Has Anyone Simulated running a Boglead portfolio through the Great Depression Period ?

[CraigTester]

Using Portfolio Visualizer (PV), below is the performance of a strawman portfolio from 1990-2023

EG: 50-50 SP500, ST Treas

                      Real CAGR  Max Drawdown

Rebal yrly           4.47%            23.89%

No Rebal:           5.2%            32.49%

100% SP500.      6.95%             50.89%              


However, since PV doesn't extend to the Great Depression period, or the late 1960's, wondering if there are any existing tools available to easily extend my above table to cover these earlier periods? (without building my own)

[toddthebod]

The Trinity study went back to 1925.

[stan1]

I'm going to change the question a little but use a Monte Carlo simulation, you can start with the one in PV then move onto ones that are more custom if you want. The future won't be like the past.

Some recommend i-ORP however while it is still on-line it does not appear to be in active development or support any longer.

[nisiprius]

It's difficult to know what to do about international stocks. The only "real" international index, the MSCI EAFE index, only goes back to 1970. The Dimson and Marsh studies go back to 1900, but I don't think you can get the actual numbers without paying for them... paying a lot.

Because of World Wars I and II, from 1900 to the present, US stocks and a few other countries--notably Australia--did quite a bit better than the rest of the world, and the Great Depression hit everything.

[dbr]

This tool shows outcomes for a rebalanced stock/bond portfolio back to 1871. I would understand the performance would be asset returns with no expenses and no taxes. I don't know of you consider this a Boglehead portfolio, but it seems close enough to me. Note FireCalc also does this range.

https://engaging-data.com/visualizing-4-rule/

A person retiring in 1929 with a 60/40 portfolio had a safe withdrawal rate of 4.8% and 1930 5.0%. The bad years to retire at 4% were 1965-1969.

People make a lot of presumptions about the conditions that affect retirement success that are just not so.

[JoMoney]

https://firecalc.com/
Is a monte carlo simulation using U.S. market data going back to 1871

[Jungle]

It's been done with firecalc, FIcalc etc. But it's something of a moot point, as total stock market index funds & bond funds didn't exist until the 70s, and didn't become popular til the 80s. Active mutual funds were just getting started then, and had higher expenses than today, plus many (MITTX, Wellington, etc) actually underperformed the S&P.

Most likely, an investor back then would have held a basket of individual stocks, plus some treasury bonds.

[bh1]

Chuck 85% of your stock value and crank up the stock growth for the next ten years?

This is why I have gov't bonds.

[CraigTester]

Thanks for all the quick responses....

The PV tool allows you to address various what-if type questions in a very flexible manner, that I find appealing.....whereas firecalc, trinity-study, etc, are more designed to answer specific questions within a less flexible template.....

Does anyone know why PV doesn't use data prior to the early 70's for things like treasuries and US stocks...?

I understand why they don't include pre 1970's Int'l data, as explained above by nisiprius...and I understand the apples-n-oranges data issues comparing current US data to historical, but sources like Shiller, Firecalc, etc, just provide the appropriate disclaimers and proceed anyway....

Maybe Bogleheads is big enough to formally request the PV folks to expand their data...?

With as many references to PV that I see on this site, I'm sure that expanded historical data would be much appreciated by a lot of BH's....

Or perhaps there's still someone out there that knows of another tool with more historical reach....?

[texasfight]

great depression is irrelevant - that was back when couldn't debase the currency

now the ONLY risk to retirement planning is inflation

we had a bull market while the entire global economy was shut down because we printed 40% of the money supply, gave massive stimulus, etc.

Deflation will only be allowed briefly (mainy with regards to wage inflation). Notice how the Fed had no problem with inflation until wages.

cycle to cycle this will be considered a 25% debasement, and we will get many more like this due to US debt load (30T+) + unfunded liabilities (300 trillion plus)

Fed balance sheet hits 30-40 trillion by end of the decade

[marcopolo]

https://firecalc.com/
Is a monte carlo simulation using U.S. market data going back to 1871

i don't think firecalc is a monte carlo simulation.
It only uses actual historical data, with no statistical sampling.

[dbr]

https://firecalc.com/
Is a monte carlo simulation using U.S. market data going back to 1871

i don't think firecalc is a monte carlo simulation.
It only uses actual historical data, with no statistical sampling.

