GMO's 7‐Year Asset Class Real Return Forecasts

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Beliavsky
Posts: 761
Joined: Sun Jun 29, 2014 10:21 am

GMO's 7‐Year Asset Class Real Return Forecasts

Post by Beliavsky » Mon Jan 14, 2019 2:03 pm

GMO has released to the public its 7‐Year Asset Class Real Return Forecasts. As of December 31, 2018
they are

Stocks
-2.5% U.S. Large
1.3% U.S. Small
2.0% Intl Large
2.8% Intl Small
4.7% Emerging
8.2% Emerging Value

Bonds
-0.7% U.S. Bonds
-2.6% Intl Hedged
2.7% Emerging debt
-0.1% U.S. Inflation Linked Bonds
0.6% U.S. Cash

As of December 31, 2017 they were

Stocks
-4.7% U.S. Large
-3.0% U.S. Small
-1.7% Intl Large
-1.0% Intl Small
-2.0% Emerging
1.5% Emerging Value

Bonds
-1.0% U.S. Bonds
-2.5% Intl Hedged
0.5% Emerging debt
-0.4% U.S. Inflation Linked Bonds
0.4% U.S. Cash

GMO is bullish on emerging market stocks and debt. Is anyone willing to venture their long-term forecasts of asset class returns?
Last edited by Beliavsky on Mon Jan 14, 2019 4:21 pm, edited 1 time in total.

User avatar
Tycoon
Posts: 1353
Joined: Wed Mar 28, 2012 7:06 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by Tycoon » Mon Jan 14, 2019 2:12 pm

What did the GMO predict on January 31, 2008?
Be still, sad heart! and cease repining; | Behind the clouds is the sun still shining; | Thy fate is the common fate of all, | Into each life some rain must fall, | Some days must be dark and dreary. | Longfellow

megabad
Posts: 966
Joined: Fri Jun 01, 2018 4:00 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by megabad » Mon Jan 14, 2019 2:42 pm

Tycoon wrote:
Mon Jan 14, 2019 2:12 pm
What did the GMO predict on January 31, 2008?
Just for posterity sake. Nominal 2% was GMO predicted for S&P500 I think. Missed it by a factor of 3 or so.

I am willing to venture my long term prediction is that GMO will be wrong again.

letsgobobby
Posts: 11956
Joined: Fri Sep 18, 2009 1:10 am

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by letsgobobby » Mon Jan 14, 2019 3:05 pm

Beliavsky wrote:
Mon Jan 14, 2019 2:03 pm
GMO has released to the public its 7‐Year Asset Class Real Return Forecasts. As of December 31, 2018
They are

Stocks
-2.5% U.S. Large
1.3% U.S. Small
2.0% Intl Large
2.8% Intl Small
4.7% Emerging
8.2% Emerging Value

Bonds
-0.7% U.S. Bonds
-2.6% Intl Hedged
2.7% Emerging debt
-0.1% U.S. Inflation Linked Bonds
0.6% U.S. Cash

As of December 31, 2017 they were

Stocks
-4.7% U.S. Large
-3.0% U.S. Small
-1.7% Intl Large
-1.0% Intl Small
-2.0% Emerging
1.5% Emerging Value

Bonds
-1.0% U.S. Bonds
-2.5% Intl Hedged
0.5% Emerging debt
-0.4% U.S. Inflation Linked Bonds
0.4% U.S. Cash

GMO is bullish on emerging market stocks and debt. Is anyone willing to venture their long-term forecasts of asset class returns?
what is their forecast for inflation?

MotoTrojan
Posts: 3142
Joined: Wed Feb 01, 2017 8:39 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by MotoTrojan » Mon Jan 14, 2019 3:10 pm

letsgobobby wrote:
Mon Jan 14, 2019 3:05 pm

what is their forecast for inflation?
Why does that matter? Real return is the best normalizer across history/conditions, no?

columbia
Posts: 1287
Joined: Tue Aug 27, 2013 5:30 am

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by columbia » Mon Jan 14, 2019 3:21 pm

Even if these are accurate, expected returns will be higher once the markets crash. ;)

reformed.trader
Posts: 51
Joined: Sat Apr 15, 2017 11:14 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by reformed.trader » Mon Jan 14, 2019 3:25 pm

Annualized 7 year returns for Emerging Markets went from -2% to +4.7% in one year? Seems odd.

