When Cash Was King

[hocus]

The thread at this link reports on one of the more surprising findings that JWR1945 has put forward in the research he posts to the SWR Research Group board (at NoFeeBoards.com)

http://www.nofeeboards.com/boards/viewtopic.php?t=2234

JWR1945: "Look back into the years just before the twentieth century. Commercial paper portfolios by themselves supported inflation matched withdrawal rates in excess of 6%.... Commercial paper seems to have gotten a bum rap. It has done quite well except for retirements that were started during the depression."

 

[unclemick]

And that's why I have a tendency to bad mouth SWR unless it remains in context of an overall calculation with the ground rules spelled out.

Commercial paper during the gold standard and before the Central Bank got really rolling - is an entirely different critter.

 

[JWR1945]

Check out 1965, 1966, 1968 and 1969. Commercial paper by itself did better than a portfolio with 80% stocks and 20% commercial paper.

Have fun.

John R.

 

[haha]

Check out 1965, 1966, 1968 and 1969.  Commercial paper by itself did better than a portfolio with 80% stocks and 20% commercial paper.

In the context that you use it does commercial paper have a given maturity or average maturity? Does it pay coupon interest, or is it in effect discount bills?

And if you were to use this asset in your allocation plan, what specifically would you buy? A mutual fund with a given duration, or what?

Thanks, Mikey

 

[intercst]

see next post

 

[intercst]

Check out 1965, 1966, 1968 and 1969.  Commercial paper by itself did better than a portfolio with 80% stocks and 20% commercial paper.

Have fun.

John R.

Here's a study that looked at the optimal asset allocation (S&P500/cash) for each of the 102 thirty-year overlapping pay out periods from 1871-2002. About 75% of the time, a 100% stock portfolio resulted in the highest SWR. An allocation of less than 50% stock was optimal less than 10% of the time. A portfolio of 100% cash was not optimal in any 30-year period on record. Here's the results.

% -----No. of
S&P---- Periods

100% 74
95% 4
90% -
85% 2
80% 2
75% 2
70% 4
65% 1
60% 2
55% 1
50% 2
45% 2
40% -
35% 1
30% -
25% 2
20% 1
15% -
10% 2
5% -
0% -
--------
Total = 102

If you want to check the arithmetic, you can download the spreadsheet from this link.

http://www.retireearlyhomepage.com/re2000SWRasset.xls

I wouldn't want to trust my retirement to the market-timing voodoo required to identify the precious few periods where cash slightly outperforms a diversified portfolio at the efficient frontier. There's damn little upside, and a good chance you'll significantly underperform if you can't correctly time this low probability event.

intercst

 

[hocus]

I wouldn't want to trust my retirement to the market-timing voodoo required to identify the precious few periods where cash slightly outperforms a diversified portfolio at the efficient frontier. There's damn little upside....

The Data-Based SWR Tool is a descriptive tool, not a prescriptive tool. The purpose is to look at how the various asset classes have performed in the past so that you can develop insights into how they may perform in the future. The upside of looking at the data is that you learn things you did not know before you did so.

 

[Cut-Throat]

The upside of looking at the data is that you learn things you did not know before you did so.

I learned alot from looking at intercst's table.

 

[hocus]

I learned a lot from looking at intercst's table.

That's good. Learning is what we are here for.

A lot of people have learned from looking at JWR1945's research too. That's why I put up the thread-starter. I want even more people to look at his work and to learn from it.

 

[unclemick]

Heh, heh,heh

Talk about being out of sync - the 60's period cited was when I first started investing - made with I thought was a killing with help of a Dean Witter broker - bought a new 1968 Datsun 2000 roadster, rented the penthouse in our apartment building, chased dirty blonds and generally expressed confidence that 'mine didn't stink' - a true legend in my own mind. Subsequent history was more humbling. Had I ownly known.

Oh well.

 

[JWR1945]

In the context that you use it does commercial paper have a given maturity or average maturity? Does it pay coupon interest, or is it in effect discount bills?

And if you were to use this asset in your allocation plan, what specifically would you buy? A mutual fund with a given duration, or what?

Thanks, Mikey

Commercial paper is similar to a money market fund. It is what both the Retire Early Safe Withdrawal Calculator and the FIRECalc call commercial paper. It is what is in Professor Shiller's database.

