DFA - Where's the alpha?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
Blue
Posts: 1161
Joined: Sat Jul 12, 2008 10:18 pm

DFA - Where's the alpha?

Post by Blue » Wed Oct 30, 2013 7:56 pm

Thanks to the good work of pvguy, performing factor load analysis and alpha for mutual funds has become readily accessible to individual investors. Portfoliovisualizer.com Perform Factor Analysis.

I was curious if DFA funds had alpha beyond factor loads through their securities lending, patient trading, momentum screens, use of proprietary indices and all of the other intangibles sometimes attributed to DFA.

I randomly selected funds from both DFA and Vanguard to compare alpha's, selecting funds that represented a range of asset classes and styles. This is in no way rigorous, but it was an attempt to satisfy my curiosity and perhaps be a starting point for more discussion.

I ran seven DFA funds through the three factor model against Fama-French returns since inception with monthly alphas of DFFVX (-0.30%), DFEOX (-0.06%), DFVEX (-0.13%), DFTCX (-0.11%), DFSVX (0.08%), DFSTX (-0.03%), DTMEX (-0.09%). Simple average of seven DFA funds (-0.09% monthly)

Same methodology for five Vanguard funds VTSMX (-0.03%), VFIAX (-0.05%), VDAIX (0.08%), VIMAX (0.06%), VISVX (0.00%). Simple average of five Vanguard funds (+0.01% monthly).

edit: DFVEX (not DFEVX)

This seems to suggest, there is no measurable benefit to DFA funds beyond factor loads?
Last edited by Blue on Thu Oct 31, 2013 4:14 am, edited 1 time in total.

Jebediah
Posts: 579
Joined: Tue Aug 28, 2012 9:19 pm
Location: Austin TX

Re: DFA - Where's the alpha?

Post by Jebediah » Wed Oct 30, 2013 8:39 pm

I was noticing this yesterday as well. And also the lameness of the factor loads in the core funds.

It would help if you could post a) the fund names b) the analysis dates c) which of the two datasets you used.

The differences between the two datasets are significant enough (and not in ways you'd expect) that I kind of gave up on the reliability of the whole thing (no offense to pvguy, it's a great site).

Also, I don't think international funds are applicable here.

Calm Man
Posts: 2917
Joined: Wed Sep 19, 2012 9:35 am

Re: DFA - Where's the alpha?

Post by Calm Man » Wed Oct 30, 2013 9:04 pm

I have it figured out !!!!!
If somebody actually could generate "alpha" why the heck would they tell anybody else?
I suspect the difference between DFA and a risk-adjusted Vanguard fund if such a comparison existed would be negative at exactly the difference between the higher DFA expense ratio and the lower Vanguard expense ratio. The difference is in the pockets of the DFA folks.

Bradley
Posts: 486
Joined: Tue May 01, 2007 2:41 pm

Re: DFA - Where's the alpha?

Post by Bradley » Wed Oct 30, 2013 9:22 pm

Blue wrote:
This seems to suggest, there is no measurable benefit to DFA funds beyond factor loads?

This could very well be the case. The vast majority of my portfolio is in Vanguard funds but I have owned DFCEX (DFA”S emerging market fund) and DISVX (Int’l SmCap Value) since 2006 and they are by far the most expensive funds in the portfolio. The monthly performance reports comparing my portfolio to different DFA allocation from 20-100% equity often show my portfolio outperforming. As it stands today my portfolio growth since 2006 is the same as a model portfolio of DFA funds containing 80% equity from inception. There have been stretches of time including 1 year and more in which the Vanguard portfolio has surpassed the growth of all the model portfolios made up of all DFA funds. I started with a low 60’s equity allocation and it is now 70%.
DFA is a great company with some very good funds. Vanguard is a great company with some very good funds. An all DFA portfolio may outperform an all Vanguard fund for one period of time while the Vanguard will outperform over another period. You do not need DFA funds to invest wisely and reach your goals. IMO, the benefits of DFA funds are oversold.
You can sum up any active fund manager’s presentation at an investor conference in one sentence: “We’re doing well, all things considered.”

pvguy
Posts: 27
Joined: Mon Sep 23, 2013 7:54 pm
Location: Austin, TX
Contact:

Re: DFA - Where's the alpha?

Post by pvguy » Wed Oct 30, 2013 10:47 pm

I would caution against jumping to too quick conclusions based on factor regression results or making any investment decisions solely based on it. It can be very misleading to simply look at the alpha values in isolation, or to average them, without considering how well the factor analysis explains the fund returns. For example, one of the funds mentioned above (DFEVX) is an emerging markets fund whereas the Fama-French factors are based on US stock portfolios. If you look at the correlation coefficient, std. error and t-stat values for DFEVX, you can see that the model does not explain the results of the fund very well compared to US stock funds, which is expected. To emphasize this point, you could run Fama-French factor regression on a total bond fund like BND and notice that there seems to be positive alpha (>0.3%), but the result is meaningless and has no statistical significance since the Fama-French three factor model does not explain bond fund returns. As I mentioned in an earlier post, very small differences in a fund's monthly returns can impact the factor analysis results and the factor exposure for many funds fluctuates over time. Also, even if you do happen to find a fund with a nice positive alpha and a good statistical fit to the model, you would have no way to know whether the alpha will persist in the future (past performance does not guarantee future returns).

Jebediah
Posts: 579
Joined: Tue Aug 28, 2012 9:19 pm
Location: Austin TX

Re: DFA - Where's the alpha?

Post by Jebediah » Thu Oct 31, 2013 12:19 am

True, but a significantly negative past alpha across the board with DFA funds would definitely be of interest given how hard their alpha-adding capabilities (patient trading, securities lending, IPO screens, etc) have been sold here.

feature request for pvguy: add alpha x 12 to rolling factor charts, use second y axis?

User avatar
Dale_G
Posts: 3269
Joined: Tue Feb 20, 2007 5:43 pm
Location: Central Florida - on the grown up side of 82

Re: DFA - Where's the alpha?

Post by Dale_G » Thu Oct 31, 2013 1:19 am

I am not surprised there is little, or no, or even negative alpha after accounting for "factors". Patient trading (always makes me laugh) and momentum screens, which seem to counteract "patient treading", and avoiding IPOs, are techniques that are hardly unknown to other investors and investment managers.

It all comes down to expenses.

If anyone is paying AUM fees to get DFA funds, they should realize they are paying for advice, not performance.

Dale
Volatility is my friend

User avatar
Topic Author
Blue
Posts: 1161
Joined: Sat Jul 12, 2008 10:18 pm

Re: DFA - Where's the alpha?

