Small value underperformance, DFA

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Taylor Larimore
Advisory Board
Posts: 26899
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Joking?

Post by Taylor Larimore » Sat Jan 09, 2016 4:17 pm

The alternative is to be smart and listen to sales people pitch one market beating, risk mitigating, strategy after another: don't give up on the opportunity to beat the market, or to be safer than the market, using security selection. The proof is in the backtest.
Zotty:

I hope that you are joking. In my opinion, none of the above is true.

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

zotty
Posts: 841
Joined: Tue Sep 15, 2009 5:18 pm
Location: DFW, Texas

Re: Joking?

Post by zotty » Sat Jan 09, 2016 4:30 pm

Taylor Larimore wrote:
The alternative is to be smart and listen to sales people pitch one market beating, risk mitigating, strategy after another: don't give up on the opportunity to beat the market, or to be safer than the market, using security selection. The proof is in the backtest.
Zotty:

I hope that you are joking. In my opinion, none of the above is true.

Best wishes.
Taylor
yes. joking. I'm a firm believer in simplicity and total market.

i was just making a somewhat melodramatic point about the alternatives to total market investing. Any strategy that tilts from the market is an attempt to beat the market. it's a matter of degrees on the strategies discussed here. if costs and behaviors are under control, it's likely to be a successful plan.
Nadie Sabe Nada

User avatar
Taylor Larimore
Advisory Board
Posts: 26899
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Thankfully, it was a joke.

Post by Taylor Larimore » Sat Jan 09, 2016 4:56 pm

zotty:

Thank you for your reply. I was concerned that you might be serious. You described what lots of people believe. :(

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

paul merriman
Posts: 26
Joined: Fri Oct 25, 2013 1:00 am

Re: Small value underperformance, DFA

Post by paul merriman » Sat Jan 09, 2016 7:21 pm

It has been far too civil on the boards lately, so I thought it'd be fun to stir it up a little bit!

I imagine many investors did not become fans of small value tilting until about 2003.
I suspect that on an actual asset weighted basis, a lot of money has been lost/foregone by chasing the elusive small value premium. The underperformance over the last 10 years is pretty remarkable. What is particularly noteworthy is the strong performance of small growth over 10 years!
I don't have DFAs asset levels, but imagine their growth is a good indicator of the trend following. They have $380 billion in AUM today. I imagine it was far less in 2003.

Thoughts?!


---------------------------------------------- 1-Year 3-Year 5-Year 10-Year 15-Year
DFSVX (DFA small Value)---------------- -7.22-- 8.71 -- 8.08 -- 5.80 -- 10.16
VISGX (Vanguard small growth)-------- -2.90-- 9.89 -- 9.28 -- 7.72-- 8.73
VSCIX (Vangaurd small blend)---------- -3.36-- 10.72 --9.78 -- 7.53-- 8.61
VTSMX (Vangaurd total stock market)- 0.75-- 12.93-- 11.40 -- 7.02-- 5.63
S&P 500------------------------------------ 1.93--- 13.56 --11.93-- 6.95-- 5.03
How many years of performance are sufficient to make judgements about the returns? I suspect the academics would require a lot more than 15 years. I think it's important to understand what the academics believe about small cap value vs. S&P 500 returns. First, they make no predictions about how much small cap value and large cap blend (S&P 500) are likely to make. What they have said is there is a long term premium of small cap value over large cap stocks. The expectation for the small cap value return, compared to the S&P 500, is approximately 5%. It is interesting to note that over the 15 year period the difference between the S&P and small cap value return is about 5%. For short periods of time it is not surprising that large stocks beat small, or growth beats value. One year at a time small beats large about 53% of the time and value beats growth about 65% of the time.

lack_ey
Posts: 6454
Joined: Wed Nov 19, 2014 11:55 pm

Re: Small value underperformance, DFA

Post by lack_ey » Sat Jan 09, 2016 7:36 pm

That 5% difference came despite an unusually wide value spread to start the period. Start from around 1996 or 2003 instead (where it's similar to where it's at today) and the performance gap is considerably smaller, more like 2-3% between the S&P 500 and DFA Small Value.

As always, these things are sensitive to endpoints. And it's not very controversial to say that something weird was up in 2001. It's like backtesting long bonds since 1982. Something is obviously different about the starting conditions than the current conditions.

It's well possible that SV has been unusually bad over this period other than the 2000-2002 deflation of tech because of how inflation and growth turned out, but if taking 2000-2015 seriously then you really need to be honest about what happened there.

User avatar
bertilak
Posts: 5883
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Small value underperformance, DFA

Post by bertilak » Sat Jan 09, 2016 7:39 pm

paul merriman wrote:How many years of performance are sufficient to make judgements about the returns? I suspect the academics would require a lot more than 15 years. I think it's important to understand what the academics believe about small cap value vs. S&P 500 returns. First, they make no predictions about how much small cap value and large cap blend (S&P 500) are likely to make. What they have said is there is a long term premium of small cap value over large cap stocks. The expectation for the small cap value return, compared to the S&P 500, is approximately 5%. It is interesting to note that over the 15 year period the difference between the S&P and small cap value return is about 5%. For short periods of time it is not surprising that large stocks beat small, or growth beats value. One year at a time small beats large about 53% of the time and value beats growth about 65% of the time.
Paul,

As someone who is 70 years old, retired, and making no new contributions I think SCV tilt is not appropriate for me as I don't want to wait too long to turn an ongoing loss into an eventual profit. I may have spent it all before the profit shows up. This thinking goes for other "tilts" like EM.

It seems there is more to risk than just risk vs return -- but something to do with how long is long term.

Am I thinking the right way about this? If so, is there a more formal way to express, or label, this?
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker, the Cowboy Poet

User avatar
William Million
Posts: 544
Joined: Wed May 05, 2010 4:41 am
Location: A Deep Mountain

Re: Small value underperformance, DFA

Post by William Million » Sat Jan 09, 2016 8:00 pm

On the other hand, now is an opportune time for someone who believes in SV premium to jump in.

In the past, I often pointed out that although I actually believe in the SV premium, it is not right for the vast majority of investors. Some will give up after 5 years of under-performance, some after 10 years, some after 15 years . . . . They won't be around to benefit for the big gains in year 18.

Theoretical
Posts: 1242
Joined: Tue Aug 19, 2014 10:09 pm

Re: Small value underperformance, DFA

Post by Theoretical » Sun Jan 10, 2016 12:30 am

Conceptually, I think small value makes sense as the counterweight to the exuberance of the new "paradigm-shifting" "tech stocks" of whatever era you are in. If you were in the 1870s, it was the railroads. In the 1900s, it was the big manufacturing and oil companies. Then came the consumer industry with the baby boom. Most recently, you've had the tech boom and now the more mature Internet boom.

Next could be biotech or some crazy technology we've not thought up, but there almost certainly will be another. Could it be that, value, perhaps more than small, is a contrarian counterweight in those times?

lack_ey
Posts: 6454
Joined: Wed Nov 19, 2014 11:55 pm

Re: Small value underperformance, DFA

Post by lack_ey » Sun Jan 10, 2016 12:56 am

Theoretical wrote:Conceptually, I think small value makes sense as the counterweight to the exuberance of the new "paradigm-shifting" "tech stocks" of whatever era you are in. If you were in the 1870s, it was the railroads. In the 1900s, it was the big manufacturing and oil companies. Then came the consumer industry with the baby boom. Most recently, you've had the tech boom and now the more mature Internet boom.

