## SEC 30 day yield vs Distribution yield

### SEC 30 day yield vs Distribution yield

I have an Ally CD maturing on 9/5. As of a 2 weeks ago when I talked with them, the best they had to offer was a 15 month Select CD with a 0.95% rate and with the loyalty reward of 5 basis points it would be 1%. This also is about money in a rollover IRA.

I have looked at 3 bond funds at Vanguard (Vanguard Intermediate-Term Tax-Exempt Fund Admiral Shares (VWIUX), Vanguard Short-Term Investment-Grade Fund Admiral Shares (VFSUX), Vanguard Intermediate-Term Investment-Grade Fund Admiral Shares (VFIDX)) and the SEC yield is always way less than the Distribution yield. I assume you get the Distribution yield at month end. I realize the Distribution yield on these 3 funds has dropped a bit each month, but it is way higher than the SEC and way higher than the Ally CD.

I know a CD has FDIC insurance and that the bond fund can lose or gain in nav effecting the yield each month but it seems like I'd earn far more in one of the bond funds. This is for taxable money I don't have a use for and for money in a tax deferred account.

So which yield on the bond funds is accurate, SEC or Distribution or neither!

I have looked at 3 bond funds at Vanguard (Vanguard Intermediate-Term Tax-Exempt Fund Admiral Shares (VWIUX), Vanguard Short-Term Investment-Grade Fund Admiral Shares (VFSUX), Vanguard Intermediate-Term Investment-Grade Fund Admiral Shares (VFIDX)) and the SEC yield is always way less than the Distribution yield. I assume you get the Distribution yield at month end. I realize the Distribution yield on these 3 funds has dropped a bit each month, but it is way higher than the SEC and way higher than the Ally CD.

I know a CD has FDIC insurance and that the bond fund can lose or gain in nav effecting the yield each month but it seems like I'd earn far more in one of the bond funds. This is for taxable money I don't have a use for and for money in a tax deferred account.

So which yield on the bond funds is accurate, SEC or Distribution or neither!

### Re: SEC 30 day yield vs Distribution yield

Zaplunken,

I use distribution yield with VWIUX...because that's what you are actually paid. I watch it every month. If it goes down far enough, I may or may not move on.

Look at Vanguard's formula to figure distribution yield. One of the largest factors is the price.

Also look at the amount paid per share. Note the amount paid for share hasn't gone down much in the last 18 months. The real number to watch might be the amount paid per share.

Once you own the fund, you might want to figure your own personal distribution yield using the price you paid?

The SEC yield is not what you get paid. It's great for comparing funds.

I use distribution yield with VWIUX...because that's what you are actually paid. I watch it every month. If it goes down far enough, I may or may not move on.

Look at Vanguard's formula to figure distribution yield. One of the largest factors is the price.

Also look at the amount paid per share. Note the amount paid for share hasn't gone down much in the last 18 months. The real number to watch might be the amount paid per share.

Once you own the fund, you might want to figure your own personal distribution yield using the price you paid?

The SEC yield is not what you get paid. It's great for comparing funds.

### Re: SEC 30 day yield vs Distribution yield

Thanks. I take it your comments apply to only the tax exempt bond fund? That would be where I am considering put the maturing CD money since it is in a taxable account but that is just under $30k so I'd have to get the Investor shares, I talked about the Admiral shares in my OP.hudson wrote: ↑Sun Aug 30, 2020 3:06 pm Zaplunken,

I use distribution yield with VWIUX...because that's what you are actually paid. I watch it every month. If it goes down far enough, I may or may not move on.

Look at Vanguard's formula to figure distribution yield. One of the largest factors is the price.

Also look at the amount paid per share. Note the amount paid for share hasn't gone down much in the last 18 months. The real number to watch might be the amount paid per share.

Once you own the fund, you might want to figure your own personal distribution yield using the price you paid?

The SEC yield is not what you get paid. It's great for comparing funds.

Where do I find Vanguard's formula to figure distribution yield?

The IRA money probably would go to the 2 Investment Grade funds. I always heard it makes no sense to put tax free funds into a tax deferred account since the distributions will be taxed when RMDs occur but for a good yield I might consider the tax exempt for the IRA though right now the other 2 have a better yield.

### Re: SEC 30 day yield vs Distribution yield

Distribution yield is cash distributions.

SEC 30 day yield is essentially a 30 day average yield to maturity. Among other things, it takes into account that you might pay more or less than the face amount of the bond, but that at maturity the bond will be redeemed at the face amount (although funds don't usually hold to maturity).

Which is a better indicia of future returns is the subject of debate. Many look at both.

Vanguard reports the numbers on the web page for the relevant fund.

SEC 30 day yield is essentially a 30 day average yield to maturity. Among other things, it takes into account that you might pay more or less than the face amount of the bond, but that at maturity the bond will be redeemed at the face amount (although funds don't usually hold to maturity).

Which is a better indicia of future returns is the subject of debate. Many look at both.

Vanguard reports the numbers on the web page for the relevant fund.

### Re: SEC 30 day yield vs Distribution yield

Most people consider SEC yield, which is based on the yield to maturity of the bonds in the fund, as a better indicator of expected total return than distribution yield.

You can see the distribution history for the last 18 months on the Distributions tab of the fund's webpage. Here it is for VFIDX: https://investor.vanguard.com/mutual-fu ... ions/vfidx. Click on the "Distribution yield" column header to see Vanguard's definition of it:

Say you buy a 1-year bond with a 2% yield to maturity (YTM) and a 2% coupon rate. You will pay 100 (100% of face value), because that's the price of a bond for which the coupon rate equals the YTM. The current yield also is 2%, which is the annual interest of 2 (per 100) divided by the price, so 2/100 = 2%.

Say the next day the YTM of 1-year bonds drops to 1%. The price of your bond will increase to about 101. The current yield still is about 2% (2/101 = 1.98%), but that's not what you, or someone you sell the bond to, will earn over the next year. You, or whoever buys the bond, will earn 2% from the coupon payments but lose about 1% in capital, when the bond matures at 100. It should be clear that the 1% YTM is much more relevant at this point than the current yield of 2%.

Looking at only "what you get paid" in interest is only part of your return. The other part is what you lose in capital as the bond drops from 101 today to 100 at maturity.

This is exactly what's going on with bonds in a fund for which the distribution yield is higher than the SEC yield (which again, is based on the YTM of the bonds in the fund). The bonds have increased in value since purchase due to a decrease in yields, resulting in in current yields that are higher than YTMs, and distribution yield that is higher than SEC yield.

Over time, the bond prices are pulled toward par (100) as the bonds approach maturity, which contributes to a decline in NAV over time. Of course in the shorter term, this effect can be swamped out by further decreases in yields, leading to temporarily higher prices and NAV.

There can also be a rolldown return component of return, which is a potential increase in price as bonds roll down a positively-sloped yield curve, and since most bond funds sell bonds before maturity, this can push prices up as a counter force to the pull toward par effect. This can be a reason that SEC yield or average YTM might understate expected return by some amount, and possibly contribute to distribution yields remaining higher than SEC yields for longer than one might expect.

Kevin

You can see the distribution history for the last 18 months on the Distributions tab of the fund's webpage. Here it is for VFIDX: https://investor.vanguard.com/mutual-fu ... ions/vfidx. Click on the "Distribution yield" column header to see Vanguard's definition of it:

To understand why SEC yield is more relevant than distribution yield in terms of expected total return, it's useful to understand that distribution yield for a bond fund is similar to current yield for an individual bond. It's easier to understand why yield to maturity is more relevant than current yield with respect to the total expected return of the bond.The fund's current monthly income dividend per share, annualized by dividing by the number of days in the month and multiplying by 365, and shown as a percentage of the fund's average net asset value (NAV) during the month.

Say you buy a 1-year bond with a 2% yield to maturity (YTM) and a 2% coupon rate. You will pay 100 (100% of face value), because that's the price of a bond for which the coupon rate equals the YTM. The current yield also is 2%, which is the annual interest of 2 (per 100) divided by the price, so 2/100 = 2%.

Say the next day the YTM of 1-year bonds drops to 1%. The price of your bond will increase to about 101. The current yield still is about 2% (2/101 = 1.98%), but that's not what you, or someone you sell the bond to, will earn over the next year. You, or whoever buys the bond, will earn 2% from the coupon payments but lose about 1% in capital, when the bond matures at 100. It should be clear that the 1% YTM is much more relevant at this point than the current yield of 2%.

Looking at only "what you get paid" in interest is only part of your return. The other part is what you lose in capital as the bond drops from 101 today to 100 at maturity.

This is exactly what's going on with bonds in a fund for which the distribution yield is higher than the SEC yield (which again, is based on the YTM of the bonds in the fund). The bonds have increased in value since purchase due to a decrease in yields, resulting in in current yields that are higher than YTMs, and distribution yield that is higher than SEC yield.

Over time, the bond prices are pulled toward par (100) as the bonds approach maturity, which contributes to a decline in NAV over time. Of course in the shorter term, this effect can be swamped out by further decreases in yields, leading to temporarily higher prices and NAV.

