Why is the SEC yield the one that Vanguard displays?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
linenfort
Posts: 1997
Joined: Sat Sep 22, 2007 9:22 am
Location: #96151D

Why is the SEC yield the one that Vanguard displays?

Post by linenfort » Tue Apr 17, 2018 9:18 am

Example: Looking at Vanguard's own short-term treasury fund (VSBSX, #1942). Yield is 2.25%. You can see the definition when you click the 'A':
A — BASED ON HOLDINGS' YIELD TO MATURITY FOR PRIOR 30 DAYS;DISTRIBUTION MAY DIFFER
Ok, fair enough. There's your caveat. But if my most recent statement reads
Est. yield: 1.18%
Wouldn't be better for prospective investors to see the most recent distribution yield on the front page?
Maybe not. I'm just asking.

Earlier thread: Is SEC Yield a good way to pick a bond fund?
Is it too early for an all-bond-portfolio thread?

User avatar
Pajamas
Posts: 5084
Joined: Sun Jun 03, 2012 6:32 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by Pajamas » Tue Apr 17, 2018 9:58 am

Because fund companies are required to do so as it allows consumers to compare a standardized measure of bond fund yields.

It's also the best predictor of future yields according to this article:

https://www.wsj.com/articles/SB10001424 ... 0271646756
The SEC yield is better than historical data at estimating future income yield, says Ken Volpert, head of Vanguard Group's taxable-bond group and co-manager of Vanguard Total Bond Market. "It's a close approximation to the net yield to maturity of the fund, which is what we as bond investors think about," he says.
Matthew Tucker, head of investment strategy for fixed income at BlackRock Inc., which runs the iShares ETF brand, says: "Distribution yield is the best indicator of the current level of income a fund is paying. The yield to maturity [or SEC yield] is the best forward-looking indicator."

Chuck
Posts: 2049
Joined: Thu May 21, 2009 12:19 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by Chuck » Tue Apr 17, 2018 10:04 am

So, if you are choosing a fund, the distribution yield is the yield you "would have gotten" and the SEC yield is the yield you "probably will get."

alex_686
Posts: 3332
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by alex_686 » Tue Apr 17, 2018 10:05 am

SEC yield is forward looking, which is better in my opinion.

SEC yield is standardized. All funds have to use the same methodology. It makes comparisons easy.

I personally think that it is one of the higher quality numbers on a fund's fact sheet.

User avatar
Meg77
Posts: 2254
Joined: Fri May 22, 2009 1:09 pm
Location: Dallas, TX
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Meg77 » Tue Apr 17, 2018 11:04 am

Also, the SEC yield nets out the expense ratio which also makes it easier to compare between funds.
"An investment in knowledge pays the best interest." - Benjamin Franklin

User avatar
linenfort
Posts: 1997
Joined: Sat Sep 22, 2007 9:22 am
Location: #96151D

Re: Why is the SEC yield the one that Vanguard displays?

Post by linenfort » Tue Apr 17, 2018 11:11 am

Chuck wrote:
Tue Apr 17, 2018 10:04 am
So, if you are choosing a fund, the distribution yield is the yield you "would have gotten" and the SEC yield is the yield you "probably will get."
Looks like holders of VSBSX, defined above, will get half the yield of the SEC figure, if things remain the same.

Still, if that's the requirement, as Pajamas posted, then I guess that's that.
Is it too early for an all-bond-portfolio thread?

alex_686
Posts: 3332
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by alex_686 » Tue Apr 17, 2018 11:16 am

linenfort wrote:
Tue Apr 17, 2018 11:11 am
Looks like holders of VSBSX, defined above, will get half the yield of the SEC figure, if things remain the same.

Still, if that's the requirement, as Pajamas posted, then I guess that's that.
I think I know the trick here. The SEC yield covers all income, less expenses. So both coupon and capital appreciation. The fund only distributes coupon income.

User avatar
linenfort
Posts: 1997
Joined: Sat Sep 22, 2007 9:22 am
Location: #96151D

Re: Why is the SEC yield the one that Vanguard displays?

Post by linenfort » Tue Apr 17, 2018 2:05 pm

Thanks Alex, everyone.
Is it too early for an all-bond-portfolio thread?

BigJohn
Posts: 1523
Joined: Wed Apr 02, 2014 11:27 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by BigJohn » Tue Apr 17, 2018 3:21 pm

alex_686 wrote:
Tue Apr 17, 2018 11:16 am
I think I know the trick here. The SEC yield covers all income, less expenses. So both coupon and capital appreciation. The fund only distributes coupon income.
Are you sure? I’m no expert but how can any fund company know what the capital appreciation will be looking forward 30 days?

