U.S. dollars)
250
Single Name
200
Net notional (billions of U.S. dollars)
25
150
15
100
10
50
5
0
0
Gross notional (billions of U.S. dollars)
150
Index
100
20
Neither
Single Name
Net notional (billions of U.S. dollars)
25
Index
20
15
10
50
5
0
0
2010
2012
2014
2016
2018
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the duration-adjusted monthly average gross notional of outstanding positions
in single-name and index contracts by type of index. The right column shows the duration-adjusted monthly
average net notional of new positions in single-name and index contracts by type of index. Notionals are
measured in U.S. dollar billion equivalents; positive net notional indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
37
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix (Continued)
2: Risk-Adjusted Positions (Continued)
Chart 2D
Duration-Adjusted Monthly Average Gross and Net Positions by Clearing Eligibility
Eligible
Gross notional (billions of U.S. dollars)
250
Single name
200
Non-eligible
Net notional (billions of U.S. dollars)
30
Single name
20
150
100
10
50
0
0
Gross notional (billions of U.S. dollars)
150
Index
100
Net notional (billions of U.S. dollars)
25
Index
20
15
10
50
5
0
0
2010
2012
2014
2016
2018
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the duration-adjusted monthly average gross notional of outstanding positions
in single-name and index contracts by clearing eligibility, determined as described in Appendix 1. The right
column shows the duration-adjusted monthly average net notional of new positions in single-name and index
contracts by clearing eligibility. Notionals are measured in U.S. dollar billion equivalents; positive net notional
indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
38
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix (Continued)
2: Risk-Adjusted Positions (Continued)
Chart 2E
Duration-Adjusted Monthly Average Gross and Net Positions by ISDA Master Agreement
ISDA 2003
Gross notional (billions of U.S. dollars)
250
Single name
200
Net notional (billions of U.S. dollars)
25
150
15
100
10
50
5
0
0
Gross notional (billions of U.S. dollars)
150
Index
20
ISDA 2014
Single name
Net notional (billions of U.S. dollars)
20
Index
15
100
10
50
5
0
0
2010
2012
2014
2016
2018
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the duration-adjusted monthly average gross notional of outstanding positions in
single-name and index contracts by ISDA master agreement. The right column shows the duration-adjusted monthly
average net notional of new positions in single-name and index contracts by ISDA master agreement. Notionals are
measured in U.S. dollar billion equivalents; positive net notional indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
39
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix (Continued)
2: Risk-Adjusted Positions (Continued)
Chart 2F
Duration-Adjusted Monthly Average Gross and Net Positions by Credit Rating
Single-name:
Index:
Both:
AAA
Investment grade
Unrated
Percent new gross notional
100
Single Name
80
(AA,A)
High-yield
BB
Net notional (billions of U.S. dollars)
500
400
60
300
40
200
20
100
0
0
Percent new gross notional
100
BBB
Below BB
Single Name
Net notional (billions of U.S. dollars)
1,500
Index
80
Index
1,000
60
40
500
20
0
2010
2012
2014
2016
2018
0
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW);
Markit (credit rating information).
Notes: The left column shows the percentage of duration-adjusted monthly average gross notional of new
positions in single-name and index contracts for different credit rating categories. The right column shows
the duration-adjusted monthly average net notional of new positions in single-name and index contracts
for different credit rating categories. Notionals are measured in U.S. dollar billion equivalents; positive net
notional indicates net buying of protection.
Charts 2F-2I plot the duration-weighted net notional of new positions by credit rating of
the underlying reference entity, initial maturity of the contract, reference entity sector, and
currency of the contract, together with the duration-weighted percent of new gross notional
amounts traded represented by each category. For new positions, the duration-adjusted positions
are similar across the board to the unadjusted positions.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
40
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix 2: Risk-Adjusted Positions (Continued)
Chart 2G
Duration-Adjusted Monthly Average Gross and Net Positions by Initial Maturity
(0,1 years]
Percent new gross notional
100
(1-5 years)
Single Name
80
>10 years
(5-10 years)
Net notional (billions of U.S. dollars)
600
Single Name
400
60
40
200
20
0
0
Percent new gross notional
100
Index
80
60
Net notional (billions of U.S. dollars)
2,000
Index
1,500
1,000
40
500
20
0
2010
5 years
2012
2014
2016
2018
0
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the percentage of duration-adjusted monthly average gross notional of new positions
in single-name and index contracts across initial maturity buckets. The right column shows the duration-adjusted
monthly average net notional of new positions in single-name and index contracts across initial maturity buckets.
