Long term low returns predicted

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
sixtyforty
Posts: 656
Joined: Tue Nov 25, 2014 11:22 am
Location: USA

Re: Long term low returns predicted

Post by sixtyforty »

It's actually Just the opposite for me. When everyone is predicting high returns in the future, I get very worried.
"Simplicity is the ultimate sophistication" - Leonardo Da Vinci
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Long term low returns predicted

Post by willthrill81 »

sixtyforty wrote: Thu Sep 23, 2021 11:44 am It's actually Just the opposite for me. When everyone is predicting high returns in the future, I get very worried.
When everyone is predicting low future returns, it gives me at least a little assurance that we're probably not in a bubble. Most folks in the late 1990s were crowing about how great stocks were going to perform in the 2000s.
The Sensible Steward
toomanysidehustles
Posts: 585
Joined: Tue Oct 06, 2020 10:09 am

Re: Long term low returns predicted

Post by toomanysidehustles »

alex123711 wrote: Wed Sep 22, 2021 8:50 pm Seeing a lot of predictions (including on this forum) about low returns for a prolonged period (30+ yrs) some predictions are only 1% - 2% above inflation/ CPI. If this does occur then I think index funds will be a bad investment for that time period as 2% over a 30 yr period would not even double your money. At what point does it become 'not worth it' and worth spending the money on other things/ investments/ real estate/ business?
This is why I am equally invested in real estate as index funds, with a growing interest (the last 18 months) in converting to AirBnB. Two of our AirBnB's combined produce a net income for our family of just over 90K, so my focus is on that currently. The income allows me to buy more properties, or more index funds. :D
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

So I want to push back a little on the idea that if the next decade or so is relatively weak for stocks with high valuations, they will end with low valuations and then the next decade or so after that might be better.

That is certainly one possible scenario that has happened fairly frequently in the past, but it is definitely not the only possible scenario.

Another possible scenario is valuations don't actually change much, but there isn't enough real earnings growth to provide more than a low return at those valuations. There is no real limit to how long such a period can last. In fact, it is essentially equivalent to assuming a permanent reduction in the equity risk premium, and there is no particular reason to assume that couldn't happen.

This relates to another issue I see with some of these discussions, which is to point out how there is indeed a recent track record of various people looking at historically high valuations for something like the SP500 (or equivalent), predicting low returns, and then returns being pretty good anyway. Which is then used to suggest all such predictions should be viewed with inherent skepticism.

The issue I see is those predictions were "wrong" largely because the SP500's valuation keeps on going up--with lots of ups and downs, of course, but overall it has been in a largely upward trend since the early 1990s (or indeed longer depending on when you want to start).

Which is fine, but I think everyone needs to have in the back of their mind the question of whether these valuation increases will keep happening indefinitely. Again, not necessarily crash back (or not for long), but simply level out at some point in a newer, higher range.

Because again, if that happens, then it would become pretty easy to have historically low average returns indefinitely.
atdharris
Posts: 2091
Joined: Wed Jan 02, 2019 2:18 pm

Re: Long term low returns predicted

Post by atdharris »

skierincolorado wrote: Thu Sep 23, 2021 10:48 am
atdharris wrote: Thu Sep 23, 2021 10:46 am No one knows anything. I believe the same thing has been said for a decade now, and the last decade of returns has been pretty good in my opinion
People might have said the same thing a decade ago but the argument was weak. Valuations like cape were much lower in 2011. Even a few years ago valuations were still much lower.
Maybe this time they are right, who knows. But I have a 30 year time horizon, so even if we see lower returns, I anticipate all things will even out.
barnaclebob
Posts: 5564
Joined: Thu Aug 09, 2012 10:54 am

Re: Long term low returns predicted

Post by barnaclebob »

I had to check the date on this post to make sure it wasnt resurrected from 2012.
User avatar
HomerJ
Posts: 21246
Joined: Fri Jun 06, 2008 12:50 pm

Re: Long term low returns predicted

Post by HomerJ »

NiceUnparticularMan wrote: Thu Sep 23, 2021 12:57 pm So I want to push back a little on the idea that if the next decade or so is relatively weak for stocks with high valuations, they will end with low valuations and then the next decade or so after that might be better.

That is certainly one possible scenario that has happened fairly frequently in the past, but it is definitely not the only possible scenario.

Another possible scenario is valuations don't actually change much, but there isn't enough real earnings growth to provide more than a low return at those valuations. There is no real limit to how long such a period can last. In fact, it is essentially equivalent to assuming a permanent reduction in the equity risk premium, and there is no particular reason to assume that couldn't happen.
You are correct that this is possible, but I wouldn't go so far to say "there is no particular reason it couldn't happen".

Growth and returns aren't just random independent events. Other variables could show up if we have low growth and returns for a while.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
imak
Posts: 212
Joined: Fri Mar 22, 2019 5:18 pm
Location: Austin, TX

Re: Long term low returns predicted

Post by imak »

Most of the concerns in valuation are misplaced.

As per Bogle's expected return formula:
Future Market Returns = Dividend Yield + Earnings Growth +/- Change in P/E Ratio

Consider Dividend yield is about 2% on average.
Earnings growth is approximately equal to GDP growth (3%-5%).

The "Change in P/E ratio" component is the part which most people are worried about.
Bogle called this component "speculative return", because nobody can predict this change in earnings multiple.

High CAPE ratio is often said to "predict" future P/E compression but that prediction itself is still a speculation.

It might very well be the case that earnings grow rapidly while price trends sideways for a long time. In this case there is no bear market.

For all lifecycle investing purposes, its best to simply consider 5% equity risk premium and tune out the noise.
AlohaJoe
Posts: 6609
Joined: Mon Nov 26, 2007 1:00 pm
Location: Saigon, Vietnam

Re: Long term low returns predicted

Post by AlohaJoe »

jeffyscott wrote: Thu Sep 23, 2021 9:25 am I don't think there is anyone more pessimistic than GMO? And I don't think even they would estimate 30 year expected returns for US stocks at under 2% real.
Exactly. Some people are predicting low returns, yes. But only for 7 or 10 years. Not 30. And definitely not "30+". The OP just made that up out of whole cloth. Who is making 40 years forecasts :confused
RA has an option to use a "yield and growth" model for expected returns. I think that model, unlike their (10 year) "valuation dependent" model, is actually independent of the time period and so could be their estimate of 30 year expected returns?
Even their "yield and growth" model is only good for 10ish years if you look at their methodology paper.

For the "yield" they just assume current yields continue but then spend several pages explaining how yields change over time (see Table 4), so it isn't independent of time period.

