"value factor" doesn't seem reliable to me

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

"value factor" doesn't seem reliable to me

Post by trademil » Thu Apr 18, 2013 7:27 am

First, I'll state my problem: due to local regulations, a significant portion of my retirement savings is out of my control, and is actively managed with a stocks/bonds ratio of about 30/70, which is way too conservative for me (I'm in my 20s). So I put 100% of the assets that are under my control (execpt for a decent emergency fund) in stocks (TSM+TISM), but It still leaves me with lower risk then I'm able and willing to take, so I'm considering ways to tilt my portfolio to higher risk - higher reward assets in an efficient way.

It seems that many here consider "small-value" as "the" tilt exactly for situations like that. I can partly understand the reasoning behind the "small" part, but it seems most sources claim that the "value" factor is the more significant one.

The issue is, that while "small" factor is based on the market value of the company, which is a straight-forward information reveled by the market, the "value" factor is based on the "book value" of the company, which is not a market-determined information.

A company might report "book value" which have nothing to do with realty, because they don't have to mark their assets to market daily. The book could say an asset of the company is worth what it was worth years ago, while it's actual (market, current) value is far lower.

This is especially true for financial companies, where a major part of the company assets are securities, many of them might be Illiquid, not-listed and so poorly valuated. And what do you know, 40% of Vanguard Small-Cap Value ETF (VBR) is concentrated in the Financials sector:

https://personal.vanguard.com/us/funds/ ... =INT#tab=2

My reasoning is that if the book value is not reliable then you can't split the companies to "value" and "growth" categories because you don't have the data to do so, and therefore "value" factor can't be reliably used for investments.

Furthermore, the historical data on which the whole "value" analyses based might be irrelevant since the effect I have described might have been much weaker prior to the 2008 financial melt-down, when many of the financial assets lost much of their market value and created the big gap from the book value. Maybe "value" funds wasn't always a concentrated bet on the financial sector (where can I see how much of VBR was in the financials sector in 2007?), so it might just be the case that the value factor did have something to offer back then, but not anymore (you know, "past performance is no guarantee of future results").

Thoughts?

EDN
Posts: 528
Joined: Sat Dec 29, 2012 7:55 pm
Location: Oklahoma City, OK

Re: "value factor" doesn't seem reliable to me

Post by EDN » Thu Apr 18, 2013 8:08 am

I just skimmed, something like this comes up weekly. "Value", or more appropriately the high book to market effect identified by FF is just a way to isolate high from low priced (relative to fundamentals) stocks--a distinct expected return from the overall market effect. A ton of empiracle evidence on this. Price is what matters, what you scale price by matters less. Book value just produces the most stability in ratios, leading to very low turnover. This becomes important when you also introduce other sorts, such as one on relative profitability, which has also been shown to have explainatory power.

The Vanguard fund you reference holds REITS and financial stocks, but makes no distinction between the two. It is not necessary to have a heavy concentration in financials if you tilt to small and value. DFA actually excludes REITS from their equity funds, and looking at their total market Core funds, which have increased exposure to the size and value factors relative to the market, we see US Core 2 has 20% financials vs 14% for Russell 3000. Int'l Core has 26% vs 24% for MSCI World ex US, and EM Core has 26% vs 28% for MSCI EM index. Of course, some overweighting of lower priced sectors is natural, as is some underweight of higher priced sectors (like Tech in 1990s).

Eric

trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

Re: "value factor" doesn't seem reliable to me

Post by trademil » Thu Apr 18, 2013 8:43 am

EDN wrote:"Value", or more appropriately the high book to market effect identified by FF is just a way to isolate high from low priced (relative to fundamentals) stocks
I understand that, but I have a problem with the underlying assumption that we know the fundamentals. And if we don't know the fundamentals we can't isolate the stocks that are priced low relative to the fundamentals.

User avatar
Bogle101
Posts: 306
Joined: Mon Oct 01, 2012 10:14 am

Re: "value factor" doesn't seem reliable to me

Post by Bogle101 » Thu Apr 18, 2013 9:01 am

trademil wrote:
EDN wrote:"Value", or more appropriately the high book to market effect identified by FF is just a way to isolate high from low priced (relative to fundamentals) stocks
I understand that, but I have a problem with the underlying assumption that we know the fundamentals. And if we don't know the fundamentals we can't isolate the stocks that are priced low relative to the fundamentals.
Then just go with a regular small-cap fund, that indexes all small caps. Or if you can tolerate it, choose a micro cap fund since you are so heavy in bonds relative to your age
25% S&P 500 | 25% Extended Market | 20% International | 10% REIT | 10% Sector Funds | 10% Cash

gt4715b
Posts: 363
Joined: Mon Jun 11, 2007 10:29 am

Re: "value factor" doesn't seem reliable to me

Post by gt4715b » Thu Apr 18, 2013 9:14 am

trademil wrote:I understand that, but I have a problem with the underlying assumption that we know the fundamentals. And if we don't know the fundamentals we can't isolate the stocks that are priced low relative to the fundamentals.
Book value at its very basic is assets minus liabilities. Putting aside financials (for which book value still works) are you saying that we can't compute what a business owns and what it owes? Of course, there is no one TRUE value for a company at any given moment but it doesn't have to be that precise. For example the Russell 3000 Value index has a P/B ratio of 1.63 whereas the Russell 3000 Growth index has a P/B ratio of 4.38. I think we have a good shot of nailing down book value at that level of precision.