Next could be biotech or some crazy technology we've not thought up, but there almost certainly will be another. Could it be that, value, perhaps more than small, is a contrarian counterweight in those times?
If this were the case and the primary driving effect you'd probably expect the value effect to be relatively nonlinear in book/market with the bottom quintile or so (most growth) being significantly worse than the others with not a huge difference between say the middle quintile and the top quintile (most value). But that's not what the data really looks like.

Erwin
Posts: 1929
Joined: Fri Apr 27, 2007 11:16 pm

Re: Small value underperformance, DFA

Post by Erwin » Sun Jan 10, 2016 3:30 am

bertilak wrote:
paul merriman wrote:How many years of performance are sufficient to make judgements about the returns? I suspect the academics would require a lot more than 15 years. I think it's important to understand what the academics believe about small cap value vs. S&P 500 returns. First, they make no predictions about how much small cap value and large cap blend (S&P 500) are likely to make. What they have said is there is a long term premium of small cap value over large cap stocks. The expectation for the small cap value return, compared to the S&P 500, is approximately 5%. It is interesting to note that over the 15 year period the difference between the S&P and small cap value return is about 5%. For short periods of time it is not surprising that large stocks beat small, or growth beats value. One year at a time small beats large about 53% of the time and value beats growth about 65% of the time.
Paul,

As someone who is 70 years old, retired, and making no new contributions I think SCV tilt is not appropriate for me as I don't want to wait too long to turn an ongoing loss into an eventual profit. I may have spent it all before the profit shows up. This thinking goes for other "tilts" like EM.

It seems there is more to risk than just risk vs return -- but something to do with how long is long term.

Am I thinking the right way about this? If so, is there a more formal way to express, or label, this?
right on!
Erwin

larryswedroe
Posts: 15613
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Small value underperformance, DFA

Post by larryswedroe » Sun Jan 10, 2016 8:54 am

Bertilak
I often hear that line of thinking, but I suggest another way to think about it as IMO it's EXACTLY backwards.

Over the long term the differences in returns between asset classes is relatively small with the S&P 500 underperforming SV by say 4% a year.

The shorter the horizon the wider the dispersion of returns gets. So take for example 2000-02 when SV outperformed by 26% a year, or 2000-09 when it outperformed by over 13% a year.

Thus, the shorter the horizon the more important diversification across asset classes becomes and thus investing in TSM leaves you with no diversification across the factors.

It's the same thing with retirees, they should be the very ones who use the Larry Portfolio as it cuts tail risks which increase when in withdrawal phase, and you need to tilt to SV to allow for holding less beta.

I hope that is helpful
Larry

User avatar
Taylor Larimore
Advisory Board
Posts: 26899
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Re: Small value underperformance, DFA

Post by Taylor Larimore » Sun Jan 10, 2016 10:01 am

Bogleheads:

We can read Mr. Bogle's thoughts about "Slice & Dice" and "Small Value" here:

THE TELLTALE CHART.

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

User avatar
bertilak
Posts: 5883
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Small value underperformance, DFA

Post by bertilak » Sun Jan 10, 2016 10:23 am

larryswedroe wrote:It's the same thing with retirees, they should be the very ones who use the Larry Portfolio as it cuts tail risks which increase when in withdrawal phase, and you need to tilt to SV to allow for holding less beta.

I hope that is helpful
Larry
It might help with a little follow-up! I's just not yet clear in my mind.

I can't reconcile that with Paul Merriman's "requir[ing] a lot more than 15 years" to see the SCV premium. Is this a disagreement between you and Paul or (more likely) my misunderstanding?

Thanks.
B.
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker, the Cowboy Poet

randomguy
Posts: 5548
Joined: Wed Sep 17, 2014 9:00 am

Re: Small value underperformance, DFA

Post by randomguy » Sun Jan 10, 2016 10:44 am

bertilak wrote:
larryswedroe wrote:It's the same thing with retirees, they should be the very ones who use the Larry Portfolio as it cuts tail risks which increase when in withdrawal phase, and you need to tilt to SV to allow for holding less beta.

I hope that is helpful
Larry
It might help with a little follow-up! I's just not yet clear in my mind.

I can't reconcile that with Paul Merriman's "requir[ing] a lot more than 15 years" to see the SCV premium. Is this a disagreement between you and Paul or (more likely) my misunderstanding?

Thanks.
B.
Imagine that these are 2 return possibilities for the next 10 years
a) total returns 10%, SV returns 0%
b) SV returns 10%, total returns 0%

which portfolio would you rather hold
1) 100% total market
2) 100% SV
3) 50/50 total/SV
?

3 is low risk play (you get 5% no matter what) while 1+2 are the high risk plays (get 10% or 0%). And before you go that is a made up example that could never happen, look at : https://www.portfoliovisualizer.com/bac ... entage=0.0

User avatar
Robert T
Posts: 2509
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: Small value underperformance, DFA

Post by Robert T » Sun Jan 10, 2016 10:52 am

.
I would caution views that the equity premium is more immune to underperformance than the value premium. According to the earlier linked Fama-French paper Volatility and Premiums – even over a 50-year investment lifetime, an investor has a one-in-twenty chance of a negative realized equity premium, which is even higher than the one-in-thirty five chance of a negative realized value premium (if you believe the Fama-French research). And over the last 16 years (2000-2015) Vanguard Total Bond Market has outperformed Vanguard Total Stock market by 0.65% annualized ( 5.11% vs. 4.46%) obviously highly period dependent – but that’s the point – it can happen.

And if I look at the table Taylor posted for the 15 years to 6-30-2000, Small cap value returned 11.3% while large cap blend returned 15.2%. And Larry posted the returns 15-years to date with SV = 9.9%, S&P500 = 4.8%. So across the two periods SV had returns of 11.3% and 9.9%, while large cap = 15.2% vs. 4.8%. To me, SV provided more consistent returns over the two periods, even if it underperformed LB in the 15 yrs to mid-2000 (I will take an 11.3% return any day). Over these periods the value premium had low correlation with the equity premium providing a diversification benefit (i.e. lower deviations in returns over the two periods). However, the challenge is that correlations across factors premiums varies over time, and in some cases the correlations become increasingly positive just at the wrong time – as in 1929-1932.

Here are some returns over three bear markets: 1929-1932, 1973-74, 2000-2002, together with returns for the 10 years since the start of the bear market.

1929-1932: LV and SV declined by more than the overall market (deflation increased the real cost of debt which is higher in LV companies, and even more so in SV companies). Over the 10 year period from 1929-1938, nominal annualized returns of LV was still negative but better than the overall market (-0.9%, and -1.6% annualized), and both had positive real returns given the -2.0% inflation rate over the period. Small Value lagged both, with a -6.7% nominal annualized return. In sum: large value had better performance than small value over this period, declining less in 1929-1932 and having higher 10 yr annualized returns from 1929-1932 than both SV and the overall market.