It is not a Monte Carlo simulation but rather a tally of actual historical results. The idea is to use the variable outcomes over 100+ years as a surrogate for a hypothetical distribution of future probabilities. As a heuristic this is appealing because it side-steps having to assume what distributions to pick and what value to assign to the parameters, as well as any other assumptions about serial correlation or anything else about the data. The obvious downside is that you have to assume the future to be a statistical extrapolation of the past. This may seem to be a great error until one tangles with what to do instead.

[MJS]

Historical Investment Calculator:

https://financial-calculators.com/histo ... calculator

Mix its S&P500 with ... cash?

[CraigTester]

Historical Investment Calculator:

https://financial-calculators.com/histo ... calculator

Mix its S&P500 with ... cash?

Thanks MJS...!

Will check it out....

[junior]

Maybe Bogleheads is big enough to formally request the PV folks to expand their data...?

I suspect if PV knew how to get data going back to the depression they would have done it already. Index funds didn't exist back then so I think you'd have to simulate them. Some people might have already done this, but presumably these simulations are proprietary and need a license to use.

[seajay]

This tool shows outcomes for a rebalanced stock/bond portfolio back to 1871. I would understand the performance would be asset returns with no expenses and no taxes. I don't know of you consider this a Boglehead portfolio, but it seems close enough to me. Note FireCalc also does this range.

https://engaging-data.com/visualizing-4-rule/

A person retiring in 1929 with a 60/40 portfolio had a safe withdrawal rate of 4.8% and 1930 5.0%. The bad years to retire at 4% were 1965-1969.

People make a lot of presumptions about the conditions that affect retirement success that are just not so.

Pre 1915 and most investors held bonds. 1794 to 1914 and bonds supported a 30 year 5% SWR

Money was gold (silver) and people lent that gold to the state/banks in return for interest. Gold in being finite saw median (broad) inflation tending to equal 0% - so bond interest was like a real rate of return.

The Roaring 20's introduced stocks widely, up to then they were primarily seen as being for speculators. The 1920's however enticed many new stock investors - as 'sure things'. People even borrowed heavily to buy stocks for the 'rewards they provided'.

In the 1930's money was no longer directly convertible/equal to gold. Over time circumstances transitioned to where the state could print/spend money, banks could just create money to add to borrowers accounts, neither really need't to borrow other peoples gold/money anymore and as such the interest rates offered in net real terms declined to where investors needed to hold some stocks, bonds alone would no longer sustain 5% SWR type rewards.

If you back-test using actual data/60-40 for pre 1940's I suspect you'd be employ somewhat unrealistic data even though actual data exists, such that you might just as well employ Monte-Carlo simulated data. If 'everyone' is into bonds (or stocks in the 20's and back to bonds again in the 1930's (after Wall Street Crash losses)) the prices of assets will tend to be distorted.

[Mr.BB]

This tool goes back to the 1800's. You can put in a 60/40 portfolio and it gives you numerous choices on withdrawal techniques.
www.ficalc.app

[abc132]

All tools are flawed so the important thing is knowing the flaws and what the tool can and cannot do.

Historical SWR's focus only on what worked in the worst historical period which makes them over-value the things that happened to work in that one case.

Historical simulations only tell us what did work.

Monte Carlo simulations have trouble with what parameters we should use for the future.

4% SWR has been great for the US (including great depression), 3% SWR internationally, and 2% SWR or less if your country is going to be devastated by war. The future is unknown but we can decide to what degree we think it will rhyme with the past.

[Harmanic]

Another interesting approach is to look at index-like funds from the time. Index funds could not exist because they rely on computers, but the Massachusetts Investors Trust went through depression holding stocks that mimicked the "market averages" of the time. There were few redemptions during the depression and the fees were very low. In fact, redemptions exceeded deposits in only one quarter during the entire depression.

It was also the mutual fund that inspired Jack Bogle's thesis in the 1940s.

[CraigTester]

This tool goes back to the 1800's. You can put in a 60/40 portfolio and it gives you numerous choices on withdrawal techniques.
www.ficalc.app

Thanks Mr. BB

Will have a look.

