Are Dimensional ETFs worth it?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Always passive
Posts: 942
Joined: Fri Apr 14, 2017 4:25 am
Location: Israel

Are Dimensional ETFs worth it?

Post by Always passive »

This forum has typically recommended total market ETFs, and more specifically Vanguard funds. For me, living overseas, that means VTI and VXUS. Now that Dimensional ETFs are available to all, are they worth it when compared to Vanguard's?
livesoft
Posts: 76884
Joined: Thu Mar 01, 2007 8:00 pm

Re: Are Dimensional ETFs worth it?

Post by livesoft »

I am sure that for some people Dimensional ETFs will be "worth it" to them and for some other people they will not be. I am trying to figure out how to answer your question better than that, but I cannot for the life of me give you a better answer.
Wiki This signature message sponsored by sscritic: Learn to fish.
tomsense76
Posts: 650
Joined: Wed Oct 14, 2020 1:52 am

Re: Are Dimensional ETFs worth it?

Post by tomsense76 »

I found nisi's post on this ( viewtopic.php?p=5608502#p5608502 ) instructive. My takeaway from that is the ETFs currently don't cover the things people would look to Dimensional for (like small cap value). By contrast Avantis, run by ex-DFA people, does supply ETFs that do target those factors. So if I were to hold ETFs that did that tilt (I do not currently and unclear if I would), I'd probably go with Avantis. There seem to be several people that already have (see recent post in this thread viewtopic.php?t=282533 and others).
"Anyone who claims to understand quantum theory is either lying or crazy" -- Richard Feynman
Topic Author
Always passive
Posts: 942
Joined: Fri Apr 14, 2017 4:25 am
Location: Israel

Re: Are Dimensional ETFs worth it?

Post by Always passive »

livesoft wrote: Thu Jun 24, 2021 1:09 pm I am sure that for some people Dimensional ETFs will be "worth it" to them and for some other people they will not be. I am trying to figure out how to answer your question better than that, but I cannot for the life of me give you a better answer.
What is it that make Dimensional funds, now some turned into ETFs so attractive that some investors are trying to get them by having financial advisors manage their money?
The literature points at the funds being developed using the most up to date academic research.
livesoft
Posts: 76884
Joined: Thu Mar 01, 2007 8:00 pm

Re: Are Dimensional ETFs worth it?

Post by livesoft »

Always passive wrote: Thu Jun 24, 2021 1:24 pmWhat is it that make Dimensional funds, now some turned into ETFs so attractive that some investors are trying to get them by having financial advisors manage their money?
I think it is marketing hype since the Dimensional secret sauce has been copied by others nowadays.

One might as well ask: Which is worth it, Toyota RAV4 or Honda CR-V?
Wiki This signature message sponsored by sscritic: Learn to fish.
absolute zero
Posts: 969
Joined: Thu Dec 29, 2016 4:59 pm

Re: Are Dimensional ETFs worth it?

Post by absolute zero »

Are they worth what? Question too vague to answer.
exodusNH
Posts: 411
Joined: Wed Jan 06, 2021 8:21 pm

Re: Are Dimensional ETFs worth it?

Post by exodusNH »

Always passive wrote: Thu Jun 24, 2021 1:01 pm This forum has typically recommended total market ETFs, and more specifically Vanguard funds. For me, living overseas, that means VTI and VXUS. Now that Dimensional ETFs are available to all, are they worth it when compared to Vanguard's?
My understanding is the ETFs they currently offer are not for the funds that people actually want access to.
UpperNwGuy
Posts: 5711
Joined: Sun Oct 08, 2017 7:16 pm

Re: Are Dimensional ETFs worth it?

Post by UpperNwGuy »

Always passive wrote: Thu Jun 24, 2021 1:24 pm
livesoft wrote: Thu Jun 24, 2021 1:09 pm I am sure that for some people Dimensional ETFs will be "worth it" to them and for some other people they will not be. I am trying to figure out how to answer your question better than that, but I cannot for the life of me give you a better answer.
What is it that make Dimensional funds, now some turned into ETFs so attractive that some investors are trying to get them by having financial advisors manage their money?
The literature points at the funds being developed using the most up to date academic research.
I've never bought into the claim that Dimensional has access to academic research that no other fund family has access to. I think it's mostly marketing talk.
bigskyguy
Posts: 320
Joined: Sat Jan 24, 2015 4:59 pm

Re: Are Dimensional ETFs worth it?

Post by bigskyguy »

Dimensional has historically been a fund company that adhere's to the Fama-French three factor model (small cap, value, market risk) of investing. That is a distinct (and presently somewhat out of favor) model. It is in contrast to the singular indexing model (the Bogle model), the five factor model (includes momentum), and I suppose an array of other models. I am sure that each has its attractions and its proponents.

As far as models go, the Dimensional ETFs are relatively low cost, and consistent in philosophy and implementation. The same can be said for simple index investing. Other approaches not so much (my sense).

Is it worth it? That is a judgement question, not a fact based question. That means the answer for you is likely to be different than the answer for me. Since none of us can predict the unpredictable (the future that is), there is in all honesty no objective answer to your question. There are infinite subjective answers.

My two cents worth.
absolute zero
Posts: 969
Joined: Thu Dec 29, 2016 4:59 pm

Re: Are Dimensional ETFs worth it?

