Investing in the new and improved value factor(s)

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
strakert
Posts: 85
Joined: Sat Oct 02, 2021 12:06 pm

Investing in the new and improved value factor(s)

Post by strakert »

The value factor is dead.
The original HML value factor is redundant in the new Fama-French 5 factor model (subsumed by investment and profitability factors) and absent in the Hou-Xue-Zhang 4 factor and 5 factor models.

Several recent industry papers exploring the 13+ year drawdown of value similarly conclude that the original HML definition is obsolete and needs further adjustments to be useful.
  1. AQR: Fact, Fiction, and Value Investing (2017)
  2. AQR: Is (Systematic) Value Investing Dead? (Apr 2020)
  3. Research Affiliates: Reports of Value’s Death May Be Greatly Exaggerated (Oct 2020)
  4. Robeco: Resurrecting the Value Premium (Oct 2020)
Long live the value factor!
  1. All of the papers above recommend using multiple metrics, typically a (different) combination of Book/Price, Earnings/Price, Future Earnings/Price, Sales/Price, Buyback-adjusted Dividend Yield/Price etc. to construct a new value factor.
  2. All of them recommend additional (but different) improvements, including adjusting the accounting metrics to account for intangibles, using more timely metrics, sector neutrality, excluding financial sector, combining with other factors like Momentum and Profitability in either trading strategies or in screens, minimizing trading costs etc.
  3. While the HML value factor has only been present in small-cap stocks post-discovery (Fama-French (2020)), some of them re-expand the universe to large and mid caps given the redefinition, or change from cap-weighting to equal weighting.
  4. Finally, all of the research and backtests are for long-short factors, while most value indexes and ETF are long only.
Existing indexes and funds are outdated!
But existing Small Cap Value Indices don't make most of the adjustments recommended above.
  1. The S&P 600 Small Cap Value Index only uses an average of Book/Price, Earnings/Price and Sales/Price with no further screens or adjustments. State Street's SLYV, Vanguard's VIOV and iShares IJS passively follow this index.
  2. The S&P 600 Small Cap Pure Value Index is a more "concentrated" version one above. It only includes ~33% of the 600 small-cap stocks ranked by their value score instead of ~50% in the one above. The stocks are also weighted by their value score, unlike their cap-weight in the one above. Invesco's RZV follows this index.
  3. Similarly, the CRSP US Small Cap Value Index uses Book/Price, Future Earnings/Price, Earnings/Price, Dividend/Price and Sales/Price with no further screens or adjustments. Its investment universe seems to include mid-cap value as well (it has ~1000 stocks). Vanguard's VBR and iShares ISCV passively follow this index.
  4. Finally, Avantis AVUV is actively managed. According to it's prospectus, "To identify small capitalization, high profitability, or value companies, the portfolio managers may use reported and estimated company financials and market data including, but not limited to, shares outstanding, book value and its components, cash flows, revenue, expenses, accruals and income ... The portfolio managers may also consider other factors when selecting a security including, industry classification, the past performance of the security relative to other securities, its liquidity, its float, and tax, governance or cost considerations, among others.". Emphasis mine - plenty of wiggle room and hard to predict or backtest performance.
My questions are:
  1. Are there any indices and passive ETFs that track the "new and improved" value factors?
  2. If not, is there any further research on the performance of existing indices w.r.t the new value factors? I.e. how should I adjust my performance expectations for investing in them w.r.t the results in the research papers?
  3. Are there any actively managed funds that publish their methodology and are also affordable for individual investors (for minimums and ERs)?
  4. Which value factor ETFs do you actually invest in given the considerations above?
Last edited by strakert on Sun Oct 17, 2021 5:41 pm, edited 1 time in total.
User avatar
Vulcan
Posts: 2996
Joined: Sat Apr 05, 2014 11:43 pm

Re: Investing in the new and improved value factor(s)

Post by Vulcan »

See signature
If you torture the data long enough, it will confess to anything. ~Ronald Coase
FiveFactor
Posts: 100
Joined: Thu Apr 08, 2021 8:47 pm

Re: Investing in the new and improved value factor(s)

Post by FiveFactor »

Hi. Signature.
Small/Value/Profitability: | 30% AVUV | 30% AVDV | 30% AVES | Momentum: | 5% QMOM | 5% IMOM | Volatility: | 0.1% PUTW | Term: | 0.1% BND
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Investing in the new and improved value factor(s)

Post by 000 »

All of these tweaks to the 'factors' reveal them to be nothing more than active management IMHO.

If I were going to go down the factor road it seems a proudly active fund might be the best way to go. Or just buy individual stocks, which I do.
User avatar
Beensabu
Posts: 5657
Joined: Sun Aug 14, 2016 3:22 pm

Re: Investing in the new and improved value factor(s)

Post by Beensabu »

You're making it too complicated.

The value factor is not dead. It just might not necessarily be able to be captured via a "value" fund. You might have to work a little harder than that for it. But not that hard.

The size factor definitely isn't dead.
"The only thing that makes life possible is permanent, intolerable uncertainty; not knowing what comes next." ~Ursula LeGuin
Topic Author
strakert
Posts: 85
Joined: Sat Oct 02, 2021 12:06 pm

Re: Investing in the new and improved value factor(s)

Post by strakert »

Beensabu wrote: Sun Oct 17, 2021 6:41 pm The value factor is not dead.
The original Fama-French HML definition is. See the references above from all the major firms selling quantitative value factor funds. The redefinition is fine, I'm asking how to benefit from it.
Beensabu wrote: Sun Oct 17, 2021 6:41 pm
It just might not necessarily be able to be captured via a "value" fund. You might have to work a little harder than that for it. But not that hard.
How can I do that with investable indices / ETFs?
Beensabu wrote: Sun Oct 17, 2021 6:41 pm The size factor definitely isn't dead.
Apparently it was never alive, and has been given a similar "Ship of Theseus" treatment:
AQR: Size Matters, If You Control Your Junk
AQR: Fact, Fiction, and the Size Effect
Robeco: Settling the Size Matter

Let's keep this discussion about the value factor though.
User avatar
nisiprius
Advisory Board
Posts: 52211
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: Investing in the new and improved value factor(s)

Post by nisiprius »

Beensabu wrote: Sun Oct 17, 2021 6:41 pm...The size factor definitely isn't dead...
In 1981, in "The Relationship Between Return and Market Value of Common Stocks," Rolf W. Banz wrote:
It is found that smaller firms have had higher risk adjusted returns, on average, than larger firms. This ‘size effect’ has been in existence for at least forty years and is evidence that the capital asset pricing model is misspecified.
The DFA Micro Cap Portfolio had inception almost immediately after the paper was published.

During the 40 years of data since inception, compared to an S&P 500 index fund, the DFA Micro Cap Portfolio has had
  • A virtually identical average return, 11.90% versus 11.93%
  • A considerably higher standard deviation, 19.76% versus 14.99%
  • And, thus, a considerably lower risk-adjusted return, a Sharpe ratio of 0.49 versus 0.59
Image

The important hope or expectation was higher risk-adjusted return. It hasn't been fulfilled.

Less important, but often talked about, was an expectation of a "premium," that is, simply higher return, even if nothing more than a reward for taking more risk. But that hasn't been fulfilled, either.

Banz said that the size effect had been in existence for forty years. Well, it now has been out of existence for forty years.

If that isn't "dead," it is hard to imagine what "dead" would look like.
Last edited by nisiprius on Mon Oct 18, 2021 4:45 am, edited 1 time in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Investing in the new and improved value factor(s)

Post by nedsaid »

I keep reading that the Value funds that I picked were the "wrong" ones, particularly Vanguard Small Cap Value Index ETF (VBR), and the "wrong" Value funds have performed pretty well, thank you. I also own the iShares S&P 600 Small Cap Value ETF (IJS), and it has also done well relative to other funds in its category. What we are discussing here is tweaking and combining factors and finding the managers that are up to the latest research. If you aren't satisfied with the "wrong" factor products, the Avantis ETF family is a good place to look.

Nisiprius commented on the Size premium, I saw a Morningstar article that said that the premium from Small Cap stocks seems to have gone away, but when you add the Quality factor or at least screen out the junkier stocks, the Size premium returns with a vengeance. They specifically mentioned the S&P 600 Small Cap Index as a good place to invest for the Size factor as S&P requires that stocks in their indexes actually have earnings. The research shows that Quality helps both Size and Value.

There is yet another tweak that I have become aware of that seems to work, screening for Cash Flow. I have looked at the Pacer Cash Cows ETF with clever ticker symbols. Their Small Cap Cash Cow ETF (CALF) has done very well, it represents deep Value and is almost in the Micro-Cap area of the market with a portfolio of just 100 stocks. CALF has even cheaper P/E, P/Cash Flow than the iShares or Avantis products. So another thing for us Small Cap Value fanatics to look at.
A fool and his money are good for business.
User avatar
Beensabu
Posts: 5657
Joined: Sun Aug 14, 2016 3:22 pm

Re: Investing in the new and improved value factor(s)

Post by Beensabu »

strakert wrote: Sun Oct 17, 2021 7:26 pm How can I do that with investable indices / ETFs?
I dunno. I just picked a sector I thought was undervalued and went with a tilt to that. It's been okay so far. Who knows how it'll go.

It's more not looking for something that has "value" in the description, but rather something that has holdings that you consider to fall under your chosen definition of "value".
nisiprius wrote: Sun Oct 17, 2021 8:02 pm If that isn't "dead," it is hard to imagine what "dead" would look like.
Small doesn't have to be small/micro value. Factors don't have to be combined. They can be independently alive.

Size doesn't have to mean small, even though small was the size factor that was identified previously.

I get that particular factors were identified as having provided a risk-adjusted premium in the past.

