RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
diyfp
Posts: 50
Joined: Wed Mar 09, 2016 2:22 am

RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by diyfp » Mon Mar 14, 2016 3:12 am

Hi everyone.

As I mentioned in my first post to this forum, I'm considering splitting my international holdings 50/50 between Vanguard and RAFI.

While doing my research, I found that since 2009, RAFI's International Small Index has beaten Vanguard's International Small Index by 3% annually and with lower volatility. It also beat DFA's International Small (which had much higher volatility) although by less (0.61% annually).

https://www.portfoliovisualizer.com/asset-correlations?s=y&numTradingDays=60&s=y&endDate=03%2F13%2F2016&timePeriod=2&symbols=VFSVX+SFILX+DISVX

How do we explain this?? Can someone do (and/or show me how to do) a factor analysis on the three funds to explain the performance difference please?

Thanks.

User avatar
David Jay
Posts: 3981
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by David Jay » Mon Mar 14, 2016 3:43 am

What is RAFI?

In your link, what does RAFI have to do with Schwab?
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius

diyfp
Posts: 50
Joined: Wed Mar 09, 2016 2:22 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by diyfp » Mon Mar 14, 2016 3:45 am

David Jay wrote:What is RAFI?

In your link, what does RAFI have to do with Schwab?


RAFI makes the index that the Schwab fund SFILX follows.

RAFI is Research Affiliates Fundamental Index.

User avatar
David Jay
Posts: 3981
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by David Jay » Mon Mar 14, 2016 3:57 am

Okay, so Schwab follows a different index than Vanguard. That easily explains the difference. The specific mix of companies is different so the results are different.

Remember, 6 years is not very long. Past performance is not a reliable indicator of future performance.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius

diyfp
Posts: 50
Joined: Wed Mar 09, 2016 2:22 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by diyfp » Mon Mar 14, 2016 4:02 am

Yes, the three funds all follow different indices.

My question was whether factor analysis could explain the performance difference between the 3 since 2009.

Especially between RAFI and Vanguard.

User avatar
in_reality
Posts: 4274
Joined: Fri Jul 12, 2013 6:13 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by in_reality » Mon Mar 14, 2016 4:14 am

diyfp wrote:Yes, the three funds all follow different indices.

My question was whether factor analysis could explain the performance difference between the 3 since 2009.

Especially between RAFI and Vanguard.


That Schwab Fundamental fund loads on small and value and doesn't have negative momentum.

See viewtopic.php?t=184501 Towards the end of the thread Larry Swedroe has his take on it. Larry is well regarded so you might want to read that part too.

Anyway, what I know about those Fundamental funds from holding them is that they are not less risky. FNDE and FNDX (emerging and large cap US ETFs) rebalanced into energy more than traditional value indexes. Maybe it will pay off, maybe it won't, maybe it will have lower volatility and maybe it won't, but surely that is increasing the risk compared to market cap weighting.

Brogleski
Posts: 37
Joined: Fri Jul 12, 2013 4:23 pm

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Brogleski » Mon Mar 14, 2016 8:50 am

diyfp wrote:Hi everyone.

As I mentioned in my first post to this forum, I'm considering splitting my international holdings 50/50 between Vanguard and RAFI.

While doing my research, I found that since 2009, RAFI's International Small Index has beaten Vanguard's International Small Index by 3% annually and with lower volatility. It also beat DFA's International Small (which had much higher volatility) although by less (0.61% annually).

https://www.portfoliovisualizer.com/asset-correlations?s=y&numTradingDays=60&s=y&endDate=03%2F13%2F2016&timePeriod=2&symbols=VFSVX+SFILX+DISVX

How do we explain this?? Can someone do (and/or show me how to do) a factor analysis on the three funds to explain the performance difference please?