Post by absolute zero »

bigskyguy wrote: Thu Jun 24, 2021 2:11 pm Dimensional has historically been a fund company that adhere's to the Fama-French three factor model (small cap, value, market risk) of investing. That is a distinct (and presently somewhat out of favor) model. It is in contrast to.....the five factor model (includes momentum), and I suppose an array of other models.
Not to nitpick, but these comments are somewhat inaccurate. The Fama-French five factor model does not include momentum, though other asset pricing models do. Regardless, DFA does utilize momentum in their funds as a screening tool. They also utilize additional factors beyond the original FF 3 factor model, e.g. they target the profitability factor in a number of their funds. For what it's worth, I do not invest in any DFA funds.
bigskyguy wrote: Thu Jun 24, 2021 2:11 pm It is in contrast to the singular indexing model (the Bogle model)
Did you make that up? I've never read the words "singular indexing model." John Bogle never created any asset pricing models i.e. models that attempt to explain the cross section of returns. His philosophy was to ignore the cross section.
absolute zero
Posts: 969
Joined: Thu Dec 29, 2016 4:59 pm

Re: Are Dimensional ETFs worth it?

Post by absolute zero »

UpperNwGuy wrote: Thu Jun 24, 2021 2:01 pm
Always passive wrote: Thu Jun 24, 2021 1:24 pm
livesoft wrote: Thu Jun 24, 2021 1:09 pm I am sure that for some people Dimensional ETFs will be "worth it" to them and for some other people they will not be. I am trying to figure out how to answer your question better than that, but I cannot for the life of me give you a better answer.
What is it that make Dimensional funds, now some turned into ETFs so attractive that some investors are trying to get them by having financial advisors manage their money?
The literature points at the funds being developed using the most up to date academic research.
I've never bought into the claim that Dimensional has access to academic research that no other fund family has access to. I think it's mostly marketing talk.
Have they made that claim before? If so, I've never seen it. Though I do agree that there are occasionally some forum members that think DFA has some sort of secret sauce. I don't really buy into the secret sauce idea though.
bigskyguy
Posts: 320
Joined: Sat Jan 24, 2015 4:59 pm

Re: Are Dimensional ETFs worth it?

Post by bigskyguy »

absolute zero wrote: Thu Jun 24, 2021 2:38 pm
bigskyguy wrote: Thu Jun 24, 2021 2:11 pm Dimensional has historically been a fund company that adhere's to the Fama-French three factor model (small cap, value, market risk) of investing. That is a distinct (and presently somewhat out of favor) model. It is in contrast to.....the five factor model (includes momentum), and I suppose an array of other models.
Not to nitpick, but these comments are somewhat inaccurate. The Fama-French five factor model does not include momentum, though other asset pricing models do. Regardless, DFA does utilize momentum in their funds as a screening tool. They also utilize additional factors beyond the original FF 3 factor model, e.g. they target the profitability factor in a number of their funds. For what it's worth, I do not invest in any DFA funds.
bigskyguy wrote: Thu Jun 24, 2021 2:11 pm It is in contrast to the singular indexing model (the Bogle model)
Did you make that up? I've never read the words "singular indexing model." John Bogle never created any asset pricing models i.e. models that attempt to explain the cross section of returns. His philosophy was to ignore the cross section.
Thanks for your comments. DFA was founded upon the concepts developed by Fama, French, Scholes, and Merton, which at their core are predicated on the three factor model, and has slowly evolved to incorporate additional factors, as you have noted. By incorporating these factors, they have departed from the Bogle approach of index investing, suggesting these various factors provide additional return in excess of additional risk. Their inherent assumption is that historical trends can predict future trends. Bogle, on the other hand, has a more agnostic approach, preferring to accept what the market as a whole provides. These are very distinctive approaches. The term I used above is an acknowledgement of John Bogles singular reliance on market returns for his preferred investment approach, nothing more.

Truth in advertising, we have historically invested with DFA, still do to a very limited extent (our HSAs), but have the vast majority of our equity portfolio in Vanguard Total World Stock Market ETF (VT). I find both approaches well founded, and have no clue which one will provide superior returns over time.
NiceUnparticularMan
Posts: 2526
Joined: Sat Mar 11, 2017 7:51 am

Re: Are Dimensional ETFs worth it?

Post by NiceUnparticularMan »

absolute zero wrote: Thu Jun 24, 2021 2:38 pm
bigskyguy wrote: Thu Jun 24, 2021 2:11 pm Dimensional has historically been a fund company that adhere's to the Fama-French three factor model (small cap, value, market risk) of investing. That is a distinct (and presently somewhat out of favor) model. It is in contrast to.....the five factor model (includes momentum), and I suppose an array of other models.
Not to nitpick, but these comments are somewhat inaccurate. The Fama-French five factor model does not include momentum, though other asset pricing models do. Regardless, DFA does utilize momentum in their funds as a screening tool. They also utilize additional factors beyond the original FF 3 factor model, e.g. they target the profitability factor in a number of their funds. For what it's worth, I do not invest in any DFA funds.
bigskyguy wrote: Thu Jun 24, 2021 2:11 pm It is in contrast to the singular indexing model (the Bogle model)
Did you make that up? I've never read the words "singular indexing model." John Bogle never created any asset pricing models i.e. models that attempt to explain the cross section of returns. His philosophy was to ignore the cross section.
I'm not sure I would call it "the Bogle model," but I do think what many would call a simple Boglehead approach is basically a CAPM approach. And I do think generally that approach more or less assumes only one type of risk, known in the literature as market risk or Beta. The plan is you get exposure to Beta with total market index stock funds, then modulate your risk with some fixed percentage of low-to-no-risk bonds, and that is then it for planning.

