CAPE: A much stronger predictor of stock returns than many think

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: CAPE: A much stronger predictor of stock returns than many think

Post by vineviz »

abc132 wrote: Mon Dec 14, 2020 12:17 pm I'm much more curious why you used PE and declared it highly predictive earlier in the thread. If you only believe in E/P regression, why would you post a P/E regression and declare it highly predictive?
I don't ONLY believe in the E/P regression. What I said is that it has a more sound theoretical basis than a P/E regression: it is methodologically superior, for reasons that have been known and discussed for literally 50+ years.
abc132 wrote: Mon Dec 14, 2020 12:17 pm
Valuation ratios moved up in the year 2000 to levels that were absolutely unprecedented, and are still nearly as high as of this writing at the beginning of 2001. Even allowing for the possibility that the economy and financial markets have undergone some struc- tural changes, these ratios imply a stronger case for a poor stock market outlook than has ever been seen before.
And so far they have all been wrong - we haven't seen those poor returns, so people might wonder about those declaring them highly predictive absent the results they predicted. [/quote]

Nonsense. They wrote that at the end of 2000, when CAPE was over 44, and were specifically referring to the subsequent 10 years (i.e. 2001 to 2010). They weren't saying the outlook in early 2001 was the worst it had ever been, just that the case for the outlook being poor was stronger than it had ever been. Indeed, the performance of the S&P 500 from 2001 to 2010 was "poor" by anyone's measure (they don't call it the "lost decade" for nothing).
vineviz wrote: Mon Dec 14, 2020 9:55 am You don't need any training to see that the annual return has to go to infinity as price approaches zero, so that's a definite reason to go with E/P regression. You don't need a CAPE model to predict that higher valuations portend lower future returns.
If you think that higher valuations " higher valuations portend lower future returns" then you need some way to measure valuations, right?

If not CAPE, then what? Price/book? Dividend yield? EV/EBTIDA? All of those work to some degree, with pros and cons relative to CAPE. Implementing any of them also results in the same kind of methodological issues and questions that CAPE does. So I can agree that you don't NEED to use CAPE, but - as you say - you've got to use something. What?
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
abc132
Posts: 2435
Joined: Thu Oct 18, 2018 1:11 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by abc132 »

vineviz wrote: Mon Dec 14, 2020 1:17 pm
abc132 wrote: Mon Dec 14, 2020 12:17 pm I'm much more curious why you used PE and declared it highly predictive earlier in the thread. If you only believe in E/P regression, why would you post a P/E regression and declare it highly predictive?
I don't ONLY believe in the E/P regression. What I said is that it has a more sound theoretical basis than a P/E regression: it is methodologically superior, for reasons that have been known and discussed for literally 50+ years.
Well that puts us in a bind, because you seem to want us to use both models to make our plans, and they give different results at high CAPE10 ratios. Which one should we be using for our planning? -2% or 5% at CAPE10=40?

How many additional years would one need to work if they chose one over the other, and how much of our disposable income should we give up in the present to meet these predictions? It's a very significant difference.

abc132 wrote: Mon Dec 14, 2020 12:17 pm
Valuation ratios moved up in the year 2000 to levels that were absolutely unprecedented, and are still nearly as high as of this writing at the beginning of 2001. Even allowing for the possibility that the economy and financial markets have undergone some structural changes, these ratios imply a stronger case for a poor stock market outlook than has ever been seen before.


And so far they have all been wrong - we haven't seen those poor returns, so people might wonder about those declaring them highly predictive absent the results they predicted.
vineviz wrote: Mon Dec 14, 2020 9:55 am Nonsense. They wrote that at the end of 2000, when CAPE was over 44, and were specifically referring to the subsequent 10 years (i.e. 2001 to 2010). They weren't saying the outlook in early 2001 was the worst it had ever been, just that the case for the outlook being poor was stronger than it had ever been. Indeed, the performance of the S&P 500 from 2001 to 2010 was "poor" by anyone's measure (they don't call it the "lost decade" for nothing).
Wasn't there irrational exuberance in 1996? You have to look at all of the predictions about future returns, and not cherry pick instances. Historically high CAPE10 ratios have not produced lower returns from 2000 to 2020.

Separate out the predictability at high and low CAPE10 ratios and tell us where the future predictability is coming from - it's the low end of CAPE10 ratios. It would only be rational to look at where the methods FORWARD predictability has come from.

vineviz wrote: Mon Dec 14, 2020 9:55 am If you think that higher valuations " higher valuations portend lower future returns" then you need some way to measure valuations, right?

If not CAPE, then what? Price/book? Dividend yield? EV/EBTIDA? All of those work to some degree, with pros and cons relative to CAPE. Implementing any of them also results in the same kind of methodological issues and questions that CAPE does. So I can agree that you don't NEED to use CAPE, but - as you say - you've got to use something. What?
I think the reliability of any of these methods is marginal at best. It's just that tough to predict the future. I would recommend one make broad judgements based on historical returns and variability, diversify to better weather any changes, plan for what percentage of worst case scenarios they are willing to work additional years to resolve, and then course correct as their plan plays out.

It's really that simple absent better models. The actual forward predictability of these models are within what you would already be planning for given all but the most extreme risk tolerances.
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: CAPE: A much stronger predictor of stock returns than many think

Post by vineviz »

abc132 wrote: Mon Dec 14, 2020 2:01 pm Well that puts us in a bind, because you seem to want us to use both models to make our plans, and they give different results at high CAPE10 ratios. Which one should we be using for our planning? -2% or 5% at CAPE10=40?
If you're in a bind it's only because you choose to be there. I've already explained why I think the E/P specification should be preferred over P/E.

And if CAPE is 40 and the 10-year US Treasury is yielding 0.90% (as it currently is) then if you have ANY faith that markets are even a LITTLE BIT efficient you should be able to detect that using an expected return for stocks of -2% is probably wrong.

However, either specification is arguably going to give you a more accurate estimate than a naive "let's just take the long-run average return" or approach. If CAPE is 40 then you should be able to detect that using an expected return for stocks of +12% is probably also wrong.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
abc132
Posts: 2435
Joined: Thu Oct 18, 2018 1:11 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by abc132 »

vineviz wrote: Mon Dec 14, 2020 3:03 pm
abc132 wrote: Mon Dec 14, 2020 2:01 pm Well that puts us in a bind, because you seem to want us to use both models to make our plans, and they give different results at high CAPE10 ratios. Which one should we be using for our planning? -2% or 5% at CAPE10=40?
If you're in a bind it's only because you choose to be there. I've already explained why I think the E/P specification should be preferred over P/E.

And if CAPE is 40 and the 10-year US Treasury is yielding 0.90% (as it currently is) then if you have ANY faith that markets are even a LITTLE BIT efficient you should be able to detect that using an expected return for stocks of -2% is probably wrong.

However, either specification is arguably going to give you a more accurate estimate than a naive "let's just take the long-run average return" or approach. If CAPE is 40 then you should be able to detect that using an expected return for stocks of +12% is probably also wrong.
You lose me on
1. I believe in the P/E model (but prefer E/P)
2. I reject what the P/E model says (-2% nominal at CAPE10=40)
but I think I should just accept your conflicting statements at this point of the thread.

Preparing for historical returns and variability includes preparing for lower than average returns. It's well spelled out in my prior posts if you want to take another read. Many of us prepare for something closer to 2% real returns when we use 25-30x earnings/expenses as a guideline for a 30 year retirement - largely because of sequence of returns risk. We do this regardless of the CAPE10 ratio, so no there is not a wrong 12% assumption to returns when using historical data for planning purposes.

I'm using 6% nominal returns as my baseline for preparation for retirement, but I was doing this throughout every CAPE10 ratio since I have been working. That doesn't mean it is the return I expect, just that it is a reasonable/conservative enough assumption to successfully prepare for most early retirement scenarios.

I've made no prediction about 10 years returns. I think doing so is a zero sum game.
dont_know_mind
Posts: 94
Joined: Sun Nov 22, 2020 7:14 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by dont_know_mind »

abc132 wrote: Mon Dec 14, 2020 5:28 pm
vineviz wrote: Mon Dec 14, 2020 3:03 pm
abc132 wrote: Mon Dec 14, 2020 2:01 pm Well that puts us in a bind, because you seem to want us to use both models to make our plans, and they give different results at high CAPE10 ratios. Which one should we be using for our planning? -2% or 5% at CAPE10=40?
If you're in a bind it's only because you choose to be there. I've already explained why I think the E/P specification should be preferred over P/E.

And if CAPE is 40 and the 10-year US Treasury is yielding 0.90% (as it currently is) then if you have ANY faith that markets are even a LITTLE BIT efficient you should be able to detect that using an expected return for stocks of -2% is probably wrong.

However, either specification is arguably going to give you a more accurate estimate than a naive "let's just take the long-run average return" or approach. If CAPE is 40 then you should be able to detect that using an expected return for stocks of +12% is probably also wrong.
You lose me on
1. I believe in the P/E model (but prefer E/P)
2. I reject what the P/E model says (-2% nominal at CAPE10=40)
but I think I should just accept your conflicting statements at this point of the thread.