P&C actuary
Posts: 109
Joined: Fri Jul 19, 2013 12:27 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by P&C actuary » Tue Apr 17, 2018 3:43 pm

alex_686 wrote:
Tue Apr 17, 2018 11:16 am
I think I know the trick here. The SEC yield covers all income, less expenses. So both coupon and capital appreciation. The fund only distributes coupon income.
It is the other way around. Distribution yield includes interest/dividends and capital gains.

longinvest
Posts: 2786
Joined: Sat Aug 11, 2012 8:44 am

Re: Why is the SEC yield the one that Vanguard displays?

Post by longinvest » Tue Apr 17, 2018 3:52 pm

SEC yield or yield to maturity (YTM) doesn't predict the future return of a bond fund: How can Bond Funds perform better than yield

The future returns of a broad-market bond index fund can't be predicted.
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VLB/ZRR

alex_686
Posts: 3332
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by alex_686 » Tue Apr 17, 2018 4:11 pm

BigJohn wrote:
Tue Apr 17, 2018 3:21 pm
alex_686 wrote:
Tue Apr 17, 2018 11:16 am
I think I know the trick here. The SEC yield covers all income, less expenses. So both coupon and capital appreciation. The fund only distributes coupon income.
Are you sure? I’m no expert but how can any fund company know what the capital appreciation will be looking forward 30 days?
I am a bit rusty on the subject, but I think the key is that discount on discount bonds is not passed through.
P&C actuary wrote:
Tue Apr 17, 2018 3:43 pm
It is the other way around. Distribution yield includes interest/dividends and capital gains.
But capital gains - which IIRC includes the discount on discount bonds - are not distributed until year end. Hence why the OP is seeing a lower than expected monthly distribution yield.
Last edited by alex_686 on Tue Apr 17, 2018 4:21 pm, edited 1 time in total.

kaneohe
Posts: 4707
Joined: Mon Sep 22, 2008 12:38 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by kaneohe » Tue Apr 17, 2018 4:17 pm

Meg77 wrote:
Tue Apr 17, 2018 11:04 am
Also, the SEC yield nets out the expense ratio which also makes it easier to compare between funds.
Are you saying distribution yield does not net out expense ratio?

User avatar
alpenglow
Posts: 590
Joined: Tue May 31, 2011 12:02 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by alpenglow » Tue Apr 17, 2018 7:06 pm


User avatar
Kevin M
Posts: 9419
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Kevin M » Tue Apr 17, 2018 9:18 pm

BigJohn wrote:
Tue Apr 17, 2018 3:21 pm
alex_686 wrote:
Tue Apr 17, 2018 11:16 am
I think I know the trick here. The SEC yield covers all income, less expenses. So both coupon and capital appreciation. The fund only distributes coupon income.
Are you sure? I’m no expert but how can any fund company know what the capital appreciation will be looking forward 30 days?
SEC yield is a prior 30-day average yield to maturity (YTM) less expenses. YTM factors in coupon payments (income return) and change in bond price from today until the bond matures at par value of 100 on its maturity date (capital return).

For an individual bond, YTM is a good estimate of total annualized return for a bond held to maturity, with some uncertainty due to uncertainty in the reinvestment rate. However, a bond fund may not hold bonds until maturity, but may sell them at a capital gain or loss, then reinvest the proceeds in other bonds.

Also, although the YTM of a 5-year bond, for example, is a good estimate of annualized return over the next five years, the return over the next one year could be very different for several reasons. First, with a positively sloped yield curve, there could be a larger one-year capital return component than for the bond held to maturity. Next, the capital return component could be positive or negative due to changes in yields. So even with a single individual bond, you don't know what your 1-year return will be unless the bond has only one year until maturity (much less your 30-day return).

Since a bond fund never matures, as with a rolling bond ladder, SEC yield or its cousin average YTM do not provide certainty in the return of a bond fund over any particular holding period, including one equal to the duration of the fund. It's an OK ballpark estimate for a holding period equal to the duration, but the realized annualized return over that period can easily vary by +/- 0.5 percentage points, and even by +/- 1 percentage point or more.

Distribution yield as published by Vanguard (and they don't publish it for VSBSX by the way--not sure why) is average annualized income for the previous calendar month. Income is what you actually receive as dividends. Income includes bond coupon payments minus bond premium amortization plus bond discount accretion, or at least this is what the prospectus says for bond funds I've looked at. So based on what I've seen in the prospectuses, the Oblivious Investor blog post is not exactly correct on this particular point, but this requires more explanation.

The bond premium or discount accrual that is combined with coupon payments to determine income distributions only relates to premium or discount upon purchase. You are required to factor in these accruals in reporting income for tax purposes for an individual bond, but the fund takes care of that for you. Premiums or discounts to par value that arise as a result of market changes after purchase are not factored into income distributions.