Post by Blue » Thu Oct 31, 2013 4:24 am

Funds I used are (using inception dates through Sept 30, 2013 on 3 Factor model against Fama-French benchmark). Note, I had a typo on DFVEX/DFEVX in my OP (now fixed). Limited my "back of the envelope" search to US only funds.

DFA US Targeted Value, DFFVX, 12/1/2000, alpha -0.30%
DFA US Core Equity I, DFEOX, 12/1/2005, alpha -0.06%
DFA US Vector Equity, DFVEX, 12/1/2006, alpha -0.13%
DFA US T.A. Core Equity II, DFTCX, 12/3/2007, alpha -0.11%
DFA US Small Cap, DFSTX, 7/1/1996, alpha -0.03%
DFA Tax Managed US Equity Portfolio, DTMEX, 11/1/2001, alpha -0.09%

Vanguard Total Stock Market, VTSMX, 7/1/1996, alpha -0.03%
Vanguard SP 500 Admiral Shares, VFIAX, 12/1/2000, -0.05%
Vanguard Dividend Appreciation Index, VDAIX, 6/1/2006, +0.08%
Vanguard Mid-Cap Index Admiral Shares, VIMAX, 12/3/2001, +0.06%
Vanguard Small Cap Value Index, VISVX, 6/1/1998, +0.00%

User avatar
nisiprius
Advisory Board
Posts: 39444
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 - Where's the alpha?

Post by nisiprius » Thu Oct 31, 2013 5:52 am

Dale_G wrote:I am not surprised there is little, or no, or even negative alpha after accounting for "factors". Patient trading (always makes me laugh) and momentum screens, which seem to counteract "patient treading", and avoiding IPOs, are techniques that are hardly unknown to other investors and investment managers.
But alpha is relative to an unmanaged index, not to the actively managed competition, isn't it?

The mythos is that the managers of index funds are shackled by a silly inflexible need to track an index, and that while the managers of a Vanguard fund may well have considerable "transactional skill," they still can't afford to be very "patient" because that mean old index is always after them and won't let up for an instant. While DFA, instead of insisting that managers charge directly in hot pursuit of factor loads, allows them to take their time, hide in the bushes, and sneak up on them. What's the point in doing that unless it provides alpha? If it doesn't cost you alpha to have your factor loading and your index, too, why wouldn't you want both?

By the way, noting Jack Bogle's acerb remarks about being inspired by Samuelson, not by Fama, I do get more and more curious about interpersonal relations between the principals of DFA and Vanguard. Bogle hates it whenever anyone tries to claim (as DFA does) that the origin of the index fund was to be found in the Samsonite Luggage Fund in 1971.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
asset_chaos
Posts: 1658
Joined: Tue Feb 27, 2007 6:13 pm
Location: Melbourne

Re: DFA - Where's the alpha?

Post by asset_chaos » Thu Oct 31, 2013 5:55 am

Does DFA claim to generate alpha in their marketing materials? I don't follow DFA closely, but I'd have expected DFA to claim that they captured beta on known risk factors and to perhaps spin that in marketing to say they produced above market average returns, which over time I suppose would be strictly true though somewhat misleading (their returns shouldn't be above factor risk adjusted returns). All the trading techniques I'd expect to reduce the costs of running their funds, which is a good thing. On the other hand, I'd be surprised if they could make the trading costs so far negative that they would cover the management fee plus even more to account for genuine alpha. The OP's results seem to hint at Bogle's cost matter hypothesis as the determining, irreducible factor.
Regards, | | Guy

User avatar
JoMoney
Posts: 7786
Joined: Tue Jul 23, 2013 5:31 am

Re: DFA - Where's the alpha?

Post by JoMoney » Thu Oct 31, 2013 6:29 am

nisiprius wrote: ...
By the way, noting Jack Bogle's acerb remarks about being inspired by Samuelson, not by Fama, I do get more and more curious about interpersonal relations between the principals of DFA and Vanguard. Bogle hates it whenever anyone tries to claim (as DFA does) that the origin of the index fund was to be found in the Samsonite Luggage Fund in 1971.
I'm sure there's lots of comments both in-between the lines, and more forward with their jabs
Eugene Fama wrote: http://www.dfaus.com/2009/05/the-new-indexing.html
This might seem like "sector betting" to traditional indexers like Vanguard founder John Bogle, who still believe that market risk primarily determines performance and that small stocks and value stocks aren't separate sources of risk and return. The academic community is arriving at a different consensus, one that recognizes multiple independent risks. Investors might even have natural combinations of the different risk exposures that best suit their individual time horizons and preferences. As long as the portfolios they use to gain these exposures are index funds, and as long as the exposures are consistent and not timed to predict markets, this sort of portfolio structuring is not a "sector bet"—it's the new face of indexing.
[EDIT]... I find it interesting, that their claim is to deviate from the market portfolio... but don't try to time it. Which I'm sure is a big concern due to the liquidity problems of small-cap stocks. Investors withdrawing funds at an in-opportune time "flight to liquidity" was one of the things the LTCM catastrophe was blamed on (something I'm sure the folks at DFA are intimately familiar with).
... but here we are today, with a possibly overvalued small-cap market, and some investors questioning how long it will last.. I hope the intermediaries that DFA forces their investors to go through aren't using the same pitch competing small-cap funds are:
T. Rowe Price Small-Cap Stock Fund wrote: http://www.sec.gov/Archives/edgar/data/ ... /srscs.htm
Relative to large-caps, small-caps still trade at a P/E premium ... above historic levels. Small-caps tend to trade at a discount to large, well-established companies due to less liquidity and greater perceived risk. They undoubtedly will again. Currently, earnings growth for small-caps continues to lag large-cap shares, presenting us with another major headwind to outperformance.
http://www.sec.gov/Archives/edgar/data/ ... s_ncsr.htm [/size]
We believe small-caps will face a significant challenge in outperforming larger-cap shares. Estimates for 20% small-cap earnings growth in 2013 certainly appear optimistic given our economic environment. Valuations remain at premium levels on most relative measures, including the price-to-book and price-to-earnings ratios. A January 2, 2013, small-cap mutual fund performance study by Bank of America Merrill Lynch concluded that less than 40% of active small-cap funds outperformed their passive benchmarks. This surely explains the continued negative cash flow in the small-cap space, which is a clear negative for performance.
http://www.sec.gov/Archives/edgar/data/ ... _ncsrs.htm
In our view, small-caps still face significant headwinds, as valuations remain at a premium versus large-caps. Second, the Russell 2000 Index’s forward price-to-earnings (P/E) ratio of 17 is above its historical average of 15. Small-cap earnings growth would need to continue at the rate of the previous two quarters’ trend of outperformance versus larger-caps in order to sustain their current premium valuations. First-quarter earnings for small-caps were solid, but the largest-cap companies had higher earnings growth on the back of improved margins. Research conducted by Bank of America Merrill Lynch suggests that small-cap growth expectations of 13% are more reasonable than the 20% pace predicted in January. Many have suggested that a 20% growth rate would be needed to support the current premium P/E levels.
Does "Stay The Course" apply when you're on an off-kilter path?
Last edited by JoMoney on Thu Oct 31, 2013 7:07 am, edited 3 times in total.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

User avatar
nisiprius
Advisory Board
Posts: 39444
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 - Where's the alpha?