Please note: all of these calculations are on the Retire Early Safe Withdrawal Calculator, Version 1.61, Revised: November 07, 2002. I just verified my numbers using a copy of what I originally downloaded.

Have fun.

John R.

 

[JWR1945]

I made a comparison of two portfolios. One consisted entirely of commercial paper. The other consisted of 80% stocks and 20% commercial paper. I started with an initial balance of $100000. I looked at year 30. I set expense equal to 0.20%. All other values were the default settings on the Retire Early Safe Withdrawal Calculator. I just took these data using what I originally downloaded. It is Version 1.61, revised Novermber 7, 2002.

You can reconstruct my results easily.

To do this for yourself, look at column S, Z and AB. Column S has dates. Column Z has nominal balances at year 30. Column AB has real balances at year 30. The year 1965 is on row 109. The year 1966 is on row 110. The year 1968 is on row 112. The year 1969 is on row 113.

I increased the withdrawal rate in increments of 0.1%. At a withdrawal rate of 3.9%, both portfolios had a positive balance in year 30. At a withdrawal rate of 4.0%, the 80% stock portfolio failed in 1966. At a withdrawal rate of 4.1%, the 80% stock portfolio failed in 1965 and 1966. At a withdrawal rate of 4.2%, the 80% stock portfolio failed in 1965, 1966, 1968 and 1969.

At withdrawal rates of 4.3% and lower, the 100% commercial paper portfolios all survived in 1965, 1966, 1968 and 1969. They all failed (that is, they had negative balances at year 30) at a withdrawal rate of 4.4%.

Remember, these are the outputs from a version of the Retire Early Safe Withdrawal Calculator that I left unmodified.

Have fun.

John R.

 

[JWR1945]

I just took similar data on FIRECalc.

Remember, there are differences between the Retire Early Safe Withdrawal Calculator and FIRECalc, especally in terms of the default settings. In particular, the FIRECalc makes one withdrawal at one time each year. The Retire Early Safe Withdrawal Calculator splits the withdrawal equally (using the default setting) into two parts, one at the beginning of the year and one at the end of the year.

With $100000 initial balance, 0.20% expenses, CPI selected for inflation and with other numbers at their default settings (including making the first withdrawal at the end of the first year, not at the beginning), these are my results:

With 80% stock and 20% commercial paper, the portfolio failed when withdrawing $4100 in 1966. It survived at a $4000 withdrawal amount. The 80% stock portfolios all survived for the full 30 years when withdrawing $4200 in 1965, 1966, 1968 and 1969, but it failed when withdrawing $4300.

With 100% commercial paper (and 0% stocks), all portfolios survived with $4200 withdrawal amounts in 1965, 1966, 1968 and 1969. They all failed when the withdrawal amount was increased to $4300.

Keep in mind that the Retire Early Safe Withdrawal Calculator and the FIRECalc have different default settings.

BTW, you may be surprised to see how well commercial paper did in the days of the Gold Standard and after we were solidly out of the Great Depression. You can see this yourself with the FIRECalc.

Have fun.

John R.

 

[JWR1945]

Re: When Cash

I wouldn't want to trust my retirement to the market-timing voodoo required to identify the precious few periods where cash slightly outperforms a diversified portfolio at the efficient frontier. There's damn little upside....

The Data-Based SWR Tool is a descriptive tool, not a prescriptive tool. The purpose is to look at how the various asset classes have performed in the past so that you can develop insights into how they may perform in the future. The upside of looking at the data is that you learn things you did not know before you did so.

Remember that this voodoo is produced by the Retire Early Safe Withdrawal Calculator.

Reasonable people learn from what the data have to say. Stocks have done poorly when prices have been sky high. They have even underperformed commercial paper (a cash equivalent) at such times.

These are days of sky high stock valuations.

Have fun.

John R.

 

[intercst]

Re: When Cash

Remember that this voodoo is produced by the Retire Early Safe Withdrawal Calculator.

Reasonable people learn from what the data have to say. Stocks have done poorly when prices have been sky high. They have even underperformed commercial paper (a cash equivalent) at such times.

These are days of sky high stock valuations.

Have fun.

John R.

That's correct. And I believe the creator of the Retire Early Safe Withdrawal Calculator recognized the stock-switching studies as voodoo when it was produced and said so.

