Why factor investing isn't working

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
schooner
Posts: 388
Joined: Sun Jun 09, 2019 8:27 am

Re: Why factor investing isn't working

Post by schooner »

larryswedroe wrote: Wed Jul 10, 2019 11:20 am schooner
I have shown you how it is improved results in both articles and my books. As I said, just read Reducing the Risk of Black Swans to find the historical evidence, using live funds, not indices. I've also shown you that over the past 20+ years factor based funds that are well designed have significantly outperformed TSM counterparts, but of course that will not be true over all periods, nor is it a guarantee it will in future, but based on the evidence, and IMO the logic it definitely puts the odds in your favor.

If not familiar I suggest you run the following as a simple test using the portfoliovisualizer tool you are familiar with
Portfolio A Vanguard US TSM 60%/Vanguard Intermediate Treasury 40%
Portfolio B DFSVX 40%/Vanguard Intermediate Treasury 60%

Run it from inception of DFSVX so you have the longest period possible, now over 25 years. Check the returns/SD, SR, factor loadings and best and worst case years. Then ask yourself which portfolio you would rather have owned.

Again, this is just example. I would not own such a portfolio as would do it globally including EM. But it's example of moving toward risk parity.
Then you can decide if you believe benefits of TSM (lowest costs and most tax efficient) more than the benefits of the other portfolio. And you cannot argue A is simpler, they both are just two fund portfolios, and globally can also do with just two equity funds.

Again, no right answer, because there are trade offs.

BTW, Note I have tried patiently to answer all your questions but I note you almost never answer the simple ones I ask you. Like why you believe 10 years is not long enough to give up on market beta but thought 10 years demonstrated value and size were not real factors. Just asking (:-)) It appears to me there are many inconsistencies in your explanations, like you believe markets are efficient but the much more volatile small and value stocks should not be expected to have higher returns, not higher risk adjusted returns, but higher returns? And the inconsistency on the 10 year issue.

Larry
That's completely fair - I'm not answering your questions because I don't know the answers. The argument is a bit asymmetric because you are the professional and I am the novice.

I just get a bit nervous about all of this factor stuff taking place at the institutional level - since a small share of that is money that's invested on my behalf. And if there is a change in strategy, it's only fair for me to ask the tough questions. I am a stakeholder.

The back tests looks great but the actual track record looks a bit shaky.

Again, if Nedsaid wants play factors or Hedgefundie is using leveraged ETFs to try and beat the market, more power to them. It's a whole different thing when it's being played on an institutional level.
TropikThunder
Posts: 3918
Joined: Sun Apr 03, 2016 5:41 pm

Re: Why factor investing isn't working

Post by TropikThunder »

schooner wrote: Wed Jul 10, 2019 11:39 am That's completely fair - I'm not answering your questions because I don't know the answers. The argument is a bit asymmetric because you are the professional and I am the novice.
Again, you’re being inconsistent. You’re perfectly content to make a judgement on factors based on a 10 year stretch (which I pointed out to you was really just a 5 year stretch) despite being a novice. But then you say you can’t evaluate TSM v Treasuries on a 10 year stretch because you’re a novice.
schooner wrote: Wed Jul 10, 2019 11:39 am The back tests looks great but the actual track record looks a bit shaky.
What does this mean? The back tests Larry showed you are with actual funds, not indexes. In other words, the actual track record.
schooner wrote: Wed Jul 10, 2019 11:39 am Again, if Nedsaid wants play factors or Hedgefundie is using leveraged ETFs to try and beat the market, more power to them. It's a whole different thing when it's being played on an institutional level.
We're 150 posts into this thread, and now you mention that your concern is factor tilting at the institutional level, not whether it may be appropriate for individuals to make that choice. Might have been nice to know at the beginning that that was your concern, otherwise it feels like you’re just moving the goalposts. As a counter however, long-term institutional investors are much more well-placed than individuals to take advantage of premia that can take decades to manifest.
schooner
Posts: 388
Joined: Sun Jun 09, 2019 8:27 am

Re: Why factor investing isn't working

Post by schooner »

TropikThunder wrote: Wed Jul 10, 2019 11:46 am
schooner wrote: Wed Jul 10, 2019 11:39 am The back tests looks great but the actual track record looks a bit shaky.
What does this mean? The back tests Larry showed you are with actual funds, not indexes. In other words, the actual track record.
schooner wrote: Wed Jul 10, 2019 11:39 am Again, if Nedsaid wants play factors or Hedgefundie is using leveraged ETFs to try and beat the market, more power to them. It's a whole different thing when it's being played on an institutional level.
We're 150 posts into this thread, and now you mention that your concern is factor tilting at the institutional level, not whether it may be appropriate for individuals to make that choice. Might have been nice to know at the beginning, otherwise it feels like you’re just moving the goalposts.
This entire thread is about why factor investing isn't working. The byline of the article is: "Pension funds and endowments aren’t getting the factor premiums they desire — just uncompensated risk."

Some of that institutional money is my money. It's only fair that I can voice my own interest. I am a stakeholder. I can change the goal posts. I can ask the tough questions. Because at the end of the day, folks like Larry are not playing around with their own money, they are playing around with my money.

If you want to play around with factors yourself, that's fine. The thread is just an intellectual exercise, an outlet of spirited debate :-)
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Why factor investing isn't working

Post by nedsaid »

schooner wrote: Wed Jul 10, 2019 11:39 am
larryswedroe wrote: Wed Jul 10, 2019 11:20 am schooner
I have shown you how it is improved results in both articles and my books. As I said, just read Reducing the Risk of Black Swans to find the historical evidence, using live funds, not indices. I've also shown you that over the past 20+ years factor based funds that are well designed have significantly outperformed TSM counterparts, but of course that will not be true over all periods, nor is it a guarantee it will in future, but based on the evidence, and IMO the logic it definitely puts the odds in your favor.

If not familiar I suggest you run the following as a simple test using the portfoliovisualizer tool you are familiar with
Portfolio A Vanguard US TSM 60%/Vanguard Intermediate Treasury 40%
Portfolio B DFSVX 40%/Vanguard Intermediate Treasury 60%

Run it from inception of DFSVX so you have the longest period possible, now over 25 years. Check the returns/SD, SR, factor loadings and best and worst case years. Then ask yourself which portfolio you would rather have owned.

Again, this is just example. I would not own such a portfolio as would do it globally including EM. But it's example of moving toward risk parity.
Then you can decide if you believe benefits of TSM (lowest costs and most tax efficient) more than the benefits of the other portfolio. And you cannot argue A is simpler, they both are just two fund portfolios, and globally can also do with just two equity funds.

Again, no right answer, because there are trade offs.

BTW, Note I have tried patiently to answer all your questions but I note you almost never answer the simple ones I ask you. Like why you believe 10 years is not long enough to give up on market beta but thought 10 years demonstrated value and size were not real factors. Just asking (:-)) It appears to me there are many inconsistencies in your explanations, like you believe markets are efficient but the much more volatile small and value stocks should not be expected to have higher returns, not higher risk adjusted returns, but higher returns? And the inconsistency on the 10 year issue.

Larry
That's completely fair - I'm not answering your questions because I don't know the answers. The argument is a bit asymmetric because you are the professional and I am the novice.

I just get a bit nervous about all of this factor stuff taking place at the institutional level - since a small share of that is money that's invested on my behalf. And if there is a change in strategy, it's only fair for me to ask the tough questions. I am a stakeholder.

The back tests looks great but the actual track record looks a bit shaky.

Again, if Nedsaid wants play factors or Hedgefundie is using leveraged ETFs to try and beat the market, more power to them. It's a whole different thing when it's being played on an institutional level.
My take is that during the long secular bear markets that diversifying across factors helps about 2/3 of the time. The Small/Value tilting would have helped during 1968-1984 and during 2000-2012. It would not have helped from 1929-1946. If you look at it in another way, just looking at the bear markets where you were down 50% or more: factor tilting would not have helped in the aftermath of 1929, would have helped in 1973-74, helped in 2000-2002, but did not help in 2008-2009. So it appears that depending on how you look at it, Small/Value factor tilting will help you 2/3 of the time during the long secular bear markets (where markets are essentially flat for years with big drops and rallies in between) or 1/2 of the time when the markets drop 50% or more.

So market history seems to indicate the potential for downside protection when markets go south, though as we can see it doesn't always work. Note the chart that shows decade by decade performance of the S&P 500, Large Value, Small Caps, and Small-Value stocks. You can see why Larry Swedroe makes the diversification by factors argument.

https://www.marketwatch.com/story/8-les ... 2014-11-19
Last edited by nedsaid on Wed Jul 10, 2019 12:09 pm, edited 2 times in total.
A fool and his money are good for business.
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: Why factor investing isn't working

Post by HomerJ »

TropikThunder wrote: Wed Jul 10, 2019 11:46 am
schooner wrote: Wed Jul 10, 2019 11:39 am The back tests looks great but the actual track record looks a bit shaky.
What does this mean? The back tests Larry showed you are with actual funds, not indexes. In other words, the actual track record.
Anyone can look backwards and find a portfolio that worked well.

The question is what were people's actual returns?
If not familiar I suggest you run the following as a simple test using the portfoliovisualizer tool you are familiar with
Portfolio A Vanguard US TSM 60%/Vanguard Intermediate Treasury 40%
Portfolio B DFSVX 40%/Vanguard Intermediate Treasury 60%

Run it from inception of DFSVX so you have the longest period possible, now over 25 years. Check the returns/SD, SR, factor loadings and best and worst case years. Then ask yourself which portfolio you would rather have owned.
A back-test on a fund should NEVER EVER be run from inception. Larry shouldn't post that.

Here's how a lot of funds work. A couple new funds come out. They are small, very few people are invested in them. One of them does terrible, it's closed a few years later, one does okay, stays open, but only grows slowly. One does great, has amazing 3-year or 5-year returns, and gets written up in a magazine or on the Internet as a "hot new fund".

A million people pile into it. The ACTUAL results for those people is the results from the day they invested, not inception. Very few people capture the returns "from inception".