Thanks.
Not that a six year time period should influence your investment decisions, but the comparison as presented isn't very meaningful. VFSVX contains emerging markets while SFILX and DISVX don't. Below is a more granny smith to fuji comparison.

https://www.portfoliovisualizer.com/backtest-portfolio?s=y&endDate=03%2F13%2F2016&allocation2_2=80&symbol4=BND&lastMonth=12&allocation4_3=100&symbol1=VSS&endYear=2016&symbol3=SFENX&frequency=4&symbol2=SFILX&inflationAdjusted=true&annualAdjustment=0&showYield=false&startYear=1985&rebalanceType=1&timePeriod=4&annualPercentage=0.0&allocation1_1=100&allocation3_2=20&annualOperation=0&firstMonth=1&reinvestDividends=true&initialAmount=10000

Don't worry. Despite an unfortunate 6-year under performance relative to Vanguard, those RAFI funds are still fine investments.
:beer
Last edited by Brogleski on Mon Mar 14, 2016 9:46 am, edited 1 time in total.
“A good plan, violently executed now, is better than a perfect plan next week.” - George S. Patton

User avatar
in_reality
Posts: 4274
Joined: Fri Jul 12, 2013 6:13 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by in_reality » Mon Mar 14, 2016 9:03 am

Brogleski wrote:Don't worry. Despite an unfortunate 6-year under performance relative to Vanguard, those RAFI funds are still fine investments.
:beer


VSS isn't 20% emerging though is it? Shows more like 12-13%.

Keeping the emerging to about the same allocation shows:

VSS 3.88% CAGR

SFILX SFENX 4.33% CAGR
2.07% lower Std.Dev. too!

Brogleski
Posts: 37
Joined: Fri Jul 12, 2013 4:23 pm

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Brogleski » Mon Mar 14, 2016 9:44 am

in_reality wrote:
Brogleski wrote:Don't worry. Despite an unfortunate 6-year under performance relative to Vanguard, those RAFI funds are still fine investments.
:beer


VSS isn't 20% emerging though is it? Shows more like 12-13%.

Keeping the emerging to about the same allocation shows:

VSS 3.88% CAGR

SFILX SFENX 4.33% CAGR
2.07% lower Std.Dev. too!


Vanguard says it's currently 18%, and I think it was higher in 2010 than it is now. I also picked SFENX, because RAFI doesn't have an emerging market small cap fund. Perhaps DGS would work better.

Either way, the inclusion of emerging markets has probably played a bigger role in the performance difference between VSS and SFILX than factor-style has.
“A good plan, violently executed now, is better than a perfect plan next week.” - George S. Patton

User avatar
in_reality
Posts: 4274
Joined: Fri Jul 12, 2013 6:13 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by in_reality » Mon Mar 14, 2016 10:07 am

Brogleski wrote:Vanguard says it's currently 18%, and I think it was higher in 2010 than it is now.


Well just because they want to put developed Asia into the emerging category doesn't make it so. Morningstar and the FTSE index it follows disagree.

Modern Indexer
Posts: 21
Joined: Sun Mar 06, 2016 10:04 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Modern Indexer » Mon Mar 14, 2016 11:39 am

fundamentally indexing is not really old enough to know if there is really alpa there....it's really a question of what form of indexing you are most comfortable with a traditional market cap weighting or company valuation weighting with the traditional measures of yield, p/b etc.

Brogleski
Posts: 37
Joined: Fri Jul 12, 2013 4:23 pm

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Brogleski » Mon Mar 14, 2016 11:46 am

in_reality wrote:
Brogleski wrote:Vanguard says it's currently 18%, and I think it was higher in 2010 than it is now.


Well just because they want to put developed Asia into the emerging category doesn't make it so. Morningstar and the FTSE index it follows disagree.
Fair enough. VSS vs. SFILX/FNDC is still a meaningless comparison (without adding some emerging markets on the SFILX/FNDC side).
“A good plan, violently executed now, is better than a perfect plan next week.” - George S. Patton

Modern Indexer
Posts: 21
Joined: Sun Mar 06, 2016 10:04 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Modern Indexer » Mon Mar 14, 2016 12:18 pm

David Jay wrote:Okay, so Schwab follows a different index than Vanguard. That easily explains the difference. The specific mix of companies is different so the results are different.

