Morningstar: What to Consider When Picking Multifactor ETFs

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Morningstar: What to Consider When Picking Multifactor ETFs

Post by vineviz »

Morningstar recently published a paper called "A Framework for Analyzing Multifactor Funds", which I don't think I've seen discussed here.

Christine Benz and Alex Bryan, from Morningstar.com, discuss it in this video. Bryan makes a case that taking a careful look at fund expenses matters a lot, just as with other investment vehicles, and that integrated multifactor funds generally make more sense for most investors.

https://www.morningstar.com/videos/8707 ... -etfs.html

"The basic idea behind a multifactor fund is that it's not a bad idea to diversify across these different factor strategies, because although each factor strategy, like targeting stocks with low valuations, small market capitalization, high-quality, good momentum, all those things have tended to work well over the very long term, but they each go through their own cycle of outperformance and underperformance. The basic idea is by putting these different factors together in a portfolio you can diversify your risk reducing the risk of underperforming for an extended period of time, and we think that that would make it easier to stick with a multifactor fund than to try to go it alone with a single factor."
Here's a link to the full paper: http://corporate1.morningstar.com/Resea ... tor-funds/ Not a great deal of ground breaking going on, but its a decent basic introduction to the concept.

The framework Bryan outlines:

1. What is the fund’s selection universe?
2. Which factors does the fund target?
3. How does the fund measure its targeted factors?
4. How does the fund combine its targeted factors?
5. How aggressively does the fund pursue its targeted factors?
6. Do the fund’s factor tilts wash out, or does one factor dominate?
7. Are there any constraints on the portfolio?
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by vineviz »

In case anyone is interested in a short list of the kinds of multifactor funds that might function as a core holding in a three or four fund portfolio, I offer the following (decidedly NOT comprehensive) list:
  • Hartford Multifactor US Equity ETF (ROUS)
  • iShares Edge MSCI Multifactor USA ETF (LRGF)
  • JPMorgan Diversified Return US Eq ETF (JPUS)
  • Vanguard US Multifactor ETF (VFMF)
  • Xtrackers Russell 1000 Cmprhsv Fac ETF (DEUS)
These are funds that meet a balance of criteria similar to the ones presented in the M* report: they have strong factor exposure, track the total stock market reasonably well, have expenses at 20 bps or less, and are either large cap or multi cap US focused.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
jhfenton
Posts: 4754
Joined: Sat Feb 07, 2015 10:17 am
Location: Ohio

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by jhfenton »

vineviz wrote: Tue Jul 03, 2018 8:08 am In case anyone is interested in a short list of the kinds of multifactor funds that might function as a core holding in a three or four fund portfolio, I offer the following (decidedly NOT comprehensive) list:
  • Hartford Multifactor US Equity ETF (ROUS)
  • iShares Edge MSCI Multifactor USA ETF (LRGF)
  • JPMorgan Diversified Return US Eq ETF (JPUS)
  • Vanguard US Multifactor ETF (VFMF)
  • Xtrackers Russell 1000 Cmprhsv Fac ETF (DEUS)
These are funds that meet a balance of criteria similar to the ones presented in the M* report: they have strong factor exposure, track the total stock market reasonably well, have expenses at 20 bps or less, and are either large cap or multi cap US focused.
It's not an ETF, but Vanguard offers Vanguard US Multifactor Admiral Shares (VFMFX) as well. It is a twin to VFMF, but unlike most of Vanguard's ETF and funds, they are not share classes of the same fund.

Based solely on design and initial portfolio characteristics, I consider VFMF and VFMFX to be the clear winner in the space. I bought VFMFX on the first day. I started a thread shortly after launch discussing the portfolio characteristics of Vanguard's new factor funds. I really like the strong factor characteristics and the all-cap design. (They end up roughly 1/3 large, 1/3 mid, 1/3 small.)

I also bought VFVA (Vanguard US Value Factor ETF) after it settled into trading a bit.
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by vineviz »

jhfenton wrote: Tue Jul 03, 2018 8:41 am Based solely on design and initial portfolio characteristics, I consider VFMF and VFMFX to be the clear winner in the space.
I also bought VFVA (Vanguard US Value Factor ETF) after it settled into trading a bit.
VFMF is my personal preference as well.

I started with iShares Core S&P Total US Stock Mkt ETF (ITOT) as my core US index holding, but have built up VFMF to roughly a 50/50 split with ITOT as I get more comfortable with the multifactor fund's track record.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by SimpleGift »

vineviz wrote: Tue Jul 03, 2018 8:55 am
jhfenton wrote: Tue Jul 03, 2018 8:41 am Based solely on design and initial portfolio characteristics, I consider VFMF and VFMFX to be the clear winner in the space.
I also bought VFVA (Vanguard US Value Factor ETF) after it settled into trading a bit.
VFMF is my personal preference as well.

I started with iShares Core S&P Total US Stock Mkt ETF (ITOT) as my core US index holding, but have built up VFMF to roughly a 50/50 split with ITOT as I get more comfortable with the multifactor fund's track record.
Just out of curiosity, what are typical or recommended allocations to these multi-factor ETFs, within an otherwise diversified portfolio of risk assets? In other words, how much does one reasonably carve out from the equity potion of the portfolio to devote to these multi-factors funds? Any research on this?
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by vineviz »

SimpleGift wrote: Tue Jul 03, 2018 10:35 am Just out of curiosity, what are typical or recommended allocations to these multi-factor ETFs, within an otherwise diversified portfolio of risk assets? In other words, how much does one reasonably carve out from the equity potion of the portfolio to devote to these multi-factors funds? Any research on this?
I think there are lots of ways to approach that question, and I don't know that there is a "typical" recommendation.

