Skip to main content
Update
Non-financial boards have been closed.

Non-financial boards have been closed but will continue to be accessible in read-only form. If you're disappointed, we understand. Thank you for being an active participant in this community. We have more community features in development that we look forward to sharing soon.

Fool.com | The Motley Fool Community
Message Font: Serif | Sans-Serif
 
No. of Recommendations: 6
Reference: Galagan's 4-14-03 3:04 pm post number 101797 on the Re: The Safe Withdrawal Rate Debate Thread

It is improper for me to post directly on that thread since only you, hocus and intercst are part of the debate. However, I have pulled up a couple of posts from my files. I have placed them on the http://nofeeboards.com site. There are many other threads related to this subject.

This one was meant to become an input to developing FAQS related to Safe Withdrawal Rates.

Alternative Methods of Determining Safe Withdrawal Rates (12-10-02)

A. The Historical Sequence Approach
1. This method uses historical data directly.
a) It relies on minimal assumptions. Ideally, it would make none. But the historical record is incomplete in some respects. Best estimates fill in the gaps. For example, the SP500 index itself has not been around for long and index funds have been around even less. It is necessary to extend the historical record.
b) To a large extent the limitations of the historical record are common to all approaches. However, their effects are more pronounced when using historical sequences. Although it is possible to synthesize a reasonable (backwards) extension of the SP500 index, that is not true for many asset classes. Quite a few have extremely short histories. REITS constitute a major asset class that is very new. Income producing properties such as apartments have been around for a long time. But meaningful statistics are quite limited. Most records are very short. Most of the information available is strictly local in its application.
c) The historical sequence approach is inefficient in extracting information. The effective number of data points is quite limited. This is also an advantage. The historical sequence approach is inefficient in extracting false information as well.
d) The historical sequence approach has a credible basis. It cannot project future events that could never happen. After all, the sequences themselves have all happened once.
e) The credibility of the historical sequence approach is easily exaggerated. There are many qualitative changes that have taken place over the years. Some of these are dramatic enough to exclude certain projections from the historical record. Examples include the introduction of the Federal Reserve, the move away from the gold standard and both the introduction and numerous changes in income taxes.
f) The historical sequence approach produces a unique set of answers. They can be replicated. At a very detailed level there are differences among models. For example, results differ for different choices of days, months and years in making calculations. Results from investing each month differ from investing once a year. Results from investing at the beginning of a month differ from those at the end of the month.
g) The historical sequence approach is transparent. It is easy to identify assumptions.
h) The historical sequence approach often makes it easy to identify cause and effect. One can look into the general history of events to interpret what happens in any particular sequence. It is always a great advantage to know cause and effect relationships (as opposed to a simple correlation of events). It helps you estimate the reliability of your projections.
B. Deterministic Models
a) Deterministic models are based on information extracted from the historical record.
b) Examples include estimates of the periods and magnitudes of stock market cycles.
c) It is common to introduce a degree of randomness. In its simplest form randomness is added at a model's output and it is easily understood. Eventually, there can be enough random elements that the model becomes a Monte Carlo model.
d) Some of the elements in a deterministic model are based on cause and effect relationships. Others are from simple correlations. Their root causes are not known. That makes their projections are unreliable…even though they may turn out to be accurate.
e) There are many successful deterministic models. Often, the most successful are also the simplest and most transparent. One example is the projection that the stock market grows at 10 to11% annualized in nominal dollars or 6% to 7% in real dollars. Another example involves seasonal variations. There are good days to buy and to sell for each month. There are good months to own stocks and there are not so good months.
f) Transparency is important since it can help you avoid making serious errors. An example of a hidden assumption is the Federal Reserve's model for determining the fair value of stocks. It is based on limited data. During the short time history behind those data, the effects of inflation on the stock market were embedded in the treasury bond yield curve. Whether that will continue into the future is uncertain.
C. Monte Carlo Models
a) Monte Carlo models are based on information extracted from the historical record.
b) There are different degrees of transparency among different Monte Carlo models.
c) Monte Carlo models introduce randomness in a complex fashion. There can be many elements with random components.
d) With a Monte Carlo model a random number generator provides a value for each element with a random component during each run. Those values vary from run to run. After making a large number of runs, you look at the distribution of the results. They form your estimate of the true probability distribution.
e) Monte Carlo models can produce answers with as great a statistical confidence level as you desire.
f) Using Monte Carlo models does not guarantee that their results are accurate. Accuracy depends on how well information is extracted from the historical record.
g) Using a well-designed Monte Carlo model, you can make reasonable projections at the extremes of a probability distribution because of their large number of runs.
h) It is necessary to determine when the outputs from a Monte Carlo model can be relied upon. That determination is non-trivial.
i) There are several excellent Monte Carlo models.
j) Raddr's advanced Monte Carlo model emphasizes what he calls reversion to the mean. The definition of that phrase is somewhat ambiguous. What raddr's model takes into account is that the spread of actual stock market returns is much less than is usually projected. Projections usually decrease the spread (standard deviation) by a factor approximated by 1/SQRT(N), where SQRT means “square root” and N is the number of years of the projection.
k) Gummy's (Professor Ponzo's) Monte Carlo model emphasizes the historical sequences of inflation. Year-to-year stock market fluctuations are treated as random (with a historically based non-normal distribution). Inflation sequences are selected at random but not year-to-year inflation fluctuations.
m) I would personally like to see a Monte Carlo model that uses a long-term cyclical variation to account for behavior of the spread in actual returns. We know that there are some factors in the stock market that vary slowly. The SP500 price to earnings ratio (using Professor Shiller's P/E10 numbers) has varied slowly. Most recently, that multiple increased over a period of two decades, the 1980s and the 1990s.
D. The Reliability of Projections
a) It is a serious mistake to rely on either the historical sequence approach or a Monte Carlo model by itself. Both approaches have something to add.
b) I use the dory36 calculator for my investigations. It is great for rejecting things that do not work. Not only is it fast, it is quite credible when indicating failure. It helps me understand the results since I know when effects would have occurred. By running a variety of conditions, I can gain insights as to possible cause and effect relationships.
c) I am not satisfied with the details of any positive result that I get from the dory36 calculator. I am comfortable with certain broadly defined conclusions when I have cause and effect. But the numbers themselves (such as for a safe withdrawal rate or a number of failures) are uncertain. There may be hidden sensitivities. Exact sequences can cause results different from typical sequences.
d) One thing that raddr has done is to match his Monte Carlo model results to the entire distribution in the historical record. It comes close enough to give me confidence in small extrapolations outside of the historical range. His results are traceable to a large number of small variations. Those results do not depend on one or two critically significant events in isolation.
e) The historical sequence approach has an inherent granularity built into it. There are only three or four completely independent sequences in the historical record. There are many partially independent sequences. There are 130 data points that form these sequences. The statistical behavior lies between those extremes. It is better than just having 3 data points. It is worse than having 130 data points. The historical sequences form an estimate of an underlying distribution. When you desire a high degree of safety, you are looking at a tail of the underlying distribution. You have very few relevant sequences with limited independence when you make your estimate.
f) There is another factor to consider whenever making a projection. It has to do with the underlying distribution. This question can never be answered entirely in advance but it must be addressed. What characteristics of the underlying distribution will remain the same and what characteristics will change in the future? For example, we have been well outside of the historical range for valuations since the mid-1990s. Should we ignore that fact? I think otherwise. Then, there is a question about the sequences themselves. To what extent is it necessary to consider the recent past? If you look at 50 year sequences, all of the sequences since 1950 (1952 actually) are incomplete. I can certainly believe the failures among the partial sequences. But I cannot believe the successes.