There might be some dancing around with which bonds to use, what counts as the total market, emergency funds or other savings not included in this portfolio, and so on. But I do feel that is basically a CAPM sort of understanding of portfolio theory.
NiceUnparticularMan
Posts: 2526
Joined: Sat Mar 11, 2017 7:51 am

Re: Are Dimensional ETFs worth it?

Post by NiceUnparticularMan »

bigskyguy wrote: Thu Jun 24, 2021 3:37 pm
absolute zero wrote: Thu Jun 24, 2021 2:38 pm
bigskyguy wrote: Thu Jun 24, 2021 2:11 pm Dimensional has historically been a fund company that adhere's to the Fama-French three factor model (small cap, value, market risk) of investing. That is a distinct (and presently somewhat out of favor) model. It is in contrast to.....the five factor model (includes momentum), and I suppose an array of other models.
Not to nitpick, but these comments are somewhat inaccurate. The Fama-French five factor model does not include momentum, though other asset pricing models do. Regardless, DFA does utilize momentum in their funds as a screening tool. They also utilize additional factors beyond the original FF 3 factor model, e.g. they target the profitability factor in a number of their funds. For what it's worth, I do not invest in any DFA funds.
bigskyguy wrote: Thu Jun 24, 2021 2:11 pm It is in contrast to the singular indexing model (the Bogle model)
Did you make that up? I've never read the words "singular indexing model." John Bogle never created any asset pricing models i.e. models that attempt to explain the cross section of returns. His philosophy was to ignore the cross section.
Thanks for your comments. DFA was founded upon the concepts developed by Fama, French, Scholes, and Merton, which at their core are predicated on the three factor model, and has slowly evolved to incorporate additional factors, as you have noted. By incorporating these factors, they have departed from the Bogle approach of index investing, suggesting these various factors provide additional return in excess of additional risk. Their inherent assumption is that historical trends can predict future trends. Bogle, on the other hand, has a more agnostic approach, preferring to accept what the market as a whole provides. These are very distinctive approaches. The term I used above is an acknowledgement of John Bogles singular reliance on market returns for his preferred investment approach, nothing more.

Truth in advertising, we have historically invested with DFA, still do to a very limited extent (our HSAs), but have the vast majority of our equity portfolio in Vanguard Total World Stock Market ETF (VT). I find both approaches well founded, and have no clue which one will provide superior returns over time.
So I think that is a bit inaccurate.

Anyone who believes, say, that a diversified portfolio of stocks is likely to have higher risk and higher returns than low-risk bonds in the long run is already doing something similar to a multi-factor model. Again, roughly speaking most such people are doing something broadly consistent with the CAPM model.

Multi-factor models just then say the number of such risk factors is not just one, it is more than one.

And the initial evidence for there being an "equity risk premium" is more or less the same as the initial evidence for there being more than one equity risk premium--it is observable historically.

Conversely, if you were TRULY agnostic about the existence of an equity risk premium, you'd have no particular reason to invest in stocks over low-risk IPS. Sure, for example, history says U.S. stocks returned more than TIPS would have returned. And that is explainable by an equity risk premium. But if you ignore that history and don't accept that conclusion, why hold U.S. stocks when TIPS are available?
User avatar
nisiprius
Advisory Board
Posts: 44068
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: Are Dimensional ETFs worth it?

Post by nisiprius »

Always passive wrote: Thu Jun 24, 2021 1:01 pm This forum has typically recommended total market ETFs, and more specifically Vanguard funds. For me, living overseas, that means VTI and VXUS. Now that Dimensional ETFs are available to all, are they worth it when compared to Vanguard's?
I wouldn't make any judgement based on the brand name.

To be specific, DFA is not currently offering ETFs that correspond to the DFA US Micro-Cap Portfolio fund, nor to the DFA US Small-Cap Value Portfolio fund.

Let me describe a certain investing strategy. I will describe it by example. A typical example is the set of the model portfolios presented by Larry Swedroe in The Only Guide to a Winning Investment Strategy You'll Ever Need. It is characterized by a portfolio of about ten funds, including small and value tilts, often including REITs; small and value tilts within the international stock elements, too; and a use of several bond funds to exploit the Fama-French bond factors as well as the Fama-French stock factors.

A fair number of people believe that

a) this strategy is better than the four-fund untilted portfolios which Vanguard uses in its LifeStrategy funds;

b) implementing this strategy requires funds that are special, factor-oriented tools for the job;

c) DFA offers specific funds that are better tools for this job than Vanguard's equivalents. Their small-cap fund reaches down into smaller-caps, their value fund loads higher on the value factor, and they get an edge due to having access to Fama and French themselves, and to their decision to be passive but not actually indexed.

