"Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

"Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

Index Fund Advisors (IFA.com) has a chart on it's site comparing various Vanguard and DFA -based portfolios, from lower risk to higher risk...

The chart, labeled "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios; 16 Years (1/1/1999 - 12/31/2014)", is embedded in the middle of the following article...

"The Value of a Passive Advisor" (IFA.com/Mark Hebner, Sep 17, 2013): https://www.ifa.com/articles/value_pass ... lashID1340

Vertical axis: "Annualized Return (%)"

Horizontal axis: "Risk: Annualized Standard Deviation (%)"

The DFA-based and Vanguard-based portfolios have similar risk (SD)/return profiles at lower risk levels, but diverge noticeably at higher risk levels.

Can anyone help deconstruct and explain the data in this specific chart, which seems to be saying/implying "DFA funds are superior to Vanguard funds"?

I actually talked with Mark Hebner on the phone a few months ago ... about IFA in general and multi-factor investing specifically ... and then followed up with an e-mail asking him about the relationship between the two curves displayed in that chart.

In particular, it seemed to me that standard deviation probably wasn't the best measure of risk when you're evaluating portfolios based on DFA funds, which are premised on multi-factor asset pricing model theory. Seems to me like using beta (or betas) as the risk measure would be better, and would likely reduce the (perceived) superiority of the DFA-based portfolios over the Vanguard-based portfolios.

Mark H. didn't have time for a follow-up conversation at the time due to other commitments, and I've not tried to follow-up with him since then.

Any thoughts or comments from the Bogleheads on the chart and the gap between the DFA & Vanguard-based portfolios displayed? [I believe IFA.com also has companion charts explaining the specific content/construction of the various portfolios in the chart I asking about above.]

Thanks!
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
User avatar
nisiprius
Advisory Board
Posts: 52105
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: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by nisiprius »

Oh, boy... you opened a Frequently Opened Can Of Worms. For background, first read the Wiki article on DFA.

To add a few relevant notes:

IFA is, in fact, an advisory service: "Index Fund Advisors, Inc. (IFA) is a fee-only advisory and wealth management firm that provides risk-appropriate, returns-optimized, globally-diversified and tax-managed investment strategies with a fiduciary standard of care."

They make money by charging fees. The services they perform include recommending and managing portfolios of mutual funds.

For the most part, DFA mutual funds are available only through advisors.

DFA mutual funds are described by DFA as "passive," but they are not index funds. They do not attempt to track any index precisely.

An advisor who offers DFA funds has an incentive to present them as superior to straight index funds because--or, as tools that a skilled advisor can use to build superior portfolios--because, in IFA's own words, once you accept passive investing,
...the next logical question is, "Why should I seek the help of an advisor when I can just create a portfolio of indexes on my own using index funds at a place like Vanguard or purchase ETFs through a low cost broker like Charles Schwab?"
People who use DFA funds do so because they think they are superior to straight index funds. People who use advisors who offer DFA funds do so because they think the advisor can build a better portfolio using DFA funds than they can build for themselves using Vanguard funds.

They could be right. But it's not a slam dunk. You have to decide for yourself.

All of the "past performance" arguments--DFA funds and portfolios using them have "consistently" outperformed, etc. can be used equally well by advocates of actively managed funds. For example, what can anyone say about this chart? During the time DFA Core Equity (DFEOX, orange) has existed, a $10,000 investment in that fund earned $60 more than it would have earned in Vanguard Total Stock Market Index (VTSMX, blue), but it would have earned $3,500 more in Fidelity Contrafund than it would have in either of the two "passive" funds. Are you better off with DFA than with Vanguard? Are you better off with a good active fund than with either passive fund? Beats me. My investing life has gotten a lot calmer since I started to think in terms of "good enough" than "must pick the best."

Image

When comparing DFA and Vanguard funds, don't forget that you have to deduct the cost of the advisory fees when calculating the DFA portfolios' performance, unless you are fully convinced that the advisors completely earn their fees by providing you with other valuable services.
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.
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

nisiprius,

Thanks for your reply post. I think I understand the (general) approaches and business models of both IFA and DFA. I've actually talked on the phone with Mark Hebner and a couple of the client-facing people at DFA in Austin,

My original post and main question is specific to the single chart I referenced.

If anyone in the BH community can shed some light on the methodology used by IFA.com in that specific graph, and connect it to the underlying DFA and Vanguard funds used, I would love to know your thoughts.

I suspect the diverging lines (DFA vs. Vanguard in the IFA-generated portfolios) are driven primarily by the use of standard deviation, rather than beta/s, on the horizontal axis.

I think what I may do is forward the URL link to this post/thread and give Mark Hebner another opportunity to clarify, explain, educate me, etc.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
User avatar
TomP10
Posts: 238
Joined: Tue Mar 18, 2014 4:13 am

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by TomP10 »

A couple of comments. First, there is very little difference between DFA and Vanguard bond funds. For example, DFA's short term gov't bond fund holds the same type of bonds as Vanguard's short term gov't bond fund. So, when looking at portfolios with heavy bond allocation, the DFA and Vanguard portfolios are going to look similar.

Second, please keep in mind when looking at the IFA charts that IFA assumes you pay the 0.9% IFA advisor fee even if you build a comparable Vanguard portfolio. My guess is that many people who use Vanguard are "do it yourself" investors and therefore don't pay anything but the fund expense ratio.

So, if my view you should add 0.9% return to the Vanguard efficient frontier. That would make Vanguard portfolio superior for high bond allocations and (perhaps) slightly inferior for low bond (high equity) allocations.

Tom
"It is remarkable how much long term advantage people like us have gotten by trying to be consistently not stupid, instead of trying to be very intelligent." -- Charlie Munger
garlandwhizzer
Posts: 3562
Joined: Fri Aug 06, 2010 3:42 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by garlandwhizzer »

Note that the comparison charts show DFA and IFA portfolios NET OF FEES. IFA which authored the piece charges a fee of 0.90%/yr. for the first 500K of investment portfolio to employ their DFA based portfolios in addition to the DFA fund costs. The total cost of their portfolios is therefore well over 1%/yr. and the chart comparisons are therefore misleading. Also the comparisons are not apples to apples since the DFA and IFA funds which presumably represent all cap weights with a strong slant toward SCV are compared to the S&P 500, a large cap blend index. Historical comparisons of similar cap weights and factor weights between low cost index funds and the DFA-IFA funds would again not look so impressive. This piece is merely marketing and advertisement designed primarily to take advantage of unsophisticated investors. Misleading comparisons are the pseudo-science marketing tools of those who hawk their costly wares, driven primarily by the desire to maximize their own income rather than their clients.

The only comparisons that matter are after cost comparisons between appropriately matched asset classes and the total data on this does not in my opinion make a convincing case for active management or smart beta or factor indexing relative to a properly measured benchmark.

Garland Whizzer
SpaceCowboy
Posts: 1006
Joined: Sun Aug 12, 2012 12:35 am

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by SpaceCowboy »

I think that IFA is a pretty reputable DFA shop. I also think their fees are on the high side for DFA advisors. I'm pretty familiar with them, as I used to have an office in the same building as IFA. If you ask they will send you a huge book that basically preaches buy & hold passive index investing, which is pretty much the BH philosophy.
I'm pretty sure that the difference in the returns shown at higher equity, "risk," levels is due to DFA's equity funds having higher small-cap and value weightings than the corresponding Vanguard funds used in the analysis. I'm also slightly suspicious of the word "New" in the New IFA Portfolios, as to me this implies that they have changed the construction of the portfolios during the period and there is backtesting and not just actual results included in the analysis. Somewhere on their site, they should have the actual returns achieved by their different risk level portfolios, at least they did in the past.
Use of standard deviation is perfectly accepted financial practice in this type of analysis. It is basically showing that you can achieve a superior Sharpe ratio at the indicated risk levels using their portfolios.
User avatar
Taylor Larimore
Posts: 32839
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

Thoughts ?