[Mr. Rumples]

Deflation was also an issue: https://www.frbsf.org/economic-research ... deflation/

Clothing cost less, at least in VA. Dresses were made of old animal feed bags, homemade three tier depression cooking pots became common to save energy; there was a shortage of apple cider presses which was used for bootlegging, but can't find data on those prices. Cotton prices fell from 22 cents lb. in 1925 to 6 cents in 1931. Tobacco fell from 15 cents per lb to 11.6 cents in 1932 - most folks rolled their own cigarettes. Housing costs fell 7%.

[dcabler]

Simba Spreadsheet available on this forum has a wealth of return data and a host of built in tools and is updated annually.

viewtopic.php?p=7049298#p7049298

Cheers

[dkturner]

For the period 1929-1932 a portfolio of 60% S&P500 and 40% 5year Treasuries (rebalanced annually) would have declined in value by 39.4% in nominal terms, but “only” 19.4% in inflation adjusted terms, due to the substantial deflation over the period. A total “real” four year purchasing power loss of only 19.4% for a balanced portfolio is a pretty good result for the worst of the Great Depression.

By contrast, for the more recent two year period of 1973-1974 the same portfolio would have experienced a nominal decline of 20.4% but an inflation adjusted decline of 37.5%, because of the substantial inflation for those years.

[martincmartin]

Deflation was also an issue:

When looking at total real returns, deflation actually helps. The stock market was down 89% in nominal dollars, but less in real dollars because of inflation. And deflation also helped real bond yields, since you add the deflation (as a positive number) to nominal bond returns to get real bond returns.

If you imagine retiring at the peak of the market in 1929, it also makes sense that deflation takes the edge off your problems: you have less money, but things also cost less, so your situation isn't as bad as it seems.

The Great Depression is the reason you don't have 100% stocks in retirement, even if you have the risk tolerance. The deflation means bonds saved you from the sequence of returns risk.

[JoMoney]

...
Does anyone know why PV doesn't use data prior to the early 70's for things like treasuries and US stocks...?
...

If you subscribe to the paid service (or free trial), you can import your own data sets to use.
The data they provide is data the creator believes reliable and was available for them to distribute, they used to offer some different data series, like MSCI index data, but I believe they had to take that down or pay for a license, which is why the current international data available on there doesn't go back as far as the MSCI index does.

[nisiprius]

great depression is irrelevant - that was back when couldn't debase the currency

now the ONLY risk to retirement planning is inflation...

The highest inflation in the history of the CPI occurred from 6/30/1915 to 6/30/1920. when the United States was on the gold standard and the currency was not "debased."

It averaged 15.66% per year.

Overall inflation over that time period was 2.07X and the purchasing power of $1 dropped to $0.48.

It stole the value of Henry Ford's revolutionary $5/day wage out of workers' pockets, with resulting labor unrest and a dimming of once-wide admiration for "Fordism." The satire on Ford and Fordism seemed odd to me when I encountered it as a kid in Aldous Huxley's Brave New World, because by the sixties you hardly heard of "Fordism" any more; but there was a period of time when "Fordism" was lauded and Henry Ford was lionized, almost universally.

Of course $35 still bought an ounce of gold at the end of the period, so if the definition of inflation is the price of gold there was no inflation at all.

[Jungle]

Deflation was also an issue: https://www.frbsf.org/economic-research ... deflation/

Clothing cost less, at least in VA. Dresses were made of old animal feed bags, homemade three tier depression cooking pots became common to save energy; there was a shortage of apple cider presses which was used for bootlegging, but can't find data on those prices. Cotton prices fell from 22 cents lb. in 1925 to 6 cents in 1931. Tobacco fell from 15 cents per lb to 11.6 cents in 1932 - most folks rolled their own cigarettes. Housing costs fell 7%.

This is an excellent point. A lot of people point to the ~85% drop in index points, which does sound very bad. But companies tended to pay out much larger dividends back then, like 3-4%. Plus, the economy was so rough that there was significant deflation through the 30s.

Accounting for those two factors, the "real" drop in the value of the market is more like 50-55% - which is still scary, but not really that different than what we saw in the 2000s. One could have retired in 1929, 100% stock portfolio, with a 3.4% inflation (or deflation) adjusted withdrawal rate, and had a successful retirement all the way out to 1979. Adding bonds to that helped.

