100% Equities Challenge Scenario

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
hoops777
Posts: 4609
Joined: Sun Apr 10, 2011 12:23 pm
Location: Behind the 3 point line

Re: 100% Equities Challenge Scenario

Post by hoops777 »

EliteZags wrote: Mon Apr 01, 2024 4:01 pm so basically if you assume having to retire within 10 years then having 2 major historic level market crashes within the first few years of retirement, then it makes sense to have bonds

ok cool


the rest of us will be over here building wealth
You may be right…..and you may not.
When stocks are at all time highs and had a great 14 year run the enthusiasm is always at its highest.
We had a pandemic, wars going on, inflation and nothing can stop the market, or so it seems. Hmmmmm.
K.I.S.S........so easy to say so difficult to do.
StillGoing
Posts: 370
Joined: Mon Nov 04, 2019 3:43 am
Location: U.K.

Re: 100% Equities Challenge Scenario

Post by StillGoing »

abc132 wrote: Mon Apr 01, 2024 5:26 pm
StillGoing wrote: Mon Apr 01, 2024 4:20 am While the extensive lifecycle thread has been cited upthread, it might be worth noting that there are a very small number (*see below) of historical 40 year accumulation periods starting between about 1885 and 1905 where holding 100% stocks did not result in the largest portfolio at accumulation.

Using the Simba dataset (TSM and TBM), with 1 inflation-adjusted dollar invested at the start of each year, the accumulation period starting in 1902 is given for stock allocations of 0, 20, 40, 60, 80, and 100% (where appropriate, rebalanced each year).

Interesting to note that there's not a lot of difference in the first twenty years (possibly because the amount saved was a significant fraction of the portfolio value), and that the run up that stocks enjoyed before the crash at the end of the 20s was wiped out by that crash.

The terminal values (in real $) were, for stock allocations of 0% to 100%: 78.5, 91.3, 101.4, 107.0, 106.9, and 100.6

i.e., the value peaked somewhere between a stock allocation of 60% and 80% (FWIW, redoing the analysis with finer steps in stock allocation indicated the peak lay close to 70%).

* The fraction of retirements where the 40 year terminal value was greater for a lower stock allocation than for an allocation of 100% was 0%, 0%, 1.8%, 3.6%, and 7.1% for allocations of 0%, 20%, 40%, 60%, and 80%, respectively. I should also point out that on the occasions where 100% allocation did not give the highest amount, the difference was not that great (about $6) compared to the 25th percentile, where 100% stocks beat 80% by $12, 60% by $30, and 40% by $48.

Of course, this is for domestic US stocks, adding international stocks may, or may not, have changed these results.

cheers
StillGoing
The worst decumulation sequences are typically some of the best accumulation sequences. The worst accumulation sequences typically have great returns in decumulation. This is what my backtesting shows - in the rare case accumulation is worse you still tend to get rewarded in decumulation.

The best odds tend to be stock heavy accumulation and risk reduction for decumulation. You simply don't know ahead of time if you are a good or bad accumulation sequence so taking the 3.6% chance of outperformance with 60/40 and 96.4% chance of underperformance is typically an increase in your personal risk.

When you include accumulation and decumulation there is not a great risk based argument for something like 60/40 in accumulation. Of course we can take on more risk for behavioral reasons but the 60/40 crowd may be taking more risk than the 80/20 crowd.
For the US, I'd agree that the best odds were obtained by having a high stock allocation. However, taking Japan since 1950 as an extreme example, gives a different picture (returns from macrohistory.net - I note that this database currently only goes as far as 2020) for a 40 year accumulation period.

Image

The 100% stock allocation often gave the worst portfolio value (note the logarithmic y-axis) - statistically, an 80% allocation gave a higher portfolio value than 100% in 67% of cases and for allocations of 40% and 60% in 60% of cases.

Of course, this ignores your point that bad accumulation periods can be followed by good decumulation periods (and vice versa) and that for Japanese investors who held international stocks at cap weighting the outcomes would have been better for high stock allocations than a purely domestic portfolio.

What would be really interesting to model would be the psychological effect of strong declines in portfolio value on the ability of people to 'stick with the plan'. Even on these boards, there were a number of capitulation (Plan B?) threads during the global financial crisis and cases where people really did sell out of stocks and consequently struggled to get back into the market.

cheers
StillGoing
User avatar
TimeIsYourFriend
Posts: 569
Joined: Sun Dec 17, 2023 11:19 am

Re: 100% Equities Challenge Scenario

Post by TimeIsYourFriend »

steve r wrote: Mon Apr 01, 2024 6:29 pm
TimeIsYourFriend wrote: Sun Mar 31, 2024 12:39 pm
steve r wrote: Sun Mar 31, 2024 11:25 am
TimeIsYourFriend wrote: Sun Mar 31, 2024 9:49 am The 3-fund portfolio with nominal bonds looks really good from a tailwind of falling interest rates for decades. The paper in question doesn't mean you have to adopt 100% equities to get anything out of the conclusion. It could be that you exchange nominals for TIPS for example, although, not as optimal as 100% equities but that doesn't matter. What matters is gliding to a nominal bond heavy portfolio in retirement can get murdered with high inflation and the longer you hold it, the more likely that will happen.
This is probably true in theory and in the long run. This was NOT true in rising rate / inflationary 2022-2023. Intermediate term treasures did better than total bond and TIPs. (presumably because the real return portion of TIPs went from negative to positive 2 percentage points +/-).

