Avantis Funds: what's so special about them?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Northern Flicker
Posts: 15363
Joined: Fri Apr 10, 2015 12:29 am

Re: Avantis Funds: what's so special about them?

Post by Northern Flicker »

asturzu wrote: Sun Jul 03, 2022 2:17 pm I believe the explanation is here:
https://www.aqr.com/Insights/Perspectiv ... ng-the-Way
If you do enough backtests, you will find what worked. The investor who uses it runs the out of sample clinical trial.
Apathizer
Posts: 2507
Joined: Sun Sep 26, 2021 2:56 pm

Re: Avantis Funds: what's so special about them?

Post by Apathizer »

Northern Flicker wrote: Sun Jul 03, 2022 4:09 pm Factors are not independent despite claims to the contrary. A major limitation of factor theory is the lack of agreement in how to define factors other than size or market.
No, it's clear they've existed in the past. Whether they will persist in the future isn't, but there are good reasons to think they will at least to some extent. While markets are probably reasonably efficient, many investors aren't. This can result in some stocks being unreasonably expensive. A factor-slant might hedge against this.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
User avatar
typical.investor
Posts: 5263
Joined: Mon Jun 11, 2018 3:17 am

Re: Avantis Funds: what's so special about them?

Post by typical.investor »

asturzu wrote: Sun Jul 03, 2022 2:17 pm I believe the explanation is here:
https://www.aqr.com/Insights/Perspectiv ... ng-the-Way
Having read that, I still see nothing I'd call explanatory there.

To recap, AVUS (avantis small cap value) shows no momentum using HML (Fama-French value) but strong a strong momentum loading using HML-DEV.

This makes no sense knowing what we do about the Avantis methodology which has two momentum components:

1) it delays the purchase of stocks with large negative six-month returns and avoids the sale of stocks with large positive six-month returns.
2) it uses an adjusted book-to-market ratio that lags price by three months helps mitigate negative momentum

(see p6 https://www.avantisinvestors.com/conten ... esting.pdf)

The AQR paper you cited defines HML-DEV and states that the original Fama-French measure of HML obscures MOM (called UMD in their paper). the state "the version of momentum implied in original HML isn’t the real momentum factor, but the avoidance of shorting momentum that comes from using a six to eighteen month lag in price."

I really don't see how HML-DEV is going to result in a more accurate measure of momentum in Avantis funds.
Northern Flicker
Posts: 15363
Joined: Fri Apr 10, 2015 12:29 am

Re: Avantis Funds: what's so special about them?

Post by Northern Flicker »

typical_investor wrote: I really don't see how HML-DEV is going to result in a more accurate measure of momentum in Avantis funds.
It may be more accurate, less accurate, or momentum may not really exist. All three explanations are possible.

If you change the definition of a variable, a regression will fit the data to a different hyperplane.

I don’t know which definition of value is better, but I doubt HML-DEV is in common usage.
ElGalloRico
Posts: 20
Joined: Wed Aug 03, 2022 3:00 pm

Re: Avantis Funds: what's so special about them?

Post by ElGalloRico »

Ed 2 wrote: Fri Jul 01, 2022 10:32 pm
JSPECO9 wrote: Fri Jul 01, 2022 10:24 pm
Ed 2 wrote: Fri Jul 01, 2022 10:18 pm
JSPECO9 wrote: Fri Jul 01, 2022 10:03 pm What's the big difference between Vanguard funds and Avantis Funds? I was listening to the Bogleheads podcast and heard the interview with Eduardo Repetto. He was saying all the right things but wanted to check with the Bogleheads first.

He was saying how index funds are great but a defect they have is that they have to rush to buy when there's a new inclusion to the index? Leading to buying overpriced stocks. And how Avantis funds don't do this?

What's the big difference between funds like VTI vs AVUS. Or AVUV (which I keep seeing here repeatedly) vs VIOV/VBR. Is there a Bogleheads consensus on Avantis funds? I only ask because I keep seeing them mentioned on this forum.
Simple explanation. It’s like if your Toyota Corolla instead of four wheels have six.
Can you explain? Does this mean Avantis funds has something Vanguard funds don't have?

Thanks
They buy more small value companies than Vanguard Total Stock Market Index fund has , BUT their funds has higher expense ratio and high turnover ratio , plus I don’t see benefits investing in more small companies for the long term . This funds are more volatile and not everyone will handle such volatility. They claim they have “ scientific “ prove that over long term small value companies will outperform large and so on . To me personally I’d prefer go with the market and happy with results.
Hello,

Actually, Small Cap Value destroys Total Market and Large Cap - and it's not even close. They introduce more volatility and the last decade has shifted significantly more towards Large Cap, but it's hard to ignore the history. These things being, to a degree, cyclical, it is not unreasonable to think that Value stocks will reward the risk premium again.

While I personally maintain the core of my portfolio on VTI, I do use Avantis International and US small cap value ETF's instead of something like VBR which has a Core tilt.

Therein lies what to me, is the potential of Avantis (following Dimensional who they basically spawned out of) - in tilting risk towards Small Cap Values in ways that increase expected returns.
Postings are my opinion, not financial advice. | Taxable/CC Rewards: VTI-70 | VXUS-30 | IRA: AVGE-100 | Tax Deferred: FXAIX-40 | VSIIX-10 | VTIAX-30 | FXNAX-20
User avatar
drumboy256
Posts: 673
Joined: Sat Jun 06, 2020 2:21 pm

Re: Avantis Funds: what's so special about them?

Post by drumboy256 »

ElGalloRico wrote: Wed Aug 03, 2022 3:11 pm
Ed 2 wrote: Fri Jul 01, 2022 10:32 pm
JSPECO9 wrote: Fri Jul 01, 2022 10:24 pm
Ed 2 wrote: Fri Jul 01, 2022 10:18 pm
JSPECO9 wrote: Fri Jul 01, 2022 10:03 pm What's the big difference between Vanguard funds and Avantis Funds? I was listening to the Bogleheads podcast and heard the interview with Eduardo Repetto. He was saying all the right things but wanted to check with the Bogleheads first.

He was saying how index funds are great but a defect they have is that they have to rush to buy when there's a new inclusion to the index? Leading to buying overpriced stocks. And how Avantis funds don't do this?

What's the big difference between funds like VTI vs AVUS. Or AVUV (which I keep seeing here repeatedly) vs VIOV/VBR. Is there a Bogleheads consensus on Avantis funds? I only ask because I keep seeing them mentioned on this forum.
Simple explanation. It’s like if your Toyota Corolla instead of four wheels have six.
Can you explain? Does this mean Avantis funds has something Vanguard funds don't have?

Thanks
They buy more small value companies than Vanguard Total Stock Market Index fund has , BUT their funds has higher expense ratio and high turnover ratio , plus I don’t see benefits investing in more small companies for the long term . This funds are more volatile and not everyone will handle such volatility. They claim they have “ scientific “ prove that over long term small value companies will outperform large and so on . To me personally I’d prefer go with the market and happy with results.
Hello,

Actually, Small Cap Value destroys Total Market and Large Cap - and it's not even close. They introduce more volatility and the last decade has shifted significantly more towards Large Cap, but it's hard to ignore the history. These things being, to a degree, cyclical, it is not unreasonable to think that Value stocks will reward the risk premium again.

While I personally maintain the core of my portfolio on VTI, I do use Avantis International and US small cap value ETF's instead of something like VBR which has a Core tilt.

Therein lies what to me, is the potential of Avantis (following Dimensional who they basically spawned out of) - in tilting risk towards Small Cap Values in ways that increase expected returns.
Piggy backing on that, Value, as a holistic viewpoint of small cap and even large cap propositions, have ironically been stronger than total market because of the spread of investment of industries. Fidelity has been yelling at me that I'm underweight Technology by almost 19% (which, it doesn't have the intelligence to the fact that I'm running a mHFEA in my accounts) so I think capturing the total market is within the means. That said, Avantis and even DFA to an extent, really offer a great way to passively track factored areas of the market for a fraction of the price.
Promise is one thing. Fulfilling that promise is quite another. - Sir Alex Ferguson | 20% IVV / 40% IBIT / 20% IXUS / 20% VGLT + chill
Ed 2
Posts: 2692
Joined: Sat May 15, 2010 9:34 am

Re: Avantis Funds: what's so special about them?

Post by Ed 2 »

drumboy256 wrote: Thu Aug 04, 2022 11:28 am
ElGalloRico wrote: Wed Aug 03, 2022 3:11 pm
Ed 2 wrote: Fri Jul 01, 2022 10:32 pm
JSPECO9 wrote: Fri Jul 01, 2022 10:24 pm
Ed 2 wrote: Fri Jul 01, 2022 10:18 pm

Simple explanation. It’s like if your Toyota Corolla instead of four wheels have six.
Can you explain? Does this mean Avantis funds has something Vanguard funds don't have?

Thanks
They buy more small value companies than Vanguard Total Stock Market Index fund has , BUT their funds has higher expense ratio and high turnover ratio , plus I don’t see benefits investing in more small companies for the long term . This funds are more volatile and not everyone will handle such volatility. They claim they have “ scientific “ prove that over long term small value companies will outperform large and so on . To me personally I’d prefer go with the market and happy with results.
Hello,

Actually, Small Cap Value destroys Total Market and Large Cap - and it's not even close. They introduce more volatility and the last decade has shifted significantly more towards Large Cap, but it's hard to ignore the history. These things being, to a degree, cyclical, it is not unreasonable to think that Value stocks will reward the risk premium again.

While I personally maintain the core of my portfolio on VTI, I do use Avantis International and US small cap value ETF's instead of something like VBR which has a Core tilt.

Therein lies what to me, is the potential of Avantis (following Dimensional who they basically spawned out of) - in tilting risk towards Small Cap Values in ways that increase expected returns.
Piggy backing on that, Value, as a holistic viewpoint of small cap and even large cap propositions, have ironically been stronger than total market because of the spread of investment of industries. Fidelity has been yelling at me that I'm underweight Technology by almost 19% (which, it doesn't have the intelligence to the fact that I'm running a mHFEA in my accounts) so I think capturing the total market is within the means. That said, Avantis and even DFA to an extent, really offer a great way to passively track factored areas of the market for a fraction of the price.
As older I get as more I realise that Jack Bogle’s teaching about simplicity when you investing long term is crucial for success. There is theory and there is a real life where many things can happen . Three funds portfolio is the ultimate to have returns close what most people need. Those who trying to “ beat the market “, pretending or believe that they know more than collective wisdom of the market eventually will realise otherwise. But what do I know. Time will tell. ;)
Cheers, Ed.
"The fund industry doesn't have a lot of heroes, but he (Bogle) is one of them," Russ Kinnel
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Avantis Funds: what's so special about them?

Post by vineviz »

typical.investor wrote: Sun Jul 03, 2022 1:04 am
What is perplexing to me, is why the significant positive loading on momentum in the AQR model disappears when using HML instead of HML-DEV. If factor returns are actually independent, why does changing the HML definition change MOM so much?
The "independence" of factors shouldn't (and doesn't) imply that all the factors have perfectly zero correlations, nor does it imply that every possible method of constructing the factor-mimicking portfolios is equally effective.

One goal in asset pricing model construction is to design factor specifications which minimize autocorrelations, but total elimination generally isn't possible.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Gaston
Posts: 1220
Joined: Wed Aug 21, 2013 7:12 pm

Re: Avantis Funds: what's so special about them?

Post by Gaston »

ElGalloRico wrote: Wed Aug 03, 2022 3:11 pm Actually, Small Cap Value destroys Total Market and Large Cap - and it's not even close.
My apologies for the pedantic comment, but I would change the tense of the above verb to destroyed, because the data refer to what happened in the past. If French, Fama and others are correct, the outperformance of small cap value will repeat itself. But whether it does so, and whether it does so within a time span that is meaningful to our lives, only time will tell.
“My opinions are just that - opinions.”
Apathizer
Posts: 2507
Joined: Sun Sep 26, 2021 2:56 pm

Re: Avantis Funds: what's so special about them?

Post by Apathizer »

Gaston wrote: Sun Aug 07, 2022 8:37 am
ElGalloRico wrote: Wed Aug 03, 2022 3:11 pm Actually, Small Cap Value destroys Total Market and Large Cap - and it's not even close.
My apologies for the pedantic comment, but I would change the tense of the above verb to destroyed, because the data refer to what happened in the past. If French, Fama and others are correct, the outperformance of small cap value will repeat itself. But whether it does so, and whether it does so within a time span that is meaningful to our lives, only time will tell.
Exactly. To me it seems likely not only based on historical returns, but also for theoretical and behavioral reasons.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
ElGalloRico
Posts: 20
Joined: Wed Aug 03, 2022 3:00 pm

Re: Avantis Funds: what's so special about them?