Post by nisiprius » Thu Oct 31, 2013 6:46 am

An interesting thing: Googling on DFA "patient trading" I get the impression that DFA itself is extremely cautious about making any claims in writing. It is mostly other people making claims on behalf of DFA. Thus, Daniel R. Solin, who is "director of investor advocacy for The BAM ALLIANCE and a wealth advisor with Buckingham Asset Management," writes, in The Smartest 401(k) Book You'll Ever Read
While a traditional small-cap index fund that tracks the Russell 2000 index has to kick out its small-cap stocks when these little guys outgrow the benchmark, DFA creates its own construction and trading rules that allow for patient trading strategies in order to achieve lower costs and increase returns for its shareholders.
If that isn't a claim of alpha, I don't know what it is. DFA's (very low) expense ratios are, nevertheless, higher than Vanguard's, so it better be a claim of alpha. It would be pretty lame to describe a strategy that incurs a higher expense ratio as a strategy that "achieving lower costs" unless there is alpha in excess of added costs.
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.

rkhusky
Posts: 7565
Joined: Thu Aug 18, 2011 8:09 pm

Re: DFA - Where's the alpha?

Post by rkhusky » Thu Oct 31, 2013 7:13 am

Jebediah wrote:True, but a significantly negative past alpha across the board with DFA funds would definitely be of interest given how hard their alpha-adding capabilities (patient trading, securities lending, IPO screens, etc) have been sold here.

feature request for pvguy: add alpha x 12 to rolling factor charts, use second y axis?
Another feature request: add a rolling error plot. I found that the error for VISVX (Vanguard Small Value) was much larger in the first few years of the fund (40%) than in the last few years (10%).

larryswedroe
Posts: 16002
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: DFA - Where's the alpha?

Post by larryswedroe » Thu Oct 31, 2013 7:54 am

Few thoughts
First as you know all funds have expense ratios. So if you can be generating "alpha" relative to the benchmark but it won't show up in the regressions until it exceeds the expense ratio. Also note that DFA screens out REITS and regulated utilities and that can have significant impact on "alpha" calculation. So you would need to look at the performance of those sectors/industries during the period measured.
Second, DFA did not begin using MOM screens until AFTER 2002.
Third, those who 'mock" patient trading simply have no clue about trading costs. But patient trading while can make some significant difference in micro cap type stocks won't make much difference in any large cap type fund. And the screens have definitely added value as well. As perfect example they have screened out the extreme small growth stocks and their performance has been very poor. Note that the ability to add value in patient trading has come down with the advent of HFT and algo trading (though it's still there) as amount of block trading done now is less.
Fourth, you can look at the results of the fund vs the benchmark and other asset class funds over the periods since inception and see the differences--It's the exposure to the loading factors and the other construction rules that matter and who up in the results vs. the benchmarks.

So I just ran two of the funds to see actual results. The longest running fund is the DFA micro fund which was for a long time a CRSP 9-10 fund and that is still a reasonable benchmark. 1/82- 8/13 Fund returned 12.15 vs 11.59 for the benchmark. And that includes of course the expense ratio which was higher but now 52 bp. (I ran similar data for the Small fund which is a 6-10 fund and it underperformed its benchmark by about 70bp for the full period, with an expense ratio of now 37 but was higher).

The small value fund since inception returned 12.33 and beat the MSCI SV index by 1.3% a year and exactly matched the FF small value ex utilities index--so fully covered expense ratios


I would note that the core funds clearly add value by reducing trading costs internally and reducing turnover (estimate about 12bp in trading costs) which improves tax efficiency and importantly for taxable accounts eliminates the need to rebalance between the asset classes, saving K gains.

I would also add that the factor loads are the most important issue, with the other factors (like focus on trading costs and screens adding some value).

Oh yes, and any value type fund will have negative exposure to the MOM factor, which if MOM was positive will then show up in a three factor regression as negative alpha. Remember alpha is the unexplained return of the portfolio (the return not explained by factor exposure). The deeper the value loading the more negative the MOM will be unless you screen for it (and even if you do about best you can get to is close to zero or you lose value exposure)
Hope that is helpful

Larry

rkhusky
Posts: 7565
Joined: Thu Aug 18, 2011 8:09 pm

Re: DFA - Where's the alpha?

Post by rkhusky » Thu Oct 31, 2013 9:11 am

larryswedroe wrote: Remember alpha is the unexplained return of the portfolio (the return not explained by factor exposure).
Just to clarify, even with the alpha parameter, there is still unexplained return when using the FF model(s). No model is perfect and there is always error between the predictions and reality. The 3-factor FF model has roughly a 20% RMS error in explaining monthly returns of the FF portfolios (roughly corresponding to r2=0.96 in these cases).

User avatar
camontgo
Posts: 324
Joined: Tue Oct 27, 2009 1:46 pm
Location: Chicago, IL
Contact:

Re: DFA - Where's the alpha?

Post by camontgo » Thu Oct 31, 2013 9:46 am

JoMoney wrote:
Eugene Fama wrote: http://www.dfaus.com/2009/05/the-new-indexing.html
This might seem like "sector betting" to traditional indexers like Vanguard founder John Bogle, who still believe that market risk primarily determines performance and that small stocks and value stocks aren't separate sources of risk and return. The academic community is arriving at a different consensus, one that recognizes multiple independent risks. Investors might even have natural combinations of the different risk exposures that best suit their individual time horizons and preferences. As long as the portfolios they use to gain these exposures are index funds, and as long as the exposures are consistent and not timed to predict markets, this sort of portfolio structuring is not a "sector bet"—it's the new face of indexing.
Those comments are from Gene Fama, Jr....not the Gene Fama.
nisiprius wrote:
While a traditional small-cap index fund that tracks the Russell 2000 index has to kick out its small-cap stocks when these little guys outgrow the benchmark, DFA creates its own construction and trading rules that allow for patient trading strategies in order to achieve lower costs and increase returns for its shareholders.
If that isn't a claim of alpha, I don't know what it is. DFA's (very low) expense ratios are, nevertheless, higher than Vanguard's, so it better be a claim of alpha. It would be pretty lame to describe a strategy that incurs a higher expense ratio as a strategy that "achieving lower costs" unless there is alpha in excess of added costs.
Costs include trading costs as well as expense ratios. Bid-ask spreads become quite high if you reach for high SMB loading. This results in negative alpha...even if the expense ratio is low. So, reducing the bid-ask spread through patient trading isn't necessarily a claim of positive alpha. I would interpret this as saying they can get a higher factor loading without lagging the FF3F benchmark....something that Russell 2000 tracking indexes have not done consistently.