Remember, 6 years is not very long. Past performance is not a reliable indicator of future performance.



these index's started by Research Affiliates were backtested much farther than 6 years before the developed the index's and they found increased return in any fundamental measurement available (pick your poison) in comparison to market cap weighted index's. adding such an etf to your portfolio is like adding a value fund based on long term outperformance, who knows what will win out in the future.

larryswedroe
Posts: 15598
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by larryswedroe » Mon Mar 14, 2016 12:31 pm

Haven't looked at this in long time but this might be helpful

http://www.cbsnews.com/news/how-serendipity-plays-a-role-in-returns/

diyfp
Posts: 50
Joined: Wed Mar 09, 2016 2:22 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by diyfp » Mon Mar 14, 2016 9:04 pm

So it looks there are 3 theories for how RAFI had higher returns with lower volatility over this period than Vanguard and DFA.

1. Small/Value
in_reality wrote:
That Schwab Fundamental fund loads on small and value and doesn't have negative momentum.



The small/value theory doesn't seem to hold much water based on Morningstar. When I look at the 3 funds (SFILX, VFSVX & DISVX), SFILX has a larger average market cap than the other two. It is also less valuey (using all 5 value measures on Morningstar) than the other two.

What is negative momentum? And where can I find that information for a fund?

2. Emerging Markets
Brogleski wrote:

...the comparison as presented isn't very meaningful. VFSVX contains emerging markets while SFILX and DISVX don't.



I had not realized this. Valid point. Emerging markets have underperformed quite a bit recently. As in_reality brings up though, this doesn't seem to explain all the difference...

3. They Got Lucky


Possible. Why only on their international small fund though? Shouldn't this style drift over the course of the year affect their international large and emerging large as well? These last two (which I'm more seriously considering investing in than the international small) have seriously underperformed their Vanguard counterparts.

User avatar
in_reality
Posts: 4274
Joined: Fri Jul 12, 2013 6:13 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by in_reality » Mon Mar 14, 2016 11:08 pm

diyfp wrote:
3. They Got Lucky


Possible. Why only on their international small fund though? Shouldn't this style drift over the course of the year affect their international large and emerging large as well? These last two (which I'm more seriously considering investing in than the international small) have seriously underperformed their Vanguard counterparts.


I don't consider it style drift. It's the result of their weighting. If there is a large difference between Growth and Value valuations, there will be more value exposure. But after a period of value over-performance and valuations are closer there will be less.

Larry correctly points out that the effect is not seasonal. It just happened to coincide with the seasons. Still, the value effect does come and go and research has been done on how it is mean reverting. Thus, can you really say a fund that adjusts it value exposure to catch the mean reversion is lucky. To some degree perhaps but ...

http://www.researchaffiliates.com/Our%2 ... rting.aspx

Anyway, if you look at the morningstar 9box, you will see the "small value" Fundamental Indexes have a fair amount in mid-cap growth? Again, I don't consider it style drift. There are companies in non-value sectors (finance and energy tend to be more valuey) that still have attractive valuations based on their economic footprint. So looking at size/value measures in morningstar will make it seem that the Fundamental Indexes have less exposure. I will take a relatively valuey company in a non-valuey sector anyway. Anyway, Robert T has calculated the factor loadings on the index over a long time, and they do show a strong value tilt.

Negative momentum? In general value indexes have negative momentum loads. Momentum is when things going up keep going up. I may be wrong in the exact relationship. RAFI funds do have negative momentum, maybe even worse negative momentum but then higher alpha (perhaps by their counter trading) --

viewtopic.php?t=140656#p2084678
viewtopic.php?f=10&t=126899#p1863397

Those are for a different fund that no longer exists. It's RAFI but pure value (sticking only to the value side). Schwab fund includes even growth companies with relatively favorable valuations (for their economic footprint). So not sure how relevant those threads are.

