I checked the Bloomberg for the USD swap rate and US giv strip curve with the same maturity, but swap rate is about 40bps above US gov strip rate( for instance 20 yr maturity). I do not get it why there is spread between them. What I know, swap rate is calculated by the forward rate and forward rate can be derived from zero coupon yield curve(US gov strip curve), so the swap rate curve should be very tight to zero coupon rate curve. I think that I miss something, if someone can point out it, I really appreciate !Thank in advance

Your question a bit sounds mixed up. You are referring to 3 different curves: 1) the USD par swap curve, the swap forward rate curve and the US Treasury strip curve. For curves of the SAME credit quality there is a deterministic relationship between the par, zero and forward curves. Given any one of these (for the same credit quality) you can solve for the other two.It sounds like you are comparing a 20-year swap rate (presumably a par or coupon-paying rate) to a 20-year zero coupon government strip curve rate. Two things are different here - credit quality and duration (the strip point having much longer duration). The comparison tells you little, if anything.You should be comparing par rates with par rates, zero rates with zero rates and forward rates with forward rates.Perhaps you can rephrase your question more clearly? It would help to include a description of the specific Bloomberg screens you are looking at.

Thank you for your help.Based on my understanding on par swap rate, I think it is the fix rate for interest rate swap. If you go Bloomberg, type YCRV(yield curve analysis), put S23(US Dollar swaps(30/360)) and I25 (US Treasury Actives), you will see there is a spread between two curves, and it gets bigger as maturity gets longer.By using upslope US Treasury curve to calculate the swap rate,which I got is lower than US Treasury curve at the same maturity. But Bloomberg's US Dollar swap curve is higher than US Treasury curve. I do not why this is happend.If this is still confusing, would you like to tell me how to build the US Dollar swaps curve, and any realtionship between US Dollar Swap rate and Zero Cupon Yield curve?Thank you.

i think the essence of your question is why a swap rate and a treasury rate of the same maturity have different values.... the reason for this, as DavidJN alluded to, is that the two rates represent different underlying credits... the treasury rate represents the credit of the US government, whereas the swap rate loosely represents the credit of A-AA rated multi-national banks. the reason that the swap rate represents bank credit is that the underlying short rate of a swap is LIBOR (London Interbank Offered Rate) which is the rate at which financial institutions in europe extend unsecured, term loans (eurodollars) to each other. it should be clear, then, that the credit of AA banks is less than that of the government, so the bank rate (swap rate) must be higher.... the concept is no different than the spread between corporate bonds and treasuries.....

Thank you johnself!If so, what curve should we use to calculate the swap rate? if it is convient for you, would you like to give me a description of Bloomberg screen you are using .Thanks

ok - here's basically how it works.... interest rate swap markets are now by far the largest and most liquid interest rate product there is.... so for even, fundamental maturities - 2y,3y,5y,10y,30y the swap rate is the GIVEN, not the RESULT. the market sets these rates second by second and trades 100's of millions at those prices in a period of a few minutes... so the only way to "calculate" the "base" swap rates is to look at the market at any given moment and these rates set exact landmarks as to the rate in 2y,5y, etc.... the tricky part is that swaps also trade (much less frequently) in "odd" years (16y, 17y) and customers more often than not want to execute a swap that matures on a very specific, uneven date (say 6.25y)... so one must decide on some way to interpolate between these liquid, given points in order to create a tradeable, continuos curve of discount factors so that any cashflow(s) on any given date become just a function of the 5-6 base rates along with the interp method... there are numerous posts on this site which go into the specifics of interpolation theories so i'll leave you to investigate that.... now to answer your bbg question, i'm not that familiar with their swap analytics, but:- an even year "spot" swap rate, eg the 9y market swap rate, type USSWAP9 <Curncy> - bbg also keeps track of swap spread to like-maturitiy treasurys (the topic of your original question).... to get the 10y swap spread type $$SWAP10 <iindex>for odd maturity swaps you will have to learn the yield curve functionality on bbg and sorry i cant' help there....

I remember reading from some book (probably fixed income securities), the credit risk between the Libor (A-AA) and Gov is literally zero. Historically there is no such a instance of a bank with credit AAA goes to bankcrupcy. There is one difference, liquidity. Gov Bond is much widely used in the Repo market, it is as good as cash. But this is not true for a loan. However, the difference should not be very big.

Repo spread helps drive a wedge between swap rate and treasury rate. In Europe, the GC curve is around Libor-10bp, which is roughly the spread between swap rate and govie bond yield. The GC spread is much wider in the States, hence the much lower treasury yield vs. swap rates

Johnself,I still do not get it. For calibrate the swap, what curve should I use? You mentioned the "landmarks", what is that?cksh2005,what does GC stand for? And why repo spread cause a big spread between the swap rate and treasury rate?Thanks a lot

GC - general collateral: terminology used for the pool of collateral eligible to secure funding. If a bond is not GC, then it could be trading special - in high demand hence at a special/lower repo rate.Think of it in this way:Say you want to invest in a 10yr maturity fixed income security, you can either: (1) do a 10yr swap (in this case, think of swap as just a par bond with financing/repo at Libor flat), "repo" the swap at Libor flat, or(2) buy a 10yr German Bund, repo the bond at Libor-10bpsFor the Bund, you're actually "picking up" an extra 10bps from favourable financing. Well, all thing being equal, where do you think the Bund should yield? Around 10bps lower.