EDIT:

Here is a comparison of DFA Small Value and Vanguard Small Value. DFA's alpha is 0.02% worse on a monthly basis (close to the expense ratio difference..though these numbers fluctuate based on the time period). However, DFA does get a higher SMB loading. All else equal I'd pick the DFA fund as a SCV option. However, this analysis doesn't consider the additional cost of a DFA advisor. I personally stick with non-DFA options because I don't think the DFA funds are worth it after this additional added cost and hassle.

Code: Select all

DFSVX Expense Ratio: 0.52%
VISVX Expense Ratio: 0.24%
                                     RMRF    SMB     HML      Alpha     R^2
DFSVX 	10/01/2003 	09/30/2013 	1.07 	0.91 	0.43 	-0.05% 	0.98
VISVX 	10/01/2003 	09/30/2013 	1.01 	0.61 	0.42 	-0.03% 	0.97
Last edited by camontgo on Thu Oct 31, 2013 10:20 am, edited 2 times in total.
"Essentially, all models are wrong, but some are useful." - George E. P Box

berntson
Posts: 1366
Joined: Mon Oct 29, 2012 12:10 pm

Re: DFA - Where's the alpha?

Post by berntson » Thu Oct 31, 2013 9:50 am

Blue wrote: This seems to suggest, there is no measurable benefit to DFA funds beyond factor loads?
I agree. DFA has higher fees, and it is a non-trivial accomplishment to pay for those fees with added value. But that still does not leave investors with any more money in their pockets. I don't care if a fund generating alpha that is going to the managers.

Larry could be right. Maybe DFA is generating "hidden" alpha that is being masked by market conditions or other features of their funds (i.e. excluding REITs or Utilities or whatnot). If so, the positive alpha will likely show up in the fund data in the future. Until then, reasonable investors have no reason to pay a premium for DFA funds over and above whatever they should pay for additional factor loadings. No one should be paying for hidden alpha.

Take-home for investors: Ignore both DFA's extra fees and its superior trading practices and focus on loadings. Those factors just cancel out. But unless you need especially extreme loadings for your portfolio, you can achieve those loadings with publicly available ETFs and mutual funds.

User avatar
Rick Ferri
Posts: 8935
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: DFA - Where's the alpha?

Post by Rick Ferri » Thu Oct 31, 2013 9:56 am

berntson wrote:Take-home for investors: Ignore both DFA's extra fees and its superior trading practices and focus on loadings. Those factors just cancel out. But unless you need especially extreme loadings for your portfolio, you can achieve those loadings with publicly available ETFs and mutual funds.
This is a true statement - and I use a few DFA funds!

It remains to be seen whether the unique way DFA funds are structured and traded will overcome their higher cost. I believe their momentum and new profitably screens will add value - but will it be enough? Perhaps, but there is also the cost of hiring an adviser. There isn't any way the value added from DFA funds will make up the average adviser fee, so the real take-home for investors is don't go to an adviser just to gain access to DFA funds.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.

vesalius
Posts: 794
Joined: Tue Jul 13, 2010 7:00 pm
Location: Texas

Re: DFA - Where's the alpha?

Post by vesalius » Thu Oct 31, 2013 10:32 am

Rick Ferri wrote:
berntson wrote:Take-home for investors: Ignore both DFA's extra fees and its superior trading practices and focus on loadings. Those factors just cancel out. But unless you need especially extreme loadings for your portfolio, you can achieve those loadings with publicly available ETFs and mutual funds.
This is a true statement - and I use a few DFA funds!

It remains to be seen whether the unique way DFA funds are structured and traded will overcome their higher cost. I believe their momentum and new profitably screens will add value - but will it be enough? Perhaps, but there is also the cost of hiring an adviser. There isn't any way the value added from DFA funds will make up the average adviser fee, so the real take-home for investors is don't go to an adviser just to gain access to DFA funds.

Rick Ferri
Not that he needs it in any way, but I second Rick's last sentence. If I am already paying advisor fees because I chose to go with an advisor for myriad other valid reasons then I would be happy to be in DFA funds, but I would not waste the money on advisor fees simply to gain access to DFA. It simply isn't worth it.

Random Walker
Posts: 4188
Joined: Fri Feb 23, 2007 8:21 pm

Re: DFA - Where's the alpha?

Post by Random Walker » Thu Oct 31, 2013 10:40 am

I'm curious if people feel the advisor/DFA versus DIY/VG decision is at all affected by whether the assets to be invested are predominantly in taxable or tax deferred accounts. To me, it seems the more an individual interested in slicing and dicing is in taxable accounts the more one might consider the advisor/DFA route.

Dave

User avatar
nisiprius
Advisory Board
Posts: 39444
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 - Where's the alpha?

Post by nisiprius » Thu Oct 31, 2013 11:18 am

berntson wrote:Larry could be right. Maybe DFA is generating "hidden" alpha that is being masked by market conditions or other features of their funds (i.e. excluding REITs or Utilities or whatnot). If so, the positive alpha will likely show up in the fund data in the future. Until then, reasonable investors have no reason to pay a premium for DFA funds over and above whatever they should pay for additional factor loadings. No one should be paying for hidden alpha.
To be perfectly fair, you probably can't time the market for hidden alpha. If you are absolutely sure that the hidden alpha is really there, you might feel sure that someday it will be unveiled--you just don't know when. If you wait before committing, you may miss it. The way to be sure of seeing it personally is to pay for it well in advance and just sit there watching, for decades if necessary, until it does appear.

(Ooooooh! Hidden alpha! I LOVE your coinage! Hidden alpha, hidden alpha, hidden alpha!
-->Alpha alpha alpha alpha alpha alpha alpha alpha alpha alpha<--
-->α α α α α α α α α α α α α α α α α α α α α α α α α α α α α α α α<--
It's the greatest thing since sliced beta!)
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
kenyan
Posts: 2986
Joined: Thu Jan 13, 2011 12:16 am

Re: DFA - Where's the alpha?