There can also be a rolldown return component of return, which is a potential increase in price as bonds roll down a positively-sloped yield curve, and since most bond funds sell bonds before maturity, this can push prices up as a counter force to the pull toward par effect. This can be a reason that SEC yield or average YTM might understate expected return by some amount, and possibly contribute to distribution yields remaining higher than SEC yields for longer than one might expect.

Kevin

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

### Re: SEC 30 day yield vs Distribution yield

Is there support for saying "most"? I haven't seen survey data and there are many people who focus on distribution yield.

Ignoring amortization of premium/discount doesn't seem wise, but that's a different question.

### Re: SEC 30 day yield vs Distribution yield

Wow. Thanks for the explanation, Kevin. I've been wanted to ask this same question. I have pasted your post into a Word doc and will re-read many times as I try to understand (have always had problems understanding bond funds).Kevin M wrote: ↑Sun Aug 30, 2020 4:36 pm Most people consider SEC yield, which is based on the yield to maturity of the bonds in the fund, as a better indicator of expected total return than distribution yield.

You can see the distribution history for the last 18 months on the Distributions tab of the fund's webpage. Here it is for VFIDX: https://investor.vanguard.com/mutual-fu ... ions/vfidx. Click on the "Distribution yield" column header to see Vanguard's definition of it:To understand why SEC yield is more relevant than distribution yield in terms of expected total return, it's useful to understand that distribution yield for a bond fund is similar to current yield for an individual bond. It's easier to understand why yield to maturity is more relevant than current yield with respect to the total expected return of the bond.The fund's current monthly income dividend per share, annualized by dividing by the number of days in the month and multiplying by 365, and shown as a percentage of the fund's average net asset value (NAV) during the month.

Say you buy a 1-year bond with a 2% yield to maturity (YTM) and a 2% coupon rate. You will pay 100 (100% of face value), because that's the price of a bond for which the coupon rate equals the YTM. The current yield also is 2%, which is the annual interest of 2 (per 100) divided by the price, so 2/100 = 2%.

Say the next day the YTM of 1-year bonds drops to 1%. The price of your bond will increase to about 101. The current yield still is about 2% (2/101 = 1.98%), but that's not what you, or someone you sell the bond to, will earn over the next year. You, or whoever buys the bond, will earn 2% from the coupon payments but lose about 1% in capital, when the bond matures at 100. It should be clear that the 1% YTM is much more relevant at this point than the current yield of 2%.

Looking at only "what you get paid" in interest is only part of your return. The other part is what you lose in capital as the bond drops from 101 today to 100 at maturity.

This is exactly what's going on with bonds in a fund for which the distribution yield is higher than the SEC yield (which again, is based on the YTM of the bonds in the fund). The bonds have increased in value since purchase due to a decrease in yields, resulting in in current yields that are higher than YTMs, and distribution yield that is higher than SEC yield.

Over time, the bond prices are pulled toward par (100) as the bonds approach maturity, which contributes to a decline in NAV over time. Of course in the shorter term, this effect can be swamped out by further decreases in yields, leading to temporarily higher prices and NAV.

There can also be a rolldown return component of return, which is a potential increase in price as bonds roll down a positively-sloped yield curve, and since most bond funds sell bonds before maturity, this can push prices up as a counter force to the pull toward par effect. This can be a reason that SEC yield or average YTM might understate expected return by some amount, and possibly contribute to distribution yields remaining higher than SEC yields for longer than one might expect.

Kevin

Not to hijack the OP's thread - but is it correct to assume that the yield of Intermediate Bond funds in a no-Fed-interest-rate-increase future will continue to decrease?

### Re: SEC 30 day yield vs Distribution yield

Yes, thanks Kevin. I wish I understood this as well as you do. I can follow some of your post but honestly it is confusing. I need to reread it a few times to see if I can fully understand.

Bottom line is I want to put some money in both a taxable and IRA account that pays a better yield than what I see in other vehicles. Honestly, I am looking for a better interest rate than 1%. This isn't going to be a long term holding like 5 or 10 years. I understand the nav can decrease and that can offset the interest rate looking so much better but no one know how much or fast rates will change. It seems like a no brainer that the tax exempt fund for the maturing CD in a taxable account at Ally is so much better re the interest vs another Ally CD. For the IRA, any of the 3 bond funds look better than a fund whose yield is 1% or less. I know there is a right and a wrong choice here and I prefer to make the right one that's why I am asking.

Bottom line is I want to put some money in both a taxable and IRA account that pays a better yield than what I see in other vehicles. Honestly, I am looking for a better interest rate than 1%. This isn't going to be a long term holding like 5 or 10 years. I understand the nav can decrease and that can offset the interest rate looking so much better but no one know how much or fast rates will change. It seems like a no brainer that the tax exempt fund for the maturing CD in a taxable account at Ally is so much better re the interest vs another Ally CD. For the IRA, any of the 3 bond funds look better than a fund whose yield is 1% or less. I know there is a right and a wrong choice here and I prefer to make the right one that's why I am asking.

### Re: SEC 30 day yield vs Distribution yield

Zaplunken,

I think that VWITX is a good choice for taxable. It's not as safe as an FDIC CD or savings account. I've owned it (VWIUX) for years. I think that it's OK. William Bernstein said that it's OK to hold munis as long as you have some treasuries and CDs to back them up. I watch VWIUX closely; as long as it's a good deal, I'm going to stay.

KevinM's explanation is exactly right. He and Grabiner have explained it to me several times. I'm starting to understand. I still think that if you follow the payouts and the share price, you'll be OK. I always compare VWIUX with CDs because if I didn't own VWIUX, I would move the funds to a CD. Right now VWIUX is riskier than a CD, but it pays you to take the risk.

I think that VWITX is a good choice for taxable. It's not as safe as an FDIC CD or savings account. I've owned it (VWIUX) for years. I think that it's OK. William Bernstein said that it's OK to hold munis as long as you have some treasuries and CDs to back them up. I watch VWIUX closely; as long as it's a good deal, I'm going to stay.

KevinM's explanation is exactly right. He and Grabiner have explained it to me several times. I'm starting to understand. I still think that if you follow the payouts and the share price, you'll be OK. I always compare VWIUX with CDs because if I didn't own VWIUX, I would move the funds to a CD. Right now VWIUX is riskier than a CD, but it pays you to take the risk.

- abuss368
**Posts:**21515**Joined:**Mon Aug 03, 2009 2:33 pm**Location:**Where the water is warm, the drinks are cold, and I don't know the names of the players!-
**Contact:**

### Re: SEC 30 day yield vs Distribution yield

I believe the SEC 30 yield is the more important of the two. Kevin has a very good example upthread.

John C. Bogle: “Simplicity is the master key to financial success."

### Re: SEC 30 day yield vs Distribution yield

Thanks hudson. I do have 2 treasury funds at Vanguard and several CDs at Ally. All the CDs have fairly decent interest rates from 1.50% to 2.82% (that unfortunately is the one maturing). Going to 0.90% or 0.95% or 1%, well it seems like there must be a better way abet one that carries more risk. The maturing CD isn't needed for living expenses so I'm comfortable chasing a bit more yield and if worse comes to worse I can always sell shares should the need arise.hudson wrote: ↑Sun Aug 30, 2020 6:54 pm Zaplunken,

I think that VWITX is a good choice for taxable. It's not as safe as an FDIC CD or savings account. I've owned it (VWIUX) for years. I think that it's OK. William Bernstein said that it's OK to hold munis as long as you have some treasuries and CDs to back them up. I watch VWIUX closely; as long as it's a good deal, I'm going to stay.

KevinM's explanation is exactly right. He and Grabiner have explained it to me several times. I'm starting to understand. I still think that if you follow the payouts and the share price, you'll be OK. I always compare VWIUX with CDs because if I didn't own VWIUX, I would move the funds to a CD. Right now VWIUX is riskier than a CD, but it pays you to take the risk.

### Re: SEC 30 day yield vs Distribution yield

Did you mean that you are going to hold for at least 5-10 years? The duration of intermediate-term funds is in the 5-6 year range, and we commonly see advice to not hold a fund with a duration longer than your expected holding period.zaplunken wrote: ↑Sun Aug 30, 2020 6:23 pm Yes, thanks Kevin. I wish I understood this as well as you do. I can follow some of your post but honestly it is confusing. I need to reread it a few times to see if I can fully understand.

Bottom line is I want to put some money in both a taxable and IRA account that pays a better yield than what I see in other vehicles. Honestly, I am looking for a better interest rate than 1%. This isn't going to be a long term holding like 5 or 10 years. I understand the nav can decrease and that can offset the interest rate looking so much better but no one know how much or fast rates will change. It seems like a no brainer that the tax exempt fund for the maturing CD in a taxable account at Ally is so much better re the interest vs another Ally CD. For the IRA, any of the 3 bond funds look better than a fund whose yield is 1% or less. I know there is a right and a wrong choice here and I prefer to make the right one that's why I am asking.

