Current expected return of stocks

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
305pelusa
Posts: 1266
Joined: Fri Nov 16, 2018 10:20 pm

Current expected return of stocks

Post by 305pelusa » Wed Feb 27, 2019 3:23 pm

Hello,
Could anyone point me to good resources for estimating a crude approximation of the forward-looking, long term equity premium?

I get that no one knows for certain. I just want to know a reasonable ballpark number to help guide my Asset Allocation decision a bit more.

Thank you

User avatar
David Jay
Posts: 7501
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Current expected return of stocks

Post by David Jay » Wed Feb 27, 2019 5:31 pm

There are no “good” resources, but lots of people are taking a WAG at it.

For the very long term (30+ years) the total historical performance is probably a starting point. The real total return of the SP500 for the 60 years from 1950 to 2009 is 7%.

Here is another AA resource from Vanguard (note that these numbers are nominal, not real):
https://www.vanguard.com/us/insights/sa ... llocations
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius

bloom2708
Posts: 7089
Joined: Wed Apr 02, 2014 2:08 pm
Location: Fargo, ND

Re: Current expected return of stocks

Post by bloom2708 » Wed Feb 27, 2019 5:53 pm

It is better to match savings with a number that will allow you to meet your goals.

Say that is 4% or Pi% (3.14159%). If you can get to 25x or 30x with savings at some reasonable rate, then you can get there with a higher long term return.

If your plan only works with a 7% or 8% return, it may not work.

Save enough and see what happens.
"People want confirmation, not advice" Unknown | "We are here to provoke thoughtfulness, not agree with you" Unknown | Four words: Whole food, plant based

User avatar
Svensk Anga
Posts: 629
Joined: Sun Dec 23, 2012 5:16 pm

Re: Current expected return of stocks

Post by Svensk Anga » Wed Feb 27, 2019 6:02 pm

FWIW, Prof. Aswath Damodaran estimates the equity risk premium here: http://aswathdamodaran.blogspot.com/201 ... inder.html

User avatar
Topic Author
305pelusa
Posts: 1266
Joined: Fri Nov 16, 2018 10:20 pm

Re: Current expected return of stocks

Post by 305pelusa » Wed Feb 27, 2019 6:31 pm

bloom2708 wrote:
Wed Feb 27, 2019 5:53 pm
It is better to match savings with a number that will allow you to meet your goals.

Say that is 4% or Pi% (3.14159%). If you can get to 25x or 30x with savings at some reasonable rate, then you can get there with a higher long term return.

If your plan only works with a 7% or 8% return, it may not work.

Save enough and see what happens.
I'm not asking to try to figure out how little I could get away saving and still get there. I ask to help guide how much of my portfolio I'd feel comfortable placing in stocks going forward. Think of it this way:

You and I play a coin game. If it's heads, you lose 1 dollar. If it's tails, you earn 1.01 dollars. This seems like a game I think most would have no issues playing (I know I certainly would).
Now say instead that if it's heads, you lose 1 Million dollars (say that's all you have). If it's tails, you earn 1 Million and 10k. Would you play that game? I most definitely would not risk all of my wealth like that. The proportional return is the same but I require a far bigger premium to even consider such a game.

So knowing a rough approximation of the expected returns of stocks is, at least to me personally, very relevant in determining AA. If the premium was, say, 4% then I might be willing to risk, say 30% of my wealth to that volatility. Make the premium 5% and I would reconsider to perhaps 50%. These are just example numbers but I hope they explain my point of view here. :happy
Svensk Anga wrote:
Wed Feb 27, 2019 6:02 pm
FWIW, Prof. Aswath Damodaran estimates the equity risk premium here: http://aswathdamodaran.blogspot.com/201 ... inder.html
Whoah that was fascinating and 100% the kind of thing I was looking for. Thank you very much.

bgf
Posts: 1170
Joined: Fri Nov 10, 2017 9:35 am

Re: Current expected return of stocks

Post by bgf » Wed Feb 27, 2019 7:32 pm

the 2018 semper augustus client letter does a good job of breaking down and explaining the historical ROE (return on equity) of the sp500. it also goes into great detail as to the downward adjustments that should be made to current publicized sp500 earnings.

if you buy the sp500 at multiple "x" and assume you can sell it later down the road at the same multiple, over a long time period you will earn the ROE, minus certain drags like writedowns, reinvested dividends, and share buybacks performed at a premium to book value.

if you arrive at an adjusted earnings number that you feel comfortable with and find the ROE, then you can calculate the expected return based on the premium you paid for the index above book value.

for example, if the ROE is 15% and you pay 3x book value, your expected return is 5%. over a long time, your return will approach the 15% number, minus the drags.

historically, sp500 ROE has been a stable 13-14%, according to the client letter. the drags however, have recently been quite large... also, sp500 price to book value is currently 3.3x. YIKES
Last edited by bgf on Wed Feb 27, 2019 7:37 pm, edited 1 time in total.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Current expected return of stocks

Post by Thesaints » Wed Feb 27, 2019 7:36 pm

305pelusa wrote:
Wed Feb 27, 2019 3:23 pm
Hello,
Could anyone point me to good resources for estimating a crude approximation of the forward-looking, long term equity premium?

I get that no one knows for certain. I just want to know a reasonable ballpark number to help guide my Asset Allocation decision a bit more.

Thank you
The main thing to understand is that such a ballpark figure has to be a range of values. Whoever gives you a single number is fooling you and perhaps himself as well.
The range might be too wide to be of practical use to you, but maybe not.

User avatar
Topic Author
305pelusa
Posts: 1266
Joined: Fri Nov 16, 2018 10:20 pm

Re: Current expected return of stocks

Post by 305pelusa » Wed Feb 27, 2019 7:47 pm

Thesaints wrote:
Wed Feb 27, 2019 7:36 pm
305pelusa wrote:
Wed Feb 27, 2019 3:23 pm
Hello,
Could anyone point me to good resources for estimating a crude approximation of the forward-looking, long term equity premium?

I get that no one knows for certain. I just want to know a reasonable ballpark number to help guide my Asset Allocation decision a bit more.

Thank you
The main thing to understand is that such a ballpark figure has to be a range of values. Whoever gives you a single number is fooling you and perhaps himself as well.
The range might be too wide to be of practical use to you, but maybe not.
Fundamentally, the true expectation shouldn't be a range; it should just be a value. Not getting that return simply reflects the volatile (st. dev) nature of stocks. You'll even get to find out the value once the time period has passed.

However, in practice, I imagine there are multiple accepted ways of estimating the future returns. Different methods would predict different singular values so then you'd get a range of possible future-looking values. I'm hoping that, in general, most methods arrive at similar enough values that this exercise will be useful.

