Expected Return For Bond Funds

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
aristotelian
Posts: 6447
Joined: Wed Jan 11, 2017 8:05 pm

Expected Return For Bond Funds

Post by aristotelian » Fri Nov 08, 2019 12:16 pm

Total return of bond funds includes both capital return and income return. The market price fluctuates purely as a function of changes in the interest rate environment. Is there any reason to expect positive capital return over time? If I assume that interests rates have equal chance of going higher or lower, is my expected return equivalent to the current yield? If I were to buy either a 10 year Treasury or a 10 year Treasury fund, both would have the same expected return, correct?

If that is the case, when projecting expected returns, we should disregard past capital return (even averaged over long periods) and look exclusively at yield, correct?

User avatar
Tyler Aspect
Posts: 1568
Joined: Mon Mar 20, 2017 10:27 pm
Location: California
Contact:

Re: Expected Return For Bond Funds

Post by Tyler Aspect » Fri Nov 08, 2019 12:33 pm

aristotelian wrote:
Fri Nov 08, 2019 12:16 pm
Total return of bond funds includes both capital return and income return. The market price fluctuates purely as a function of changes in the interest rate environment. Is there any reason to expect positive capital return over time? If I assume that interests rates have equal chance of going higher or lower, is my expected return equivalent to the current yield? If I were to buy either a 10 year Treasury or a 10 year Treasury fund, both would have the same expected return, correct?

If that is the case, when projecting expected returns, we should disregard past capital return (even averaged over long periods) and look exclusively at yield, correct?
A 10 year Treasury note has one maturity date, while a Treasury bond fund would have internally a range of maturity dates of its holdings. Individual Treasury note pays dividend twice a year, while Treasury bond funds pay dividend every month.

We do not know if our bond purchase will result in a capital gain or loss at the time of purchase. Our purchase decision can be based on bond yields and if the target bond duration is a good match with our investment horizon.
Past result does not predict future performance. Mentioned investments may lose money. Contents are presented "AS IS" and any implied suitability for a particular purpose are disclaimed.

User avatar
Kevin M
Posts: 11250
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Expected Return For Bond Funds

Post by Kevin M » Fri Nov 08, 2019 1:10 pm

A lot of folks, including me, think the SEC yield is a reasonable indicator of expected return for a bond fund over a holding period in the ballpark of the fund duration. But keep in mind that expected return is just the midpoint of a range of probable returns, so the realized 5-year or 10-year return of an intermediate-term bond fund could be higher or lower than the SEC yield. Historically, it easily varies by +/-0.5%, and it can vary by +/-1% or more compared to the initial SEC yield.

By contrast, a 10-year coupon-bearing Treasury held to maturity will return something quite close to the initial yield to maturity, with the only uncertainty being the reinvestment rates on the coupons. A zero-coupon Treasury eliminates the reinvestment uncertainty. A direct CD with interest reinvested also eliminates most uncertainty, although there is a small chance of bank or credit union failure, in which case the CD could be closed out before maturity.

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

Scooter57
Posts: 1065
Joined: Thu Jan 24, 2013 9:20 am

Re: Expected Return For Bond Funds

Post by Scooter57 » Fri Nov 08, 2019 1:24 pm

The SEC yield is fine and dandy unless bonds in the fund default. This is an issue with High Yield Bond Funds and could be an issue with funds holding corporate bonds with lower ratings should their ratings drop out of the Investment Grade class.

RAchip
Posts: 382
Joined: Sat May 07, 2016 7:31 pm

Re: Expected Return For Bond Funds

Post by RAchip » Sun Nov 10, 2019 3:52 pm

“We do not know if our bond purchase will result in a capital gain or loss at the time of purchase.“

Not totally correct. If you buy a bond and intend to and do hold to maturity, you know EXACTLY what you will get (what your return will be) assuming no default. You get the interest payments specified in the bond plus the return of your principal (par value) at maturity.

The market value of bonds in the secondary market fluctuates. If you decide to sell the bond prior to maturity you may have a capital gain or loss.

User avatar
Kevin M
Posts: 11250
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Expected Return For Bond Funds

Post by Kevin M » Sun Nov 10, 2019 5:04 pm

RAchip wrote:
Sun Nov 10, 2019 3:52 pm
“We do not know if our bond purchase will result in a capital gain or loss at the time of purchase.“

Not totally correct. If you buy a bond and intend to and do hold to maturity, you know EXACTLY what you will get (what your return will be) assuming no default. You get the interest payments specified in the bond plus the return of your principal (par value) at maturity.
You may know the cash flows exactly, but you don't know the return exactly, unless you buy a zero-coupon bond, because you don't know the reinvestment rate on the coupon payments. Granted, at current low interest rates, the interest on interest component does not make a large difference.

