Over what time period should stocks NOT lose money

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
SpartanBull
Posts: 127
Joined: Fri Jun 06, 2014 12:31 am

Over what time period should stocks NOT lose money

Post by SpartanBull » Sun Jan 08, 2017 9:25 pm

Hello,
While there are no certainties, bogleheads clearly believe that over time, staying invested is a prudent investment. Time is your friend. If forced to answer "over what time period will stocks NOT lose money....in real terms, what would you say?". If forced to give an answer on this I would say 10 years. I'm curious what others think.

User avatar
Phineas J. Whoopee
Posts: 6598
Joined: Sun Dec 18, 2011 6:18 pm

Re: Over what time period should stocks NOT lose money

Post by Phineas J. Whoopee » Sun Jan 08, 2017 9:32 pm

The subject reads should, and the body reads will.

I don't know there is any period of time over which we can be confident stocks won't lose money. As for should, that's not for me to say.

PJW
Last edited by Phineas J. Whoopee on Mon Jan 09, 2017 12:13 am, edited 1 time in total.

User avatar
arcticpineapplecorp.
Posts: 2198
Joined: Tue Mar 06, 2012 9:22 pm

Re: Over what time period should stocks NOT lose money

Post by arcticpineapplecorp. » Sun Jan 08, 2017 9:35 pm

The link below is a good response from Nisprius. I believe the answer was it took around 14 years, not 25 to recover from the Great Depression. But that's based on a lot of "ifs" which Nisiprius points out. And the next Great Depression might look different from the last one. So the answer is nobody really knows. It's been said that the market can stay down longer than you can stay solvent. And therein lies the rub.

You might want to read the link below and other similar threads already on this site (I did a simple search using the google search bar at the top right to find the link below):

viewtopic.php?p=368850#p368850
"Invest we must." -- Jack Bogle | “The purpose of investing is not to simply optimise returns and make yourself rich. The purpose is not to die poor.” -- William Bernstein

qwertyjazz
Posts: 1026
Joined: Tue Feb 23, 2016 4:24 am

Re: Over what time period should stocks NOT lose money

Post by qwertyjazz » Sun Jan 08, 2017 9:39 pm

30 years in the modern era
viewtopic.php?t=120819

Now idea though when my Tulip stocks will rebound though / or more recently my Japanese index fund from the 80's

So if forced to answer 30 years - but my personal answer is I have no idea if it will be positive but I hate it less than all other saving tools
G.E. Box "All models are wrong, but some are useful."

qwertyjazz
Posts: 1026
Joined: Tue Feb 23, 2016 4:24 am

Re: Over what time period should stocks NOT lose money

Post by qwertyjazz » Sun Jan 08, 2017 9:41 pm

Phineas J. Whoopee wrote:The subject reads should, and the body reads will.

I don't there is any period of time over which we can be confident stocks won't lose money. As for should, that's not for me to say.

PJW

+1
I will change my answer to one day then. I should not have to deal with volatility - at least my doctor says it is bad for my indigestion or agita
G.E. Box "All models are wrong, but some are useful."

AlohaJoe
Posts: 2431
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Over what time period should stocks NOT lose money

Post by AlohaJoe » Sun Jan 08, 2017 9:45 pm

SpartanBull wrote:If forced to answer "over what time period will stocks NOT lose money....in real terms, what would you say?". If forced to give an answer on this I would say 10 years. I'm curious what others think.


Here's the actual data for both stocks & bonds:

Image

KlangFool
Posts: 6761
Joined: Sat Oct 11, 2008 12:35 pm

Re: Over what time period should stocks NOT lose money

Post by KlangFool » Sun Jan 08, 2017 9:46 pm

OP,

A) At any point of time, your equity portion could lose 50% of its value.

With (A) in mind, please tell me what do you mean by stock will not lose money?

i) Not lose money in REAL term?

ii) Not lose money in a nominal term?

KlangFool

SpartanBull
Posts: 127
Joined: Fri Jun 06, 2014 12:31 am

Re: Over what time period should stocks NOT lose money

Post by SpartanBull » Sun Jan 08, 2017 10:53 pm

KlangFool wrote:OP,

A) At any point of time, your equity portion could lose 50% of its value.

With (A) in mind, please tell me what do you mean by stock will not lose money?

i) Not lose money in REAL term?

ii) Not lose money in a nominal term?

KlangFool


Klangfool,
Real return. The question was over what period is it reasonable to assume that you will have a real return that is not below 0%.

lack_ey
Posts: 5628
Joined: Wed Nov 19, 2014 11:55 pm

Re: Over what time period should stocks NOT lose money

Post by lack_ey » Sun Jan 08, 2017 11:00 pm

With what kind of probability? 100%?

User avatar
grayfox
Posts: 4555
Joined: Sat Sep 15, 2007 4:30 am
Location: Earth

Re: Over what time period should stocks NOT lose money

Post by grayfox » Sun Jan 08, 2017 11:05 pm

SpartanBull wrote:Hello,
While there are no certainties, bogleheads clearly believe that over time, staying invested is a prudent investment. Time is your friend. If forced to answer "over what time period will stocks NOT lose money....in real terms, what would you say?". If forced to give an answer on this I would say 10 years. I'm curious what others think.


I considered this question back in 2013. Rather than using historical data, I used a stochastic model. Returns were modeled as a Constant Expected Return (CER) Model, i.e. the Mean Return μ and Standard Deviation σ were constant. Here are some results:

Code: Select all

(μ,    σ)        5% chance of loss at
(0.10, 0.20)   130 months (10.8 years)
(0.05, 0.10)   130 months (10.8 years)   <- period is determined by the ratio σ / μ
(0.07, 0.20)   266 months (22.2 years)   <- smaller μ lengthens period
(0.10, 0.10)    33 months (2.75 years)   <- smaller σ shortens period

                 1% chance of loss at
(0.10, 0.20)   266 months (22 years)


There were other interesting results, like how the max drawdown varied with μ and σ.
Also, the worst drawdown was typically at about 5 years. After that, things improved.
Last edited by grayfox on Sun Jan 08, 2017 11:06 pm, edited 1 time in total.
Gott mit uns.