So, if a Vanguard fund buys a bond at 100, the only income will be from the coupons. Say that bond later increases to 102 because of a decrease in yields or because of the shape of the yield curve as the bond "rolls down the yield curve". Income for that bond still will be coupon payments only, but you now have received a 2% capital return. If Vanguard now buys more of that same bond at 102, the income for that bond will consist of the same coupon payments, but will be reduced by a premium amortization amount (it would be amortized from 102 to 100 over the remaining lifetime of the bond if it were held to maturity). So the income distribution from the second holding would be lower than the first, even though they are the same bond with the same coupon rate.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

venkman
Posts: 538
Joined: Tue Mar 14, 2017 10:33 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by venkman » Tue Apr 17, 2018 9:20 pm

Let's say a Treasury note is issued today with a 2% coupon. Tomorrow, the interest rate on Treasuries of that maturity jumps to 4%. You could then buy the 2%-coupon note at a discount from its face value. The total return you would get on that discounted note, including all interest payments and return of principal, would be equal to the total return you could get from buying a 4%-coupon note at face value.

For the life of the note, you would continue to get coupon payments of 2%. That's your distribution yield. (Your yield would slightly higher than 2%, because you bought at a discount.)

At maturity, you would get back the full face value of the note, which will be more than what you paid for it. Your total return will be 4%. That's the SEC yield.

VSBSX is essentially an aggregation of bonds with lower-than-market coupon rates that you can currently buy at a discount.

User avatar
Kevin M
Posts: 9419
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Kevin M » Tue Apr 17, 2018 9:33 pm

The last paragraph in my previous reply illustrates one reason SEC yield is preferable to distribution yield in comparing bond funds. Modify that example to be two different bond funds, one that bought the bond some time ago at 100, and one that bought it some time later at 102. The yield to maturity of the bond is exactly the same for each fund, and thus will have the same contribution to the SEC yield calculation, and yield to maturity is more relevant when considering total return. However, the income component for the second fund will be lower than for the first, due to the premium amortization, so the contribution to distribution yield also will be lower for the second fund, even though the coupon rates are the same.

I believe one reason that SEC mandates reporting SEC yield according to a specified formula is that bond funds played games with distribution yields and perhaps other yields to make them look better to customers. Even distribution yield is not standardized, so you can't necessarily use it to compare funds. Vanguard uses a trailing 30-day yield, while Morningstar and others use a trailing 12-month yield. If yields have changed a lot in the last year, then a trailing 12-month yield isn't very meaningful.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Meg77
Posts: 2254
Joined: Fri May 22, 2009 1:09 pm
Location: Dallas, TX
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Meg77 » Wed Apr 18, 2018 11:19 am

kaneohe wrote:
Tue Apr 17, 2018 4:17 pm
Meg77 wrote:
Tue Apr 17, 2018 11:04 am
Also, the SEC yield nets out the expense ratio which also makes it easier to compare between funds.
Are you saying distribution yield does not net out expense ratio?
That's my understanding. Per Investopedia:
"The distribution yield of a security is calculated by dividing the distributions paid (yearly, monthly, etc.) by its cost or net asset value."
"The [SEC] yield figure reflects the dividends and interest earned during the period after the deduction of the fund's expenses."
"An investment in knowledge pays the best interest." - Benjamin Franklin

alex_686
Posts: 3332
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by alex_686 » Wed Apr 18, 2018 11:23 am

Meg77 wrote:
Wed Apr 18, 2018 11:19 am
kaneohe wrote:
Tue Apr 17, 2018 4:17 pm
Meg77 wrote:
Tue Apr 17, 2018 11:04 am
Also, the SEC yield nets out the expense ratio which also makes it easier to compare between funds.
Are you saying distribution yield does not net out expense ratio?
That's my understanding. Per Investopedia:
"The distribution yield of a security is calculated by dividing the distributions paid (yearly, monthly, etc.) by its cost or net asset value."
"The [SEC] yield figure reflects the dividends and interest earned during the period after the deduction of the fund's expenses."
Distributions are reduced by expenses. You don't pay taxes on the fund's gross income, you pay taxes on the fund's net income - income after expenses.

First, realize that distributions occur so the IRS know what to tax you on. Second, that distributions are based on income, SEC yield is based on forward expected "Total Returns".

User avatar
linenfort
Posts: 1997
Joined: Sat Sep 22, 2007 9:22 am
Location: #96151D

Re: Why is the SEC yield the one that Vanguard displays?

Post by linenfort » Wed Apr 18, 2018 12:06 pm

Makes sense, Kevin.
Thanks for the clear explanation.
Is it too early for an all-bond-portfolio thread?