For the "growth" it is based on the last 10-years. ("The growth forecast is derived by first taking the difference between the 10-year smoothed earnings yield (average 10-year earnings divided by price) and the 10-year smoothed dividend yield (average 10-year dividends divided by price).") So that's also not independent of time period.
stocknoob4111
Posts: 3495
Joined: Sun Jan 07, 2018 11:52 am

Re: Long term low returns predicted

Post by stocknoob4111 »

if you look at the predictions in the historical articles you notice one consistent pattern.. most of them were wildly inaccurate

Also if you're diversified into other classes besides US Large you should do fairly well.. International is beaten down and S&P600 Small Caps with a current PE of 16 is below historical median.
Last edited by stocknoob4111 on Thu Sep 23, 2021 11:31 pm, edited 1 time in total.
absolute zero
Posts: 1244
Joined: Thu Dec 29, 2016 3:59 pm

Re: Long term low returns predicted

Post by absolute zero »

AlohaJoe wrote: Thu Sep 23, 2021 11:07 pm
jeffyscott wrote: Thu Sep 23, 2021 9:25 am I don't think there is anyone more pessimistic than GMO? And I don't think even they would estimate 30 year expected returns for US stocks at under 2% real.
Exactly. Some people are predicting low returns, yes. But only for 7 or 10 years. Not 30. And definitely not "30+". The OP just made that up out of whole cloth. Who is making 40 years forecasts :confused
RA has an option to use a "yield and growth" model for expected returns. I think that model, unlike their (10 year) "valuation dependent" model, is actually independent of the time period and so could be their estimate of 30 year expected returns?
Even their "yield and growth" model is only good for 10ish years if you look at their methodology paper.

For the "yield" they just assume current yields continue but then spend several pages explaining how yields change over time (see Table 4), so it isn't independent of time period.

For the "growth" it is based on the last 10-years. ("The growth forecast is derived by first taking the difference between the 10-year smoothed earnings yield (average 10-year earnings divided by price) and the 10-year smoothed dividend yield (average 10-year dividends divided by price).") So that's also not independent of time period.
GMO is pretty impressively bearish. They say -7.5% real CAGR for next 7 years for US equities. If you follow those 7 years with 23 years of 6% real (just a made up assumption) it comes out to 2.5% real over the full 30 year period. I don’t know what GMO thinks will happen over 30 years (nor do I care to be honest) but based off this simple math I would not be surprised if their forecast for even a long time period was in the ballpark of 2% real.
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Long term low returns predicted

Post by 000 »

absolute zero wrote: Thu Sep 23, 2021 11:30 pm GMO is pretty impressively bearish. They say -7.5% real CAGR for next 7 years for US equities. If you follow those 7 years with 23 years of 6% real (just a made up assumption) it comes out to 2.5% real over the full 30 year period. I don’t know what GMO thinks will happen over 30 years (nor do I care to be honest) but based off this simple math I would not be surprised if their forecast for even a long time period was in the ballpark of 2% real.
That would only be a ~42% drop taking us to a VTI price level of 133, which was seen in March 2020, Q4 2018, and Q4 2017.
User avatar
jeffyscott
Posts: 13438
Joined: Tue Feb 27, 2007 8:12 am

Re: Long term low returns predicted

Post by jeffyscott »

absolute zero wrote: Thu Sep 23, 2021 11:30 pm
AlohaJoe wrote: Thu Sep 23, 2021 11:07 pm
jeffyscott wrote: Thu Sep 23, 2021 9:25 am I don't think there is anyone more pessimistic than GMO? And I don't think even they would estimate 30 year expected returns for US stocks at under 2% real.
Exactly. Some people are predicting low returns, yes. But only for 7 or 10 years. Not 30. And definitely not "30+". The OP just made that up out of whole cloth. Who is making 40 years forecasts :confused
RA has an option to use a "yield and growth" model for expected returns. I think that model, unlike their (10 year) "valuation dependent" model, is actually independent of the time period and so could be their estimate of 30 year expected returns?
Even their "yield and growth" model is only good for 10ish years if you look at their methodology paper.

For the "yield" they just assume current yields continue but then spend several pages explaining how yields change over time (see Table 4), so it isn't independent of time period.

For the "growth" it is based on the last 10-years. ("The growth forecast is derived by first taking the difference between the 10-year smoothed earnings yield (average 10-year earnings divided by price) and the 10-year smoothed dividend yield (average 10-year dividends divided by price).") So that's also not independent of time period.
GMO is pretty impressively bearish. They say -7.5% real CAGR for next 7 years for US equities. If you follow those 7 years with 23 years of 6% real (just a made up assumption) it comes out to 2.5% real over the full 30 year period. I don’t know what GMO thinks will happen over 30 years (nor do I care to be honest) but based off this simple math I would not be surprised if their forecast for even a long time period was in the ballpark of 2% real.
Actually, as I mentioned above, GMO's 7 year forecast for US large caps is -8.4% real. Not that it makes a huge difference, as I had posted, my guess at a hypothetical GMO 30 year forecast was about 2.25% real.
User avatar
jeffyscott
Posts: 13438
Joined: Tue Feb 27, 2007 8:12 am

Re: Long term low returns predicted

Post by jeffyscott »

AlohaJoe wrote: Thu Sep 23, 2021 11:07 pm
RA has an option to use a "yield and growth" model for expected returns. I think that model, unlike their (10 year) "valuation dependent" model, is actually independent of the time period and so could be their estimate of 30 year expected returns?
Even their "yield and growth" model is only good for 10ish years if you look at their methodology paper.

For the "yield" they just assume current yields continue but then spend several pages explaining how yields change over time (see Table 4), so it isn't independent of time period.

For the "growth" it is based on the last 10-years. ("The growth forecast is derived by first taking the difference between the 10-year smoothed earnings yield (average 10-year earnings divided by price) and the 10-year smoothed dividend yield (average 10-year dividends divided by price).") So that's also not independent of time period.
Thanks for the info on their methodology. It's at least a model that is more reasonable to use as a long term projection than the valuation dependent one.

Breaking it down they have 1.3% yield + 1.2% (real) growth. I think the current yield is generally considered to be a real yield, so expected to grow at the inflation rate. Is there a better figure than their 1.2% to use for the expected real growth rate over the next 30 years?
Young Boglehead
Posts: 321
Joined: Wed Jan 02, 2019 4:11 pm

Re: Long term low returns predicted

Post by Young Boglehead »

Don't many of these predictions also have a range of something negative to something many here would consider too optimistic? And then they settle at some low number in the middle. Bernstein has mentioned this. If so, i have no clue why anyone would focus on the singular number when it's really a range - and a range so wide as to be completely useless. And even then sometimes its wrong!
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

HomerJ wrote: Thu Sep 23, 2021 4:04 pm
NiceUnparticularMan wrote: Thu Sep 23, 2021 12:57 pm So I want to push back a little on the idea that if the next decade or so is relatively weak for stocks with high valuations, they will end with low valuations and then the next decade or so after that might be better.

That is certainly one possible scenario that has happened fairly frequently in the past, but it is definitely not the only possible scenario.

Another possible scenario is valuations don't actually change much, but there isn't enough real earnings growth to provide more than a low return at those valuations. There is no real limit to how long such a period can last. In fact, it is essentially equivalent to assuming a permanent reduction in the equity risk premium, and there is no particular reason to assume that couldn't happen.
You are correct that this is possible, but I wouldn't go so far to say "there is no particular reason it couldn't happen".

Growth and returns aren't just random independent events. Other variables could show up if we have low growth and returns for a while.
Sure, but if you do a reasonable ground up model of how things could look going forward . . . .

Real aggregate corporate earnings tend to track up with real GDP over the long run, which is unsurprising. But then real earnings per share lag significantly (like around half or so historically), largely because of a "dilution" effect were so much of "new" corporate earnings go to businesses without outstanding shares (yet).

OK, so for various reasons including stagnant working age population growth, the end of increasing female labor force participation circa 2000, and so on, these days most longer-term real GDP growth forecasts for the U.S. are pretty mild--like around 2% I think, give or take. And I don't think that is necessarily an unsustainable sort of scenario.