But the new DFA ETFs have nothing to do with that. They are not equivalent to the DFA funds that are used to implement the classic "DFA-style" strategies. They are just some irrelevant ETFs that happen to be provided by Dimensional.
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.
Gaston
Posts: 198
Joined: Wed Aug 21, 2013 7:12 pm

Re: Are Dimensional ETFs worth it?

Post by Gaston »

NiceUnparticularMan wrote: Thu Jun 24, 2021 3:47 pm I'm not sure I would call it "the Bogle model," but I do think what many would call a simple Boglehead approach is basically a CAPM approach.
I bow to your expertise on the above, but will add my simple view of the Bogle model.

1. Keep costs as low as possible.

2. Buy the whole market.

3. Enjoy the returns. Based on Morningstar, Standard & Poor's and most other data I've seen, this Bogle model has beaten 70-80% of all other investment strategies (themes, tilts, factors, actively-managed) on a pre-tax basis, and 80-90% on on post-tax basis, over a long-term 20-30 year horizon. Whether it will continue to do so in the future is unknown, but it's a great model to place one's bet on.

The problem with French-Fama and other academic models is that they do not accommodate behavioral risk. People, especially males, have a tough time controlling their tendencies to tinker, and end up frittering away whatever advantage an academic model might offer. I'm just as guilty as anyone else on this, but am trying to do better by resisting the temptation to tinker with various tilts. I know that after 10 years of underperformance with a particular tilt, I will become impatient and give up on it.
bigskyguy
Posts: 320
Joined: Sat Jan 24, 2015 4:59 pm

Re: Are Dimensional ETFs worth it?

Post by bigskyguy »

NiceUnparticularMan wrote: Thu Jun 24, 2021 3:52 pm
Multi-factor models just then say the number of such risk factors is not just one, it is more than one.
Again, thanks for your comments. Maybe I'm a bit too simplistic in the way I think about indexing and John Bogle, but my sense is that he looked at the investing world, and came to the conclusion that the average investor was best served by index investing, equities and fixed income inclusive. I've recently read the newest edition of Charlie Ellis' book Winning the Loser's Game, and I cannot argue with his basic premise. No question that looking backward, one can readily identify non-index portfolios that have superior risk adjusted returns. That is essentially what DFA and all non-index investing approaches are based upon. The simplistic beauty of owning the market (equities and fixed income) is very hard to argue with.

That being said, the analytics employed by DFA are indeed eloquent and vey attractive.

I harken back to the movie Moneyball. No question the methodology that led to the success of the Oakland A's in the 70s were impressive. But it didn't last. Similarly, the vast sums spent by LA Dodgers and NY Yankees didn't lead to a World Series in 34 of the last 40 years.

You got to show up and play the game. And one can't predict a priori how it will work out.

Bummer, isn't it?

So getting back to the initial question, is DFA worth it? Beats me!!
Carol88888
Posts: 451
Joined: Wed Jan 24, 2018 2:24 am

Re: Are Dimensional ETFs worth it?

Post by Carol88888 »

I thought you had to have a personal advisor in order to get access to Dimensional funds and ETFs. In that case, I would say not worth it because you are adding an extra level of expense.
absolute zero
Posts: 969
Joined: Thu Dec 29, 2016 4:59 pm

Re: Are Dimensional ETFs worth it?

Post by absolute zero »

Carol88888 wrote: Thu Jun 24, 2021 5:05 pm I thought you had to have a personal advisor in order to get access to Dimensional funds and ETFs. In that case, I would say not worth it because you are adding an extra level of expense.
True for mutual funds, but not for ETF’s. There’s no way to limit access to an ETF - it can be purchased by anyone.
absolute zero
Posts: 969
Joined: Thu Dec 29, 2016 4:59 pm

Re: Are Dimensional ETFs worth it?

Post by absolute zero »

NiceUnparticularMan wrote: Thu Jun 24, 2021 3:47 pm
absolute zero wrote: Thu Jun 24, 2021 2:38 pm
bigskyguy wrote: Thu Jun 24, 2021 2:11 pm Dimensional has historically been a fund company that adhere's to the Fama-French three factor model (small cap, value, market risk) of investing. That is a distinct (and presently somewhat out of favor) model. It is in contrast to.....the five factor model (includes momentum), and I suppose an array of other models.
Not to nitpick, but these comments are somewhat inaccurate. The Fama-French five factor model does not include momentum, though other asset pricing models do. Regardless, DFA does utilize momentum in their funds as a screening tool. They also utilize additional factors beyond the original FF 3 factor model, e.g. they target the profitability factor in a number of their funds. For what it's worth, I do not invest in any DFA funds.
bigskyguy wrote: Thu Jun 24, 2021 2:11 pm It is in contrast to the singular indexing model (the Bogle model)
Did you make that up? I've never read the words "singular indexing model." John Bogle never created any asset pricing models i.e. models that attempt to explain the cross section of returns. His philosophy was to ignore the cross section.
I'm not sure I would call it "the Bogle model," but I do think what many would call a simple Boglehead approach is basically a CAPM approach. And I do think generally that approach more or less assumes only one type of risk, known in the literature as market risk or Beta. The plan is you get exposure to Beta with total market index stock funds, then modulate your risk with some fixed percentage of low-to-no-risk bonds, and that is then it for planning.