User avatar
aburntoutcase
Posts: 147
Joined: Tue Mar 13, 2018 12:06 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by aburntoutcase » Mon Jan 14, 2019 3:34 pm

At the end of 2010 this is what GMO forecast. US large cap was going to have 0.4% real returns, developed intl large 2.1% and emerging markets 4.1%. They were quite wrong.

Image

randomguy
Posts: 7075
Joined: Wed Sep 17, 2014 9:00 am

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by randomguy » Mon Jan 14, 2019 3:35 pm

reformed.trader wrote:
Mon Jan 14, 2019 3:25 pm
Annualized 7 year returns for Emerging Markets went from -2% to +4.7% in one year? Seems odd.
Makes sense. Imagine the difference between 1 year of -20% and 6 years of 4% returns and 7 years of 5% returns. Now that we had a bad year, the bull market can start up:)

The issue isn't so much the GMO being wrong. Nobody should predict super accurate estimates. But have they every been wrong on the bullish side? Some one needs to dredge up the 7 year predcations from say 1998-2003. Now maybe we have gotten lucky but at a certain point you have to consider that their model is just wrong.

User avatar
aburntoutcase
Posts: 147
Joined: Tue Mar 13, 2018 12:06 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by aburntoutcase » Mon Jan 14, 2019 3:39 pm

Subsequent realized 7 year real returns were 10% for the S&P 500 while Emerging Markets by my rough math were down -2.5% real.

User avatar
aburntoutcase
Posts: 147
Joined: Tue Mar 13, 2018 12:06 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by aburntoutcase » Mon Jan 14, 2019 3:41 pm

The problem with their forecast returns is that they keep building in a reversion to mean of US large cap company profitability that hasn't happened.

User avatar
alpine_boglehead
Posts: 278
Joined: Fri Feb 17, 2017 9:51 am
Location: Austria

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by alpine_boglehead » Mon Jan 14, 2019 4:13 pm

Beliavsky wrote:
Mon Jan 14, 2019 2:03 pm

Stocks
4.7% Emerging
8.2% Emerging Value

GMO is bullish on emerging market stocks and debt. Is anyone willing to venture their long-term forecasts of asset class returns?
I wouldn't call 4.7% for emerging market bullish - when investing in something as risky as emerging markets, expecting only a 4.7% return doesn't seem optimistic.

Ben Carlson on "a wealth of common sense" has already analyzed their forecast in mid-2018.

The truth is as always, nobody knows nothing. Their guess is as good as anyone's.

Maybe we're in for the gilded age for all of earth's population (and the moon's then, too) with amazing technological advances coming to fruition, giving everyone a good life, and a fortune to investors.
Maybe things will chug along so-so.
Maybe a virus or some political madman will destabilize modern society.
Maybe in a few minutes a cosmic gamma ray blast will sterilize the face of the planet.

All of these scenarios are possible. But we don't know which one (or any other of the truly infinite other possibilities) it will be. And no one can. It's in the category of "thoroughly unknowable". Even not accounting for the really unknowable (random cosmic events), the economy is already an extremely complex beast with billions of moving parts. And it's tied into human society and all other biological systems. Straight down to quantum physics butterfly effects. A genetic mutation could have made <insert random history-altering person> a different personality. Good luck predicting that.

asif408
Posts: 1568
Joined: Sun Mar 02, 2014 8:34 am
Location: Florida

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by asif408 » Mon Jan 14, 2019 4:34 pm

