The Worst U.S. Portfolio Disasters — A Historical Comparison

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by SimpleGift » Fri Jul 06, 2018 4:37 pm

"I was lucky enough to see with my own eyes the recent stock market crash...
Never as then, amid suicides, hysteria, and groups of fainting people,
have I felt the sensation of real death, death without hope..."
- Federico Garcia Lorca, October 1929.


What's the worst that's likely to happen to our investment portfolios? Honestly, no one can say. But armed with knowledge of the worst U.S. portfolio disasters that have happened over the past century, we are perhaps better prepared for the financial shocks of the future. This post briefly compares the performance of different stock/bond mixes during the Great Depression of the 1930s, the oil price and inflation shocks of the 1970s, and the Tech Crash & Financial Crisis of the 2000s.

Technical note: In the three charts below, each portfolio's performance is indexed to the pre-crisis market peak. All returns are geometric total returns, inflation-adjusted, with dividends reinvested. Each chart is drawn to the same scale, with black circles showing the point at which each portfolio permanently regained its real pre-crisis value.

1. The Great Depression of the 1930s
The Depression Era performance of stocks and bonds is widely known. Stock-heavy portfolios dropped up to 75% over 3 years and took 12-15 years to finally regain their real value (chart below). Due to strong deflation, especially in 1931-33, bond-heavy portfolios regained their real value within just 4 years, with modest positive returns over the period.
  • Image
    Note: The 100% stock portfolio did briefly dip below its 1929 high several times during 1948-49.
    Sources: Monthly S&P stock returns from Shiller; monthly 10-year Treasury returns from Medium
2. The Oil and Inflation Shocks of the 1970s
The real price of imported oil in the U.S. increased 6-fold in the 1970s, leading to runaway inflation and stagnant economic growth. Initially, stock-heavy portfolios were hit the hardest, declining up to 50% within 18 months (chart below). Later, as inflation accelerated, bond-heavy portfolios suffered as well, declining up to 40% by 1981. Once inflation subsided, the various stock/bond mixes all regained their previous real values more or less together after 12 years.
  • Image
    Sources: Monthly S&P stock returns from Shiller; monthly 10-year Treasury returns from Medium
3. The Tech Crash & Financial Crisis of the 2000s
As you perhaps know well, the 2000s saw both the collapse of the dot.com bubble and an international banking crisis. Stock-heavy portfolios declined up to 40%-50% in both events and did not permanently recover their real pre-crisis values for 10-13 years (chart below). Due to declining interest rates and flight-to-quality, bond-heavy portfolios barely suffered during the period, but with modest returns.
  • Image
    Sources: Monthly S&P stock returns from Shiller; monthly 10-year Treasury returns from Medium
Your thoughts on any of this?

PS. Many thanks to Forum member siamond for instruction on the use of geometric rather than arithmetic total returns.
Last edited by SimpleGift on Sun Jul 08, 2018 6:56 am, edited 1 time in total.
Cordially, Todd

Random Poster
Posts: 1680
Joined: Wed Feb 03, 2010 10:17 am

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by Random Poster » Fri Jul 06, 2018 5:17 pm

Looks to me like a 50/50 split would have many advantages---perhaps it is the best way to ensure that one gets an average of the average returns?

snarlyjack
Posts: 780
Joined: Fri Aug 28, 2015 12:44 pm
Location: Montana

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by snarlyjack » Fri Jul 06, 2018 5:25 pm

SimpleGift,

Interesting charts as usual...Thank You.

I do know of a couple that had a 60/40 portfolio
going through the 2008/2009 debacle & were dca'ing
as much as they could through the whole thing.

When all was said & done, they dca their way into
a $1.5 Million portfolio. I thought that was interesting
considering they were using a 60/40 balanced fund.

User avatar
willthrill81
Posts: 6436
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by willthrill81 » Fri Jul 06, 2018 5:30 pm

A poll that Mebane Faber did not long ago asked investors what the maximum drawdown of bonds in real terms for U.S. investors over the last 100 years was. Nearly all of them dramatically underestimated it. You've correctly pointed out that it was approximately 40% over a five year period. Those 'safe' bonds may not be so safe after all.

The 1970s seem to make a compelling case that inflation-linked bonds should comprise a substantial part of one's fixed income holdings. Of course, one counter-argument to that is that if bonds are meant to smooth out portfolio volatility and offer stability, inflation-linked bonds are less than ideal as they are quite volatile compared to other bonds.

Logically, I can see a strong case being made for splitting one's fixed income between TIPS and short-term Treasuries. The former provides protection from unexpected inflation, while the latter provides stability and liquidity.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by SimpleGift » Fri Jul 06, 2018 5:54 pm

willthrill81 wrote:
Fri Jul 06, 2018 5:30 pm
A poll that Mebane Faber did not long ago asked investors what the maximum drawdown of bonds in real terms for U.S. investors over the last 100 years was. Nearly all of them dramatically underestimated it. You've correctly pointed out that it was approximately 40% over a five year period. Those 'safe' bonds may not be so safe after all.
One motivation for the OP analysis was to shine more light specifically on the portfolio disaster of the 1970s. The Great Depression is already well-known and many investors personally lived through the Tech Crash and Financial Crisis of the 2000s.

Do I think that central banks around the world are likely to ever let inflation get that out-of-control again? No, but the 1970s showed the devastation that high inflation can wreak upon both stock-heavy and bond-heavy portfolios. Your suggestion of a TIPS allocation seems like a rather low-cost but valuable insurance policy.
Cordially, Todd

staythecourse
Posts: 6197
Joined: Mon Jan 03, 2011 9:40 am

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by staythecourse » Fri Jul 06, 2018 5:55 pm

Simplegift, great work as usual.

Hopefully, someone can help me out as one poster sometime ago posted a presentation by DFA using Dimson Marsh data for 101 years as to stretch this discussion FURTHER out over multiple countries and periods of history. I remember an interesting observation as one of the slides showed the top 5-6 largest single country equity downtruns throughout HISTORY and the results were quite interesting. I remember cash and bonds took longer to recover in those situations then a high equity allocation. So the teaching point may not be so obvious in "worst case scenarios". Now hope someone can find the slide I am talking about.

What I take out of the excellent chart is what I have said for sometime... If one is going to be HIGH in equities one better: Have a rock solid occupation, long time horizon, no need for liquidity of that money, and ability to stay the course. Shakespeare is correct "Time heals all wounds". The question then becomes does the investor have the time to recover if one of these droughts occur.

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

columbia
Posts: 1100
Joined: Tue Aug 27, 2013 5:30 am

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by columbia » Fri Jul 06, 2018 6:11 pm

It seems like a good argument for a 50/50 portfolio over the long run.

marcopolo
Posts: 1270
Joined: Sat Dec 03, 2016 10:22 am

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by marcopolo » Fri Jul 06, 2018 6:37 pm

columbia wrote:
Fri Jul 06, 2018 6:11 pm
It seems like a good argument for a 50/50 portfolio over the long run.
Maybe.

