s-l-o-w-l-y try to explain why an investor, going forward, should invest in either taxable or municipal bond funds in their portfolio's fixed income sleeve instead of say, 5-yr, 4.50%, non-callable CDs.
But please leave out the widely understood part about past performance being no guarantee of future results. Got that part.
s-l-o-w-l-y? I don't know how one
writes slowly, but if you're asking for gory details, I'm happy to oblige.
Let's start with the question. It gives as one option a 5-yr, 4.5% non-callable CD. Even non-callable CDs may be redeemed early by a debtor. Should a bank fail (no longer an unexpected event), high yielding CDs may be redeemed by the FDIC or reset to a lower rate by an acquiring bank.
Thus actual rate of return, though highly likely to be as stated, is not certain. Some posters have addressed this risk by saying they would only buy CDs from well-managed banks. You did not. (I discovered that by reading your post
s-l-o-w-l-y.)
The question carries an implicit assumption that an investor is absolutely certain that they will not want to withdraw money early. Any possibility of pulling money out would expose the investor to the same interest rate risk as experienced by a bond fund.
Further, the investor would have less flexibility in selling off a CD (basically, all or nothing on a per-CD basis) as opposed to a bond fund where one can sell as little as 0.001 shares. In addition, the investor would take a big hit on the bid-ask spread that isn't present when selling bond fund shares.
So already we have a reason - flexibility - for an investor to consider using a bond fund rather than a CD with the same (or even marginally lower) expectation value of rate of return. Maybe you wouldn't, but the question was asked about any investor.
That brings us to the expected rate of return going forward. As explained (slowly) above, aside from minor risks expected rate of return is pretty well though not quite 100% certain for the CD. The challenge is to figure out at a minimum what the expected rate of return of a bond fund is. Ideally one would want to estimate not only the expected return but the dispersion of possible outcomes. That plays into risk analysis, which I'll (slowly) get to.
What you did was look at past 5 year returns. In do so, you acknowledged but disregarded the fact that past returns may be poor predictors of future returns. Putting that problem aside, you also disregarded that fact that
all rates were lower over the past five years than they are now. One could make at least a passing attempt at compensating for this this by looking at 5 year CD rates 5 years ago vs. now and adjusting the question's comparison accordingly.
According to depositaccounts.com, 5 years ago the best one could do was
about 3.3%, while now it is, as you stated,
about 4.5%. So if we use your method of estimating bond fund returns going forward based on past performance, we should adjust those past performance figures upward by about 1.2%. I'll leave that as an exercise for the reader.
But situations change, and as already noted, past returns may be poor predictors.
bond funds returning 7%-8% over any LT period of time ... just doesn't happen.This is easy to disprove by counter example. It does sometimes happen. From inception (12/31/1986) through the end of 2002, VBMFX had an
annualized return of 7.85%. More generally, looking at 10 year rolling averages, AAA corporate bonds had annualized returns ranging between 7.30% and 11.29%(!) for
every 10 year period ending between 1975 and 2000. So, over a period of
at least 30 years (three non-overlapping ten year periods between 1965 and 1995) the average return on AAA corporates was more than 7.3%. That seems long term enough.
Baa corporates did even better. Their 10 year average rate of return surpassed 7.26% every 10 year period ending between 1973 and 2005, topping out at a 10 year average return of 12.84%. Interested in T-bonds? The same analysis shows that 10 year T-bonds had ten year rolling average returns exceeding 7.12% for every period ending between 1984 and 2002.
Source is
spreadsheet from NYU/Stern, whose ultimate data source is FRED.
Even though you asked for an explanation given
s-l-o-w-l-y, I can understand your quick and dirty search for 5 year bond fund returns. I can understand your saying that there were 1921 funds even though 138 of them didn't have five year records. I can understand your excluding the 114 funds that are closed at Fidelity, even if some of them were open five years ago.
I can understand your using Fidelity's screener though it gives fewer than half the number taxable bond funds with 5 year records that
Portfolio Visualizer's screener returns. Because a reasonable (though unverified) assumption is that the funds currently open and sold by Fidelity are representative of all the bond funds available five years ago.
But when it comes to expected returns going forward, one is going to have to do better than assert 7%-8% LT returns just don't happen.
So far, most of what I've done is explain why some of the data presented is either unhelpful, biased, or simply wrong. I've also provided one rationale for preferring bond funds to broker-sold CDs, viz. flexibility.
Implicit in your reasoning (and that of most others) is that investors are risk averse. Someone who is truly risk indifferent will consider a bond fund with an expected 4.5% return to be just as good - not better, not worse - than a CD at that rate. (As I explained before, given the additional risk of possibly needing access to the money, someone who is risk indifferent would demand a higher rate from the CD than from the bond fund.)
An investor who is only slightly risk averse will not need a much higher expected rate of return to choose the bond fund. So the question comes down to: what is a reasonable expectation for five year returns of some bond funds? Past performance used blindly clearly is not a good approach to answer this; there have been extended periods of time when bonds have returned well in execess of 7%. It could happen again.
The question is not what has happened before, but what (and why) one expects going forward.
Others have offered some explanations for better returns going forward - based on their expectations for interest rates. I could dig up a bunch of papers explaining that over particular long terms, what one should expect from bond funds (total return) is determined by their current yields. That's how I look at bond funds, assuming that I'll hold for a long period of time.
Checking out current SEC yields, it's not hard to find several familiar funds yielding above 4.5%. Many multisector funds sport yields above 6% (i.e. 1.5% or more above the CD) such as DBLNX (8.69%) and MWFSX (7.69%). PIMIX (5.86%) comes in just below 6%, but still well above the CD rate. The core plus fund TGLMX has a 6.17% yield. Even a fund as conservative as FCNVX has a yield above 5% and can serve as a dynamic (flexible) cash backup.
(These are not recommendations; just a listing a few familiar funds.)
We've
seen this question before:
RPHYX/RPHIX vs. 6 mo T-bills. As here, I used current data, not past performance (i.e. 2022 or earlier). What one gleans from past performance is general behavior of a fund, not performance that can be easily extrapolated.