But it's kind of a theoretical point, since there wasn't much of a way to "own the market" in a practical fashion, in the 20s and 30s

[enad]

The problem will be relying on the data assuming you can find it. Also one would need to consider indices that did not exist at the time or anything that resembles them today. Then the question, is why? Even if you could you could not necessarily use it to predict another depression on that scale.

[dkturner]

great depression is irrelevant - that was back when couldn't debase the currency

now the ONLY risk to retirement planning is inflation...

The highest inflation in the history of the CPI occurred from 6/30/1915 to 6/30/1920. when the United States was on the gold standard and the currency was not "debased."

It averaged 15.66% per year.

Overall inflation over that time period was 2.07X and the purchasing power of $1 dropped to $0.48.

It stole the value of Henry Ford's revolutionary $5/day wage out of workers' pockets, with resulting labor unrest and a dimming of once-wide admiration for "Fordism." The satire on Ford and Fordism seemed odd to me when I encountered it as a kid in Aldous Huxley's Brave New World, because by the sixties you hardly heard of "Fordism" any more; but there was a period of time when "Fordism" was lauded and Henry Ford was lionized, almost universally.

Of course $35 still bought an ounce of gold at the end of the period, so if the definition of inflation is the price of gold there was no inflation at all.

I took a look at a data base (going back to 1900) from Cambridge Associates. It only reports annual changes. What I found interesting is that for the four calendar years 1916-1919 the inflation adjusted total return of the (reconstructed) S&P 500 index was a negative 39.0%. Normally one hides in Treasury bills during periods of excessive inflation. Treasury bills produces an inflation adjusted total return of a negative 38.8% for those four years.

[warner25]

Sure, there are plenty of ways to run the numbers. On my spreadsheet, I've used the following series of real (as in after-inflation) returns for an annually rebalanced 50/50 US stock and bond portfolio from 1929 to 1948.

Code: Select all

0.97
0.94
0.96
1.1179
1.34415
0.96
1.2714
1.17745
0.81
1.12865
1.04995
0.98
0.89
1.01045
1.07045
1.09485
1.19305
0.87
0.87

The first thing to notice is that, on paper, deflation (like -10% in 1932) helped tremendously with the bonds. I've read that the reality of this is more complicated; severe deflation meant that almost nobody had the liquidity to rebalance a portfolio or buy stocks at bargain prices. So in this scenario, losing your income source(s) is probably your bigger problem.

More importantly, as others have implied, I would just question of the relevance of market data from a time before the SEC, FDIC, leaving the gold standard, 401(k) plans, IRAs, global markets, etc... I've learned that the Federal government didn't even track the unemployment rate at that time; what we know about unemployment was reconstructed years later by researchers. It's like looking at baseball statistics from the dead-ball era to understand the modern version of the game.

[smooth_rough]

Vanguard Wellington fund inception date 1929. Or maybe that's well known to bogleheads.

[CraigTester]

...
Does anyone know why PV doesn't use data prior to the early 70's for things like treasuries and US stocks...?
...

If you subscribe to the paid service (or free trial), you can import your own data sets to use.
The data they provide is data the creator believes reliable and was available for them to distribute, they used to offer some different data series, like MSCI index data, but I believe they had to take that down or pay for a license, which is why the current international data available on there doesn't go back as far as the MSCI index does.

Thanks JoMoney.

This is an interesting idea...Have you experimented at all....?

If it's as simple as importing data from say, Shiller's website, this might be the path of least resistance to leveraging their features with data (I think) I trust....

[nisiprius]

My posting, "4% rule" withdrawals from real mutual funds explores the survival of 4%-first-year-then-COLA-adjusted withdrawals on these portfolios. In each case, the exploration starts at inception of the fund or the portfolio, so most of them don't go back to before the Great Depression, but three of them, marked with asterisks, do. MITTX is 100% stocks, the rest are balanced funds with both stocks and bonds.

Only the Fidelity Puritan Fund was able to sustain 4% withdrawals for every 30-year period within its lifetime, but that only goes back to 1948.