Obviously there are other ways to hedge inflation. Own stuff (i.e. a home) or iBonds are two examples.
I don’t expect any portfolio hedges everything over a 2 year period.
Good point. However, it is the only inflationary period where we could observe how TIPs performed. So, I guess it depends on how much faith you place on their performance during inflation matching one's expectations. I would expect they do well in such a protracted scenario.
I'm seeing a different result when I run it through PV: https://www.portfoliovisualizer.com/bac ... t0V8N7MhGg

Vanguard's intermediate TIPS beat Vanguard Total Bond and Vanguard Intermediate Treasures from Jan 2021 to present, from Jan 2021 to Dec 2023, from Jan 2021 to Jan 2023. This is despite the TIPs fund having a longer avg. duration of 6.8 years vs 5.3 of the other two.

Portfolio performance statistics (Jan 2021 to present)

Code: Select all

Portfolio	                       Initial	Final	CAGR	Stdev	Best Year	Worst Year	Max. Drawdown	Sharpe Ratio	Sortino Ratio	Market Correlation
Vanguard Interm-Term Treasury Inv	$10,000	$9,041 	-3.05% 	5.73%	4.07%	-10.43%	-14.20% 	-0.97	-1.15	0.56
Vanguard Total Bond Market Index Adm	$10,000	$8,955 	-3.34% 	6.98%	5.70%	-13.16%	-17.16% 	-0.83	-1.03	0.67
Vanguard Inflation-Protected Secs Adm	$10,000	$9,653 	-1.08% 	6.85%	5.68%	-11.89%	-13.42% 	-0.49	-0.61	0.74
"Time is your friend; impulse is your enemy." - John C. Bogle
Target2019
Posts: 918
Joined: Sat Mar 03, 2007 4:30 pm

Re: 100% Equities Challenge Scenario

Post by Target2019 »

Investor 1: Is a "Boglehead" style investor. They split their portfolio into 60% equities, 20% Bond Index Funds, and 20% TIPS.

Investor 1 rebalances once per year maintain a strict 75% in equities and 25% in bond index funds every year, and withdraws using the same proportion. Investor 1 holds TIPS as a standalone investment, initially in a TIPS fund, and then, at retirement, in a TIPS ladder. This means that Investor 1 will not rebalance the TIPS portion of their portfolio.
Which is it? 60/40 or 75/25 for Investor 1?
User avatar
firebirdparts
Posts: 4447
Joined: Thu Jun 13, 2019 4:21 pm
Location: Southern Appalachia

Re: 100% Equities Challenge Scenario

Post by firebirdparts »

aj76er wrote: Sat Mar 30, 2024 8:18 pm What you are not taking into account is that the 100% equity investor is going to have a much larger starting portfolio after many decades of accumulating. So it is not fair to assume both portfolios start at 25x. The 100% equity portfolio would likely be much larger at the start of retirement.
This X 1000, obviously. If you're going to make up a single scenario, though, you can do whatever you want. Your choice to start in 2025, when a 100% equity investor would be light years ahead.

There is sequence of return risk and there is sequence of return bonanza. It's not popular to recognize bonanza, but risk is a constant source of human attention.
This time is the same
Morse Code
Posts: 380
Joined: Wed Mar 14, 2007 12:02 pm
Location: Northern Michigan

Re: 100% Equities Challenge Scenario

Post by Morse Code »

firebirdparts wrote: Tue Apr 02, 2024 7:34 am
aj76er wrote: Sat Mar 30, 2024 8:18 pm What you are not taking into account is that the 100% equity investor is going to have a much larger starting portfolio after many decades of accumulating. So it is not fair to assume both portfolios start at 25x. The 100% equity portfolio would likely be much larger at the start of retirement.
This X 1000, obviously. If you're going to make up a single scenario, though, you can do whatever you want. Your choice to start in 2025, when a 100% equity investor would be light years ahead.

There is sequence of return risk and there is sequence of return bonanza. It's not popular to recognize bonanza, but risk is a constant source of human attention.
For sure. This scenario could only be possible if both investors were 100% equity until retirement and then one investor de-risked on retirement day. I have suggested this would not be a bad strategy, assuming one could choose when to retire, but I have been shouted down on the forum for this.
Livin' the dream
SnowBog
Posts: 4742
Joined: Fri Dec 21, 2018 10:21 pm

Re: 100% Equities Challenge Scenario

Post by SnowBog »

Target2019 wrote: Tue Apr 02, 2024 6:58 am
Investor 1: Is a "Boglehead" style investor. They split their portfolio into 60% equities, 20% Bond Index Funds, and 20% TIPS.

Investor 1 rebalances once per year maintain a strict 75% in equities and 25% in bond index funds every year, and withdraws using the same proportion. Investor 1 holds TIPS as a standalone investment, initially in a TIPS fund, and then, at retirement, in a TIPS ladder. This means that Investor 1 will not rebalance the TIPS portion of their portfolio.
Which is it? 60/40 or 75/25 for Investor 1?
I had the same question at first...

It's actually both - at the same time.

If you re-read it, you'll note they are not rebalancing TIPS. That initial 20% allocation to TIPS "stands alone" (won't be rebalanced).

Which leaves 60% in equities and 20% into Bond index funds, making up the remaining 80%. Or viewed in isolation, equities make up 75% of the remaining allocation (60/80) with the bond funds being the last 25% (20/80) - once you remove the TIPS funds (which again "stand alone").
abc132
Posts: 2471
Joined: Thu Oct 18, 2018 1:11 am

Re: 100% Equities Challenge Scenario

Post by abc132 »

StillGoing wrote: Tue Apr 02, 2024 4:27 am For the US, I'd agree that the best odds were obtained by having a high stock allocation. However, taking Japan since 1950 as an extreme example, gives a different picture (returns from macrohistory.net - I note that this database currently only goes as far as 2020) for a 40 year accumulation period.

Image

The 100% stock allocation often gave the worst portfolio value (note the logarithmic y-axis) - statistically, an 80% allocation gave a higher portfolio value than 100% in 67% of cases and for allocations of 40% and 60% in 60% of cases.

