Value Factor Investing is not Value Investing

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
packer16
Posts: 863
Joined: Sat Jan 04, 2014 2:28 pm

Value Factor Investing is not Value Investing

Post by packer16 » Thu Aug 31, 2017 7:15 am

I think there is some confusion on what value factor investing is and is not. Value factor investing is a trading strategy based upon buying stocks the market attributes low multiples to due to in most part to their high leverage and low margins (reflecting a poor competitive position) across a wide spectrum of securities. This is many ways is the opposite of value investing. Value investing is selecting securities that trade a significant discount to their estimated intrinsic value using conservative assumptions. This value can be more easily estimated for stock that do not have the characteristics of value factor stocks. As a matter of fact, may value investors will not even try to value value factor stocks because their values are so uncertain. I would characterize value factor stocks as crappy stocks in part due to the above characteristics and in part due to the lack of due diligence and relying on diversification to protect the investor from mistakes in selection. So value investing is a exclusive process, rejecting most securities except those who sell at a discount, and value factor investing in inclusive process, including all securities that have high value factor values. Also note value investing is dependent on only a few securities selling at a discount to intrinsic value while value factor investing is dependent upon a more broad base systematic effect occurring. IMO based upon markets becoming more efficient over time specific discounts to intrinsic value are more likely than broad based systematic effects occurring.

The value factor strategy is also based upon a theoretical construct built on past performance, which reflects how the market weighed the theoretical construct at times in the past. We know the market is adaptive so the weight of the construct in the future will be different how different will be shown in future performance. Contrast this with value investing who estimates an independent economic intrinsic value of a security based upon the cash flow or asset value of the asset underlying the security. One is based upon a screen of factors that have generated returns in the past & the other is based upon a valuation of a firm which includes both past performance & future expectations.

Also value investing considers the growth of a firm in its valuation while value factor investing does not. This is what made me skeptical of value factor investing. How can you invest in a strategy that does not consider the growth of a firm? To me this is a cornerstone of valuation but is completely absent in the value factor approach to investing. So when folks say value factor stocks are cheap, what does that mean? Compared to what as growth is not considered in that statement? It only makes sense to say something is cheap compared intrinsic value but with value factor stocks intrinsic value is never estimated. The one historical example of value factor investing before the recent spate of value factors funds beginning in 1993 with DFA is the Graham Rea Fund that used Ben Graham's screens in the Intelligent Investor. This fund was a failure & eventually shut down.

I think understanding the difference between value investing & value factor investing is key because they are not the same thing. Value investing properly implemented can provide returns in excess of the average value factor investing IMO can not as I view value factor investing as sector investing and economically should not result in excess returns over time as it is easily implementable and its risks can be diversified away.

Packer
Buy cheap and something good might happen

DaufuskieNate
Posts: 198
Joined: Wed May 28, 2014 11:53 am

Re: Value Factor Investing is not Value Investing

Post by DaufuskieNate » Thu Aug 31, 2017 8:04 am

There are differences in approaches to what you call "value factor investing." DFA and Bridgeway both use quality screens. An example of a quality screen is the QmJ (Quality minus Junk) factor. This factor includes quantitative measures of profitability, growth, safety and payout. Another way to frame this discussion is to think about it as the difference between quantitative investing and traditional security analysis. An excellent book on this subject is Quantitative Value by Wesley Gray and Tobias Carlisle.

User avatar
packer16
Posts: 863
Joined: Sat Jan 04, 2014 2:28 pm

Re: Value Factor Investing is not Value Investing

Post by packer16 » Thu Aug 31, 2017 8:18 am

Do you know if the quality screens are derived using a similar technique as the value factor, namely deriving a factor based upon historical returns. If so it is subject to the same issues as value in that there is no independent determination of value just what has provided high returns in the past. This can very easily lead to data mining. I think one larger point not discussed much is the underlying assumption in value factor investing that higher returning or lower risk securities can be identified across a large number of securities (a difficult if not impossible task IMO) versus a small number required for traditional value investing to work.

I have read the book & talked with others who do screen extensively and from their and from the author's funds experience, you can see how adaptive the market is in the poor performance of those who attempt to use these screens to invest.

You are correct in that I think Value Factor investing belongs to more in quant investing group versus the traditional value investing group.

Packer
Buy cheap and something good might happen

analogsavior
Posts: 26
Joined: Thu May 19, 2016 3:22 pm

Re: Value Factor Investing is not Value Investing

Post by analogsavior » Thu Aug 31, 2017 8:19 am

Thanks for sharing your thoughts Packer. I was recently thinking about this and came to very similar conclusions. To use your terms, value investing and value factor investing are fundamentally two different things. I don't believe there is a way to invest in value (value investing) that is passive given these differences.

Park
Posts: 452
Joined: Sat Nov 06, 2010 4:56 pm

Re: Value Factor Investing is not Value Investing

Post by Park » Thu Aug 31, 2017 8:47 am

I have great respect for those who are successful at value investing. They deserve the alpha that they get. I'll never be able to do that.

However, I think I may be able to value factor invest. Why be interested in this inferior form of investing? My greatest fear as an investor is a secular bear market. The data I"ve seen is consistent with value factor investing being able to mitigate that risk.

User avatar
Simplegift
Posts: 2553
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Value Factor Investing is not Value Investing

Post by Simplegift » Thu Aug 31, 2017 8:51 am

packer16 wrote:
Thu Aug 31, 2017 7:15 am
I think understanding the difference between value investing & value factor investing is key because they are not the same thing.
Agreed. There is often confusion in Forum discussions between passive "value factor tilting" (ala Fama-French) and active "fundamental value investing" (ala Ben Graham). But these are two completely different investing approaches:
  • To Fama-French, value is determined strictly by a database screen that sorts companies by book value and price. It's assumed that no one can do better than average at discerning how well a company is likely to perform in the future. Because it involves simple database screens, value factor investing lends itself to a mostly passive, index-like approach.
  • In contrast, Ben Graham's fundamental analysis values specific businesses on a bottom-up basis. The value investor’s goal is to estimate a company’s future cash flows and buy it when its share price is trading significantly lower than the intrinsic value implied by these cash flows. It's the epitome of active stock picking.
This is not to make any judgement about the efficacy of these two approaches, but just to observe they are quite different.
Last edited by Simplegift on Thu Aug 31, 2017 9:02 am, edited 1 time in total.
Cordially, Todd

Random Walker
Posts: 1921
Joined: Fri Feb 23, 2007 8:21 pm

Re: Value Factor Investing is not Value Investing

Post by Random Walker » Thu Aug 31, 2017 9:01 am

Packer,
You mention backward looking historical performance supporting value factor. But it is also important to acknowledge forward looking Risk and behavioral based reasons for the value factor to exist. Also, as stated above, I believe a profitability /quality screen on top of the value factor explains much of the difference between your "value factor investing" and your "value investing". My understanding is that Buffet effectively used the profitability screen before the academics caught on to it.