Post by Taylor Larimore »

If anyone in the BH community can shed some light on the methodology used by IFA.com in that specific graph, and connect it to the underlying DFA and Vanguard funds used, I would love to know your thoughts.
Scott:

"Thoughts":

Were fees included?

"Past performance does not forecast future performance."

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
User avatar
amphora
Posts: 235
Joined: Fri Jun 26, 2015 10:44 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by amphora »

You can buy DFA funds in some 529 plans so in that case you can avoid high advisory fees.
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

Thanks for the additional thoughts and comments everyone.

On the fee structure in the IFA comparison ... as I understand the IFA chart, it assumes IFA will get paid their advisory fee in either scenario, using DFA or Vanguard products; this seems appropriate to make an apples-to-apples comparison. If the fee is added to both sides/lines, then the net effect in in the comparison is zero/neutral.

If you disallow the fee, then the comparison is basically moot, because you can not invest in both alternatives; the DIY with Vanguard and no advisory fee has no practical corollary in the DFA world.

I think, for the most part, the only way you can dine at the DFA table [at least for the good stuff] is by getting access through one of their approved front-end advisor partners.

On the variance/SD as a measure of risk... this, I think, gets to the very core of my main questions about factor-based investing. I could definitely be wrong, but my current working theory is that anyone using a factor-based approach to build a portfolio and model expected (and historical) returns on a truly risk-adjusted basis has to use a multiple-regression approach, with multiple betas and an alpha remainder, or the analysis is incomplete.

I'm doing my best to disprove this hypothesis, but I haven't seen any evidence to do that yet. I hope there is an answer out there somewhere.

Does anyone know of a source where I can get a definitive answer to this factor-based investing question on risk-adjusted returns and beta vs. standard deviation/variance measure of risk? Thanks.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
User avatar
privatefarmer
Posts: 779
Joined: Mon Sep 08, 2014 2:45 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by privatefarmer »

scottj19707 wrote:Index Fund Advisors (IFA.com) has a chart on it's site comparing various Vanguard and DFA -based portfolios, from lower risk to higher risk...

The chart, labeled "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios; 16 Years (1/1/1999 - 12/31/2014)", is embedded in the middle of the following article...

"The Value of a Passive Advisor" (IFA.com/Mark Hebner, Sep 17, 2013): https://www.ifa.com/articles/value_pass ... lashID1340

Vertical axis: "Annualized Return (%)"

Horizontal axis: "Risk: Annualized Standard Deviation (%)"

The DFA-based and Vanguard-based portfolios have similar risk (SD)/return profiles at lower risk levels, but diverge noticeably at higher risk levels.

Can anyone help deconstruct and explain the data in this specific chart, which seems to be saying/implying "DFA funds are superior to Vanguard funds"?

I actually talked with Mark Hebner on the phone a few months ago ... about IFA in general and multi-factor investing specifically ... and then followed up with an e-mail asking him about the relationship between the two curves displayed in that chart.

In particular, it seemed to me that standard deviation probably wasn't the best measure of risk when you're evaluating portfolios based on DFA funds, which are premised on multi-factor asset pricing model theory. Seems to me like using beta (or betas) as the risk measure would be better, and would likely reduce the (perceived) superiority of the DFA-based portfolios over the Vanguard-based portfolios.

Mark H. didn't have time for a follow-up conversation at the time due to other commitments, and I've not tried to follow-up with him since then.

Any thoughts or comments from the Bogleheads on the chart and the gap between the DFA & Vanguard-based portfolios displayed? [I believe IFA.com also has companion charts explaining the specific content/construction of the various portfolios in the chart I asking about above.]

Thanks!
I have been studying DFA vs Vanguard for the past several weeks now. FPL capital management will "sell" you DFA access for $1000/year, regardless of portfolio size. For me that comes out to about 30bp/year right now.

I have been comparing DFA funds to vanguard for as far back as we can go (usually about 15 years), and in ALMOST every case DFA has beaten vangaurd by ~1.5%/year. What I did was compare DFA US SCV to Vanguard US SCV, DFA US SCB to vanguard US SCB etc. It is very evident to ME that DFA has outperformed vanguard in the funds that "tilt" towards small and value. Since almost my entire portfolio is constructed of assets that are tilted to small or value this is significant. The only equity asset class that I found Vanguard kept up w/ DFA on was US small cap blend. However, apaprently NAESX (vanguards index fund) was actively managed until '94, so if you backtest starting from '94 then DFA actually beats it...

Does any of this mean anything? We could certainly find plenty of actively managed funds that have beaten the index over the last 15 years. I would not bet that they would continue to beat the index. But in basically every case, when looking at equity funds that tilt towards small/value, DFA has beaten vanguard going back as far as we can (~15-20 years in most cases. international small cap only goes back ~5 years for vanguard and DFA has beaten it w/ international small cap value. vanguard does not have a int. SCV ETF/fund).

So in my case, I already expect to pay ~20-30bp/year more in ER for DFA funds, plus ~30bp/yr to access DFA through FPL. Thats a total of ~60bp/yr over vanguard. Again, over the last 15-20 years, DFA has beaten vanguard by ~1.5%/yr on average in the "tilted" funds. So I am still debating this for myself. I am leaning towards switching to DFA just bc I do think they do a better job at 'harvesting" the small/value "premium" but whether or not they can continue that is unknown.
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by Random Walker »

It can get pretty tricky trying to do the DFA v. VG comparison. The downside of the DFA approach is certain and obvious: the increased ERs and the advisor AUM fee. The upsides for DFA are potential and much less obvious. It only captures part of the potential DFA benefit to for example compare like named SV funds. What one really needs to do is compare similar portfolios with similar factor tilts. Since DFA has stronger tilts to small and value than VG in the asset class funds, it takes less of those more expensive funds to create a portfolio with the same tilts as VG. Also for taxable investors there is the benefit of tax managed value funds for DFA, while VG has none. The DFA core funds help provide tilt while minimizing rebalancing costs. Also DFA offsets their higher ERs with securities lending. DFA also has an asset class or two that I believe VG doesn't have: international small value (which I believe is a better diversifier than EM), international REITS, CCFs. So a portfolio comprised of DFA funds has the potential to be marginally more efficient.
I am not nearly sophisticated enough to put numbers to these potential benefits of DFA, but since I am mainly invested in taxable accounts I have gone the advisor/DFA route and plunged into a few of the more esoteric asset classes. Interested in other's thoughts. I'm always reevaluating my decision because as I said, AUM fee and higher ERs of this approach are certain costs.

So the negatives of DFA approach: AUM fee, higher portfolio ER
The potential positives of DFA approach: higher return, improved portfolio efficiency, better tax management, securities lending, lower rebalancing costs, advisor helps minimize behavioral errors, other advisor benefits like tax loss harvesting on specific lot basis throughout the year.

