I must admit, I never thought it would actually happen. But the most liquid single-name CDS are now expected to move to fixed coupons with upfront payments in the not-too-distant future. The fixed coupons will supposedly be set at 100bps and 500bps (depending on credit quality). Currently single-name CDS trade on a "running spread" or "par spread" basis—the protection buyer pays for protection by making regular spread payments (premiums) to the protection seller until the contract matures or there is a credit event. So for example, if a 5-year CDS on $10 million notional is quoted at 100bps, the protection buyer will pay $100,000 a year (usually broken into quarterly payments). Crucially, no money is exchanged upfront—this is why CDS are leveraged bets. Now the market is moving to trading on a "points upfront" basis. Contracts will have fixed coupons of either 100bps or 500bps, and upfront payments will be made at initiation to reflect the change in price. The amount paid upfront will equal the present value of the difference between the current market spread and the fixed coupon. So if a contract with a fixed coupon of 100bps is trading at 150bps, the protection buyer would make an upfront payment equal to the present value of the difference between 150bps and 100bps. If the market spread is lower than the fixed coupon, then the protection seller would have to pay the difference between the two spreads upfront. The most popular index contracts (CDX and iTraxx) trade on a similar points upfront basis. Each index series has a fixed coupon (expressed as a spread) that's set when the series is launched, and when the market spread trades away from the fixed spread (known as the "deal spread"), the parties make an upfront payment equal to the present value of the difference between the two spreads. The deal spread on the CDX IG11 is 150bps, so when the IG11 is trading around 190bps, like it is now, the protection buyer has to pay the difference between the spreads upfront. Okay, so why the move to upfront plus fixed coupons? For one thing, it will allow more effective netting between index and single-name contracts by making the cash flows more similar. This is especially important as index and single-name CDS move onto central clearinghouses—the ability to net down outstanding CDS exposures is one of the main benefits of a clearinghouse. Second, upfront plus fixed coupons also makes CDS cash flows more like bonds. Third, and most importantly, the move will help the dealers reduce their exposure to large spread movements ("jump risk"). The unholy spread widening in the aftermath of Lehman's collapse savaged some dealers as they tried to keep a net flat position. (Yes, I know, a change in the CDS market that benefits the dealers—what were the odds?) To illustrate jump risk, imagine that a hedge fund—let's call it Max Power Capital (MPC)—purchased 5-year CDS protection on GE at 100bps back in October, and that the same contract is trading at 400bps today. The trade is now "in-the-money," and MPC wants to take its profit and go home. The most common way to do this is for MPC to sell offsetting CDS protection on GE that matures on the same date as the original contract (December 20, 2013, since it was purchased in October 2008). That way, MPC would collect a profit of 300bps a year, since it would be paying annual premiums of 100bps but collecting annual premiums of 400bps. The problem with this approach is that MPC is still exposed to some default risk—the premium cash flows will stop if GE defaults before the contracts mature. To account for this leftover default risk, MPC will enter into an offsetting CDS contract with a dealer in which the dealer pays MPC the difference between the spreads (i.e., MPC's profit) upfront. Since dealers generally try to run matched books, the dealer now has another position it needs to hedge, so it has to sell an offsetting CDS with another counterparty. The problem is that the CDS on GE that mature on December 20, 2013 are now "off-the-run" — the current 5-year contract referencing GE, which is "on-the-run" because 5-year contracts are the most liquid, matures on March 20, 2013. Moreover, to match the cash flows of the two contracts, the dealer has to find a counterparty willing to make a sizable upfront payment to offset the upfront payment the dealer made to MPC. Needless to say, it's difficult and often expensive to find a counterparty willing to buy an off-the-run contract at an off-market price. When spreads suddenly gap out significantly, dealers end up making substantial upfront payments to protection buyers looking to close out their trades and take their profits. Moving to upfront CDS with fixed coupons will help the dealers in this situation because upfront payments make the contracts less sensitive to mark-to-market spread movements. It will also make it much easier for dealers to find offsetting contracts with similar upfront payments. It remains to be seen how much upfront payments reduce the liquidity of single-name CDS. Opinion in the CDS market on this issue seems to be roughly split. I suspect upfront payments won't significantly reduce liquidity for the 100 or so most heavily traded references, since the most liquid CDS contracts are the index contracts that already trade with points upfront. We shall see though.