1973-74: LV declined by less than the overall market (-12.4% vs. -22.7% annualized), while SV decline by slightly more than the overall market (-24.3% vs. -22.7% annualized). Given the +10% inflation rate over this period, real returns were much worse. Over the 10 year period 1973-1982, SV and LV both outperformed the market (20.2% and 13.5% vs. 7.5% annualized return). And with the +8.7% annualized inflation rate over this period, the real annualized market return was still negative after 10 years. In sum: in this period, LV provided greater downside diversification in 1973-74, while SV had higher annualized returns over 1973-1982.

2000-02: Both LV and SV had positive annualized returns over this period (+0.2% vs. +12.1%) while the market declined (-14.7% annualized). Over the 10yr period 2000-2009, the annualized nominal market return was -0.9%, which was lower in real terms given the +2.5% annualized inflation rate, while the return from LV and SV were +4.0% and +12.7% respectively. In sum: in this this period SV provided greater downside diversification in 2000-02, and provide higher annualized returns over the 2000-09.

Annualized return (%)

………..…...…TSM….……LV…….….SV……Inflation
1929-1932 -25.3% / -31.2% / -37.9% / -6.5%
1973-1974 -22.7% / -12.4% / -24.3% / +10.5%
2000-2002 -14.7% / +0.2% / +12.1% / +2.4%

………..…...…TSM….……LV…….….SV……Inflation
1929-1938 ..-1.6% / ...-0.9% / ..-6.7% / -2.0%
1973-1982 ..+7.5% / +13.5% / +20.2% / +8.7%
2000-2009 ..-0.3% / ..+4.0% / +12.7% / +2.5%

TSM = CRSP1-10
LV = Dimensional US Large Value Index
SV = Dimensional US Small Value Index

What do I conclude from the above?: The diversification benefits of LV and SV have varied across market downturns and over different timeframes (i.e. during actual downturn vs. 10yrs from start of downturn). LV provided better 10yr diversification from 1929-1938 than SV, but both amplified 1929-1932 declines. LV provided better diversification in 1973-74, while SV had best 1973-1982 returns. SV provided best downside diversification in 2000-02 and higher 10 year returns from 2000-09. Given these differing effects, personally, I take a middle (value) path - with 50:50 LV:SV type combination – rather than all LV or all SV (or all TSM). For what its worth a 50:50 LV:SV combination had higher annualized returns than each individually in the 17 years since 1927 when inflation was above 5 percent. This broadly equates to a value load about double the size load. And adding longer term bonds (personally would not extend beyond intermediate) can help in 1929-1932 type periods.

We each need make our own choices. If TSM is what you are comfortable with and can stick with it - then go with that - but would caution thinking that the equity premium is more immune to underperformance than the value premium, as this could lead to disappointment.

Obviously no guarantees.

Robert
.

User avatar
bertilak
Posts: 5883
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Small value underperformance, DFA

Post by bertilak » Sun Jan 10, 2016 11:46 am

randomguy wrote: Imagine that these are 2 return possibilities for the next 10 years
a) total returns 10%, SV returns 0%
b) SV returns 10%, total returns 0%

which portfolio would you rather hold
1) 100% total market
2) 100% SV
3) 50/50 total/SV
?
Which is more likely, a) or b)? I think that may factor in to the answer.

I guess it boils down to whether or not SCV is inherently more risky, and by how much. Everyone who promotes its use says it is precisely because it's more risky that the returns are higher. Yes, some say there is at least a little free lunch there, or as Larry puts it, a free trip to the salad bar, but I think most agree it is a risk story that produces most of the extra return.

I am wondering if I want to accept risk (even fully, or possibly even excessively, rewarded risk) if the corresponding reward is most likely to arrive beyond my investment horizon. Also, if it reduces left-side fat tail is that reduction ALSO beyond my investment horizon?

Note: I am not trying to argue here (I know it sounds that way) I'm just trying to sort out the concepts, which I know I have not fully internalized. (Some might remember that notorious DCA thread, the one before the "ultimate" DCA thread. I initially felt that DCA must be the thing to do as it obviously reduced risk. It took me a while to "get it." Could be I'm in the same situation here.)
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker, the Cowboy Poet

User avatar
Kenkat
Posts: 4058
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: Small value underperformance, DFA

Post by Kenkat » Sun Jan 10, 2016 11:56 am

I want to take a quick minute to thank all of the posters on this thread for all the effort put into this discussion - both pro and con. Truly invaluable stuff to think about here.

User avatar
Ketawa
Posts: 1918
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Small value underperformance, DFA

Post by Ketawa » Sun Jan 10, 2016 12:05 pm

Aish wrote:
Ketawa wrote:
gwrvmd wrote:Daniel is the only person I have read who is willing to call it out and go against the tide. To get the real Small Value Tilt you have to go through a DFA Advisor and buy DFA funds to get the "Secret Sauce" which then does not outperform plain indexing
Criticism of small and value tilts is common on the forum, as well as other tilts.

There are many ways to get a small value tilt. Robert T has long had an example ETF portfolio set up in Morningstar designed to mirror the factor tilts of a DFA portfolio. The performance has been basically identical. So, tilting using ETFs is completely viable. And now there are funds from AQR, which is a fund provider similar to DFA, that are accessible to retail investors and held by many on the board.
Could you provide a link to Robert T's example portfolio? Thanks in advance.
They are linked from his Collective thoughts [investing mini-reference] thread.

ETF portfolio: http://socialize.morningstar.com/NewSoc ... C1FF7784FC
DFA portfolio: http://socialize.morningstar.com/NewSoc ... 7926991271

randomguy
Posts: 5548
Joined: Wed Sep 17, 2014 9:00 am

Re: Small value underperformance, DFA

Post by randomguy » Sun Jan 10, 2016 12:29 pm

bertilak wrote:
randomguy wrote: Imagine that these are 2 return possibilities for the next 10 years
a) total returns 10%, SV returns 0%
b) SV returns 10%, total returns 0%

which portfolio would you rather hold
1) 100% total market
2) 100% SV
3) 50/50 total/SV
?
Which is more likely, a) or b)? I think that may factor in to the answer.

To some extent it doesn't matter. Lets say 1) happens 75% of the time an b happens 25% . If you could play the game 1000 times you would go with portfolio 1 as your expected value would be 7.5% versus the 5% of 3. However you would have to decide if the 25% cases where you got 0% was worth the added gain.

Obviously in the real world things are a lot messier than that simple example but it shows why you might want to diversify across size, value (or any other factor you like) versus betting it all on beta. It isn't going to totally smooth out volatility but it helps a bit.

lack_ey
Posts: 6454
Joined: Wed Nov 19, 2014 11:55 pm

Re: Small value underperformance, DFA

Post by lack_ey » Sun Jan 10, 2016 12:31 pm

bertilak wrote:I am wondering if I want to accept risk (even fully, or possibly even excessively, rewarded risk) if the corresponding reward is most likely to arrive beyond my investment horizon. Also, if it reduces left-side fat tail is that reduction ALSO beyond my investment horizon?
SCV is riskier. If nothing else, it is more concentrated in fewer stocks, but there's much more to it than that. The tails are longer, if only by the nature of higher volatility. As such, you're not reducing the left-side tail.