Post by kenyan » Thu Oct 31, 2013 11:40 am

"Alpha" that goes to the fund managers' pockets is entirely unimpressive to me, and is common enough in actively managed funds. I would consider using DFA funds (where available) if I really wanted to tweak my factor loadings. I don't have that desire currently; the Vanguard funds are doing a good enough job.
Retirement investing is a marathon.

berntson
Posts: 1366
Joined: Mon Oct 29, 2012 12:10 pm

Re: DFA - Where's the alpha?

Post by berntson » Thu Oct 31, 2013 11:42 am

[double post]
Last edited by berntson on Thu Oct 31, 2013 11:52 am, edited 1 time in total.

berntson
Posts: 1366
Joined: Mon Oct 29, 2012 12:10 pm

Re: DFA - Where's the alpha?

Post by berntson » Thu Oct 31, 2013 11:50 am

Random Walker wrote:I'm curious if people feel the advisor/DFA versus DIY/VG decision is at all affected by whether the assets to be invested are predominantly in taxable or tax deferred accounts. To me, it seems the more an individual interested in slicing and dicing is in taxable accounts the more one might consider the advisor/DFA route.
I doubt it makes much difference. On the one hand, DFA can use tax management to reduce turnover and lower potential capital gains. On the other hand, Vanguard has an ETF share class that it can use to distribute capital gains more efficiently. These features seem to be a wash--take DFSVX (DFA small value) and VBR (Vanguard small value) for instance. Morningstar reports that they had tax ratios of .54 and .53 respectively over the last five years (Though over the last year, DFSVX had a tax ratio of 1.36 and VBR a tax ratio of .53. I'm not sure what happened there.).
nisiprius wrote: (Ooooooh! Hidden alpha! I LOVE your coinage! Hidden alpha, hidden alpha, hidden alpha!
-->Alpha alpha alpha alpha alpha alpha alpha alpha alpha alpha<--
-->α α α α α α α α α α α α α α α α α α α α α α α α α α α α α α α α<--
It's the greatest thing since sliced beta!)
Ha! We should start a fund company and make millions. Hidden alpha is clearly the next big thing.

vesalius
Posts: 794
Joined: Tue Jul 13, 2010 7:00 pm
Location: Texas

Re: DFA - Where's the alpha?

Post by vesalius » Thu Oct 31, 2013 11:50 am

berntson wrote:
Random Walker wrote:I'm curious if people feel the advisor/DFA versus DIY/VG decision is at all affected by whether the assets to be invested are predominantly in taxable or tax deferred accounts. To me, it seems the more an individual interested in slicing and dicing is in taxable accounts the more one might consider the advisor/DFA route.
I doubt it makes much difference. On the one hand, DFA can use tax management to reduce turnover lower potential capital gains. On the other hand, Vanguard has an ETF share class that it can use to distribute capital gains more efficiently. These features seem to be a wash--take DFSVX (DFA small value) and VBR (Vanguard small value) for instance. Morningstar reports that they had tax ratios of .54 and .53 respectively over the last five years.
and if you are intersted in more concentrated factor loads than VG offers several other ETFs offer that with similar tax efficiency to both DFA and VG, at least per morningstar. I would still side with DFA not worth it unless you have chosen to go with an advisor for any other reason.

User avatar
1210sda
Posts: 1525
Joined: Wed Feb 28, 2007 8:31 am

Re: DFA - Where's the alpha?

Post by 1210sda » Thu Oct 31, 2013 12:53 pm

This is the best advisor/DFA versus DIY/VG discussion that I've seen in a long time. It tells it like it is....
And, Rick's comments about "the real take-home for investors is don't go to an adviser just to gain access to DFA funds" is spot on.

Thanks Blue.

1210

Jebediah
Posts: 579
Joined: Tue Aug 28, 2012 9:19 pm
Location: Austin TX

Re: DFA - Where's the alpha?

Post by Jebediah » Thu Oct 31, 2013 1:08 pm

camontgo wrote:All else equal I'd pick the DFA fund as a SCV option. However, this analysis doesn't consider the additional cost of a DFA advisor. I personally stick with non-DFA options because I don't think the DFA funds are worth it after this additional added cost and hassle.
vesalius wrote: I would still side with DFA not worth it unless you have chosen to go with an advisor for any other reason.
Rick Ferri wrote:There isn't any way the value added from DFA funds will make up the average adviser fee, so the real take-home for investors is don't go to an adviser just to gain access to DFA funds.
Rick Ferri
This conventional wisdom about adviser fees is not necessarily true. One can hire a "hands-off" advisor with DFA access for 1K a year. That's 20 bp on a 500K portfolio, 10 bp on a 1M portfolio.

Code Commit
Posts: 138
Joined: Mon Mar 04, 2013 11:42 am

Re: DFA - Where's the alpha?

Post by Code Commit » Thu Oct 31, 2013 1:33 pm

Rick Ferri wrote:..., so the real take-home for investors is don't go to an adviser just to gain access to DFA funds.
+1.

I have always wondered where this advisor requirement comes from and why it is still there. Is it DFA that has always insisted on restricting access to their funds through approved advisors? But, then what about DFA funds that get offered in few 401k and 529 plans, without any advisor? Or, is it the advisors that lobby DFA to keep this requirement? (I am just wondering. Not really trying to insinuate anyone).

Disclosure: I use DFA small value, large value and international value funds in Utah 529 (without advisor). For everything else, it is Vanguard funds/ETFs.

User avatar
camontgo
Posts: 324
Joined: Tue Oct 27, 2009 1:46 pm
Location: Chicago, IL
Contact:

Re: DFA - Where's the alpha?

Post by camontgo » Thu Oct 31, 2013 1:49 pm

Jebediah wrote:This conventional wisdom about adviser fees is not necessarily true. One can hire a "hands-off" advisor with DFA access for 1K a year. That's 20 bp on a 500K portfolio, 10 bp on a 1M portfolio.
Yes, it can be a small cost on a large portfolio. However, personally, I believe there is only a portion of my asset allocation where the DFA alternative to the low-cost ETF might provide a modest benefit.