I don't know why people think that past premiums have to persist. That's not the point. The point is that it is possible to obtain such a premium, but the "factor" that provides it at any particular point in time will vary. Future research/literature will probably determine that we have gone through a period of a large growth premium. Nothing stays the same, and investing as if it does and thinking you can obtain a premium by using a fund oriented to a certain style based on a past premium-providing factor is silly. That's just marketing.
"The only thing that makes life possible is permanent, intolerable uncertainty; not knowing what comes next." ~Ursula LeGuin
Topic Author
strakert
Posts: 85
Joined: Sat Oct 02, 2021 12:06 pm

Re: Investing in the new and improved value factor(s)

Post by strakert »

Thanks nedsaid.
nedsaid wrote: Sun Oct 17, 2021 8:22 pm What we are discussing here is tweaking and combining factors and finding the managers that are up to the latest research.
From my reading the issue is more serious than that. It's not just that they've found a better version (there's plenty of that too), it's that original HML value factor has lost its explanatory power since discovery (just like the original SMB factor), completely in large-cap and largely in small-cap. This is blamed mostly on book value not accounting for intangibles (R&D, branding). These are much more important in tech-heavy 21st century firms than old and capital intensive industrials and utilities, leading to an unintentional and unproductive sector tilt.

It may be that the SCV indices are not affected as strongly by this and other issues, e.g., because they already use multiple metrics. If so, I'm looking for more information on *how much* of a non-issue this is. I.e., where do current indices fall on the spectrum of "our value factor definition is useless going forward" to "our hyper-tweaked definition captures 110% of value premium". Also, does anyone know if the construction of these indices has changed over time to accommodate newer research?
Topic Author
strakert
Posts: 85
Joined: Sat Oct 02, 2021 12:06 pm

Re: Investing in the new and improved value factor(s)

Post by strakert »

Beensabu wrote: Sun Oct 17, 2021 9:01 pm
strakert wrote: Sun Oct 17, 2021 7:26 pm How can I do that with investable indices / ETFs?
I dunno. I just picked a sector I thought was undervalued and went with a tilt to that. It's been okay so far. Who knows how it'll go.

It's more not looking for something that has "value" in the description, but rather something that has holdings that you consider to fall under your chosen definition of "value".
I have no chosen definition of "value" and no opinion on "valuations". I'm not look to do "Value Investing" in the style of Graham and Buffet. I'm looking to invest in the "value factor" as established by the APT based factor models.
Beensabu wrote: Sun Oct 17, 2021 9:01 pm
nisiprius wrote: Sun Oct 17, 2021 8:02 pm If that isn't "dead," it is hard to imagine what "dead" would look like.
Small doesn't have to be small/micro value. Factors don't have to be combined. They can be independently alive.
The links I shared and the nisiprius' chart are not about small value, they are about small as in the small-cap stocks and Small-Minus-Big factor.
Beensabu wrote: Sun Oct 17, 2021 9:01 pm I get that particular factors were identified as having provided a risk-adjusted premium in the past.

I don't know why people think that past premiums have to persist. That's not the point. The point is that it is possible to obtain such a premium, but the "factor" that provides it at any particular point in time will vary. Future research/literature will probably determine that we have gone through a period of a large growth premium. Nothing stays the same, and investing as if it does and thinking you can obtain a premium by using a fund oriented to a certain style based on a past premium-providing factor is silly. That's just marketing.
But this is not why factor investing / SCV believers do it. They're (mostly) not just chasing the newest factor premium. They do believe that Size/Value/Momentum/Quality etc. explain something systematic and fundamental about the nature of market risk and returns that will not be arbitraged away. That's the whole point of factor investing.

We can debate if that's a good point, and there are hundreds of threads doing just that. But for the purpose of this discussion let's agree with this belief and explore how one can invest in the "value factor" as defined by latest and greatest factor research.
Massdriver
Posts: 233
Joined: Tue Jan 02, 2018 11:05 am

Re: Investing in the new and improved value factor(s)

Post by Massdriver »

The S&P 600 small cap value index only selects companies with positive earnings which helps. You may have missed it because those screens are done in the 600 small cap index which the value index selects from.

Factor-heads seem to prefer funds from Avantis with broader exposure (e.g. AVUV, AVDV), Alpha Architect with concentrated exposure (e.g. QVAL, IVAL) and even some Vanguard factor funds. I prefer Avantis myself. The active label given to these funds doesn’t mean the same thing as an active fund that stock picks. They are using systematic criterion to determine which stocks are included in the fund just as the S&P 600 small cap value index is. They are just considered active because they can rebalance and adjust weights more often.
User avatar
typical.investor
Posts: 5263
Joined: Mon Jun 11, 2018 3:17 am

Re: Investing in the new and improved value factor(s)

Post by typical.investor »

strakert wrote: Sun Oct 17, 2021 4:46 pm
Several recent industry papers exploring the 13+ year drawdown of value similarly conclude that the original HML definition is obsolete and needs further adjustments to be useful.

[*]Research Affiliates: Reports of Value’s Death May Be Greatly Exaggerated (Oct 2020)
Sure, that paper constructs an iHML factor following the same rules we previously used to construct the regular B/P-based HML factor, with only one change. Instead of using the book-to-market ratio to define value, they use the ratio of our intangibles-adjusted book value to market value (iBook to Market) to define the value factor.

They note the the outperformance of the iHML factor, relative to B/P-based HML, is statistically significant at the 5% significance level and notably beats B/P-based HML by nearly a twofold margin after 1990.

They also look earnings-to-price (E/P) ratio and sales-to-price (S/P) as value measures and note that the "improved" iHML measure of value has also recently suffered a large drawdown, and post-2007 is still not as good as S/P or E/P.
strakert wrote: Sun Oct 17, 2021 4:46 pm My questions are:
  1. Are there any indices and passive ETFs that track the "new and improved" value factors?
  2. Which value factor ETFs do you actually invest in given the considerations above?
I don't know of anything targeting iHML, but I am satisfied by the methodology in the Russell Rafi Series available via Schwab and use FNDA, FNDC, FNDE etc.

They use Adjusted Sales instead of sales as shown effective in the paper above, but that's fine as it is just reducing exposure to highly leveraged firms. If those companies' use of leverage is profitable, I expect it'll show up in the Retained Operating Cash Flow, that the series also weights by, which is operating cash flow from operations less dividends and buybacks. Finally it uses Dividends plus Buybacks.

So it's weighted on both earnings and sales.

A similar value definition is also used in the PIMCO RAFI Dynamic Multi-Factor strategy takes time-varying exposures to four return factors; value, low volatility, quality, and momentum. I have some invested there. It'll weight more on factors that are cheap (and low volatility is know to have better returns when cheap) and also on factor momentum. Oh, factors have momentum which means poor value returns tend to persist and do good ones. I'm not completely sold that the time frame for these factor reversals will meet up with my investment horizon and the ER is a little higher so I have much less exposure here.

DFA is screen on profitability now too and have for some time, so they aren't HML only anymore.

Avantis is a current darling with their new funds. They state that:
To identify small capitalization, high profitability, or value companies, the portfolio managers may use reported and estimated company financials and market data including, but not limited to, shares outstanding, book value and its components, cash flows, revenue, expenses, accruals and income. Value companies may be defined as those with lower price relative to book value ratio or other fundamental value. High profitability companies may be defined as those with higher cash based operating profitability. The portfolio managers may also consider other factors when selecting a security including, industry classification, the past performance of the security relative to other securities, its liquidity, its float, and tax, governance or cost considerations, among others.
At the end of the day, different value funds will likely perform better or worse in different periods. Anything with some measure of profitability as opposed to simply book-to-market will do a decent job I think. It's hard to know in any given future holding period what the optimal methodology will be. That's my opinion anyway.
User avatar
nisiprius
Advisory Board
Posts: 52211
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: Investing in the new and improved value factor(s)

Post by nisiprius »

Beensabu wrote: Sun Oct 17, 2021 9:01 pmSmall doesn't have to be small/micro value. Factors don't have to be combined. They can be independently alive...
Sorry, typo in the caption. DFSCX, which is what is charted, is the DFA US Micro Cap Portfolio, not the Small-cap Value Portfolio. Now corrected.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
martincmartin
Posts: 900
Joined: Wed Jul 02, 2014 3:04 pm
Location: Boston, MA USA

Re: Investing in the new and improved value factor(s)

Post by martincmartin »

There are new ETFs from Dimensional Fund Advisors, e.g. DFAC and and DFAX. They implement the factor model, claim tax efficiency and don't contain REITs, which means you can keep all your REITs in sheltered. On the minus side, they have relatively high expense ratios, and average to low Morningstar ratings.
User avatar
Forester
Posts: 2400
Joined: Sat Jan 19, 2019 1:50 pm
Location: UK

Re: Investing in the new and improved value factor(s)

Post by Forester »

strakert wrote: Sun Oct 17, 2021 4:46 pm My questions are:
  1. Are there any indices and passive ETFs that track the "new and improved" value factors?
  2. If not, is there any further research on the performance of existing indices w.r.t the new value factors? I.e. how should I adjust my performance expectations for investing in them w.r.t the results in the research papers?
  3. Are there any actively managed funds that publish their methodology and are also affordable for individual investors (for minimums and ERs)?
  4. Which value factor ETFs do you actually invest in given the considerations above?
QVAL / IVAL are large cap value funds with transparent public rules, using only the enterprise multiple, around since 2014.
Amateur Self-Taught Senior Macro Strategist
secondopinion
Posts: 6011
Joined: Wed Dec 02, 2020 12:18 pm

Re: Investing in the new and improved value factor(s)

Post by secondopinion »

nisiprius wrote: Sun Oct 17, 2021 8:02 pm
Beensabu wrote: Sun Oct 17, 2021 6:41 pm...The size factor definitely isn't dead...
In 1981, in "The Relationship Between Return and Market Value of Common Stocks," Rolf W. Banz wrote:
It is found that smaller firms have had higher risk adjusted returns, on average, than larger firms. This ‘size effect’ has been in existence for at least forty years and is evidence that the capital asset pricing model is misspecified.
The DFA Micro Cap Portfolio had inception almost immediately after the paper was published.