Dave
SpaceCowboy
Posts: 1006
Joined: Sun Aug 12, 2012 12:35 am

Re: Thoughts ?

Post by SpaceCowboy »

Taylor Larimore wrote:
If anyone in the BH community can shed some light on the methodology used by IFA.com in that specific graph, and connect it to the underlying DFA and Vanguard funds used, I would love to know your thoughts.
Scott:

"Thoughts":

Were fees included?

"Past performance does not forecast future performance."

Best wishes.
Taylor
Fees were included in a special way. The chart is clearly labelled "Net of IFA and DFA fees." Now the other side is that Vanguard was penalized as being advisor bought and labelled as "Net of IFA and Vanguard fees." So, the DFA funds are being penalized for their higher expense ratio as compared to Vanguard, but not for the advisor required to access it. The performance difference is due to the larger small/value tilt in the DFA portfolio as compared to the Vanguard one.
gkaplan
Posts: 7034
Joined: Sat Mar 03, 2007 7:34 pm
Location: Portland, Oregon

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by gkaplan »

Gordon
User avatar
nedsaid
Posts: 19249
Joined: Fri Nov 23, 2012 11:33 am

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by nedsaid »

My take on this is that relative performance between Vanguard and DFA depend on market conditions.

DFA is largely built around the "small" and "value" premiums that the academics talk about. According to Larry Swedroe, DFA now includes "momentum" in their stock screens. I don't know that the academic researchers found anything new, my suspicion is that professional investors have known about this for many years. The academics have provided the research backing and have been better able to articulate those concepts.

I went to a couple of the Paul Merriman seminars where I was introduced to small/value tilting. They said that in a Large/Growth decade like the 1990's that a Vanguard portfolio would win out. In a Small/Value decade like the 2000's, DFA would do better than Vanguard.

So to take a DFA approach and to pay an advisor, you have to believe a few things. First, that the small/value premium is persistent over long periods of time. Second, that small/value tilting will give you enough of a performance premium to pay the advisor and still have excess performance above the market averages. Merriman claimed that they could outperform Vanguard by 1% even after charging their 1% management fee. Third, you have to believe that DFA products will outperform similar Vanguard products. This seems plausible to me as Vanguard in their heart of hearts probably do not believe in factor investing as DFA does.

What I mean by my third point is illustrated by the Vanguard Small Cap Value Index. Does this product capture the small/value premium as well as the DFA product? From what I have read, it does not. DFA seems to do a better job of screening their stocks for small and value characteristics. Merriman said that DFA products had smaller average market capitalization and better value characteristics than similar Vanguard products. I think they were right.

It was disheartening to learn all of this after I had tilted my portfolio. But I determined that what I had was good enough and that I didn't want to pay an advisor 1% of assets a year.

If you believe that Value and Growth simply switch places in the performance derby and that their long term performance is about the same, stick with Vanguard. Ditto for Large and Small. If you believe that the premiums are real and persistent over long periods of time AND if your time horizon is long enough, then invest the DFA way.
A fool and his money are good for business.
User avatar
steve roy
Posts: 1855
Joined: Thu May 13, 2010 5:16 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by steve roy »

I've said this before, but whatever asset allocation you use will not turn out to be Numero Uno year in and year out. And two decades from now, you'll look back and see that you would hae been (somewhat) better off doing some other allocation.

So, build your portfolio, stick with it, and over time you'll come out fine. Just be aware that since you're not psychic, you won't get maximum returns.
User avatar
nisiprius
Advisory Board
Posts: 52105
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: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by nisiprius »

With regard to the IFA chart, I think Taylor's new thread about a J. P. Morgan performance claim, is vaguely relevant.

Don't trust (self-serving) Past Performance Figures.
It references:
We Tried to Re-Create JPMorgan’s Mutual Fund Returns and Gave Up.

Summary:
According to the bank’s 2014 presentation, 97 percent of alternative assets beat their benchmarks for the previous 10 years. In 2015 that figure rose to 100 percent. Working from their own, more limited data on JPMorgan’s alternative assets, Morningstar and Lipper got different numbers. Morningstar said 33 percent of JPMorgan’s alternative assets beat their benchmarks over the previous 10 years. Lipper’s figures show that only 14 percent did.
In general, JPMorgan made a number of impressive claims:
Pie charts in the asset management unit’s presentation showed the percentage of money invested in funds that ranked in the top half of their categories: In fixed-income funds with 10-year records, the figure was 85 percent; for stock funds with 10-year records, it was 83 percent.
Morningstar simply couldn't verify these numbers. The end result is that JPMorgan was claiming much better performance than Morningstar could verify, with no real explanation of the difference.

What's the relevance? If you are interested in comparing portfolios built from DFA funds versus Vanguard funds, do it yourself. Don't waste time looking at a chart prepared by someone who makes a living out of designing portfolios using DFA funds.
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.
Kevin K
Posts: 354
Joined: Sun Aug 26, 2007 7:47 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by Kevin K »

One of the best (and lowest-cost) advisory firms that uses DFA funds is Evanson Asset Management, and Steven Evanson has a worthwhile discussion of exactly how the DFA funds differ from index funds and ETFs on their web site:

http://www.evansonasset.com/index.cfm?Page=20

It seems to me that if you're a dyed-in-the wool believer in MPT and tilting AND have at least a 50% allocation to equities and a portfolio of a at least 500K (though many advisors want twice that) going with DFA makes all kinds of sense. They are as structurally committed to their approach as Vanguard is to indexing, and advisors like Evanson (and FPL, apparently) who provide access for a fixed fee rather than a percentage of assets make it an affordable choice (unlike IFA, which I've never believed did anything to merit its high fee apart from maintaining a colorful website).
User avatar
Taylor Larimore
Posts: 32839
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

"The Telltale Chart"

Post by Taylor Larimore »

Scott:

In his wonderful speech, The Telltale Chart, Mr. Bogle said:
So Slice and Dice is what you make it. Like all other investment strategies ever devised by the mind of man, sometimes it works and sometimes it doesn't. Uncertainty rules. Even if the overall program appears to outpace the Index, over a long inevitably period-dependent span of years, don't forget how little (!) it costs to emulate the total stock market in the real world nor how much (!) it costs to use active funds to fill the S&D boxes, and even to use passive funds to do so. If we take the extra risk into account, there's a real question about whether the game is worth the candle. And even if you don't accept my challenge to S&D, I urge you, before you plunge into a 4x25 portfolio, to put more than 25% in the total market—say 55%. Then put just 15% in the three slices that you dice, thereby taking much of the risk out of your decision. Think then, about a 1 x 55% + 3 x 15% portfolio. If it is true, as Dr. Fama (and most other academics, to say nothing of many, many practitioners) says, that "for most people, the market portfolio is the most sensible decision," (underline mine) you might as well make the most of it.
Think about it.

The Telltale Chart

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
pkcrafter
Posts: 15461
Joined: Sun Mar 04, 2007 11:19 am
Location: CA
Contact:

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by pkcrafter »

Scott, the composition of these portfolios was no where to be found-at least I could not find it. Nothing but performance charts. That's a pretty poor way to present a case for superior funds. I finally located composition for the portfolio 80, meaning 80/20 on a different site.

http://www.ifaarchive.com/pdf/ifa_portf ... et_p80.pdf

The IFA/DFA funds can't honestly be called index funds.

Are they superior to Vanguard funds on a risk adjusted basis? ??, but they have done better during the short time frame that was used, which is not surprising. I would draw no conclusion from IFA's pitch.