You need to consider risk and expected total return, not just interest rates. One risk factor to consider is maximum drawdown, which for VWITX (int-term muni, investor shares) was -6.8%, with total under water period of 11 months. Source: Portfolio Visualizer backtest.

Another thing we can look at is the dispersion of 5-year or 10-year returns compared to initial SEC yield (since it's so common to use SEC yield to estimate expected return). A commonly quoted statistic is that there's something like 95% correlation between initial yield to maturity and subsequent 10-year return for the aggregate US bond market, but my research indicates that the high correlation doesn't translate into high prediction accuracy. For example, here's a chart showing 10-year annualized return minus initial SEC yield for Vanguard total bond fund (VBMFX) for as many years as we have SEC yield data for the fund:

For example, the SEC yield on 12/29/2000 was 6.67%, and the 10-year annualized return for 2001-2010 was 5.57%, so one percentage point lower than the initial yield. We see that as the blue bar for 2001 going down to about -1.00% in the chart.

A way to look at the dispersion of returns is to plot a histogram of the 10-year returns minus initial yields. Using bins of 10 basis points, here's what I come up with:

If 10-year returns were clustered around the initial yield, we'd see something that looked more like a typical bell curve, with lots of occurrences near the mean, which happens to be -0.2% for this data sample, but we don't see that. Instead we see return deltas scattered more uniformly from the low to the high, apparently with about as much chance as being about 1 percentage point above or below the mean as very near the mean.

So with an initial SEC yield in the 1% ballpark (it's 0.92% for VWITX), we might expect to earn a 10-year annualized return anywhere from about 0% to 2%, and it's tough to say that any return in that range is more likely than any other. It's somewhat harder to do after-tax or taxable-equivalent return estimates, since the dividends are tax exempt but the capital return is subject to capital gains taxes, and we are taxed on the distribution yield, not the SEC yield.

The high distribution yields do make the bond funds seem more tempting--I've argued the case myself about distribution yields appearing to remain stubbornly higher than one would think they would based on bond math, but it's hard to sort through all of the variables causing this to be the case, such as continuously changing yields, rolldown return, etc. But the bond math applies to the bonds in the fund, so it's hard to see how the NAV can't eventually fall as the bond premiums bleed off as the bonds approach maturity.

As another example, consider these characteristics of VWITX as of 7/31/2020:

YTM: 1.1%

Average coupon: 4.4%

Average maturity: 8.9 years

Note that the average coupon is much higher than the YTM, which means the fund holds on average bonds with large price premiums (values above 100). Plugging the values above into a PV function, we get a price of 127.83 and a current yield of 3.44% (and the high current yield contributes to a high distribution yield). But we know that over 8.9 years, the value of this bond would fall to 100 at maturity. The combination of the 4.4 per 100 of interest per year and the negative capital return of 27.83 per 100 over 8.9 years gets us to a 1.1% return, equivalent to the initial yield.

In contrast to a bond fund, for which there is a fairly wide range of possible returns, a CD gives you an almost certain nominal return for the term of the CD. A 5-year CD at 1.5% will almost certainly earn 1.5% over five years. A more appropriate comparison for a CD is to a Treasury of the same term, and the yield on a 5-year Treasury is 0.28%, so you are much better off with the CD if you plan to hold to maturity. Any additional yield above the Treasury yield comes with additional risk.

Kevin

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

### Re: SEC 30 day yield vs Distribution yield

zaplunken

It looks like SEC yield and distribution yield have been explored pretty well. I don't think anyone has brought up the part where part of VWITX/VWIUX's payout is from the principal. That usually comes up in this discussion. The simple answer to that is yes; Vanguard doesn't do it to artificially pump up distribution yield. It has something to do with the purchase price of a bond. If the fund profits because of bond pricing, it is required to pay that profit as interest. Somebody correct me if I'm wrong.

KevinM's posts have helped me slide into to great deals on CDs through the years. You only asked about SEC vs distribution yields. You didn't ask what the best options might be. Maybe what are the best FDIC options and what are the best mutual fund options? If I had a CD maturing on Sep. 5, I would search for the best 5 year CD available and put that up against VWITX/VWIUX....if the math worked out as far as tax brackets. I would also look at inflation protect products. If I was forced to decide right at this moment, I would dump it all in VWIUX while I made up my mind.

VWITX/VWIUX https://investor.vanguard.com/mutual-fu ... file/VWITX

Note: to find Vanguard's definition of Distribution Yield, go to the Distributions Tab; go to the far right column, and click on Distribution Yield.

Personal Distribution Yield: After you buy a fund, you can take the above formula and plug in the price you paid per share instead of the "fund's average net asset value."The fund's current monthly income dividend per share, annualized by dividing by the number of days in the month and multiplying by 365, and shown as a percentage of the fund's average net asset value (NAV) during the month.

### Re: SEC 30 day yield vs Distribution yield

Thanks Kevin and hudson. As usual, when looking at and deciding about whether to select a bond fund or which bond fund, it is confusing and loaded with all different what ifs.

At this point, I am only looking for a place to put the cash from the Ally taxable account CD and to rebalance my AA by exchanging some equity into FI in my IRA. I can earn more in equities but I don't want to have my equity allocation be higher than it is. It is 46/54 and I am going to rebalance back to 40/60 which is why I want to select a bond fund exchanging from the IRA TSMI holding. How long will I keep the money in whatever fund or funds in the taxable and IRA? I don't know. Probably my best answer is until I find something that is better or until the selected fund or funds develop problems where they aren't competitive or my AA drops below 40/60 and I can rebalance some of that FI back into the TSMI.

TSMI is Vanguard's Total Stock Market Index

At this point, I am only looking for a place to put the cash from the Ally taxable account CD and to rebalance my AA by exchanging some equity into FI in my IRA. I can earn more in equities but I don't want to have my equity allocation be higher than it is. It is 46/54 and I am going to rebalance back to 40/60 which is why I want to select a bond fund exchanging from the IRA TSMI holding. How long will I keep the money in whatever fund or funds in the taxable and IRA? I don't know. Probably my best answer is until I find something that is better or until the selected fund or funds develop problems where they aren't competitive or my AA drops below 40/60 and I can rebalance some of that FI back into the TSMI.

TSMI is Vanguard's Total Stock Market Index

### Re: SEC 30 day yield vs Distribution yield

I agree that is true, and I think it's very perceptive, but I've had a hard time finding data or research to support that conclusion in the context of bond funds. Are you aware of any?

Last edited by mptfan on Mon Aug 31, 2020 1:15 pm, edited 1 time in total.

### Re: SEC 30 day yield vs Distribution yield

deleted

Last edited by Seasonal on Mon Aug 31, 2020 8:40 am, edited 1 time in total.

### Re: SEC 30 day yield vs Distribution yield

The SEC Yield is an indicator of what you may or might receive as a distribution some time in the future.. not necessarily sooner than later either. The present SEC on VWIUX is 1.00%, last month the distribution was 2.39% tax-exempt. Personally, I evaluate on what I'm going to receive for my money, not some statistical number that may or may not be true in the future.

### Re: SEC 30 day yield vs Distribution yield

I generally agree with this.midareff wrote: ↑Mon Aug 31, 2020 8:33 am The SEC Yield is an indicator of what you may or might receive as a distribution some time in the future.. not necessarily sooner than later either. The present SEC on VWIUX is 1.00%, last month the distribution was 2.39% tax-exempt. Personally, I evaluate on what I'm going to receive for my money, not some statistical number that may or may not be true in the future.

### Re: SEC 30 day yield vs Distribution yield

The excess of the distribution yield over the SEC yield is offset by a decline in the NAV of the fund. It's an illusion to think you are earning an above market yield because the fund happens to hold bonds with a higher coupon than current market rates. Part of the distribution yield is a return of principal (and not interest income). I own dozens of individual muni bonds and have seen this play out for years now.midareff wrote: ↑Mon Aug 31, 2020 8:33 am The SEC Yield is an indicator of what you may or might receive as a distribution some time in the future.. not necessarily sooner than later either. The present SEC on VWIUX is 1.00%, last month the distribution was 2.39% tax-exempt. Personally, I evaluate on what I'm going to receive for my money, not some statistical number that may or may not be true in the future.

Kevin did a great job explaining this. I've tried to do the same thing myself a number of times, such as in this thread:

viewtopic.php?f=1&t=305142&p=5056107&hi ... d#p5056107

Real Knowledge Comes Only From Experience

### Re: SEC 30 day yield vs Distribution yield

If rates don't move, the market value of a bond will move towards its face amount, because that's what the holder will get at maturity.

The complication is that rates do move. In particular, if rates drop the market value of the bond will increase, perhaps enough to offset the decline in value as the bond moves towards maturity. If the bond is sold at that point, you may get both the distribution yield and a price at or above face amount. The decline in rates means that newly purchased bonds will have a lower stated yield than the prior bonds.

Rates have been dropping for a while, but they are not likely to continue dropping forever.