During the 40 years of data since inception, compared to an S&P 500 index fund, the DFA Micro Cap Portfolio has had
  • A virtually identical average return, 11.90% versus 11.93%
  • A considerably higher standard deviation, 19.76% versus 14.99%
  • And, thus, a considerably lower risk-adjusted return, a Sharpe ratio of 0.49 versus 0.59
Image

The important hope or expectation was higher risk-adjusted return. It hasn't been fulfilled.

Less important, but often talked about, was an expectation of a "premium," that is, simply higher return, even if nothing more than a reward for taking more risk. But that hasn't been fulfilled, either.

Banz said that the size effect had been in existence for forty years. Well, it now has been out of existence for forty years.

If that isn't "dead," it is hard to imagine what "dead" would look like.
There is some ebb and flow. There would be some merit to hold a tiny amount of microcap.

It is not dead, but the factor is not for increased returns but variance of these returns.

Also, does DFA Micro Cap Portfolio cost more in ER?
Passive investing: not about making big bucks but making profits. Active investing: not about beating the market but meeting goals. Speculation: not about timing the market but taking profitable risks.
thenextguy
Posts: 717
Joined: Wed Mar 25, 2009 12:58 am

Re: Investing in the new and improved value factor(s)

Post by thenextguy »

secondopinion wrote: Mon Oct 18, 2021 11:40 am
nisiprius wrote: Sun Oct 17, 2021 8:02 pm
Beensabu wrote: Sun Oct 17, 2021 6:41 pm...The size factor definitely isn't dead...
In 1981, in "The Relationship Between Return and Market Value of Common Stocks," Rolf W. Banz wrote:
It is found that smaller firms have had higher risk adjusted returns, on average, than larger firms. This ‘size effect’ has been in existence for at least forty years and is evidence that the capital asset pricing model is misspecified.
The DFA Micro Cap Portfolio had inception almost immediately after the paper was published.

During the 40 years of data since inception, compared to an S&P 500 index fund, the DFA Micro Cap Portfolio has had
  • A virtually identical average return, 11.90% versus 11.93%
  • A considerably higher standard deviation, 19.76% versus 14.99%
  • And, thus, a considerably lower risk-adjusted return, a Sharpe ratio of 0.49 versus 0.59
Image

The important hope or expectation was higher risk-adjusted return. It hasn't been fulfilled.

Less important, but often talked about, was an expectation of a "premium," that is, simply higher return, even if nothing more than a reward for taking more risk. But that hasn't been fulfilled, either.

Banz said that the size effect had been in existence for forty years. Well, it now has been out of existence for forty years.

If that isn't "dead," it is hard to imagine what "dead" would look like.
There is some ebb and flow. There would be some merit to hold a tiny amount of microcap.

It is not dead, but the factor is not for increased returns but variance of these returns.

Also, does DFA Micro Cap Portfolio cost more in ER?
Excellent point. And yes. It would be interesting to see the results without fees. That seems like the proper way to do a comparison. It might reveal that a size premium exists, but capturing it is another story.
garlandwhizzer
Posts: 3565
Joined: Fri Aug 06, 2010 3:42 pm

Re: Investing in the new and improved value factor(s)

Post by garlandwhizzer »

Nisi, I think it you nailed it on the so called "small" factor. DFA has 2 Nobel Prize winners on its board of directors as well as other acclaimed academics on its payroll. One assumes that DFA is state of the art at creating a portfolio to harvest the small premium. One also assumes that micro-cap would maximize that premium since cap weight is precisely how you measure size and micro-cap is the smallest. While there is huge controversy and no certainty whatsoever about how to currently measure value, it's very easy to measure size. So the DFA Microcap Fund should massively outperform the S&P 500 which is dominated by large and mega-cap over a long time span long term if the small factor is actually harvestable after costs in the real world. Failing to outperform and doing so with much more volatility over a 40 year time span, which is long term in anybody's book, should put this factor in its well deserved grave, at least for those who are actually interested in risk adjusted returns rather than a nice narrative based on an unrealistic factor model.

Humans facing uncertainty are hard wired to create and embrace a seemingly convincing narrative that seeks to unravel that uncertainty. Uncertainty is precisely what our brains cannot tolerate. Being wrong is much easier to tolerate. When shown to be wrong, we simply create a new narrative to explain it as is currently going on with the value factor. We love our narratives and embrace them more than real facts.

Factor models have been in a constant state of evolution and change for 13 years. This has happened in response to research which demonstrates for example that the original definition of value PB doesn't work now. We are always in the process of finding a new approach for defining value. That new and improved approach may turn out likewise not work over the next 13 years. At that point yet another new approach will appear and the true believers will embrace the latest one.True believers in the factor model will not lose faith. Just like today's failure of PB. The failure of multi-factor funds to deliver and the failure of the small factor as nisi has pointed out--these hard facts do not dissuade those who embrace the model. They simply alter the model as time goes by.

Personally, I don't know whether cap weight or factor approaches will win out long term for risk adjusted returns. I am willing to admit my ignorance rather than pretending to know up front the future winner. I suspect both will do well if adhered to but that is as far my belief goes.

Garland Whizzer
Topic Author
strakert
Posts: 85
Joined: Sat Oct 02, 2021 12:06 pm

Re: Investing in the new and improved value factor(s)

Post by strakert »

strakert wrote: Sun Oct 17, 2021 9:41 pm
nedsaid wrote: Sun Oct 17, 2021 8:22 pm What we are discussing here is tweaking and combining factors and finding the managers that are up to the latest research.
Also, does anyone know if the construction of these indices has changed over time to accommodate newer research?
To answer my own question: Doesn't look like the indices themselves have changed. However VBR has changed indices not once but twice! It went from the S&P to MSCI to CRSP index. As mentioned above, these are fairly different.
Massdriver wrote: Sun Oct 17, 2021 11:31 pm The S&P 600 small cap value index only selects companies with positive earnings which helps. You may have missed it because those screens are done in the 600 small cap index which the value index selects from.
Thanks for the clarification.
Massdriver wrote: Sun Oct 17, 2021 11:31 pm Factor-heads seem to prefer funds from Avantis with broader exposure (e.g. AVUV, AVDV), Alpha Architect with concentrated exposure (e.g. QVAL, IVAL) and even some Vanguard factor funds. I prefer Avantis myself.
Forester wrote: Mon Oct 18, 2021 6:52 am QVAL / IVAL are large cap value funds with transparent public rules, using only the enterprise multiple, around since 2014.
martincmartin wrote: Mon Oct 18, 2021 6:13 am There are new ETFs from Dimensional Fund Advisors, e.g. DFAC and and DFAX. They implement the factor model, claim tax efficiency and don't contain REITs, which means you can keep all your REITs in sheltered. On the minus side, they have relatively high expense ratios, and average to low Morningstar ratings.
Thanks for the pointers. Will do more research on these.
alex_686
Posts: 13320
Joined: Mon Feb 09, 2015 1:39 pm

Re: Investing in the new and improved value factor(s)

Post by alex_686 »

strakert wrote: Mon Oct 18, 2021 4:16 pm To answer my own question: Doesn't look like the indices themselves have changed. However VBR has changed indices not once but twice! It went from the S&P to MSCI to CRSP index. As mentioned above, these are fairly different.
We are talking about 2 very different things.

The original and primary purpose of a index is for measuring a portfolio's performance.

VBR switched from one index to the next mainly for licensing fees. Yes, the fund has to pay the index provider to use their index. I would not call these "fairly different". It is like switching from yards to meters. The changes are fairly small. And the impact of a fund switching indexes tend to be pretty small.

On the other hand you really should not muck around with a index's methodology. Imagine if the length of a yard stick changed depending on what year you were using it. It would wreck the purpose of the index. Changes to indexes are major and you note the changes when presenting historical data.

I think that the Value factor is in trouble. In part because it is hard to handle intangible assets. Minor revisions will not solve the answer. Major revisions mean a whole new measuring stick.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
Topic Author
strakert
Posts: 85
Joined: Sat Oct 02, 2021 12:06 pm

Re: Investing in the new and improved value factor(s)

Post by strakert »

typical.investor wrote: Mon Oct 18, 2021 2:02 am
strakert wrote: Sun Oct 17, 2021 4:46 pm
Several recent industry papers exploring the 13+ year drawdown of value similarly conclude that the original HML definition is obsolete and needs further adjustments to be useful.

[*]Research Affiliates: Reports of Value’s Death May Be Greatly Exaggerated (Oct 2020)
Sure, that paper constructs an iHML factor following the same rules we previously used to construct the regular B/P-based HML factor, with only one change. Instead of using the book-to-market ratio to define value, they use the ratio of our intangibles-adjusted book value to market value (iBook to Market) to define the value factor.

They note the the outperformance of the iHML factor, relative to B/P-based HML, is statistically significant at the 5% significance level and notably beats B/P-based HML by nearly a twofold margin after 1990.

They also look earnings-to-price (E/P) ratio and sales-to-price (S/P) as value measures and note that the "improved" iHML measure of value has also recently suffered a large drawdown, and post-2007 is still not as good as S/P or E/P.
Thanks for pointing this out, I'd skipped over the table. I also found this post with graphs showing relative performance of some metrics, confirming that E/P and S/P handily beat B/P, but a multi-metric definition did even better. On the other hand, in the table the composite metric had worse performance than any of the 3 individual metrics (except in 2007-2020 where B/P was worse). I'm guessing that the difference is that the paper creates a normalized average score of all metrics and then ranks the stocks, while the blog post ranks the stocks independently by each metric and picks stocks in the intersection. So apparently not only the metrics, but how they're combined matters as well. FWIW, I think the S&P 600 SCV takes the former approach, didn't check the others.
typical.investor wrote: Mon Oct 18, 2021 2:02 am
strakert wrote: Sun Oct 17, 2021 4:46 pm My questions are:
  1. Are there any indices and passive ETFs that track the "new and improved" value factors?
  2. Which value factor ETFs do you actually invest in given the considerations above?
I don't know of anything targeting iHML, but I am satisfied by the methodology in the Russell Rafi Series available via Schwab and use FNDA, FNDC, FNDE etc.