Once again, the actual returns over the time period might not be anywhere near that value. But I'll chuck that up to volatility and risk instead of having started with an incorrect estimate. Does that kinda make sense? Or perhaps I'm not following what you're saying well hehe.

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Current expected return of stocks

Post by Thesaints » Wed Feb 27, 2019 7:53 pm

I'm saying the method itself has to generate a range of values. For instance, that's what Vanguard's proprietary model does.
A model that generates a single value is flawed at start. It is kind of saying that the next roll of a dice will show "3.5"

Beliavsky
Posts: 933
Joined: Sun Jun 29, 2014 10:21 am

Re: Current expected return of stocks

Post by Beliavsky » Wed Feb 27, 2019 8:35 pm

Thesaints wrote:
Wed Feb 27, 2019 7:53 pm
I'm saying the method itself has to generate a range of values. For instance, that's what Vanguard's proprietary model does.
A model that generates a single value is flawed at start. It is kind of saying that the next roll of a dice will show "3.5"
What if the method generates two numbers, the forecasted mean and standard deviation of annual stock market returns?

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Current expected return of stocks

Post by Thesaints » Wed Feb 27, 2019 8:42 pm

Beliavsky wrote:
Wed Feb 27, 2019 8:35 pm
Thesaints wrote:
Wed Feb 27, 2019 7:53 pm
I'm saying the method itself has to generate a range of values. For instance, that's what Vanguard's proprietary model does.
A model that generates a single value is flawed at start. It is kind of saying that the next roll of a dice will show "3.5"
What if the method generates two numbers, the forecasted mean and standard deviation of annual stock market returns?
I'd still need to know something about the probability distribution.

User avatar
Topic Author
305pelusa
Posts: 1266
Joined: Fri Nov 16, 2018 10:20 pm

Re: Current expected return of stocks

Post by 305pelusa » Wed Feb 27, 2019 8:54 pm

Thesaints wrote:
Wed Feb 27, 2019 7:53 pm
I'm saying the method itself has to generate a range of values. For instance, that's what Vanguard's proprietary model does.
A model that generates a single value is flawed at start. It is kind of saying that the next roll of a dice will show "3.5"
Huh? The expected value of a dice roll IS 3.5. So that model generated a single value and it was correct. That disproves that a model that generates a singular value for an expected value is "flawed at start". Ironically, you seemed to provide the counterexample.

Beliavsky
Posts: 933
Joined: Sun Jun 29, 2014 10:21 am

Re: Current expected return of stocks

Post by Beliavsky » Wed Feb 27, 2019 9:02 pm

Thesaints wrote:
Wed Feb 27, 2019 8:42 pm
Beliavsky wrote:
Wed Feb 27, 2019 8:35 pm
Thesaints wrote:
Wed Feb 27, 2019 7:53 pm
I'm saying the method itself has to generate a range of values. For instance, that's what Vanguard's proprietary model does.
A model that generates a single value is flawed at start. It is kind of saying that the next roll of a dice will show "3.5"
What if the method generates two numbers, the forecasted mean and standard deviation of annual stock market returns?
I'd still need to know something about the probability distribution.
The normal distribution is used most often and is natural when the forecasting regression is fit via least squares.

bgf
Posts: 1170
Joined: Fri Nov 10, 2017 9:35 am

Re: Current expected return of stocks

Post by bgf » Wed Feb 27, 2019 9:05 pm

305pelusa wrote:
Wed Feb 27, 2019 8:54 pm
Thesaints wrote:
Wed Feb 27, 2019 7:53 pm
I'm saying the method itself has to generate a range of values. For instance, that's what Vanguard's proprietary model does.
A model that generates a single value is flawed at start. It is kind of saying that the next roll of a dice will show "3.5"
Huh? The expected value of a dice roll IS 3.5. So that model generated a single value and it was correct. That disproves that a model that generates a singular value for an expected value is "flawed at start". Ironically, you seemed to provide the counterexample.
you can't take the ensemble average for compounding stock returns. the time average is what matters. the expected value of a dice roll might be 3.5 over 10000 independent rolls, but that means nothing when your wealth compounds after each individual roll and there is risk of total loss.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Current expected return of stocks

Post by Thesaints » Wed Feb 27, 2019 9:11 pm

305pelusa wrote:
Wed Feb 27, 2019 8:54 pm
Huh? The expected value of a dice roll IS 3.5. So that model generated a single value and it was correct. That disproves that a model that generates a singular value for an expected value is "flawed at start". Ironically, you seemed to provide the counterexample.
Sure. See how much that is gonna help you.

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Current expected return of stocks

Post by Thesaints » Wed Feb 27, 2019 9:12 pm

Beliavsky wrote:
Wed Feb 27, 2019 9:02 pm
The normal distribution is used most often and is natural when the forecasting regression is fit via least squares.
Unfortunately, stock returns ain't normal and left tails kill instantly.

User avatar
Matigas
Posts: 100
Joined: Sun Dec 26, 2010 3:34 pm

Re: Current expected return of stocks

Post by Matigas » Wed Feb 27, 2019 9:16 pm

Dave Ramsey says we can expect a 12% return on stocks.

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Current expected return of stocks

Post by Thesaints » Wed Feb 27, 2019 9:16 pm

Matigas wrote:
Wed Feb 27, 2019 9:16 pm
Dave Ramsey says we can expect a 12% return on stocks.
When ? :)

User avatar
Matigas
Posts: 100
Joined: Sun Dec 26, 2010 3:34 pm

Re: Current expected return of stocks

Post by Matigas » Wed Feb 27, 2019 9:21 pm

That is just the average, so we just need to wait.

User avatar
knpstr
Posts: 2399
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: Current expected return of stocks

Post by knpstr » Wed Feb 27, 2019 9:21 pm

305pelusa wrote:
Wed Feb 27, 2019 3:23 pm
Hello,
Could anyone point me to good resources for estimating a crude approximation of the forward-looking, long term equity premium?

I get that no one knows for certain. I just want to know a reasonable ballpark number to help guide my Asset Allocation decision a bit more.

Thank you
Up
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

Ferdinand2014
Posts: 1001
Joined: Mon Dec 17, 2018 6:49 pm

Re: Current expected return of stocks

Post by Ferdinand2014 » Wed Feb 27, 2019 9:35 pm

https://www.cnbc.com/2017/11/20/jack-bo ... eyond.html


Jack Bogle in this article suggests about 6% nominal and 4% real returns for the U.S. equity market over the next 10 years.