OK, so 2% real GDP growth, around the same in terms of aggregate corporate earnings, but dilute that down for real earnings per share . . . we are only talking mild growth contributions on the order of 1% a year.

Of course part of the return on stocks isn't growth in real earnings per share, it is just actual earnings per share (however they are then "yielded" to common stock owners, whether it be by dividends, stock buybacks, business expansions, or so on).

But that is where valuations kick in, because at high prices in terms of earnings per share, that "yield" is low in rate terms.

Combine that low "yield" with low growth in real earnings per share, and valuations stabilizing without further structural increases, and voila--you have an indefinite period of low real returns on U.S. stocks.

And again, none of this implies some sort of dramatic crisis, or is generally unsustainable in nature. It is just the predictable result given that set of reasonably plausible assumptions.
User avatar
HomerJ
Posts: 21246
Joined: Fri Jun 06, 2008 12:50 pm

Re: Long term low returns predicted

Post by HomerJ »

absolute zero wrote: Thu Sep 23, 2021 11:30 pm GMO is pretty impressively bearish. They say -7.5% real CAGR for next 7 years for US equities. If you follow those 7 years with 23 years of 6% real (just a made up assumption) it comes out to 2.5% real over the full 30 year period. I don’t know what GMO thinks will happen over 30 years (nor do I care to be honest) but based off this simple math I would not be surprised if their forecast for even a long time period was in the ballpark of 2% real.
People always say that, but that's never how it works.

It COULD work that way in the future, but, so far, we haven't ever gotten 23 years of "average".

If get -50% over the next 7 years, some of the following years, based on history, will be in 12%-30% range, not just 6%

The long-term average, so far, of 6%-7% real, INCLUDES all the bad years.

So the good years, so far, are very good, not average.

But who knows what the future holds? Not GMO, not me, not you, not anyone. It definitely could be different from past.
Last edited by HomerJ on Fri Sep 24, 2021 9:39 am, edited 2 times in total.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
HomerJ
Posts: 21246
Joined: Fri Jun 06, 2008 12:50 pm

Re: Long term low returns predicted

Post by HomerJ »

Young Boglehead wrote: Fri Sep 24, 2021 7:30 am Don't many of these predictions also have a range of something negative to something many here would consider too optimistic? And then they settle at some low number in the middle. Bernstein has mentioned this. If so, i have no clue why anyone would focus on the singular number when it's really a range - and a range so wide as to be completely useless. And even then sometimes its wrong!
This is correct.. The error bars of these predictions are very large. Like plus/minus 6% or 8%.

I don't need a PhD to state that 10-year returns will be somewhere between -2% and 14%

They state the middle number as "the" prediction, but we don't even know if the distribution is normal. Returns aren't some random independent variable. If we get into negative returns for a while, the Fed and other entities may step in and do something which will affect returns.

There are thousands of variables, including human laws, and even human emotions.

No one can accurately predict the future, and even when they DO give themselves plus/minus 6% or 8%, the "experts" still often end up wrong.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
make_a_better_world
Posts: 292
Joined: Mon Mar 18, 2013 11:55 pm

Re: Long term low returns predicted

Post by make_a_better_world »

I cannot predict what will happen over a short period of time but it is not difficult to predict the general trend of the longer time horizon.

Man began with a pointed stick and an 18-hour-day job chasing lizards. He has increased his wealth many, many fold.

Around 1900 many major cities were mostly fields. Today there are incredible skyscrapers, transportation systems, communication systems, and immense wealth. Simply look at a photo of the skyline of any major city one hundred years ago compared to today.

In our lifetime, artificial intelligence and robotics will increase wealth to a greater extent than what happened in the Industrial Revolution in 1760. We are moving from human minds operating simple mechanical muscles to smarter mechanical minds operating sophisticated mechanical muscles.

This new wealth will not be homogeneously distributed. Many in the workforce will be displaced. Those who participate as owners or investors in the companies that benefit from this technology will benefit in spectacular fashion. Those who do not will move further towards the other side of gap inequality.

I will be holding index funds of these companies. Good luck.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

imak wrote: Thu Sep 23, 2021 8:01 pm Most of the concerns in valuation are misplaced.

As per Bogle's expected return formula:
Future Market Returns = Dividend Yield + Earnings Growth +/- Change in P/E Ratio

Consider Dividend yield is about 2% on average.
Earnings growth is approximately equal to GDP growth (3%-5%).

The "Change in P/E ratio" component is the part which most people are worried about.
Bogle called this component "speculative return", because nobody can predict this change in earnings multiple.

High CAPE ratio is often said to "predict" future P/E compression but that prediction itself is still a speculation.

It might very well be the case that earnings grow rapidly while price trends sideways for a long time. In this case there is no bear market.

For all lifecycle investing purposes, its best to simply consider 5% equity risk premium and tune out the noise.
So in another post I just used a very similar model, but the difference is in the details on that bolded part.

First, the dividend yield on the SP500 is currently 1.3%, not 2% That is a direct and predictable consequence of higher valuations.

Second, again real long-term GDP growth forecasts for the U.S. these days are more in the 2% range (see CBO forecasts and such if you want examples). Historic real GDP growth rates in the U.S. have typically been higher, but that was either during the U.S.'s industrialization phase, or at least when we had a combination of faster working age population growth and increasing female participation in the labor force. So, the labor supply, and therefore real GDP, was growing much faster, and without those factors potential real GDP growth has now slowed considerably.

And then there is the dilution effect where real GDP growth does approximately equal real corporate earnings growth, but a lot of that growth in earnings is not captured by existing common stock investors.

So again, a better estimate where you have 3-5% is actually more like 1% (in real terms) these days.

Which means by this sort of model logic, you should really be revising down that 5% real return estimate to more like 2.3% or so.

Of course as others have pointed out, this is nothing new. Valuations have been trending up, and the SP500 dividend yield has been trending down, for decades. So has real potential (and ultimately actual) GDP growth.

And people making predictions of lower US stock returns as a result, using precisely this sort of model, have nonetheless been wrong--so far.

Why? Precisely because valuations just kept on increasing.

But if valuations stabilize, and all the rest of these long-term trends remain the same, then these models may stop being wrong. Indefinitely.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

jeffyscott wrote: Fri Sep 24, 2021 7:23 am Breaking it down they have 1.3% yield + 1.2% (real) growth. I think the current yield is generally considered to be a real yield, so expected to grow at the inflation rate. Is there a better figure than their 1.2% to use for the expected real growth rate over the next 30 years?
So I came up with a similar number just doing a back of the envelope calculation, and I suspect the answer to your question is basically no. Obviously no one knows for sure what U.S. real GDP growth, real growth in aggregate corporate earnings, and real earnings per share growth will look like over the next 30 years. But I think pretty much any sensible model given current conditions and trends is going to give you an answer in that range.

This is getting repetitive, but this is a sort of macroeconomic perspective that does seem to have been largely ignored by a lot of personal investors who simply assume historic rates of return will continue to apply. If you believe, as you should, that the long-term returns on U.S. stocks will be linked to U.S. real GDP growth, then the observed and predicted slowdown in U.S. real GDP growth implies lower rates of returns on stocks.

The typical answer again is that people have been saying this for years and been wrong so far. But they have basically only been wrong because of valuation increases.