User avatar
Kevin M
Posts: 9419
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Kevin M » Wed Apr 18, 2018 12:48 pm

alex_686 wrote:
Wed Apr 18, 2018 11:23 am
Meg77 wrote:
Wed Apr 18, 2018 11:19 am
kaneohe wrote:
Tue Apr 17, 2018 4:17 pm
Meg77 wrote:
Tue Apr 17, 2018 11:04 am
Also, the SEC yield nets out the expense ratio which also makes it easier to compare between funds.
Are you saying distribution yield does not net out expense ratio?
That's my understanding. Per Investopedia:
"The distribution yield of a security is calculated by dividing the distributions paid (yearly, monthly, etc.) by its cost or net asset value."
"The [SEC] yield figure reflects the dividends and interest earned during the period after the deduction of the fund's expenses."
Distributions are reduced by expenses. You don't pay taxes on the fund's gross income, you pay taxes on the fund's net income - income after expenses.
Exactly. A fund isn't going to pay you dividends without taking their cut first.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Meg77
Posts: 2254
Joined: Fri May 22, 2009 1:09 pm
Location: Dallas, TX
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Meg77 » Thu Apr 19, 2018 11:11 am

Kevin M wrote:
Wed Apr 18, 2018 12:48 pm
alex_686 wrote:
Wed Apr 18, 2018 11:23 am
Meg77 wrote:
Wed Apr 18, 2018 11:19 am
kaneohe wrote:
Tue Apr 17, 2018 4:17 pm
Meg77 wrote:
Tue Apr 17, 2018 11:04 am
Also, the SEC yield nets out the expense ratio which also makes it easier to compare between funds.
Are you saying distribution yield does not net out expense ratio?
That's my understanding. Per Investopedia:
"The distribution yield of a security is calculated by dividing the distributions paid (yearly, monthly, etc.) by its cost or net asset value."
"The [SEC] yield figure reflects the dividends and interest earned during the period after the deduction of the fund's expenses."
Distributions are reduced by expenses. You don't pay taxes on the fund's gross income, you pay taxes on the fund's net income - income after expenses.
Exactly. A fund isn't going to pay you dividends without taking their cut first.

Kevin
I understand that, but my point is that when you are trying to figure out where to park your money - let's say you are looking at several different money market funds for example - comparing the SEC yield easily tells you your expected distribution net of the expense ratio. Looking at the simple "distribution yield" you would have to do the math yourself, which many people don't realize. I didn't realize it myself until I was googling this recently for that very purpose. I used to look at the SEC yield of VG funds and then subtract the expense ratio, having no idea that was already factored in.
"An investment in knowledge pays the best interest." - Benjamin Franklin

User avatar
Kevin M
Posts: 9419
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Kevin M » Thu Apr 19, 2018 2:11 pm

Meg77 wrote:
Thu Apr 19, 2018 11:11 am
I understand that, but my point is that when you are trying to figure out where to park your money - let's say you are looking at several different money market funds for example - comparing the SEC yield easily tells you your expected distribution net of the expense ratio.
This still is not correct. SEC yield includes expected capital return if all bonds were held to maturity, and this component of capital return is not included in dividend distributions used to calculate distribution yield. SEC yield can be far different than distribution yield, and typically the previous month distribution yield is a better indicator of the next month distribution yield, although I just saw a huge jump in distribution yield for the short-term Treasury fund when I looked yesterday.

You can receive some capital return in the form of capital gain distributions. Other than that, the capital return depends on the yield curve and changes in yields, and is reflected in the change in share price. It is not "distributed" unless you sell shares.
Looking at the simple "distribution yield" you would have to do the math yourself, which many people don't realize. I didn't realize it myself until I was googling this recently for that very purpose.
Not sure what math you would do. The most recent distribution yield shown by Vanguard tells you your annualized distribution for the previous calendar month, again, net of all expenses. That's it.

I think your main point is that SEC yield is a better estimate of long-term expected return than distribution yield, and this is correct, just like yield to maturity is a much better indicator of expected return for an individual bond held to maturity than is current yield (coupon rate divided by price). But both yields are net of expenses, so that's not the issue.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Meg77
Posts: 2254
Joined: Fri May 22, 2009 1:09 pm
Location: Dallas, TX
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Meg77 » Thu Apr 19, 2018 3:36 pm

Kevin M wrote:
Thu Apr 19, 2018 2:11 pm
Meg77 wrote:
Thu Apr 19, 2018 11:11 am
I understand that, but my point is that when you are trying to figure out where to park your money - let's say you are looking at several different money market funds for example - comparing the SEC yield easily tells you your expected distribution net of the expense ratio.
This still is not correct. SEC yield includes expected capital return if all bonds were held to maturity, and this component of capital return is not included in dividend distributions used to calculate distribution yield. SEC yield can be far different than distribution yield, and typically the previous month distribution yield is a better indicator of the next month distribution yield, although I just saw a huge jump in distribution yield for the short-term Treasury fund when I looked yesterday.

You can receive some capital return in the form of capital gain distributions. Other than that, the capital return depends on the yield curve and changes in yields, and is reflected in the change in share price. It is not "distributed" unless you sell shares.
Looking at the simple "distribution yield" you would have to do the math yourself, which many people don't realize. I didn't realize it myself until I was googling this recently for that very purpose.
Not sure what math you would do. The most recent distribution yield shown by Vanguard tells you your annualized distribution for the previous calendar month, again, net of all expenses. That's it.

