Safe Withdrawal Rate versus P/E10 Data

Research on Safe Withdrawal Rates

Moderator: hocus2004

JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Safe Withdrawal Rate versus P/E10 Data

Post by JWR1945 »

Safe Withdrawal Rate versus P/E10 Data

These tables provide the data to replace the Safe Withdrawal Rate versus P/E10 graph that BenSolar frequently mentions. It is at the bottom of this page: http://rehphome.tripod.com/pestudy1.html

I have listed each year from 1921 through 1980 in the order of increasing values of P/E10. Each table has 20 entries. They are the year, the value of P/E10 as of January of that year, the dory36 FIRECalc result with a 50% stock allocation and the FIRECalc result with an 80% stock allocation.

The inputs to the FIRECalc safe withdrawal rate calculator were an initial balance of $1000, a portfolio lifespan of 30 years, and expense of 0.20%, an inflation adjustment to match the CPI and the selection of commercial paper for the cash equivalent portion of the portfolio. The withdrawal amounts were varied for stock allocations of 50% and 80%. All other entries were left at their default values.

Each entry in the table is the withdrawal amount that caused the first portfolio failure for the specified start year. Those entries are in increments of $2. Since the initial balance is always $1000, you just have to insert a decimal point to convert the withdrawal amount to a withdrawal percentage. For example, if the withdrawal amount in the table is $54, it is 5.4% of the initial balance. The portfolio failed at that value ($54 or 5.4% but it survived at $2 less: $52...or 5.2%...was safe.)

The FIRECalc safe withdrawal rate calculator is available at:
http://capn-bill.com/fire/. It is exceptionally easy to use. It uses the historical sequence methodology.

The inputs to FIRECalc were generated by Yale Professor Shiller at www.econ.yale.edu/~shiller/. You can go directly to his historical S&P 500 data by using www.econ.yale.edu/~shiller/data/ie_data.xls. I used only the January entries for each year (which are written as xxxx.01) and I used his values of P/E10 (truncated to the indicated precision).

My selection of 30 years was for purposes of data analysis. With longer portfolio lifespans it is possible to see the effects from two different but significant causes. I included the partial sequences starting in the 1970s through 1980 because some of the failures that occurred early in the 1970s are important for making an analysis.

The tables show that the relationship between the safe withdrawal rate and P/E10 has been strong from 1921 onward.

The qualitative differences that separate the 1881-1920 era from the later 1921-1980 period include the effects of the Federal Reserve and its ability to influence interest rates versus being on the gold standard. There are additional considerations that I am not prepared to address concerning the reliability of the older data. It is extremely difficult to gather and there is an on-going challenge from economists of the Austrian school (visit www.mises.org) involving distortions (not all of it intentional) in the financial record. The gist of those challenges is similar to our objections of a cost of living comparison among nations. Several months ago, we saw a cost of living comparison that showed, for example, that the USA was much less expensive than the UAE. It is only when you read the fine print that you learn that taxes and housing expenses were excluded from the comparisons. Of course, those are the two biggest items in most household budgets. In any event this challenge has been taking place in academia for decades.

It is likely that these results will be sufficient to prove hocus's hypothesis to a large extent. The exact details will require great care in defining exactly what is established and in specifying the full context. I am hopeful that these results will advance our understanding of peteyperson's new approach, rooted in the concepts of intrinsic value and intelligent management of a cash buffer account. I think that he will be able to reach withdrawal rates much, much higher than 2% with high levels of safety. I expect us to make great progress.

Have fun.

John R.
Last edited by JWR1945 on Wed May 19, 2004 6:08 am, edited 1 time in total.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

1921-1980 PE10 Table

These are the twenty years with the lowest values of P/E10. The withdrawal amounts are listed under the 50% and 80% headings. They are based upon an initial balance of $1000. The allocations are 50% stocks and 80% stocks with the remainder in commercial paper (which is similar to money market funds).