Which in turn have pushed down real yields.

Anyway, food for thought.
User avatar
HomerJ
Posts: 21246
Joined: Fri Jun 06, 2008 12:50 pm

Re: Long term low returns predicted

Post by HomerJ »

NiceUnparticularMan wrote: Fri Sep 24, 2021 9:42 am But if valuations stabilize, and all the rest of these long-term trends remain the same, then these models may stop being wrong. Indefinitely.
Yeah, if all the thousands of variables stop being "variable" and instead become "constant", you may be right. Forever.

:oops:
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Long term low returns predicted

Post by vanbogle59 »

HomerJ wrote: Fri Sep 24, 2021 9:37 am The error bars of these predictions are very large. Like plus/minus 6% or 8%.
As reference, I just asked firecalc what the distribution of returns for 100/0 looked like in its historical data.
Roughly:
Median - 6.5%
Low - 3.2%
High - 10.2%
That's probably as reasonable "prediction" for the next 30 years as anyone could talk me into.

Throw in these 2 "certainties" (both of which appear in the firecalc time range) and I'll buy in:
1) the advancement of society/technology/culture in ways that makes all of our lives dramatically better
2) horrible, tragic events that threaten millions of lives, including war, pestilence....
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

make_a_better_world wrote: Fri Sep 24, 2021 9:41 am I cannot predict what will happen over a short period of time but it is not difficult to predict the general trend of the longer time horizon.

Man began with a pointed stick and an 18-hour-day job chasing lizards. He has increased his wealth many, many fold.

Around 1900 many major cities were mostly fields. Today there are incredible skyscrapers, transportation systems, communication systems, and immense wealth. Simply look at a photo of the skyline of any major city one hundred years ago compared to today.

In our lifetime, artificial intelligence and robotics will increase wealth to a greater extent than what happened in the Industrial Revolution in 1760. We are moving from human minds operating simple mechanical muscles to smarter mechanical minds operating sophisticated mechanical muscles.

This new wealth will not be homogeneously distributed. Many in the workforce will be displaced. Those who participate as owners or investors in the companies that benefit from this technology will benefit in spectacular fashion. Those who do not will move further towards the other side of gap inequality.

I will be holding index funds of these companies. Good luck.
I would suggest this is the most uncertain part of your prediction.

One obvious point is just that sociopolitical developments could interfere with this sort of outcome, at least in the long run.

A less obvious, but potentially more profound, point is that in many ways the scenario you are positing could represent the end of material scarcity on a human scale.

The very concept of relative wealth depends on the assumption of material scarcity, such that there is only so much consumption and control of material resources to go around, and therefore there is some sort of zero-sum game between the haves and have-nots.

But the fundamental cause of material scarcity has been the scarcity of human labor. The world (or solar system, since we are looking forward now) is full of things like energy, raw materials, and so on--way more than humans could conceivably use. The issue has been obtaining, processing, storing, transporting, trading, and so on such resources, which has historically taken critical contributions from human labor.

And then some people have gotten some of the fruits of other people's labor, and some people have only gotten some of the fruits of their own labor, and that has defined the difference between the wealthy and that not wealthy.

But if we actually do create AI and robots and such who essentially can supply unlimited amounts of labor, then we might well just end material scarcity in general. In which case it will essentially be meaningless to be wealthy, because you having whatever you want to use and enjoy will not in anyway limit my ability to have anything I want to use and enjoy.

All that said, even if that is the final destination, getting from here to there could make for "interesting" times, and who really knows on what schedule. So I would not in fact advise stopping the process of wealth accumulation at this moment in history.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

HomerJ wrote: Fri Sep 24, 2021 9:56 am
NiceUnparticularMan wrote: Fri Sep 24, 2021 9:42 am But if valuations stabilize, and all the rest of these long-term trends remain the same, then these models may stop being wrong. Indefinitely.
Yeah, if all the thousands of variables stop being "variable" and instead become "constant", you may be right. Forever.

:oops:
Well, it is really only a handful of variables, and 10-30 years.

But sure, it would be foolish for anyone to predict any of this with certainty.

My point is more just that what it would take for U.S. real stock returns to go from something like 5% over long periods to something like 2.3% over long periods is nothing terribly dramatic or necessarily unsustainable. Hence why there is no PARTICULAR reason to believe it won't happen. Just the general sort of unpredictability of the future.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

vanbogle59 wrote: Fri Sep 24, 2021 9:59 am
HomerJ wrote: Fri Sep 24, 2021 9:37 am The error bars of these predictions are very large. Like plus/minus 6% or 8%.
As reference, I just asked firecalc what the distribution of returns for 100/0 looked like in its historical data.
Roughly:
Median - 6.5%
Low - 3.2%
High - 10.2%
That's probably as reasonable "prediction" for the next 30 years as anyone could talk me into.
I know lots of people do it, but I never quite understand why someone would assume something as complicated and variable as financial market returns would stay bounded by the range of what has happened in the past so far.

It is a comforting thought, that we have already seen everything that could happen in the future, in approximately the ratios it is likely to happen in the future.

But it just seems obvious to me that such a thought is fundamentally unfounded.
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Long term low returns predicted

Post by vanbogle59 »

NiceUnparticularMan wrote: Fri Sep 24, 2021 10:09 am My point is more just that what it would take for U.S. real stock returns to go from something like 5% over long periods to something like 2.3% over long periods is nothing terribly dramatic or necessarily unsustainable. Hence why there is no PARTICULAR reason to believe it won't happen. Just the general sort of unpredictability of the future.
Isn't the old saying: "You know the economists are lying when they include a decimal point?"

The title of the thread should be modified to "Long term low returns predicted - by the very nobody who knows nothing"
Then all BHs can agree.
:sharebeer
User avatar
HomerJ
Posts: 21246
Joined: Fri Jun 06, 2008 12:50 pm

Re: Long term low returns predicted

Post by HomerJ »

If you look at the max GDP graphs at the Fed (1947-2021), you will indeed see that the good years in the past were higher. Like 5% growth in the good years. And the good years in the past 20 years have been more in the 3% range.

But interestingly, total GDP growth has stayed the same (maybe even accelerated), even with the smaller year-to-year GDP growth.

Because in the old days (1950s and 1960s), we'd get 3-5 years of 5%, then a couple of bad years, then 3-5 years of 5%, then a few bad years.

Since 2001, we've gotten like 7 years of 3%, with a couple years of bad during 2008-2009, and then 12 years of 3%.

https://fred.stlouisfed.org/graph/?g=eUmi (Click on Max above the graph to see 1947-2021)

If you look at a graph of just real total GDP, you'll see a fairly straight line over time, with the last 20 years matching (or even exceeding) the slope of the "high growth" years.

Because we have longer periods of steady growth now, instead of giant jumps back and forth between great growth and no growth.

https://fred.stlouisfed.org/series/GDPC1

I have no idea what this means... Just found the graph interesting. There's more to the story than just saying "GDP growth has dropped from 5% to 3%"
Last edited by HomerJ on Fri Sep 24, 2021 10:29 am, edited 3 times in total.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Long term low returns predicted

Post by vanbogle59 »

NiceUnparticularMan wrote: Fri Sep 24, 2021 10:12 am
vanbogle59 wrote: Fri Sep 24, 2021 9:59 am
HomerJ wrote: Fri Sep 24, 2021 9:37 am The error bars of these predictions are very large. Like plus/minus 6% or 8%.
As reference, I just asked firecalc what the distribution of returns for 100/0 looked like in its historical data.
Roughly:
Median - 6.5%
Low - 3.2%
High - 10.2%
That's probably as reasonable "prediction" for the next 30 years as anyone could talk me into.
I know lots of people do it, but I never quite understand why someone would assume something as complicated and variable as financial market returns would stay bounded by the range of what has happened in the past so far.