For example, say I had a $500,000 equity portfolio with 50% VTI, 25% VXUS, and 25% IJS. The only thing I would ask of the hands-off advisor would be to move the $125,000 in IJS to DFSVX....and I'd consider the $1000 to be an 80 bp cost increase on that portion of the portfolio. Add that to the already higher DFA expense ratio, and I don't think its worth it....even though I do believe the DFA funds, in some categories, are a bit better.
Last edited by camontgo on Thu Oct 31, 2013 2:53 pm, edited 2 times in total.
"Essentially, all models are wrong, but some are useful." - George E. P Box

User avatar
camontgo
Posts: 324
Joined: Tue Oct 27, 2009 1:46 pm
Location: Chicago, IL
Contact:

Re: DFA - Where's the alpha?

Post by camontgo » Thu Oct 31, 2013 2:08 pm

Code Commit wrote:I have always wondered where this advisor requirement comes from and why it is still there. Is it DFA that has always insisted on restricting access to their funds through approved advisors? But, then what about DFA funds that get offered in few 401k and 529 plans, without any advisor? Or, is it the advisors that lobby DFA to keep this requirement? (I am just wondering. Not really trying to insinuate anyone).
IMO, DFA's biggest added value is their ability to manage liquidity in portfolios of small stocks. These stocks typically have high trading costs due to the large bid-ask spreads for big blocks of stock.

DFA can get a higher factor loading without most of the negative alpha that the trading costs in this area create for other funds by managing their liquidity and trading very carefully. I think requiring investors to go through a DFA trained advisor limits "hot money" inflows and outflows that can be very costly when you specialize in illiquid sectors of the market. 401ks and 529s are also likely to be relatively stable sources of funds. I think this is a genuine advantage for DFA investors...but, only if the additional advisor costs can be kept very low.
"Essentially, all models are wrong, but some are useful." - George E. P Box

Jebediah
Posts: 579
Joined: Tue Aug 28, 2012 9:19 pm
Location: Austin TX

Re: DFA - Where's the alpha?

Post by Jebediah » Thu Oct 31, 2013 2:45 pm

Blue wrote:Funds I used are (using inception dates through Sept 30, 2013 on 3 Factor model against Fama-French benchmark). Note, I had a typo on DFVEX/DFEVX in my OP (now fixed). Limited my "back of the envelope" search to US only funds.

DFA US Targeted Value, DFFVX, 12/1/2000, alpha -0.30%
DFA US Core Equity I, DFEOX, 12/1/2005, alpha -0.06%
DFA US Vector Equity, DFVEX, 12/1/2006, alpha -0.13%
DFA US T.A. Core Equity II, DFTCX, 12/3/2007, alpha -0.11%
DFA US Small Cap, DFSTX, 7/1/1996, alpha -0.03%
DFA Tax Managed US Equity Portfolio, DTMEX, 11/1/2001, alpha -0.09%

Vanguard Total Stock Market, VTSMX, 7/1/1996, alpha -0.03%
Vanguard SP 500 Admiral Shares, VFIAX, 12/1/2000, -0.05%
Vanguard Dividend Appreciation Index, VDAIX, 6/1/2006, +0.08%
Vanguard Mid-Cap Index Admiral Shares, VIMAX, 12/3/2001, +0.06%
Vanguard Small Cap Value Index, VISVX, 6/1/1998, +0.00%
I don't think this analysis is accurate enough to draw any conclusions, for multiple reasons:

- mismatched time periods
- low R^2 values in the Vanguard funds
- should use the U.S. Research dataset
- t-stats on the alphas are too low to indicate significance
- different time periods yield very different results (as pvguy has pointed out)

For example, here is DFA Small Value vs Vanguard Small Value since 6/1/1998 (inception of VISVX)

Code: Select all

Ticker    Start    Market  Size    Value    Mom      Alpha    Alpha t-stat    R2
VISVX    6/1/98    0.96    0.59    0.61     -0.10   -0.02%   -0.17            0.94
DFSVX    6/1/98    1.01    0.82    0.66     -0.07    0.02%    0.20            0.97

User avatar
nisiprius
Advisory Board
Posts: 39444
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 - Where's the alpha?

Post by nisiprius » Thu Oct 31, 2013 3:14 pm

Jebediah wrote:This conventional wisdom about adviser fees is not necessarily true. One can hire a "hands-off" advisor with DFA access for 1K a year. That's 20 bp on a 500K portfolio, 10 bp on a 1M portfolio.
Really? I thought the whole point of DFA working through advisors was that they were not supposed to be hands-off, but were supposed to educate the clients to make sure that they understand the approach, and coach them into holding the funds for the long term.
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.

Jebediah
Posts: 579
Joined: Tue Aug 28, 2012 9:19 pm
Location: Austin TX

Re: DFA - Where's the alpha?

Post by Jebediah » Thu Oct 31, 2013 3:18 pm

larryswedroe wrote: I would note that the core funds clearly add value by reducing trading costs internally and reducing turnover (estimate about 12bp in trading costs) which improves tax efficiency and importantly for taxable accounts eliminates the need to rebalance between the asset classes, saving K gains.

Larry
Larry, I disagree that DFA's Core funds offer any advantage on the domestic side. Here's an example:

DFA US Core 2 Tax Managed (DFTCX) vs 50/50 Vanguard SCV (VBR) + Vanguard TSM (VTI)

Since inception of DFTCX (12/03/2007):

DFTCX
Market: 1.03
Size: 0.25
Value: 0.15
Mom: -0.02
R2: 1.0
5-Year Tax adjusted return: 10.63%
5-Year "Tax Cost Ratio": 0.58

50/50 VBR+VTI
Market: 1.0
Size: 0.27
Value: 0.20
Mom: -0.04
R2: 0.98
5-Year Tax adjusted return: 10.59%
5-Year "Tax Cost Ratio": 0.46

-

larryswedroe
Posts: 16002
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: DFA - Where's the alpha?

Post by larryswedroe » Thu Oct 31, 2013 3:24 pm

jeb
You can disagree all you want but in taxable accounts, there is a big advantage of a core fund in that YOU don't have to rebalance between asset classes with YOUR money and pay taxes when doing so. The fund rebalances with OTHER PEOPLE's money.
And obviously core funds save on transactions costs and some taxes by reducing need to trade as stocks migrate from one asset class to another.
Best
Larry

larryswedroe
Posts: 16002
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: DFA - Where's the alpha?

Post by larryswedroe » Thu Oct 31, 2013 3:26 pm

Forgot to mention another point that is important to remember about regression analysis. The funds are long only and the regressions are based on long-short factors--they are uninvestable.
Larry

Jebediah
Posts: 579
Joined: Tue Aug 28, 2012 9:19 pm
Location: Austin TX

Re: DFA - Where's the alpha?