I think your main point is that SEC yield is a better estimate of long-term expected return than distribution yield, and this is correct, just like yield to maturity is a much better indicator of expected return for an individual bond held to maturity than is current yield (coupon rate divided by price). But both yields are net of expenses, so that's not the issue.

Kevin
Kevin, thanks for the explanation. I was talking about money market funds, not bonds, but I get your point.
"An investment in knowledge pays the best interest." - Benjamin Franklin

User avatar
Kevin M
Posts: 9419
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Kevin M » Fri Apr 20, 2018 1:46 am

Meg77 wrote:
Thu Apr 19, 2018 3:36 pm
Kevin, thanks for the explanation. I was talking about money market funds, not bonds, but I get your point.
OK, but the OP was about a bond fund.

For MM funds the SEC yield is a 7-day average of income, as opposed to 30-day average yield to maturity for bond funds. So SEC yield for MM funds tells you what you earned over the last 7 days, annualized. It doesn't tell you much about what you'll earn over the next year, since the yields can change a lot quite quickly. They've rising steadily for some time now, especially the taxable MM funds. The muni MM funds have been more erratic, but have been rising steadily recently.

With yields rising as they have been, recent MM fund distribution yields are much lower than SEC yields. For the CA muni MM fund that I'm using, March distribution yield was 1.09%, but SEC yield now is 1.43%. SEC yield is much more relevant in deciding where I want to hold my cash today. SEC yield of the fund was only 0.95% at beginning of March, so no big surprise March distribution yield was only 1.09%. SEC yield at beginning of April was 1.27%, so unless yield really drops, distribution yield for April will be significantly higher than it was for March.

Both SEC yield and distribution yield are net of expenses for MM funds too.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

Prudence
Posts: 345
Joined: Fri Mar 09, 2012 4:55 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by Prudence » Fri Apr 20, 2018 7:50 am

Does SEC yield assume no changes in interest rates (since they cannot be predicted)? So, suppose I buy a bond fund with a 3% SEC Yield and average duration of six years. If market interest rates drop one percent and I sell my position one year later, will my yield be negative for the time I held the fund?

alex_686
Posts: 3332
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by alex_686 » Fri Apr 20, 2018 8:31 am

Prudence wrote:
Fri Apr 20, 2018 7:50 am
Does SEC yield assume no changes in interest rates (since they cannot be predicted)? So, suppose I buy a bond fund with a 3% SEC Yield and average duration of six years. If market interest rates drop one percent and I sell my position one year later, will my yield be negative for the time I held the fund?
It assumes that bonds are held to maturity. Duration calculations do not come into play. SEC Yield is a yield question, duration is a risk question. They lie in 2 different dimensions.

As for not being able to predict interest rate movement, we can do a pretty decent job of that. Link to a simple tool on how to do so for the simple of case of Fed's fund rates. Now, can we predict with a higher rate of accuracy than the market?

http://www.cmegroup.com/trading/interes ... fomc.html/

User avatar
Kevin M
Posts: 9419
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Kevin M » Fri Apr 20, 2018 3:47 pm

Prudence wrote:
Fri Apr 20, 2018 7:50 am
Does SEC yield assume no changes in interest rates (since they cannot be predicted)? So, suppose I buy a bond fund with a 3% SEC Yield and average duration of six years. If market interest rates drop one percent and I sell my position one year later, will my yield be negative for the time I held the fund?
You are confusing yield and return. Yield factors in coupon rate and change in price from today until maturity. Yield is not at all predictive of 1-year return for an intermediate-term bond fund, or for an intermediate-term individual bond for that matter.

If the yield of a fund with duration of six years drops by one percentage point very quickly, say from 3% to 2%, the share price will increase by about 6%. So you will have a capital gain, but your expected return going forward from that point is lower (closer to 2% than 3%, but again, actual return can vary widely from this).

Your total return will be a combination of your capital return, positive or negative, and the income return, always positive. The income return is basically the coupon payments of the bonds in the fund (but there also is some accrual of discounts or premiums when the bonds were purchased). The capital return depends on the shape of the yield curve that's relevant to the bonds in the fund, and changes to the yields of the bonds in the fund.

Note that it is not "market interest rates", but yield of the fund that matters. "Market interest rates" has no meaning without further specification. The federal funds rate (FFR) can go up, while the 5-year or 10-year Treasury yield goes down. What matters is the net effect of what happens to the yields of the bonds the fund holds.
alex_686 wrote:
Fri Apr 20, 2018 8:31 am
[As for not being able to predict interest rate movement, we can do a pretty decent job of that. Link to a simple tool on how to do so for the simple of case of Fed's fund rates. Now, can we predict with a higher rate of accuracy than the market?

http://www.cmegroup.com/trading/interes ... fomc.html/
This is the FFR only. Even if we have high confidence that the FFR will increase on a certain date, that tells us little about what will happen to yields of bonds of longer maturities, and to yields of different types of bonds (Treasuries, munis, corporates, etc.).

