DFA funds no better than Vanguard?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
barodian
Posts: 5
Joined: Thu Sep 17, 2015 2:35 pm

DFA funds no better than Vanguard?

Post by barodian »

Our company 401k plan is in the process of being transferred to a new company. We have access to DFA funds in the current setup, but not in the new one. New company representative says that after expense, DFA funds do no better than Vanguard funds. I have invested in DFA funds, as according to some investors, advisors on this board have shown otherwise.

Is there any merit to the new company stance?

Thanks,
Sandy
livesoft
Posts: 86075
Joined: Thu Mar 01, 2007 7:00 pm

Re: DFA funds no better than Vanguard?

Post by livesoft »

Yes, I think so. DFA funds do not always do better than Vanguard funds. When one picks Vanguard funds with the same "factor loadings" as DFA funds, then I think they are pretty hard to tell apart. Have you made the comparisons yourself?
Wiki This signature message sponsored by sscritic: Learn to fish.
Topic Author
barodian
Posts: 5
Joined: Thu Sep 17, 2015 2:35 pm

Re: DFA funds no better than Vanguard?

Post by barodian »

Livesoft,

Yes. But with my consumer tools, sometime one and sometime the other has done better. Before, I started investing in DFA the information I gathered was that DFA would do 1 to 1.5% better that an equivalent Vanguard fund, as DFA was more valuey, and its small was smaller that Vanguard.
User avatar
JoMoney
Posts: 16260
Joined: Tue Jul 23, 2013 5:31 am

Re: DFA funds no better than Vanguard?

Post by JoMoney »

If you're talking about expenses, Vanguard is better on just about every front.
If you're talking about other criteria, like who is better at concentrating on certain FF Factors, probably DFA.
What does "better" mean? I think the DFA icons are EMH proponents that would probably say the market is always "Efficient" DFA just builds portfolios that focus on taking on more of the distinct risk factors.

It's not always a good thing to load heavier on small stocks or value stocks. If it's a strategy you're trying to follow (for better or worse), maybe DFA does a "better" job of loading on those factors. Vanguard does a better job of keeping costs low.
Last edited by JoMoney on Thu Sep 17, 2015 3:14 pm, edited 1 time in total.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham
livesoft
Posts: 86075
Joined: Thu Mar 01, 2007 7:00 pm

Re: DFA funds no better than Vanguard?

Post by livesoft »

This might interest you then. DFA has a 60/40 asset allocation fund with ticker DGSIX. Vanguard has 60/40 asset allocation fund with ticker VSMGX. What happens if one compares the performances of the two at morningstar.com ?

Answer here:
Image

Image

I don't see a 1% to 1.5% better annual performance of the DFA fund over the Vanguard fund over the years, do you?

In any event, one can create a Vanguard fund portfolio that is small-cap and value tilted with ease. Do not go by the fund names, but go by the factor loadings or even a Morningstar 9-box style grid. This will mean that one will need more of the Vanguard Small-cap value index fund than one would need of the DFSVX fund.

Way too often when an advisor is trying to make DFA look good, they simply use the same percentages of fund names and do not look at what the funds actually contain. This is a less-than apples-to-apples comparison.
Last edited by livesoft on Thu Sep 17, 2015 3:18 pm, edited 1 time in total.
Wiki This signature message sponsored by sscritic: Learn to fish.
User avatar
Mel Lindauer
Moderator
Posts: 35782
Joined: Mon Feb 19, 2007 7:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Re: DFA funds no better than Vanguard?

Post by Mel Lindauer »

barodian wrote:Livesoft,

Yes. But with my consumer tools, sometime one and sometime the other has done better. Before, I started investing in DFA the information I gathered was that DFA would do 1 to 1.5% better that an equivalent Vanguard fund, as DFA was more valuey, and its small was smaller that Vanguard.
As Livesoft pointed out, you need to compare apples to apples to get a valid comparison. And you also need to remember that the DFA funds normally come with an advisor fee (often ~1%), so that reduces the investor's return.

Finally, you may well have gotten the "DFA would be 1 to 1.5% better" from a DFA sales person who was trying to overcome his/her fee. And while DFA advisors often state that their "handholding" fee helps investors stay the course in bad times, in 2014, Morningstar reported that in six of the top 10 DFA funds, investors actually UNDERPERFORMED the very funds they were invested in which seems to shoot that argument down.
Best Regards - Mel | | Semper Fi
Topic Author
barodian
Posts: 5
Joined: Thu Sep 17, 2015 2:35 pm

Re: DFA funds no better than Vanguard?

Post by barodian »

Came across this article from 2010 which agrees with you gentlemen.

http://www.cbsnews.com/news/dfa-vs-vang ... is-better/

Another poster who had posted earlier on this forum:


Random Walker

I Switched to DFA

Report this post
Quote

Postby Random Walker » Tue Apr 20, 2010 1:48 pm
I think the VG v. DFA decision can easily come down to personal circumstance. I have read all the books and am quite psychologically disciplined. My compulsiveness reaches its limit when it comes to keeping sufficient records for efficient TLH. I am about 85% taxable and 15% tax deferred. And I have come to believe in small and value tilts as primarily risk story and perhaps a bit behavioral too.
So I considered the following in making the jump from VG to DFA for myself:
1. VG lacked tax managed value funds. The DFA core funds and tax managed funds give me access to small and value tilts with my taxable money.
1a. VG index funds effectively behave as growth funds because of the index construction. With VG there is a major tradeoff between tax efficiency and diversification across risk factors for taxable money
2. With DFA I would therefore have a more efficient portfolio with risk spread better across the 3 risk factors: beta, small, and value. My portfolio is more efficient and more torqued and in a tax efficient manner.
3. DFA has access to international small value which may be a better contributor to a portfolio than EM
4. Securities lending by DFA somewhat offsets the ER difference between DFA and VG
5. The advisor fee of 0.5-1% can potentially be offset by aggressive efficient TLH by advisor throughout the year which I would not do as well on my own.
6. My advisor manages a ladder of individual bonds which I would not do on my own. This presents additonal TLH opportunities and avoids the ER of a bond fund

So i roughly calculated
ER difference perhaps -0.3%
Advisor fee -0.7%
Small &Value tilt + 1.5%
TLH benefit + 0.7%
_______
Total potential advantage ~1.2% per year

I think this is conservative. The small and value tilt might be larger long term. The improved efficiency of the portfolio might result in a smoother ride for the portfolio bringing the annualized return closer to the average annual return, but I can't guesstimate that number. My advisor is managing a ladder of individual bonds as well (which I would not do on my own). This presents additional TLH opportunities that I might not have with bond funds. Lastly there are benefits to advisor that I'll learn about over time that I definitely cannot quantitate now.
So overall, I am just pointing out that the DFA VG decision can come down to personal circumstance and what the advisor does to justify his fee. I think the above are significant issues to consider in the decision making process. I really surprised myself. After reading all the books and enjoying investing, I have effectively handed it all over to an advisor!
livesoft
Posts: 86075
Joined: Thu Mar 01, 2007 7:00 pm

Re: DFA funds no better than Vanguard?

Post by livesoft »

Random Walker posted this thread recently: viewtopic.php?t=172559

I will say that one can do small-cap and value-tilting with Vanguard rather easily nowadays. In 2009 (year before your quoted stuff), Vanguard was just getting going with a small-cap foreign index fund (VSS / VFSVX), so it has that covered now. Vanguard's international index fund did not contain small-caps at that time as well. So tilting to small-caps is easier than before.

As for value-tilting, this is where examining what is in a fund can be important and not just the name of the fund. As I mentioned, one will need less "blend" funds like Total US Stock Market Index and more "value" fund assets like Small-cap value index or Mid-cap value index to match some of the factor loadings that a DFA large-cap value fund might have. See posts by Robert T for help with that.

As for tax-loss harvesting mentioned by Random Walker, I find that easy to do myself, but it would not apply to a 401(k) anyways.

I will admit that back in 2000, I lusted after DFA funds, too. But I could not bring myself to pay an advisor fee to get them. Instead, I did some research and found some small-cap foreign funds that were available at that time. I have not regretted my decision not to use DFA.
Wiki This signature message sponsored by sscritic: Learn to fish.
User avatar
JoMoney
Posts: 16260
Joined: Tue Jul 23, 2013 5:31 am

Re: DFA funds no better than Vanguard?

Post by JoMoney »

barodian wrote:... Postby Random Walker » Tue Apr 20, 2010 1:48 pm
...
Small &Value tilt + 1.5%
...
What's nice about the Costs Matter Hypothesis, is it reliably performs as expected. Theoretical strategies we hope will outperform don't always work the way we want.
Image
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham
User avatar
Mel Lindauer
Moderator
Posts: 35782
Joined: Mon Feb 19, 2007 7:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Re: DFA funds no better than Vanguard?

Post by Mel Lindauer »

JoMoney wrote:
barodian wrote:... Postby Random Walker » Tue Apr 20, 2010 1:48 pm
...
Small &Value tilt + 1.5%
...
What's nice about the Costs Matter Hypothesis, is it reliably performs as expected. Theoretical strategies we hope will outperform don't always work the way we want.
Image
And then you still have to subtract the annual advisor fees over that time frame which results in even less money for the investor.
Best Regards - Mel | | Semper Fi
pingo
Posts: 2638
Joined: Sat Sep 19, 2009 8:24 pm

Re: DFA funds no better than Vanguard?

Post by pingo »

I won't say that one is better than the other and JoMoney basically beat me to the punch, but I'll post anyway.

I'm a huge fan of Vanguard, a respecter of DFA, and I have no quarrel with DFA products in and of themselves. Costs and desired asset allocation are the first considerations in fund selection, but there are others. As to costs, DFA funds offered through a 401k often avoid additional advisor costs. Going to the point of asset allocation and Livesoft's point, I'll post a clickable chart comparing the Vanguard 60/40 LifeStrategy Moderate Growth Fund (VSMGX ER 0.16%) and DFA Global Allocation 60/40 Fund (DGSIX ER 0.29%):

Image

A little over a year ago, we'd be saying that DGSIX was beating Vanguard by a small margin. One might even try to argue that the tendency of DGSIX, based on the chart, is to separate from the broad market-oriented VSMGX for some level of outperformance. I'd even note that the value-tilted DGSIX was outperforming VSMGX, which includes more growth stocks, during a period when growth has basically been outperforming value.