My study of their asset class forecast returns is that, typically, when the spread between asset class returns is tight (say 3-4%), it's probably not much worth following. When the spread between asset classes is large, however, they have been pretty accurate about the best asset class to be in going forward. They don't necessarily get the absolute return numbers correct, but they do appear to get the relative performance between asset classes correct, which is important to me, a long-term investor with many years ahead of me investing. For instance, in Dec. 1999 their expected returns forecast for US REITs and Emerging Markets had about a 10-12% higher expected returns than large cap US stocks: https://goo.gl/images/QDf6gR. They were the overwhelming winners in the following 5-10 years, and even as of today EM and US REITs have outperformed the total US stock market: http://quotes.morningstar.com/chart/etf ... 2%3A955%7D

As another poster showed, their 2010 forecasts were pretty close together. In the current one, there is an almost 11% spread between the best and worst asset class. So unlike other posters, I think GMO probably knows something. I'm prone to believe the current forecast has a better chance of being correct than the 2010 forecast. But we are talking probabilities here and I probably wouldn't put all my money in EM Value stocks. But maybe half.

User avatar
Tycoon
Posts: 1353
Joined: Wed Mar 28, 2012 7:06 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by Tycoon » Mon Jan 14, 2019 4:39 pm

aburntoutcase wrote:
Mon Jan 14, 2019 3:34 pm
At the end of 2010 this is what GMO forecast. US large cap was going to have 0.4% real returns, developed intl large 2.1% and emerging markets 4.1%. They were quite wrong.

Image
I especially get a kick out of the +/- 10.5% estimated range for emerging markets. That will get one's heart racing.
Be still, sad heart! and cease repining; | Behind the clouds is the sun still shining; | Thy fate is the common fate of all, | Into each life some rain must fall, | Some days must be dark and dreary. | Longfellow

longinvest
Posts: 3257
Joined: Sat Aug 11, 2012 8:44 am

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by longinvest » Mon Jan 14, 2019 5:56 pm

aburntoutcase wrote:
Mon Jan 14, 2019 3:34 pm
At the end of 2010 this is what GMO forecast. US large cap was going to have 0.4% real returns, developed intl large 2.1% and emerging markets 4.1%. They were quite wrong.

Image
For the 7 years that followed that prediction or, more exactly, from December 31, 2010 to December 31, 2017, US Large Caps* returned 13.6% annualized.

* Vanguard Large-Cap ETF (VV).

The prediction, for US Large Caps, was 0.4% +/- 6.5% which is a range of -6.1% to 6.9%. Actual returns were 6.7% higher than the top of their prediction range (6.9%).

Despite a large error range, the prediction was completely off by more than twice the predicted error!

What I like about the Bogleheads investment philosophy is that it doesn't rely on future return predictions. It simply tells me to invest my money into assets with an internal return driver (stocks which can generate profits, bonds which pay coupons and their principal back) using index funds covering broad markets. It tells me not to take too much or too little risk, to keep fees low, and to accept market returns by staying the course with my chosen allocation. Investing doesn't have to be more complicated than this.
Bogleheads investment philosophy

User avatar
nisiprius
Advisory Board
Posts: 37623
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by nisiprius » Mon Jan 14, 2019 6:17 pm

These were their forecasts 6/30/11-6/30/18

Image

Here's my chart of how their forecasts compared with the actual results:

Image

The correlation coefficient between the forecasts and the observations 0.05. That is to say, virtually zero. The forecast had virtually no predictive power.

Why do people pay any attention to GMO's forecasts?
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

ge1
Posts: 475
Joined: Sat Apr 28, 2012 8:15 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by ge1 » Mon Jan 14, 2019 8:29 pm

Exactly Nisi.

I have been following them closely for more than 10 years and even though I enjoy reading their quarterly letter, the return forecast is - there is no other way of saying it - totally useless. My favorite was when they suddenly decreased the expected return for EM by several % points from one forecast to the other, as they somehow realized they significantly overestimated the EM returns.

At some point they will be right again and people will tout the accuracy of their forecast. Amazing how quickly people forget.