Post by ElGalloRico »

Apathizer wrote: Sun Aug 07, 2022 1:55 pm
Gaston wrote: Sun Aug 07, 2022 8:37 am
ElGalloRico wrote: Wed Aug 03, 2022 3:11 pm Actually, Small Cap Value destroys Total Market and Large Cap - and it's not even close.
My apologies for the pedantic comment, but I would change the tense of the above verb to destroyed, because the data refer to what happened in the past. If French, Fama and others are correct, the outperformance of small cap value will repeat itself. But whether it does so, and whether it does so within a time span that is meaningful to our lives, only time will tell.
Exactly. To me it seems likely not only based on historical returns, but also for theoretical and behavioral reasons.
Sure, but you have to basically go to the last decade or so for SVC to "under-perform" the Large Cap growth we have been witnessing.

That said, you are right Gaston - Destroyed :happy

I lean to agree with Apathizer thou.
Postings are my opinion, not financial advice. | Taxable/CC Rewards: VTI-70 | VXUS-30 | IRA: AVGE-100 | Tax Deferred: FXAIX-40 | VSIIX-10 | VTIAX-30 | FXNAX-20
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

Northern Flicker wrote: Sat Jul 02, 2022 1:45 am
DanFFA wrote: The differentiator in my mind between the factor portfolios of Vanguard and Avantis & Alpha Architect, DFA, etc. is that Vanguard is following indexes that largely track funds composed by committees. While this is largely passive, there is some active bent inherent to that committee (minor but it exists, think of how Tesla wasn't allowed in the S&P 500 for a long time and sometimes unprofitable companies take a long time to leave the S&P 500 for better or worse). On the other hand, the quant funds, while 'active' by legal classification for tax purposes, are grounded in systematic, rules-based trading methodologies with robust academic backing while also considering practical trading costs.
You are describing the S&P500, and it is not really a correct description of that. The committee is not making active stock-picking decisions. But rather than derail the thread by diving into that topic, I’ll mention that this is not the process CRSP uses. It is a rule-based process with randomized bands to make index changes unpredictable. VBR does not have the deep factor loadings that AVUV has, but it tracks a well managed index.

I think the benefit of AVUV or VFVA is that they can reconstitute at any time instead of on a quarterly schedule. Using your definition, that could be construed as a very minor active decision that VBR lacks, but I don’t think it matters.

The main benefits of VBR are lower cost and the transparency of tracking an index managed by a 3rd party.
Am I understanding you correctly that CRSP has a randomized element in their process, while the S&P500 has not? Do you happen to know how the different processes translate into different drags from front-running?
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

Apathizer wrote: Sat Jul 02, 2022 2:32 pm
Ed 2 wrote: Sat Jul 02, 2022 2:19 pm
Northern Flicker wrote: Sat Jul 02, 2022 2:01 pm
Ed 2 wrote: Sat Jul 02, 2022 10:28 am It seams ones an a while Wall Street comes out with new shiny objects . It’s up to you to go simple and rational way to invest or try to “ beat the market “. I am not arguing that Avantis doing something wrong or ripping off investors. As above post this are actively traded funds and this is your choice. A year ago this humble forum was flooded with discussion about ARK funds , some time ago about crypto , before than about investing in gold ....
Factor strategies are based on research on systematic risk factors of stocks. They are not analogous to crypto or ARK funds that make concentrated sector bets and managed concentrated active portfolios.

What is true is that the finance industry marketing machines regularly push what has been hot recently, encouraging listeners/readers to go look up the product to confirm how great it is. They rarely say that the made the recommendation before the hot runup, because generally they didn’t do so.

Factor investing is not a short-term endeavor. It requires the resolve to stick with it through long periods of underperformance. I do think the benefits of factor investing tend to be a bit oversold by factor proponents.
Yes, yes and yes! I agree with you . The only my personal belief that most of us are humans with tendencies to make wrong investments decision wrong time. This funds are high volatility funds , majority can’t stand high volatility for a long time . Plus all this” scientific “ research makes me a little smile. There are many unpredictable things during investment period . In the beginning of this year majority of vanguard investors were predicting many things what may come , only small minority was worried about geopolitics and it’s important on the market.
They only might be more volatile in the very short-term compared to a cap weight index. Remember, the market is a factor itself, so in terms of return sources you're actually more diversified with light-moderate factor slant. This seems likely to reduce overall risk-adjusted returns by adding more potential return sources.

For instance, if you care the year to year returns of AVUS and VTI, I would argue AVUS has been more consistent. In 2020 VTI out-performed since the pandemic lock-downs benefited large growth companies that are heavily weighted in VTI. AVUS still did just fine since it has a fair amount of market exposure. For the last two years as large growth has tanked, AVUS has out-performed VTI.

Again, this is exactly what I was hoping for. I don't expect factors to always out-perform the market, but improve consistency.
https://www.portfoliovisualizer.com/bac ... ion2_2=100
This has been speculated about in many other threads. But as long as it is not even clear or agreed on if any of the factors have positive or negative expected returns in the future, this entire theory of independence between factors may or may not collapse. You could compare any data mined subset of the market given any criteria, to the market itself, and whenever that subset out- or underperformed in the past, you discovered a "factor", and that factor will likely have peaks and troughs different from the peaks and troughs of the market. As long as it is not clear that the futures is similar to the past in relation to the "factor", the relative independence of the "factor" returns to the market returns gives you exactly nothing for your asset allocation.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

grabiner wrote: Sat Jul 02, 2022 5:04 pm
DanFFA wrote: Fri Jul 01, 2022 11:06 pm The differentiator in my mind between the factor portfolios of Vanguard and Avantis & Alpha Architect, DFA, etc. is that Vanguard is following indexes that largely track funds composed by committees. While this is largely passive, there is some active bent inherent to that committee (minor but it exists, think of how Tesla wasn't allowed in the S&P 500 for a long time and sometimes unprofitable companies take a long time to leave the S&P 500 for better or worse). On the other hand, the quant funds, while 'active' by legal classification for tax purposes, are grounded in systematic, rules-based trading methodologies with robust academic backing while also considering practical trading costs.
Vanguard has some of its own factor funds. You can get US value factor exposure at a slightly lower cost with VFVA (0.13% expenses) rather than a 50-50 split of AVLV/AVUV (0.15% and 0.25%). This is my choice for US stock. I use Avantis for foreign stock because Vanguard doesn't have a value ETF option there at all.
Quick question: You are saving 0.07% with Vanguard, assuming a 50/50 split between small and large. But if there is even a relatively small chance that the more sophisticated screens of Avantis, and/or the fact that they don't have to religiously follow a 3rd party index, results in a relatively small outperformance, then your 0.07% will be dwarfed, or not? What I'm saying, isn't the likely risk/return better with Avantis than with Vanguard? The max you lose is 0.07% (unless you expect the additional screens to have negative impact in the long run).
Not to mention the stronger tilt of Avantis, that may result in more efficient portfolio allocation.
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Avantis Funds: what's so special about them?

Post by Nathan Drake »

comeinvest wrote: Sun Sep 11, 2022 2:55 am
Apathizer wrote: Sat Jul 02, 2022 2:32 pm
Ed 2 wrote: Sat Jul 02, 2022 2:19 pm
Northern Flicker wrote: Sat Jul 02, 2022 2:01 pm
Ed 2 wrote: Sat Jul 02, 2022 10:28 am It seams ones an a while Wall Street comes out with new shiny objects . It’s up to you to go simple and rational way to invest or try to “ beat the market “. I am not arguing that Avantis doing something wrong or ripping off investors. As above post this are actively traded funds and this is your choice. A year ago this humble forum was flooded with discussion about ARK funds , some time ago about crypto , before than about investing in gold ....
Factor strategies are based on research on systematic risk factors of stocks. They are not analogous to crypto or ARK funds that make concentrated sector bets and managed concentrated active portfolios.

What is true is that the finance industry marketing machines regularly push what has been hot recently, encouraging listeners/readers to go look up the product to confirm how great it is. They rarely say that the made the recommendation before the hot runup, because generally they didn’t do so.

Factor investing is not a short-term endeavor. It requires the resolve to stick with it through long periods of underperformance. I do think the benefits of factor investing tend to be a bit oversold by factor proponents.
Yes, yes and yes! I agree with you . The only my personal belief that most of us are humans with tendencies to make wrong investments decision wrong time. This funds are high volatility funds , majority can’t stand high volatility for a long time . Plus all this” scientific “ research makes me a little smile. There are many unpredictable things during investment period . In the beginning of this year majority of vanguard investors were predicting many things what may come , only small minority was worried about geopolitics and it’s important on the market.
They only might be more volatile in the very short-term compared to a cap weight index. Remember, the market is a factor itself, so in terms of return sources you're actually more diversified with light-moderate factor slant. This seems likely to reduce overall risk-adjusted returns by adding more potential return sources.

For instance, if you care the year to year returns of AVUS and VTI, I would argue AVUS has been more consistent. In 2020 VTI out-performed since the pandemic lock-downs benefited large growth companies that are heavily weighted in VTI. AVUS still did just fine since it has a fair amount of market exposure. For the last two years as large growth has tanked, AVUS has out-performed VTI.

Again, this is exactly what I was hoping for. I don't expect factors to always out-perform the market, but improve consistency.
https://www.portfoliovisualizer.com/bac ... ion2_2=100
This has been speculated about in many other threads. But as long as it is not even clear or agreed on if any of the factors have positive or negative expected returns in the future, this entire theory of independence between factors may or may not collapse. You could compare any data mined subset of the market given any criteria, to the market itself, and whenever that subset out- or underperformed in the past, you discovered a "factor", and that factor will likely have peaks and troughs different from the peaks and troughs of the market. As long as it is not clear that the futures is similar to the past in relation to the "factor", the relative independence of the "factor" returns to the market returns gives you exactly nothing for your asset allocation.
The same (faulty) logic could be extended to the market factor over bonds
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

Nathan Drake wrote: Sun Sep 11, 2022 3:16 am
comeinvest wrote: Sun Sep 11, 2022 2:55 am
Apathizer wrote: Sat Jul 02, 2022 2:32 pm
Ed 2 wrote: Sat Jul 02, 2022 2:19 pm
Northern Flicker wrote: Sat Jul 02, 2022 2:01 pm
Factor strategies are based on research on systematic risk factors of stocks. They are not analogous to crypto or ARK funds that make concentrated sector bets and managed concentrated active portfolios.

What is true is that the finance industry marketing machines regularly push what has been hot recently, encouraging listeners/readers to go look up the product to confirm how great it is. They rarely say that the made the recommendation before the hot runup, because generally they didn’t do so.

Factor investing is not a short-term endeavor. It requires the resolve to stick with it through long periods of underperformance. I do think the benefits of factor investing tend to be a bit oversold by factor proponents.
Yes, yes and yes! I agree with you . The only my personal belief that most of us are humans with tendencies to make wrong investments decision wrong time. This funds are high volatility funds , majority can’t stand high volatility for a long time . Plus all this” scientific “ research makes me a little smile. There are many unpredictable things during investment period . In the beginning of this year majority of vanguard investors were predicting many things what may come , only small minority was worried about geopolitics and it’s important on the market.
They only might be more volatile in the very short-term compared to a cap weight index. Remember, the market is a factor itself, so in terms of return sources you're actually more diversified with light-moderate factor slant. This seems likely to reduce overall risk-adjusted returns by adding more potential return sources.

For instance, if you care the year to year returns of AVUS and VTI, I would argue AVUS has been more consistent. In 2020 VTI out-performed since the pandemic lock-downs benefited large growth companies that are heavily weighted in VTI. AVUS still did just fine since it has a fair amount of market exposure. For the last two years as large growth has tanked, AVUS has out-performed VTI.

Again, this is exactly what I was hoping for. I don't expect factors to always out-perform the market, but improve consistency.
https://www.portfoliovisualizer.com/bac ... ion2_2=100
This has been speculated about in many other threads. But as long as it is not even clear or agreed on if any of the factors have positive or negative expected returns in the future, this entire theory of independence between factors may or may not collapse. You could compare any data mined subset of the market given any criteria, to the market itself, and whenever that subset out- or underperformed in the past, you discovered a "factor", and that factor will likely have peaks and troughs different from the peaks and troughs of the market. As long as it is not clear that the futures is similar to the past in relation to the "factor", the relative independence of the "factor" returns to the market returns gives you exactly nothing for your asset allocation.
The same (faulty) logic could be extended to the market factor over bonds
No, as has been pointed out many times. Permanently negative expected market returns would halt the economy and by implication the country, as nobody would invest a dollar in any business for negative expected returns, and the remaining businesses would shut down.
Negative value premium or quality premium? What bad things would happen to the country? Not much, right? Some smarty-pants dudes who looked at data-mining exercises without real fundamental explanation and were betting on continued value or quality outperformance, would lose some dollars to counterparties who were overweighing growth. That's it.
User avatar
grabiner
Advisory Board
Posts: 35307
Joined: Tue Feb 20, 2007 10:58 pm
Location: Columbia, MD

Re: Avantis Funds: what's so special about them?

