How to think about expected returns--continuing discussion

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: How to think about expected returns--continuing discussi

Post by Rodc »

larryswedroe wrote:Rod
Very much disagree about its usefulness. Example, if you see reducing your spending rate by 5% increases your odds of success by 10% that is helpful. If decreasing your equity allocation by 10% increases your odds of not running out of money by 10% but decreases your odds of leaving an estate of say $1mm by 20%, that is useful information. And similarly seeing how tilting impacts your odds of success, or allocation more to EM or international. Models are just that, like 3 factor or five factor or 7 factor models, they help us make more informed decisions in the presence of uncertainty. Understanding the flaws is also important.
You use the best tools that you have
Larry
Those sorts of sensitivity analyses are not bad as long as one remembers there is at best one significant digit of accuracy. Those are really just illustrations rather than accurate plans in my opinion, again like MPT is useful for illustrations.

But all those examples have nothing to do with the complaints about your opening post and thus nothing really to do with this thread.

You claimed MC was key to making use of time varying valuation based expected return models but then claim that you don't need to even think about time varying MC because the effect of time varying valuations washes out in the noise.

This is very simple. If you start and hold a low or high expected return in your MC you introduce a distinct bias relative to a real markets where the valuations vary considerably over time. Consider starting a MC back in 2001 at the top of the market boom. The expected return would be something extremely low. Now consider putting in $10,000 every year for 30 years and each year using that starting 2001 expected mean. How can that make any sense?
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

Rod
First it is important as I said to recognize that the MCS is a tool, and IMO and the opinion of many highly qualified math professionals, world class mathematicians, that is the best tool we have. Just like any model of the world it is flawed (or wrong), but that doesn't mean it doesn't have value, because it is still the best we have. No different than using CAPE 10 as the best tool to estimate returns. You use the best tools one has, and understand it's limitations
Second, your example of say starting in 2000 (I assume you meant that not 2001) you do have low expected returns, and nothing changes based on what we know after the fact. The outcome we know occurred was surely in the potential dispersion just as the 2008 period was in the bottom 5% of the distribution we used prior to that. One accepts the possibility of that outcome
Third, one doesn't or shouldn't set it and forget it. You run an MCS and then when returns are actually significantly different in future than the mean, you run a new one to give you the new best estimate of the odds of success based on the NEW information you now have.
So bottom line MCS isn't just useful in illustrations but in helping decide the right AA in the first place, the right goal, the right withdrawal and so on. And then you run them whenever the future looks significantly different than your expectation. You use the new information to adapt your plan
Now you can certainly have a different opinion
Larry
james22
Posts: 1992
Joined: Tue Aug 21, 2007 2:22 pm

Re: How to think about expected returns--continuing discussi

Post by james22 »

I appreciate your reply, Larry.

How about those who do not attempt to determine the right mean and/or time the correction but simply wait to recognize a correction after the fact?

Buffett has said he is waiting for a “fat pitch,” his phrase for an opportunity to buy a stock at a favorable price. He said this month in Fortune’s Most Powerful Women Summit that most of the time, stocks tend to trade in what he calls “the zone of reasonableness” and that he’ll get a chance every five to ten years to make a definitive statement.

“There is a big zone of reasonableness and anybody who thinks they can pinpoint it is crazy,” Buffett said during the conference.


http://www.bloomberg.com/news/2014-10-2 ... -rout.html
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

James
First Buffett advises YOU not to do that. Why not take his advice. Peter Lynch never tried to time market. And no evidence of people having that skill.
Bottom line is that IMO that is a loser's game. So why play? That is basically what Buffett said in the quote you cite. And Buffett hasn't sold either. Though he likes to build up some cash reserves for a possible crash, he doesn't sell the rest of his holdings.
Even in 1998 when I sold all my non value stocks, I still used proceeds to buy value stocks and eventually moved to the LP of small value. That was based on what I viewed as extreme valuations in growth but value wasn't anywhere near extreme valuations.
Larry
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: How to think about expected returns--continuing discussi

Post by Rodc »

Larry,

Yes, typo 2000.

The argument of simply saying well we have world class mathematicians is about as weak an argument as one can make.

I have a PhD in math. I work with world class mathematicians, computer scientists, etc and have for decades. The Monte Carlo simulation tools we design, implement and use, as are the simulations used in many fields, are far more sophisticated than used in retail finance, and the issues they are used to address are of far greater concern. I don't use the argument "trust me I'm have a PhD", preferring to focus on actual content. Your arguments would carry more weight if you took the same approach.

At this point I'll quit as we are just running in circles. Readers will have to make their own judgement on the issues presented.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

Rod, I was only pointing out that people with very high levels of skills in the field come to a different conclusion than you do. Doesn't make you wrong and them right, but that there is another viewpoint than yours with people who have the skills to make an informed decision. Nothing more was implied.
Larry
Beliavsky
Posts: 1233
Joined: Sun Jun 29, 2014 10:21 am

Re: How to think about expected returns--continuing discussi

Post by Beliavsky »

larryswedroe wrote:Peter Lynch never tried to time market.
I don't think this is true. Did you read his book "Beating the Street"? I remember his writing that in the early 1980s, he had a hefty stake in Treasury bonds in Fidelity Magellan, because yields were so high. Using the Google books preview feature, I see this on p112:
By October, my bond position was whittled down to 5% of Magellan's assets. The great bull market had begun in earnest. Interest rates had started to come down, and the economy showed signs of revival.
So earlier (in 1982 or 1983?), Magellan had more than 5% on bonds, and I think it was a lot more.

Warren Buffett closed his stock investing partnership in the late 1960s because he thought prices were too high. That was market timing.
Beliavsky
Posts: 1233
Joined: Sun Jun 29, 2014 10:21 am

Re: How to think about expected returns--continuing discussi

Post by Beliavsky »

Rodc wrote:This is very simple. If you start and hold a low or high expected return in your MC you introduce a distinct bias relative to a real markets where the valuations vary considerably over time. Consider starting a MC back in 2001 at the top of the market boom. The expected return would be something extremely low. Now consider putting in $10,000 every year for 30 years and each year using that starting 2001 expected mean. How can that make any sense?
Over time, simulations will become more sophisticated. The simplest way to model stock market returns is to assume a constant drift and volatility, with annual returns having mean and volatility of say 8% and 20%. But volatility certainly varies over time, which is why VIX fluctuates so much. It is likely that expected returns also fluctuate. So you can simulate from a "stochastic volatility" model with time-varying expected returns. I know some Bogleheads will be wary of the value of more complex models, but computer models of many things in everyday life, for example cars, have become more complex over time and have resulted in better quality products.
grayfox
Posts: 5569
Joined: Sat Sep 15, 2007 4:30 am

Re: How to think about expected returns--continuing discussi

Post by grayfox »

I don't really care about expected returns. At least not directly. What I want to know is: How much money I will have in 10 years?

