Treasury yields - Why are zeros lower?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Pacinotti
Posts: 17
Joined: Wed May 15, 2019 1:37 pm

Treasury yields - Why are zeros lower?

Post by Pacinotti »

For, presumably, the same level of risk, why are yields of zeros lower than non-zeros?

This is easy to see on the Fidelity yield table:

Image
secondopinion
Posts: 3196
Joined: Wed Dec 02, 2020 1:18 pm

Re: Treasury yields - Why are zeros lower?

Post by secondopinion »

A good amount of the yield difference is that the bid-ask spread is wider (they quote asks).
Passive investing: not about making big bucks but making profits. Active investing: not about beating the market but meeting goals. Speculation: not about timing the market but taking profitable risks.
Topic Author
Pacinotti
Posts: 17
Joined: Wed May 15, 2019 1:37 pm

Re: Treasury yields - Why are zeros lower?

Post by Pacinotti »

The yields displayed on the table are ask yields. Why would I buy a zero with lower yield for the same horizon.

I am a rank amateur when it comes to bonds so this likely is a bonehead question. I must be missing something, like maybe coupon payment/yield is not included on non-zeros, but I believe it is...
User avatar
grabiner
Advisory Board
Posts: 32917
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: Treasury yields - Why are zeros lower?

Post by grabiner »

A minor factor is that the yield curve is currently inverted, with yields from 1-3 years higher than longer-term yields. A 10-year Treasury bond gets some of its value from coupons in the first three years, and those coupons are effectively short-term zeros with a higher yield than longer-term zeros.

However, liquidity is probably more important. There is essentially no difference between a three-month Treasury bill and a three-month zero (which might be a stripped coupon from a longer-term bond), but the zero has far fewer investors who want to trade it.
Wiki David Grabiner
User avatar
Kevin M
Posts: 13959
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Treasury yields - Why are zeros lower?

Post by Kevin M »

The differences in yields for stripped interest and stripped coupon STRIPS can be quite large, and it depends on maturity as to how they compare to non-STRIPS:

Image

The red vertical bars are stripped interest at the top and stripped coupon at the bottom. Example: 2/15/2041 (18.25-year) stripped interest yield is 4.22% while the stripped coupon yield maturing the same date is 3.97%. The stripped interest yield is higher than the yield for the 4.75% coupon bond maturing the same date, for which the yield is 4.01% (blue dot on the vertical red line at about 18 years on the chart).

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
JoMoney
Posts: 14561
Joined: Tue Jul 23, 2013 5:31 am

Re: Treasury yields - Why are zeros lower?

Post by JoMoney »

grabiner wrote: Mon Nov 21, 2022 10:15 pm... There is essentially no difference between a three-month Treasury bill and a three-month zero (which might be a stripped coupon from a longer-term bond), but the zero has far fewer investors who want to trade it.
I believe that's true, but shouldn't less demand for it push it to a higher yield?
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham
User avatar
jeffyscott
Posts: 11843
Joined: Tue Feb 27, 2007 9:12 am

Re: Treasury yields - Why are zeros lower?

Post by jeffyscott »

If you click through to the actual quotes, the yields are even stranger. For ~3 mo. there are two STRIPS with ask yield of 3.872 and 3.778 maturing on 2/15/23. Maturing on the same day there are two notes at 4.162 and 4.091. The Bid yields are also lower on the STRIPS than those two notes. The lowest and highest spreads are the two notes, but they all range from 0.133 to 0.282

But the real outlier is a 7.125% bond, CUSIP 912810EP9 maturing on 2/15 just like the rest, but the ask is only 3.306% and the bid is 4.312% a spread of more than 1%.
And so it goes, And so it goes, And so it goes, And so it goes, But where it's goin' no one knows
User avatar
Kevin M
Posts: 13959
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Treasury yields - Why are zeros lower?