Post by Jebediah » Thu Oct 31, 2013 3:29 pm

nisiprius wrote:
Jebediah wrote:This conventional wisdom about adviser fees is not necessarily true. One can hire a "hands-off" advisor with DFA access for 1K a year. That's 20 bp on a 500K portfolio, 10 bp on a 1M portfolio.
Really? I thought the whole point of DFA working through advisors was that they were not supposed to be hands-off, but were supposed to educate the clients to make sure that they understand the approach, and coach them into holding the funds for the long term.
Probably so, but nonetheless the small-cost, access-only variety is out there. Which is a good thing, IMO. Even so, there doesn't seem to be much, if any, "DFA advantage" on the domestic side. The international side is probably another story. But as Camontgo points out, the cost goes up if you're only using DFA on a subset of the portfolio.

Jebediah
Posts: 579
Joined: Tue Aug 28, 2012 9:19 pm
Location: Austin TX

Re: DFA - Where's the alpha?

Post by Jebediah » Thu Oct 31, 2013 3:38 pm

larryswedroe wrote:jeb
You can disagree all you want but in taxable accounts, there is a big advantage of a core fund in that YOU don't have to rebalance between asset classes with YOUR money and pay taxes when doing so. The fund rebalances with OTHER PEOPLE's money.
And obviously core funds save on transactions costs and some taxes by reducing need to trade as stocks migrate from one asset class to another.
Best
Larry
On the first point, I seriously doubt this is a "big advantage". How often do you really need to rebalance between SCV and TSM? Most of one's rebalancing will be in the form of selling stock to buy bonds anyway, so whether in an up year you're selling say 20% of your DFTCX allocation vs 8% of your VTI + 12% of your VBR really amounts to very little difference.

On the second point, it doesn't show up in the numbers. VBR+VTI appears to be just as tax efficient as DFA's tax-managed core fund. According to M*, even more so.

larryswedroe
Posts: 16002
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: DFA - Where's the alpha?

Post by larryswedroe » Thu Oct 31, 2013 4:44 pm

Jeb
but when SV outperforms TSM as it does most years you have to sell SV to buy TSM while the core fund doesn't. That can be over time significant advantage. It's the same principle of why Vanguard combined EM and international and DFA did also before that.

Larry

Jebediah
Posts: 579
Joined: Tue Aug 28, 2012 9:19 pm
Location: Austin TX

Re: DFA - Where's the alpha?

Post by Jebediah » Thu Oct 31, 2013 5:06 pm

Seems to me that there would be very few years where the VBR+VTI investor found themselves having to rebalance SCV in which the DFTCX investor did not have to also rebalance.

But speaking of International funds-- Are DFA and Vanguard really rebalancing EM and Developed in their Ex-US funds according to set allocations? I thought due to cap weighting they just let them be in whatever ratio the market takes them. For instance, If VGTSX starts at 20% EM and then EM loses 50% one year, then doesn't VGTSX just carry on with 10% EM instead of rebalancing it back?

larryswedroe
Posts: 16002
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: DFA - Where's the alpha?

Post by larryswedroe » Thu Oct 31, 2013 5:21 pm

Jeb
That's the point, they don't have to rebalance between them when you combine them in one fund
Larry

Jebediah
Posts: 579
Joined: Tue Aug 28, 2012 9:19 pm
Location: Austin TX

Re: DFA - Where's the alpha?

Post by Jebediah » Thu Oct 31, 2013 5:30 pm

I guess my question wasn't clear. Do these International Core funds rebalance developed and EM internally to a set allocation (such as 80/20) or do they just let the developed:EM ratio within the fund drift to wherever the market takes it?

Bradley
Posts: 486
Joined: Tue May 01, 2007 2:41 pm

Re: DFA - Where's the alpha?

Post by Bradley » Thu Oct 31, 2013 6:31 pm

larryswedroe wrote:Jeb
but when SV outperforms TSM as it does most years you have to sell SV to buy TSM while the core fund doesn't. That can be over time significant advantage.
Larry

The advantage could go either way depending on the rebalancing method/time and the random nature returns. If you have to rebalance every year your rebalancing bands may not be wide enough and/or your frequency may be too often. Bill Bernstein shows the advantage went to 4 year rebalancing vs shorter periods. His findings over the 28 year period studied can lead one to conclude that monthly/quarterly and yearly rebalancing may not be nearly as beneficial as some appear to believe. Below are some of his findings. Other studies may find small benefits to more frequent rebalancing. Anyways, IMO the additional costs incurred in owning DFA funds eat away at the benefits of owning DFA core funds.



“rebalancing amounts to a bet that last year's above/below average return will reverse next year. If this is not the case, then there is no sense in rebalancing. There is overwhelming evidence that there is short-term persistence in asset class returns, so it is a good idea not to be too hasty pulling the trigger.” -------------- Bill Bernstein

“There are small rewards to increasing one's rebalancing frequency from quarterly up to several years, but this comes at the price of increased portfolio risk.
You makes your choice and you takes your chances, but don't sweat this one too much. The returns differences among various rebalancing strategies are quite small in the long run.”------------Bill Bernstein


http://www.efficientfrontier.com/ef/100/rebal100.htm
You can sum up any active fund manager’s presentation at an investor conference in one sentence: “We’re doing well, all things considered.”

User avatar
Robert T
Posts: 2596
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: DFA - Where's the alpha?

Post by Robert T » Fri Nov 01, 2013 4:28 am

.
pvguy – nice factor analysis tool – saves a lot of time doing FF3F analyses!

On DFA alpha:

1. I had set up two sets of portfolios in M*, with approximately the same ex-ante factor loads. These portfolios now have close to 5 year returns.
  • Annualized return since inception: 11-06-08
    ETF = 14.4%
    DFA TM =14.3%

    Annualized return since: 12-31-09
    RAFI = 15.2%
    DFA Vector/Core = 15.3%

    So far, no 1-3% in alpha beyond factor loads.
2. As Don Keim’s earlier paper on “An analysis of mutual fund design – the case of investing in small-cap stocks” – DFA screen rules and trading strategies largely act to change the factor loads, not add alpha.
  • For example - statistically significant effects on factor load changes (from the paper):
    Trading strategy – significantly increased the size load
    8th decile hold ranges – significantly reduced the size load
    IPO exclusion – significantly increased the value load
    Low price/size exclusion – significantly reduced the size and value load
3. My view is DFA do offer some advantages – perhaps closer to 0.2% - 0.3% (my earlier estimate) than the sometimes claimed 1-3%. If you do a rolling regression on DFA vector, for example, it shows a more consistent value load over time than some of the other funds. Will the momentum and profitability screens/sorts add more alpha - time will tell. Is this enough advantage to pay an advisor fee for access? Personally I don’t think so. An advisor fee is certain, the DFA advantage is less certain. A 1% advisor fee would account for 2/3 of my 1.5% in expected above market returns (for a greater risk [size-value] tilt). Pay an advisor for other things, not DFA access.