In many cases, funds do much better those first couple of years when they are small, with less assets. Easier to find winning overlooked small stocks when the fund is small itself. Lots of times, as the fund bloats, it does less well going forward. Or for factor funds, when they are first in the market, capturing market inefficiencies. As more players jump in, the easy pickings become harder to find.

DFSVX only did okay in the 1990s compared to Total Stock Market (since the stock market in general was doing so well)

Where DFSVX really shined was in the early 2000s. The portfolio with DFSVX trounced the one with TSM for a few years (plus it was heavier on bonds and bonds did pretty well in the early 2000s)

So it was the fund that only did okay for a few years, but then it did finally have a hot 5 years where it did great.

So when did people pile into it? Before the great out-performance in the early 2000s? Or after? Because anyone who invested in it after 2006 because it did so well from 2000-2005, didn't capture any of that out-performance.

That's what we mean by "actual" results.

Larry deserves some kudos for recommending small-value in the 1990s (Did he? I seem to recall he did).

Recommending it in the mid to late 2000s (and through today) hasn't paid off (yet?) for actual investors who followed that advice.
Last edited by HomerJ on Wed Jul 10, 2019 12:29 pm, edited 1 time in total.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
DaufuskieNate
Posts: 605
Joined: Wed May 28, 2014 11:53 am

Re: Why factor investing isn't working

Post by DaufuskieNate »

Schooner - not trying to be a smart Alek here, but I don’t think I could have a strong opposing view on a topic where I considered myself a novice without answers to basic questions of logic.
stan1
Posts: 14246
Joined: Mon Oct 08, 2007 4:35 pm

Re: Why factor investing isn't working

Post by stan1 »

If it just comes down to "I don't agree and want to do something different". Fine, go ahead.
Warning: I am about 80% satisficer (accepting of good enough) and 20% maximizer
schooner
Posts: 388
Joined: Sun Jun 09, 2019 8:27 am

Re: Why factor investing isn't working

Post by schooner »

DaufuskieNate wrote: Wed Jul 10, 2019 12:36 pm Schooner - not trying to be a smart Alek here, but I don’t think I could have a strong opposing view on a topic where I considered myself a novice without answers to basic questions of logic.
I'm a novice relative to Larry Swedroe. He is an investment profession with decades of experience.

I have read numerous books by Bogle and follow the Boglehead philosophy. The low cost, passive approach of investing in market-cap indexes has helped millions of people just like me save for nearly 40 years.

I've also read interviews with both Sharpe and Fama where they discuss factors. And may I point out, they both acknowledge the academic nature of these factors. I believe Fama even recommended investors stick with a simple market weighted index (I'll look for the cite on that one).

So not a complete novice on the subject ;-)
DaufuskieNate
Posts: 605
Joined: Wed May 28, 2014 11:53 am

Re: Why factor investing isn't working

Post by DaufuskieNate »

schooner wrote: Wed Jul 10, 2019 12:43 pm
DaufuskieNate wrote: Wed Jul 10, 2019 12:36 pm Schooner - not trying to be a smart Alek here, but I don’t think I could have a strong opposing view on a topic where I considered myself a novice without answers to basic questions of logic.
I'm a novice relative to Larry Swedroe. He is an investment profession with decades of experience.

I have read numerous books by Bogle and follow the Boglehead philosophy. The low cost, passive approach of investing in market-cap indexes has helped millions of people just like me save for nearly 40 years.

I've also read interviews with both Sharpe and Fama where they discuss factors. And may I point out, they both acknowledge the academic nature of these factors. I believe Fama even recommended investors stick with a simple market weighted index (I'll look for the cite on that one).

So not a complete novice on the subject ;-)
So perhaps the answer is to just be content in your current understanding and decisions. If you want to learn more about factors you could expand your reading. Larry’s books on the subject are very good.
jbranx
Posts: 1544
Joined: Thu Feb 09, 2017 5:57 pm

Re: Why factor investing isn't working

Post by jbranx »

{For a second day in a row, this thread is requiring an inordinate amount of Moderator time because some members are focusing on insulting each other rather than on the topic. To keep the thread open, remember Rule 2(a):}
Discussions should be conducted without fondness for dispute or desire for victory. - Benjamin Franklin

This is a moderated forum. We expect this forum to be a place where people can feel comfortable asking questions and where debates and discussions are conducted in civil tones. Discussions are about issues, not people. If you disagree with an idea, go ahead and marshal all your forces against it. But do not confuse ideas with the person posting them. At all times we must conduct ourselves in a respectful manner to other posters. Attacks on individuals, insults, name calling, trolling, baiting or other attempts to sow dissension are not acceptable. If you feel that someone has attacked you or otherwise violated the policies of this forum, do not respond in kind. Instead, please click the report button on the offending post. This is the quickest method to notify the site moderators.

Remember that you can edit your own posts even after submitting them, so comments made in anger or haste can be rephrased before the problem escalates or a moderator has to step in.
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Why factor investing isn't working

Post by vineviz »

schooner wrote: Wed Jul 10, 2019 12:43 pm I have read numerous books by Bogle and follow the Boglehead philosophy. The low cost, passive approach of investing in market-cap indexes has helped millions of people just like me save for nearly 40 years.
I think it's great that you've read Bogle's books: there's a great deal of wisdom in them, and you're right that millions of people have benefited from his pioneering work in the field of low-cost passive investing.

On the other hand, there's a lot of great information about finance and investing that is NOT found in Bogle's books. Many of us are familiar with THAT information too and want to discuss it without being trolled by people who aren't sincerely curious.

I'll be the last one to say that investors need to go beyond basic market-cap weighted investing, but I DO think that people who are truly curious might benefit from learning from sources other than Bogle. As Harold Francis Callahan once said, "A man's got to know his limitations."
schooner wrote: Wed Jul 10, 2019 12:43 pmI've also read interviews with both Sharpe and Fama where they discuss factors. And may I point out, they both acknowledge the academic nature of these factors.
Then you also know that Eugene Fama is on DFA's Board of Directors and is one of their consultants.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
schooner
Posts: 388
Joined: Sun Jun 09, 2019 8:27 am

Re: Why factor investing isn't working

Post by schooner »

vineviz wrote: Wed Jul 10, 2019 1:04 pm
schooner wrote: Wed Jul 10, 2019 12:43 pm I have read numerous books by Bogle and follow the Boglehead philosophy. The low cost, passive approach of investing in market-cap indexes has helped millions of people just like me save for nearly 40 years.
I think it's great that you've read Bogle's books: there's a great deal of wisdom in them, and you're right that millions of people have benefited from his pioneering work in the field of low-cost passive investing.

On the other hand, there's a lot of great information about finance and investing that is NOT found in Bogle's books. Many of us are familiar with THAT information too and want to discuss it without being trolled by people who aren't sincerely curious.

I'll be the last one to say that investors need to go beyond basic market-cap weighted investing, but I DO think that people who are truly curious might benefit from learning from sources other than Bogle. As Harold Francis Callahan once said, "A man's got to know his limitations."
schooner wrote: Wed Jul 10, 2019 12:43 pmI've also read interviews with both Sharpe and Fama where they discuss factors. And may I point out, they both acknowledge the academic nature of these factors.
Then you also know that Eugene Fama is on DFA's Board of Directors and is one of their consultants.
Fama himself has warned against going beyond the 3 factors in anything more than an academic way:

"Many of these factors are not robust across countries and across time, or they are not actionable by real-world investors because they exist in micro-cap stocks that are highly illiquid."

https://blogs.cfainstitute.org/investor ... c-factors/

Furthermore, Sharpe and others take issue with the underlying data examined that reveal even these factors (I have included the entire quote for more nuance and fairness):

"There's a whole industry of turning out papers showing things "wrong" and "partially wrong" with the Fama- French study. I have not been part of that industry. I would only point out that during that period in the United States, value stocks did much better than growth. In the bear market of 1973 and 1974, people thought the world was coming to an end. It didn't come to an end. Surprise. The stocks that had been beaten down came back, and they came back a lot more than some of the growth stocks.

Maybe in an efficient market, small stocks would do better because they're illiquid, and people demand a premium for illiquidity That gets to be less compelling if you start thinking about mutual funds that package a bunch of small stocks and therefore make the illiquid liquid. As people figured that out, they'd put money into those funds, which would drive up the price of small stocks, and there goes the premium. For the value/growth effect, there's the behaviorist story that people overextrapolate. I have quite a bit of sympathy with that. I'm a bit of a fan of behavioral finance—the psychology of markets—so I don't dismiss that argument out of hand.

Since the studies about the size effect were published, small stocks have not done better than large stocks on average. Since the publicity about the value effect, value stocks haven't done as well as before around the world. So there's always the possibility that whatever these things were may have gone away, and that the publication of these studies may have helped them go away. It's too early to tell. It's a short data period. One would not want to infer too much, except that rushing to embrace those strategies has not turned out to be a very good idea, recently, certainly in the United States."

Source: https://web.stanford.edu/~wfsharpe/art/djam/djam.htm

Yes, the second source is 20 years old. But I'd argue we haven't learned too much since that point - small cap and value have had some periods of over performance and periods of under performance. We've had two bear markets and two bull markets. So what are we left with?
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Why factor investing isn't working

Post by vineviz »

schooner wrote: Wed Jul 10, 2019 2:11 pm Fama himself has warned against going beyond the 3 factors in anything more than an academic way:
That's not what he said. What he said is the same thing that Larry said earlier: don't invest using factors that fail to past reasonable robustness checks. There are definitely more than three that pass these tests.
schooner wrote: Wed Jul 10, 2019 2:11 pm Furthermore, Sharpe and others take issue with the underlying data examined that reveal even these factors.
In the late 1990s, there was a cottage industry built around claiming that factors were dead. Sharpe (who is not an impartial observer, it should be noted) was merely one of many making similar claims. What do things look like since this interview was published in May, 1998?