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

Thanks, everyone, for the additional replies to my post.

I'll try to touch on a couple of the comments, in sequential posts...

I'm not sure I have the hang of the quoting functionality, but here goes...

From "privatefarmer":
I have been comparing DFA funds to vanguard for as far back as we can go (usually about 15 years), and in ALMOST every case DFA has beaten vangaurd by ~1.5%/year.
Again, over the last 15-20 years, DFA has beaten vanguard by ~1.5%/yr on average in the "tilted" funds.
My take is that any performance comparison really has to be done on a risk-adjusted basis, using some type of multiple-regression analysis based on factor-specific betas for each factor included in the asset pricing model; the remainder is the portfolio's alpha, either + or -.

I think this is the crux of my main question re: factor-based investing, and also relates to IFA's use of standard deviation/variance (vs. beta/s) in the comparison chart that was the basis for my OP.

Thanks for your thoughts, privatefarmer. Best wishes.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

From Dave, "Random Walker":
So the negatives of DFA approach: AUM fee, higher portfolio ER
The potential positives of DFA approach: higher return, improved portfolio efficiency, better tax management, securities lending, lower rebalancing costs, advisor helps minimize behavioral errors, other advisor benefits like tax loss harvesting on specific lot basis throughout the year.
"Higher return" vs. risk-adjusted returns? Right now I'm thinking I should focus on risk-adjusted returns, based on the appropriate beta/s for each individual factor, to do this type of comparison.

Do you agree?

Thanks for your post and comments, Dave. Best wishes.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

From "gkaplan":
Postby gkaplan » Sun Aug 09, 2015 12:36 pm

http://www.altruistfa.com/dfavanguard.htm might be useful here.
Thanks for the link to the DFA vs. Vanguard reference comparison chart from Altruist, gkaplan. Best wishes.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by Random Walker »

Scott,
I wouldn't take credit for any free lunches. I would say a DFA portfolio is quite likely going to be more heavily tilted to the risk factors that provide higher expected returns. One can achieve the same tilts with VG funds, but I doubt most people would tilt as heavily with VG as they would with DFA.

Dave
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

From "nedsaid":
Post by nedsaid » Sun Aug 09, 2015 1:54 pm
...

I went to a couple of the Paul Merriman seminars where I was introduced to small/value tilting. They said that in a Large/Growth decade like the 1990's that a Vanguard portfolio would win out. In a Small/Value decade like the 2000's, DFA would do better than Vanguard.

So to take a DFA approach and to pay an advisor, you have to believe a few things. First, that the small/value premium is persistent over long periods of time. Second, that small/value tilting will give you enough of a performance premium to pay the advisor and still have excess performance above the market averages. Merriman claimed that they could outperform Vanguard by 1% even after charging their 1% management fee. Third, you have to believe that DFA products will outperform similar Vanguard products. This seems plausible to me as Vanguard in their heart of hearts probably do not believe in factor investing as DFA does.
I subscribe to Paul Merriman's e-newsletter, and read most of his articles posted at MArketWatch.com. My perception is that Mr. Merriman doesn't emphasize enough the concept of risk-adjusted returns (betas, not SD/var) when he writes about portfolio construction, portfolio tilts, etc.
Third, you have to believe that DFA products will outperform similar Vanguard products. This seems plausible to me as Vanguard in their heart of hearts probably do not believe in factor investing as DFA does.
How do you define outperformance? Based on what I've read to date, a proper performance comparison should be done via multiple-regression analysis, using factor-specific beta/s.

I totally agree with you that Vanguard, for whatever reason, is not gung-ho about factor-based investing. It's almost like trying to get a cat to bark, or a horse to cluck... you know what I mean?

I'm not sold yet on the idea myself. My focus right now if understanding the principles and approaches, then eventually make a (hopefully) well-informed decision on portfolio construction wrt to plain-vanilla or multi-factor (beta/s) investing.
If you believe that Value and Growth simply switch places in the performance derby and that their long term performance is about the same, stick with Vanguard. Ditto for Large and Small. If you believe that the premiums are real and persistent over long periods of time AND if your time horizon is long enough, then invest the DFA way.
This is a good point, which I believe both Mr. Bogle and Rick Ferri often emphasize. The ebb and flow of financial markets, past history vs. the unknown future, the long-term nature of many of the factor premiums, etc.

Thanks, nedsaid, for your thoughts and comments. Best wishes.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by Random Walker »

Additional DFA potential benefits: momentum and profitability screens

Dave
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

From: "Kevin K":
Postby Kevin K » Sun Aug 09, 2015 4:45 pm

One of the best (and lowest-cost) advisory firms that uses DFA funds is Evanson Asset Management, and Steven Evanson has a worthwhile discussion of exactly how the DFA funds differ from index funds and ETFs on their web site:

http://www.evansonasset.com/index.cfm?Page=20
Thanks for the URL link, Kevin K. I'll put that one with the others I'm accumulating.

BTW, do you have some of your personal money invested via Evanson Asset Management? How did you find out about them?

Best wishes.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

From Taylor Larimore:
"The Telltale Chart"

Postby Taylor Larimore » Sun Aug 09, 2015 6:05 pm

Scott:

In his wonderful speech, The Telltale Chart, Mr. Bogle said:


So Slice and Dice is what you make it. Like all other investment strategies ever devised by the mind of man, sometimes it works and sometimes it doesn't. Uncertainty rules. Even if the overall program appears to outpace the Index, over a long inevitably period-dependent span of years, don't forget how little (!) it costs to emulate the total stock market in the real world nor how much (!) it costs to use active funds to fill the S&D boxes, and even to use passive funds to do so. If we take the extra risk into account, there's a real question about whether the game is worth the candle. And even if you don't accept my challenge to S&D, I urge you, before you plunge into a 4x25 portfolio, to put more than 25% in the total market—say 55%. Then put just 15% in the three slices that you dice, thereby taking much of the risk out of your decision. Think then, about a 1 x 55% + 3 x 15% portfolio. If it is true, as Dr. Fama (and most other academics, to say nothing of many, many practitioners) says, that "for most people, the market portfolio is the most sensible decision," (underline mine) you might as well make the most of it.

Think about it.

The Telltale Chart

Best wishes.
Taylor
Thanks, Taylor, for your reply and the excerpt from Mr. Bogle's speech.

I especially like this part: "If we take the extra risk into account, there's a real question about whether the game is worth the candle."

I think he's talking about risk-adjusted returns, but just not in the same language that the academics like Fama, French, Cochrane, Harvey, Gray, etc. would use.

Both you and Mr. Bogle seem very wise in these (somewhat) murky waters, which I totally respect and appreciate.

When Mr. Bogle refers to "4x25 portfolio", what is he talking about? The context is missing. Is he commenting on allocating a portfolio between 4 factors? Or is that along the lines of the ever-trendy risk parity approach?

Best wishes.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by lack_ey »

scottj19707 wrote:When Mr. Bogle refers to "4x25 portfolio", what is he talking about? The context is missing. Is he commenting on allocating a portfolio between 4 factors? Or is that along the lines of the ever-trendy risk parity approach?
4x25 seems to have been a fairly commonly discussed slice-and-dice allocation for (US) equities. That's equal parts large blend, large cap value, small blend, small cap value. There is a huge megathread here about it, but I can't recall the title and am having no luck searching for it.

