Impact of Called Bonds on Bond Funds

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Scooter57
Posts: 1605
Joined: Thu Jan 24, 2013 9:20 am

Impact of Called Bonds on Bond Funds

Post by Scooter57 »

Years ago I sent a note to the Richelson's after reading their bond book, and ever since they send me an infrequent newsletter. Today's version had what I found to be very interesting and detailed descriptions of the call features found in both Muni and Corporate Bonds.

Reading it suggested to me that the actual yields we will receive going forward from our bond funds may not be what we expect based on SEC yields, as according to the Richelsons, there are significant numbers of bonds for which a Yield to Worst can't be calculated and which can be called at any time or replaced by bonds paying lower interest rates.

https://www.allbondportfolios.com/why_a ... 0-2020.pdf

We get precious little information about the holdings in the bond funds we buy--as in almost none. So I have no way of evaluating what impact it could have on the bond funds I own and wonder to what extent the calling of bonds could provide an unpleasant surprise going forward with the Fed continuing to keep rates near zero.
alex_686
Posts: 7446
Joined: Mon Feb 09, 2015 2:39 pm

Re: Impact of Called Bonds on Bond Funds

Post by alex_686 »

I would be relaxed about this.

The SEC Yield is a high quality number. It is robust. The weakness it has is also its virtue. It is a mechanical calculation and all funds have to use the same calculations.

Yield to Worst is not a very good metric. The probability of this happening is low.

There are better valuation techniques out there. You can download the list of CUSIPs from the latest filings and run the calculations yourself. The problem is that you have to model future changes in the yield curve. Not just the Treasury Yield curve, but also the credit spread and the probability of default. This is computationally intense. This is not the issue. The problem is that model has lots of subjective assumptions. Don't get me wrong, you still get high quality results. The problem is that it is not a standardized model that the SEC can force a fund to use.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
humblecoder
Posts: 256
Joined: Thu Aug 06, 2020 8:46 am

Re: Impact of Called Bonds on Bond Funds

Post by humblecoder »

This wiki page touches on this issue briefly. Not sure if it helps or not: https://www.bogleheads.org/wiki/SEC_Yield

For mortgage-backed bonds such as GNMAs, the SEC yield does not reflect prepayment risk. If mortgage rates decline, homeowners will refinance their mortgages, causing the principal to be paid out and forcing the fund to reinvest at lower yields. This is analogous to the call provision on many other bonds. The SEC yield can adjust for callable bonds by using the yield to call if a bond is likely to be called, but it cannot adjust for mortgage-backed bonds because mortgages can be called at any time (upon refinance, or when the home is sold).

I do understand where you are coming from. If a bond is likely to be called, the yield to maturity does not mean anything since the bond will never make it to maturity. If rates have dropped and the bond is likely to get called, it is more accurate to use the yield to call in this case. In a falling interest rate environment like we are in now, it is likely that many callable bonds are going to be called. It would be irrational for a borrower to continue to pay a higher interest rate when they can exercise their "get out of jail free" card to refinance at a lower rate.

I know I didn't truly answer your question, but I do acknowledge that your logic seems sound that this is something that should be taken into account somehow.
alex_686
Posts: 7446
Joined: Mon Feb 09, 2015 2:39 pm

Re: Impact of Called Bonds on Bond Funds

Post by alex_686 »

humblecoder wrote: Wed Nov 11, 2020 1:31 pm This wiki page touches on this issue briefly. Not sure if it helps or not: https://www.bogleheads.org/wiki/SEC_Yield

For mortgage-backed bonds such as GNMAs, the SEC yield does not reflect prepayment risk. If mortgage rates decline, homeowners will refinance their mortgages, causing the principal to be paid out and forcing the fund to reinvest at lower yields. This is analogous to the call provision on many other bonds. The SEC yield can adjust for callable bonds by using the yield to call if a bond is likely to be called, but it cannot adjust for mortgage-backed bonds because mortgages can be called at any time (upon refinance, or when the home is sold).

I do understand where you are coming from. If a bond is likely to be called, the yield to maturity does not mean anything since the bond will never make it to maturity. If rates have dropped and the bond is likely to get called, it is more accurate to use the yield to call in this case. In a falling interest rate environment like we are in now, it is likely that many callable bonds are going to be called. It would be irrational for a borrower to continue to pay a higher interest rate when they can exercise their "get out of jail free" card to refinance at a lower rate.

I know I didn't truly answer your question, but I do acknowledge that your logic seems sound that this is something that should be taken into account somehow.
I need to add 2 points to your post.

The SEC Yield does factor in pre-payments for MBS. Everybody knows that MBS will be called early. Home owner moves, refinances, or makes extra payments. So MBS comes with a standardized prepayment speed. That goes into the SEC yield calculations. So prepayments are not the issue - it is the change in the speed of the prepayments. It is a technical point but a important one.

The second is something that I alluded to above - credit spreads. Or as we are currently in, a "Double Gamma" event. Yes, treasury yields are down. This pushes corporate yields down. However, credit spreads are up. Many companies have had their bond ratings downgraded. A company is only going to call its bond if it can issue a lower yielding bond.