Of course, this ignores your point that bad accumulation periods can be followed by good decumulation periods (and vice versa) and that for Japanese investors who held international stocks at cap weighting the outcomes would have been better for high stock allocations than a purely domestic portfolio.

What would be really interesting to model would be the psychological effect of strong declines in portfolio value on the ability of people to 'stick with the plan'. Even on these boards, there were a number of capitulation (Plan B?) threads during the global financial crisis and cases where people really did sell out of stocks and consequently struggled to get back into the market.

cheers
StillGoing
If I am reading your graph correctly, your extreme example is a 70% lower stock portfolio right before the Nikkei share average (.N225) started 1980 at 6,867 and ended the decade at 38,915. So we clearly had major stock outperformance by living past age 60 for the accumulator that started work at 20. In this case I see the 70% lower portfolio value as a non-issue because you would have had to die while working to end up behind and portfolio failure was thus a non-issue.

As to psychology, I am almost certain conservative investors capitulate more often - our comments are typically a reflection of ourselves and their comments are always very fearful. "Evidence" against stocks is never a fair assessment of comparing action A vs action B. A common example is not tying accumulation and decumulation together and instead showing bad decumulation or bad accumulation results which will misrepresent what actually happens.

Our risks is what actually happens in accumulation and decumulation and any analysis that starts with equal portfolios while searching for bad decumulation sequences is biased to the point of misrepresenting what actually happens. I suspect we have people with accumulation AA's based on these decumulation studies. It's unfortunate we are not given information representative of what actually happens.
triz
Posts: 67
Joined: Fri Jul 21, 2023 4:12 pm

Re: 100% Equities Challenge Scenario

Post by triz »

I was in the 100% equities camp, but I think the risk level for a crash is higher than what a lot of people are thinking. Obviously stay the course – everyone should have a portfolio they’d feel comfortable with in that scenario. It's just the attitude so many people have around the market now -- the greed levels are quite high IMO. That doesn't mean I know when a crash is going to happen, but I do feel good about trend following strategies/funds going forward to hopefully (not guaranteed to work) limit downside risk and I appreciate that so many people cannot trend follow cause of the extreme tracking variance.
alluringreality
Posts: 1520
Joined: Tue Nov 12, 2019 9:59 am

Re: 100% Equities Challenge Scenario

Post by alluringreality »

firebirdparts wrote: Tue Apr 02, 2024 7:34 am
aj76er wrote: Sat Mar 30, 2024 8:18 pm What you are not taking into account is that the 100% equity investor is going to have a much larger starting portfolio after many decades of accumulating. So it is not fair to assume both portfolios start at 25x. The 100% equity portfolio would likely be much larger at the start of retirement.
This X 1000, obviously. If you're going to make up a single scenario, though, you can do whatever you want. Your choice to start in 2025, when a 100% equity investor would be light years ahead.
Well, there was a recent thread where someone 10 years from retirement asked if they should increase equity, based on the paper. Granted the context of the paper amounts to a couple with 40 years for a 10% base rate savings with some uncertainty, although seemingly never withdrawing before retirement, yet realistically people are simply going to ask about out of simulation implications. Some comments may encourage even later revisions into their strategy (Stocks/I), for example in section 5.2 the authors claim:

"For investors who adopt constant-weight strategies and have no ability to short an asset class, the optimal weight in bonds is 0%. Strategies that hold 5% or 10% in bonds only during the retirement period produce negligible gains relative to Stocks/I, and Stocks/I dominates any strategy with a retirement bond weight of 15% or higher. "

"When domestic equity does poorly, bonds and bills also tend to
do poorly. Fixed income does not offer an adequate safe haven against poor equity outcomes over the long run.
"
viewtopic.php?p=7793435#p7793435
45% US Indexes, 25% Ex-US Indexes, 30% Fixed Income - Buy & Hold
User avatar
steve r
Posts: 1310
Joined: Mon Feb 13, 2012 7:34 pm
Location: Connecticut

Re: 100% Equities Challenge Scenario

Post by steve r »

TimeIsYourFriend wrote: Tue Apr 02, 2024 6:33 am
steve r wrote: Mon Apr 01, 2024 6:29 pm
TimeIsYourFriend wrote: Sun Mar 31, 2024 12:39 pm
steve r wrote: Sun Mar 31, 2024 11:25 am
TimeIsYourFriend wrote: Sun Mar 31, 2024 9:49 am The 3-fund portfolio with nominal bonds looks really good from a tailwind of falling interest rates for decades. The paper in question doesn't mean you have to adopt 100% equities to get anything out of the conclusion. It could be that you exchange nominals for TIPS for example, although, not as optimal as 100% equities but that doesn't matter. What matters is gliding to a nominal bond heavy portfolio in retirement can get murdered with high inflation and the longer you hold it, the more likely that will happen.
This is probably true in theory and in the long run. This was NOT true in rising rate / inflationary 2022-2023. Intermediate term treasures did better than total bond and TIPs. (presumably because the real return portion of TIPs went from negative to positive 2 percentage points +/-).

Obviously there are other ways to hedge inflation. Own stuff (i.e. a home) or iBonds are two examples.
I don’t expect any portfolio hedges everything over a 2 year period.
Good point. However, it is the only inflationary period where we could observe how TIPs performed. So, I guess it depends on how much faith you place on their performance during inflation matching one's expectations. I would expect they do well in such a protracted scenario.
I'm seeing a different result when I run it through PV: https://www.portfoliovisualizer.com/bac ... t0V8N7MhGg

Vanguard's intermediate TIPS beat Vanguard Total Bond and Vanguard Intermediate Treasures from Jan 2021 to present, from Jan 2021 to Dec 2023, from Jan 2021 to Jan 2023. This is despite the TIPs fund having a longer avg. duration of 6.8 years vs 5.3 of the other two.

