Sovereign credit default swap (CDS) market, including the liquidity of the product, the major participants, the factors driving spreads, and the implications for Treasury
The presenting member began by describing some of the standard features of CDS contracts, noting that CDS can have reference entities that are either single name sovereigns/corporates or indices. Typical maturities on sovereign CDS are the 5- and 10-year points. Over 85 percent of CDS trades on sovereigns are denominated in US dollars. Recently, there have been efforts to provide standardization and transparency to the market in an effort to facilitate trade netting and provide more information to regulators. On a year-over-year basis, the sovereign CDS market has been growing at about a 30 percent pace.
The presenting member noted that the gross notional amount for sovereign CDS is $2.3 trillion, while the net notional amount- which reflects the actual settlement exposure in the event of default -is $230 billion. The presenting member noted that the corporate CDS market is about eight times larger than the sovereign market.
The member noted that the sovereign CDS market is a small fraction of the size of sovereign cash markets, which limits the impact that sovereign CDS has on sovereign yields. In the corporate market, however, the opposite is the case. Average monthly trading volume for Western European sovereign CDS is conservatively estimated at about $34 billion, with trades of $25 to $50 million common. Bid-ask spreads on higher quality sovereigns are typically less than 5 basis points, while spreads on lesser-quality sovereigns can be as wide as 100 basis points or more.
The presenting member noted that while there are natural buyers of sovereign CDS there are no natural sellers. This hurts the overall liquidity of the sovereign CDS market. Sellers of CDS tend to be market participants with a particular macro view on a sovereign credit or relative value traders.
Natural buyers include counterparty hedging desks (50 percent market share), while hedge funds and proprietary trading desks (30 percent market share) and real money investors (20 percent market share) often buy and sell depending on market conditions.
Counterparty hedging desks are the largest natural buyer. These entities purchase sovereign CDS to hedge sovereign exposure related to swap transactions. Proprietary trading desks and hedge funds use CDS to make macro trades or regional basis trades, while real money accounts use CDS as substitutes for cash bonds.
The presenting member stated that investor positioning in the last 18 months has shifted from a net short position to a net long position, and during this period the net-notional outstanding has increased.
The presenting member identified 5 drivers of sovereign CDS spreads, with the predominant factor being macro-economic fundamentals on the underlying sovereign. Sovereign CDS spreads are an indicator of default risk and other macro factors such as current account and fiscal balances, debt-to-GDP, and stability and credibility of government policies. Market sentiment and risk tolerance are also drivers of CDS prices. CDS spreads also widen as counterparty risk increases, a feature that would be mitigated if this market moved from an OTC market to an exchange with centralized clearing. There is currency risk embedded in sovereign CDS as well. Finally, basis trading between the cash and CDS market can also impact CDS prices, but there are not many investors currently putting on basis trades.
The member noted that there appears to be no evidence that CDS movements are driving borrowing costs for sovereigns. Instead, CDS spreads appear to move alongside cash market spreads. Some members questioned this conclusion.
The presenter then highlighted some benefits of sovereign CDS including the use of CDS as a risk management tool for a wide variety of market participants. CDS additionally provide a low cost means of gaining or reducing exposure to a particular sovereign or index. The product also allows market participants to short sovereign markets more efficiently than is possible in the corresponding cash market. CDS spreads also serve a signaling effect by improving the price discovery process and alerting participants to perceived risks. Typically movements in CDS lead changes in credit ratings.
In terms of risk, the presenter suggested that if position transparency on CDS is low and counterparty risk is high, defaults could lead to contagion. The presenter suggested that standardizing contracts and implementing a central clearing mechanism could mitigate these risks.
For sovereigns, CDS spreads serve as a good indicator of default risk and should be monitored and used to inform policy decisions and actions. Widening of CDS spreads is typically accompanied by increases in wholesale funding costs for banks, leading to markdowns for banks that hold government bonds in their portfolios.
With regards to implications for the U.S. Treasury, the presenter stated that that CDS on the US is not very actively traded, amounting to 0.03 percent of US debt held by the public. Net notional outstanding has been very stable in recent years, and given the small size of the market, it should not have any significant implications for Treasury. A host of macro-economic factors, however, could drive US CDS wider including rising or unsustainable deficits, increasing debt-to-GDP ratios, deflation or high inflation, and concerns about the status of the US dollar as a reserve currency.
A brief discussion followed the presentation with one member stating that CDS serve as a substitute for collateral, given that sovereign counterparties don’t typically post collateral.
Another member pointed out that CDS is not a very meaningful indicator for sovereigns that have control over the currency in which their debt is denominated.
A discussion ensued concerning whether CDS served a greater good beyond the needs of investment professionals following one member’s observation that there was often a lack of basis convergence between CDS and cash. One member stated that some investors were avoiding the CDS market altogether, fearful of reputational and regulatory risks associated from profiting from bad news. Some members opined that CDS helped to make markets more efficient. Others suggested that CDS were vehicles that allowed investors to express a contrary view on asset bubbles, and that this was beneficial.
A few members suggested that the unlimited open interest in CDS, where contracts could potentially exceed the underlying cash instruments, could create market distortions and that exchange limits might mitigate some of this risk. Standardization in contracts, centralized clearing, and position limits might improve the market over the long-run. One member stated that the negative carry on CDS should serve to limit how big open interest would grow.
Recent narrowing of long-dated swap spreads to Treasuries and the implications for Treasury
The presenting member began by noting that supply and demand dynamics between the swaps market and the Treasury market can be independent. Recent drivers of the spread between swaps and Treasuries include the supply of government debt, balance sheet capacity, relative funding costs, corporate issuance hedging, pension fund demand, mortgage convexity hedging, a reduction in the use of derivatives, hedging of currency linked and curve linked notes, and quarter-end dynamics. Supply and convexity hedging were cited as the most important factors.
The presenter then stated that the collapse of swap spreads since late 2008 was likely precipitated by two events. First, the LIBOR/OIS spread hit an all time peak following the Lehman collapse and the repair of the LIBOR market has been a major driver in the compression of swap spreads (especially in the short-end). Second, Treasury increased supply, starting in October and November of 2008 with the $40 billion in tap issues and the reintroduction of the 3-year note. The rapid rise in coupon issuance continued for another year.
The presenting member then focused on the two primary drivers of swap spreads, supply and rapid interest rate moves. The close relationship between the increase in the supply of government debt (as measured by the budget deficit as a percentage of nominal GDP) and the narrowing of swap spreads was noted. The presenter also pointed to the influence of the significant relative decline in corporate versus Treasury issuance on the level of swap spreads.
In the past, rapid moves in interest rates (as measured by the difference in rate level versus the trailing 90-day moving average) influenced the level of swap spreads through convexity hedging needs. The presenting member noted the anomalous situation this year when swap spreads narrowed during the sell-off in rates.
Regarding the recent price action at the end of March 2010 when 10-year swap spreads broke the zero barrier and turned negative, the presenter attributed the phenomenon to the following factors: quarter-end pressures on dealers to pare down balance sheet, which led to selling of Treasuries and replacement with swaps, and large swapped corporate issuance activity that left dealers with long swap spread positions.
The presenting member downplayed the possibility that fears about sovereign risk played a role in the narrowing of swap spreads and pointed to stable U.S. sovereign CDS levels during that period as evidence. [1]
As for future trends in longer-dated swap spreads, the presenter expressed the opinion that fiscal shifts and regulatory changes would have the greatest influence. The peaking of Treasury coupon issuance and the supply reductions going forward should normalize the swap spread curve over time. However, the effects of regulatory changes were more difficult to handicap.
Source: TBAC Minutes