Image
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
schooner
Posts: 388
Joined: Sun Jun 09, 2019 8:27 am

Re: Why factor investing isn't working

Post by schooner »

vineviz wrote: Wed Jul 10, 2019 2:34 pm
schooner wrote: Wed Jul 10, 2019 2:11 pm Fama himself has warned against going beyond the 3 factors in anything more than an academic way:
That's not what he said. What he said is the same thing that Larry said earlier: don't invest using factors that fail to past reasonable robustness checks. There are definitely more than three that pass these tests.
schooner wrote: Wed Jul 10, 2019 2:11 pm Furthermore, Sharpe and others take issue with the underlying data examined that reveal even these factors.
In the late 1990s, there was a cottage industry built around claiming that factors were dead. Sharpe (who is not an impartial observer, it should be noted) was merely one of many making similar claims. What do things look like since this interview was published in May, 1998?

Image
1) The article states clearly: "Whatever the approach, Fama returns to the same core components of underlying risk factors — market, size, and value, with market risk measured by the standard market beta, size by the relative market capitalization, and value by the book-to-market ratio."

2) I can't argue with the performance of that specific time frame. But one observation - the small cap fund you reference (NAESX) has returned 10.6%/year since it's inception in 1960. The SP500 has returned 10.25%/year during the same time frame. Is that evidence of a small cap factor? Maybe?
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Why factor investing isn't working

Post by vineviz »

schooner wrote: Wed Jul 10, 2019 2:49 pm 1) The article states clearly . . .
That's someone telling you what THEY think Fama said. Why rely on that when you can read what he ACTUALLY said?
schooner wrote: Wed Jul 10, 2019 2:49 pmBut one observation - the small cap fund you reference (NAESX) has returned 10.6%/year since it's inception in 1960. The SP500 has returned 10.25%/year during the same time frame. Is that evidence of a small cap factor? Maybe?
Hard to say: the fund was not managed by Vanguard nor was it an index fund before the mid-198Os.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: Why factor investing isn't working

Post by larryswedroe »

Schooner
Either you are just cherry picking and using what you like or you are not aware of what Fama does in practice
Fama himself has warned against going beyond the 3 factors in anything more than an academic way:

"Many of these factors are not robust across countries and across time, or they are not actionable by real-world investors because they exist in micro-cap stocks that are highly illiquid."
Fama was actually the first to write about the profitability factor and DFA incorporated it into their strategies under his direction. You are totally mischaracterizing what he was saying, again choosing to read what you want. Fama was saying the exact same thing I was saying, many if not most factors don't pass all the tests you would want for investability. So be careful. Under Fama Dimensional also uses momentum.

Larry
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: Why factor investing isn't working

Post by larryswedroe »

nedsaid
The factors will help you perhaps even all the time in down markets if used the way I have suggested. That is consider lower equity exposure as the equities you do hold have higher expected returns allowing you to hold less equity and more safe bonds. Moving more to that risk parity idea. And that is the whole premise of Reducing the Risk of Black Swans
Larry
schooner
Posts: 388
Joined: Sun Jun 09, 2019 8:27 am

Re: Why factor investing isn't working

Post by schooner »

vineviz wrote: Wed Jul 10, 2019 3:13 pm
schooner wrote: Wed Jul 10, 2019 2:49 pm 1) The article states clearly . . .
That's someone telling you what THEY think Fama said. Why rely on that when you can read what he ACTUALLY said?
schooner wrote: Wed Jul 10, 2019 2:49 pmBut one observation - the small cap fund you reference (NAESX) has returned 10.6%/year since it's inception in 1960. The SP500 has returned 10.25%/year during the same time frame. Is that evidence of a small cap factor? Maybe?
Hard to say: the fund was not managed by Vanguard nor was it an index fund before the mid-198Os.
Re Fama: It's the CFA Institute. I'll trust their reporting of the interview.

Re Performance Returns: Exactly my point. Well, Sharpe's point. The data is messy and there's not very much of it. Maybe factors exist. But they seem to be pretty ephemeral. Like mirages in a journey through the desert.

DFA takes a market cap index and tilts it towards these factors. DFA funds are slightly more expensive than the traditional Vanguard Funds. DFA also requires most of its customers to work through an advisor who charges an additional fee. Even assuming these factors exist, is it worth the extra cost to investors?

The performance and cost hurdles are real. And I don't think the factor track record has passed either one of these hurdles.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: Why factor investing isn't working

Post by larryswedroe »

Schooner
T
hat's completely fair - I'm not answering your questions because I don't know the answers. The argument is a bit asymmetric because you are the professional and I am the novice.

I just get a bit nervous about all of this factor stuff taking place at the institutional level - since a small share of that is money that's invested on my behalf. And if there is a change in strategy, it's only fair for me to ask the tough questions. I am a stakeholder.

The back tests looks great but the actual track record looks a bit shaky.
Two things, the questions I asked do not require any knowledge of investing really beyond what you have. I was simply pointing out inconsistencies in your arguments, like is 10 years long enough or not.

As to the back tests I showed--they are not backtests but live portfolios of funds I personally was investing in and those who read my books were ableo t invest in. My first book came out in 1998 and I was using DFA funds for several years before that. Homer's comments are just wrong (though they might be applicable to active funds as you get that survivorship bias problem and how would you know to choose them when they first came out). With passive funds it is very different. Just like you could use any Vanguard index fund from inception (:-))

Let me just close (hopefully) with this, It's very important to be skeptical. I think the fact that I've told you that among the 600 factors in the zoo of factors I believe only about a handful are worth considering. That's a 99% percent rejection rate. Would you agree that shows skepticism? The others pass all the test with flying colors. Should that not also tell you that they are not the results of data mining as well as white collar declared? What are the odds that value (buying cheap outperforms buying expensive) would work not only in virtually every country in the world but across sectors as well, and even across asset classes, and even across many different metrics? That's how you gain confidence it is LIKELY to work in future, just like you have with market beta.

And if you really are serious about this stuff try reading my factor book, we summarize the findings of more than 100 academic papers in one place for you. So you can make the informed decision.


And finally it seems hard for you to admit that the portfolio I had suggested in Kiplingers underperformed NOT BECAUSE of factor exposure as you claimed, but because it happened to have a market cap weighting to international. Others showed you that. Note also I did not "walk back" my recommendation, only pointed out that I was restricted in my recommendation by the reporter. It was not the portfolio I would have recommended. With that said, the results would have been similar, as the factors were negative in US but positive internationally, about a net wash.
Best wishes
Larry
User avatar
Taylor Larimore
Posts: 32842
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

Questionable chart

Post by Taylor Larimore »

vineviz:

Sorry, but the graph you posted does not reflect the bear market of 2000-2002 when Vanguard's 500 Index Fund plunged -44.8% during the 2000-2002 bear market. That terrible bear market looks like a minor decline in the chart?

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
schooner
Posts: 388
Joined: Sun Jun 09, 2019 8:27 am

Re: Why factor investing isn't working

Post by schooner »

larryswedroe wrote: Wed Jul 10, 2019 3:31 pm Schooner
T
hat's completely fair - I'm not answering your questions because I don't know the answers. The argument is a bit asymmetric because you are the professional and I am the novice.

I just get a bit nervous about all of this factor stuff taking place at the institutional level - since a small share of that is money that's invested on my behalf. And if there is a change in strategy, it's only fair for me to ask the tough questions. I am a stakeholder.

The back tests looks great but the actual track record looks a bit shaky.
Two things, the questions I asked do not require any knowledge of investing really beyond what you have. I was simply pointing out inconsistencies in your arguments, like is 10 years long enough or not.

As to the back tests I showed--they are not backtests but live portfolios of funds I personally was investing in and those who read my books were ableo t invest in. My first book came out in 1998 and I was using DFA funds for several years before that. Homer's comments are just wrong (though they might be applicable to active funds as you get that survivorship bias problem and how would you know to choose them when they first came out). With passive funds it is very different. Just like you could use any Vanguard index fund from inception (:-))

Let me just close (hopefully) with this, It's very important to be skeptical. I think the fact that I've told you that among the 600 factors in the zoo of factors I believe only about a handful are worth considering. That's a 99% percent rejection rate. Would you agree that shows skepticism? The others pass all the test with flying colors. Should that not also tell you that they are not the results of data mining as well as white collar declared? What are the odds that value (buying cheap outperforms buying expensive) would work not only in virtually every country in the world but across sectors as well, and even across asset classes, and even across many different metrics? That's how you gain confidence it is LIKELY to work in future, just like you have with market beta.

And if you really are serious about this stuff try reading my factor book, we summarize the findings of more than 100 academic papers in one place for you. So you can make the informed decision.


And finally it seems hard for you to admit that the portfolio I had suggested in Kiplingers underperformed NOT BECAUSE of factor exposure as you claimed, but because it happened to have a market cap weighting to international. Others showed you that. Note also I did not "walk back" my recommendation, only pointed out that I was restricted in my recommendation by the reporter. It was not the portfolio I would have recommended. With that said, the results would have been similar, as the factors were negative in US but positive internationally, about a net wash.
Best wishes
Larry
Larry, I appreciate the response and will check out your book. And kudos for jumping into the lion's den on this one. I may not agree with you but I respect you for that. Also, I have no problem with other's posting your writings on Bogleheads. Always good to be challenged on your ideas. It is a discussion board after all ;-)

Sincerely,
Skeptical
schooner
Posts: 388
Joined: Sun Jun 09, 2019 8:27 am

Re: Questionable chart

Post by schooner »

Taylor Larimore wrote: Wed Jul 10, 2019 3:34 pm vineviz:

Sorry, but the graph you posted does not reflect the bear market of 2000-2002 when Vanguard's 500 Index Fund plunged -44.8% during the 2000-2002 bear market. That terrible bear market looks like a minor decline in the chart?