Portfolio performance statistics (Jan 2021 to present)

Code: Select all

Portfolio	                       Initial	Final	CAGR	Stdev	Best Year	Worst Year	Max. Drawdown	Sharpe Ratio	Sortino Ratio	Market Correlation
Vanguard Interm-Term Treasury Inv	$10,000	$9,041 	-3.05% 	5.73%	4.07%	-10.43%	-14.20% 	-0.97	-1.15	0.56
Vanguard Total Bond Market Index Adm	$10,000	$8,955 	-3.34% 	6.98%	5.70%	-13.16%	-17.16% 	-0.83	-1.03	0.67
Vanguard Inflation-Protected Secs Adm	$10,000	$9,653 	-1.08% 	6.85%	5.68%	-11.89%	-13.42% 	-0.49	-0.61	0.74
Good.

I picked 2022 with Simba which is annual data, but inflation spiked mid year 2021. So your dates are preferred. (I just looked it up, it hit over 5 percent in April 2021, and started a steady rise from 5 percent in October, peaking the following June at over 9 percent.) If you end in June 2022 when inflation started to decline, TIPS outperformance is even greater -- actually much greater. Sort of cherry picking the dates, but based on inflation rates, not performance.
"Owning the stock market over the long term is a winner's game. Attempting to beat the market is a loser's game. ..Don't look for the needle in the haystack. Just buy the haystack." Jack Bogle
Topic Author
SantaClaraSurfer
Posts: 751
Joined: Tue Feb 19, 2019 10:09 am

Re: 100% Equities Challenge Scenario

Post by SantaClaraSurfer »

Thanks to everyone who has responded. Disagreement (even it has been blunt sometimes) helps make arguments stronger.

I've rerun the numbers to provide an alternate scenario that addresses some of the arguments I've received.

Scenario 2 Givens:

Average Annual Inflation: 4.5% (up from 3.2%)
Bond Index Fund Nominal / Real Return: 4% nominal / -.5% real. (down -1.7% real)
Equities (outside of the challenge years) have a 10.25% nominal and a 5.75% real return (up 1.5% and .25%)
TIPS are available at a 1.5% fixed rate (down -.5%)
The "Challenge Years" or "Stress Test" is significantly less stark: it is now only 2 Bear Markets of -25% within five years of one another and the full recovery, still in the eighth year, is to +33% (an increase of 33%, and the two shocks only add up to -50% versus -67%)

So our two investors (who both decided to COAST FIRE at 25x at age 55, btw, with one holding 100% equities and the other switching their portfolio to 60% Equities, 20% Bond Index Funds, and 20% TIPS) are now facing more inflation, significantly less equity downside in the stress test, much weaker bond index fund returns (-.5% real) and enjoy more robust equity returns through the whole period.

Let's see where they end up at age 75:

Code: Select all

Year	Investor 1 Total	Investor 2 Total
2025	$1,000,000.00	$1,000,000.00
2026	$1,042,507.50	$1,052,887.50
2027	$1,086,904.40	$1,108,572.09
2028	$1,133,279.16	$1,167,201.69
2029	$1,181,724.60	$1,228,932.07
2030	$1,232,338.14	$1,293,927.22
2031	$1,285,222.06	$1,362,359.79
2032	$1,377,709.18	$1,434,411.60
2033	$1,436,874.57	$1,510,274.04
2034	$1,216,816.31	$1,081,733.78
2035	$1,281,723.38	$1,157,022.45
2036	$1,291,040.97	$1,166,202.52
2037	$1,376,272.13	$1,242,581.92
2038	$1,403,217.17	$1,278,998.97
2039	$1,170,287.38	$877,883.01
2040	$1,174,909.31	$871,440.43
2041	$1,351,808.54	$1,126,915.85
2042	$1,373,519.85	$1,148,315.61
2043	$1,403,031.37	$1,170,847.15
2044	$1,427,301.44	$1,194,570.33
2045	$1,452,831.81	$1,219,548.17
Pretty much a similar place as Scenario 1, despite the new parameters favoring equities and boosting returns.

Our Boglehead Investor 1 holds enough equities, combined with 20% TIPS (combatting the strong average annual inflation rate) to offset the drag of the negative returns for bond index funds. Investor 2, holding 100% equities (I like to think of them holding a Vanguard Total World fund to keep it simple and in line with Prof. Cederburg) still gets hit very hard in early retirement withdrawing from a 100% equity portfolio. They fall significantly behind Investor 1, and don't recover the gap even by age 75.

Both portfolios survive Scenario 2, but it's easy to see that Investor 1 had the much easier 20 year experience; their portfolio never drops below $1.15 million after year 3, whereas Investor 2 fell from $1.5 million two years prior to retirement to $877,000 four years after retirement. As many have pointed out, that's more or less the Sequence of Returns Risk that all 100% equity investors may face at retirement; a normal bear market is painful, but two bear markets in close succession is a stark outcome. It may take well into their 80s, and assuming no further bear markets, for Investor 2 to pass Investor 1 in this scenario, despite the upside of holding 100% stocks.

Yes, if you backtest 100% equities it can look very tempting to take Prof. Cederburg's findings as your new Asset Allocation.

My advice, after running this test for myself, is to construct your own forward-looking scenario in a spreadsheet that you control.

You can use that forward projection to think about how you would handle the stress test of your choice: inflation, poor nominal bond performance, negative TIPS rates, a market crash with a delayed recovery.

What I think you'll see is that, given the all-time market highs we've been enjoying, any forward-looking stress test you come up with will teach you more about how you feel about your own asset allocation than simply running a backtest in portfolio visualizer based on 2024 numbers.