Preparing for historical returns and variability includes preparing for lower than average returns. It's well spelled out in my prior posts if you want to take another read. Many of us prepare for something closer to 2% real returns when we use 25-30x earnings/expenses as a guideline for a 30 year retirement - largely because of sequence of returns risk. We do this regardless of the CAPE10 ratio, so no there is not a wrong 12% assumption to returns when using historical data for planning purposes.

I'm using 6% nominal returns as my baseline for preparation for retirement, but I was doing this throughout every CAPE10 ratio since I have been working. That doesn't mean it is the return I expect, just that it is a reasonable/conservative enough assumption to successfully prepare for most early retirement scenarios.

I've made no prediction about 10 years returns. I think doing so is a zero sum game.
Falling for the CAPE narrative is I think a rookie error.
The finance and the consultancy industry market an attribute based investing perspective.
From what I have seen (but I could be wrong), if you are going to tilt or market-time or take a positional view (or why else would CAPE be useful) you get completely caned using CAPE or at least I did when I tried to apply it.

What I have learnt is that valuation is contextual.
How could it not be ?
Looking for any historical correlation is misunderstanding the game.
It would be like looking for the relationship between returns and horse characteristics. Even if there is a historical relationship with a statistical significance that horses with high odds are positive expectancy at a certain racetrack, what does it mean ?

Either you think there is a universal truth to be gleaned : there is actually a zebra there and attribute based perspective is valid or you don't.
I really can't see how valuation cannot be anything but contextual.
It will depend on the situation and always has. This is the only invariant thing I can find about markets.
My ramblings are probably rubbish, biased by my position and talking my own book. No-one should invest based on my views. Some days I wonder whether I've had some skill or just been a lucky gambler.
User avatar
dogagility
Posts: 3237
Joined: Fri Feb 24, 2017 5:41 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by dogagility »

dont_know_mind wrote: Wed Dec 16, 2020 3:34 am Looking for any historical correlation is misunderstanding the game.
It would be like looking for the relationship between returns and horse characteristics. Even if there is a historical relationship with a statistical significance that horses with high odds are positive expectancy at a certain racetrack, what does it mean ?
Perhaps the belief in CAPE predictivity is akin to the satirical Pastafarianism belief that global warming is caused by a decrease in the number of pirates. https://en.wikipedia.org/wiki/Flying_Spaghetti_Monster :twisted:

My takeaway from this thread is that there may be no specific forward predictivity of CAPE because data plotted on the x-axis of marcopolo's graph appears to shift to the right over time. viewtopic.php?p=5644874#p5644874. It seems we have no idea where today's conditions should be placed on the x-axis.
Make sure you check out my list of certifications. The list is short, and there aren't any. - Eric 0. from SMA
Jeff Albertson
Posts: 904
Joined: Sat Apr 06, 2013 7:11 pm
Location: Springfield

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Jeff Albertson »

10-year Annualized REAL Returns Since 1881 (relative to starting CAPE)
Image
https://ritholtz.com/2020/12/10-wednesd ... um=twitter
Seasonal
Posts: 3045
Joined: Sun May 21, 2017 1:49 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Seasonal »

vineviz wrote: Mon Dec 14, 2020 12:57 pm Although I am sympathetic to the view that CAPE generates return estimates which are imprecise, I come back to the fundamental problem: ALL goal-based financial planning decisions require the use of SOME sort of return forecasts, and for equities there simply are not any alternatives to CAPE which are clearly more accurate or more precise.

Historically, using CAPE to inform savings rate decisions and asset allocation decisions would have typically produced "better" outcomes than ignoring it. The improvements haven't ALWAYS been large, but sometimes they have been. Using CAPE didn't ALWAYS produce better outcomes, but usually it did.

Investing involves making decisions under conditions of uncertainty, and there is no avoiding it. Ignoring a model that can reduce that uncertainty a little bit is irrational, especially when there really aren't better alternatives.
I haven't seen any plausible models that are more accurate than CAPE, but that doesn't necessarily mean CAPE is useful. As a Vanguard study found, CAPE and p/e do better than other models they tested for a 10 year horizon, but they only explained about 40% of the variation in real returns.

Please provide evidence that CAPE usually produced better outcomes regarding savings rate and asset allocation than ignoring it.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

Seasonal wrote: Wed Dec 16, 2020 10:41 am
vineviz wrote: Mon Dec 14, 2020 12:57 pm Although I am sympathetic to the view that CAPE generates return estimates which are imprecise, I come back to the fundamental problem: ALL goal-based financial planning decisions require the use of SOME sort of return forecasts, and for equities there simply are not any alternatives to CAPE which are clearly more accurate or more precise.

Historically, using CAPE to inform savings rate decisions and asset allocation decisions would have typically produced "better" outcomes than ignoring it. The improvements haven't ALWAYS been large, but sometimes they have been. Using CAPE didn't ALWAYS produce better outcomes, but usually it did.

Investing involves making decisions under conditions of uncertainty, and there is no avoiding it. Ignoring a model that can reduce that uncertainty a little bit is irrational, especially when there really aren't better alternatives.
I haven't seen any plausible models that are more accurate than CAPE, but that doesn't necessarily mean CAPE is useful. As a Vanguard study found, CAPE and p/e do better than other models they tested for a 10 year horizon, but they only explained about 40% of the variation in real returns.

Please provide evidence that CAPE usually produced better outcomes regarding savings rate and asset allocation than ignoring it.
You're saying that a variable that you believe is capable of explaining 40% of the variation in returns is worthless?
The Sensible Steward
Seasonal
Posts: 3045
Joined: Sun May 21, 2017 1:49 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Seasonal »

willthrill81 wrote: Wed Dec 16, 2020 11:06 am
Seasonal wrote: Wed Dec 16, 2020 10:41 am
vineviz wrote: Mon Dec 14, 2020 12:57 pm Although I am sympathetic to the view that CAPE generates return estimates which are imprecise, I come back to the fundamental problem: ALL goal-based financial planning decisions require the use of SOME sort of return forecasts, and for equities there simply are not any alternatives to CAPE which are clearly more accurate or more precise.

Historically, using CAPE to inform savings rate decisions and asset allocation decisions would have typically produced "better" outcomes than ignoring it. The improvements haven't ALWAYS been large, but sometimes they have been. Using CAPE didn't ALWAYS produce better outcomes, but usually it did.

Investing involves making decisions under conditions of uncertainty, and there is no avoiding it. Ignoring a model that can reduce that uncertainty a little bit is irrational, especially when there really aren't better alternatives.
I haven't seen any plausible models that are more accurate than CAPE, but that doesn't necessarily mean CAPE is useful. As a Vanguard study found, CAPE and p/e do better than other models they tested for a 10 year horizon, but they only explained about 40% of the variation in real returns.

Please provide evidence that CAPE usually produced better outcomes regarding savings rate and asset allocation than ignoring it.
You're saying that a variable that you believe is capable of explaining 40% of the variation in returns is worthless?
vineviz made a statement. I'm asking for evidence. If you agree with the statement, please provide evidence that CAPE usually produced better outcomes regarding savings rate and asset allocation than ignoring it.
User avatar
Uncorrelated
Posts: 1237
Joined: Sun Oct 13, 2019 3:16 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Uncorrelated »

willthrill81 wrote: Wed Dec 16, 2020 11:06 am
Seasonal wrote: Wed Dec 16, 2020 10:41 am
vineviz wrote: Mon Dec 14, 2020 12:57 pm Although I am sympathetic to the view that CAPE generates return estimates which are imprecise, I come back to the fundamental problem: ALL goal-based financial planning decisions require the use of SOME sort of return forecasts, and for equities there simply are not any alternatives to CAPE which are clearly more accurate or more precise.

Historically, using CAPE to inform savings rate decisions and asset allocation decisions would have typically produced "better" outcomes than ignoring it. The improvements haven't ALWAYS been large, but sometimes they have been. Using CAPE didn't ALWAYS produce better outcomes, but usually it did.

Investing involves making decisions under conditions of uncertainty, and there is no avoiding it. Ignoring a model that can reduce that uncertainty a little bit is irrational, especially when there really aren't better alternatives.
I haven't seen any plausible models that are more accurate than CAPE, but that doesn't necessarily mean CAPE is useful. As a Vanguard study found, CAPE and p/e do better than other models they tested for a 10 year horizon, but they only explained about 40% of the variation in real returns.

Please provide evidence that CAPE usually produced better outcomes regarding savings rate and asset allocation than ignoring it.
You're saying that a variable that you believe is capable of explaining 40% of the variation in returns is worthless?
We don't know if it explains 40% of returns. The 95% confidence interval is probably that it explains between (-20%, 90%) of returns. Not exactly useful.

In my experiments with machine learning models I've discovered that explaining 40% of future variance out of sample is still not sufficient to beat a sample mean forecast after transaction costs. The R^2 needed to be even higher in order to obtain some CER gains. Therefore, even if CAPE actually explains 40% of returns, it's not obvious whether that actually results in tangible improvements.


Goyal and Welch have investigated the CER (certainty equivalent return) gains of using various P/E forecasts and found that all of them perform worse than a simple mean forecast out-of-sample.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

Seasonal wrote: Wed Dec 16, 2020 11:16 am
willthrill81 wrote: Wed Dec 16, 2020 11:06 am
Seasonal wrote: Wed Dec 16, 2020 10:41 am
vineviz wrote: Mon Dec 14, 2020 12:57 pm Although I am sympathetic to the view that CAPE generates return estimates which are imprecise, I come back to the fundamental problem: ALL goal-based financial planning decisions require the use of SOME sort of return forecasts, and for equities there simply are not any alternatives to CAPE which are clearly more accurate or more precise.