Dave

WoodSpinner
Posts: 231
Joined: Mon Feb 27, 2017 1:15 pm

Re: Value Factor Investing is not Value Investing

Post by WoodSpinner » Thu Aug 31, 2017 9:07 am

OP,

Very interesting post--thanks for helping me grasp the differences.

Are there low cost Mutual Funds or ETFs that focus on Value Investing? Or is this mainly done by an investor though detailed analysis of various individual stocks?

Thanks

:beer

User avatar
privatefarmer
Posts: 266
Joined: Mon Sep 08, 2014 2:45 pm

Re: Value Factor Investing is not Value Investing

Post by privatefarmer » Thu Aug 31, 2017 9:08 am

appreciate the insight. I have been pondering factor-based investing for several years, have become more and more skeptical of it. I think if you wanted to determine the "intrinsic value" of a company it's very subjective as you'd be analyzing not just accounting numbers but what you think the future of the company is. It also would need to be individually computed for each company and would most likely look at different aspects of each company to determine the value. Simply buying a fund that broadly purchases companies based on one or two accounting matrixes does not seem like a way to capture discounted companies based on "intrinsic value". Intrinsic value seems like a far more complex calculation and if 100 people came up with the intrinsic value of a company you'd likely get 100 different answers.

snarlyjack
Posts: 364
Joined: Fri Aug 28, 2015 12:44 pm
Location: Montana

Re: Value Factor Investing is not Value Investing

Post by snarlyjack » Thu Aug 31, 2017 9:37 am

I totality agree with this "Value Factor Investing is not Value Investing".

In my world...I like dividends, for the "Snowball" effect.
But dividend paying companies can be value or growth stocks.
A lot of people think that only value companies pay a dividend
and by wanting dividends make you a value investor which is
not true. Their are high tech growth companies that also pay
a dividend (looking at you, Microsoft).

The question is...are dividends a "Value Factor"? That question
is open for debate. I totality agree with the premise that "value
factor investing is not value investing".

This should be a fun interesting conversation...Thanks Packer16.
This conversation is also hitting on Warren Buffett's wheelhouse.
The most successful investor of all time. I think I just read his
personal asset's are now $77 Billion. He must be doing something right.

DaufuskieNate
Posts: 198
Joined: Wed May 28, 2014 11:53 am

Re: Value Factor Investing is not Value Investing

Post by DaufuskieNate » Thu Aug 31, 2017 9:53 am

Data mining is, of course, something to be on the lookout for with any factor-based investment. On the other hand, there are a lot of steps that can be taken to address this issue. The quality factor has been studied with a number of different definitions. It has been shown to exist in multiple markets and over multiple timeframes. As for looking at historical results, I would ask what investment strategy has ever been developed that didn't look at actual results to test its validity? The problem I see with so-called value investing is that very small changes in assumptions can have a huge impact on the intrinsic value calculation. Most of the value in a stock is in cash flows that take place far into the future. A small change in the growth assumption in the terminal value causes wild swings in intrinsic value. Perhaps this is why different analysts can come up with completely different intrinsic values for a firm. So, how are we to actually employ a good value investing strategy? Run the numbers for yourself? Hard to imagine that most people could come up with winning assumptions on other than a random basis. Look at the historical performance of good analysts? Well, here we are again basing a decision on historical results. However, in this case its been shown that good stock picker performance in one period does not easily translate to future performance.

User avatar
David Jay
Posts: 3792
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Value Factor Investing is not Value Investing

Post by David Jay » Thu Aug 31, 2017 10:01 am

WoodSpinner wrote:
Thu Aug 31, 2017 9:07 am
OP,

Very interesting post--thanks for helping me grasp the differences.

Are there low cost Mutual Funds or ETFs that focus on Value Investing? Or is this mainly done by an investor though detailed analysis of various individual stocks?

Thanks

:beer
By definition, value investing is an active process. I do not believe that there can be a passive "index" fund for value investing.

Costs can still be controlled, so the question becomes: Is the cost of running an active value fund covered by the alpha generated by the specific fund manager.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius

User avatar
Simplegift
Posts: 2553
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Value Factor Investing is not Value Investing

Post by Simplegift » Thu Aug 31, 2017 10:32 am

WoodSpinner wrote:
Thu Aug 31, 2017 9:07 am
Are there low cost Mutual Funds or ETFs that focus on Value Investing? Or is this mainly done by an investor though detailed analysis of various individual stocks?
Low cost and true value investing don't really go together. It takes a lot of work to research and analyze individual companies, including traveling to meet with a company’s customers, suppliers and competitors. One mutual fund company that has been doing this kind of work and following the traditional Ben Graham, intrinsic value approach for decades is Tweedy Browne.

They have several funds, including a Value Fund, a Global Value Fund, and a Global High Dividend Fund. But they are not cheap, with expense ratios in the 1.35%-1.40% range. With these high costs, I'd expect their expenses to eat up most of any alpha they might be able to generate over the long term.

Interestingly, before there were inexpensive small value index funds available, the Tweedy Browne Value Fund used to show up in several of William Bernstein’s asset allocation models in the early 1990s.
Cordially, Todd

User avatar
bertilak
Posts: 5566
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Value Factor Investing is not Value Investing

Post by bertilak » Thu Aug 31, 2017 10:46 am

Packer,

You are 100% correct.

Charlie Munger (of Berkshire Hathaway fame) has a lot to say about this and is quoted and paraphrased in Tren Griffin's book Charlie Munger: The Complete Investor. The difference between the two is clearly distinguished. There is a whole chapter called "Value Investing vs, Factor Investing." A quote:
  • "It is important to draw a clear and simple definitional distinction between value as a statistical factor (Fama/French) and value as an analytical style or goal (Ben Graham)."
It is further pointed out that the two are not simply two ways (top down vs bottom up) to answer the same question but are looking to answer different questions.

Griffin makes an analogy, about how to build a better basketball team. One way is to hire the tallest people available (the Fama/French way). Another way is to hold tryouts and evaluate each prospect's performance (the Ben Graham way). Both methods may result in good teams, but for different reasons.