The complication is that rates do move. In particular, if rates drop the market value of the bond will increase, perhaps enough to offset the decline in value as the bond moves towards maturity. If the bond is sold at that point, you may get both the distribution yield and a price at or above face amount. The decline in rates means that newly purchased bonds will have a lower stated yield than the prior bonds.

Rates have been dropping for a while, but they are not likely to continue dropping forever.

### Re: SEC 30 day yield vs Distribution yield

And if rates rise the opposite happens.Seasonal wrote: ↑Mon Aug 31, 2020 9:38 am If rates don't move, the market value of a bond will move towards its face amount, because that's what the holder will get at maturity.

The complication is that rates do move. In particular, if rates drop the market value of the bond will increase, perhaps enough to offset the decline in value as the bond moves towards maturity. If the bond is sold at that point, you may get both the distribution yield and a price at or above face amount. The decline in rates means that newly purchased bonds will have a lower stated yield than the prior bonds.

Rates have been dropping for a while, but they are not likely to continue dropping forever.

If one wants to invest based upon their ability to predict the future movement of interest rates that is their choice. I am not aware of anyone who has demonstrated an ability to do that accurately and/or consistently.

Regardless, the bonds price will as you rightly say ultimately move to its par value as it nears and then reaches its maturity date. This is inevitable since the bond will only pay par at maturity.

Real Knowledge Comes Only From Experience

### Re: SEC 30 day yield vs Distribution yield

Of course.MikeG62 wrote: ↑Mon Aug 31, 2020 10:12 amAnd if rates rise the opposite happens.Seasonal wrote: ↑Mon Aug 31, 2020 9:38 am If rates don't move, the market value of a bond will move towards its face amount, because that's what the holder will get at maturity.

The complication is that rates do move. In particular, if rates drop the market value of the bond will increase, perhaps enough to offset the decline in value as the bond moves towards maturity. If the bond is sold at that point, you may get both the distribution yield and a price at or above face amount. The decline in rates means that newly purchased bonds will have a lower stated yield than the prior bonds.

Rates have been dropping for a while, but they are not likely to continue dropping forever.

Did someone suggest investing based on interest rate predictions? Someone asked for historical evidence of bonds dropping toward par as they mature and the general decline in rates over the past decades has muddied the evidence for the reasons set forth above.

See my post at 5:35pm yesterday and Kevin's immediately following post for the same thought.

### Re: SEC 30 day yield vs Distribution yield

Just want to say thanks!Kevin M wrote: ↑Sun Aug 30, 2020 8:45 pmDid you mean that you are going to hold for at least 5-10 years? The duration of intermediate-term funds is in the 5-6 year range, and we commonly see advice to not hold a fund with a duration longer than your expected holding period.zaplunken wrote: ↑Sun Aug 30, 2020 6:23 pm Yes, thanks Kevin. I wish I understood this as well as you do. I can follow some of your post but honestly it is confusing. I need to reread it a few times to see if I can fully understand.

Bottom line is I want to put some money in both a taxable and IRA account that pays a better yield than what I see in other vehicles. Honestly, I am looking for a better interest rate than 1%. This isn't going to be a long term holding like 5 or 10 years. I understand the nav can decrease and that can offset the interest rate looking so much better but no one know how much or fast rates will change. It seems like a no brainer that the tax exempt fund for the maturing CD in a taxable account at Ally is so much better re the interest vs another Ally CD. For the IRA, any of the 3 bond funds look better than a fund whose yield is 1% or less. I know there is a right and a wrong choice here and I prefer to make the right one that's why I am asking.

You need to consider risk and expected total return, not just interest rates. One risk factor to consider is maximum drawdown, which for VWITX (int-term muni, investor shares) was -6.8%, with total under water period of 11 months. Source: Portfolio Visualizer backtest.

Another thing we can look at is the dispersion of 5-year or 10-year returns compared to initial SEC yield (since it's so common to use SEC yield to estimate expected return). A commonly quoted statistic is that there's something like 95% correlation between initial yield to maturity and subsequent 10-year return for the aggregate US bond market, but my research indicates that the high correlation doesn't translate into high prediction accuracy. For example, here's a chart showing 10-year annualized return minus initial SEC yield for Vanguard total bond fund (VBMFX) for as many years as we have SEC yield data for the fund:

For example, the SEC yield on 12/29/2000 was 6.67%, and the 10-year annualized return for 2001-2010 was 5.57%, so one percentage point lower than the initial yield. We see that as the blue bar for 2001 going down to about -1.00% in the chart.

A way to look at the dispersion of returns is to plot a histogram of the 10-year returns minus initial yields. Using bins of 10 basis points, here's what I come up with:

If 10-year returns were clustered around the initial yield, we'd see something that looked more like a typical bell curve, with lots of occurrences near the mean, which happens to be -0.2% for this data sample, but we don't see that. Instead we see return deltas scattered more uniformly from the low to the high, apparently with about as much chance as being about 1 percentage point above or below the mean as very near the mean.

So with an initial SEC yield in the 1% ballpark (it's 0.92% for VWITX), we might expect to earn a 10-year annualized return anywhere from about 0% to 2%, and it's tough to say that any return in that range is more likely than any other. It's somewhat harder to do after-tax or taxable-equivalent return estimates, since the dividends are tax exempt but the capital return is subject to capital gains taxes, and we are taxed on the distribution yield, not the SEC yield.

The high distribution yields do make the bond funds seem more tempting--I've argued the case myself about distribution yields appearing to remain stubbornly higher than one would think they would based on bond math, but it's hard to sort through all of the variables causing this to be the case, such as continuously changing yields, rolldown return, etc. But the bond math applies to the bonds in the fund, so it's hard to see how the NAV can't eventually fall as the bond premiums bleed off as the bonds approach maturity.

As another example, consider these characteristics of VWITX as of 7/31/2020:

YTM: 1.1%

Average coupon: 4.4%

Average maturity: 8.9 years

Note that the average coupon is much higher than the YTM, which means the fund holds on average bonds with large price premiums (values above 100). Plugging the values above into a PV function, we get a price of 127.83 and a current yield of 3.44% (and the high current yield contributes to a high distribution yield). But we know that over 8.9 years, the value of this bond would fall to 100 at maturity. The combination of the 4.4 per 100 of interest per year and the negative capital return of 27.83 per 100 over 8.9 years gets us to a 1.1% return, equivalent to the initial yield.

In contrast to a bond fund, for which there is a fairly wide range of possible returns, a CD gives you an almost certain nominal return for the term of the CD. A 5-year CD at 1.5% will almost certainly earn 1.5% over five years. A more appropriate comparison for a CD is to a Treasury of the same term, and the yield on a 5-year Treasury is 0.28%, so you are much better off with the CD if you plan to hold to maturity. Any additional yield above the Treasury yield comes with additional risk.

Kevin

### Re: SEC 30 day yield vs Distribution yield

Most of the above is pretty accurate stuff! Great discussion!

I still think that a knowledgeable investor who knows what he's getting into can be served well at the present time by VWITX/VWIUX.

Here's some history for VWIUX.

On Oct. 15, 2018

Share price $13.87

Nov. 1, 2018 Dividend: .03329

On March 1, 2019 VWITX

Share price $14.03

Dividend payout per share: .02977

On August 3, 2020

Share price: $14.78

Dividend payout per share: .02876

Therefore, VWITX/VWIUX is still a good deal. (for now )

Bottom Line:

I believe that if you want to hold VWITX/VWIUX and forget about it, use SEC yield.

If you want to buy it and watch it, use the payout/distribution yield.

I still think that a knowledgeable investor who knows what he's getting into can be served well at the present time by VWITX/VWIUX.

Here's some history for VWIUX.

On Oct. 15, 2018

Share price $13.87

Nov. 1, 2018 Dividend: .03329

On March 1, 2019 VWITX

Share price $14.03

Dividend payout per share: .02977

On August 3, 2020

Share price: $14.78

Dividend payout per share: .02876

Therefore, VWITX/VWIUX is still a good deal. (for now )

Bottom Line:

I believe that if you want to hold VWITX/VWIUX and forget about it, use SEC yield.

If you want to buy it and watch it, use the payout/distribution yield.

### Re: SEC 30 day yield vs Distribution yield

You can explain all you like Mike... simply look at the historical NAV of the fund and you will see how wrong that statement is. The NAV was roughly $14.20 a share in January 2012 and is now a bit higher than that after tracking close to it for a long time. Sure it goes up some and down some... I basically see it in the range from $13.72 as a low to $14.78 last month. My personal data on this fund goes back 9 years.MikeG62 wrote: ↑Mon Aug 31, 2020 9:13 amThe excess of the distribution yield over the SEC yield is offset by a decline in the NAV of the fund. It's an illusion to think you are earning an above market yield because the fund happens to hold bonds with a higher coupon than current market rates. Part of the distribution yield is a return of principal (and not interest income). I own dozens of individual muni bonds and have seen this play out for years now.midareff wrote: ↑Mon Aug 31, 2020 8:33 am The SEC Yield is an indicator of what you may or might receive as a distribution some time in the future.. not necessarily sooner than later either. The present SEC on VWIUX is 1.00%, last month the distribution was 2.39% tax-exempt. Personally, I evaluate on what I'm going to receive for my money, not some statistical number that may or may not be true in the future.