Post by grabiner »

comeinvest wrote: Sun Sep 11, 2022 3:02 am
grabiner wrote: Sat Jul 02, 2022 5:04 pm
DanFFA wrote: Fri Jul 01, 2022 11:06 pm The differentiator in my mind between the factor portfolios of Vanguard and Avantis & Alpha Architect, DFA, etc. is that Vanguard is following indexes that largely track funds composed by committees. While this is largely passive, there is some active bent inherent to that committee (minor but it exists, think of how Tesla wasn't allowed in the S&P 500 for a long time and sometimes unprofitable companies take a long time to leave the S&P 500 for better or worse). On the other hand, the quant funds, while 'active' by legal classification for tax purposes, are grounded in systematic, rules-based trading methodologies with robust academic backing while also considering practical trading costs.
Vanguard has some of its own factor funds. You can get US value factor exposure at a slightly lower cost with VFVA (0.13% expenses) rather than a 50-50 split of AVLV/AVUV (0.15% and 0.25%). This is my choice for US stock. I use Avantis for foreign stock because Vanguard doesn't have a value ETF option there at all.
Quick question: You are saving 0.07% with Vanguard, assuming a 50/50 split between small and large. But if there is even a relatively small chance that the more sophisticated screens of Avantis, and/or the fact that they don't have to religiously follow a 3rd party index, results in a relatively small outperformance, then your 0.07% will be dwarfed, or not? What I'm saying, isn't the likely risk/return better with Avantis than with Vanguard? The max you lose is 0.07% (unless you expect the additional screens to have negative impact in the long run).
Not to mention the stronger tilt of Avantis, that may result in more efficient portfolio allocation.
VFVA doesn't religiously follow a third-party index (as, say, VBR, Vanguard Small-Cap Value ETF, does). From its prospectus:
"The advisor uses a quantitative model to evaluate all of the securities in an investment universe comprised of U.S. large-, mid-, and small-capitalization stocks and to construct a U.S. equity portfolio that seeks to achieve exposure to securities with lower prices relative to fundamental measures of value subject to a rules-based screen designed to promote diversification and to mitigate exposure to certain less liquid stocks. Securities with lower prices relative to fundamental value may be identified by measures such as book to price and earnings to price ratios."

I don't know whether Avantis will be superior in managing the funds. However, it does appear that Avantis doesn't have a stronger value tilt; the P/E and P/B of VFVA are lower than that of the AVLV/AVUV combination, and the reported value exposure for VFVA is stronger (according to Morningstar's analysis; on ETF.com, AVLV is too new to have factor data, but VFVA has almost as strong a value factor exposure as AVUV).
Wiki David Grabiner
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Avantis Funds: what's so special about them?

Post by burritoLover »

comeinvest wrote: Sun Sep 11, 2022 4:18 am
Nathan Drake wrote: Sun Sep 11, 2022 3:16 am
comeinvest wrote: Sun Sep 11, 2022 2:55 am
Apathizer wrote: Sat Jul 02, 2022 2:32 pm
Ed 2 wrote: Sat Jul 02, 2022 2:19 pm

Yes, yes and yes! I agree with you . The only my personal belief that most of us are humans with tendencies to make wrong investments decision wrong time. This funds are high volatility funds , majority can’t stand high volatility for a long time . Plus all this” scientific “ research makes me a little smile. There are many unpredictable things during investment period . In the beginning of this year majority of vanguard investors were predicting many things what may come , only small minority was worried about geopolitics and it’s important on the market.
They only might be more volatile in the very short-term compared to a cap weight index. Remember, the market is a factor itself, so in terms of return sources you're actually more diversified with light-moderate factor slant. This seems likely to reduce overall risk-adjusted returns by adding more potential return sources.

For instance, if you care the year to year returns of AVUS and VTI, I would argue AVUS has been more consistent. In 2020 VTI out-performed since the pandemic lock-downs benefited large growth companies that are heavily weighted in VTI. AVUS still did just fine since it has a fair amount of market exposure. For the last two years as large growth has tanked, AVUS has out-performed VTI.

Again, this is exactly what I was hoping for. I don't expect factors to always out-perform the market, but improve consistency.
https://www.portfoliovisualizer.com/bac ... ion2_2=100
This has been speculated about in many other threads. But as long as it is not even clear or agreed on if any of the factors have positive or negative expected returns in the future, this entire theory of independence between factors may or may not collapse. You could compare any data mined subset of the market given any criteria, to the market itself, and whenever that subset out- or underperformed in the past, you discovered a "factor", and that factor will likely have peaks and troughs different from the peaks and troughs of the market. As long as it is not clear that the futures is similar to the past in relation to the "factor", the relative independence of the "factor" returns to the market returns gives you exactly nothing for your asset allocation.
The same (faulty) logic could be extended to the market factor over bonds
No, as has been pointed out many times. Permanently negative expected market returns would halt the economy and by implication the country, as nobody would invest a dollar in any business for negative expected returns, and the remaining businesses would shut down.
Negative value premium or quality premium? What bad things would happen to the country? Not much, right? Some smarty-pants dudes who looked at data-mining exercises without real fundamental explanation and were betting on continued value or quality outperformance, would lose some dollars to counterparties who were overweighing growth. That's it.
There are numerous examples of countries where the market risk premium was negative over many decades - even in the US (see below). Those economies did not grind to a halt.

Image
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Avantis Funds: what's so special about them?

Post by Nathan Drake »

comeinvest wrote: Sun Sep 11, 2022 4:18 am
Nathan Drake wrote: Sun Sep 11, 2022 3:16 am
comeinvest wrote: Sun Sep 11, 2022 2:55 am
Apathizer wrote: Sat Jul 02, 2022 2:32 pm
Ed 2 wrote: Sat Jul 02, 2022 2:19 pm

Yes, yes and yes! I agree with you . The only my personal belief that most of us are humans with tendencies to make wrong investments decision wrong time. This funds are high volatility funds , majority can’t stand high volatility for a long time . Plus all this” scientific “ research makes me a little smile. There are many unpredictable things during investment period . In the beginning of this year majority of vanguard investors were predicting many things what may come , only small minority was worried about geopolitics and it’s important on the market.
They only might be more volatile in the very short-term compared to a cap weight index. Remember, the market is a factor itself, so in terms of return sources you're actually more diversified with light-moderate factor slant. This seems likely to reduce overall risk-adjusted returns by adding more potential return sources.

For instance, if you care the year to year returns of AVUS and VTI, I would argue AVUS has been more consistent. In 2020 VTI out-performed since the pandemic lock-downs benefited large growth companies that are heavily weighted in VTI. AVUS still did just fine since it has a fair amount of market exposure. For the last two years as large growth has tanked, AVUS has out-performed VTI.

Again, this is exactly what I was hoping for. I don't expect factors to always out-perform the market, but improve consistency.
https://www.portfoliovisualizer.com/bac ... ion2_2=100
This has been speculated about in many other threads. But as long as it is not even clear or agreed on if any of the factors have positive or negative expected returns in the future, this entire theory of independence between factors may or may not collapse. You could compare any data mined subset of the market given any criteria, to the market itself, and whenever that subset out- or underperformed in the past, you discovered a "factor", and that factor will likely have peaks and troughs different from the peaks and troughs of the market. As long as it is not clear that the futures is similar to the past in relation to the "factor", the relative independence of the "factor" returns to the market returns gives you exactly nothing for your asset allocation.
The same (faulty) logic could be extended to the market factor over bonds
No, as has been pointed out many times. Permanently negative expected market returns would halt the economy and by implication the country, as nobody would invest a dollar in any business for negative expected returns, and the remaining businesses would shut down.
Negative value premium or quality premium? What bad things would happen to the country? Not much, right? Some smarty-pants dudes who looked at data-mining exercises without real fundamental explanation and were betting on continued value or quality outperformance, would lose some dollars to counterparties who were overweighing growth. That's it.
Stocks have gone through essentially periods of an investor's lifetime with a negative premium relative to bonds, as BurritoLover pointed out below. Economic growth essentially has no correlation to stock market returns either.

Remember that when you claim something is "data mined" that can apply to the overall stock market. Why should we invest based on market capitalization? Is that because some people data mined and found that by having this level of diversification, as weighted by the market, that you tend to receive an adequate 5-6% real return over the risk free rate over LONG periods? How is this any different than some people data mining the market looking for alternative ways to capture additional risk premiums? It's not.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

Nathan Drake wrote: Sun Sep 11, 2022 12:21 pm
comeinvest wrote: Sun Sep 11, 2022 4:18 am
Nathan Drake wrote: Sun Sep 11, 2022 3:16 am
comeinvest wrote: Sun Sep 11, 2022 2:55 am
Apathizer wrote: Sat Jul 02, 2022 2:32 pm
They only might be more volatile in the very short-term compared to a cap weight index. Remember, the market is a factor itself, so in terms of return sources you're actually more diversified with light-moderate factor slant. This seems likely to reduce overall risk-adjusted returns by adding more potential return sources.

For instance, if you care the year to year returns of AVUS and VTI, I would argue AVUS has been more consistent. In 2020 VTI out-performed since the pandemic lock-downs benefited large growth companies that are heavily weighted in VTI. AVUS still did just fine since it has a fair amount of market exposure. For the last two years as large growth has tanked, AVUS has out-performed VTI.

Again, this is exactly what I was hoping for. I don't expect factors to always out-perform the market, but improve consistency.
https://www.portfoliovisualizer.com/bac ... ion2_2=100
This has been speculated about in many other threads. But as long as it is not even clear or agreed on if any of the factors have positive or negative expected returns in the future, this entire theory of independence between factors may or may not collapse. You could compare any data mined subset of the market given any criteria, to the market itself, and whenever that subset out- or underperformed in the past, you discovered a "factor", and that factor will likely have peaks and troughs different from the peaks and troughs of the market. As long as it is not clear that the futures is similar to the past in relation to the "factor", the relative independence of the "factor" returns to the market returns gives you exactly nothing for your asset allocation.
The same (faulty) logic could be extended to the market factor over bonds
No, as has been pointed out many times. Permanently negative expected market returns would halt the economy and by implication the country, as nobody would invest a dollar in any business for negative expected returns, and the remaining businesses would shut down.
Negative value premium or quality premium? What bad things would happen to the country? Not much, right? Some smarty-pants dudes who looked at data-mining exercises without real fundamental explanation and were betting on continued value or quality outperformance, would lose some dollars to counterparties who were overweighing growth. That's it.
Stocks have gone through essentially periods of an investor's lifetime with a negative premium relative to bonds, as BurritoLover pointed out below. Economic growth essentially has no correlation to stock market returns either.

Remember that when you claim something is "data mined" that can apply to the overall stock market. Why should we invest based on market capitalization? Is that because some people data mined and found that by having this level of diversification, as weighted by the market, that you tend to receive an adequate 5-6% real return over the risk free rate over LONG periods? How is this any different than some people data mining the market looking for alternative ways to capture additional risk premiums? It's not.
I don't take sides if the factor models are data mining or have predictive capability. I'm saying that the confidence in the market premium is significantly higher than the confidence in factor premia. As we recently discussed in another thread, factor premia have essentially no explanation. All "explanations" were made up ad-hoc and are purely descriptive, a desperate attempt at "explaining" describing the historical data.

I still think that if expected real equity risk premia were negative, the economy and the country would come to a standstill. Nobody would invest a dollar into a business with the expectation that it will be worth for example $0.98 a year later. As a result, any and all innovation would be halted, businesses would die. If you disagree, tell me how the economy would work, why anybody would have any incentive to run a business. If nationalization of productive assets were imminent like during the October revolution in Russia, the economy and the country as we know it would effectively cease to exist.

I'm not sure what precisely you are trying to show with the above chart. We all know that there was an equity market drawdown in the 1929+ period, and a bond market drawdown in the 1970ies to early 1980ies because of a dramatic unexpected rise in interest rates. None of that says that expected ex-ante returns were negative, or that long-term returns were ever expected to be negative. Most major market swings in history were to a large extent due to valuation changes, meaning expected returns arguably moved opposite to the market, if anything.

Another thought experiment: The predictive capability (or lack of it) of factors is heavily debated in this forum. But regardless if a risk-adjusted outperformance could be achieved with factors or not, I think most participants in this forum would be happy to invest 100% in the stock market index with no factor tilts (e.g. if "factors" or factor tilts didn't exist). On the other hand, how many Bogleheads would be confident investing 100% of their lifetime retirement savings in a long/short factor fund with no market exposure? I wouldn't.
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Avantis Funds: what's so special about them?