User avatar
White Coat Investor
Posts: 12804
Joined: Fri Mar 02, 2007 9:11 pm
Location: Greatest Snow On Earth

Re: Over what time period should stocks NOT lose money

Post by White Coat Investor » Sun Jan 08, 2017 11:06 pm

I think 10 years is a pretty safe bet, but the odds certainly aren't zero. That's one good reason to hold bonds besides volatility issues. It is possible for bonds to outperform stocks for your entire investment horizon.
1) Invest you must 2) Time is your friend 3) Impulse is your enemy | 4) Basic arithmetic works 5) Stick to simplicity 6) Stay the course

User avatar
tludwig23
Posts: 1652
Joined: Thu Dec 30, 2010 3:27 pm
Location: 48deg46"23"N 122deg28'21"W

Re: Over what time period should stocks NOT lose money

Post by tludwig23 » Sun Jan 08, 2017 11:10 pm

KlangFool wrote:
A) At any point of time, your equity portion could lose 50% of its value.


Then logically at any point in time your equity portion could lose all of its value.
That's what I do: I drink, and I know things. --Tyrion Lannister

User avatar
raven15
Posts: 290
Joined: Sun Nov 30, 2014 8:01 pm

Re: Over what time period should stocks NOT lose money

Post by raven15 » Sun Jan 08, 2017 11:12 pm

It also matters a lot if you are contributing, withdrawing, or doing nothing. This is a point which is often ignored but is important. If you are dollar cost averaging into the market as a result of your career your "duration" is effectively much shorter, and I doubt you would see more than 5-10 years of negative returns depending on the value of contributions relative to existing investments. If you are doing nothing then the usual answer of "probably not more than 20 years" applies. If you are withdrawing then you may very well require as long as 50 years to get back up to real starting principal, depending on your rate of withdrawal. If you ever do get back up to the real starting amount. All of the above assume global diversification.
It's Time. Adding Interest.

grok87
Posts: 7516
Joined: Tue Feb 27, 2007 9:00 pm

Re: Over what time period should stocks NOT lose money

Post by grok87 » Sun Jan 08, 2017 11:51 pm

SpartanBull wrote:Hello,
While there are no certainties, bogleheads clearly believe that over time, staying invested is a prudent investment. Time is your friend. If forced to answer "over what time period will stocks NOT lose money....in real terms, what would you say?". If forced to give an answer on this I would say 10 years. I'm curious what others think.

Perhaps you will allow me to rephrase your question?

"Over what time period would it be prudent to invest in a long term equity oriented portfolio of 55% stocks: 15% real estate: 15% treasuries: 15% tips?

David Swensen's answer is 10 years [as per his book "unconventionAl success"]
"...people always live for ever when there is any annuity to be paid them"- Jane Austen

User avatar
saltycaper
Posts: 2091
Joined: Thu Apr 24, 2014 8:47 pm
Location: The Tower

Re: Over what time period should stocks NOT lose money

Post by saltycaper » Mon Jan 09, 2017 1:03 am

tludwig23 wrote:
KlangFool wrote:
A) At any point of time, your equity portion could lose 50% of its value.


Then logically at any point in time your equity portion could lose all of its value.


Undoubtedly true, if we're allowing for any probability greater than zero.
"I guess I should warn you, if I turn out to be particularly clear, you've probably misunderstood what I've said." --Alan Greenspan

lgs88
Posts: 210
Joined: Tue Jun 07, 2016 7:48 am

Re: Over what time period should stocks NOT lose money

Post by lgs88 » Mon Jan 09, 2017 6:48 am

Image

I saw that image bandied about on these forums a couple of months ago. Ostensibly, it is a Monte Carlo simulation of the S&P 500 since 1871. I confess I'm a little leery of it, since that particular index has not been around nearly that long. I'd appreciate it if another forum member could common-sense check it for me.
merely an interested amateur

User avatar
happyisland
Posts: 378
Joined: Thu Oct 03, 2013 1:36 pm

Re: Over what time period should stocks NOT lose money

Post by happyisland » Mon Jan 09, 2017 7:11 am

qwertyjazz wrote:
Now idea though when my Tulip stocks will rebound though


Man, I would love to see that a version of that Nisi post above, but with Tulip stocks during the bubble instead of stocks in the Great Depression. I wonder how long their dividend- and inflation-adjusted returns took to make it back to break even?

Levett
Posts: 4155
Joined: Fri Feb 23, 2007 2:10 pm
Location: upper Midwest

Re: Over what time period should stocks NOT lose money

Post by Levett » Mon Jan 09, 2017 7:33 am

This part of the OPs statement I agree with:

"staying invested is a prudent investment"

How you allocate, and on what basis, that's on you (or us, as the case may be).

Lev

delrinson
Posts: 90
Joined: Tue Sep 06, 2016 11:46 am

Re: Over what time period should stocks NOT lose money

Post by delrinson » Mon Jan 09, 2017 8:21 am

I was playing around with some data the other day...looking at recoveries in the S&P 500 from bear markets with dividends always being reinvested. The one scenario that was most sobering is that from the peak in 2000 to the end of 2016 the annualized return was 4.x%....about what I'd expect from a bond ETF.

Other scenarios (starting at a market peak, always reinvesting dividends, and seeing how long it takes to realize a certain annualized return) were much better.

User avatar
rustymutt
Posts: 3632
Joined: Sat Mar 07, 2009 12:03 pm
Location: Oklahoma

Re: Over what time period should stocks NOT lose money

Post by rustymutt » Mon Jan 09, 2017 9:07 am

You should be ready to loose money anytime you're in the markets. There is a right and wrong way to invest. Slow DCA into a plan or investment, or lump sum in, you've got to be in the market, to get market returns, and yet there is academic proof that based on when you get into a plan, or timing that's playing a huge factor in results. There is some faith required of honest investors. Nobody knows the answer to your question. You should read up on books listed on this forum. Learn about investing from those whose done it, and are doing it. :sharebeer
Knowledge is knowing that the Tomato is a fruit. Wisdom is knowing better than to put the tomato in a fruit salad.