But, i fear that looking at the data starting just before the meltdown periods may lead to the wrong conclusions.
Not sure about the oil/stagflation shock of the 70s, but both the great depression and the tech bubble were proceeded by huge run up in equities.
The plots show each of the asset allocation cases starting at the same point. But, if one had been investing prior to that arbitrary starting point with the same allocation, the starting value would have been vastly different. So, even though a more stock heavy portfolio may have lost more during the down turn, it would have started at a much higher beginning value. I have not done the analysis to see if in the end that would have changed the story much, but i would just caution drawing too much from looking at just the period of draw down. We have to live with both draw downs and melt ups.
Once in a while you get shown the light, in the strangest of places if you look at it right.

Nate79
Posts: 3710
Joined: Thu Aug 11, 2016 6:24 pm
Location: Delaware

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by Nate79 » Fri Jul 06, 2018 6:48 pm

What impact does the choice of "stocks" and "bonds" make in the discussion? It looks like the "stocks" is S&P500 and bonds is 10 year treasury bonds?

S&P500 was not really the S&P500 and only included, what, 90 stocks in the early days until 1957. Would a total stock market fund make any difference?

And how does the 10 year treasury bond return compare to something that most people may invest in (total bond market fund or intermediate bond fund)?

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by SimpleGift » Fri Jul 06, 2018 6:50 pm

staythecourse wrote:
Fri Jul 06, 2018 5:55 pm
What I take out of the excellent chart is what I have said for sometime... If one is going to be HIGH in equities one better: Have a rock solid occupation, long time horizon, no need for liquidity of that money, and ability to stay the course.
An often overlooked but vital point. Beyond stock and bond portfolios, financial crises of course impact the real economy, employment, business and consumer confidence, etc. It's one thing to have one's stock-heavy portfolio lose 50% of its value and take 10-12 years to recover. It's quite another to also lose one's job, and possibly one's home, at the same time:
Thus the standard Boglehead recommendations for adequate bond allocations and emergency funds are not just empty bromides, but are based on real-life experiences of past portfolio disasters.
Last edited by SimpleGift on Fri Jul 06, 2018 6:52 pm, edited 1 time in total.
Cordially, Todd

User avatar
willthrill81
Posts: 6436
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by willthrill81 » Fri Jul 06, 2018 6:51 pm

Nate79 wrote:
Fri Jul 06, 2018 6:48 pm
What impact does the choice of "stocks" and "bonds" make in the discussion? It looks like the "stocks" is S&P500 and bonds is 10 year treasury bonds?

S&P500 was not really the S&P500 and only included, what, 90 stocks in the early days until 1957. Would a total stock market fund make any difference?

And how does the 10 year treasury bond return compare to something that most people may invest in (total bond market fund or intermediate bond fund)?
I don't believe that there is much debate about the viability of the stock data used in the period referenced in the OP. Ten year Treasuries are more volatile than total bond market or intermediate term Treasuries, but likely still a decent proxy.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
willthrill81
Posts: 6436
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by willthrill81 » Fri Jul 06, 2018 6:55 pm

SimpleGift wrote:
Fri Jul 06, 2018 6:50 pm
It's one thing to have one's stock-heavy portfolio lose 50% of its value and take 10-12 years to recover. It's quite another to also lose one's job, and possibly one's home, at the same time:
The fact that this happened has happened to millions of people in the U.S. and not just in one or two limited time periods leads me to believe that intentionally using your home as an instrument to gain investment leverage is riskier than many believe. Historically, the odds are high that it will work out fine, but that tail risk may be gnarly.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

AlohaJoe
Posts: 3919
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by AlohaJoe » Fri Jul 06, 2018 9:08 pm

Nate79 wrote:
Fri Jul 06, 2018 6:48 pm
S&P500 was not really the S&P500 and only included, what, 90 stocks in the early days until 1957. Would a total stock market fund make any difference?
Using a total stock market fund for the Great Depression would make things look worse -- small cap stocks got obliterated in the Great Depression -- though I'm not sure the difference is big enough to change any conclusions or cause people to change their investing strategies.

Image

You see that over the 1924-1940 period there are 4 different historical data series and the total percentage return varies from -15.49% to -24.39%. "S&P Historical" is the 90-stock index you're talking about, so you can see that it has the "best" return (only -15.49%) whereas all the others had worse returns. A 10% difference sounds huge but that's over 16 years. It is really just a 0.7% difference each year that adds up; and I'd argue that's not the kind of difference people really notice in the real world.

User avatar
siamond
Posts: 4305
Joined: Mon May 28, 2012 5:50 am

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by siamond » Fri Jul 06, 2018 11:20 pm

marcopolo wrote:
Fri Jul 06, 2018 6:37 pm
But, i fear that looking at the data starting just before the meltdown periods may lead to the wrong conclusions.
Not sure about the oil/stagflation shock of the 70s, but both the great depression and the tech bubble were proceeded by huge run up in equities.
The plots show each of the asset allocation cases starting at the same point. But, if one had been investing prior to that arbitrary starting point with the same allocation, the starting value would have been vastly different. So, even though a more stock heavy portfolio may have lost more during the down turn, it would have started at a much higher beginning value. [..]
Yup, exactly. Those charts are essentially cherry-picking, I am afraid. Not only do they start in the year where two of the worst stock crises started (1929, 1973, 2000), but they even start at the very exact right month to make a stock-heavy portfolio look as bad as possible, totally ignoring the big ramp-up that preceded the crisis. This isn't terribly meaningful. We could assemble charts like that, with a different starting point, where a bonds-heavy portfolio would look incredibly bad. Notably somewhere in 1941 where bonds started on a drop deeper than 40% (in real terms) which lasted more than 40 years before recovery. Which in my humble opinion is by far the worst US portfolio disaster. And this would be equally meaningless, a single data point in history which has no chance of happening again in the same manner.

In the Simba backtesting spreadsheet, one can quickly assemble portfolio cycles charts, varying the starting year, and comparing two portfolios according to a given metric. Here is an example, showing the CAGR (in real terms) of two realistic portfolios over a cycle of the 30 years, varying the starting year. Then we can truly appreciate the diversity of outcomes. Yes, no monthly returns in there, so the Sep 1929 'worst case' is missing, but how meaningful is that?

Image

Then, much more significantly in my opinion, one can easily look at a drawdown analysis and an SWR analysis. I went on a long rant on this precise topic in this recent blog... Here is the SWR chart, which illustrates much better how bad the various crises were, by accounting for sequence of returns issues and regular withdrawals (e.g. for retirees).

Image

My overall feedback to the OP is that, by overly focusing on a few cherry-picked starting points, one can end up with very flawed conclusions about their own strategy for the future.

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by SimpleGift » Sat Jul 07, 2018 12:02 am

siamond wrote:
Fri Jul 06, 2018 11:20 pm
Those charts are essentially cherry-picking, I am afraid. Not only do they start in the year where two of the worst stock crises started (1929, 1973, 2000), but they even start at the very exact right month to make a stock-heavy portfolio look as bad as possible, totally ignoring the big ramp-up that preceded the crisis.
Well, the whole point of the OP and this thread is to discuss what the very worst portfolio losses (magnitude and duration) during crisis periods have been over the last century. Thus they have to start at the peak month of the stock market in the years of the crashes — in 1929, 1973 and 2000. It's a feature, not a bug!