The funds I listed above are conservative enough, IMHO, that - for a sophisticated investor - they could totally replace Vanguard Total Stock Market Index Fund or the equivalent in a portfolio. They have strong factor exposure, for sure, but aren't taking extreme bets as far as I can tell (none of these funds are terrible old).

I think if you wanted to ADD a multifactor fund to an otherwise diversified portfolio then you probably are looking for something slightly different: higher factor exposure (to offset the "blandness" of Vanguard Total Stock Market Index) and possibly lower market beta (to get the most bang for your buck, so to speak). In that kind of situation, I'd probably use a simple two-factor fund (e.g.iShares S&P Small-Cap 600 Value ETF (IJS)) or single factor fund (e.g. Vanguard US Liquidity Factor ETF (VFLQ)).
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
maximuum
Posts: 33
Joined: Mon Jun 16, 2008 3:34 pm

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by maximuum »

Does the fact that VFMFX (mutual fund) is not a share class of VFMF (ETF) mean that VFMF should preferentially be used in taxable accounts?
User avatar
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by SimpleGift »

vineviz wrote: Tue Jul 03, 2018 10:52 am The funds I listed above are conservative enough, IMHO, that - for a sophisticated investor - they could totally replace Vanguard Total Stock Market Index Fund or the equivalent in a portfolio. They have strong factor exposure, for sure, but aren't taking extreme bets as far as I can tell (none of these funds are terrible old).

I think if you wanted to ADD a multifactor fund to an otherwise diversified portfolio then you probably are looking for something slightly different: higher factor exposure (to offset the "blandness" of Vanguard Total Stock Market Index) and possibly lower market beta (to get the most bang for your buck, so to speak).
Thank you. I didn't realize that the multi-factor funds could be considered a replacement for Total Stock Market Index Fund — but it makes some sense, since they seem to be diversified across the range of market caps (large to small), as well as the various factor exposures.
User avatar
jhfenton
Posts: 4754
Joined: Sat Feb 07, 2015 10:17 am
Location: Ohio

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by jhfenton »

I am essentially using VFMFX as a replacement for a Total Stock fund along with VFVA for an additional value tilt. I actually took the new funds as an opportunity to simplify. At this point, they are the only U.S. funds we own in our individual accounts. We have a large chunk in VMVAX (Vanguard Mid Cap Value Admiral) and DFSTX (DFA US Small Cap I) in my 401(k). Those are our only four U.S. stock funds (other than a trivial amount in VIOV / Vanguard S&P 600 Small Cap Value in taxable).

If we didn't have 401(k) constraints, I might have all of our U.S. allocation in VFMFX and VFVA.
maximuum wrote: Tue Jul 03, 2018 11:16 am Does the fact that VFMFX (mutual fund) is not a share class of VFMF (ETF) mean that VFMF should preferentially be used in taxable accounts?
I think so. I haven't added any of them in taxable yet because they are brand new funds, I want to see a record of solid tax efficiency on a fund before I consider buying it in taxable. But I expect I would opt for VFMF over VFMFX, even if they both look good the first couple of years. Beyond tax efficiency, VFMF would be portable. VFMFX might not be portable.
User avatar
whodidntante
Posts: 13114
Joined: Thu Jan 21, 2016 10:11 pm
Location: outside the echo chamber

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by whodidntante »

I think it is reasonable to use a multifactor fund as a replacement for a total market, float weighted index fund. Up to 100% if practical. I can't do that because of limitations in my 401k, and capital gains in taxable. I like the integrated multifactor funds.
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by vineviz »

It’s just a hunch, but the design of the Vanguard Multifactor finds makes me think they are hoping it will act as a core fund for some clients.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
MJW
Posts: 724
Joined: Sun Jul 03, 2016 7:40 pm
Location: Pacific Northwest

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by MJW »

vineviz wrote: Tue Jul 03, 2018 12:57 pm It’s just a hunch, but the design of the Vanguard Multifactor finds makes me think they are hoping it will act as a core fund for some clients.
Do you have any thoughts on ishares Global Multifactor ETF (ACWI)? I'm still too "convictionless" about investing to adopt a factor-based approach, but I've taken an interest in at least learning more about them.
User avatar
Taylor Larimore
Posts: 32842
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

Multi-Factor funds vs. Total Stock Market

Post by Taylor Larimore »

whodidntante wrote: Tue Jul 03, 2018 12:31 pm I think it is reasonable to use a multifactor fund as a replacement for a total market, float weighted index fund. Up to 100% if practical. I can't do that because of limitations in my 401k, and capital gains in taxable. I like the integrated multifactor funds.
whodidntante:

The oldest multi-factor fund I can find is PNC Multi-Factor Small-Cap Value C (PSVCX). Let's compare it with Vanguard Total Stock Market (VTSAX) as of June 30, 2018:

Fund-----------15-year return------After-tax

PSVCX-------------7.01%---------------5.71%

VTSAX-------------9.77%---------------9.29%

In my opinion, it could be a serious mistake to replace Total Stock Market Index Fund with a multifactor fund.