Fryxell
Posts: 69
Joined: Sat Mar 02, 2013 10:08 pm
Location: Dallas

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Fryxell » Mon Mar 14, 2016 11:37 pm

diyfp wrote:Yes, the three funds all follow different indices.

My question was whether factor analysis could explain the performance difference between the 3 since 2009.

Especially between RAFI and Vanguard.


A 3-factor factor analysis on PortfolioVisualizer using Global ex-US using the same dates shows alpha for SFILX, but it is not statistically significant. So it could very well be a fluke. If you run it as a 4-factor (including momentum) it is barely statically significant (0.016 p-value). It could still be a fluke, or it could suggest that negative momentum may be partially counteracting the benefit of any alpha that the fund may have.

The DFA fund and SFILX are close in overall returns, while the Vanguard fund lags. This is probably because the Vanguard fund includes emerging markets.

Alex Frakt
Founder
Posts: 10730
Joined: Fri Feb 23, 2007 1:06 pm
Location: Chicago
Contact:

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Alex Frakt » Tue Mar 15, 2016 1:04 am

diyfp wrote:So it looks there are 3 theories for how RAFI had higher returns with lower volatility over this period than Vanguard and DFA.

1. Small/Value
2. Emerging Markets
3. They Got Lucky

4. They bury the losers.

RAFI has changed its index construction methodology when actual returns failed to live up to the backtest. Of course the newly constituted "fundamental" formula always shows superior backtested results, but it does make you wonder how they define the word "fundamental." They have also shut down indices that have underperformed, such as the entire RAFI Fundamental US Style Index Series.

larryswedroe
Posts: 15598
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by larryswedroe » Tue Mar 15, 2016 7:55 am

few things
First, for in_reality. MOM as used by RAFI (and DFA) is cross-sectional MOM, not time series, so it's not absolute but relative----which means you can be buying stocks that are going down, just not going down as much as the rest of their peers in the asset class.

Second, there are papers on timing value ---switching to growth when spreads of btm are smaller than average and vice versa--and it doesn't work for same reason timing equity premium in similar way doesn't work=----there's always an ex-ante value premium because you are buying the cheaper stocks.

Larry

User avatar
in_reality
Posts: 4274
Joined: Fri Jul 12, 2013 6:13 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by in_reality » Tue Mar 15, 2016 8:59 am

larryswedroe wrote:few things
First, for in_reality. MOM as used by RAFI (and DFA) is cross-sectional MOM, not time series, so it's not absolute but relative----which means you can be buying stocks that are going down, just not going down as much as the rest of their peers in the asset class.

Second, there are papers on timing value ---switching to growth when spreads of btm are smaller than average and vice versa--and it doesn't work for same reason timing equity premium in similar way doesn't work=----there's always an ex-ante value premium because you are buying the cheaper stocks.

Larry


Thanks for your input Larry. It's always good to hear your views.

I didn't mean to characterize RAFI as switching to growth.

This is what RAFI asserts and which makes sense to me.

When the mean-reversion effect shows evidence of mean reversion, it makes sense to dollar cost average contrarian bets.2 Let me show this by comparing two portfolios: one which allocates a constant tracking error to low P/B stocks, and another which dynamically allocates more tracking error when the gap between growth and value P/B ratios widens. The first portfolio is akin to the traditional value strategies, which tilt toward cheap stocks in order to generate outperformance. The second portfolio is similar to fundamentals-weighted and other simpler Smart Beta indices, whose rebalancing heuristics implicitly contain dollar cost averaging.


http://www.researchaffiliates.com/Our%2 ... rting.aspx

KyleAAA
Posts: 6322
Joined: Wed Jul 01, 2009 5:35 pm
Contact:

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by KyleAAA » Tue Mar 15, 2016 9:28 am

Random chance could easily explain such a difference over a short period of time.