They use Adjusted Sales instead of sales as shown effective in the paper above, but that's fine as it is just reducing exposure to highly leveraged firms. If those companies' use of leverage is profitable, I expect it'll show up in the Retained Operating Cash Flow, that the series also weights by, which is operating cash flow from operations less dividends and buybacks. Finally it uses Dividends plus Buybacks.

So it's weighted on both earnings and sales.
Thanks for the pointer again. I don't have the accounting knowledge necessary to understand (much less evaluate) their strategy, but I'll keep learning more.
typical.investor wrote: Mon Oct 18, 2021 2:02 am A similar value definition is also used in the PIMCO RAFI Dynamic Multi-Factor strategy takes time-varying exposures to four return factors; value, low volatility, quality, and momentum. I have some invested there. It'll weight more on factors that are cheap (and low volatility is know to have better returns when cheap) and also on factor momentum. Oh, factors have momentum which means poor value returns tend to persist and do good ones. I'm not completely sold that the time frame for these factor reversals will meet up with my investment horizon and the ER is a little higher so I have much less exposure here.
I can barely understand a single factor, so multi-factor models with factor timing are out of scope for now.
typical.investor wrote: Mon Oct 18, 2021 2:02 am At the end of the day, different value funds will likely perform better or worse in different periods. Anything with some measure of profitability as opposed to simply book-to-market will do a decent job I think. It's hard to know in any given future holding period what the optimal methodology will be. That's my opinion anyway.
Good to know that including a profitability metric is the important thing. What do you think about the other main objection re: sector diversification? Do you think inclusion of other metrics provides that already?
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Investing in the new and improved value factor(s)

Post by Nathan Drake »

alex_686 wrote: Mon Oct 18, 2021 4:33 pm
strakert wrote: Mon Oct 18, 2021 4:16 pm To answer my own question: Doesn't look like the indices themselves have changed. However VBR has changed indices not once but twice! It went from the S&P to MSCI to CRSP index. As mentioned above, these are fairly different.
We are talking about 2 very different things.

The original and primary purpose of a index is for measuring a portfolio's performance.

VBR switched from one index to the next mainly for licensing fees. Yes, the fund has to pay the index provider to use their index. I would not call these "fairly different". It is like switching from yards to meters. The changes are fairly small. And the impact of a fund switching indexes tend to be pretty small.

On the other hand you really should not muck around with a index's methodology. Imagine if the length of a yard stick changed depending on what year you were using it. It would wreck the purpose of the index. Changes to indexes are major and you note the changes when presenting historical data.

I think that the Value factor is in trouble. In part because it is hard to handle intangible assets. Minor revisions will not solve the answer. Major revisions mean a whole new measuring stick.
Plenty of ways to quantify value absent P/B.

Avantis takes this into account. I’m not concerned, the value factor is robust
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
Topic Author
strakert
Posts: 85
Joined: Sat Oct 02, 2021 12:06 pm

Re: Investing in the new and improved value factor(s)

Post by strakert »

alex_686 wrote: Mon Oct 18, 2021 4:33 pm
strakert wrote: Mon Oct 18, 2021 4:16 pm To answer my own question: Doesn't look like the indices themselves have changed. However VBR has changed indices not once but twice! It went from the S&P to MSCI to CRSP index. As mentioned above, these are fairly different.
We are talking about 2 very different things.

VBR switched from one index to the next mainly for licensing fees. Yes, the fund has to pay the index provider to use their index. I would not call these "fairly different". It is like switching from yards to meters. The changes are fairly small. And the impact of a fund switching indexes tend to be pretty small.

I think that the Value factor is in trouble. In part because it is hard to handle intangible assets. Minor revisions will not solve the answer. Major revisions mean a whole new measuring stick.
Makes sense that indices are stable. However the difference between VBR (CRSP) and VIOV (S&P 600) is significant if you look at them as standalone SCV funds. VBR has almost 1000 companies while VIOV has 500. Median market cap of VBR is ~2x that of VIOV. Avg. P/B of VBR is 25% higher. Their top 10 holdings have 0 overlap. If someone is looking for implementing the value factor specifically, I'd imagine they'd have a strong opinion about this change.

OTOH, looking at portfolio visualizer their overall performance since 2013 has been about the same. That somewhat addresses my concerns about performance sensitivity w.r.t. these 2 indices - it doesn't seem to matter that much. Thanks for making me look!
muffins14
Posts: 5528
Joined: Wed Oct 26, 2016 4:14 am
Location: New York

Re: Investing in the new and improved value factor(s)

Post by muffins14 »

nisiprius wrote: Sun Oct 17, 2021 8:02 pm
Beensabu wrote: Sun Oct 17, 2021 6:41 pm...The size factor definitely isn't dead...
In 1981, in "The Relationship Between Return and Market Value of Common Stocks," Rolf W. Banz wrote:
It is found that smaller firms have had higher risk adjusted returns, on average, than larger firms. This ‘size effect’ has been in existence for at least forty years and is evidence that the capital asset pricing model is misspecified.
The DFA Micro Cap Portfolio had inception almost immediately after the paper was published.

During the 40 years of data since inception, compared to an S&P 500 index fund, the DFA Micro Cap Portfolio has had
  • A virtually identical average return, 11.90% versus 11.93%
  • A considerably higher standard deviation, 19.76% versus 14.99%
  • And, thus, a considerably lower risk-adjusted return, a Sharpe ratio of 0.49 versus 0.59
Image

The important hope or expectation was higher risk-adjusted return. It hasn't been fulfilled.

Less important, but often talked about, was an expectation of a "premium," that is, simply higher return, even if nothing more than a reward for taking more risk. But that hasn't been fulfilled, either.

Banz said that the size effect had been in existence for forty years. Well, it now has been out of existence for forty years.

If that isn't "dead," it is hard to imagine what "dead" would look like.

The existence of factors is supposed to help explain movements in returns, not to necessarily guarantee that a specific mutual fund has a higher return, right?

It's entirely possible that the starting and ending values are exactly the same for two hypothetical portfolios A and B but the path between A and B could be very different depending on the types of stocks involved. If the size variable explains how you got from point A to point B better than no variable at all, then that is the size factor.

The factor having explanatory power is not the same as it having higher risk-adjusted returns, or a premium at all, I believe.
Crom laughs at your Four Winds
Topic Author
strakert
Posts: 85
Joined: Sat Oct 02, 2021 12:06 pm

Re: Investing in the new and improved value factor(s)

Post by strakert »

muffins14 wrote: Mon Oct 18, 2021 5:27 pm
nisiprius wrote: Sun Oct 17, 2021 8:02 pm
Beensabu wrote: Sun Oct 17, 2021 6:41 pm...The size factor definitely isn't dead...
In 1981, in "The Relationship Between Return and Market Value of Common Stocks," Rolf W. Banz wrote:
It is found that smaller firms have had higher risk adjusted returns, on average, than larger firms. This ‘size effect’ has been in existence for at least forty years and is evidence that the capital asset pricing model is misspecified.
The DFA Micro Cap Portfolio had inception almost immediately after the paper was published.

During the 40 years of data since inception, compared to an S&P 500 index fund, the DFA Micro Cap Portfolio has had
  • A virtually identical average return, 11.90% versus 11.93%
  • A considerably higher standard deviation, 19.76% versus 14.99%
  • And, thus, a considerably lower risk-adjusted return, a Sharpe ratio of 0.49 versus 0.59
Image

The important hope or expectation was higher risk-adjusted return. It hasn't been fulfilled.

Less important, but often talked about, was an expectation of a "premium," that is, simply higher return, even if nothing more than a reward for taking more risk. But that hasn't been fulfilled, either.

Banz said that the size effect had been in existence for forty years. Well, it now has been out of existence for forty years.

If that isn't "dead," it is hard to imagine what "dead" would look like.

The existence of factors is supposed to help explain movements in returns, not to necessarily guarantee that a specific mutual fund has a higher return, right?

The factor having explanatory power is not the same as it having higher risk-adjusted returns, or a premium at all, I believe.
Yes, by construction the factors are meant to explain past returns, not predict future returns. But there's been a lot of additional research on the major factors that did explain past results well, to establish if there are any structural reasons they can be expected to persist in the future. This takes the form of economic or risk premium based theories - establishing that the factor premiums arise from underlying economic structures or non-diversifiable risks that investors should be rewarded for; or behavioral theories - establishing that the factor premiums arise from systematic biases and vagaries of human nature. In either case, the expectation is that these sources of factor premiums are NOT something can be arbitraged away once the factors are known. Following this process, the only factors worth investing in are those that have convincing persistence research supporting them.

There are different opinions about this "persistence research" and how true/useful it is. Some say it's well founded, others say it's just post-facto justifications. If you're interested, I'm trying to answer this question for myself separately in this thread.

But factor believers DO believe that the factors will persist. Let's accept this belief for the purposes of this thread.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Investing in the new and improved value factor(s)

Post by nedsaid »

strakert wrote: Sun Oct 17, 2021 9:41 pm Thanks nedsaid.
nedsaid wrote: Sun Oct 17, 2021 8:22 pm What we are discussing here is tweaking and combining factors and finding the managers that are up to the latest research.
From my reading the issue is more serious than that. It's not just that they've found a better version (there's plenty of that too), it's that original HML value factor has lost its explanatory power since discovery (just like the original SMB factor), completely in large-cap and largely in small-cap. This is blamed mostly on book value not accounting for intangibles (R&D, branding). These are much more important in tech-heavy 21st century firms than old and capital intensive industrials and utilities, leading to an unintentional and unproductive sector tilt.