Historically, using CAPE to inform savings rate decisions and asset allocation decisions would have typically produced "better" outcomes than ignoring it. The improvements haven't ALWAYS been large, but sometimes they have been. Using CAPE didn't ALWAYS produce better outcomes, but usually it did.

Investing involves making decisions under conditions of uncertainty, and there is no avoiding it. Ignoring a model that can reduce that uncertainty a little bit is irrational, especially when there really aren't better alternatives.
I haven't seen any plausible models that are more accurate than CAPE, but that doesn't necessarily mean CAPE is useful. As a Vanguard study found, CAPE and p/e do better than other models they tested for a 10 year horizon, but they only explained about 40% of the variation in real returns.

Please provide evidence that CAPE usually produced better outcomes regarding savings rate and asset allocation than ignoring it.
You're saying that a variable that you believe is capable of explaining 40% of the variation in returns is worthless?
vineviz made a statement. I'm asking for evidence. If you agree with the statement, please provide evidence that CAPE usually produced better outcomes regarding savings rate and asset allocation than ignoring it.
I'm not trying to be rude or snippy at all, but if you believe that CAPE is capable of explaining 40% of market returns, then I think that the burden of proof is on you to show that it's not useful.
The Sensible Steward
Seasonal
Posts: 3045
Joined: Sun May 21, 2017 1:49 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Seasonal »

Uncorrelated wrote: Wed Dec 16, 2020 11:21 am<quote snipped>
[1]We don't know if it explains 40% of returns. The 95% confidence interval is probably that it explains between (-20%, 90%) of returns. Not exactly useful.

[2]In my experiments with machine learning models I've discovered that explaining 40% of future variance out of sample is still not sufficient to beat a sample mean forecast after transaction costs. The R^2 needed to be even higher in order to obtain some CER gains. Therefore, even if CAPE actually explains 40% of returns, it's not obvious whether that actually results in tangible improvements.


Goyal and Welch have investigated the CER (certainty equivalent return) gains of using various P/E forecasts and found that all of them perform worse than a simple mean forecast out-of-sample.
[1] I agree that point estimates should be accompanied by confidence intervals.

[2] Sounds right, but lets see if the proponents have anything to say besides trying to shift the burden of proof.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

Seasonal wrote: Wed Dec 16, 2020 11:53 amSounds right, but lets see if the proponents have anything to say besides trying to shift the burden of proof.
The burden of proof lies with the one making the claim. Yes, vineviz made a claim, but you have made a claim yourself. How can a statistically significant variable explain 40% of the variance and not be useful in some way?
The Sensible Steward
Seasonal
Posts: 3045
Joined: Sun May 21, 2017 1:49 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Seasonal »

willthrill81 wrote: Wed Dec 16, 2020 11:44 am<quote snipped>
I'm not trying to be rude or snippy at all, but if you believe that CAPE is capable of explaining 40% of market returns, then I think that the burden of proof is on you to show that it's not useful.
<second post>
The burden of proof lies with the one making the claim. Yes, vineviz made a claim, but you have made a claim yourself. How can a statistically significant variable explain 40% of the variance and not be useful in some way?
I have no idea if CAPE can consistently explain 40% of returns. I referred to a paper in which Vanguard found that it did over a defined period and didn't find anything better. Something can easily be the best, especially over a specific historic period, and not continue to even hit that level. My crystal ball is cloudy. It might or might not work going forward.

In other words, I'm not claiming anything beyond that Vanguard made a statement.

I did not say whether or not CAPE is useful. vineviz made a clear statement that it's useful. See above. If vineviz does not belief it is in fact useful, I'd appreciate a clarification.

As you note, the burden of proof is on the one who makes a claim.
secondopinion
Posts: 6011
Joined: Wed Dec 02, 2020 12:18 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by secondopinion »

Jeff Albertson wrote: Wed Dec 16, 2020 8:09 am 10-year Annualized REAL Returns Since 1881 (relative to starting CAPE)
Image
https://ritholtz.com/2020/12/10-wednesd ... um=twitter
The variation of returns is still wide. Given how few entries exist in the current range, we cannot accurately claim much. I plan for less returns and hope for more returns.
Passive investing: not about making big bucks but making profits. Active investing: not about beating the market but meeting goals. Speculation: not about timing the market but taking profitable risks.
GAAP
Posts: 2556
Joined: Fri Apr 08, 2016 12:41 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by GAAP »

HomerJ wrote: Sat Dec 12, 2020 7:00 pm Anyone who changed their AA based on valuations over the past 30 years has LESS money than those of us who ignored valuations.
Seasonal wrote: Wed Dec 16, 2020 10:41 am Please provide evidence that CAPE usually produced better outcomes regarding savings rate and asset allocation than ignoring it.
dont_know_mind wrote: Wed Dec 16, 2020 3:34 am From what I have seen (but I could be wrong), if you are going to tilt or market-time or take a positional view (or why else would CAPE be useful) you get completely caned using CAPE or at least I did when I tried to apply it.
So AA changes, market-timing, and market-tilting are the only reasons to estimate future returns? 465 replies so far, and still ignoring other reasons to estimate future returns.

Your own personal need/desire for future returns may only be for those three reasons -- that doesn't mean that CAPE-based estimates are not useful for other reasons. Quite frankly, unless the naysayers explain why CAPE used for other reasons is not useful, this thread is also not useful anymore.
“Adapt what is useful, reject what is useless, and add what is specifically your own.” ― Bruce Lee
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

Seasonal wrote: Wed Dec 16, 2020 12:02 pm
willthrill81 wrote: Wed Dec 16, 2020 11:44 am<quote snipped>
I'm not trying to be rude or snippy at all, but if you believe that CAPE is capable of explaining 40% of market returns, then I think that the burden of proof is on you to show that it's not useful.
<second post>
The burden of proof lies with the one making the claim. Yes, vineviz made a claim, but you have made a claim yourself. How can a statistically significant variable explain 40% of the variance and not be useful in some way?
I have no idea if CAPE can consistently explain 40% of returns. I referred to a paper in which Vanguard found that it did over a defined period and didn't find anything better.
Okay, I was taking that as an implication that CAPE was a significant predictor of future returns. Because if it is, then it follows that it must be useful to some extent for something.
The Sensible Steward
scout1
Posts: 234
Joined: Thu Oct 18, 2018 3:26 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by scout1 »

GAAP wrote: Wed Dec 16, 2020 2:28 pm
HomerJ wrote: Sat Dec 12, 2020 7:00 pm Anyone who changed their AA based on valuations over the past 30 years has LESS money than those of us who ignored valuations.
Seasonal wrote: Wed Dec 16, 2020 10:41 am Please provide evidence that CAPE usually produced better outcomes regarding savings rate and asset allocation than ignoring it.
dont_know_mind wrote: Wed Dec 16, 2020 3:34 am From what I have seen (but I could be wrong), if you are going to tilt or market-time or take a positional view (or why else would CAPE be useful) you get completely caned using CAPE or at least I did when I tried to apply it.
So AA changes, market-timing, and market-tilting are the only reasons to estimate future returns? 465 replies so far, and still ignoring other reasons to estimate future returns.

Your own personal need/desire for future returns may only be for those three reasons -- that doesn't mean that CAPE-based estimates are not useful for other reasons. Quite frankly, unless the naysayers explain why CAPE used for other reasons is not useful, this thread is also not useful anymore.
There is no value in estimating future returns incorrectly, and historically the CAPE has been incorrect. Now if you could estimate future returns even a little bit accurately, there would be a LOT of value in that.
Last edited by scout1 on Wed Dec 16, 2020 2:39 pm, edited 2 times in total.
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

willthrill81 wrote: Wed Dec 16, 2020 11:06 am You're saying that a variable that you believe is capable of explaining 40% of the variation in returns is worthless?
So far it has been...

So far, the other 60% of variables showed up.

Anyone making changes based on CAPE has less money than those of us who did not.
"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: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

willthrill81 wrote: Wed Dec 16, 2020 2:34 pm
Seasonal wrote: Wed Dec 16, 2020 12:02 pm
willthrill81 wrote: Wed Dec 16, 2020 11:44 am<quote snipped>
I'm not trying to be rude or snippy at all, but if you believe that CAPE is capable of explaining 40% of market returns, then I think that the burden of proof is on you to show that it's not useful.
<second post>
The burden of proof lies with the one making the claim. Yes, vineviz made a claim, but you have made a claim yourself. How can a statistically significant variable explain 40% of the variance and not be useful in some way?
I have no idea if CAPE can consistently explain 40% of returns. I referred to a paper in which Vanguard found that it did over a defined period and didn't find anything better.
Okay, I was taking that as an implication that CAPE was a significant predictor of future returns. Because if it is, then it follows that it must be useful to some extent for something.
40% (if that's even true) of predicting a range of plus/minus 8%.

It has not been useful.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

HomerJ wrote: Wed Dec 16, 2020 2:37 pm Anyone making changes based on CAPE has less money than those of us who did not.
How do you figure that? If CAPE 'tells you' that expected returns are 5%, you decide that you need to save more to meet your goals, and actual returns are 10%, you would have more than if you had used 7% real (close to the historic average) as an estimate instead.