Appreciate your analysis of historical 30-year portfolio CAGRs over various starting periods — but that's really not the point of this thread, which focuses on worst case, portfolio crisis scenarios in history.
Cordially, Todd

AlohaJoe
Posts: 3919
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by AlohaJoe » Sat Jul 07, 2018 12:06 am

marcopolo wrote:
Fri Jul 06, 2018 6:37 pm
columbia wrote:
Fri Jul 06, 2018 6:11 pm
It seems like a good argument for a 50/50 portfolio over the long run.
Maybe.

But, i fear that looking at the data starting just before the meltdown periods may lead to the wrong conclusions.
Not sure about the oil/stagflation shock of the 70s, but both the great depression and the tech bubble were proceeded by huge run up in equities.
To elaborate on marcopolo's point, here's a series of charts showing the years before 1929.

Let's start with the usual 1929 chart, showing 100% stocks (the blue line) and a 60/40 portfolio (the orange line). This is inflation-adjusted portfolio values assuming 4% inflation-adjusted withdrawals.

Image

That's the kind of chart many of us have seen many times. But let's wind back the clock a little bit. What if someone didn't retire in 1929. What if they retired in 1922, just a few years earlier?

Image

You can still see the steep drawdown of the 1929 crash. But you can also see how big the runup was. They retired with $1 million and it got up to $4 million. Even after the worst equity crash in US, the 100% stock portfolio solidly trounces the 60/40 portfolio. (And that's without playing second guessing games like, "Wow, when it got to $3 million they wouldn't have bought a $1 million annuity or something with all of that extra windfall?)

Now let's look at a few more years.

Image
Image
Image
Image
Image
Image

From 1925 and earlier, the 100% equity portfolio does at least as well, and usually much better, than the 60/40 portfolio despite suffering the worst equity crash in US history.

And even in 1925 and (to a lesser extent) 1926 -- the portfolio value of the 100% equity portfolio never really drops below it started. (In 1925 it never drops below the starting value; in 1926 it drops to 88% of its starting value.) So if pause a moment in our "keeping up with the Jones's, compare to a benchmark portfolio" mindset (which I do all the time, too, so I'm not casting aspersions) -- think about it solely from the retiree's perspective. Despite being 100% equities and retiring in 1926 -- 30-40 months before the biggest stock market crash in US history -- the portfolio was never under any stress and the retiree had no real reason for concern. (That's not to say that people wouldn't have felt concerned. After all unemployment was 25% but it wouldn't exactly have been a rational concern.)

So this is something that I personally struggle with. When I look at charts of the massive runup in the US & Japan before their huge equity crashes, part of me wonders whether any retiree would really have retired anywhere close to the peak at exactly Their Number. On the one hand, I know how easy it is for humans to get swept up in hysteria and manias. On the other hand...(at least on this board) it seems that everyone suffers from One More Year syndrome and pads Their Number out quite a bit.

And from what I can tell, at least in the US, as long you retired +/- 24 months away from the absolute worst month you are basically okay no matter what your asset allocation.

User avatar
HomerJ
Posts: 11919
Joined: Fri Jun 06, 2008 12:50 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by HomerJ » Sat Jul 07, 2018 12:23 am

willthrill81 wrote:
Fri Jul 06, 2018 5:30 pm
A poll that Mebane Faber did not long ago asked investors what the maximum drawdown of bonds in real terms for U.S. investors over the last 100 years was. Nearly all of them dramatically underestimated it. You've correctly pointed out that it was approximately 40% over a five year period. Those 'safe' bonds may not be so safe after all.

The 1970s seem to make a compelling case that inflation-linked bonds should comprise a substantial part of one's fixed income holdings. Of course, one counter-argument to that is that if bonds are meant to smooth out portfolio volatility and offer stability, inflation-linked bonds are less than ideal as they are quite volatile compared to other bonds.

Logically, I can see a strong case being made for splitting one's fixed income between TIPS and short-term Treasuries. The former provides protection from unexpected inflation, while the latter provides stability and liquidity.
You can also use shorter term bonds to mitigate the effects of inflation. Most charts like the ones above use 10-year Treasury bonds. If one is using Total Bond Market (6.1 years average duration), one should do better.

Or one can use even shorter term bonds or CDs.
The J stands for Jay

User avatar
willthrill81
Posts: 6436
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by willthrill81 » Sat Jul 07, 2018 12:28 am

HomerJ wrote:
Sat Jul 07, 2018 12:23 am
willthrill81 wrote:
Fri Jul 06, 2018 5:30 pm
A poll that Mebane Faber did not long ago asked investors what the maximum drawdown of bonds in real terms for U.S. investors over the last 100 years was. Nearly all of them dramatically underestimated it. You've correctly pointed out that it was approximately 40% over a five year period. Those 'safe' bonds may not be so safe after all.

The 1970s seem to make a compelling case that inflation-linked bonds should comprise a substantial part of one's fixed income holdings. Of course, one counter-argument to that is that if bonds are meant to smooth out portfolio volatility and offer stability, inflation-linked bonds are less than ideal as they are quite volatile compared to other bonds.

Logically, I can see a strong case being made for splitting one's fixed income between TIPS and short-term Treasuries. The former provides protection from unexpected inflation, while the latter provides stability and liquidity.
You can also use shorter term bonds to mitigate the effects of inflation. Most charts like the ones above use 10-year Treasury bonds. If one is using Total Bond Market (6.1 years average duration), one should do better.

Or one can use even shorter term bonds or CDs.
Yep. Reducing the duration has helped to deal with short-term inflation risks. I believe that even the lowly T-bill has only lagged the current CPI by more than 1% only once or twice in history for more than a few months.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by SimpleGift » Sat Jul 07, 2018 12:29 am

Nate79 wrote:
Fri Jul 06, 2018 6:48 pm
And how does the 10 year treasury bond return compare to something that most people may invest in (total bond market fund or intermediate bond fund)?
To keep the focus on how assets have performed historically during worst case crisis periods:

Beside 10-year Treasuries, the only other monthly bond series going back to the early 1900s that I'm aware of is the Moody's Seasoned Aaa Corporate Bond data (found here). These are high-quality corporate bonds with maturities 20 years and above. To see how they performed compared with 10-year Treasuries during our three portfolio crises, the charts below show their real monthly total returns indexed over each period.

During the Great Depression, 10-year Treasuries outperformed by a wide margin, but corporates mostly held their value.
  • Image

During the Oil and Inflation Shocks, corporates massively outperformed, while 10-year Treasuries lost 40% of their value.
  • Image

During the Tech Crash & Financial Crisis, 10-year Treasuries outperformed and high-quality corporates did just OK.
  • Image
Bottom line: A mix of Treasuries and high-quality corporates like Total Bond or Intermediate Bond Index Fund would likely have performed fine during the three crisis periods — but TIPS replacing nominal Treasuries would have been better in 1973.
Cordially, Todd

Whakamole
Posts: 790
Joined: Wed Jan 13, 2016 9:59 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by Whakamole » Sat Jul 07, 2018 12:45 am

AlohaJoe wrote:
Sat Jul 07, 2018 12:06 am
marcopolo wrote:
Fri Jul 06, 2018 6:37 pm
columbia wrote:
Fri Jul 06, 2018 6:11 pm
It seems like a good argument for a 50/50 portfolio over the long run.
Maybe.