Have fun.

John R.


Print the post Back To Top
No. of Recommendations: 5
Here is another of my posts.

FIRE Consistent fallacies (1-20-03) by JWR1945

Consistent fallacies

1. mhtyler identified three major flaws related to the Retire Early Safe Withdrawal Rate study in the How do we handle dividends? thread. They relate to the applicability of the study and not the study itself (when taken in isolation). This draws attention to two fallacies that I have seen repeatedly.

2. The first fallacy is that we have a relevant and reliable historical record. There are some features of the historical record that we can rely upon. There are others that we cannot. The most important feature is that the last few decades are not part of the historical record involving safe withdrawal rate calculations. More precisely, there are no complete historical sequences that include current stock market valuations. As we look back in history, we see even more qualitative changes. Some are highly significant such as going off the gold standard and introducing the federal reserve banking system. Others are highly relevant but probably less significant, such as the availability of index funds, the availability of modern REITS and major changes in the Federal Income Tax laws. The easiest way to summarize this is to say that it really is different this time in many ways and it is not different this time in others.

3. The other fallacy is the assertion that nobody can time the market successfully over a long period of time. This has been proved untrue. Mark Hulbert of the Hulbert Financial Digest has demonstrated this by following investment newsletters. Recently, a couple of university doctoral candidates [this is a correction] have expanded upon Hulbert's findings. They have used Hulbert's database to prove the point using several dozen criteria for success in timing the market.