Now if the criticism was that CAPE was consistently overestimating forward returns, there would be more potential for investors to be materially disadvantaged.
HomerJ wrote: Wed Dec 16, 2020 2:40 pm 40% (if that's even true) of predicting a range of plus/minus 8%.

It has not been useful.
If you're only estimating something like 3-4% real in the first place, then you're already 'erring' on the conservative side of at least historic returns, and you don't need a metric to tell you that returns are likely to be that high or higher. But the average investor (or the average FA, I think) isn't planning around 3% real.
The Sensible Steward
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

willthrill81 wrote: Wed Dec 16, 2020 2:53 pm
HomerJ wrote: Wed Dec 16, 2020 2:37 pm Anyone making changes based on CAPE has less money than those of us who did not.
How do you figure that? If CAPE 'tells you' that expected returns are 5%, you decide that you need to save more to meet your goals
That's not how most people use CAPE.

They are told that the U.S, market has "high" valuations, and low expected returns, and International has low valuations, and higher expected returns, and they change their Asset Allocation to tilt more to International.

And then, more times than not since CAPE was discovered, they end up with less money.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

HomerJ wrote: Wed Dec 16, 2020 3:07 pm
willthrill81 wrote: Wed Dec 16, 2020 2:53 pm
HomerJ wrote: Wed Dec 16, 2020 2:37 pm Anyone making changes based on CAPE has less money than those of us who did not.
How do you figure that? If CAPE 'tells you' that expected returns are 5%, you decide that you need to save more to meet your goals
That's not how most people use CAPE.
Perhaps it's how they should use CAPE.
The Sensible Steward
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

willthrill81 wrote: Wed Dec 16, 2020 3:08 pm
HomerJ wrote: Wed Dec 16, 2020 3:07 pm
willthrill81 wrote: Wed Dec 16, 2020 2:53 pm
HomerJ wrote: Wed Dec 16, 2020 2:37 pm Anyone making changes based on CAPE has less money than those of us who did not.
How do you figure that? If CAPE 'tells you' that expected returns are 5%, you decide that you need to save more to meet your goals
That's not how most people use CAPE.
Perhaps it's how they should use CAPE.
No, they should assume low returns and save accordingly.

Then, adjust AFTER the returns come in.

Would you really tell someone to save LESS if CAPE was low? "Go ahead, buy a new car, don't worry, CAPE says we're "likely" to get great returns going forward - save half as much as you used to - it's probably all good!"

No, people should change their spending or retirement plans AFTER the ACTUAL returns come in...

No one should be making saving or spending decisions based on CAPE, and predictions of future returns. 40% predictive power combined with plus/minus 8% error bands is not very useful.
Last edited by HomerJ on Wed Dec 16, 2020 3:14 pm, edited 1 time in total.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
rich126
Posts: 4475
Joined: Thu Mar 01, 2018 3:56 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by rich126 »

I've seen several people mention sub 2% annual nominal returns for the next 10 years. Unfortunately I'm at work now and can't get a link to one article.
Will be interesting to see what happens. Those with high US stock allocations may be in a rough ride if that is true, or if wrong, may be in good shape :)

I'm going with more international stocks, and more cash even if earning low returns and will use the cash when buying opportunities occur. Along with a few other things to tweak returns a bit.
----------------------------- | If you think something is important and it doesn't involve the health of someone, think again. Life goes too fast, enjoy it and be nice.
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

rich126 wrote: Wed Dec 16, 2020 3:13 pm I'm going with more international stocks, and more cash even if earning low returns and will use the cash when buying opportunities occur. Along with a few other things to tweak returns a bit.
And there you go, will.

rich126 isn't going to start saving more... He's going to find ways to "tweak returns".
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

HomerJ wrote: Wed Dec 16, 2020 3:11 pm
willthrill81 wrote: Wed Dec 16, 2020 3:08 pm
HomerJ wrote: Wed Dec 16, 2020 3:07 pm
willthrill81 wrote: Wed Dec 16, 2020 2:53 pm
HomerJ wrote: Wed Dec 16, 2020 2:37 pm Anyone making changes based on CAPE has less money than those of us who did not.
How do you figure that? If CAPE 'tells you' that expected returns are 5%, you decide that you need to save more to meet your goals
That's not how most people use CAPE.
Perhaps it's how they should use CAPE.
No, they should assume low returns and save accordingly.

Then, adjust AFTER the returns come in.

Would you really tell someone to save LESS if CAPE was low? "Go ahead, buy a new car, don't worry, CAPE says we're "likely" to get great returns going forward - save half as much as you used to - it's probably all good!"

No, people should change their spending or retirement plans AFTER the ACTUAL returns come in...

No one should be making saving or spending decisions based on CAPE, and predictions of future returns.
That's a matter of opinion. It's certainly an informed position, and I don't fault you for it, but I think that it's wrong to say that that's what investors should do, implying that it's the only good or the best option.

In the decade after CAPE hit 40+, U.S. stocks lost money after inflation. Japan's stock market implosion can be attributed in no small part to absolutely stratospheric valuations. You're saying that investors would have done better to ignore that.

Personally, I would recommend that an investor plan on the lesser of something like 5% real returns OR a CAPE-derived estimate (i.e. a regression model, 1/CAPE, etc.), whichever is less. But I realize that that's merely my own take on the issue and not what anyone else should do. My current plan is to use the ABW method in retirement, and I think that I'll cap expected returns at 5% real, no matter what CAPE or any other metric estimates.
HomerJ wrote: Wed Dec 16, 2020 3:16 pm
rich126 wrote: Wed Dec 16, 2020 3:13 pm I'm going with more international stocks, and more cash even if earning low returns and will use the cash when buying opportunities occur. Along with a few other things to tweak returns a bit.
And there you go, will.

rich126 isn't going to start saving more... He's going to find ways to "tweak returns".
I agree that the data I've seen do not suggest that changing one's AA on the basis of valuations is an effective strategy. And I have never recommended that any investor take that approach.
The Sensible Steward
junior
Posts: 1452
Joined: Wed Sep 10, 2008 6:14 pm
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by junior »

HomerJ wrote: Wed Dec 16, 2020 3:07 pm
That's not how most people use CAPE.

They are told that the U.S, market has "high" valuations, and low expected returns, and International has low valuations, and higher expected returns, and they change their Asset Allocation to tilt more to International.
People have been saying valuations are better for international for maybe the last 10 years or so, but CAPE wasn't necessarily the only metric used to track that.

At any rate I'd say the jury is still out on whether tilting based on valuations will work, based on CAPE or anything else.

I'll be in the market for 50 or so years, so I'm not sure that 10 years of international underperformance means tilting doesn't work, though of course if you hang out on an investment board and track your portfolio daily you might get frustrated after 1, 2, or even 20 years of feeling that tilting based on valuations hasn't yet worked. It doesn't prove it won't work out over the course of the investor's lifespan.
Last edited by junior on Wed Dec 16, 2020 3:31 pm, edited 1 time in total.
GAAP
Posts: 2556
Joined: Fri Apr 08, 2016 12:41 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by GAAP »

scout1 wrote: Wed Dec 16, 2020 2:37 pm There is no value in estimating future returns incorrectly, and historically the CAPE has been incorrect. Now if you could estimate future returns even a little bit accurately, there would be a LOT of value in that.
No value -- really?

Historically, if you constrained CAPE values by rounding up anything under 10.5to 10.5, you would have overestimated 1-year returns about two-thirds of the time. If your planning is revisited every year, you don't really even care about longer terms.

If a planning effort revolves around estimating a date to hit "a number", underestimation reduces disappointment and provides a much better forward path. If a planning effort revolves around a withdrawal calculation, then underestimation reduces portfolio failure and smooths withdrawals. In either case, a CAPE-based estimate that is generally pessimistic has quite a bit of value.
“Adapt what is useful, reject what is useless, and add what is specifically your own.” ― Bruce Lee
GAAP
Posts: 2556
Joined: Fri Apr 08, 2016 12:41 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by GAAP »

HomerJ wrote: Wed Dec 16, 2020 3:07 pm
willthrill81 wrote: Wed Dec 16, 2020 2:53 pm
HomerJ wrote: Wed Dec 16, 2020 2:37 pm Anyone making changes based on CAPE has less money than those of us who did not.
How do you figure that? If CAPE 'tells you' that expected returns are 5%, you decide that you need to save more to meet your goals
That's not how most people use CAPE.

They are told that the U.S, market has "high" valuations, and low expected returns, and International has low valuations, and higher expected returns, and they change their Asset Allocation to tilt more to International.

And then, more times than not since CAPE was discovered, they end up with less money.
"Most people" chase the hot stocks, regardless of what CAPE says. I would think/hope that the discussion here would be about what intelligent, thinking people would do -- not the madding crowd.
“Adapt what is useful, reject what is useless, and add what is specifically your own.” ― Bruce Lee
scout1
Posts: 234
Joined: Thu Oct 18, 2018 3:26 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by scout1 »

GAAP wrote: Wed Dec 16, 2020 3:31 pm
scout1 wrote: Wed Dec 16, 2020 2:37 pm There is no value in estimating future returns incorrectly, and historically the CAPE has been incorrect. Now if you could estimate future returns even a little bit accurately, there would be a LOT of value in that.
No value -- really?