So not a slam dunk that a company can issue lower yielding bonds. Most call provisions require that the company pay some premium over par. Or the call is the future, maybe with a lower premium. Will the company's ability to issue bonds in the future increase or decrease?
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
User avatar
vineviz
Posts: 8492
Joined: Tue May 15, 2018 1:55 pm

Re: Impact of Called Bonds on Bond Funds

Post by vineviz »

Scooter57 wrote: Wed Nov 11, 2020 12:50 pm We get precious little information about the holdings in the bond funds we buy--as in almost none. So I have no way of evaluating what impact it could have on the bond funds I own and wonder to what extent the calling of bonds could provide an unpleasant surprise going forward with the Fed continuing to keep rates near zero.
For most bond funds, the impact is going to be very low.

Treasury bonds, of course, are not callable to begin with and many investors own either a dedicated Treasury fund or an aggregate (e.g. "total bond" fund that holds a large percentage of assets in Treasury bonds).

Moreover, the difference for most corporate bond funds isn't significant. Take for instance the iShares iBoxx $ Investment Grade Corporate Bond ETF (LQD), which is the largest corporate bond ETF in the market.

iShares reports both the "30 Day SEC Yield" (2.00%) and the "Average Yield to Maturity" (2.13%) for this fund, and YTM in this case is yield-to-worst for callable bonds. The SEC yield is actually LOWER than the reported yield-to-worst, so if every single callable bond in that fund were to be called at the first possible moment any surprise that you received would be PLEASANT and not unpleasant.

Your bond funds probably report YTW somewhere, so you likely DO have a way of evaluating your exposure but I'm highly confident it's minimal or non-existent. It's not as if callable bonds are a new phenomenon. And bond yields have been declining for 40 years, so the market is full of investors with first-hand experience with bonds being called. There simply aren't any "gotcha" moments for bond fund holders in this area.

Finally, even though bonds are usually priced AS IF they will produce the YTW it remains return that just because a bond CAN be called that doesn't mean it WILL be called. I own a municipal bond with coupon rate of 5.9% which has been callable since at least January but which has not been called yet.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
rockstar
Posts: 1121
Joined: Mon Feb 03, 2020 6:51 pm

Re: Impact of Called Bonds on Bond Funds

Post by rockstar »

It's same if you held the bonds yourself. You'd receive your funds now, and you'd deploy them into bonds most likely paying a lower yield, which would push down the yield overall. This is why treasuries are good when they pay good yields. They're not called early.
Topic Author
Scooter57
Posts: 1605
Joined: Thu Jan 24, 2013 9:20 am

Re: Impact of Called Bonds on Bond Funds

Post by Scooter57 »

Getting back to non-Treasury funds, if it isn't possible to calculate Yield to Worst for the category the article calls "extraordinary calls" which could happen at any time or to account for what the article calls "scoop and toss" in municipal bonds, couldn't we see sudden, unexpected, and significant drops in the yields of corporate-holding and municipal bond funds.

The section about states authorizing muni bonds to be called back (not within the provisions of the original issue, it sounds like) and replaced with bonds paying less with a longer duration is particularly troubling. This isn't the same as an individual bond holder getting a bond refunded and deciding to reinvest in one paying less. The investor has the option of buying a bond from a different issuer, or another kind of bond, or going for a different term. In this case, the bondholder sounds like he or she is being FORCED to buy a bond from the same issuer who just refunded it, but one that is paying less and lasts for a longer term. The article cites examples of where this has already happened in Virginia and Texas.
User avatar
vineviz
Posts: 8492
Joined: Tue May 15, 2018 1:55 pm

Re: Impact of Called Bonds on Bond Funds

Post by vineviz »

Scooter57 wrote: Thu Nov 12, 2020 11:02 am Getting back to non-Treasury funds, if it isn't possible to calculate Yield to Worst for the category the article calls "extraordinary calls" which could happen at any time ....
Of course such a calculation IS possible, and trivially easy to do."Yield to Worst" is precisely that, the yield you'd get if the worst possible outcome occurred. Buying a continuously callable bond at a price greater than par value simply results in a negative YTW calculation. Unless you really get a kick of of speculating in bonds, don't do that.
Scooter57 wrote: Thu Nov 12, 2020 11:02 am ...or to account for what the article calls "scoop and toss" in municipal bonds, ....
This phrase is simply scare-mongering by the authors. Of course issuers would prefer to call bonds with high coupon rates and issue new debt with lower coupon rates. This kind of flexibility is the entire reason that callable bonds exist to begin with. And investors know the risks involved with the bonds being called, which is why they demand higher yields (all else equal) for a callable bond than a non-callable bond.