At the end of the day, your asset allocation is your choice, and how you feel about it is the most important factor to consider.
abc132
Posts: 2471
Joined: Thu Oct 18, 2018 1:11 am

Re: 100% Equities Challenge Scenario

Post by abc132 »

SantaClaraSurfer wrote: Tue Apr 02, 2024 3:08 pm Thanks to everyone who has responded. Disagreement (even it has been blunt sometimes) helps make arguments stronger.

I've rerun the numbers to provide an alternate scenario that addresses some of the arguments I've received.

Scenario 2 Givens:

Average Annual Inflation: 4.5% (up from 3.2%)
Bond Index Fund Nominal / Real Return: 4% nominal / -.5% real. (down -1.7% real)
Equities (outside of the challenge years) have a 10.25% nominal and a 5.75% real return (up 1.5% and .25%)
TIPS are available at a 1.5% fixed rate (down -.5%)
The "Challenge Years" or "Stress Test" is significantly less stark: it is now only 2 Bear Markets of -25% within five years of one another and the full recovery, still in the eighth year, is to +33% (an increase of 33%, and the two shocks only add up to -50% versus -67%)

So our two investors (who both decided to COAST FIRE at 25x at age 55, btw, with one holding 100% equities and the other switching their portfolio to 60% Equities, 20% Bond Index Funds, and 20% TIPS) are now facing more inflation, significantly less equity downside in the stress test, much weaker bond index fund returns (-.5% real) and enjoy more robust equity returns through the whole period.

Let's see where they end up at age 75:

Code: Select all

Year	Investor 1 Total	Investor 2 Total
2025	$1,000,000.00	$1,000,000.00
2026	$1,042,507.50	$1,052,887.50
2027	$1,086,904.40	$1,108,572.09
2028	$1,133,279.16	$1,167,201.69
2029	$1,181,724.60	$1,228,932.07
2030	$1,232,338.14	$1,293,927.22
2031	$1,285,222.06	$1,362,359.79
2032	$1,377,709.18	$1,434,411.60
2033	$1,436,874.57	$1,510,274.04
2034	$1,216,816.31	$1,081,733.78
2035	$1,281,723.38	$1,157,022.45
2036	$1,291,040.97	$1,166,202.52
2037	$1,376,272.13	$1,242,581.92
2038	$1,403,217.17	$1,278,998.97
2039	$1,170,287.38	$877,883.01
2040	$1,174,909.31	$871,440.43
2041	$1,351,808.54	$1,126,915.85
2042	$1,373,519.85	$1,148,315.61
2043	$1,403,031.37	$1,170,847.15
2044	$1,427,301.44	$1,194,570.33
2045	$1,452,831.81	$1,219,548.17
Pretty much a similar place as Scenario 1, despite the new parameters favoring equities and boosting returns.

Our Boglehead Investor 1 holds enough equities, combined with 20% TIPS (combatting the strong average annual inflation rate) to offset the drag of the negative returns for bond index funds. Investor 2, holding 100% equities (I like to think of them holding a Vanguard Total World fund to keep it simple and in line with Prof. Cederburg) still gets hit very hard in early retirement withdrawing from a 100% equity portfolio. They fall significantly behind Investor 1, and don't recover the gap even by age 75.

Both portfolios survive Scenario 2, but it's easy to see that Investor 1 had the much easier 20 year experience; their portfolio never drops below $1.15 million after year 3, whereas Investor 2 fell from $1.5 million two years prior to retirement to $877,000 four years after retirement. As many have pointed out, that's more or less the Sequence of Returns Risk that all 100% equity investors may face at retirement; a normal bear market is painful, but two bear markets in close succession is a stark outcome. It may take well into their 80s, and assuming no further bear markets, for Investor 2 to pass Investor 1 in this scenario, despite the upside of holding 100% stocks.

Yes, if you backtest 100% equities it can look very tempting to take Prof. Cederburg's findings as your new Asset Allocation.

My advice, after running this test for myself, is to construct your own forward-looking scenario in a spreadsheet that you control.

You can use that forward projection to think about how you would handle the stress test of your choice: inflation, poor nominal bond performance, negative TIPS rates, a market crash with a delayed recovery.

What I think you'll see is that, given the all-time market highs we've been enjoying, any forward-looking stress test you come up with will teach you more about how you feel about your own asset allocation than simply running a backtest in portfolio visualizer based on 2024 numbers.

At the end of the day, your asset allocation is your choice, and how you feel about it is the most important factor to consider.
If you can time a switch into bonds before a downturn without any loss of accumulation no doubt you can be ahead for a good while. The 100% stock investor with great returns certainly has the option to de-risk. In my book that is really an argument for the value of high stock allocation in accumulation. Interestingly enough, retirement portfolio size matters more than AA.

If you want a comfortable decumulation you should plan on 30x expenses or hitting 25-27x within a decline. Interestingly enough, a 50/50 portfolio won't recover much faster than a 100/0 portfolio despite the lower decline (think 10 years vs 10.5 years to recover). Start investor 2 at 1.2 million and take a look at the results (30x is 20% bigger than 25x). Pretty comfortable!

Want a 100/0 AA? It's pretty simple, just work until 30x and ignore the 25x backtests with equal retirement portfolios. Add some bonds if you want to de-risk. Even 10-20% bonds will really improve your portfolio safety for a 30x portfolio.
lostdog
Posts: 5378
Joined: Thu Feb 04, 2016 1:15 pm

Re: 100% Equities Challenge Scenario

Post by lostdog »

Catalina25 wrote: Mon Apr 01, 2024 9:18 pm
SantaClaraSurfer wrote: Sun Mar 31, 2024 10:11 am The alternative, staying the course and gritting it out with 100% stocks, may actually have paid off in the past per the research, but that's asking a great deal of patience from, as you put it, "the average schmo."
Last Friday, I finally forced myself to throttle back my portfolio from 100% equities, which I'd been riding for quite a long time, to what I've come to realize is a more rational, diverse investment appproach now that I'm retired (actually just over two years ago at 59). Not gonna lie, the last 15 years were quite a ride with the S&P rocking nominal total returns over 15%, and it was with great angst that I reallocated down to my now 75/25 portflio. I rode 100% equities over the last few decades as my fear of not accumulating enough greatly exceeded my concern over normal market fluctuations/corrections.