Kevin did a great job explaining this. I've tried to do the same thing myself a number of times, such as in this thread:

viewtopic.php?f=1&t=305142&p=5056107&hi ... d#p5056107

### Re: SEC 30 day yield vs Distribution yield

The first underlined part is correct. You will find a statement like this in the "Notes to Financial Statements" in the annual or semi-annual report:hudson wrote: ↑Mon Aug 31, 2020 6:15 am <snip>

I don't think anyone has brought up the part where part of VWITX/VWIUX's payout is from the principal. That usually comes up in this discussion. The simple answer to that is yes; Vanguard doesn't do it to artificially pump up distribution yield. It has something to do with the purchase price of a bond. If the fund profits because of bond pricing, it is required to pay that profit as interest. Somebody correct me if I'm wrong.

Anyone who has purchased a discount or premium bond, and done the tax returns, knows that this is referring to the premium or discount at purchase, not any premiums or discounts that arise after purchase.12. Other: Interest income includes income distributions received from Vanguard Market Liquidity Fund and is accrued daily. Premiums and discounts on debt securities are amortized and accreted, respectively, to interest income over the lives of the respective securities, except for premiums on certain callable debt securities that are amortized to the earliest call date.

Example.

Say you buy a 10-year bond on the secondary market at 110 (110% of par). You will amortize the $10 (per $100 of bond value) over a 10-year period, so roughly $1 per year, although that's not exactly how it's calculated. A fund will subtract this amortization from the interest (coupon) payments it receives in calculating the income to distribute as dividends.

Say yields fall further, and the bond value increases to 115. You, or the fund, will continue amortizing the original $10 per $100; the extra $5 due to price change after purchase does not affect the amortization or the income the fund distributes. It just contributes to an increase in NAV.

With this understanding, the second underlined phrase in the first quote above does not make sense to me. In the example above, the profit would be the $5 per $100 that the bond increased in price after purchase, and the fund does not "pay that profit as interest".

This makes no sense, but I think it is a common misunderstanding that somehow your yield relates to what you paid for an asset. What you paid makes no difference looking forward, and yields are forward-looking values. It's the forward looking values that are relevant to decisions about making portfolio changes, such as selling a bond fund to buy a CD or another bond fund.

Example. I bought a bunch of individual Treasuries with yields in the 2% ballpark (numbers are rough, for illustration purposes). Yields fell to the 1% ballpark, and the prices of the Treasuries increased correspondingly. I sold the Treasuries to invest in a CD based on the 1% Treasury yields. It would have been foolish of me to think that I was earning 2% on the Treasuries, just because that was the yield based on my purchase prices. Part of that 2% had been paid to me in advance in the form of price increases, so I could capitalize on that by selling at a profit, and buying something else with a higher yield, rather than only earning 1% going forward.

Kevin

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

### Re: SEC 30 day yield vs Distribution yield

As I think I mentioned earlier, it can be difficult to isolate the pull toward par effect from other effects, such as price change due to market yield changes. We can see that by looking at an individual bond example, using roughly the 2012 yield and the 2020 yield of the fund.midareff wrote: ↑Mon Aug 31, 2020 4:04 pmYou can explain all you like Mike... simply look at the historical NAV of the fund and you will see how wrong that statement is. The NAV was roughly $14.20 a share in January 2012 and is now a bit higher than that after tracking close to it for a long time. Sure it goes up some and down some... I basically see it in the range from $13.72 as a low to $14.78 last month. My personal data on this fund goes back 9 years.MikeG62 wrote: ↑Mon Aug 31, 2020 9:13 amThe excess of the distribution yield over the SEC yield is offset by a decline in the NAV of the fund. It's an illusion to think you are earning an above market yield because the fund happens to hold bonds with a higher coupon than current market rates. Part of the distribution yield is a return of principal (and not interest income). I own dozens of individual muni bonds and have seen this play out for years now.

Kevin did a great job explaining this. I've tried to do the same thing myself a number of times, such as in this thread:

viewtopic.php?f=1&t=305142&p=5056107&hi ... d#p5056107

On Jan 3, 2012, VWIUX SEC yield was 2.28% (and price was $14.03). On Aug 28, 2020, SEC yield was 1.00% (and price was $14.73). Let's look at an individual bond to see how the price increase due to decline in yield can swamp out the decrease due to the pull toward par effect over quite a long time. This example is more complicated than it needs to be to just show this effect, since it also includes bond premium amortization, but perhaps that also will be informative.

Say in 2012 the fund buys a 20-year bond at 2.30% yield and 3.56% coupon. The price will be about 120, so over 20 years the fund will amortize $20 per $100 of bond value (all subsequent values are per $100). For simplicity of illustration, assume straight-line amortization, which means that each year the fund will subtract about $1 in amortized premium from the interest payment of $3.56, for a net income distribution of $2.56 per year.

Now in 2020 the fund has a 12-year bond with a yield of 1.00%, still paying the coupon interest of $3.56 per year, with $1 per year subtracted for amortized bond premium. Although the fund has amortized $8 in bond interest, reducing the basis to 112 (120 - 8), since the 1% yield is much less than the 3.56% interest payment, the price now is 128.81. So we have a price increase of $8.81. After subtracting the $8 in amortized bond premium, we still have a net increase of $0.81. This would contribute to a higher NAV for the fund, even though the fund has amortized $8 in amortized bond premiums, reducing its basis from 120 to 112.

Incidentally, at this point, the distribution yield contribution for this bond would be about 2%, which is 3.56 in interest minus 1.00 in amortized bond premium divided by the 128.81 price; so distribution yield would be about 2X yield to maturity.

Fast forward to when the bond has three years remaining until maturity, and say the yield of the 3-year bond is 1% at that time. The price will then be 107.53, which of course is less than the 120 we paid for the bond and the 128.81 it is valued at today. At this point we would finally see the negative impact on bond fund NAV. It can take a long time, but we eventually see the impact of the pull toward par effect on NAV if the fund continues to hold the bond.

I think this is an example of what's been going on in bond funds as yields have been declining.

At any rate, unless you think the fund managers are performing some kind of magic, I think it's incumbent on anyone who doesn't believe that bond premiums being pulled toward par at maturity will eventually result in decreased NAV, unless yields continue to fall, to show us the bond math that can make that happen. Again, a bond fund is just a collection of bonds, so the bond math that applies inexorably to each individual bond must apply in aggregate to the fund.

Kevin

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

### Re: SEC 30 day yield vs Distribution yield

Thanks Kevin!Kevin M wrote: ↑Mon Aug 31, 2020 7:04 pm I think it's incumbent on anyone who doesn't believe that bond premiums being pulled toward par at maturitywill eventually result in decreased NAV,unless yields continue to fall, to show us the bond math that can make that happen. Again, a bond fund is just a collection of bonds, so the bond math that applies inexorably to each individual bond must apply in aggregate to the fund.

Kevin

I always thought that at least with VWITX/VWIUX that the distribution yield would eventually decline towards the SEC yield.

How long that will take. I don't know. It looks like the run will continue for another year or more.

### Re: SEC 30 day yield vs Distribution yield

If the yield is that good for a year or so then it fits my need for a year or so, at that point I'll decide what to do. Other than when the Prime Money Market had a good yield, I have not put any money into my Vanguard taxable account for at least 3 or 4 years. All that PMM money went to Ally when they had good rates. Now that CD will leave Ally and go to the investor shares in the taxable account.

### Re: SEC 30 day yield vs Distribution yield

Right. Distribution yield declining toward SEC yield is a slightly different topic than NAV eventually falling as bond premiums bleed off (the pull toward par effect), but I've done quite a bit of analysis of that in past years as well. I haven't updated the results, but here's a relevant chart from this thread from early 2015: Total Bond: 23 years of SEC yield and distribution yieldhudson wrote: ↑Mon Aug 31, 2020 7:19 pmThanks Kevin!Kevin M wrote: ↑Mon Aug 31, 2020 7:04 pm I think it's incumbent on anyone who doesn't believe that bond premiums being pulled toward par at maturitywill eventually result in decreased NAV,unless yields continue to fall, to show us the bond math that can make that happen. Again, a bond fund is just a collection of bonds, so the bond math that applies inexorably to each individual bond must apply in aggregate to the fund.

Kevin

I always thought that at least with VWITX/VWIUX that the distribution yield would eventually decline towards the SEC yield.

How long that will take. I don't know. It looks like the run will continue for another year or more.

Since yield generally had been falling, distribution yield usually was higher than SEC yield, but we see a few points on the chart when yields had risen and distribution yield was less than SEC yield.

This makes sense in thinking about bond turnover. Assuming bonds are issued at prices close to par (price = 100), coupon rates on new bonds will decline as yields decline, and the coupon rate is a dominant factor in determining distribution yields. As older bonds mature or are sold as yields decline, the newer bonds will tend to have lower coupon rates.