Suppose I want to put some money away to purchase a SPIA in 10 years. It could be a substantial amount of money, say $100K or $500K. To keep from having to many zeros, let's say I put $10,000 into some investment how much will I end up with. I look up 10-year TIPS and it is yielding 0.66%. This COMPOUND INTEREST CALCULATOR says it will compound to $10,679.96
Not much growth. But this is after inflation, so it outpaces inflation.

If an alternative is the S&P 500, how much will a $10K S&P 500 have in 10 years?
Back on the previous page, the estimate is 4.4% real return. The calculator says I end up with $15,381.72.
Again after inflation.

That looks like a slam dunk for the S&P 500, right? 4.4% vs. 0.66%
$15,381 vs. $10,680. 8-)

Not so fast. :annoyed
Further down the page there is the fine print. It's 4.4% +/- 8%
So it's not really 4.4% after all. It -3.6% to +12.4%.
$10,000 investment ends up between $6,931 and $32,166

That's a horse of a different color. Now it seems that I could end up with substantially less than the 10-year TIPS.

Here is my graphic illustration

Code: Select all

       $6,931 +++++++++++++++*+++stock+++++++++++++++++++++++++ $32,186
                     * $10,680 (TIPS)
|__________________|___________________|__________________|__________________|__________________|
0                 $10K                $20K               $30K               $40K               $50K 
Sure, there's a lot of potential upside with stocks. :greedy
But you can also end up with only 70% of what you started with. :oops:

So which would you go with for 10 years, the TIPS or the SP 500?
Last edited by grayfox on Wed Oct 22, 2014 12:51 pm, edited 2 times in total.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

Beliavsky
From my collection, four quotes from Peter Lynch that are applicable

To the rash and impetuous stockpicker who chases hot tips and rushes in and out of his of equities, an ‘investment’ in stocks is no more reliable than throwing away paychecks on the horse with the prettiest mane, or the jockey with the purple silks…[But] when you lose [at the racetrack, at least] you’ll be able to say you had a great time doing it.—Peter Lynch

Your ultimate success or failure will depend on your ability to ignore the worries of the world long enough to allow your investments to succeed. It isn’t the head, but the stomach that determines [your] fate. Peter Lynch, Beating the Street, Simon Publishing (1993).

[Investors] think of the so-called professionals as having all the advantages. That is total crap. They'd be better off in an index fund. —Peter Lynch, “Is There Life After Babe Ruth,” Barron’s, April 2, 1990

I can’t recall ever once having seen the name of a market timer on Forbes’s annual list of the richest people in the world. If it were truly possible to predict corrections, you’d think somebody would have made billions by doing it.— Peter Lynch


Larry
yukon50
Posts: 334
Joined: Thu Jan 13, 2011 7:23 pm

Re: How to think about expected returns--continuing discussi

Post by yukon50 »

grayfox wrote:I don't really care about expected returns. At least not directly. What I want to know is: How much money I will have in 10 years?

Suppose I want to put some money away to purchase a SPIA in 10 years. It could be a substantial amount of money, say $100K or $500K. To keep from having to many zeros, let's say I put $10,000 into some investment how much will I end up with. I look up 10-year TIPS and it is yielding 0.66%. This COMPOUND INTEREST CALCULATOR says it will compound to $10,679.96
Not much growth. But this is after inflation, so it outpaces inflation.

If an alternative is the S&P 500, how much will a $10K S&P 500 have in 10 years?
Back on the previous page, the estimate is 4.4% real return. The calculator says I end up with $15,381.72.
Again after inflation.

That looks like a slam dunk for the S&P 500, right? 4.4% vs. 0.66%
$15,381 vs. $10,680. 8-)

Not so fast. :annoyed
Further down the page there is the fine print. It's 4.4% +/- 8%
So it's not really 4.4% after all. It -3.6% to +12.4%.
$10,000 investment ends up between $6,931 and $32,166

That's a horse of a different color. Now it seems that I could end up with substantially less than the 10-year TIPS.

Here is my graphic illustration

Code: Select all

       $6,931 +++++++++++++++*+++stock+++++++++++++++++++++++++ $32,186
                     * $10,680 (TIPS)
|__________________|___________________|__________________|__________________|__________________|
0                 $10K                $20K               $30K               $40K               $50K 
Sure, there's a lot of potential upside with stocks. :greedy
But you can also end up with only 70% of what you started with. :oops:

So which would you go with for 10 years, the TIPS or the SP 500?

For a young investor, it seems like there is a lot more risk with bonds in your example. With a bond investment you could miss out on 21k in gains, with a stock investment you could sustain 4k in losses...? hmm
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: How to think about expected returns--continuing discussi

Post by Rodc »

larryswedroe wrote:Rod, I was only pointing out that people with very high levels of skills in the field come to a different conclusion than you do. Doesn't make you wrong and them right, but that there is another viewpoint than yours with people who have the skills to make an informed decision. Nothing more was implied.
Larry
Fair enough. My apologies.
Last edited by Rodc on Wed Oct 22, 2014 2:04 pm, edited 1 time in total.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
grayfox
Posts: 5569
Joined: Sat Sep 15, 2007 4:30 am

Re: How to think about expected returns--continuing discussi

Post by grayfox »

yukon50 wrote:
grayfox wrote:

Code: Select all

       $6,931 +++++++++++++++*+++stock+++++++++++++++++++++++++ $32,186
                     * $10,680 (TIPS)
|__________________|___________________|__________________|__________________|__________________|
0                 $10K                $20K               $30K               $40K               $50K 

For a young investor, it seems like there is a lot more risk with bonds in your example. With a bond investment you could miss out on 21k in gains, with a stock investment you could sustain 4k in losses...? hmm
I would think that a younger investor might have a longer time horizon. Depending on current age and planned retirement age maybe 20 or 30 years.