For index options, maturity is the maturity of the option, not the underlying index. Notionals are measured in
U.S. dollar billion equivalents; positive net notional indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
41
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix 2: Risk-Adjusted Positions (Continued)
Chart 2H
Duration-Adjusted Monthly Average Gross and Net Positions by Sector
Manufacturing
Percent new gross notional
100
80
Energy
Utility/Telecommunications
Gross notional
Financial services
Other
Net notional (in billions of U.S. dollars)
400
Net notional
300
60
200
40
100
20
0
0
2010
2012
2014
2016
2018
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW);
Markit (reference entity sector information).
Notes: The left panel shows the percentage of duration-adjusted monthly average gross notional of new positions in
single-name contracts by reference entity sector. The right panel shows the duration-adjusted monthly average net
notional of new positions in single-name contracts by reference entity sector. Notionals are measured in U.S. dollar
billion equivalents; positive net notional indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
42
The Long and Short of It: The Post-Crisis Corporate CDS Market
Appendix (Continued)
2: Risk-Adjusted Positions (Continued)
Chart 2I
Duration-Adjusted Monthly Average Gross and Net Positions by Contract Currency
U.S. dollar
Percent new gross notional
100
80
Single Name
Net notional (billions of U.S. dollars)
500
400
60
300
40
200
20
100
0
0
Percent new gross notional
100
Index
80
Euro
Other
Single Name
Net notional (billions of U.S. dollars)
1,500
Index
1,000
60
40
500
20
0
2010
0
2012
2014
2016
2018
2010
2012
2014
2016
2018
Sources: Depository Trust and Clearing Corporation (DTCC); DTCC’s Trade Information Warehouse (TIW).
Notes: The left column shows the percentage of duration-adjusted monthly average gross notional of new positions
in single-name and index contracts by contract currency. The right column shows the duration-adjusted monthly
average net notional of new positions in single-name and index contracts by contract currency. Notionals are
measured in U.S. dollar billion equivalents; positive net notional indicates net buying of protection.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
43
The Long and Short of It: The Post-Crisis Corporate CDS Market
Notes
Acknowledgments: The authors thank Benjamin Marrow, Anna Sanfilippo, and Matthew Yeaton for excellent
research assistance and the Depository Trust and Clearing Corporation for providing the data on CDS positions
and transactions. For comments on previous drafts of the paper, they also thank Caren Cox, Giulia Iori,
Dragon Tang, Johanna Schwab, and participants at the 2015 annual meeting of the Society for Economic Dynamics,
the Banque du France/Federal Reserve Bank of Cleveland Conference on Endogenous Financial Networks and
Equilibrium Dynamics, the Chicago Initiative in Theory and Empirics 2015, the Marti G. Subrahmanyam Festschrift,
the Macro Financial Modeling Group Winter 2019 Meeting, and Chicago Booth Alumni Insight 2019.
1
Additional changes to the regulatory environment not discussed here include changes to capital charges for derivative
positions; the introduction of liquidity requirements, which are also affected by the amount of derivative positions
an institution holds; and the introduction of the Volcker Rule, which restricts banks from participating in
proprietary trading and owning or investing in hedge funds and private equity funds.
2
DTCC estimates that the TIW covers about 98 percent of globally traded CDS.
3
The coverage of index tranche trades between dealers and customers has declined to about 75 percent at the end
of 2018. This suggests that customers have shifted their activity to institutions not regulated by the Federal Reserve.
Our results pertaining to index tranche contracts are robust to this coverage decline.
4
See http://www.creditfixings.com to view auction results.
5
For a detailed discussion of the auction mechanism and its efficiency, see Helwege et al. (2009), Gupta and
Sundaram (2015), Chernov et al. (2013), and Du and Zhu (2017).
6
See the ISDA SwapsInfo website at http://swapsinfo.org.
7
For example, as we discuss next, index tranches were a popular product. The attachment
and detachment points—that is, the range of percentage losses on reference entities the tranche absorbs—of the
new version of the index tranches can only be determined when the auction results are finalized.
8
For an analysis of the effects on the single-name CDS spreads of entering and exiting the index, see Bai
and Shachar (2015).
9
For comparison, while there is no periodic adjustment of the S&P 500, changes are made when needed,
including removal of a company from the index when it violates one or more inclusion criteria or when it is
involved in a bankruptcy, merger, takeover, or other significant corporate restructuring. As a consequence,
25–50 index replacements take place every year, which represents a 5–10 percent turnover of the index composition.
10
Markit creates a “Liquidity List” after each publication of the “6 month Analysis Top 1,000 Single Names” report by
DTCC. The list is used to determine roll exclusions and inclusions.
11
For current index rules for CDX.NA.IG, CDX.NA.HY, and iTraxx, see Markit’s website.
12
The rapid growth of the CDS market in the early 2000s was reflected not only in the enormous levels of
gross notional amount outstanding, but also in an operational backlog. In response, the CDS contract and its trading
conventions were changed in April 2009 as part of the Big Bang Protocol in order to create a more standardized
contract. Standardization streamlines netting across trades and facilitates centralized clearing.