There might be some dancing around with which bonds to use, what counts as the total market, emergency funds or other savings not included in this portfolio, and so on. But I do feel that is basically a CAPM sort of understanding of portfolio theory.
I think it’s a decent analogy, though I do have one issue with it. CAPM is an obsolete pricing model, and yet there are sophisticated/educated investors who still choose to ignore factor investing and instead invest in total market funds only. Why? They either (A) do not wish to take on the risk of the additional factors or, more likely, (B) they acknowledge that the FF 3 factor has a high degree of explanatory power, but they believe that it’s unlikely that the factor premiums will persist.

I think it’s worth separating (1) the explanatory power of a model and (2) the belief that the premiums associated with its risk factors will continue to be positive in the future. It’s possible that factor investors won’t be rewarded over the next, say, 30 years, but I find it extremely unlikely that CAPM will come anywhere close to a 3 factor model when it comes to explaining the cross section of returns, even over that hypothetical future 30 year period.

To reject the FF 3 factor model and instead say that CAPM is a better model, one must believe that small stocks and big stocks will cease to behave differently from one another. One must believe that you could take a diversified portfolio of 200 random small cap stocks and compare that to a portfolio of 200 large cap stocks, and find no significant difference in performance between the two portfolios in the future. Personally I doubt whether a small cap premium will persist, but I thinks it’s extremely likely that the size factor (and the value factor too) will continue to explain returns.
Gaston
Posts: 198
Joined: Wed Aug 21, 2013 7:12 pm

Re: Are Dimensional ETFs worth it?

Post by Gaston »

absolute zero wrote: Thu Jun 24, 2021 5:26 pm To reject the FF 3 factor model and instead say that CAPM is a better model, one must believe that small stocks and big stocks will cease to behave differently from one another. One must believe that you could take a diversified portfolio of 200 random small cap stocks and compare that to a portfolio of 200 large cap stocks, and find no significant difference in performance between the two portfolios in the future.
Sorry if I am repeating myself. I don’t reject factor models, but I think that most investors lack the disciple to carry them out over the long term. Me included. Waiting perhaps 10 years for small cap value to have its day in the sun is hard. Behavioral risk steps in, people tinker with their tilts, and the investor ends up with returns that are below the “total market” index model.
User avatar
Marmot
Posts: 530
Joined: Sun Oct 10, 2010 1:44 pm
Location: Phoenix, AZ

Re: Are Dimensional ETFs worth it?

Post by Marmot »

Always passive wrote: Thu Jun 24, 2021 1:01 pm This forum has typically recommended total market ETFs, and more specifically Vanguard funds. For me, living overseas, that means VTI and VXUS. Now that Dimensional ETFs are available to all, are they worth it when compared to Vanguard's?
Sounds like something you would find in the Matrix. Just saying.
Marty....don't go to the year 2020....Dr. Emmett Brown
NiceUnparticularMan
Posts: 2526
Joined: Sat Mar 11, 2017 7:51 am

Re: Are Dimensional ETFs worth it?

Post by NiceUnparticularMan »

Gaston wrote: Thu Jun 24, 2021 4:26 pm The problem with French-Fama and other academic models is that they do not accommodate behavioral risk. People, especially males, have a tough time controlling their tendencies to tinker, and end up frittering away whatever advantage an academic model might offer. I'm just as guilty as anyone else on this, but am trying to do better by resisting the temptation to tinker with various tilts. I know that after 10 years of underperformance with a particular tilt, I will become impatient and give up on it.
I completely agree that behavioral risk is a huge issue in general, and specifically a risk for any individual with a portfolio which departs significantly from the most popular indices in an individual's broader community--in the U.S., this would be something like the S&P 500.

Indeed, it is possible this behavioral risk is at least some of the systemic risk actually associated with the factors. In that sense I think it might be a bit unfair to say the models themselves do not accommodate such risks, but investment plans based on the models may fail to take those risks sufficiently seriously in application.

So yeah, I agree that most people are probably ill-advised to invest with a meaningful exposure to these risks. Generally, most people should just put their long-term investments in something like a popular Target fund and be done with it. And Target funds are generally not tilted, probably not least because if they had an extended period of underperforming untilted Target funds, they would become unpopular.
NiceUnparticularMan
Posts: 2526
Joined: Sat Mar 11, 2017 7:51 am

Re: Are Dimensional ETFs worth it?

Post by NiceUnparticularMan »

bigskyguy wrote: Thu Jun 24, 2021 4:43 pm
NiceUnparticularMan wrote: Thu Jun 24, 2021 3:52 pm
Multi-factor models just then say the number of such risk factors is not just one, it is more than one.
Again, thanks for your comments. Maybe I'm a bit too simplistic in the way I think about indexing and John Bogle, but my sense is that he looked at the investing world, and came to the conclusion that the average investor was best served by index investing, equities and fixed income inclusive. I've recently read the newest edition of Charlie Ellis' book Winning the Loser's Game, and I cannot argue with his basic premise. No question that looking backward, one can readily identify non-index portfolios that have superior risk adjusted returns. That is essentially what DFA and all non-index investing approaches are based upon. The simplistic beauty of owning the market (equities and fixed income) is very hard to argue with.
So I don't think much necessarily depends on this, but index investing and factor investing are not incompatible. An index can be of the total market, but it can also be of subsets of the total market. Indeed, the most popular index in the U.S., the S&P500, is a subset index itself. And it was the basis of the first Vanguard index fund, not a total market index.