Historically, if you constrained CAPE values by rounding up anything under 10.5to 10.5, you would have overestimated 1-year returns about two-thirds of the time. If your planning is revisited every year, you don't really even care about longer terms.

If a planning effort revolves around estimating a date to hit "a number", underestimation reduces disappointment and provides a much better forward path. If a planning effort revolves around a withdrawal calculation, then underestimation reduces portfolio failure and smooths withdrawals. In either case, a CAPE-based estimate that is generally pessimistic has quite a bit of value.
Yes, pulling a random number out of thin air is of no value to anyone. That's exactly what CAPE is doing.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

GAAP wrote: Wed Dec 16, 2020 3:31 pm
scout1 wrote: Wed Dec 16, 2020 2:37 pm There is no value in estimating future returns incorrectly, and historically the CAPE has been incorrect. Now if you could estimate future returns even a little bit accurately, there would be a LOT of value in that.
No value -- really?

Historically, if you constrained CAPE values by rounding up anything under 10.5to 10.5, you would have overestimated 1-year returns about two-thirds of the time. If your planning is revisited every year, you don't really even care about longer terms.

If a planning effort revolves around estimating a date to hit "a number", underestimation reduces disappointment and provides a much better forward path. If a planning effort revolves around a withdrawal calculation, then underestimation reduces portfolio failure and smooths withdrawals. In either case, a CAPE-based estimate that is generally pessimistic has quite a bit of value.
I agree that if the argument is that CAPE consistently underestimates returns, that's much less 'sinful' than the opposite. It's better for most investors to have too much than not enough. That said, I think that the jury is still out on whether making AA adjustments on the basis of valuations is warranted. Even if it is beneficial in the very long-term (e.g. 30+ years), I'm not sure that it would be generally beneficial in shorter but still very relevant periods.
The Sensible Steward
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

scout1 wrote: Wed Dec 16, 2020 3:36 pm
GAAP wrote: Wed Dec 16, 2020 3:31 pm
scout1 wrote: Wed Dec 16, 2020 2:37 pm There is no value in estimating future returns incorrectly, and historically the CAPE has been incorrect. Now if you could estimate future returns even a little bit accurately, there would be a LOT of value in that.
No value -- really?

Historically, if you constrained CAPE values by rounding up anything under 10.5to 10.5, you would have overestimated 1-year returns about two-thirds of the time. If your planning is revisited every year, you don't really even care about longer terms.

If a planning effort revolves around estimating a date to hit "a number", underestimation reduces disappointment and provides a much better forward path. If a planning effort revolves around a withdrawal calculation, then underestimation reduces portfolio failure and smooths withdrawals. In either case, a CAPE-based estimate that is generally pessimistic has quite a bit of value.
Yes, pulling a random number out of thin air is of no value to anyone. That's exactly what CAPE is doing.
If you think that CAPE (or 1/CAPE or a CAPE derived regression model) is a random number, then you're wrong, plain and simple, and very possibly just trolling.
The Sensible Steward
GAAP
Posts: 2556
Joined: Fri Apr 08, 2016 12:41 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by GAAP »

HomerJ wrote: Wed Dec 16, 2020 3:11 pm No, they should assume low returns and save accordingly.
Absolutely.
HomerJ wrote: Wed Dec 16, 2020 3:11 pm Then, adjust AFTER the returns come in.
Certainly.
HomerJ wrote: Wed Dec 16, 2020 3:11 pm Would you really tell someone to save LESS if CAPE was low? "Go ahead, buy a new car, don't worry, CAPE says we're "likely" to get great returns going forward - save half as much as you used to - it's probably all good!"
No.
HomerJ wrote: Wed Dec 16, 2020 3:11 pm No, people should change their spending or retirement plans AFTER the ACTUAL returns come in...
Which would happen as part of an annual planning process...
HomerJ wrote: Wed Dec 16, 2020 3:11 pm No one should be making saving or spending decisions based on CAPE, and predictions of future returns. 40% predictive power combined with plus/minus 8% error bands is not very useful.
This depends to a great deal upon the type of decision. If you're hoping that the magic CAPE 8 ball will tell you how little you can get away with savings then you need to start thinking. If you're trying to estimate whether or not (or when) you'll be able to retire, then you need some value to estimate with. CAPE is one reasonable way to do that -- again as part of a planning process that is regularly revisited.
“Adapt what is useful, reject what is useless, and add what is specifically your own.” ― Bruce Lee
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

junior wrote: Wed Dec 16, 2020 3:30 pm
HomerJ wrote: Wed Dec 16, 2020 3:07 pm
That's not how most people use CAPE.

They are told that the U.S, market has "high" valuations, and low expected returns, and International has low valuations, and higher expected returns, and they change their Asset Allocation to tilt more to International.
People have been saying valuations are better for international for maybe the last 10 years or so, but CAPE wasn't necessarily the only metric used to track that.

At any rate I'd say the jury is still out on whether tilting based on valuations will work, based on CAPE or anything else.
I'd accept it if people said the "jury is still out"... I wouldn't agree, but I wouldn't throw a fit about that statement.

The title of this thread is that CAPE is a STRONG predictor of stock returns.

When the opposite has been true in real life.

That is what makes me crazy...

I think the disconnect is that CAPE appears to be strongly correlated to 10-year returns, but it's been terrible as a predictor in real life because the correlation keeps changing.
Last edited by HomerJ on Wed Dec 16, 2020 3:45 pm, edited 1 time 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: CAPE: A much stronger predictor of stock returns than many think

Post by vineviz »

Seasonal wrote: Wed Dec 16, 2020 12:02 pm
willthrill81 wrote: Wed Dec 16, 2020 11:44 am<quote snipped>
I'm not trying to be rude or snippy at all, but if you believe that CAPE is capable of explaining 40% of market returns, then I think that the burden of proof is on you to show that it's not useful.
<second post>
The burden of proof lies with the one making the claim. Yes, vineviz made a claim, but you have made a claim yourself. How can a statistically significant variable explain 40% of the variance and not be useful in some way?
I have no idea if CAPE can consistently explain 40% of returns. I referred to a paper in which Vanguard found that it did over a defined period and didn't find anything better. Something can easily be the best, especially over a specific historic period, and not continue to even hit that level. My crystal ball is cloudy. It might or might not work going forward.

In other words, I'm not claiming anything beyond that Vanguard made a statement.

I did not say whether or not CAPE is useful. vineviz made a clear statement that it's useful. See above. If vineviz does not belief it is in fact useful, I'd appreciate a clarification.

As you note, the burden of proof is on the one who makes a claim.
I've lost faith that anyone who things CAPE is useless can be persuaded by evidence at this point, but here is some anyway.

Imagine an investor who must decide at the beginning of each year how much to invest for their retirement in 30 years. Call this the savings rate. For the sake of simplicity, let's assume they are 100% invested in equities and that each year they make the decision mechanically using as inputs three numbers: the current balance in the account, the number of years remaining until retirement, and an estimate of future real returns. I use real returns so I can simulate over multiple time periods without bias. The target retirement wealth is $100,000.

There are two options for estimating future real returns. Option 1 is to use a CAPE regression which incorporates only the data available as of 12/31 of the prior year (i.e. no look-ahead bias). Option 2 is to just use the long-run average return as of 12/31 of the prior year. I compared the outcomes for all 81 rolling 30-year periods starting in 1910. Option 1 (using CAPE) produced a higher retirement wealth on average (about 6% higher) with smoother contributions (about 8% less volatility) than not using CAPE.

By way of illustration, here are the annual contributions for the 1989 start years for the investor using CAPE (blue line) and not using CAPE (orange line). Ignoring valuations in formulating the financial plan, as the orange line investor did, produced the kind of sudden and severe adjustments in savings rate that investors generally prefer to avoid.

Image

Now imagine the reverse scenario. A retired investor starts with $100,000 and must decide at the beginning of each year how much of their portfolio to withdraw. Using the same options, with a goal of drawing down the portfolio to zero at the end of 30 years, the investor using CAPE to inform their withdrawal rates had less volatile swings in retirement consumption (about 11% less) than the investor who did not use CAPE to inform their withdrawal rate.

By way of illustration, here are the annual withdrawals for the 1965 start years for the investor using CAPE (blue line) and not using CAPE (orange line). Ignoring valuations in formulating the financial plan, as the orange line investor did, produced the kind of sudden and severe curtailments in consumption that investors generally prefer to avoid.