You're only potentially reducing the left-side tail if you replace X stocks with say 0.6X SCV and 0.4X bonds or similar. This is really more a testament to diversification and holding more safe assets than it is to SCV. In general if you suppose SCV return is higher then there is some mix of SCV and bonds that will have the same expected returns as total stock. But if you do this, then you're not actually expected to have higher returns, just less volatile ones, and even that depends on the distributions you assume for all the asset classes involved.

In reality you're looking at an asset allocation that already has both stocks and bonds vs. one tilted to SCV and more bonds, where the extra diversification effect from the extra bonds is lower. But still, unless your starting position is something like 30% stocks or maybe lower, there's room for extra bonds to improve risk/return, under some understandings of that. (And also, SCV mixes with bonds in a slightly different way than total stock does.)

You mention the horizon multiple times but this is largely confusing strategy with outcome. If you don't get a good outcome in your investment horizon, that's bad luck. But that can happen for any risk asset, including total stock. The left-side tail is just a statistical representation of some bad outcomes. If you can shift the distribution of possible futures to be more favorable, that is a good strategy and prudent now, regardless of whatever outcome actually happens. In fact, you hope to not get the left tail regardless of your asset allocation.

If you believe that future returns for SCV relative to the total market will be similar to what many people here say, then "the corresponding reward" is around more likely than not (higher than 50%) to arrive in any investment horizon. It is just increasingly likely the longer the horizon is.

larryswedroe
Posts: 15613
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Small value underperformance, DFA

Post by larryswedroe » Sun Jan 10, 2016 1:03 pm

bertilak
First SV is clearly more risky in isolation. The SD is about 35% vs. 20% for the market. That's one reason why SV should have higher expected returns.
Another is SV risks are more correlated with the economy, and value stocks in general have more leverage and higher SD of earnings. And there is more liqudity risk in small stocks.

Now as I pointed out, when you combine SV with high quality bonds and use the higher expected returns of SV to lower your overall equity allocation then the PORTFOLIO has been less risky, and that is what matters. Now that's not a guarantee of course that the future will be the same. But IMO the logic is there to have that expectation due to the way assets mix. Again, might want to read Reducing the Risks of Black Swans. I don't know what the returns will be but the left tail risk will almost certainly be lower, and likely much lower (which is why IMO retirees should consider the Larry Portfolio strategy even more than others).

As to me and Paul. There is no disagreement. There is no guarantee that SV will outperform over any period, any more than there is with TSM outperforming safe Treasuries. In fact based on my observations there are much longer periods of TSM underperforming than SV underperforming. Try 40 years from 69-08 when large stocks underperformed long-term Treasuries. There's no 40 year period when SV underperformed, or even close to it.

This line of thinking is IMO the mistake so many make. They think of risks as "one sided coin", whether it comes to SV vs. TSM, US stocks vs. international stocks, currency risk, etc. As example, your focus is on risk of SV underperforming when there is equal or greater risk of TSM underperforming. Just suggesting that one should consider both sides of the risk coin.

So the longer the time frame, the MORE LIKELY (but never certain) that a higher expected returning asset class will outperform. But also the shorter the horizon the WIDER the potential DISPERSION of outcomes is. Thus diversification is IMO always the prudent strategy, always, but in fact it's more important IMO the shorter the horizon because the dispersion of possible outcomes is wider.

I hope that clarifies things for you.

Best wishes
Larry

User avatar
bertilak
Posts: 5883
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Small value underperformance, DFA

Post by bertilak » Sun Jan 10, 2016 2:59 pm

larryswedroe wrote:So the longer the time frame, the MORE LIKELY (but never certain) that a higher expected returning asset class will outperform. But also the shorter the horizon the WIDER the potential DISPERSION of outcomes is. Thus diversification is IMO always the prudent strategy, always, but in fact it's more important IMO the shorter the horizon because the dispersion of possible outcomes is wider.

I hope that clarifies things for you.

Best wishes
Larry
Thanks again. Things are getting clearer (MPT, right?) but another question: Why is wider DISPERSION of outcomes good?

I thought that by lessening the dispersion one could avoid (to some degree) devastating, sever, downside losses at the expense of gigantic wins on the upside. (potentials for; no guarantees)
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker, the Cowboy Poet

lack_ey
Posts: 6454
Joined: Wed Nov 19, 2014 11:55 pm

Re: Small value underperformance, DFA

Post by lack_ey » Sun Jan 10, 2016 3:09 pm

bertilak wrote:
larryswedroe wrote:So the longer the time frame, the MORE LIKELY (but never certain) that a higher expected returning asset class will outperform. But also the shorter the horizon the WIDER the potential DISPERSION of outcomes is. Thus diversification is IMO always the prudent strategy, always, but in fact it's more important IMO the shorter the horizon because the dispersion of possible outcomes is wider.

I hope that clarifies things for you.

Best wishes
Larry
Thanks again. Things are getting clearer (MPT, right?) but another question: Why is wider DISPERSION of outcomes good?

I thought that by lessening the dispersion one could avoid (to some degree) devastating, sever, downside losses at the expense of gigantic wins on the upside. (potentials for; no guarantees)
You're reading that backwards. Reducing dispersion (which is what diversification does) is more important when dispersion is wider.

larryswedroe
Posts: 15613
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Small value underperformance, DFA

Post by larryswedroe » Sun Jan 10, 2016 3:29 pm

bertilak
As Lack_ey stated you had it backwards, wide dispersion outcomes is what DIVERSIFICATION REDUCES. So if your horizon is short, when wide dispersions of outcomes is greatest, and you want to minimize that you should diversify across asset classes

Larry

User avatar
ogd
Posts: 4854
Joined: Thu Jun 14, 2012 11:43 pm

Re: Small value underperformance, DFA

Post by ogd » Sun Jan 10, 2016 9:09 pm

lack_ey wrote: But what's the mechanism by which the market would smarten up and arbitrage away any value effect over perhaps what might be expected from extra risk? Any such move should be reflected in a reduction of the value spread—value companies getting bid up to higher valuations than before, with growth companies getting bid down. The ratio of BE/ME of value compared to the BE/ME of growth should be smaller now by that view, for example. Pick your own value metrics. However, that ratio is now at typical levels (was very wide during the tech bubble), not compressed relative to history.
Sorry, but I don't believe in watching historical metrics to guess whether value is still a good deal. (Or stocks for that matter). This immediately leads to market timing. When the market kills value, it might do it at the same time as it makes your ratio meaningless. Many stock/bond measures have suffered the same fate.

Nor do I think that value is some kind of tautology -- that if the market overvalues a value company, it will cease to be value and thus the value buyer is still good. The market can err with deep enough value that has some room before it emerges from value. Room it wouldn't have if its value-ness wasn't a thing (e.g. bum company).

lack_ey
Posts: 6454
Joined: Wed Nov 19, 2014 11:55 pm

Re: Small value underperformance, DFA

Post by lack_ey » Sun Jan 10, 2016 9:41 pm

ogd wrote:
lack_ey wrote: But what's the mechanism by which the market would smarten up and arbitrage away any value effect over perhaps what might be expected from extra risk? Any such move should be reflected in a reduction of the value spread—value companies getting bid up to higher valuations than before, with growth companies getting bid down. The ratio of BE/ME of value compared to the BE/ME of growth should be smaller now by that view, for example. Pick your own value metrics. However, that ratio is now at typical levels (was very wide during the tech bubble), not compressed relative to history.
Sorry, but I don't believe in watching historical metrics to guess whether value is still a good deal. (Or stocks for that matter). This immediately leads to market timing. When the market kills value, it might do it at the same time as it makes your ratio meaningless. Many stock/bond measures have suffered the same fate.