Quark
Posts: 1045
Joined: Sun Nov 01, 2015 5:32 pm

Re: Over what time period should stocks NOT lose money

Post by Quark » Mon Jan 09, 2017 9:19 am

Are you assuming the past 90 years of US market data are a reliable guide to the future?

KlangFool
Posts: 6761
Joined: Sat Oct 11, 2008 12:35 pm

Re: Over what time period should stocks NOT lose money

Post by KlangFool » Mon Jan 09, 2017 9:45 am

saltycaper wrote:
tludwig23 wrote:
KlangFool wrote:
A) At any point of time, your equity portion could lose 50% of its value.


Then logically at any point in time your equity portion could lose all of its value.


Undoubtedly true, if we're allowing for any probability greater than zero.


You are correct.

KlangFool

KlangFool
Posts: 6761
Joined: Sat Oct 11, 2008 12:35 pm

Re: Over what time period should stocks NOT lose money

Post by KlangFool » Mon Jan 09, 2017 9:51 am

SpartanBull wrote:
KlangFool wrote:OP,

A) At any point of time, your equity portion could lose 50% of its value.

With (A) in mind, please tell me what do you mean by stock will not lose money?

i) Not lose money in REAL term?

ii) Not lose money in a nominal term?

KlangFool


Klangfool,
Real return. The question was over what period is it reasonable to assume that you will have a real return that is not below 0%.


SpartanBull,

In the context of personal finance, a typical person has 20 to 30 years to accumulate and 20 to 30 years to spend. If a person is unlucky enough that when he/she start to spend, the stock went down by 50% a few times, the real return will be below 0%. So, for a 100% stock portfolio, in term of personal finance, over the 20 to 30 years time period, the real return could be below 0%.

In term of personal finance again, we are not a statistic. We need to take a micro-view of the world. The average does not work for us. We only care about the specific return sequence that affects us.

KlangFool

User avatar
nisiprius
Advisory Board
Posts: 33944
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: Over what time period should stocks NOT lose money

Post by nisiprius » Mon Jan 09, 2017 9:57 am

SpartanBull wrote:Hello,
While there are no certainties, bogleheads clearly believe that over time, staying invested is a prudent investment. Time is your friend. If forced to answer "over what time period will stocks NOT lose money....in real terms, what would you say?". If forced to give an answer on this I would say 10 years. I'm curious what others think.
Stocks have lost money over the following 10-year periods:

All figures "real" (inflation-adjusted), and including reinvested dividends.
Year range, inclusive; CAGR; total cumulative percentage loss

1965-1974 -31.86% -3.76%
1966-1975 -20.82% -2.31%
1968-1977 -22.33% -2.49%
1969-1978 -28.42% -3.29%
1970-1979 -13.23% -1.41%
1972-1981 -18.17% -1.99%
1973-1982 -16.87% -1.83%
1999-2008 -32.17% -3.81%
2000-2009 -29.17% -3.39%
2001-2010 -8.66% -0.90%
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.

dbr
Posts: 23391
Joined: Sun Mar 04, 2007 9:50 am

Re: Over what time period should stocks NOT lose money

Post by dbr » Mon Jan 09, 2017 9:59 am

lgs88 wrote:Image

I saw that image bandied about on these forums a couple of months ago. Ostensibly, it is a Monte Carlo simulation of the S&P 500 since 1871. I confess I'm a little leery of it, since that particular index has not been around nearly that long. I'd appreciate it if another forum member could common-sense check it for me.


I would say that picture is a pretty good representation of how this works. I would not use that picture to extract an absolute statement such as "There is no possibility of losing money if held for more than 20.88 years." or something. One thing about this kind of analysis is that there are no error bands on the result. To get that you have to run a sensitivity analysis on the input assumptions. That is also equivalent to considering that investment probabilities for the next hundred years will not come from the same statistics we have been sampling the last hundred years or more. One needs to assume some statistics for what the differences might be.

The basic question "Over what time period . . ." has to be considered as a probabilistic question and has no answer in definite numbers.

User avatar
Aptenodytes
Posts: 3741
Joined: Tue Feb 08, 2011 8:39 pm

Re: Over what time period should stocks NOT lose money

Post by Aptenodytes » Mon Jan 09, 2017 10:02 am

I think the OP's question is a very useful way to frame risk -- it has much more intuitive punch than things like standard deviations and it relates very directly to an aspect of the portfolio that has real decision-relevance. It also lends itself well to visualization.

Below is a chart made from the 1871-1916 S&P data in the Simba spreadsheet -- converted to real returns using the inflation data also contained in the spreadsheet and then made into total returns for trailing 10, 15 and 20 year periods. The 10 and 15 year periods each dipped negative, but the 20 never did. [edited to clarify that graph has misleading label -- it charts "return in percentage" not total dollars. Sorry -- moving too fast"]
Image
Last edited by Aptenodytes on Mon Jan 09, 2017 10:05 am, edited 1 time in total.

dbr
Posts: 23391
Joined: Sun Mar 04, 2007 9:50 am

Re: Over what time period should stocks NOT lose money

Post by dbr » Mon Jan 09, 2017 10:04 am

I like the above plot because it shows how dependent the result is on when one starts. That is a reason I like looking at those retirement outcomes in FireCalc as well for a demonstration of this phenomenon.

User avatar
nisiprius
Advisory Board
Posts: 33944
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: Over what time period should stocks NOT lose money

Post by nisiprius » Mon Jan 09, 2017 10:04 am

I explored this a bit in Does Siegel ever define "long run?" (Jeremy Siegel, author of the influential book Stocks for the Long Run.)