My take: One way is easy and anyone can do it. The other way is hard and takes both rare skill and lots of work. You can decide for yourself which is more effective or, at least, is more appropriate for you but do not get confused by the terminology.
Listen very carefully. I shall say this only once. (There! I've said it.)

dkturner
Posts: 1265
Joined: Sun Feb 25, 2007 7:58 pm

Re: Value Factor Investing is not Value Investing

Post by dkturner » Thu Aug 31, 2017 10:58 am

Interesting discussion. I prefer the Vanguard Equity Income, Wellesley and Wellington funds for my immersion in "value" investing. Although classified as value funds, none of them are quantitatively driven, like the various value index funds, but historically they have managed to perform well in both "growth" markets and "value" markets, due to the latitude given to the investment managers in defining "value".

User avatar
alec
Posts: 2829
Joined: Fri Mar 02, 2007 2:15 pm

Re: Value Factor Investing is not Value Investing

Post by alec » Thu Aug 31, 2017 12:23 pm

snarlyjack wrote:
Thu Aug 31, 2017 9:37 am
I totality agree with this "Value Factor Investing is not Value Investing".

In my world...I like dividends, for the "Snowball" effect.
But dividend paying companies can be value or growth stocks.
A lot of people think that only value companies pay a dividend
and by wanting dividends make you a value investor which is
not true. Their are high tech growth companies that also pay
a dividend (looking at you, Microsoft).

The question is...are dividends a "Value Factor"? That question
is open for debate. I totality agree with the premise that "value
factor investing is not value investing".

This should be a fun interesting conversation...Thanks Packer16.
This conversation is also hitting on Warren Buffett's wheelhouse.
The most successful investor of all time. I think I just read his
personal asset's are now $77 Billion. He must be doing something right.
If Microsoft is a high tech growth company then why do a lot of the index and actively managed value funds own it? For example Vanguards LC value index has it as the top holding, but not the LC growth index fund.
"It is difficult to get a man to understand something, when his salary depends upon his not understanding it!" - Upton Sinclair

avalpert
Posts: 5678
Joined: Sat Mar 22, 2008 4:58 pm

Re: Value Factor Investing is not Value Investing

Post by avalpert » Thu Aug 31, 2017 2:03 pm

alec wrote:
Thu Aug 31, 2017 12:23 pm
snarlyjack wrote:
Thu Aug 31, 2017 9:37 am
I totality agree with this "Value Factor Investing is not Value Investing".

In my world...I like dividends, for the "Snowball" effect.
But dividend paying companies can be value or growth stocks.
A lot of people think that only value companies pay a dividend
and by wanting dividends make you a value investor which is
not true. Their are high tech growth companies that also pay
a dividend (looking at you, Microsoft).

The question is...are dividends a "Value Factor"? That question
is open for debate. I totality agree with the premise that "value
factor investing is not value investing".

This should be a fun interesting conversation...Thanks Packer16.
This conversation is also hitting on Warren Buffett's wheelhouse.
The most successful investor of all time. I think I just read his
personal asset's are now $77 Billion. He must be doing something right.
If Microsoft is a high tech growth company then why do a lot of the index and actively managed value funds own it? For example Vanguards LC value index has it as the top holding, but not the LC growth index fund.
Yep, it is the largest weighted security in the CRSP Large Value index and #4 in MSCI's Large Value index but isn't in S&P or Russell's Large value indexes - methodology matters in the end.

As for dividends, well they are just a distraction and the 'snowball' effect a concept for marketing investment schemes, not investment theory.

rkhusky
Posts: 4205
Joined: Thu Aug 18, 2011 8:09 pm

Re: Value Factor Investing is not Value Investing

Post by rkhusky » Thu Aug 31, 2017 2:18 pm

WoodSpinner wrote:
Thu Aug 31, 2017 9:07 am
OP,

Very interesting post--thanks for helping me grasp the differences.

Are there low cost Mutual Funds or ETFs that focus on Value Investing? Or is this mainly done by an investor though detailed analysis of various individual stocks?

Thanks

:beer
Vanguard's Wellington and Wellesley funds are low-cost balanced funds, whose stock component is Large Value.

rkhusky
Posts: 4205
Joined: Thu Aug 18, 2011 8:09 pm

Re: Value Factor Investing is not Value Investing

Post by rkhusky » Thu Aug 31, 2017 2:26 pm

David Jay wrote:
Thu Aug 31, 2017 10:01 am

By definition, value investing is an active process. I do not believe that there can be a passive "index" fund for value investing.
I think if you had a source of automated company-specific information, you could feed that information into a computer program to choose companies in which to invest, which I suppose is what is called a quant fund. I would consider it passive investing and should be low cost, if you could just set it going and let it do its work without much day to day management.

User avatar
nisiprius
Advisory Board
Posts: 33784
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: Value Factor Investing is not Value Investing

Post by nisiprius » Thu Aug 31, 2017 3:08 pm

Agreed, they are different, and this represents a problem for beginning investors. The same thing is true of growth. There are a large number of older actively-managed funds from the pre-Fama-French days that have the worlds "growth" and "value" in their names that practice active stock-picking according to managers' analysis of fundamentals, and newer ones that invest in stocks that load heavily on the growth or value factors.

Another issue is that while you probably can find some objective agreement as to what "value factor investing" is, you will not find any such agreement about value investing. There is a further problem of the "no true Scotsman fallacy" kind in that value investing advocates will say that any value investing strategy that didn't work was not truly a value investing strategy. This is very notable in the case of Bill Miller and the Legg Mason Value Trust. Miller was universally regarded as the epitome of value investors during the period when his fund beat the S&P 500 fifteen years running, and his habit of buying more of his chosen stocks when they tanks was praised as being the indicator of a real value investor. When his fund blew up, people then said that they never thought Miller was really a value investor, and began describing the same behavior they'd praised before as "doubling down on his bad bets."
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.

Case59
Posts: 51
Joined: Fri Dec 30, 2011 12:31 pm

Re: Value Factor Investing is not Value Investing

Post by Case59 » Thu Aug 31, 2017 3:47 pm

Thanks, OP, this is a really instructive thread.
"Most quotations on the internet are incorrect."-Mark Twain

stlutz
Posts: 3919
Joined: Fri Jan 02, 2009 1:08 am

Re: Value Factor Investing is not Value Investing

Post by stlutz » Thu Aug 31, 2017 7:22 pm

Actually, I think I partially disagree with packer's post. When I think "value investing", the place I start is Graham and Dodd. The thing about them is that for their time they were highly quantitative in their approach. They were more focused on the facts of the numbers than on more intangible questions like whether they thought the CEO was doing a good job. Calculating a number of ratios and then selecting the companies that are the "best" is very much in line with what they did.