However, even if DGSIX should outperform in the long(er) run, theoretically, it can only do so by taking on greater risk--at least that is my understanding. Well, risk occasionally shows up, as it did in the form of lower returns year-to-date which translate into DGSIX's orange line falling back down to the Vanguard's blue line. Whether DGSIX will recover or fall below the Vanguard line, is unknowable until future returns move into the realm of past performance. Sometimes risk shows up to the tune of long-term underperformance, even if the theories behind DFA's value/factor strategies hold up. Assuming for a moment they hold up, will those long-term periods of DFA underperformance occur during our investment lifetimes and will they make the periods of DFA outperformance insufficient to favor DGSIX? Or, will it be a wash between the two? Or will DFA outperformance prevail? Again, unknowable. That's the risk of it.

Given that it is so difficult to beat low cost beta during the course of one's particular investment lifetime, the least expensive option increases in favor inasmuch as the cost difference widens.
Last edited by pingo on Thu Sep 17, 2015 5:29 pm, edited 15 times in total.
User avatar
nisiprius
Advisory Board
Posts: 52211
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: DFA funds no better than Vanguard?

Post by nisiprius »

(Judging from their website, they want people to call them "Dimensional" rather than "DFA" now?)

It is my personal opinion that if you had access to Dimensional funds before and you have access to Vanguard funds now--assuming it's a reasonable collection of funds--that you are fortunate indeed.

It is my honest opinion that nobody can say which funds are better, and anyone who says they can is grinding an axe. Yes, Vanguard's expense ratios are lower than Dimensional's, but Dimensional's can honestly be called low (if you're not paying an advisor fee).

It's Coke versus Pepsi, PC versus Mac, Dunkin' Donuts versus Starbucks. You have two firms that are both firmly grounded in "passive" investing. I don't have access to Dimensional funds so this could be sour grapes, but if had a free choice between Dimensional and Vanguard I'd choose Vanguard--but that's mostly brand loyalty or Bogle fandom on my part. I do have a rationale or a rationalization--understand better what an index fund is doing, majesty of simplicity, etc. But it's more like a flavor preference than a strong conviction.

Seriously: look at the two charts. Blue is DFA US Core Equity 1 Fund, DFEOX. Orange is Vanguard Total Stock Market Index Fund, Admiral shares, VTSAX. Do you really think the difference between them matters much? If you put $10,000 into each fund at inception of DFEOX, you'd have $39.95 more in DFEOX today. That's over almost exactly ten years and with numbers that small we can approximate and say it's an annual difference of 0.04% annualized.
Source: Morningstar

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.
breeks
Posts: 50
Joined: Fri Sep 11, 2015 4:49 pm

Re: DFA funds no better than Vanguard?

Post by breeks »

If I were you, I would probably not roll over the old 401k, to maintain access to those DFA funds -- unless the account maintenance fees go up dramatically for ex-employees. They have a number of funds that don't have a Vanguard analog, such as international small-cap value.

EDIT: Just reread your post, and it sounds like that's not an option since you're not changing employers. Well, be thankful that you have Vanguard offerings in your new 401k.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: DFA funds no better than Vanguard?

Post by larryswedroe »

The following is bit dated as from my Shrinking Alpha book, but easily updated by anyone interested

Morningstar Percentile Ranking (as of October 31 2014)
Fund 10 Years 15 Years
Domestic
Vanguard 500 Index (VFINX) 29 50
DFA U.S. Large (DFUSX) 24 48
Vanguard Value Index (VIVAX) 36 64
DFA U.S. Large Value III (DFUVX) 7 6
Vanguard Small Cap Index (NAESX) 14 59
DFA U.S. Small (DFSTX) 16 36
DFA U.S, Micro Cap (DFSCX) 39 24
Vanguard S.C. Value Index (VISVX) 27 58
DFA U.S. Small Value (DFSVX) 30 20
Vanguard REIT Index (VGSIX) 34 34
DFA Real Estate (DFREX) 42 27
International
Vanguard Developed Markets Index (VTMGX) 40 46
DFA International Large (DFALX) 39 43
DFA International Value III (DFVIX) 18 11
DFA International Small (DFISX) 28 12
DFA International Small Value (DISVX) 1 7
Vanguard Emerging Markets Index (VEIEX) 31 47
DFA Emerging Markets II (DFEMX) 22 40
DFA Emerging Markets Value (DFEVX) 14 17
DFA Emerging Markets Small (DEMSX) 4 12
Average Vanguard Ranking 30 51
Average DFA Ranking 22 23
Note: A percentile ranking of 1 represents the best performance, 100 the worst.

For the seven Vanguard index funds, the average 10-year and 15-year rankings were 30 percent and 51 percent, respectively. Again, keep in mind the impact of survivorship bias on the long-term rankings. If survivorship bias were accounted for, it is highly likely that the 15-year ranking for the Vanguard funds would be well below 50 percent. As for the 13 passively managed DFA funds, the average 10-year and 15 year rankings were 22 percent and 23 percent, respectively. Outperforming 78 percent and 77 percent of the surviving funds is hardly an average performance. And if Morningstar accounted for survivorship bias, their rankings would almost certainly be considerably higher.
livesoft
Posts: 86075
Joined: Thu Mar 01, 2007 7:00 pm

Re: DFA funds no better than Vanguard?

Post by livesoft »

Mr Swedroe, that's nice, but how do you explain the comparison of DGSIX and VSMGX please?
Wiki This signature message sponsored by sscritic: Learn to fish.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: DFA funds no better than Vanguard?