In particular, not Siegel but Peter Bernstein, author of the introduction to the third edition:
we must keep in mind that Professor Siegel did not lightly choose Stocks for the Long Run as the title of his book. The operative number is 20. Volatility of returns is high in periods of less than 20 years.
However, Siegel never really says. There is some internal evidence that he was thinking "30 years." For example:
Never in the past 150 years has the buyer of a newly-issued 30-year government bond who held it to maturity achieved greater gains than an investor who held a diversified portfolio of common stocks over the same period.
In the 4th edition, he makes a similar statement. In the 5th, however, he writes:
In the first four editions of Stocks for the Long Run, I noted that the last 30-year period when the return on long-term bonds beat stocks ended in 1861, at the onset of the Civil War. That is no longer true. Because of the large drop in government bond yields over the past decade, the 11.03 percent annual returns on long-term government bonds surpassed the 10.98 percent on stocks for the 30-year period from January 1, 1982, through the end of 2011.
Finally, one really must not just ignore Pastor and Stambaugh:

Are Stocks Really Less Volatile in the Long Run?
According to conventional wisdom, annualized volatility of stock returns is lower over long horizons than over short horizons, due to mean reversion induced by return predictability. In contrast, we find that stocks are substantially more volatile over long horizons from an investor's perspective. This perspective recognizes that parameters are uncertain, even with two centuries of data, and that observable predictors imperfectly deliver the conditional expected return. Mean reversion contributes strongly to reducing long-horizon variance, but it is more than offset by various uncertainties faced by the investor, especially uncertainty about the expected return. The same uncertainties reduce desired stock allocations of long-horizon investors contemplating target-date funds.
That's the elephant in the room. Nobody ever talks about it.
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.

User avatar
rustymutt
Posts: 3632
Joined: Sat Mar 07, 2009 12:03 pm
Location: Oklahoma

Re: Over what time period should stocks NOT lose money

Post by rustymutt » Mon Jan 09, 2017 10:49 am

nisiprius wrote:
SpartanBull wrote:Hello,
While there are no certainties, bogleheads clearly believe that over time, staying invested is a prudent investment. Time is your friend. If forced to answer "over what time period will stocks NOT lose money....in real terms, what would you say?". If forced to give an answer on this I would say 10 years. I'm curious what others think.
Stocks have lost money over the following 10-year periods:

All figures "real" (inflation-adjusted), and including reinvested dividends.
Year range, inclusive; CAGR; total cumulative percentage loss

1965-1974 -31.86% -3.76%
1966-1975 -20.82% -2.31%
1968-1977 -22.33% -2.49%
1969-1978 -28.42% -3.29%
1970-1979 -13.23% -1.41%
1972-1981 -18.17% -1.99%
1973-1982 -16.87% -1.83%
1999-2008 -32.17% -3.81%
2000-2009 -29.17% -3.39%
2001-2010 -8.66% -0.90%



I find the charts showing results can be manipulated by cherry picking periods of time. Shifting simply 1 year at times can result in positive growth or negative depending on time periods. Not claiming that about you post figures nisiprius, but just in general studies. So when you get in, is as important, of factor, as most. That is if your IPO has you all out of stocks at some time. Mine can't. That's why the longer you buy and hold, the risk associated with this factor start to diminish over time. Makes even catastrophic market changes more bearable for some.
Knowledge is knowing that the Tomato is a fruit. Wisdom is knowing better than to put the tomato in a fruit salad.

KlangFool
Posts: 6761
Joined: Sat Oct 11, 2008 12:35 pm

Re: Over what time period should stocks NOT lose money

Post by KlangFool » Mon Jan 09, 2017 11:08 am

nisiprius wrote:
Are Stocks Really Less Volatile in the Long Run?
According to conventional wisdom, annualized volatility of stock returns is lower over long horizons than over short horizons, due to mean reversion induced by return predictability. In contrast, we find that stocks are substantially more volatile over long horizons from an investor's perspective. This perspective recognizes that parameters are uncertain, even with two centuries of data, and that observable predictors imperfectly deliver the conditional expected return. Mean reversion contributes strongly to reducing long-horizon variance, but it is more than offset by various uncertainties faced by the investor, especially uncertainty about the expected return. The same uncertainties reduce desired stock allocations of long-horizon investors contemplating target-date funds.
That's the elephant in the room. Nobody ever talks about it.


nisiprius,

+ 1,000,000

<< Mean reversion contributes strongly to reducing long-horizon variance, but it is more than offset by various uncertainties faced by the investor>>

<<That's the elephant in the room. Nobody ever talks about it.>>

Let's assume that stock in X number of years has a positive return. But, that fact is meaningless and useless for an individual investor. The reason is as an individual investor, the person cannot guarantee that he/she can stay invested for X number of years. Life happened.

KlangFool

691175002
Posts: 352
Joined: Tue Mar 17, 2015 7:22 pm

Re: Over what time period should stocks NOT lose money

Post by 691175002 » Mon Jan 09, 2017 11:12 am

FWIW I think that this question should be answered in nominal terms (or against riskfree). It doesn't make sense to calculate it in real terms because breakeven is a psychological concept, and a person asking this question is likely deciding between investing in stocks or cash - not some hypothetical inflation tracking security.

I would also argue that it is highly unlikely that an investors required withdrawals will increase during an extended market drawdown* regardless of what CPI says. Inflation has been concentrated in goods like education and housing, a retiree who relies on their investments for income will not be purchasing those items.

*This of course ignores hyperinflation or other catastrophic events that will ruin you regardless of asset allocation.

grok87
Posts: 7516
Joined: Tue Feb 27, 2007 9:00 pm

Re: Over what time period should stocks NOT lose money

Post by grok87 » Mon Jan 09, 2017 11:33 am

nisiprius wrote:
SpartanBull wrote:Hello,
While there are no certainties, bogleheads clearly believe that over time, staying invested is a prudent investment. Time is your friend. If forced to answer "over what time period will stocks NOT lose money....in real terms, what would you say?". If forced to give an answer on this I would say 10 years. I'm curious what others think.
Stocks have lost money over the following 10-year periods:

All figures "real" (inflation-adjusted), and including reinvested dividends.
Year range, inclusive; CAGR; total cumulative percentage loss

1965-1974 -31.86% -3.76%
1966-1975 -20.82% -2.31%
1968-1977 -22.33% -2.49%
1969-1978 -28.42% -3.29%
1970-1979 -13.23% -1.41%
1972-1981 -18.17% -1.99%
1973-1982 -16.87% -1.83%
1999-2008 -32.17% -3.81%
2000-2009 -29.17% -3.39%
2001-2010 -8.66% -0.90%

Thanks. Very sobering. Looking at 65-74, 73-82 and 99-08
"...people always live for ever when there is any annuity to be paid them"- Jane Austen

Clive
Posts: 1951
Joined: Sat Jun 13, 2009 5:49 am

Re: Over what time period should stocks NOT lose money

Post by Clive » Mon Jan 09, 2017 1:00 pm

World, equal weight, around 20 years
Image
If you concentrate the chances are you'll lag the equal weight average.