He basis this on a dividend yield of the S&P 500 of about 2% and earnings yield of 4%. Subtract 2% inflation and you get 4% real returns. Plus or minus where the PE lands 10 years from now. This of course could be completely wrong or right. This is also misleading as along the way, you could see +50% or -50% in any given year. In the end, It's fair to say, over the long haul it's going up. With huge bumps along the way. Other than that, who knows.

The equity risk premium may be in the range of 4% nominal based on the T-Bill returns currently. However, some feel (Jeremy Siegel) the short TIPS yield as a better 'risk-less' asset risk premium comparison.

In the end nobody knows
“You only find out who is swimming naked when the tide goes out.“ — Warren Buffett

NibbanaBanana
Posts: 247
Joined: Sun Jan 22, 2017 10:34 pm

Re: Current expected return of stocks

Post by NibbanaBanana » Wed Feb 27, 2019 9:54 pm

10 year annualized returns from Vanguards latest. https://personal.vanguard.com/pdf/ISGVEMO_122018.pdf
3-5% US stocks
3-5% US bonds
6-8% international stocks

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Current expected return of stocks

Post by Thesaints » Wed Feb 27, 2019 9:57 pm

Ferdinand2014 wrote:
Wed Feb 27, 2019 9:35 pm
https://www.cnbc.com/2017/11/20/jack-bo ... eyond.html


Jack Bogle in this article suggests about 6% nominal and 4% real returns for the U.S. equity market over the next 10 years.

He basis this on a dividend yield of the S&P 500 of about 2% and earnings yield of 4%. Subtract 2% inflation and you get 4% real returns. Plus or minus where the PE lands 10 years from now.
That part in bold alone easily means a ±3.5% on top of that 4%, which is not set in stone either. So, at a minimum we are talking of an about zero to +7.5% indetermination in the estimate. Add a little noise on dividends yield and EGR and we are talking -1, maybe -2% to almost 10%.
Incidentally, iirc -2% would correspond to the worst decade ever, while 10% is about the (very) long term average.
In other words, Bogle's forecast means returns could be anywhere from average to the worst ever.

iamlucky13
Posts: 1580
Joined: Sat Mar 04, 2017 5:28 pm
Location: Western Washington

Re: Current expected return of stocks

Post by iamlucky13 » Wed Feb 27, 2019 9:57 pm

As always, that is extremely hard to call. The usual starting point is long term historical data, like these Vanguard return charts:
https://www.vanguard.com/us/insights/sa ... llocations

Of course, there are countless attempts to look not only backwards, but forward as well, based on current conditions. Even Vanguard does this. Their economic and market outlook is here:
https://advisors.vanguard.com/iwe/pdf/ISGVEMO.pdf

The extremely short summary is that they expect near term inflation to stick around 2%, and nominal investment returns over the next decade to be in the range of 2.5%-4.5% for bonds.

They expect global equities to average 4.5%-6.5%, which would be the combination of US-market returns of 3-5% and non-US market returns of 6-8%. The equities forecast is based mainly on the currently high valuations, as I understand it.

The positive range of the forecast over the decade indicates the average over that period. It does not suggest a lack of risk of a an economic recession and/or major market drop during that time.

alex_686
Posts: 5153
Joined: Mon Feb 09, 2015 2:39 pm

Re: Current expected return of stocks

Post by alex_686 » Wed Feb 27, 2019 10:17 pm

David Jay wrote:
Wed Feb 27, 2019 5:31 pm
For the very long term (30+ years) the total historical performance is probably a starting point. The real total return of the SP500 for the 60 years from 1950 to 2009 is 7%.
I would not use historical returns. Returns are based on the underlying structure of the market, which is based on savings rate, capital returns, etc. The risk / return profile undergoes a change every 5 to 15 years. Easy to see in retrospect with the right statistical tools. Harder to act on it.

I will suggest Shiller's Cape 10. It has a has a predictive power of 60%, which I find amazing for such simple and easily understood model. Schiller's model is predicting 3 %to 4% real returns over the next 10 years.

alex_686
Posts: 5153
Joined: Mon Feb 09, 2015 2:39 pm

Re: Current expected return of stocks

Post by alex_686 » Wed Feb 27, 2019 10:22 pm

Thesaints wrote:
Wed Feb 27, 2019 9:11 pm
305pelusa wrote:
Wed Feb 27, 2019 8:54 pm
Huh? The expected value of a dice roll IS 3.5. So that model generated a single value and it was correct. That disproves that a model that generates a singular value for an expected value is "flawed at start". Ironically, you seemed to provide the counterexample.
Sure. See how much that is gonna help you.
I guess you have never played a fighter wielding a 2 handed great sword in D&D? Or tried to figure out how many spaces you are going to go in Monopoly when rolling 2 dice?

The correct answer is a single value with either error bars, a confidence interval, a standard distribution, or some other probability function.

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Current expected return of stocks

Post by Thesaints » Wed Feb 27, 2019 10:38 pm

alex_686 wrote:
Wed Feb 27, 2019 10:22 pm
The correct answer is a single value with either error bars, a confidence interval, a standard distribution, or some other probability function.
Yep. That's exactly what I said in my initial post, except that the distribution doesn't necessarily have to be normal.

Knowing that the expected value of a roll is 3.5 (and nothing else), does not help you a bit with a single roll. First of all, it would be the same for an ordinary 1-6 dice and for a dice with "3.5" on all of its faces. Secondly, on an ordinary dice, you can be sure 3.5 will never come up, while all the possible results are equally probable.
Assigning a single value to the outcome of a stochastic process is fundamentally flawed. By itself it removes its stochastic nature and induces the unwary observer into forgetting about it entirely.

User avatar
Topic Author
305pelusa
Posts: 1266
Joined: Fri Nov 16, 2018 10:20 pm

Re: Current expected return of stocks

Post by 305pelusa » Wed Feb 27, 2019 11:34 pm

bgf wrote:
Wed Feb 27, 2019 9:05 pm
305pelusa wrote:
Wed Feb 27, 2019 8:54 pm
Thesaints wrote:
Wed Feb 27, 2019 7:53 pm
I'm saying the method itself has to generate a range of values. For instance, that's what Vanguard's proprietary model does.
A model that generates a single value is flawed at start. It is kind of saying that the next roll of a dice will show "3.5"
Huh? The expected value of a dice roll IS 3.5. So that model generated a single value and it was correct. That disproves that a model that generates a singular value for an expected value is "flawed at start". Ironically, you seemed to provide the counterexample.
you can't take the ensemble average for compounding stock returns. the time average is what matters. the expected value of a dice roll might be 3.5 over 10000 independent rolls, but that means nothing when your wealth compounds after each individual roll and there is risk of total loss.
Sure you can. If side 1 meant you earned 1%, side 2 meant 2%... side 6 meant 6% then the compounded return after 10000 will be approximately 3.48%. In this example, it's nearly identical to 3.5%.