So he would be interested in a different chart that reflects the longer timeline. All the numbers and ranges of possible outcomes would be different.
Cruick
Posts: 18
Joined: Fri Jan 10, 2014 7:14 am

Re: How to think about expected returns--continuing discussi

Post by Cruick »

Larry, James:

I too share the same questions as James. Klarman, Buffet etc plus many of the "golden rules" seem to suggest not putting money in at tops and holding large %s of cash for better times. But as you say Larry, what are tops? Who can identify tops? Who can identify high valuations? What is normal? There are corrections though and the market will go down again so the idea of waiting for better opportunities would seem to have some merit.

I've been a picture of poor investor behavior (psychology) in that I keep trying to time the market ... waiting to get in, but always waiting for a better opportunity figuring the fat pitch will far outperform riding capital loss through. I still like to believe that is the case, but I'm never sure if it will work out. The more thoughts the better on this Larry esp as it pertains to constructing your LP. Having too much cash sitting around, I went into the Euro area markets and emerging markets and bought those while leaving the US markets unbought figuring I was capturing better opportunities.

I'm sure you've hit on these ideas ad nauseum.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

Cruick/James
So let's assume you like Buffett's suggestion to not sell, but wait for buying opportunities, building up cash
Here's the problem. IMO you are unlikely to be able to pull the trigger as Buffett is able. Reason is while Buffett can keep his head while everyone around him is losing their's, I doubt that most people (meaning you) can do that. More likely is that while Buffett sees a light at end of tunnel, you'll think the light is the truck coming the other way. Almost impossible for most people to buy during severe bear markets.
Even though we do extensive "training" with our clients, persistently educating them on this issue and making sure they will rebalance during bear markets, showing them history,etc., during 2008 about 1/3 of clients failed to do so, though at least they didn't panic and sell. So without someone holding your hand I think it's unlikely that most would be able to buy even if they had the cash to do so. That is what makes Buffett special. There's nothing special about his stock picking skill as the latest research has shown. It's really his ability to avoid panicked selling and being able to buy during panics that separates him from the crowd. So bottom line: If you see Buffett when you look in the mirror, great. Go ahead and try it. If not, don't. Better IMO simply to adopt the Larry Portfolio and then you won't worry so much about bear markets as losses will be minimized.
Hope that is helpful
Larry
james22
Posts: 1992
Joined: Tue Aug 21, 2007 2:22 pm

Re: How to think about expected returns--continuing discussi

Post by james22 »

larryswedroe wrote:First Buffett advises YOU not to do that.
For "non-professionals" he most recently recommended a low-cost S&P 500 index fund, yes.

He's many times previously recommended investors wait for a fat pitch.
larryswedroe wrote:So bottom line: If you see Buffett when you look in the mirror, great. Go ahead and try it.
What about those who take to the idea of buying dollar bills for 40 cents?

One sidelight here: it is extraordinary to me that the idea of buying dollar bills for 40 cents takes immediately to people or it doesn't take at all. It's like an inoculation. If it doesn't grab a person right away, I find that you can talk to him for years and show him records, and it doesn't make any difference. They just don't seem able to grasp the concept, simple as it is. A fellow like Rick Guerin, who had no formal education in business, understands immediately the value approach to investing and he's applying it five minutes later. I've never seen anyone who became a gradual convert over a ten-year period to this approach. It doesn't seem to be a matter of IQ or academic training. It's instant recognition, or it is nothing.

http://www.tilsonfunds.com/superinvestors.html

And because of that they've the ability "to avoid panicked selling and being able to buy during panics"? Knowing: If you buy a dollar bill for 60 cents, it's riskier than if you buy a dollar bill for 40 cents, but the expectation of reward is greater in the latter case. The greater the potential for reward in the value portfolio, the less risk there is.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

James
That idea of buying something at 40 cents that's worth a dollar---well if that was the case and you could do that it would be easy to beat the market on risk-adjusted basis. Unfortunately there is little evidence of such superhuman abilities. If it's selling at 40cents the market believes it's worth 40 cents. It's just that you think market is stupid/wrong.
Having said that what Buffett is saying really IMO is that he buys value companies that tend to also be quality (expanded definition of the profitability premium). Well it turns out that Buffett didn't really exhibit such superhuman stock picking skills. What he exhibited was the ability to identify the TYPE of stock to buy. Two recent papers found that once you adjusted for the factors including quality BRK's public security holdings show no statistically significant alpha. Now that doesn't take anything away from Buffett--he discovered the secret sauce 50 years before the academics. But now we all know. FWIW, BRK has underperformed both DFA SV and DFA LV for past 15 calendar years by fair amount.

And again if you look in mirror and see Buffett, go ahead and try to do what he does. But there are many pros who have tried to replicate his results without success, even though his formula is well known. Discipline matters. And few have it, especially the ability to buy when the world looks darkest

Larry
IlliniDave
Posts: 2388
Joined: Fri May 17, 2013 7:09 am

Re: How to think about expected returns--continuing discussi

Post by IlliniDave »

Personally, I'm a big fan of MC simulations, and operate a home-spun one for things financial. I've thought about trying to capture valuations as an input somehow but abandoned the idea because it made my head hurt. What I do instead is somewhat arbitrarily dial down the mean result below the typical historical values used (often I listen to whatever Mr. Bogle's most recent stock outlook is and decrease it by 0.5%-1.0%). So my current self-forecasts are a little on the bleak side. Fortunately my lifestyle needs are very modest!

The great thing about Monte Carlo simulations is that you can easily spin off thousands of trials in a few minutes. I always focus my attention on the worst results, and I think it helps sober my opinion of how "secure" I will be. I'm going to have to be careful and willing to let my consumption spending vary were the two main takeaways.
Don't do something. Just stand there!
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: How to think about expected returns--continuing discussi

Post by nedsaid »

I am not a quant. In college, I got a D and C in Calculus and I worked pretty darned hard for that D. But I do respect numbers, ratios, statistical analysis. I am an accountant so I work with numbers. It is just that I understand the limitations.