You can see a number of so-called lazy portfolios using it to some extent, especially the one from 4.1 Bill Schultheis:
http://www.bogleheads.org/wiki/Lazy_portfolios

The returns of 60/40 (US stock/bond), 4x25/40 (above split/bond), 30/30/40 (TSM/SCV/bond) can be seen here for 1972-2014:
https://www.portfoliovisualizer.com/bac ... pBlend2=15

I think some people these days would just take the third allocation if they believe in tilting.

EDIT: wait, as it says in the graph, he's defining it as S&P 500, large value, small value, and CRSP 9-10 (9th and 10th deciles by size, i.e. small blend). So more or less what I said.
Last edited by lack_ey on Tue Aug 11, 2015 10:29 am, edited 1 time in total.
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

Dear Taylor,

I forgot to comment on this section of the Bogle speech excerpt you included in this thread:
If it is true, as Dr. Fama (and most other academics, to say nothing of many, many practitioners) says, that "for most people, the market portfolio is the most sensible decision," (underline mine) you might as well make the most of it.
I think that is also consistent with a risk-adjusted comparison, using beta/s as the measure of risk.

What's seems a little contradictory (at least to me) is Dr. Fama's central role in the creation an evolution of DFA over the last 35 years.

Perhaps the "for most people" means anyone who's not a current or potential DFA client? Or maybe it's similar to Ralphie and his new toy in the movie "A Christmas Story"... and his mother's dire warning?

There sure is an awful lot of energy behind factor-based investing these days. As I said in an earlier reply post, my main goal is fully understanding both sides of the topic, so I can then make a (hopefully) informed decision re: future portfolio construction.

Until then, I sit next to you using a 3 fund portfolio approach.

Cheers!
Last edited by scottj19707 on Tue Aug 11, 2015 10:26 am, edited 1 time in total.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
bigred77
Posts: 2049
Joined: Sat Jun 11, 2011 4:53 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by bigred77 »

scottj19707 wrote:From Taylor Larimore:
"The Telltale Chart"

Postby Taylor Larimore » Sun Aug 09, 2015 6:05 pm

Scott:

In his wonderful speech, The Telltale Chart, Mr. Bogle said:


So Slice and Dice is what you make it. Like all other investment strategies ever devised by the mind of man, sometimes it works and sometimes it doesn't. Uncertainty rules. Even if the overall program appears to outpace the Index, over a long inevitably period-dependent span of years, don't forget how little (!) it costs to emulate the total stock market in the real world nor how much (!) it costs to use active funds to fill the S&D boxes, and even to use passive funds to do so. If we take the extra risk into account, there's a real question about whether the game is worth the candle. And even if you don't accept my challenge to S&D, I urge you, before you plunge into a 4x25 portfolio, to put more than 25% in the total market—say 55%. Then put just 15% in the three slices that you dice, thereby taking much of the risk out of your decision. Think then, about a 1 x 55% + 3 x 15% portfolio. If it is true, as Dr. Fama (and most other academics, to say nothing of many, many practitioners) says, that "for most people, the market portfolio is the most sensible decision," (underline mine) you might as well make the most of it.

Think about it.

The Telltale Chart

Best wishes.
Taylor
Thanks, Taylor, for your reply and the excerpt from Mr. Bogle's speech.

I especially like this part: "If we take the extra risk into account, there's a real question about whether the game is worth the candle."

I think he's talking about risk-adjusted returns, but just not in the same language that the academics like Fama, French, Cochrane, Harvey, Gray, etc. would use.

Both you and Mr. Bogle seem very wise in these (somewhat) murky waters, which I totally respect and appreciate.

When Mr. Bogle refers to "4x25 portfolio", what is he talking about? The context is missing. Is he commenting on allocating a portfolio between 4 factors? Or is that along the lines of the ever-trendy risk parity approach?

Best wishes.
I believe the 4x25 portfolio described in the tell tale chart refers to the following portfolio:

25% S&P 500
25% International Developed (EAFE)
25% Small Cap Blend
25% Gold
bigred77
Posts: 2049
Joined: Sat Jun 11, 2011 4:53 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by bigred77 »

bigred77 wrote:
scottj19707 wrote:From Taylor Larimore:
"The Telltale Chart"

Postby Taylor Larimore » Sun Aug 09, 2015 6:05 pm

Scott:

In his wonderful speech, The Telltale Chart, Mr. Bogle said:


So Slice and Dice is what you make it. Like all other investment strategies ever devised by the mind of man, sometimes it works and sometimes it doesn't. Uncertainty rules. Even if the overall program appears to outpace the Index, over a long inevitably period-dependent span of years, don't forget how little (!) it costs to emulate the total stock market in the real world nor how much (!) it costs to use active funds to fill the S&D boxes, and even to use passive funds to do so. If we take the extra risk into account, there's a real question about whether the game is worth the candle. And even if you don't accept my challenge to S&D, I urge you, before you plunge into a 4x25 portfolio, to put more than 25% in the total market—say 55%. Then put just 15% in the three slices that you dice, thereby taking much of the risk out of your decision. Think then, about a 1 x 55% + 3 x 15% portfolio. If it is true, as Dr. Fama (and most other academics, to say nothing of many, many practitioners) says, that "for most people, the market portfolio is the most sensible decision," (underline mine) you might as well make the most of it.

Think about it.

The Telltale Chart

Best wishes.
Taylor
Thanks, Taylor, for your reply and the excerpt from Mr. Bogle's speech.

I especially like this part: "If we take the extra risk into account, there's a real question about whether the game is worth the candle."

I think he's talking about risk-adjusted returns, but just not in the same language that the academics like Fama, French, Cochrane, Harvey, Gray, etc. would use.

Both you and Mr. Bogle seem very wise in these (somewhat) murky waters, which I totally respect and appreciate.

When Mr. Bogle refers to "4x25 portfolio", what is he talking about? The context is missing. Is he commenting on allocating a portfolio between 4 factors? Or is that along the lines of the ever-trendy risk parity approach?

Best wishes.
edit... They discussed 2 different portfolios:

4x25 portfolio
25% S&P 500
25% Large Value
25% Small Value
25% Small Cap stocks in the bottom 20% of CRSP market cap


Alternative 4x25 portfolio
25% S&P 500
25% International Developed (EAFE)
25% Small Cap Blend
25% Gold
hoops777
Posts: 4588
Joined: Sun Apr 10, 2011 12:23 pm
Location: Behind the 3 point line

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by hoops777 »

Nisprius made one of the wisest comments I have seen on this website......he has settled on good enough vs going for the very best.How much time,effort and stress is involved trying to get the highest return possible when it probably amounts to a meaningless amount of money when all is said and done.In the end with all the extra time and effort spent to hopefully get that fraction of a pct.....it is kind of ironic that things change and you might end up with less.
K.I.S.S........so easy to say so difficult to do.
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Unread postby hoops777 » Tue Aug 11, 2015 10:08 am

Nisprius made one of the wisest comments I have seen on this website......he has settled on good enough vs going for the very best.How much time,effort and stress is involved trying to get the highest return possible when it probably amounts to a meaningless amount of money when all is said and done.In the end with all the extra time and effort spent to hopefully get that fraction of a pct.....it is kind of ironic that things change and you might end up with less.
Yes, I think that's where the judgement and wisdom starts to come into play.

I'm still very much a 3 fund guy... VTI, VXUS, BND.