In terms of it continuing, sure, as long as yields are stable or decline further. But unless you believe yields can go deeply negative, the closer we get to 0% yields, the less room there is for yields to decline further. And once coupon rates get close to 0%, distribution yield also will approach 0%.

So I would say that making investment decisions (whether it's to buy or hold) based on distribution yields being higher than SEC yields is a bit of market timing, which is something I've done myself in the past.

Kevin

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

### Re: SEC 30 day yield vs Distribution yield

Just remember that maximum drawdown of more than 6% I mentioned earlier. I'm not saying not to do it, but just be aware that the price decline could more than offset the income distributions, resulting in a negative total return. Just looking at calendar years, in 2013 the capital return for VWITX was -4.59% and the income return was 3.03%, for a net total return of -1.56%. In 2008 capital return was -4.15% and income return was 4.01%, for a total return of -0.14%. Source: https://investor.vanguard.com/mutual-fu ... ve-returns.zaplunken wrote: ↑Mon Aug 31, 2020 7:59 pm If the yield is that good for a year or so then it fits my need for a year or so, at that point I'll decide what to do. Other than when the Prime Money Market had a good yield, I have not put any money into my Vanguard taxable account for at least 3 or 4 years. All that PMM money went to Ally when they had good rates. Now that CD will leave Ally and go to the investor shares in the taxable account.

Kevin

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

### Re: SEC 30 day yield vs Distribution yield

Thanks again KevinM!

Anything that works is good technique!

Anything that works is good technique!

### Re: SEC 30 day yield vs Distribution yield

Thank you Kevin, I appreciate all your posts. I realize there is risk in everything even treasuries have risk. I'm willing to chance it that VWITX will have a better interest rate for the time being, it only has to beat 75 to 100 basis points!

### Re: SEC 30 day yield vs Distribution yield

SEC Yield on the wiki has examples which show how this works. If two bonds have equal SEC yield and duration, they will have the same total return whether rates rise, fall, or stay the same. (In the example on the wiki, the two bonds do not have equal duration, so the one with higher duration gains slightly more if rates fall.)Seasonal wrote: ↑Mon Aug 31, 2020 9:38 am If rates don't move, the market value of a bond will move towards its face amount, because that's what the holder will get at maturity.

The complication is that rates do move. In particular, if rates drop the market value of the bond will increase, perhaps enough to offset the decline in value as the bond moves towards maturity. If the bond is sold at that point, you may get both the distribution yield and a price at or above face amount. The decline in rates means that newly purchased bonds will have a lower stated yield than the prior bonds.

Rates have been dropping for a while, but they are not likely to continue dropping forever.

### Re: SEC 30 day yield vs Distribution yield

" I think it's incumbent on anyone who doesn't believe that bond premiums being pulled toward par at maturity will eventually result in decreased NAV, unless yields continue to fall, to show us the bond math that can make that happen."Kevin M wrote: ↑Mon Aug 31, 2020 7:04 pmAs I think I mentioned earlier, it can be difficult to isolate the pull toward par effect from other effects, such as price change due to market yield changes. We can see that by looking at an individual bond example, using roughly the 2012 yield and the 2020 yield of the fund.midareff wrote: ↑Mon Aug 31, 2020 4:04 pmYou can explain all you like Mike... simply look at the historical NAV of the fund and you will see how wrong that statement is. The NAV was roughly $14.20 a share in January 2012 and is now a bit higher than that after tracking close to it for a long time. Sure it goes up some and down some... I basically see it in the range from $13.72 as a low to $14.78 last month. My personal data on this fund goes back 9 years.MikeG62 wrote: ↑Mon Aug 31, 2020 9:13 am

Kevin did a great job explaining this. I've tried to do the same thing myself a number of times, such as in this thread:

viewtopic.php?f=1&t=305142&p=5056107&hi ... d#p5056107

On Jan 3, 2012, VWIUX SEC yield was 2.28% (and price was $14.03). On Aug 28, 2020, SEC yield was 1.00% (and price was $14.73). Let's look at an individual bond to see how the price increase due to decline in yield can swamp out the decrease due to the pull toward par effect over quite a long time. This example is more complicated than it needs to be to just show this effect, since it also includes bond premium amortization, but perhaps that also will be informative.

Say in 2012 the fund buys a 20-year bond at 2.30% yield and 3.56% coupon. The price will be about 120, so over 20 years the fund will amortize $20 per $100 of bond value (all subsequent values are per $100). For simplicity of illustration, assume straight-line amortization, which means that each year the fund will subtract about $1 in amortized premium from the interest payment of $3.56, for a net income distribution of $2.56 per year.

Now in 2020 the fund has a 12-year bond with a yield of 1.00%, still paying the coupon interest of $3.56 per year, with $1 per year subtracted for amortized bond premium. Although the fund has amortized $8 in bond interest, reducing the basis to 112 (120 - 8), since the 1% yield is much less than the 3.56% interest payment, the price now is 128.81. So we have a price increase of $8.81. After subtracting the $8 in amortized bond premium, we still have a net increase of $0.81. This would contribute to a higher NAV for the fund, even though the fund has amortized $8 in amortized bond premiums, reducing its basis from 120 to 112.

Incidentally, at this point, the distribution yield contribution for this bond would be about 2%, which is 3.56 in interest minus 1.00 in amortized bond premium divided by the 128.81 price; so distribution yield would be about 2X yield to maturity.

Fast forward to when the bond has three years remaining until maturity, and say the yield of the 3-year bond is 1% at that time. The price will then be 107.53, which of course is less than the 120 we paid for the bond and the 128.81 it is valued at today. At this point we would finally see the negative impact on bond fund NAV. It can take a long time, but we eventually see the impact of the pull toward par effect on NAV if the fund continues to hold the bond.

I think this is an example of what's been going on in bond funds as yields have been declining.

At any rate, unless you think the fund managers are performing some kind of magic, I think it's incumbent on anyone who doesn't believe that bond premiums being pulled toward par at maturity will eventually result in decreased NAV, unless yields continue to fall, to show us the bond math that can make that happen. Again, a bond fund is just a collection of bonds, so the bond math that applies inexorably to each individual bond must apply in aggregate to the fund.

Kevin

Of course they are, for bonds purchased at a premium. .. which assumes they are all bought at a premium rather than at par or at a discount. For a fund that holds 10343 issues purchased over a time period of multiple years that's not a call I would make with any degree of confidence.

To continue, if bond NAV is being amortized into the distribution payment resulting in the bond being eventually sold at a loss of NAV, it would seem there would be some form of distributions labeled capital loss in this scenario. I don't see any on record since my records of this fund since 2011 so what's happening with that?

### Re: SEC 30 day yield vs Distribution yield

It doesn't assume that any bonds are bought at a premium. As I mentioned, my example was complicated by the bond premium amortization stuff, which is not required to understand how bonds currently trading at a premium will be pulled toward par as they approach maturity. We can look at a non-amortized example, but first, let's address how we know that bonds in the fund are trading at a premium on average, hopefully understanding that it's the average that matters, not that every single bond is trading at a premium.midareff wrote: ↑Tue Sep 01, 2020 7:23 am " I think it's incumbent on anyone who doesn't believe that bond premiums being pulled toward par at maturity will eventually result in decreased NAV, unless yields continue to fall, to show us the bond math that can make that happen."

Of course they are, for bonds purchased at a premium. .. which assumes they are all bought at a premium rather than at par or at a discount. For a fund that holds 10343 issues purchased over a time period of multiple years that's not a call I would make with any degree of confidence.

I mentioned the key bond characteristics for VWIUX/VWITX as of July 31 in an earlier post, but let's take a look at them on the Vanguard Portfolio & Management page for VWIUX:

The key numbers are the Yield to maturity and Average coupon. We see that the average coupon of 4.4% is much higher than the YTM of 1.1%. Intuitively you might be able to understand that if the coupon rate of a bond is higher than the yield (to maturity), the bond is valued above par, but if you can't understand it intuitively, you can use the spreadsheet PRICE function, PV function, or a financial calculator to verify it. For example:

=PRICE("1/1/2000", "1/1/2010", 2%, 2%, 100, 2)

returns the value 100.00, while

=PRICE("1/1/2000", "1/1/2010", 2%, 1%, 100, 2)

returns the value 109.49. The first percentage is the coupon rate, the second is the yield.

Using the PV function for the 2% yield and 1% coupon, but assuming only one coupon payment for year for simplicity:

=-PV(1%, 10, 2%, 1)*100

returns 109.47, with the small difference due to the coupon frequency assumption difference (changing the frequency parameter from 2 to a 1 in the PRICE function gives the same value of 109.47).

Plugging the numbers from the fund into the PV function:

=-PV(1.1%, 8.9, 4.4%, 1)*100

which returns 127.83. But applying bond pricing to these averages might not give us quite the right number.