EDN
Posts: 528
Joined: Sat Dec 29, 2012 7:55 pm
Location: Oklahoma City, OK

Re: "value factor" doesn't seem reliable to me

Post by EDN » Thu Apr 18, 2013 9:19 am

trademil wrote:
EDN wrote:"Value", or more appropriately the high book to market effect identified by FF is just a way to isolate high from low priced (relative to fundamentals) stocks
I understand that, but I have a problem with the underlying assumption that we know the fundamentals. And if we don't know the fundamentals we can't isolate the stocks that are priced low relative to the fundamentals.
There's nothing there. Any discrepancy in the value of assets on an individual basis is certainly washed out across a portfolio of thousands of value companies.

Further, even if there was some flaw in how we measure book value, portfolios sorted on this flawed factor still produce the expected results. Value portfolios using other metrics produce similar returns. Across countries, who have different accounting standards, the results persist.

There are a lot of things to worry about in investing. This just isn't one of them.

Eric

EDN
Posts: 528
Joined: Sat Dec 29, 2012 7:55 pm
Location: Oklahoma City, OK

Re: "value factor" doesn't seem reliable to me

Post by EDN » Thu Apr 18, 2013 9:25 am

I find it a bit ironic today when I read folks believe or buy into the size effect, but question the value premium. 5 or 10 years ago, it was the opposite. What changed? Simple: small premium has been high lately, value premium has been more muted. So it's a plain and simple recency effect with rationale created after the fact.

You see somewhat of the same discussion indirectly around the equity premium (negative for 13 years) in terms of constant mention of "stocks as risky assets", ultra-low equity allocations, and this new found interest in immediate annuities and the extremely high cost of certainty. Best to just tune out all the noise, conspiracies, and new paradigms. Traditional investing is tough to beat. There are no concepts in finance that we can explain with 100% certainty, so you just have to accept it.

Eric

YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: "value factor" doesn't seem reliable to me

Post by YDNAL » Thu Apr 18, 2013 9:34 am

trademil wrote:It seems that many here consider "small-value" as "the" tilt exactly for situations like that. I can partly understand the reasoning behind the "small" part, but it seems most sources claim that the "value" factor is the more significant one.

The issue is, that while "small" factor is based on the market value of the company, which is a straight-forward information reveled by the market, the "value" factor is based on the "book value" of the company, which is not a market-determined information.

A company might report "book value" which have nothing to do with realty, because they don't have to mark their assets to market daily. The book could say an asset of the company is worth what it was worth years ago, while it's actual (market, current) value is far lower.

This is especially true for financial companies, where a major part of the company assets are securities, many of them might be Illiquid, not-listed and so poorly valuated.
This is true, "book value" of one specific company may be [edit typo] imprecise to the Nth degree as you described.

That said, the difference in "Growth" companies (HIGH market price over book) in an index vs. "Value" companies (LOW market price over book) is significant. For instance, staying with Vanguard since you mentioned Vanguard Small-Cap Value ETF (VBR) in your OP:
  • 1. VBR has a Price/book ratio of 1.5x
    2. VBK (Growth) has a Price/book ratio of 3.2x - more than twice VBR's Index.
Now, you either believe the LOW price/book in "Value" companies, dictated by market forces, is risk-based with corresponding expected return, or you don't believe that.
Last edited by YDNAL on Thu Apr 18, 2013 11:40 am, edited 2 times in total.
Landy | Be yourself, everyone else is already taken -- Oscar Wilde

richard
Posts: 7961
Joined: Tue Feb 20, 2007 3:38 pm
Contact:

Re: "value factor" doesn't seem reliable to me

Post by richard » Thu Apr 18, 2013 9:42 am

The underlying issue is the value premium is not reliable. Value is a risk factor and sometimes gets a higher return due to higher risk.

If we could clearly identify value stocks and have a high degree of confidence in their performance, then they wouldn't be riskier and therefore there would be no reason to think they might earn more. Only the real chance of substantial under-performance or losses gives the hope of higher performance.

staythecourse
Posts: 5874
Joined: Mon Jan 03, 2011 9:40 am

Re: "value factor" doesn't seem reliable to me

Post by staythecourse » Thu Apr 18, 2013 9:54 am

In the end folks need to just decide do they believe something exists or not. In medicine we often say if one does not want to believe in something no matter how many studies come out will not be enough to connvinc them otherwise.

You have to decide that yourself.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

EDN
Posts: 528
Joined: Sat Dec 29, 2012 7:55 pm
Location: Oklahoma City, OK

Re: "value factor" doesn't seem reliable to me

Post by EDN » Thu Apr 18, 2013 10:03 am

richard wrote:The underlying issue is the value premium is not reliable. Value is a risk factor and sometimes gets a higher return due to higher risk.

If we could clearly identify value stocks and have a high degree of confidence in their performance, then they wouldn't be riskier and therefore there would be no reason to think they might earn more. Only the real chance of substantial under-performance or losses gives the hope of higher performance.
No source of higher EXPECTED return is reliable (read: guaranteed), not stocks over bonds, or long bonds over short ones. That is not specific to this conversation.

Eric

richard
Posts: 7961
Joined: Tue Feb 20, 2007 3:38 pm
Contact:

Re: "value factor" doesn't seem reliable to me

Post by richard » Thu Apr 18, 2013 10:31 am

EDN wrote:No source of higher EXPECTED return is reliable (read: guaranteed), not stocks over bonds, or long bonds over short ones. That is not specific to this conversation.