FWIW – I don’t use DFA funds for retirement investments, but do use them for 529 (Utah - lower cost, and more able to achieve factor load targets than with other plans).

Robert
.

larryswedroe
Posts: 16002
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: DFA - Where's the alpha?

Post by larryswedroe » Fri Nov 01, 2013 8:00 am

Few thoughts
First, as Robert noted much of what DFA does in screens changes factor loads--like cutting out extreme small growth stocks would load more on profitability factor.
Second, the trading and other things like securities loading--I completely agree with Robert's estimate--they add value on margin, it's the factor loadings that matter--the fund construction rules. And what is important is that the higher the loadings the less equity risk you need to get the same expected return--which cuts tail risks. Or you can earn higher returns with same equity allocation. It's always been about loadings as the big things and the other things as value adds, on the margin benefits, but benefits that in long term matter.
Third, what is simply amazing to me is that people dispute the benefits of a core holding--that is likely arguing against owning TSM and instead own the components in the same proportion-which of course no one would do. The benefits of TSM are that migration doesn't lead to trading and you don't have to rebalance. Well those are the same benefits of a core fund.
Best wishes
Larry

Bradley
Posts: 486
Joined: Tue May 01, 2007 2:41 pm

Re: DFA - Where's the alpha?

Post by Bradley » Fri Nov 01, 2013 10:29 am

I don't believe anyone has disputed the benefits of DFA’s core funds only the idea that they have significant advantage or certain advantage over separate non core holdings which are rebalanced according to an individual IPS. The random nature of returns, correlation and RTM can and do allow for outperformance of either core or separate holdings depending on the time period measured. We have all benefitted from the research of FF and strategies implemented by DFA much in the same way the auto industry benefitted from Henry Ford and the assembly line. Once discovered as a value added strategy other car companies copied and improved the process. Any alpha that may be created by DFA’s core fund is eaten away by the increased costs ( 4 to 6 times Vanguard’s TSM fund VTI ) of those funds. Once you add in advisory fees, IMO, the disciplined, informed investor using a well constructed diversified portfolio has every expectation of matching and/or exceeding any perceived “alpha” from a managed DFA core or non core portfolio.
You can sum up any active fund manager’s presentation at an investor conference in one sentence: “We’re doing well, all things considered.”

larryswedroe
Posts: 16002
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: DFA - Where's the alpha?

Post by larryswedroe » Fri Nov 01, 2013 11:48 am

Just gotta love it when people make statements like DFA core funds cost 4-6x as much as say VTI. Yes it's true but that is irrelevant. It's not like VTI costs 20bp and DFA costs 80-120bp.

What matters is the absolute differential. DFA core funds cost 22bp. Now a very reasonable estimate of the reduced trading costs of less migration impact (stocks leaving asset class) is say 12bp, That brings the net cost to 10bp. That's vs 5bp for VTI. Now screens and MOM and other issues like profitability tilt and patient trading--think safe to say should address that. And then you have the rebalancing issue for taxable accounts.
There's one right way to look at things

Best wishes
Larry

smpatel
Posts: 155
Joined: Mon Jan 16, 2012 10:30 am

Re: DFA - Where's the alpha?

Post by smpatel » Fri Nov 01, 2013 12:12 pm

We should not forget advisory fee for access to DFA funds on top of the expense ratio. This is a risk vs reward with best value story. It does appear that the value is in DIY investor using publicly accessible low cost ETF/Vanguard funds.

larryswedroe
Posts: 16002
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: DFA - Where's the alpha?

Post by larryswedroe » Fri Nov 01, 2013 1:17 pm

smpatel
Of course one should not but then you have to assign the value (if any) an advisor brings. So unless you think it's zero, that's not the whole story.
Larry

User avatar
Epsilon Delta
Posts: 8090
Joined: Thu Apr 28, 2011 7:00 pm

Re: DFA - Where's the alpha?

Post by Epsilon Delta » Fri Nov 01, 2013 1:24 pm

larryswedroe wrote:smpatel
Of course one should not but then you have to assign the value (if any) an advisor brings. So unless you think it's zero, that's not the whole story.
Larry
Do we have to stop at zero? Some advisors are dedicated to exploring the rest of the number line.

User avatar
matjen
Posts: 2128
Joined: Sun Nov 20, 2011 11:30 pm

Re: DFA - Where's the alpha?

Post by matjen » Fri Nov 01, 2013 2:00 pm

Robert T wrote: 3. My view is DFA do offer some advantages – perhaps closer to 0.2% - 0.3% (my earlier estimate) than the sometimes claimed 1-3%. If you do a rolling regression on DFA vector, for example, it shows a more consistent value load over time than some of the other funds. Will the momentum and profitability screens/sorts add more alpha - time will tell. Is this enough advantage to pay an advisor fee for access? Personally I don’t think so. An advisor fee is certain, the DFA advantage is less certain. A 1% advisor fee would account for 2/3 of my 1.5% in expected above market returns (for a greater risk [size-value] tilt). Pay an advisor for other things, not DFA access.

FWIW – I don’t use DFA funds for retirement investments, but do use them for 529 (Utah - lower cost, and more able to achieve factor load targets than with other plans).

Robert
.
Thanks to Robert's earlier posts and obviously Larry's, I have started to move part of our portfolio into DFA. Do I expect to do 3% better than an all Vanguard portfolio...no, but I don't expect to do any worse and hope the newer factors add a bit more and will probably migrate to more of a "Larry Portfolio."

Every time we have one of these threads we get the same tired old and patently false argument about how you have to pay 1% or close to that for an advisor, etc. In this thread it is stated even after folks said they are paying only 1K a year for access to the funds. To be 100% clear, you can get access to the funds (not in a 529 or 401k) for 1k per year PER HOUSEHOLD. If you have two highish earners who are bogleheads then the accounts can get fairly large and that 1k is absolutely minimal (as was pointed out earlier). I will also add that if you have a self-directed PCRA option in your 401(k) account (a lot of Schwab 401(k)s have this feature) then you can get into the DFA family without any tax consequences since they will now be part of your options. Having said all of that, if you need or want an advisor then a good and fair one is worth his/her fee just to keep you from doing stupid things.
A man is rich in proportion to the number of things he can afford to let alone.

Post Reply