I'm not sure if I can get down to the prevailing wisdom of the 60/40 portolio as I like having a bit of growth still as a hedge against inflation, but who knows, maybe if a drink more of the Kool-Aid I will get there.

And a shout out to the people of this forum for helping me see the light!
Grats! for making the right decision. With all these "why not 100% equities" posts lately, I bet it would be mentally hard to avoid the greed.
World Stocks(VT)-70% || World Bonds(BNDW)-30%
StillGoing
Posts: 370
Joined: Mon Nov 04, 2019 3:43 am
Location: U.K.

Re: 100% Equities Challenge Scenario

Post by StillGoing »

abc132 wrote: Tue Apr 02, 2024 8:42 am
StillGoing wrote: Tue Apr 02, 2024 4:27 am For the US, I'd agree that the best odds were obtained by having a high stock allocation. However, taking Japan since 1950 as an extreme example, gives a different picture (returns from macrohistory.net - I note that this database currently only goes as far as 2020) for a 40 year accumulation period.

Image

The 100% stock allocation often gave the worst portfolio value (note the logarithmic y-axis) - statistically, an 80% allocation gave a higher portfolio value than 100% in 67% of cases and for allocations of 40% and 60% in 60% of cases.

Of course, this ignores your point that bad accumulation periods can be followed by good decumulation periods (and vice versa) and that for Japanese investors who held international stocks at cap weighting the outcomes would have been better for high stock allocations than a purely domestic portfolio.

What would be really interesting to model would be the psychological effect of strong declines in portfolio value on the ability of people to 'stick with the plan'. Even on these boards, there were a number of capitulation (Plan B?) threads during the global financial crisis and cases where people really did sell out of stocks and consequently struggled to get back into the market.

cheers
StillGoing
If I am reading your graph correctly, your extreme example is a 70% lower stock portfolio right before the Nikkei share average (.N225) started 1980 at 6,867 and ended the decade at 38,915. So we clearly had major stock outperformance by living past age 60 for the accumulator that started work at 20. In this case I see the 70% lower portfolio value as a non-issue because you would have had to die while working to end up behind and portfolio failure was thus a non-issue.

As to psychology, I am almost certain conservative investors capitulate more often - our comments are typically a reflection of ourselves and their comments are always very fearful. "Evidence" against stocks is never a fair assessment of comparing action A vs action B. A common example is not tying accumulation and decumulation together and instead showing bad decumulation or bad accumulation results which will misrepresent what actually happens.

Our risks is what actually happens in accumulation and decumulation and any analysis that starts with equal portfolios while searching for bad decumulation sequences is biased to the point of misrepresenting what actually happens. I suspect we have people with accumulation AA's based on these decumulation studies. It's unfortunate we are not given information representative of what actually happens.
Just to clarify, the graph was showing the final portfolio value after a 40 year accumulation period that started in the marked year (i.e. the results at the extreme left of the graph are for an accumulation period starting in 1950 and ending in 1990 with 1 real currency unit (Yen in this case) invested at the beginning of each year. My apologies - I didn't explain this at all well.

I think the point is that while 100% stocks for a US investor has historically given the highest accumulated portfolio in the vast majority of cases, there are a small number of starting years where this is not the case. Furthermore, in those cases, the shortfall from holding 100% stocks was small compared to the outperformance in the vast majority of cases. There are also cases outside the US where holding 100% domestic stocks did not result in the largest accumulated portfolio and in the Japan case those starting in the late 1960s onwards were particularly poor for 100% stocks.

I'd agree - from a historical point of view, a lifecycle analysis covering a period of roughly 30-40 years in the accumulation phase and 30-40 years in retirement (e.g. Pfau, Safe Savings rates: A new Approach to retirement Planning over the life cycle, 2011 - although he used 30 years for each phase) is an excellent approach. However, for Monte-Carlo approaches or bootstrapping, the good sequence followed by bad sequence that has been evident historically (at least in the US) may no longer hold for all lifecycles.

cheers
StillGoing
User avatar
papiper
Posts: 168
Joined: Tue May 28, 2013 10:36 am

Re: 100% Equities Challenge Scenario

Post by papiper »

So... you use "average equity returns" then artificially have two crashes - and the recoveries are at the average return? If you impose a crash, and then ignore the speed that almost all equities recover soon after. So on the equity side you handicapping returns much more than history would support. Of course the equity side looks worse if you make those two negative assumption.

If you want equities too look risky, just look at all the downside and ignore any upside performance. If you leave the data alone - 100% equities deliver more over the long haul.
abc132
Posts: 2471
Joined: Thu Oct 18, 2018 1:11 am

Re: 100% Equities Challenge Scenario

Post by abc132 »

StillGoing wrote: Wed Apr 03, 2024 4:08 am Just to clarify, the graph was showing the final portfolio value after a 40 year accumulation period that started in the marked year (i.e. the results at the extreme left of the graph are for an accumulation period starting in 1950 and ending in 1990 with 1 real currency unit (Yen in this case) invested at the beginning of each year. My apologies - I didn't explain this at all well.
Thanks! That is helpful.
StillGoing wrote: Wed Apr 03, 2024 4:08 am I think the point is that while 100% stocks for a US investor has historically given the highest accumulated portfolio in the vast majority of cases, there are a small number of starting years where this is not the case. Furthermore, in those cases, the shortfall from holding 100% stocks was small compared to the outperformance in the vast majority of cases. There are also cases outside the US where holding 100% domestic stocks did not result in the largest accumulated portfolio and in the Japan case those starting in the late 1960s onwards were particularly poor for 100% stocks.

