Which is more diversified? TSM or TSM+value?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
acegolfer
Posts: 1092
Joined: Tue Aug 25, 2009 9:40 am

Which is more diversified? TSM or TSM+value?

Post by acegolfer » Sun Aug 19, 2018 9:27 am

The following is my main question on value investing. (I'm not against value investing. Just want to know what your objective is.)

I invest in value (or any other factor such as size, momentum, profitability, investment) because

1) to achieve a higher expected return + lower stdev than TSM (which is true in multi-factor efficient frontier http://faculty.chicagobooth.edu/john.co ... 3q99_4.pdf)

2) or something else?

AlohaJoe
Posts: 3821
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Which is more diversified? TSM or TSM+value?

Post by AlohaJoe » Sun Aug 19, 2018 9:33 am

What's your definition of "diversified"? What metric does it have? What does that metric say about your two options?

(Why is "most diversified" an investing goal for you?)

User avatar
vineviz
Posts: 2004
Joined: Tue May 15, 2018 1:55 pm

Re: Which is more diversified? TSM or TSM+value?

Post by vineviz » Sun Aug 19, 2018 10:12 am

acegolfer wrote:
Sun Aug 19, 2018 9:27 am
The following is my main question on value investing. (I'm not against value investing. Just want to know what your objective is.)

I invest in value (or any other factor such as size, momentum, profitability, investment) because

1) to achieve a higher expected return + lower stdev than TSM (which is true in multi-factor efficient frontier http://faculty.chicagobooth.edu/john.co ... 3q99_4.pdf)

2) or something else?
You need to be careful in your question to match your definitions with the capital asset pricing model you are using.

In the original single factor CAPM, the only "risk" in the theory is volatility. Starting from a market portfolio, it is theoretically possible to either increase expected returns or reduce expected volatility by changing the weights of the assets, but it isn't possible to do both simultaneously. Even under such a model, though, the "most diversified" portfolio isn't necessarily the one with the lowest volatility.

In a multifactor world, volatility is no longer the only risk. You could theoretically increase returns AND lower volatility, but only at the expense of increasing some OTHER risk.

Diversification, in the modern understanding which embraces multiple kinds of risk, is the process of spreading the total risk of the portfolio more evenly across multiple independent sources. In that sense, TSM+value would be more diversified than just TSM even if TSM+value had higher volatility than just TSM.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
whodidntante
Posts: 4153
Joined: Thu Jan 21, 2016 11:11 pm

Re: Which is more diversified? TSM or TSM+value?

Post by whodidntante » Sun Aug 19, 2018 10:22 am

The market beta premium is the main source of risk and return for Boglehead portfolios. Bonds, as some would say, are for safety. So the idea is to make your money via the market beta premium and squirrel some away into bonds as your human capital decreases so you just don't care about potential future losses in the stock market. However, the market beta premium does not explain all stock market returns, which is where the other factors identified in academia come in.

By diversifying across factors you can achieve another form of diversification, that is, factor diversification. This creates a portfolio that is more diversified, and will perform differently than a portfolio that can only capture market beta.

From Swedroe's book, "Your Complete Guide to Factor-Based Investing"
Image

You can also chose a less diversified approach, e.g. heavily overweighting value stocks, which may outperform a cap weighted portfolio over your lifetime. Then you are exposed to a different source of risk and return than market beta alone. It's a choice you can make, but be aware that you might not have the payday when you need it. I would still argue that this portfolio is more diversified, because it captures the value factor and the market beta factor.

acegolfer
Posts: 1092
Joined: Tue Aug 25, 2009 9:40 am

Re: Which is more diversified? TSM or TSM+value?

Post by acegolfer » Sun Aug 19, 2018 10:59 am

vineviz wrote:
Sun Aug 19, 2018 10:12 am
In a multifactor world, volatility is no longer the only risk. You could theoretically increase returns AND lower volatility, but only at the expense of increasing some OTHER risk.

Diversification, in the modern understanding which embraces multiple kinds of risk, is the process of spreading the total risk of the portfolio more evenly across multiple independent sources. In that sense, TSM+value would be more diversified than just TSM even if TSM+value had higher volatility than just TSM.
To anyone who invest in value, can you define what the "OTHER" risk is?

User avatar
whodidntante
Posts: 4153
Joined: Thu Jan 21, 2016 11:11 pm

Re: Which is more diversified? TSM or TSM+value?