It may be that the SCV indices are not affected as strongly by this and other issues, e.g., because they already use multiple metrics. If so, I'm looking for more information on *how much* of a non-issue this is. I.e., where do current indices fall on the spectrum of "our value factor definition is useless going forward" to "our hyper-tweaked definition captures 110% of value premium". Also, does anyone know if the construction of these indices has changed over time to accommodate newer research?
I guess it is because of my background in accounting that I have always looked at such things as Price/Earnings, Price/Cash Flow, Price/Sales. In other words, I take mostly an Income Statement approach to valuation as stock prices are largely a valuation of the cash flows of the company. Assets such as Property, Plant, and Equipment are valued at historical cost minus depreciation on the balance sheet. I am more interested in liquidity and debt. Unless the company is very asset rich, I don't take more than a shorter look at the Balance Sheet. Now that the economy is more High Tech, the value of such things as Property, Plant, Equipment and thus Book Value are less meaningful. Intangible assets are becoming more and more important. No one really cares the value of Microsoft's Property, Plant, and Equipment. Because of the trend towards intangible assets, which are difficult for accountants to measure, Price to Book is less important than it used to be.

My number one question is how much are investors paying for a dollar of earnings? For the most part, stocks are valued according to the value of the cash flows that a corporation generates. The balance sheet has a lesser but important effect, for example more cash on the balance sheet tends toward higher P/E ratios. Debt to market capitalization is another consideration. Better to have a strong balance sheet rather than a weak balance sheet.
A fool and his money are good for business.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Investing in the new and improved value factor(s)

Post by nedsaid »

garlandwhizzer wrote: Mon Oct 18, 2021 1:26 pm Nisi, I think it you nailed it on the so called "small" factor. DFA has 2 Nobel Prize winners on its board of directors as well as other acclaimed academics on its payroll. One assumes that DFA is state of the art at creating a portfolio to harvest the small premium. One also assumes that micro-cap would maximize that premium since cap weight is precisely how you measure size and micro-cap is the smallest. While there is huge controversy and no certainty whatsoever about how to currently measure value, it's very easy to measure size. So the DFA Microcap Fund should massively outperform the S&P 500 which is dominated by large and mega-cap over a long time span long term if the small factor is actually harvestable after costs in the real world. Failing to outperform and doing so with much more volatility over a 40 year time span, which is long term in anybody's book, should put this factor in its well deserved grave, at least for those who are actually interested in risk adjusted returns rather than a nice narrative based on an unrealistic factor model.

Humans facing uncertainty are hard wired to create and embrace a seemingly convincing narrative that seeks to unravel that uncertainty. Uncertainty is precisely what our brains cannot tolerate. Being wrong is much easier to tolerate. When shown to be wrong, we simply create a new narrative to explain it as is currently going on with the value factor. We love our narratives and embrace them more than real facts.

Factor models have been in a constant state of evolution and change for 13 years. This has happened in response to research which demonstrates for example that the original definition of value PB doesn't work now. We are always in the process of finding a new approach for defining value. That new and improved approach may turn out likewise not work over the next 13 years. At that point yet another new approach will appear and the true believers will embrace the latest one.True believers in the factor model will not lose faith. Just like today's failure of PB. The failure of multi-factor funds to deliver and the failure of the small factor as nisi has pointed out--these hard facts do not dissuade those who embrace the model. They simply alter the model as time goes by.

Personally, I don't know whether cap weight or factor approaches will win out long term for risk adjusted returns. I am willing to admit my ignorance rather than pretending to know up front the future winner. I suspect both will do well if adhered to but that is as far my belief goes.

Garland Whizzer
The Death of Value in my view is greatly exaggerated, until the summer of 2020 the stock market had been in a Growth trend since 2009. The Value Premium will look a lot smaller after 11 years during which Growth predominated. Small Value actually kept pace with the S&P 500 until the last 3-4 years. As mentioned about the Size factor, combining it with Quality really helps. What has changed as far as defining what Value is the declining importance of Book Value.
A fool and his money are good for business.
Topic Author
strakert
Posts: 85
Joined: Sat Oct 02, 2021 12:06 pm

Re: Investing in the new and improved value factor(s)

Post by strakert »

nedsaid wrote: Mon Oct 18, 2021 5:47 pm
strakert wrote: Sun Oct 17, 2021 9:41 pm Thanks nedsaid.
nedsaid wrote: Sun Oct 17, 2021 8:22 pm What we are discussing here is tweaking and combining factors and finding the managers that are up to the latest research.
From my reading the issue is more serious than that. It's not just that they've found a better version (there's plenty of that too), it's that original HML value factor has lost its explanatory power since discovery (just like the original SMB factor), completely in large-cap and largely in small-cap. This is blamed mostly on book value not accounting for intangibles (R&D, branding). These are much more important in tech-heavy 21st century firms than old and capital intensive industrials and utilities, leading to an unintentional and unproductive sector tilt.

It may be that the SCV indices are not affected as strongly by this and other issues, e.g., because they already use multiple metrics. If so, I'm looking for more information on *how much* of a non-issue this is. I.e., where do current indices fall on the spectrum of "our value factor definition is useless going forward" to "our hyper-tweaked definition captures 110% of value premium". Also, does anyone know if the construction of these indices has changed over time to accommodate newer research?
I guess it is because of my background in accounting that I have always looked at such things as Price/Earnings, Price/Cash Flow, Price/Sales. In other words, I take mostly an Income Statement approach to valuation as stock prices are largely a valuation of the cash flows of the company. Assets such as Property, Plant, and Equipment are valued at historical cost minus depreciation on the balance sheet. I am more interested in liquidity and debt. Unless the company is very asset rich, I don't take more than a shorter look at the Balance Sheet. Now that the economy is more High Tech, the value of such things as Property, Plant, Equipment and thus Book Value are less meaningful. Intangible assets are becoming more and more important. No one really cares the value of Microsoft's Property, Plant, and Equipment. Because of the trend towards intangible assets, which are difficult for accountants to measure, Price to Book is less important than it used to be.

My number one question is how much are investors paying for a dollar of earnings? For the most part, stocks are valued according to the value of the cash flows that a corporation generates. The balance sheet has a lesser but important effect, for example more cash on the balance sheet tends toward higher P/E ratios. Debt to market capitalization is another consideration. Better to have a strong balance sheet rather than a weak balance sheet.
I see what you mean, and typical.investor made a similar point about importance of profitability. Thanks for explaining.

Let me clarify what I'm asking for and why:

I (and presumably most new value factor investors) do not have a background in accounting. I don't know what the metrics represent, what the relationships between them are, or how they should (even in theory) affect a stock's returns. To me, they are some opaque functions of a firm's fundamentals, of which a very specific combination (HML) had been analyzed and shown to generate extra returns using a methodology I trust, and research had been done to establish that this combination of metrics would continue doing so. Now the latest research says that the previously analyzed combination is no longer predictive, and there are better metrics and better ways to combine them.

Because of my ignorance, I'm not in a position to evaluate the sensitivity of previous results to these differences in metrics, transformations of these metrics, or ways to combine these metrics. However I do have enough knowledge to know that they can be very sensitive to such differences. As an example, see my reply to simple.investor above about how two slightly different "composites" of similar metrics have outperformed all of them in one case, and underperformed all of them in another!

So my question is essentially: where can I find the backtests and factor analysis for the value factor as implemented by actual indices or funds. I'm sure the index / ETF providers must have done this analysis themselves to construct the indices / funds. Is this information available publicly?

Alternatively, is there a website / tool where I can run regressions on the new value factor definition for available ETFs and see how well they capture it? Note that if that original HML B/P definition is not useful, the regressions on Fama-French 3-factor models on portfolio visualizer etc. should also not be useful.

The general theme of responses so far seems to be that all the existing indices capture "true value" sufficiently well and are robust to these redefinitions. Unfortunately this being the internet I also cannot tell whether this belief is based on faith or on deep knowledge of accounting, factor analysis and index construction, hence the skepticism.
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Investing in the new and improved value factor(s)

Post by Nathan Drake »

I'd be more concerned about the "Value is Dead" argument if earnings for Value significantly trailed the market. But the truth is that Value had almost identical earnings over the past decade. The humongous returns for the market were due to valuation expansion (speculative return).
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
Topic Author
strakert
Posts: 85
Joined: Sat Oct 02, 2021 12:06 pm

Re: Investing in the new and improved value factor(s)

Post by strakert »

Nathan Drake wrote: Mon Oct 18, 2021 8:21 pm I'd be more concerned about the "Value is Dead" argument if earnings for Value significantly trailed the market. But the truth is that Value had almost identical earnings over the past decade. The humongous returns for the market were due to valuation expansion (speculative return).
"Size/Value is dead" above should be read as "the previously discovered premium for having invested in SMB or HML portfolios is no longer expected going forward".

This says nothing about whether returns for continuing to invest in SMB or HML will be higher or lower than investing in the market. Just like it says nothing about investing in an arbitrary portfolio of stocks.

It turns out that it is not even possible to invest in the academic definition of SMB or HML, since no ETF/funds implement size/value exactly as defined, so YMMV. But by the same logic, it's also not correct to claim that investing in "small cap value" ETFs is capturing the "size or value premium" defined by Fama and French and supported by other academic research.
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Investing in the new and improved value factor(s)

Post by Nathan Drake »

strakert wrote: Mon Oct 18, 2021 8:39 pm
Nathan Drake wrote: Mon Oct 18, 2021 8:21 pm I'd be more concerned about the "Value is Dead" argument if earnings for Value significantly trailed the market. But the truth is that Value had almost identical earnings over the past decade. The humongous returns for the market were due to valuation expansion (speculative return).
"Size/Value is dead" above should be read as "the previously discovered premium for having invested in SMB or HML portfolios is no longer expected going forward".