Best wishes.
Taylor
I think it's because the Y axis is not presented on a log scale. Not incorrect, but just a different way of looking at it.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: Why factor investing isn't working

Post by larryswedroe »

Skeptical
One of the reasons I used to post here all the time was that debating these issues helps you learn, being forced to defend your ideas. It also allows me to see all of what are IMO generally biases people have simply because they are humans and subject to those biases. What helps people prevent the biases is being aware of them. Hopefully I have helped people by pointing the biases out, as well as writing about them in my book Investment Mistakes Even Smart People Make and how to avoid them. If interested in behavioral finance you might check that out as well.
Also note I am always happy to answer questions from interested readers who can email me at lswedroe@bamadvisor.com

Best wishes
Larry
abc132
Posts: 2435
Joined: Thu Oct 18, 2018 1:11 am

Re: Why factor investing isn't working

Post by abc132 »

nedsaid wrote: Tue Jul 09, 2019 12:43 pm
abc132 wrote: Tue Jul 09, 2019 12:28 pm I'd like to see a reply to the idea that for anyone tilting a particular direction, someone is also tilting in the opposite direction.

Thanks in advance.
Actually, I have posted on this topic many times. A month or two ago, I looked at the 10 year performance of the S&P 500, Vanguard Value Index, Vanguard Growth Index, Vanguard Small Value Index, and Vanguard Small Growth Index. Going from memory Vanguard Value Index had returns of 14%, the S&P 500 15%, and Vanguard Growth Index 16%. All of the Vanguard Growth and Value indexes outperformed the S&P 500 except for the Vanguard Value Index which represents Large Value. The excess performance of the S&P 500 and the Growth Index over the Value index comes from the High Tech/Internet parts of the market, particularly the FAANG stocks. In fact, the Vanguard Growth Index is now in non-diversified status because of heavy weightings in just a few stocks. We have been in a Large Growth stock market since the 2008-2009 financial crisis and bear market.

The Vanguard Small Value Index and the Vanguard Small Growth Indexes also beat the S&P 500 but not by very much. So the idea that Small/Value tilts have been awful is just ridiculous, they have pretty much tracked the S&P 500 over the last 10 years. What has happened is that Small/Value has not done so well over the last five years. Over the last 10 years, Large Value has trailed the broad market indexes by about 1% and Large Growth by 2% a year.

We have been in a Large Growth market and that is the other side of the trade.
Thanks for the reply. If I exclude the performance comparison, which is not related to my comments and is within the 10 year period Larry says we can't use, the point I was interested in clarifying is that someone is on the opposite side of every tilt. In a scenario where we all have access to the same information, it simply comes down to who has the best strategy, which is difficult to know in advance, for our own individual time horizon. We should all be open to the idea that factor tilting may work out moving forward, and it may not. It likely will depend on how you do it, as some tilts will work out and some will not, so there is nuance. I certainly appreciate the time/effort spent here to educate Bogleheads and discuss factors.

The following is certainly not directed at any individual:

I wish there were experts willing to make their recommendations in a clearly testable manner moving forward, without all the ambiguity about why a particular past or future recommendation should or shouldn't count for anything. That's why the scientific method was created - so that what is right or wrong is more than ever-changing opinions and rationalizations for why ones own position is correct no matter what the results. There is too much contradiction between stating we can't look at data within 10 or 20 or 30 years and then routinely using data within that same time period as if it means something. Data seams to mean something when it supports ones own opinion, and can be ignored when it does not. It's far too easy to grab 10 or 20 or 30 year data that shows exactly the opposite results, depending on the data selection.

A starting point would be that every recommendation made by someone declaring themselves an expert should come with a way to test or score any future recommendation (when and how and over what time period it can be tested), and they should be evaluated based on the prediction at the time is was made, and not on some after the fact rationalization. Nobody should be put in to the expert category that is not willing to do this. Just my two cents that if someone wants to get paid for their success, they ought to be able to numerically define with metrics in advance what that success means, and be held accountable to their own metrics.

The actionable portion here is to clearly predefine the future success or failure of a method, and to really limit our individual desire to equate only limited and selected portions of the past to the future in ways that fit our own narrative and predispositions. It's too easy to fool ourselves, with the evidence being thousands of people with contrary positions all believing they are correct, and the others are the fools, each with data to back up their position.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Why factor investing isn't working

Post by nedsaid »

abc132 wrote: Wed Jul 10, 2019 4:52 pm
nedsaid wrote: Tue Jul 09, 2019 12:43 pm
abc132 wrote: Tue Jul 09, 2019 12:28 pm I'd like to see a reply to the idea that for anyone tilting a particular direction, someone is also tilting in the opposite direction.

Thanks in advance.
Actually, I have posted on this topic many times. A month or two ago, I looked at the 10 year performance of the S&P 500, Vanguard Value Index, Vanguard Growth Index, Vanguard Small Value Index, and Vanguard Small Growth Index. Going from memory Vanguard Value Index had returns of 14%, the S&P 500 15%, and Vanguard Growth Index 16%. All of the Vanguard Growth and Value indexes outperformed the S&P 500 except for the Vanguard Value Index which represents Large Value. The excess performance of the S&P 500 and the Growth Index over the Value index comes from the High Tech/Internet parts of the market, particularly the FAANG stocks. In fact, the Vanguard Growth Index is now in non-diversified status because of heavy weightings in just a few stocks. We have been in a Large Growth stock market since the 2008-2009 financial crisis and bear market.

The Vanguard Small Value Index and the Vanguard Small Growth Indexes also beat the S&P 500 but not by very much. So the idea that Small/Value tilts have been awful is just ridiculous, they have pretty much tracked the S&P 500 over the last 10 years. What has happened is that Small/Value has not done so well over the last five years. Over the last 10 years, Large Value has trailed the broad market indexes by about 1% and Large Growth by 2% a year.

We have been in a Large Growth market and that is the other side of the trade.
Thanks for the reply. If I exclude the performance comparison, which is not related to my comments and is within the 10 year period Larry says we can't use, the point I was interested in clarifying is that someone is on the opposite side of every tilt. In a scenario where we all have access to the same information, it simply comes down to who has the best strategy, which is difficult to know in advance, for our own individual time horizon. We should all be open to the idea that factor tilting may work out moving forward, and it may not. It likely will depend on how you do it, as some tilts will work out and some will not, so there is nuance. I certainly appreciate the time/effort spent here to educate Bogleheads and discuss factors.

Nedsaid: We know that investors as a whole own the market. For me, that means that I own more Small-Cap stocks than the market 17% vs. 6% for Total Market, my Mid-Caps are 19% vs. 18% for Total Market so obviously I am underweight Large-Cap. I also hold 23% Large Value, 5% Mid Value, and 5% Small Value vs. 25% Large Value in Total Market, 5% Mid Value in Total Market, and 2% for Small Value in Total Market. In reality, I have a Mid/Small-Cap tilt but I have the same Value weightings as the Market itself other than I have a bit less Large Value and a bit more Small Value. Over time my portfolio lost its Value tilt, at one time 8% of my stocks were Small Value, now it is 5%. I hold individual stocks and some of those have drifted from Value to Core. One reason I am in a Growth to Value rebalancing program, trying to retilt my portfolio. Who is on the other side of my tilts, boy that is hard to say. My guess it would be other individuals who presumably have a taste for Large Cap Growth.

The following is certainly not directed at any individual:

I wish there were experts willing to make their recommendations in a clearly testable manner moving forward, without all the ambiguity about why a particular past or future recommendation should or shouldn't count for anything. That's why the scientific method was created - so that what is right or wrong is more than ever-changing opinions and rationalizations for why ones own position is correct no matter what the results. There is too much contradiction between stating we can't look at data within 10 or 20 or 30 years and then routinely using data within that same time period as if it means something. Data seams to mean something when it supports ones own opinion, and can be ignored when it does not. It's far too easy to grab 10 or 20 or 30 year data that shows exactly the opposite results, depending on the data selection.

A starting point would be that every recommendation made by someone declaring themselves an expert should come with a way to test or score any future recommendation (when and how and over what time period it can be tested), and they should be evaluated based on the prediction at the time is was made, and not on some after the fact rationalization. Nobody should be put in to the expert category that is not willing to do this. Just my two cents that if someone wants to get paid for their success, they ought to be able to numerically define with metrics in advance what that success means, and be held accountable to their own metrics.

Nedsaid: For whatever it is worth, you can check my New 'Doo thread where I post about my own portfolio. You can see what I have done and how I have been performing. I have been recommending Large Value for probably 3 years now and recently and somewhat jokingly have put a buy on Large Value, Mid Value, and Small Value in true Jim Cramer fashion hitting the buy button. Obviously, Large Growth continues to outperform and that could continue for a while. As Yogi Berra said, "Making predictions is hard, particularly about the future."

The actionable portion here is to clearly predefine the future success or failure of a method, and to really limit our individual desire to equate only limited and selected portions of the past to the future in ways that fit our own narrative and predispositions. It's too easy to fool ourselves, with the evidence being thousands of people with contrary positions all believing they are correct, and the others are the fools, each with data to back up their position.

Nedsaid: Hard to know even if you have been successful, had quite a discussion with Larry Swedroe about benchmarking and he warned me against resulting. You can check that out in my New 'Doo thread as I summarized that discussion hoping someone would gain from that. I compared my results vs. the 3 fund portfolio and did additional factor analysis for 2017 and 2018. Pretty much seeing if my portfolio is performing as one would expect. It has. But factor tilts don't work in a Large Growth environment.

I have to tell you that measuring my investment performance was difficult enough, particularly with Compound Annual Growth Rate which I calculate with Excel. Quicken calculates Internal Rate of Return and does a good job but fumbles Growth of $10,000 which gets confused as you add money to investment accounts. Benchmarking was difficult, partly because my asset allocation drifts a bit. I also benchmarked vs. 3 fund portfolio but Larry said that was all wrong. It did give me insight as to the effect of factors. Also have done comparisons with the "Lazy Portfolios" on MarketWatch. Exact comparisons are hard to make partly because pre-2013, I had a very, very relaxed attitude towards rebalancing, before then asset allocation drifted. To it is like comparing a moving target allocation vs. a fixed allocation and that isn't an accurate comparison. So pretty much a lot of eyeballing but not a terrifically good process. But that was the best I could do.