Post by larryswedroe »

Livesoft
I'll assume you asked in spirit of trying to understand what happened.
So the returns are virtually identical over the most recent 10 year period, according to M*. And actually a bitt surprising as DFA has significant tilt to value and size and the unexpected happened in that there was negative size and value premium over the period. So you got much better than expected returns relatively speaking from DFA, given higher expenses and more exposure to factors with negative premiums. If one knew that ahead of time I would have expected Vanguard to outperform by far more than it did, which was a few basis points. Even more surprising is that with Vanguard's bond portfolio having quite a bit longer maturity and rates collapsed that would only have added to the expectation that Vanguard would have outperformed by a good margin.

My conclusion based on the quick review is that DFA did much better relatively speaking than what I would have expected knowing what we know about premiums that actually happened. Now if the "expected" happened and there were size and value premiums and bond rates say were stable or rose, one would then expect DFA to outperform by perhaps a wide margin. Clearly they appear to have added more value with things like patient trading and screens. At least based on this quick review. So surprisingly even in an environment that favored Vanguard, surprisingly they didn't really outperform.

I hope that is helpful
Larry
livesoft
Posts: 86075
Joined: Thu Mar 01, 2007 7:00 pm

Re: DFA funds no better than Vanguard?

Post by livesoft »

That is helpful, thank you very much.
Wiki This signature message sponsored by sscritic: Learn to fish.
edge
Posts: 3833
Joined: Mon Feb 19, 2007 6:44 pm
Location: NY

Re: DFA funds no better than Vanguard?

Post by edge »

For the core offerings VG is better but DFA offers good products that do not have VG equivalents e.g. EM value.
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: DFA funds no better than Vanguard?

Post by pascalwager »

edge wrote:For the core offerings VG is better but DFA offers good products that do not have VG equivalents e.g. EM value.
I don't think the core offerings of one fund company are better than the other, unless you're referring to VG being cheaper. The VG core is capitalization proportioned total market and the DFA core contains most of the same companies but is over-weighted with respect to size and value. Both are solid products.
VT 60% / VFSUX 20% / TIPS 20%
Day9
Posts: 1000
Joined: Mon Jun 11, 2012 6:22 pm

Re: DFA funds no better than Vanguard?

Post by Day9 »

Anyone remember this thread?

PXSV has changed indexes

This is a small-value ETF. Many investors were disappointed when it changed indexes. Many people believe one of DFA's advantages is that they will 'stay the course' with their systematic approach.
I'm just a fan of the person I got my user name from
heyyou
Posts: 4461
Joined: Tue Feb 20, 2007 3:58 pm

Re: DFA funds no better than Vanguard?

Post by heyyou »

@ those 1% DFA fees
My flat fee advisor offering DFA and VG funds has a few billion dollars of AUM. The flat fee works out to be .18% of my portfolio value so I can't compete in the "mine is smaller than yours" threads on total ERs. That flat fee is less than VG's new advisory service fee of .3% of personal AUM, but total ER could be less at VG. I'm doing what suits me and my innumerate spouse.

Since the 2000 Crash, I no longer see cap weighted large cap index funds as any more advantageous than the other equity sub-asset classes. My AA looks more like a Callan Periodic Table with extra bonds.

livesoft's advice to use the M* nine box analyzer when seeking small cap and value exposure using VG funds is important.
afan
Posts: 8191
Joined: Sun Jul 25, 2010 4:01 pm

Re: DFA funds no better than Vanguard?

Post by afan »

Does anyone have data on the income from securities lending from DFA as compared to Vanguard? As a percent of the returns or assets of the funds? DFA talks about its lending a lot, Vanguard does not have much to say, but both do it.

Thanks
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
garlandwhizzer
Posts: 3565
Joined: Fri Aug 06, 2010 3:42 pm

Re: DFA funds no better than Vanguard?

Post by garlandwhizzer »

Larry wrote:

So the returns are virtually identical over the most recent 10 year period, according to M*. And actually a bitt surprising as DFA has significant tilt to value and size and the unexpected happened in that there was negative size and value premium over the period.
This suggests two possibilities. First, that more than a decade of patience (something few of us have) may be required to harvest factor outperformance. Second, that factor outperformance may be absent or markedly less than in the distant past, not only in the last decade but in the future as well. In short that the "unexpected" as Larry calls it may once again rear its ugly head. What do factor investors do if another decade passes and the SCV factor premium is still either a no show or so small that execution costs and fees swallow it up? I believe this second possibility is a less probable but entirely possible outcome.

I like the concept of value investing and the concept of small cap investing, both of which have risk and behavioral stories that sound entirely plausible. Like everyone else, I like good plausible stories that seem to make sense. However, I have noticed in my investing career that things change over time, that periods of value outperformance alternate with periods of growth outperformance (which is where we are now). Likewise small cap and large cap go through alternating cycles of outperformance/underperformance.

Is it destined that ultimately SCV must always outperform? I'm not so sure any more that the market is as fond of plausible stories as I am. It's important to remember that the low volatility, low beta strategy is entirely 100% counterintuitive (more return, less risk) and yet it is real, persistent, and worldwide. It seems to me that factor investing is more art than science and hence less assuredly predictable.

Garland Whizzer
livesoft
Posts: 86075
Joined: Thu Mar 01, 2007 7:00 pm

Re: DFA funds no better than Vanguard?