I'd agree - from a historical point of view, a lifecycle analysis covering a period of roughly 30-40 years in the accumulation phase and 30-40 years in retirement (e.g. Pfau, Safe Savings rates: A new Approach to retirement Planning over the life cycle, 2011 - although he used 30 years for each phase) is an excellent approach.
I think we are in agreement - and your graphs are very helpful in the context of these statements.

StillGoing wrote: Wed Apr 03, 2024 4:08 am However, for Monte-Carlo approaches or bootstrapping, the good sequence followed by bad sequence that has been evident historically (at least in the US) may no longer hold for all lifecycles.

cheers
StillGoing
While true, I think the conclusion is different if you analyze the results. The possibility of failure is not what you care about, it is the probability of failure when comparing action A vs action B. What actually happens is that conservative investing often does not pay off historically or based on simulations - there can be a net increase in failure rate because of that conservative choice. The most important variable is portfolio size - which of course is typically ignored in decumulation studies and therefore the value of the decumulation studies is limited or even misleading.

I will take a specific scenario as an example. Assume we have a 30 year retirement and we are waiting 5 more years before getting social security which covers 30% of our expenses. If we are okay with 5% failure rate then 65/35 is the optimal AA and we need a portfolio of 19.1x our annual expenses. Anything more conservative or aggressive than this increases risk. I have a 9.7% chance of failure at 30/70 and 6.9% chance of failure at 100/0 with this same 19.1x expenses. 100/0 is safer than 30/70.

People pick 30/70 because backtesting shows that less than 30% stocks increases risk. What backtesting does not show is that 30/70 AA drops off that same cliff if future stock returns are lower than past stock returns. You actually need more than 30% stocks if you think stocks will perform poorly. 40/60 would be the better AA choice if you look at CAPE, think we will have lower returns, and want to minimize volatility while still preserving your success chance.

50/50 AA is safer than 30/70 even for large portfolios unless you form a complete TIPS ladder. Thinking stocks are "risky" can lead to behavioral choices that increase portfolio risk.

The benefit of being or staying somewhat aggressive AA is that portfolio size provides safety even while dealing with volatility. That 65/35 portfolio at 19.1x with a 5% failure rate drops to a failure rate of 0.1% at 30.9x expenses. Getting a 62% bigger portfolio decreases the failure rate by 50x. Aggressive accumulators that achieve 30% bigger portfolios can be 10x safer if they de-risk. They can be safer with that bigger portfolio even with 10-20% bonds. Some risk reduction is always helpful in retirement but the "greedy" choice is often the safer choice.

Conservative investors would learn a lot if they studied portfolio size. Unfortunately it may go against their behavior preference and we like to ignore data that doesn't fit our preference. We can say that aggressive accumulation is helpful, that fixed income is helpful in retirement, that portfolio size matters, and that obtaining a big enough portfolio to build a TIPS ladder is the ultimate risk reducer. That leaves a lot of room for both aggressive and conservative choices, even from the same investor.
Topic Author
SantaClaraSurfer
Posts: 751
Joined: Tue Feb 19, 2019 10:09 am

Re: 100% Equities Challenge Scenario

Post by SantaClaraSurfer »

papiper wrote: Wed Apr 03, 2024 7:53 am So... you use "average equity returns" then artificially have two crashes - and the recoveries are at the average return? If you impose a crash, and then ignore the speed that almost all equities recover soon after. So on the equity side you handicapping returns much more than history would support. Of course the equity side looks worse if you make those two negative assumption.

If you want equities too look risky, just look at all the downside and ignore any upside performance. If you leave the data alone - 100% equities deliver more over the long haul.
First, I appreciate the challenge, but I think you are mischaracterizing the stress test. The downturns I included are in the ballpark of previous US bear markets. Yes, there is lower equity performance overall for this 20 year period. That's precisely the point of this stress test, and how it is different from a backtest looking at the last twenty years, for example, which were great for equities.

This was a stress test emphasizing lower than expected returns in equity markets to let me see how two different allocations would perform over 20 years; I set it up by default to examine a challenging scenario for equities for two portfolios that were already at 25x.

In Scenario 1 the base return outside the challenge period is 8.7%. However, the overall nominal equity return over the 20 year period is 5.4% since it includes a -24% bear market in 2033 and a -43% market crash in 2036. The recovery sequence from the downturns gets back to 2033 levels by 2040, including two years at +12% and one year at +40%. This is not a great depression scenario as the recovery to 2033 levels happens within 7 years.

Code: Select all

2033	-24.00%
2034	+12.00%
2035	+9.00%
2036	-43.00%
2037	+12.00%
2038	+11.00%
2039	+8.00%
2040	+40.00%
In Scenario 2 the base return (outside the challenge period) is 10.25% and, in this case, the overall nominal equity return over the 20 year period is 6.4% including two -25% bear markets, one in 2033 and the other in 2036. The recovery sequence is more robust, and the equity market exceeds 2033 levels in 2040 by 33%.

Code: Select all

2033	-25.00%
2034	+12.00%
2035	+9.00%
2036	+15.00%
2037	+11.00%
2038	-25.00%
2039	+8.50%
2040	+40.00%
What really impacts equities in both scenarios is that the real return is impacted by inflation: 3.2% average annual inflation in Scenario 1, and 4.5% annual inflation in Scenario 2. I included these inflation rates because this is specifically what the the Cederburg study calls out as causing nominal bond portfolios to fail over the long run.