Post by Nathan Drake »

comeinvest wrote: Sun Sep 11, 2022 4:55 pm
Nathan Drake wrote: Sun Sep 11, 2022 12:21 pm
comeinvest wrote: Sun Sep 11, 2022 4:18 am
Nathan Drake wrote: Sun Sep 11, 2022 3:16 am
comeinvest wrote: Sun Sep 11, 2022 2:55 am

This has been speculated about in many other threads. But as long as it is not even clear or agreed on if any of the factors have positive or negative expected returns in the future, this entire theory of independence between factors may or may not collapse. You could compare any data mined subset of the market given any criteria, to the market itself, and whenever that subset out- or underperformed in the past, you discovered a "factor", and that factor will likely have peaks and troughs different from the peaks and troughs of the market. As long as it is not clear that the futures is similar to the past in relation to the "factor", the relative independence of the "factor" returns to the market returns gives you exactly nothing for your asset allocation.
The same (faulty) logic could be extended to the market factor over bonds
No, as has been pointed out many times. Permanently negative expected market returns would halt the economy and by implication the country, as nobody would invest a dollar in any business for negative expected returns, and the remaining businesses would shut down.
Negative value premium or quality premium? What bad things would happen to the country? Not much, right? Some smarty-pants dudes who looked at data-mining exercises without real fundamental explanation and were betting on continued value or quality outperformance, would lose some dollars to counterparties who were overweighing growth. That's it.
Stocks have gone through essentially periods of an investor's lifetime with a negative premium relative to bonds, as BurritoLover pointed out below. Economic growth essentially has no correlation to stock market returns either.

Remember that when you claim something is "data mined" that can apply to the overall stock market. Why should we invest based on market capitalization? Is that because some people data mined and found that by having this level of diversification, as weighted by the market, that you tend to receive an adequate 5-6% real return over the risk free rate over LONG periods? How is this any different than some people data mining the market looking for alternative ways to capture additional risk premiums? It's not.
I don't take sides if the factor models are data mining or have predictive capability. I'm saying that the confidence in the market premium is significantly higher than the confidence in factor premia. As we recently discussed in another thread, factor premia have essentially no explanation. All "explanations" were made up ad-hoc and are purely descriptive, a desperate attempt at "explaining" describing the historical data.

I still think that if expected real equity risk premia were negative, the economy and the country would come to a standstill. Nobody would invest a dollar into a business with the expectation that it will be worth for example $0.98 a year later. As a result, any and all innovation would be halted, businesses would die. If you disagree, tell me how the economy would work, why anybody would have any incentive to run a business. If nationalization of productive assets were imminent like during the October revolution in Russia, the economy and the country as we know it would effectively cease to exist.

I'm not sure what precisely you are trying to show with the above chart. We all know that there was an equity market drawdown in the 1929+ period, and a bond market drawdown in the 1970ies to early 1980ies because of a dramatic unexpected rise in interest rates. None of that says that expected ex-ante returns were negative, or that long-term returns were ever expected to be negative. Most major market swings in history were to a large extent due to valuation changes, meaning expected returns arguably moved opposite to the market, if anything.

Another thought experiment: The predictive capability (or lack of it) of factors is heavily debated in this forum. But regardless if a risk-adjusted outperformance could be achieved with factors or not, I think most participants in this forum would be happy to invest 100% in the stock market index with no factor tilts (e.g. if "factors" or factor tilts didn't exist). On the other hand, how many Bogleheads would be confident investing 100% of their lifetime retirement savings in a long/short factor fund with no market exposure? I wouldn't.
I don't think anyone is necessarily disagreeing with the idea that the market premium may have more confidence. After all, it explains 60% of returns. But the other 40% of returns that are explained by additional factors are not insignificant.

There are plenty of rational explanations for factors. The idea that these were made up "after" they were discovered seems like it's not much different than equities. Back in the early 1900s and prior, stocks did not have an explanation. In fact bonds were seen as just as risky as stocks. What changed? Maybe markets matured, in other words they started pricing the risk of stocks (versus bonds) better. But, like the factor explanation, it's just something theorized AFTER THE FACT. That doesn't mean it's foundation is weak.

If the expected risk premium is negative, it means that equities are priced way out of balance relative to other assets. This usually coincides with a euphoric rise in valuations and "this time is different" type thinking where stocks will always outperform now and forever. Markets don't work like that. There's constant re-pricing to adjust for newly perceived risks. That's why these things go through cycles of underperformance. If there were not periods of underperformance with negative premiums, there would be no positive premiums to begin with.

So let's look at 1966-1982 - an example where starting valuations were eurphoric. What brought massive inflows into stocks brought investors great pleasure. Valuations became stretched. Geopolitical and inflationary risks showed up. Valuations dropped. The equity risk premium went negative. This environment brought investors great pain. Nobody wanted to invest in a market after it had a significant re-valuation. But that re-pricing to account for a higher risk premium is precisely why it became such a good time to invest. Markets aren't static. Eventually investors will demand more return for the risk they are taking on.

The thought experiment doesn't really hold any significance to me. It's comparing apples to oranges. You could make the same thought experiment about going long total stock market and short bonds. Investment professionals like DFA, etc. don't advocate going 100 long/short because the short side of things can really hurt performance if you happen to go through a long cycle of underperformance. A long only portfolio doesn't have such risk at all. And there are many people that are comfortable being nearly 100% SCV (long only).
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
Apathizer
Posts: 2507
Joined: Sun Sep 26, 2021 2:56 pm

Re: Avantis Funds: what's so special about them?

Post by Apathizer »

comeinvest wrote: Sun Sep 11, 2022 4:55 pm Another thought experiment: The predictive capability (or lack of it) of factors is heavily debated in this forum. But regardless if a risk-adjusted outperformance could be achieved with factors or not, I think most participants in this forum would be happy to invest 100% in the stock market index with no factor tilts (e.g. if "factors" or factor tilts didn't exist). On the other hand, how many Bogleheads would be confident investing 100% of their lifetime retirement savings in a long/short factor fund with no market exposure? I wouldn't.
It's heavily debated here, but there still seems to be enough consensus among academics that factors have delivered a positive risk premium. The issue is whether this will continue. This was discussed recently on the Rational Reminder podcast. The academic guesstimate is that as more investors employ factor slants their premium will some degree of arbitrage decrease, but they will still persist to some extent. Considering how easy it is now to construct a factor-slanted portfolio this approach seems reasonable, though yes, a cap weight total market portfolio is also fine.

The highlighted portion seems puzzling and hyperbolic. Since factors are components of the market, how exactly it possible to invest in them without market exposure? That doesn't make sense. Also, few if any of us here are arguing for massive slants in one or a few factors. Most of are arguing for a total market approach with some degree of factor slant. This increases diversification across risk factors, which increases likelihood of higher returns and consistency, though there's also risk of under-performance. If slants are light-moderate that possible under-performance will be fairly modest.
https://www.pwlcapital.com/five-factor- ... with-etfs/
Last edited by Apathizer on Sun Sep 11, 2022 11:38 pm, edited 1 time in total.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Avantis Funds: what's so special about them?

Post by vineviz »

comeinvest wrote: Sun Sep 11, 2022 4:55 pm I don't take sides if the factor models are data mining or have predictive capability. I'm saying that the confidence in the market premium is significantly higher than the confidence in factor premia.
If that's your opinion, that's fine. Just know that there's not really any evidence to back it up.
comeinvest wrote: Sun Sep 11, 2022 4:55 pm I still think that if expected real equity risk premia were negative, the economy and the country would come to a standstill.
No one would EVER invest in a risky asset if the expected risk premium was negative. That's not unique to the market risk factor.
comeinvest wrote: Sun Sep 11, 2022 4:55 pm
On the other hand, how many Bogleheads would be confident investing 100% of their lifetime retirement savings in a long/short factor fund with no market exposure? I wouldn't.
The equity risk premium is, itself, a long-short portfolio: long stocks and short the risk-free asset.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

vineviz wrote: Sun Sep 11, 2022 7:37 pm
comeinvest wrote: Sun Sep 11, 2022 4:55 pm I don't take sides if the factor models are data mining or have predictive capability. I'm saying that the confidence in the market premium is significantly higher than the confidence in factor premia.
If that's your opinion, that's fine. Just know that there's not really any evidence to back it up.
comeinvest wrote: Sun Sep 11, 2022 4:55 pm I still think that if expected real equity risk premia were negative, the economy and the country would come to a standstill.
No one would EVER invest in a risky asset if the expected risk premium was negative. That's not unique to the market risk factor.
comeinvest wrote: Sun Sep 11, 2022 4:55 pm
On the other hand, how many Bogleheads would be confident investing 100% of their lifetime retirement savings in a long/short factor fund with no market exposure? I wouldn't.
The equity risk premium is, itself, a long-short portfolio: long stocks and short the risk-free asset.
With none of your 3 replies you are engaging only to the slightest with my analysis, my comments, or with my thought experiments.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

Nathan Drake wrote: Sun Sep 11, 2022 6:23 pm
comeinvest wrote: Sun Sep 11, 2022 4:55 pm
Nathan Drake wrote: Sun Sep 11, 2022 12:21 pm
comeinvest wrote: Sun Sep 11, 2022 4:18 am
Nathan Drake wrote: Sun Sep 11, 2022 3:16 am

The same (faulty) logic could be extended to the market factor over bonds
No, as has been pointed out many times. Permanently negative expected market returns would halt the economy and by implication the country, as nobody would invest a dollar in any business for negative expected returns, and the remaining businesses would shut down.
Negative value premium or quality premium? What bad things would happen to the country? Not much, right? Some smarty-pants dudes who looked at data-mining exercises without real fundamental explanation and were betting on continued value or quality outperformance, would lose some dollars to counterparties who were overweighing growth. That's it.
Stocks have gone through essentially periods of an investor's lifetime with a negative premium relative to bonds, as BurritoLover pointed out below. Economic growth essentially has no correlation to stock market returns either.

Remember that when you claim something is "data mined" that can apply to the overall stock market. Why should we invest based on market capitalization? Is that because some people data mined and found that by having this level of diversification, as weighted by the market, that you tend to receive an adequate 5-6% real return over the risk free rate over LONG periods? How is this any different than some people data mining the market looking for alternative ways to capture additional risk premiums? It's not.
I don't take sides if the factor models are data mining or have predictive capability. I'm saying that the confidence in the market premium is significantly higher than the confidence in factor premia. As we recently discussed in another thread, factor premia have essentially no explanation. All "explanations" were made up ad-hoc and are purely descriptive, a desperate attempt at "explaining" describing the historical data.

I still think that if expected real equity risk premia were negative, the economy and the country would come to a standstill. Nobody would invest a dollar into a business with the expectation that it will be worth for example $0.98 a year later. As a result, any and all innovation would be halted, businesses would die. If you disagree, tell me how the economy would work, why anybody would have any incentive to run a business. If nationalization of productive assets were imminent like during the October revolution in Russia, the economy and the country as we know it would effectively cease to exist.

I'm not sure what precisely you are trying to show with the above chart. We all know that there was an equity market drawdown in the 1929+ period, and a bond market drawdown in the 1970ies to early 1980ies because of a dramatic unexpected rise in interest rates. None of that says that expected ex-ante returns were negative, or that long-term returns were ever expected to be negative. Most major market swings in history were to a large extent due to valuation changes, meaning expected returns arguably moved opposite to the market, if anything.

Another thought experiment: The predictive capability (or lack of it) of factors is heavily debated in this forum. But regardless if a risk-adjusted outperformance could be achieved with factors or not, I think most participants in this forum would be happy to invest 100% in the stock market index with no factor tilts (e.g. if "factors" or factor tilts didn't exist). On the other hand, how many Bogleheads would be confident investing 100% of their lifetime retirement savings in a long/short factor fund with no market exposure? I wouldn't.
I don't think anyone is necessarily disagreeing with the idea that the market premium may have more confidence. After all, it explains 60% of returns. But the other 40% of returns that are explained by additional factors are not insignificant.

There are plenty of rational explanations for factors. The idea that these were made up "after" they were discovered seems like it's not much different than equities. Back in the early 1900s and prior, stocks did not have an explanation. In fact bonds were seen as just as risky as stocks. What changed? Maybe markets matured, in other words they started pricing the risk of stocks (versus bonds) better. But, like the factor explanation, it's just something theorized AFTER THE FACT. That doesn't mean it's foundation is weak.

If the expected risk premium is negative, it means that equities are priced way out of balance relative to other assets. This usually coincides with a euphoric rise in valuations and "this time is different" type thinking where stocks will always outperform now and forever. Markets don't work like that. There's constant re-pricing to adjust for newly perceived risks. That's why these things go through cycles of underperformance. If there were not periods of underperformance with negative premiums, there would be no positive premiums to begin with.

So let's look at 1966-1982 - an example where starting valuations were eurphoric. What brought massive inflows into stocks brought investors great pleasure. Valuations became stretched. Geopolitical and inflationary risks showed up. Valuations dropped. The equity risk premium went negative. This environment brought investors great pain. Nobody wanted to invest in a market after it had a significant re-valuation. But that re-pricing to account for a higher risk premium is precisely why it became such a good time to invest. Markets aren't static. Eventually investors will demand more return for the risk they are taking on.