What I can say is that I have a sense that I underperformed the 3 fund portfolio during the 1990's because of my Value orientation and refusal to chase High Tech/Internet. I outperformed from 2000-2007 and have underperformed since the 2008-2009 financial crisis. Hardly surprising as I have a Value orientation and don't performance chase.

So my answers aren't all that good, best I can do is give you a sense based upon my experience and my feeble attempts at benchmarking.
A fool and his money are good for business.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: Why factor investing isn't working

Post by larryswedroe »

abc
it simply comes down to who has the best strategy, which is difficult to know in advance
This is the wrong way to frame the issue. And it is easy to demonstrate why as it is the mistake known as resulting. The "best" or right strategy is known in advance, what we don't know is if it will result in the best outcome because we cannot know the future.

Consider the following: You are 35 married with two young kids. You decide to not buy life insurance because you learn the odds of you dying in the next 10 years are extremely low (knowing that if you did die early the family would not be able to have acceptable lifestyle). You don't die. Was the strategy good or bad? Resulting would say yes, good strategy. But that clearly is wrong because we don't know which universe will show up.

Or consider the employee at Intel who in 1999 knowing what a great company he worked for puts 100% of his assets in the stock, which immediately collapsed and fell about 85% and has never recovered to even the nominal value. Was strategy good or bad? Did the outcome matter in judging the quality of the decision? If he had put all the money in google instead would that have made it a good decision?

That's the problem, sadly most people base their judgment on the outcome, which is clearly wrong when you don't know the future. That's what the book Thinking In Bets is all about, by poker expert.

In a world of uncertainty the only right way to judge the merits of a strategy is before you know the outcome, not after.

Here's one of my favorite quotes, about managing risks.

“Individual decisions can be badly thought through, and yet be successful, or exceedingly well thought through, but be unsuccessful, because the recognized possibility of failure in fact occurs. But over time, more thoughtful decision-making will lead to better results, and more thoughtful decision-making can be encouraged by evaluating decisions on how well they were made rather than on outcome.”

– Robert Rubin,
Former co-chairman of Goldman Sachs

I've spent my entire career either managing risks for some of the largest companies in the world or advising some of the largest companies on managing financial risks, and now for the last 25 years individuals and institutional investors. And that is one of the most important lesson I think can be taught--to judge a strategy by it's quality of thinking, not by its outcome

Hope that is helpful
Larry
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Why factor investing isn't working

Post by vineviz »

larryswedroe wrote: Wed Jul 10, 2019 8:04 pm That's the problem, sadly most people base their judgment on the outcome, which is clearly wrong when you don't know the future. That's what the book Thinking In Bets is all about, by poker expert.
This is a great book, by the way, and one that I think everyone should read.

For those who listen to Podcasts, The Unmistakable Creative Podcast had author Annie Duke on in May: https://unmistakablecreative.com/podcas ... nnie-duke/
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: Why factor investing isn't working

Post by HomerJ »

larryswedroe wrote: Wed Jul 10, 2019 8:04 pm abc
it simply comes down to who has the best strategy, which is difficult to know in advance
This is the wrong way to frame the issue. And it is easy to demonstrate why as it is the mistake known as resulting. The "best" or right strategy is known in advance, what we don't know is if it will result in the best outcome because we cannot know the future.
You are WAY too confident of your models.

It absolutely is correct that your models may be right, and we just have had good luck.. That's absolutely possible. Maybe even probable.

You haven't made any promises of certainty... You've just postulated probabilities.

You predicted 4.5% real a year going forward in 2011, and instead we've gotten 12.5% real... BUT you also said (when pressed) "plus/minus 8%" so technically, you're still technically correct.

But there's ALSO a chance your models are missing a variable or a variable has changed, and the "best" or right strategy is no longer what it was 10 years ago.

The fact that you cannot conceive of the possibility that your models are wrong is what keeps me from following them.

You've changed your mind in the past because of new data (CCFs).... Why are you so certain new data won't appear in the future that will change your mind about factors?

I respect a lot of your writing, but you are NEVER WRONG. No economist should ever consider themselves never wrong. You should have some doubt in your models.

If you create a model of flooding, and a 100-year flood happens 3 times in a decade, a reasonable person is going to evaluate that model. Maybe it has a flaw, or maybe it was constructed perfectly based on past data, but something changed?

I've spent my entire career either managing risks for some of the largest companies in the world or advising some of the largest companies on managing financial risks, and now for the last 25 years individuals and institutional investors. And that is one of the most important lesson I think can be taught--to judge a strategy by it's quality of thinking, not by its outcome
I hear what you're saying, but I can't retire on the quality of your thinking.

Respectfully...
Last edited by HomerJ on Wed Jul 10, 2019 10:02 pm, edited 2 times in total.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Why factor investing isn't working

Post by vineviz »

HomerJ wrote: Wed Jul 10, 2019 8:36 pm But there's ALSO a chance your models are missing a variable or a variable has changed, and the "best" or right strategy is no longer what it was 10 years ago.
Yes. And?

If you make the best decision that you can today, based on the facts you have today, then your strategy is the optimal strategy.

Bad luck tomorrow doesn't mean the strategy chosen today was wrong, nor does the appearance of new facts the day after tomorrow.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: Why factor investing isn't working

Post by HomerJ »

vineviz wrote: Wed Jul 10, 2019 8:46 pm
HomerJ wrote: Wed Jul 10, 2019 8:36 pm But there's ALSO a chance your models are missing a variable or a variable has changed, and the "best" or right strategy is no longer what it was 10 years ago.
Yes. And?

If you make the best decision that you can today, based on the facts you have today, then your strategy is the optimal strategy.

Bad luck tomorrow doesn't mean the strategy chosen today was wrong, nor does the appearance of new facts the day after tomorrow.
"New" facts may indeed just mean the "old" facts were incorrect, even at the time.

Nothing is certain when dealing with economics. There is not enough data. "New" facts may just mean you have more data now, and you never should have been so SURE with the old data in the first place.

I disagree that any economist can make the "best" decision today. There are too many unknown variables. Too many known variables with wide dispersions. No one knows the true odds.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Why factor investing isn't working

Post by vineviz »

HomerJ wrote: Wed Jul 10, 2019 8:53 pm I disagree that any economist can make the "best" decision today. There are too many unknown variables. Too many known variables with wide dispersions. No one knows the true odds.
I don't think you're using the word "best" the way that the rest of the world uses that word. Making the best possible decision doesn't require perfection, and no one here is implying that the best decision is perfect. That requires omniscience and prescience, neither of which anyone possesses. Not even economists.

Instead, what I think Larry was suggesting is that the only reasonable way to evaluate the quality of a decision is based on the information available at the time the decision was made.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: Why factor investing isn't working

Post by HomerJ »

vineviz wrote: Wed Jul 10, 2019 9:06 pm
HomerJ wrote: Wed Jul 10, 2019 8:53 pm I disagree that any economist can make the "best" decision today. There are too many unknown variables. Too many known variables with wide dispersions. No one knows the true odds.
I don't think you're using the word "best" the way that the rest of the world uses that word. Making the best possible decision doesn't require perfection, and no one here is implying that the best decision is perfect. That requires omniscience and prescience, neither of which anyone possesses. Not even economists.

Instead, what I think Larry was suggesting is that the only reasonable way to evaluate the quality of a decision is based on the information available at the time the decision was made.
But if you KNOW the information for sure isn't 100% complete, should you really move a couple million dollars around based on it?

What if you SUSPECT the information isn't even 50% complete?

Maybe you should just stick with "good enough most of the time" instead of going for "best based on the information known today that may possibly change tomorrow"

What makes it REALLY hard is if ten years go by, it starts to look like the "best" strategy from 10 years ago isn't working, maybe information has changed, you think, but the guy who sold it to you tells you you're wrong to even question it.

It doesn't matter what's actually happening, just that the plan 10 years ago was the "best" one based on the data known at that time. That's all that matters, you're told, not the result.

That's a really hard investment plan to stick with.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Why factor investing isn't working

Post by willthrill81 »

HomerJ wrote: Wed Jul 10, 2019 12:17 pmDFSVX only did okay in the 1990s compared to Total Stock Market (since the stock market in general was doing so well)

Where DFSVX really shined was in the early 2000s. The portfolio with DFSVX trounced the one with TSM for a few years (plus it was heavier on bonds and bonds did pretty well in the early 2000s)
When I compare DFSVX to VISVX from 1999-2019, the former had .6% higher annual returns but was more volatile and had a deeper max. drawdown as well. The Sharpe ratio was slightly higher for the VISVX. So it seems that the DFA fund just won out a bit by increasing the volatility.
The Sensible Steward
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Why factor investing isn't working

Post by vineviz »

HomerJ wrote: Wed Jul 10, 2019 9:25 pm But if you KNOW the information for sure isn't 100% complete, should you really move a couple million dollars around based on it?
That's information available at the time the decision was made.
HomerJ wrote: Wed Jul 10, 2019 9:25 pm What if you SUSPECT the information isn't even 50% complete?
That's information available at the time the decision was made.
HomerJ wrote: Wed Jul 10, 2019 9:25 pmMaybe you should just stick with "good enough most of the time" instead of going for "best based on the information known today that may possibly change tomorrow"
Wouldn't it be likely that those are, in fact, just two ways to describe the same decision?
HomerJ wrote: Wed Jul 10, 2019 9:25 pmWhat makes it REALLY hard is if ten years go by, it starts to look like the "best" strategy from 10 years ago isn't working, maybe information has changed, you think, but the guy who sold it to you tells you you're wrong to even question it.
Wouldn't that be LESS hard if you had made the decision using all the information available to you at the time you made the decision, rather than just buying what the salesman was selling?
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: Why factor investing isn't working

Post by HomerJ »

willthrill81 wrote: Wed Jul 10, 2019 9:37 pm
HomerJ wrote: Wed Jul 10, 2019 12:17 pmDFSVX only did okay in the 1990s compared to Total Stock Market (since the stock market in general was doing so well)

Where DFSVX really shined was in the early 2000s. The portfolio with DFSVX trounced the one with TSM for a few years (plus it was heavier on bonds and bonds did pretty well in the early 2000s)
When I compare DFSVX to VISVX from 1999-2019, the former had .6% higher annual returns but was more volatile and had a deeper max. drawdown as well. The Sharpe ratio was slightly higher for the VISVX. So it seems that the DFA fund just won out a bit by increasing the volatility.
Yes, you started the backtest right before it had it's great run (2000-2005).