larryswedroe
Posts: 15598
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by larryswedroe » Tue Mar 15, 2016 9:44 am

In_reality
What I've tried to point out is that I don't buy the RAFI strategy of weighting more to value when spreads are wider and less when narrower. The reason is same you don't switch from stocks to bonds when P/Es are higher. A smaller spread just means you have an expected smaller value premium. Studies on subject confirm this.
Larry

Theoretical
Posts: 1068
Joined: Tue Aug 19, 2014 10:09 pm

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Theoretical » Tue Mar 15, 2016 10:01 am

in_reality wrote:
larryswedroe wrote:few things
First, for in_reality. MOM as used by RAFI (and DFA) is cross-sectional MOM, not time series, so it's not absolute but relative----which means you can be buying stocks that are going down, just not going down as much as the rest of their peers in the asset class.

Second, there are papers on timing value ---switching to growth when spreads of btm are smaller than average and vice versa--and it doesn't work for same reason timing equity premium in similar way doesn't work=----there's always an ex-ante value premium because you are buying the cheaper stocks.

Larry


Thanks for your input Larry. It's always good to hear your views.

I didn't mean to characterize RAFI as switching to growth.

This is what RAFI asserts and which makes sense to me.

When the mean-reversion effect shows evidence of mean reversion, it makes sense to dollar cost average contrarian bets.2 Let me show this by comparing two portfolios: one which allocates a constant tracking error to low P/B stocks, and another which dynamically allocates more tracking error when the gap between growth and value P/B ratios widens. The first portfolio is akin to the traditional value strategies, which tilt toward cheap stocks in order to generate outperformance. The second portfolio is similar to fundamentals-weighted and other simpler Smart Beta indices, whose rebalancing heuristics implicitly contain dollar cost averaging.


http://www.researchaffiliates.com/Our%2 ... rting.aspx


Ok, I have absolutely no idea how actively variable exposure to the value premium is dollar cost averaging. Wouldn't DCA be more consistent with plugging the same amount to static value whether the premium is strong or weak? I think they are some of the better funds (especially internationally) for going after small/value, especially for taxable accounts, but isn't the point of DCA that you invest incrementally whether it is expensive or cheap to do so?

larryswedroe
Posts: 15598
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by larryswedroe » Tue Mar 15, 2016 11:43 am

Theoretical
FWIW
Constantinides of U of Chicago wrote paper about 50 years ago showing that DCA isn't optimal, investing all at once is, and that hasn't been challenged since. DCA can be helpful in supplying a psychological crutch which at least allows people to invest in the face of uncertainty, which is easier for many to do with small amounts than large amounts
Larry

garlandwhizzer
Posts: 1623
Joined: Fri Aug 06, 2010 3:42 pm

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by garlandwhizzer » Tue Mar 15, 2016 1:08 pm

Larry wrote:
DCA can be helpful in supplying a psychological crutch which at least allows people to invest in the face of uncertainty, which is easier for many to do with small amounts than large amounts


Do not underestimate the importance of psychological crutches in making market timing moves. To put it mildly market timing is an inexact science and having a psychological crutch can be very handy when you've picked the wrong time to invest lump sum invest. When the market is richly valued and yet you wish to increase equity exposure, wading in slowly is easier than diving off a high platform. If the market declines during the DCA period, you wind up buying more shares than an initial lump sum investment. Plus with DCA you can stop altogether if unexpected disaster happens (think 9/11) and wait until the dust settles at a much lower market level to do your purchase. Whatever the academic studies say, these are points to consider. As to which is best, DCA or lump sum, I don't believe it's a one size fits all simple answer. Rather it depends on the exact circumstances of the market at that particular time, whether it is richly valued or a bargain, and also the specifics of the investor at that time, his risk tolerance, needs for liquidity, and what percentage of his assets are to be invested. If it's a small investment relative to his asset base, lump sum is preferred. If it's 50% of his asset base, a psychological crutch and wading in slowly may be entirely rational.