The thought experiment doesn't really hold any significance to me. It's comparing apples to oranges. You could make the same thought experiment about going long total stock market and short bonds. Investment professionals like DFA, etc. don't advocate going 100 long/short because the short side of things can really hurt performance if you happen to go through a long cycle of underperformance. A long only portfolio doesn't have such risk at all. And there are many people that are comfortable being nearly 100% SCV (long only).
Most of your paragraphs just re-phrase what I was said before, except your dismissal of my thought experiments. You didn't address it. To be clear, by "expected premium" I always meant just that: *expected*, i.e. a priori premium. All your paragraphs about historical drawdowns describe realized premia, not expected premia. I pointed out myself in my last few comments that realized premia can and always will be negative at times, of course - markets go up and down. What does this have to do with anything. My thought experiment is about when *expected* long-term market term premia are negative. I believe that the economy and the country as we know it would cease to exist, for the reasons that I cited: Neither you nor I would invest a dollar in any business with the expectation to get back $0.98 in real terms a year later, or maybe $0.50 in real terms 30 years later. In contrast to the newer (possibly data mined - the jury is still out) "factors", which would have near zero effect if they were negative starting now for the remaining existence of humanity, except for some folks who were betting on the continuation of historical patterns. I feel safer putting my eggs in the basket that assumes the economy and the country as we know it will survive just a little longer; I think if it does not, then we have bigger problems than our investment portfolios.

Let's do some straight talk. Equity market returns come from the profits of businesses in the real economy (adjusted for valuation changes and "slippage" from new competing capital, of course). "Factor" returns are mathematical artifacts that showed up in some descriptive analyses of historical returns, and that must mathematically net to zero (e.g. value vs. growth investors). Fundamentally different concepts, I think.
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Avantis Funds: what's so special about them?

Post by Nathan Drake »

comeinvest wrote: Sun Sep 11, 2022 9:17 pm
Nathan Drake wrote: Sun Sep 11, 2022 6:23 pm
comeinvest wrote: Sun Sep 11, 2022 4:55 pm
Nathan Drake wrote: Sun Sep 11, 2022 12:21 pm
comeinvest wrote: Sun Sep 11, 2022 4:18 am

No, as has been pointed out many times. Permanently negative expected market returns would halt the economy and by implication the country, as nobody would invest a dollar in any business for negative expected returns, and the remaining businesses would shut down.
Negative value premium or quality premium? What bad things would happen to the country? Not much, right? Some smarty-pants dudes who looked at data-mining exercises without real fundamental explanation and were betting on continued value or quality outperformance, would lose some dollars to counterparties who were overweighing growth. That's it.
Stocks have gone through essentially periods of an investor's lifetime with a negative premium relative to bonds, as BurritoLover pointed out below. Economic growth essentially has no correlation to stock market returns either.

Remember that when you claim something is "data mined" that can apply to the overall stock market. Why should we invest based on market capitalization? Is that because some people data mined and found that by having this level of diversification, as weighted by the market, that you tend to receive an adequate 5-6% real return over the risk free rate over LONG periods? How is this any different than some people data mining the market looking for alternative ways to capture additional risk premiums? It's not.
I don't take sides if the factor models are data mining or have predictive capability. I'm saying that the confidence in the market premium is significantly higher than the confidence in factor premia. As we recently discussed in another thread, factor premia have essentially no explanation. All "explanations" were made up ad-hoc and are purely descriptive, a desperate attempt at "explaining" describing the historical data.

I still think that if expected real equity risk premia were negative, the economy and the country would come to a standstill. Nobody would invest a dollar into a business with the expectation that it will be worth for example $0.98 a year later. As a result, any and all innovation would be halted, businesses would die. If you disagree, tell me how the economy would work, why anybody would have any incentive to run a business. If nationalization of productive assets were imminent like during the October revolution in Russia, the economy and the country as we know it would effectively cease to exist.

I'm not sure what precisely you are trying to show with the above chart. We all know that there was an equity market drawdown in the 1929+ period, and a bond market drawdown in the 1970ies to early 1980ies because of a dramatic unexpected rise in interest rates. None of that says that expected ex-ante returns were negative, or that long-term returns were ever expected to be negative. Most major market swings in history were to a large extent due to valuation changes, meaning expected returns arguably moved opposite to the market, if anything.

Another thought experiment: The predictive capability (or lack of it) of factors is heavily debated in this forum. But regardless if a risk-adjusted outperformance could be achieved with factors or not, I think most participants in this forum would be happy to invest 100% in the stock market index with no factor tilts (e.g. if "factors" or factor tilts didn't exist). On the other hand, how many Bogleheads would be confident investing 100% of their lifetime retirement savings in a long/short factor fund with no market exposure? I wouldn't.
I don't think anyone is necessarily disagreeing with the idea that the market premium may have more confidence. After all, it explains 60% of returns. But the other 40% of returns that are explained by additional factors are not insignificant.

There are plenty of rational explanations for factors. The idea that these were made up "after" they were discovered seems like it's not much different than equities. Back in the early 1900s and prior, stocks did not have an explanation. In fact bonds were seen as just as risky as stocks. What changed? Maybe markets matured, in other words they started pricing the risk of stocks (versus bonds) better. But, like the factor explanation, it's just something theorized AFTER THE FACT. That doesn't mean it's foundation is weak.

If the expected risk premium is negative, it means that equities are priced way out of balance relative to other assets. This usually coincides with a euphoric rise in valuations and "this time is different" type thinking where stocks will always outperform now and forever. Markets don't work like that. There's constant re-pricing to adjust for newly perceived risks. That's why these things go through cycles of underperformance. If there were not periods of underperformance with negative premiums, there would be no positive premiums to begin with.

So let's look at 1966-1982 - an example where starting valuations were eurphoric. What brought massive inflows into stocks brought investors great pleasure. Valuations became stretched. Geopolitical and inflationary risks showed up. Valuations dropped. The equity risk premium went negative. This environment brought investors great pain. Nobody wanted to invest in a market after it had a significant re-valuation. But that re-pricing to account for a higher risk premium is precisely why it became such a good time to invest. Markets aren't static. Eventually investors will demand more return for the risk they are taking on.

The thought experiment doesn't really hold any significance to me. It's comparing apples to oranges. You could make the same thought experiment about going long total stock market and short bonds. Investment professionals like DFA, etc. don't advocate going 100 long/short because the short side of things can really hurt performance if you happen to go through a long cycle of underperformance. A long only portfolio doesn't have such risk at all. And there are many people that are comfortable being nearly 100% SCV (long only).
Most of your paragraphs just re-phrase what I was said before, except your dismissal of my thought experiments. You didn't address it. To be clear, by "expected premium" I always meant just that: *expected*, i.e. a priori premium. All your paragraphs about historical drawdowns describe realized premia, not expected premia. I pointed out myself in my last few comments that realized premia can and always will be negative at times, of course - markets go up and down. What does this have to do with anything. My thought experiment is about when *expected* long-term market term premia are negative. I believe that the economy and the country as we know it would cease to exist, for the reasons that I cited: Neither you nor I would invest a dollar in any business with the expectation to get back $0.98 in real terms a year later, or maybe $0.50 in real terms 30 years later. In contrast to the newer (possibly data mined - the jury is still out) "factors", which would have near zero effect if they were negative starting now for the remaining existence of humanity, except for folks who were betting on the continuation of historical patterns.

Let's do some straight talk. Equity market returns come from the profits of businesses in the real economy (adjusted for valuation changes and "slippage" from new competing capital, of course). "Factor" returns are mathematical artifacts that showed up in some descriptive analyses of historical returns, and that must mathematically net to zero (e.g. value vs. growth investors). Fundamentally different concepts, I think.
I did address your thought experiment. The market is a factor above the risk free rate. People do not invest long/short Mkt - Rfr because that's trying to selectively capture the difference in premium, knowing that premiums don't exist unless there are periods where they are negative. So you are exposing yourself to potentially more risk if you invest this way with 100% of your assets. Some people do invest a part of their portfolio in such a way, through long/short levered funds like AQR. But for many, it's a little too much risk they aren't comfortable with. And that's fine.

Expected premiums being NEGATIVE simply does not happen as a consequence of how markets work. Markets may be highly valued, and they may encounter times when that high valuation is not warranted, so what does the market do? It drops in price to reflect greater risks. What was once a LOW expected (not negative expected) premium turned into a HIGHER expected premium as the pricing action dropped.

SCV having a negative EXPECTED premium is a mathematical impossibility because it's selectively seeking stocks with LOWER valuations. The valuations would somehow have to be higher than the marker, which again, does not make sense as the sorting process is rebalanced over time.

Equity returns *DO NOT* come from the profits of businesses in the real economy. They come from the PRICE YOU PAY for the EXPECTED FUTURE CASH FLOWS as priced by the market, and the net change over time of that SPECULATIVE pricing valuation versus actual cash flows accrued by the time you liquidate your position. "Factor" funds are no different other than the fact that they selectively seek assets with higher expected returns (via higher discount rates) so that you are entitled to a larger share of future cashflows than the overall market. It is the exact same principle.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

Apathizer wrote: Sun Sep 11, 2022 6:43 pm
comeinvest wrote: Sun Sep 11, 2022 4:55 pm Another thought experiment: The predictive capability (or lack of it) of factors is heavily debated in this forum. But regardless if a risk-adjusted outperformance could be achieved with factors or not, I think most participants in this forum would be happy to invest 100% in the stock market index with no factor tilts (e.g. if "factors" or factor tilts didn't exist). On the other hand, how many Bogleheads would be confident investing 100% of their lifetime retirement savings in a long/short factor fund with no market exposure? I wouldn't.
It's heavily debated here, but there still seems to be enough consensus among academics that factors have delivered a positive risk premium. The issue is whether this will continue. This was discussed recently on the Rational Reminder podcast. The academic guesstimate is that as more investors employ factor slants their premium will some degree of arbitrage decrease, but they will still persist to some extent. Considering how easy it is now to construct a factor-slanted portfolio this approach seems reasonable, though yes, a cap weight total market portfolio is also fine.

The highlighted portion seems puzzling and hyperbolic. Since factors are components of the market, how exactly it possible to invest in them without market exposure? That doesn't make sense. Also, few if any of us here are arguing for massive slants in one a few factors. Most of are arguing for a total market approach with some degree of factor slant. This increases diversification across risk factors, which increases likelihood of higher returns and consistency, though there's also risk of under-performance. If slants are light-moderate that possible under-performance will be fairly modest.
https://www.pwlcapital.com/five-factor- ... with-etfs/
You are just reiterating what I was trying to say, more or less. Most of us have much higher confidence in the continuation of the equity risk premium than in the continuation of the alternative factor premia. And yes, factors can be implemented at low expense these days, which might make it worthwhile just slanting somewhat, "just in case" so we don't miss anything that "others" harvest, if we think that the likelihood of factors outperforming, or the likelihood that the "experts" advocating for it are right, is higher than the likelihood of underperformance. It looks like the opinions regarding this differ here in this forum as well as among "experts".
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

Nathan Drake wrote: Sun Sep 11, 2022 9:41 pm
comeinvest wrote: Sun Sep 11, 2022 9:17 pm
Nathan Drake wrote: Sun Sep 11, 2022 6:23 pm
comeinvest wrote: Sun Sep 11, 2022 4:55 pm
Nathan Drake wrote: Sun Sep 11, 2022 12:21 pm

Stocks have gone through essentially periods of an investor's lifetime with a negative premium relative to bonds, as BurritoLover pointed out below. Economic growth essentially has no correlation to stock market returns either.

Remember that when you claim something is "data mined" that can apply to the overall stock market. Why should we invest based on market capitalization? Is that because some people data mined and found that by having this level of diversification, as weighted by the market, that you tend to receive an adequate 5-6% real return over the risk free rate over LONG periods? How is this any different than some people data mining the market looking for alternative ways to capture additional risk premiums? It's not.
I don't take sides if the factor models are data mining or have predictive capability. I'm saying that the confidence in the market premium is significantly higher than the confidence in factor premia. As we recently discussed in another thread, factor premia have essentially no explanation. All "explanations" were made up ad-hoc and are purely descriptive, a desperate attempt at "explaining" describing the historical data.

I still think that if expected real equity risk premia were negative, the economy and the country would come to a standstill. Nobody would invest a dollar into a business with the expectation that it will be worth for example $0.98 a year later. As a result, any and all innovation would be halted, businesses would die. If you disagree, tell me how the economy would work, why anybody would have any incentive to run a business. If nationalization of productive assets were imminent like during the October revolution in Russia, the economy and the country as we know it would effectively cease to exist.

I'm not sure what precisely you are trying to show with the above chart. We all know that there was an equity market drawdown in the 1929+ period, and a bond market drawdown in the 1970ies to early 1980ies because of a dramatic unexpected rise in interest rates. None of that says that expected ex-ante returns were negative, or that long-term returns were ever expected to be negative. Most major market swings in history were to a large extent due to valuation changes, meaning expected returns arguably moved opposite to the market, if anything.

Another thought experiment: The predictive capability (or lack of it) of factors is heavily debated in this forum. But regardless if a risk-adjusted outperformance could be achieved with factors or not, I think most participants in this forum would be happy to invest 100% in the stock market index with no factor tilts (e.g. if "factors" or factor tilts didn't exist). On the other hand, how many Bogleheads would be confident investing 100% of their lifetime retirement savings in a long/short factor fund with no market exposure? I wouldn't.
I don't think anyone is necessarily disagreeing with the idea that the market premium may have more confidence. After all, it explains 60% of returns. But the other 40% of returns that are explained by additional factors are not insignificant.