But, i fear that looking at the data starting just before the meltdown periods may lead to the wrong conclusions.
Not sure about the oil/stagflation shock of the 70s, but both the great depression and the tech bubble were proceeded by huge run up in equities.
To elaborate on marcopolo's point, here's a series of charts showing the years before 1929.

Let's start with the usual 1929 chart, showing 100% stocks (the blue line) and a 60/40 portfolio (the orange line). This is inflation-adjusted portfolio values assuming 4% inflation-adjusted withdrawals.

Image

That's the kind of chart many of us have seen many times. But let's wind back the clock a little bit. What if someone didn't retire in 1929. What if they retired in 1922, just a few years earlier?

Image

You can still see the steep drawdown of the 1929 crash. But you can also see how big the runup was. They retired with $1 million and it got up to $4 million. Even after the worst equity crash in US, the 100% stock portfolio solidly trounces the 60/40 portfolio. (And that's without playing second guessing games like, "Wow, when it got to $3 million they wouldn't have bought a $1 million annuity or something with all of that extra windfall?)

Now let's look at a few more years.

Image
Image
Image
Image
Image
Image

From 1925 and earlier, the 100% equity portfolio does at least as well, and usually much better, than the 60/40 portfolio despite suffering the worst equity crash in US history.

And even in 1925 and (to a lesser extent) 1926 -- the portfolio value of the 100% equity portfolio never really drops below it started. (In 1925 it never drops below the starting value; in 1926 it drops to 88% of its starting value.) So if pause a moment in our "keeping up with the Jones's, compare to a benchmark portfolio" mindset (which I do all the time, too, so I'm not casting aspersions) -- think about it solely from the retiree's perspective. Despite being 100% equities and retiring in 1926 -- 30-40 months before the biggest stock market crash in US history -- the portfolio was never under any stress and the retiree had no real reason for concern. (That's not to say that people wouldn't have felt concerned. After all unemployment was 25% but it wouldn't exactly have been a rational concern.)

So this is something that I personally struggle with. When I look at charts of the massive runup in the US & Japan before their huge equity crashes, part of me wonders whether any retiree would really have retired anywhere close to the peak at exactly Their Number. On the one hand, I know how easy it is for humans to get swept up in hysteria and manias. On the other hand...(at least on this board) it seems that everyone suffers from One More Year syndrome and pads Their Number out quite a bit.

And from what I can tell, at least in the US, as long you retired +/- 24 months away from the absolute worst month you are basically okay no matter what your asset allocation.
Except when are people more likely to retire? You pointed out that the market in 1922 quadrupled by the time it got to 1929. I would not have even thought to retire if I held $250K in 1922. In 1929 though, now I have $1M. It's not a matter of it being at the peak per se, there's nothing special about the year. It may be more likely for people to pick 1929 because:
- more people have piled into the market, making it richly valued. That includes new investors, existing investors deciding to go 100% stock, people deciding to quit their jobs and trade stocks for a living (saw a lot of this in the run-up to 2000, like we saw professional house-flippers in 2007), etc. All that increases the value of their portfolio and brings more people to their number.
- One More Year syndrome (if you can save any percentage of your income, then working another year lets you invest even more money into the market, and hopefully have a better quality of life in retirement. Make hay while the sun shines.)
- The number of people who could retire because they believed their portfolio could sustain them dropped with the stock market crash.

Benjamin Roth's "The Great Depression: A Diary" (recommended a few times on this forum) really brought to life how it wasn't that nobody knew when the stock market would recover; they didn't know if the stock market would recover. It's easy for us to look at how the market did for 20 years but all anyone saw after the crash was the blue line dropping like a rock.

AlohaJoe
Posts: 3919
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by AlohaJoe » Sat Jul 07, 2018 1:14 am

Whakamole wrote:
Sat Jul 07, 2018 12:45 am
Except when are people more likely to retire? You pointed out that the market in 1922 quadrupled by the time it got to 1929. I would not have even thought to retire if I held $250K in 1922. In 1929 though, now I have $1M. It's not a matter of it being at the peak per se, there's nothing special about the year. It may be more likely for people to pick 1929 because
Stock market increases have surprisingly little impact on causing people to retire. Not "no impact". But more of a marginal impact and a pretty small one. After all, most people tend to be constrained by things like "I need Social Security to make up 20-50% of my retirement income, so I can't retire before 62 unless my portfolio gets MASSIVELY larger" or "I'm not going to retire, no matter what, until the kids are out of the house/graduated from college" or "I'm not going to retire until after the mortgage is paid off" or "I'm not going to retire until my pension is fully vested" or "I'm not going to retire until I get promoted to G-14 and have the higher pension payout" or "no point retiring, since my spouse isn't going to retire for 2 more years" and so on.

The only real data I'm aware of on the subject focused on the stock market bubble of the 1990s and its impact on retirement. That multi-year bubble
At its peak in 1999, the percent retired in the sample is about 3.3 percentage points higher than it would be without the extra stock returns.
In context this means at age 62 the percent of those retired went from ~49% to ~52%, or about a 5-7% relative increase.
The effect on the average retirement age can be found by adding the effects at the various ages. When this is done, the conclusion is that the average retirement age for the entire sample is decreased by 0.25 years, or about three months.
Though we have to keep in mind, as always, that averages can be deceiving. The average person has <$100,000 so of course a massive bull market isn't going to allow them to retire 5 years early or anything.

Overall, while the narrative of more retirements at the top of a bubble sounds plausible on the surface, I've never found anything like actual data to support the view that it is anything but a small, marginal effect outweighed by many other personal circumstances.

jbranx
Posts: 264
Joined: Thu Feb 09, 2017 6:57 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by jbranx » Sat Jul 07, 2018 1:42 am

Thanks again, Simplegift, for a good depiction of what the worst can look like. I'm an early on boomer, so I grew up with a set of great grandparents, grandparents, and parents who had experienced the G. Depression. Tales of stock losses, home losses, job losses, bankruptcies, suicides, and begging were a part of many a discussion so the younger ones would know what could happen. One set of my great grandparents never trusted banks again and ran their own little loan operation out of their farmhouse. None of them trusted stocks, except the younger ones too late in life to really benefit.

I bought my first investments in 1970 and know what torture it was to stay invested. I remember one fund I bought in '70 that I sold in '83 for a loss, two years into the recovery. It's hard for anyone not living through inflation/poor stock returns/poor bond returns to really understand how misery compounds. Folks who lived through the lines at gas stations and all the inflation didn't get as traumatized as those who lived through the depression, however. "Timing newsletters" proliferated in the eighties to tell people when the next crash was coming. I sat in front of a quote machine in Oct. 19, '87 and saw price declines that were simply beyond belief. Stocks gaping down high percentages on the next trade. Lots of dire headlines were written by folks my age who had heard their grandparent's tales of woe. Happily, '87 ended as an up year.