This says nothing about whether returns for continuing to invest in SMB or HML will be higher or lower than investing in the market. Just like it says nothing about investing in an arbitrary portfolio of stocks.

It turns out that it is not even possible to invest in the academic definition of SMB or HML, since no ETF/funds implement size/value exactly as defined, so YMMV. But by the same logic, it's also not correct to claim that investing in "small cap value" ETFs is capturing the "size or value premium" defined by Fama and French and supported by other academic research.
You don't need to. Value factor is robust. It works for numerous measures of value. You can't construct a long/short portfolio, but you can capture the long value loadings. Returns won't be as high, but they are expected to, and have, outperformed the market.

Valuation expansion gives more credence to the idea that it's not dead, just growing spreads.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
Topic Author
strakert
Posts: 85
Joined: Sat Oct 02, 2021 12:06 pm

Re: Investing in the new and improved value factor(s)

Post by strakert »

Nathan Drake wrote: Mon Oct 18, 2021 8:48 pm
strakert wrote: Mon Oct 18, 2021 8:39 pm
Nathan Drake wrote: Mon Oct 18, 2021 8:21 pm I'd be more concerned about the "Value is Dead" argument if earnings for Value significantly trailed the market. But the truth is that Value had almost identical earnings over the past decade. The humongous returns for the market were due to valuation expansion (speculative return).
"Size/Value is dead" above should be read as "the previously discovered premium for having invested in SMB or HML portfolios is no longer expected going forward".

This says nothing about whether returns for continuing to invest in SMB or HML will be higher or lower than investing in the market. Just like it says nothing about investing in an arbitrary portfolio of stocks.

It turns out that it is not even possible to invest in the academic definition of SMB or HML, since no ETF/funds implement size/value exactly as defined, so YMMV. But by the same logic, it's also not correct to claim that investing in "small cap value" ETFs is capturing the "size or value premium" defined by Fama and French and supported by other academic research.
You don't need to. Value factor is robust. It works for numerous measures of value. You can't construct a long/short portfolio, but you can capture the long value loadings. Returns won't be as high, but they are expected to, and have, outperformed the market.

Valuation expansion gives more credence to the idea that it's not dead, just growing spreads.
To clarify, I'm not claiming that value factor will not work going forward. I'm looking for any papers or analysis that show that it is indeed robust to different measures of value. Valuation spread expansion as an explanation for underperformance makes sense. The RAFI/Robeco papers above make that case convincingly.
muffins14
Posts: 5528
Joined: Wed Oct 26, 2016 4:14 am
Location: New York

Re: Investing in the new and improved value factor(s)

Post by muffins14 »

garlandwhizzer wrote: Mon Oct 18, 2021 1:26 pm One also assumes that micro-cap would maximize that premium since cap weight is precisely how you measure size and micro-cap is the smallest. While there is huge controversy and no certainty whatsoever about how to currently measure value, it's very easy to measure size. So the DFA Microcap Fund should massively outperform the S&P 500 which is dominated by large and mega-cap over a long time span long term if the small factor is actually harvestable after costs in the real world. Failing to outperform and doing so with much more volatility over a 40 year time span, which is long term in anybody's book, should put this factor in its well deserved grave,
Garland Whizzer
This feels a little lazy to me

If I go to portfolio visualizer and make an 80/20 stock/bond portfolio, one with a the DFA micro cap fund DFSCX plus the oldest share class of Vanguard total bond, and another portfolio with total US stock (or S&P500) plus total bond , the data go back 38.5 years to May 1992

The micro cap portfolio is ahead almost the entire time, except 1999-2001. Why does that mean it’s an unharvestable, dead, failure?

Even a straight up 100% stock comparison between your micro cap fund and the S&P 500 shows that the micro cap fund has outperformed since May 1992. Did you extend the data from 38.5 years to 40 years and get different results with different total market funds?
Crom laughs at your Four Winds
MotoTrojan
Posts: 11259
Joined: Wed Feb 01, 2017 7:39 pm

Re: Investing in the new and improved value factor(s)

Post by MotoTrojan »

I'd suggest you come join some of us (HML_Compounder there) on the RR forum where factors are discussed in-depth.

https://community.rationalreminder.ca/

I personally use a combination of Avantis (AVUV/AVDV/AVES) and Alpha Architect (QVAL/IVAL) products to get some diversity in my value methodology. Avantis is probably the best out there (DFA as well but more expensive and less accessible, plus I prefer Avantis' greater emphasis on profitability) if you want to follow the more classic Fama French 5 factor type research/methodology. Alpha Architect takes a different approach with less emphasis on small-caps, but with concentration (50 stocks) and the use of EV/EBIT as a value metric which I am a fan of. I think the combo of both is a great way to go as it gives you exposure to cheap assets, and cheap profits.

And don't forget global diversification! Value has done exceptionally well in developed and emerging markets with ~30 and ~60bp/month premiums respectively since publication.

Think Robert T pointed this out on here recently but holding DFA's US SCV, DM SCV, and EM Value close to market-cap or fundamental weight (this is my allocation here, 40/40/20) resulted in a higher CAGR and Sharpe ratio than any of the individual funds, with close to the minimum drawdown.

https://www.portfoliovisualizer.com/bac ... tion3_1=20
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Investing in the new and improved value factor(s)

Post by nedsaid »

strakert wrote: Mon Oct 18, 2021 8:10 pm
nedsaid wrote: Mon Oct 18, 2021 5:47 pm
strakert wrote: Sun Oct 17, 2021 9:41 pm Thanks nedsaid.
nedsaid wrote: Sun Oct 17, 2021 8:22 pm What we are discussing here is tweaking and combining factors and finding the managers that are up to the latest research.
From my reading the issue is more serious than that. It's not just that they've found a better version (there's plenty of that too), it's that original HML value factor has lost its explanatory power since discovery (just like the original SMB factor), completely in large-cap and largely in small-cap. This is blamed mostly on book value not accounting for intangibles (R&D, branding). These are much more important in tech-heavy 21st century firms than old and capital intensive industrials and utilities, leading to an unintentional and unproductive sector tilt.

It may be that the SCV indices are not affected as strongly by this and other issues, e.g., because they already use multiple metrics. If so, I'm looking for more information on *how much* of a non-issue this is. I.e., where do current indices fall on the spectrum of "our value factor definition is useless going forward" to "our hyper-tweaked definition captures 110% of value premium". Also, does anyone know if the construction of these indices has changed over time to accommodate newer research?
I guess it is because of my background in accounting that I have always looked at such things as Price/Earnings, Price/Cash Flow, Price/Sales. In other words, I take mostly an Income Statement approach to valuation as stock prices are largely a valuation of the cash flows of the company. Assets such as Property, Plant, and Equipment are valued at historical cost minus depreciation on the balance sheet. I am more interested in liquidity and debt. Unless the company is very asset rich, I don't take more than a shorter look at the Balance Sheet. Now that the economy is more High Tech, the value of such things as Property, Plant, Equipment and thus Book Value are less meaningful. Intangible assets are becoming more and more important. No one really cares the value of Microsoft's Property, Plant, and Equipment. Because of the trend towards intangible assets, which are difficult for accountants to measure, Price to Book is less important than it used to be.

My number one question is how much are investors paying for a dollar of earnings? For the most part, stocks are valued according to the value of the cash flows that a corporation generates. The balance sheet has a lesser but important effect, for example more cash on the balance sheet tends toward higher P/E ratios. Debt to market capitalization is another consideration. Better to have a strong balance sheet rather than a weak balance sheet.
I see what you mean, and typical.investor made a similar point about importance of profitability. Thanks for explaining.

Let me clarify what I'm asking for and why:

I (and presumably most new value factor investors) do not have a background in accounting. I don't know what the metrics represent, what the relationships between them are, or how they should (even in theory) affect a stock's returns. To me, they are some opaque functions of a firm's fundamentals, of which a very specific combination (HML) had been analyzed and shown to generate extra returns using a methodology I trust, and research had been done to establish that this combination of metrics would continue doing so. Now the latest research says that the previously analyzed combination is no longer predictive, and there are better metrics and better ways to combine them.

Because of my ignorance, I'm not in a position to evaluate the sensitivity of previous results to these differences in metrics, transformations of these metrics, or ways to combine these metrics. However I do have enough knowledge to know that they can be very sensitive to such differences. As an example, see my reply to simple.investor above about how two slightly different "composites" of similar metrics have outperformed all of them in one case, and underperformed all of them in another!

So my question is essentially: where can I find the backtests and factor analysis for the value factor as implemented by actual indices or funds. I'm sure the index / ETF providers must have done this analysis themselves to construct the indices / funds. Is this information available publicly?

Alternatively, is there a website / tool where I can run regressions on the new value factor definition for available ETFs and see how well they capture it? Note that if that original HML B/P definition is not useful, the regressions on Fama-French 3-factor models on portfolio visualizer etc. should also not be useful.

The general theme of responses so far seems to be that all the existing indices capture "true value" sufficiently well and are robust to these redefinitions. Unfortunately this being the internet I also cannot tell whether this belief is based on faith or on deep knowledge of accounting, factor analysis and index construction, hence the skepticism.
It isn't too hard, you have to know the accounting assumptions behind the numbers. If you want to know the actual cash that a business generates after expenses, look at Cash Flow. Earnings take into account non-cash charges like depreciation and depletion. Depreciation takes into account that things wear out and need replacing over time. Depletion is the reduction of known resources through production such as oil deposits. Even though depreciation and depletion are non-cash expenses, they are actual expenses over time. It is a way for companies to recognize the costs over time and hopefully reserve for replacement.

Assets are booked under historical cost. You might have depletion/depreciation/amortization but you add back the cost of improvements. So over time, a building, like the Empire State Building is valued at historical costs minus depreciation plus improvements. The improvements get depreciated as well. But over time, the actual value of the Empire State Building might get grossly understated on the books.