To be fair, yes, I understand your point. I agree the forecasted compound return of stocks would be even more useful to know. I just figured the expected or average is close enough and I figured it'd be much easier to guesstimate.
Thesaints wrote:
Wed Feb 27, 2019 10:38 pm
alex_686 wrote:
Wed Feb 27, 2019 10:22 pm
The correct answer is a single value with either error bars, a confidence interval, a standard distribution, or some other probability function.
Knowing that the expected value of a roll is 3.5 (and nothing else), does not help you a bit with a single roll.
That's true. But I'm not planning on investing in the market for one year. I'm looking to do it for 40. If I was planning on rolling the dice 40 times and I had a reasonable idea of its standard deviation (which I can take an educated guess for stocks), then knowing the expected value is 3.5 is very useful.

Thesaints wrote:
Wed Feb 27, 2019 10:38 pm

Secondly, on an ordinary dice, you can be sure 3.5 will never come up, while all the possible results are equally probable.
This is a weak argument IMO. If you had a 5-sided dice, then its expected value could also come up. Apparently, in that case, you would feel like the expected value is a more useful number. Stock returns behave like 5-sided dice in the sense that any return could happen so any reasonable guess can in fact occur.

So your critique that forecasting expected stock returns is "flawed" because it's like a 6 sided die where the expected value cannot ever come up... is not well found since any expected value, like 5.297%, could very well occur in reality.


It sounds like you think I'm going to reduce the entire process to one singular number; the expected value. I won't. I will be taking volatility/standard deviation into account as well in my decision-making. Perhaps what you refer to as "range of expected values", I'm simply calling volatility. It's the probability that I do not encounter that expected return.

I think we're all on the same page

User avatar
Topic Author
305pelusa
Posts: 1266
Joined: Fri Nov 16, 2018 10:20 pm

Re: Current expected return of stocks

Post by 305pelusa » Wed Feb 27, 2019 11:40 pm

Thesaints wrote:
Wed Feb 27, 2019 9:57 pm
Ferdinand2014 wrote:
Wed Feb 27, 2019 9:35 pm
https://www.cnbc.com/2017/11/20/jack-bo ... eyond.html


Jack Bogle in this article suggests about 6% nominal and 4% real returns for the U.S. equity market over the next 10 years.

He basis this on a dividend yield of the S&P 500 of about 2% and earnings yield of 4%. Subtract 2% inflation and you get 4% real returns. Plus or minus where the PE lands 10 years from now.
That part in bold alone easily means a ±3.5% on top of that 4%, which is not set in stone either. So, at a minimum we are talking of an about zero to +7.5% indetermination in the estimate. Add a little noise on dividends yield and EGR and we are talking -1, maybe -2% to almost 10%.
Incidentally, iirc -2% would correspond to the worst decade ever, while 10% is about the (very) long term average.
In other words, Bogle's forecast means returns could be anywhere from average to the worst ever.
The +-3.5% is likely to happen either direction. So it doesn't affect the expected value. The expected value is still 4% in that example. The standard deviation (a measure that quantifies how probable you could miss that expected value, and by how much) is a different number that can also be computed. That one will take that +-3.5% into account.

There's no "range of expected values". There is an expected value (a weighted average) and then there are factors that can make it likely to get more or less than that. That one gets computed by knowing the range of possible values.

Ok, that's the final word on this. Hope it's clear. I appreciate the Bogle reference Ferdinand!
Last edited by 305pelusa on Wed Feb 27, 2019 11:50 pm, edited 1 time in total.

klaus14
Posts: 335
Joined: Sun Nov 25, 2018 7:43 pm

Re: Current expected return of stocks

Post by klaus14 » Wed Feb 27, 2019 11:50 pm

Ferdinand2014 wrote:
Wed Feb 27, 2019 9:35 pm
https://www.cnbc.com/2017/11/20/jack-bo ... eyond.html


Jack Bogle in this article suggests about 6% nominal and 4% real returns for the U.S. equity market over the next 10 years.

He basis this on a dividend yield of the S&P 500 of about 2% and earnings yield of 4%. Subtract 2% inflation and you get 4% real returns. Plus or minus where the PE lands 10 years from now. This of course could be completely wrong or right. This is also misleading as along the way, you could see +50% or -50% in any given year. In the end, It's fair to say, over the long haul it's going up. With huge bumps along the way. Other than that, who knows.

The equity risk premium may be in the range of 4% nominal based on the T-Bill returns currently. However, some feel (Jeremy Siegel) the short TIPS yield as a better 'risk-less' asset risk premium comparison.

In the end nobody knows
I don't understand why do you add dividend yield to earnings. Isn't dividend after earnings?
Also, why subtract inflation from earnings yield?

But in the end, i reach to the same number.
Earnings yield = real returns.
This assumes PE will stay the same and earnings will grow similar to inflation.

corporate profits to GDP ratio is at record levels so i don't think earnings will grow faster.
PE is relatively high, i don't think it will get higher.

So there you have it: expect 3.2% real return based on optimistic assumptions as of today.
5% ACWV | 10% VFMF, 15% NTSX | 10% ISCF, 5% EFAV | 5% FNDE, 5% EMGF, 5% VEGBX, 5% LEMB | 15% EDV, 5% CD (5y), 5% I/EE Bonds | 10% GLDM

User avatar
Topic Author
305pelusa
Posts: 1266
Joined: Fri Nov 16, 2018 10:20 pm

Re: Current expected return of stocks

Post by 305pelusa » Thu Feb 28, 2019 12:19 am

bgf wrote:
Wed Feb 27, 2019 7:32 pm
the 2018 semper augustus client letter does a good job of breaking down and explaining the historical ROE (return on equity) of the sp500. it also goes into great detail as to the downward adjustments that should be made to current publicized sp500 earnings.

if you buy the sp500 at multiple "x" and assume you can sell it later down the road at the same multiple, over a long time period you will earn the ROE, minus certain drags like writedowns, reinvested dividends, and share buybacks performed at a premium to book value.

if you arrive at an adjusted earnings number that you feel comfortable with and find the ROE, then you can calculate the expected return based on the premium you paid for the index above book value.

for example, if the ROE is 15% and you pay 3x book value, your expected return is 5%. over a long time, your return will approach the 15% number, minus the drags.

historically, sp500 ROE has been a stable 13-14%, according to the client letter. the drags however, have recently been quite large... also, sp500 price to book value is currently 3.3x. YIKES
Very interesting. I will take a look at it. Thank you very much!
Ferdinand2014 wrote:
Wed Feb 27, 2019 9:35 pm
https://www.cnbc.com/2017/11/20/jack-bo ... eyond.html


Jack Bogle in this article suggests about 6% nominal and 4% real returns for the U.S. equity market over the next 10 years.