Image
Last edited by vineviz on Wed Dec 16, 2020 3:49 pm, edited 1 time in total.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
GAAP
Posts: 2556
Joined: Fri Apr 08, 2016 12:41 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by GAAP »

willthrill81 wrote: Wed Dec 16, 2020 3:38 pm I agree that if the argument is that CAPE consistently underestimates returns, that's much less 'sinful' than the opposite. It's better for most investors to have too much than not enough. That said, I think that the jury is still out on whether making AA adjustments on the basis of valuations is warranted. Even if it is beneficial in the very long-term (e.g. 30+ years), I'm not sure that it would be generally beneficial in shorter but still very relevant periods.
I wouldn't make AA adjustments based upon CAPE for any period of time. I'm also not too sure that doing so over a long term would be any more beneficial than the short term. Things change all the time -- the potential dispersion over 30 years would be huge.
“Adapt what is useful, reject what is useless, and add what is specifically your own.” ― Bruce Lee
User avatar
Uncorrelated
Posts: 1237
Joined: Sun Oct 13, 2019 3:16 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Uncorrelated »

GAAP wrote: Wed Dec 16, 2020 2:28 pm
HomerJ wrote: Sat Dec 12, 2020 7:00 pm Anyone who changed their AA based on valuations over the past 30 years has LESS money than those of us who ignored valuations.
Seasonal wrote: Wed Dec 16, 2020 10:41 am Please provide evidence that CAPE usually produced better outcomes regarding savings rate and asset allocation than ignoring it.
dont_know_mind wrote: Wed Dec 16, 2020 3:34 am From what I have seen (but I could be wrong), if you are going to tilt or market-time or take a positional view (or why else would CAPE be useful) you get completely caned using CAPE or at least I did when I tried to apply it.
So AA changes, market-timing, and market-tilting are the only reasons to estimate future returns? 465 replies so far, and still ignoring other reasons to estimate future returns.

Your own personal need/desire for future returns may only be for those three reasons -- that doesn't mean that CAPE-based estimates are not useful for other reasons. Quite frankly, unless the naysayers explain why CAPE used for other reasons is not useful, this thread is also not useful anymore.
Provided that the CAPE10 has a "true" R^2 of at least 0.5% on a monthly horizon and there are no other indicators that dominate CAPE10, it would be useful and economically significant for all those things. This is recognized in many papers even those critical of economical indicators, Campbell and Thompson even have a direct formula to calculate the expected sharpe ratio increase from R^2.

However, the fact that CAPE10 has failed to actually obtain superior performance in the past casts doubt that the true R^2 is indeed that high, and suggests that these high R^2's are the result of statistical errors rather than a genuine relationship.

Some suggest that an indicator can be useful for long term planning but not market timing. I simply don't see how this is even possible within any rational optimization model (i.e. merton's portfolio model). But the opposite, where an indicator can be used for market timing but not long-term planning, is possible. For example, if you have insider information that has not yet reached the market.
GAAP
Posts: 2556
Joined: Fri Apr 08, 2016 12:41 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by GAAP »

HomerJ wrote: Wed Dec 16, 2020 3:42 pm The title of this thread is that CAPE is a STRONG predictor of stock returns.
"CAPE: A much stronger predictor of stock returns than many think" DOES NOT EQUAL "CAPE: a strong predictor of stock returns".
“Adapt what is useful, reject what is useless, and add what is specifically your own.” ― Bruce Lee
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

HomerJ wrote: Wed Dec 16, 2020 3:42 pm The title of this thread is that CAPE is a STRONG predictor of stock returns.

When the opposite has been true in real life.
In the terminology of regression analysis, which was used in the OP, the terms 'independent variable' and 'predictor' are interchangeable.

A variable being a strong predictor of another variable in this type of analysis merely means that there was a strong linear correlation between the variables in the data being analyzed. Nothing more, nothing less.

And from 1989-2010, forward 10 year returns were very strongly correlated with the starting CAPE. That was 'real life' and is not up for debate.

What marcopolo correctly pointed out is that from 1989-2010, the specific nature of the relationship between CAPE and forward 10 year returns shifted in such a way that returns were higher for a given level of CAPE than before (i.e. there was a substantial shift in the intercept of the regression model).

Your claim has been and continues to be since Shiller put CAPE forward, specific predictions involving CAPE using data available at the time of the prediction have consistently underestimated returns. That is true, but it is also true that CAPE was very strongly negatively related to forward returns. As you have said yourself, the general trend that CAPE has been negatively related to forward returns is quite clear.
The Sensible Steward
scout1
Posts: 234
Joined: Thu Oct 18, 2018 3:26 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by scout1 »

willthrill81 wrote: Wed Dec 16, 2020 3:40 pm
scout1 wrote: Wed Dec 16, 2020 3:36 pm
GAAP wrote: Wed Dec 16, 2020 3:31 pm
scout1 wrote: Wed Dec 16, 2020 2:37 pm There is no value in estimating future returns incorrectly, and historically the CAPE has been incorrect. Now if you could estimate future returns even a little bit accurately, there would be a LOT of value in that.
No value -- really?

Historically, if you constrained CAPE values by rounding up anything under 10.5to 10.5, you would have overestimated 1-year returns about two-thirds of the time. If your planning is revisited every year, you don't really even care about longer terms.

If a planning effort revolves around estimating a date to hit "a number", underestimation reduces disappointment and provides a much better forward path. If a planning effort revolves around a withdrawal calculation, then underestimation reduces portfolio failure and smooths withdrawals. In either case, a CAPE-based estimate that is generally pessimistic has quite a bit of value.
Yes, pulling a random number out of thin air is of no value to anyone. That's exactly what CAPE is doing.
If you think that CAPE (or 1/CAPE or a CAPE derived regression model) is a random number, then you're wrong, plain and simple, and very possibly just trolling.
It predicts the future as accurately as flipping a coin. Take it from the people in this thread who tried using the CAPE ten years ago to estimate future returns. They're telling you that it didn't work. The CAPE isn't new but somehow we bogleheads have discovered a highly predictive tool that no one uses? Being even a tiny bit predictive is worth a huge amount of money to the financial industry. Here's a random Krugman quote that was quoted by Scott Sumner that describes a little bit about this thread and why it's 10 pages.

I’m actually kind of reluctant to even get into this, because any discussion of these issues brings out the people who believe that they have discovered the hidden secrets of the monetary universe, somehow missed by generations of economists. But here goes anyway.
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

willthrill81 wrote: Wed Dec 16, 2020 3:53 pm
HomerJ wrote: Wed Dec 16, 2020 3:42 pm The title of this thread is that CAPE is a STRONG predictor of stock returns.

When the opposite has been true in real life.
In the terminology of regression analysis, which was used in the OP, the terms 'independent variable' and 'predictor' are interchangeable.

A variable being a strong predictor of another variable in this type of analysis merely means that there was a strong linear correlation between the variables in the data being analyzed. Nothing more, nothing less.

And from 1989-2010, forward 10 year returns were very strongly correlated with the starting CAPE. That was 'real life' and is not up for debate.

What marcopolo correctly pointed out is that from 1989-2010, the specific nature of the relationship between CAPE and forward 10 year returns shifted in such a way that returns were higher for a given level of CAPE than before (i.e. there was a substantial shift in the intercept of the regression model).

Your claim has been and continues to be since Shiller put CAPE forward, specific predictions involving CAPE using data available at the time of the prediction have consistently underestimated returns. That is true, but it is also true that CAPE was very strongly negatively related to forward returns. As you have said yourself, the general trend that CAPE has been negatively related to forward returns is quite clear.
I edited one of my posts to say
I think the disconnect is that CAPE appears to be strongly correlated to 10-year returns, but it's been terrible as a predictor in real life because the correlation keeps changing.
Which you stated above (far better than me)...

So I think we agree...

It's very important to always state that CAPE has done a poor job in real-time PREDICTING returns... even though it does seem very correlated with stock market returns, the actual predictions have shifted a large amount over the years. Your title should talk about correlation, not predictions.

CAPE of 25 was "very high" and predicted 0% real in 1996... And correctly so, based on 1926-1995 data.

CAPE of 25 today is considered "near normal" and predicts 6% real.... And correctly so, based on 1926-2020 data.

But that is a huge change over a fairly short period of time, AND both predictions still have the plus/minus 8% problem as well.

Very difficult and dangerous to make large changes to Asset Allocation based on CAPE, considering the massive changes in predictions it has gone through since it was discovered.
Last edited by HomerJ on Wed Dec 16, 2020 4:03 pm, edited 2 times in total.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by marcopolo »

vineviz wrote: Wed Dec 16, 2020 3:45 pm
Seasonal wrote: Wed Dec 16, 2020 12:02 pm
willthrill81 wrote: Wed Dec 16, 2020 11:44 am<quote snipped>
I'm not trying to be rude or snippy at all, but if you believe that CAPE is capable of explaining 40% of market returns, then I think that the burden of proof is on you to show that it's not useful.
<second post>
The burden of proof lies with the one making the claim. Yes, vineviz made a claim, but you have made a claim yourself. How can a statistically significant variable explain 40% of the variance and not be useful in some way?
I have no idea if CAPE can consistently explain 40% of returns. I referred to a paper in which Vanguard found that it did over a defined period and didn't find anything better. Something can easily be the best, especially over a specific historic period, and not continue to even hit that level. My crystal ball is cloudy. It might or might not work going forward.

In other words, I'm not claiming anything beyond that Vanguard made a statement.

I did not say whether or not CAPE is useful. vineviz made a clear statement that it's useful. See above. If vineviz does not belief it is in fact useful, I'd appreciate a clarification.

As you note, the burden of proof is on the one who makes a claim.
I've lost faith that anyone who things CAPE is useless can be persuaded by evidence at this point, but here is some anyway.