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
User avatar
jhfenton
Posts: 4754
Joined: Sat Feb 07, 2015 10:17 am
Location: Ohio

Re: Multi-Factor funds vs. Total Stock Market

Post by jhfenton »

Taylor Larimore wrote: Tue Jul 03, 2018 1:38 pm The oldest multi-factor fund I can find is PNC Multi-Factor Small-Cap Value C (PSVCX). Let's compare it with Vanguard Total Stock Market (VTSAX) as of June 30, 2018:
Sigh. Taylor, As some of have pointed out a few times in response to your raising the specter of PSVCX previously, it is a completely inappropriate comparison. First, PSVCX has an expense ratio of 1.85%. Second, for most of the time period, it was not a multi-factor fund. They added that more recently as a buzz word. Third, it has an expense ratio of 1.85%. Fourth, it is literally the worst performing small-cap value fund over the trailing 15 years. Fifth, it has an expense ratio of 1.85%.

Vanguard's VFMFX has an initial expense ratio of 0.18% that should come down further as the fund attracts assets. VFMFX is a true multi-cap, multi-factor fund providing broad U.S. market coverage. I understand and respect that you have no interest in owning it.
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Multi-Factor funds vs. Total Stock Market

Post by vineviz »

Taylor Larimore wrote: Tue Jul 03, 2018 1:38 pm The oldest multi-factor fund I can find is PNC Multi-Factor Small-Cap Value C (PSVCX). Let's compare it with Vanguard Total Stock Market (VTSAX) as of June 30, 2018:

Fund-----------15-year return------After-tax

PSVCX-------------7.01%---------------5.71%

VTSAX-------------9.77%---------------9.29%

In my opinion, it could be a serious mistake to replace Total Stock Market Index Fund with a multifactor fund.
Hey, if we're going to play the "pick the fund with good past performance game" let's let loose. DFA U.S. Targeted Value Portfolio (DFFVX) is a multi-factor fund with a 15 year track record. Let's compare it with Vanguard Total Stock Market (VTSAX) as of June 30, 2018:

Fund-----------15-year return------After-tax

DFFVX-------------11.38%---------------9.94%

VTSAX-------------9.77%---------------9.29%

If Taylor is consistent in his beliefs, I presume he thinks it would be a serious mistake to replace DFA U.S. Targeted Value Portfolio with a total market index fund.
Last edited by vineviz on Tue Jul 03, 2018 7:26 pm, edited 1 time in total.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
HippoSir
Posts: 269
Joined: Tue Jul 03, 2018 2:56 pm

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by HippoSir »

maximuum wrote: Tue Jul 03, 2018 11:16 am Does the fact that VFMFX (mutual fund) is not a share class of VFMF (ETF) mean that VFMF should preferentially be used in taxable accounts?
I'm very curious as to the answer to this question as well, I very much like the construction of VFMF/VFMFX, and I've moved to it in my nontaxable, but I'm worried about it throwing off too much capital gains for use in taxable.
TM90
Posts: 132
Joined: Mon Mar 06, 2017 1:13 am

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by TM90 »

I bought a multifactor etf, could handle everything except the tracking error, went back to regular indexing after a month.. Cost me a lot of transaction costs!

Stay the course
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by vineviz »

Here's Morningstar's style box diagram for U.S. Multifactor ETF (VFMF).

Image

The style box only maps two factors, but that's a core fund in my mind.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
whodidntante
Posts: 13114
Joined: Thu Jan 21, 2016 10:11 pm
Location: outside the echo chamber

Re: Multi-Factor funds vs. Total Stock Market

Post by whodidntante »

Taylor Larimore wrote: Tue Jul 03, 2018 1:38 pm
whodidntante wrote: Tue Jul 03, 2018 12:31 pm I think it is reasonable to use a multifactor fund as a replacement for a total market, float weighted index fund. Up to 100% if practical. I can't do that because of limitations in my 401k, and capital gains in taxable. I like the integrated multifactor funds.
whodidntante:

The oldest multi-factor fund I can find is PNC Multi-Factor Small-Cap Value C (PSVCX). Let's compare it with Vanguard Total Stock Market (VTSAX) as of June 30, 2018:

Fund-----------15-year return------After-tax

PSVCX-------------7.01%---------------5.71%

VTSAX-------------9.77%---------------9.29%

In my opinion, it could be a serious mistake to replace Total Stock Market Index Fund with a multifactor fund.

Best wishes.
Taylor
What would a factor thread be without this disingenuous comparison?
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Multi-Factor funds vs. Total Stock Market

Post by vineviz »

whodidntante wrote: Tue Jul 03, 2018 5:22 pm What would a factor thread be without this disingenuous comparison?
Someday I hope to meet someone who truly understands factor investing AND thinks it is a blatantly flawed approach.