User avatar
aburntoutcase
Posts: 147
Joined: Tue Mar 13, 2018 12:06 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by aburntoutcase » Tue Jan 15, 2019 6:51 pm

I think the basic flaw in their methodology is that they are forecasting significant decline in S&P 500 operating profit margins and a valuation multiple decline. While latter seems justified, I am not so sure on former. I totally get their POV that profit margins are the most mean reverting series in capitalism but I think sector composition is not uniform across the decades, and certain sectors do have higher profit margins structurally.

staythecourse
Posts: 6573
Joined: Mon Jan 03, 2011 9:40 am

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by staythecourse » Tue Jan 15, 2019 7:43 pm

aburntoutcase wrote:
Tue Jan 15, 2019 6:51 pm
I think the basic flaw in their methodology is that they are forecasting significant decline in S&P 500 operating profit margins and a valuation multiple decline. While latter seems justified, I am not so sure on former. I totally get their POV that profit margins are the most mean reverting series in capitalism but I think sector composition is not uniform across the decades, and certain sectors do have higher profit margins structurally.
Funny I thought their flaw was believing everyone out there is so gullible to believe anything they say as fact. I can guarantee they don't invest their personal monies using their forecasts.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

staythecourse
Posts: 6573
Joined: Mon Jan 03, 2011 9:40 am

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by staythecourse » Tue Jan 15, 2019 7:46 pm

nisiprius wrote:
Mon Jan 14, 2019 6:17 pm
These were their forecasts 6/30/11-6/30/18

Image

Here's my chart of how their forecasts compared with the actual results:

Image

The correlation coefficient between the forecasts and the observations 0.05. That is to say, virtually zero. The forecast had virtually no predictive power.

Why do people pay any attention to GMO's forecasts?
Nisi,

This may be a great side gig for you. Setup a website and go back and evaluate every major call that is made and see how it turns out. My guess is eventually Wall Street or CNBC will come in and purchase the site for millions just to shut it down. Would be a great way to make a TON of money.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

reformed.trader
Posts: 51
Joined: Sat Apr 15, 2017 11:14 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by reformed.trader » Tue Jan 15, 2019 10:27 pm

randomguy wrote:
Mon Jan 14, 2019 3:35 pm
reformed.trader wrote:
Mon Jan 14, 2019 3:25 pm
Annualized 7 year returns for Emerging Markets went from -2% to +4.7% in one year? Seems odd.
Makes sense. Imagine the difference between 1 year of -20% and 6 years of 4% returns and 7 years of 5% returns. Now that we had a bad year, the bull market can start up:)

The issue isn't so much the GMO being wrong. Nobody should predict super accurate estimates. But have they every been wrong on the bullish side? Some one needs to dredge up the 7 year predcations from say 1998-2003. Now maybe we have gotten lucky but at a certain point you have to consider that their model is just wrong.
If the index was at 100 to start 2017, -2% for 7 years would get you 86.81.

The index was down about 20%, so that would be starting at 80. If you are getting 4.7% on 80 for 6 years that would give you 105.38.

They aren't even close. Something major had to of changed other than price. Is the OP sure he is pulling the data correctly? All other estimates I have seen back in 2017 all called for EM to generate 5%+. Estimates are in line now with the rest of the research groups I follow.

dkturner
Posts: 1400
Joined: Sun Feb 25, 2007 7:58 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by dkturner » Wed Jan 16, 2019 12:21 pm

aburntoutcase wrote:
Mon Jan 14, 2019 3:34 pm
At the end of 2010 this is what GMO forecast. US large cap was going to have 0.4% real returns, developed intl large 2.1% and emerging markets 4.1%. They were quite wrong.

Image
Historically GMO has never had any luck divining what magnitude of return any asset class would deliver over the ensuing 7-10 year period. They had been fairly successful in picking the relative positions of various asset classes among each other.

Taking the chart you posted for year end 2010 GMO predicted that large-cap US equities would perform better than small-cap US equities by 2.3 percentage points. They did, by 2.8 percentage points. Pretty good call on relative performance.

Looking at large-cap US equities versus large-cap international equities GMO predicted that The US equities would UNDER perform the international equities by 2.5 percentage points. Between 2011 and 2018 The US equities actually OUT performed international equities by 8.8 percentage points. Ouch!