I do agree that they had more of a "multi factor" approach as opposed to only looking for low price-to-X ratios. But low-price-to-X ratios were pretty key to their stock selection approach.

Their approach was also based on more on a theoretical construction than a backtesting based one, so that is a potential difference. But that's probably not so much due to philosophy than the fact that you couldn't do quantitative backtesting back then.

They also thought that they were limiting risk with their approaches; the Fama/French disciples today believe that they are maximizing their risk by investing in value stocks. They might be doing similar things, but their understanding of what they are doing is quite different.

Going through 10-Ks and finding raw data to calculate ratios sounds like detailed analysis. Nowadays somebody pulls up their financial database and can get dozens of calculations on a stock in a matter of seconds. Does that mean that the later is not doing "value investing?". I would say no.

User avatar
siamond
Posts: 3295
Joined: Mon May 28, 2012 5:50 am

Re: Value Factor Investing is not Value Investing

Post by siamond » Thu Aug 31, 2017 7:57 pm

Thank you, packer16, I never really formalized this distinction in my mind. Very useful.

So... just to clarify... How would you classify this ETF: iShares Edge MSCI USA Value Factor ETF (VLUE)? :wink:

User avatar
nedsaid
Posts: 8262
Joined: Fri Nov 23, 2012 12:33 pm

Re: Value Factor Investing is not Value Investing

Post by nedsaid » Fri Sep 01, 2017 12:13 am

Simplegift wrote:
Thu Aug 31, 2017 8:51 am
packer16 wrote:
Thu Aug 31, 2017 7:15 am
I think understanding the difference between value investing & value factor investing is key because they are not the same thing.
Agreed. There is often confusion in Forum discussions between passive "value factor tilting" (ala Fama-French) and active "fundamental value investing" (ala Ben Graham). But these are two completely different investing approaches:
  • To Fama-French, value is determined strictly by a database screen that sorts companies by book value and price. It's assumed that no one can do better than average at discerning how well a company is likely to perform in the future. Because it involves simple database screens, value factor investing lends itself to a mostly passive, index-like approach.
  • In contrast, Ben Graham's fundamental analysis values specific businesses on a bottom-up basis. The value investor’s goal is to estimate a company’s future cash flows and buy it when its share price is trading significantly lower than the intrinsic value implied by these cash flows. It's the epitome of active stock picking.
This is not to make any judgement about the efficacy of these two approaches, but just to observe they are quite different.
My understanding is that the academics are using more than price to book, but also price to earnings, price to sales, and price to cash flow. DFA now screens to set momentum to neutral and also now screens for profitability/quality. So it seems that DFA is moving towards a definition of value that is closer to Warren Buffett and Charlie Munger. My contention is that the academic approach and the Benjamin Graham approach are converging. Buffett under the influence of Munger is now primarily looking for Quality, but Quality at a good price. Buffett is not strictly a Benjamin Graham disciple, Munger was also a huge influence, hence the emphasis on Quality.
A fool and his money are good for business.

venkman
Posts: 177
Joined: Tue Mar 14, 2017 10:33 pm

Re: Value Factor Investing is not Value Investing

Post by venkman » Fri Sep 01, 2017 12:31 am

Doesn't the idea of value investing conflict with the Efficient Markets Hypothesis, at least to some degree?

Assuming I'm an expert (which I'm not), if I can identify a stock that is currently undervalued, why can't all the other experts? And if the other experts CAN identify the undervalued stock, why aren't they buying it until it's no longer undervalued?

AlohaJoe
Posts: 2324
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Value Factor Investing is not Value Investing

Post by AlohaJoe » Fri Sep 01, 2017 1:24 am

venkman wrote:
Fri Sep 01, 2017 12:31 am
Doesn't the idea of value investing conflict with the Efficient Markets Hypothesis, at least to some degree?
No one believes in the caricature of the Efficient Markets Hypothesis anymore -- and they haven't for probably 20 years, though that hasn't necessarily been widely understood. Robert Shiller became famous before his Irrational Exuberance and his CAPE because of his work in the early 1980s showing that the Efficient Markets Hypothesis didn't hold. Among other things, someone has to do price/information discovery, provide liquidity, and perform arbitrage. All of those things will provide some level of profit to the person performing them. Most academics today probably believe that markets are what Lasse Pedersen calls "Efficiently Inefficient":
Rather, they are inefficient enough that money managers can be compensated for their costs through the profits of their trading strategies and efficient enough that the profits after costs do not encourage additional active investing.
venkman wrote:
Fri Sep 01, 2017 12:31 am
Assuming I'm an expert (which I'm not), if I can identify a stock that is currently undervalued, why can't all the other experts? And if the other experts CAN identify the undervalued stock, why aren't they buying it until it's no longer undervalued?
Other experts can. And they will buy it until it is no longer undervalued. But that doesn't happen instantly at every moment.

Plus, there are other factors in play like liquidity, limits of leverage, behavioural biases, market regulations, manager career risk (buying Apple even if they think it is overvalued because they know they're suffer redemptions if they're the only ones who don't buy Apple), misaligned incentives (managers getting money on gains but not losing money on losses), and so on.

User avatar
packer16
Posts: 863
Joined: Sat Jan 04, 2014 2:28 pm

Re: Value Factor Investing is not Value Investing

Post by packer16 » Fri Sep 01, 2017 7:31 am

Thanks for the continued insight. Another difference between the two is the question each is trying to answer. Value investing per Graham is trying to answer the question of what makes a low risk high return investment and factors answer the question of what has driven difference in returns cross sectionally in the past. So value investing is focused on finding good investment opportunities today versus factors explaining the theoretical constructs that have worked good in the past.

What has happened with factor investing is folks have taken the constructs used to explain the differences in past returns & apply them to the market today. This will work fine with a set of variables that do not change or are slowly changing. The statistical tools used to derive factors are based are based upon this assumption. However, market are are adaptive & thus have variables that change rapidly so what has generated returns in the past is not what will generate returns in the future. One way to think about it is the market is changing the factor weights all the time & the past weights will be different than the future weight. As an observation, factors are interesting but using it as a basis for an investment strategy leads all kinds of issues versus value investing which was designed as a strategy to begin with. These issues include the changing weights the markets are putting on factors & the resulting largely cyclical behavior of factor based funds & the large slippage between past performance & future performance. There is also the issue that the set of factors is not complete, thus IMO diversification is difficult if not impossible without a complete set. If you are missing a factor, no matter how diversified you are amongst factors you will have more volatility and less return than the market portfolio that includes all know & unkown factors.

