VFMF (Vanguard Multifactor) == Closet Value Index Fund

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Topic Author
hdas
Posts: 874
Joined: Thu Jun 11, 2015 8:24 am

VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by hdas » Tue Jun 11, 2019 12:00 pm

I've been following the performance (I don't own the fund) of VFMF (Vanguard Multifactor). And trying to figure out why the terrible performance, given that Momentum and Quality have been doing ok.....Mr. Gray from Alpha Architect seems to be commenting about the same issue in twitter.

Image

Image

Cheers :greedy
"whenever there is a randomized way of doing something, then there is a nonrandomized way that delivers better performance but requires more thought" ET Jaynes

User avatar
whodidntante
Posts: 5455
Joined: Thu Jan 21, 2016 11:11 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index

Post by whodidntante » Tue Jun 11, 2019 12:04 pm

I personally want a strong value loading in a multi-factor fund, because value has had the second largest factor premium, followed by market. But as you noted, this can cause significant deviation from market performance. I guess that's also the point. If it had been 10 points ahead I don't think anyone would complain. :happy

User avatar
vineviz
Posts: 4012
Joined: Tue May 15, 2018 1:55 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by vineviz » Tue Jun 11, 2019 12:12 pm

hdas wrote:
Tue Jun 11, 2019 12:00 pm
I've been following the performance (I don't own the fund) of VFMF (Vanguard Multifactor). And trying to figure out why the terrible performance, given that Momentum and Quality have been doing ok.
I wouldn't necessarily take that as a given.

Since the inception of VFMF, the monthly factor premiums (using Alpha Architects' own data) have been negative for size, value, momentum, and quality. At least they were through the end of April.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
Topic Author
hdas
Posts: 874
Joined: Thu Jun 11, 2015 8:24 am

Re: VFMF (Vanguard Multifactor) == Closet Value Index

Post by hdas » Tue Jun 11, 2019 12:14 pm

whodidntante wrote:
Tue Jun 11, 2019 12:04 pm
I personally want a strong value loading in a multi-factor fund, because value has had the second largest factor premium, followed by market. But as you noted, this can cause significant deviation from market performance. I guess that's also the point. If it had been 10 points ahead I don't think anyone would complain. :happy
I'm not questioning the deviation from market performance, but instead the overwhelming effect of the value factor performance on the supposedly "multifactor" fund. Momentum has been on a tear, quality doing ok.....I guess I don't see the "balancing" effect of the multifactor approach in this case. :greedy
"whenever there is a randomized way of doing something, then there is a nonrandomized way that delivers better performance but requires more thought" ET Jaynes

pdavi21
Posts: 1057
Joined: Sat Jan 30, 2016 4:04 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by pdavi21 » Tue Jun 11, 2019 12:24 pm

There are risks that tend to be higher with a tilted fund:
1. Index Choice
2. Factor Methodology
3. Expenses

There could, theoretically be two multi-factor funds that have vastly differing performance over longer term intervals because one decided to hold a few high climbing stocks while one decided not to.

I recently saw someone quote a joke, "Fruit is good, cake is good. So fruitcake must be good." I think this can also be an interesting analogy for multifactor funds. If Momentum, quality, value, (and size?) or whatever factors the fund is targeting perform a certain way separately, it is not 100% guaranteed that combing those factors will yield an aggregate performance that is comparable to the weighted average of each factor's performance. It could be that the best momentum stocks have a high P/E and are not value stocks. Or that only large momentum stocks do better, while small momentum stocks perform poorly, for example.

That being said, over this tiny term interval, probably the factors targeted did poorly in aggregate.
"We spend a great deal of time studying history, which, let's face it, is mostly the history of stupidity." -Stephen Hawking

User avatar
Topic Author
hdas
Posts: 874
Joined: Thu Jun 11, 2015 8:24 am

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by hdas » Tue Jun 11, 2019 12:31 pm

vineviz wrote:
Tue Jun 11, 2019 12:12 pm
hdas wrote:
Tue Jun 11, 2019 12:00 pm
I've been following the performance (I don't own the fund) of VFMF (Vanguard Multifactor). And trying to figure out why the terrible performance, given that Momentum and Quality have been doing ok.
I wouldn't necessarily take that as a given.