Eric
People often fail to see the forest for the trees. It can be useful to point out the forest.

I continue to find the term "expected" misleading for many investors. As I said in another thread (again, not specific to this conversation):
-------
Expected return is one of the more misleading terms in investing. Large numbers of investors appear to believe that expected returns are returns they should expect to receive.

An investment with a 99% chance of returning nothing and a 1% chance of returning $1 million has an expected return of $10,000, but $10,000 is not at all a likely return.

Another problem is that we have no reasonable way of estimating the odds for most investments, which makes the entire notion of expected return even less useful.
------

trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

Re: "value factor" doesn't seem reliable to me

Post by trademil » Thu Apr 18, 2013 10:53 am

gt4715b wrote: For example the Russell 3000 Value index has a P/B ratio of 1.63 whereas the Russell 3000 Growth index has a P/B ratio of 4.38.
YDNAL wrote: For instance, staying with Vanguard since you mentioned Vanguard Small-Cap Value ETF (VBR) in your OP:
  • 1. VBR has a Price/book ratio of 1.5x
    2. VBK (Growth) has a Price/book ratio of 3.2x - more than twice VBR's Index.
You have a point here, but, lets add:
3.Vanguard Small-Cap ETF (VB) has a Price/book ratio of 2.1x:

https://personal.vanguard.com/us/funds/ ... =INT#tab=2

So it seems that most of the gap is between the growth and VB, and only the minority is between VB and value. The gap between 2.1 and 1.5 doesn't look that decisive to me.
Given that the financial sector is 40% of VBR, theoretically, it might just be the case that if those financial institutions will mark their securities to market and correct their books accordingly their book value will drop by 70%, which will be a drop of 28% (70*0.4) for the whole fund, which will put it's Price/book of the fund at 2.1, just like VB.

YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: "value factor" doesn't seem reliable to me

Post by YDNAL » Thu Apr 18, 2013 11:51 am

trademil wrote:
YDNAL wrote: For instance, staying with Vanguard since you mentioned Vanguard Small-Cap Value ETF (VBR) in your OP:
  • 1. VBR has a Price/book ratio of 1.5x
    2. VBK (Growth) has a Price/book ratio of 3.2x - more than twice VBR's Index.
You have a point here, but, lets add:
3.Vanguard Small-Cap ETF (VB) has a Price/book ratio of 2.1x:

https://personal.vanguard.com/us/funds/ ... =INT#tab=2

So it seems that most of the gap is between the growth and VB, and only the minority is between VB and value. The gap between 2.1 and 1.5 doesn't look that decisive to me.
Given that the financial sector is 40% of VBR, theoretically, it might just be the case that if those financial institutions will mark their securities to market and correct their books accordingly their book value will drop by 70%, which will be a drop of 28% (70*0.4) for the whole fund, which will put it's Price/book of the fund at 2.1, just like VB.
There is no such "gap." What happens is that 1,440 Stocks in VB (CRSP US Small Cap Index) - at market weights - do not comparatively match the sum of the parts in 1,004 Stocks currently in VBR (CRSP US Small Cap Value Index) and 949 Stocks in VBK (CRSP US Small Cap Growth Index).
Last edited by YDNAL on Thu Apr 18, 2013 11:55 am, edited 1 time in total.
Landy | Be yourself, everyone else is already taken -- Oscar Wilde

EDN
Posts: 528
Joined: Sat Dec 29, 2012 7:55 pm
Location: Oklahoma City, OK

Re: "value factor" doesn't seem reliable to me

Post by EDN » Thu Apr 18, 2013 11:55 am

richard wrote:
EDN wrote:No source of higher EXPECTED return is reliable (read: guaranteed), not stocks over bonds, or long bonds over short ones. That is not specific to this conversation.

Eric
People often fail to see the forest for the trees. It can be useful to point out the forest.

I continue to find the term "expected" misleading for many investors. As I said in another thread (again, not specific to this conversation):
-------
Expected return is one of the more misleading terms in investing. Large numbers of investors appear to believe that expected returns are returns they should expect to receive.

An investment with a 99% chance of returning nothing and a 1% chance of returning $1 million has an expected return of $10,000, but $10,000 is not at all a likely return.

Another problem is that we have no reasonable way of estimating the odds for most investments, which makes the entire notion of expected return even less useful.
------
Strange to read this, as I have been the one who has had to point out what expected returns are about a dozen times (a weighted average of potential outcomes), as posters want to use "expected" and "worst case" in the same sentence, as if they have no association with one another.

And you can draw up any sort of lottery-ticket like probabilities you want, but we are talking about investing in capital markets where there is a positive and probable expected return on investment--if not, prices would fall to the point where there is. That's how the system works. Because you cannot precisely estimate the probabilities to the third decimal point means next to nothing. The quest for certainty will land you in "0% return land", or an immediate annuity, which is close.

We have a very good idea of what expected returns are and the distribution around those returns: small chance (say 5%) of a really disappointing return or an absolutely stratospheric return, slightly larger chance (say 15%) of a small amount of underperformance our outperformance, and a majority probability (say 60%) of a few percent (5% to 7%) of outperformance, as in the case of SV>TSM or stocks>bonds respectively.