Post by livesoft »

I was hoping Mr Swedroe would have had another answer even though his answer was very clever (DFA did better than expected through the negative premium years). And that other answer could have been: DFA didn't rebalance in clever enough ways or manage the portfolio of different asset classes so well during those years, so they didn't do as well as could be expected.

For myself, I have had a small-cap and value-tilted portfolio for many years now. I keep tabs on performance and benchmark performance. I am pleased that small-cap and value-tilted portfolio outperforms the Vanguard Lifestrategy Moderate Growth fund benchmark.
Wiki This signature message sponsored by sscritic: Learn to fish.
saurabh
Posts: 319
Joined: Thu Aug 20, 2015 11:56 am

Re: DFA funds no better than Vanguard?

Post by saurabh »

garlandwhizzer wrote: Is it destined that ultimately SCV must always outperform? I'm not so sure any more that the market is as fond of plausible stories as I am. It's important to remember that the low volatility, low beta strategy is entirely 100% counterintuitive (more return, less risk) and yet it is real, persistent, and worldwide. It seems to me that factor investing is more art than science and hence less assuredly predictable.
In general this forum tends to divide among the classic cap weighted indexers and the tilters with their quant based approachs to identify value. What you will rarely find is someone looking at the actual stocks the small and value funds are invested in in all the obsessive talk about "factor premiums". Take a look at the Vanguard Small Cap Value fund. It is 30% invested in financials and 20% in industrials. It's really more a concentrated sector bet more than anything. Personally I would not be comfortable with strategies that overweight SCV. I know some on this board feel that they can reduce equity exposure and lower risk that way, but at the end of the day I am not comfortable with such a huge sector tilt and that too in small caps, which to begin with are riskier than large caps. The Vanguard large cap value fund is less concentrated as Financials are 23% of the fund and Top 2 sectors are only 35% of the fund unlike 50% for Small Cap value. Also another thing to keep in mind is that a simplistic value strategy like P/B will tend to favor sectors that trade at low premiums to book value (like Financials) or asset heavy companies. Just because that worked in the past doesn't mean it will work in a more modern economy. Yes I know that earnings multiples are also used in determining value companies, but then you start getting into how the sausage is made and you realize that you are essentially relying on computerized backtests and pretty vulnerable to data mining, implementation issues and design error. There is plenty of human element behind the scenes even with quant approaches. I mention this because many on this board shudder at human risk with active management, but don't realize that there is definitely human risk with quant factor investing too as each fund will have a different implementation. I think if you are going to tilt then focus more on large caps as your tracking error relative to cap weighting will be less and I think fundamentally it is just a less volatile strategy compared to the low quality small cap companies that abound.
thx1138
Posts: 1164
Joined: Fri Jul 12, 2013 2:14 pm

Re: DFA funds no better than Vanguard?

Post by thx1138 »

As others mentioned, define "better" would be the first step. If you want to very heavily tilt then certain DFA funds are "better" as they are more tilted.

After that it is very much "costs matter". To date the additional DFA "screens" seem to provide a small increase in return, but of course this is subject to the usual risk that "performance" might just be "tracking error" and today's "plus" may get paid back as tomorrow's "minus". Or that the screens could stop having positive benefit in the future.

Assume there is a constant persistent increased performance for DFA then the next question is how much is that performance increase compared to any additional costs the investor might incur by investing in a DFA fund. There is no consistent answer to that because the costs are not the same to everyone.

Case in point, many people point out you might need an advisor for access to DFA and that will likely cost you 1%. On the other hand you might have DFA in a 401k at no cost to you (the cost might be there in some other way as part of plan fees, but the point is you don't incur the cost by selecting the DFA fund compared to a non-DFA fund in your plan). Or you might be in my perverse situation in which one of the DFA funds in my 401k is actually the *cheapest* one by ER, it is even cheaper than the SP500 fund offered. So of course I go ahead and and use the DFA fund because it is certainly no worse than most any other fund by performance and in this case it is also the cheapest fund in my plan.

So the question to me really is, what is the cost for DFA? That answer really depends on the situation.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: DFA funds no better than Vanguard?

Post by larryswedroe »

Few things
First, Vanguard reconstitutes indexes annually so you tend to lose exposure to a factor over the year, DFA, Bridgeway, AQR do it monthly, so loadings tend to be more stable

Second, of course there is risk that you could go another decade with SV underperforming, you have risks show up. We have had deflationary environment with very low economic growth so you should expect that to occur if you knew we would have that.

Larry
User avatar
nisiprius
Advisory Board
Posts: 52211
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: DFA funds no better than Vanguard?

Post by nisiprius »

thx1138 wrote:...Case in point, many people point out you might need an advisor for access to DFA and that will likely cost you 1%. On the other hand you might have DFA in a 401k at no cost to you (the cost might be there in some other way as part of plan fees, but the point is you don't incur the cost by selecting the DFA fund compared to a non-DFA fund in your plan)...
Barodian said that the DFA funds are in the 401(k) plan he's currently in, presumably with no advisory fee?

Barodian, could you clarify? And do you have the option of continuing to hold the DFA funds when they switch to the new plan? Possibly, rolling them out into a rollover IRA?