That's fractal. And a major reason IMO why cap weighted has a tendency to lag (which is more commonly highlighted as being equal weighted/smaller cap ...etc 'rewarding more').

That's total returns, adjusting for inflation. Add costs, taxes and withdrawals and the odds worsen. Also less if you don't ignore survivorship (for example that table doesn't include Russia or China). The indications are however that survivorship bias can be relatively small (negligible)

http://www.cfapubs.org/doi/pdf/10.2470/rf.v2011.n4.5 page 42 (pdf page 11)
For the World index, our estimate of the annualized historical equity premium relative to bills is 4.5 percent. This estimate is based on the 19 countries in the DMS database, all of which survived from 1900 to 2011. These 19 countries accounted for an estimated 89 percent of the world equity market in 1900. The remaining 11 percent came from markets that existed in 1900 but for which we have been unable to obtain data. Some of these omitted markets failed to survive, and in cases like Russia in 1917 and China in 1949, investors lost all of their money. To quantify the maximum possible impact of omitted markets on the magnitude of the historical equity risk premium, we make an extreme assumption. We assume that all omitted markets became valueless and that this outcome occurred for every omitted country in a single disastrous year, rather than building up gradually. We then ask what risk premium investors would have earned if in 1900, they had purchased a holding in the entire World market, including countries omitted from the DMS database, and held this portfolio for 111 years. At the start of the period, their portfolio would have comprised an 89 percent holding in the DMS World index and an 11 percent holding in countries that we have assumed were all destined to become valueless.

Given these extreme assumptions, we demonstrate (see Dimson, Marsh, and Staunton 2008) that survivorship bias could, at most, give rise to an overstatement of the geometric mean risk premium on the World equity index by about one-tenth of a percentage point. If omitted markets did not all become valueless—and we know that very many did not—the magnitude of survivorship bias would be smaller still. Although debate continues about the precise impact of the bias because some, but not all, of these equity markets experienced a total loss of value, the net impact on the worldwide geometric mean equity premium is no more than 0.1 percent. The effect on the arithmetic mean is similar. The intuition involves the disappearance of 11 percent of the value of the market over 111 years, which represents a loss of value averaging 0.1 percent per year. We conclude that survivorship bias in world stock market returns is negligible.

User avatar
grayfox
Posts: 4555
Joined: Sat Sep 15, 2007 4:30 am
Location: Earth

Re: Over what time period should stocks NOT lose money

Post by grayfox » Mon Jan 09, 2017 1:56 pm

lgs88 wrote:Image

I saw that image bandied about on these forums a couple of months ago. Ostensibly, it is a Monte Carlo simulation of the S&P 500 since 1871. I confess I'm a little leery of it, since that particular index has not been around nearly that long. I'd appreciate it if another forum member could common-sense check it for me.


That looks reasonable and about matches my simple mental model for chance of nominal loss over various holding periods:

33% chance of loss over 1 year
20% chance of loss over 3 years
10% chance of loss over 5 years
5% chance of loss over 10 years
1% chance of loss over 20 years
0.25% chance of loss over 30 years.

That is barring a major calamity like:
1. War that destroys much of the nations infrastructure - some places 2-3 per century.
2. Nuclear War
3. Communist takeover
4. Hyperinflation - 2 per century?
5. Great Depression - 1 per century?
6. Ice Age - every 100,000 years?
6. Asteroid strike - 1 per 50,000 years?
7. Massive Extinction event - 1 per 26 million years?
8. Zombie Apocalypse, etc

But its difficult to assign a probability to those kind of rare events.
What's the Probability of Nuclear War over the next 20 years? .1%, 1%, 10%, 100%? Who knows?

BTW, that is probably using Shiller's Data which might be called a reconstructed S&P Composite.
Last edited by grayfox on Mon Jan 09, 2017 2:49 pm, edited 1 time in total.
Gott mit uns.

taguscove
Posts: 101
Joined: Sun Dec 11, 2016 6:49 pm
Location: Boston, MA

Re: Over what time period should stocks NOT lose money

Post by taguscove » Mon Jan 09, 2017 2:23 pm

The answers above are solid based on a Monte Carlo approach on historic US equity returns. 10 years, including dividends, is about the longest drawdown period for the USA.

There is a huge survivorship bias though, that the USA was a dominant successful country in the 20-21st century. Germany, Argentina, UK, France, and China are prominent examples of major markets that experienced major disruptions from WW1 and WW2. The most recent hundred years had the biggest increase in human prosperity in the history of humanity.

betting on a diversified portfolio of stocks is inherently a bet on the continued prosperity of humanity. For the most recent hundred years, that has definitely been a good bet. Going forward is anyone's guess, and I would be very wary of anyone who seems too confident. A diversified equity/bond portfolio is a solid default position.

My personal opinion is that people are generally too complacent towards paradigm shifts. In these paradigm shifts, water, food, shelter, medicine, weapons, gold, and cash were what historically paid off. There's a very good chance that financial markets will freeze up when a true disaster occurs. Mostly equity & bonds, with a small position in the essentials is my recommendation.

Dirghatamas
Posts: 506
Joined: Fri Jan 01, 2016 6:18 pm

Re: Over what time period should stocks NOT lose money

Post by Dirghatamas » Mon Jan 09, 2017 4:23 pm

taguscove wrote:betting on a diversified portfolio of stocks is inherently a bet on the continued prosperity of humanity. For the most recent hundred years, that has definitely been a good bet. Going forward is anyone's guess, and I would be very wary of anyone who seems too confident. A diversified equity/bond portfolio is a solid default position.