I have since watched my generation gain great confidence in our financial system. Most have not been traumatized by market crashes in 2000 and 2008-9 like my ancestors were. In fact, most people I deal with, even if they think the market is overvalued, would be delighted to get back in with a good correction, or so they think. I'm with those who say we retired ones probably are best off with a 50/50 or something close allocation to broad stock indices and low to intermediate duration bonds, a good portion of those inflation protected. Not because we think we know what is going to happen, but for the reason Peter Bernstein said he preferred 60/40: we don't know. The "worst" you show very likely won't happen again--there are good reasons for higher confidence in our financial systems--but nothing guarantees that something cannot sail out of the mist and wreck havoc on the best portfolios.

User avatar
oldcomputerguy
Posts: 3530
Joined: Sun Nov 22, 2015 6:50 am
Location: In the middle of five acres of woods

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by oldcomputerguy » Sat Jul 07, 2018 5:53 am

willthrill81 wrote:
Fri Jul 06, 2018 5:30 pm
Logically, I can see a strong case being made for splitting one's fixed income between TIPS and short-term Treasuries. The former provides protection from unexpected inflation, while the latter provides stability and liquidity.
In fact Paul Merriman uses a mixture of intermediate-term, short-term, and inflation-protected Treasury bonds in his recommended portfolios. I always wondered why he was completely eschewing corporate bonds, I suppose he also prefers taking all risk on the equity side.
It’s taken me a lot of years, but I’ve come around to this: If you’re dumb, surround yourself with smart people. And if you’re smart, surround yourself with smart people who disagree with you.

thx1138
Posts: 784
Joined: Fri Jul 12, 2013 2:14 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by thx1138 » Sat Jul 07, 2018 6:30 am

SimpleGift wrote:
Sat Jul 07, 2018 12:02 am
siamond wrote:
Fri Jul 06, 2018 11:20 pm
Those charts are essentially cherry-picking, I am afraid. Not only do they start in the year where two of the worst stock crises started (1929, 1973, 2000), but they even start at the very exact right month to make a stock-heavy portfolio look as bad as possible, totally ignoring the big ramp-up that preceded the crisis.
Well, the whole point of the OP and this thread is to discuss what the very worst portfolio losses (magnitude and duration) during crisis periods have been over the last century. Thus they have to start at the peak month of the stock market in the years of the crashes — in 1929, 1973 and 2000. It's a feature, not a bug!

Appreciate your analysis of historical 30-year portfolio CAGRs over various starting periods — but that's really not the point of this thread, which focuses on worst case, portfolio crisis scenarios in history.
I think the takeaway still is that you didn’t base the analysis on a “worst case” for *portfolios* but rather a worst case for *stocks* because the dates were primarily based on stock crises.

The same analysis starting in late 1940 would be an interesting one. Bonds won’t look so “safe” when you include peak start dates for bond crises and not just stock crises.

Thanks for doing the analysis by the way, good stuff!

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by SimpleGift » Sat Jul 07, 2018 7:06 am

thx1138 wrote:
Sat Jul 07, 2018 6:30 am
I think the takeaway still is that you didn’t base the analysis on a “worst case” for *portfolios* but rather a worst case for *stocks* because the dates were primarily based on stock crises.

The same analysis starting in late 1940 would be an interesting one. Bonds won’t look so “safe” when you include peak start dates for bond crises and not just stock crises.
Point taken. My focus in this thread was primarily on "stock crises," since they tend to be the most emotionally dramatic and frightening for beginning and veteran investors alike ("death without hope...").

Perhaps it would be worthwhile to do a follow-up thread one of these days that focuses on various portfolio performances during "bond crises" (though we may have to come up with a less oxymoronic name!). Thank you for the suggestion.
Cordially, Todd

grok87
Posts: 8501
Joined: Tue Feb 27, 2007 9:00 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by grok87 » Sat Jul 07, 2018 8:44 am

this was really useful thanks!

so probably very difficult to do, but in case you are up for a challenge i'd be interested in seeing the same for Japan starting in say late 1991 for the next two decades, sometimes referred to as the Japanese lost decade or Japan lost "score" (as in four score and 7 years ago...)
https://en.wikipedia.org/wiki/Lost_Decade_(Japan)
happy to help with the data...
Keep calm and Boglehead on. KCBO.

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by SimpleGift » Sat Jul 07, 2018 9:06 am

grok87 wrote:
Sat Jul 07, 2018 8:44 am
so probably very difficult to do, but in case you are up for a challenge i'd be interested in seeing the same for Japan starting in say late 1991 for the next two decades, sometimes referred to as the Japanese lost decade or Japan lost "score" (as in four score and 7 years ago...)
I believe Forum member siamond already did such an analysis last year. See this Forum thread:

A Short Study of the Japan Crisis

It's a really first-class look at the 1990s crisis using excellent stock and bond data for the Japanese markets.
Last edited by SimpleGift on Sat Jul 07, 2018 9:09 am, edited 1 time in total.
Cordially, Todd

User avatar
siamond
Posts: 4305
Joined: Mon May 28, 2012 5:50 am

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by siamond » Sat Jul 07, 2018 9:09 am

SimpleGift wrote:
Sat Jul 07, 2018 12:02 am
siamond wrote:
Fri Jul 06, 2018 11:20 pm
Those charts are essentially cherry-picking, I am afraid. Not only do they start in the year where two of the worst stock crises started (1929, 1973, 2000), but they even start at the very exact right month to make a stock-heavy portfolio look as bad as possible, totally ignoring the big ramp-up that preceded the crisis.
Well, the whole point of the OP and this thread is to discuss what the very worst portfolio losses (magnitude and duration) during crisis periods have been over the last century. Thus they have to start at the peak month of the stock market in the years of the crashes — in 1929, 1973 and 2000. It's a feature, not a bug!

Appreciate your analysis of historical 30-year portfolio CAGRs over various starting periods — but that's really not the point of this thread, which focuses on worst case, portfolio crisis scenarios in history.
Sure, I was keenly aware of the fact that this thread, by design, focuses on worst US stock crises (which is actually NOT the same thing as the worst portfolio disasters). And this is my entire point. I am afraid this kind of perspective is terribly misleading. I don't know, it seems that some people somehow like to frighten themselves with this type of cherry-picking of 'worst cases'. Trouble is those are a couple of extreme singularities. What real lesson can you learn from those? It isn't difficult to find worst singularities (e.g. from international stock history AND from bonds history). So should we look for those? But then the stock market closed for decades in some countries, which is even worse. Where do you draw the lines, notably in your expectations about the future?

PS. as a side note, 1965/66 are actually worse starting dates than 1973...

User avatar
Kenkat
Posts: 4430
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by Kenkat » Sat Jul 07, 2018 9:11 am

This makes a great argument for why 100% stocks has never made sense for me at least - although in 1973 there was really no good place to hide in the U.S.

CarpeDiem22
Posts: 139
Joined: Tue May 22, 2018 11:20 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by CarpeDiem22 » Sat Jul 07, 2018 9:30 am

OP, you missed to include those portfolio disaster years when stocks heavily outperformed bonds. Being 100% bonds would have been a disaster for your portfolio. As Warren Buffet says, whatever notional gains he misses, he considers them a loss.