It is a comforting thought, that we have already seen everything that could happen in the future, in approximately the ratios it is likely to happen in the future.

But it just seems obvious to me that such a thought is fundamentally unfounded.
OK. I don't disagree.
My only point is that such a prediction was "reasonable".
And, if someone has something else, they have to beat that standard.
User avatar
HomerJ
Posts: 21246
Joined: Fri Jun 06, 2008 12:50 pm

Re: Long term low returns predicted

Post by HomerJ »

NiceUnparticularMan wrote: Fri Sep 24, 2021 10:12 am
vanbogle59 wrote: Fri Sep 24, 2021 9:59 am
HomerJ wrote: Fri Sep 24, 2021 9:37 am The error bars of these predictions are very large. Like plus/minus 6% or 8%.
As reference, I just asked firecalc what the distribution of returns for 100/0 looked like in its historical data.
Roughly:
Median - 6.5%
Low - 3.2%
High - 10.2%
That's probably as reasonable "prediction" for the next 30 years as anyone could talk me into.
I know lots of people do it, but I never quite understand why someone would assume something as complicated and variable as financial market returns would stay bounded by the range of what has happened in the past so far.

It is a comforting thought, that we have already seen everything that could happen in the future, in approximately the ratios it is likely to happen in the future.

But it just seems obvious to me that such a thought is fundamentally unfounded.
Of course, the future doesn't have to match the past exactly. But it seems even more unfounded to say something that has NEVER happened is somewhat as likely as stuff that HAS happened.

And the past includes some pretty crazy bad times, so it seems somewhat safe to use those times as a bottom (But nothing is guaranteed, you're right about that).
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

vanbogle59 wrote: Fri Sep 24, 2021 10:15 am
NiceUnparticularMan wrote: Fri Sep 24, 2021 10:09 am My point is more just that what it would take for U.S. real stock returns to go from something like 5% over long periods to something like 2.3% over long periods is nothing terribly dramatic or necessarily unsustainable. Hence why there is no PARTICULAR reason to believe it won't happen. Just the general sort of unpredictability of the future.
Isn't the old saying: "You know the economists are lying when they include a decimal point?"

The title of the thread should be modified to "Long term low returns predicted - by the very nobody who knows nothing"
Then all BHs can agree.
:sharebeer
So I guess why I keep banging on about this is while some Bogleheads really do seem to open-minded about not knowing what the future will hold, others default to making it a planning assumption that U.S. financial asset returns going forward will look like past returns, at least in the long run.

Being more in the first camp, I definitely cannot say with certainty that returns going forward will NOT look like past returns.

But I also think people should take very seriously the POSSIBILITY they will not.

And not just as some abstraction. I think people should understand what we are seeing in the U.S. in terms of working age population dynamics, labor force participation dynamics, valuations, and so on are not just the same as prior historic conditions. And there is at least a strong possibility that will mean different results for financial assets in the long run going forward.

And I guess I don't accept the right answer to that sort of observation is just to throw up your hands and call it all completely unknowable. It is definitely not predictable with certainty. But changing conditions logically merit reconsidering planning assumptions.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

HomerJ wrote: Fri Sep 24, 2021 10:16 am If you look at the max GDP graphs at the Fed (1947-2021), you will indeed see that the good years in the past were higher. Like 5% growth in the good years. And the good years in the past 20 years have been more in the 3% range.

But interestingly, total GDP growth has stayed the same (maybe even accelerated), even with the smaller year-to-year GDP growth.

Because in the old days (1950s and 1960s), we'd get 3-5 years of 5%, then a couple of bad years, then 3-5 years of 5%, then a few bad years.

Since 2001, we've gotten like 7 years of 3%, with a couple years of bad during 2008-2009, and then 12 years of 3%.

https://fred.stlouisfed.org/graph/?g=eUmi (Click on Max above the graph to see 1947-2021)

If you look at a graph of just real total GDP, you'll see a fairly straight line over time, with the last 20 years matching (or even exceeding) the slope of the "high growth" years.

Because we have longer periods of steady growth now, instead of giant jumps back and forth between great growth and no growth.

https://fred.stlouisfed.org/series/GDPC1

I have no idea what this means... Just found the graph interesting. There's more to the story than just saying "GDP growth has dropped from 5% to 3%"
So for these purposes, it is generally best to look at what is known as potential real GDP growth. The basic point of doing that is to smooth out all the cyclical and event-driven ups and downs, and for that purpose it tends to work pretty well.

And here is what that looks like (this is year over year percentage increases on an annual basis):

Image

And now the structural slowdown is pretty obvious.

OK, now we can turn to things like the CBO's economic forecast (boldly out to 2031!):

https://www.cbo.gov/publication/56965

They also separate out actual from potential GDP growth, and all they are really doing is assuming actual will converge on potential--as it tends to in the long run.

If you look at the figure, and you can also click on the link to the underlying spreadsheet, they have real potential GDP growth starting around 1.9 in 2022, then sliding down to around 1.7 by about 2031.

Again, none of this is very controversial in macroeconomic circles. It is widely know that structural real GDP growth has slowed down in the U.S. in recent decades. It is widely expected that it will continue on at this lower level.

I just wonder if enough personal investors know all this, and have taken due account of what it possibly means.
Last edited by NiceUnparticularMan on Fri Sep 24, 2021 10:44 am, edited 1 time in total.
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Long term low returns predicted

Post by vanbogle59 »

NiceUnparticularMan wrote: Fri Sep 24, 2021 10:32 am So I guess why I keep banging on about this is while some Bogleheads really do seem to open-minded about not knowing what the future will hold, others default to making it a planning assumption that U.S. financial asset returns going forward will look like past returns, at least in the long run.
Wait, you are saying that predicting historically unprecedented low returns is the "open minded" position?
And that it is open minded to do that over an unknowable future lifetime with such precision that you can say something "like 2.3%"?
And that it is open minded to say that without qualifying it with something like "in the range of -2% to +14%" ???

For me, the standard BH position is: equities will provide real growth over long periods, in wide ranges comparable to the last hundred years
Even if I toss in the patriotic "don't bet against the US", I still think I qualify for the open minded merit badge.

Somebody who claims to know otherwise has the burden of proof. And good luck with that.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

vanbogle59 wrote: Fri Sep 24, 2021 10:17 am
NiceUnparticularMan wrote: Fri Sep 24, 2021 10:12 am
vanbogle59 wrote: Fri Sep 24, 2021 9:59 am
HomerJ wrote: Fri Sep 24, 2021 9:37 am The error bars of these predictions are very large. Like plus/minus 6% or 8%.
As reference, I just asked firecalc what the distribution of returns for 100/0 looked like in its historical data.
Roughly:
Median - 6.5%
Low - 3.2%
High - 10.2%
That's probably as reasonable "prediction" for the next 30 years as anyone could talk me into.
I know lots of people do it, but I never quite understand why someone would assume something as complicated and variable as financial market returns would stay bounded by the range of what has happened in the past so far.

