Why not go all small cap value for equities?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
stan1
Posts: 8528
Joined: Mon Oct 08, 2007 4:35 pm

Re: Why not go all small cap value for equities?

Post by stan1 » Fri Dec 28, 2018 10:34 am

I view the all small cap value discussion as mostly a theoretical one. Very few people actually would choose this due to tracking error and the potential that you might underperform the market for the wrong decades out of a 50-70 investing horizon most of us have. Also at the end of the day I find it hard to rationalize not owning stock in the most successful companies in America. Yes there are a few Enrons and GEs but that's why you buy the S&P 500 not the Dow with just 30 companies. You have to have a lot of confidence in the theory to eschew investing in Apple, Exxon, Chase, Walmart, and most of the other products you see at home and work.

The Top 10 Holdings in VIOV (S&P 600 Value) accounting for 7.8% right now are:
Spire Inc.
Community Bank System Inc.
Wolverine World Wide Inc.
American Equity Investment Life Holding Co.
SkyWest Inc.
Old National Bancorp
Finisar Corp.
Simmons First National Corp.
FTI Consulting Inc.
Monro Inc.

I'm willing to overweight a large group of small value stocks but not willing to invest just in these companies.

User avatar
vineviz
Posts: 6723
Joined: Tue May 15, 2018 1:55 pm

Re: Why not go all small cap value for equities?

Post by vineviz » Fri Dec 28, 2018 10:36 am

Random Walker wrote:
Fri Dec 28, 2018 10:07 am
Privatefarmer,
You know I’m going to disagree with you on DFA being active versus passive. I view DFA as passive. It is formulaic, and has no individual stock picking or market timing. It simply provides passive access to the factors that have been shown to explain equity returns. It tries to add value by systematically screening out some stocks that have historically performed poorly like IPOs and small growth, and by performing securities lending, optimizing trades with patient buying and selling, and using buy/hold ranges to capitalize a bit on momentum. But all that is formulaic and agnostic to individual stocks or market timing.

Dave
I think anyone familiar with how active funds are managed would agree with you.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

Random Walker
Posts: 4522
Joined: Fri Feb 23, 2007 8:21 pm

Re: Why not go all small cap value for equities?

Post by Random Walker » Fri Dec 28, 2018 10:40 am

I realize that can’t deny what’s in writing, but the organization’s culture also needs to get some credit. A lot of what’s in the prospectus is simply legaleze. Although the prospectus provides a lot of wiggle room (as I think all do!), the company as far as I know adheres diligently to its rules based strategies.

I can see where it’s necessary for a prospectus to provide leeway. Let’s say DFA decided to change from using B/M alone as it’s value metric to several different value metrics in combination. Is that really being active, or just improving the passive process based on accumulated learning? Different people can view it differently I suppose.

Dave

stan1
Posts: 8528
Joined: Mon Oct 08, 2007 4:35 pm

Re: Why not go all small cap value for equities?

Post by stan1 » Fri Dec 28, 2018 10:45 am

vineviz wrote:
Fri Dec 28, 2018 10:36 am
Random Walker wrote:
Fri Dec 28, 2018 10:07 am
Privatefarmer,
You know I’m going to disagree with you on DFA being active versus passive. I view DFA as passive. It is formulaic, and has no individual stock picking or market timing. It simply provides passive access to the factors that have been shown to explain equity returns. It tries to add value by systematically screening out some stocks that have historically performed poorly like IPOs and small growth, and by performing securities lending, optimizing trades with patient buying and selling, and using buy/hold ranges to capitalize a bit on momentum. But all that is formulaic and agnostic to individual stocks or market timing.

Dave
I think anyone familiar with how active funds are managed would agree with you.
So what's the boundary between a factor based algorithm (passive) and a quantitative algorithm (active)?

Random Walker
Posts: 4522
Joined: Fri Feb 23, 2007 8:21 pm

Re: Why not go all small cap value for equities?

Post by Random Walker » Fri Dec 28, 2018 11:16 am

stan1 wrote:
Fri Dec 28, 2018 10:45 am
vineviz wrote:
Fri Dec 28, 2018 10:36 am
Random Walker wrote:
Fri Dec 28, 2018 10:07 am
Privatefarmer,
You know I’m going to disagree with you on DFA being active versus passive. I view DFA as passive. It is formulaic, and has no individual stock picking or market timing. It simply provides passive access to the factors that have been shown to explain equity returns. It tries to add value by systematically screening out some stocks that have historically performed poorly like IPOs and small growth, and by performing securities lending, optimizing trades with patient buying and selling, and using buy/hold ranges to capitalize a bit on momentum. But all that is formulaic and agnostic to individual stocks or market timing.