As to faith in an approach, I think you need to understand what you are having faith in. Faith in value investing is faith in the market will reflect the value of future cash flows of a group of assets. I have pretty good faith in this & this can be what Graham called investment. Faith in the value factor, is faith that the market will weigh the value factor by the same weights as in the past. This IMO is speculation not investment. In value investing, the variables are more in the control of management of what to do with the resulting cash flows so the exercise becomes one of determining how good management is at capital allocation. There may be a temporary misunderstanding by the market but given a pretty efficient market this will be resolved pretty quickly. In factor investing, the results are more dependent upon factors outside the control of the investor so the increase in timing risk observed.

As to iShares Edge MSCI USA Value Factor ETF, it is an interesting offering as it value weighs companies by industry group. This I think is moving in the right direction as it reduces the differences in growth issue between industries but not within industries. However, it still has the factor weight timing issue & has a high turnover at 81%.

As to the behavioral or risk based explanations for factor based investing, IMO this is rationalization not explanation. This how you get the perverse result of a given stock with a lower P/E being more risky than the same stock with a higher P/E. This does not make sense. Think about what you are being asked to believe in in factor investing - that the factors will be weighted in the same manner as in the past so you can obtain the same returns. Also that you will obtain more diversification from an incomplete set of factors. IMO these are pretty big things to believe in versus the value investing issues of management correctly allocating capital and the market rewarding cash flows. Now the purveyors have of factors have tried everything out there to magnify the effects of factors from leverage to long/short to take advantage of these observations & to my knowledge none have been as successful as plain old value investing.

Packer
Buy cheap and something good might happen

Random Walker
Posts: 1921
Joined: Fri Feb 23, 2007 8:21 pm

Re: Value Factor Investing is not Value Investing

Post by Random Walker » Fri Sep 01, 2017 7:56 am

Packer16,
At least the size, value, Momentum, and market factors have risk based and / or behavioral based forward looking reasons to expect them to persist. It's not only backwards looking data.
The TSM portfolio may have all stocks, but no net exposure to factors. That is one of the biggest risks of a TSM portfolio, its volatility. And volatility drag costs real money. Diversifying across factors and keeping expected return constant should decrease portfolio volatility compared to TSM portfolio.

Dave

User avatar
packer16
Posts: 863
Joined: Sat Jan 04, 2014 2:28 pm

Re: Value Factor Investing is not Value Investing

Post by packer16 » Fri Sep 01, 2017 8:23 am

Dave,

IMO the key word is your statement is should. Also as shown above many of the risk/behavioral explanations are not explanations but rationalization of why factors have disappointed & why you should hang in there with this strategy despite its weaknesses. IMO I think it is important for folks to understand what they have faith in with the value factor is not value investing but something very different. I agree It has provided diversification in the past with the weights the markets have put on factors but we know the weights will change in the future. IMO it is speculation that the weights will change to provide a benefit versus a drawback. The drawback is what we have seen with the value factor over the past few years.

Packer
Buy cheap and something good might happen

DaufuskieNate
Posts: 198
Joined: Wed May 28, 2014 11:53 am

Re: Value Factor Investing is not Value Investing

Post by DaufuskieNate » Fri Sep 01, 2017 8:47 am

Every concept in investing should be prefaced with the word should, including the 3-fund approach that is so comforting to many here.

User avatar
siamond
Posts: 3295
Joined: Mon May 28, 2012 5:50 am

Re: Value Factor Investing is not Value Investing

Post by siamond » Fri Sep 01, 2017 9:20 am

packer16 wrote:
Fri Sep 01, 2017 7:31 am
What has happened with factor investing is folks have taken the constructs used to explain the differences in past returns & apply them to the market today. This will work fine with a set of variables that do not change or are slowly changing. The statistical tools used to derive factors are based are based upon this assumption. However, market are are adaptive & thus have variables that change rapidly so what has generated returns in the past is not what will generate returns in the future.
Well, although I see your point, I am not so convinced that factor variables are changing that rapidly. Could you please elaborate on this, providing some evidence? Also you might be underestimating human nature, greed and foolishness. If there is one thing that does NOT change, this is it.

Now, say you're right, and the pace of changes is indeed quite fast and unpredictable, then this would be the exact reason to NOT do any form of value investing, because nobody would be prescient enough to make it work, whether they use backwards-looking metrics or forward-looking metrics or an ever evolving combo. And fact is, besides Ben Graham and Warren Buffet, many people tried value investing and royally failed to sustainably make a difference.

Random Walker
Posts: 1921
Joined: Fri Feb 23, 2007 8:21 pm

Re: Value Factor Investing is not Value Investing

Post by Random Walker » Fri Sep 01, 2017 9:33 am

I tend to agree with Siamond. I don't think human behavior is going to change much. Also, I believe Risk and reward should be related. The market prices risk. So I think there is risk based reason to believe in market and small, behavioral to believe in Momentum, and both behavioral and risk for value. I'm not sure what to make of profitability/quality.

Dave

User avatar
nedsaid
Posts: 8262
Joined: Fri Nov 23, 2012 12:33 pm

Re: Value Factor Investing is not Value Investing

Post by nedsaid » Fri Sep 01, 2017 11:02 am

venkman wrote:
Fri Sep 01, 2017 12:31 am
Doesn't the idea of value investing conflict with the Efficient Markets Hypothesis, at least to some degree?

Assuming I'm an expert (which I'm not), if I can identify a stock that is currently undervalued, why can't all the other experts? And if the other experts CAN identify the undervalued stock, why aren't they buying it until it's no longer undervalued?
I actually think that markets are fairly efficient and that Wall Street has legions of very smart and talented people to value stocks. It really is about human behavior and human nature, the old fear and greed thing. Investors like what is popular and what has recently performed well. Investors as a whole make behavioral errors over and over again and that is why factors like Value exist.
A fool and his money are good for business.

Random Walker
Posts: 1921
Joined: Fri Feb 23, 2007 8:21 pm

Re: Value Factor Investing is not Value Investing

Post by Random Walker » Fri Sep 01, 2017 12:01 pm

I've believed in value factor as risk story from the start. As time has passed, I've gained more and more belief in the behavioral component. It's great that value has both risk based and behavioral explanations. Perhaps this is why it is nearly as persistent as the market factor.