4. The timing assertion is especially galling because it is applied without any restrictions or thought whatsoever. It is true that the definition that Hulbert uses to measure the effects of market timing does identify all allocation changes as market timing. That includes allocation changes according to stock market valuations. In that sense, Hulbert's definition includes Sir John Templeton and Warren Buffett and Benjamin Graham. Not only is the timing assertion false, it maligns great investors of both the past and present.

Have fun.

John R.
Print the post Back To Top
No. of Recommendations: 4
I considered this exchange to be especially enlightening. You should not interpret this as being all inclusive. But it brings up many critical factors.

Have fun.

John R.

FIRE FMO to raddr and SWR definition (1-20-03)

From the How do we handle dividends? thread
on the http://nofeeboards.com site.

raddr:

Not only that but no one has ever actually lived by the premise of the study all the way to death. Since the S&P500 was invented only about 45 years ago and index funds only 25 years ago the time period is too short for anyone to have actually lived out a relatively long retirement according to REHP mantra. My contention is that few, if any, ever will. It is simply too mechanical, risky, and undiversified for mere mortals to live with after the paychecks stop.

FMO

I agree completely. My point was that it is unproductive to challenge the results of the study directly. It is the assumptions built into the intercst model which are the actual bone of contention. In my view, this is what in large part leads to the formation of the circus environment when hocus posts at the REHP Board. He keeps challenging the accuracy of the study results with the result being that his arguments get rammed down his throat since the results are analytical and follow directly from the model assumptions. It doesn't help as well that many of the board regulars resort to demagoguery, deception and ridicule to muddy the waters.

Instead of directly challenging the model results it is more useful to challenge and debate its assumptions and limitations. They are actually very numerous and provide quite fertile ground for investigating the validity of the results which follow.

1. The study assumes an almost superhuman ability to follow a mechanical regimen which would have even experienced, sophisticated investors wetting their pants.

2. The study assumes investment expense ratios which were not practically available to investors over the entirety of the period which the study seeks to address.

3. The study is by definition incomplete and probably sub-optimal since it can only address the use of asset classes for which 130 years of data exists.

4. The data which do exist are not unquestionable.

5. As you have noted, no single individual as of yet has proven successful at following the dictates of the study.

6. The study has recently accumulated data representing a valuation condition which cannot be assessed due to the lack of available data.



Print the post Back To Top
No. of Recommendations: 3
JWR1945 writes,

4. The timing assertion is especially galling because it is applied without any restrictions or thought whatsoever. It is true that the definition that Hulbert uses to measure the effects of market timing does identify all allocation changes as market timing. That includes allocation changes according to stock market valuations. In that sense, Hulbert's definition includes Sir John Templeton and Warren Buffett and Benjamin Graham. Not only is the timing assertion false, it maligns great investors of both the past and present.


I don't think that anyone disputes the success of Buffett, Templeton et al, but we have yet to see evidence that the average investor can improve returns using some form of market timing or trading.

Those with an interest in the academic research on the subject may find this April 2000 paper by UC Berkeley Professor Terrance Odean illuminating:

http://faculty.haas.berkeley.edu/odean/papers/returns/Individual_Investor_Performance_Final.pdf

ABSTRACT

Individual investors who hold common stocks directly pay a tremendous perfor-mance penalty for active trading. Of 66,465 households with accounts at a large discount broker during 1991 to 1996, those that trade most earn an annual return of 11.4 percent, while the market returns 17.9 percent. The average household earns an annual return of 16.4 percent, tilts its common stock investment toward high-beta, small, value stocks, and turns over 75 percent of its portfolio annually. Overconfidence can explain high trading levels and the resulting poor performance of individual investors. Our central message is that trading is hazardous to your wealth.


intercst
Print the post Back To Top
No. of Recommendations: 5
John -

I appreciate the response very much. My hope was that the debate would generate side threads like this, and I would hope that normal posting rules apply and everyone will feel comfortable contributing. Certainly I don't want to be considered a moderator for this.

Let me pose the following initial thoughts and questions:

1. To the extent that one introduces deterministic/Monte Carlo methodology, one must make assumptions about the distribution from which random samples are drawn. Many people who favor the intercst approach seem to think that these assumptions can never be relied on, and so they feel safer ignoring the question. But it seems to me that one could at least use a Monte Carlo study to figure out whether a given number (e.g., 4 percent) is anywhere near the right ballpark. In other words, rather than trying to employ the method to come up with its own number, instead use the method to see if you can disprove the hypothesis developed from using pure historical methods.

2. How can one address the granularity of the data without punting? I mean, we only have the historical record that we have. And frankly, I'm skeptical that data from the 19th century is particularly useful - it seems to me the distribution of possible returns would change over time.