Even if you knew what the distributions of future outcomes looked like exactly, what would you do with the information? Markets would know too, and prices would reflect it. And you still couldn't predict the future as in when certain outcomes would materialize. As an investor, all you do is plan the best you can using all available information, and make sure you have the flexibility to make adjustments should actual not = expected.

You can try to make things more complicated or abstract than this, but it's not necessary and in most cases ill-advised.

Eric

YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: "value factor" doesn't seem reliable to me

Post by YDNAL » Thu Apr 18, 2013 12:29 pm

richard wrote:I continue to find the term "expected" misleading for many investors. As I said in another thread (again, not specific to this conversation):

Expected return is one of the more misleading terms in investing. Large numbers of investors appear to believe that expected returns are returns they should expect to receive....
Richard,

There should be nothing in the term/word "expected" to indicate to anyone that investing in the Markets (Equity & Debt) comes with any guarantee of what one will receive.

That said, and since we invest for expected growth and not expected loss, I do believe that there is a tendency to think of positive potential outcomes - like 3% to 7% expected return. We don't typically see -10% to +10% expected return. :D
Landy | Be yourself, everyone else is already taken -- Oscar Wilde

User avatar
LadyGeek
Site Admin
Posts: 47401
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: "value factor" doesn't seem reliable to me

Post by LadyGeek » Thu Apr 18, 2013 3:01 pm

This thread is now in the Investing - Theory, News & General forum (theory).
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.

trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

Re: "value factor" doesn't seem reliable to me

Post by trademil » Thu Apr 18, 2013 6:06 pm

YDNAL wrote:
trademil wrote:
YDNAL wrote: For instance, staying with Vanguard since you mentioned Vanguard Small-Cap Value ETF (VBR) in your OP:
  • 1. VBR has a Price/book ratio of 1.5x
    2. VBK (Growth) has a Price/book ratio of 3.2x - more than twice VBR's Index.
You have a point here, but, lets add:
3.Vanguard Small-Cap ETF (VB) has a Price/book ratio of 2.1x:

https://personal.vanguard.com/us/funds/ ... =INT#tab=2

So it seems that most of the gap is between the growth and VB, and only the minority is between VB and value. The gap between 2.1 and 1.5 doesn't look that decisive to me.
Given that the financial sector is 40% of VBR, theoretically, it might just be the case that if those financial institutions will mark their securities to market and correct their books accordingly their book value will drop by 70%, which will be a drop of 28% (70*0.4) for the whole fund, which will put it's Price/book of the fund at 2.1, just like VB.
There is no such "gap." What happens is that 1,440 Stocks in VB (CRSP US Small Cap Index) - at market weights - do not comparatively match the sum of the parts in 1,004 Stocks currently in VBR (CRSP US Small Cap Value Index) and 949 Stocks in VBK (CRSP US Small Cap Growth Index).
I stand corrected. But I still don't see how does it conflict with my argument.

stlutz
Posts: 4655
Joined: Fri Jan 02, 2009 1:08 am

Re: "value factor" doesn't seem reliable to me

Post by stlutz » Thu Apr 18, 2013 8:55 pm

Probably worth noting that Vanguard's index funds are not split between value and growth based only on book value. They use multi-characteristic sorts.

Most academic research relies on price/book. Most practitioners or people who backtest for a living tend not to use P/B because there are other metrics that backtest so much better.

YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: "value factor" doesn't seem reliable to me

Post by YDNAL » Fri Apr 19, 2013 6:27 am

trademil wrote:
YDNAL wrote:There is no such "gap." What happens is that 1,440 Stocks in VB (CRSP US Small Cap Index) - at market weights - do not comparatively match the sum of the parts in 1,004 Stocks currently in VBR (CRSP US Small Cap Value Index) and 949 Stocks in VBK (CRSP US Small Cap Growth Index).
I stand corrected. But I still don't see how does it conflict with my argument.
What is your argument?... please specify.

VB (CRSP US Small Cap Index) should be expected to have a HIGHER price/book ratio than VBR (CRSP US Small Cap Value Index).
  • But, you seem to think that its P/B should be exactly (or near) 1/2 way between VBR/VBK's individual P/B ratios.
  • That is not necessarily the result when a representative sample of the Small Cap universe (at market weights) is put together.
Landy | Be yourself, everyone else is already taken -- Oscar Wilde

trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

Re: "value factor" doesn't seem reliable to me

Post by trademil » Fri Apr 19, 2013 7:35 am

My argument is that the difference between 2.1 to 1.5 is not significant enough and might be explained only (or mostly) by obsolete numbers in the books of the companies (mainly in the books of financial companies).

larryswedroe
Posts: 15691
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: "value factor" doesn't seem reliable to me

Post by larryswedroe » Fri Apr 19, 2013 8:38 am

trademil
First, the evidence says that the difference is significant, it matters in returns
Second, the issues with BtM is why many believe a multi screen strategy which seeks to make sure a stock is value based on several metrics does better at finding value stocks. I believe that is the case
But BtM does a very good job
Cash flow measures are one preferred method and seem to capture some of the quality factor as well (along with MOM) at least in small stocks
Larry

YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: "value factor" doesn't seem reliable to me

Post by YDNAL » Fri Apr 19, 2013 9:30 am

trademil wrote:My argument is that the difference between 2.1 to 1.5 is not significant enough and might be explained only (or mostly) by obsolete numbers in the books of the companies (mainly in the books of financial companies).
trademil, as Larry just said, 28.6% LOWER market valuation (price) to book value between VBR and VB is significant. Additionally, 53.1% LOWER market valuation (price) to book value between VBR and VBK (CRSP Small Growth) is extremely significant.