That day is not today.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Ron Scott
Posts: 1090
Joined: Tue Apr 05, 2016 5:38 am

Re: Multi-Factor funds vs. Total Stock Market

Post by Ron Scott »

vineviz wrote: Tue Jul 03, 2018 5:32 pm
Someday I hope to meet someone who truly understands factor investing AND thinks it is a blatantly flawed approach.
My guess is if you found someone who didn't think factor investing was an improvement over the basic 2-fund approach you'd conclude they didn't truly understand it.
Retirement is a game best played by those prepared for more volatility in the future than has been seen in the past. The solution is not to predict investment losses but to prepare for them.
palaheel
Posts: 627
Joined: Wed Mar 22, 2017 7:35 am

Re: Multi-Factor funds vs. Total Stock Market

Post by palaheel »

vineviz wrote: Tue Jul 03, 2018 2:39 pm
Hey, if we're going to play the "pick the fund with good past performance game" let's let loose. DFA U.S. Targeted Value Portfolio (DFFVX) is a multi-factor fund with a 15 year track record. Let's compare it with Vanguard Total Stock Market (VTSAX) as of June 30, 2018:

Fund-----------15-year return------After-tax

DTMVX-------------11.38%---------------9.94%

VTSAX-------------9.77%---------------9.29%

If Taylor is consistent in his beliefs, I presume he thinks it would be a serious mistake to replace DFA U.S. Targeted Value Portfolio with a total market index fund.
I'm confused (easily). You mention DFFVX (which I own, fwiw), but then list DTMVX in the comparison. Is there a typo, or missing comparison?
Nothing to say, really.
User avatar
jhfenton
Posts: 4754
Joined: Sat Feb 07, 2015 10:17 am
Location: Ohio

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by jhfenton »

HippoSir wrote: Tue Jul 03, 2018 3:00 pm
maximuum wrote: Tue Jul 03, 2018 11:16 am Does the fact that VFMFX (mutual fund) is not a share class of VFMF (ETF) mean that VFMF should preferentially be used in taxable accounts?
I'm very curious as to the answer to this question as well, I very much like the construction of VFMF/VFMFX, and I've moved to it in my nontaxable, but I'm worried about it throwing off too much capital gains for use in taxable.
I would rate it as a strong possibility that VFMFX will not be tax-friendly. With a mandate to maintain factor exposure leading to what I suspect will be consistent moderate turnover, I can see substantial capital gains distributions in its future. I purchased it in my Roth IRA and my wife's rollover IRA. If the time comes when I decide to add some in taxable, I will buy VFMF.

(I wonder whether there was a regulatory quirk that led Vanguard to keep their first "active" ETF separate from its companion mutual fund. The ETF-as-a-share-class structure is quirky in itself and unique to/patented by Vanguard. I wonder if they ran into an incompatibility of some sort trying to combine the two structures.)
TM90 wrote: Tue Jul 03, 2018 3:31 pm I bought a multifactor etf, could handle everything except the tracking error, went back to regular indexing after a month.. Cost me a lot of transaction costs!

Stay the course
In my entire investing career dating back to the late '90s--1998 was the start of our non-401(k) investing--I have never at any time held anything close to the U.S. market portfolio. I have always tilted us to small value and emerging markets. And since I believed in tilting, I have never paid significant attention to "tracking error" compared to what are clearly irrelevant benchmarks.

Clearly, in 1998 and 1999, my small value tilted portfolio underperformed the tech bubble-led S&P 500. And to be honest, that just reinforced my decision. I didn't trust the valuations in the tech-dominated market-cap indices one bit. And while the "market" crashed in 2000-2001, our portfolio was up in 2000 and 2001.
Angst
Posts: 2968
Joined: Sat Jun 09, 2007 11:31 am

Re: Multi-Factor funds vs. Total Stock Market

Post by Angst »

I'm not so sanguine about the notion of comparing a TSM investment with a Multi-factor fund. Much in the way not so long ago we were hearing about all the new funds that could give us "equity-like returns" without the volatility, I'm more than content to just take a wait and see approach to things. Mixing culinary metaphors here, I say the proof is in the pudding but this cake ain't baked yet. It smells pretty good though. :) Good enough for me to be holding some of AQR's Small Cap Multi-Style fund (not to mention QSPIX as well), but only time will tell. My big wonder is how will things play out as component stocks move in and out of these fund's portfolios? Over time, there has to be some "factor drift" turnover. Hopefully that shouldn't mean selling low and buying high?

It all just makes my head spin though when I try to conceptualize how the algorithms are going to manage everything when any given stock may easily have varying amounts of both positive and negative exposure to the desired factors. It's enough to make me think that certain fund managers may be a lot better at running a multi-factor fund than other managers. Then again, I have to come up with something to justify my ~.5% ER differential at AQR!
:oops:
User avatar
triceratop
Posts: 5838
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: Multi-Factor funds vs. Total Stock Market

Post by triceratop »

vineviz wrote: Tue Jul 03, 2018 5:32 pm
whodidntante wrote: Tue Jul 03, 2018 5:22 pm What would a factor thread be without this disingenuous comparison?
Someday I hope to meet someone who truly understands factor investing AND thinks it is a blatantly flawed approach.