Code: Select all

Year PE10 50% 80%
1921 5.1 98 118
1922 6.2 86 104
1924 8.0 80 96
1923 8.1 78 92
1933 8.7 60 86
1980 8.8 96 106
1975 8.9 72 84
1978 9.2 82 94
1979 9.2 88 98
1932 9.3 62 82
1925 9.6 74 86
1942 10.1 62 92
1943 10.1 68 94
1949 10.2 80 112
1948 10.4 78 110
1950 10.7 74 104
1944 11.0 62 88
1976 11.1 68 74
1926 11.3 70 78
1935 11.4 54 78
Have fun.

John R.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

1921-1980 PE10 Table (continued)

These are the twenty years with the middle values of P/E10. The withdrawal amounts are listed under the 50% and 80% headings. They are based upon an initial balance of $1000. The allocations are 50% stocks and 80% stocks with the remainder in commercial paper (which is similar to money market funds).

Code: Select all

Year PE10 50% 80%
1947 11.4 70 96
1977 11.4 70 76
1951 11.8 72 96
1945 11.9 60 84
1954 12.0 70 92
1952 12.5 70 92
1934 13.0 50 64
1953 13.0 68 88
1927 13.1 68 74
1938 13.5 50 68
1974 13.5 58 60
1958 13.7 60 72
1941 13.9 52 72
1939 15.5 48 62
1946 15.6 54 70
1955 15.9 60 70
1940 16.3 48 62
1971 16.4 50 70
1931 16.7 52 58
1957 16.7 54 62
Have fun.

John R.
Last edited by JWR1945 on Thu Jul 24, 2003 9:11 am, edited 1 time in total.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

1921-1980 PE10 Table (continued)

These are the twenty years with the highest values of P/E10. The withdrawal amounts are listed under the 50% and 80% headings. They are based upon an initial balance of $1000. The allocations are 50% stocks and 80% stocks with the remainder in commercial paper (which is similar to money market funds).

Code: Select all

Year PE10 50% 80%
1936 17.0 46 56
1970 17.0 50 50
1972 17.2 50 50
1959 17.9 52 56
1956 18.2 54 60
1960 18.3 52 54
1961 18.4 52 54
1973 18.7 48 48
1928 18.8 58 60
1963 19.2 50 52
1967 20.4 46 46
1962 21.1 48 50
1969 21.1 44 44
1968 21.6 44 44
1937 21.6 40 48
1964 21.6 48 48
1930 22.3 52 50
1965 23.2 44 44
1966 24.0 44 42
1929 27.0 48 44
Have fun.

John R.
peteyperson
**** Heavy Hitter
Posts: 525
Joined: Tue Nov 26, 2002 6:46 am

Post by peteyperson »

ADVERTISEMENT!!

Low on optimism?

Call on John for a dose of the good stuff!

:P
JWR1945 wrote:I am hopeful that these results will advance our understanding of peteyperson's new approach, rooted in the concepts of intrinsic value and intelligent management of a cash buffer account. I think that he will be able to reach withdrawal rates much, much higher than 2% with high levels of safety. I expect us to make great progress.
peteyperson
**** Heavy Hitter
Posts: 525
Joined: Tue Nov 26, 2002 6:46 am

Post by peteyperson »

John,

No time to read or reply at the mo. Will try to get to this tomorrow, meanwhile have you seen my new thread?

Incidentally, the return on historical analysis on a 60/40 split was 3.6% gross, 2.4% was after deducting taxes in cash interest and dividends. I was looking for a realistic spendable cash from the withdrawal but it is helpful to indicate the gross return. Taxes sucks so much of that away as Bogle did indicate.

A 70/30 split would push it up to 4% gross, 2.8% net (UK taxes).

Petey
JWR1945 wrote:Safe Withdrawal Rate versus P/E10 Data

These tables provide the data to replace the Safe Withdrawal Rate versus P/E10 graph that BenSolar frequently mentions. It is at the bottom of this page: http://rehphome.tripod.com/pestudy1.html
Last edited by peteyperson on Thu Jul 24, 2003 10:27 am, edited 2 times in total.
User avatar
BenSolar
*** Veteran
Posts: 242
Joined: Mon Nov 25, 2002 5:46 am
Location: Western NC

Post by BenSolar »

Very nice stuff, John :D

This will be very useful! :great:
"Do not spoil what you have by desiring what you have not; remember that what you now have was once among the things only hoped for." - Epicurus
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

The tables show that the relationship between the safe withdrawal rate and P/E10 has been strong from 1921 onward.