Dave
I think anyone familiar with how active funds are managed would agree with you.
So what's the boundary between a factor based algorithm (passive) and a quantitative algorithm (active)?
I’m likely not qualified to answer that, but I’ll give it a shot anyways. I think there are 5 or 6 factors that have been shown by academics to explain about 95% of an equity portfolio’s returns. If some computers formulaicly access just those factors, then I’d consider that passive factor based investing. Anything fancier, more complicated, accessing other metrics, timing signals, might be considered over the line into another realm.

Dave

User avatar
vineviz
Posts: 6723
Joined: Tue May 15, 2018 1:55 pm

Re: Why not go all small cap value for equities?

Post by vineviz » Fri Dec 28, 2018 11:53 am

stan1 wrote:
Fri Dec 28, 2018 10:45 am
vineviz wrote:
Fri Dec 28, 2018 10:36 am
Random Walker wrote:
Fri Dec 28, 2018 10:07 am
Privatefarmer,
You know I’m going to disagree with you on DFA being active versus passive. I view DFA as passive. It is formulaic, and has no individual stock picking or market timing. It simply provides passive access to the factors that have been shown to explain equity returns. It tries to add value by systematically screening out some stocks that have historically performed poorly like IPOs and small growth, and by performing securities lending, optimizing trades with patient buying and selling, and using buy/hold ranges to capitalize a bit on momentum. But all that is formulaic and agnostic to individual stocks or market timing.

Dave
I think anyone familiar with how active funds are managed would agree with you.
So what's the boundary between a factor based algorithm (passive) and a quantitative algorithm (active)?
Who says there is a boundary between those particular descriptions? The active/passive distinction is primarily one of security—by-security manager discretion, not the use of factors and/or quantitative techniques.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

Valuethinker
Posts: 40259
Joined: Fri May 11, 2007 11:07 am

Re: Why not go all small cap value for equities?

Post by Valuethinker » Fri Dec 28, 2018 12:02 pm

ChinchillaWhiplash wrote:
Thu Dec 27, 2018 6:27 pm
Sorry if this has been covered, but I did not find it in a search. Judging by the available data going back to 1972 for asset classes on Portforlio Visualizer, sure seems like it is worth heavily or completely tilting towards US small cap value. If equities are supposed to be your risk, is it not worth having a bunch of your portfolio in this asset class? Maybe even replacing total US market? I know the future is unknown, but if past is any indicator :twisted: If you are in it for the long haul, sure makes sense to hold SCV to me.
Now there was a poster SmallHi? Some variation on that.

Haven't heard from him in years.

You are taking a very big bet on what, 4% of US market?

Worth investigating the DFA funds, as they have refined this strategy down to a fine art. The trick is things like dealing costs - actually getting those down. Dealing and market impact costs are what will kill you in implementing this strategy.

Valuethinker
Posts: 40259
Joined: Fri May 11, 2007 11:07 am

Re: Why not go all small cap value for equities?

Post by Valuethinker » Fri Dec 28, 2018 12:07 pm

ChinchillaWhiplash wrote:
Thu Dec 27, 2018 6:27 pm
Sorry if this has been covered, but I did not find it in a search. Judging by the available data going back to 1972 for asset classes on Portforlio Visualizer, sure seems like it is worth heavily or completely tilting towards US small cap value. If equities are supposed to be your risk, is it not worth having a bunch of your portfolio in this asset class? Maybe even replacing total US market? I know the future is unknown, but if past is any indicator :twisted: If you are in it for the long haul, sure makes sense to hold SCV to me.
Note the research suggests you really only get the SV effect if you long the bottom decile of price to book (top decile of book to market) and short the top decile of price to book.

You have to be able to go long and short to really capitalize on the effect.

I think, due to momentum effects (and dealing costs), you are actually more successful if you long the bottom decile and *hold* until they converge onto the median (recover) and ditto the other way on the shorts (collapse).

It's a strategy that will hurt you most of the time, and then you will get 2000-03 and you will make a generation's worth of excess return in about 36 months.

Conversely remember "Bonfire of the Quants" the implosion of this strategy in early August 2007. Worth reading Scott Patterson ("The Quants") about those days, if nowhere else.

Valuethinker
Posts: 40259
Joined: Fri May 11, 2007 11:07 am

Re: Why not go all small cap value for equities?