However, when persistent inflation combines with lower average nominal returns for equities, a portfolio that includes TIPS can more than hold its own versus a 100% equity portfolio over 20 years. This can be true even when the scenario that includes TIPS also includes nominal bonds earning a negative real return for 20 years.

When persistent inflation is coupled with a bear market, equities can take a long time to recover their real value even when they grow at nominal rates that look normal. Conversely, TIPS with a fixed real return really help in this scenario.

The contrast between equities and TIPS in this situation is clear when you look at it via spreadsheet formulas.

Starting at $100, with 5% inflation, and a 25% drawdown, equity returns at 100*(1+(-.25) = $75 and then the real return with inflation 75*(1+(-.05)= $71.25.

That same year, TIPS with a 1.5% real fixed rate, starting at $100: 100*(1+(.015)=$101.5 and then the real return with the inflation adjustment 101.5*(1+(.05)) = $106.6.

A $200,000 allocation to TIPS really helps Investor 1.

If anything, as I've mentioned above, this stress test emphasizes that inflation adjusted bonds with a fixed real return can be a good value and complement a 60% equity portfolio even if you don't use them to rebalance. The current I Bond at 1.3%, for example, will grow at 1.3% plus the inflation rate for the next 30 years, and never lose value. That's not sufficient for a retirement plan, of course, but it is a strong complement to equities in the 20 year scenarios I ran above.
abc132
Posts: 2471
Joined: Thu Oct 18, 2018 1:11 am

Re: 100% Equities Challenge Scenario

Post by abc132 »

SantaClaraSurfer wrote: Wed Apr 03, 2024 11:49 am
papiper wrote: Wed Apr 03, 2024 7:53 am So... you use "average equity returns" then artificially have two crashes - and the recoveries are at the average return? If you impose a crash, and then ignore the speed that almost all equities recover soon after. So on the equity side you handicapping returns much more than history would support. Of course the equity side looks worse if you make those two negative assumption.

If you want equities too look risky, just look at all the downside and ignore any upside performance. If you leave the data alone - 100% equities deliver more over the long haul.
First, I appreciate the challenge, but I think you are mischaracterizing the stress test. The downturns I included are in the ballpark of previous US bear markets. Yes, there is lower equity performance overall for this 20 year period. That's precisely the point of this stress test, and how it is different from a backtest looking at the last twenty years, for example, which were great for equities.

This was a stress test emphasizing lower than expected returns in equity markets to let me see how two different allocations would perform over 20 years; I set it up by default to examine a challenging scenario for equities for two portfolios that were already at 25x.

In Scenario 1 the base return outside the challenge period is 8.7%. However, the overall nominal equity return over the 20 year period is 5.4% since it includes a -24% bear market in 2033 and a -43% market crash in 2036. The recovery sequence from the downturns gets back to 2033 levels by 2040, including two years at +12% and one year at +40%. This is not a great depression scenario as the recovery to 2033 levels happens within 7 years.

Code: Select all

2033	-24.00%
2034	+12.00%
2035	+9.00%
2036	-43.00%
2037	+12.00%
2038	+11.00%
2039	+8.00%
2040	+40.00%
In Scenario 2 the base return (outside the challenge period) is 10.25% and, in this case, the overall nominal equity return over the 20 year period is 6.4% including two -25% bear markets, one in 2033 and the other in 2036. The recovery sequence is more robust, and the equity market exceeds 2033 levels in 2040 by 33%.

Code: Select all

2033	-25.00%
2034	+12.00%
2035	+9.00%
2036	+15.00%
2037	+11.00%
2038	-25.00%
2039	+8.50%
2040	+40.00%
What really impacts equities in both scenarios is that the real return is impacted by inflation: 3.2% average annual inflation in Scenario 1, and 4.5% annual inflation in Scenario 2. I included these inflation rates because this is specifically what the the Cederburg study calls out as causing nominal bond portfolios to fail over the long run.

However, when persistent inflation combines with lower average nominal returns for equities, a portfolio that includes TIPS can more than hold its own versus a 100% equity portfolio over 20 years. This can be true even when the scenario that includes TIPS also includes nominal bonds earning a negative real return for 20 years.

When persistent inflation is coupled with a bear market, equities can take a long time to recover their real value even when they grow at nominal rates that look normal. Conversely, TIPS with a fixed real return really help in this scenario.

The contrast between equities and TIPS in this situation is clear when you look at it via spreadsheet formulas.

Starting at $100, with 5% inflation, and a 25% drawdown, equity returns at 100*(1+(-.25) = $75 and then the real return with inflation 75*(1+(-.05)= $71.25.

That same year, TIPS with a 1.5% real fixed rate, starting at $100: 100*(1+(.015)=$101.5 and then the real return with the inflation adjustment 101.5*(1+(.05)) = $106.6.

A $200,000 allocation to TIPS really helps Investor 1.

If anything, as I've mentioned above, this stress test emphasizes that inflation adjusted bonds with a fixed real return can be a good value and complement a 60% equity portfolio even if you don't use them to rebalance. The current I Bond at 1.3%, for example, will grow at 1.3% plus the inflation rate for the next 30 years, and never lose value. That's not sufficient for a retirement plan, of course, but it is a strong complement to equities in the 20 year scenarios I ran above.
Reverse your sequences and see if you are actually benefitting because of the bears or because you got lucky in timing the TIPS/bonds.

Hint: The bond portfolio is not necessarily safer because of your backtest, it just benefitted from earlier downturn while the stock portfolio benefits from an earlier good sequence. I think you have fooled yourself into a false conclusion as the worst decumulation sequences typically follow good sequences - you need the good and bad to compare the portfolios fairly. We don't know if the first few years will be good or bad, but if we did it would help us pick the better portfolio.
Post Reply