There are plenty of rational explanations for factors. The idea that these were made up "after" they were discovered seems like it's not much different than equities. Back in the early 1900s and prior, stocks did not have an explanation. In fact bonds were seen as just as risky as stocks. What changed? Maybe markets matured, in other words they started pricing the risk of stocks (versus bonds) better. But, like the factor explanation, it's just something theorized AFTER THE FACT. That doesn't mean it's foundation is weak.

If the expected risk premium is negative, it means that equities are priced way out of balance relative to other assets. This usually coincides with a euphoric rise in valuations and "this time is different" type thinking where stocks will always outperform now and forever. Markets don't work like that. There's constant re-pricing to adjust for newly perceived risks. That's why these things go through cycles of underperformance. If there were not periods of underperformance with negative premiums, there would be no positive premiums to begin with.

So let's look at 1966-1982 - an example where starting valuations were eurphoric. What brought massive inflows into stocks brought investors great pleasure. Valuations became stretched. Geopolitical and inflationary risks showed up. Valuations dropped. The equity risk premium went negative. This environment brought investors great pain. Nobody wanted to invest in a market after it had a significant re-valuation. But that re-pricing to account for a higher risk premium is precisely why it became such a good time to invest. Markets aren't static. Eventually investors will demand more return for the risk they are taking on.

The thought experiment doesn't really hold any significance to me. It's comparing apples to oranges. You could make the same thought experiment about going long total stock market and short bonds. Investment professionals like DFA, etc. don't advocate going 100 long/short because the short side of things can really hurt performance if you happen to go through a long cycle of underperformance. A long only portfolio doesn't have such risk at all. And there are many people that are comfortable being nearly 100% SCV (long only).
Most of your paragraphs just re-phrase what I was said before, except your dismissal of my thought experiments. You didn't address it. To be clear, by "expected premium" I always meant just that: *expected*, i.e. a priori premium. All your paragraphs about historical drawdowns describe realized premia, not expected premia. I pointed out myself in my last few comments that realized premia can and always will be negative at times, of course - markets go up and down. What does this have to do with anything. My thought experiment is about when *expected* long-term market term premia are negative. I believe that the economy and the country as we know it would cease to exist, for the reasons that I cited: Neither you nor I would invest a dollar in any business with the expectation to get back $0.98 in real terms a year later, or maybe $0.50 in real terms 30 years later. In contrast to the newer (possibly data mined - the jury is still out) "factors", which would have near zero effect if they were negative starting now for the remaining existence of humanity, except for folks who were betting on the continuation of historical patterns.

Let's do some straight talk. Equity market returns come from the profits of businesses in the real economy (adjusted for valuation changes and "slippage" from new competing capital, of course). "Factor" returns are mathematical artifacts that showed up in some descriptive analyses of historical returns, and that must mathematically net to zero (e.g. value vs. growth investors). Fundamentally different concepts, I think.
I did address your thought experiment. The market is a factor above the risk free rate. People do not invest long/short Mkt - Rfr because that's trying to selectively capture the difference in premium, knowing that premiums don't exist unless there are periods where they are negative. So you are exposing yourself to potentially more risk if you invest this way with 100% of your assets. Some people do invest a part of their portfolio in such a way, through long/short levered funds like AQR. But for many, it's a little too much risk they aren't comfortable with. And that's fine.
Not sure how this is related to my thought experiment. Maybe you don't realize that you can only invest purely in a factor (and not simultaneously in the market, i.e. market neutral) via a long-short portfolio, because like I said before, the "factors" are nothing but mathematical artifacts, i.e. they don't represent claims on tangible or intangible assets of the real economy like a business. By contrast, in the market you can invest with a long-only portfolio. If you implement a factor tilted portfolio, you can decompose it into a market index portfolio and a long-short overlay.
Nathan Drake wrote: Sun Sep 11, 2022 9:41 pm Expected premiums being NEGATIVE simply does not happen as a consequence of how markets work. Markets may be highly valued, and they may encounter times when that high valuation is not warranted, so what does the market do? It drops in price to reflect greater risks. What was once a LOW expected (not negative expected) premium turned into a HIGHER expected premium as the pricing action dropped.
Exactly what I said before. Markets go up and down (what a surprise). And valuations change over time. What does it have to do with anything.
Nathan Drake wrote: Sun Sep 11, 2022 9:41 pm SCV having a negative EXPECTED premium is a mathematical impossibility because it's selectively seeking stocks with LOWER valuations. The valuations would somehow have to be higher than the marker, which again, does not make sense as the sorting process is rebalanced over time.
You completely misunderstand the mechanics and the nature of factor premia. SCV is just one example. The "value" outperformance does NOT directly come from the lower valuation of "value" companies. There are risk and behavioral explanations, all of which are more "descriptions", not explanations; and there is a possibility that it's just data mining. Companies are valued based on expected cash flows and expected growth, and under a naive no-arbitrage assumption or a simplistic capital asset pricing model, expected risk-adjusted returns would be equal. It would be 100% thinkable that "growth" companies outperform "value" companies for the remaining existence of humanity, and nothing material would change to our economy.
Nathan Drake wrote: Sun Sep 11, 2022 9:41 pm Equity returns *DO NOT* come from the profits of businesses in the real economy. They come from the PRICE YOU PAY for the EXPECTED FUTURE CASH FLOWS as priced by the market, and the net change over time of that SPECULATIVE pricing valuation versus actual cash flows accrued by the time you liquidate your position. "Factor" funds are no different other than the fact that they selectively seek assets with higher expected returns (via higher discount rates) so that you are entitled to a larger share of future cashflows than the overall market. It is the exact same principle.
Your second sentence repeats exactly what I said in my last post: "Equity market returns come from the profits of businesses in the real economy (adjusted for valuation changes and "slippage" from new competing capital, of course)."
Your third sentence just re-phrases what I pointed out is a fundamental difference: Factor returns come from assumptions on continuation of the outperformance of stocks with certain criteria, i.e. from the belief that historical patterns continue. A stake in a factor or "exposure" to a factor does not represent a tangible asset in the real economy and can only be implemented with a long-short bet, unless combined with the market factor into a tilted long-only portfolio. Before we get lost in terminology about whether the difference is "fundamental", please read my entire post in context to arrive at my definition of "fundamentally" different: I am using the word "fundamental" in the context and with the semantics of the dramatically different outcomes if either expected return premia were to be negative in the future: The economy and the country as we know it ceasing to exist, vs. some dudes betting on some pattern losing some dollars.
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Avantis Funds: what's so special about them?

Post by Nathan Drake »

I’m not sure why you’re theorizing hypothetical doomsday scenarios, but the question of whether factors are more “data mined” than the market itself would be the least of my worries should the expected premiums of equities turn sharply negative and remain there in perpetuity. That would represent a cataclysmic circumstance to which there would never be chance for the situation to improve.

It seems obvious to me that different pricings in the market represent different risks. Your understanding of market pricing seems to harken back to an older understanding with far less certainty in the explanation of returns. Behavioral explanations in addition to risk based explanations doesn’t change that fact.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

Nathan Drake wrote: Sun Sep 11, 2022 10:57 pm I’m not sure why you’re theorizing hypothetical doomsday scenarios, but the question of whether factors are more “data mined” than the market itself would be the least of my worries should the expected premiums of equities turn sharply negative and remain there in perpetuity. That would represent a cataclysmic circumstance to which there would never be chance for the situation to improve.

It seems obvious to me that different pricings in the market represent different risks. Your understanding of market pricing seems to harken back to an older understanding with far less certainty in the explanation of returns. Behavioral explanations in addition to risk based explanations doesn’t change that fact.
Glad that what the community in this forum is very divided on and what is heavily debated even among experts, is so obvious to you.

I read probably 100's of threads on the subject of factor tilting in this forum and 100's of papers outside this forum, and I for myself implement some "small value" tilts, "just in case" or "not to miss out". But like many others, I'm not extremely confident and don't necessarily expect any risk-adjusted outperformance or diversification of sources of returns from my factor tilts, not even in the long run.
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Avantis Funds: what's so special about them?

Post by Nathan Drake »

comeinvest wrote: Sun Sep 11, 2022 11:06 pm
Nathan Drake wrote: Sun Sep 11, 2022 10:57 pm I’m not sure why you’re theorizing hypothetical doomsday scenarios, but the question of whether factors are more “data mined” than the market itself would be the least of my worries should the expected premiums of equities turn sharply negative and remain there in perpetuity. That would represent a cataclysmic circumstance to which there would never be chance for the situation to improve.

It seems obvious to me that different pricings in the market represent different risks. Your understanding of market pricing seems to harken back to an older understanding with far less certainty in the explanation of returns. Behavioral explanations in addition to risk based explanations doesn’t change that fact.
Glad that what the community in this forum is very divided on and what is heavily debated even among experts, is so obvious to you.

I read probably 100's of threads on the subject of factor tilting in this forum and 100's of papers outside this forum, and I for myself implement some small value tilts, "just in case". But like many others, I'm not extremely confident and don't necessarily expect any risk-adjusted outperformance or diversification of sources of returns from my factor tilts, not even in the long run.
If you’re not extremely confident after reading so much research, and seeing so much data that clearly shows a long term outperformance (with higher risk) in addition to diversification benefits, then why are you investing in it at all?

The evidence is compelling to me. It doesn’t discredit someone with less conviction being perfectly happy in total stock market funds, but the evidence that leads us to believe in the long term outperformance of the market over bonds is the same evidence that leads us to believe in higher discounted stocks having a higher risk/return profile over the market.

It’s really not too far removed from the idea of corporate bonds with higher yields over treasuries having a higher risk/return as priced by the market.

Why is the value idea for stocks more hotly debated? It simply seems like some people want to believe that a total stock market fund is a perfect product that always bears the best return. The historical evidence refutes that idea.
Last edited by Nathan Drake on Sun Sep 11, 2022 11:17 pm, edited 1 time in total.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
User avatar
whodidntante
Posts: 13114
Joined: Thu Jan 21, 2016 10:11 pm
Location: outside the echo chamber

Re: Avantis Funds: what's so special about them?

Post by whodidntante »

There is going to be an interview with the big cheese at Avantis by the rational reminder podcast. Y’all could head over to their forum and put 9,784 upvotes on the question I posed about the Avantis value proposition versus DFA funds and Ben Felix will ask him about it.
Apathizer
Posts: 2507
Joined: Sun Sep 26, 2021 2:56 pm

Re: Avantis Funds: what's so special about them?

Post by Apathizer »

comeinvest wrote: Sun Sep 11, 2022 10:06 pmIt would be 100% thinkable that "growth" companies outperform "value" companies for the remaining existence of humanity, and nothing material would change to our economy.
[Profanity removed by moderator oldcomputerguy] Any concept or possibility is 100% thinkable. All we can do as investors is try to figure out approach that seems most likely to produce the highest, most consistent returns. Since they have more diverse potential return sources, a factor-slanted total market portfolio seems the most likely approach.

Yes, there isn't unanimous consensus among academics, but that's the case for almost anything. A small minority of scientists dismiss anthropogenic climate change, but the vast majority don't. While the consensus on factors probably isn't as overwhelming, to me the evidence is strong enough I'm willing employ them. If you're unconvinced and prefer a cap weight portfolio, that's fine, but there's also a reasonable case for factors.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

Apathizer wrote: Sun Sep 11, 2022 11:58 pm
comeinvest wrote: Sun Sep 11, 2022 10:06 pmIt would be 100% thinkable that "growth" companies outperform "value" companies for the remaining existence of humanity, and nothing material would change to our economy.
[Profanity removed by moderator oldcomputerguy] Any concept or possibility is 100% thinkable. All we can do as investors is try to figure out approach that seems most likely to produce the highest, most consistent returns. Since they have more diverse potential return sources, a factor-slanted total market portfolio seems the most likely approach.

Yes, there isn't unanimous consensus among academics, but that's the case for almost anything. A small minority of scientists dismiss anthropogenic climate change, but the vast majority don't. While the consensus on factors probably isn't as overwhelming, to me the evidence is strong enough I'm willing employ them. If you're unconvinced and prefer a cap weight portfolio, that's fine, but there's also a reasonable case for factors.
I pointed out that there is a real possibility of a "growth premium" in the future, in response to your assertion that "SCV having a negative EXPECTED premium is a mathematical impossibility", based on your obvious lack of understanding of the mechanics and causes of the value premium and of sources of equities returns in general. I did not say that a growth premium is more likely than a value premium, but that it is a scenario that is definitely not a mathematical impossibility. I'm not sure why you need to use profanity in response. As you are not engaging with my thoughts or with the original subject that let to them, we better end this conversation here.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

Nathan Drake wrote: Sun Sep 11, 2022 11:15 pm
comeinvest wrote: Sun Sep 11, 2022 11:06 pm
Nathan Drake wrote: Sun Sep 11, 2022 10:57 pm I’m not sure why you’re theorizing hypothetical doomsday scenarios, but the question of whether factors are more “data mined” than the market itself would be the least of my worries should the expected premiums of equities turn sharply negative and remain there in perpetuity. That would represent a cataclysmic circumstance to which there would never be chance for the situation to improve.