*60/40 allocation to SBBI "large-company stocks" and "intermediate-term government bonds"
Fidelity Puritan Fund, FPURX
George Putnam Balanced Fund Class Y (PGEYX)
Dodge & Cox Balanced Fund (DODBX)
T. Rowe Price Balanced Fund (RPBAX)
*MFS Massachusetts Investors Trust (MITTX)
*Vanguard Wellington Fund (VWELX)

[CraigTester]

Sure, there are plenty of ways to run the numbers. On my spreadsheet, I've used the following series of real (as in after-inflation) returns for an annually rebalanced 50/50 US stock and bond portfolio from 1929 to 1948.

Code: Select all

0.97
0.94
0.96
1.1179
1.34415
0.96
1.2714
1.17745
0.81
1.12865
1.04995
0.98
0.89
1.01045
1.07045
1.09485
1.19305
0.87
0.87

The first thing to notice is that, on paper, deflation (like -10% in 1932) helped tremendously with the bonds. I've read that the reality of this is more complicated; severe deflation meant that almost nobody had the liquidity to rebalance a portfolio or buy stocks at bargain prices. So in this scenario, losing your income source(s) is probably your bigger problem.

More importantly, as others have implied, I would just question of the relevance of market data from a time before the SEC, FDIC, leaving the gold standard, 401(k) plans, IRAs, global markets, etc... I've learned that the Federal government didn't even track the unemployment rate at that time; what we know about unemployment was reconstructed years later by researchers. It's like looking at baseball statistics from the dead-ball era to understand the modern version of the game.

Thanks Warner25...

I've done a bit of what-if's using excel as well....One of the most concerning realizations has been understanding that on an inflation adjusted basis, the SP500 (per Shiller's proxy data), required 20 years to "permanently" re-attain it's 1929 value... (assuming dividends reinvested)...

But I didn't go on to see how many years it would take if that performance was inter-played with like you say, 50% bonds, rebalanced every year.....

Are you able to answer that question?

I started looking into building my own tool to look at questions like this but hit pause to see if I'm just rebuilding a mousetrap that's already been built...

And yes, as you say, the world was a bit different back then in a number of very important ways... but perhaps still useful to look at this period to "calibrate" our expectations of what's possible....

[CraigTester]

My posting, "4% rule" withdrawals from real mutual funds explores the survival of 4%-first-year-then-COLA-adjusted withdrawals on these portfolios. In each case, the exploration starts at inception of the fund or the portfolio, so most of them don't go back to before the Great Depression, but three of them, marked with asterisks, do. MITTX is 100% stocks, the rest are balanced funds with both stocks and bonds.

Only the Fidelity Puritan Fund was able to sustain 4% withdrawals for every 30-year period within its lifetime, but that only goes back to 1948.

*60/40 allocation to SBBI "large-company stocks" and "intermediate-term government bonds"
Fidelity Puritan Fund, FPURX
George Putnam Balanced Fund Class Y (PGEYX)
Dodge & Cox Balanced Fund (DODBX)
T. Rowe Price Balanced Fund (RPBAX)
*MFS Massachusetts Investors Trust (MITTX)
*Vanguard Wellington Fund (VWELX)

That's a great post, Nisiprius, thanks for pointing it out....!

Several questions,

1) For the 60/40 SBBI, am I reading your chart correctly to say that your first "candidate" retired in 1930 (per the x-axis)..? I know you put an asterisks on this one to specify it predates depression, so maybe I'm just missing something...?

2) Not to create more work, but did you happen to capture the "portfolio-lows" reached by the 60/40 along the way.... ? This is an issue I am zeroing in on at the moment....There's a lot of angst when one's portfolio loses 50%, 60%, 80% of its value in the middle of a 30 year run....And even if they eventually "made it", it would be interesting to see how white knuckled they got along the way....

3) Not a question, but just a comment to highlight the great value in your post of highlighting the drag from expenses in funds...! So nice to have access these days to index funds....and to analysis like this to illuminate the impact...

[warner25]

... Are you able to answer that question?

Fiddling with the rough numbers on my spreadsheet right now: If I start with $100k in the beginning of 1929, I'm almost back to $100k by the end of 1932, and it never dips below that again. By the end of 1945, the original sum has more than doubled, but then 1946-1948 was another down market followed by a big inflation spike that destroyed returns.

Edited to say that I'm no longer confident in what I just wrote above, and this exercise is making me want to re-check everything to be sure... I'll edit again later. Edited again to remove my previous strikethrough. Looking more closely at the numbers and seeing SimpleGift's post below gave me confidence in my answer again.