Most investors would have jumped in after it made the "best 5-year return" list in 2006, not in 1999 when it was trailing Total Market.

2006-2019, it didn't do very well compared to Total Market.

Yes, if you get into a fund right BEFORE it starts its good run, you get good results.

That's not how the real world works. But that's how every BACKTEST works.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Why factor investing isn't working

Post by willthrill81 »

HomerJ wrote: Wed Jul 10, 2019 9:45 pm
willthrill81 wrote: Wed Jul 10, 2019 9:37 pm
HomerJ wrote: Wed Jul 10, 2019 12:17 pmDFSVX only did okay in the 1990s compared to Total Stock Market (since the stock market in general was doing so well)

Where DFSVX really shined was in the early 2000s. The portfolio with DFSVX trounced the one with TSM for a few years (plus it was heavier on bonds and bonds did pretty well in the early 2000s)
When I compare DFSVX to VISVX from 1999-2019, the former had .6% higher annual returns but was more volatile and had a deeper max. drawdown as well. The Sharpe ratio was slightly higher for the VISVX. So it seems that the DFA fund just won out a bit by increasing the volatility.
Yes, you started the backtest right before it had it's great run (2000-2005).

Most investors would have jumped in after it made the "best 5-year return" list in 2006, not in 1999 when it was trailing Total Market.

2006-2019, it didn't do very well compared to Total Market.

Yes, if you get into a fund right BEFORE it starts its good run, you get good results.

That's not how the real world works. But that's how every BACKTEST works.
To be fair, my backtest was using all available info in Portfolio Visualizer, but your point is well taken. Mine was that over the last 20 years, there hasn't been anything special about DFSVX compared to the lower cost and more available Vanguard equivalent.
The Sensible Steward
fennewaldaj
Posts: 1097
Joined: Sun Oct 22, 2017 11:30 pm

Re: Why factor investing isn't working

Post by fennewaldaj »

abc132 wrote: Wed Jul 10, 2019 4:52 pm That's why the scientific method was created - so that what is right or wrong is more than ever-changing opinions and rationalizations for why ones own position is correct no matter what the results. There is too much contradiction between stating we can't look at data within 10 or 20 or 30 years and then routinely using data within that same time period as if it means something. Data seams to mean something when it supports ones own opinion, and can be ignored when it does not. It's far too easy to grab 10 or 20 or 30 year data that shows exactly the opposite results, depending on the data selection.
If you ever look into philosophy of science you will learn their is really no straightforward scientific method even in the natural sciences.
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: Why factor investing isn't working

Post by HomerJ »

vineviz wrote: Wed Jul 10, 2019 9:42 pm
HomerJ wrote: Wed Jul 10, 2019 9:25 pmWhat makes it REALLY hard is if ten years go by, it starts to look like the "best" strategy from 10 years ago isn't working, maybe information has changed, you think, but the guy who sold it to you tells you you're wrong to even question it.
Wouldn't that be LESS hard if you had made the decision using all the information available to you at the time you made the decision, rather than just buying what the salesman was selling?
We don't have all the information. What we have is a summary of 100 academic research papers that was written for us.

99.9% of us aren't going to read the 100 academic research papers ourselves AND be able to understand them well enough to make the decision with all the information that is available at the time.

We don't have all the information. We have the summary of all the information.

(Information that might be end up being wrong later).

We're told to trust the summary, and when we question the results 10 years later, we're told we should trust that the strategy was the best at the time.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: Why factor investing isn't working

Post by HomerJ »

fennewaldaj wrote: Wed Jul 10, 2019 9:51 pm
abc132 wrote: Wed Jul 10, 2019 4:52 pm That's why the scientific method was created - so that what is right or wrong is more than ever-changing opinions and rationalizations for why ones own position is correct no matter what the results. There is too much contradiction between stating we can't look at data within 10 or 20 or 30 years and then routinely using data within that same time period as if it means something. Data seams to mean something when it supports ones own opinion, and can be ignored when it does not. It's far too easy to grab 10 or 20 or 30 year data that shows exactly the opposite results, depending on the data selection.
If you ever look into philosophy of science you will learn their is really no straightforward scientific method even in the natural sciences.
Repeatable experiments is something I would suggest.

But I admit that not all sciences have that available. Geology, for instance. Yet we still trust discoveries from geologists about the ancient geological eras, for instance.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Why factor investing isn't working

Post by willthrill81 »

HomerJ wrote: Wed Jul 10, 2019 9:59 pm
fennewaldaj wrote: Wed Jul 10, 2019 9:51 pm
abc132 wrote: Wed Jul 10, 2019 4:52 pm That's why the scientific method was created - so that what is right or wrong is more than ever-changing opinions and rationalizations for why ones own position is correct no matter what the results. There is too much contradiction between stating we can't look at data within 10 or 20 or 30 years and then routinely using data within that same time period as if it means something. Data seams to mean something when it supports ones own opinion, and can be ignored when it does not. It's far too easy to grab 10 or 20 or 30 year data that shows exactly the opposite results, depending on the data selection.
If you ever look into philosophy of science you will learn their is really no straightforward scientific method even in the natural sciences.
Repeatable experiments is something I would suggest.

But I admit that not all sciences have that available. Geology, for instance. Yet we still trust discoveries from geologists about the ancient geological eras, for instance.
Actually, the problem you refer to is a known and unresolved one in many scientific fields. People make hypotheses about past events, but there's often little or no means of actually testing those hypotheses, especially when they were formed by those with access to all available information, precluding a meaningful test from occurring.

There is actually a similar problem in the investment field. When a 'factor' is identified post hoc, it's not appropriate to take the same data used to identify the factor to somehow 'test' how good a factor was at something. It's like watching a sporting event, finding out what happened, forming a hypothesis as to why it happened, then watching the same (now recorded) event later to test your hypothesis. The most, perhaps only, reliable means of testing a hypothesis is to do so with new data (i.e. data not used in the formulation of the hypothesis).
The Sensible Steward
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Why factor investing isn't working

Post by vineviz »

HomerJ wrote: Wed Jul 10, 2019 9:56 pm We're told to trust the summary, and when we question the results 10 years later, we're told we should trust that the strategy was the best at the time.
I'm not sure I see the point you're trying to make. Am I supposed to feel sorry for people who intentionally choose to make uninformed decisions and then second-guess them later? I don't get it.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
ThereAreNoGurus
Posts: 970
Joined: Fri Jan 24, 2014 10:41 pm

Re: Why factor investing isn't working

Post by ThereAreNoGurus »

Ken Fisher seems to agree with the OP's title.
“Small-cap value stocks are just better than other stocks.”
THIS IS A MYTH MANY professional investors and hardcore enthusiasts tend to fall for—the belief that small-cap value is inherently better and prone to long-term superiority moving forward, forever and ever, world without end, Amen.

....

But just because something has led for a long time doesn’t mean it must lead for a long time ahead.

As an example, since 1926, small-cap stocks have annualized 11.9% to the S&P 500’s 9.9%. Proof small cap is perma-better? Not really. A lot of that outperformance ignores huge bid–ask spreads common in small-cap stocks in the 1930s and 1940s—sometimes up to 30% of the purchase price. If you actually bought and sold small-cap stocks then, costs ate up a major chunk of your return—but that’s not captured in long-term index returns.
http://www.kenfisher.com/books/little-b ... hapter-09/
Trade the news and you will lose.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Why factor investing isn't working

Post by nedsaid »

vineviz wrote: Wed Jul 10, 2019 10:19 pm
HomerJ wrote: Wed Jul 10, 2019 9:56 pm We're told to trust the summary, and when we question the results 10 years later, we're told we should trust that the strategy was the best at the time.
I'm not sure I see the point you're trying to make. Am I supposed to feel sorry for people who intentionally choose to make uninformed decisions and then second-guess them later? I don't get it.
Larry changed his mind on a couple of issues and he is berated for lack of consistency. For heaven's sake, we all learn and we refine our thinking over time. I would hope that my ideas and my posts have gotten better over time. I only claim that I am right about anything around here every 6 months or so. All the other times, I am wrong about absolutely everything.

The other thing I try to get across is that we deal here with uncertainty and also from not knowing everything. If you study military history, you realize very fast that field commanders had to make decisions based upon incomplete information. What is obvious 156 years after the Battle of Gettysburg and after many books have been written on the topic was not obvious to the battlefield commanders at the time. There are a lot of armchair Generals and Admirals around here, most of whom have never written a book doing a lot of second guessing of authors years after their book was written.

Markets and the economy are dynamic. Just the way it is.
A fool and his money are good for business.
abc132
Posts: 2435
Joined: Thu Oct 18, 2018 1:11 am

Re: Why factor investing isn't working

Post by abc132 »

nedsaid wrote: Wed Jul 10, 2019 6:29 pm
abc132 wrote: Wed Jul 10, 2019 4:52 pm
nedsaid wrote: Tue Jul 09, 2019 12:43 pm
abc132 wrote: Tue Jul 09, 2019 12:28 pm I'd like to see a reply to the idea that for anyone tilting a particular direction, someone is also tilting in the opposite direction.

Thanks in advance.
Actually, I have posted on this topic many times. A month or two ago, I looked at the 10 year performance of the S&P 500, Vanguard Value Index, Vanguard Growth Index, Vanguard Small Value Index, and Vanguard Small Growth Index. Going from memory Vanguard Value Index had returns of 14%, the S&P 500 15%, and Vanguard Growth Index 16%. All of the Vanguard Growth and Value indexes outperformed the S&P 500 except for the Vanguard Value Index which represents Large Value. The excess performance of the S&P 500 and the Growth Index over the Value index comes from the High Tech/Internet parts of the market, particularly the FAANG stocks. In fact, the Vanguard Growth Index is now in non-diversified status because of heavy weightings in just a few stocks. We have been in a Large Growth stock market since the 2008-2009 financial crisis and bear market.