He basis this on a dividend yield of the S&P 500 of about 2% and earnings yield of 4%. Subtract 2% inflation and you get 4% real returns. Plus or minus where the PE lands 10 years from now. This of course could be completely wrong or right. This is also misleading as along the way, you could see +50% or -50% in any given year. In the end, It's fair to say, over the long haul it's going up. With huge bumps along the way. Other than that, who knows.

The equity risk premium may be in the range of 4% nominal based on the T-Bill returns currently. However, some feel (Jeremy Siegel) the short TIPS yield as a better 'risk-less' asset risk premium comparison.

In the end nobody knows
Ok so earnings yield + dividend yield. Got it, thanks! I thought it was dividend yield + dividend growth (Gordon's equation?), but these two give different answers.
alex_686 wrote:
Wed Feb 27, 2019 10:17 pm
David Jay wrote:
Wed Feb 27, 2019 5:31 pm
For the very long term (30+ years) the total historical performance is probably a starting point. The real total return of the SP500 for the 60 years from 1950 to 2009 is 7%.
I would not use historical returns. Returns are based on the underlying structure of the market, which is based on savings rate, capital returns, etc. The risk / return profile undergoes a change every 5 to 15 years. Easy to see in retrospect with the right statistical tools. Harder to act on it.

I will suggest Shiller's Cape 10. It has a has a predictive power of 60%, which I find amazing for such simple and easily understood model. Schiller's model is predicting 3 %to 4% real returns over the next 10 years.
How are you getting that prediction from the Shiller's Cape 10? Is it just the inverse?

Thank you very much!

User avatar
Topic Author
305pelusa
Posts: 1266
Joined: Fri Nov 16, 2018 10:20 pm

Re: Current expected return of stocks

Post by 305pelusa » Thu Feb 28, 2019 12:22 am

klaus14 wrote:
Wed Feb 27, 2019 11:50 pm
Ferdinand2014 wrote:
Wed Feb 27, 2019 9:35 pm
https://www.cnbc.com/2017/11/20/jack-bo ... eyond.html


Jack Bogle in this article suggests about 6% nominal and 4% real returns for the U.S. equity market over the next 10 years.

He basis this on a dividend yield of the S&P 500 of about 2% and earnings yield of 4%. Subtract 2% inflation and you get 4% real returns. Plus or minus where the PE lands 10 years from now. This of course could be completely wrong or right. This is also misleading as along the way, you could see +50% or -50% in any given year. In the end, It's fair to say, over the long haul it's going up. With huge bumps along the way. Other than that, who knows.

The equity risk premium may be in the range of 4% nominal based on the T-Bill returns currently. However, some feel (Jeremy Siegel) the short TIPS yield as a better 'risk-less' asset risk premium comparison.

In the end nobody knows
I don't understand why do you add dividend yield to earnings. Isn't dividend after earnings?
Also, why subtract inflation from earnings yield?

But in the end, i reach to the same number.
Earnings yield = real returns.
This assumes PE will stay the same and earnings will grow similar to inflation.

corporate profits to GDP ratio is at record levels so i don't think earnings will grow faster.
PE is relatively high, i don't think it will get higher.

So there you have it: expect 3.2% real return based on optimistic assumptions as of today.
Could you tell me where to find this data? I tried http://www.multpl.com/s-p-500-earnings-yield
But that seems to say 4.67% as of today.

Any ways, yes, these kind of estimations with dividends, earnings, etc are the kind of thing that I'm looking for. Just some vague approximations, ideally multiple ones of them, so that I could always make some approximations in the future. Before starting the thread, I just had no guideline at all.

klaus14
Posts: 335
Joined: Sun Nov 25, 2018 7:43 pm

Re: Current expected return of stocks

Post by klaus14 » Thu Feb 28, 2019 12:28 am

305pelusa wrote:
Thu Feb 28, 2019 12:22 am
Could you tell me where to find this data? I tried http://www.multpl.com/s-p-500-earnings-yield
But that seems to say 4.67% as of today.

Any ways, yes, these kind of estimations with dividends, earnings, etc are the kind of thing that I'm looking for. Just some vague approximations, ideally multiple ones of them, so that I could always make some approximations in the future. Before starting the thread, I just had no guideline at all.
You have the right number. I accidentally used shiller pe instead of trailing 12 months pe.
So it's 4.67%.
5% ACWV | 10% VFMF, 15% NTSX | 10% ISCF, 5% EFAV | 5% FNDE, 5% EMGF, 5% VEGBX, 5% LEMB | 15% EDV, 5% CD (5y), 5% I/EE Bonds | 10% GLDM

User avatar
HomerJ
Posts: 13561
Joined: Fri Jun 06, 2008 12:50 pm

Re: Current expected return of stocks

Post by HomerJ » Thu Feb 28, 2019 1:06 am

305pelusa wrote:
Wed Feb 27, 2019 11:40 pm
The +-3.5% is likely to happen either direction.
This is not a known fact. There are not enough data points to tell if stock market returns are a normal distribution.
The J stands for Jay

User avatar
HomerJ
Posts: 13561
Joined: Fri Jun 06, 2008 12:50 pm

Re: Current expected return of stocks

Post by HomerJ » Thu Feb 28, 2019 1:14 am

305pelusa wrote:
Wed Feb 27, 2019 3:23 pm
I get that no one knows for certain. I just want to know a reasonable ballpark number to help guide my Asset Allocation decision a bit more.
305pelusa wrote:
Wed Feb 27, 2019 6:31 pm
So knowing a rough approximation of the expected returns of stocks is, at least to me personally, very relevant in determining AA. If the premium was, say, 4% then I might be willing to risk, say 30% of my wealth to that volatility. Make the premium 5% and I would reconsider to perhaps 50%. These are just example numbers but I hope they explain my point of view here. :happy
The ballpark is too big for you to make large changes to your AA based on a 1% difference in expected returns.

What you are trying to do is time the market. You want to invest more when the market is likely to do well going forward, and invest less when the market is likely to do poorly.

It is not that easy.

The long-term historical return of the market is 7% real. That INCLUDES the crashes and the bad years. Read that again.

You didn't have to avoid the bad times to become wealthy in the stock market in the past.