That day is not today.
You’ll only find someone like that in Scotland.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Multi-Factor funds vs. Total Stock Market

Post by vineviz »

palaheel wrote: Tue Jul 03, 2018 6:09 pm
vineviz wrote: Tue Jul 03, 2018 2:39 pm
Hey, if we're going to play the "pick the fund with good past performance game" let's let loose. DFA U.S. Targeted Value Portfolio (DFFVX) is a multi-factor fund with a 15 year track record. Let's compare it with Vanguard Total Stock Market (VTSAX) as of June 30, 2018:

Fund-----------15-year return------After-tax

DTMVX-------------11.38%---------------9.94%

VTSAX-------------9.77%---------------9.29%

If Taylor is consistent in his beliefs, I presume he thinks it would be a serious mistake to replace DFA U.S. Targeted Value Portfolio with a total market index fund.
I'm confused (easily). You mention DFFVX (which I own, fwiw), but then list DTMVX in the comparison. Is there a typo, or missing comparison?
It's a typo: the returns are for DFFVX. Thanks for pointing out the error.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Multi-Factor funds vs. Total Stock Market

Post by vineviz »

Ron Scott wrote: Tue Jul 03, 2018 5:54 pm
vineviz wrote: Tue Jul 03, 2018 5:32 pm
Someday I hope to meet someone who truly understands factor investing AND thinks it is a blatantly flawed approach.
My guess is if you found someone who didn't think factor investing was an improvement over the basic 2-fund approach you'd conclude they didn't truly understand it.
That's certainly possible. Either way, I hope we find out someday.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by vineviz »

jhfenton wrote: Tue Jul 03, 2018 6:10 pm (I wonder whether there was a regulatory quirk that led Vanguard to keep their first "active" ETF separate from its companion mutual fund. The ETF-as-a-share-class structure is quirky in itself and unique to/patented by Vanguard. I wonder if they ran into an incompatibility of some sort trying to combine the two structures.)
It's possible they were anticipating the new SEC rules that came out this week. I'm not sure if any of the changes made this arrangement better than the old one, but it seems plausible.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
Taylor Larimore
Posts: 32842
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

There is more than one road to Dublin

Post by Taylor Larimore »

Bogleheads:
There is more than one road to Dublin.
Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
User avatar
jhfenton
Posts: 4754
Joined: Sat Feb 07, 2015 10:17 am
Location: Ohio

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by jhfenton »

vineviz wrote: Tue Jul 03, 2018 7:30 pm
jhfenton wrote: Tue Jul 03, 2018 6:10 pm (I wonder whether there was a regulatory quirk that led Vanguard to keep their first "active" ETF separate from its companion mutual fund. The ETF-as-a-share-class structure is quirky in itself and unique to/patented by Vanguard. I wonder if they ran into an incompatibility of some sort trying to combine the two structures.)
It's possible they were anticipating the new SEC rules that came out this week. I'm not sure if any of the changes made this arrangement better than the old one, but it seems plausible.
I was also wondering whether the new SEC rules this week would have simplified some of the exemptive relief Vanguard had to get to offer "active" ETFs. One aspect of the rules this week had to do with custom creation baskets, which I believe Vanguard is using with the new factor ETFs.

I believe Vanguard also uses custom creation baskets to some degree with many of its ordinary index ETFs. It allows a Total Stock Market ETF to have a manageable creation basket on a daily basis while still achieving full index replication. With the hybrid ETF-traditional index fund (TIF) structure they can also fill in portfolio gaps in the creation basket with fund flows on the TIF side of the ledger.

That's why I've been speculating as to the reason VFMFX and VFMF are separate funds. Could Vanguard not get the SEC to approve rolling all of that into one package? An "active" ETF pooled with an "active" mutual fund is something that hasn't existed before. I don't see why it couldn't work, but it would have been novel.
User avatar
nisiprius
Advisory Board
Posts: 52216
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by nisiprius »

I'm adding inception dates, or something close (picked off the Morningstar chart by hitting "max") in red:
vineviz wrote: Tue Jul 03, 2018 8:08 am
  • Hartford Multifactor US Equity ETF (ROUS) 2/25/2015
  • iShares Edge MSCI Multifactor USA ETF (LRGF) 4/27/2015
  • JPMorgan Diversified Return US Eq ETF (JPUS) 9/29/2015
  • Vanguard US Multifactor ETF (VFMF) 2/14/2018
  • Xtrackers Russell 1000 Cmprhsv Fac ETF (DEUS) 11/24/2015
If three years is enough time to justify an almost-recommendation of these as core holdings, in place of a non-factor-based core holding, then it is enough time to justify inspection of actual performance. Since four of these are about three years old, let's look at those four.

Let's say it's 2015 and a group of people decide to commit to investing in one of these four, but they can't agree on which, so they each pick one at random. The average results for the group will be the same as that of an equal-weighted, un-rebalanced* portfolio of 25% of each. That's portfolio 1, blue. Portfolio 2 is 100% plain old Vanguard Total Stock Market Index Fund.

https://www.portfoliovisualizer.com/bac ... 100]Source
Image

Pick one of the interpretations you like or choose your own:
  • It just didn't matter. You might as well have been in the Total Stock fund, so keep it simple.
  • It just didn't matter. There was no convincing evidence that it did any harm to be in the multifactor funds, so why not?
  • The cap-weighted total market index fund outperformed the multifactor funds, with less risk. And just imagine the life you could be leading with an extra 1.2% CAGR compounded over fifty years.
*Not rebalancing is appropriate here, but if "not rebalancing" just bothers you, or if you think you could get a rebalancing bonus by intentionally following a strategy of using all four and rebalancing, then use the PortfolioVisualizer control to choose a rebalancing regime. The results are not very different.
Last edited by nisiprius on Wed Jul 04, 2018 7:29 am, edited 5 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
jhfenton
Posts: 4754
Joined: Sat Feb 07, 2015 10:17 am
Location: Ohio