To tell the truth, if I were in your situation, and if there were no fees on top of the DFA expense ratio, and I could keep the old funds, and I didn't mind clutter in my 401(k) I'd just hang onto the existing DFA funds, put new money into Vanguard funds, and call it all good. Make no decisions about the existing DFA holdings because there's no need to, make no decisions about new contributions because Vanguard's the only (low-cost, passive) available choice, and call it good.
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.
Hank Moody
Posts: 279
Joined: Fri Oct 17, 2014 9:41 pm

Re: DFA funds no better than Vanguard?

Post by Hank Moody »

Larry,

Aren't Vanguard funds (at least those following CRSP) reconstituted quarterly?

http://www.crsp.com/indexes-pages/index ... g-calendar

Hank
larryswedroe wrote:Few things
First, Vanguard reconstitutes indexes annually so you tend to lose exposure to a factor over the year, DFA, Bridgeway, AQR do it monthly, so loadings tend to be more stable

Second, of course there is risk that you could go another decade with SV underperforming, you have risks show up. We have had deflationary environment with very low economic growth so you should expect that to occur if you knew we would have that.

Larry
-HM
Hank Moody
Posts: 279
Joined: Fri Oct 17, 2014 9:41 pm

Re: DFA funds no better than Vanguard?

Post by Hank Moody »

Larry,

This is to understand, not challenge. When I look at 10 years through 12/2014, I'm seeing positive premiums for both size and value.

The FF US HML Research factor is 0.24%/annual while the FF US SmB (3-factor) is 1.13%/annual. I'm pulling this from Returns.

Can you help me understand where you're seeing negative size and value premiums?

-HM
larryswedroe wrote:Livesoft
I'll assume you asked in spirit of trying to understand what happened.
So the returns are virtually identical over the most recent 10 year period, according to M*. And actually a bitt surprising as DFA has significant tilt to value and size and the unexpected happened in that there was negative size and value premium over the period.
Larry
-HM
saurabh
Posts: 319
Joined: Thu Aug 20, 2015 11:56 am

Re: DFA funds no better than Vanguard?

Post by saurabh »

Hank Moody wrote:Larry,

This is to understand, not challenge. When I look at 10 years through 12/2014, I'm seeing positive premiums for both size and value.

The FF US HML Research factor is 0.24%/annual while the FF US SmB (3-factor) is 1.13%/annual. I'm pulling this from Returns.

Can you help me understand where you're seeing negative size and value premiums?

-HM
larryswedroe wrote:Livesoft
I'll assume you asked in spirit of trying to understand what happened.
So the returns are virtually identical over the most recent 10 year period, according to M*. And actually a bitt surprising as DFA has significant tilt to value and size and the unexpected happened in that there was negative size and value premium over the period.
Larry
Not Larry but I think for a long-only fund you have to cut the spread above in half to approximate what the tilt would get you over the cap weighting. So that would make it closer to a wash.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: DFA funds no better than Vanguard?

Post by larryswedroe »

hank
Sorry should not have generalized as depends on what index, many indices like RAFI for example and Russell, do it annually. Some do it more frequently. The more frequent the better. Forgot Vanguard using CRSP now
Larry
Last edited by larryswedroe on Fri Sep 18, 2015 4:28 pm, edited 1 time in total.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: DFA funds no better than Vanguard?

Post by larryswedroe »

saurabh/Hank
I ran the data for the GLOBAL size and value premiums (my guess is you looked at US), since these are global (not US) funds. I also ran it through August and 10 years. The annualized global size effect 9/05-8/15 was -0.7 and annualized global value effect was -2.6.
If you want the annual premiums they were -1.8 and +0.1

BTW- through August US 10 year annual average premium was -1.3 for value (research factor) and 1.1 for size
Larry
Hank Moody
Posts: 279
Joined: Fri Oct 17, 2014 9:41 pm

Re: DFA funds no better than Vanguard?

Post by Hank Moody »

In any event, is this data enough to support that Vanguard would have done worse than DFA if it had the same tilts?
larryswedroe wrote:saurabh/Hank
I ran the data for the GLOBAL size and value premiums (my guess is you looked at US), since these are global (not US) funds. I also ran it through August and 10 years. The annualized global size effect 9/05-8/15 was -0.7 and annualized global value effect was -2.6.
If you want the annual premiums they were -1.8 and +0.1

BTW- through August US 10 year annual average premium was -1.3 for value (research factor) and 1.1 for size
Larry
-HM
Leeraar
Posts: 4109
Joined: Tue Dec 10, 2013 7:41 pm
Location: Nowhere

Re: DFA funds no better than Vanguard?

Post by Leeraar »

It's Coke versus Pepsi
Faygo in Detroit
PC versus Mac
Chromebook
Dunkin' Donuts versus Starbucks.
McDonald's (Coffee) or Panera

Given your choices, I don't think it matters.

And yes, some will always say, the difference is there, it just didn't show up yet.

L. :)
You can get what you want, or you can just get old. (Billy Joel, "Vienna")
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: DFA funds no better than Vanguard?

Post by larryswedroe »

Hank cannot answer for certain but given what I looked at IMO the answer is yes. Certainly alone the longer bond duration should have helped them quite a bit. And obviously the greater tilts for DFA were drags, and Vanguard has lower expenses. So something was going on that prevented Vanguard from outperforming by more than the 3bp it did when I looked
Larry
small_index
Posts: 631
Joined: Sun Aug 16, 2015 2:47 am

Re: DFA funds no better than Vanguard?