Post by Valuethinker » Fri Dec 28, 2018 12:10 pm

privatefarmer wrote:
Fri Dec 28, 2018 10:25 am
Random Walker wrote:
Fri Dec 28, 2018 10:07 am
Privatefarmer,
You know I’m going to disagree with you on DFA being active versus passive. I view DFA as passive. It is formulaic, and has no individual stock picking or market timing. It simply provides passive access to the factors that have been shown to explain equity returns. It tries to add value by systematically screening out some stocks that have historically performed poorly like IPOs and small growth, and by performing securities lending, optimizing trades with patient buying and selling, and using buy/hold ranges to capitalize a bit on momentum. But all that is formulaic and agnostic to individual stocks or market timing.

Dave
The problem though is what they state in their own prospectus. From DFSVX :

“Advisor may adjust the representation in the U.S. Small Cap Value Portfolio of an eligible company, or exclude a company, after considering such factors as free float, momentum, trading strategies, liquidity, value, profitability, and other factors that the Advisor determines to be appropriate, given market conditions... The criteria the Advisor uses for assessing value or profitability are subject to change from time to time.”

To me that sounds a LOT like active management versus passively following an index. Their turnover is not the highest I’ve seen and their costs are not as high as most active managed funds, and to some extent ANY index is at least in part activeley managed/created. However I would say DFA is far closer to actively managed that purely passive.
Yes DFA is not a "passive" manager.

They'd be classified as "Active Quant" in the jargon of institutional investing. Often criticized as a group for their "black box" nature - Swensen says he doesn't back many of those types of funds at Yale Endowment, because you can't see inside the box - proprietary algorithms. By the time you know if the strategy really works, they probably have too much money under management to implement effectively.

Market impact costs are the key - the bottom decile of Price to Book & Size are very difficult to deal in.

Also see my post above re Bonfire of the Quants. Those were SCV strategies that imploded when a major hedge fund had to liquidate its portfolio.

MikeMak27
Posts: 129
Joined: Fri Jul 03, 2015 4:37 am

Re: Why not go all small cap value for equities?

Post by MikeMak27 » Fri Dec 28, 2018 12:19 pm

My equity allocation is 55% small cap value / small cap (IJR, IJS, FSEMX, VBR (best option used for each account), and 45% Emerging market IEMG, VWO, FPADX. They seem to balance each other with regards to dollar strength. It has been frustrating in 2018 with the Nasdaq QQQ outperforming, but once you pick a strategy you have to stick with it. I’m confident over the long haul I’ll be rewarded.

So yes you can do it, but you must be patient.
Mak 3 fund portfolio: 50% US small cap value & US Small cap (IJS, IJR), 40% Emerging Markets (IEMG, VWO, FPADX), 10% US REIT (VNQ)

User avatar
Earl Lemongrab
Posts: 7270
Joined: Tue Jun 10, 2014 1:14 am

Re: Why not go all small cap value for equities?

Post by Earl Lemongrab » Fri Dec 28, 2018 4:50 pm

MotoTrojan wrote:
Thu Dec 27, 2018 6:31 pm
If it were this easy/full-proof then everyone would do it and prices would go up until it was no longer a premium. I tilt 36% of US equity to SCV fwiw.
Only if the SCV premium is largely or exclusively a behavioral one would that work. But it is or is theorized to be a risk premium. Those don't get arbitraged away. Otherwise, the stock premium would evaporate and we should all be in bonds.

User avatar
Earl Lemongrab
Posts: 7270
Joined: Tue Jun 10, 2014 1:14 am

Re: Why not go all small cap value for equities?

Post by Earl Lemongrab » Fri Dec 28, 2018 4:53 pm

AHTFY wrote:
Thu Dec 27, 2018 7:40 pm
Very simple answer: Because you'll be increasing your risk by a greater degree than your potential return.

Note: The increase in risk comes from two areas:
1) You are less diversified both in terms of owning fewer stocks and different types of stocks.
2) Small caps are [supposedly] more risky than large or mid cap stocks.
I think it's a big stretch to say that there is any significant risk due to lack of diversification. While not as diversified as large segments of the market, they aren't that low in diversification. Where did you get the numbers that show that they have a lower risk-adjusted return than other stocks? If that were true then you could AVOID SCV and have a similar return with lower risk.

KJVanguard
Posts: 222
Joined: Fri Oct 05, 2018 3:39 am

Re: Why not go all small cap value for equities?

Post by KJVanguard » Fri Dec 28, 2018 5:17 pm

If you want to go all SCV for your US equity allocation I wouldn't not criticize such a decision.