Now, to stop beating a dead horse, I ask you..... have you taken the time to evaluate individual holdings in VBR vs. VBK vs. VB to understand what you are looking at (and make decisions) ? I'll help by giving you top 50 holdings - sorted alphabetically - in VB vs. VBR where I see lots of similarities.

Code: Select all

VB (Small Cap Index)	Market Value		VBR (Small Cap Value Index)	Market Value
American Capital Ltd	$64,731,576 		Allied World Assuran	$26,507,690 
Apartment Investment	$65,443,013 		American Capital Ltd	$36,646,632 
Ariad Pharmaceutical	$52,967,143 		AOL Inc.	$26,170,147 
Atmos Energy Corp.	$52,625,950 		Apartment Investment	$24,075,363 
BRE Properties Inc.	$64,868,331 		Assured Guaranty Ltd	$23,375,735 
Cadence Design Syste	$61,709,289 		Atmos Energy Corp.	$29,786,888 
Carlisle Cos. Inc.	$61,082,665 		BioMed Realty Trust 	$28,033,197 
CBL & Associates Pro	$53,037,429 		BRE Properties Inc.	$36,720,812 
Cheniere Energy Inc.	$56,815,153 		Broadridge Financial	$26,878,280 
Chico's FAS Inc.	$50,922,888 		Cabot Corp.	$23,683,923 
Clean Harbors Inc.	$51,212,440 		Carlisle Cos. Inc.	$34,593,599 
Concur Technologies 	$52,474,991 		CBL & Associates Pro	$30,051,537 
Corrections Corp. of	$58,968,875 		Cinemark Holdings In	$25,268,460 
Dean Foods Co.	$50,699,899 		Community Health Sys	$26,298,746 
Douglas Emmett Inc.	$51,356,881 		Compass Minerals Int	$23,259,987 
Extra Space Storage 	$62,912,961 		Cytec Industries Inc	$28,323,270 
Genesee & Wyoming In	$55,950,525 		Douglas Emmett Inc.	$29,090,959 
GNC Holdings Inc. Cl	$54,915,395 		East West Bancorp In	$28,689,537 
Graco Inc.	$51,773,298 		First Niagara Financ	$26,302,303 
Great Plains Energy 	$51,782,904 		Flowers Foods Inc.	$27,321,051 
Hanesbrands Inc.	$58,040,507 		Great Plains Energy 	$29,310,864 
Home Properties Inc.	$50,839,907 		Hancock Holding Co.	$24,041,622 
Jack Henry & Associa	$53,394,189 		Highwoods Properties	$23,932,906 
JDS Uniphase Corp.	$52,183,011 		Home Properties Inc.	$28,772,476 
Kennametal Inc.	$53,247,040 		Ingram Micro Inc.	$23,884,204 
Kilroy Realty Corp.	$58,196,745 		Kennametal Inc.	$30,139,720 
Kirby Corp.	$51,717,141 		Kilroy Realty Corp.	$32,938,777 
Lincoln Electric Hol	$64,222,842 		MFA Financial Inc.	$27,268,553 
MEDNAX Inc.	$65,690,915 		Mid-America Apartmen	$25,028,012 
National Retail Prop	$56,038,382 		National Retail Prop	$31,748,683 
NCR Corp.	$67,297,394 		NCR Corp.	$24,783,300 
Packaging Corp. of A	$62,737,376 		Old Republic Interna	$24,688,713 
Penn National Gaming	$53,132,747 		Omega Healthcare Inv	$24,376,059 
PerkinElmer Inc.	$60,161,107 		Oshkosh Corp.	$23,000,335 
Questar Corp.	$57,690,091 		Packaging Corp. of A	$35,512,042 
RPM International In	$64,391,824 		PerkinElmer Inc.	$34,054,417 
Shaw Group Inc.	$51,120,450 		ProAssurance Corp.	$23,117,757 
Signature Bank/New Y	$54,714,070 		Questar Corp.	$32,656,186 
Sirona Dental System	$59,046,134 		Regal-Beloit Corp.	$28,027,469 
SolarWinds Inc.	$51,723,777 		RPM International In	$36,452,378 
Tanger Factory Outle	$53,358,361 		Ryder System Inc.	$24,004,247 
Tenet Healthcare Cor	$56,216,935 		Service Corp. Intern	$27,912,882 
Teradyne Inc.	$52,589,701 		Starwood Property Tr	$28,698,622 
Terex Corp.	$51,619,517 		Teleflex Inc.	$27,403,720 
Triumph Group Inc.	$54,179,671 		Terex Corp.	$29,214,273 
tw telecom inc Class	$63,960,722 		Trinity Industries I	$26,614,224 
Two Harbors Investme	$51,398,790 		Two Harbors Investme	$29,114,140 
United Rentals Inc.	$59,602,932 		Visteon Corp.	$24,001,029 
Valmont Industries I	$54,273,163 		WESCO International 	$27,728,632 
Westar Energy Inc.	$60,071,777 		Westar Energy Inc.	$34,003,937
Landy | Be yourself, everyone else is already taken -- Oscar Wilde

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: "value factor" doesn't seem reliable to me

Post by Rodc » Fri Apr 19, 2013 10:33 am

trademil wrote:My argument is that the difference between 2.1 to 1.5 is not significant enough and might be explained only (or mostly) by obsolete numbers in the books of the companies (mainly in the books of financial companies).
In addition to what others have said, these values move more or less continuously, just as P/E or any other metric moves with time. Unless you are specifically market timing I suggest a broader point of view than a simple snap shot of one particular time.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

Re: "value factor" doesn't seem reliable to me

Post by trademil » Fri Apr 19, 2013 5:47 pm

YDNAL wrote: trademil, as Larry just said, 28.6% LOWER market valuation (price) to book value between VBR and VB is significant. Additionally, 53.1% LOWER market valuation (price) to book value between VBR and VBK (CRSP Small Growth) is extremely significant.