Dave

avalpert
Posts: 5678
Joined: Sat Mar 22, 2008 4:58 pm

Re: Value Factor Investing is not Value Investing

Post by avalpert » Fri Sep 01, 2017 1:23 pm

packer16 wrote:
Fri Sep 01, 2017 7:31 am

As to faith in an approach, I think you need to understand what you are having faith in. Faith in value investing is faith in the market will reflect the value of future cash flows of a group of assets. I have pretty good faith in this & this can be what Graham called investment.
I think you are missing an article of faith in there - faith that your assessment of future cash-flows is correct.

garlandwhizzer
Posts: 1559
Joined: Fri Aug 06, 2010 3:42 pm

Re: Value Factor Investing is not Value Investing

Post by garlandwhizzer » Fri Sep 01, 2017 2:08 pm

Very interesting discussion. Packer has raised some good questions, pointing out that value investors like Warren Buffett are not fans of value indexes and rather determine intrinsic value based on thorough research of a single company. No one argues with the concept that cheap beats expensive in the long run, but there is disagreement about how best to define cheap and capacity restraints as more money seeks cheap, driving up its cost.

As for the SCV factor, one thing that is clear on backtesting is that the ride has not been a smooth one. During times of robust economic growth with moderate but not extreme inflation, the typically more indebted and more vulnerable companies in SCV indexes are disproportionally lifted by the rising economic tide. Their top and bottom lines grow and their future fixed debt obligations decline in real inflation-adjusted terms, both of which raise their ugly duckling prices. During recessions and difficult macroeconomic times SCV tends to suffer more and underperform growth. Even in difficult times, growing companies with strong balance sheets can still generate profits and often grow them.

Many analysts currently believe that, largely due to the duo of demographics and debt, the US has now entered the "new normal," a secular period of slow economic growth and persistently low inflation resistant to maximal monetary policy stimulus. How long that will last is unknown but currently that is likely the foreseeable future. Japan has been stuck in that situation for 28 years and continues to struggle with it following the collapse of the largest bubble in modern history. It is noteworthy that we have ourselves been through the popping of 2 asset bubbles, tech and real estate. In general post-bubble recoveries are prolonged and difficult (Great Depression, Japan, Great Recession's still ongoing to some extent). If we continue to struggle with sluggish growth and microscopic inflation for long years into the future how will that affect expected SCV returns? My guess is that absent a return to robust economic growth and increasing inflation, SCV will continue to struggle just as it has under these circumstances in recent years. DFSVX has currently underperformed VTI for the last 13 years and there could be significantly more to come. If you don't have incredible patience, a rock-solid belief in factor investing, and a very long time frame SCV may not be for you.

Choosing a market-cap-weighted index portfolio does not automatically mean that you're destined to underperform factor based strategies going forward IMO. It merely means that you place more faith in the market's ability to set asset prices than a backtesting-derived index model designed to separate future winners from losers.

Garland Whizzer

User avatar
Robert T
Posts: 2432
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: Value Factor Investing is not Value Investing

Post by Robert T » Fri Sep 01, 2017 2:35 pm

.
May be of interest – an earlier related take on value investing viewtopic.php?t=52397

Ben Graham [“Value”] = exploiting mis-pricing e.g. 'buying a dollar for 50 cents' [compensated when everyone else realizes the market got the pricing wrong], need belief that: (i) mis-pricing will continue in the future, and (ii) an investor can identify where these exist (or can select a fund managers who applies the Ben Graham approach successfully).

Fama/French [“Value (risk) factor”]
= exposure to greater risk that is somewhat different to overall ‘market risk’ [compensated with an expected risk premium], need belief that: (i) capital markets will on average reward greater risk with higher returns over the long-term, and (ii) that this risk will continue to be somewhat different from overall ‘market risk’. “It all comes down to do you think its risk or not” - 5:40 minute mark on interview with Fama - https://www.youtube.com/watch?v=HIKO-t4vU6Q

Robert
.

DaufuskieNate
Posts: 198
Joined: Wed May 28, 2014 11:53 am

Re: Value Factor Investing is not Value Investing

Post by DaufuskieNate » Fri Sep 01, 2017 2:37 pm

garlandwhizzer wrote:
Fri Sep 01, 2017 2:08 pm

Choosing a market-cap-weighted index portfolio does not automatically mean that you're destined to underperform factor based strategies going forward IMO. It merely means that you place more faith in the market's ability to set asset prices than a backtesting-derived index model designed to separate future winners from losers.
I think there is another way of looking at this. By investing in SCV you ARE placing faith in the market's ability to set asset prices. For all the reasons you mention, SCV is riskier and the market appropriately sets prices to generate a higher return.

stlutz
Posts: 3919
Joined: Fri Jan 02, 2009 1:08 am

Re: Value Factor Investing is not Value Investing

Post by stlutz » Fri Sep 01, 2017 4:48 pm

I think there is another way of looking at this. By investing in SCV you ARE placing faith in the market's ability to set asset prices. For all the reasons you mention, SCV is riskier and the market appropriately sets prices to generate a higher return.
SCV is generally categorized as something in the range of 3-7% of the market. So let's just assume 5% for now.

It is *of course* the case that investing in 5% of the market is riskier than investing in 100% of the market. The real question is whether SCV is riskier than any other 5% slice of the market--say, smallcap growth or the materials sector.

The logic of the SCV as riskier understanding actually goes the opposite way to what is sometimes presented in this forum--that is:

--Riskier investments should usually outperform
--SCV has outperformed other segments of the market in historical backtests
--Ergo, SCV was riskier than other segment of the market

You can't then reverse the direction of this approach to claim the SCV will therefore outperform in the future because it's riskier. SCV being riskier is not something that has been observed; it's something that's been concluded based on an assumption about risk premiums and an observation about performance.
Last edited by stlutz on Fri Sep 01, 2017 5:30 pm, edited 1 time in total.

User avatar
siamond
Posts: 3295
Joined: Mon May 28, 2012 5:50 am

Re: Value Factor Investing is not Value Investing

Post by siamond » Fri Sep 01, 2017 5:02 pm

stlutz wrote:
Fri Sep 01, 2017 4:48 pm
[...] SCV being riskier is not something that has been observed; it's something that's been concluded based on an assumption about risk premiums and an observation about performance.
That is an intriguing way to put it. Well, one observation is that backtesting shows that SCV Sharpe/Sortino ratios were usually better than TSM (at least in the US). Why? Well, because std-deviation was not much higher than TSM, and returns were definitely better. Ergo, I would observe that SCV has been LESS risky than TSM, if one accepts the (very dubious!) definition of risk from Sharpe and co.

stlutz
Posts: 3919
Joined: Fri Jan 02, 2009 1:08 am

Re: Value Factor Investing is not Value Investing

Post by stlutz » Fri Sep 01, 2017 5:30 pm

That is an intriguing way to put it. Well, one observation is that backtesting shows that SCV Sharpe/Sortino ratios were usually better than TSM (at least in the US). Why? Well, because std-deviation was not much higher than TSM, and returns were definitely better. Ergo, I would observe that SCV has been LESS risky than TSM, if one accepts the (very dubious!) definition of risk from Sharpe and co.
Over the past 50 years or so, if you compare value to growth (as opposed to TSM), you of course find that value has had less volatility and higher returns. So, to me the question is whether that is an example of the risks of value not showing up, in which case I shouldn't really expect a future repeat of higher returns/lower risk. Or is it the case that value is in fact less risky, which might suggest that the pattern will repeat?