Nor do I think that value is some kind of tautology -- that if the market overvalues a value company, it will cease to be value and thus the value buyer is still good. The market can err with deep enough value that has some room before it emerges from value. Room it wouldn't have if its value-ness wasn't a thing (e.g. bum company).
So what is your non-belief in value, if we want to call it, based on? If it's just the story of "oh, everyone knows about it and too many people are chasing it now" then there are lots of stories for all kinds of things in investing. But where's the data? What is the thought process and line of reasoning?

Off the top of my head, here are some ways value in stocks could effectively disappear:
1. The excess returns in the past were a statistical fluke
2. Underlying economic conditions that relatively favor value companies over growth become less prevalent than in the past
3. Companies with higher value scores (higher BE/ME by Fama-French) get bid up compared to growth companies, equalizing out future returns

Can you think of more? For now I'm moving on.

If the market either got smart or for whatever reason a lot of people chase value, then we expect value companies to become more expensive, lowering the BE/ME relative to growth companies, relative to an alternate universe where there isn't a rush into value.

I'm just saying that, by the most direct and naive measure possible, this doesn't appear to have happened. As I said before, it could be a number of things, and maybe it actually has happened but just has been masked by another effect (today's distribution of companies being different from in the past, which is entirely believable). I'm not even saying that this measure is very predictive of future value returns at all. Just that the data on casual inspection doesn't quite seem to fit the narrative of people having figured out value and overbought (or correctly bought) in.

On the other hand, you can check the data and see that today, high dividend payers appear to be bought up relative to the market (they're less tilted to value than typical), which fits into the narrative of investors reaching for yield, bidding these prices up. Of course, maybe there's something else going on there. Cloudy crystal balls and all, regardless.

If anyone has alternate explanations and theories, I'd like to hear them because I miss a lot of stuff on my own.

User avatar
nisiprius
Advisory Board
Posts: 35704
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: Small value underperformance, DFA

Post by nisiprius » Sun Jan 10, 2016 10:39 pm

lack_ey wrote:Vanguard has switched indexes multiple times. In fact, index hopping in of itself is one of the more frustrating aspects to deal with for tilters.
That really shouldn't matter.

If there's any reality to the value factor, then competent index providers should produce indexes that are reasonably consistent, equivalent, and interchangeable. If they aren't, then there's something wrong with the concept of "value." Flipping from one implementation of the value factor to another shouldn't completely wipe out the value premium unless that premium was pretty fragile to begin with. If the value factor is so badly defined that different authorities can't agree how to measure it, or what should be included in it, or if the definition keeps changing over time, then it's not very credible.

It's like saying that the reason car X didn't get better fuel economy than car Y is that the driver of car X changed gasoline brands several times. No, if claims that car X has superior fuel economy are valid, then as long as the gasoline meets the maker's octane specifications, you should get the improved fuel economy.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
timboktoo
Posts: 698
Joined: Sun Mar 31, 2013 3:42 pm

Re: Small value underperformance, DFA

Post by timboktoo » Sun Jan 10, 2016 10:42 pm

I don't do small value, but there seems to be good support for it. Every asset has its day.

- Tim

zotty
Posts: 841
Joined: Tue Sep 15, 2009 5:18 pm
Location: DFW, Texas

Re: Small value underperformance, DFA

Post by zotty » Sun Jan 10, 2016 11:14 pm

lack_ey wrote: 1. The excess returns in the past were a statistical fluke
I'll explain what i mean by fluke.

if past performance is used to predict future returns, this is an example of a predictive model.

In most non-trivial domains, building a predictive model is tough work. the sample set is critical if I want my model to perform above random. It's a high bar and gets short tilt by people who haven't been burned by it. For those of us who get burned by this regularly, it even has a name: garbage in, garbage out. practical statistics work is often not about numbers, but about creatively designing the universe/sample space.

This almost always involves trying to reduce data dependence to an absolute minimum. If dependence is too high, the predictive capability of the model is damaged.

Relying on annual data seems unwise for that reason. At best, the predictive system would reduce to modelling business cycles with everything else discarded as noise. Typically, the model will end up skewed toward the most dependent samples. in either case, the sample size, if effectively quantized to independent samples, is way too small to develop a predictive model any nontrivial domain.

"business cycles" feels like the right sample quantization to minimize data dependence and construct a non-garbage statistical model, but 15 isn't enough. honestly, a 100 independent data points is not enough, but it's more than 15.
Nadie Sabe Nada

lack_ey
Posts: 6454
Joined: Wed Nov 19, 2014 11:55 pm

Re: Small value underperformance, DFA

Post by lack_ey » Sun Jan 10, 2016 11:34 pm

nisiprius wrote:
lack_ey wrote:Vanguard has switched indexes multiple times. In fact, index hopping in of itself is one of the more frustrating aspects to deal with for tilters.
That really shouldn't matter.

If there's any reality to the value factor, then competent index providers should produce indexes that are reasonably consistent, equivalent, and interchangeable. If they aren't, then there's something wrong with the concept of "value." Flipping from one implementation of the value factor to another shouldn't completely wipe out the value premium unless that premium was pretty fragile to begin with. If the value factor is so badly defined that different authorities can't agree how to measure it, or what should be included in it, or if the definition keeps changing over time, then it's not very credible.

It's like saying that the reason car X didn't get better fuel economy than car Y is that the driver of car X changed gasoline brands several times. No, if claims that car X has superior fuel economy are valid, then as long as the gasoline meets the maker's octane specifications, you should get the improved fuel economy.
Okay, first of all you're addressing a different issue than I was; my original response was taken a bit out of context and to be honest a one-liner like that does leave itself up to misunderstandings.

You're focusing on interpretations of value itself. The argument is that different value funds should be able to capture the value effect, if there is anything to be captured. If there is outperformance to be had, then outperformance should ensue, with any reasonable implementation, right? Most formulations, whether you use price/book, price/earnings, some combination of a bunch of stuff, etc. are all highly correlated and produce similar but definitely different returns. Historically, because Fama-French first used it because of its performance record over the period studied, price/book (book/market, actually) was used. This is covered in many articles including this one:
http://www.etf.com/sections/index-inves ... nopaging=1

However, differences between value funds go well beyond just the measure of value used. Should a value fund own the highest-value third of a market or a half? And by "half" do we mean half of the companies or half by market capitalization? In Four Pillars (2002), Bill Bernstein complains about "blend contamination" in Vanguard value funds. And... should a value fund try to have zero momentum or should it just focus on value purely and let everything else fall as it will?

Nowadays, the fundamental index types aren't cap weighting constituents but are getting value exposure weighting by fundamentals. Some value funds split sectors in half so as to be sector neutral, while oldschool formulations take on significant sector tilts. It goes on and on.