It seems obvious to me that different pricings in the market represent different risks. Your understanding of market pricing seems to harken back to an older understanding with far less certainty in the explanation of returns. Behavioral explanations in addition to risk based explanations doesn’t change that fact.
Glad that what the community in this forum is very divided on and what is heavily debated even among experts, is so obvious to you.

I read probably 100's of threads on the subject of factor tilting in this forum and 100's of papers outside this forum, and I for myself implement some small value tilts, "just in case". But like many others, I'm not extremely confident and don't necessarily expect any risk-adjusted outperformance or diversification of sources of returns from my factor tilts, not even in the long run.
If you’re not extremely confident after reading so much research, and seeing so much data that clearly shows a long term outperformance (with higher risk) in addition to diversification benefits, then why are you investing in it at all?

The evidence is compelling to me. It doesn’t discredit someone with less conviction being perfectly happy in total stock market funds, but the evidence that leads us to believe in the long term outperformance of the market over bonds is the same evidence that leads us to believe in higher discounted stocks having a higher risk/return profile over the market.

It’s really not too far removed from the idea of corporate bonds with higher yields over treasuries having a higher risk/return as priced by the market.

Why is the value idea for stocks more hotly debated? It simply seems like some people want to believe that a total stock market fund is a perfect product that always bears the best return. The historical evidence refutes that idea.
I'm glad that the continued outperformance of factors is obvious to you. As evidenced by the many threads on the subject, it is not obvious to many participants in this forum, some of which have studied and followed the topic for decades.
You keep reiterating your convictions, and you don't engage in the discourse. That's perfectly fine.
User avatar
typical.investor
Posts: 5263
Joined: Mon Jun 11, 2018 3:17 am

Re: Avantis Funds: what's so special about them?

Post by typical.investor »

comeinvest wrote: Mon Sep 12, 2022 12:55 am
Nathan Drake wrote: Sun Sep 11, 2022 11:15 pm
comeinvest wrote: Sun Sep 11, 2022 11:06 pm
Nathan Drake wrote: Sun Sep 11, 2022 10:57 pm I’m not sure why you’re theorizing hypothetical doomsday scenarios, but the question of whether factors are more “data mined” than the market itself would be the least of my worries should the expected premiums of equities turn sharply negative and remain there in perpetuity. That would represent a cataclysmic circumstance to which there would never be chance for the situation to improve.

It seems obvious to me that different pricings in the market represent different risks. Your understanding of market pricing seems to harken back to an older understanding with far less certainty in the explanation of returns. Behavioral explanations in addition to risk based explanations doesn’t change that fact.
Glad that what the community in this forum is very divided on and what is heavily debated even among experts, is so obvious to you.

I read probably 100's of threads on the subject of factor tilting in this forum and 100's of papers outside this forum, and I for myself implement some small value tilts, "just in case". But like many others, I'm not extremely confident and don't necessarily expect any risk-adjusted outperformance or diversification of sources of returns from my factor tilts, not even in the long run.
If you’re not extremely confident after reading so much research, and seeing so much data that clearly shows a long term outperformance (with higher risk) in addition to diversification benefits, then why are you investing in it at all?

The evidence is compelling to me. It doesn’t discredit someone with less conviction being perfectly happy in total stock market funds, but the evidence that leads us to believe in the long term outperformance of the market over bonds is the same evidence that leads us to believe in higher discounted stocks having a higher risk/return profile over the market.

It’s really not too far removed from the idea of corporate bonds with higher yields over treasuries having a higher risk/return as priced by the market.

Why is the value idea for stocks more hotly debated? It simply seems like some people want to believe that a total stock market fund is a perfect product that always bears the best return. The historical evidence refutes that idea.
I'm glad that the continued outperformance of factors is obvious to you. As evidenced by the many threads on the subject, it is not obvious to many participants in this forum, some of which have studied and followed the topic for decades.
You keep reiterating your convictions, and you don't engage in the discourse. That's perfectly fine.
In any case, even assuming historical evidence shows value outperformance and assuming that outperformance will be eternal, there is still risk for me as a value investor in that the outperformance may not coincide with my holding period.

Also, I don't see equity premium relative to bonds as being identical to the value premium over growth. Sure, they are conceptually similar but the mechanism are very different. I simply don't see the frequency, duration and magnitude of stock under-performance relative to bonds as being directly comparable to the frequency, duration and magnitude of value under-performance relative to growth.

Expecting stocks to outperform bonds simply doesn't mean we should expect value to outperform growth especially in any given holding period. Otherwise it would be saying we found gold in the hills so there must be oil too. For all my might, I just can't see the logic in that.

OK, maybe you figure that stocks out-perform because they are riskier and similarly perhaps value will out-perform growth because it is riskier, but it's clear as day that not everything with more risk is going to outperform in your holding period. It just doesn't work like that otherwise it'd simply be a race to take on the most risk imaginable.

I'm a value investor and think the risk is worth it but no guarantees and probably less certainty than stocks outperforming bonds.
Apathizer
Posts: 2507
Joined: Sun Sep 26, 2021 2:56 pm

Re: Avantis Funds: what's so special about them?

Post by Apathizer »

comeinvest wrote: Mon Sep 12, 2022 12:46 am I pointed out that there is a real possibility of a "growth premium" in the future, in response to your assertion that "SCV having a negative EXPECTED premium is a mathematical impossibility"
No, I never said that, you're confusing me with someone else, Nathan Drake I think. It's certainly possible, but highly improbable. You seem fixated on purely conceptual possibilities rather than rationally considering evidence, and seem to treat cap weight indexing as the unequivocal best approach when evidence shows otherwise.
comeinvest wrote: Mon Sep 12, 2022 12:46 amI did not say that a growth premium is more likely than a value premium, but that it is a scenario that is definitely not a mathematical impossibility.
That's possible, but highly improbable. In poker sometimes 66 will beat AA, but that's improbable. The same is true of growth and value. Sometimes growth will out-perform value, but there are good theoretical and empirical reasons to expect value will usually out-perform growth. While almost anything is possible, not all things are equally plausible.

I'll explain it in simple mathematical terms. It's far more likely that one improbable event will happen than 4 equally improbable events will all happen simultaneously. With investing there's about a 20% chance the market will under-perform over a decade. To keep things simple, let's say it's the same probability for the other 4 most relevant factors: size, value, profitability and investment. The probability all 5 factors will under-perform over the same period is 0.032%. The actual math is more complex; I've simplified it to illustrate a point.

Do you understand how diversifying across factors is likely to increase overall returns and consistency? That's all we're doing with factor-slants.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
User avatar
typical.investor
Posts: 5263
Joined: Mon Jun 11, 2018 3:17 am

Re: Avantis Funds: what's so special about them?

Post by typical.investor »

Apathizer wrote: Mon Sep 12, 2022 4:13 am Do you understand how diversifying across factors is likely to increase overall returns and consistency? That's all we're doing with factor-slants.
Not cominvest, but yeah I understand that you think that.

For your example though, it assumes a normal distribution on factor returns. Real data will show that a factor diversified portfolio has returns - especially during drawdowns- where it exhibits far from what you'd expect. Think corporate bonds and equities correlations. Correlations change especially when you don't want them to.

Also, factors are serially correlated. This means that they have momentum. You can see this in a value drawdown that seemingly goes on forever.

Not that these things should scare you off, but I am not really sure if your portfolio construction is going to achieve what you want.

In any case, I'd be wary of 'evidence' showing factor outperformance (either in returns or lowered volatility) that are based on long/short portfolios.

The simple reason is that that long only factor portfolios are not beta neutral. You can see this in a study of disappointing factor returns from June 2016 to June 2019 which turns out isn't that much of an anomaly. While Size, Value and Momentum delivered much worse
than the average negative performance observed, examination of longer shows these periods do happen. Of course, since we are looking at six factor portfolios, the other factors should have saved our bacon. Right!?!

What's interesting though, is that the explanation given for underperformance is not that factors themselves overall did poorly, but rather that long only portfolios actually reduce your market exposure such that even if your factor diversification is net positive, your returns could be lower due to lowered beta exposure.

I found it interesting at least to find that long only factor portfolios underperformed in a bull market and that perhaps this is to be expected.
It is clearly apparent that over the last 3 years, these indices without market-beta control have considerably underperformed
their equivalents with a market-neutral market beta long/short overlay. Ultimately, both in the US and Developed ex-US regions, the long-only multi-factor assembly does not allow the broad cap weighted index to be outperformed, as was the case previously.
Ok, that was in a bull. Not so good. Will a benefit be apparent in other types of markets? Research Affiliates data shows a factor diversified portfolio crashing much more than you'd expect (if in were normally diversified). Maybe think corporate bonds in a crash again. Similar effect.

So I am left guessing that perhaps it's neither boom nor bust where factor exposure does well and you could see outperformance.

Anyway, I am a factor investor but I do not believe the data necessarily shows that going forward we are in for a smoother ride and better returns. I think that it will kind of matter on what the market does. This makes sense. If factor diversification were sure to smooth and boost returns, why wouldn't everyone want it. I do believe there is the possibility the the market is such that factor diversification could underperform.

In any case, I do not want to hear promised for a long only portfolio that is based on long/short returns. I don't think just saying the premiums will just be weaker is accurate.

I didn't actually know this before now but was looking for one study Larry cited that I interpreted differently than him (because the only diversification benefit was really in momentum which may be hard to catch in the real world).


See https://www.institutional-money.com/con ... v_2019.pdf
dcabler
Posts: 4541
Joined: Wed Feb 19, 2014 10:30 am
Location: TX

Re: Avantis Funds: what's so special about them?

Post by dcabler »

Morik wrote: Fri Jul 01, 2022 10:38 pm They apply screening for additional factors on top of a size and/or value tilt.
DFA also does this, and some indices do some screening (e.g., I've read that S&P 600 value does a profitability screen).

The main thing for me is that I can get Avantis funds without an advisor, whereas DFA funds require either an advisor or sometimes they are offered in some retirement plans.
Many of the DFA funds today now have ETF equivalents available that you can access without an advisor.
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Avantis Funds: what's so special about them?

Post by vineviz »

typical.investor wrote: Mon Sep 12, 2022 3:23 am
OK, maybe you figure that stocks out-perform because they are riskier and similarly perhaps value will out-perform growth because it is riskier, but it's clear as day that not everything with more risk is going to outperform in your holding period. It just doesn't work like that otherwise it'd simply be a race to take on the most risk imaginable.
We expect stocks to outperform bonds, because stocks represent a risk that some investors will pay to avoid.

We expect value stocks to outperform growth stocks, because value stocks represent a risk that some investors will pay to avoid.

Neither of these statements imply that "everything with more risk is going to outperform in your holding period": if we could clearly know that, there'd be much less risk involved.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Avantis Funds: what's so special about them?

Post by vineviz »

comeinvest wrote: Sun Sep 11, 2022 9:07 pm
vineviz wrote: Sun Sep 11, 2022 7:37 pm
comeinvest wrote: Sun Sep 11, 2022 4:55 pm I don't take sides if the factor models are data mining or have predictive capability. I'm saying that the confidence in the market premium is significantly higher than the confidence in factor premia.
If that's your opinion, that's fine. Just know that there's not really any evidence to back it up.
comeinvest wrote: Sun Sep 11, 2022 4:55 pm I still think that if expected real equity risk premia were negative, the economy and the country would come to a standstill.
No one would EVER invest in a risky asset if the expected risk premium was negative. That's not unique to the market risk factor.
comeinvest wrote: Sun Sep 11, 2022 4:55 pm
On the other hand, how many Bogleheads would be confident investing 100% of their lifetime retirement savings in a long/short factor fund with no market exposure? I wouldn't.
The equity risk premium is, itself, a long-short portfolio: long stocks and short the risk-free asset.
With none of your 3 replies you are engaging only to the slightest with my analysis, my comments, or with my thought experiments.
Your thought experiment is uninteresting to me so, yes, I'm not addressing it.

I'm pointing out some material errors in your analysis: you have stated opinion as if it is fact, despite an absence of evidence to support that opinion; you've confused expected return with realized return; and you've overlooked the fact that ALL factors (including market) are measured by constructing long/short portfolios.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Apathizer
Posts: 2507
Joined: Sun Sep 26, 2021 2:56 pm

Re: Avantis Funds: what's so special about them?

Post by Apathizer »

typical.investor wrote: Mon Sep 12, 2022 5:57 am For your example though, it assumes a normal distribution on factor returns. Real data will show that a factor diversified portfolio has returns - especially during drawdowns- where it exhibits far from what you'd expect. Think corporate bonds and equities correlations. Correlations change especially when you don't want them to.
Sometimes factors will have steeper draw-downs than the market, but they will also sometimes have higher returns during recoveries. As you know there have also been extended periods some factors have out-performed the market. I don't know if this is the case, but it seems likely this will happen at least occasionally.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

typical.investor wrote: Mon Sep 12, 2022 5:57 am
Apathizer wrote: Mon Sep 12, 2022 4:13 am Do you understand how diversifying across factors is likely to increase overall returns and consistency? That's all we're doing with factor-slants.
Not cominvest, but yeah I understand that you think that.