As far as intangible assets, these are harder to value. For example, Microsoft has tremendous value in its Windows and Office franchises, but hard for the accountants to put a valuation on these. You can see the limitations of accounting. Intangible assets such as goodwill are amortized over time. Goodwill is the purchase of a business above its Owner's Equity on the Balance Sheet. So if you buy a business for $5 million and the accountants say it is worth $4 million, you have one million in goodwill.

Marketable securities are marked up to market value and the company will have a loss or income depending upon fluctuation.

So stuff like that.

You can use Portfolio Visualizer for backtests with actual funds. There are limitations for this as Vanguard has changed its Indexes that particular funds follow over time. Some of their funds switched to CRSP Indexes. So you don't get exact comparisons but with actual funds that is the best you can do. The tool also will give information on factor loading. There are other tools as well.
A fool and his money are good for business.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Investing in the new and improved value factor(s)

Post by nedsaid »

Other things to know about valuing stocks. Wall Street likes consistent earnings growth, doesn't matter if the growth is relatively slow or relatively fast, as long as it is consistent growth the Stock Market will put a premium on that growth with higher P/E ratios. An earnings stream that is volatile will command lower P/E ratios than earnings streams that are more consistent. Consistency uber Alles.

Obviously, Wall Street likes faster growth, these are the Aggressive Growth stocks and will command high P/E ratios, even higher if the growth rates are consistent. Sort of like a successful Restaurant chain that opens up in one area of the country and continuously expands across the nation. These stocks are riskier because if an earnings report disappoints then the stock price can drop by a lot, it is called being priced to perfection. Microsoft was a classic Growth stock in the later 1980's and the 1990's, consistent 15% or more earnings growth year after year. These type of stocks can make you a mint.

Consumer Staples stocks like Proctor and Gamble, Colgate Palmolive, Coke, Pepsi, General Mills, all produce stuff that people want and need and demand is consistent even during recessions. Even though earnings growth is slower than a classic Growth stock, these stocks are rarely ever cheap because these companies are so consistent. Many of these are the so-called Low Volatility stocks because the ups and downs in stock price are steadier than the market itself, people also like them because they don't drop in price as much during recessions. These companies have strong balance sheets and thus have the financial resources to deal with tough economic times. They also are dividend payers.

So in an environment of 2% inflation and 2-3% economic growth, the low volatility stocks might grow earnings at 4-6% a year but consistently. Slow growth but the market loves them because the earnings are so darned steady. Blue Chips, the sweet spot of the market, have steady growth rates more like 8% to 9%. Aggressive growth has growth rates of perhaps 10% or more, nirvana for Aggressive Growth are consistent growth rates of 15% or more.

Value stocks tend to have higher leverage and more volatile earnings than the market as a whole, so their P/E ratios will be lower. What you don't want is a "Value trap", a dying company in a dying industry is the ultimate nightmare. A big trick to Value investing is to buy companies which are down in price but whose problems are temporary in nature, it isn't always easy to tell the difference between temporary or permanent. If you are a Value investor, you will have your disappointments. What you really want is a good company which is out of favor and/or experiencing temporary problems. Strength of management is something else you want to look at. Sometimes stocks are cheap because the company has terrible management.

Strong balance sheets tend towards higher P/E ratios and weak balance sheets tend towards lower P/E ratios. So hopefully this answers some questions.
Last edited by nedsaid on Wed Oct 20, 2021 12:25 am, edited 1 time in total.
A fool and his money are good for business.
garlandwhizzer
Posts: 3565
Joined: Fri Aug 06, 2010 3:42 pm

Re: Investing in the new and improved value factor(s)

Post by garlandwhizzer »

muffins14 wrote:

The micro cap portfolio is ahead almost the entire time, except 1999-2001. Why does that mean it’s an unharvestable, dead, failure?
If you look at nisi's graph DFSVX versus S&P 500 you'll see that there are alternating periods of outperformance and underperformance by both the DFA micro-cap fund and the S&P500 over the span of 40 years. The S&P 500 outperformed from 1986 - 2002, 16 years running, and it also outperformed 2008 - 2020, another 12 years running. Micro-cap had its own runs. In the end after 40 years it was a wash, they were almost exactly dead even. S&P 500 clearly had lower volatility over the 40 year time span resulting in better risk adjusted returns as measured by the Sharpe ratio. But the S&P 500 is merely beta for large cap stocks and it does not claim to generate a "premium" like the size factor does. Premium means excess returns to me. Clearly there was no premium whatsoever during this period, only a bumpier ride for the micro-cap fund.

In Larry Swedroe's book, YOUR COMPLETE GUIDE TO FACTOR-BASED INVESTING, Larry an acknowledged expert on factor approaches, estimated the size factor to produce a premium of 3.3% annually based on backtesting from 1927 - 2015. That is based on index returns, not SC fund returns. Important distinction because in real funds, as nisi pointed out it was zero for 40 years. How can this be?

I really don't care in the least about the results of an index that has no management fee, does cost-free long/short, has no trading fees, no trading frictions, and no liquidity restraints. That is a theoretical entity only. Trading costs and frictions as well as liquidity restraints are massive in the small/micro-cap space. When fund management tries to selling 10,000 shares from their micro-cap stock fund, the per share sales price collapses. There are no large bid orders, only large sell orders. The buyers are not idiots and they pay less and less as the fund tries to unload a large block of stock. The opposite happens when the fund wants to buy 10,000 shares of a micro-cap stock. The guys on the other side of the trade extract a pound of flesh from the fund with every time they buys or sell stock not to mention the sales commission per share. These problems are severe in the illiquid SC/micro-cap space which is one reason why LC stocks which are very liquid and have few trading frictions are more popular and expensive. You can buy or sell millions of dollars of LC stocks quickly without much impact on sales price.

Small cap and micro-cap Indexes completely ignore all these problems and they add an extra boost, cost-free long/short. They assume you can sell the whole lot for whatever the current price is at any moment and do at no cost. Hence results of SC indexes are very misleading and do not translate into the real world of investing. Indexes are designed to create impressive results that will draw in investment dollars. It's not hard to do if you get to write all the rules. However, the so called "size premium" disappears completely in the real world when you subtract management fees, trading costs, trading frictions, and liquidity restraints of real funds. It is important also as a starting point to cut factor results in half up front because factor funds are typically long-only, having learned already that long/short is very expensive, difficult, and counterproductive especially in the SC space.

Garland Whizzer
Massdriver
Posts: 233
Joined: Tue Jan 02, 2018 11:05 am

Re: Investing in the new and improved value factor(s)

Post by Massdriver »

MotoTrojan wrote: Tue Oct 19, 2021 8:26 am I'd suggest you come join some of us (HML_Compounder there) on the RR forum where factors are discussed in-depth.

https://community.rationalreminder.ca/

I personally use a combination of Avantis (AVUV/AVDV/AVES) and Alpha Architect (QVAL/IVAL) products to get some diversity in my value methodology. Avantis is probably the best out there (DFA as well but more expensive and less accessible, plus I prefer Avantis' greater emphasis on profitability) if you want to follow the more classic Fama French 5 factor type research/methodology. Alpha Architect takes a different approach with less emphasis on small-caps, but with concentration (50 stocks) and the use of EV/EBIT as a value metric which I am a fan of. I think the combo of both is a great way to go as it gives you exposure to cheap assets, and cheap profits.

And don't forget global diversification! Value has done exceptionally well in developed and emerging markets with ~30 and ~60bp/month premiums respectively since publication.

Think Robert T pointed this out on here recently but holding DFA's US SCV, DM SCV, and EM Value close to market-cap or fundamental weight (this is my allocation here, 40/40/20) resulted in a higher CAGR and Sharpe ratio than any of the individual funds, with close to the minimum drawdown.

https://www.portfoliovisualizer.com/bac ... tion3_1=20
I figured that was you on RR given your cheerful but analytic posting style. I’ve read a lot of content on RR. Thanks for your contributions as usual! :sharebeer
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Investing in the new and improved value factor(s)

Post by Nathan Drake »

nedsaid wrote: Tue Oct 19, 2021 5:49 pm Other things to know about valuing stocks. Wall Street likes consistent earnings growth, doesn't matter if the growth is relatively slow or relatively fast, as long as it is consistent growth the Stock Market will put a premium on that growth with higher P/E ratios. An earnings stream that is volatile will command lower P/E ratios than earnings streams that are more consistent. Consistency uber Alles.

Obviously, Wall Street likes faster growth, these are the Aggressive Growth stocks and will command high P/E ratios, even higher if the growth rates are consistent. Sort of like a successful Restaurant chain that opens up in one area of the country and continuously expands across the nation. These stocks are riskier because if an earnings report disappoints then the stock price can drop by a lot, it is called being priced to perfection. Microsoft was a classic Growth stock in the later 1980's and the 1990's, consistent 15% or more earnings growth year after year. These type of stocks can make you a mint.

Consumer Staples stocks like Proctor and Gamble, Colgate Palmolive, Coke, Pepsi, General Mills, all produce stuff that people want and need and demand is consistent even during recessions. Even though earnings growth is slower than a classic Growth stock, these stocks are rarely ever cheap because these companies are so consistent. Many of these are the so-called Low Volatility stocks because the ups and downs in stock price are steadier than the market itself, people also like them because they don't drop in price as much during recessions. These companies have strong balance sheets and thus have the financial resources to deal with tough economic times. They also are dividend payers.

So in an environment of 2% inflation and 2-3% earnings growth, the low volatility stocks might grow earnings at 4-6% a year but consistently. Slow growth but the market loves them because the earnings are so darned steady. Blue Chips, the sweet spot of the market, have steady growth rates more like 8% to 9%. Aggressive growth has growth rates of perhaps 10% or more, nirvana for Aggressive Growth are consistent growth rates of 15% or more.