Another thing we can do is download the holdings from the Vanguard financial advisor's website (currently available for 6/30/2020), load them into a spreadsheet, and calculate total market value = 74,078,382,172, total face value = 65,038,800,142, and average coupon = 4.78% (the latter is not a weighted average, just a simple average). The ratio of market value to face value is 1.14, which would translate to an average price of 114. Note that these values are two months old, and prices have risen since then.

Anyway, either way we estimate it, average price is quite a bit higher than 100.

This is not true. The fund can't distribute a loss in the form of a negative dividend--that would mean they'd be sending you a bill. But forget about distributions, and let's just consider if the fund must necessarily realize a loss.midareff wrote: ↑Tue Sep 01, 2020 7:23 am To continue, if bond NAV is being amortized into the distribution payment resulting in the bond being eventually sold at a loss of NAV, it would seem there would be some form of distributions labeled capital loss in this scenario. I don't see any on record since my records of this fund since 2011 so what's happening with that?

This brings us to a simpler example with no bond premium amortization. Let's revisit the example using the 2.3% yield from 2012 and 1.0% yield for 2020, but this time let's assume the fund bought the 20-year bond at par (price = 100), which means that the coupon rate was 2.3% when the bond was purchased. At the 1% yield in 2020, the price of what now is a 12-year bond would be 114.63, and the current yield would be 2.01% (2.3/114.63), and since there is no bond premium amortization, the distribution yield would be 2.01% (maybe less due to expenses, but in the ballpark). So we're still looking at a distribution yield that is 2X the yield to maturity.

In this case, the basis of the bond is 100, so unless the price drops below 100, there will never be a loss. But there is 14.63 of bond premium that will eventually decline to 0 if the bond is held to maturity. If the bond were sold at 3-year maturity and 1% yield, the price would be 103.82, so that's a capital gain of 3.82, but still 10.81 of price that has been lost, relative to current market value, as the bond price is pulled toward par.

Going back to the example with bond premium amortization, we saw that at 12-year maturity (today), the amortized value of the bond was 112, which is the basis, but the market value was 128.81. At 3-year maturity (9 years from now), amortized value would be about 103 (120 - 17), but the market value would be 107.53, so if the bond were sold at that point, there would be a capital gain of about 4.53 (= 107.53 - 103), but 21.28 of price loss relative to the current price.

Coming back to the relationship between distribution yield and NAV decline due to bond premiums bleeding off as yields are pulled toward par at maturity. The fund manager could avoid some NAV decline by selling bonds before the premiums decrease too much, and buying newer bonds with lower coupon rates (and thus lower premiums). This would lower the distribution yield. So it does seem like the fund managers can tweak holdings to balance NAV decline with lower distribution yield.

Kevin

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

### Re: SEC 30 day yield vs Distribution yield

I wanted to mention that I learned most of what I understand about how bond premiums affect fund income distributions from posts on this forum over the years, supplemented of course with much spreadsheet analysis on my own.

One thing I was scratching my head over last night was the following. If the fund is reducing its income distributions due to bond premium amortization, isn't that a double whammy, since the price of the bond also will fall toward par at maturity? The answer is "no", and I found a very clear explanation in this forum post from 2014. I think it's so good that I'm just going to quote part of it here. For context, the scenario is a 1-year bond with 0.25% yield and 4.25% coupon, priced at 104.

Kevin

One thing I was scratching my head over last night was the following. If the fund is reducing its income distributions due to bond premium amortization, isn't that a double whammy, since the price of the bond also will fall toward par at maturity? The answer is "no", and I found a very clear explanation in this forum post from 2014. I think it's so good that I'm just going to quote part of it here. For context, the scenario is a 1-year bond with 0.25% yield and 4.25% coupon, priced at 104.

Looking at the Prem column, what I was not considering is that even though the fund doesn't distribute the amortized premium, it collects and retains it in the form of the coupon payments, which contributes to the NAV.#Cruncher wrote: ↑Thu Feb 27, 2014 9:16 pm <snip>grabineris correct that the interest income of the 4.25% bond will depend on the price the fund paid for it. For example, if it bought it at par sometime in the past, then the entire 4.25% coupon will be interest income and will be distributed. However, if it bought the bond today at the current market price of 104%, then the interest income over the final year would be only 0.25% (4.25% coupon less amortization of the 4% purchase price premium).

Here is a comparison of that portion of the fund represented by $100 face value of the bond in question. The "Par" case assumes the bond was bought at a price of 100%. The "Prem" case assumes it was bought at a premium price of 104%.In theCode: Select all

`Par Prem ------ ------ 104 104 Net Assets at beginning of year (market value of bond) 4.25 4.25 Coupon interest collected 0.00 (4.00) Amortization of Purchase Premium 4.25 0.25 Interest income distributed 100 100 Cash from Redemption 100 104 Net Assets at end of year (cash)`

Parcase, the fund distributes $4.25 and loses $4.00 in net assets. * In thePremcase it distributes only $0.25 but loses nothing in net assets because $4.00 of the coupon interest it receives offsets the decline from $104 market value of the bond at the beginning of the year to the $100 redemption.

<snip>

Kevin

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

### Re: SEC 30 day yield vs Distribution yield

midareff,midareff wrote: ↑Mon Aug 31, 2020 4:04 pm

Kevin did a great job explaining this. I've tried to do the same thing myself a number of times, such as in this thread:

viewtopic.php?f=1&t=305142&p=5056107&hi ... d#p5056107

There are a lot of things going on under the water line of the NAV. One is the amortization of the bond premium. Another is changes in market interest rates. A third, is the buying and selling of bonds within the portfolio.

Let me ask you this simple question. If you think the current distribution yield for VWIUX (2.29% for the month of August) represents the annualized interest income based upon the August distribution, then how is it possible that this fund could be offering a yield "

**way above**" market interest rates(for comparable term and credit risk) and that fact not have been exploited (arbitraged away) by professional money managers?

This site shows the current national muni bond yields for single A through AAA rated bonds at durations ranging from 10 to 30 years.

https://www.fmsbonds.com/market-yields/

Here are the 10-Year Bond Yields (to save you the time of looking):

AAA = 0.70%

AA = 0.85%

A = 1.05%

Based upon the weighting of the bonds in VWIUX (~20% AAA, ~50% AA, ~22% A and ~8% below A), the current weighted average market yield would be somewhere around 1.0% (and interestingly, that is spot on the 30-day SEC yield for VWIUX published by Vanguard).

So again, you would suggest we believe that VWIUX offers a real interest rate that is 2.3 times current comparable muni market rates? This rate is over 3 times the yield for a 10-year treasury and the muni's have the added tax benefit of being federally tax exempt.

Real Knowledge Comes Only From Experience

### Re: SEC 30 day yield vs Distribution yield

Seeing as the NAV has remained relatively constant over the last 9+ years I have to assume they are not distributing the NAV. If they are distributing an amortized bong purchase premium as a tax-exempt dividend that is another thing. One day there will be other alternatives to a 2.3% tax exempt dividend, which is steadily dropping, ... and when it drops to the point there are other viable alternatives those will have to be evaluated at that time. For now, I'm taking a 2.3% tax-exempt return.MikeG62 wrote: ↑Wed Sep 02, 2020 8:56 ammidareff,

Kevin did a great job explaining this. I've tried to do the same thing myself a number of times, such as in this thread:

viewtopic.php?f=1&t=305142&p=5056107&hi ... d#p5056107

There are a lot of things going on under the water line of the NAV. One is the amortization of the bond premium. Another is changes in market interest rates. A third, is the buying and selling of bonds within the portfolio.

Let me ask you this simple question. If you think the current distribution yield for VWIUX (2.29% for the month of August) represents the annualized interest income based upon the August distribution, then how is it possible that this fund could be offering a yield "way above" market interest rates(for comparable term and credit risk) and that fact not have been exploited (arbitraged away) by professional money managers?

This site shows the current national muni bond yields for single A through AAA rated bonds at durations ranging from 10 to 30 years.

https://www.fmsbonds.com/market-yields/

Here are the 10-Year Bond Yields (to save you the time of looking):

AAA = 0.70%

AA = 0.85%

A = 1.05%

Based upon the weighting of the bonds in VWIUX (~20% AAA, ~50% AA, ~22% A and ~8% below A), the current weighted average market yield would be somewhere around 1.0% (and interestingly, that is spot on the 30-day SEC yield for VWIUX published by Vanguard).

So again, you would suggest we believe that VWIUX offers a real interest rate that is 2.3 times current comparable muni market rates? This rate is over 3 times the yield for a 10-year treasury and the muni's have the added tax benefit of being federally tax exempt.

### Re: SEC 30 day yield vs Distribution yield

The listed yields are YTMs. " These rates reflect the approximate yield to maturity that an investor can earn in today’s tax-free municipal bond market as of 09/02/2020".MikeG62 wrote: ↑Wed Sep 02, 2020 8:56 am Let me ask you this simple question. If you think the current distribution yield for VWIUX (2.29% for the month of August) represents the annualized interest income based upon the August distribution, then how is it possible that this fund could be offering a yield "way above" market interest rates(for comparable term and credit risk) and that fact not have been exploited (arbitraged away) by professional money managers?