User avatar
siamond
Posts: 4305
Joined: Mon May 28, 2012 5:50 am

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by siamond » Sat Jul 07, 2018 9:31 am

Another point (AlohaJoe hinted at it).

If one's portfolio grows by 20% in a few months, then drops 50%, does it REALLY feel like a 50% drop? Or less than that? Yes, we're all subject to anchoring bias, but to which extent? If the portfolio grew 20% at some point, then stayed stable for a couple of years, then yeah, a sudden drop of 50% would feel like it (the anchor is set). But it would definitely not feel like that in the first scenario (steep rise, steep drop). Which is basically what happened in 1929 or 2002 or 2008 (while the 1973 crisis was markedly different). I have long thought that drawdown charts should depict the drop from the lowest point reached in the past year, or something like that. But even that isn't quite right. Human emotions are just so hard to quantify!

At the end, I strongly believe that overly focusing on portfolio value is very unfortunate. Focusing on actual income (current and future expectations) is a much sounder perspective. I can't recommend enough for every Boglehead to read this marvel of a book:
Devil Take the Hindmost: A History of Financial Speculation, by Edward Chancellor

The author really drives home that most of those crises are fully due to human foolishness, which does NOT change much over the centuries. And then, maybe, we can become a little more detached from those 'worst crises/disasters', and focus on what matters much more in life, decent income and enjoying life while we can.

grok87
Posts: 8501
Joined: Tue Feb 27, 2007 9:00 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by grok87 » Sat Jul 07, 2018 9:47 am

SimpleGift wrote:
Sat Jul 07, 2018 9:06 am
grok87 wrote:
Sat Jul 07, 2018 8:44 am
so probably very difficult to do, but in case you are up for a challenge i'd be interested in seeing the same for Japan starting in say late 1991 for the next two decades, sometimes referred to as the Japanese lost decade or Japan lost "score" (as in four score and 7 years ago...)
I believe Forum member siamond already did such an analysis last year. See this Forum thread:

A Short Study of the Japan Crisis

It's a really first-class look at the 1990s crisis using excellent stock and bond data for the Japanese markets.
thanks
Keep calm and Boglehead on. KCBO.

Nate79
Posts: 3710
Joined: Thu Aug 11, 2016 6:24 pm
Location: Delaware

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by Nate79 » Sat Jul 07, 2018 9:52 am

While I think the results are interesting to see what happens to a portfolio during these major downturns I disagree with the posters that somehow by looking at the charts you can conclude what your asset allocation should be. Most people are investing for decades and decades and even in retirement you may have 30+ years. These charts only look at the negative years and ignore the great increases before and after which would have been missed by an overly conservative portfolio.

User avatar
willthrill81
Posts: 6436
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by willthrill81 » Sat Jul 07, 2018 9:57 am

oldcomputerguy wrote:
Sat Jul 07, 2018 5:53 am
willthrill81 wrote:
Fri Jul 06, 2018 5:30 pm
Logically, I can see a strong case being made for splitting one's fixed income between TIPS and short-term Treasuries. The former provides protection from unexpected inflation, while the latter provides stability and liquidity.
In fact Paul Merriman uses a mixture of intermediate-term, short-term, and inflation-protected Treasury bonds in his recommended portfolios. I always wondered why he was completely eschewing corporate bonds, I suppose he also prefers taking all risk on the equity side.
Yes, his avoidance of corporate bonds would be right up Larry Swedroe's alley. And frankly, it makes sense to me. Corporate bonds behave like a mix of mostly Treasuries and some stock.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
willthrill81
Posts: 6436
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by willthrill81 » Sat Jul 07, 2018 9:58 am

Kenkat wrote:
Sat Jul 07, 2018 9:11 am
This makes a great argument for why 100% stocks has never made sense for me at least - although in 1973 there was really no good place to hide in the U.S.
Short-term Treasuries and T-bills looked better than about anything else I've seen, but positive real returns were pretty scarce in that period.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

grok87
Posts: 8501
Joined: Tue Feb 27, 2007 9:00 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by grok87 » Sat Jul 07, 2018 10:06 am

willthrill81 wrote:
Sat Jul 07, 2018 9:58 am
Kenkat wrote:
Sat Jul 07, 2018 9:11 am
This makes a great argument for why 100% stocks has never made sense for me at least - although in 1973 there was really no good place to hide in the U.S.
Short-term Treasuries and T-bills looked better than about anything else I've seen, but positive real returns were pretty scarce in that period.
if tips were around they may have been a good place to be
Keep calm and Boglehead on. KCBO.

GAAP
Posts: 686
Joined: Fri Apr 08, 2016 12:41 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by GAAP » Sat Jul 07, 2018 10:18 am

siamond wrote:
Sat Jul 07, 2018 9:31 am
Another point (AlohaJoe hinted at it).

If one's portfolio grows by 20% in a few months, then drops 50%, does it REALLY feel like a 50% drop? Or less than that?
That probably depends to a great deal on how closely and frequently you look at portfolio value. Too much attention leads to bad decisions...

InvMoney
Posts: 19
Joined: Mon Jun 11, 2018 3:14 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by InvMoney » Sat Jul 07, 2018 10:21 am

When we're in a stock bull market such as we currently are, some investors loose site of the consequences of a bear market on a stock heavy, indexed portfolio that by its very nature takes no steps to mitigate losses during bear markets.
Last edited by InvMoney on Sat Jul 07, 2018 10:59 am, edited 1 time in total.

Theoretical
Posts: 1439
Joined: Tue Aug 19, 2014 10:09 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by Theoretical » Sat Jul 07, 2018 10:38 am

I look at these in a different way. These are not the worst case portfolio disasters. The real beasts are the single stock risk calamities (or single country in case of the Axis and Soviet Bloc). Those guys made even the 100% stock portfolios look great in comparison.

GAAP
Posts: 686
Joined: Fri Apr 08, 2016 12:41 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by GAAP » Sat Jul 07, 2018 11:20 am

willthrill81 wrote:
Fri Jul 06, 2018 5:30 pm
Logically, I can see a strong case being made for splitting one's fixed income between TIPS and short-term Treasuries. The former provides protection from unexpected inflation, while the latter provides stability and liquidity.
Does anyone know how much of the inflation was "unexpected" and therefore would have been protected by TIPS, had such securities been available?