Post by whodidntante » Sun Aug 19, 2018 11:01 am

acegolfer wrote:
Sun Aug 19, 2018 10:59 am
vineviz wrote:
Sun Aug 19, 2018 10:12 am
In a multifactor world, volatility is no longer the only risk. You could theoretically increase returns AND lower volatility, but only at the expense of increasing some OTHER risk.

Diversification, in the modern understanding which embraces multiple kinds of risk, is the process of spreading the total risk of the portfolio more evenly across multiple independent sources. In that sense, TSM+value would be more diversified than just TSM even if TSM+value had higher volatility than just TSM.
To anyone who invest in value, can you define what the "OTHER" risk is?
I just did that, one post up.

acegolfer
Posts: 1092
Joined: Tue Aug 25, 2009 9:40 am

Re: Which is more diversified? TSM or TSM+value?

Post by acegolfer » Sun Aug 19, 2018 11:02 am

AlohaJoe wrote:
Sun Aug 19, 2018 9:33 am
(Why is "most diversified" an investing goal for you?)
If you are asking me, diversification is not a goal for me. It's a mean to achieve an objective. I started this thread to understand what "your" investing objective is.

acegolfer
Posts: 1092
Joined: Tue Aug 25, 2009 9:40 am

Re: Which is more diversified? TSM or TSM+value?

Post by acegolfer » Sun Aug 19, 2018 11:04 am

whodidntante wrote:
Sun Aug 19, 2018 11:01 am

I just did that, one post up.
I'm sorry. I can't find your definition of "OTHER" risk. Can you quote it? TY.

User avatar
bertilak
Posts: 6120
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Which is more diversified? TSM or TSM+value?

Post by bertilak » Sun Aug 19, 2018 6:46 pm

acegolfer wrote:
Sun Aug 19, 2018 10:59 am
vineviz wrote:
Sun Aug 19, 2018 10:12 am
In a multifactor world, volatility is no longer the only risk. You could theoretically increase returns AND lower volatility, but only at the expense of increasing some OTHER risk.

Diversification, in the modern understanding which embraces multiple kinds of risk, is the process of spreading the total risk of the portfolio more evenly across multiple independent sources. In that sense, TSM+value would be more diversified than just TSM even if TSM+value had higher volatility than just TSM.
To anyone who invest in value, can you define what the "OTHER" risk is?
Value IS the other risk. Another currently-favored risk is size. I am not a value investor but I think it boils down to this:
  • By tilting to various factors you are actually ADDING risk to your portfolio and decreasing diversification in the original beta-volatility sense.
  • The theory is (I think) that some risks (e.g. value and size) have a better (lower) risk/return ratio. Therefore, adding some of these favored risks gives you better risk-adjusted return. This allows you to perhaps dial back on the overall risk while maintaining the original return potential.
The fact that, above, I emphasize think (in the sense of unsure) is why I am not a factor slice-n-dicer.
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker, the Cowboy Poet

acegolfer
Posts: 1092
Joined: Tue Aug 25, 2009 9:40 am

Re: Which is more diversified? TSM or TSM+value?

Post by acegolfer » Sun Aug 19, 2018 7:04 pm

bertilak wrote:
Sun Aug 19, 2018 6:46 pm
Value IS the other risk. Another currently-favored risk is size. I am not a value investor but I think it boils down to this:
  • By tilting to various factors you are actually ADDING risk to your portfolio and decreasing diversification in the original beta-volatility sense.
  • The theory is (I think) that some risks (e.g. value and size) have a better (lower) risk/return ratio. Therefore, adding some of these favored risks gives you better risk-adjusted return. This allows you to perhaps dial back on the overall risk while maintaining the original return potential.
The fact that, above, I emphasize think (in the sense of unsure) is why I am not a factor slice-n-dicer.
1. I understand there is a value factor. What I want to know is what that risk is. As a comparison, in CAPM with a single market factor, the risk is "volatility" (as vineviz stated above). What is the corresponding "risk" for the value factor?
2. I totally agree that by tilting, we are "ADDING" not reducing other risks. I read many posts suggesting otherwise. My impression with factor investors is that they are not aware of the other risks they are adding to their portfolios. (This is what I'm trying to answer.)
3. Yes, in a multi-factor model, one can theoretically lower the ptf volatility without lowering E(r). (However, as you stated, it will increase other risks)

User avatar
patrick013
Posts: 2405
Joined: Mon Jul 13, 2015 7:49 pm

Re: Which is more diversified? TSM or TSM+value?