And there are now a variety of subset indices with factor tilts, and index funds based on these indices, including as offered by Vanguard. So, one can use exclusively index funds to implement a multi-factor strategy.

That said, generally these tilted index funds do have a somewhat higher cost than a total market or S&P 500 fund. So, at a minimum you have to ask whether whatever you are doing with them is worth the extra cost.

The other thing I would just keep emphasizing is that depending on your view on what multi-factor models mean, a tilted portfolio may have higher expected returns than an untilted portfolio with the same BETA risk, but NOT "superior risk adjusted returns" when you ALSO adjust for the DIFFERENT risks associated with other factors.

This is a very important point! If you believe the factors are associated with risk premiums, to get the "premium" in the sense of higher expected returns, you have to add the risk too!

On the one hand, this makes these factor models much more robust in the sense they are compatible with things like the efficient market hypothesis. But on the other, it means people who tilt are exposing themselves to more risks--and it is dangerous to do that without understanding that is what you are doing.
Gaston
Posts: 198
Joined: Wed Aug 21, 2013 7:12 pm

Re: Are Dimensional ETFs worth it?

Post by Gaston »

NiceUnparticularMan wrote: Fri Jun 25, 2021 8:06 am I think it might be a bit unfair to say the models themselves do not accommodate such risks, but investment plans based on the models may fail to take those risks sufficiently seriously in application.
Yes, it might be unfair, but would welcome your view on the following. Some quick examples:

- I have not seen an academic study that looks at how long the average investor holds an equity investment (a stock, a fund, or an ETF). But this seems important to me. If, on average, an investor holds a fund for 5 years before moving to a different tilt, then advocating a tilt that requires a 10-15 year timeframe seems, to me, to be a non-starter.

- I also have not seen an academic study that adequately compares the average return of a tilt (eg, small cap value) versus what actual investors earned with the tilt. For example, let’s say Fund ABC returned 10% per year over 10 years. But what did the actual investors in the fund earn (who entered the fund at different times, exited at various dates, gave up altogether on the tilt, etc)? I bet it’s a much lower number.

- I’ve also not seen a good study that focuses on after-tax returns which, for many of us, is the key number. You will know that every shift into or out of a fund carries a tax consequence. My guess it that chasing tilts greatly reduces after-returns.

BTW, I don’t want to give the impression that I’ve read lots of studies, but I have read a few of the more famous ones. And I’m happy to be further educated on the above.
-
NiceUnparticularMan
Posts: 2526
Joined: Sat Mar 11, 2017 7:51 am

Re: Are Dimensional ETFs worth it?

Post by NiceUnparticularMan »

absolute zero wrote: Thu Jun 24, 2021 5:26 pm
NiceUnparticularMan wrote: Thu Jun 24, 2021 3:47 pm
absolute zero wrote: Thu Jun 24, 2021 2:38 pm
bigskyguy wrote: Thu Jun 24, 2021 2:11 pm Dimensional has historically been a fund company that adhere's to the Fama-French three factor model (small cap, value, market risk) of investing. That is a distinct (and presently somewhat out of favor) model. It is in contrast to.....the five factor model (includes momentum), and I suppose an array of other models.
Not to nitpick, but these comments are somewhat inaccurate. The Fama-French five factor model does not include momentum, though other asset pricing models do. Regardless, DFA does utilize momentum in their funds as a screening tool. They also utilize additional factors beyond the original FF 3 factor model, e.g. they target the profitability factor in a number of their funds. For what it's worth, I do not invest in any DFA funds.
bigskyguy wrote: Thu Jun 24, 2021 2:11 pm It is in contrast to the singular indexing model (the Bogle model)
Did you make that up? I've never read the words "singular indexing model." John Bogle never created any asset pricing models i.e. models that attempt to explain the cross section of returns. His philosophy was to ignore the cross section.
I'm not sure I would call it "the Bogle model," but I do think what many would call a simple Boglehead approach is basically a CAPM approach. And I do think generally that approach more or less assumes only one type of risk, known in the literature as market risk or Beta. The plan is you get exposure to Beta with total market index stock funds, then modulate your risk with some fixed percentage of low-to-no-risk bonds, and that is then it for planning.

There might be some dancing around with which bonds to use, what counts as the total market, emergency funds or other savings not included in this portfolio, and so on. But I do feel that is basically a CAPM sort of understanding of portfolio theory.
I think it’s a decent analogy, though I do have one issue with it. CAPM is an obsolete pricing model, and yet there are sophisticated/educated investors who still choose to ignore factor investing and instead invest in total market funds only. Why? They either (A) do not wish to take on the risk of the additional factors or, more likely, (B) they acknowledge that the FF 3 factor has a high degree of explanatory power, but they believe that it’s unlikely that the factor premiums will persist.