On the other hand, I don't think it's (necessarily) a good idea to ignore new evidence, assuming it makes sense and there's a good rationale and basis in fact.

Care, caution, and extra thought are certainly in order wrt to factor-based investing.

That's why I've spent so much time on due diligence, background checks, gathering evidence, etc. Information and understanding before action.

I'm actually very interested in opposing opinions and counterfactual arguments, sort of Charlie Munger's general approach to investing and life. Richard Feynman is, hopefully, smiling down on me with a twinkle in his eyes.

Thanks for your thoughts & comments, hoops777. Best wishes.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
User avatar
Taylor Larimore
Posts: 32839
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

The Three Fund Portfolio

Post by Taylor Larimore »

Bogleheads:

It is amazing to me how much time we spend trying to "beat the market" when The Three Fund Portfolio outperforms (return and risk) most of us who try.

A Case for Index Fund Portfolios by Rick Ferri, CFA
"The enemy of a good plan is the dream of a perfect plan." -- Jack Bogle
Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

Hi Taylor,

Thanks for your last reply, especially the link to Rick Ferri's article/paper.

I've had visibility to that paper pretty much since it was published. I re-scanned it just now, focusing mainly on the "risk-adjusted returns" aspect, and much less so on the "active vs. passive" aspect (which seems like the main focus of the overall paper).

Before I make a comment on the paper, I just want to be clear on two things:

First, I view Rick Ferri as a golden resource, both for the Bogleheads forum and for me personally. I really, really respect him and, as far as I can tell, he seems like a top-flight (all puns intended) type of person. I read as much of what he writes as possible, and benefit greatly as a result.

Second, I do not claim to be an expert in the area of factor-based investing. I'm just trying to understand the material better, both pros and cons, so I can (hopefully) make informed decisions about asset allocation, portfolio construction, etc.

Back to Rick's paper...

On page 18, under the section labeled "Scenario 5: Risk-Adjusted Performance", this is the direct quote from the 2nd paragraph...

"The risk metric used was the Sharpe ratio."

That single sentence loops me right back to my core question... what is the proper method/technique to analyze risk-adjusted returns/performance, especially in the context of factor-based investing (specifically using a multi-factor asset pricing model/approach).

My view, at least based on what I know right now, is that beta/s is the right approach. Using SD/variance, and by association a Sharpe ratio, is less good, less appropriate.

I believe the Treynor ratio incorporates beta component, right?

I need to go back to the paper and re-read it again completely, from top to bottom, and will certainly post again if I conclude something different after I do that.

Also, if you know Rick, which I'm guessing you probably do, could you maybe give him a heads-up re: my post here and ask him to comment, straighten me out on the details of his paper and how it relates to factor-based analysis? I would greatly appreciate it. I'm not intending in any way to leave any loose ends, etc. with respect to Rick's work.

It's very possible (probable?) I'm not correctly understanding or interpreting the design of the study, the intent, etc.

Thanks and best regards.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
afan
Posts: 8169
Joined: Sun Jul 25, 2010 4:01 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by afan »

Responding on two issues.

Differences in factor tilts: As others have noted, the DFA funds tend to be tilted to small and value, and more recently toward other factors. If you control for the tilts, then the DFA funds do not outperform. But this comparison does not mean that you can actually get a Vanguard portfolio with similar factor tilts. So, one can say the the DFA advantage lies entirely in the factor tilts, but if you want those tilts, you still would prefer DFA.

This result does suggest that the other things DFA promotes (patient investing, securities lending, etc) either are not as unique as they like to claim, or whatever advantages they may produce disappear in higher expenses or other things that Vanguard does better.

The DFA equity portfolio has not consistently outperformed the Vanguard equity portfolio. Over the entire life of the DFA equity portfolio, its return is less than that of the Vanguard equity portfolio, even without adding in advisor and transaction fees for the DFA portfolio. However, for the period of the start of the growth stock bubble (end of 1995) through August 31, 2009, DFA, even with an advisor and transaction fee of up to 1%/year, out-returned the Vanguard portfolio.

How has Vanguard matched up with two recent innovations in enhanced
indexing? From the end of 2007 through the end of August 2009, Vanguard’s equity portfolio has out returned
WisdomTree’s equity portfolio by a hair. Since the end of 2006, Vanguard’s equity return has fallen short of the DFA Core Equity return, although adding in a 1% annual fee for DFA reverses the ranking

http://poseidon01.ssrn.com/delivery.php ... 93&EXT=pdf
Other topic: What is the appropriate way to account for risk?
The IFA chart uses standard deviation, but as others have pointed out, this is at best an incomplete metric of risk. Specifically, since Kraus and Litzenberger it has been reliably shown that investors prefer positive skew and demand higher returns to hold negatively skewed portfolios. Certain factors, such as size and value, select for negatively skewed portfolios. If you include this negative skew, the size and value anomalies go away.

Tilting to size and value produces higher risk adjusted return only for investors whose preference for positive skew is weaker than "average". That is, if you are happier with negatively skewed portfolios than are most people, then you will be happy with the mean, variance and skew that these negatively skewed portfolios produce. If you are average in your skew preference, then these portfolios hold no advantage. If you are like almost everyone except a small number of professors of behavioral finance and don't know your skew preferences, then the safest move is to assume you are average, and ignore the temptation to tilt.
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
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

Thanks very much, "afan", for your recent post.
Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Unread postby afan » Tue Aug 11, 2015 1:15 pm.
On this section of your post...
Responding on two issues.

Differences in factor tilts: As others have noted, the DFA funds tend to be tilted to small and value, and more recently toward other factors. If you control for the tilts, then the DFA funds do not outperform. But this comparison does not mean that you can actually get a Vanguard portfolio with similar factor tilts. So, one can say the the DFA advantage lies entirely in the factor tilts, but if you want those tilts, you still would prefer DFA.

This result does suggest that the other things DFA promotes (patient investing, securities lending, etc) either are not as unique as they like to claim, or whatever advantages they may produce disappear in higher expenses or other things that Vanguard does better.
... I think I generally understand your points and agree with you.

On this section...
The DFA equity portfolio has not consistently outperformed the Vanguard equity portfolio. Over the entire life of the DFA equity portfolio, its return is less than that of the Vanguard equity portfolio, even without adding in advisor and transaction fees for the DFA portfolio. However, for the period of the start of the growth stock bubble (end of 1995) through August 31, 2009, DFA, even with an advisor and transaction fee of up to 1%/year, out-returned the Vanguard portfolio.

How has Vanguard matched up with two recent innovations in enhanced
indexing? From the end of 2007 through the end of August 2009, Vanguard’s equity portfolio has out returned
WisdomTree’s equity portfolio by a hair. Since the end of 2006, Vanguard’s equity return has fallen short of the DFA Core Equity return, although adding in a 1% annual fee for DFA reverses the ranking

http://poseidon01.ssrn.com/delivery.php ... 93&EXT=pdf
... it's hard to tell from your copy/paste above how exactly "outperformance" is being measured. I've downloaded the source paper from SSRN and need to spend some quality time with it. Thanks for the link... that one is new to me at this point.

On this section...
Other topic: What is the appropriate way to account for risk?

The IFA chart uses standard deviation, but as others have pointed out, this is at best an incomplete metric of risk. Specifically, since Kraus and Litzenberger it has been reliably shown that investors prefer positive skew and demand higher returns to hold negatively skewed portfolios. Certain factors, such as size and value, select for negatively skewed portfolios. If you include this negative skew, the size and value anomalies go away.