Since the inception of VFMF, the monthly factor premiums (using Alpha Architects' own data) have been negative for size, value, momentum, and quality. At least they were through the end of April.
Ok, fair enough, that doesn't explain why the fund is less balanced than expected, or why is mostly value. And I think the question is still open if this specific multifactor implementation is good.

Code: Select all

Portfolio performance statistics
Note: The time period was automatically adjusted based on the available data (Mar 2018 - May 2019) for the selected asset: Vanguard US Multifactor ETF (VFMF)

        CAGR	Stdev	BestY	WorstY	Max. DD
VFMF	-4.45% 	17.86%	5.82%	-10.73%	-18.66% 	
VFVA	-7.72% 	22.73%	5.93%	-14.62%	-20.67% 	
QUAL	3.45% 	16.51%	12.31%	-7.11%	-14.61% 	
QMOM   -0.79% 	23.91%	17.48%	-15.72%	-25.76% 	
VTI      2.97% 	16.57%	10.83%	-6.41%	-14.20%
"whenever there is a randomized way of doing something, then there is a nonrandomized way that delivers better performance but requires more thought" ET Jaynes

pdavi21
Posts: 1057
Joined: Sat Jan 30, 2016 4:04 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by pdavi21 » Tue Jun 11, 2019 1:59 pm

hdas wrote:
Tue Jun 11, 2019 12:31 pm
vineviz wrote:
Tue Jun 11, 2019 12:12 pm
hdas wrote:
Tue Jun 11, 2019 12:00 pm
I've been following the performance (I don't own the fund) of VFMF (Vanguard Multifactor). And trying to figure out why the terrible performance, given that Momentum and Quality have been doing ok.
I wouldn't necessarily take that as a given.

Since the inception of VFMF, the monthly factor premiums (using Alpha Architects' own data) have been negative for size, value, momentum, and quality. At least they were through the end of April.
Ok, fair enough, that doesn't explain why the fund is less balanced than expected, or why is mostly value. And I think the question is still open if this specific multifactor implementation is good.

Code: Select all

Portfolio performance statistics
Note: The time period was automatically adjusted based on the available data (Mar 2018 - May 2019) for the selected asset: Vanguard US Multifactor ETF (VFMF)

        CAGR	Stdev	BestY	WorstY	Max. DD
VFMF	-4.45% 	17.86%	5.82%	-10.73%	-18.66% 	
VFVA	-7.72% 	22.73%	5.93%	-14.62%	-20.67% 	
QUAL	3.45% 	16.51%	12.31%	-7.11%	-14.61% 	
QMOM   -0.79% 	23.91%	17.48%	-15.72%	-25.76% 	
VTI      2.97% 	16.57%	10.83%	-6.41%	-14.20%
I think you are making a mistake in assuming that combining factors will combine performance. A portfolio that is part VFVA/QUAL/QMOM/VTI will contain vastly different holdings than one that applies all four factors.
"We spend a great deal of time studying history, which, let's face it, is mostly the history of stupidity." -Stephen Hawking

User avatar
nisiprius
Advisory Board
Posts: 38263
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: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by nisiprius » Wed Jun 12, 2019 7:02 am

The great thing about multifactor funds from the point of view of investment firms is that they have solved the problem of benchmarking. That is, they have solved the awkward problem of being able to compare funds to a benchmark, and to each other. No two of them are alike. The actual strategies themselves are often described as "rules-based," but the actual rules are not disclosed in detail.

I don't think there is even a way for an ordinary retail investor, relying on regulatory documents, to verify whether or not one of these funds is even following the rules it says it is following. Hey, OK, they probably are, but still.

For a traditional index fund, even one that is factor-based, the objective is simple and it is easy to tell how well the fund is achieving it. For example, the Vanguard Value Index Fund (according to the "strategy and policy" link from the "portfolio & management" tab at VIVAX) is simple:
The fund employs an indexing investment approach designed to track the performance of the CRSP US Large Cap Value Index, a broadly diversified index predominantly made up of value stocks of large U.S. companies. The fund attempts to replicate the target index by investing all, or substantially all, of its assets in the stocks that make up the index, holding each stock in approximately the same proportion as its weighting in the index.
How well has it done it? This well:
Image

How did it do compared with a competitor's large-cap value ETF--SPYV, iShares S&P 500 Value? Here's the comparison, and it's reasonable apples-to-apples:
Source