13
The emerging market CDS index, 12 CDX.EM, has semiannual payments.
14
For further details, see Acharya et al. (2010).
15
That is, bilateral positions between non-CCP market participants.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
44
The Long and Short of It: The Post-Crisis Corporate CDS Market
Notes (Continued)
16
For a comprehensive review of the literature, see Augustin et al. (2014).
17
Comparing the gross notional of contracts reported by the DTCC and the gross notional of contracts that banks
and dealers voluntarily reported in a Bank for International Settlements (BIS) survey, ECB (2009) concludes that
while the DTCC covers 98 percent of CDS contracts involving a dealer, it captures only 29 percent of contracts reported
to the BIS that do not involve a dealer.
18
19
See BIS (2013) for a consultative report on authorities’ access to centralized trade repository data.
An index type (for example, CDX.NA.IG) is counted as a single reference entity, regardless of the series and version.
20
The TIW also reports the number of contracts and gross notional for customer-to-customer and customer-to-CCP trades.
Since the supervisory DTCC data do not cover these trades, we omit them from Chart 2.
21
This represents, however, 45 percent of the gross notional of index contracts exchanged between a dealer and a CCP.
22
As of July 2018, the total of BBB-rated corporate bonds outstanding was $2.56 trillion, bonds rated higher than
BBB totaled $2.55 trillion, and high-yield bonds totaled $1.21 trillion. Source: Bloomberg Opinion, “The Corporate
Bond Market is Getting Junkier,” July 10, 2018, https://www.bloomberg.com/opinion/articles/2018-07-10/corporatebonds-are-getting-junkier.
23
We confirmed with the DTCC that this factor should indeed be taken into account. We do not know, however,
whether this data issue affects other regulatory agencies using these historical data.
24
Advanced European economies include Austria, Belgium, Denmark, England and Wales, Finland, France, Germany,
Greece, Iceland, Ireland, Italy, Luxembourg, the Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, Scotland,
and the United Kingdom.
25
See https://www.cftc.gov/PressRoom/PressReleases/pr6607-13.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
45
The Long and Short of It: The Post-Crisis Corporate CDS Market
References
Acharya, V. V., O. Shachar, M. Subrahmanyam, T. F. Cooley, M. Richardson, and I. Walter. 2010.
Regulating Wall Street: The Dodd-Frank Act and the New Architecture of Global Finance.
Hoboken, N.J.: John Wiley & Sons, Inc.
Arora, N., P. Gandhi, and F. A. Longstaff. 2012. “Counterparty Credit Risk and the Credit Default Swap
Market.” Journal of Financial Economics 103, no. 2 (February): 280–93.
Augustin, P., M. G. Subrahmanyam, D. Y. Tang, S. Q. Wang. 2014. “Credit Default Swaps:
A Survey.” Foundations and Trends in Finance 9, no. 1–2 (December): 1–196.
http://dx.doi.org/10.1561/0500000040.
Bai, J. and O. Shachar. 2015. “The Effects of Entering and Exiting a Credit Default Swap Index.”
Federal Reserve Bank of New York Liberty Street Economics (blog), March 30, 2015,
https://libertystreeteconomics.newyorkfed.org/2015/03/the-effects-of-entering-and-exiting-a-credit
-default-swap-index.html.
Bank for International Settlements. 2013. “Authorities’ Access to Trade Repository Data.” CPMI Papers,
no. 110, August. https://www.bis.org/cpmi/publ/d110.pdf.
——— 2018. “Statistical Release: OTC Derivatives Statistics at End-June 2018.” Oct. 31, 2018.
https://www.bis.org/publ/otc_hy1810.pdf.
Chen, K., M. J. Fleming, J. P. Jackson, A. Li, and A. Sarkar. 2011. “An Analysis of CDS Transactions:
Implications for Public Reporting.” Federal Reserve Bank of New York Staff Reports,
no. 517, September.
Chernov, M., A. S. Gorbenko, and I. Makarov. 2013. “CDS Auctions.” Review of Financial Studies 26, no. 3
(March): 768–805.
Collin-Dufresne, P., R. S. Goldstein, and F. Yang. 2012. “On the Relative Pricing of LongMaturity Index Options and Collateralized Debt Obligations.” Journal of Finance 67, no. 6
(December): 1983–2014.
Collin-Dufresne, P., B. Junge, and A. B. Trolle. 2018. “Market Structure and Transaction Costs
of Index CDSs.” Swiss Finance Institute Research Paper Series, no. 18–40, October.
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2786907.
Coval, J. D., J. W. Jurek, and E. Stafford. 2009. “Economic Catastrophe Bonds.” American Economic
Review 99, no. 3 (June): 628–66.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
46
The Long and Short of It: The Post-Crisis Corporate CDS Market
References (Continued)
Du, S., and H. Zhu. 2017. “Are CDS Auctions Biased and Inefficient?” Journal of Finance, 72, no. 6
(December): 2589–628.