It is a comforting thought, that we have already seen everything that could happen in the future, in approximately the ratios it is likely to happen in the future.

But it just seems obvious to me that such a thought is fundamentally unfounded.
OK. I don't disagree.
My only point is that such a prediction was "reasonable".
And, if someone has something else, they have to beat that standard.
Sure. I think we have, though. It is reasonable to do ground up models of the sort we are discussing here, which get you to a prediction of around 2 to 2.5% real return on U.S. stocks in light of current valuations and consensus macroeconomic predictions.
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Long term low returns predicted

Post by vanbogle59 »

NiceUnparticularMan wrote: Fri Sep 24, 2021 10:46 am
vanbogle59 wrote: Fri Sep 24, 2021 10:17 am
NiceUnparticularMan wrote: Fri Sep 24, 2021 10:12 am
vanbogle59 wrote: Fri Sep 24, 2021 9:59 am
HomerJ wrote: Fri Sep 24, 2021 9:37 am The error bars of these predictions are very large. Like plus/minus 6% or 8%.
As reference, I just asked firecalc what the distribution of returns for 100/0 looked like in its historical data.
Roughly:
Median - 6.5%
Low - 3.2%
High - 10.2%
That's probably as reasonable "prediction" for the next 30 years as anyone could talk me into.
I know lots of people do it, but I never quite understand why someone would assume something as complicated and variable as financial market returns would stay bounded by the range of what has happened in the past so far.

It is a comforting thought, that we have already seen everything that could happen in the future, in approximately the ratios it is likely to happen in the future.

But it just seems obvious to me that such a thought is fundamentally unfounded.
OK. I don't disagree.
My only point is that such a prediction was "reasonable".
And, if someone has something else, they have to beat that standard.
Sure. I think we have, though. It is reasonable to do ground up models of the sort we are discussing here, which get you to a prediction of around 2 to 2.5% real return on U.S. stocks in light of current valuations and consensus macroeconomic predictions.
"consensus macroeconomic predictions" - you know these have a pretty horrible track record, right?

I will take the firecalc, stupid, no analysis position: the next 30 years, equities will return between 3% and 10%.
You take yours: the next 30 years, equities will return between 2% and 2.5%

I will qualify mine with nobody knows nothin.
You qualify yours with "consensus macroeconomic predictions".

I'm happy to allow a 3rd party arbitrator to declare who is more open minded.
And I'm happy to allow the future to award the winner.

Best of luck to you.
:sharebeer
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

HomerJ wrote: Fri Sep 24, 2021 10:19 amOf course, the future doesn't have to match the past exactly. But it seems even more unfounded to say something that has NEVER happened is somewhat as likely as stuff that HAS happened.
Well, we do that all the time.

My son was once 4'6". Then he was 4'7". At that point, it was considerably more likely he would be 4'8" then 4'6" in the near future.

Of course this is a trivial example, because we are familiar with human growth trends from observing humans in general. But it is a useful example in the sense we didn't need to observe this particular human previously being in a certain state to know it was possible for that human to enter that state. Not all predictable change is just variation within a previously-established range.
And the past includes some pretty crazy bad times, so it seems somewhat safe to use those times as a bottom (But nothing is guaranteed, you're right about that).
I truly don't think that is even remotely safe. I mean, just as a trivial point, we know from observing other countries that worse things can happen than have happened in the U.S. so far, so already that is a really bad assumption.

But on top of that, all this depends on conditions we KNOW have changed. Valuations are different. Macroeconomic conditions, and the underlying factors that drive them, are different. The whole world is different. When the sorts of factors that CAUSE financial events are out of their historic ranges, such that we have never seen a combination of conditions like this before, it just seems crazy to me to insist the RESULTS will be bounded by what happened under different conditions.

Again, to me this is like a person inside a NOAA hurricane cone for a Cat 5 arguing they can safely disregard an evacuation order on the reasoning that a Cat 5 hasn't hit that area before, and therefore it is unlikely there will really be a storm in that area any worse than has happened before.

It truly would be comforting if the world worked like that, if the future was so easy to predict, at least probabilistically, as you are suggesting.

Regrettably, it is not. And as a result we actually have to keep our eyes open for evidence that things that haven't happened before are getting increasingly likely to happen in the future.
User avatar
Beensabu
Posts: 5618
Joined: Sun Aug 14, 2016 3:22 pm

Re: Long term low returns predicted

Post by Beensabu »

HomerJ wrote: Fri Sep 24, 2021 9:56 am
NiceUnparticularMan wrote: Fri Sep 24, 2021 9:42 am But if valuations stabilize, and all the rest of these long-term trends remain the same, then these models may stop being wrong. Indefinitely.
Yeah, if all the thousands of variables stop being "variable" and instead become "constant", you may be right. Forever.
It's just one possible path. It does no good to ignore the possibility. I've certainly been assuming that valuations would eventually realign with actual growth, but maybe they won't all the way. Maybe there are certain factors preventing a complete correction (i.e. certain factors affecting the speculative element of return). I didn't think about that. Or maybe that realignment will occur slowly over the very long-term. I didn't think about that either.

When you feel like people are overlooking something obvious, and you care, you say something. You know all about that.

I think they're doing a really good job of helping people think about what goes into creating return and why actual return has been different from that forecasted by these models. Not in order to predict anything in particular (their numbers are simply a correction to the numbers provided by others), but to broaden perspective so that people are able to see more possibilities and potential paths.
"The only thing that makes life possible is permanent, intolerable uncertainty; not knowing what comes next." ~Ursula LeGuin
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Long term low returns predicted

Post by vanbogle59 »

NiceUnparticularMan wrote: Fri Sep 24, 2021 11:00 am Again, to me this is like a person inside a NOAA hurricane cone for a Cat 5 arguing they can safely disregard an evacuation order on the reasoning that a Cat 5 hasn't hit that area before, and therefore it is unlikely there will really be a storm in that area any worse than has happened before.
I think you have put your finger on precisely why you and I will never agree.

I think NOAA is actually VERY GOOD at predicting hurricanes.
I think economists (stockpickers, historians, gurus, fortune tellers) are completely useless when it comes to making a 30 year prediction.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

vanbogle59 wrote: Fri Sep 24, 2021 10:44 am
NiceUnparticularMan wrote: Fri Sep 24, 2021 10:32 am So I guess why I keep banging on about this is while some Bogleheads really do seem to open-minded about not knowing what the future will hold, others default to making it a planning assumption that U.S. financial asset returns going forward will look like past returns, at least in the long run.
Wait, you are saying that predicting historically unprecedented low returns is the "open minded" position?
And that it is open minded to do that over an unknowable future lifetime with such precision that you can say something "like 2.3%"?
And that it is open minded to say that without qualifying it with something like "in the range of -2% to +14%" ???
That is very much not what I am saying.

I try to be careful to make clear that what I am saying is given how things have observably changed over time, people should be open-minded to a substantial possibility of unprecedented returns. But "predicting" such returns implies a degree of certainty I would very much not endorse.

But similarly insisting that the possibility of such unprecedented returns is so low as to be negligible for planning purposes, despite knowing relevant conditions have changed, is equally lacking in open-mindedness.