Also, I was really speaking to the broader issue of tilting, which includes multifactor funds, where now you have disagreements on measures for multiple factors! There's also an additional layer of complication in multifactor funds: do you split the N-factor portfolio into N mini-portfolios each focusing on a different factor, or do you have one giant pile that's only the securities that each rank highly on a composite of multiple factors? Some funds do it the first way, and others do it the second way.
---
It is not a problem that there are multiple competing definitions for value. It does not diminish the impact of any original or other definition of value. When writing papers, sure, it is customary to use Fama-French for consistency with the literature and honesty of statistical tests. But for example index providers have an incentive to try to come up with differentiating details and hopefully capture a better representation of the underlying economic effect. It's fine as long as we know what they're doing. The issue for an investor is when your fund's new index is doing something you don't want and you can't switch out without selling and taking the capital gains tax hit. Which is why I'm not a huge fan of tilting in taxable.
---
The comparison everyone likes to bring up is Vanguard small-cap value index fund vs. Vanguard small-cap growth index fund, which is what my original response was targeted at. As pointed out by Robert T, the latter has managed to have positive value exposure over its lifetime (by Fama-French 3-factor regression). The size loading was a little different too. Exposures change over time, especially with changing indexes, and it does matter some. Some of this explains some of the difference between raw (long-short) factor returns and those of any given tilted fund. There are plenty of other reasons why that happens too.

There can be a lot more than just gasoline changes going on, in any case.

zotty wrote:
lack_ey wrote: 1. The excess returns in the past were a statistical fluke
I'll explain what i mean by fluke.

if past performance is used to predict future returns, this is an example of a predictive model.

In most non-trivial domains, building a predictive model is tough work. the sample set is critical if I want my model to perform above random. It's a high bar and gets short tilt by people who haven't been burned by it. For those of us who get burned by this regularly, it even has a name: garbage in, garbage out. practical statistics work is often not about numbers, but about creatively designing the universe/sample space.

This almost always involves trying to reduce data dependence to an absolute minimum. If dependence is too high, the predictive capability of the model is damaged.

Relying on annual data seems unwise for that reason. At best, the predictive system would reduce to modelling business cycles with everything else discarded as noise. Typically, the model will end up skewed toward the most dependent samples. in either case, the sample size, if effectively quantized to independent samples, is way too small to develop a predictive model any nontrivial domain.

"business cycles" feels like the right sample quantization to minimize data dependence and construct a non-garbage statistical model, but 15 isn't enough. honestly, a 100 independent data points is not enough, but it's more than 15.
Are even business cycles independent?

Financial data has too few samples to begin with, so is going from monthly/quarterly/annual to business cycles going to help? You can clean up the data some but now you basically don't have any samples and have no power. We can't build any actually good models with what we have, but that doesn't mean every poorly constructed model is completely wrong and chasing noise.

With value you can also look at results in other countries and in asset classes other than stocks, but all these things are somewhat or a lot dependent. It's hard to disentangle.

zotty
Posts: 841
Joined: Tue Sep 15, 2009 5:18 pm
Location: DFW, Texas

Re: Small value underperformance, DFA

Post by zotty » Mon Jan 11, 2016 12:48 am

lack_ey wrote: Are even business cycles independent?
definitely not, but it should be less than annual, quarterly, intraday, etc. business cycle bookends have an inflectional characteristic to them. A marker that something significant changed about the market and economy. data points in the middle of a cycle have more in common with each other than do data points before/after an inflection (using business cycle as a proxy)

Using international data seems like a reasonable thing to do with some functional work, like incorporating samples from countries whose markets were least correlated to the US market at the time of sampling.

interesting.
Nadie Sabe Nada

Call_Me_Op
Posts: 6793
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Re: Small value underperformance, DFA

Post by Call_Me_Op » Mon Jan 11, 2016 6:47 am

bertilak wrote:
larryswedroe wrote:It's the same thing with retirees, they should be the very ones who use the Larry Portfolio as it cuts tail risks which increase when in withdrawal phase, and you need to tilt to SV to allow for holding less beta.

I hope that is helpful
Larry
It might help with a little follow-up! I's just not yet clear in my mind.

I can't reconcile that with Paul Merriman's "requir[ing] a lot more than 15 years" to see the SCV premium. Is this a disagreement between you and Paul or (more likely) my misunderstanding?
All Paul is saying is you have to wait for at least 15 years (or whatever the number is, probably closer to 20 years) to be guaranteed that you will realize the SV premium (based upon historical performance). It may show-up in 1 year, though.
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein

larryswedroe
Posts: 15613
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Small value underperformance, DFA

Post by larryswedroe » Mon Jan 11, 2016 8:27 am

Nisiprius
If there's any reality to the value factor, then competent index providers should produce indexes that are reasonably consistent, equivalent, and interchangeable. If they aren't, then there's something wrong with the concept of "value."
IMO this is incorrect. For example, consider that small factor has generally been used as return of top half minus return of bottom half. Why, I don't know. But value by academics has been top and bottom 30%, while middle is referred to as core. But for example many split value in same way that they split small and large, so top and bottom half. And different indices will get you different loadings on value factor as well.

Also funds can be run very differently depending on their fund construction rules. Some might choose top and bottom halfs, others top and bottom 30%, and others top and bottom 25%, and so on. Doesn't make one right or wrong, just different.

One therefore must be careful in reading papers to make sure they know/understand how value is defined, or size, or any other factor, and also choose funds based on how much exposure/diversification, etc one wants to each factor.

Larry

User avatar
abuss368
Posts: 12381
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!

Re: Small value underperformance, DFA

Post by abuss368 » Mon Jan 11, 2016 4:08 pm

Taylor Larimore wrote:Bogleheads:

We can read Mr. Bogle's thoughts about "Slice & Dice" and "Small Value" here:

THE TELLTALE CHART.

Best wishes.
Taylor
One of my favorite speeches!
John C. Bogle: "You simply do not need to put your money into 8 different mutual funds!" | | Disclosure: Three Fund Portfolio + U.S. & International REITs

User avatar
abuss368
Posts: 12381
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!

Re: Small value underperformance, DFA

Post by abuss368 » Mon Jan 11, 2016 4:10 pm

zotty wrote:
garlandwhizzer wrote: A non-tilted portfolio never performs optimally but it does manage to pick a more comfortable middle ground, with both growth and value, large and small, which presumably offers a smoother ride through market cycles when one segment after another outperforms or underperforms.
i completely agree with sentiment expressed in your post, but i have a problem with this line. it hits a nerve because it's a fairly constant sentiment on this board. the snicker-worthy 3 fund portfolio for the simple man. it bothers me.

The alternative is to be smart and listen to sales people pitch one market beating, risk mitigating, strategy after another: don't give up on the opportunity to beat the market, or to be safer than the market, using security selection. The proof is in the backtest, accurate to 5 significant digits with the caveat that nobody knows the future.

The market is an ever evolving thing and random is streaky. tilting might continue to streak, the streak might be over. To me, that's the heart of gamble.