Institutional bond investors know about probabilities of FFR increases or decreases, and have already factored this into today's bond prices/yields. You can see this in comparing the 1-month Treasury yield, a very short-term yield that does tend to track FFR, and the FFR itself. One-month yield tends to rise in anticipation of the increase in FFR, but there is little correlation between changes in FFR and changes in longer-term Treasuries, especially over periods of a few years or less. Here is a chart of FFR, 1-month Treasury, and 10-year Treasury:

Image

Does this make the point clear? Note that FFR is higher than at the beginning of the period, but 10-year yield is lower, and bounced all over the place while FFR and 1-month Treasury were basically flat at close to 0% for years.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

Prudence
Posts: 345
Joined: Fri Mar 09, 2012 4:55 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by Prudence » Sat Apr 21, 2018 9:52 am

Very clear as usual Kevin. If rates (say ten year Treasury) are expected to increase one percent in one year and they do, does this mean that the share price (return) of the fund should not be adversely impacted?

User avatar
Kevin M
Posts: 9419
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Kevin M » Sat Apr 21, 2018 12:41 pm

Prudence wrote:
Sat Apr 21, 2018 9:52 am
Very clear as usual Kevin. If rates (say ten year Treasury) are expected to increase one percent in one year and they do, does this mean that the share price (return) of the fund should not be adversely impacted?
I just picked the 10-year Treasury for comparison as it's widely quoted, and clearly not highly correlated with changes in FFR.

For an intermediate-term Treasury fund a 5-year or 7-year Treasury yield would be more relevant, as these would be closer to the duration of the fund. However, the fund holds a range of maturities, from about 3-10 years (for the actively manged fund), and yields can change differently at different points on that segment of the yield curve.

For a non-Treasury fund, the yields may change differently than Treasury yields, so it would be the yields of those bonds that would matter.

The duration rule of thumb is that share price changes inversely by duration percent for a one percentage point change in yield, assuming parallel shift of the relevant yield curve--i.e., all yields move up or down the same amount. So if all yields increase by one percentage point, share price of a fund with duration of five years would decrease by about 5%. Over a one-year time period, the income would offset that to some extent, so total return is capital return (price change) plus income return (dividends). If fund distributed 2% in dividends, total return would be about -3%.

Vanguard provides a breakdown by year of capital return and income return for its funds. Here it is for intermediate-term Treasury Admiral shares: https://personal.vanguard.com/us/funds/ ... rue#tab=1a. Look at 2013 for an example of negative capital return of -4.50% (due to increasing yields) plus income return of 1.51% (dividend distributions) providing a total return of -2.99%.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

Prudence
Posts: 345
Joined: Fri Mar 09, 2012 4:55 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by Prudence » Sat Apr 21, 2018 6:04 pm

I agree Kevin. (I was confused by your previous comments that institutional investors would have already anticipated and priced in the future increase in rates; I thought this meant that the bond fund price would not be negatively impacted when the rate increase actually occurred).

User avatar
Kevin M
Posts: 9419
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Kevin M » Sun Apr 22, 2018 10:53 am

Prudence wrote:
Sat Apr 21, 2018 6:04 pm
I agree Kevin. (I was confused by your previous comments that institutional investors would have already anticipated and priced in the future increase in rates; I thought this meant that the bond fund price would not be negatively impacted when the rate increase actually occurred).
What I meant was to the extent there is any likely impact of increase on the very short-term FFR rate, investors have already factored it into likely impact on longer-term yields. I think history shows that there isn't necessarily much correlation between change in FFR and change in longer-term yields, but if institutional investors disagree, they will factor that into their thinking when they bid on longer-term bonds.

So a bond fund's price indeed may not be negatively impacted when the FFR increases. It will be negatively impacted when the yields of the bonds it holds increase. We can't just talk about "rate increase", because there are many different rates (or yields), and one is not necessarily directly related to another.

The Treasury yield curve looks somewhat different than the CD yield curve which looks somewhat different from the AA muni yield curve. These yield curves will change differently over time, and the changes may not have much relationship to the change in FFR, other than at the very short end of the Treasury yield curve.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

alex_686
Posts: 3332
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by alex_686 » Sun Apr 22, 2018 2:08 pm

Kevin M wrote:
Fri Apr 20, 2018 3:47 pm
alex_686 wrote:
Fri Apr 20, 2018 8:31 am
[As for not being able to predict interest rate movement, we can do a pretty decent job of that. Link to a simple tool on how to do so for the simple of case of Fed's fund rates. Now, can we predict with a higher rate of accuracy than the market?

http://www.cmegroup.com/trading/interes ... fomc.html/
This is the FFR only. Even if we have high confidence that the FFR will increase on a certain date, that tells us little about what will happen to yields of bonds of longer maturities, and to yields of different types of bonds (Treasuries, munis, corporates, etc.).