Also, looking at the reply including the quoted statement, it's not clear if it's about individual bonds or bond funds. The thread is about bond funds, and the quoted statement is true for bond funds.

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

jebmke
Posts: 9829
Joined: Thu Apr 05, 2007 2:44 pm

Re: Expected Return For Bond Funds

Post by jebmke » Sun Nov 10, 2019 5:08 pm

Scooter57 wrote:
Fri Nov 08, 2019 1:24 pm
The SEC yield is fine and dandy unless bonds in the fund default. This is an issue with High Yield Bond Funds and could be an issue with funds holding corporate bonds with lower ratings should their ratings drop out of the Investment Grade class.
Wouldn't the bonds be prices based on some expectation of default rates? If so, then the expected return would incorporate that expectation. Of course, actual returns can vary but that isn't known in advance.
When you discover that you are riding a dead horse, the best strategy is to dismount.

User avatar
Kevin M
Posts: 11250
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Expected Return For Bond Funds

Post by Kevin M » Sun Nov 10, 2019 5:23 pm

jebmke wrote:
Sun Nov 10, 2019 5:08 pm
Scooter57 wrote:
Fri Nov 08, 2019 1:24 pm
The SEC yield is fine and dandy unless bonds in the fund default. This is an issue with High Yield Bond Funds and could be an issue with funds holding corporate bonds with lower ratings should their ratings drop out of the Investment Grade class.
Wouldn't the bonds be prices based on some expectation of default rates? If so, then the expected return would incorporate that expectation. <snip>
Right, I think you're saying the same thing. The point is that SEC yield, which is a type of yield to maturity, is based only on the bond yields (which are the flip side of bond prices). Bond yields will be higher (prices lower) for bonds with more credit risk, but if bonds default, the return will be less than the initial yield. Therefore SEC yield, based on yield to maturity of the bonds in the fund, would be higher than expected return.

I don't recall looking at a high-yield bond fund, but for total bond and some other intermediate-term bond funds I looked at, SEC yield was a decent proxy for expected return. As I've mentioned, there is a relatively wide dispersion of 5-year and 10-year returns around the initial SEC yield, so we need to remember that expected return or SEC yield is not necessarily what you'll earn.

The realized returns have tended to be higher, on average, than initial SEC yield, since yields have generally fallen since Vanguard bond funds were created, adding a positive capital return component to the total return. This is related to the generally positive roll-down return component, since the yield curve has generally been positively sloped, so if bond yields remain the same, fall, or don't increase too much, the bonds sold before maturity will contribute a positive capital return component. Some have argued, with merit, that we should add a bit to SEC yield to incorporate positive expected roll-down return.

Note that a positive roll-down return would tend to offset negative capital return from defaults.

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

Scooter57
Posts: 1065
Joined: Thu Jan 24, 2013 9:20 am

Re: Expected Return For Bond Funds

Post by Scooter57 » Sun Nov 10, 2019 5:28 pm

I wouldn't assume that bond ratings are up to date. The Financial crisis a decade ago showed just how faulty bond ratings could be, and we have no reason to think that is no longer true.

Vast numbers of bonds are bought and held by funds and ETFs that follow indexes, without any human analysis of the quality of the bond being purchased. The current desperate hunt for yield has fueled investor hunger for anything paying over 2%, and many reasonable people who write about finance are warning that the risk of many bonds is not being reflected accurately.

User avatar
patrick013
Posts: 2755
Joined: Mon Jul 13, 2015 7:49 pm

Re: Expected Return For Bond Funds

Post by patrick013 » Sun Nov 10, 2019 6:50 pm

SEC yield should be the main metric but it assumes the market will stay somewhat static. So if yields increase you should get slightly more than SEC yield if the bond(s) are held to maturity and slightly less if yield decreases and the bond(s) are held to maturity. That's with reinvestment. Ticker EDV has one of the highest SD's of all bond funds so market movements can alter market prices greatly. Standard deviations can portray price movements and returns of a bond investment.
age in bonds, buy-and-hold, 10 year business cycle

User avatar
Kevin M
Posts: 11250
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Expected Return For Bond Funds

Post by Kevin M » Sun Nov 10, 2019 7:12 pm

patrick013 wrote:
Sun Nov 10, 2019 6:50 pm
SEC yield should be the main metric but it assumes the market will stay somewhat static.
Not necessarily. We are discussing expected return, which is a probability-weighted average of all possible returns. If we assume a "somewhat static" market in estimating probable returns, then the dispersion of possible returns would be narrower than if the market were not assumed to be somewhat static; the expected return may or may not change under different assumptions that change the dispersion of possible returns.
patrick013 wrote:
Sun Nov 10, 2019 6:50 pm
So if yields increase you should get slightly more than SEC yield if the bond(s) are held to maturity and slightly less if yield decreases and the bond(s) are held to maturity.
Not necessarily.