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by jhfenton »

nisiprius wrote: Wed Jul 04, 2018 6:50 am Pick one of the interpretations you like or choose your own:
  • It just didn't matter. You might as well have been in the simple Total Stock fund.
  • There was no convincing evidence that it did any harm to be in the multifactor funds, so why not use one of them?
  • The cap-weighted total market index fund outperformed the multifactor funds, with less risk.
You forgot the obvious interpretation:
  • Those finds were all obviously flawed in some way, so of course they didn’t perform well. This new multifactor fund X is much better designed. :P
And, yes, I am being both serious and self-mocking.
User avatar
nisiprius
Advisory Board
Posts: 52216
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by nisiprius »

jhfenton wrote: Wed Jul 04, 2018 7:20 am
nisiprius wrote: Wed Jul 04, 2018 6:50 am Pick one of the interpretations you like or choose your own:
  • It just didn't matter. You might as well have been in the simple Total Stock fund.
  • There was no convincing evidence that it did any harm to be in the multifactor funds, so why not use one of them?
  • The cap-weighted total market index fund outperformed the multifactor funds, with less risk.
You forgot the obvious interpretation:
  • Those finds were all obviously flawed in some way, so of course they didn’t perform well. This new multifactor fund X is much better designed. :P
And, yes, I am being both serious and self-mocking.
How about "It's just what you'd expect during anomalous market conditions like those of the last three years?"
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
Topic Author
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by vineviz »

nisiprius wrote: Wed Jul 04, 2018 6:50 am If three years is enough time to justify an almost-recommendation of these as core holdings in place of a non-factor-based core holding, then it is enough time to justify inspection of actual performance. Since four of these are about three years old, let's look at those four.
I think this is a fair observation. None of these funds have a very long track record, and I do think that people should take that into account.

To some degree, the list was designed to produce a result like the one you got: these particular funds were selected in part because they have (for their short lives) tracked the Vanguard Total Stock Market Index fund relatively closely (r-squared greater than .85). That means they are unlikely to ever dramatically lead or lag the index fund.

That said, while these funds have a short track record, we do have much longer evidence for the underlying factors that the funds are using. I think it's reasonable to expect that the multifactor funds will outperform the core index fund on an absolute basis over the investment horizon for most investors. But I also think it is entirely reasonable for someone to want a longer track record before deciding to go this route.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Park
Posts: 738
Joined: Sat Nov 06, 2010 4:56 pm

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by Park »

Nisiprius is comparing multifactor ETFs to a total market cap index ETF. How about comparing multifactor ETFs to value ETFs? Is there any research on that topic?

An individual factor can underperform for many years. So there should be less tracking error with a multifactor ETF.

OTOH, multifactor ETFs commonly include momentum. Momentum is not easy to make money on . IIRC, Rob Arnott has looked at momentum funds, and find that they don't do well.

Also, the greatest weakness of market cap indexing, IMO, is its susceptibility to bubbles. The more a stock is overpriced, the more you own it. Quantitative value investing made the 2000 bear market much more palatable than market cap indexing. How would multifactor investing have done?

My impression is that DFA funds are multifactor funds. However, my impression is that they tilt to value first, and then apply tilts to other factors, with the value tilt dominating.

Once you tilt to value along with a momentum screen, how much do the other factors add?
User avatar
whodidntante
Posts: 13114
Joined: Thu Jan 21, 2016 10:11 pm
Location: outside the echo chamber

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by whodidntante »

Park wrote: Wed Jul 04, 2018 8:14 am Nisiprius is comparing multifactor ETFs to a total market cap index ETF. How about comparing multifactor ETFs to value ETFs? Is there any research on that topic?

An individual factor can underperform for many years. So there should be less tracking error with a multifactor ETF.

OTOH, multifactor ETFs commonly include momentum. Momentum is not easy to make money on . IIRC, Rob Arnott has looked at momentum funds, and find that they don't do well.

Also, the greatest weakness of market cap indexing, IMO, is its susceptibility to bubbles. The more a stock is overpriced, the more you own it. Quantitative value investing made the 2000 bear market much more palatable than market cap indexing. How would multifactor investing have done?

My impression is that DFA funds are multifactor funds. However, my impression is that they tilt to value first, and then apply tilts to other factors, with the value tilt dominating.

Once you tilt to value along with a momentum screen, how much do the other factors add?
I agree that DFA funds are multifactor and have posted that before. This new breed of multifactor funds adds additional factor diversification and gives more weight to momentum than DFA. It's not the track record of the funds that makes them worth considering, it is the evidence for the factor premia, acceptance of the cyclical nature of factors, low costs, and confidence in the implementation. I like the way the MSCI factor indexes are built, for example, and I also have confidence that Vanguard will get it right.

If an investor only dedicates a small portion of their portfolio to factors, then buying DFA funds or their advisor free doppelgangers is likely a good choice, because you are buying more extreme exposure in your limited space. I admit this conclusion is due to reasoning, not research. Paul Merriman goes halfsies between value and core, for example, and he frequently praises DFA. I'm not saying that a half allocation to value is a small allocation, but there is a reason he holds back from 100% value.
garlandwhizzer
Posts: 3565
Joined: Fri Aug 06, 2010 3:42 pm

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by garlandwhizzer »

The Morningstar article points out the theoretical advantages of a multi-factor approach and mentions its 2 favorite ETFs which combine a good approach with low costs, GSLC and LRGF. According to Morningstar graphs, neither of these ETFs has departed significantly during their approximately 3 year lifespans from either VTI (TSM) or VOO (S&P 500). 3 years is too short a lifespan to make solid decisions. That works both ways: too short to say the multi-factor strategy demonstrably works in the real world, also too short to say it fails to do so. Whether you choose a multi-factor approach or not depends at present more on faith in the reliability of backtesting research than on demonstrated alpha in real funds. So far, these two ETFs perform almost identically to beta and bear in mind these two are the favorites of Christine Benz and Alex Bryan of Morningstar. IMO it has not been settled in the real investing world whether multi-factor will produce the magical combo of increased returns with less risk that in theory it promises. Beta may not be "dumb" after all.