So while I have opined that successful investing is mostly behavioral and that all the math has its limits, it doesn't mean that I dismiss what the quants have to say.

I have employed financial planners to run Monte Carlo simulations as part of putting together my financial plan. And of course there are return assumptions built in. These are useful as they give me an idea of the possible variation of outcomes as well as my chances for success. Beyond this, I don't give this to much thought. I don't spend hours trying to calculate the expected returns of my portfolio. I make my investments based on the best information that I have at the time and hope for the best.

This is a way of saying that the markets will do what the markets will do. My needs, desires, future plans, hopes and dreams have no bearing on how the markets will perform. I do the very best I can with the knowledge and resources I have. I agree with Larry that valuations matter and matter a lot. Valuations and future expected returns impact how I build my portfolio. I realize this is a flawed and inexact science but I would rather have imperfect tools than none at all.
A fool and his money are good for business.
Cruick
Posts: 18
Joined: Fri Jan 10, 2014 7:14 am

Re: How to think about expected returns--continuing discussi

Post by Cruick »

Larry,

Thanks for your response. I cant remember seeing too much in your books, but maybe remember a brief mention in a forum or blog, but what are your thoughts do about allocation within the LP framework based on valuation or maybe we should say expected returns. In other words, the idea of changing the allocation from or towards emerging, international, domestic based on expected returns/opportunity?

In this case we simple we add a layer of activeness on top of just rebalancing. Now we become more macro tactical in a way and make subtle shifts to capture macro trends.

Thanks.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

Cruick
IMO basically set it and forget it re allocations between US, developed and EM. Now you should take into account current valuations when determining expected returns, with EM now the highest. But once you decide on the allocations, and to me world market cap is good starting place and should have strong reason to deviate, then just rebalance. I would suggest something like 50% US, 37.5% Developed and 12.5% EM. But I would say that for all portfolios, not just the lP
Larry
grayfox
Posts: 5569
Joined: Sat Sep 15, 2007 4:30 am

Re: How to think about expected returns--continuing discussi

Post by grayfox »

My main complaint about most expected return forecasts is that the most important thing is buried in a footnote to Table Z.11 in the Appendix on page 167. Which is where you find that the range is +/- 8%. The debate centers on whether the mean return is 4.0% or 4.4% or 4.5%, but the interval is 20x or 80x larger than the minute adjustments being made to the mean forecast. :confused

Actually, most forecasts I see never mention the error or confidence interval. On their 7-year forecast, GMO used to put +/- 6.5% at the bottom of the chart. Now they leave that off their chart. Why is that?

It's misleading to not include some kind of error or range with a forecast. Unless you are Pythia, there should always be some kind of interval or range. And I don't think it should be left as an exercise for the reader to do the calculations.
User avatar
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: How to think about expected returns--continuing discussi

Post by SimpleGift »

grayfox wrote:It's misleading to not include some kind of error or range with a forecast. Unless you are Pythia, there should always be some kind of interval or range. And I don't think it should be left as an exercise for the reader to do the calculations.
I'd just point out that Pythia has a much easier time forecasting the fundamental return of stocks, that is dividends (in yellow below) plus earnings growth (in green), which are somewhat stable over long periods. What confounds her and most other prognosticators is the speculative return, that is the change in market multiples (in pink) — though the speculative return does tend to revert back to the fundamental return over time.

So it seems the range in errors could be broken down between the fundamental return (small error range, higher predictability) and speculative return (much larger error range, low predictability, helped by mean reversion).

Image
Source: Business Insider

(This just happens to be what the Gordon Equation does, when used for 30-year and longer stock holding periods.) :wink:
Last edited by SimpleGift on Fri Oct 24, 2014 11:54 am, edited 1 time in total.
User avatar
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: How to think about expected returns--continuing discussi

Post by SimpleGift »

Just to show an example of the difference in error ranges between the factors of stock returns, we can look at the recent return forecasts published by Research Affiliates. They use an earnings growth model to estimate expected real stock returns for U.S. large cap stocks over next 10 years (same method as Mr. Bogle):

Image
  • Current Yield………….……1.8% — Known
    Earnings Growth Rate…….1.3% — Small Error Range
    Valuation Change……..….(2.4%) — Wild-ass Guess
    REAL TOTAL RETURN….....0.7%
The point is that the large majority of the error in the forecast is contained in the estimate of valuation changes. This is where the error ranges need to be added — or one can sidestep the issue entirely by only forecasting long-term fundamental return (30-years or longer) and assume the valuation changes (plus and minus) will cancel out.
User avatar
Abe
Posts: 2572
Joined: Fri Sep 18, 2009 5:24 pm
Location: Earth in the Milky Way Galaxy

Re: How to think about expected returns--continuing discussi

Post by Abe »

This is how I see it. The correlation coefficient in a linear association has a negative association, so there are three axioms of probability. This constitutes a base rate fallacy in my opinion, and I am more inclined to look at the binomial distribution when projecting expected returns. So my conclusion is: nobody knows nothing. Hope this helps. :wink:
Slow and steady wins the race.
Hank Moody
Posts: 279
Joined: Fri Oct 17, 2014 9:41 pm

Re: How to think about expected returns--continuing discussi

Post by Hank Moody »

Larry,

I asked this earlier and I don't know if you ever responded, but I'm curious you are focused on ER within the work you do at BAM? I realize this has been a very long conversation and I'm fairly new here. It's difficult for me to understand if the discussion is purely academic or if you apply it somehow to the advice you give to the firm's clients. Can you give me some context, please?