This site shows the current national muni bond yields for single A through AAA rated bonds at durations ranging from 10 to 30 years.

https://www.fmsbonds.com/market-yields/

Here are the 10-Year Bond Yields (to save you the time of looking):

AAA = 0.70%

AA = 0.85%

A = 1.05%

Based upon the weighting of the bonds in VWIUX (~20% AAA, ~50% AA, ~22% A and ~8% below A), the current weighted average market yield would be somewhere around 1.0% (and interestingly, that is spot on the 30-day SEC yield for VWIUX published by Vanguard).

So again, you would suggest we believe that VWIUX offers a real interest rate that is 2.3 times current comparable muni market rates? This rate is over 3 times the yield for a 10-year treasury and the muni's have the added tax benefit of being federally tax exempt.

Nice to see that the Vanguard fund 30-day SEC yield (essentially an average YTM) approximates market YTMs.

### Re: SEC 30 day yield vs Distribution yield

Yup, and one is not earning 2.30% as one might think from looking at the distribution yield (which includes a return of principal component).Seasonal wrote: ↑Wed Sep 02, 2020 9:53 amThe listed yields are YTMs. " These rates reflect the approximate yield to maturity that an investor can earn in today’s tax-free municipal bond market as of 09/02/2020".MikeG62 wrote: ↑Wed Sep 02, 2020 8:56 am Let me ask you this simple question. If you think the current distribution yield for VWIUX (2.29% for the month of August) represents the annualized interest income based upon the August distribution, then how is it possible that this fund could be offering a yield "way above" market interest rates(for comparable term and credit risk) and that fact not have been exploited (arbitraged away) by professional money managers?

This site shows the current national muni bond yields for single A through AAA rated bonds at durations ranging from 10 to 30 years.

https://www.fmsbonds.com/market-yields/

Here are the 10-Year Bond Yields (to save you the time of looking):

AAA = 0.70%

AA = 0.85%

A = 1.05%

Based upon the weighting of the bonds in VWIUX (~20% AAA, ~50% AA, ~22% A and ~8% below A), the current weighted average market yield would be somewhere around 1.0% (and interestingly, that is spot on the 30-day SEC yield for VWIUX published by Vanguard).

So again, you would suggest we believe that VWIUX offers a real interest rate that is 2.3 times current comparable muni market rates? This rate is over 3 times the yield for a 10-year treasury and the muni's have the added tax benefit of being federally tax exempt.

Nice to see that the Vanguard fund 30-day SEC yield (essentially an average YTM) approximates market YTMs.

If it had outsized earnings relative to the market, it would have to be taking on considerably more risk. Otherwise, the upside opportunity would be exploited by people far smarter than us and it would quickly disappear.

Real Knowledge Comes Only From Experience

### Re: SEC 30 day yield vs Distribution yield

True, although if the claim is that distribution yield, under the circumstances, is as useful or more useful than SEC yield or, essentially equivalently, yield to maturity, looking at other YTMs doesn't really answer the question.

It appears that NAV has been stable due to the effects of falling interest rates and that interest rates can't fall forever, but interest rates are notoriously tricky to predict.

It appears that NAV has been stable due to the effects of falling interest rates and that interest rates can't fall forever, but interest rates are notoriously tricky to predict.

### Re: SEC 30 day yield vs Distribution yield

It's true that looking at just the YTM may not help us understand how how distribution yield is related to SEC yield (or YTM), but looking YTM and other relevant characteristics of an individual bond can indeed help us understand it, if we are actually willing to invest the time and mental effort to try. Rather than look at averages, let's look at an actual 10-year bond shown at Fidelity.

ALABAMA ST HWY FIN CORP SPL OBLIG REV

Rating: AA2 (Moody's)

Maturity date: 08/01/2030

Coupon 5.00%

Price: 136.779

YTM/YTW (call protected): 1.077%

The yield (to maturity or to worst) of 1.077% factors in both the coupon payments of 5% and the fact that the bond will mature at 100, losing 36.779 relative to the purchase price (pull toward par effect).

The current yield is 3.66% (=5/136.779). If the bond were bought today, or at any time in the past a premium (price > 100), bond amortization would cause the distribution yield to be less than current yield by the amortization amount. For example, if the bond were bought on 8/1/2015 (as a 15-year bond) at a price of 130.00, the amortization amount would average $2 per $100 of value (30 / 15), reducing the average distribution from 5 to 3, giving us a distribution yield of about 2.19% (= 3/136.779).

(Individual bond jocks know that I'm simplifying the annual bond amortization amount, since for bonds issued after September 27, 1985, you must amortize bond premium using a constant yield method, as opposed to a straight-line method. For the above example, this means that annual amortization would be less than $2 per $100 in the earlier years, and be more than $2 per $100 in the later years. But this wouldn't have a huge impact on the example distribution yield, and the conclusions are the same.)

Anyone who owns this bond would be foolish to think that they're going to earn 5% or 3.66% or 2.19% if they hold the bond to maturity. They are going to earn about 1%, the YTM/YTW, with the only uncertainty being the return they earn on the reinvested coupon payments.

Of course there is uncertainty in what they would earn if they sold the bond before maturity and reinvest the proceeds in another bond, as that would depend on the yield/price of the bond when they sold. But if yields are about the same when they sell/buy as they are now, they are not going to be able to magically convert a 1% yield into a 2% return (although they might earn a bit more than 1% due to rolldown return).

Only magical thinking can lead to thinking you will actually earn more than 2% over the long term when yields are 1%, assuming yields stay about the same. Of course yields don't stay the same, which leads back to the historical data I've shown that indicates that with a starting yield of 1%, you are likely to earn something between 0% and 2% over the next 10 years in an intermediate-term bond fund, with not much more likelihood that it will be closer to the high end than the low end. And over a much shorter term, it's even more of a crapshoot.

Agreed. Ignoring the impact of yields having generally fallen in past years seems to have led to some serious misunderstandings about the relationship between SEC yield (or YTM), distribution yield, and NAV.

Yields certainly could fall further, and it could be years until any such misunderstandings are eventually corrected by reality. Even if yields remain about where they are, it could take a few years for NAV and/or distribution yields to come down enough that people will start to get it. And we shouldn't ignore the possibility that yields could increase enough at some point that total returns could be negative for extended periods of time.

Kevin

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

### Re: SEC 30 day yield vs Distribution yield

Agreed. My point was that trying to prove that YTM is a better measure than distribution yield by showing that a fund's YTM is approximately equal to the average YTM of the underlying securities just begs the question.Kevin M wrote: ↑Wed Sep 02, 2020 6:27 pmIt's true that looking at just the YTM may not help us understand how how distribution yield is related to SEC yield (or YTM), but looking YTM and other relevant characteristics of an individual bond can indeed help us understand it, if we are actually willing to invest the time and mental effort to try. Rather than look at averages, let's look at an actual 10-year bond shown at Fidelity.

They key here is that NAV has remained stable, and therefore return has approximated distribution yield rather than YTM, due to falling interest rates. Which measure will work better in the future depends on future interest rates. Under the default assumption that rates will remain stable, YTM (or SEC yield) will work better. Or at least as well - if bonds are bought at par, remain at par and are redeemed at par, the two measures should be the same. If the yield curve is not flat, then selling prior to maturity can complicate things.

### Re: SEC 30 day yield vs Distribution yield

^Emphasis added.Kevin M wrote: ↑Wed Sep 02, 2020 6:27 pm ...Rather than look at averages, let's look at an actual 10-year bond shown at Fidelity.

ALABAMA ST HWY FIN CORP SPL OBLIG REV

Rating: AA2 (Moody's)

Maturity date: 08/01/2030

Coupon 5.00%

Price: 136.779

YTM/YTW (call protected): 1.077%

The yield (to maturity or to worst) of 1.077% factors in both the coupon payments of 5% and the fact that the bond will mature at 100, losing 36.779 relative to the purchase price (pull toward par effect).

The current yield is 3.66% (=5/136.779). If the bond were bought today, or at any time in the past a premium (price > 100), bond amortization would cause the distribution yield to be less than current yield by the amortization amount. For example, if the bond were bought on 8/1/2015 (as a 15-year bond) at a price of 130.00, the amortization amount would average $2 per $100 of value (30 / 15), reducing the average distribution from 5 to 3, giving us a distribution yield of about 2.19% (= 3/136.779).

(Individual bond jocks know that I'm simplifying the annual bond amortization amount, since for bonds issued after September 27, 1985, you must amortize bond premium using a constant yield method, as opposed to a straight-line method. For the above example, this means that annual amortization would be less than $2 per $100 in the earlier years, and be more than $2 per $100 in the later years. But this wouldn't have a huge impact on the example distribution yield, and the conclusions are the same.)

Anyone who owns this bond would be foolish to think that they're going to earn 5% or 3.66% or 2.19% if they hold the bond to maturity. They are going to earn about 1%, the YTM/YTW, with the only uncertainty being the return they earn on the reinvested coupon payments.

Kevin

Real Knowledge Comes Only From Experience