Post by Kevin M »

jeffyscott wrote: Tue Nov 22, 2022 11:57 am If you click through to the actual quotes, the yields are even stranger. For ~3 mo. there are two STRIPS with ask yield of 3.872 and 3.778 maturing on 2/15/23. Maturing on the same day there are two notes at 4.162 and 4.091.
Yes, those would be the top and bottom of the red vertical lines in the chart I shared. One is stripped interest, the other is stripped coupon.
If I make a calculation error, #Cruncher probably will let me know.
MikeG62
Posts: 4758
Joined: Tue Nov 15, 2016 3:20 pm
Location: New Jersey

Re: Treasury yields - Why are zeros lower?

Post by MikeG62 »

Spoke to a fixed income specialist at Fidelity this morning (on an unrelated matter). Asked him why the yields on the Zero's are lower (across most of the yield curve). He said very limited supply of these Strips so not much activity (most of these strips are from Treasuries issued a very long time ago). When I asked why anyone would buy one of these when comparable term Treasury yields are 15-30bps higher, he said unless someone needed a Treasury with a very specific maturity, most of these won't sell (at the yields shown anyway).

I've bought a lot of Treasuries this year (almost all in the secondary market) and I've never even considered Treasury Zero's. Yield has simply not competitive with straight up Treasuries.
Real Knowledge Comes Only From Experience
JayB
Posts: 349
Joined: Sat May 28, 2022 9:57 am

Re: Treasury yields - Why are zeros lower?

Post by JayB »

MikeG62 wrote: Wed Nov 23, 2022 9:22 am Spoke to a fixed income specialist at Fidelity this morning (on an unrelated matter). Asked him why the yields on the Zero's are lower (across most of the yield curve). He said very limited supply of these Strips so not much activity (most of these strips are from Treasuries issued a very long time ago). When I asked why anyone would buy one of these when comparable term Treasury yields are 15-30bps higher, he said unless someone needed a Treasury with a very specific maturity, most of these won't sell (at the yields shown anyway).

I've bought a lot of Treasuries this year (almost all in the secondary market) and I've never even considered Treasury Zero's. Yield has simply not competitive with straight up Treasuries.
I think there's more to it than that. For instance, let's compare a 2-year coupon Treasury (4.59% YTM) to a 2-year Zero (STRIP) (4.40% YTM). Potential purchasers of this maturity may very well be thinking of holding their bond to maturity and, for the coupon bond, reinvesting the coupon payments to mature at the same time as the principal. The terminal value of the STRIP in 2 years is known ($1K), but the total value of the coupon bond at maturity, with coupons reinvested, is not. So the STRIP might be a more (or equally) compelling value to the coupon bond -- notwithstanding the differences in purchase YTM -- if investors in the aggregate believe that the front end of the yield curve will decline and coupon reinvestment rates will be lower than they are today. In short, Zero coupon bonds do not have reinvestment risk prior to maturity, but coupon bonds do; theoretically, this may be reflected in the pricing and purchase YTMs relative to expected interest rates.
MikeG62
Posts: 4758
Joined: Tue Nov 15, 2016 3:20 pm
Location: New Jersey

Re: Treasury yields - Why are zeros lower?

Post by MikeG62 »

JayB wrote: Wed Nov 23, 2022 9:59 am
MikeG62 wrote: Wed Nov 23, 2022 9:22 am Spoke to a fixed income specialist at Fidelity this morning (on an unrelated matter). Asked him why the yields on the Zero's are lower (across most of the yield curve). He said very limited supply of these Strips so not much activity (most of these strips are from Treasuries issued a very long time ago). When I asked why anyone would buy one of these when comparable term Treasury yields are 15-30bps higher, he said unless someone needed a Treasury with a very specific maturity, most of these won't sell (at the yields shown anyway).

I've bought a lot of Treasuries this year (almost all in the secondary market) and I've never even considered Treasury Zero's. Yield has simply not competitive with straight up Treasuries.
I think there's more to it than that. For instance, let's compare a 2-year coupon Treasury (4.59% YTM) to a 2-year Zero (STRIP) (4.40% YTM). Potential purchasers of this maturity may very well be thinking of holding their bond to maturity and, for the coupon bond, reinvesting the coupon payments to mature at the same time as the principal. The terminal value of the STRIP in 2 years is known ($1K), but the total value of the coupon bond at maturity, with coupons reinvested, is not. So the STRIP might be a more (or equally) compelling value to the coupon bond -- notwithstanding the differences in purchase YTM -- if investors in the aggregate believe that the front end of the yield curve will decline and coupon reinvestment rates will be lower than they are today. In short, Zero coupon bonds do not have reinvestment risk prior to maturity, but coupon bonds do; theoretically, this may be reflected in the pricing and purchase YTMs relative to expected interest rates.
That is true.