Garland Whizzer

larryswedroe
Posts: 15598
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by larryswedroe » Tue Mar 15, 2016 3:37 pm

Garland
I don't underestimate it's value, I use the example all the time of how it can help prevent people from being paralyzed. Now with that said one should know what the "correct" answer is in terms of the strategy that is MOST LIKELY to have the best results. Once you have that knowledge you can make an informed decision, knowing that DCA isn't optimal but it's better than alternative which might be paralysis.
Larry

User avatar
Rick Ferri
Posts: 8202
Joined: Mon Feb 26, 2007 11:40 am
Location: Vagabond, USA, Twitter: @Rick_Ferri
Contact:

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Rick Ferri » Tue Mar 15, 2016 3:59 pm

It would be a mistake to say one value strategy is "better" or "worse" based on what happened over the past six years.

All total stock market index funds have similar returns because they all hold the roughly the same securities in roughly the same cap weighted amount. Value investing is different. Value index providers have different security selection schemes and different weighing schemes. RAFI differs from DFA which differs from Vanguard, etc. Since value is in the eyes of the beholder, there will be different performance among the value funds at different times. This will rotate around. Sometimes one value strategy will outperform and then underperform. There is no consistency. Hence, it's not possible to predict which value strategy will outperform going forward.

Rick Ferri
The views expressed by Rick Ferri are strictly his own as a private investor and author and do not reflect the views of any entity or other persons.

Angst
Posts: 1641
Joined: Sat Jun 09, 2007 11:31 am
Location: St Louis, MO

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Angst » Tue Mar 15, 2016 8:23 pm

larryswedroe wrote:Theoretical
FWIW
Constantinides of U of Chicago wrote paper about 50 years ago showing that DCA isn't optimal, investing all at once is, and that hasn't been challenged since. DCA can be helpful in supplying a psychological crutch which at least allows people to invest in the face of uncertainty, which is easier for many to do with small amounts than large amounts
Larry

I agree with the notion that DCA can be an effective tool to enable certain investors overcome anxieties that otherwise might preclude their investing anything at all, but I daresay that the river in which some of us must timidly dip our toes rather than jump in all at once is quite different from that river of 50 years ago. High speed trading, high speed volatility and flash crashes are modern phenomena, and generally speaking, that which has long been considered correct often can reasonably be questioned again with the passage of time. That notwithstanding, I've always maintained that unless one is buying only one specific security or fund, going "all-in at once" is technically only a hypothetical or conceptual choice, not a real one. Ok, purchasing a combination of mutual funds all at today's closing price works, but if e.g. you're buying multiple ETFs today... well, if the market's flat today, perhaps that's close to "all-at-once", however if the market's steadily dropping, or maybe it's "flashing" about like a fish out of water, I say one cannot literally go "all-in at once". At least another Greek professor might agree with me here.

Angie

User avatar
in_reality
Posts: 4274
Joined: Fri Jul 12, 2013 6:13 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by in_reality » Tue Mar 15, 2016 11:16 pm

larryswedroe wrote:In_reality
What I've tried to point out is that I don't buy the RAFI strategy of weighting more to value when spreads are wider and less when narrower.


OK.

larryswedroe wrote:The reason is same you don't switch from stocks to bonds when P/Es are higher. A smaller spread just means you have an expected smaller value premium. Studies on subject confirm this.


Sure. Of course.

In my view, most are DCAing into cheap stocks already. We do this when stocks crash (boom), have a low (high) price, we are out of balance, and we rebalance. It's effectively incrementally increasing (decreasing) our exposure. When they boom, of course I think it would be ridiculous to sell out of stocks. I don't think selling out of stocks is analogous to having less value exposure after a value run though. It's more like rebalancing out of stocks after a good run which is a common practice.