Tilting to size and value produces higher risk adjusted return only for investors whose preference for positive skew is weaker than "average". That is, if you are happier with negatively skewed portfolios than are most people, then you will be happy with the mean, variance and skew that these negatively skewed portfolios produce. If you are average in your skew preference, then these portfolios hold no advantage. If you are like almost everyone except a small number of professors of behavioral finance and don't know your skew preferences, then the safest move is to assume you are average, and ignore the temptation to tilt
... I need to get a copy of the "Skewness Preference and the Valuation of Risk Assets" paper by

Alan Kraus and Robert H. Litzenberger
The Journal of Finance
Vol. 31, No. 4 (Sep., 1976), pp. 1085-1100
Published by: Wiley for the American Finance Association
DOI: 10.2307/2326275
Stable URL: http://www.jstor.org/stable/2326275
Page Count: 16

I don't have personal access to JSTOR, so I think I'll need to visit my local library to get a legit copy... that will take a little while to obtain and read.

The K&L article is also new to me. It looks like it's been around or a while (since 1976)... I'm curious why I didn't regularly see that one in the references of the many other papers I've read/scanned recently? Maybe I overlooked it in my prior searches to date.

On the surface, the concept you describe in your post sounds like it's getting at the "lottery effect" many people refer to, right?

Thanks for your help and your insightful comments and links.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
User avatar
Taylor Larimore
Posts: 32839
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by Taylor Larimore »

Vanguard Funds vs. Dimensional Funds ?
Bogleheads:

I don't have the time to evaluate the past performance* of Vanguard and Dimensional funds. However, I do know that over the past 10 years 92% of Vanguard funds have outperformed their peer group average. That's good enough for me.

Vanguard Performance Report

* Past performance does not forecast future performance.

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
afan
Posts: 8169
Joined: Sun Jul 25, 2010 4:01 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by afan »

Well, people marketing negatively skewed portfolios don't like to talk about skewness since it makes their anomalies go away. Many people have enough trouble wrapping their minds around variance so they get headaches when people bring up higher moments.

K&L, as far as I know, were the first to bring this up, but many people have followed on since then.

A recent paper with a good brief literature review is
"Market skewness risk and the cross section of stock returns"
Bo YoungChang a, PeterChristoffersen b,c,n, KrisJacobs d,e

Journal of Financial Economics 107 (2013)46–68

Also,

"Asset pricing when returns are nonnormal: Fama-French factors versus higher-order systematic comoments"
Chung, YP; Johnson, H; Schill, MJ
JOURNAL OF BUSINESS Volume: 79 Issue: 2 Pages: 923-940 Published: MAR 2006

"A growing literature contends that, since returns are not normal, higher order comoments matter to risk-averse investors. Fama and French (1993, 1995) find that nonmarket risk factors based on size and book-to-market ratio are priced by investors. We test the hypothesis that the Fama-French factors simply proxy for the pricing of higher order comoments. Using portfolio returns over various time horizons, we show that adding a set of systematic comoments (but not standard moments) of order 3-10 reduces the explanatory power of the Fama-French factors to insignificance in almost every case."

That should get you on your way. Really interesting, and not something those who sell tilting advice like to discuss.
Last edited by afan on Tue Aug 11, 2015 4:10 pm, edited 1 time in total.
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
User avatar
Rick Ferri
Posts: 9703
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by Rick Ferri »

scottj19707 wrote: Back to Rick's paper...

On page 18, under the section labeled "Scenario 5: Risk-Adjusted Performance", this is the direct quote from the 2nd paragraph...

"The risk metric used was the Sharpe ratio."

That single sentence loops me right back to my core question... what is the proper method/technique to analyze risk-adjusted returns/performance, especially in the context of factor-based investing (specifically using a multi-factor asset pricing model/approach).

My view, at least based on what I know right now, is that beta/s is the right approach. Using SD/variance, and by association a Sharpe ratio, is less good, less appropriate.

I believe the Treynor ratio incorporates beta component, right?
The Sharpe ratio is a simple method to compare the risk and return of different portfolios. Simply subtract the risk-free T-bill return from a portfolio return, and then divide by the annualized standard deviation of the portfolio. The result is the Sharp ratio. It tells you the excess return you received per unit of risk taken (when volatility is used as risk). The Sharp ratio of one fund is compared to similar funds, including index funds, and the funds with the highest Sharp ratios have provided the best risk adjusted return. This was a proper comparison in the context of our white paper.

You can also use the Sharp ratio to compare multi-factor funds other multi-factor funds and indexes, but you'll run into a lack of data. Most multi-factor funds don't have a long track record, so you're not going to get much information out of the comparison. There are many ways to attempt to measure the risk and return in a portfolio, but again, all the "smart beta" stuff is newly conceived and based on optimal back-testing. Whatever live data there is going back 5-years has not been impressive.

That brings us to the problem of data-snooping biases. There are long periods of time when the additional risk factors do not outperform a single factor model (smart beta does not beat the market). Nonetheless, the sellers of financial products conveniently leave those periods out of their analysis. This is why most sellers of smart beta products only go back to the year 2000 when making the case for higher risk-adjusted returns. Going back to 1990 blows their case out of the water.

There is not such thing as a free lunch on Wall Street. If some fund is achieving a high risk adjusted return, and anyone can invest in that fund, then there is a risk hidden in there someplace. We may know what it is and ignore it, or perhaps we just don't know what it is yet.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
afan
Posts: 8169
Joined: Sun Jul 25, 2010 4:01 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by afan »

Sharpe ratio is appealing because it produces a unique ranking of portfolio performance. All risk averse investors would prefer the higher Sharpe ratio portfolio. At least they would if you assume they can borrow at the risk free rate to leverage their investments to get to as high a risk as some might want.

As soon as you introduce a third return factor, skewness, then this nice property goes away. Investors with different utility functions may prefer portfolios with different combinations of return, variance, and skewness. Although one can infer the utility function of investors collectively from the investment decisions they make, there is no suggestion that everyone has the same function. If you had a report with the mean, variance and skewness of multiple portfolios there is not a consensus statistic that would tell you which was best. Of portfolios with the same skewness, one would prefer the higher Sharpe ratio. Among portfolios with the same Sharpe ratio, one would prefer the most positive skewness. If they differ in mean, variance and skew, there is no answer right now for which is best.

It is clear that it makes no sense to ignore variance and risk in determining optimal. Thus, saying that small company value portfolios had a higher mean (over some time period when this was true) without inquiring into their variance or skewness, would tell one nothing about how they did relative to some other portfolio- the market portfolio or anything else.
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
Kevin K
Posts: 354
Joined: Sun Aug 26, 2007 7:47 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by Kevin K »

My apologies to the OP for taking awhile to reply to your question to me about Evanson and Associates. I did have 100% of my portfolio with them for about 5 years and don't regret the experience a bit. I think that for many their market insights and hand-holding during times of crisis could be worth many times their modest fees.

Over time (and after seeing what happened to my very complex, slice-and-dice, thoroughly tilted and supposedly conservative portfolio during the '08 market meltdown) I came to see that (a) I prefer simplicity and am willing to accept the market return at as low a cost as possible; (b) with 60% of my portfolio in bonds it makes little sense to pay for access to a complex DFA stock portfolio.