Imagine an investor who must decide at the beginning of each year how much to invest for their retirement in 30 years. Call this the savings rate. For the sake of simplicity, let's assume they are 100% invested in equities and that each year they make the decision mechanically using as inputs three numbers: the current balance in the account, the number of years remaining until retirement, and an estimate of future real returns. I use real returns so I can simulate over multiple time periods without bias. The target retirement wealth is $100,000.

There are two options for estimating future real returns. Option 1 is to use a CAPE regression which incorporates only the data available as of 12/31 of the prior year (i.e. no look-ahead bias). Option 2 is to just use the long-run average return as of 12/31 of the prior year. I compared the outcomes for all 81 rolling 30-year periods starting in 1910. Option 1 (using CAPE) produced a higher retirement wealth on average (about 6% higher) with smoother contributions (about 8% less volatility) than not using CAPE.

By way of illustration, here are the annual contributions for the 1989 start years for the investor using CAPE (blue line) and not using CAPE (orange line). Ignoring valuations in formulating the financial plan, as the orange line investor did, produced the kind of sudden and severe adjustments in savings rate that investors generally prefer to avoid.

Image

Now imagine the reverse scenario. A retired investor starts with $100,000 and must decide at the beginning of each year how much of their portfolio to withdraw. Using the same options, with a goal of drawing down the portfolio to zero at the end of 30 years, the investor using CAPE to inform their withdrawal rates had less volatile swings in retirement consumption (about 11% less) than the investor who did not use CAPE to inform their withdrawal rate.

By way of illustration, here are the annual withdrawals for the 1965 start years for the investor using CAPE (blue line) and not using CAPE (orange line). Ignoring valuations in formulating the financial plan, as the orange line investor did, produced the kind of sudden and severe curtailments in consumption that investors generally prefer to avoid.

Image

That is a nice academic excersize.

While I have never said CAPE is useless, this is precisely why I keep questioning how actionable it really is.

How many proponents of valuation models were telling people in the depths of the 1999/2000 recession "Hey this is great, with valuations this low, you don't need to bother saving anything for retirement!"

Most people I know just saved what they could based on income, expenses, and life circumstances, not some measure of valuation.

Do you know anyone that actually does that in real life?
Once in a while you get shown the light, in the strangest of places if you look at it right.
GAAP
Posts: 2556
Joined: Fri Apr 08, 2016 12:41 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by GAAP »

Uncorrelated wrote: Wed Dec 16, 2020 3:46 pm
GAAP wrote: Wed Dec 16, 2020 2:28 pm
HomerJ wrote: Sat Dec 12, 2020 7:00 pm Anyone who changed their AA based on valuations over the past 30 years has LESS money than those of us who ignored valuations.
Seasonal wrote: Wed Dec 16, 2020 10:41 am Please provide evidence that CAPE usually produced better outcomes regarding savings rate and asset allocation than ignoring it.
dont_know_mind wrote: Wed Dec 16, 2020 3:34 am From what I have seen (but I could be wrong), if you are going to tilt or market-time or take a positional view (or why else would CAPE be useful) you get completely caned using CAPE or at least I did when I tried to apply it.
So AA changes, market-timing, and market-tilting are the only reasons to estimate future returns? 465 replies so far, and still ignoring other reasons to estimate future returns.

Your own personal need/desire for future returns may only be for those three reasons -- that doesn't mean that CAPE-based estimates are not useful for other reasons. Quite frankly, unless the naysayers explain why CAPE used for other reasons is not useful, this thread is also not useful anymore.
Provided that the CAPE10 has a "true" R^2 of at least 0.5% on a monthly horizon and there are no other indicators that dominate CAPE10, it would be useful and economically significant for all those things. This is recognized in many papers even those critical of economical indicators, Campbell and Thompson even have a direct formula to calculate the expected sharpe ratio increase from R^2.

However, the fact that CAPE10 has failed to actually obtain superior performance in the past casts doubt that the true R^2 is indeed that high, and suggests that these high R^2's are the result of statistical errors rather than a genuine relationship.

Some suggest that an indicator can be useful for long term planning but not market timing. I simply don't see how this is even possible within any rational optimization model (i.e. merton's portfolio model). But the opposite, where an indicator can be used for market timing but not long-term planning, is possible. For example, if you have insider information that has not yet reached the market.
My question was not if CAPE was useful for those things, but whether it is useful for other things. I have no interest in using CAPE for market timing, AA changes in general, or tilting. Does that mean that I can't possibly use CAPE for something else?

In my limited reading, Merton's portfolio model requires known constant values for the risk-free rate, expected return, and volatility -- is that correct?

If so, it doesn't sound at all rational to me.

In any case, why does a planning methodology -- which if done at all well, will be regularly updated to adjust for reality -- require an optimization model at all?
“Adapt what is useful, reject what is useless, and add what is specifically your own.” ― Bruce Lee
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

marcopolo wrote: Wed Dec 16, 2020 4:02 pm
vineviz wrote: Wed Dec 16, 2020 3:45 pm
Seasonal wrote: Wed Dec 16, 2020 12:02 pm
willthrill81 wrote: Wed Dec 16, 2020 11:44 am<quote snipped>
I'm not trying to be rude or snippy at all, but if you believe that CAPE is capable of explaining 40% of market returns, then I think that the burden of proof is on you to show that it's not useful.
<second post>
The burden of proof lies with the one making the claim. Yes, vineviz made a claim, but you have made a claim yourself. How can a statistically significant variable explain 40% of the variance and not be useful in some way?
I have no idea if CAPE can consistently explain 40% of returns. I referred to a paper in which Vanguard found that it did over a defined period and didn't find anything better. Something can easily be the best, especially over a specific historic period, and not continue to even hit that level. My crystal ball is cloudy. It might or might not work going forward.

In other words, I'm not claiming anything beyond that Vanguard made a statement.

I did not say whether or not CAPE is useful. vineviz made a clear statement that it's useful. See above. If vineviz does not belief it is in fact useful, I'd appreciate a clarification.

As you note, the burden of proof is on the one who makes a claim.
I've lost faith that anyone who things CAPE is useless can be persuaded by evidence at this point, but here is some anyway.

Imagine an investor who must decide at the beginning of each year how much to invest for their retirement in 30 years. Call this the savings rate. For the sake of simplicity, let's assume they are 100% invested in equities and that each year they make the decision mechanically using as inputs three numbers: the current balance in the account, the number of years remaining until retirement, and an estimate of future real returns. I use real returns so I can simulate over multiple time periods without bias. The target retirement wealth is $100,000.

There are two options for estimating future real returns. Option 1 is to use a CAPE regression which incorporates only the data available as of 12/31 of the prior year (i.e. no look-ahead bias). Option 2 is to just use the long-run average return as of 12/31 of the prior year. I compared the outcomes for all 81 rolling 30-year periods starting in 1910. Option 1 (using CAPE) produced a higher retirement wealth on average (about 6% higher) with smoother contributions (about 8% less volatility) than not using CAPE.

By way of illustration, here are the annual contributions for the 1989 start years for the investor using CAPE (blue line) and not using CAPE (orange line). Ignoring valuations in formulating the financial plan, as the orange line investor did, produced the kind of sudden and severe adjustments in savings rate that investors generally prefer to avoid.

Image

Now imagine the reverse scenario. A retired investor starts with $100,000 and must decide at the beginning of each year how much of their portfolio to withdraw. Using the same options, with a goal of drawing down the portfolio to zero at the end of 30 years, the investor using CAPE to inform their withdrawal rates had less volatile swings in retirement consumption (about 11% less) than the investor who did not use CAPE to inform their withdrawal rate.

By way of illustration, here are the annual withdrawals for the 1965 start years for the investor using CAPE (blue line) and not using CAPE (orange line). Ignoring valuations in formulating the financial plan, as the orange line investor did, produced the kind of sudden and severe curtailments in consumption that investors generally prefer to avoid.

Image

That is a nice academic excersize.

While I have never said CAPE is useless, this is precisely why I keep questioning how actionable it really is.

How many proponents of valuation models were telling people in the depths of the 1999/2000 recession "Hey this is great, with valuations this low, you don't need to bother saving anything for retirement!"

Most people I know just saved what they could based on income, expenses, and life circumstances, not some measure of valuation.

Do you know anyone that actually does that in real life?
It sounds like you're moving the goalposts.

vineviz demonstrated one effective means of using CAPE, and you're arguing that because a lot of people weren't espousing that 20 years age or because few investors do that today that CAPE has no value.
The Sensible Steward
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

This goes back to academic vs real life.
Imagine an investor who must decide at the beginning of each year how much to invest for their retirement in 30 years. Call this the savings rate. For the sake of simplicity, let's assume they are 100% invested in equities and that each year they make the decision mechanically using as inputs three numbers: the current balance in the account, the number of years remaining until retirement, and an estimate of future real returns.
No one does this.

The problem is that, in real life, no one knows when they are going to retire, and no one even knows how much they are going to need in retirement.

No 30 year-old can calculate how large her nest egg should be in 30 years. She is still building her career, she may get married, she may have kids, she may acquire new taste for luxuries, she may decide she likes a simple life...