Now, to stop beating a dead horse, I ask you..... have you taken the time to evaluate individual holdings in VBR vs. VBK vs. VB to understand what you are looking at (and make decisions) ? I'll help by giving you top 50 holdings - sorted alphabetically - in VB vs. VBR where I see lots of similarities.
I don't see how VBK is relevant here, unless I want to short it, which I don't.
If your point is that VB and VBR actually have a very large overlap it just emphasizes the point that there is not much of an advantage to one over the other. Or was your point that Vanguards fund is not the right fund to use if you want to capture the "small value premium"?
larryswedroe wrote:trademil
First, the evidence says that the difference is significant, it matters in returns

I wonder if the historical data used to derive the notion of value premium also included this level of disparity between the P/B or the "small value" and the P/B of "small", or was it greater back then.


Are there any reasonable strategies to increase my risk (and potential return) without tilting for the small and/or value premiums (given my inability to reduce my bonds holdings)?
For example, how can I construct a "1.2 beta portfolio" with the equity part of my AA?

larryswedroe
Posts: 15691
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: "value factor" doesn't seem reliable to me

Post by larryswedroe » Sat Apr 20, 2013 8:17 am

trademil
What the data shows is what you would expect
When the gap between btm of value vs growth is wider, the value premium is wider and vice versa
Just as it is with stocks --the higher the P/E the lower the future returns (on average).
But you cannot use that information to time value vs growth because there is always a value premium
Larry

YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: "value factor" doesn't seem reliable to me

Post by YDNAL » Sat Apr 20, 2013 8:53 am

trademil wrote:I don't see how VBK is relevant here, unless I want to short it, which I don't.
trademil, one final time, VBK (Small Growth) is relevant because it holds what is included in VB (Small) - the source of your comparison to VBR (Small Value). You either choose to include constituents in the Growth Index in your Small Caps or you don't. By excluding them, you significantly DEcrease the price to book ratio (inverse of book to market) of your Small Caps.

Otherwise, I'm finished here.
Landy | Be yourself, everyone else is already taken -- Oscar Wilde

afan
Posts: 3759
Joined: Sun Jul 25, 2010 4:01 pm

Re: "value factor" doesn't seem reliable to me

Post by afan » Sat Apr 20, 2013 9:12 am

We are mixing four questions

1 How reliable is the value premium? Studies have suggested that it is robust across markets, countries, and time periods. But only if you hold to the strategy for many years.

2 Which value metric is best? Studies have shown that the multiple metrics tend to produce similar portfolios and returns. So academics focus on those that are the least ambiguous and most reliably recovered from existing databases.

3 Is pursuing the more valuey half of the market a strong enough tilt? Many studies of the phenomenon have seen the strongest effects in more extreme portfolios, with associated greater risk.

4 Are the anticipated higher returns worth the anticipated higher risk? Maybe. Depends on whether the effect will persist (no one knows), how long you can hold on during periods when it does not work, and what you will do if new studies suggest that it does not work anymore. The alternative is to settle for the total market, ignore these anomalies, and invest your efforts to earning or saving more. The classic approach to increasing rusk in the cost Free academic world is to use leverage. Buy on margin, use derivatives, or leverage up the rest of your life to free up more cash to invest. One could consider a more extreme small value slice than Vanguard offers, with higher fees and more turnover.
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama

trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

Re: "value factor" doesn't seem reliable to me

Post by trademil » Sat Apr 20, 2013 9:13 am

YDNAL,
Thank you.

trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

Re: "value factor" doesn't seem reliable to me

Post by trademil » Sat Apr 20, 2013 9:14 am

larryswedroe wrote: there is always a value premium
But it might be negative some years\decades?

stlutz
Posts: 4655
Joined: Fri Jan 02, 2009 1:08 am

Re: "value factor" doesn't seem reliable to me

Post by stlutz » Sat Apr 20, 2013 9:42 am

But it might be negative some years\decades?
The "premium" is in expected returns, not actual returns. The basic theory is that:

--one should expect riskier investments to have higher returns than less risky ones.
--value stocks have significantly outperformed the market across time and markets
--that value outpeformance was expected beforehand
--because the value stocks had higher expected returns, they are riskier.
--therefore value stocks have higher expected returns going forward.

Larry is not saying that he knows that value will have higher actual returns. That is unknown.

YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: "value factor" doesn't seem reliable to me

Post by YDNAL » Sat Apr 20, 2013 9:45 am

trademil wrote:
larryswedroe wrote: there is always a value premium
But it might be negative some years\decades?
Yes.