And yes, this is a difficult thing to deal with, a substantial possibility given changed conditions that the future won't be like anything we have seen in the past, without being at all certain about what it will look like. And yet such is life.
For me, the standard BH position is: equities will provide real growth over long periods, in wide ranges comparable to the last hundred years
Well, that second bit is unfortunate in my view. I do think common stocks are more likely than most investable securities to provide real returns in the long run. I don't think it is reasonable to simply assume the range of those returns going forward will be "comparable" to the range of those returns looking backward.

But to the extent you are saying at least many Bogleheads tend to assume that the future will be like the past in the general sense you are describing, I do think that is an accurate description of a common (not universal) attitude here.
Somebody who claims to know otherwise has the burden of proof. And good luck with that.
So that is indeed an easy strawman to defeat.

Less easy to defeat is my actual position, which is that observably changed material conditions outside of their historic ranges can create a substantial possibility that the future returns will not be bounded by their historic ranges either.

That is not a statement that I know what will happen next. It is a statement that I am open to admitting that I know even less than you are claiming to know, meaning I am dropping your "in wide ranges comparable to the last hundred years" assumption.
Last edited by NiceUnparticularMan on Fri Sep 24, 2021 11:22 am, edited 1 time in total.
User avatar
jeffyscott
Posts: 13438
Joined: Tue Feb 27, 2007 8:12 am

Re: Long term low returns predicted

Post by jeffyscott »

HomerJ wrote: Fri Sep 24, 2021 9:30 am
absolute zero wrote: Thu Sep 23, 2021 11:30 pm GMO is pretty impressively bearish. They say -7.5% real CAGR for next 7 years for US equities. If you follow those 7 years with 23 years of 6% real (just a made up assumption) it comes out to 2.5% real over the full 30 year period. I don’t know what GMO thinks will happen over 30 years (nor do I care to be honest) but based off this simple math I would not be surprised if their forecast for even a long time period was in the ballpark of 2% real.
People always say that, but that's never how it works.

It COULD work that way in the future, but, so far, we haven't ever gotten 23 years of "average".

If get -50% over the next 7 years, some of the following years, based on history, will be in 12%-30% range, not just 6%

The long-term average, so far, of 6%-7% real, INCLUDES all the bad years.

So the good years, so far, are very good, not average.

But who knows what the future holds? Not GMO, not me, not you, not anyone. It definitely could be different from past.
No one is saying that the real returns for each year will be 6%, 6%, 6%...6%. (Not sure if that is what you intended to imply there.) The suggestion was a annualized return of 6% real, basically the historical average.

So a ~40-50% decline due to the current high valuations, followed by approximately historically average returns as a median forecast, does not strike me as unreasonable. And no one, by making a forecast of a median for expected return, is saying that they know what the returns will be or that this will be the path that stock prices will follow.

RA seems to be the best at actually publishing their probability distributions and also provides the ability to use two different models, or three if you count the mixed one.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Long term low returns predicted

Post by NiceUnparticularMan »

vanbogle59 wrote: Fri Sep 24, 2021 11:09 am
NiceUnparticularMan wrote: Fri Sep 24, 2021 11:00 am Again, to me this is like a person inside a NOAA hurricane cone for a Cat 5 arguing they can safely disregard an evacuation order on the reasoning that a Cat 5 hasn't hit that area before, and therefore it is unlikely there will really be a storm in that area any worse than has happened before.
I think you have put your finger on precisely why you and I will never agree.

I think NOAA is actually VERY GOOD at predicting hurricanes.
I think economists (stockpickers, historians, gurus, fortune tellers) are completely useless when it comes to making a 30 year prediction.
I mean they don't call economics the dismal science for nothing.

The problem is people who simply assume the future range of financial asset returns will be like the past range of those same asset categories are being "economists"/"fortune tellers" too. That is a form of predictive model. It is also well-known not to be a reliable sort of model.

So relying on that sort of model isn't actually a better way of reasoning about the future than paying attention to other economists and their forecasts. It is just putting oneself into that same role, and then doing a particularly bad version of it.

The problem, of course, is once you have truly adopted a radical (and arguably appropriate) skepticism about economic predictions, INCLUDING as to your own such efforts, how on earth can you possible plan your personal financial life?

So we all have to then just accept doing the best we can, knowing it is all impossible to predict with any sort of certainty.

But to me, doing the best we can is not putting our faith in a predictive model we know is really bad. It involves being open to the possibility that model could easily be wrong, particularly when material conditions are outside of their historical ranges.
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Long term low returns predicted

Post by vanbogle59 »

Are there 2 different people posting under your ID? :D
NiceUnparticularMan wrote: Fri Sep 24, 2021 9:52 am
jeffyscott wrote: Fri Sep 24, 2021 7:23 am Is there a better figure than their 1.2% to use for the expected real growth rate over the next 30 years?
So I came up with a similar number ... I think pretty much any sensible model given current conditions and trends is going to give you an answer in that range.
NiceUnparticularMan wrote: Fri Sep 24, 2021 11:13 am That is not a statement that I know what will happen next. It is a statement that I am open to admitting that I know even less than you are claiming to know

Anyway, trying to find common ground :beer
If your point that there is a non-zero possibility that the next 30 years MIGHT BE worse than any previously recorded 30 year period, we actually agree.
User avatar
HomerJ
Posts: 21246
Joined: Fri Jun 06, 2008 12:50 pm

Re: Long term low returns predicted

Post by HomerJ »

NiceUnparticularMan wrote: Fri Sep 24, 2021 10:43 am
HomerJ wrote: Fri Sep 24, 2021 10:16 am If you look at the max GDP graphs at the Fed (1947-2021), you will indeed see that the good years in the past were higher. Like 5% growth in the good years. And the good years in the past 20 years have been more in the 3% range.

But interestingly, total GDP growth has stayed the same (maybe even accelerated), even with the smaller year-to-year GDP growth.

Because in the old days (1950s and 1960s), we'd get 3-5 years of 5%, then a couple of bad years, then 3-5 years of 5%, then a few bad years.

Since 2001, we've gotten like 7 years of 3%, with a couple years of bad during 2008-2009, and then 12 years of 3%.

https://fred.stlouisfed.org/graph/?g=eUmi (Click on Max above the graph to see 1947-2021)

If you look at a graph of just real total GDP, you'll see a fairly straight line over time, with the last 20 years matching (or even exceeding) the slope of the "high growth" years.

Because we have longer periods of steady growth now, instead of giant jumps back and forth between great growth and no growth.

https://fred.stlouisfed.org/series/GDPC1

I have no idea what this means... Just found the graph interesting. There's more to the story than just saying "GDP growth has dropped from 5% to 3%"
So for these purposes, it is generally best to look at what is known as potential real GDP growth. The basic point of doing that is to smooth out all the cyclical and event-driven ups and downs, and for that purpose it tends to work pretty well.
LOL... "Potential" GDP growth?
Potential GDP is a theoretical construct, an estimate of the value of the output that the economy would have produced if labor and capital had been employed at their maximum sustainable rates—that is, rates that are consistent with steady growth and stable inflation. Figure 1 compares the levels of real GDP and potential output over time. In general, the economy operates close to potential, but deep recessions are notable exceptions to the trend. In these episodes, GDP can lag behind potential, sometimes persistently.
Look at what actually happened. Don't show me fake charts.

I guess you didn't look at the links I posted... Fine, I'll do the work to present them.