Garland Whizzer
User avatar
Taylor Larimore
Posts: 32842
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

Advisor fees included ?

Post by Taylor Larimore »

nisiprius:

In your chart showing the Total Market Index Fund outperforming the average of four multi-factor funds, did you include the advisor's fee necessary for the multi-factor funds?

Thank you and best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
User avatar
jhfenton
Posts: 4754
Joined: Sat Feb 07, 2015 10:17 am
Location: Ohio

Re: Advisor fees included ?

Post by jhfenton »

Taylor Larimore wrote: Wed Jul 04, 2018 3:44 pm nisiprius:

In your chart showing the Total Market Index Fund outperforming the average of four multi-factor funds, did you include the advisor's fee necessary for the multi-factor funds?

Thank you and best wishes.
Taylor
Those were all ETFs, so no advisor fees would be necessary.

In my case, I own Vanguard's VFMFX, so there are also no advisor fees necessary.

The DFA funds referenced earlier would require an advisor fee or their presence in your 401(k). (I have DFA Small Cap I / DFSTX in my 401(k), but no others.)
User avatar
Taylor Larimore
Posts: 32842
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by Taylor Larimore »

Those were all ETFs, so no advisor fees would be necessary.
jhfenton:

Are you saying that anyone can purchase DFA ETFs without an advisor?

Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
User avatar
jhfenton
Posts: 4754
Joined: Sat Feb 07, 2015 10:17 am
Location: Ohio

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by jhfenton »

Taylor Larimore wrote: Wed Jul 04, 2018 7:00 pm
Those were all ETFs, so no advisor fees would be necessary.
jhfenton:

Are you saying that anyone can purchase DFA ETFs without an advisor?

Taylor
DFA doesn't directly offer ETFs. They manage a lineup of ETFs for John Hancock, but I don't find them particularly compelling.

ETFs, by nature of being traded on stock exchanges, cannot be restricted to advisors like traditional mutual funds. Vanguard is targeting advisors with its new factor ETFs too, but they are available for anyone to purchase.

None of the multifactor ETFs that nisiprius compared, though, were DFA-managed ETFs.
AlohaJoe
Posts: 6609
Joined: Mon Nov 26, 2007 1:00 pm
Location: Saigon, Vietnam

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by AlohaJoe »

jhfenton wrote: Tue Jul 03, 2018 6:10 pm I would rate it as a strong possibility that VFMFX will not be tax-friendly.
I'm less sure. All of these factor funds are more tax efficient than Vanguard's S&P 500 fund[1][2]
  • iShares Edge MSCI Multifactor USA Small-Cap ETF
  • iShares Edge MSCI USA Momentum Factor ETF
  • Goldman Sachs ActiveBeta U.S. Large Cap Equity ETF
  • iShares Edge MSCI Multifactor USA ETF
So I don't see any reason for Vanguard to do a worse job, especially since the main reason they went active (instead of passive) was to lower costs.

[1]: not a comprehensive list, I didn't check Xtracker, JPMorgan, Hartford or several others
[2]: https://docs.google.com/spreadsheets/d/ ... =153756706 I find that using triceratop's Tax Efficiency 2017 spreadsheet is better than making guesses :wink:
User avatar
jhfenton
Posts: 4754
Joined: Sat Feb 07, 2015 10:17 am
Location: Ohio

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by jhfenton »

AlohaJoe wrote: Wed Jul 04, 2018 8:37 pm
jhfenton wrote: Tue Jul 03, 2018 6:10 pm I would rate it as a strong possibility that VFMFX will not be tax-friendly.
I'm less sure. All of these factor funds are more tax efficient than Vanguard's S&P 500 fund[1][2]
  • iShares Edge MSCI Multifactor USA Small-Cap ETF
  • iShares Edge MSCI USA Momentum Factor ETF
  • Goldman Sachs ActiveBeta U.S. Large Cap Equity ETF
  • iShares Edge MSCI Multifactor USA ETF
So I don't see any reason for Vanguard to do a worse job, especially since the main reason they went active (instead of passive) was to lower costs.

[1]: not a comprehensive list, I didn't check Xtracker, JPMorgan, Hartford or several others
[2]: https://docs.google.com/spreadsheets/d/ ... =153756706 I find that using triceratop's Tax Efficiency 2017 spreadsheet is better than making guesses :wink:
Those are all ETFs. I expect VFMF will have few if any--probably zero--capital gains distributions. But VFMFX is a mutual fund without an ETF share class. It has no creation-destruction mechanism to offload appreciated securities.
User avatar
Taylor Larimore
Posts: 32842
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by Taylor Larimore »

All of these factor funds are more tax efficient than Vanguard's S&P 500 fund[1][2]
iShares Edge MSCI Multifactor USA Small-Cap ETF
iShares Edge MSCI USA Momentum Factor ETF
Goldman Sachs ActiveBeta U.S. Large Cap Equity ETF
iShares Edge MSCI Multifactor USA ETF
AlohaJoe:

I attempted to look-up each of the above factor funds on Morningstar. None were listed?