However, the reinvestment risk on the coupon on the traditional Treasury note is likely to be trivial in the overall outcome of things not because i am predicting little movement in rates but because the coupon is tiny in relation to the principal amount invested. This is especially so at the shorter end of the curve (where we are seeing the biggest delta). It is hard to imagine that is the main driver causing these 15-30bp differences in yields to maturity. I do accept that it may account for a piece (small??) of the delta.
Real Knowledge Comes Only From Experience
User avatar
jeffyscott
Posts: 11843
Joined: Tue Feb 27, 2007 9:12 am

Re: Treasury yields - Why are zeros lower?

Post by jeffyscott »

^ And, in fact, none of the 2/15/23 ones that I posted about above will have a coupon to reinvest prior to maturity. So reinvestment risk explains nothing.

But, as noted above, the bigger outlier is the bond (or was at the time that I looked):

CUSIP 912810EP9 bond maturing on 2/15/23, ask yield = 3.306%
STRIPS maturing on 2/15, ask yield = 3.872% and 3.778%
Notes maturing 2/15, ask yield = 4.162 and 4.091
And so it goes, And so it goes, And so it goes, And so it goes, But where it's goin' no one knows
JackoC
Posts: 3903
Joined: Sun Aug 12, 2018 11:14 am

Re: Treasury yields - Why are zeros lower?

Post by JackoC »

Kevin M wrote: Mon Nov 21, 2022 10:24 pm The differences in yields for stripped interest and stripped coupon STRIPS can be quite large, and it depends on maturity as to how they compare to non-STRIPS:

The red vertical bars are stripped interest at the top and stripped coupon at the bottom. Example: 2/15/2041 (18.25-year) stripped interest yield is 4.22% while the stripped coupon yield maturing the same date is 3.97%. The stripped interest yield is higher than the yield for the 4.75% coupon bond maturing the same date, for which the yield is 4.01% (blue dot on the vertical red line at about 18 years on the chart).

Kevin
To clarify which I think you also did later, the bottom of the lines is stripped principal payments, the top is stripped interest payments. One reason principal STRIPS can have a much lower yield is they are not fungible with any other instrument, ie reconstructing the original bond (to arbitrage price discrepancies) requires using the principal STRIP of that bond. For example, there are two treasury issues maturing 2/15/41, the 4.75's and the 1.875's. The principal STRIP yielding below 4% belongs to the former, an old long bond; not the latter, a 20 yr issued last year. In contrast interest STRIPS with the same maturity date are fungible, all have the save CUSIP number. Once they are stripped it doesn't matter what issue they came from and they can be attached to any issue with the same payment date to arbitrage price discrepancies. Therefore, the interest STRIP will follow the 'generic' curve but the principal STRIP will more strongly reflect anything unusual in the yield of the particular coupon bond from which it came. And those two issues themselves, the coupon bonds, have significantly different yield, the 4.75's offer yield just now on Vanguard was 3.90, the 1.875's 4.07 (the 4.75's prin STRIP 3.84, the generic interest STRIP for that date 4.09). One reason can be just the curve slope effect on different duration due to different coupon, but other reasons can be for example an issue being the 'on-the-run' or Cheapest to Deliver into one of the futures contracts. The 4.75's are now only 7th CTD into the Dec 'Bond' futures contract, ZB, but appear likely to become CTD eventually on a subsequent contract because highest coupon in the vicinity. Whether or how completely that explains the yield difference to 1.875's I don't know, but it's not necessarily a pure STRIP anomaly. It also relates to why coupon instruments of the same or very similar maturities can also have fairly different yields.
User avatar
Kevin M
Posts: 13959
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Treasury yields - Why are zeros lower?