User avatar
siamond
Posts: 4305
Joined: Mon May 28, 2012 5:50 am

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by siamond » Sat Jul 07, 2018 11:39 am

GAAP wrote:
Sat Jul 07, 2018 11:20 am
willthrill81 wrote:
Fri Jul 06, 2018 5:30 pm
Logically, I can see a strong case being made for splitting one's fixed income between TIPS and short-term Treasuries. The former provides protection from unexpected inflation, while the latter provides stability and liquidity.
Does anyone know how much of the inflation was "unexpected" and therefore would have been protected by TIPS, had such securities been available?
Some of us had long discussions on this forum on how to attempt to model TIPS if they had existed in the 70s, we read through corresponding research, developed some complicated spreadsheets, and long story short, we just couldn't figure out something half sensible & credible... :(

What is pretty clear though is that TIPS did become available as a direct reaction to the inflation/bond crisis of the late 70s and 80s.

thx1138
Posts: 784
Joined: Fri Jul 12, 2013 2:14 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by thx1138 » Sat Jul 07, 2018 12:18 pm

GAAP wrote:
Sat Jul 07, 2018 11:20 am
willthrill81 wrote:
Fri Jul 06, 2018 5:30 pm
Logically, I can see a strong case being made for splitting one's fixed income between TIPS and short-term Treasuries. The former provides protection from unexpected inflation, while the latter provides stability and liquidity.
Does anyone know how much of the inflation was "unexpected" and therefore would have been protected by TIPS, had such securities been available?
5 yr Treasury yield was 7.6% on 1/1/78

Inflation:

1978 - 9%
1979 - 13.3%
1980 - 12.5%
1981 - 8.9%
1982 - 3.8%

So I’d say the market clearly didn’t expect that inflation.

garlandwhizzer
Posts: 2057
Joined: Fri Aug 06, 2010 3:42 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by garlandwhizzer » Sat Jul 07, 2018 2:45 pm

siamond wrote:
My overall feedback to the OP is that, by overly focusing on a few cherry-picked starting points, one can end up with very flawed conclusions about their own strategy for the future.
1+ Excellent point, siamond. Great discussion also.

Graphs, like statistics, if tortured sufficiently will confess to the desired conclusion. Few things are as certain as advertised where investing is involved. Whether bonds are safer than stocks varies during a given time frame. Sometimes they are safer (1929 - 1941): sometimes they are riskier (1941 - 1982). Even things we all assume to be true like stocks reliably outperforming bonds over long time periods is open to question depending on the time period chosen. In the former 12 year period (1929 - 1941) they did not; in the later 40 year period (1941 - 1982) they did. This makes a strong case for a balanced portfolio unless you have massive wealth/totally secure income stream in which case you can do anything you wish and remain financially secure.

Historically there are long time periods in which either bonds or stocks, or in the case of stagflation both, suffer many consecutive years of low/absent/negative real returns. Predicting these periods in advance is unreliable. Having both in a portfolio in general smooths out the ride and may or may produce better returns over a given time period. In general there is a risk/reward tradeoff. In general stocks outperform bonds over the long run although how long that period is may be difficult to define. Current parameters do have some (about 40%) predictive value about future returns but, although better than nothing, that is less than a coin flip.

Investing is a tricky business with a high level of uncertainty. If you accept that up front and prepare yourself for shocks you are ahead of the game IMO. The journey for those with 100% stocks can be emotionally hazardous, but so can it be for those with 100% bonds for risk aversion. If you do so during the wrong time period, you can lose big time. The most important decision for investors to make is not which specific equity or bond funds/ETFS to use although this is what most of us obsess over.
rather the critical decision is the relative stock/bond portfolio allocation in accord with the individual's specific risk tolerance, financial goals and personal circumstances. Any low cost, broadly based, well-diversified, balanced portfolio--factor, cap-weighted index, or active--will get the job done if not abandoned during the inevitable downturns along the way.

Garland Whizzer

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by SimpleGift » Sat Jul 07, 2018 3:25 pm

garlandwhizzer wrote:
Sat Jul 07, 2018 2:45 pm
Graphs, like statistics, if tortured sufficiently will confess to the desired conclusion. Few things are as certain as advertised where investing is involved. Whether bonds are safer than stocks varies during a given time frame. Sometimes they are safer (1929 - 1941): sometimes they are riskier (1941 - 1982).
I find it hard to accept this notion that investors today should be weighing "bond crises" equally with "stock crises." Certainly, there have been long periods in the past (the 1940s, the 1970s) when high inflation seriously eroded the real value of bond-heavy portfolios. But this is why TIPS were created. Thankfully, we no longer live in a time when nominal bonds are our only choice — which makes past "bond crises" mostly an artifact of history in my view (and a bit oxymoronic today, if I may add).

In short, yes, bond-heavy portfolios were risky in the past — but stocks today are still as risky as they have ever been!
Cordially, Todd

grok87
Posts: 8501
Joined: Tue Feb 27, 2007 9:00 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by grok87 » Sun Jul 08, 2018 5:12 am

SimpleGift wrote:
Sat Jul 07, 2018 3:25 pm
garlandwhizzer wrote:
Sat Jul 07, 2018 2:45 pm
Graphs, like statistics, if tortured sufficiently will confess to the desired conclusion. Few things are as certain as advertised where investing is involved. Whether bonds are safer than stocks varies during a given time frame. Sometimes they are safer (1929 - 1941): sometimes they are riskier (1941 - 1982).
I find it hard to accept this notion that investors today should be weighing "bond crises" equally with "stock crises." Certainly, there have been long periods in the past (the 1940s, the 1970s) when high inflation seriously eroded the real value of bond-heavy portfolios. But this is why TIPS were created. Thankfully, we no longer live in a time when nominal bonds are our only choice — which makes past "bond crises" mostly an artifact of history in my view (and a bit oxymoronic today, if I may add).

In short, yes, bond-heavy portfolios were risky in the past — but stocks today are still as risky as they have ever been!
Agree that investors should pay more attention to stock crises than bond crises.

I'm probably one of the more vocal supporters of Tips. But i would not go as far as to say that TIPS have made "bond crises" a thing of the past or an artifact of history.

The risk of tips is that real interest rates will spike. That could happen for any number of reasons. US could default or be downgraded.
That's why liability matching is so important. If you can hold US government bonds to maturity you will likely be fine (barring default).

cheers,
grok
Keep calm and Boglehead on. KCBO.

User avatar
AtlasShrugged?
Posts: 613
Joined: Wed Jul 15, 2015 6:08 pm

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by AtlasShrugged? » Sun Jul 08, 2018 5:37 am

Simplegift.....Really like the history charts. A lot. :beer

Here is what struck me. What are the commonalities between the Great Depression and the 1970's Stagflation Era? And can we identify them? It seems that going heavy on bonds wound up doing much better. One poster mentioned something about huge run-ups in equities which is true of the Great Depression (Roaring Twenties, anyone?), but not the Stagflation Era.

To me, find the commonalities between the the two periods, and then we may have something very actionable indeed. What was monetary policy like before each period? Commodities? World trade climate? Asset backing of currencies?
“If you don't know, the thing to do is not to get scared, but to learn.”

AlohaJoe
Posts: 3919
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by AlohaJoe » Sun Jul 08, 2018 6:00 am

Whakamole wrote:
Sat Jul 07, 2018 12:45 am
Benjamin Roth's "The Great Depression: A Diary" (recommended a few times on this forum) really brought to life how it wasn't that nobody knew when the stock market would recover; they didn't know if the stock market would recover. It's easy for us to look at how the market did for 20 years but all anyone saw after the crash was the blue line dropping like a rock.
I guess we took away very different messages from Roth's book. When I read it, I came away with the understanding that Roth was completely convinced that the stock market would absolutely recover -- and in a relatively short period of time. He had no doubts on the matter at all. Nearly every page is him writing about how he wishes he had some money so he could buy stocks because he was so positive they would make a man rich in the next 2-3 years.

