Market Structure and Transaction Costs of Index CDSs

• DOI: http://doi.org/10.1111/jofi.12953
• Author: ANDERS B. TROLLE,BENJAMIN JUNGE,PIERRE COLLIN‐DUFRESNE
• Published date: 01 October 2020
• Date: 01 October 2020

THE JOURNAL OF FINANCE •VOL. LXXV, NO. 5 •OCTOBER 2020

Market Structure and Transaction Costs

of Index CDSs

PIERRE COLLIN-DUFRESNE, BENJAMIN JUNGE, and ANDERS B. TROLLE∗

ABSTRACT

Despite regulatory efforts to promote all-to-all trading, the post–Dodd-Frank index

credit default swap market remains two-tiered. Transaction costs are higher for

dealer-to-client than interdealer trades, but the difference is explained by the higher,

largely permanent, price impact of client trades. Most interdealer trades are liquid-

ity motivated and executed via low-cost, low-immediacy trading protocols. Dealer-to-

client trades are nonanonymous; they almost always improve upon contemporaneous

executable interdealer quotes, and dealers appear to price discriminate based on the

perceived price impact of trades. Our results suggest that the market structure is a

consequence of the characteristics of client trades: relatively infrequent, large, and

differentially informed.

THE INDEX CREDIT DEFAULT SWAP (CDS) market constitutes an important

component of the corporate credit market. Index CDSs allow banks, asset

managers, and other institutional investors to efficiently hedge and trade

∗Pierre Collin-Dufresne is at EPFL and Swiss Finance Institute. Benjamin Junge is at Capital

Fund Management. Anders B. Trolle is at HEC Paris and Copenhagen Business School. Wethank

Stefan Nagel (the Editor), two anonymous referees, Bruno Biais, Darrell Duffie, Thierry Foucault,

Larry Glosten, Michael Johannes, René Kallestrup, Laurence Lescourret, Albert Menkveld, Ioanid

Rosu, Grigory Vilkov, Søren Willemann, Hongjun Yan, Zhaodong Zhong, Alex Zhou, Haoxiang

Zhu, and seminar participants at the Bank of England, BI Norwegian Business School, Boston

University, the Commodity Futures Trading Commission, Cornerstone Research, Deutsche Bun-

desbank, EPFL, ESSEC, the European Central Bank, the Federal Reserve Bank of New York,

the Federal Reserve Board, Frankfurt School of Finance & Management, HEC Paris, Lombard

Odier, McGill University, Rutgers University, University of New South Wales, University of St.

Gallen, the 12th Annual Central Bank Conference on Microstructure of Financial Markets, the

2016 Paris Finance Meeting, the 2016 SFI Research Days, the 2017 Chicago Financial Institu-

tions Conference, the 2017 Fixed Income and Financial Institutions Conference, the 2017 Four

Nations Cup, the Midwest Finance Association 2017 meeting, the American Finance Association

2018 meeting, the Northern Finance Association 2018 meeting, the 2018 FRIC conference, the

2018 NFI conference, the 2018 Paris Microstructure Forum, and the Society for Financial Econo-

metrics 2018 conference for comments and suggestions. Collin-Dufresne acknowledges research

support from the Swiss Finance Institute. Trolle acknowledges support from the Investissements

d’Avenir Labex (ANR-11-IDEX-0003/Labex Ecodec/ANR-11-LABX-0047) and the Danish Finance

Institute. The authors do not have any conflicts of interest as identified in The Journal of Finance

disclosure policy.

Correspondence: Anders B. Trolle, HEC Paris, Finance Department, 1 Rue de la Libération,

Yveline, Jouy-en-Josas, 78350, France; e-mail: trolle@hec.fr.

DOI: 10.1111/jofi.12953

© 2020 the American Finance Association

2719

2720 The Journal of Finance®

aggregate credit risk in the economy. Unlike single-name CDSs, index CDSs

have remained popular since the financial crisis, with tens of billion dollars of

notional amount traded on a daily basis. Nevertheless, little is known about

the cost of trading in this important market.

The index CDS market is also interesting as a test case of how recent regula-

tion introduced in the wake of the financial crisis affects the structure of swap

markets. Since its inception in 2003, the index CDS market has operated as a

classic two-tiered over-the-counter (OTC) market in which global derivatives

dealers provide liquidity to clients in the dealer-to-client (D2C) segment of the

market, and dealers trade among themselves in the interdealer (D2D) segment

of the market. New swap market regulation following the Dodd-Frank Act had

the potential to change this market structure by mandating that trades in

the most liquid index CDSs be executed on so-called swap execution facilities

(SEFs).1These regulated trading platforms are required to offer trading in or-

der books, which opens the market to all-to-all trading where clients can com-

pete with dealers for liquidity provision. However, the regulation also allows

for trading via request for quote (RFQ) provided that at least three dealers are

put in competition for trades.2Interestingly, several years after the new regu-

lation was fully implemented, all-to-all trading has yet to materialize. Instead,

the two-tiered market structure persists, with D2C trades taking place on one

group of SEFs (almost exclusively via name-disclosed RFQs) and D2D trades

taking place on another group of SEFs (via a diverse set of anonymous trading

protocols).3

From both a regulatory and a market design perspective, it is important to

understand the persistence of this bifurcated market structure. Indeed, given

that index CDSs are highly standardized and centrally cleared, and the most

liquid of them trade in large volumes, the lack of all-to-all trading may seem

puzzling (see, for example, Duffie (2012)).

