Sound Investing For Every Stage of Life
ETFs Questions & Answers
by Chris Pedersen
Note: These questions from readers and listeners are in addition to many addressed in the article, Best-in-Class ETFs for Ultimate Buy & Hold Portfolio (Updated March, 2019).
Q: Why not use the Vanguard Value Factor Fund (VFVA, 0.13% expense ratio) in the best-in-class ETF recommendations?
It has a price-to-book ratio of 1.08, 766 companies and an average market cap of $6.4B.
A: There are a few reasons we didn't consider VFVA. The main reason is that it's neither a large-cap or small-cap value fund, so it doesn't fit into any of the asset classes we need for the Ultimate Buy-and-Hold portfolio. A second reason is that it's only a year old, so there's not much track record by which to judge it. The third reason is that it's only got $53.5M in assets, so it trades with a bid-ask spread of 0.10% vs. 0.07% for SLYV and 0.04% for RPV according to ETF.com. Could someone do well investing in VFVA? It's quite possible they could. Would we recommend it if they had more history and assets under management? Probably not because it doesn't fit the asset classes Paul's portfolios use.
Q: Is some mid-cap OK to have in the portfolio now? If so, would using dedicated mid-cap blend & mid-cap value funds help even more?
Adding VTI/RPV adds some mid-cap and I have always felt this could be beneficial for diversification purposes, but Paul's work has always preached against mid-cap.
A: I don't believe Paul has ever said mid-caps are bad. Rather, he's pointed out that having a portfolio constructed with less overlap gives you more chance to benefit from rebalancing. If you implement Paul's portfolio using DFA funds, you can get large diversified funds in all of the categories Paul recommends. Unfortunately, with ETFs, there are more tradeoffs. There are no LCV or SCV ETFs with as many companies and as deep a value discount as the DFA funds. To get a cost-effective deep value discount, I went with RPV. To get a larger number of companies, I added VTI. It's not a perfect implementation of Paul's Ultimate Buy-and-Hold strategy, but it's the best I could do with what's available.
Q: You removed the small-cap-value RZV ETF from the original Best-in-Class recommendations because of its negative momentum, yet you chose RPV for the latest LCV ETF, so how did you justify similar negative momentum with RPV’s construction?
From what I understand RPV is constructed the same way as RZV (except for the size).
A: When I ran the factor analysis to predict future returns based on past factor loading, RPV had a predicted return that was higher than any of the other fund options in the US LCV space. That means its historical value, quality and size attributes overcame any negative momentum, alpha or expense. The same was not true for RZV.
Q: Why is EFV still recommended for use in the Best-in-Class ETF taxable accounts given that it carries a taxable yield of 4.20%, which is about the same as VNQ the US REIT?
Is there another foreign large value ETF with lower dividend yield, or should we stick with EFV, but use it only in tax-deferred accounts?
A: This is an excellent question, and could be extended to other high-yielding equity ETFs such as DGS which yields 3.82%.
The first thing to remember is that not all investment income is taxed equally. Qualified dividend income for long-term holdings is taxed in the US at 15 to 25%. REIT income used to be taxed at an investor's marginal tax rate which was typically higher. This was the primary reason for recommending placing REITs in tax-deferred accounts, but this changed in 2018 with the Tax Cuts & Jobs Act (TCJA) which introduced a 20% reduction in REIT dividend income tax rates.
This table from Inland Real Estate Investment Corporation shows the impact of the change. The effective US tax rate on ordinary REIT dividend income now ranges from 8% to 29.6%. This neglects any state income taxes, but shows that the difference between tax liability for dividends and REITs has narrowed. So, does that mean we should no longer care whether REITs are in taxable or tax-deferred accounts? Possibly, but possibly not. Remember, many of the provisions of the TCJA will expire and revert, and tax law can change. For now, we'll keep REITs in the tax-deferred portfolios, but investors should be aware of the changes and decide what's best for their individual situations.
The second question, whether there are foreign large value ETFs with lower dividend yield is easier to answer. The few I identified were PXF, FNDF (similar to PXF, but lower expense ratio), and IVLU. Their yields are 3.10%, 3.17% and 2.60% respectively which are all lower than EFV's 4.20%. IVLU is much less liquid and, according to ETF.com, trades with an average spread of 0.16%. FNDF is interesting, but when I analyzed EFV vs. SFNNX (Schwab mutual fund similar to FNDF but with longer history), EFV had a 0.65% higher factor-predicted expected return. So, yes there are alternatives with lower yields, but whether you'll do better with them after taxes over the long-haul is unclear.
Finally, should EFV (and or DGS?) only be used in tax-deferred accounts? I don't think so. Yes, they have higher yields, but those yields receive tax-advantaged treatment. If you have the flexibility to put your highest yielding assets in tax-deferred accounts, go ahead, but if it requires selling and buying to get there, please consider the tax consequences of those transactions too before starting down that path.
Q: The “best in class portfolios” primarily target the value and small factors, so what makes these two factors superior to the other factors? Why shouldn’t I for example buy multifactor ETFs such as VFMF, INTF and ISCF?
A: Paul's Ultimate Buy-and-Hold portfolio emphasizes value and size because they historically offer premiums and they are broadly available in several different mutual funds and ETFs. Other factors such as momentum and low-volatility have larger premiums, but they are also less-widely available and may be more expensive to access.
Since the Ultimate Buy-and-Hold portfolio has 10 asset classes, including 8 that are categorized by value and size, I need to find ETFs that fit those classes AND add up to a great overall portfolio.
There are many other ways to create portfolios that would have tilts to small and value which would produce similar results, but they wouldn't be as useful to Paul's listeners. If you find through your own analysis that a multifactor ETF meets your needs, that's great. Personally, I like the idea of a multi-factor fund, but many have issues.
VFMF is relatively small (~$85M total assets), has about 0.09% bid-ask-spread, and only 13 months of history. INTF is intriguing with >$1B in assets, but 4 years of history is still not all that long. ISCF is also very small (~$80M total assets) and has a typical bid-ask-spread of 0.33%. That's a lot to pay every time you buy. These attributes may not worry you personally, but they are some of the reasons I couldn't consider them for the best-in-class ETF recommendations.