Sure, it would be be great if you could invest more when the market is doing well, and invest less when the market is going down. Many many people have tried and failed. With many different systems.

Just making decisions on "expected" returns is likely to under-perform.

Expected returns in 2011 were 4.5% real. Instead we got like 11% real since then. That's not even close to the same ballpark. That's barely in the same city.

Shiller himself predicted 0% 10-year expected returns in 1996. Instead we got 6% real, very close to the historical average.

It is very hard to time the market successfully. Nobody knows enough. There are too many variables.
The J stands for Jay

User avatar
RickBoglehead
Posts: 5608
Joined: Wed Feb 14, 2018 9:10 am
Location: In a house

Re: Current expected return of stocks

Post by RickBoglehead » Thu Feb 28, 2019 6:51 am

If one reviewed every prediction ten years ago or twenty years ago or anytime, the irrational nature of the stock market would prove that there is no model to predict the future that has any practical value.. Therefore, one figures out the return they expect and plans for that, and is wrong.
Avid user of forums on variety of interests-financial, home brewing, F-150, PHEV, home repair, etc. Enjoy learning & passing on knowledge. It's PRINCIPAL, not PRINCIPLE. I ADVISE you to seek ADVICE.

User avatar
Topic Author
305pelusa
Posts: 1266
Joined: Fri Nov 16, 2018 10:20 pm

Re: Current expected return of stocks

Post by 305pelusa » Thu Feb 28, 2019 7:14 am

HomerJ wrote:
Thu Feb 28, 2019 1:14 am

What you are trying to do is time the market. You want to invest more when the market is likely to do well going forward, and invest less when the market is likely to do poorly.
Timing the market, AFAIK, refers to short-term tactical AA changes. And it's adopted, generally, as a way to beat the market. I'm looking for forecasts of long-term stock returns to help guide my long-term AA. I'm not looking to beat the market; I will accept the market returns by buying-and-holding.

It sounds like you also use an expectation (a historical one) to help guide your AA. If you knew the stock returns were 1% for the past 130 years, with the same volatility, would you dedicate as much to stocks as now? Of course not.

If peeps use past returns to guide AA, then that's acceptable. If someone uses long-term approximate expectations, then I'm timing the market?
HomerJ wrote:
Thu Feb 28, 2019 1:14 am

You didn't have to avoid the bad times to become wealthy in the stock market in the past.

Sure, it would be be great if you could invest more when the market is doing well, and invest less when the market is going down. Many many people have tried and failed. With many different systems.
Writes Bernstein in Four Pillars of Investing:
"Recall from Chapter 2 that it is likely that long-term stock returns will not be much greater than bond returns. In such an environment, we find it hard to recommend an all-stock portfolio; 80% would seem to be a reasonable upper limit at the present time." Page 264.

Is Bernstein market timing with this recommendation? Is he telling you to "avoid the bad times"? I certainly don't see the above as a market-timing scheme aimed at trying to avoid bad times, or improve upon the market. Just a very real possibility that great historical portfolios are not as likely to be great now because forecasted returns have changed. So adjust the AA accordingly.

It sounds like you disagree and that's cool. I appreciate your thoughts as well :)

Ferdinand2014
Posts: 1001
Joined: Mon Dec 17, 2018 6:49 pm

Re: Current expected return of stocks

Post by Ferdinand2014 » Thu Feb 28, 2019 7:19 am

klaus14 wrote:
Wed Feb 27, 2019 11:50 pm
Ferdinand2014 wrote:
Wed Feb 27, 2019 9:35 pm
https://www.cnbc.com/2017/11/20/jack-bo ... eyond.html


Jack Bogle in this article suggests about 6% nominal and 4% real returns for the U.S. equity market over the next 10 years.

He basis this on a dividend yield of the S&P 500 of about 2% and earnings yield of 4%. Subtract 2% inflation and you get 4% real returns. Plus or minus where the PE lands 10 years from now. This of course could be completely wrong or right. This is also misleading as along the way, you could see +50% or -50% in any given year. In the end, It's fair to say, over the long haul it's going up. With huge bumps along the way. Other than that, who knows.

The equity risk premium may be in the range of 4% nominal based on the T-Bill returns currently. However, some feel (Jeremy Siegel) the short TIPS yield as a better 'risk-less' asset risk premium comparison.

In the end nobody knows
I don't understand why do you add dividend yield to earnings. Isn't dividend after earnings?
Also, why subtract inflation from earnings yield?

But in the end, i reach to the same number.
Earnings yield = real returns.
This assumes PE will stay the same and earnings will grow similar to inflation.

corporate profits to GDP ratio is at record levels so i don't think earnings will grow faster.
PE is relatively high, i don't think it will get higher.

So there you have it: expect 3.2% real return based on optimistic assumptions as of today.
Jack Bogle describes returns of equities estimate as (dividend yield + earnings yield) +/- speculative return (variation of the PE over time)=annual growth of the stock market. This is nominal. Inflation must also be factored. This is true over longer periods of time, but over short periods of days, months and even a few years not always true. Remember also, dividend yield is not the same as dividend growth. To maintain a dividend yield of 2% (currently for the S&P 500) you have to have a dividend growth that rises commensurate to the growth of the S&P 500. As of 12/2018 dividend growth was over 9%. The mean over the past 10 years was about 6% just to maintain a dividend yield of about 2%.

From his book Common Sense - 10th edition:

" History, if only we would take the trouble to examine it, reveals the remarkable, if essential, linkage between the cumulative long-term returns earned by U.S. business—the annual dividend yield plus the annual rate of earnings growth—and the cumulative returns earned by the stock market."

Bogle, John C.. The Little Book of Common Sense Investing: The Only Way to Guarantee Your Fair Share of Stock Market Returns (Little Books. Big Profits) (p. 10). Wiley. Kindle Edition.

"The price/earnings (P/E) ratio measures the number of dollars investors are willing to pay for each dollar of earnings. As investor confidence waxes and wanes, P/E multiples rise and fall.1 When greed holds sway, very high P/Es are likely. When hope prevails, P/Es are moderate. When fear is in the saddle, P/Es are typically very low. ​Back and forth, over and over again, swings in the emotions of investors are reflected in speculative return."

Bogle, John C.. The Little Book of Common Sense Investing: The Only Way to Guarantee Your Fair Share of Stock Market Returns (Little Books. Big Profits) (pp. 11-12). Wiley. Kindle Edition.