One view is that dealers have market power and therefore a vested inter-

est in maintaining a two-tiered market structure to limit competition from

nondealer liquidity providers (see, for example, Managed Funds Association

(2015)). An alternative view is that the two-tiered market structure is a con-

sequence of the nature of client order flow. If clients trade infrequently but in

1Other key elements of the new swap market regulation are posttrade transparency via the

immediate public dissemination of trades as well as mandatory central clearing of index CDSs

with standardized contract terms.

2RFQ is an electronic trading protocol in which executable prices for a given notional amount

are requested simultaneously from multiple dealers. In a name-disclosed RFQ, the quote requester

reveals his identity.Compared to traditional trading in OTC markets, where dealers are contacted

sequentially,the RFQ protocol increases quote competition between dealers.

3Referring to both the index CDS and the interest rate swap markets, a recent article summa-

rizes the current situation as follows: “...dealer banks still trade together privately in one segment

of the market and the buy side still executes via RFQ to the dealers in another. Proponents of this

view say that nothing really changed in terms of how firms execute swaps except that the buy side

has gone from RFQ-ing one dealer to RFQ-ing three. This appears to be in stark contrast to the

all-to-all trading model envisioned for the swaps markets by regulators under Dodd-Frank.” See

“SEFs: A Market Divided,” Profit & Loss, October 22, 2015.

Market Structure and Transaction Costs of Index CDSs 2721

large sizes, then dealers’ willingness to absorb sizeable supply and demand

imbalances may depend on the existence of an interdealer market to manage

inventory risk.4Moreover, if clients are differentially informed, dealers may

mitigate adverse selection risk by trading nonanonymously with clients and

using heterogeneous trading protocols in the anonymous interdealer market.

To help distinguish between these two views, we conduct a detailed empirical

study of client and dealer trade characteristics and transaction costs in the

post–Dodd-Frank index CDS market.

Using transaction data from October 2, 2013 (when the first SEFs started

operating), to October 16, 2015, we focus on the two most popular credit in-

dices, CDX.IG and CDX.HY, which, respectively, cover the investment-grade

and high-yield components of the North American corporate credit market.

The transaction data include execution timestamps, transaction prices, and

trade sizes up to certain notional caps. In addition, we develop algorithms that

allow us to identify the SEF on which a trade took place and the type of trade

(outright trade, index roll, curve trade, or delta hedge of an index swaption or

tranche swap). Outright trades in five-year CDSs on the most recently issued

(on-the-run) index account for the majority of trading volume. These trades are

the focus of the paper. The SEF on which the trade took place in turn reveals

whether the trade is D2C or D2D.5

We find that trading volumes are large, but much larger in the D2C than the

D2D segment. The average daily notional amounts across all types of trades in

the D2C segment are USD 9.843 billion for CDX.IG and USD 3.705 billion for

CDX.HY. In the D2D segment, the corresponding figures are USD 1.354 billion

and USD 0.402 billion. The client order flow consists of relatively few trades of

large size. For CDX.IG, for example, there are on average 114 client trades per

day with a median trade size of USD 50 million.

We next document differences between transaction costs of D2C and D2D

trades and investigate whether these can be attributed to differences in price

impacts or differences in dealer profits. We measure transaction costs using the

effective half-spread, which is the difference between the transaction price and

the mid-point of contemporaneous quotes to buy or sell protection (henceforth,

the mid-quote). We measure price impact as the change in the mid-quote over

a period of approximately 15 minutes following a trade. Throughout the paper,

we express transaction prices and quotes in terms of par spreads.6For D2 C

4This view is expressed, for instance, by the new chairman of the Commodity Futures Trad-

ing Commission (CFTC), who writes that “[dealers] compete with each other, often deriving small

profits per trade from a large volume of transactions,” “[they] offer competitive prices to their cus-

tomer base,” and “[w]ithout access to D2D markets, the risk inherent in holding swaps inventory

arguably would require dealers to charge their buy-side customers much higher prices for taking

on their liquidity risk, assuming they remained willing to do so.” See Giancarlo (2015).

5Because we identify D2C and D2D trades based on the SEF on which the trade took place,

our sample is limited to the period during which SEFs were in operation and to trades that are

executed on SEFs.

6The par spread is the annual insurance premium, as a percentage of the notional amount,

such that the present value of the index CDS is zero for both the buyer and the seller at the outset

of the trade.