I have heard this argument on this board a lot but folks seem to have some cognitive dissonance about bonds vs. stocks.

First, I agree that the future is unknowable, so it might very well happen that the economic engine of the world (stocks) doesn't make progress. The second part is whether in that case, bonds will help you out? That's the part which has never made sense to me.

Consider all the countries folks give examples of where war/revolution etc. destroyed stocks e.g. China, Russia, Japan, Germany. Guess what..bond holders got a big fat zero after that for their investments. So, if the argument is that bonds are safer than stocks due to major catastrophies like revolution or war in a given country, that is clearly not true, based on historical data.

If the argument is that we should instead be globally diversified to avoid concentrated revolution/war etc risk in any given country, I clearly agree. So, then the question becomes is a global weighted stock portfolio more risky in the long term than a globally diversified stocks AND bonds portfolio.

Consider that most bonds (by value) in the world are nominal bonds not inflation protected. Most central banks have policy guidelines of ~2% inflation. Most/all country governments and regional governments have large debts. It is in their interest to create inflation over the long term (because it reduces the real debt they have to pay through taxation). So, they have the means and the desire. Getting good returns from bonds long term given this demographics and debt burden seems a long shot to me.

If you then ignore Governmnet Bonds and focus on Corporate, again the same argument. Companies are by law responsible to get returns for their owners (stock shareholders). On average if corporate bonds are a cheaper investment than equity, then companies are doing a bad job. On average, company stocks should return better than company bonds.

I would like folks to argue under which financial case would they predict long term global bonds to return better than global stocks. I just don't see it.

taguscove
Posts: 101
Joined: Sun Dec 11, 2016 6:49 pm
Location: Boston, MA

Re: Over what time period should stocks NOT lose money

Post by taguscove » Mon Jan 09, 2017 6:20 pm

Dirghatamas wrote:
taguscove wrote:betting on a diversified portfolio of stocks is inherently a bet on the continued prosperity of humanity. For the most recent hundred years, that has definitely been a good bet. Going forward is anyone's guess, and I would be very wary of anyone who seems too confident. A diversified equity/bond portfolio is a solid default position.


I have heard this argument on this board a lot but folks seem to have some cognitive dissonance about bonds vs. stocks.

First, I agree that the future is unknowable, so it might very well happen that the economic engine of the world (stocks) doesn't make progress. The second part is whether in that case, bonds will help you out? That's the part which has never made sense to me.

Consider all the countries folks give examples of where war/revolution etc. destroyed stocks e.g. China, Russia, Japan, Germany. Guess what..bond holders got a big fat zero after that for their investments. So, if the argument is that bonds are safer than stocks due to major catastrophies like revolution or war in a given country, that is clearly not true, based on historical data.

If the argument is that we should instead be globally diversified to avoid concentrated revolution/war etc risk in any given country, I clearly agree. So, then the question becomes is a global weighted stock portfolio more risky in the long term than a globally diversified stocks AND bonds portfolio.

Consider that most bonds (by value) in the world are nominal bonds not inflation protected. Most central banks have policy guidelines of ~2% inflation. Most/all country governments and regional governments have large debts. It is in their interest to create inflation over the long term (because it reduces the real debt they have to pay through taxation). So, they have the means and the desire. Getting good returns from bonds long term given this demographics and debt burden seems a long shot to me.

If you then ignore Governmnet Bonds and focus on Corporate, again the same argument. Companies are by law responsible to get returns for their owners (stock shareholders). On average if corporate bonds are a cheaper investment than equity, then companies are doing a bad job. On average, company stocks should return better than company bonds.

I would like folks to argue under which financial case would they predict long term global bonds to return better than global stocks. I just don't see it.


The financial case for bonds in a diversified portfolio is actually far easier to make than the complete financial market collapse situations. Financial market data for the last 50+ years is available in quality and quantity. There is diversification opportunity where assets correlations are < 1.0. Bonds have relatively low correlation to stocks. Financial collapse is harder because it's less frequent and data isn't as available.

There is a big spectrum of asset returns:
[worst]
__|__ nuclear apocalypse
__|__ financial markets freeze
__|__ depression
__|__ recession
__|__ good growth
__|__ technology boom
__|__ AI revolution (humanity goes extinct)
[best]

Bonds generally outperform stocks in depressions and recessions
US: great depression, dot com bubble, 2008 financial crisis
Japan: bonds pretty much outperformed stocks from 1980s to today
Eurozone: last 5-7 years

User avatar
Toons
Posts: 11921
Joined: Fri Nov 21, 2008 10:20 am
Location: Hills of Tennessee

Re: Over what time period should stocks NOT lose money

Post by Toons » Mon Jan 09, 2017 6:31 pm

I have always used 10 years as a minimum reference point when purchasing mutual funds.
:happy
"One does not accumulate but eliminate. It is not daily increase but daily decrease. The height of cultivation always runs to simplicity" –Bruce Lee

dbr
Posts: 23391
Joined: Sun Mar 04, 2007 9:50 am

Re: Over what time period should stocks NOT lose money

Post by dbr » Mon Jan 09, 2017 6:52 pm

Of course, since investors invest in stocks to gain money, asking when they would not lose money seems an odd question. A better question would be what the chances are of not gaining as much as one needs for one's objectives, whether or not that is less than the starting amount.

Generally if one wants to exactly preserve an amount of principal one would not invest in stocks to do that.

User avatar
sperry8
Posts: 1379
Joined: Sat Mar 29, 2008 9:25 pm
Location: CA, on the way to FL soon

Re: Over what time period should stocks NOT lose money

Post by sperry8 » Mon Jan 09, 2017 7:08 pm

SpartanBull wrote:Hello,
While there are no certainties, bogleheads clearly believe that over time, staying invested is a prudent investment. Time is your friend. If forced to answer "over what time period will stocks NOT lose money....in real terms, what would you say?". If forced to give an answer on this I would say 10 years. I'm curious what others think.