Post by small_index »

Being ignorant of life strategy funds, I looked on Yahoo Finance to get more data on the two 60/40 funds mentioned earlier (DGSIX, VSMGX). Both have a little drift from 60% stocks, with DGSIX holding 55% stocks and almost 10% cash:
http://finance.yahoo.com/q/hl?s=DGSIX+Holdings
http://finance.yahoo.com/q/hl?s=VSMGX+Holdings
Hank Moody
Posts: 279
Joined: Fri Oct 17, 2014 9:41 pm

Re: DFA funds no better than Vanguard?

Post by Hank Moody »

DFA never maintains a 10% cash balance in any of their funds and I doubt Vanguard does either. There must be an error in the way Yahoo tabulated their data.
small_index wrote:Being ignorant of life strategy funds, I looked on Yahoo Finance to get more data on the two 60/40 funds mentioned earlier (DGSIX, VSMGX). Both have a little drift from 60% stocks, with DGSIX holding 55% stocks and almost 10% cash:
http://finance.yahoo.com/q/hl?s=DGSIX+Holdings
http://finance.yahoo.com/q/hl?s=VSMGX+Holdings
-HM
edge
Posts: 3833
Joined: Mon Feb 19, 2007 6:44 pm
Location: NY

Re: DFA funds no better than Vanguard?

Post by edge »

pascalwager wrote:
edge wrote:For the core offerings VG is better but DFA offers good products that do not have VG equivalents e.g. EM value.
I don't think the core offerings of one fund company are better than the other, unless you're referring to VG being cheaper. The VG core is capitalization proportioned total market and the DFA core contains most of the same companies but is over-weighted with respect to size and value. Both are solid products.
By core I mean the normal definition of it, not DFA-speak. VG is better simply because it is cheaper.
small_index
Posts: 631
Joined: Sun Aug 16, 2015 2:47 am

Re: DFA funds no better than Vanguard?

Post by small_index »

Hank Moody wrote:DFA never maintains a 10% cash balance in any of their funds and I doubt Vanguard does either. There must be an error in the way Yahoo tabulated their data.
small_index wrote:Being ignorant of life strategy funds, I looked on Yahoo Finance to get more data on the two 60/40 funds mentioned earlier (DGSIX, VSMGX). Both have a little drift from 60% stocks, with DGSIX holding 55% stocks and almost 10% cash:
http://finance.yahoo.com/q/hl?s=DGSIX+Holdings
http://finance.yahoo.com/q/hl?s=VSMGX+Holdings
Another website views it as 4.2% net cash, categorized as 10.4% long and 6.2% short:
http://money.usnews.com/funds/mutual-fu ... x/holdings
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: DFA funds no better than Vanguard?

Post by larryswedroe »

Few things,
First a fund could have cash on any given day and be FULLY INVESTED, and that is likely the case with both Vanguard and DFA.
And it would be highly intelligent to do so. Say you get cash coming in and now a pure index fund has to buy all the shares right at the close, demanding liquidity and incurring perhaps significant market impact costs, but certainly some "extra" costs as demander of liquidity. So instead the smart fund buys futures (assuming the spreads are thin) and then patiently trades the stocks over time to become a seller vs. demander of liquidity, or at least not paying up on the offered side. One fund I know in the small cap space tells me they buy something like 80% of the trades at below midpoint simply by being patient trader.

Second, saying Vanguard is better simply because it is cheaper is just plain wrong, unless the two funds are identical and both 100% pure indexers and also both use the same index. Simply choosing a "smarter" index say like the CRSP 6-10 instead of R2K could make a more expensive fund better. Simple example is Vanguard foolishly chose to use the R2k small cap index for long time because that was the "retail" index and finally Sauter convinced them to change because the R2k is a dumb index and it has significantly underperforrmed for years because active managers could exploit it. There is a laundry list of ways DFA adds value over a pure indexer, and nothing to do with higher tilts ===patient trading, exclusion of certain type securities, incorporating other factors like MOM, buy and hold ranges, tax management, etc.

Larry
bs1
Posts: 23
Joined: Sun Jun 24, 2012 7:10 pm

Re: DFA funds no better than Vanguard?

Post by bs1 »

Larry,

Bernstein has shown that the value premium tracks with the level of inflation rather than its rate of change, suggesting a negative premium can be sustained for a prolonged (even indefinite) period. do you have data on the premium across countries during different economic cycles? for example Japan since its period of deflation/poor growth began?

if the correlation between inflation and the premium is strong, it would seem that estimates of the volatility of the premium may have been skewed by an overrepresentation of inflationary periods in historical datasets. thus, a globally diversified “core” strategy that emphasizes moderate tilts to multiple factors seems prudent, since confidence in achieving positive premiums over an investing lifetime is diminished if slow growth is more common in the future as economies become “developed”. likewise, indecies that weigh country components according to valuation, such as RAFI, which can result in large deviations from the cap-weighted market portfolio, might be more risky than they appear. on the other hand, strong correlations could potentially be exploited by overweighting value among countries with inflation, since inflations levels have some persistence. are these valid considerations?

thank you
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: DFA funds no better than Vanguard?

