I am looking to build a municipal bond ladder with target date ETFs. I am wondering if there is any opinion on BulletShares versus Ibonds. The differences I notice are BulletShares has lower volume, lower assets, higher yields, higher durations, and higher expenses.
BulletShares: average volume about 1K-10K, expense ratio 0.18%
Ibonds: average volume 10K-60K, expense ratio 0.10%
For both, the longer maturity target dates tend to have lower volumes
I am not sure why, but the yields are consistently higher for BulletShares despite the same average credit quality. BulletShares s tends to have longer effective duration for the same target year. I think BulletShares might have more callable bonds (but their yield to worse still is higher than Ibonds yield to maturity)
BulletShares 2023: Effective duration 3.3 years. Yield to maturity 1.87%, yield to worst 0.84%
Ibonds 2023: effective duration 2.8 years, Yield to maturity 0.52%
BulletShares 2026: Effective duration 7.2 years, Yield to maturity 2.79%, yield to worst 1.79%
Ibonds 2026 effective duration 5.4 years Yield to maturity 0.9%
Part of me says take the better yield and since I plan to hold until the last year don’t worry about liquidity, but part of me says this is too good to be true and maybe Guggenheim’s bond pricing methodology is artificially boosting their yields, or they are taking excessive duration or call risks on callable bonds. Any input would be appreciated.
Comments
iBonds® Dec 2023 Term Muni Bond ETF prospectus.
Invesco Exchange-Traded Self-Indexed Fund Trust prospectus.
iBonds track an S&P AMT-Free Muncipal Series December 20xx Index™ while BulletShares track a proprietary Invesco BulletShares®MunicipalBond 20xx Index.
The first difference is obvious from the names of the indexes. iBonds invest in AMT-free munis. Consequently they may pay a little less interest. The last major change in tax laws made AMT moot for most investors. So generally there's no advantage in accepting a lower yield to get AMT-free payments. I suspect the difference in yields between AMT-free and private activity munis is not so large for the same reason.
The second and most important difference is callability. The S&P indexes include only "non-callable U.S. municipal bonds maturing in 20[xx]." The Invesco indexes "may include callable, puttable, and pre-funded bonds." The corresponding ETFs explicitly include call risk among their principal risks. Call risk is not mentioned in the iBonds prospectus. So it appears that the difference in call risk between the ETFs is not just quantitative (more or less call risk), but qualitative (is there call risk or not?).
The third difference, related to callability is how the maturity year is determined. Obviously with noncallable bonds (iBonds), the year of maturity is, well, the year the bonds mature. But with BulletShares, a callable bond may mature in 2024 (or even Jan 2025) and be considered a 2023 bond, so long as the first call date is within 13 months of maturity (and if the call is at par). That would seem to mean that the fund could include some 2024/2025 bonds trading at a discount (so YTW = YTM). This would make the SEC yield higher, since you're really buying 2024 bonds rather than 2023 bonds. But it would also seem to subject the bonds (and thus the ETF) to greater interest rate risk.
I haven't worked through all the differences in the yields, durations, etc. There could be more going on. Still, this is a reasonable start at identifying the fundamental differences. I would add another: S&P has much more experience in designing indexes than Invesco (or its predecessor, Accretive Asset Management [acquired by Guggenheim]), and Blackrock (with its predecessor, Barclays) has earned a reputation for well managed index funds. I don't know how well Invesco runs its index funds.
Finally, according to the fact sheets, both 2023 ETFs have the same 0.18% ER.
iBonds 2023 (IBML) fact sheet.
BulletShares 2023 (BSMN) fact sheet.
I have some fixed expenses coming up in 5-6 years and I was thinking of using these defined maturity ETFs to pay for those.