Maybe this makes me a nihilist, or maybe i just like mid caps.
Are you serious?
John C. Bogle: "You simply do not need to put your money into 8 different mutual funds!" | | Disclosure: Three Fund Portfolio + U.S. & International REITs

wolf359
Posts: 1207
Joined: Sun Mar 15, 2015 8:47 am

Re: Small value underperformance, DFA

Post by wolf359 » Mon Jan 11, 2016 4:27 pm

Dieharder wrote:You either believe in the strategy or not. Looking at past performance, especially over such a short period as 10 years, is not a great idea. Once you decide on a strategy, you should just invest and ignore all such things as past performance. What matters is whether it worked out over your lifetime of investing, typically around 40 or 50 years. Come back in 2046 :twisted:
Don't gamble; take all your savings and buy some good stock and hold it till it goes up, then sell it. If it don't go up, don't buy it.

Will Rogers

larryswedroe
Posts: 15613
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Small value underperformance, DFA

Post by larryswedroe » Mon Jan 11, 2016 4:35 pm

The Tell Tale Chart
There's only one HUGE problem with Bogle's speech, It's not measuring what it should. It doesn't show the performance of growth versus value stocks. What he did was to show the performance of ACTIVE GROWTH funds versus ACTIVE VALUE FUNDS.

Now I don't know why he would choose to do that when all the data was there to show the returns to growth versus value stocks which would have shown an entirely differently conclusion, as anyone who access the Ken French data base can see

Larry

zotty
Posts: 841
Joined: Tue Sep 15, 2009 5:18 pm
Location: DFW, Texas

Re: Small value underperformance, DFA

Post by zotty » Mon Jan 11, 2016 4:42 pm

abuss368 wrote:
zotty wrote:
garlandwhizzer wrote: A non-tilted portfolio never performs optimally but it does manage to pick a more comfortable middle ground, with both growth and value, large and small, which presumably offers a smoother ride through market cycles when one segment after another outperforms or underperforms.
i completely agree with sentiment expressed in your post, but i have a problem with this line. it hits a nerve because it's a fairly constant sentiment on this board. the snicker-worthy 3 fund portfolio for the simple man. it bothers me.

The alternative is to be smart and listen to sales people pitch one market beating, risk mitigating, strategy after another: don't give up on the opportunity to beat the market, or to be safer than the market, using security selection. The proof is in the backtest, accurate to 5 significant digits with the caveat that nobody knows the future.

The market is an ever evolving thing and random is streaky. tilting might continue to streak, the streak might be over. To me, that's the heart of gamble.

Maybe this makes me a nihilist, or maybe i just like mid caps.
Are you serious?
I am now sure this post was totally confusing. Sorry.

I'm a total market adherent. I don't think the past is prologue and i doubt that the risk adjusted outperformance of SCV is anything more than streaky random. It is unlikely to kill a retirement portfolio in the way that other market-beating sales pitches can, but qualitatively, it's not all that different than every other gimmick strategy. people do it because they want to beat the market and think they can beat the market. It doesn't strike me as wise, yet it's the total market adherents who are often treated as the simpletons in the room. it's annoying.
Nadie Sabe Nada

lack_ey
Posts: 6454
Joined: Wed Nov 19, 2014 11:55 pm

Re: Small value underperformance, DFA

Post by lack_ey » Mon Jan 11, 2016 5:13 pm

zotty wrote:I am now sure this post was totally confusing. Sorry.
IMHO if anyone read the other posts or this in context it's not ambiguous, but I guess it's better to be safe than sorry and tag sarcasm explicitly.
zotty wrote:I'm a total market adherent. I don't think the past is prologue and i doubt that the risk adjusted outperformance of SCV is anything more than streaky random. It is unlikely to kill a retirement portfolio in the way that other market-beating sales pitches can, but qualitatively, it's not all that different than every other gimmick strategy. people do it because they want to beat the market and think they can beat the market. It doesn't strike me as wise, yet it's the total market adherents who are often treated as the simpletons in the room. it's annoying.
First of all, risk-adjusted outperformance is harder to prove and much more tenuous than non-risk-adjusted outperformance, and plenty of people question the latter. It is not always clear if people are talking about risk-adjusted outperformance or just outperformance. It might be one thing or the other depending on the context. Personally, I think SCV risk-adjusted outperformance was probably legit in the past, I think it has some definitely non-negligible chance (but less than half I guess) of being real in the future, and I'm fairly certain there will be non-risk-adjusted outperformance in the long run, say the next 50 years.

With respect to asset allocation, risk-adjusted outperformance is definitely not a prerequisite for SCV targeting to be sensible for many. A higher-risk, higher-return asset effectively gives you the potential advantages of leverage—being able to target higher expected return, being able to increase diversification for a given level of expected return (the rationale behind the Larry Portfolio relative to a total market 60/40), among others—without the costs and drawbacks. So aside from the academic argument we're interested in the relative risk-adjusted performance of SCV compared to the market. As long as it's not worse by enough then there's something there that can be used.

One caution is that the expectation and starting point before we examine the data should be for SCV to indeed have worse risk-adjusted performance than the total market. A slice of the market is always going to be less diversified than the total market, so you're giving up something meaningful there.

Bob.Beeman
Posts: 85
Joined: Mon Dec 12, 2011 5:32 pm

Re: Small value underperformance, DFA

Post by Bob.Beeman » Mon Jan 11, 2016 5:33 pm

Test.

Code: Select all

Fund Name                               1-Year   3-Year   5-Year  10-Year  15-Year
-----------------------------------    -------  -------   ------  -------  -------
DFSVX (DFA small Value)                  -7.22     8.71     8.08     5.80    10.16
VISGX (Vanguard small growth)            -2.90     9.89     9.28     7.72     8.73
VSCIX (Vangaurd small blend)             -3.36    10.72     9.78     7.53     8.61
VTSMX (Vangaurd total stock market)       0.75    12.93    11.40     7.02     5.63
S&P 500                                   1.93    13.56    11.93     6.95     5.03

User avatar
ogd
Posts: 4854
Joined: Thu Jun 14, 2012 11:43 pm

Re: Small value underperformance, DFA

Post by ogd » Mon Jan 11, 2016 5:35 pm

lack_ey wrote:So what is your non-belief in value, if we want to call it, based on? If it's just the story of "oh, everyone knows about it and too many people are chasing it now" then there are lots of stories for all kinds of things in investing. But where's the data? What is the thought process and line of reasoning?

...
3. Companies with higher value scores (higher BE/ME by Fama-French) get bid up compared to growth companies, equalizing out future returns
Yes, and yes. Let me clarify on #3 that I think that the risk story will in the long term persist, like all the risk stories are entitled to. Just not the behavioral story, which is the more valuable one.

In my worldview, I actually don't think I need data for this, although I could cite the failure of value to deliver anything other than avoidance of the tech bubble[*] in the brave new factor world. It is persistent market anomalies that need ongoing support from data and it's always a surprise when the anomaly persists after being so widely known. Clearly, entire tomes have been written on this so I'm not the only one wondering how it can persist.

My expectation is that value will simply fade away and it will not deliver the returns it was supposed to based on history, with the crucial difference that makes history not applicable being the widely accepted knowledge. I might be surprised, of course.