Institutional bond investors know about probabilities of FFR increases or decreases, and have already factored this into today's bond prices/yields. You can see this in comparing the 1-month Treasury yield, a very short-term yield that does tend to track FFR, and the FFR itself. One-month yield tends to rise in anticipation of the increase in FFR, but there is little correlation between changes in FFR and changes in longer-term Treasuries, especially over periods of a few years or less. Here is a chart of FFR, 1-month Treasury, and 10-year Treasury:

...

Does this make the point clear? Note that FFR is higher than at the beginning of the period, but 10-year yield is lower, and bounced all over the place while FFR and 1-month Treasury were basically flat at close to 0% for years.

Kevin
I chose FFR because it is simple, and the CME is very kind in providing free easy to under graphs, the inputs to construct those graphs, and the white paper of the methodology with the formulas. There are similar tools out there for the 10 year treasury, corporate bonds, foreign bonds, etc. They tend to be more complex and behind paywalls. They are robust models that have been well tested.
Prudence wrote:
Sat Apr 21, 2018 9:52 am
If rates (say ten year Treasury) are expected to increase one percent in one year and they do, does this mean that the share price (return) of the fund should not be adversely impacted?
Yes - sort of. If you check out my link to the FFR you will see that there is a dispersion of results - low chance of a rate reduction, low chance of a major hike, most probabilities clustered around a modest increase. Same with the 10 year - but the results are more chaotic. Impacts are increased by a factor of 10.

User avatar
dm200
Posts: 15994
Joined: Mon Feb 26, 2007 2:21 pm
Location: Washington DC area

Re: Why is the SEC yield the one that Vanguard displays?

Post by dm200 » Sun Apr 22, 2018 3:21 pm

Vanguard also displays the distribution yield . Go to performance then distribution then list

User avatar
Kevin M
Posts: 9419
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Kevin M » Sun Apr 22, 2018 3:34 pm

alex_686 wrote:
Sun Apr 22, 2018 2:08 pm
I chose FFR because it is simple, and the CME is very kind in providing free easy to under graphs, the inputs to construct those graphs, and the white paper of the methodology with the formulas. There are similar tools out there for the 10 year treasury, corporate bonds, foreign bonds, etc. They tend to be more complex and behind paywalls. They are robust models that have been well tested.
I'm not sure what you're saying here, but if you're saying that the 1-year change in the 10-year yield can be predicted with a high level of confidence, then that goes against everything I've ever read about accuracy of economists predicting future yields.

The theories on the term structure of interest rates are not at all well supported by the empirical evidence, particularly the expectations hypothesis. Again, if anyone could do this with any reliability, they could become fabulously wealthy by just trading in the fixed income markets, and certainly wouldn't be sharing their secrets here or anywhere else.
alex_686 wrote:
Sun Apr 22, 2018 2:08 pm
Prudence wrote:
Sat Apr 21, 2018 9:52 am
If rates (say ten year Treasury) are expected to increase one percent in one year and they do, does this mean that the share price (return) of the fund should not be adversely impacted?
Yes - sort of. If you check out my link to the FFR you will see there is a dispersion of results - low chance of a rate reduction, low chance of a major hike, most probabilities clustered around a modest increase. Same with the 10 year - but the results are more chaotic. Impacts are increased by a factor of 10.
Again, I really don't know what you're trying to say here, but it seems very misleading to me. The current yield and price of the 10-year Treasury is what it is, and factors in all expectations, risk aversion, and everything else the market knows about. If a bond fund holds a 10-year Treasury, and say rolls it monthly into a new 10-year, and the 10-year yield increases by one percentage point in one year, the price will drop by the duration of the bond, so about 8%. It has nothing to do with what the market expects to happen--it just depends on what actually happens.

Here's a more precise example. If you buy a 10-year Treasury today at par (100), the yield (and coupon) is about 3%. The yield curve is pretty flat at around 10 years, so let's just assume 9-year Treasury also is 3%. If you hold the 10-year for one year, you now have a 9-year Treasury. If 9-year yield increases to 4% in one year, the value of your bond drops from 100 to 92.5, for a capital return of -7.5% (a loss). Your 3% coupon brings your total return to -4.5%.

Any fund holding this Treasury for one year will have a total return of -4.5% contributed to the return of the fund. The total return of the fund will depend on the coupon rates and what happens to the yields of the other bonds it owns.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Doc
Posts: 8134
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Why is the SEC yield the one that Vanguard displays?