The Vanguard Small Value Index and the Vanguard Small Growth Indexes also beat the S&P 500 but not by very much. So the idea that Small/Value tilts have been awful is just ridiculous, they have pretty much tracked the S&P 500 over the last 10 years. What has happened is that Small/Value has not done so well over the last five years. Over the last 10 years, Large Value has trailed the broad market indexes by about 1% and Large Growth by 2% a year.

We have been in a Large Growth market and that is the other side of the trade.
Thanks for the reply. If I exclude the performance comparison, which is not related to my comments and is within the 10 year period Larry says we can't use, the point I was interested in clarifying is that someone is on the opposite side of every tilt. In a scenario where we all have access to the same information, it simply comes down to who has the best strategy, which is difficult to know in advance, for our own individual time horizon. We should all be open to the idea that factor tilting may work out moving forward, and it may not. It likely will depend on how you do it, as some tilts will work out and some will not, so there is nuance. I certainly appreciate the time/effort spent here to educate Bogleheads and discuss factors.

Nedsaid: We know that investors as a whole own the market. For me, that means that I own more Small-Cap stocks than the market 17% vs. 6% for Total Market, my Mid-Caps are 19% vs. 18% for Total Market so obviously I am underweight Large-Cap. I also hold 23% Large Value, 5% Mid Value, and 5% Small Value vs. 25% Large Value in Total Market, 5% Mid Value in Total Market, and 2% for Small Value in Total Market. In reality, I have a Mid/Small-Cap tilt but I have the same Value weightings as the Market itself other than I have a bit less Large Value and a bit more Small Value. Over time my portfolio lost its Value tilt, at one time 8% of my stocks were Small Value, now it is 5%. I hold individual stocks and some of those have drifted from Value to Core. One reason I am in a Growth to Value rebalancing program, trying to retilt my portfolio. Who is on the other side of my tilts, boy that is hard to say. My guess it would be other individuals who presumably have a taste for Large Cap Growth.

The following is certainly not directed at any individual:

I wish there were experts willing to make their recommendations in a clearly testable manner moving forward, without all the ambiguity about why a particular past or future recommendation should or shouldn't count for anything. That's why the scientific method was created - so that what is right or wrong is more than ever-changing opinions and rationalizations for why ones own position is correct no matter what the results. There is too much contradiction between stating we can't look at data within 10 or 20 or 30 years and then routinely using data within that same time period as if it means something. Data seams to mean something when it supports ones own opinion, and can be ignored when it does not. It's far too easy to grab 10 or 20 or 30 year data that shows exactly the opposite results, depending on the data selection.

A starting point would be that every recommendation made by someone declaring themselves an expert should come with a way to test or score any future recommendation (when and how and over what time period it can be tested), and they should be evaluated based on the prediction at the time is was made, and not on some after the fact rationalization. Nobody should be put in to the expert category that is not willing to do this. Just my two cents that if someone wants to get paid for their success, they ought to be able to numerically define with metrics in advance what that success means, and be held accountable to their own metrics.

Nedsaid: For whatever it is worth, you can check my New 'Doo thread where I post about my own portfolio. You can see what I have done and how I have been performing. I have been recommending Large Value for probably 3 years now and recently and somewhat jokingly have put a buy on Large Value, Mid Value, and Small Value in true Jim Cramer fashion hitting the buy button. Obviously, Large Growth continues to outperform and that could continue for a while. As Yogi Berra said, "Making predictions is hard, particularly about the future."

The actionable portion here is to clearly predefine the future success or failure of a method, and to really limit our individual desire to equate only limited and selected portions of the past to the future in ways that fit our own narrative and predispositions. It's too easy to fool ourselves, with the evidence being thousands of people with contrary positions all believing they are correct, and the others are the fools, each with data to back up their position.

Nedsaid: Hard to know even if you have been successful, had quite a discussion with Larry Swedroe about benchmarking and he warned me against resulting. You can check that out in my New 'Doo thread as I summarized that discussion hoping someone would gain from that. I compared my results vs. the 3 fund portfolio and did additional factor analysis for 2017 and 2018. Pretty much seeing if my portfolio is performing as one would expect. It has. But factor tilts don't work in a Large Growth environment.

I have to tell you that measuring my investment performance was difficult enough, particularly with Compound Annual Growth Rate which I calculate with Excel. Quicken calculates Internal Rate of Return and does a good job but fumbles Growth of $10,000 which gets confused as you add money to investment accounts. Benchmarking was difficult, partly because my asset allocation drifts a bit. I also benchmarked vs. 3 fund portfolio but Larry said that was all wrong. It did give me insight as to the effect of factors. Also have done comparisons with the "Lazy Portfolios" on MarketWatch. Exact comparisons are hard to make partly because pre-2013, I had a very, very relaxed attitude towards rebalancing, before then asset allocation drifted. To it is like comparing a moving target allocation vs. a fixed allocation and that isn't an accurate comparison. So pretty much a lot of eyeballing but not a terrifically good process. But that was the best I could do.

What I can say is that I have a sense that I underperformed the 3 fund portfolio during the 1990's because of my Value orientation and refusal to chase High Tech/Internet. I outperformed from 2000-2007 and have underperformed since the 2008-2009 financial crisis. Hardly surprising as I have a Value orientation and don't performance chase.

So my answers aren't all that good, best I can do is give you a sense based upon my experience and my feeble attempts at benchmarking.
Always appreciate the thought and effort put into your posts. I was talking purely about what would quantify a strategy as successful moving forward, so your past results shouldn't really matter for your current decisions. A real time strategy is independent of past results, as only future results should matter to our current decision making. I was also not holding you to any standard of expertise, despite the quality of your posts. If you were to proclaim you are giving professional advice here on Bogleheads, I would then expect a different set of standards about your posts.
abc132
Posts: 2435
Joined: Thu Oct 18, 2018 1:11 am

Re: Why factor investing isn't working

Post by abc132 »

larryswedroe wrote: Wed Jul 10, 2019 8:04 pm This is the wrong way to frame the issue. And it is easy to demonstrate why as it is the mistake known as resulting. The "best" or right strategy is known in advance, what we don't know is if it will result in the best outcome because we cannot know the future.


Consider the following: You are 35 married with two young kids. You decide to not buy life insurance because you learn the odds of you dying in the next 10 years are extremely low (knowing that if you did die early the family would not be able to have acceptable lifestyle). You don't die. Was the strategy good or bad? Resulting would say yes, good strategy. But that clearly is wrong because we don't know which universe will show up.
Your point is correct that a good bet may not pay off, but factor tilting is not life insurance. Life insurance makes a guarantee in writing subject to the company going out of business. Factor tilting professionals make no such guarantees whatsoever, as a factor tilting professional can profit immensely while the customer can lose extensively. Factor tilting is simply not insurance - as it makes no guarantee at all.

What we know about a good strategy is that if enough bets are made, the probability of success can be overwhelming. This is proven in games like blackjack, where putting the odds in your favor makes it mathematically unlikely that you can lose money over the long term. Someone using the tools of scientific processes could give expected odds of winning, and be held accountable to those predictions.

larryswedroe wrote: Wed Jul 10, 2019 8:04 pm That's the problem, sadly most people base their judgment on the outcome, which is clearly wrong when you don't know the future. That's what the book Thinking In Bets is all about, by poker expert.
That may be correct for the individual, but if 100,000 people follow a strategy in thousands of permutations, the group as a whole should be likely to be successful for a good strategy. Bogleheads have proven this by following low cost diversified portfolios, to the credit and benefit of those that really advocated such instruments and made them possible. Factor investing should be able to make similar claims by demonstrating the success of the people using the strategy. You should have access to this information at your firm - how your investors are actually performing as a group. It's fair to say that the strategy should be judged over some long time interval, but a strategy that never materializes is simply not a good strategy unless most of the expected metrics are not related to performance.
larryswedroe wrote: Wed Jul 10, 2019 8:04 pm In a world of uncertainty the only right way to judge the merits of a strategy is before you know the outcome, not after.
That is what I have been saying, and yet I have seen lots of past references and nothing numerically predictive of the future. How much value do you expect your clients to gain from investing in factor tilting, and what is the value you expect to add net fees? Over what time period should they expect to see these results? I don't expect a specific answer to these questions, but the questions demonstrate the type of analysis that should be given when recommending a particular strategy. If one can't do this, they should not refer to their product as anything close to insurance.
larryswedroe wrote: Wed Jul 10, 2019 8:04 pm Here's one of my favorite quotes, about managing risks.

“Individual decisions can be badly thought through, and yet be successful, or exceedingly well thought through, but be unsuccessful, because the recognized possibility of failure in fact occurs. But over time, more thoughtful decision-making will lead to better results, and more thoughtful decision-making can be encouraged by evaluating decisions on how well they were made rather than on outcome.”

– Robert Rubin,
Former co-chairman of Goldman Sachs
An expert should quantify this numerically. How much better should the results be, in what time period should they materialize, and what performance will ultimately show the strategy as either successful or unsuccessful? We know that simply saying a strategy is good does not make it so. Without defining the specific expected metrics in advance there is no way to show a strategy was ultimately good, and way too much wiggle room for misinterpreting any result in a way that meets ones individual biases.
larryswedroe wrote: Wed Jul 10, 2019 8:04 pm I've spent my entire career either managing risks for some of the largest companies in the world or advising some of the largest companies on managing financial risks, and now for the last 25 years individuals and institutional investors. And that is one of the most important lesson I think can be taught--to judge a strategy by it's quality of thinking, not by its outcome

Hope that is helpful
Larry
And yet you said that over time more thoughtful decision making should prevail. An investor should know if the likely/expected pay off period is 10 years, 20 years, 100 years, or 500 years so that they can pick a strategy that is appropriate for their personal situation. What is true for the individual is not true for the strategy as a whole. So while our individual results may not be indicative of a good strategy, the combined results of all those using a strategy should be able to show the predicted metrics of a good strategy.