HM
larryswedroe wrote:Cruick
IMO basically set it and forget it re allocations between US, developed and EM. Now you should take into account current valuations when determining expected returns, with EM now the highest. But once you decide on the allocations, and to me world market cap is good starting place and should have strong reason to deviate, then just rebalance. I would suggest something like 50% US, 37.5% Developed and 12.5% EM. But I would say that for all portfolios, not just the lP
Larry
-HM
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

Hank
Before we invest any assets we first do a discovery meeting, learning all we can about what's important about money to the investor. Learn all their important relationships, financial and life goals, ability, willingness and need to take risk and so on. Then we produce a "proposed" allocation for discussion, showing the results of the MCS with our inputs. That allows analysis of odds of not running out of money as well as meeting secondary goals such as leaving an estate of a certain size. Then we also look at how changing spending rates up or down impacts odds of success, helping the investor decide on what is best for them including AA and spending rates. The inputs are based on our estimates of expected returns and include correlations and SDs. We also have discussions about how much tilt to small and value and international and EM to have in the portfolio. Obviously the more you tilt the higher the expected returns for the same equity allocation. But higher tilt also allows you to lower equity allocation while having same expected returns. The MCS allows analysis of the tails, especially the left tail outcomes, and then have discussions about "plan B"--actions required to be taken if the left tail shows up. And then new MCS are run whenever returns turn out to be significantly different than expected, or whenever any of the other assumptions in plans change.
We update annually the inputs. Sometimes more frequently if have major moves like in 2008. Then might do it twice.
I hope that answers your question

Larry
Hank Moody
Posts: 279
Joined: Fri Oct 17, 2014 9:41 pm

Re: How to think about expected returns--continuing discussi

Post by Hank Moody »

Got it with the financial planning piece, but would I be correct that Er has nothing to do with tactical decisions? In other words, are you letting your views about Er affect the stock/bond ratio or US/foreign stock allocation?
larryswedroe wrote:Hank
Before we invest any assets we first do a discovery meeting, learning all we can about what's important about money to the investor. Learn all their important relationships, financial and life goals, ability, willingness and need to take risk and so on. Then we produce a "proposed" allocation for discussion, showing the results of the MCS with our inputs. That allows analysis of odds of not running out of money as well as meeting secondary goals such as leaving an estate of a certain size. Then we also look at how changing spending rates up or down impacts odds of success, helping the investor decide on what is best for them including AA and spending rates. The inputs are based on our estimates of expected returns and include correlations and SDs. We also have discussions about how much tilt to small and value and international and EM to have in the portfolio. Obviously the more you tilt the higher the expected returns for the same equity allocation. But higher tilt also allows you to lower equity allocation while having same expected returns. The MCS allows analysis of the tails, especially the left tail outcomes, and then have discussions about "plan B"--actions required to be taken if the left tail shows up. And then new MCS are run whenever returns turn out to be significantly different than expected, or whenever any of the other assumptions in plans change.
We update annually the inputs. Sometimes more frequently if have major moves like in 2008. Then might do it twice.
I hope that answers your question

Larry
Last edited by Hank Moody on Fri Oct 24, 2014 9:42 pm, edited 2 times in total.
-HM
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

Hank
Not exactly
The ER does impact the need to take risk. The lower the ER the more beta or more tilt you need (or some more of both). Now that doesn't mean one should choose that alternative. Other options are to work longer, cut goals, or cut current spending to save more. Also ER can impact allocations themselves. At some point you might decide that beta premium is too low, as I did in 1998 and sold all but value (where premium was much larger than ever). Now I did that personally.
Hope that is helpful
Larry
Hank Moody
Posts: 279
Joined: Fri Oct 17, 2014 9:41 pm

Re: How to think about expected returns--continuing discussi

Post by Hank Moody »

I'm trying to understand if your capital market expectations figure into planning decisions, which obviously affects the AA decision, or if there's some subtle market timing/tactical decision making at work here.


larryswedroe wrote:Hank
Not exactly
The ER does impact the need to take risk. The lower the ER the more beta or more tilt you need (or some more of both). Now that doesn't mean one should choose that alternative. Other options are to work longer, cut goals, or cut current spending to save more. Also ER can impact allocations themselves. At some point you might decide that beta premium is too low, as I did in 1998 and sold all but value (where premium was much larger than ever). Now I did that personally.
Hope that is helpful
Larry
-HM
Hank Moody
Posts: 279
Joined: Fri Oct 17, 2014 9:41 pm

Re: How to think about expected returns--continuing discussi

Post by Hank Moody »

Larry,

Although my geographical allocations are pretty much set-and-forget, I've often wondered if this is a little shortsighted. Are we confident that international capital flows do a fairly good job of leveling out the discount rate for a country or region of the world? Or do long-range macro economic forecasts have value for passive investors?

Suppose we have developed countries A and B. Country A has fiscal and political issues that are expected to cause problems for the next 5 to 10 years, but it also has a young population and a newly established and well-regarded educational system. The long range prospects 15 to 30 years out appear positive. County B doesn't have the same short-run problems as A, but it's population is aging and its educational system is weaker that A's, and may cause slower growth beyond the next decade.

Client #1 can afford to take international equity risk, but is nearing retirement and doesn't care what happens in 15 - 30 years. Client #2 is 30 years old. He doesn't care what happens in the next 10 years. His focus starts at 15 years.

If investors have different duration preferences in stocks, can long-range forecasting play a role in our geographic allocation?
larryswedroe wrote:Cruick
IMO basically set it and forget it re allocations between US, developed and EM. Now you should take into account current valuations when determining expected returns, with EM now the highest. But once you decide on the allocations, and to me world market cap is good starting place and should have strong reason to deviate, then just rebalance. I would suggest something like 50% US, 37.5% Developed and 12.5% EM. But I would say that for all portfolios, not just the lP
Larry
-HM
grayfox
Posts: 5569
Joined: Sat Sep 15, 2007 4:30 am

Re: How to think about expected returns--continuing discussi

Post by grayfox »

Simplegift wrote:Just to show an example of the difference in error ranges between the factors of stock returns, we can look at the recent return forecasts published by Research Affiliates. They use an earnings growth model to estimate expected real stock returns for U.S. large cap stocks over next 10 years (same method as Mr. Bogle):

Image
  • Current Yield………….……1.8% — Known
    Earnings Growth Rate…….1.3% — Small Error Range
    Valuation Change……..….(2.4%) — Wild-ass Guess
    REAL TOTAL RETURN….....0.7%
The point is that the large majority of the error in the forecast is contained in the estimate of valuation changes. This is where the error ranges need to be added — or one can sidestep the issue entirely by only forecasting long-term fundamental return (30-years or longer) and assume the valuation changes (plus and minus) will cancel out.
That should be a whole separate discussion by itself: The Sources of Estimation Error in Expected Return Forecasts

Interesting report from Rob Arnott and Research Affiliates. He also left calculation of estimation error as an exercise for the reader.