JWR1945:

Thanks so much for getting this stuff out there. This is going to be a big help to me in trying to gain a firmer grasp of things.

I don't believe that I will be able to participate much here for some time. But please do not think that I will not be studying this material. I will be absorbing it all over time and asking some questions about it at a later date.
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

It is likely that these results will be sufficient to prove hocus's hypothesis to a large extent.

P.S.--Does this mean it's over?
peteyperson
**** Heavy Hitter
Posts: 525
Joined: Tue Nov 26, 2002 6:46 am

Post by peteyperson »

Hocus,

Please tell me this isn't another one of your starting a new board up and then deserting it shortly after? You've done that once before and the board died a slow death. You cannot be thinking of making the same mistake twice surely?

<shakes head in utter disbelief>

Petey
hocus wrote:I don't believe that I will be able to participate much here for some time.
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

Please tell me this isn't another one of your starting a new board up and then deserting it shortly after?

How many months of your life have you given up so that fellow aspiring early retirees could learn the truth about SWRs, Petey? For me, it's 14 months and counting. I've paid my dues on this matter.

If we had 10 people in the FIRE community (I mean both boards) who had each done one-tenth of what I have done so far in this matter, we might have a valid methodology to work with right at this moment, today. I have responsibilities outside what I do at these boards, and I am going to take care of those responsibilities. That's the way it is.

If you think it is taking too long to make progess on this matter, you might want to do something about the problems that are making it take so long. You had lots of opportunities to help out diring the time when we posted together over at the REHP board, and I don't recall you stepping forward very often and asking people to knock off the funny business. It's a community, Petety. What we all get out of it is largely a function of what we all put in. I've put in my share, and then some, and then some more. There is not one other poster who has worked this one-tenth as hard as me, or who has had to take one-tenth of the grief for doing so.

I'll be back. I'll do the work that needs to be done here. But over the next six months there will be an opportunity for some others like yourself to step forward and grab the bull by the horns. We need more people working this, so I hope you increase your level of activity. This should not be a one-man show. One reason why it has taken so long to push ahead on this is the fact that it has largely been a one-man show until now (with some significant exceptions, but not nearly enough of them).

You want something done on this, you get about the business of doing it, Pete. I'm with you all the way. It is simply not healthy that the attitude has developed in the FIRE community that if anything is ever going to be done to improve the accuracy of SWRs, it must be done by hocus. It can't work that way. It needs to be a community effort. It needs to be more than just that one voice all the time. I'm sick of hearing hocus post on this subject. I don't want to think about the reactions others have when they see my screen-name appear in connection with the SWR issue.

It's a good thing that I am taking my leave of this for six months. We need fresh voices speaking up on this matter. I hope yours is one of them, Pete.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

peteyperson
No time to read or reply at the mo. Will try to get to this tomorrow, meanwhile have you seen my new thread?
Look at those horrible tables. Think along these lines:

When stocks were cheap (and P/E10 was low), the P/E10 ranged from 5.1 to 11.4. The lowest withdrawal amount at the first failure was $54 or 5.4% with a 50% stock / 50% cash allocation. It was $74 or 7.4% with an 80% stock / 20% cash allocation. The safe withdrawal rates were either 0.1% or 0.2% lower (since I worked in $2 increments). This means that you can plan to withdraw 5.2% (gross, before taxes) with 50% stocks or 7.2% with 80% stocks...provided that stocks become cheap again (i.e., P/E10 is 11.4 or lower) some time in the future.

If stocks were to return to normal valuation, the P/E10 would range between 11.4 and 16.7. It is reasonable to expect something like this within the next decade. With 50% stocks you can plan to withdraw 4.6% (gross, before taxes) safely. With 80% stocks you can plan to withdraw 5.6% stocks safely...provided that stocks return to their normal valuations (i.e., P/E10 is 16.7 or lower) some time in the future.