US Stocks have never lost money over any 30 year period (as others above have shown in graphs). :moneybag
Certainty is a requirement of ignorance. | Humbling Contest results: | 2016: #121 of 610 | 2015: #18 of 552 | 2014: #225 of 503 | 2013: #383 of 433 | 2012: #366 of 410 | 2011: #113 of 369 | 2010: #53 of 282

Dirghatamas
Posts: 506
Joined: Fri Jan 01, 2016 6:18 pm

Re: Over what time period should stocks NOT lose money

Post by Dirghatamas » Mon Jan 09, 2017 7:40 pm

taguscove wrote:Bonds generally outperform stocks in depressions and recessions
US: great depression, dot com bubble, 2008 financial crisis
Japan: bonds pretty much outperformed stocks from 1980s to today
Eurozone: last 5-7 years


Right, I understand all this. I fully agree about diversification. However, when people talk about diversification, most are talking about stocks vs. bonds in THE SAME GEOGRAPHY. My argument has always been that Stocks normally have higher expected returns than bonds and you can reduce tail risk by diversifying globally.

My arguments are not just theoretical. This simple view of the world has allowed me to invest every month since 1992, passively with 100% stocks without ever selling any shares and I am now FI by any reasonable measure, after almost 25 years of passive investing. I am absolutely confident I would have lost my marbles if I was 100% stocks ONLY in the US when the bear markets hit. I have stayed calm because, long term, I can't see how the world would lose its economic engine (stocks) and yet provide good returns on other assets. On the other hand I have no such confidence in 100% stocks if you focus on only one country (Japanese stocks being a recent example). The world is a closed system while a country is not.

Consider your examples. Yes a Japanese investor would have done better by stocks/bonds in Japan vs. only Japanese equity but in the same time frame would have actually done even better by global equity..so why ignore that?

Similarly, take the European investor. Bond yields have been poor. Stock returns (in Euro) have been poor. Euro has deflated against the dollar in the last few years. That person would have done much better than his/her stock/bond portfolio measured in Euro by investing globally in stocks..

Similarly, from 2000 to 2008 when a US investor did poorly in US stocks AND Dollar deflated against the Euro, the International stocks (measured in USD) did quite well.

Bonds are just loans. Companies pay them (Corporate bonds) from their earnings. Govt bonds will only get paid long term because of taxes (which again come from economic engine of companies). If the economic engine dries up long term, neither (company loans or Govt loans) will get paid in real terms..so why go through all this middle man stuff (loans to companies) instead of buying them (stocks)?

As for rebalancing between stocks and bonds, to me that is basically market timing. Yes, on this board people make a HUGE distinction between market timing and rebalancing (frowning on market timing) but it all gets muddled by definitions of risk and keeping risk constant. With 100% global stocks, there is no rebalancing because you are always in balance by definition. Whether the stocks markets go up or down, I do nothing. I just buy more every month.

I will again assert (look at data Clive wrote in a post above) that a 100% global stocks portfolio is probably the safest long term portfolio. If folks disagree, show me the case (not with geographically concentrated portfolio like Japan but a globally diversified one) over the long term.

EHEngineer
Posts: 560
Joined: Sat Feb 28, 2015 4:35 pm

Re: Over what time period should stocks NOT lose money

Post by EHEngineer » Mon Jan 09, 2017 7:51 pm

Am I really the first to mention Bernstein's retirement planner from Hell? Part 3 is my favorite and IMO more valuable than statistics.
http://www.efficientfrontier.com/ef/901/hell3.htm

Also coniser reading "Deep Risk"
https://www.amazon.com/Deep-Risk-Histor ... +bernstein
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

AlohaJoe
Posts: 2431
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Over what time period should stocks NOT lose money

Post by AlohaJoe » Wed Jan 11, 2017 3:57 am

nisiprius wrote:Finally, one really must not just ignore Pastor and Stambaugh


The lack of academic engagement around the topic is surprisingly limited; only 4 papers in the past 17 years, from what I can tell.

But in the most recent paper on the subject, "On the Long Run Volatility of Stocks" (2015) the authors find

The analyses above give us enough evidence that the conventional wisdom behind larger allocations in stocks for long horizon investors is not so wrong after all and puts the results from Pastor and Stambaugh [2012] into question [...] Our view is that the results in Figure 3 are enough to validate the notion that stocks are safer for the long run


Despite skimming the paper several times, most of it is still over my head. But I think their central claim is:

  • In order to replicate Pastor & Stambaugh's finding either β needs to be large or ρuw needs to be very negative.
  • Posterior evidence makes either those seem unlikely
  • Using more plausible values for the predictive regression framework confirms the common wisdom

Unless investors possess very unusual beliefs about key quantities in the model the predictive variance per period of stocks decreases as a function of the investment horizon. This result is very robust and can be achieved in a variety of settings including the very flexible time varying volatilities and covariances.

User avatar
bottlecap
Posts: 4918
Joined: Tue Mar 06, 2007 11:21 pm
Location: Tennessee

Re: Over what time period should stocks NOT lose money

Post by bottlecap » Wed Jan 11, 2017 6:59 am

Strange duplicate post.
Last edited by bottlecap on Wed Jan 11, 2017 7:00 am, edited 1 time in total.

User avatar
bottlecap
Posts: 4918
Joined: Tue Mar 06, 2007 11:21 pm
Location: Tennessee

Re: Over what time period should stocks NOT lose money

Post by bottlecap » Wed Jan 11, 2017 7:00 am

With the understanding that there are no certainties, I'm going to say 80, perhaps 100, years gets you pretty close.