One of the things to note is that there was a huge upswing, like 35%, just from the beginning of 1929 to the peak in October. So if you invested all your savings in stocks at the actual peak, the numbers are obviously much worse, but otherwise not so bad.

[SimpleGift]

We had a Forum thread some years ago comparing the worst U.S. portfolio disasters in history. The Great Depression period is interesting, but as several other posters have pointed out, it featured strong price deflation over an extended period, which is not at all representative of the modern era and (hopefully) the future.

Using Shiller data for stocks and synthetic 10-year Treasury returns, we are able to see a rough picture of portfolio drawdowns and recoveries (chart below). Each portfolio's performance is indexed to the pre-crisis market peak. All returns are geometric total returns, inflation-adjusted, with dividends reinvested. The black circles show the point at which each portfolio permanently regained its real pre-crisis value.

• 
  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

Stock-heavy portfolios dropped up to 75% real over 3 years and took 12-15 years to finally regain their real value (chart above). 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 entire period.

[Kookaburra]

What is boglead?

[McQ]

CraigTester, Morningstar has monthly returns for Wellington from July1, 1929. You'll have to become a paying member or exploit the free trial period to download these in a spreadsheet. Once you do, you can simulate any withdrawal rate you like, in confidence that this was a real fund that real investors could own (it rather quickly became one of the largest two balanced funds). However, it was not then a 60/40 split, rather, the mindset when balanced funds began about 1929 was to go 50-50. The 60/40 standard wasn't in existence even as late as the 1950s (TIAA-CREF participants could allocate no more than 50% to stocks). I surmise that in the early 1960s, after the incredible stock returns in the 1950s (one year above 50% and another above 40%), balanced funds decided to err on the side of having somewhat more in stocks, and thus 60/40 or 65/35 was born.

Morningstar also has returns for Mass Investor Trust from the 1920s. If you don't like the Shiller-based simulation that SimpleGift offered you, and can access the CRSP database of Treasuries (I access it through Wharton Research Services through my university library), then you can create your own blend of a stock fund real investors could own (by the mid 1930s MITTX was by far the largest open-end stock fund), and a suitable Treasury bond, blended as you wish.

Inflation can be downloaded from the Bureau of Labor Statistics site.

[MnD]

https://cfiresim.com/ is my favorite for this.

The 1906 to 1935 sequence was rough as was 1937-1966.
But nothing as bad as starting a 30-year sequence in the mid-1960's.

60/40 4% rule ER=.10
https://www.cfiresim.com/0eac4fbe-5898- ... a7fed26628

[Booper]

The big issue is that index funds didn't exist back then. So even if you could get a number, it likely wouldn't mean what you would want it to mean.

[CraigTester]

We had a Forum thread some years ago comparing the worst U.S. portfolio disasters in history. The Great Depression period is interesting, but as several other posters have pointed out, it featured strong price deflation over an extended period, which is not at all representative of the modern era and (hopefully) the future.

Using Shiller data for stocks and synthetic 10-year Treasury returns, we are able to see a rough picture of portfolio drawdowns and recoveries (chart below). Each portfolio's performance is indexed to the pre-crisis market peak. All returns are geometric total returns, inflation-adjusted, with dividends reinvested. The black circles show the point at which each portfolio permanently regained its real pre-crisis value.

• 
  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

Stock-heavy portfolios dropped up to 75% real over 3 years and took 12-15 years to finally regain their real value (chart above). 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 entire period.

Thanks SimpleGift!

Looking at your graph, I was initially concerned that your "black circle", (denoting the point at which the all-stock portfolio permanently regained its real pre-crisis value), was not drawn correctly at June, 1949... But then I noticed your note below the chart acknowledging that it actually required just shy of 20 years for the SP500 to "permanently" regain it's 1929 high....

This gives me confidence that we're looking at the same math...!

And then to take it a step further as you did is brilliant...

So how do we turn your excellent chart into a parameter driven tool, available to all BH's....(perhaps similar to the PV format)....?!

It would sure reduce a lot of debates seen daily on this forum if there was one "trustworthy" place to go to get everyone on the same page.....

As Professor McQ (who comments below) has correctly identified on his recent thread, there is a great need for this type of information, presented in an easy-to-manipulate "tool"....