For your example though, it assumes a normal distribution on factor returns. Real data will show that a factor diversified portfolio has returns - especially during drawdowns- where it exhibits far from what you'd expect. Think corporate bonds and equities correlations. Correlations change especially when you don't want them to.

Also, factors are serially correlated. This means that they have momentum. You can see this in a value drawdown that seemingly goes on forever.

Not that these things should scare you off, but I am not really sure if your portfolio construction is going to achieve what you want.

In any case, I'd be wary of 'evidence' showing factor outperformance (either in returns or lowered volatility) that are based on long/short portfolios.

The simple reason is that that long only factor portfolios are not beta neutral. You can see this in a study of disappointing factor returns from June 2016 to June 2019 which turns out isn't that much of an anomaly. While Size, Value and Momentum delivered much worse
than the average negative performance observed, examination of longer shows these periods do happen. Of course, since we are looking at six factor portfolios, the other factors should have saved our bacon. Right!?!

What's interesting though, is that the explanation given for underperformance is not that factors themselves overall did poorly, but rather that long only portfolios actually reduce your market exposure such that even if your factor diversification is net positive, your returns could be lower due to lowered beta exposure.

I found it interesting at least to find that long only factor portfolios underperformed in a bull market and that perhaps this is to be expected.
It is clearly apparent that over the last 3 years, these indices without market-beta control have considerably underperformed
their equivalents with a market-neutral market beta long/short overlay. Ultimately, both in the US and Developed ex-US regions, the long-only multi-factor assembly does not allow the broad cap weighted index to be outperformed, as was the case previously.
Ok, that was in a bull. Not so good. Will a benefit be apparent in other types of markets? Research Affiliates data shows a factor diversified portfolio crashing much more than you'd expect (if in were normally diversified). Maybe think corporate bonds in a crash again. Similar effect.

So I am left guessing that perhaps it's neither boom nor bust where factor exposure does well and you could see outperformance.

Anyway, I am a factor investor but I do not believe the data necessarily shows that going forward we are in for a smoother ride and better returns. I think that it will kind of matter on what the market does. This makes sense. If factor diversification were sure to smooth and boost returns, why wouldn't everyone want it. I do believe there is the possibility the the market is such that factor diversification could underperform.

In any case, I do not want to hear promised for a long only portfolio that is based on long/short returns. I don't think just saying the premiums will just be weaker is accurate.

I didn't actually know this before now but was looking for one study Larry cited that I interpreted differently than him (because the only diversification benefit was really in momentum which may be hard to catch in the real world).


See https://www.institutional-money.com/con ... v_2019.pdf
I don't understand everything that you said, but your post and other controversial posts in this forum tell me that it's not black or white, and the value of factor investing in portfolio applications is far from obvious. I'm getting more and more confused, the deeper I try to understand factor investing and whether or not it improves my portfolio. New factors emerge, old factors are dismissed or "modified" to fit evolving data sets, and then when you put everything together in real life portfolios, some of the "evidence" is turned upside down, and results vary heavily with backtesting parameters, and then in finance even when historical evidence seems consistent, when things change it's always a gamble if mean reversion will happen or a regime change occurred.
Apathizer
Posts: 2507
Joined: Sun Sep 26, 2021 2:56 pm

Re: Avantis Funds: what's so special about them?

Post by Apathizer »

typical.investor wrote: Mon Sep 12, 2022 5:57 am Anyway, I am a factor investor but I do not believe the data necessarily shows that going forward we are in for a smoother ride and better returns. I think that it will kind of matter on what the market does. This makes sense. If factor diversification were sure to smooth and boost returns, why wouldn't everyone want it. I do believe there is the possibility the the market is such that factor diversification could underperform.
There are any number of reason everyone wouldn't necessarily want factors. Many casual investors are disinterested in financial research and minutia, so don't know anything about factors. Many people are perfectly content with an automatic cap weight approach using auto rebalancing funds, which is fine.

Psychologically, most people have a status quo bias, meaning most people are disinclined to make changes even if those changes are likely to be beneficial. This isn't so much of a problem for simple cap weight investors (compared to other aspects of life) since that approach is far more likely to succeed than traditional active management, but I doggedly try to avoid status quo bias since I like change and trying new things. In investing terms, I was an ESG investor for many years, then an index investors for a few years. I still generally favor indexing, but now employ factor-slants since there seems to be enough evidence they provide beneficial diversification.

I'll probably take a break from thinking about investing for at least a few months since it's not the be all of life to me, but will still try to keep updated on the latest research at least occasionally.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
km91
Posts: 1388
Joined: Wed Oct 13, 2021 12:32 pm

Re: Avantis Funds: what's so special about them?

Post by km91 »

vineviz wrote: Sun Sep 11, 2022 7:37 pm
comeinvest wrote: Sun Sep 11, 2022 4:55 pm I don't take sides if the factor models are data mining or have predictive capability. I'm saying that the confidence in the market premium is significantly higher than the confidence in factor premia.
If that's your opinion, that's fine. Just know that there's not really any evidence to back it up.
Can you explain what is meant here and what evidence proves or disproves this claim?

Markets seem to have had a general sense of the existence of an "equity risk premium" and of the risk/return trade off 100+ years before it was formalized in modern portfolio theory or the capm. The academic factor models have been around for 30 or so years, but I think you can argue that "value" and "momentum" were known to the market long before they were ever published in the academic papers. I don't know if this can make us more "confident" in these factors but I do think it speaks to the durability and persistence of these factors in describing stock returns.
User avatar
typical.investor
Posts: 5263
Joined: Mon Jun 11, 2018 3:17 am

Re: Avantis Funds: what's so special about them?

Post by typical.investor »

comeinvest wrote: Mon Sep 12, 2022 1:23 pm
I don't understand everything that you said, but your post and other controversial posts in this forum tell me that it's not black or white, and the value of factor investing in portfolio applications is far from obvious. I'm getting more and more confused, the deeper I try to understand factor investing and whether or not it improves my portfolio. New factors emerge, old factors are dismissed or "modified" to fit evolving data sets, and then when you put everything together in real life portfolios, some of the "evidence" is turned upside down, and results vary heavily with backtesting parameters, and then in finance even when historical evidence seems consistent, when things change it's always a gamble if mean reversion will happen or a regime change occurred.
Well, I'd say the best evidence for factor benefit we have are the actual DFA funds with a long history. Even though their methodology has been improved to incorporate profitability screens, value exposure was still useful.

As for the claims that diversified factor investing is going to smooth your ride, all I can conclude is that it might but knowing what we do about how factors actually perform - nobody should be surprised if it doesn't.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

km91 wrote: Mon Sep 12, 2022 1:41 pm
vineviz wrote: Sun Sep 11, 2022 7:37 pm
comeinvest wrote: Sun Sep 11, 2022 4:55 pm I don't take sides if the factor models are data mining or have predictive capability. I'm saying that the confidence in the market premium is significantly higher than the confidence in factor premia.
If that's your opinion, that's fine. Just know that there's not really any evidence to back it up.
Can you explain what is meant here and what evidence proves or disproves this claim?

Markets seem to have had a general sense of the existence of an "equity risk premium" and of the risk/return trade off 100+ years before it was formalized in modern portfolio theory or the capm. The academic factor models have been around for 30 or so years, but I think you can argue that "value" and "momentum" were known to the market long before they were ever published in the academic papers. I don't know if this can make us more "confident" in these factors but I do think it speaks to the durability and persistence of these factors in describing stock returns.
I'm not vineviz, but Larry Swedroe and others made convincing arguments and showed that value, profitability, and momentum showed outperformance over very long periods and across geographical regions.
But then again, many things in finance persisted for a long time, until they didn't. (It's then called "regime change", and an "explanation" is usually offered after the fact, and/or the parameters are "adjusted" for modern times.) Others keep persisting in a mean-reverting way. And then on top of it, there is that self-negating feedback loop effect after "discovery". (I know I know, there are also those risk and behavioral "explanations" descriptions.) I think it's hard to quantify the likelihood of future persistence of a pattern in terms of statistical probabilities, as distinguishing data mining from finding "fundamental" patterns would depend on the definition of the original opportunity space in which people tried to find the patterns, which was not done a priori.
I still think it would be hard to imagine a world with negative expected long-term equity risk premium, as I think it would be a world without investment in the real economy. And a world without investment is a world without investing. But I don't want to start this discussion all over again, we just abandoned that thought experiment.
I'm happy to stand corrected, but I think there has been no moment in the history of mankind when the a priori equity risk premium was negative, i.e. when the expectation was that you will get back less than the capital you put into something.
Last edited by comeinvest on Mon Sep 12, 2022 3:01 pm, edited 4 times in total.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Avantis Funds: what's so special about them?

Post by comeinvest »

typical.investor wrote: Mon Sep 12, 2022 2:28 pm
comeinvest wrote: Mon Sep 12, 2022 1:23 pm
I don't understand everything that you said, but your post and other controversial posts in this forum tell me that it's not black or white, and the value of factor investing in portfolio applications is far from obvious. I'm getting more and more confused, the deeper I try to understand factor investing and whether or not it improves my portfolio. New factors emerge, old factors are dismissed or "modified" to fit evolving data sets, and then when you put everything together in real life portfolios, some of the "evidence" is turned upside down, and results vary heavily with backtesting parameters, and then in finance even when historical evidence seems consistent, when things change it's always a gamble if mean reversion will happen or a regime change occurred.
Well, I'd say the best evidence for factor benefit we have are the actual DFA funds with a long history. Even though their methodology has been improved to incorporate profitability screens, value exposure was still useful.

As for the claims that diversified factor investing is going to smooth your ride, all I can conclude is that it might but knowing what we do about how factors actually perform - nobody should be surprised if it doesn't.
Exactly. I am most interested in seeing factual evidence of *risk-adjusted* outperformance or benefits of factor tilts within a (for example) equities/treasuries portfolio, and I think most other retail investors should be most interested in that question. And with a definition of "risk" that is not volatility or something similarly irrelevant, but with an actual meaningful personal utility based risk definition, for example maximum expected drawdown over a life cycle (still not perfect, but better). (We all know at this point that factor correlations with each other and with equities seem to often increase during crises.)
There were posts in this forum (don't have the link right now) showing that the DFA funds had no *risk-adjusted* outperformance or benefits in a typical portfolio, compared to index funds. (And also no outperformance compared to Vanguard value funds, but that's an entirely different topic.)
km91
Posts: 1388
Joined: Wed Oct 13, 2021 12:32 pm

Re: Avantis Funds: what's so special about them?

Post by km91 »

comeinvest wrote: Mon Sep 12, 2022 2:30 pm I'm happy to stand corrected, but I think there has been no moment in the history of mankind when the a priori equity risk premium was negative, i.e. when the expectation was that you will get back less than the capital you put into something.
Not sure if this statement can be true. Everyone who sells during a market crash would seem to be worried that they will get back less than they put in
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: Avantis Funds: what's so special about them?

Post by Nathan Drake »

comeinvest wrote: Mon Sep 12, 2022 2:48 pm
typical.investor wrote: Mon Sep 12, 2022 2:28 pm
comeinvest wrote: Mon Sep 12, 2022 1:23 pm
I don't understand everything that you said, but your post and other controversial posts in this forum tell me that it's not black or white, and the value of factor investing in portfolio applications is far from obvious. I'm getting more and more confused, the deeper I try to understand factor investing and whether or not it improves my portfolio. New factors emerge, old factors are dismissed or "modified" to fit evolving data sets, and then when you put everything together in real life portfolios, some of the "evidence" is turned upside down, and results vary heavily with backtesting parameters, and then in finance even when historical evidence seems consistent, when things change it's always a gamble if mean reversion will happen or a regime change occurred.
Well, I'd say the best evidence for factor benefit we have are the actual DFA funds with a long history. Even though their methodology has been improved to incorporate profitability screens, value exposure was still useful.

As for the claims that diversified factor investing is going to smooth your ride, all I can conclude is that it might but knowing what we do about how factors actually perform - nobody should be surprised if it doesn't.
Exactly. I am most interested in seeing factual evidence of *risk-adjusted* outperformance or benefits of factor tilts within a (for example) equities/treasuries portfolio, and I think most other retail investors should be most interested in that question. And with a definition of "risk" that is not volatility or something similarly irrelevant, but with an actual meaningful personal utility based risk definition, for example maximum expected drawdown over a life cycle (still not perfect, but better). (We all know at this point that factor correlations with each other and with equities seem to often increase during crises.)
There were posts in this forum (don't have the link right now) showing that the DFA funds had no *risk-adjusted* outperformance or benefits in a typical portfolio, compared to index funds. (And also no outperformance compared to Vanguard value funds, but that's an entirely different topic.)
Nobody is claiming higher risk adjusted long term returns

You don’t need higher risk adjusted returns for higher overall returns and additional diversification
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
Post Reply