John Jansen has written about this, and others have commented at his excellent blog on this topic. This is my take: there are many who want to receive fixed, and pay floating for a long time. This depresses the 30-year swap yield.

Some want to do so because they are hedging exotic instruments that do the opposite, others because they might be hedging long term guarantees for long-dated liabilities (pensions, etc.), looking to minimize losses synthetically.

I’m not sure anyone has the definitive answer. But here’s a game to play. Swap rates reflect AA counterparties. Take a 30-year Treasury. Pay fixed on the swap, receive back 3-month LIBOR. The package now pays 20 bp over 3-month LIBOR. Put it in a trust, get Moody’s and S&P to rate it, and call it “The Floating Rate Treasury Trust.” Sell participations to money market funds. This beats three-month T-bills by a little less than 60 basis points before expenses.

Now, I wouldn’t buy that if I were managing a money market fund just because of the illiquidity versus Treasuries. But there is a tweak we could try:

On top of the trust, offer protection on a basket of 100 single-A names on a five year rolling basis. Now the yield really swings; negotiate terms with the rating agencies so that it can be rated A1/P1/F1.

I think that I saw somewhere (recently) that swap spreads on the long end were negative. I assume (correct me if I’m wrong) that this is the spread over treasuries. How is this possible? What does/did it mean?

PS: Good luck to all of the CFA candidates who took there exams today?