I've argued both ways in previous threads. :P

User avatar
siamond
Posts: 3295
Joined: Mon May 28, 2012 5:50 am

Re: Value Factor Investing is not Value Investing

Post by siamond » Fri Sep 01, 2017 5:45 pm

stlutz wrote:
Fri Sep 01, 2017 5:30 pm
Over the past 50 years or so, if you compare value to growth (as opposed to TSM), you of course find that value has had less volatility and higher returns. So, to me the question is whether that is an example of the risks of value not showing up, in which case I shouldn't really expect a future repeat of higher returns/lower risk. Or is it the case that value is in fact less risky, which might suggest that the pattern will repeat?
Ah ok, I get your point better now. Thanks for clarifying.

PS. there is a a third possibility, the Bob Dylan answer, which is that times are a-changin'... Which we can never exclude. But personally, I tend to prefer historical observations to free speculation, while still taking such observations with a grain of salt.

stlutz
Posts: 3919
Joined: Fri Jan 02, 2009 1:08 am

Re: Value Factor Investing is not Value Investing

Post by stlutz » Fri Sep 01, 2017 5:47 pm

But personally, I tend to prefer historical observations to free speculation, while still taking such observations with a grain of salt.
I like that way of putting it. I agree.

User avatar
packer16
Posts: 863
Joined: Sat Jan 04, 2014 2:28 pm

Re: Value Factor Investing is not Value Investing

Post by packer16 » Sun Sep 03, 2017 9:47 am

Thanks Robert T. for posting the definitions especially the interview. Clearly Fama is talking about something different than value investing as described by B. Graham. I think the use of the term value for the "return challenged" firms is confusing as even the interviewer is confused and Fama does not correct him. Also most value investors today invest in the opposite type of securities firms with a some type of moat. In value investing, growth & earnings stability are important to determining the intrinsic value where as for the return challenged factor they are not. I find it interesting that the reason for risk (leverage and low margins) are not screened directly versus indirectly via value. Maybe it has been done & the returns are not there so there is more to the return challenged factor than leverage & low margins.

One item folks keep going back to is data before 1993 for the "return challenged" factor. I think this is just a theoretical exercise as the market is adaptive and the data before 1993 does not include the effects this strategy has on prices. In 1993, DFA opened a fund that invested in the value (aka returned challenged) factor. If we look at performance since 1993, IMO your impression of SCV will change materially. Since 1993, SCV has outperformed the S&P 500 by 0.8% per year but the swings are very large thus leading to a large timing risk. Also the pattern of returns imply that the factor may be arbed away as the returns early on to 2005 are positive but have been negative every since. So why take so much timing risk to gain 0.8% (& falling) per year of additional return?

The question of forward looking persistence IMO is still open because as each of the factors have been implemented in portfolios, the consequences has been worse than in the past & in some cases even negative, value since 2005. I just do not see the diversification argument as anything but marketing as "diversifying" with these factors once they are widely invested has lead to lower returns than if you stay put in the standard market weighted portfolio.

One Boglehead tenant not stated but implied is the reliance on the market to provide the best average aggregate estimate of weights of each security in a portfolio/index. Is not factor investing/weighing saying the market is not correctly weighing these factors and a factor weighted portfolio is better than the market? If so, I think this is a pretty big assumption & probably the most unBoglehead aspect of factor investing.

As to human behavior driving factor performance, where is the link? Is it just a residual to what cannot be explained by the risk that no one can measure accurately? I can clearly see behavior driving performance in a value investing framework, folks transacting at prices blow or above intrinsic value but in a value factor framework it is a little more difficult. This can happen on micro-level pretty easily but it IMO it is much more difficult on a macro level which is what must happen for factors to work.

Packer
Buy cheap and something good might happen

snarlyjack
Posts: 364
Joined: Fri Aug 28, 2015 12:44 pm
Location: Montana

Re: Value Factor Investing is not Value Investing

Post by snarlyjack » Sun Sep 03, 2017 11:52 am

Packer16,

(This is just me & my requirements).

I love to invest in companies that I've heard of.
(Coke-Cola, Pepsi, Microsoft, Burlington Northern, etc).

If I've never heard of the companies... I' am very skeptical of
investing in them. So my very first requirement is their
(advertising) or name recognition.

Would you rather invest in Anheuser Busch or Full Sail Brewing?
Some people would invest in Anheuser Busch & some people
would invest in Full Sail. I personally would invest in the value stock
Anheuser Busch.

avalpert
Posts: 5678
Joined: Sat Mar 22, 2008 4:58 pm

Re: Value Factor Investing is not Value Investing

Post by avalpert » Sun Sep 03, 2017 12:16 pm

snarlyjack wrote:
Sun Sep 03, 2017 11:52 am
Packer16,

(This is just me & my requirements).

I love to invest in companies that I've heard of.
(Coke-Cola, Pepsi, Microsoft, Burlington Northern, etc).

If I've never heard of the companies... I' am very skeptical of
investing in them. So my very first requirement is their
(advertising) or name recognition.

Would you rather invest in Anheuser Busch or Full Sail Brewing?
Some people would invest in Anheuser Busch & some people
would invest in Full Sail. I personally would invest in the value stock
Anheuser Busch.
Yikes, personal brand recognition sounds like an even worse criteria than chasing dividend yield.

totality
Posts: 30
Joined: Mon Aug 21, 2017 1:06 pm

Re: Value Factor Investing is not Value Investing

Post by totality » Sun Sep 03, 2017 12:24 pm

packer16 wrote:
Sun Sep 03, 2017 9:47 am
One Boglehead tenant not stated but implied is the reliance on the market to provide the best average aggregate estimate of weights of each security in a portfolio/index. Is not factor investing/weighing saying the market is not correctly weighing these factors and a factor weighted portfolio is better than the market? If so, I think this is a pretty big assumption & probably the most unBoglehead aspect of factor investing.
One question I ask myself when thinking about overweighting and underweighting relative to the market is, "Can I explain when I would want to reverse this tilt without looking at market performance?"