Image

Once you get into the world of multifactor funds, that all goes out the window. No two are alike. No two can be properly compared with each other. They don't follow any published index. They all seem to be described as "rules-based," but you don't get to see a detailed statement of the rules; the holdings of the fund may be transparent, but the fund's strategy, what it is trying to do--exactly--is not. Suddenly, the objectives turn to mush; for VFMF,
The fund invests primarily in U.S. common stocks with the potential to generate higher returns relative to the broad U.S. equity market by investing in stocks with relatively strong recent performance, strong fundamentals, and low prices relative to fundamentals as determined by the advisor. The portfolio will include a diverse mix of companies representing many different market sectors and industry groups. The advisor uses a quantitative model to evaluate all of the securities in an investment universe comprised of U.S. large, mid, and small capitalization stocks and to construct a U.S. equity portfolio that seeks to achieve exposure to multiple factors subject to a set of reasonable constraints designed to foster portfolio diversification, liquidity, and lower volatility.
How the heck can anybody tell whether or not, or how well, the advisor is doing what it claims to be doing? Apart from the claim that a "quantitative" model is used, or that it is "rules-based," it might as well be an active fund in terms of knowing what it is doing. Some fund companies actually use the word "proprietary" to describe their strategies, although Vanguard doesn't.

And the companies' performance reports for these funds use so-called "benchmarks" that are just plain stupid:

Image

The Russell 3000? That's just a total market index (and not one that Vanguard usually uses; why here?) It is doesn't make any more sense to compare VFMF to the Russell 3000 than to compare the Vanguard Value Index Fund to the Russell 3000. This isn't supposed to be a total market index ETF, why is Vanguard comparing it to one?

None of this seems to bother the factor mavens much, but it bothers me. The new breed of factor funds are pigs in a poke, and you are just as reliant on your personal trust in the managers' wisdom as you would be with an actively managed fund. People enjoy doing Kreminology in an attempt to reverse-engineer just what the managers are doing--just the way enthusiasts for actively managed funds do. One thing they do have going for them is that the costs are relatively low, and that if the strategy is rules-based and if they truly follow the rules then you shouldn't get the phenomena of a manager suddenly and capriciously making major changes in strategy.

Except when you do, of course--as when the "AQR Risk Parity" fund becomes the "AQR Multi-Asset Fund.
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.

azanon
Posts: 2360
Joined: Mon Nov 07, 2011 10:34 am
Location: Little Rock, AR
Contact:

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by azanon » Wed Jun 12, 2019 8:33 am

TBH, I'm surprised it sells at all. I get a value investor. I also get a momentum investor. I don't get both of those at the same time though given that they generally prefer polar opposite types of stocks.

columbia
Posts: 1557
Joined: Tue Aug 27, 2013 5:30 am

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by columbia » Wed Jun 12, 2019 8:49 am

“Rules based” appears to be the foundation of Wellington and Wellesley. But an investor will never really know what the rules are. It’s certainly better than turning those funds over to the new hot hand, who will eventually run out of luck.

User avatar
nisiprius
Advisory Board
Posts: 38263
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: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by nisiprius » Wed Jun 12, 2019 8:55 am

azanon wrote:
Wed Jun 12, 2019 8:33 am
TBH, I'm surprised it sells at all. I get a value investor. I also get a momentum investor. I don't get both of those at the same time though given that they generally prefer polar opposite types of stocks.
In "A Review of Equity Factor Premia," Jared Kizer wrote:
Some factor premia have negative correlation with each other, namely value and momentum. Practically, it is impossible, therefore, to own a stock portfolio that is both deeply tilted toward value and deeply tilted toward momentum. At best, stock portfolios can be moderately tilted toward both. More generally, it is not feasible to build portfolios that capture significant amounts of multiple factor premia.
I think, though, that the "multifactor" funds try to get around this by market timing factors.

Arnott and Research Associates openly advocate "timing:" factors: Timing smart beta strategies? Of course! Buy low, sell high.
A contrarian timing approach—emphasizing factors or strategies trading cheap relative to their own historical norms, and deemphasizing the more expensive factors or strategies—can improve performance, but should be used in moderation to avoid increasing portfolio risk from a loss of diversification.
Others would probably object to a characterization of "market timing," but use something that superficially sounds like it. "Time series momentum?" Indeed. And what is traditional market timing but "time series asset allocation?" And don't traditional market timers claim that what they are doing is "rules based," too?
Last edited by nisiprius on Wed Jun 12, 2019 9:06 am, edited 4 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.

pdavi21
Posts: 1057
Joined: Sat Jan 30, 2016 4:04 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by pdavi21 » Wed Jun 12, 2019 8:59 am

Also, you should be multiplying the total returns, not weighted averaging the percentages (assuming performance is 100% factor driven-which it is not).

