We first looked at the trading volume in single name corporate and sovereign credit default swaps in three blog entries in May and June 2010, immediately after the Depository Trust and Clearing Corporation began releasing volume information. In today’s blog, the first of four in the series, we look at weekly data since the week ended July 16, 2010 to measure how much of existing single name CDS trading is the exchange of positions among a small group of dealers and how much is true “end user” trading.
We first looked at single name CDS trading volume in this series of three blog entries:
In this blog, we analyze additional data from the Depository Trust and Clearing Corporation and made available from www.dtcc.com at this link (please note consent is necessary to access the data):
The most recent explanation of the data provided by the DTCC is available here:
In today’s blog, we are interested in how much of the “live trades” in the CDS trade warehouse represent transactions between two dealers and how much involve non-dealer end users on at least one side of the trade. The DTCC trade warehouse explanation above does not identify the dealers supplying data to DTCC at the current time. At the time of our June 2, 2010 blog, however, DTCC listed the 14 “families” of contributing dealers as follows:
- Bank of America Merrill Lynch
- BNP Paribas
- Credit Suisse
- Deutsche Bank
- Goldman Sachs
- Morgan Stanley
- Royal Bank of Scotland
We presume the list of contributing dealers is somewhat longer than it was in 2010, but it is not clear that the highly concentrated nature of CDS trading has changed during that time interval. We analyze that issue in a separate blog.
Table 1 is released weekly by the DTCC. It lists the number of contracts and notional values that are still “live” in the trade warehouse at that time. DTCC is careful to note that the data may not be complete if the inclusion of a particular trade would identify the parties to the transaction. We report here the composition of trades by four categories of “seller buyer” combinations in the trading of single name credit default swaps:
|Combination of Seller and Buyer||Seller||Buyer|
Based on the number of trades reported by DTCC, the breakdown among these four buyer-seller combinations are as follows:
In percentage terms, the dealer-dealer share of single name CDS contracts traded has ranged from 79.86% to 88.47% with an average of 81.68%. Trades between non-dealer parties were only 0.12% of total trades on average:
The results are similar if one looks at the dealer-dealer share of trades by notional principal. Dealer-dealer trades ranged from 73.40% to 88.36% of the total, with an average of 76.28%. Trades involving two non-dealers were only 0.16% of total notional principal traded:
Well-intentioned academics and journalists, plus many politicians, well-intentioned or not, discuss trading levels on single name credit default swaps as if such numbers were gifts from God. Instead, we see that roughly 76-82% of all single name credit default swaps are trades between Bill Smith at Goldman Sachs and John Smith at JPMorgan or other dealer firms. As a sophisticated observer of financial markets, should an investor take these traded prices as meaningful information on the credit worthiness of the underlying reference name? Is it safe to assume that Bill Smith and John Smith do not talk to each other?
Let’s look at examples from other markets in which the same dealer firms have participated for many decades:
- London interbank offered rate market. This lawsuit by Charles Schwab accuses 12 firms of colluding to manipulate libor. Nine of the 12 firms are also dealers listed by DTCC in single name credit default swaps:
- Municipal bond market. This recent article on Reuters summarizes the repeated legal violations stemming from collusion in the market for municipal bonds in the United States. 6 of the 8 firms named are also dealers in single name CDS that were listed by DTCC in 2010:
Given that the majority of dealers listed by DTCC in single name credit default swaps are, in the words of the Reuters article, “serial offenders,” a sophisticated observer should assume that both traded CDS spreads and quoted spreads are highly likely to have been affected by collusion. Any other assumption could be hazardous to your wallet.
Donald R. van Deventer
Honolulu, January 4, 2012