Post by nedsaid »

barodian wrote:Livesoft,

Yes. But with my consumer tools, sometime one and sometime the other has done better. Before, I started investing in DFA the information I gathered was that DFA would do 1 to 1.5% better that an equivalent Vanguard fund, as DFA was more valuey, and its small was smaller that Vanguard.
I went to a couple Paul Merriman seminars and yes, this was the claim that was made. And yes, DFA funds have smaller market caps than the equivalent Vanguard funds. Now that Vanguard has switched Value indexes, it is unclear to me if DFA funds are more valuey than Vanguard.
A fool and his money are good for business.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: DFA funds no better than Vanguard?

Post by larryswedroe »

bs1
I looked at the relationship between inflation and value premium in US a while ago and you get what you should expect, IMO.
Since value companies tend to be higher financial leverage companies they benefit from inflation being unexpected since real cost of debt falls. Of course deflation or less than expected inflation works the reverse way.
And that's what you see.
Now if you get VERY high inflation which would cause FED to tighten and drive real rates up to slow economy then that is BAD for these type companies as they suffer from high real rates and they tend to have high financial and operating leverage and lose both ways then

So they tend to do best with moderate growth and moderate inflation.

Larry
stlutz
Posts: 5585
Joined: Fri Jan 02, 2009 12:08 am

Re: DFA funds no better than Vanguard?

Post by stlutz »

Bernstein has shown that the value premium tracks with the level of inflation rather than its rate of change, suggesting a negative premium can be sustained for a prolonged (even indefinite) period. do you have data on the premium across countries during different economic cycles? for example Japan since its period of deflation/poor growth began?
I looked at the relationship between inflation and value premium in US a while ago and you get what you should expect, IMO. Since value companies tend to be higher financial leverage companies they benefit from inflation being unexpected since real cost of debt falls. Of course deflation or less than expected inflation works the reverse way.
bs1: I'd actually suggest just looking at the data yourself. Fama/French return data is available here: http://mba.tuck.dartmouth.edu/pages/fac ... brary.html and historical inflation data is available here: http://www.usinflationcalculator.com/in ... ion-rates/.

Your question actually provoked me to look at the data myself and I wasn't really seeing any reliable correlation--e.g. if I changed the start/end years some the results would change quite a bit. But then I logon to Bogleheads and see that Larry claims just the opposite. When the data is so easily available, you really don't need to rely on message board posters for answers--you can easily come to your own conclusion.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: DFA funds no better than Vanguard?

Post by larryswedroe »

stlutz
The problem is that the relationship between value and inflation isn't linear ( I explained why I believe that is the case), so would not use correlations to find the relationship. I haven't looked at this in long time but here is what I found when I did

When had deflation premium was -5.6%. When inflation 0-1% was 5.2% When inflation 1-2 it was 10%. When 2-3 it was -1.6. When 3-4 it was 3.7%. When 4-5 it was 11.9 and when >5 it was 6.4

Interesting when it was <3% the premium was 2.4% and when above it was 6.2%.
When below 4% it was 2.8 and when above 7.8
User avatar
nisiprius
Advisory Board
Posts: 52211
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: DFA funds no better than Vanguard?

Post by nisiprius »

Hank Moody wrote:DFA never maintains a 10% cash balance in any of their funds and I doubt Vanguard does either. There must be an error in the way Yahoo tabulated their data.
small_index wrote:Being ignorant of life strategy funds, I looked on Yahoo Finance to get more data on the two 60/40 funds mentioned earlier (DGSIX, VSMGX). Both have a little drift from 60% stocks, with DGSIX holding 55% stocks and almost 10% cash:
http://finance.yahoo.com/q/hl?s=DGSIX+Holdings
http://finance.yahoo.com/q/hl?s=VSMGX+Holdings
Morningstar is showing DGSIX as holding cash with a 10.20% long position and a 6.30% short position!

Source: Morningstar
Image

I don't think it would ever make sense to hold cash long and short at the same time, so what that suggests is that DGSIX is doing some tricky thing with leverage or futures contracts or something that Morningstar's computers can't understand, and that I probably wouldn't understand. One of the things I like about Vanguard is whenever I've looked up their portfolio on Morningstar it always seems sensible. Weird things--like 85% cash in this Fidelity fund--show up in other companies' funds, not Vanguard.

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.
livesoft
Posts: 86075
Joined: Thu Mar 01, 2007 7:00 pm

Re: DFA funds no better than Vanguard?

Post by livesoft »

^Perhaps those numbers simply reflect the "Selectively Hedged" fund in DGSIX.
Wiki This signature message sponsored by sscritic: Learn to fish.
bs1
Posts: 23
Joined: Sun Jun 24, 2012 7:10 pm

Re: DFA funds no better than Vanguard?

Post by bs1 »

Larry.

low inflation rates in the US (ex, 0-1%) may frequently reflect a period of changing inflation rather than a sustained low rate of inflation. data from other countries (eg, Japan) that went through long periods of low growth/deflation may indicate different premiums, perhaps owing to a different interest rate cycle as you mention.

regardless, thanks for sharing
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: DFA funds no better than Vanguard?

Post by larryswedroe »

bs1
But what we do know is that value companies TEND TO have higher financial leverage, which means they benefit from inflation that is greater than expected, and vice versa.
Also deflation is associated with weak economic activity and value companies also tend to have higher operating leverage and more irreversible capital thus more likely to suffer in bad economic times.
So a negative premium in deflation, which is what I found, is logical, and the higher premium in higher inflation periods also seems logical.
And I would add that the last 10 years in US we have had less than expected inflation and negative value premium--personally I don't think it a coincidence.
Larry
Post Reply