Code: Select all

Vanguard Small Value / Growth
1 Year           19.85% 16.04%
3 Year           13.41% 15.90%
5 Year            8.77% 10.35%
10 Year          11.91% 12.73%
Since Inception   7.79%  7.93% (05/21/1998 for both, 15-year period)
Landy | Be yourself, everyone else is already taken -- Oscar Wilde

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: "value factor" doesn't seem reliable to me

Post by Rodc » Sat Apr 20, 2013 9:58 am

A classic approach if you want more risk in the hopes of better returns is to use leverage.

If you don't find available mutual funds to your liking, that would be your best approach.

Best of luck.

(not that I would suggest it)
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

larryswedroe
Posts: 15691
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: "value factor" doesn't seem reliable to me

Post by larryswedroe » Sat Apr 20, 2013 12:39 pm

trademil
Risk premiums are always EXPECTED return premiums, no matter how long the period it cannot be guaranteed or there would be no risk.
Larry

User avatar
nedsaid
Posts: 10210
Joined: Fri Nov 23, 2012 12:33 pm

Re: "value factor" doesn't seem reliable to me

Post by nedsaid » Sun Apr 21, 2013 11:59 am

I have two reactions to this discussion.

First, I don't believe that Value stocks perform better because they are riskier than stocks as a whole. I truly believe that there is a Value premium independent of the risk factors. I would argue that Value stocks have less risk because there is less expectation built into them. The popular growth stocks that everyone loves are more vulnerable to earnings disappointments. Apple is Exhibit A of this.

Second, valuation metrics are imperfect. For example, assets on the books of companies are valued by accountants at historical costs less any applicable depreciation. So an asset might be valued at very low costs (if it was purchased many years ago) and depreciated even though that asset has gained in market value. So metrics involving book value are imperfect but seem to be a really good valuation metric. Cash flow is something more accurately measured by accountants and is another good metric perhaps better than book value. Earnings are a good metric but are affected by charges that do not affect actual cash (such as depreciation and amortization). We all know that earnings can be manipulated to some extent as GE used to do with timing of asset sales. It would seem to me that using a combination of several value metrics would be the best approach.

Another point that I would raise is that most market premiums (equity over fixed income, value over growth, small over large) will disappear from time to time only to reassert themselves later. I would argue that there are good behavioral reasons for these premiums that are rooted in human nature. All the math can measure the effect of these behaviors in the past but cannot predict the behaviors themselves. All we can say is that eventually things return to the norms.
A fool and his money are good for business.

User avatar
White Coat Investor
Posts: 13365
Joined: Fri Mar 02, 2007 9:11 pm
Location: Greatest Snow On Earth

Re: "value factor" doesn't seem reliable to me

Post by White Coat Investor » Sun Apr 21, 2013 3:43 pm

The fun thing about value tilting is you only have to tilt as much as you believe. If you think the value effect is real and likely to persist, make a heavy tilt. If you think it's hogwash, don't tilt at all. If you're like the rest of us, do a milder tilt.

Since you're in your 20s and presumably have a relatively tiny portfolio, keep in mind that your income and savings rate have far more to do with the stash you end up with than your value tilt.
1) Invest you must 2) Time is your friend 3) Impulse is your enemy | 4) Basic arithmetic works 5) Stick to simplicity 6) Stay the course

User avatar
nedsaid
Posts: 10210
Joined: Fri Nov 23, 2012 12:33 pm

Re: "value factor" doesn't seem reliable to me

Post by nedsaid » Sun Apr 21, 2013 4:14 pm

I do tilt towards value and small. I believe in these premiums. It should also be said that a pure indexing approach works pretty darned well. So tilt if you believe in the premiums but don't overdo it.
A fool and his money are good for business.

trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

Re: "value factor" doesn't seem reliable to me

Post by trademil » Mon Apr 22, 2013 2:16 am

Does the fact Vanguard Total International Stock has a Price/book ratio of 1.6, lower then Vanguard Total Stock Market (2.3) or even Vanguard Value (1.7) means international stocks have more value factor in them and are expected to have higher returns?

https://personal.vanguard.com/us/funds/ ... =INT#tab=2

https://personal.vanguard.com/us/funds/ ... =INT#tab=2

https://personal.vanguard.com/us/funds/ ... =INT#tab=2

larryswedroe
Posts: 15691
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: "value factor" doesn't seem reliable to me

Post by larryswedroe » Mon Apr 22, 2013 7:37 am

absolutely, it's what I have been saying and writing about on my blog
Check the PE ratios as well and see the same thing
Of course the lower valuations reflect the higher level of perceived risk---not some free lunch
Larry

User avatar
bottomfisher
Posts: 399
Joined: Fri Jan 04, 2013 9:03 am

Re: "value factor" doesn't seem reliable to me

Post by bottomfisher » Mon Apr 22, 2013 8:39 am

Does the fact Vanguard Total International Stock has a Price/book ratio of 1.6, lower then Vanguard Total Stock Market (2.3) or even Vanguard Value (1.7) means international stocks have more value factor in them and are expected to have higher returns?
A study by Vanguard suggests the lower P/E may explain about 40% of future returns. You have to pay full price for the rest of that lunch.

https://personal.vanguard.com/pdf/s338.pdf

larryswedroe
Posts: 15691
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: "value factor" doesn't seem reliable to me

Post by larryswedroe » Mon Apr 22, 2013 9:18 am

bottom fisher
Those higher returns might just be compensation for higher risks--so even if you get them at a lower price doesn't necessarily tell you whether you got good "value"
It's like saying HY is good and investment grade is bad because you paid full price for it--no one would say that but you just did about stocks