JT

Clive
Posts: 1951
Joined: Sat Jun 13, 2009 5:49 am

Re: Over what time period should stocks NOT lose money

Post by Clive » Fri Jan 13, 2017 12:10 pm

viewtopic.php?p=1820900#p1820900
examples of how long it has taken to break-even (in real terms) after significant market declines across 19 countries between 1900-2010. Some of the longest include:

Equities

Germany: 45 years
France: 43 years
Italy: 37 years (I note this is a different number to the OP article - not sure why)
Japan: 33 years
Belgium: 31 years

The longest for a globally diversified equity portfolio: 7 years

Valuethinker
Posts: 32937
Joined: Fri May 11, 2007 11:07 am

Re: Over what time period should stocks NOT lose money

Post by Valuethinker » Fri Jan 13, 2017 2:13 pm

nisiprius wrote:
SpartanBull wrote:Hello,
While there are no certainties, bogleheads clearly believe that over time, staying invested is a prudent investment. Time is your friend. If forced to answer "over what time period will stocks NOT lose money....in real terms, what would you say?". If forced to give an answer on this I would say 10 years. I'm curious what others think.
Stocks have lost money over the following 10-year periods:

All figures "real" (inflation-adjusted), and including reinvested dividends.
Year range, inclusive; CAGR; total cumulative percentage loss

1965-1974 -31.86% -3.76%
1966-1975 -20.82% -2.31%
1968-1977 -22.33% -2.49%
1969-1978 -28.42% -3.29%
1970-1979 -13.23% -1.41%
1972-1981 -18.17% -1.99%
1973-1982 -16.87% -1.83%
1999-2008 -32.17% -3.81%
2000-2009 -29.17% -3.39%
2001-2010 -8.66% -0.90%


Is it possible you have reversed the labels on col 3 & 4?

Col 3 = total cumulative percentage loss

Col 4 = compound annual growth rate (geometric mean?)

Valuethinker
Posts: 32937
Joined: Fri May 11, 2007 11:07 am

Re: Over what time period should stocks NOT lose money

Post by Valuethinker » Fri Jan 13, 2017 2:21 pm

taguscove wrote:The answers above are solid based on a Monte Carlo approach on historic US equity returns. 10 years, including dividends, is about the longest drawdown period for the USA.

There is a huge survivorship bias though, that the USA was a dominant successful country in the 20-21st century. Germany, Argentina, UK, France, and China are prominent examples of major markets that experienced major disruptions from WW1 and WW2. The most recent hundred years had the biggest increase in human prosperity in the history of humanity.


Dividends were controlled in the UK in 1939-45 but I am not clear the market did that badly over that time (or 1914-18, although there was a financial panic in London in early August 1914 which shut down the international financial payments system).

The real disruption in UK stock markets was not the World Wars. But rather the period 1972-74 when adjusted for inflation, the market fell over 80% in real terms. This being a period without a revolution, or a war, or a change in regime. But, rather, a coal miner's strike a 3 day working week in response, followed by a government losing an election.


betting on a diversified portfolio of stocks is inherently a bet on the continued prosperity of humanity. For the most recent hundred years, that has definitely been a good bet. Going forward is anyone's guess, and I would be very wary of anyone who seems too confident. A diversified equity/bond portfolio is a solid default position.


And also stock market capitalism. The moment when the Bolsheviks storm the Winter Palace and eliminate the Menshevik government is never so very far away in world history (whether you think it's October because you use a Julian calendar, or November because you use a Gregorian one ;-) ).

Whether right now is 1913, 1933 or 1968 is an interesting question. None of those were great years for the future of stock markets.

My personal opinion is that people are generally too complacent towards paradigm shifts. In these paradigm shifts, water, food, shelter, medicine, weapons, gold, and cash were what historically paid off. There's a very good chance that financial markets will freeze up when a true disaster occurs. Mostly equity & bonds, with a small position in the essentials is my recommendation.


Or the abolition of capitalism (in its stock market/ free market form). Or environmental collapse leading to the collapse of civilization (plenty of precedent for that). A relapse to barbarism is never far from the gate of any Rome.

dharrythomas
Posts: 787
Joined: Tue Jun 19, 2007 4:46 pm

Re: Over what time period should stocks NOT lose money

Post by dharrythomas » Fri Jan 13, 2017 6:45 pm

I don't think there is any timeline where anything is safe. People are incredibly bad at forecasting.

Around the world, just in the 19-somethings, without trying to be all inclusive:
Private property was confiscated in Russia, China, Eastern Europe, Cuba--foreign investment was nationalized and that was just the communists, other governments have done the same
Hyperinflation in Germany and other countries after WWI
Swiss Banks holding Jewish investments, stolen Jewish art
Of course nothing like those things could ever happen in the US (read: it's different this time :oops:)

And those were societies that didn't completely fall apart. Jared Diamond in "Collapse" talks about how quickly past civilizations fall apart and decend into chaos from their peak. The real key is that there is no good alternative investment. In a collapse the things that are valuable now are less so.

In one of his examples, he said the richer people starved last.

There is nothing you have other than your character and attitude that cannot be taken away from you tomorrow. It is not productive to dwell on this, but it is worth keeping in the back of your mind.

Good luck

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

Re: Over what time period should stocks NOT lose money

Post by Rodc » Fri Jan 13, 2017 7:20 pm

grayfox wrote:
SpartanBull wrote:Hello,
While there are no certainties, bogleheads clearly believe that over time, staying invested is a prudent investment. Time is your friend. If forced to answer "over what time period will stocks NOT lose money....in real terms, what would you say?". If forced to give an answer on this I would say 10 years. I'm curious what others think.


I considered this question back in 2013. Rather than using historical data, I used a stochastic model. Returns were modeled as a Constant Expected Return (CER) Model, i.e. the Mean Return μ and Standard Deviation σ were constant. Here are some results:

Code: Select all

(μ,    σ)        5% chance of loss at
(0.10, 0.20)   130 months (10.8 years)
(0.05, 0.10)   130 months (10.8 years)   <- period is determined by the ratio σ / μ
(0.07, 0.20)   266 months (22.2 years)   <- smaller μ lengthens period
(0.10, 0.10)    33 months (2.75 years)   <- smaller σ shortens period

                 1% chance of loss at
(0.10, 0.20)   266 months (22 years)


There were other interesting results, like how the max drawdown varied with μ and σ.
Also, the worst drawdown was typically at about 5 years. After that, things improved.


Interesting. Thanks. Reality is likely worse than Gaussian but this gives a nice somewhat optimistic answer - ie lower bound.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

Post Reply