For those who haven't read the book, June 1931 was when he started his diary, so we don't have much real information on the period from September 1929 to June 1931. All he really has to say about stocks during that period is "Immediately after the 1929 crash the speculators rushed in to buy 'bargains' but were badly mistaken because the market kept going down [...] At present time [ed: 1931] the newspapers are urging people to buy these 'bargains' but opinion is much divided as to whether or not the bottom has been reached."
  • "I think the next five years will be a period of gradual recovery and that bargains in stocks and real estates will be available through the entire recovery period." (September 1, 1931)
  • "It all seems silly to me. Everything will work itself out without these radical changes." (September 9, 1931)
  • "I wish i had some money to invest." (September 28, 1931)
  • "So far I have not been badly hurt personally -- we are all well -- I still have faith in Youngstown and in the United States in spite of the dire prophecies that are heard on every side!" (October 17, 1931)
  • "A few weeks ago [an acquaintance of Roth's] placed a new mortgage on [his house] for $5000 and invested the proceeds in good stocks for long-term investment. I think in two or three years he will show a handsome profit. It is generally believed that good stocks and bonds can now be bought at very attractive prices." (December 11, 1931)
  • "That is why the depression period presents so many opportunities to the man who has the nerve to buy stocks and real estate and businesses when the outlook is blackest." (April 14, 1932)
  • "Of course [people who bought stocks last year] will come out ahead if they can hold on long enough" (May 9, 1932)
  • "Here again an ideal situation would have been created if she had placed all her cash in government securities and then reinvested during the panic when Republic and other stocks were way down. Such foresight was apparently given to a few people." (May 16, 1932)
  • "As far as I am personally concerned I think the company [Youngstown Sheet & Tube] will survive the depression and that the opportunity is now on hand to buy stock cheap." (May 26, 1932)
  • "I personally think the upturn will come soon -- possibly by next fall after the election [...] It seems to me an ideal time to invest money but I may be mistaken again." (June 10, 1932)
  • "I cannot agree with this dismal picture. [ed: that the city of Youngstown will run out of money and the courts, schools, etc will shut down] I think the turn for the better will come within the next six months and that there will be no such breakdown as is predicted." (June 13, 1932)
  • "The stock-market has been making up for past two weeks and newspaper headlines hail it as a sign of permanent betterment." (July 28, 1932)
  • "The stock market continued upward for two weeks now in the face of unfavorable semi-annual reports and predictions are again made freely that we have turned the corner." (August 5, 1932) [ed: in fact they had turned a permanent corner in the summer of 1932]
  • "In the last 30 days the stock market has given one of the most strenuous rallies in its history." (August 8, 1932)
  • "The stock market continues to move upward. This has continued for over 30 days except for a short pause of a few days last week." (August 20, 1932)
  • "The recovery will be slow because Europe also must recover but it is certain we will succeed." (August 29, 1932)
  • "The stock market on the contrary tripled its value during August in one of the quickest climbs ever witnessed. I believe this also established a record." (September 1, 1932)
  • "I believe it can be truly said that the man who has money during this depression to invest in the highest grade investment stocks and can hold on for 2 or 3 years will be the rich man of 1935." (September 8, 1932)
  • "I believe that a fortune in wisely selected stocks could be bought very cheaply just now." (December 31, 1932)
  • People who bought "bargains" in stocks in 1931 now find they were too quick and these same stocks are now selling at 1/3 of 1931 prices. If they can hold on long enough they will come out alright but it is a soul-trying period of waiting." (March 8, 1933)
Roth rarely talks about other people's opinions and when he does he uses extremely broad terms that make it hard to draw any real conclusions from. But Roth himself was convinced from the very beginning that everything would recover. Partly that was his innate American optimism and partly it was based on his readings about previous depressions which, though they sometimes took 5 years to resolve, always did resolve and return to booms.

While the economy isn't the same as the stock market (the stock market began recovering in the summer of 1932 but the economy didn't improve until 1933), when I read his diary I was struck by how his "at the time feelings" of turning a corner (and his reports of newspaper reports to the same effect) lined up pretty well with the market beginning its recovery.

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by SimpleGift » Sun Jul 08, 2018 7:57 am

grok87 wrote:
Sun Jul 08, 2018 5:12 am
I'm probably one of the more vocal supporters of Tips. But i would not go as far as to say that TIPS have made "bond crises" a thing of the past or an artifact of history.

The risk of tips is that real interest rates will spike. That could happen for any number of reasons. US could default or be downgraded.
That's why liability matching is so important. If you can hold US government bonds to maturity you will likely be fine (barring default).
Good point about the risk of rising real interest rates to TIPS in bond-heavy portfolios. While I believe the risk of permanent U.S. government default is almost unimaginable in today's world (but not impossible of course), there have been episodes in the past of spiking real interest rates, most recently in the early 1980s (chart below):
As you suggest, the best way to achieve maximum safety with bonds today is holding individual TIPS to maturity.
Cordially, Todd

User avatar
siamond
Posts: 4305
Joined: Mon May 28, 2012 5:50 am

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by siamond » Sun Jul 08, 2018 8:21 am

AlohaJoe wrote:
Sun Jul 08, 2018 6:00 am
I guess we took away very different messages from Roth's book. When I read it, I came away with the understanding that Roth was completely convinced that the stock market would absolutely recover -- and in a relatively short period of time. He had no doubts on the matter at all. Nearly every page is him writing about how he wishes he had some money so he could buy stocks because he was so positive they would make a man rich in the next 2-3 years.
Yep, I got exactly the same perception when reading the book, which surprised me. In all fairness, he still had a job though, even if he was struggling at times. It is much easier to see a stock crisis as an opportunity when you have reasonably stable income.

Another thing that struck me was Roth's tracking of prices for basic goods. Geez, did this go up and down. Anybody telling you that they can ride inflation/deflation by just doing some level of product substitution really should read this book... Purchasing power is a real, inflation-adjusted, quantity, period.

User avatar
willthrill81
Posts: 6436
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: The Worst U.S. Portfolio Disasters — A Historical Comparison

Post by willthrill81 » Sun Jul 08, 2018 9:34 am

AlohaJoe wrote:
Sat Jul 07, 2018 12:06 am
marcopolo wrote:
Fri Jul 06, 2018 6:37 pm
columbia wrote:
Fri Jul 06, 2018 6:11 pm
It seems like a good argument for a 50/50 portfolio over the long run.
Maybe.

But, i fear that looking at the data starting just before the meltdown periods may lead to the wrong conclusions.
Not sure about the oil/stagflation shock of the 70s, but both the great depression and the tech bubble were proceeded by huge run up in equities.
To elaborate on marcopolo's point, here's a series of charts showing the years before 1929.
Thanks for the post. Those charts seem to suggest to me that having a 'side' portfolio with an extra year or two of expenses at the point of retirement might help retirees to avoid having to withdraw from a distressed portfolio in the worst ravages of a downturn. ERN recently studied this and found it to be surprisingly effective for the worst retirement starting years in the U.S., 1929 and 1966.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Post Reply