Post by Kevin M »

MikeG62 wrote: Wed Nov 23, 2022 9:22 am Spoke to a fixed income specialist at Fidelity this morning (on an unrelated matter). Asked him why the yields on the Zero's are lower (across most of the yield curve). He said very limited supply of these Strips so not much activity (most of these strips are from Treasuries issued a very long time ago). When I asked why anyone would buy one of these when comparable term Treasury yields are 15-30bps higher, he said unless someone needed a Treasury with a very specific maturity, most of these won't sell (at the yields shown anyway).

I've bought a lot of Treasuries this year (almost all in the secondary market) and I've never even considered Treasury Zero's. Yield has simply not competitive with straight up Treasuries.
The discussion was based on a false premise, since STRIPS don't all have lower yields, as I showed in my chart and explained. For maturities of 20 to 25 years, stripped interest yields are higher, and stripped coupon yields ar about the same, maybe a little higher, than the coupon issue yields. From about 13 years to 18 years, the stripped interest yield is significantly higher than the coupon issue yield, while the stripped coupon issue yield is somewhat lower. I'd say that 13 to 25 years is a pretty large chunk of the yield curve.

The more interesting question is why the large deltas between stripped interest and stripped coupon issues, which JackoC explained.
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Kevin M
Posts: 13959
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Treasury yields - Why are zeros lower?

Post by Kevin M »

JackoC wrote: Wed Nov 23, 2022 12:25 pm
Kevin M wrote: Mon Nov 21, 2022 10:24 pm The differences in yields for stripped interest and stripped coupon STRIPS can be quite large, and it depends on maturity as to how they compare to non-STRIPS:

The red vertical bars are stripped interest at the top and stripped coupon at the bottom. Example: 2/15/2041 (18.25-year) stripped interest yield is 4.22% while the stripped coupon yield maturing the same date is 3.97%. The stripped interest yield is higher than the yield for the 4.75% coupon bond maturing the same date, for which the yield is 4.01% (blue dot on the vertical red line at about 18 years on the chart).

Kevin
To clarify which I think you also did later, the bottom of the lines is stripped principal payments, the top is stripped interest payments. One reason principal STRIPS can have a much lower yield is they are not fungible with any other instrument, ie reconstructing the original bond (to arbitrage price discrepancies) requires using the principal STRIP of that bond. For example, there are two treasury issues maturing 2/15/41, the 4.75's and the 1.875's. The principal STRIP yielding below 4% belongs to the former, an old long bond; not the latter, a 20 yr issued last year. In contrast interest STRIPS with the same maturity date are fungible, all have the save CUSIP number. Once they are stripped it doesn't matter what issue they came from and they can be attached to any issue with the same payment date to arbitrage price discrepancies. Therefore, the interest STRIP will follow the 'generic' curve but the principal STRIP will more strongly reflect anything unusual in the yield of the particular coupon bond from which it came. And those two issues themselves, the coupon bonds, have significantly different yield, the 4.75's offer yield just now on Vanguard was 3.90, the 1.875's 4.07 (the 4.75's prin STRIP 3.84, the generic interest STRIP for that date 4.09). One reason can be just the curve slope effect on different duration due to different coupon, but other reasons can be for example an issue being the 'on-the-run' or Cheapest to Deliver into one of the futures contracts. The 4.75's are now only 7th CTD into the Dec 'Bond' futures contract, ZB, but appear likely to become CTD eventually on a subsequent contract because highest coupon in the vicinity. Whether or how completely that explains the yield difference to 1.875's I don't know, but it's not necessarily a pure STRIP anomaly. It also relates to why coupon instruments of the same or very similar maturities can also have fairly different yields.
Thanks for this! I had read previously that the difference between stripped interest and stripped coupon yields had to do with liquidity and reconstituting the original issue, but I've never seen it explained so thoroughly. Also appreciate the explanation about the reasons for different yields for higher coupon bonds (that can't be explained by duration).
If I make a calculation error, #Cruncher probably will let me know.
JayB
Posts: 349
Joined: Sat May 28, 2022 9:57 am

Re: Treasury yields - Why are zeros lower?