First, most bond funds don't hold bonds until maturity so the conditional "if held to maturity" is irrelevant for most bond funds.

Second, whether the return is higher or lower than the initial SEC yield has to do not only with whether yields increase or decrease, but also the timing of the changes. If yields increase (decrease) once at the beginning of the holding period, say equal to the duration of the fund, you will earn about the initial yield over the holding period, and you will earn more (less) than the initial yield if your holding period is long than the duration. This is the "point of indifference" view of duration. However, if yields increase (decrease) once toward the end of the holding period, you will earn less (more) than the initial yield.

Of course yields won't change just once at the beginning or end of the holding period--they will change every day, and yields of different maturity bonds will change differently--one yield might go up on the day that another goes down. Modeling the net effect on bond fund return is difficult.

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

User avatar
patrick013
Posts: 2755
Joined: Mon Jul 13, 2015 7:49 pm

Re: Expected Return For Bond Funds

Post by patrick013 » Sun Nov 10, 2019 9:30 pm

Kevin M wrote:
Sun Nov 10, 2019 7:12 pm
patrick013 wrote:
Sun Nov 10, 2019 6:50 pm
SEC yield should be the main metric but it assumes the market will stay somewhat static.
Not necessarily. We are discussing expected return, which is a probability-weighted average of all possible returns. If we assume a "somewhat static" market in estimating probable returns, then the dispersion of possible returns would be narrower than if the market were not assumed to be somewhat static; the expected return may or may not change under different assumptions that change the dispersion of possible returns.
patrick013 wrote:
Sun Nov 10, 2019 6:50 pm
So if yields increase you should get slightly more than SEC yield if the bond(s) are held to maturity and slightly less if yield decreases and the bond(s) are held to maturity.
Not necessarily.

Which is why you should use SD also, to get an idea of the volatility of price and returns after you have calc'd the YTM as the starting point before market yields go up or down in the future. That is the process. Look at the frequency distribution and see what a confidence interval might be. But that's trying to be clairvoyant statistically. The futures market presently thinks much, much lower rates will occur the next few years. They're entitled to their opinion of course.

People try to knock YTM but they don't have anything better. The math stays the same. Even expectations theory has been disproved so many times it's silly. Some studies reveal negative correlations with that one. So we're at the mercy of the market again. I don't think ultra low or ultra high interest yields can last long regardless of current futures pricing. Looking at probable mean returns is very subjective or can be lacking objective or proven results and methods to be otherwise. If the OP wants a reliable figure for expected return I don't think there ever will be one. I'd guess rates will stay low for another year perhaps two.
age in bonds, buy-and-hold, 10 year business cycle

User avatar
grabiner
Advisory Board
Posts: 25373
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: Expected Return For Bond Funds

Post by grabiner » Thu Nov 14, 2019 10:55 pm

In Expected return of a bond fund, I conclude that the best estimate of a bond fund return is the SEC yield, plus the average slope of the yield curve at the fund's duration, so there is a small amount of expected capital return. If a fund holds 6-10-year bonds, then in one year, it will sell a 5-year bond to buy a 10-year bond in order to maintain the same duration. Thus, if yields don't change, the fund's return will be increased by 1/5 the difference between 5-year and 10-year yields, multiplied by the fund's 8-year duration. And when the yield curve has its average slope, this implies that investors do not expect yields to change. If the yield curve is steep, investors expect rates to rise, so the gain from the steep yield curve will be reduced by rising rates.
Wiki David Grabiner

averagedude
Posts: 760
Joined: Sun May 13, 2018 3:41 pm

Re: Expected Return For Bond Funds

Post by averagedude » Thu Nov 14, 2019 11:05 pm

When projecting 10 year annualuzed returns of 10 year treasuries with some accuracy, all you need to know is the 30 day sec yield today and what the 30 day sec yield will be 10 years from now. The former is much easier to calculate.

Post Reply