Readers might also be interested in raddr's comments on the "junk science" behind these P/E stock-switching studies.

http://nofeeboards.com/boards/viewtopic.php?t=2399&highlight=switching+beware

Have Fun. <LOL>

intercst

 

[JWR1945]

Here are a couple of links to take care of the distraction.

http://nofeeboards.com/boards/viewtopic.php?t=2919

http://www.gummy-stuff.org/JWR.htm

Have fun.

John R.

 

[JWR1945]

Much more interesting is to look at what FIRECalc can tell us. It has a fantastic display.

Put these settings into FIRECalc:
Withdrawals: $3000, $4000, $5000, $6000, $7000
Starting Portfolio: $100000
Lifespan of Portfolio: 30 Years
Percentage Stocks: 0%
Where is the rest invested: Commercial Paper (this is the default condition).
Inflation estimate: CPI

Leave everything else at their default conditions.

Click on the Detailed Results button to see the display. The display shows the start years of historical sequences. Red lines show us which portfolios would have failed within 30 years.

You will see that commercial paper has behaved differently in three different eras.

With $3000 withdrawals, portfolios with 0% stocks and 100% commercial paper would have failed for sequences beginning in 1932-1946.

With $4000 withdrawals, portfolios with 0% stocks and 100% commercial paper would have failed for sequences beginning in 1897-1899, 1908-1909, 1911-1916 and 1928-1951. Notice that they would have survived in most of the years of the Gold Standard and in the modern era (starting from 1952) after the World War 2 price controls were lifted.

Think about this. From shortly after World War 2, a portfolio of 100% commercial paper and 0% stocks would have produced an inflation-adjusted withdrawal rate of 4%! Sure, there was no upside. But this is the magic 4% withdrawal rate and it is without any stocks whatsoever.

With $5000 withdrawals, portfolios with 0% stocks and 100% commercial paper would have survived all of the years from 1871-1893, 1919-1921 and in the partial sequences starting with 1976.

Think about this. In the late 1800s, a portfolio of commercial paper and no stocks would have produced an inflation-adjusted withdrawal rate of 5%! You can't get that with today's money market funds.

With $6000 withdrawals, portfolios with 0% stocks and 100% commercial paper would have survived all of the years from 1871-1888 and 1920.

Wow! A portfolio of 0% stocks and 100% commercial paper would have survived many of the years in the late 1800s at a 6% withdrawal rate.

With $7000 withdrawals, portfolios with 0% stocks and 100% commercial paper would have survived for the sequences beginning in 1871-1878 and 1880-1884.

The good times have to come to an end sometime. But this is a 7% withdrawal rate for a portfolio with 0% stocks and 100% commercial paper.

Results using the PPI for inflation instead of the CPI are similar. They differ slightly.

Have fun.

John R.

 

[JWR1945]

Repeat the previous survery using 75% stocks (the default condition).

You will notice how important the 1960s were with withdrawal rates of 5% and higher. You will notice how dangerous years cluster together as you increase withdrawal rates to 6% and 7%.

Now repeat this with withdrawal amounts of $4000, $4100, $4200, $4300 and $4400.

The portfolio with 0% stocks and 100% commercial paper survives in 1965, 1966, 1968 and 1969 when the withdrawal amount is $4200 (and lower). It survives in 1969 when the withdrawal amount is $4300. It fails in 1965, 1966, 1968 and 1969 when the withdrawal amount is $4400.

The 75% stock portfolio fails in 1966 when the withdrawal amount is $4200. It fails in 1965, 1966 and 1969 when the withdrawal amount is $4300. It fails in 1965, 1966, 1968 and 1969 when the withdrawal amount is $4400.

This proves that it is possible for a high stock allocation to make things worse. Knowing why that happened is important. It happened because valuations were sky high, almost as high as they are today.

This does not guarantee failure. It does tell us something about your odds of success.

Have fun.

John R.

 

[cute fuzzy bunny]

I have a more applicable and interesting observation:

"Look mommy...theres an airplane up in the sky!"

A lot of this discussion ranges around periods of time when we were coming off of the civil war and most investments were very weakly valued, during times when we were on the gold standard, and during times of bad fiscal and monetary policies, they probably dont have much interest in modern times.

But who knows...

We're back to "tempest in a teapot" territory.

Everyone enjoy the show. Refunds will be made available to anyone who leaves early. Free motor oil and beef jerky to anyone over the age of 18.