For stocks at high valuations, but within the historical range, the P/E10 varied from 17.0 to 27.0. With 50% stocks you can plan to withdraw 3.8% (gross, before taxes) safely. With 80% stocks you can plan to withdraw 4.0% (gross, before taxes) safely.

For stocks at valuations outside of the historical range, you must make some kind of adjustment. The details can be controversial.

The indicated withdrawal rates incorporate the effects of allocations. You do not have to reduce these numbers for that reason. You do have to include the effects of taxes. Your factor was 0.66 (i.e., a tax rate of 34%). At middle range valuations your after tax safe withdrawal rate is slightly above 3%.

What is new is that we have a reliable indicator to convert current prices to intrinsic value. P/E10 is much better than we had thought previously. The anomalies from 1881-1920 made P/E10 look like a crude indicator with great uncertainty (i.e., lots of scatter).

You now have a tool to improve your management of your cash buffer account. Since you are still young, it is highly likely (almost certain) that you will be able to buy stock (at least, the S&P 500 index) at reasonable prices. You may even be able to buy them at bargain prices.

I have read the first post on your new thread. It is outstanding.

Have fun.

John R.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

hocus
It is likely that these results will be sufficient to prove hocus's hypothesis to a large extent.

P.S.--Does this mean it's over?
This means that the worst is over. We will be able to proceed. When people demand that you provide an answer as part of asking a question (as absurd and obnoxious as that may be), you will be able to hand them that answer. The numbers will be mathematically correct.

We now know that P/E10 is an excellent indicator, suitable for the kinds of strategies that you have hinted about in the past. As an aside, it is highly likely that BenSolar's valuation based allocation changes would have enhanced calculated Safe Withdrawal Rates dramatically if it weren't for the masking effect of the 1881-1920 anomalies.

There is much to do. There are a lot of loose ends.

Have fun.

John R.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

peteyperson to hocus:
Please tell me this isn't another one of your starting a new board up and then deserting it shortly after?
That is the wrong way to look at it. hocus has things that he needs to do, but that he has put off until now. He can now leave his day-to-day level of posting for a few months because we have made sufficient progress. Your own investigations are part of that progress. You have introduced some novel ideas of great value.

Don't worry. hocus will be back and he will have some good ideas. He is not deserting us. It is nothing like that.

Have fun.

John R.
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

This means that the worst is over. We will be able to proceed.

Those are encouraging words indeed. There's a lot at stake here, and we have laid the groundwork for creation of an extremely powerful investing tool. I agree that there is a lot of work yet to be done. But it does appear that we have broken through in a big, big way on some key aspects of this thing.

We now know that P/E10 is an excellent indicator, suitable for the kinds of strategies that you have hinted about in the past.

At some point in the future, I think it would help a lot if we could put together some materials showing in practical terms the sorts of benefits that investors could expect to obtain from a valid SWR tool. You did this a bit in your post to Petey, making reference to the sorts of withdrawal rates he may be able to take advantage of if we bring the insights we have developed to fruition. I think that others might get more excited about the work we are doing if we could put forward in dollar terms the sorts of benefits that open up to aspiring early retireees when an SWR is produced showing accurately what the historical data reveals.

From looking at your posts above, it appears to me that the potential of this tool is breathtaking. We are talking about something that could provide benefits worth $100,000 to a single aspiring early retiree, are we not? Multiply that by 10,000, and you have a big number. Am I right in having these sorts of expectations for what this tool can deliver?

I'd say it's a pretty good day for a message board when a post is put up that delivers value in the many millions of dollars to the community it was created to serve.
wanderer
*** Veteran
Posts: 363
Joined: Tue Nov 26, 2002 9:33 am
Location: anytown, usa

Post by wanderer »

outstanding work JWR. :great: :great:

If stocks were to return to normal valuation, the P/E10 would range between 11.4 and 16.7. It is reasonable to expect something like this within the next decade. With 50% stocks you can plan to withdraw 4.6% (gross, before taxes) safely. With 80% stocks you can plan to withdraw 5.6% stocks safely...provided that stocks return to their normal valuations (i.e., P/E10 is 16.7 or lower) some time in the future.