Looks like previous accurate calls were merely good luck (blind squirrel found a nut), not attributable to the skill of their in-house gurus.
Last edited by dkturner on Fri Jan 18, 2019 8:44 am, edited 1 time in total.

totesmagotes
Posts: 40
Joined: Sat Nov 04, 2017 10:08 am

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by totesmagotes » Wed Jan 16, 2019 12:50 pm

nisiprius wrote:
Mon Jan 14, 2019 6:17 pm
These were their forecasts 6/30/11-6/30/18


The correlation coefficient between the forecasts and the observations 0.05. That is to say, virtually zero. The forecast had virtually no predictive power.

Why do people pay any attention to GMO's forecasts?
It's amazing to think of what changes in 7-10 years. There is an inherent uncertainty in the world. For example, in 2009, could anyone have seen the Arab Spring coming 2 years down the road? What about Brexit in 2016? What about the election of such an "unconventional" (I'm being nice here to avoid politics) administration in late 2016? There are a host of other geopolitical developments that are inherently unknowable. The only thing we know is that there WILL be unforeseen events that will significantly affect the markets (e.g., supply/demand for commodities, new or deleted laws and regulations for markets/sectors, etc.).

In 2007, did anyone envision Amazon becoming one of the largest companies in the world? Who foresaw GE plummeting? As with geopolitics, there are a host of single-company developments that are essentially impossible to know 7-10 years in advance. Of course we can cherry pick certain calls, just like one could think that someone who flips a coin and lands it heads 10 times in a row is somehow skillful. As has been shown before, these tend to be statistical "flukes" with no predictive power.

Of course, this is not to say that valuations have no impact on future earnings. Indeed, one of the main "factors" in future returns is current valuation (think growth vs. value investing). However, AFAIK, this seem to apply on a "marketwide" level, not on a sector-by-sector or more granular level.

/ I'm mostly 3-fund with some tilts to EM, small-cap value, and fallen angels, but that's to appease the side of my investing that would otherwise be tempted to pick individual stocks.

garlandwhizzer
Posts: 2159
Joined: Fri Aug 06, 2010 3:42 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by garlandwhizzer » Wed Jan 16, 2019 12:58 pm

Nisi nailed it. Best not to put too much faith in GMO or anyone else's estimates of future returns. Put simply the future is both unknown and unknowable. I believe the best approach for investors is to get a solid portfolio and pay little/no attention to forecasts and the financial media noise. Buffett suggests that most investors are best served by buying cheap broadly based index funds, ignoring such noise, and rarely even looking at their portfolios which, if you can do it, is probably a good idea.

Garland Whizzer

User avatar
jharkin
Posts: 2041
Joined: Mon Mar 28, 2016 7:14 am
Location: Boston suburbs

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by jharkin » Wed Jan 16, 2019 1:51 pm

Morningstar recently did a roundup of all the usual suspects current return predictions - GMO is in there and by FAR the most bearish.

https://www.morningstar.com/articles/90 ... rns-2.html

Interestingly, Jack's raised his bond outlook a bit, I guess to reflect the current rate trends. He now has bonds nearly at parity to stocks in the medium term.

Topic Author
Beliavsky
Posts: 761
Joined: Sun Jun 29, 2014 10:21 am

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by Beliavsky » Wed Jan 16, 2019 2:03 pm

jharkin wrote:
Wed Jan 16, 2019 1:51 pm
Morningstar recently did a roundup of all the usual suspects current return predictions - GMO is in there and by FAR the most bearish.

https://www.morningstar.com/articles/90 ... rns-2.html
Thanks. Below is what the forecasters say in the article. Many forecasters expect U.S. stocks to earn only about 1% more than bonds over the next decade. If you are risk-averse, that would suggest tilting towards bonds in tax-deferred accounts.

BlackRock Investment Institute
Highlights: 7% nominal (non-inflation-adjusted) return for U.S. large caps over the next decade; 9% for non-U.S. large caps; 3.3% for the U.S. Aggregate Bond index(December 2018).

John C. Bogle, founder of Vanguard Group
Highlights: 4%-5% returns for stocks (nominal); 4% nominal returns for bonds over the next decade (October 2018).