Du, W., S. Gadgil, M. B. Gordy, and C. Vega. 2015. “Counterparty Risk and Counterparty
Choice in the Credit Default Swap Market.” Board of Governors of the Federal Reserve
System Finance and Economics Discussion Series, no. 2016-087, September.
https://www.federalreserve.gov/econresdata/feds/2016/files/2016087pap.pdf.
Duffie, D., M. Scheicher, and G. Vuillemey. 2015. “Central Clearing and Collateral Demand.”
Journal of Financial Economics 116, no. 2 (May): 237–56.
European Central Bank. 2009. “Credit Default Swaps and Counterparty Risk.” August.
https://www.ecb.europa.eu/pub/pdf/other/creditdefaultswapsandcounterpartyrisk2009en.pdf.
Eisfeldt, A., B. Herskovic, S. Rajan, and E. Siriwardane. 2018. “OTC Intermediaries.”
U.S. Department of Treasury, Office of Financial Research OFR Working Paper, no. 18–05, August.
https://www.financialresearch.gov/working-papers/files/OFRwp-18-05_OTC-Intermediaries.pdf.
Gehde-Trapp, M., Y. Gündüz, and J. Nasev. 2015. “The Liquidity Premium in CDS Transaction Prices:
Do Frictions Matter?” Journal of Banking and Finance 61 (December): 184–205.
Gupta, S., and R. K. Sundaram. 2015. “Mispricing and Arbitrage in CDS Auctions.” Journal of
Derivatives 22, no. 4 (Summer): 79–91. https://doi.org/10.3905/jod.2015.22.4.079.
He, Z., B. Kelly, and A. Manela. 2017. “Intermediary Asset Pricing: New Evidence from Many Asset
Classes.” Journal of Financial Economics 126, no. 1 (October): 1–35.
https://www.journals.elsevier.com/journal-of-financial-economics/highlighted-articles/
intermediary-asset-pricing-new-evidence-asset-classes.
Helwege, J., S. Maurer, A. Sarkar, and Y. Wang. 2009. “Credit Default Swap Auctions
and Price Discovery.” Journal of Fixed Income 19, no. 2 (Fall): 34–42.
Junge, B., and A. B. Trolle. 2015. “Liquidity Risk in Credit Default Swap Markets.”
Research Paper 13-65, Swiss Finance Institute Research Paper Series, no. 13-65, August.
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2371374.
Loon, Y. C., and Z. K. Zhong. 2016. “Does Dodd-Frank Affect OTC Transaction Costs and Liquidity?
Evidence from Real-time CDS Trade Reports.” Journal of Financial Economics 119, no. 3
(March): 645–72.
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
47
The Long and Short of It: The Post-Crisis Corporate CDS Market
References (Continued)
Oehmke, M., and A. Zawadowski. 2017. “The Anatomy of the CDS Market.” Review of Financial
Studies 30, no. 1 (January): 80–119.
Riggs, L., E. Onur, D. Reiffen, and H. Zhu. 2017. “Mechanism Selection and Trade Formation on
Swap Execution Facilities: Evidence from Index CDS.” Commodity Futures Trading Commission,
September. https://cftc.gov/sites/default/files/idc/groups/public/@economicanalysis/documents/file/
oce_mechanism_selection.pdf.
Seo, S. B., and J.A. Wachter. 2018. “Do Rare Events Explain CDX Tranche Spreads?” Journal of Finance 73, no. 5
(October): 2343–83.
Shachar, O. (2012): “Exposing the Exposed: Intermediation Capacity in the Credit Default
Swap Market.” Working paper, Federal Reserve Bank of New York.
https://pdfs.semanticscholar.org/adef/ec8aa2c94ecaac2967b6207a4d43a78f1484
.pdf?_ga=2.153451835.2018210055.1572563686-1493049330.1572563686.
Siriwardane, E. N. 2019. “Limited Investment Capital and Credit Spreads.” Journal of Finance 74, no. 5
(October): 2303-47.
Tang, D. Y., and H. Yan (2017): “Understanding Transactions Prices in the Credit Default Swaps Market."
Journal of Financial Markets 32, Issue C: 1–27.
The Economic Policy Review is published by the Research and Statistics Group of the Federal Reserve Bank of New York. The views
expressed are those of the individual authors and do not necessarily reflect the position of the Federal Reserve Bank of New York or
the Federal Reserve System. Economic Policy Review articles may be reproduced for educational or training purposes, provided
they are reprinted in full; include credit to the author(s), the publication, and the Bank; and include the publication’s disclaimer.
© 2020 The Federal Reserve Bank of New York
Federal Reserve Bank of New York
Economic Policy Review 26, no. 3, June 2020
48
@FedFRASER