I am pretty much forced to stay in TSM that I have held for 24 years due to massive unrealized gains at this point. Today any US stock money would go towards SCV for me. And I do own a decent amount of SCV as it is.

garlandwhizzer
Posts: 2863
Joined: Fri Aug 06, 2010 3:42 pm

Re: Why not go all small cap value for equities?

Post by garlandwhizzer » Fri Dec 28, 2018 5:50 pm

If you unequivocally accept factor research as being a consistent and reliable guide not only to the market's past but to its future, and you have a very long investing time horizon (at least 2 decades), and you are confident that you can tolerate SCV underperformance which may go on for 15 years or more without selling it, and you are risk/volatility tolerant with equity exposure (SCV expected to be more volatile than beta), then it seems reasonable for you to choose 100% SCV for equity exposure. If you have strong belief that all these conditions are met, 100% SCV is for you. If not, anywhere from market weight SCV on up seems a better fit for your market belief system.

Garland Whizzer

fennewaldaj
Posts: 950
Joined: Sun Oct 22, 2017 11:30 pm

Re: Why not go all small cap value for equities?

Post by fennewaldaj » Fri Dec 28, 2018 10:26 pm

vineviz wrote:
Fri Dec 28, 2018 11:53 am
stan1 wrote:
Fri Dec 28, 2018 10:45 am
vineviz wrote:
Fri Dec 28, 2018 10:36 am
Random Walker wrote:
Fri Dec 28, 2018 10:07 am
Privatefarmer,
You know I’m going to disagree with you on DFA being active versus passive. I view DFA as passive. It is formulaic, and has no individual stock picking or market timing. It simply provides passive access to the factors that have been shown to explain equity returns. It tries to add value by systematically screening out some stocks that have historically performed poorly like IPOs and small growth, and by performing securities lending, optimizing trades with patient buying and selling, and using buy/hold ranges to capitalize a bit on momentum. But all that is formulaic and agnostic to individual stocks or market timing.

Dave
I think anyone familiar with how active funds are managed would agree with you.
So what's the boundary between a factor based algorithm (passive) and a quantitative algorithm (active)?
Who says there is a boundary between those particular descriptions? The active/passive distinction is primarily one of security—by-security manager discretion, not the use of factors and/or quantitative techniques.
It seems like there is a continuum of how active strategies are. Something like this listed from most passive to least passive cap weighting, DFA style/fundamental indexes, low turnover high position count quant, high turnover quant, fundamental stock selection

User avatar
privatefarmer
Posts: 695
Joined: Mon Sep 08, 2014 2:45 pm

Re: Why not go all small cap value for equities?

Post by privatefarmer » Sat Dec 29, 2018 12:28 am

one other thought I have is that when you compare vanguards small cap fund vs TSM since inception (~1989 I believe), they're essentially tied. Now some will argue that that is because vanguard is not as "small" as funds like DFA, they are more of a mid-cap/small-cap blend. That's a fine argument however if that is the case then that means that there supposedly is zero premium until you get to the "really small" companies? well if that's true then I don't know how one could expect to extract any premium as now there is lots of $$$ chasing those tiny stocks (which equate to <3% of the total market)...

Either you would see some sort of premium by just excluding the mega-cap stocks (which you haven't over the last ~30 years) or you would only see a premium if investing in the smallest of the small companies (which means it would quickly be sucked up and basically impossible to extract, especially after trading costs).

rkhusky
Posts: 9496
Joined: Thu Aug 18, 2011 8:09 pm

Re: Why not go all small cap value for equities?

Post by rkhusky » Sat Dec 29, 2018 12:59 am

Random Walker wrote:
Fri Dec 28, 2018 11:16 am
It is formulaic, and has no individual stock picking or market timing.
The quote from the DFA prospectus said that they adjust their methods based on market conditions. That's market timing. A method can be formulaic, but still active, if you are changing your formula every month. And who can say whether DFA or any other smart beta provider isn't?
Random Walker wrote:
Fri Dec 28, 2018 11:16 am
I think there are 5 or 6 factors that have been shown by academics to explain about 95% of an equity portfolio’s returns.
Describing a portfolio's returns with factors doesn't tell you anything about how the portfolio will perform in the future. You need to be able to predict how the factors will perform in the future to do that. Invoking factor formulism doesn't mean that one has a better chance of out-performing the market.

fennewaldaj
Posts: 950
Joined: Sun Oct 22, 2017 11:30 pm

Re: Why not go all small cap value for equities?