If Mr. Bogle's crystal ball is correct (do I need to say I trust his sense of the market more than my own?) and we're headed for a couple of decades of very low returns from both bonds and equities I feel just fine with my Vanguard 3 fund portfolio and its ~.21% ER.

As long as you're accumulating information on the age-old DFA vs. Vanguard debate you might want to be aware if you aren't already of this post from the White Coat Investor site. I still think it's the single best piece of writing on the topic I've come across:

http://whitecoatinvestor.com/dfa-vs-vanguard/
DFAMAN
Posts: 114
Joined: Sat Sep 19, 2009 6:03 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by DFAMAN »

As you might infer from my screen name, I tend to lean a particular way in terms of preference here. However, I see this as kind of like a debate between two Baptist churches across town from each other who disagree as to whether the piano should be placed on the right v. left side of the sanctuary. Most outsiders would wonder what the fuss is about here.

Really, if you are down to a comparison between Vanguard and DFA, you likely are doing better than the vast majority of your friends and colleagues in terms of overall investment well-being (as long as you are not paying an unreasonable amount for an advisor if using DFA). Truly a first-world problem to have to make decisions like this!
User avatar
nedsaid
Posts: 19249
Joined: Fri Nov 23, 2012 11:33 am

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by nedsaid »

scottj19707 said,
"How do you define outperformance? Based on what I've read to date, a proper performance comparison should be done via multiple-regression analysis, using factor-specific beta/s."

Oh man, you got me here. I never was much of a mathematician as my D and C in College Calculus proved. I am a Accountant and I analyze numbers for a living but it is not advanced math.

In my simple mind, I would compare the performance of a tilted portfolio to a similarly weighted portfolio of index funds. So let's say that over five years, (I am making this up) that a Merriman portfolio returned 10% a year and a three fund Vanguard portfolio with the same stock/bond and US/International ratios returned 8% a year, I would say that Merriman outperformed Vanguard 2% a year.

I would compare a Merriman portfolio with a Vanguard 3 fund portfolio that is 30% US Total Stock Market Index, 30% Total International Stock Market Index, and 40% US Total Bond Market with the 60/40 buy and hold portfolio that Merriman recommends. Simple as that.

I could do an evaluation of risk and ask if the Merriman portfolio was riskier than the Vanguard Indexed portfolio. I could then do measurements and see if the extra return I got was worth the extra risk I took. But Merriman says that his tilted portfolio will have less risk with more return than the Vanguard Indexed portfolio. He measures risk by standard deviation. There might be better ways to measure risk but I will leave it to the mathematicians to figure it out. But standard deviation is the best known measure.

I could do the analysis best I could but when you study this out you see that the financial data shifts all over the place. Just change your time period for comparison a bit and this can change the conclusions that you reach. I look at the big picture and the long term and try not to get too obsessed with shorter term analysis. Longer term, the small and value tilting makes a lot of sense. Who knows what will work in the short term?

I believe in small/value tilting because it makes sense in light of human nature and human behavior. Pretty much people gravitate towards what is popular and tend to chase performance. Tilting is a way of capitalizing on behavioral errors that people make over and over again. Over time small and value are less popular and less followed and thus have lower expectations. Low expectation stocks have a better chance of beating market expectations than high expectation stocks.

I look at the math and the charts but mostly I think in narrative form. I hope my answer doesn't disappoint you too much. You probably thought that I had complex programmed spreadsheets that I created to figure all this out. I don't. I don't do a lot of complicated analysis. Sorry to disappoint.
A fool and his money are good for business.
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by Random Walker »

I think the DFA route makes most sense for individuals who have aggressive asset allocations. And the diagram behind the start of this post supports this thought. Diversifying into more esoteric asset classes on the non bond side of the portfolio may have greater benefit for individuals with smaller allocation to safe bonds. The more exotic the asset class, the more expensive. But for someone with an aggressive portfolio, the extra expense may be worth it for a more efficient portfolio and a plan that is less likely to be abandoned when the going gets rough.

Dave
afan
Posts: 8169
Joined: Sun Jul 25, 2010 4:01 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by afan »

jedsaid,

The SSRN paper cited above did what you suggest, but accounted for factor weighting. It did not account for non-normal stock returns. There is more math in looking at higher moments, but you don't need calculus.
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
Topic Author
scottj19707
Posts: 226
Joined: Fri Dec 05, 2014 12:25 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by scottj19707 »

afan wrote » Sat Aug 15, 2015 7:03 pm

jedsaid,

The SSRN paper cited above did what you suggest, but accounted for factor weighting. It did not account for non-normal stock returns. There is more math in looking at higher moments, but you don't need calculus.
I was reading an article recently that talked about the various "moments" of a random variable...

1st: mean
2nd: variance/standard deviation
3rd: skew
4th: kurtosis

Is that pretty much the straight scoop from a stats perspective?

Thanks.
“To acquire knowledge, one must study; but to acquire wisdom, one must observe.”-Marilyn vos Savant; “If you can’t explain it simply, you don’t understand it well enough.”-Albert Einstein; VTI/VXUS/BND
CRTR
Posts: 351
Joined: Sat Apr 11, 2015 1:15 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by CRTR »

Well, I was off work today and bored . . . . . So, I decided to give it a go at duplicating a well-known DFA based portfolio using VANGUARD and other available ETFs. I attempted to match each component of the old IFA 100/DFA portfolio using ETFs with similar Morningstar portfolio analytics and 10 year correlations greater than 98%. Rebalancing was performed annually. I ran the two portfolios through Morningstar and posted a PDF of the result.

As can be seen, the IFA portfolio can be "cloned" easily. The correlation coefficient between the two portfolios is 99% since inception of the newest ETFs (1/2010). Not surprisingly, overall performance, alpha, beta and sharpe ratio of the two are matched as well. Though i would like to see longer than a 7 year run, given the statistical match of the two portfolios, no reason to think things will not continue similarly as time goes on.

Of note, this exercise does NOT include trading costs or the advisor fees required to access DFA funds. Transaction costs, of course, would be higher for the DFA portfolio due to higher mutual fund trading costs vs ETFs. If one were to subtract the IFA 0.9% AUM fee, the clone's performance would trounce the DFA mix as well. Given that ANYONE can create the cloned portfolio, it's hard for me to justify IFA's fee structure. As there are a number of free or ridiculously cheap ETF trading options out there, the pro-IFA argument becomes even weaker.

https://drive.google.com/file/d/0B1ll8a ... sp=sharing

As always, I welcome any thoughtful comments/criticisms/etc.
Kevin K
Posts: 354
Joined: Sun Aug 26, 2007 7:47 pm

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by Kevin K »

Thanks very much CRTR for resurrecting this old thread with such a fascinating and useful piece of work!
indexonlyplease
Posts: 1571
Joined: Thu Apr 30, 2015 12:30 pm
Location: Florida

Re: "Vanguard Funds vs. Dimensional Funds in New IFA Index Portfolios" -- What's The Deal?

Post by indexonlyplease »

I was investing with IFA for around 3 years until I found this site. The only problem I have with them is the fee they charge. I believe it is .9% of investments. My advisor was great. The problem with advisors that use index funds the don't talk about what the fees will cost you over 30 plus years of investing. So, if they decide to use Vanguard or DFA the management fees kill the deal. You also pay .3% for DFA. For a total of 1.2%.
Also, don't forget about the trading cost when you change your AA or sell funds.

I cancelled with them and did the 3 fund portfolio.
Post Reply