Anyway, I am not suggesting the methodology is superior to or will outperform a traditional value index. I am just saying there are differences between value strategies that could account for their differing performance.

stlutz
Posts: 4013
Joined: Fri Jan 02, 2009 1:08 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by stlutz » Wed Mar 16, 2016 12:56 am

Here is one way RAFI got "lucky". They rebalance in March. In 2009 this meant that they made a pretty significant move into the big banks at the exact market bottom that year. Anybody who did the same made some great stock picks. Had they rebalanced in say, October, the numbers wouldn't have been as favorable.

Theoretical
Posts: 1068
Joined: Tue Aug 19, 2014 10:09 pm

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Theoretical » Wed Mar 16, 2016 12:58 am

I wasn't writing about the merits or demerits of Dollar Cost Averaging. What I was questioning in the RAFI article where it claims that their mean-reverting approach is like Dollar Cost Averaging for the value factor. To me, that seems completely backwards.

If I invest $100 per month in iShares S&P 600 Value, and 1/3 of the time value is cheap, 1/3 it's fair value, and 1/3 it's expensive, isn't that what DCA is talking about that you ignore the noise and invest in value regardless.

What Research Affiliates appears to be claiming (which seems like a redefinition of DCA) is that your $100 will be invested in value when it's cheap (maybe even over-flowed), even when it's fair value, and a bit less in value when it's overvalued. How that's Dollar Cost Averaging is baffling.

I like the concept of having some diversification in the value tilting world, but this seems like a big over-reach to use a classic concept like DCA (whether you use it or not) as a way to describe the variable value tilt. That's all.

User avatar
in_reality
Posts: 4274
Joined: Fri Jul 12, 2013 6:13 am

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by in_reality » Wed Mar 16, 2016 1:33 am

Theoretical wrote:What Research Affiliates appears to be claiming (which seems like a redefinition of DCA) is that your $100 will be invested in value when it's cheap (maybe even over-flowed), even when it's fair value, and a bit less in value when it's overvalued. How that's Dollar Cost Averaging is baffling.


I think it is a sensible description comparing it to the static value exposure in a traditional value fund.

My Russell Fundamental Emerging fund sets fundamental weights yearly based on economic performance. It then rebalances in 4 tranches over the year. So when energy was falling for a while, it was rebalancing to keep the energy stocks at their fundamental weight. Those same Energy stocks in other funds such as total market and value would have seen their allocation decrease (as a function of decreased market capitilazation).

When they set the yearly allocation, they could change each stock position at that time in one lump yearly sum. They do it in four tranches (25% at a time). In effect, I was increasing shares of energy stocks because their price was falling and the fund was trying to keep the allocation to whatever it was set to that year.

Maybe a better way to look at it is if a stock was 2% of a RAFI fund and dropped 50% in one day and stayed there, the fund would be DCAing over the year to bring it back up to 2% of the fund. Then that 2% target gets it's yearly evaluation based on fundamentals, and again the fund will be DCAing to move the stock to the new allocation.

You can argue this isn't something you want happening in your fund and that's fine. I do see it as DCAing though.

Park
Posts: 460
Joined: Sat Nov 06, 2010 4:56 pm

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by Park » Wed Mar 16, 2016 9:02 am

larryswedroe wrote:In_reality
What I've tried to point out is that I don't buy the RAFI strategy of weighting more to value when spreads are wider and less when narrower


How do RAFI weigh more to value when spreads are wider and less when narrower? The purpose of my question is not to disagree with the above statement. Instead, the question arises from my ignorance.

WasabiOsbourne
Posts: 308
Joined: Wed Mar 16, 2016 1:48 pm

Re: RAFI Has Beaten DFA and Vanguard (3% annually!) Since 2009 on an Absolute and Risk-Adjusted Basis. HOW?!?

Post by WasabiOsbourne » Wed Mar 16, 2016 3:08 pm

grunching... i.e. haven't read the thread...

major sector weight and country weight differences usually explains a ton of stuff...

this is not what u asked but i will say EW S500 vs. the real S&P 500 has big sector weight differences (i think)....,, energy weight vs technology weight usually explains alot, especially the further you go back

Post Reply