It's because if momentum returned 3%, and value lost 8%, you'd expect momentum stocks with a value filter to return 1.03 x 0.92-1 and not 3% x 50% - 8% x 50%.

EDIT: You also have to assume that factor performance is independent from other factors or your sample pot...which is also false.

EDIT: Also VFMF has a huge size tilt. Where is your size return?
"We spend a great deal of time studying history, which, let's face it, is mostly the history of stupidity." -Stephen Hawking

User avatar
nisiprius
Advisory Board
Posts: 38263
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: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by nisiprius » Wed Jun 12, 2019 9:21 am

It's a tiny period of time, but interesting. Morningstar classifies VFMF as "mid-cap blend." Therefore, let's not blame this on the terrible unfairness of large-cap growth having broken the factor rules by doing so well for so long.

Here's a comparison of VFMF (blue) with Morningstar's category average for mid-cap blend funds (orange), and Vanguard's own single-factor ETF, VO, the Vanguard Mid-Cap Index ETF.

Source

Image
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
nisiprius
Advisory Board
Posts: 38263
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: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by nisiprius » Wed Jun 12, 2019 9:28 am

azanon wrote:
Wed Jun 12, 2019 8:33 am
...TBH, I'm surprised it sells at all...
Well, it hasn't sold a lot. $83 million-with-an-m since 2/15/2018.

What's a fair comparison? VO, Vanguard Mid-Cap Index ETF, has $25 billion, but that's after 15 years of operation. There must be a way to figure out how much has been invested in VO since 2/15/2018, but I don't know how to do it. Can anybody help?
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
vineviz
Posts: 4012
Joined: Tue May 15, 2018 1:55 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by vineviz » Wed Jun 12, 2019 9:39 am

nisiprius wrote:
Wed Jun 12, 2019 9:28 am
There must be a way to figure out how much has been invested in VO since 2/15/2018, but I don't know how to do it. Can anybody help?
Vanguard Mid-Cap ETF $1,818.59 million

That’s about 22x VFMF.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

azanon
Posts: 2360
Joined: Mon Nov 07, 2011 10:34 am
Location: Little Rock, AR
Contact:

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by azanon » Wed Jun 12, 2019 10:09 am

nisiprius wrote:
Wed Jun 12, 2019 9:28 am
azanon wrote:
Wed Jun 12, 2019 8:33 am
...TBH, I'm surprised it sells at all...
Well, it hasn't sold a lot. $83 million-with-an-m since 2/15/2018.

What's a fair comparison? VO, Vanguard Mid-Cap Index ETF, has $25 billion, but that's after 15 years of operation. There must be a way to figure out how much has been invested in VO since 2/15/2018, but I don't know how to do it. Can anybody help?
And the one I own has even less (VFVA - Vanguard Value Factor) - 6x million or thereabouts.

Granted, two things quickly come to mind as to why neither are selling great yet: 1. Growth outperformance since launch and 2. (this one by far the more important) Vanguard barely promoting their existence, and not suggesting their use by individual investors. Compare that to the Global Wellesley/wellington products that were advertised front-and-center on their homepage last year just prior to launch.

Brogleski
Posts: 40
Joined: Fri Jul 12, 2013 4:23 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by Brogleski » Wed Jun 12, 2019 10:57 am

nisiprius wrote:
Wed Jun 12, 2019 9:28 am
azanon wrote:
Wed Jun 12, 2019 8:33 am
...TBH, I'm surprised it sells at all...
Well, it hasn't sold a lot. $83 million-with-an-m since 2/15/2018.

What's a fair comparison? VO, Vanguard Mid-Cap Index ETF, has $25 billion, but that's after 15 years of operation. There must be a way to figure out how much has been invested in VO since 2/15/2018, but I don't know how to do it. Can anybody help?
https://www.etf.com/etfanalytics/etf-fund-flows-tool
The enemy of a good plan is the dream of a perfect plan.