I'm certainly not holding you personally accountable for answering these specific questions, but I would expect the factor tilt experts to be able to do so as a whole.
Last edited by abc132 on Thu Jul 11, 2019 4:44 am, edited 3 times in total.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Why factor investing isn't working

Post by nedsaid »

abc132 wrote: Wed Jul 10, 2019 10:54 pm
Always appreciate the thought and effort put into your posts. I was talking purely about what would quantify a strategy as successful moving forward, so your past results shouldn't really matter for your current decisions. A real time strategy is independent of past results, as only future results should matter to our current decision making. I was also not holding you to any standard of expertise, despite the quality of your posts. If you were to proclaim you are giving professional advice here on Bogleheads, I would then expect a different set of standards about your posts.
One thing is to look at what is working really well right now and assuming that this trend will not continue forever. So if the FAANG stocks are really, really hot right now, the prudent thing would be not to chase them. So you can at least avoid behavioral errors. Buying high and selling low is not a winning strategy but many investors do this nevertheless.

You could try to find cheaper asset classes and do what Dr. Bill Bernstein calls overbalancing, that is you would rebalance into cheaper assets to a greater degree than what you would do with normal rebalancing.

You also could try to lighten up on expensive assets, particularly if you see euphoria. So if the baggage handlers at the airport start giving you stock tips, it might be a sign to sell some of your stocks.

What I am saying here is that valuations and sentiment matter. Stocks have higher future expected returns when they are cheap than they do when they are expensive. Also low expectations are much easier to beat than high expectations. Pretty much common sense when you think it through and is also consistent with human nature and human behavior. The old thing about the cycles of greed and fear.

The whole idea of academic research is to ascertain which types of stock have the best performance over very long periods of time and to have more stocks with the characteristics that you like and fewer stocks with characteristics you don't like. This is the factor tilting we are discussing here. So pretty much, you have to decide if you buy into what the academics are telling us.

The problem with calculating real time investment results is that short term, almost anything can happen. Performance over the last three months essentially tells you nothing, particularly if market volatility up or down is due to something like massive short covering or a hedge fund manager unwinding a losing bet. Markets do really, really weird things in the shorter term and it isn't always easy to determine why.

The best you can do in measuring the effectiveness of your strategy is to look at your portfolio components and seeing if they are acting as expected. If you construct what you think is a low volatility portfolio and yet the portfolio is more volatile than your benchmarks, something is wrong with your strategy. You have to do some benchmarking.

The whole point of this thread is that Small/Value tilting has not worked over the last decade. Well, as I said, such a strategy just isn't going to beat the market when that market is led by the Large Growth stocks. If your performance is in line with a blended benchmark, that is a benchmark that includes the same percentage of Small/Value as what is in your portfolio, you haven't failed. A thoroughbred horse is just not going to win a drag race, even if he won the Triple Crown. Success is determine by relative performance compared to your benchmark. As I mentioned above, picking the appropriate benchmark is not always easy.

Problem is that there is no way to know for sure what is going to work in the future. None of us have prophetic powers. What we can do is look at the past and make reasonable guesses about what might happen in the future.

In 1999, Jack Bogle calculated the future estimated returns of stocks to be 2% for the next decade and that the returns for bonds would be 6%-7%. He was uncannily right except that stocks returned about zero from 2000-2012. His model took into account economic (business) return and speculative return. Economic return is relatively easy to calculate, pretty much earnings growth which historically is 5%-6% and adding the dividend yield which is about 2%. So that implies economic return of 7% to 8% a year. Then there is the element of speculative return which is a measure of what a market will pay for $1 of earnings. P/E ratio expansion adds to return and P/E ratio compression detracts from the returns. In normal times, P/E ratios are about 16-17 and they got up to about 32X earnings based upon future estimate earnings and 45X earnings based upon historical earnings. Bogle thought there would be a reversion of the mean back to more normal Price to Earnings ratios and he was right.

What Bogle saw was that stocks had very little future expected return with lots of risk attached. Bonds had relatively fat returns available at very little risk. Bogle said that bonds were the steal of the century. He lighted up on his stocks and he bought more bonds. A couple of versions to the story and he isn't around anymore to ask but the point was he did some calculations and acted upon what he saw. Bogle probably also was alarmed by the euphoria and all the talk about the new paradigm.

What actually happened during the 2000's was that earnings doubled during that decade, so you have about 7% earnings growth plus the 2% dividend yield. But investor enthusiasm for stocks waned, Price to Earnings ratios compressed, so returns of the S&P 500 over that decade was -1%. Fantastic earnings could not compensate for decreased investor enthusiasm. In other words, negative speculative return cancelled out the economic returns of stocks for that decade.

Of course what happened was that the market ran out of buyers in early 2000 and the markets fell. Markets need new buyers to come into the markets to push prices higher. You need pessimists who can change their mind and become optimists. When the very last pessimist throws in the towel and throws his money at the stock market, the rally is over. Everyone is optimistic, no one out there to change their mind and become a buyer and the market falls. This is an oversimplification but that is the way it works. One reason that market sentiment is a good contrary indicator.

Problem is that you might be right but just right too early. Or worse yet, right way too early.
A fool and his money are good for business.
AlohaJoe
Posts: 6609
Joined: Mon Nov 26, 2007 1:00 pm
Location: Saigon, Vietnam

Re: Why factor investing isn't working

Post by AlohaJoe »

nedsaid wrote: Wed Jul 10, 2019 11:32 pm
In 1999, Jack Bogle calculated the future estimated returns of stocks to be 2% for the next decade and that the returns for bonds would be 6%-7%. He was uncannily right except that stocks returned about zero from 2000-2012. His model took into account economic (business) return
And in 2001 he calculated that future estimated returns would be -2% but they were actually +2.9%, so he was off by 5 percentage points.

In 2002 he calculated that future returns would be 2.3% but they were actually 7.1%.

In 1989 he calculated that future returns would be 9.4% but they were actually 18.2%

Bogle's forecasting model has been a bigger failure than factor investing. He data fit a model that backtested well and has consistently failed since publication in 1990.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Why factor investing isn't working

Post by nedsaid »

Abc132, what I am trying to tell you is you are asking for the impossible. One reason being is that there is unpredictability to human emotion, people can do really weird and irrational stuff in the short term and there is just no way to predict that. As I said above, economic or business return is relatively easier to predict. Speculative return, you might as well flip a coin. All you know with speculative return is that there is a reversion to the mean that will occur sooner or later. Trouble is no one knows how soon or how late.

You are desiring precision that just isn't there. Shoot, earnings themselves are not a precise number. Accounting involves judgment and frankly educated guesses, particularly when preparing accruals. My example of John Bogle's projections for the 2000's were pretty good but he missed it. He projected 2% annual returns for the US Stock Market and as I recall, returns were -1% a year. He got it mostly right and that is as good as you are going to get.

Investing isn't about certainties, investing is about probabilities. About the best you can do is to tilt the odds in your favor best you can but there are just no guarantees. Some folks just can't get past this. All we can do is study the past and from that make reasonable projections about the future. An educated guess is a lot better than a purely uninformed guess.

So sorry sports fans out there if this disillusions you. Investing is just not a math problem or an engineering problem to be solved. One big reason is human nature and human behavior. Sorry to break this news to you but you may as well find this out now.
A fool and his money are good for business.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Why factor investing isn't working

Post by nedsaid »

AlohaJoe wrote: Wed Jul 10, 2019 11:54 pm
nedsaid wrote: Wed Jul 10, 2019 11:32 pm
In 1999, Jack Bogle calculated the future estimated returns of stocks to be 2% for the next decade and that the returns for bonds would be 6%-7%. He was uncannily right except that stocks returned about zero from 2000-2012. His model took into account economic (business) return
And in 2001 he calculated that future estimated returns would be -2% but they were actually +2.9%, so he was off by 5 percentage points.

In 2002 he calculated that future returns would be 2.3% but they were actually 7.1%.

In 1989 he calculated that future returns would be 9.4% but they were actually 18.2%

Bogle's forecasting model has been a bigger failure than factor investing. He data fit a model that backtested well and has consistently failed since publication in 1990.
There is just no way to calculate speculative return correctly. Bogle's model is as good as any out there but even then the best he can do is predict a range of returns. People can be more optimistic than what you would expect and they are can be more pessimistic than what you expect.
A fool and his money are good for business.
AlohaJoe
Posts: 6609
Joined: Mon Nov 26, 2007 1:00 pm
Location: Saigon, Vietnam

Re: Why factor investing isn't working

Post by AlohaJoe »

nedsaid wrote: Thu Jul 11, 2019 12:04 am
AlohaJoe wrote: Wed Jul 10, 2019 11:54 pm
nedsaid wrote: Wed Jul 10, 2019 11:32 pm
In 1999, Jack Bogle calculated the future estimated returns of stocks to be 2% for the next decade and that the returns for bonds would be 6%-7%. He was uncannily right except that stocks returned about zero from 2000-2012. His model took into account economic (business) return
And in 2001 he calculated that future estimated returns would be -2% but they were actually +2.9%, so he was off by 5 percentage points.

In 2002 he calculated that future returns would be 2.3% but they were actually 7.1%.

In 1989 he calculated that future returns would be 9.4% but they were actually 18.2%

Bogle's forecasting model has been a bigger failure than factor investing. He data fit a model that backtested well and has consistently failed since publication in 1990.
There is just no way to calculate speculative return correctly. Bogle's model is as good as any out there but even then the best he can do is predict a range of returns. People can be more optimistic than what you would expect and they are can be more pessimistic than what you expect.
That's true but I'm making a different point. Go look at his model vs. reality for 1905-1990. It doesn't always get it right but it is never wrong for long and rarely wrong by more than 1%. As an example from 1952-1956 his model has a cumulative error of just 2.8% over 5 years.

What's more his errors were symmetrical. Just as likely to overestimate as underestimate.

Around 1990 -- when he first published his model -- all of that changed. His model never underestimates any more. The errors are larger and more frequent.

His model has clearly failed out of sample.
Post Reply