In Figure 4, they show Forecasted Real Returns for next ten years:

Current.Dividend.Yield + Real.Earnings.Growth + delta.Valuation + Currency Adjustment = 2.0 + 1.4 + (-2.4) + 0 = 1.0

That would compare to the 4.4% forecast on the previous page in this thread.
That's a big difference in forecasts from two different forecasters.
Even the mean forecast has a lot of variation, depending on who is you ask.

:?: Which leads to my second question about expected return forecasts: Are these forecasts fact or simply someone's opinion?
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

Hank
IMO the default --in the absence of evidence to the contrary, or a clear sign of bubble (like e/p less than TIPS yield) should assume markets are highly efficient at allocating capital
Having said that if you tilt to regions with lower valuations you clearly have higher expected returns. But I would assume that's compensation for risk, not free lunch. Thus you need to have strong stomach and deep conviction to decide to bet on anything else. No different that those that make the mistake of home country bias and bet against the market's wisdom
Larry
richard
Posts: 7961
Joined: Tue Feb 20, 2007 2:38 pm
Contact:

Re: How to think about expected returns--continuing discussi

Post by richard »

grayfox wrote:<> :?: Which leads to my second question about expected return forecasts: Are these forecasts fact or simply someone's opinion?
What do you mean?

How would (or wouldn't) a prediction of the future be a fact?

"Simply someone's opinion" is a rather dismissive characterization. Forecasting methodologies and those who forecast have track records and some are better than others. Whether past forecasting success predicts future forecasting success is not known.

In looking at any forecast, you might want to consider the logic of the forecast, its track record, possible data mining and whether conditions that led to the track record have changed.
jaab
Posts: 112
Joined: Thu Aug 07, 2014 12:52 pm

Re: How to think about expected returns--continuing discussi

Post by jaab »

Let's not forget that we are talking about long term to very long term forecasts here. 7/10, 20, 30+ years. In contrast to those over a few months or a few years only, which are basically useless. Even LT forecasts have a high uncertanity, but at least they are one of the better tools/ideas around, when linked to fundamentals like yield, growth, valuation, ...
User avatar
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: How to think about expected returns--continuing discussi

Post by SimpleGift »

grayfox wrote:Which leads to my second question about expected return forecasts: Are these forecasts fact or simply someone's opinion?
No doubt they are just opinions — but they are instructive opinions, in my view, that have planning utility for the average investor.

For example, below are the current forecasts of real returns for U.S. large cap stocks by prominent commentators, ranging from the perma-bears to the perma-bulls. Yes, the individual return forecasts are all over the map — but the average forecast return is less than half of the recent historical average return — and that's instructive for planning purposes.
  • Jeremy Grantham…….…7 years…........-1.7%
    Research Affiliates........10 years...….…..0.7%
    William Bernstein………10 years……..….2.0%
    PIMCO…………..…..……10 years……..…3.5%
    Rick Ferri…………….…..30 years…..……5.0%
    Schwab Advisory….......20 years….....….6.0%
    AVERAGE FORECAST RETURN….......……2.6%

    AVERAGE RETURN, 1980-2013…......…..6.2%
I'd just add that the utility of return forecasts is especially high today, when we know we are living through an atypical investment period. Asset yields all across the board are at or near historical lows and asset valuations are very high. So historical averages are not going to be of any help. Better to take an ahistorical approach and to make reasonable planning estimates based on today's asset return forecasts — despite their known limitations.
Hank Moody
Posts: 279
Joined: Fri Oct 17, 2014 9:41 pm

Re: How to think about expected returns--continuing discussi

Post by Hank Moody »

In 1982, how many forecasters were correct about the next 10 years?
Simplegift wrote:
grayfox wrote:Which leads to my second question about expected return forecasts: Are these forecasts fact or simply someone's opinion?
No doubt they are just opinions — but they are instructive opinions, in my view, that have planning utility for the average investor.

For example, below are the current forecasts of real returns for U.S. large cap stocks by prominent commentators, ranging from the perma-bears to the perma-bulls. Yes, the individual return forecasts are all over the map — but the average forecast return is less than half of the recent historical average return — and that's instructive for planning purposes.
  • Jeremy Grantham…….…7 years…........-1.7%
    Research Affiliates........10 years...….…..0.7%
    William Bernstein………10 years……..….2.0%
    PIMCO…………..…..……10 years……..…3.5%
    Rick Ferri…………….…..30 years…..……5.0%
    Schwab Advisory….......20 years….....….6.0%
    AVERAGE FORECAST RETURN….......……2.6%

    AVERAGE RETURN, 1980-2013…......…..6.2%
I'd just add that the utility of return forecasts is especially high today, when we know we are living through an atypical investment period. Asset yields all across the board are at or near historical lows and asset valuations are very high. So historical averages are not going to be of any help. Better to take an ahistorical approach and to make reasonable planning estimates based on today's asset return forecasts — despite their known limitations.
-HM
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

Hank
Anyone using the valuation methods standard in finance would have predicted very high returns, well above historical averages
Larry
SnowSkier
Posts: 268
Joined: Fri Mar 16, 2012 12:21 pm
Location: Austin

Re: How to think about expected returns--continuing discussi

Post by SnowSkier »

Simplegift wrote:No doubt they are just opinions — but they are instructive opinions, in my view, that have planning utility for the average investor.

For example, below are the current forecasts of real returns for U.S. large cap stocks by prominent commentators, ranging from the perma-bears to the perma-bulls. Yes, the individual return forecasts are all over the map — but the average forecast return is less than half of the recent historical average return — and that's instructive for planning purposes.
  • Jeremy Grantham…….…7 years…........-1.7%
    Research Affiliates........10 years...….…..0.7%
    William Bernstein………10 years……..….2.0%
    PIMCO…………..…..……10 years……..…3.5%
    Rick Ferri…………….…..30 years…..……5.0%
    Schwab Advisory….......20 years….....….6.0%
    AVERAGE FORECAST RETURN….......……2.6%

    AVERAGE RETURN, 1980-2013…......…..6.2%
I'd just add that the utility of return forecasts is especially high today, when we know we are living through an atypical investment period. Asset yields all across the board are at or near historical lows and asset valuations are very high. So historical averages are not going to be of any help. Better to take an ahistorical approach and to make reasonable planning estimates based on today's asset return forecasts — despite their known limitations.
+1
User avatar
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: How to think about expected returns--continuing discussi

Post by SimpleGift »

Hank Moody wrote:In 1982, how many forecasters were correct about the next 10 years?
Actually, 1982 is a good example, because it was at the opposite extreme of today's asset yields and valuations. In 1982, 10-year Treasury bonds were yielding 14% and the Shiller PE10 was at 7. Today, the 10-year Treasury is yielding 2.3% and the Shiller PE10 is at 26.