Post by JayB »

Kevin M wrote: Wed Nov 23, 2022 12:43 pm
MikeG62 wrote: Wed Nov 23, 2022 9:22 am Spoke to a fixed income specialist at Fidelity this morning (on an unrelated matter). Asked him why the yields on the Zero's are lower (across most of the yield curve). He said very limited supply of these Strips so not much activity (most of these strips are from Treasuries issued a very long time ago). When I asked why anyone would buy one of these when comparable term Treasury yields are 15-30bps higher, he said unless someone needed a Treasury with a very specific maturity, most of these won't sell (at the yields shown anyway).

I've bought a lot of Treasuries this year (almost all in the secondary market) and I've never even considered Treasury Zero's. Yield has simply not competitive with straight up Treasuries.
The discussion was based on a false premise, since STRIPS don't all have lower yields, as I showed in my chart and explained. For maturities of 20 to 25 years, stripped interest yields are higher, and stripped coupon yields ar about the same, maybe a little higher, than the coupon issue yields. From about 13 years to 18 years, the stripped interest yield is significantly higher than the coupon issue yield, while the stripped coupon issue yield is somewhat lower. I'd say that 13 to 25 years is a pretty large chunk of the yield curve.

The more interesting question is why the large deltas between stripped interest and stripped coupon issues, which JackoC explained.
Traditionally, STRIPS across the entire maturity spectrum have had higher YTMs than same-maturity coupon Treasurys. For medium and longer maturities, this has often been on the order of +20 to +60 bp of yield, and for the past 20 years or so, I have used this disparity to build higher-yielding bond ladders with Zeros only than I could with coupon bonds. I have readily understood that STRIPS have a greater duration than same-maturity coupon Treasurys, and have observed markedly larger bid-ask spreads, especially on long bonds -- hence the yield premium on STRIPS as perhaps compensation for these.

The $64K question is: What recent change(s) have disrupted that relationship across many maturities? Is it mostly a matter of having an inverted yield curve with the prospect of more Fed rate hikes? Elevated arbitrage activity? Or what? And when might more traditional YTM relationships between STRIPS and coupon-paying Treasurys return?
MikeG62
Posts: 4758
Joined: Tue Nov 15, 2016 3:20 pm
Location: New Jersey

Re: Treasury yields - Why are zeros lower?

Post by MikeG62 »

Kevin M wrote: Wed Nov 23, 2022 12:43 pm
MikeG62 wrote: Wed Nov 23, 2022 9:22 am Spoke to a fixed income specialist at Fidelity this morning (on an unrelated matter). Asked him why the yields on the Zero's are lower (across most of the yield curve). He said very limited supply of these Strips so not much activity (most of these strips are from Treasuries issued a very long time ago). When I asked why anyone would buy one of these when comparable term Treasury yields are 15-30bps higher, he said unless someone needed a Treasury with a very specific maturity, most of these won't sell (at the yields shown anyway).

I've bought a lot of Treasuries this year (almost all in the secondary market) and I've never even considered Treasury Zero's. Yield has simply not competitive with straight up Treasuries.
The discussion was based on a false premise, since STRIPS don't all have lower yields, as I showed in my chart and explained. For maturities of 20 to 25 years, stripped interest yields are higher, and stripped coupon yields ar about the same, maybe a little higher, than the coupon issue yields. From about 13 years to 18 years, the stripped interest yield is significantly higher than the coupon issue yield, while the stripped coupon issue yield is somewhat lower. I'd say that 13 to 25 years is a pretty large chunk of the yield curve.

The more interesting question is why the large deltas between stripped interest and stripped coupon issues, which JackoC explained.
I was discussing with the Fidelity rep maturities similar to those brought up by the OP in his original post (maturities ranging from 3 months out to 2-3 years). Perhaps my characterization as across most of the yield curve was misleading and inaccurate. For those maturities the yield on zero’s was considerably below other Treasuries.
Real Knowledge Comes Only From Experience
User avatar
Kevin M
Posts: 13959
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Treasury yields - Why are zeros lower?

Post by Kevin M »

^Yeah, I said something like this in another thread, and was corrected. So, I built the spreadsheet and chart to show the STRIPS yields separately from the non-STRIPS yields, and learned about the differences at different maturities.
If I make a calculation error, #Cruncher probably will let me know.
Post Reply