User avatar
nisiprius
Advisory Board
Posts: 38263
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: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by nisiprius » Wed Jun 12, 2019 11:04 am

Brogleski wrote:
Wed Jun 12, 2019 10:57 am
nisiprius wrote:
Wed Jun 12, 2019 9:28 am
azanon wrote:
Wed Jun 12, 2019 8:33 am
...TBH, I'm surprised it sells at all...
Well, it hasn't sold a lot. $83 million-with-an-m since 2/15/2018.

What's a fair comparison? VO, Vanguard Mid-Cap Index ETF, has $25 billion, but that's after 15 years of operation. There must be a way to figure out how much has been invested in VO since 2/15/2018, but I don't know how to do it. Can anybody help?
https://www.etf.com/etfanalytics/etf-fund-flows-tool
Thank you! Vineviz already posted the actual answer, but now I know a way to do it for myself.

Image
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.

HEDGEFUNDIE
Posts: 2566
Joined: Sun Oct 22, 2017 2:06 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by HEDGEFUNDIE » Wed Jun 12, 2019 11:10 am

nisiprius wrote:
Wed Jun 12, 2019 7:02 am

The Russell 3000? That's just a total market index (and not one that Vanguard usually uses; why here?) It is doesn't make any more sense to compare VFMF to the Russell 3000 than to compare the Vanguard Value Index Fund to the Russell 3000. This isn't supposed to be a total market index ETF, why is Vanguard comparing it to one?
Because if you are a factor investor, VFMF is supposed to be your substitute for a total market fund.

cheezit
Posts: 196
Joined: Sat Jul 14, 2018 7:28 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by cheezit » Wed Jun 12, 2019 11:11 am

hdas wrote:
Tue Jun 11, 2019 12:31 pm
vineviz wrote:
Tue Jun 11, 2019 12:12 pm
hdas wrote:
Tue Jun 11, 2019 12:00 pm
I've been following the performance (I don't own the fund) of VFMF (Vanguard Multifactor). And trying to figure out why the terrible performance, given that Momentum and Quality have been doing ok.
I wouldn't necessarily take that as a given.

Since the inception of VFMF, the monthly factor premiums (using Alpha Architects' own data) have been negative for size, value, momentum, and quality. At least they were through the end of April.
Ok, fair enough, that doesn't explain why the fund is less balanced than expected, or why is mostly value. And I think the question is still open if this specific multifactor implementation is good.

Code: Select all

Portfolio performance statistics
Note: The time period was automatically adjusted based on the available data (Mar 2018 - May 2019) for the selected asset: Vanguard US Multifactor ETF (VFMF)

        CAGR	Stdev	BestY	WorstY	Max. DD
VFMF	-4.45% 	17.86%	5.82%	-10.73%	-18.66% 	
VFVA	-7.72% 	22.73%	5.93%	-14.62%	-20.67% 	
QUAL	3.45% 	16.51%	12.31%	-7.11%	-14.61% 	
QMOM   -0.79% 	23.91%	17.48%	-15.72%	-25.76% 	
VTI      2.97% 	16.57%	10.83%	-6.41%	-14.20%
VFQY and VFMO make more sense than QUAL and QMOM as points of comparison, given that they're run by Vanguard and use the same ranking-within-cap-buckets methodology as VFMF and VFVA. Re-reading the VFMF-and-friends whitepaper will probably make it more obvious why the weightings are what they are.

afan
Posts: 4167
Joined: Sun Jul 25, 2010 4:01 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by afan » Wed Jun 12, 2019 11:24 am

I don't tilt, so I have not followed the fund. That said, I don't recall Vanguard promising a particular, fixed, weighting of factors. Perhaps it did and I missed it.

If, like other funds that chase multiple factors, this fund can and does change its factor weighting over time, the only way to benchmark it would be against a comparable mix of factors, matched to the fund weightings. That would require the fund to tell you each time the weightings were changed. Since I don't think they do this, one is not going to be able to determine how much of its performance is due to the, varying, weights across time.

If it uses simple index like approaches to the individual positions used to achieve the desired factor loads, then one could say that all of the difference between the fund's performance and a mix of factors would be the extent to which the funds weights deviated from whatever mix you used as a benchmark.

If you believe in factors, the maybe VTI would be an appropriate benchmark, since it is supposed to be neutral to factors other than beta.

Or save money and drama by putting your money into the benchmark instead of chasing factors.
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama

Post Reply