[*] One could always say that this was by design rather than accidental. Maybe; it's not the only strategy that would have avoided the crash and certainly hadn't been sold as such. But the next bubble might be in value (2008 was very slightly so), whereby companies that are on the brink are held up by the belief in value.

larryswedroe
Posts: 15613
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Small value underperformance, DFA

Post by larryswedroe » Mon Jan 11, 2016 7:55 pm

ogd
You might certainly be correct, but to be correct you have to believe that investors will change their behavior, because the limits to arbitrage aren't going away, so that means stocks can be overpriced persistently, leading to the behavioral part of the value premium. What the academics have found in the research is that the anomalies persist in the most illiquid stocks, the ones most expensive to trade, like small growth stocks, and penny stocks, and stocks in bankruptcy. So fwiw IMO those are likely to persist.
And then of course you do have the risk story

Best wishes
Larry

User avatar
bertilak
Posts: 5883
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Small value underperformance, DFA

Post by bertilak » Mon Jan 11, 2016 8:10 pm

larryswedroe wrote:ogd
You might certainly be correct, but to be correct you have to believe that investors will change their behavior, because the limits to arbitrage aren't going away, so that means stocks can be overpriced persistently, leading to the behavioral part of the value premium. What the academics have found in the research is that the anomalies persist in the most illiquid stocks, the ones most expensive to trade, like small growth stocks, and penny stocks, and stocks in bankruptcy. So fwiw IMO those are likely to persist.
And then of course you do have the risk story

Best wishes
Larry
Is this where someone should jump in and say "The market can remain irrational longer than you can remain solvent."

OR is this where the SCV investor waits for his SCV stocks to grow into LCG stocks?
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker, the Cowboy Poet

larryswedroe
Posts: 15613
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Small value underperformance, DFA

Post by larryswedroe » Tue Jan 12, 2016 8:46 am

Bertilak
That quotation ONLY applies if you are SHORT stocks, not long. And that is why, or one of the reasons, why limits to arbitrage allows stocks to be persistently overvalued and we get these anomalies that cannot be explained by rational behavior. It's much easier to correct an underpricing, just go long. But correcting an overvaluation is not only more costly, but far more risky
Best wishes
Larry

User avatar
William Million
Posts: 544
Joined: Wed May 05, 2010 4:41 am
Location: A Deep Mountain

Re: Small value underperformance, DFA

Post by William Million » Tue Jan 12, 2016 10:12 am

Larry-

I understand you believe adding a small value tilt enables an investor to lower stock allocation. However, why is it desirable to do so? Why is it better to be 50/50 with a s/v tilt than 60/40 without the tilt?

User avatar
Slick8503
Posts: 195
Joined: Tue Nov 16, 2010 10:25 am

Re: Small value underperformance, DFA

Post by Slick8503 » Tue Jan 12, 2016 10:59 am

I believe Larry will say that it has proven to provide better risk adjusted returns. It allows one to protect left tail risk better because of the ability to hold more high quality bonds while not sacrificing return. In other words, it has been the more efficient portfolio.

larryswedroe
Posts: 15613
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Small value underperformance, DFA

Post by larryswedroe » Tue Jan 12, 2016 11:42 am

William, the long answer is to read the book Reducing the Risks of Black Swans

The short answer is this, using rough numbers to make math easy. S&P 500 (or TSM) TSM about 10%, and SV about 14% and bonds about 6. So you could have gotten 10% by being 100% S&P 500 or 50/50 with SV and bonds. Now think about what happens in bad years, like 2008. SV likely underperforms but not by enough to offset that 50% lower allocation. Say S&P went down 37% and SV went down 45%, but the safe bonds went up say 10% and you only had 50% in SV so your portfolio was down only about say 18%. Now take reverse like in 2013 when S&P went up say 32% but SV went up 45%, and bonds say flat. Well SV outperformed but not by enough to offset the lower equity allocation. Bottom line is the strategy cuts both tails. The news is even better though because of how correlations work. In good years bonds have about 0 correlation with stocks, but in really bad years SAFE bonds have correlation that tends to turn sharply negative. So the left tail tends to get cut by more than the right tail, again if use safe bonds

If you look at the historical evidence the efficiency of the Larry Portfolio is much higher in long term.
It's also similar to the concept of risk parity strategies which are proposed for the same reasons, with similar amounts of risks in all the factors while TSM has only one factor exposure, market beta, and a 60/40 portfolio has way more than 60% of its RISK in stocks, way more, something most investors fail to understand. And finally it's also similar to Taleb's strategy of "barbelling", with the risks you take own the riskiest assets and with the rest own only the safest.

Clearly not for everyone as has massive tracking error regret mix. But those who have adopted it have benefited and certainly were sleeping lot better in 2008 and all the intermediate periods of crises.

Hope that helps

Larry

zotty
Posts: 841
Joined: Tue Sep 15, 2009 5:18 pm
Location: DFW, Texas

Re: Small value underperformance, DFA

Post by zotty » Tue Jan 12, 2016 11:47 am

Slick8503 wrote:proven
nothing has been proven. It's all theory that may falsify in the future or may hold true.
Nadie Sabe Nada

whadyaknow
Posts: 122
Joined: Sun Mar 10, 2013 11:51 am

Re: Small value underperformance, DFA

Post by whadyaknow » Tue Jan 12, 2016 11:47 am

larryswedroe wrote:William, the long answer is to read the book Reducing the Risks of Black Swans

The short answer is this, using rough numbers to make math easy. S&P 500 (or TSM) TSM about 10%, and SV about 14% and bonds about 6. So you could have gotten 10% by being 100% S&P 500 or 50/50 with SV and bonds. Now think about what happens in bad years, like 2008. SV likely underperforms but not by enough to offset that 50% lower allocation. Say S&P went down 37% and SV went down 45%, but the safe bonds went up say 10% and you only had 50% in SV so your portfolio was down only about say 18%. Now take reverse like in 2013 when S&P went up say 32% but SV went up 45%, and bonds say flat. Well SV outperformed but not by enough to offset the lower equity allocation. Bottom line is the strategy cuts both tails. The news is even better though because of how correlations work. In good years bonds have about 0 correlation with stocks, but in really bad years SAFE bonds have correlation that tends to turn sharply negative. So the left tail tends to get cut by more than the right tail, again if use safe bonds

If you look at the historical evidence the efficiency of the Larry Portfolio is much higher in long term.
It's also similar to the concept of risk parity strategies which are proposed for the same reasons, with similar amounts of risks in all the factors while TSM has only one factor exposure, market beta, and a 60/40 portfolio has way more than 60% of its RISK in stocks, way more, something most investors fail to understand. And finally it's also similar to Taleb's strategy of "barbelling", with the risks you take own the riskiest assets and with the rest own only the safest.

Clearly not for everyone as has massive tracking error regret mix. But those who have adopted it have benefited and certainly were sleeping lot better in 2008 and all the intermediate periods of crises.

Hope that helps

Larry
Thanks for the simple explanation Larry. I'm a Total Market Investor myself and don't plan to change my strategy. However, I understand the theoretical benefits of SV investing now.
80/20 Stock/Bond

User avatar
Slick8503
Posts: 195
Joined: Tue Nov 16, 2010 10:25 am

Re: Small value underperformance, DFA

Post by Slick8503 » Tue Jan 12, 2016 12:03 pm

zotty wrote:
Slick8503 wrote:proven
nothing has been proven. It's all theory that may falsify in the future or may hold true.
I should say "has been proven in the past..."

Post Reply