Post by Doc » Sun Apr 22, 2018 5:27 pm

Kevin M wrote:
Sun Apr 22, 2018 3:34 pm
alex_686 wrote: ↑Sun Apr 22, 2018 2:08 pm
I chose FFR because it is simple, and the CME is very kind in providing free easy to under graphs, the inputs to construct those graphs, and the white paper of the methodology with the formulas. There are similar tools out there for the 10 year treasury, corporate bonds, foreign bonds, etc. They tend to be more complex and behind paywalls. They are robust models that have been well tested.
I'm not sure what you're saying here, but if you're saying that the 1-year change in the 10-year yield can be predicted with a high level of confidence, then that goes against everything I've ever read about accuracy of economists predicting future yields.
The FedWatch Tool from the CME group is a prediction of the Federal Funds Rate for the upcoming Open Market Committee meetings with probability of the FFR being in a certain 25 bps range. That's the range the Open Market Committee uses as its target. It is a near term projection not a ten year projection. I am not familiar with the "similar tool out there" that alex mentioned for longer term projections.

Check out the link. If nothing else it's interesting.
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

alex_686
Posts: 3332
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by alex_686 » Mon Apr 23, 2018 9:46 am

So let us start at square one.

Can we predict future changes in interest rates? Yes. Or maybe it would be better to say that we can derive the market expectations of interest rate changes and that these changes are already Incorporated into the market price.

The tool that I pointed to is for short term rates. I think it is a good demonstration of the principle.

However the methodology used for short term rates can be used for long term rates. Same method, different inputs. There are canned reports on Bloomberg Terminals. I have built spreadsheets that do this.

The methodology relies on "rational exceptions", not "term structure". Not sure if we want to enter into that philosophical debate in this thread.

You might want to revisit your econ books. It is a little like weather reporting. It can tell me that there is a 97% chance of rain tomorrow at noon, and a 60% chance of rain 5 days from now. In both cases the models are well calibrated. Weather forecasting wins, but interest rates do a pretty good job. Another analogy is earnings forecast. Once again the market is pretty good with those. Consistently optimistic, sometimes very wrong, but better than trailing PE ratios. Just because we have accurate forecasted earnings does not mean we can make free money.

There are vast sums to be made in the bond market and it often comes down to a few bps. Everybody knows how to calculate a volatility surface - another canned report that Bloomberg offers.

User avatar
Kevin M
Posts: 9419
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Kevin M » Mon Apr 23, 2018 11:11 am

You can use all the fancy jargon you want, and read all the econ books you want, but your answer to the question asked about impact on bond fund value was just plain wrong. Regardless of how well you think you can predict change in the 10-year yield, if the 10-year yield increases by 1 percentage point in one year, a 10-year bond bought or valued today will lose value, and that loss in value will be reflected in any bond fund that holds it.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

lack_ey
Posts: 6450
Joined: Wed Nov 19, 2014 11:55 pm

Re: Why is the SEC yield the one that Vanguard displays?

Post by lack_ey » Mon Apr 23, 2018 11:31 am

Right, the prompt does not matter. If the yield increases, that is equivalent to saying the price dropped.
Prudence wrote:
Sat Apr 21, 2018 9:52 am
Very clear as usual Kevin. If rates (say ten year Treasury) are expected to increase one percent in one year and they do, does this mean that the share price (return) of the fund should not be adversely impacted?
The thing is that this is an expectation that doesn't make much sense, and maybe that is throwing people off.

While sure, some may expect the 10-year Treasury yield to rise by 1% in a year, this would be odd as a market-wide average opinion. If that were really true, many investors (maybe not central banks holding these for policy reasons or some very short-term traders using as collateral or something else, but those seeking profit and with some flexibility) would want to sell their 10-year Treasuries and buy them again in a year when the rate is higher, avoiding the price drop of 7-8% along the way. And if many people want to sell at the current price and are even willing to take lower prices, what happens? The price goes down, and the yield goes up...

That really shouldn't happen unless with strange risk pricing. Like people accept that they'll probably lose money but take the current prices because the upside on the small (significantly sub-50%) chance that rates don't change or even go down matters a lot more to them.

It's quite another thing to have expectations of very short rates changing by 1%. That's not supposing that bonds lose value, or at least not by much. In of itself it doesn't, but it depends on the yields across the yield curve. Generally if short rates are expected to go up by 1%, we should see that the 1-year Treasury is priced accordingly, with yield higher than the 1-month Treasury such that owning 12 different 1-month Treasuries over the course of a year isn't obviously way better than holding the 1-year.

User avatar
Kevin M
Posts: 9419
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Why is the SEC yield the one that Vanguard displays?

Post by Kevin M » Tue Apr 24, 2018 3:21 pm

lack_ey wrote:
Mon Apr 23, 2018 11:31 am
Prudence wrote:
Sat Apr 21, 2018 9:52 am
Very clear as usual Kevin. If rates (say ten year Treasury) are expected to increase one percent in one year and they do, does this mean that the share price (return) of the fund should not be adversely impacted?
The thing is that this is an expectation that doesn't make much sense, and maybe that is throwing people off.<snip>
Exactly--to this and the rest of your reply. I had intended to come back and discuss this as well, but you did a great job of it.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

Post Reply