So here's how GDP actually changed year to year.. You can still see the lower growth rates in recent years, so your original point was correct, but notice there are a lot more recessions and low GDP years alongside the high GDP years in the 1950s and 1960s and 1970s..

Image

A bunch of high and low years average out about the same as a bunch of medium years.

Look at actual GDP growth over time, and tell me if you can see it slowing down. The slope actually looks to be increasing in later years, not getting slower.

Image

I think the real data is more important than made-up "potential" GDP "data".
Again, none of this is very controversial in macroeconomic circles. It is widely know that structural real GDP growth has slowed down in the U.S. in recent decades. It is widely expected that it will continue on at this lower level.
It is "widely known"? Then explain that Fed chart above where GDP appears to be growing at the same pace as always (if not faster). Draw a line from 1950-2000 and see where it ends up in 2021. About the same spot.

Like I said, slow steady growth equals (or exceeds) the same total growth as the boom and bust cycles of the past.

There may be predictions that GDP total growth will slow down in the future, but it hasn't happened yet.

Economists exist to make astrologers look good.
Last edited by HomerJ on Fri Sep 24, 2021 12:04 pm, edited 1 time in total.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
jeffyscott
Posts: 13438
Joined: Tue Feb 27, 2007 8:12 am

Re: Long term low returns predicted

Post by jeffyscott »

NiceUnparticularMan wrote: Fri Sep 24, 2021 11:21 am
vanbogle59 wrote: Fri Sep 24, 2021 11:09 am
NiceUnparticularMan wrote: Fri Sep 24, 2021 11:00 am Again, to me this is like a person inside a NOAA hurricane cone for a Cat 5 arguing they can safely disregard an evacuation order on the reasoning that a Cat 5 hasn't hit that area before, and therefore it is unlikely there will really be a storm in that area any worse than has happened before.
I think you have put your finger on precisely why you and I will never agree.

I think NOAA is actually VERY GOOD at predicting hurricanes.
I think economists (stockpickers, historians, gurus, fortune tellers) are completely useless when it comes to making a 30 year prediction.
I mean they don't call economics the dismal science for nothing.

The problem is people who simply assume the future range of financial asset returns will be like the past range of those same asset categories are being "economists"/"fortune tellers" too. That is a form of predictive model. It is also well-known not to be a reliable sort of model.

So relying on that sort of model isn't actually a better way of reasoning about the future than paying attention to other economists and their forecasts. It is just putting oneself into that same role, and then doing a particularly bad version of it.

The problem, of course, is once you have truly adopted a radical (and arguably appropriate) skepticism about economic predictions, INCLUDING as to your own such efforts, how on earth can you possible plan your personal financial life?

So we all have to then just accept doing the best we can, knowing it is all impossible to predict with any sort of certainty.

But to me, doing the best we can is not putting our faith in a predictive model we know is really bad. It involves being open to the possibility that model could easily be wrong, particularly when material conditions are outside of their historical ranges.
I would assume that no one is relying on or expecting future bond returns to be similar to historical returns (5-6% :?: ). It seems easier for people to recognize that bonds are at very high valuations and what that means for expected returns, than it is to do so for stocks. Though they may not even realize that that is what they doing, when they estimate that the expected return for the total bond market fund is about equal to the SEC yield (with, of course, a range of possible actual outcomes, but not as wide as that for stocks).
User avatar
jeffyscott
Posts: 13438
Joined: Tue Feb 27, 2007 8:12 am

Re: Long term low returns predicted

Post by jeffyscott »

vanbogle59 wrote: Fri Sep 24, 2021 11:33 am Are there 2 different people posting under your ID? :D
NiceUnparticularMan wrote: Fri Sep 24, 2021 9:52 am
jeffyscott wrote: Fri Sep 24, 2021 7:23 am Is there a better figure than their 1.2% to use for the expected real growth rate over the next 30 years?
So I came up with a similar number ... I think pretty much any sensible model given current conditions and trends is going to give you an answer in that range.
NiceUnparticularMan wrote: Fri Sep 24, 2021 11:13 am That is not a statement that I know what will happen next. It is a statement that I am open to admitting that I know even less than you are claiming to know

Anyway, trying to find common ground :beer
If your point that there is a non-zero possibility that the next 30 years MIGHT BE worse than any previously recorded 30 year period, we actually agree.
I'm not sure about my alter-ego, but the non-zero would be more like a very strong possibility in my case.

I am just going by the RA range, where they have a 25% chance of real returns above 4.2%. Which would still be below historical average. And it looks like they have it at maybe a 10-15% chance of average (~6% real) or above.
rgs92
Posts: 3436
Joined: Mon Mar 02, 2009 7:00 pm

Re: Long term low returns predicted

Post by rgs92 »

1 to 2 percent over inflation is a pretty good return over the long term.
User avatar
HomerJ
Posts: 21246
Joined: Fri Jun 06, 2008 12:50 pm

Re: Long term low returns predicted

Post by HomerJ »

jeffyscott wrote: Fri Sep 24, 2021 11:45 am I am just going by the RA range, where they have a 25% chance of real returns above 4.2%. Which would still be below historical average. And it looks like they have it at maybe a 10-15% chance of average (~6% real) or above.
What does RA stand for again?

What did they predict in 2011?
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Long term low returns predicted

Post by vanbogle59 »

jeffyscott wrote: Fri Sep 24, 2021 11:41 am I would assume that no one is relying on or expecting future bond returns to be similar to historical returns (5-6% :?: ). It seems easier for people to recognize that bonds are at very high valuations and what that means for expected returns, than it is to do so for stocks. Though they may not even realize that that is what they doing, when they estimate that the expected return for the total bond market fund is about equal to the SEC yield (with, of course, a range of possible actual outcomes, but not as wide as that for stocks).
historical returns (5-6% :?: ) :confused

I just played the firecalc game again.
This time 0/100 for 30 years
Low: -1.5%
Median: 2%
High: 6.2%

Seems like a "reasonable" place to start to me :beer
stocknoob4111
Posts: 3495
Joined: Sun Jan 07, 2018 11:52 am

Re: Long term low returns predicted

Post by stocknoob4111 »

I was curious to see how Vanguard did with their predictions for 2010-2019. I found this source: https://retirementincomejournal.com/art ... e-returns/

Asset class: US Equities
VG Prediction: 6% Real
Actual: 11.34% Real

Asset class: Bonds
VG Prediction: 0-2% Real
Actual: 1.78% Real

While they nailed the Bond returns they were horribly wrong about Equities which performed twice as good as their projections which essentially renders their projection useless.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Long term low returns predicted

Post by willthrill81 »

stocknoob4111 wrote: Fri Sep 24, 2021 12:21 pm I was curious to see how Vanguard did with their predictions for 2010-2019. I found this source: https://retirementincomejournal.com/art ... e-returns/

Asset class: US Equities
VG Prediction: 6% Real
Actual: 11.34% Real

Asset class: Bonds
VG Prediction: 0-2% Real
Actual: 1.78% Real

While they nailed the Bond returns they were horribly wrong about Equities which performed twice as good as their projections which essentially renders their projection useless.
Bond performance is much easier to predict because starting yields have been highly predictive of future returns. That's why everyone should be expecting bonds to return somewhere around -2% - 0% real over the next decade.

But yes, their prediction of U.S. stock performance was far from actual.
The Sensible Steward
Post Reply