Scooter57 wrote: Thu Nov 12, 2020 11:02 am ... couldn't we see sudden, unexpected, and significant drops in the yields of corporate-holding and municipal bond funds.
No, in part because the SEC yields of bond funds are already using YTW in calculating their SEC yields and in part because bond investors aren't ignorant.
Scooter57 wrote: Thu Nov 12, 2020 11:02 amIn this case, the bondholder sounds like he or she is being FORCED to buy a bond from the same issuer who just refunded it, but one that is paying less and lasts for a longer term.
I think this is a misunderstanding. When a bond is called, the bondholder gets cash in exchange for the bond. The original bondholders NEVER get a replacement bond automatically. Again, what the authors call "scoop and toss" is just the normal way that callable bonds work.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
alex_686
Posts: 7446
Joined: Mon Feb 09, 2015 2:39 pm

Re: Impact of Called Bonds on Bond Funds

Post by alex_686 »

vineviz wrote: Thu Nov 12, 2020 12:46 pm
Scooter57 wrote: Thu Nov 12, 2020 11:02 am Getting back to non-Treasury funds, if it isn't possible to calculate Yield to Worst for the category the article calls "extraordinary calls" which could happen at any time ....
Of course such a calculation IS possible, and trivially easy to do."Yield to Worst" is precisely that, the yield you'd get if the worst possible outcome occurred. Buying a continuously callable bond at a price greater than par value simply results in a negative YTW calculation. Unless you really get a kick of of speculating in bonds, don't do that.
To extend a bit, the article you are reading is fear mongering and correct. If you read the article it is saying that if it is a callable bond you can't calculate the yield with absolute mechanical certainty. Maybe you will get YTM, maybe YTW. Who knows?

Well, there are lots of sold methods to generate a estimated probability or a call, thus one can generate multiple paths a bond can take. So the world is a uncertain place that does not offer rock hard certain path. Unless you invest in non-callable bonds. Of course, those have lower yields.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
alex_686
Posts: 7446
Joined: Mon Feb 09, 2015 2:39 pm

Re: Impact of Called Bonds on Bond Funds

Post by alex_686 »

vineviz wrote: Thu Nov 12, 2020 12:46 pm
Scooter57 wrote: Thu Nov 12, 2020 11:02 amIn this case, the bondholder sounds like he or she is being FORCED to buy a bond from the same issuer who just refunded it, but one that is paying less and lasts for a longer term.
I think this is a misunderstanding. When a bond is called, the bondholder gets cash in exchange for the bond. The original bondholders NEVER get a replacement bond automatically. Again, what the authors call "scoop and toss" is just the normal way that callable bonds work.
I will second Vineviz here.

The article makes a semi-valid point. If a municipality "scoop and toss" - that is calls the loan and reissues a longer duration loan - it could be a sign that the underlying project has issues. Or maybe it is just the Treasury refinancing their loan to get a better rate, just like the many people refi their house for a better rate. So investors in the new issue should read the issuing documents. Then again, you should always read the issuing documents.

That being said, if you own the old muni bond nobody is forcing you to buy the new muni bond. I think the only time this can happen is when a bankruptcy judge is reorganizing a failed municipality.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
JackoC
Posts: 1940
Joined: Sun Aug 12, 2018 11:14 am

Re: Impact of Called Bonds on Bond Funds

Post by JackoC »

alex_686 wrote: Wed Nov 11, 2020 1:04 pm I would be relaxed about this.

The SEC Yield is a high quality number. It is robust. The weakness it has is also its virtue. It is a mechanical calculation and all funds have to use the same calculations.

Yield to Worst is not a very good metric. The probability of this happening is low.

There are better valuation techniques out there. You can download the list of CUSIPs from the latest filings and run the calculations yourself. The problem is that you have to model future changes in the yield curve. Not just the Treasury Yield curve, but also the credit spread and the probability of default. This is computationally intense. This is not the issue. The problem is that model has lots of subjective assumptions. Don't get me wrong, you still get high quality results. The problem is that it is not a standardized model that the SEC can force a fund to use.
I would be relatively relaxed about this for comparing bond funds which have similar durations and stated maturities (both the absolute number and relation between them). When it's comparing a fund with callable/prepayment risk bonds to treasuries which aren't callable or CD which is puttable (assuming early withdrawal with interest penalty, depending the terms), it can't be neglected and still arrive at a correct analysis. The embedded options you're short in the corporate, MBS or muni fund could be worth double digit basis points per annum against you. The interest rate option you hold with a CD which allows early withdrawal with penalty could be a similar amount in your favor. Some fund sponsors provide a model based 'option adjusted spread' (OAS) to treasuries number for their non-t bond funds (Vanguard doesn't) but that would still leave most people without modelling resources to calculate the option value on a CD EWP feature of particular terms.

Sometimes we are tempted to minimize 'known unknowns' because we lack the resources to quantify them, though others have those resources. In direct CD v 'low' risk bond fund case the difference in relative attractiveness considering or failing consider optionality is unfortunately non negligible. Although we can use rough rules of thumb without modeling.

I agree on the other statements. Yield to Worst is a basically useless measure from the era before interest rate options markets. Options modeling is needed to get an accurate answer, keeping in mind that what options models do is basically interpolate/extrapolate from real numbers where people are willing to buy/sell standalone options to the implied prices of options with different particulars and/or embedded in other instruments. There's some subjectivity at the margin, but not to the degree where it's plausible to assume the embedded options aren't worth anything, which is what you'd be doing to compare say CD to corp or muni bond fund just on after tax SEC v. CD yields.
Post Reply