No one is or really should be changing their saving rate based on CAPE.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
GAAP
Posts: 2556
Joined: Fri Apr 08, 2016 12:41 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by GAAP »

marcopolo wrote: Wed Dec 16, 2020 4:02 pm
vineviz wrote: Wed Dec 16, 2020 3:45 pm
Seasonal wrote: Wed Dec 16, 2020 12:02 pm
willthrill81 wrote: Wed Dec 16, 2020 11:44 am<quote snipped>
I'm not trying to be rude or snippy at all, but if you believe that CAPE is capable of explaining 40% of market returns, then I think that the burden of proof is on you to show that it's not useful.
<second post>
The burden of proof lies with the one making the claim. Yes, vineviz made a claim, but you have made a claim yourself. How can a statistically significant variable explain 40% of the variance and not be useful in some way?
I have no idea if CAPE can consistently explain 40% of returns. I referred to a paper in which Vanguard found that it did over a defined period and didn't find anything better. Something can easily be the best, especially over a specific historic period, and not continue to even hit that level. My crystal ball is cloudy. It might or might not work going forward.

In other words, I'm not claiming anything beyond that Vanguard made a statement.

I did not say whether or not CAPE is useful. vineviz made a clear statement that it's useful. See above. If vineviz does not belief it is in fact useful, I'd appreciate a clarification.

As you note, the burden of proof is on the one who makes a claim.
I've lost faith that anyone who things CAPE is useless can be persuaded by evidence at this point, but here is some anyway.

Imagine an investor who must decide at the beginning of each year how much to invest for their retirement in 30 years. Call this the savings rate. For the sake of simplicity, let's assume they are 100% invested in equities and that each year they make the decision mechanically using as inputs three numbers: the current balance in the account, the number of years remaining until retirement, and an estimate of future real returns. I use real returns so I can simulate over multiple time periods without bias. The target retirement wealth is $100,000.

There are two options for estimating future real returns. Option 1 is to use a CAPE regression which incorporates only the data available as of 12/31 of the prior year (i.e. no look-ahead bias). Option 2 is to just use the long-run average return as of 12/31 of the prior year. I compared the outcomes for all 81 rolling 30-year periods starting in 1910. Option 1 (using CAPE) produced a higher retirement wealth on average (about 6% higher) with smoother contributions (about 8% less volatility) than not using CAPE.

By way of illustration, here are the annual contributions for the 1989 start years for the investor using CAPE (blue line) and not using CAPE (orange line). Ignoring valuations in formulating the financial plan, as the orange line investor did, produced the kind of sudden and severe adjustments in savings rate that investors generally prefer to avoid.

Image

Now imagine the reverse scenario. A retired investor starts with $100,000 and must decide at the beginning of each year how much of their portfolio to withdraw. Using the same options, with a goal of drawing down the portfolio to zero at the end of 30 years, the investor using CAPE to inform their withdrawal rates had less volatile swings in retirement consumption (about 11% less) than the investor who did not use CAPE to inform their withdrawal rate.

By way of illustration, here are the annual withdrawals for the 1965 start years for the investor using CAPE (blue line) and not using CAPE (orange line). Ignoring valuations in formulating the financial plan, as the orange line investor did, produced the kind of sudden and severe curtailments in consumption that investors generally prefer to avoid.

Image

That is a nice academic excersize.

While I have never said CAPE is useless, this is precisely why I keep questioning how actionable it really is.

How many proponents of valuation models were telling people in the depths of the 1999/2000 recession "Hey this is great, with valuations this low, you don't need to bother saving anything for retirement!"

Most people I know just saved what they could based on income, expenses, and life circumstances, not some measure of valuation.

Do you know anyone that actually does that in real life?
Did I say that? No.

Did I save as much as I could as often as I could? Yes. Did I also use CAPE to estimate when I might be able to retire? Yes.

Do I use CAPE to set a withdrawal estimate that drives a budget now that I am retired? Yes.

AFAIK, I'm living a real life.
“Adapt what is useful, reject what is useless, and add what is specifically your own.” ― Bruce Lee
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by marcopolo »

willthrill81 wrote: Wed Dec 16, 2020 4:08 pm
marcopolo wrote: Wed Dec 16, 2020 4:02 pm
vineviz wrote: Wed Dec 16, 2020 3:45 pm
Seasonal wrote: Wed Dec 16, 2020 12:02 pm
willthrill81 wrote: Wed Dec 16, 2020 11:44 am<quote snipped>
I'm not trying to be rude or snippy at all, but if you believe that CAPE is capable of explaining 40% of market returns, then I think that the burden of proof is on you to show that it's not useful.
<second post>
The burden of proof lies with the one making the claim. Yes, vineviz made a claim, but you have made a claim yourself. How can a statistically significant variable explain 40% of the variance and not be useful in some way?
I have no idea if CAPE can consistently explain 40% of returns. I referred to a paper in which Vanguard found that it did over a defined period and didn't find anything better. Something can easily be the best, especially over a specific historic period, and not continue to even hit that level. My crystal ball is cloudy. It might or might not work going forward.

In other words, I'm not claiming anything beyond that Vanguard made a statement.

I did not say whether or not CAPE is useful. vineviz made a clear statement that it's useful. See above. If vineviz does not belief it is in fact useful, I'd appreciate a clarification.

As you note, the burden of proof is on the one who makes a claim.
I've lost faith that anyone who things CAPE is useless can be persuaded by evidence at this point, but here is some anyway.

Imagine an investor who must decide at the beginning of each year how much to invest for their retirement in 30 years. Call this the savings rate. For the sake of simplicity, let's assume they are 100% invested in equities and that each year they make the decision mechanically using as inputs three numbers: the current balance in the account, the number of years remaining until retirement, and an estimate of future real returns. I use real returns so I can simulate over multiple time periods without bias. The target retirement wealth is $100,000.

There are two options for estimating future real returns. Option 1 is to use a CAPE regression which incorporates only the data available as of 12/31 of the prior year (i.e. no look-ahead bias). Option 2 is to just use the long-run average return as of 12/31 of the prior year. I compared the outcomes for all 81 rolling 30-year periods starting in 1910. Option 1 (using CAPE) produced a higher retirement wealth on average (about 6% higher) with smoother contributions (about 8% less volatility) than not using CAPE.

By way of illustration, here are the annual contributions for the 1989 start years for the investor using CAPE (blue line) and not using CAPE (orange line). Ignoring valuations in formulating the financial plan, as the orange line investor did, produced the kind of sudden and severe adjustments in savings rate that investors generally prefer to avoid.

Image

Now imagine the reverse scenario. A retired investor starts with $100,000 and must decide at the beginning of each year how much of their portfolio to withdraw. Using the same options, with a goal of drawing down the portfolio to zero at the end of 30 years, the investor using CAPE to inform their withdrawal rates had less volatile swings in retirement consumption (about 11% less) than the investor who did not use CAPE to inform their withdrawal rate.

By way of illustration, here are the annual withdrawals for the 1965 start years for the investor using CAPE (blue line) and not using CAPE (orange line). Ignoring valuations in formulating the financial plan, as the orange line investor did, produced the kind of sudden and severe curtailments in consumption that investors generally prefer to avoid.

Image

That is a nice academic excersize.

While I have never said CAPE is useless, this is precisely why I keep questioning how actionable it really is.

How many proponents of valuation models were telling people in the depths of the 1999/2000 recession "Hey this is great, with valuations this low, you don't need to bother saving anything for retirement!"

Most people I know just saved what they could based on income, expenses, and life circumstances, not some measure of valuation.

Do you know anyone that actually does that in real life?
It sounds like you're moving the goalposts.

vineviz demonstrated one effective means of using CAPE, and you're arguing that because a lot of people weren't espousing that 20 years age or because few investors do that today that CAPE has no value.

So what is the actionable thing?

You say don't use it for AA.
You say no one uses it for deciding how much to invest.
Maybe you are suggesting they should? Do You?

What do you suggest, specifically, not some hand waving theoretical, but what specific action should an investor take based on CAPE? Both, when it is high and low.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
Uncorrelated
Posts: 1237
Joined: Sun Oct 13, 2019 3:16 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Uncorrelated »

willthrill81 wrote: Wed Dec 16, 2020 3:53 pm
HomerJ wrote: Wed Dec 16, 2020 3:42 pm The title of this thread is that CAPE is a STRONG predictor of stock returns.

When the opposite has been true in real life.
In the terminology of regression analysis, which was used in the OP, the terms 'independent variable' and 'predictor' are interchangeable.

A variable being a strong predictor of another variable in this type of analysis merely means that there was a strong linear correlation between the variables in the data being analyzed. Nothing more, nothing less.

And from 1989-2010, forward 10 year returns were very strongly correlated with the starting CAPE. That was 'real life' and is not up for debate.
I tried to reproduce your graphs but failed. I was able to reproduce marcopolo's graphs which, at least to my eyes, showed significant robustness failures.

Your conclusion only appears to be true when considering in-sample data. Using the numbers from the oft-quoted Gordon and Welch paper, E/P10 has an R^2 of −2.85% at predicting the forward 5y returns starting from 1902 to 2005, and an R^2 of −25.65% when predicting from 1965 to 2005. For each prediction, the regression parameters where determined based on the data that was actually available in that year (since 1882). This suggests that either the relation is extremely weak, or that access to future data points (data snooping bias) is required in order to get acceptable results.

Image

(there are some numbers that are statistically significant, but that's actually what you would expect given the large numbers of tests performed).

I agree this isn't up for debate, but it appears we disagree on which position is the obvious one.
Post Reply