Post by patrick013 » Sun Aug 19, 2018 8:23 pm

acegolfer wrote:
Sun Aug 19, 2018 9:27 am
The following is my main question on value investing. (I'm not against value investing. Just want to know what your objective is.)

I invest in value (or any other factor such as size, momentum, profitability, investment) because

1) to achieve a higher expected return + lower stdev than TSM (which is true in multi-factor efficient frontier http://faculty.chicagobooth.edu/john.co ... 3q99_4.pdf)

2) or something else?
I use Beta to control risk as opposed to the efficient frontier concept exclusively
and expect similar results adjusting for risk tolerance so If value or other factors
do it so be it. Interesting the discussion of dividends and I still wonder if the
regression results would justify dividends becoming a valid factor. He acknowledges
the regression results then comes up with possible reasons to doubt and later
wants to eliminate dividend's usefulness in market timing. Dividend stocks usually
have lower volatility and value characteristics. Apart from market timing dividends
do a wonderful job of predicting stocks prices of certain indexes IMO.

The Conclusion right before the appendix is worth reading.

You should probably stay with Value and the more known Fama factors for long
term horizons. I don't think the others have enough history yet. But see what
it may do to total return, std deviation, and beta in a backtest. Or just stay with
the 2 fund theorem or 3 fund or whatever. :)
age in bonds, buy-and-hold, 10 year business cycle

rkhusky
Posts: 5699
Joined: Thu Aug 18, 2011 8:09 pm

Re: Which is more diversified? TSM or TSM+value?

Post by rkhusky » Sun Aug 19, 2018 8:53 pm

TSM and TSM+value are equally diversified.

I have slight tilts to value and small to slightly outperform TSM and to be slightly different than TSM in case someone figures out how to game TSM. But not too different from TSM, in case it turns out to be the optimal portfolio.

User avatar
vineviz
Posts: 2004
Joined: Tue May 15, 2018 1:55 pm

Re: Which is more diversified? TSM or TSM+value?

Post by vineviz » Sun Aug 19, 2018 9:49 pm

acegolfer wrote:
Sun Aug 19, 2018 7:04 pm

2. I totally agree that by tilting, we are "ADDING" not reducing other risks. I read many posts suggesting otherwise. My impression with factor investors is that they are not aware of the other risks they are adding to their portfolios. (This is what I'm trying to answer.)
I guess the question is whether people understand that adding risks to a portfolio is the way we achieve diversification.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

MJW
Posts: 570
Joined: Sun Jul 03, 2016 7:40 pm
Location: Pacific Northwest

Re: Which is more diversified? TSM or TSM+value?

Post by MJW » Sun Aug 19, 2018 10:11 pm

acegolfer wrote:
Sun Aug 19, 2018 7:04 pm
1. I understand there is a value factor. What I want to know is what that risk is.
Since this question hasn't been answered directly I'll take a stab at it and maybe someone more knowledgeable can chime in and correct me if needed.

Essentially, the risk you run with value is failure to be rewarded for buying the stock at a depressed price. My terminology may be clumsy here, but the basic idea is that you are buying value stocks under the premise that these companies are "better" than their stock price would suggest. In other words, the market got it wrong and the bar for them is being set too low. You buy the stocks on the belief that the market will eventually catch on and "correct" the price upwards (hopefully to a consequential degree), rewarding you for your wisdom in the process.

However, if the perception of the company doesn't change, you aren't going to be rewarded to the same degree, and if the company performs poorly you will do even worse by having more assets concentrated into the stock. So, value stocks exhibit both beta AND the risk I described above, yet the value risk does not necessarily depend on market risk (beta) to manifest itself.

By investing in a value index fun, the investor is betting that, over time, companies with solid fundamentals that are seemingly under-priced will exceed expectations more often than not. I believe that is the crux of value investing.

Okay - now someone tell me how badly I botched this. :)

acegolfer
Posts: 1092
Joined: Tue Aug 25, 2009 9:40 am

Re: Which is more diversified? TSM or TSM+value?

Post by acegolfer » Mon Aug 20, 2018 11:17 am

MJW wrote:
Sun Aug 19, 2018 10:11 pm
acegolfer wrote:
Sun Aug 19, 2018 7:04 pm
1. I understand there is a value factor. What I want to know is what that risk is.
Since this question hasn't been answered directly I'll take a stab at it and maybe someone more knowledgeable can chime in and correct me if needed.