GMO (requires login)
Highlights: negative 4.1% real (inflation-adjusted) returns for U.S. large caps over the next seven years; negative 0.2% real returns for U.S. bonds; 4.4% real returns for emerging-markets equities; 2.9% real returns for emerging-markets debt (November 2018).

J.P. Morgan Asset Management
Highlights: 5.25% return assumption (nominal) for U.S. equities over a 10- to 15-year horizon; 4.5% nominal return assumption for U.S. investment-grade corporate bonds over 10- to 15-year holding period (October 2018).

Morningstar Investment Management (login required)
Highlights: 1.8% 10-year nominal returns for U.S. stocks; 3.3% 10-year nominal returns for U.S. bonds (Sept. 30, 2018).

Research Affiliates
Highlights: 0.7% real returns for U.S. large caps during the next 10 years; 0.5% real returns for the Barclays U.S. Aggregate Bond Index (Dec. 31, 2018; valuation-dependent model).

Vanguard
Highlights: Nominal U.S. equity-market returns in the 3% to 5% range during the next decade; 6% to 8% returns for non-U.S. equities; 2.5% to 4.5% expected returns for global fixed-income markets (December 2018).

garlandwhizzer
Posts: 2159
Joined: Fri Aug 06, 2010 3:42 pm

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by garlandwhizzer » Wed Jan 16, 2019 9:35 pm

Jack's expected bond returns are higher than most because his bond portfolio employs some corporate bonds and longer duration bonds, both of which have higher yields but greater potential risk than Treasuries. By the way, The Great One passed away today. It is estimated that his creation of the index fund and Vanguard and the shift among active funds toward lower costs in response to Vanguard's overwhelming success has saved investors a trillion dollars over his lifetime. All investors in mutual funds and ETFs owe a great deal to Jack Bogle whether they know it or not.

Garland Whizzer

longinvest
Posts: 3257
Joined: Sat Aug 11, 2012 8:44 am

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by longinvest » Fri Jan 18, 2019 8:03 am

longinvest wrote:
Mon Jan 14, 2019 5:56 pm
aburntoutcase wrote:
Mon Jan 14, 2019 3:34 pm
At the end of 2010 this is what GMO forecast. US large cap was going to have 0.4% real returns, developed intl large 2.1% and emerging markets 4.1%. They were quite wrong.

Image
For the 7 years that followed that prediction or, more exactly, from December 31, 2010 to December 31, 2017, US Large Caps* returned 13.6% annualized.

* Vanguard Large-Cap ETF (VV).

The prediction, for US Large Caps, was 0.4% +/- 6.5% which is a range of -6.1% to 6.9%. Actual returns were 6.7% higher than the top of their prediction range (6.9%).

Despite a large error range, the prediction was completely off by more than twice the predicted error!

What I like about the Bogleheads investment philosophy is that it doesn't rely on future return predictions. It simply tells me to invest my money into assets with an internal return driver (stocks which can generate profits, bonds which pay coupons and their principal back) using index funds covering broad markets. It tells me not to take too much or too little risk, to keep fees low, and to accept market returns by staying the course with my chosen allocation. Investing doesn't have to be more complicated than this.
It's a mistake to look at annualized returns to assess the magnitude of a multi-year prediction error. I made this mistake above. Let me fix this.

At the end of 2010, GMO predicted that US Large Caps would return 0.4% +/- 6.5% in the upcoming 7 years.

A better way to state this prediction is this:

At the end of 2010, GMO predicted that $10,000 invested into US Large Caps would grow on a total return basis to a value between $6,437 and $15,953 and centered on $10,283. This is a huge range! $15,953 is two and half times as big as $6,437.

During this 7 year period, from the end of 2010 to the end of 2017, US Large Caps had an an annualized 13.6% return; $10,000 grew to $24,414.

I think that a chart is useful to visualize how wrong the prediction and its range were:

Image

GMO predicted that $10,000 invested into US Large Caps could lose $3,563 in the worst case, and gain $5,953 in the best case. In real life, $10,000 gained 14,414, 2.5 times as much as the predicted best case!
Bogleheads investment philosophy

asif408
Posts: 1568
Joined: Sun Mar 02, 2014 8:34 am
Location: Florida

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by asif408 » Fri Jan 18, 2019 8:47 am

longinvest wrote:
Fri Jan 18, 2019 8:03 am

It's a mistake to look at annualized returns to assess the magnitude of a multi-year prediction error. I made this mistake above. Let me fix this.