Another very simplified way to guesstimate growth is to take the inverse percent of the PE. Since 1802, According to Jeremy Siegel (Stocks for the Long Run) the average real return of the U.S. stock market was 1/15 or 6.6%. As it turns out the average PE since 1802 was 15. So a PE of 19 might suggest a return of 5.2% real. This is generally true over the very long term. He also points out that long term PE ratios might have a new higher plateau for many reasons outlined in his book including ironically lower fees and costs of trading.
“You only find out who is swimming naked when the tide goes out.“ — Warren Buffett

MathWizard
Posts: 3712
Joined: Tue Jul 26, 2011 1:35 pm

Re: Current expected return of stocks

Post by MathWizard » Thu Feb 28, 2019 7:47 am

The Boglehead wiki has an article historical and expected returns with lots of predictions, including one from Rick Ferri's former company that I like.

The predictions are from 2015, but for a 30 yr prediction you can adjust for the last 3-4 years.

tibbitts
Posts: 9402
Joined: Tue Feb 27, 2007 6:50 pm

Re: Current expected return of stocks

Post by tibbitts » Thu Feb 28, 2019 8:39 am

305pelusa wrote:
Wed Feb 27, 2019 3:23 pm
Hello,
Could anyone point me to good resources for estimating a crude approximation of the forward-looking, long term equity premium?

I get that no one knows for certain. I just want to know a reasonable ballpark number to help guide my Asset Allocation decision a bit more.

Thank you
One issue is whether "long term" is longer than your remaining investing lifetime (I'd say so, no matter your age), and if so, are you really investing for the long term or for some short or intermediate term?

User avatar
Svensk Anga
Posts: 629
Joined: Sun Dec 23, 2012 5:16 pm

Re: Current expected return of stocks

Post by Svensk Anga » Thu Feb 28, 2019 9:18 am

Ferdinand2014 wrote:
Thu Feb 28, 2019 7:19 am

Jack Bogle describes returns of equities estimate as (dividend yield + earnings yield) +/- speculative return (variation of the PE over time)=annual growth of the stock market. This is nominal. Inflation must also be factored.
This has been repeated a couple times up thread but it is not quite right. The formulation is dividend yield + earnings growth rate +/- speculative return. The earnings growth rate can be either real or nominal and the result either real or nominal stock return. Some places use dividend growth rate instead of earnings growth rate, but if the payout ratio remains constant, they are the same. The first two terms are known as the Gordon equation and last term was added by Bogle to account for gyrations in PE.

User avatar
HomerJ
Posts: 13561
Joined: Fri Jun 06, 2008 12:50 pm

Re: Current expected return of stocks

Post by HomerJ » Thu Feb 28, 2019 10:22 am

305pelusa wrote:
Thu Feb 28, 2019 7:14 am
HomerJ wrote:
Thu Feb 28, 2019 1:14 am

What you are trying to do is time the market. You want to invest more when the market is likely to do well going forward, and invest less when the market is likely to do poorly.
Timing the market, AFAIK, refers to short-term tactical AA changes. And it's adopted, generally, as a way to beat the market. I'm looking for forecasts of long-term stock returns to help guide my long-term AA. I'm not looking to beat the market; I will accept the market returns by buying-and-holding.
Okay, so you're looking for a long-term prediction for a long-term buy-and-hold AA. That indeed is not market-timing.

"Expected" return forecasts however, are all shorter-term, and have large standard deviations. There is no "expert" ballpark figure within 1% for long-term returns.

The historical average is probably your best bet, for a long-term prediction.

Current "expected" returns are based on the fact that valuations are high, and people expect them to return to the mean. That is not unreasonable. The market has indeed moved in cycles in the past. Bull market, followed by bear market, followed by bull market, etc.

But periods of low return have been followed by periods of high return. People are predicting low returns in the near-term. That doesn't mean there won't be another bull market with ten years of 14% returns following that bad period. Giving you the historical average.

If you are indeed looking for a long-term AA, then historical average is probably what you want to look at.

I certainly wouldn't COUNT on getting the historical average.. I'd plug in more conservative numbers to make sure my savings were enough to hit my goals even if long-term returns were lower.

But your hope of getting a solid long-term return prediction is unreasonable. NO ONE can tell you if the long-term stock return is more likely to be 4% or 5% (using your examples). That type of precision does not exist.

You're going to have use a different method to come up with an Asset Allocation.
The J stands for Jay

dbr
Posts: 31213
Joined: Sun Mar 04, 2007 9:50 am

Re: Current expected return of stocks

Post by dbr » Thu Feb 28, 2019 10:40 am

What has to be predicted is a probability distribution of returns. Doing that requires postulating the form of the distribution and estimates of the parameters of the distribution (expected return, expected variability of return) The estimates of the parameters are subject to estimation error. As an example 5% and 4% for expected return are probably estimates that are within estimation error of each other.

Actual results will vary from the average expected due to being a random sample from a distribution and due to the estimates of what that distribution is being themselves uncertain. So the question is what kind of predictions are being attempted from this data?

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Current expected return of stocks

Post by Thesaints » Thu Feb 28, 2019 11:22 am

klaus14 wrote:
Wed Feb 27, 2019 11:50 pm
I don't understand why do you add dividend yield to earnings. Isn't dividend after earnings?.
You are correct, but the actual formula says “earnings growth”, not “earnings”.
Clearly, if earnings stay constant and P/E does not change all one gets is the div.

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Current expected return of stocks

Post by Thesaints » Thu Feb 28, 2019 11:35 am

305pelusa wrote:
Wed Feb 27, 2019 11:40 pm
Thesaints wrote:
Wed Feb 27, 2019 9:57 pm
Ferdinand2014 wrote:
Wed Feb 27, 2019 9:35 pm
https://www.cnbc.com/2017/11/20/jack-bo ... eyond.html


Jack Bogle in this article suggests about 6% nominal and 4% real returns for the U.S. equity market over the next 10 years.

He basis this on a dividend yield of the S&P 500 of about 2% and earnings yield of 4%. Subtract 2% inflation and you get 4% real returns. Plus or minus where the PE lands 10 years from now.
That part in bold alone easily means a ±3.5% on top of that 4%, which is not set in stone either. So, at a minimum we are talking of an about zero to +7.5% indetermination in the estimate. Add a little noise on dividends yield and EGR and we are talking -1, maybe -2% to almost 10%.
Incidentally, iirc -2% would correspond to the worst decade ever, while 10% is about the (very) long term average.
In other words, Bogle's forecast means returns could be anywhere from average to the worst ever.
The +-3.5% is likely to happen either direction. So it doesn't affect the expected value. The expected value is still 4% in that example. The standard deviation (a measure that quantifies how probable you could miss that expected value, and by how much) is a different number that can also be computed. That one will take that +-3.5% into account.

There's no "range of expected values". There is an expected value (a weighted average) and then there are factors that can make it likely to get more or less than that. That one gets computed by knowing the range of possible values.