Post by fennewaldaj » Sat Dec 29, 2018 1:49 am

privatefarmer wrote:
Sat Dec 29, 2018 12:28 am
one other thought I have is that when you compare vanguards small cap fund vs TSM since inception (~1989 I believe), they're essentially tied.
Well the inception was 1999 but the small value fund has won by a lot 9.48% CAGR for small value vs 6.56% for TSM. Granted this was a very good starting point for small value

https://www.portfoliovisualizer.com/bac ... ion2_2=100

MotoTrojan
Posts: 9952
Joined: Wed Feb 01, 2017 8:39 pm

Re: Why not go all small cap value for equities?

Post by MotoTrojan » Sat Dec 29, 2018 7:53 am

Earl Lemongrab wrote:
Fri Dec 28, 2018 4:50 pm
MotoTrojan wrote:
Thu Dec 27, 2018 6:31 pm
If it were this easy/full-proof then everyone would do it and prices would go up until it was no longer a premium. I tilt 36% of US equity to SCV fwiw.
Only if the SCV premium is largely or exclusively a behavioral one would that work. But it is or is theorized to be a risk premium. Those don't get arbitraged away. Otherwise, the stock premium would evaporate and we should all be in bonds.
Fair. Probably should replace my “easy” with “risk-free”.

Random Walker
Posts: 4522
Joined: Fri Feb 23, 2007 8:21 pm

Re: Why not go all small cap value for equities?

Post by Random Walker » Sat Dec 29, 2018 10:35 am

rkhusky wrote:
Sat Dec 29, 2018 12:59 am
Describing a portfolio's returns with factors doesn't tell you anything about how the portfolio will perform in the future. You need to be able to predict how the factors will perform in the future to do that. Invoking factor formulism doesn't mean that one has a better chance of out-performing the market.
That is absolutely true! But at least the investor knows the risks he is taking and the explanatory factors he is exposed to. No one can predict what the factors, including market beta, will do in the future. It’s not about out performing the market. It’s about building an efficient portfolio and passively accessing factors known to overwhelmingly explain the behavior of assets and portfolios. There is a difference between an efficient market and an efficient portfolio. The whole market, represented by TSM is highly efficient. But a portfolio comprised of TSM exclusively is highly likely something that can be improved upon.

Dave

User avatar
whodidntante
Posts: 8334
Joined: Thu Jan 21, 2016 11:11 pm
Location: outside the echo chamber

Re: Why not go all small cap value for equities?

Post by whodidntante » Sat Dec 29, 2018 10:57 am

rkhusky wrote:
Sat Dec 29, 2018 12:59 am

Describing a portfolio's returns with factors doesn't tell you anything about how the portfolio will perform in the future. You need to be able to predict how the factors will perform in the future to do that.
Yep, true, hat's off. That's also true of the market factor. But stocks beating bonds seems like a solid bet to me, even though it's far from certain.

rkhusky
Posts: 9496
Joined: Thu Aug 18, 2011 8:09 pm

Re: Why not go all small cap value for equities?

Post by rkhusky » Sat Dec 29, 2018 10:59 am

Random Walker wrote:
Sat Dec 29, 2018 10:35 am
But at least the investor knows the risks he is taking and the explanatory factors he is exposed to.
This is the benefit of factor analysis. An investor can examine a fund and see whether its past performance is due to expertise (or lack thereof) of the fund manager, or whether it's because the fund is invested in particular classes of stocks. The investor can then determine if there is a less expensive way to access those particular classes of stocks.
Random Walker wrote:
Sat Dec 29, 2018 10:35 am
It’s about building an efficient portfolio and passively accessing factors known to overwhelmingly explain the behavior of assets and portfolios.
Factor analysis does not tell you whether you will have an efficient portfolio or not, where efficiency is measured in return vs risk.
Random Walker wrote:
Sat Dec 29, 2018 10:35 am
But a portfolio comprised of TSM exclusively is highly likely something that can be improved upon.
Factor analysis does not tell you this. I have seen nothing that would give you the numerical likelihood of outperforming the market, in terms of risk-adjusted return.

User avatar
Earl Lemongrab
Posts: 7270
Joined: Tue Jun 10, 2014 1:14 am

Re: Why not go all small cap value for equities?

Post by Earl Lemongrab » Sat Dec 29, 2018 12:14 pm

whodidntante wrote:
Sat Dec 29, 2018 10:57 am
rkhusky wrote:
Sat Dec 29, 2018 12:59 am

Describing a portfolio's returns with factors doesn't tell you anything about how the portfolio will perform in the future. You need to be able to predict how the factors will perform in the future to do that.
Yep, true, hat's off. That's also true of the market factor. But stocks beating bonds seems like a solid bet to me, even though it's far from certain.
Yeah, part of risk is the risk that it will underperform some other investment. Maybe for an extended period.

Post Reply