The point is that, at these extremes of asset yields and valuations (like 1952 or 1982 or 2012, see chart below), historical average returns are not going to be helpful to investors for planning purposes. That's exactly when reasonable forecasts of future asset returns are the most useful!

Image
Source: Money Architects
Last edited by SimpleGift on Sat Oct 25, 2014 11:48 am, edited 1 time in total.
grayfox
Posts: 5569
Joined: Sat Sep 15, 2007 4:30 am

Re: How to think about expected returns--continuing discussi

Post by grayfox »

richard wrote:
grayfox wrote:<> :?: Which leads to my second question about expected return forecasts: Are these forecasts fact or simply someone's opinion?
What do you mean?

How would (or wouldn't) a prediction of the future be a fact?

"Simply someone's opinion" is a rather dismissive characterization. Forecasting methodologies and those who forecast have track records and some are better than others. Whether past forecasting success predicts future forecasting success is not known.

In looking at any forecast, you might want to consider the logic of the forecast, its track record, possible data mining and whether conditions that led to the track record have changed.
If I were to ask 10 physicists what is the speed of light in a vacuum. I'm sure they would all tell me 3.0 x 10^8 meters per second orthereabouts. Maybe some would give it in different units or more decimal places, but they would all give essential the same answer because it's a scientific fact that they all agree on. I looked it up on wolframalpha.com and it says 2.988 x 10^8 m/s.

The speed of light c is useful to know because things like GPS only work because we know the value of c, and how the speed changes when it goes through the ionosphere or troposphere, etc.

If I were to ask 10 financial experts what will be the return of a 10-year Treasury held 10 years to maturity, they would all say 2.29% p.a., the current yield on the 10-year Treasury. Now that's a forecast, i.e. prediction of the future, but it's also a fact and they all agree on it.

But if I were to ask 10 financial experts what the return of the S&P 500 will be over the next ten years, I would get 10 different answers. Simplegift proved that with his table:

Jeremy Grantham…….…7 years…........-1.7%
Research Affiliates........10 years...….…..0.7%
William Bernstein………10 years……..….2.0%
PIMCO…………..…..……10 years……..…3.5%
Rick Ferri…………….…..30 years…..……5.0%
Schwab Advisory….......20 years….....….6.0%
AVERAGE FORECAST RETURN….......……2.6%

Facts, people must agree on. But everyone is entitled to their opinion and opinions can vary.
Expected return = X can not be a fact, since no one agrees on what X is.
So it must be an opinion. And one guy's opinion is as good as another. There is no right or wrong opinion.

If someone's opinion is that expected return is the historical mean return, and valuations don't matter, that's their opinion. You may have a different opinion. I bet Jeremy Siegel still says that stocks have constant expected return = 6.7% real

I'm not even sure that Expected return exists. For all we know, it may just be a figment of our imaginations.
User avatar
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: How to think about expected returns--continuing discussi

Post by SimpleGift »

grayfox wrote:Facts, people must agree on. But everyone is entitled to their opinion and opinions can vary.
Expected return = X can not be a fact, since no one agrees on what X is.
It's unrealistic to expect scientific precision from financial data — it's just not there and it's never going to be there. But that does not mean that analysis of financial data, as imprecise as it is, is useless or has no utility for forward planning purposes. There are useful principles and tendencies in financial data that can help inform planning decisions by the average investor — despite the known limitations.

For the mathematicians and statisticians on this Forum who are so troubled by the imprecise nature of these asset return forecasts, my suggestion is to just not use them.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: How to think about expected returns--continuing discussi

Post by Rodc »

Simplegift wrote:
grayfox wrote:Facts, people must agree on. But everyone is entitled to their opinion and opinions can vary.
It's unrealistic to expect scientific precision from financial data — it's just not there and it's never going to be there. But that does not mean that analysis of financial data, as imprecise as it is, is useless or has no utility for forward planning purposes. There are useful principles and tendencies in financial data that can help inform planning decisions by the average investor — despite the limitations.

For the mathematicians and statisticians on this Forum who are troubled by the imprecise nature of these asset return forecasts, my suggestion is to just not use them.
And we suggest the same so I guess we are in agreement. :D

Actually, one can use them but they should use them in a way that reflects the huge level of uncertainty. If using them to adjust a portfolio or use them in a Monte Carlo simulation as opposed to just saving more or putting off retirement one would be wise to at least get the math more or less correct.

At the very least a Monte Carlo simulation should not sample from a fixed distribution but rather address both the random nature of returns due to non-zero standard deviation and the random errors in our knowledge of the mean of the distribution. This could be done by running say 10,000 sims with one mean and standard deviation and then repeat 10,000 times sampling means from a distribution. One should really do the same with standard deviation as well since we don't know that with any precision either, but that might be over kill. Or at the very least perform several MC runs with various plausible means.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
User avatar
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: How to think about expected returns--continuing discussi

Post by SimpleGift »

^^^ “The best is the enemy of the good.” ― Voltaire (French philosopher, 1694-1778)
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

grayfox
There is a big difference between forecasts based on the "science" (valuations) vs those like Grantham's based on pure opinions. And IMO you are mixing the two together.

Second, fwiw, 1982 assuming the posted CAPE 10 of 7 is right, the forecast would be for 14% real returns. Well next 10 years real return was then 13.2% for S&P. I would say that happened to be pretty good. Now that is rare that you get that close. But that was the question asked

Third, thinking you can get it right on is absurd because the speculative return part is unknowable. But that doesn't mean that there isn't valuable information in the valuation part. And IMO only fools ignore valuable information.