I think it’s worth separating (1) the explanatory power of a model and (2) the belief that the premiums associated with its risk factors will continue to be positive in the future. It’s possible that factor investors won’t be rewarded over the next, say, 30 years, but I find it extremely unlikely that CAPM will come anywhere close to a 3 factor model when it comes to explaining the cross section of returns, even over that hypothetical future 30 year period.
So personally, I plan to "untilt" during our transition to retirement, and the reasoning is purely (A). I have a more heightened sense of behavioral risk than I used to, and I think that avoiding behavioral risk by doing that is worth the cost in expected returns.

And in no way do I think looking back 30 years later will a single-factor model do a better job explaining the cross-section of returns than a multi-factor model. And in fact, I don't think it is likely there will be negative rather than positive signs on small or value, meaning I expect the "premiums" for those factors will more likely than not occur over that period. I just think it will be worth forgoing those premiums to minimize behavioral risk.
To reject the FF 3 factor model and instead say that CAPM is a better model, one must believe that small stocks and big stocks will cease to behave differently from one another. One must believe that you could take a diversified portfolio of 200 random small cap stocks and compare that to a portfolio of 200 large cap stocks, and find no significant difference in performance between the two portfolios in the future. Personally I doubt whether a small cap premium will persist, but I thinks it’s extremely likely that the size factor (and the value factor too) will continue to explain returns.
So just to be clear--to make this a fair CAPM test, we have to make sure your two subset portfolios have the same Beta exposure. If it happened that filtering by size caused a Beta difference, then different returns for those portfolios would not necessarily be inconsistent with CAPM.

But yes, if an otherwise random small portfolio and an otherwise random large portfolio with the same Beta exposure have significantly different returns, then there is something wrong with CAPM.

Still, I think it is worth emphasizing either those returns for the small portfolio will be higher or lower than the returns for the large portfolio. If higher, that is a "small premium". If lower, then that is a "large premium".

And personally, I would bet on a small premium. I would not guarantee it. But I think it is more likely than not there will be a small premium than a large premium, and I would take the field against specifically zero.
Last edited by NiceUnparticularMan on Fri Jun 25, 2021 8:50 am, edited 1 time in total.
NiceUnparticularMan
Posts: 2526
Joined: Sat Mar 11, 2017 7:51 am

Re: Are Dimensional ETFs worth it?

Post by NiceUnparticularMan »

Gaston wrote: Fri Jun 25, 2021 8:33 am
NiceUnparticularMan wrote: Fri Jun 25, 2021 8:06 am I think it might be a bit unfair to say the models themselves do not accommodate such risks, but investment plans based on the models may fail to take those risks sufficiently seriously in application.
Yes, it might be unfair, but would welcome your view on the following. Some quick examples:

- I have not seen an academic study that looks at how long the average investor holds an equity investment (a stock, a fund, or an ETF). But this seems important to me. If, on average, an investor holds a fund for 5 years before moving to a different tilt, then advocating a tilt that requires a 10-15 year timeframe seems, to me, to be a non-starter.

- I also have not seen an academic study that adequately compares the average return of a tilt (eg, small cap value) versus what actual investors earned with the tilt. For example, let’s say Fund ABC returned 10% per year over 10 years. But what did the actual investors in the fund earn (who entered the fund at different times, exited at various dates, gave up altogether on the tilt, etc)? I bet it’s a much lower number.

- I’ve also not seen a good study that focuses on after-tax returns which, for many of us, is the key number. You will know that every shift into or out of a fund carries a tax consequence. My guess it that chasing tilts greatly reduces after-returns.

BTW, I don’t want to give the impression that I’ve read lots of studies, but I have read a few of the more famous ones. And I’m happy to be further educated on the above.
-
So I don't know the full extent of what is available either, but I will just note one general problem is that the very existence of low-cost vehicles for factor investing is extremely new in the broader scheme. And it is constantly evolving--for example, ETFs in general only date back to the 1990s, for example, and having a lot of tilted ETFs has been true over even less of a time frame.

My point is that there really isn't much of a record to use for the sort of practical investor studies you are describing, and it is a moving target anyway.

So yes, these may be significant cost/risk factors in a practical context. But it may be difficult to study them due to the lack of a long, stable record.
caklim00
Posts: 2389
Joined: Mon May 26, 2008 10:09 am

Re: Are Dimensional ETFs worth it?

Post by caklim00 »

No, Avantis ETFs are available for cheaper, have better factor exposure, and have already amassed a significant asset base within the past 2 years. AVUV now has 1.4B in assets and is quickly even catching up to VIOV with 1.5B in assets. I suppose DFA could serve as a decent tax loss harvest partner but then again you could just go with state street, blackrock, or vanguard for cheaper.

I have funds invested with all of the above (except blackrock at the moment). I'll consider moving out of the last DFA holdout for me DFEVX (DFA EM Value) when Avantis releases AVES (Avantis EM Value). Hopefully I can find a broker bonus to make it worthwhile.
User avatar
nisiprius
Advisory Board
Posts: 44068
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: Are Dimensional ETFs worth it?

Post by nisiprius »

NiceUnparticularMan wrote: Fri Jun 25, 2021 8:37 am...And personally, I would bet on a small premium. I would not guarantee it. But I think it is more likely than not there will be a small premium than a large premium, and I would take the field against specifically zero...
But you are leaving risk out of the picture. The essential question is does increasing risk by overweighting small-caps do anything all that different, and all that better, than increasing risk simply by increasing your stock allocation?