Ok, that's the final word on this. Hope it's clear. I appreciate the Bogle reference Ferdinand!
I don’t think you appreciate the issue.
At the end of those 10 years you will get one single value for the average yearly market return.
Yet, at the beginning of those 10 years, you don’t know with certainty what that value will turn out to be.
You want a single value ex-ante as well, which means erasing the fact that the process is stochastic and seem happy with that estimate that says “4%”, “4.67%”, or whatever.
Reflect on what that means. A few posts above we discussed how Bogle’s forecast implicitly means that the expected return is around 4%, but with a range of possible outcomes spanning from almost 10% down to -2%.
The expected return value only means that if we were able to replicate the next 10 years over and over again, the average return would be near 4%. But we can’t. You only get to experience a single possible outcome for the next 10 years, so the average value is meaningless.
Nor is the 4% value any more "typical" than others. Yes, an interval, let’s say 3.5-4.5% is a likely more probable than any other 1% wide interval, but even if we assign it a very generous 20-25% chance it only means that there is a 75-80% chance actual return won’t be around 4%.
So, I’m asking again, what good the expected return information alone does to you ?
Last edited by Thesaints on Thu Feb 28, 2019 2:15 pm, edited 3 times in total.

alex_686
Posts: 5153
Joined: Mon Feb 09, 2015 2:39 pm

Re: Current expected return of stocks

Post by alex_686 » Thu Feb 28, 2019 1:17 pm

305pelusa wrote:
Thu Feb 28, 2019 12:19 am
How are you getting that prediction from the Shiller's Cape 10? Is it just the inverse?
First, to give this better context, the earnings yield is the E/P ratio, which is just the inverse of the P/E. I think the E/P yield is better - it lets you easily compare a stock's yield to a bond's yield and isolate the risk / return aspect of the stock. It kind of a historical quirk that we user P/E

Yes, that is how you do it for all P/E ratios - forward, backwards, CAPE10, etc. Or at least that is the first step. There are extra steps that can refine the calculations. But these require more estimates, add more complexity, and the extra value might not be worth it.

User avatar
Topic Author
305pelusa
Posts: 1266
Joined: Fri Nov 16, 2018 10:20 pm

Re: Current expected return of stocks

Post by 305pelusa » Thu Feb 28, 2019 2:15 pm

alex_686 wrote:
Thu Feb 28, 2019 1:17 pm
305pelusa wrote:
Thu Feb 28, 2019 12:19 am
How are you getting that prediction from the Shiller's Cape 10? Is it just the inverse?
First, to give this better context, the earnings yield is the E/P ratio, which is just the inverse of the P/E. I think the E/P yield is better - it lets you easily compare a stock's yield to a bond's yield and isolate the risk / return aspect of the stock. It kind of a historical quirk that we user P/E

Yes, that is how you do it for all P/E ratios - forward, backwards, CAPE10, etc. Or at least that is the first step. There are extra steps that can refine the calculations. But these require more estimates, add more complexity, and the extra value might not be worth it.
Awesome thank you very much. This is a useful metric to first order.

User avatar
Topic Author
305pelusa
Posts: 1266
Joined: Fri Nov 16, 2018 10:20 pm

Re: Current expected return of stocks

Post by 305pelusa » Thu Feb 28, 2019 2:22 pm

dbr wrote:
Thu Feb 28, 2019 10:40 am
What has to be predicted is a probability distribution of returns. Doing that requires postulating the form of the distribution and estimates of the parameters of the distribution (expected return, expected variability of return) The estimates of the parameters are subject to estimation error. As an example 5% and 4% for expected return are probably estimates that are within estimation error of each other.

Actual results will vary from the average expected due to being a random sample from a distribution and due to the estimates of what that distribution is being themselves uncertain. So the question is what kind of predictions are being attempted from this data?
I will assume a normal distribution, will take the expected forecast returns and standard deviation (which is fairly constant throughout history). From there, I can figure out what I think is a reasonable AA.

If that AA recommends exposing me to less stocks than a regular "age in bonds" recommended by most here, then I'll go with my more conservative prediction. If it recommends exposing me to more, I'll have to reconsider.

I know this will make me more conservative than all of you, but like someone said before, I can make it up with extra savings.

Thesaints
Posts: 2921
Joined: Tue Jun 20, 2017 12:25 am

Re: Current expected return of stocks

Post by Thesaints » Thu Feb 28, 2019 2:40 pm

305pelusa wrote:
Thu Feb 28, 2019 2:22 pm
I will assume a normal distribution,
That would be a mistake.
The -2% lower boundary in the 10-year return needs:
- lower end of P/E at period's end estimate (P/E≈13)
- lower end of dividend yield (≈ 1.5%)
- lower end of real earnings growth (≈ 1%)

One could say that it would take a lot of things to go wrong in order to achieve that -2% figure, which makes it very unlikely and when tails are unlikely using the normal distribution is not that bad, etc. etc.
That someone would be in fact committing the same error those packaging and rating MBS did in 2007. If real growth rates end up to be at the lower end of the forecast, dividend yield almost certainly will also be lower than expected and final P/E will also be low. That makes the tails of the distribution a lot fatter (same argument applies to the other, rosiest, side) and using a normal distribution, were the expected return value is also the most likely outcome and the farther we move away from it the less likely an outcome becomes, is quite wrong.
I know this will make me more conservative than all of you, but like someone said before, I can make it up with extra savings.
Your estimate is actually rather optimistic.

User avatar
HomerJ
Posts: 13561
Joined: Fri Jun 06, 2008 12:50 pm

Re: Current expected return of stocks

Post by HomerJ » Thu Feb 28, 2019 2:44 pm

305pelusa wrote:
Thu Feb 28, 2019 2:22 pm
dbr wrote:
Thu Feb 28, 2019 10:40 am
What has to be predicted is a probability distribution of returns. Doing that requires postulating the form of the distribution and estimates of the parameters of the distribution (expected return, expected variability of return) The estimates of the parameters are subject to estimation error. As an example 5% and 4% for expected return are probably estimates that are within estimation error of each other.

Actual results will vary from the average expected due to being a random sample from a distribution and due to the estimates of what that distribution is being themselves uncertain. So the question is what kind of predictions are being attempted from this data?
I will assume a normal distribution, will take the expected forecast returns and standard deviation (which is fairly constant throughout history). From there, I can figure out what I think is a reasonable AA.
What's your algorithm for determining long-term AA based on short-term expected returns?

Are you going to change your AA if the "experts" change their "expected" return forecasts? (and they will).
The J stands for Jay

Post Reply