Fourth we know the historical return is not the expected and it's absolute nonsense to believe that it is, unless current valuations happen to coincide with history.

Larry
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: How to think about expected returns--continuing discussi

Post by Rodc »

Simplegift wrote:^^^ “The best is the enemy of the good.” ― Voltaire (French philosopher, 1694-1778)
The above is not best, it is merely an attempt to not do something really badly. One might as well not do badly when good is easy enough.

Surely if you can run one MC you can run a handful to get a sense of the sensitivity to your forecast errors. To do any less seems to me to be an exercise in willful ignorance.

Note just to try to be super clear, if one truly knew the distribution with a lot of accuracy then one run of a MC is fine because what you care about is not the fact that you do not know the distribution but rather the fact that your draws from that distribution are random. You want to know the effect of that randomness.

Here we do not know the distribution itself with any real degree of accuracy which introduces an entirely different and large source of errors. One ought to care about this source of errors because it is quite large. And it is not hard to get a sense of the size and impact of this error: just run a few MC runs with different distributions, say the distributions for the set of experts up steam in this thread.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
User avatar
oneleaf
Posts: 2562
Joined: Mon Feb 19, 2007 4:48 pm

Re: How to think about expected returns--continuing discussi

Post by oneleaf »

larryswedroe wrote:Hank
IMO the default --in the absence of evidence to the contrary, or a clear sign of bubble (like e/p less than TIPS yield) should assume markets are highly efficient at allocating capital
Having said that if you tilt to regions with lower valuations you clearly have higher expected returns. But I would assume that's compensation for risk, not free lunch. Thus you need to have strong stomach and deep conviction to decide to bet on anything else. No different that those that make the mistake of home country bias and bet against the market's wisdom
Larry
I agree with this way of looking at it, but would you agree that overweighting cheap countries and underweighting expensive while reducing equity exposure, is another way to not overly rely in the equity risk premium? Similar to what you do with small value.

I have read Ilmanen and also read a lot of Cliff Asness' writings and interviews, and interested in incorporating Mom, carry, long-cheap/short-expensive. But as a retail investor without access to AQR and DFA, it is not exactly easy to get any meaningful access to these strategies.

I know tactical asset allocation is a slippery slope, but for the most disciplined rebalancer who does his homework, can a country oriented overweight cheap/underweight expensive, while reevaluating every couple years, be a prudent way to boost returns, while reducing equity exposure and tail risk?
User avatar
siamond
Posts: 6010
Joined: Mon May 28, 2012 5:50 am

Re: How to think about expected returns--continuing discussi

Post by siamond »

grayfox wrote:But if I were to ask 10 financial experts what the return of the S&P 500 will be over the next ten years, I would get 10 different answers. Simplegift proved that with his table:

Jeremy Grantham…….…7 years…........-1.7%
Research Affiliates........10 years...….…..0.7%
William Bernstein………10 years……..….2.0%
PIMCO…………..…..……10 years……..…3.5%
Rick Ferri…………….…..30 years…..……5.0%
Schwab Advisory….......20 years….....….6.0%
AVERAGE FORECAST RETURN….......……2.6%

Facts, people must agree on. But everyone is entitled to their opinion and opinions can vary.
Expected return = X can not be a fact, since no one agrees on what X is.
So it must be an opinion. And one guy's opinion is as good as another. There is no right or wrong opinion.

If someone's opinion is that expected return is the historical mean return, and valuations don't matter, that's their opinion. You may have a different opinion. I bet Jeremy Siegel still says that stocks have constant expected return = 6.7% real

I'm not even sure that Expected return exists. For all we know, it may just be a figment of our imaginations.
I certainly agree that expected returns forecasts vary much more than a physicist describing the speed of light... Nature of the beast... But it would really help if we'd stop mixing apples and oranges. First, there are two types of expected returns, what Jack Bogle calls Fundamental Returns (ignoring speculative/valuation factors), and what Jack calls Speculative Returns (add a speculative correction factor to the fundamental returns). Next, a 10-years forecast is very different in nature to a 30-years forecast, not only for the obvious reason (10 ain't 30), but also because one can reasonably ignore the speculative factor for a 30-yrs forecast, while it seems quite unreasonable to do so for a 10-yrs forecast. And then, it also needs to be anchored on a given date. When the market goes down 5% in a week (as we just saw), well, this changes the speculative math in a non-negligible manner.

This is really the core reason for which we see low returns in SimpleGift's list vs higher ones (3.5% and less vs 5% and more). Because of the mix between apples and oranges. Not because the experts can't be trusted to display a modicum of consistency.

I certainly agree that even without such mix-up, there is still a lot of variation. Grantham's prediction is a tad mind-boggling. I am just saying that before casting too many stones to such forecasts, one should be a bit more specific about what we're speaking of...
User avatar
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: How to think about expected returns--continuing discussi

Post by SimpleGift »

Rodc wrote:
Simplegift wrote:^^^ “The best is the enemy of the good.” ― Voltaire (French philosopher, 1694-1778)
The above is not best, it is merely an attempt to not do something really badly. One might as well not do badly when good is easy enough.

Surely if you can run one MC you can run a handful to get a sense of the sensitivity to your forecast errors. To do any less seems to me to be an exercise in willful ignorance.
Your recommendation to run multiple Monte Carlo simulations with various plausible means ("the best") may be useful to financial advisers or researchers who have access to the expertise and software to make these calculations. However, for the average investor who reads these Forums and is trying to estimate future assets returns for the own financial planning, it's not going to be much help.

The simplified Gordon Equation or the CAPE earnings yield (the "good") are about as sophisticated as most of us mere mortal investors are going to get. Perhaps the mathematical and statistical expertise on this Forum could help to improve and refine these simplified methods, so they would be more useful to the average investor? That would be a constructive approach.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: How to think about expected returns--continuing discussi

Post by larryswedroe »

Oneleaf
First, re your currency question, that is what is well known as the carry trade, and part of the QSPIX strategy. And note it's like value strategy, buy cheap and sell expensive. Similar in commodities, going long backwardation and short contango, and currencies, long the high yield and short the low yield, and similarly with growth and value stocks.
Second AQR funds are I believe available to anyone. But not sure. But others offer these type strategies, although don't think in one fund
Larry
Post Reply