Once you phrase the question that way, things become much murkier.

Far too many presentations take the easy way out of simply looking at raw return, as in too-often-reproduced versions of an Ibbotson chart, look! more :moneybag :moneybag :moneybag

Image

(This one even plots inflation separately, rather than plotting real return, giving the visual impression of uninterrupted upward movement in stocks rather than showing a flat spot in 1966-1982.)

The increased risk of small-caps is plain enough to the eye on this chart if you look for it. However, this chart gives no way of judging the relationship. You can't tell by eye whether small stocks or large stocks have had higher risk-adjusted return.

The case for small-caps and for small-cap value having higher risk-adjusted return is murky, and the case for them getting much benefit from imperfect correlation with the rest of the market is also murky.
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.
NiceUnparticularMan
Posts: 2526
Joined: Sat Mar 11, 2017 7:51 am

Re: Are Dimensional ETFs worth it?

Post by NiceUnparticularMan »

nisiprius wrote: Fri Jun 25, 2021 9:45 am
NiceUnparticularMan wrote: Fri Jun 25, 2021 8:37 am...And personally, I would bet on a small premium. I would not guarantee it. But I think it is more likely than not there will be a small premium than a large premium, and I would take the field against specifically zero...
But you are leaving risk out of the picture. The essential question is does increasing risk by overweighting small-caps do anything all that different, and all that better, than increasing risk simply by increasing your stock allocation?
So I have tried to be very careful to explain that in my view increasing expected returns with factor tilts means increasing risk, and that adding additional factor exposure does not mean you are increasing your risk-adjusted returns. Instead, it just means you are adding more risk of a form which is not captured by Beta, aka market risk.

And I agree it is a fair question to ask whether adding more risk just with more Beta could be an equally good or better idea to adding more non-Beta risk with factor tilts. I don't think that is an easy question to answer, though, because it would take a really good understanding of the different risks, and how those intersected with that specific personal investor's other circumstances, including but not limited to their use of various available risk management tools.

And the context of the statements you quoted is I am explaining that on the one hand, I can believe it is more likely than not there will be something like a small premium going forward, and on the other, I can rationally choose not to pursue it with my planned retirement portfolio. And that isn't irrational because I fully accept that any small premium comes with additional risk.
The case for small-caps and for small-cap value having higher risk-adjusted return is murky, and the case for them getting much benefit from imperfect correlation with the rest of the market is also murky.
Again, to be very clear, I would personally never argue SV-tilted portfolios have a higher risk-adjusted expected return. I do continue to believe they will more likely than not have a higher return, but only because they expose investors to more systemic risk.
hnd
Posts: 601
Joined: Mon Jun 22, 2020 11:43 am

Re: Are Dimensional ETFs worth it?

Post by hnd »

exodusNH wrote: Thu Jun 24, 2021 1:58 pm
Always passive wrote: Thu Jun 24, 2021 1:01 pm This forum has typically recommended total market ETFs, and more specifically Vanguard funds. For me, living overseas, that means VTI and VXUS. Now that Dimensional ETFs are available to all, are they worth it when compared to Vanguard's?
My understanding is the ETFs they currently offer are not for the funds that people actually want access to.
This.
Gaston
Posts: 198
Joined: Wed Aug 21, 2013 7:12 pm

Re: Are Dimensional ETFs worth it?

Post by Gaston »

NiceUnparticularMan wrote: Fri Jun 25, 2021 8:47 am So yes, these may be significant cost/risk factors in a practical context. But it may be difficult to study them due to the lack of a long, stable record.
Sounds right.
Wade Garrett
Posts: 180
Joined: Wed Jul 10, 2013 2:41 pm
Location: Jasper, MO/Double Deuce

Re: Are Dimensional ETFs worth it?

Post by Wade Garrett »

IMO the DFA ETFs that are currently available are not worth it. IF DFA releases non-watered down ETFs of its strongly tilted strategies (USSCV, DISCV, EMV/EMSC) they MAY be worth it IF DFA competes on ER.
"I'm not an inventor. I'm an improver. I see things that are wrong, and I improve them." - Larry David, Curb Your Enthusiasm
FoolStreet
Posts: 1308
Joined: Fri Sep 07, 2012 12:18 am

Re: Are Dimensional ETFs worth it?

Post by FoolStreet »

Always passive wrote: Thu Jun 24, 2021 1:24 pm
livesoft wrote: Thu Jun 24, 2021 1:09 pm I am sure that for some people Dimensional ETFs will be "worth it" to them and for some other people they will not be. I am trying to figure out how to answer your question better than that, but I cannot for the life of me give you a better answer.
What is it that make Dimensional funds, now some turned into ETFs so attractive that some investors are trying to get them by having financial advisors manage their money?
The literature points at the funds being developed using the most up to date academic research.
It is really hard to find a financial advisor who will invest in simple 3 fund portfolios. I think that advisors want to charge more, so they have more complicated index funds that slice and dice into smaller subclass index funds and use that to make it sound like things are more complicated and demand more advisor value, hence fees they can charge. I don’t think the slice and dice adds more value than the standard macro slices of. A 3 fund portfolio. I don’t like the lock in and I don’t like the high AUM fees of advisors who manage ‘index funds’.
Post Reply