What is your source for the above statement?

Thank you and best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
User avatar
jhfenton
Posts: 4754
Joined: Sat Feb 07, 2015 10:17 am
Location: Ohio

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by jhfenton »

Taylor Larimore wrote: Wed Jul 04, 2018 10:17 pm AlohaJoe:

I attempted to look-up each of the above factor funds on Morningstar. None were listed?

What is your source for the above statement?

Thank you and best wishes.
Taylor
SMLF/iShares Edge MSCI Multifactor USA Small-Cap ETF
MTUM/iShares Edge MSCI USA Momentum Factor ETF
GSLC/Goldman Sachs ActiveBeta U.S. Large Cap Equity ETF
LRGF/iShares Edge MSCI Multifactor USA ETF

All are listed in M*. SMLF is hard to find on M* because they butcher the name with abbreviations.

Just looking at M*'s Tax Cost Ratio, SMLF, MTUM, and LRGF were more tax-efficient over the trailing 3 years than VOO (Vanguard 500). GSLC did not have a 3-year track record, but was slightly less-tax efficient than VOO over the trailing 1-year. (M*'s Tax Cost Ratio is generally a pretty bad measure of tax efficiency.)

I won't be buying any of them, but they would all be reasonable holdings in a taxable account. I expect the same will be true for Vanguard's VFMF. I am not confident that VFMFX will be OK in a taxable account.
User avatar
CULater
Posts: 2832
Joined: Sun Nov 13, 2016 9:59 am
Location: Hic sunt dracones

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by CULater »

I heard there's a new fund about to be listed. They take every known factor and mix 'em all up together. They're going to call it the Total Market Fund.
On the internet, nobody knows you're a dog.
User avatar
whodidntante
Posts: 13114
Joined: Thu Jan 21, 2016 10:11 pm
Location: outside the echo chamber

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by whodidntante »

CULater wrote: Thu Jul 05, 2018 6:21 pm I heard there's a new fund about to be listed. They take every known factor and mix 'em all up together. They're going to call it the Total Market Fund.
There is a lot of confusion in this post.
User avatar
triceratop
Posts: 5838
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by triceratop »

CULater wrote: Thu Jul 05, 2018 6:21 pm I heard there's a new fund about to be listed. They take every known factor and mix 'em all up together. They're going to call it the Total Market Fund.
If I buy two products which are short the market and long the market, respectively, I make money no matter what the market does, right?
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."
User avatar
CULater
Posts: 2832
Joined: Sun Nov 13, 2016 9:59 am
Location: Hic sunt dracones

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by CULater »

When you think about the multifactor idea, it sort of stretches credibility a bit doesn't it? The idea is that the returns from different pieces of the market (factors) are uncorrelated or negatively correlated, so you can "diversify" your equity holdings by owning each of these little pieces; ie. when value is down, MOM is up, etc. But, stocks is stocks ain't they? They are all part of the same pack and they tend to roam around together especially during the good times and the bad times. Just seems like when I assemble all these pieces, in the aggregate they should pretty much act like the market acts. I know, I know, we have backtesting fishing that shows the little pieces are different, but....
On the internet, nobody knows you're a dog.
User avatar
jhfenton
Posts: 4754
Joined: Sat Feb 07, 2015 10:17 am
Location: Ohio

Re: Morningstar: What to Consider When Picking Multifactor ETFs

Post by jhfenton »

CULater wrote: Thu Jul 05, 2018 6:39 pm When you think about the multifactor idea, it sort of stretches credibility a bit doesn't it? The idea is that the returns from different pieces of the market (factors) are uncorrelated or negatively correlated, so you can "diversify" your equity holdings by owning each of these little pieces; ie. when value is down, MOM is up, etc. But, stocks is stocks ain't they? They are all part of the same pack and they tend to roam around together especially during the good times and the bad times. Just seems like when I assemble all these pieces, in the aggregate they should pretty much act like the market acts. I know, I know, we have backtesting fishing that shows the little pieces are different, but....
You can think of a multifactor fund constructed like Vanguard's as a "market minus" fund: minus junk, minus expensive stocks, minus sinking stocks. The fund is going to be filled with reasonably-valued stocks, with solid balance sheets, and reasonable price momentum.

With VFMFX and VFMF, Vanguard is being fairly strict and only keeping the top 20% or so of their total market universe, and they're weighting that 20% by how high they score rather than market cap. The result is about 1/3 large, 1/3 mid, 1/3 small.

To be honest, it's ultimately not that different from what many active managers have done over the years, it's just quantitative and cheap (18 bp at launch) compared to traditional active funds.

(The other way to construct a multifactor fund is to run multiple mini-funds--value, quality, momentum, etc.--inside the fund. Goldman Sachs does this with their "multifactor" fund.)
CULater wrote: Thu Jul 05, 2018 6:21 pm I heard there's a new fund about to be listed. They take every known factor and mix 'em all up together. They're going to call it the Total Market Fund.
I'm guessing that this is only funny if you don't understand what factors are.
Post Reply