Essentially, the risk you run with value is failure to be rewarded for buying the stock at a depressed price. My terminology may be clumsy here, but the basic idea is that you are buying value stocks under the premise that these companies are "better" than their stock price would suggest. In other words, the market got it wrong and the bar for them is being set too low. You buy the stocks on the belief that the market will eventually catch on and "correct" the price upwards (hopefully to a consequential degree), rewarding you for your wisdom in the process.
value stocks = under-priced stocks?

User avatar
patrick013
Posts: 2405
Joined: Mon Jul 13, 2015 7:49 pm

Re: Which is more diversified? TSM or TSM+value?

Post by patrick013 » Mon Aug 20, 2018 11:32 am

acegolfer wrote:
Mon Aug 20, 2018 11:17 am

value stocks = under-priced stocks?
Yes but not always. Also low growth but stable market share
revenue-wise. Also quality of earnings is desirable when
estimates are close to reality and higher Sharpe Ratio is
observed.

Value stocks could be "value traps" where lower price ratios
foresee decreasing market share and earnings but the index
selection should uncover that.
age in bonds, buy-and-hold, 10 year business cycle

MJW
Posts: 570
Joined: Sun Jul 03, 2016 7:40 pm
Location: Pacific Northwest

Re: Which is more diversified? TSM or TSM+value?

Post by MJW » Mon Aug 20, 2018 12:46 pm

acegolfer wrote:
Mon Aug 20, 2018 11:17 am
value stocks = under-priced stocks?
Presumably under-priced. They're only under-priced if the market decides later that they deserve to be priced higher. You are hoping you will have purchased the stock before that happens.

Obviously I'm over-simplifying it but I believe that's the basic gist. No one has responded yet to tell me I'm an idiot, so either I have it right or nobody else cares. :happy

acegolfer
Posts: 1092
Joined: Tue Aug 25, 2009 9:40 am

Re: Which is more diversified? TSM or TSM+value?

Post by acegolfer » Mon Aug 20, 2018 2:10 pm

MJW wrote:
Mon Aug 20, 2018 12:46 pm
acegolfer wrote:
Mon Aug 20, 2018 11:17 am
value stocks = under-priced stocks?
Presumably under-priced. They're only under-priced if the market decides later that they deserve to be priced higher. You are hoping you will have purchased the stock before that happens.

Obviously I'm over-simplifying it but I believe that's the basic gist. No one has responded yet to tell me I'm an idiot, so either I have it right or nobody else cares. :happy
My next question to you is, do you think efficient market hypothesis is wrong? Investopedia explains EMH "stocks always trade at their fair value on stock exchanges, making it impossible for investors to either purchase undervalued stocks or sell stocks for inflated prices."

MJW
Posts: 570
Joined: Sun Jul 03, 2016 7:40 pm
Location: Pacific Northwest

Re: Which is more diversified? TSM or TSM+value?

Post by MJW » Mon Aug 20, 2018 3:30 pm

acegolfer wrote:
Mon Aug 20, 2018 2:10 pm
My next question to you is, do you think efficient market hypothesis is wrong? Investopedia explains EMH "stocks always trade at their fair value on stock exchanges, making it impossible for investors to either purchase undervalued stocks or sell stocks for inflated prices."
Before I respond, I want to caveat that I generally try to avoid getting too deep into these types of conversations on this forum. For one, I’m not remotely as well-versed in any of this compared to many of our fellow members. But more importantly, I don’t care enough about arguing with people over the Internet to get immersed in the drawn out, intellectual sparring over semantics or tangential subjects that typically takes place around here. It’s dumb.

Since you asked, I will do my best to answer your question in high-level form and maybe someone else can bail me out and get deeper into it if you are interested. I don’t find EMH to be a particular compelling or useful concept beyond a general sense. As it relates to value, the pricing of the stock (or any stock for that matter) is supposedly assigned according to the “best information available at the time.” I see this as highly subjective. So at the time, who is to say that what we consider a value stock is actually under-priced? It just is what it is because that’s what the market decided at the time. Later, more information comes along and the price changes. Was the market “wrong” before? I’m not sure whether that really matters, because wrong is a human judgment. What matters in the context of value investing is that information comes along to alter the market’s perception about the company in question. If that doesn’t happen, then there’s no value premium for you. The risk didn’t pay off.

The nature of the information and the market’s subsequent reaction to it is laden with behavioral quirks, potentially erroneous assumptions and possibly inaccurate data. All of these things play a role in the movement you see in the market from one day to the next and over longer periods of time. It’s a messy affair. I don't know if it invalidates EMH, but it also isn't important to me whether it does or doesn't.

Post Reply