Value stocks tend to have higher leverage and more volatile earnings than the market as a whole, so their P/E ratios will be lower. What you don't want is a "Value trap", a dying company in a dying industry is the ultimate nightmare. A big trick to Value investing is to buy companies which are down in price but whose problems are temporary in nature, it isn't always easy to tell the difference between temporary or permanent. If you are a Value investor, you will have your disappointments. What you really want is a good company which is out of favor and/or experiencing temporary problems. Strength of management is something else you want to look at. Sometimes stocks are cheap because the company has terrible management.

Strong balance sheets tend towards higher P/E ratios and weak balance sheets tend towards lower P/E ratios. So hopefully this answers some questions.
Growth Traps are probably even more dangerous (small cap growth historically being a terrible performer). When you have assets priced like Growth when they fail to deliver on the lofty expectations.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Investing in the new and improved value factor(s)

Post by 000 »

garlandwhizzer wrote: Tue Oct 19, 2021 7:47 pm I really don't care in the least about the results of an index that has no management fee, does cost-free long/short, has no trading fees, no trading frictions, and no liquidity restraints. That is a theoretical entity only. Trading costs and frictions as well as liquidity restraints are massive in the small/micro-cap space. When fund management tries to selling 10,000 shares from their micro-cap stock fund, the per share sales price collapses. There are no large bid orders, only large sell orders. The buyers are not idiots and they pay less and less as the fund tries to unload a large block of stock. The opposite happens when the fund wants to buy 10,000 shares of a micro-cap stock. The guys on the other side of the trade extract a pound of flesh from the fund with every time they buys or sell stock not to mention the sales commission per share. These problems are severe in the illiquid SC/micro-cap space which is one reason why LC stocks which are very liquid and have few trading frictions are more popular and expensive. You can buy or sell millions of dollars of LC stocks quickly without much impact on sales price.
All the more reason that factorheads should never have stolen the term 'value' from individual stock investors.

IMHO if one wishes to be a value investor that is going to require more work than following Larry's twitter and jumping on the newest fund.
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Investing in the new and improved value factor(s)

Post by Nathan Drake »

000 wrote: Tue Oct 19, 2021 8:22 pm
garlandwhizzer wrote: Tue Oct 19, 2021 7:47 pm I really don't care in the least about the results of an index that has no management fee, does cost-free long/short, has no trading fees, no trading frictions, and no liquidity restraints. That is a theoretical entity only. Trading costs and frictions as well as liquidity restraints are massive in the small/micro-cap space. When fund management tries to selling 10,000 shares from their micro-cap stock fund, the per share sales price collapses. There are no large bid orders, only large sell orders. The buyers are not idiots and they pay less and less as the fund tries to unload a large block of stock. The opposite happens when the fund wants to buy 10,000 shares of a micro-cap stock. The guys on the other side of the trade extract a pound of flesh from the fund with every time they buys or sell stock not to mention the sales commission per share. These problems are severe in the illiquid SC/micro-cap space which is one reason why LC stocks which are very liquid and have few trading frictions are more popular and expensive. You can buy or sell millions of dollars of LC stocks quickly without much impact on sales price.
All the more reason that factorheads should never have stolen the term 'value' from individual stock investors.

IMHO if one wishes to be a value investor that is going to require more work than following Larry's twitter and jumping on the newest fund.
What work? You can just buy a DFA or Avantis fund and call it a day. These are smartly designed funds that limit frictional costs.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Investing in the new and improved value factor(s)

Post by 000 »

Nathan Drake wrote: Tue Oct 19, 2021 8:25 pm What work? You can just buy a DFA or Avantis fund and call it a day. These are smartly designed funds that limit frictional costs.
Value investing means buying what's a good value i.e. underpriced securities not passively buying random stuff.

Now the OP said:
But factor believers DO believe that the factors will persist. Let's accept this belief for the purposes of this thread.
And I can respect that. My take is that factors may exist at least to some extent over some time frames and think investors seeking to harvest them should probably consider doing their own active management.
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Investing in the new and improved value factor(s)

Post by Nathan Drake »

000 wrote: Tue Oct 19, 2021 8:44 pm
Nathan Drake wrote: Tue Oct 19, 2021 8:25 pm What work? You can just buy a DFA or Avantis fund and call it a day. These are smartly designed funds that limit frictional costs.
Value investing means buying what's a good value i.e. underpriced securities not passively buying random stuff.

Now the OP said:
But factor believers DO believe that the factors will persist. Let's accept this belief for the purposes of this thread.
And I can respect that. My take is that factors may exist at least to some extent over some time frames and think investors seeking to harvest them should probably consider doing their own active management.
Buying a DFA or Avantis fund is not buying "random stuff". The same arguments for largely passive TSM funds applies to largely passive Value funds.

I would hope you would believe factors "may exist to some extent" otherwise maybe you shouldn't invest in the market.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Investing in the new and improved value factor(s)

Post by 000 »

Nathan Drake wrote: Tue Oct 19, 2021 8:47 pm Buying a DFA or Avantis fund is not buying "random stuff". The same arguments for largely passive TSM funds applies to largely passive Value funds.

I would hope you would believe factors "may exist to some extent" otherwise maybe you shouldn't invest in the market.
Sorry but the real world results 30+ years and running since founding of DFA are not promising in the ability of passive 'value' funds to deliver absolute or vol-adjusted positive premia. This seems to be a common pattern with factor approaches, like the Schwab RAFI funds and Vanguard multifactor fund, that they significantly underperform historical patterns once real funds are available.

In the context of this thread, what is the difference between running from one definition of the value factor to the next, and running from one active manager to the next? We can create the same narrative that the new manager finally has things figured out.

If you're going to be a passive investor you shouldn't change to the new and improved value factor(s) as the thread title suggests; instead you should stay the course with the value factor that existed when you started, no?
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Investing in the new and improved value factor(s)

Post by Nathan Drake »

000 wrote: Tue Oct 19, 2021 8:59 pm
Nathan Drake wrote: Tue Oct 19, 2021 8:47 pm Buying a DFA or Avantis fund is not buying "random stuff". The same arguments for largely passive TSM funds applies to largely passive Value funds.

I would hope you would believe factors "may exist to some extent" otherwise maybe you shouldn't invest in the market.
Sorry but the real world results 30+ years and running since founding of DFA are not promising in the ability of passive 'value' funds to deliver absolute or vol-adjusted positive premia. This seems to be a common pattern with factor approaches, like the Schwab RAFI funds and Vanguard multifactor fund, that they significantly underperform historical patterns once real funds are available.

In the context of this thread, what is the difference between running from one definition of the value factor to the next, and running from one active manager to the next? We can create the same narrative that the new manager finally has things figured out.

If you're going to be a passive investor you shouldn't change to the new and improved value factor(s) as the thread title suggests; instead you should stay the course with the value factor that existed when you started, no?
Real world results:
DFSVX vs VTSMX - 1% Premium since inception (period ending in largest Growth vs Value spread in history)
DISVX vs VGTSX - 2.5% Premium since inception (period ending in largest Growth vs Value spread in history)
DFEVX vs VEIEX - 2% Premium since inception (period ending in largest Growth vs Value spread in history)

Value factor is robust - works for many different formulations of "Value". There's no need to continually chase different funds. Pick a loading with low fees and forget about it.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Investing in the new and improved value factor(s)

Post by 000 »

Nathan Drake wrote: Tue Oct 19, 2021 9:16 pm Real world results:
DFSVX vs VTSMX - 1% Premium since inception (period ending in largest Growth vs Value spread in history)
DISVX vs VGTSX - 2.5% Premium since inception (period ending in largest Growth vs Value spread in history)
DFEVX vs VEIEX - 2% Premium since inception (period ending in largest Growth vs Value spread in history)
You forgot the 1% AUM fee, vol-adjusted returns, etc.
Value factor is robust - works for many different formulations of "Value". There's no need to continually chase different funds. Pick a loading with low fees and forget about it.
This whole thread is about the new and improved value factors though.
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Investing in the new and improved value factor(s)

Post by Nathan Drake »

000 wrote: Tue Oct 19, 2021 9:42 pm
Nathan Drake wrote: Tue Oct 19, 2021 9:16 pm Real world results:
DFSVX vs VTSMX - 1% Premium since inception (period ending in largest Growth vs Value spread in history)
DISVX vs VGTSX - 2.5% Premium since inception (period ending in largest Growth vs Value spread in history)
DFEVX vs VEIEX - 2% Premium since inception (period ending in largest Growth vs Value spread in history)
You forgot the 1% AUM fee, vol-adjusted returns, etc.
Value factor is robust - works for many different formulations of "Value". There's no need to continually chase different funds. Pick a loading with low fees and forget about it.
This whole thread is about the new and improved value factors though.
The premiums are after fees have been deducted from each fund. Fees for Avantis are now MUCH lower than DFA, so captured premium should be more favorable.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Investing in the new and improved value factor(s)

Post by 000 »

Nathan Drake wrote: Tue Oct 19, 2021 9:46 pm
000 wrote: Tue Oct 19, 2021 9:42 pm
Nathan Drake wrote: Tue Oct 19, 2021 9:16 pm Real world results:
DFSVX vs VTSMX - 1% Premium since inception (period ending in largest Growth vs Value spread in history)
DISVX vs VGTSX - 2.5% Premium since inception (period ending in largest Growth vs Value spread in history)
DFEVX vs VEIEX - 2% Premium since inception (period ending in largest Growth vs Value spread in history)
You forgot the 1% AUM fee, vol-adjusted returns, etc.
Value factor is robust - works for many different formulations of "Value". There's no need to continually chase different funds. Pick a loading with low fees and forget about it.
This whole thread is about the new and improved value factors though.
The premiums are after fees have been deducted from each fund. Fees for Avantis are now MUCH lower than DFA, so captured premium should be more favorable.
You couldn't and still can't buy those DFA funds except through an advisor.
Post Reply