At the end of 2010, GMO predicted that US Large Caps would return 0.4% +/- 6.5% in the upcoming 7 years.

A better way to state this prediction is this:

At the end of 2010, GMO predicted that $10,000 invested into US Large Caps would grow on a total return basis to a value between $6,437 and $15,953 and centered on $10,283. This is a huge range! $15,953 is two and half times as big as $6,437.

During this 7 year period, from the end of 2010 to the end of 2017, US Large Caps had an an annualized 13.6% return; $10,000 grew to $24,414.

I think that a chart is useful to visualize how wrong the prediction and its range were:

Image

GMO predicted that $10,000 invested into US Large Caps could lose $3,563 in the worst case, and gain $5,953 in the best case. In real life, $10,000 gained 14,414, 2.5 times as much as the predicted best case!
So when I look at the other estimates for foreign large, EM, & US gov't bonds (which most Bogleheads hold in some form) from 2011 through the end of 2017, they are within range: https://www.portfoliovisualizer.com/bac ... ion3_3=100

US (using VTI): 11.59% real return, estimate was 0.4% +/-6.5
EAFE (using EFA): 4.11% real return, estimate was 2.1% +/-6.5
EM (using VWO): 0.23% real return, estimate was 4.1% +/-10.5
Intermediate term gov't bonds (using VFITX): 0.8% real return, estimate was 0.5% +/- 4.5

So do you think the outlier of the US was a statistical anomaly or do you use that to say their overall forecasts are not worth following?

longinvest
Posts: 3257
Joined: Sat Aug 11, 2012 8:44 am

Re: GMO's 7‐Year Asset Class Real Return Forecasts

Post by longinvest » Fri Jan 18, 2019 8:55 am

asif408 wrote:
Fri Jan 18, 2019 8:47 am
longinvest wrote:
Fri Jan 18, 2019 8:03 am

It's a mistake to look at annualized returns to assess the magnitude of a multi-year prediction error. I made this mistake above. Let me fix this.

At the end of 2010, GMO predicted that US Large Caps would return 0.4% +/- 6.5% in the upcoming 7 years.

A better way to state this prediction is this:

At the end of 2010, GMO predicted that $10,000 invested into US Large Caps would grow on a total return basis to a value between $6,437 and $15,953 and centered on $10,283. This is a huge range! $15,953 is two and half times as big as $6,437.

During this 7 year period, from the end of 2010 to the end of 2017, US Large Caps had an an annualized 13.6% return; $10,000 grew to $24,414.

I think that a chart is useful to visualize how wrong the prediction and its range were:

Image

GMO predicted that $10,000 invested into US Large Caps could lose $3,563 in the worst case, and gain $5,953 in the best case. In real life, $10,000 gained 14,414, 2.5 times as much as the predicted best case!
So when I look at the other estimates for foreign large, EM, & US gov't bonds (which most Bogleheads hold in some form) from 2011 through the end of 2017, they are within range: https://www.portfoliovisualizer.com/bac ... ion3_3=100

US (using VTI): 11.59% real return, estimate was 0.4% +/-6.5
EAFE (using EFA): 4.11% real return, estimate was 2.1% +/-6.5
EM (using VWO): 0.23% real return, estimate was 4.1% +/-10.5
Intermediate term gov't bonds (using VFITX): 0.8% real return, estimate was 0.5% +/- 4.5

So do you think the outlier of the US was a statistical anomaly or do you use that to say their overall forecasts are not worth following?
They predicted that $10,000 invested into intermediate government bonds would lose 25% ($2,485) in the worst case, or gain 41% ($4,071) in the best case, over a 7 year period.

I know that I won't base my investing decisions on GMO's or anybody else's predictions. I've explained why, along with a proof, a few years ago.
Bogleheads investment philosophy

Post Reply