Now if you think it's all a behavioral story then yes it's a free lunch

Best wishes
Larry

User avatar
bottomfisher
Posts: 399
Joined: Fri Jan 04, 2013 9:03 am

Re: "value factor" doesn't seem reliable to me

Post by bottomfisher » Mon Apr 22, 2013 10:00 am

bottom fisher
Those higher returns might just be compensation for higher risks--so even if you get them at a lower price doesn't necessarily tell you whether you got good "value"
It's like saying HY is good and investment grade is bad because you paid full price for it--no one would say that but you just did about stocks

Now if you think it's all a behavioral story then yes it's a free lunch
I understand your point. But I wasn't necessarily implying that. My intent was to provide research supporting the fact that P/E may explain a certain amount of equity returns. The miscommunication is likely in the semantics; but lets let it ride in effort not to hijack an otherwise informative thread.

larryswedroe
Posts: 15691
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: "value factor" doesn't seem reliable to me

Post by larryswedroe » Mon Apr 22, 2013 10:10 am

bottomfisher
There is no doubt in the literature that valuations provide important information about returns.
Anyone taking the time for example to read Damodaran's papers on the subject would agree
It's just that there is still a lot of random noise in the data because there are risks that can show up that are by their nature unpredictable
Best wishes
Larry

User avatar
grayfox
Posts: 4932
Joined: Sat Sep 15, 2007 4:30 am

Re: "value factor" doesn't seem reliable to me

Post by grayfox » Mon Apr 22, 2013 10:17 am

trademil wrote:"value factor" doesn't seem reliable to me.
Thoughts?
I think you got it backwards. The "value" factor, HmL has been very reliable. It has almost always been positive, except for the late 1990s.

It is the "small' factor, SmB, that has not been so reliable. There have been long periods when it's been negative.

leonard
Posts: 5993
Joined: Wed Feb 21, 2007 11:56 am

Re: "value factor" doesn't seem reliable to me

Post by leonard » Mon Apr 22, 2013 11:16 am

If the "value factor" was completely reliable, then you wouldn't receive as much risk premium for it. The very fact it can underperform or is "unreliable" is part of the reason you are compensated for taking that risk.
Leonard | | Market Timing: Do you seriously think you can predict the future? What else do the voices tell you? | | If employees weren't taking jobs with bad 401k's, bad 401k's wouldn't exist.

trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

Re: "value factor" doesn't seem reliable to me

Post by trademil » Tue Apr 23, 2013 2:15 am

leonard wrote:If the "value factor" was completely reliable, then you wouldn't receive as much risk premium for it. The very fact it can underperform or is "unreliable" is part of the reason you are compensated for taking that risk.
What doesn't seem reliable to me is the methodology of basing a decision on numbers supplied by the companies (and not by the market).

trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

Re: "value factor" doesn't seem reliable to me

Post by trademil » Tue Apr 23, 2013 2:47 am

grayfox wrote:
I think you got it backwards. The "value" factor, HmL has been very reliable. It has almost always been positive, except for the late 1990s.

"Value" also lagged for the past few years:


Image

larryswedroe
Posts: 15691
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: "value factor" doesn't seem reliable to me

Post by larryswedroe » Tue Apr 23, 2013 7:31 am

What is amusing to me is that people question value premium based on persistence. But same people don't question stock premium --yet value premium has been more persistent than stock premium. Last time I looked this was true for all periods looked at and the longer the period the more true it was
But haven't looked in a while
Larry

trademil
Posts: 204
Joined: Tue May 22, 2012 3:14 am

Re: "value factor" doesn't seem reliable to me

Post by trademil » Tue Apr 23, 2013 8:38 am

larryswedroe wrote:What is amusing to me is that people question value premium based on persistence. But same people don't question stock premium --yet value premium has been more persistent than stock premium. Last time I looked this was true for all periods looked at and the longer the period the more true it was
But haven't looked in a while
Larry
I do not question value premium based on persistence, but on 2 other bases:

1.
The equity premium is derived from the well-documented and well-explained fact that equities bear more risk.
The value premium is an empirical finding (=might be random), and the conclusion that "value stocks must bear more risk" followed, but (to my knowledge) not proven. The direction of the reasoning is backwards.

2.
The definition of "stock" vs "bond" is very clear and is not dependent on human judgement or on data that might be flawed or distorted (intentionally or unintentionally).
The definition of "value" vs "growth" is arbitrary, require judgement, and based on reports by the companies.

larryswedroe
Posts: 15691
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: "value factor" doesn't seem reliable to me

Post by larryswedroe » Tue Apr 23, 2013 9:20 am

trademil
That's fair.
So we have two theories. First that value stocks are more risky. There are plenty of papers showing very clear and logical explanations for their greater risk. Also risks that show up in bad times theory states should have high risk premiums. That seems true of value stocks which tend to do worst in deflationary recessions, the worst kinds

The other theory is behavioral, that investors simply overprice growth stocks for variety of reasons. If that is true then there are two theories. First, by discovering this it will go away or, Second, that human beings don't change.

Given the huge body of evidence on the risk story I think it's hard to make the case it's all behavioral, which means it it as likely to persist and the stock premium. And if it is at least partly behavioral either
A) it will persist, and then no problem
B) it will go away and thus you should buy the value stocks TODAY before their prices rise to eliminate the free lunch

Hope that helps
Larry

Post Reply