Your chart is outstanding....!

Thank you for sharing.

[CraigTester]

https://cfiresim.com/ is my favorite for this.

The 1906 to 1935 sequence was rough as was 1937-1966.
But nothing as bad as starting a 30-year sequence in the mid-1960's.

60/40 4% rule ER=.10
https://www.cfiresim.com/0eac4fbe-5898- ... a7fed26628

Thank you, MnD....!

I'm not familiar with this one either.... will check it out....

[CraigTester]

CraigTester, Morningstar has monthly returns for Wellington from July1, 1929. You'll have to become a paying member or exploit the free trial period to download these in a spreadsheet. Once you do, you can simulate any withdrawal rate you like, in confidence that this was a real fund that real investors could own (it rather quickly became one of the largest two balanced funds). However, it was not then a 60/40 split, rather, the mindset when balanced funds began about 1929 was to go 50-50. The 60/40 standard wasn't in existence even as late as the 1950s (TIAA-CREF participants could allocate no more than 50% to stocks). I surmise that in the early 1960s, after the incredible stock returns in the 1950s (one year above 50% and another above 40%), balanced funds decided to err on the side of having somewhat more in stocks, and thus 60/40 or 65/35 was born.

Morningstar also has returns for Mass Investor Trust from the 1920s. If you don't like the Shiller-based simulation that SimpleGift offered you, and can access the CRSP database of Treasuries (I access it through Wharton Research Services through my university library), then you can create your own blend of a stock fund real investors could own (by the mid 1930s MITTX was by far the largest open-end stock fund), and a suitable Treasury bond, blended as you wish.

Inflation can be downloaded from the Bureau of Labor Statistics site.

Thank you, Professor!

[CraigTester]

What is boglead?

A non-fully-formed Boglehead.....

[SimpleGift]

Thanks SimpleGift!

You're welcome, and your kind words are appreciated. Keep in mind, though, that the chart is very much a rough draft of history, with numerous deviations from real-life portfolios, including synthetic Treasury bond returns, no portfolio rebalancing, etc.

It would be interesting to compare the chart with the returns of a real-life balanced fund available at the time, like the 50/50 Wellington Fund suggested by McQ upthread — if someone is ever inspired to pursue its monthly Morningstar data for the 1929-1945 period.

[HomerJ]

Bonds did fine during the Great Depression.

Stocks did terrible.

100% stocks portfolio did not last 30 years if you were pulling 4% a year.

But a 60/40 portfolio did last 30 years

[CraigTester]

Bonds did fine during the Great Depression.

Stocks did terrible.

100% stocks portfolio did not last 30 years if you were pulling 4% a year.

But a 60/40 portfolio did last 30 years

Encouraging that the 60-40 ultimately lasted 30 years, but that poor fella woke up one morning, just 3 years into his retirement, with half his entire portfolio gone...

And this is before he made any withdrawals....

[Mr. Buzzkill]

I used the simba spreadsheet to back test some simple TSM/Treasuries asset allocations to just before the Great Depression, to roughly determine my emotional tolerance for various levels of drawdown

I also checked the after-DEflation drawdowns of the Depression to roughly account for increased buying power of dollar denominated portfolio value to determine how much real drawdown I could tolerate.

Then I did the same for after-inflation portfolio value during the period of 1966-1982

Of course, there is no guarantee that past history represents the worst that could ever happen.

I personally think another situation of intensity and duration like either Great Depression or Great Inflation, while possible, is not likely in my lifetime. But I’m comfortable with my odds of managing through such things given my IPS, modest living expenses, and the back testing

[HomerJ]

Bonds did fine during the Great Depression.

Stocks did terrible.

100% stocks portfolio did not last 30 years if you were pulling 4% a year.

But a 60/40 portfolio did last 30 years

Encouraging that the 60-40 ultimately lasted 30 years, but that poor fella woke up one morning, just 3 years into his retirement, with half his entire portfolio gone...

And this is before he made any withdrawals....

Not really, dividends from both stocks and bonds were much higher back then... like 5%-6%, and there was deflation at the time, so everything cost less.

I'm sure it wasn't easy.

But when there is 25% unemployment, and food lines, and people living in tents, one would probably still feel pretty lucky with a paid-off house and 50% of your money still intact.