For example, I (and I think 99.9% of Bogleheads) do not hold a ratio of stocks and bonds at market weight. Rather, we consider our investment timeline and manage our "percent in bonds" in accordance our personal circumstance. And furthermore most of use would say that someone retiring tomorrow should not be 100/0, regardless of how great stocks have been doing or how poorly bonds have been doing.

But when you consider, for example, a tilt to small-cap value, (as seems popular these days), it's not clear to me when the person advocating the tilt would reverse the tilt. If someone thinks SCV outperforms the market because it has higher risk, are they going to underweight SCV as they get closer to retirement? Or if they are okay with more risk, why is overweighting SCV any better than just increasing their overall equity allocation? Or if they believe that SCV outperforms without additional risk, how long do they think that outperformance will continue? How will they know when it's time to go back to market weight, or maybe even under weight?

I can't help but feel like it's return chasing a lot of the time. We soberly and rationally select our AA, and then we immediately start trying to increase our return, either by throwing in lower quality bonds (more corporates, or even high yield), or tilting to more volatile sectors of the stock market. Sure, you can maybe get a better return that way, but you took on more risk, so is it really better than if you had just kept to a three fund portfolio and boosted your equity allocation by a few percent?

snarlyjack
Posts: 364
Joined: Fri Aug 28, 2015 12:44 pm
Location: Montana

Re: Value Factor Investing is not Value Investing

Post by snarlyjack » Sun Sep 03, 2017 12:33 pm

Avalpert,

Ok...I get it.
Let's call it "Quality".
I like quality stocks that I've heard of.

I like "warm fuzzes" & sleep at night factors.
I like the phrase my portfolio "is spinning on a stick
spitting out cash" or "I can sit back in the weeds &
wait it out without losing any sleep".

Those are my "sleep well at night factors" ...

avalpert
Posts: 5678
Joined: Sat Mar 22, 2008 4:58 pm

Re: Value Factor Investing is not Value Investing

Post by avalpert » Sun Sep 03, 2017 12:36 pm

snarlyjack wrote:
Sun Sep 03, 2017 12:33 pm
Avalpert,

Ok...I get it.
Let's call it "Quality".
I like quality stocks that I've heard of.

I like "warm fuzzes" & sleep at night factors.
I like the phrase my portfolio "is spinning on a stick
spitting out cash" or "I can sit back in the weeds &
wait it out without losing any sleep".

Those are my "sleep well at night factors" ...
But there is little reason to expect a relationship between the quality of a company from an investment perspective and whether you happened to hear of the brand or not. So calling it 'quality' is just a way of misleading yourself into thinking it has something to do with the quality factor.

Your 'sleep well at night factors' may be doing you a huge disservice.

betablocker
Posts: 251
Joined: Mon Jan 11, 2016 1:26 pm

Re: Value Factor Investing is not Value Investing

Post by betablocker » Sun Sep 03, 2017 1:18 pm

packer16 wrote:
Sun Sep 03, 2017 9:47 am
Thanks Robert T. for posting the definitions especially the interview. Clearly Fama is talking about something different than value investing as described by B. Graham. I think the use of the term value for the "return challenged" firms is confusing as even the interviewer is confused and Fama does not correct him. Also most value investors today invest in the opposite type of securities firms with a some type of moat. In value investing, growth & earnings stability are important to determining the intrinsic value where as for the return challenged factor they are not. I find it interesting that the reason for risk (leverage and low margins) are not screened directly versus indirectly via value. Maybe it has been done & the returns are not there so there is more to the return challenged factor than leverage & low margins.

One item folks keep going back to is data before 1993 for the "return challenged" factor. I think this is just a theoretical exercise as the market is adaptive and the data before 1993 does not include the effects this strategy has on prices. In 1993, DFA opened a fund that invested in the value (aka returned challenged) factor. If we look at performance since 1993, IMO your impression of SCV will change materially. Since 1993, SCV has outperformed the S&P 500 by 0.8% per year but the swings are very large thus leading to a large timing risk. Also the pattern of returns imply that the factor may be arbed away as the returns early on to 2005 are positive but have been negative every since. So why take so much timing risk to gain 0.8% (& falling) per year of additional return?

The question of forward looking persistence IMO is still open because as each of the factors have been implemented in portfolios, the consequences has been worse than in the past & in some cases even negative, value since 2005. I just do not see the diversification argument as anything but marketing as "diversifying" with these factors once they are widely invested has lead to lower returns than if you stay put in the standard market weighted portfolio.

One Boglehead tenant not stated but implied is the reliance on the market to provide the best average aggregate estimate of weights of each security in a portfolio/index. Is not factor investing/weighing saying the market is not correctly weighing these factors and a factor weighted portfolio is better than the market? If so, I think this is a pretty big assumption & probably the most unBoglehead aspect of factor investing.

As to human behavior driving factor performance, where is the link? Is it just a residual to what cannot be explained by the risk that no one can measure accurately? I can clearly see behavior driving performance in a value investing framework, folks transacting at prices blow or above intrinsic value but in a value factor framework it is a little more difficult. This can happen on micro-level pretty easily but it IMO it is much more difficult on a macro level which is what must happen for factors to work.

Packer
Good discussion. To the diversification point, factors are just persistent calculations that capture risk and/or behavioral mistakes that investors have and will most likely continue to make. Why don't smart investors arbitrage market risk? Because its risky and the market sets an equilibrium return that compensates you for the risk called the equity risk premium. Holding TSM is simply exposing yourself to that factor: market risk. Holding bonds exposes you to credit and inflation risk. Those are all factors but because they didn't explain all performance the academics started looking if there were systematic things that did and they found them: momentum, value, size, etc. those factors are present across all investments: domestic, international, they show up in currencies, the art market, etc. they have persisted for decades and in some cases hundreds of years (acknowledging the limitations of data).

In some cases they persist because those factors are risky and in other cases because people keep doing dumb things and professionals can't arbitrage away the behavior. The only way I see a factor disappearing is if people or markets fundamentally change in some way. Judging by the behavior of most people I know that is unlikely. And there won't be too many more factors. I think we're up to explaining about 98% of performance with factors. Buffett is great and he figured out factors before others did. Going back to an earlier poster, he fundamentally analyzed the basketball players bottom up and made great choices. Turns out those choices were always choosing a slight taller player who didn't run as fast, shot a lot of three pointers, and played great defense. It wasn't really the individual but the set of characteristics that made up that player. See the AQR paper: http://docs.lhpedersen.com/BuffettsAlpha.pdf.

Again while I love Buffett and Munger. They are certified geniuses but there's also a sort of religion they've built around themselves and Berkshire complete with an annual revival. You could find all the science in the world and the cult of Graham and Doddsville just ain't going to listen.

Post Reply