It seems to me there are individual equity issues (and alternative investments - well-chosen real estate opportunities) that would easily put one in the above range. If you could focus on those sorts of purchases, and count on some portion of SSI being left after the dirigistas pick it over, your take out rate would feel like 5.2%-7.2%. good lord I've worked too long! ;)
regards,

wanderer

The field has eyes / the wood has ears / I will see / be silent and hear
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

outstanding work JWR.

Wanderer:

It would have been better for the FIRE community in general and this site in particular if you had added something to the effect of: "I now see what hocus has been saying from the first day, that calculating the SWR properly would permit thousands to achieve their dreams of financial independence sooner and cause no downside whatsoever. I hope that the entire FIRE community now recognizes why hocus was willing to give up 14 months of his life to help us with this, and I hope that some on this board who have made unfair comments about him will come forward today with the appropriate apologies. As for myself, I wish that I had spoken up sooner when things started getting out of hand."
wanderer
*** Veteran
Posts: 363
Joined: Tue Nov 26, 2002 9:33 am
Location: anytown, usa

Post by wanderer »

It would have been better for the FIRE community in general and this site in particular if you had added something to the effect of: "I now see what hocus has been saying from the first day, that calculating the SWR properly would permit thousands to achieve their dreams of financial independence sooner and cause no downside whatsoever. I hope that the entire FIRE community now recognizes why hocus was willing to give up 14 months of his life to help us with this, and I hope that some on this board who have made unfair comments about him will come forward today with the appropriate apologies. As for myself, I wish that I had spoken up sooner when things started getting out of hand."

:shock:

You got a lotta nerve JWR/hocus/whoever wrote that post (the author appears to have changed).

BTW, what happened to the feel good "why can't we stop tearing at each other" tenor of hocus' recent post? Does the fact that I think hocus has his head up one of his softer apertures on many of these points explain my behavior at all?
regards,

wanderer

The field has eyes / the wood has ears / I will see / be silent and hear
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

Does the fact that I think hocus has his head up one of his softer apertures on many of these points explain my behavior at all?

It's over, Wanderer.

You described the JWR1945 post as "outstanding work." That post states that: "It is likely that these results will be sufficient to prove hocus's hypothesis to a large extent." The claims that I put forward were important claims and those claims have been vindicated.

Asked to amplify, JWR1945 stated: "This means that the worst is over. We will be able to proceed. When people demand that you provide an answer as part of asking a question (as absurd and obnoxious as that may be), you will be able to hand them that answer. The numbers will be mathematically correct. We now know that P/E10 is an excellent indicator, suitable for the kinds of strategies that you have hinted about in the past "

A numbers guy has studied the data closely. He found that the data says what I said it said all along. No one else had ever said it. I did. I was right. I told the FIRE community the truth about SWRs and it was an important truth to tell.

Everyone benefits from the insight. I provided the insight. That's what happened. There's no charge. But a "thank you" would be nice.
User avatar
BenSolar
*** Veteran
Posts: 242
Joined: Mon Nov 25, 2002 5:46 am
Location: Western NC

Post by BenSolar »

wanderer wrote::shock:

You got a lotta nerve JWR/hocus/whoever wrote that post (the author appears to have changed).

BTW, what happened to the feel good "why can't we stop tearing at each other" tenor of hocus' recent post? Does the fact that I think hocus has his head up one of his softer apertures on many of these points explain my behavior at all? < big wide completely serious grin >
LOL

I wondered what kind of reaction we'd see from wanderer. :D

BTW - I discovered the < big wide completely serious grin > thing only when I did the quote - the posting software apparently considers anything directly within < > to be code and doesn't display it.

Also - I only saw the author as hocus. Did you really see it as JWR? Strange, if so.
"Do not spoil what you have by desiring what you have not; remember that what you now have was once among the things only hoped for." - Epicurus
Post Reply