3. You'll clearly need to exercise judgment in making some decisions in analysis. But in some cases, one might want to do an analysis both ways to see what happens. For instance, if we have been outside of normal ranges of valuations for an extended period, it might be nice to see not only what happens if valuations revert to normal, but also what happens if you assume the value distribution is permanently changed (the "new paradigm" argument). You wouldn't necessarily believe what that analysis told you, but it would give you a sense of what the extremes look like.

4. Speaking of extremes, it seems to me that Monte Carlo analysis gives the advantage of providing information on extremely improbable "tail" results. This can give the extremely conservative investor whatever confidence they seek. The references in the REHP study to a 100-percent "confidence" level have always seemed misleading to me.

5. Although I will admit that some outliers challenge the assertion that market timing is not possible, do you believe a study can be reasonably based on the assumption that the participant is an "above-average" investor? Obviously not everyone can be above average, and so results based on such an assumption might give a false sense of security to those who are merely average in their investing.

6. I see some value in challenging and questioning the assumptions and limitations of the study. Many will not. Many want an answer, and they will accept a flawed answer over a correct non-answer. Then given that flawed answer, if things don't work out the way they hope, then they will make modifications.

That last has been my major problem with the whole SWR question. If someone is comfortable with a 4 percent withdrawal and then their portfolio drops 50 percent, will they really keep taking withdrawals based on the original portfolio value? It seems to me that nearly everyone would look for ways to cut back. Similarly, if their portfolio does extremely well, they would spend more. If a study doesn't take that into account, how much practical good does it do?

Again, thanks for your thoughts.

dan
Print the post Back To Top
No. of Recommendations: 3
The Berkeley study is particularly stupid. It took what a certain bunch of people did and demonstrated that they did not do it well. They also did this in a bull market, when buying and holding just about anyting worked.

The question is not what the "average investor" can do. Nobody with any sense wants to be "average". I cannot imagine why, when it comes to the stock market, people use phrases like "for the average investor...", as if they are trying to teach you to be average.

In any other discipline (if you call it a discipline), if you took a class and the instructor told you that his class would teach you to be "average", would you think your tuition was well spent? I certainly would not.

There are a number of ways to time the market which work overall. There is the stuff I do. Then consider the CANSLIM method, for example, which nobody here ever wants to discuss here. It is a particular discipline which provides superior results. But CANSLIM, like the other methods, requires some work, and people that are lazy are simply not interested in the work required.

So they are happy to be average, even if that means losing 40% or so. Then they get downright hostile when somebody points out that they did not have to do that. After all, it is not as if God said, "Well, it is time for everybody to start losing money, so here goes." He might have expected people to exercise their intelligence and get out of the way of the falling market just like they would get out of the way of a bunch of falling rocks. Only it is easier to get out of the market. You don't have to be in shape to run away.
Print the post Back To Top
No. of Recommendations: 1
Our central message is that trading is hazardous to your wealth.

I'll only point out that market timing does not necessarily imply a high portfolio turnover rate. I think all the investors cited by JWR1945 would agree that excessive trading is not advised.

dan
Print the post Back To Top
No. of Recommendations: 2
OK....I scanned 5000 words by others....

So, the bottom line

FOr a 30 year withdrawal on a 60/40, 70/30, and 50/50 portfolio, done by an 'average investor', in both a tax deferred and a taxable account, please provide answers for the 6 scenarios involved to the single question:

What is the Safe Withdrawal rate for a retiree, that would have been 99.9% safe and survived any 30 year period in the past?

Then convince us it is going to be different for any 30 year period in the future.



Complain all you like about the existing model, but if you can't come up with a new one, then all this stuff doesn't appear to be going anywhere soon, right?

Print the post Back To Top
No. of Recommendations: 1
Joel:The question is not what the "average investor" can do. Nobody with any sense wants to be "average". I cannot imagine why, when it comes to the stock market, people use phrases like "for the average investor...", as if they are trying to teach you to be average.

I agree. In the report, on page 19, there is a very interesting table that reveals that 43.4% of accounts beat the market over the reviewed time period and 25% of accounts beat the market by more than 6% annually, all net of transaction costs. Furthermore, the top 10% of accounts beat the market by over 18% annually, again, net of transaction costs.

Then consider the CANSLIM method, for example, which nobody here ever wants to discuss here. It is a particular discipline which provides superior results. But CANSLIM, like the other methods, requires some work, and people that are lazy are simply not interested in the work required.

I think CANSLIM is an excellent 'next step' for many here at the Fool because it builds upon Fundamental Analysis by adding a little technical and market indicators to increase your probability of buying the right stock. It does nothing fancy, it doesn't predict the future, but I think it improves the odds.

Perhaps you and I should have a weekly discussion on a different way to invest. I won't pretend to know everything (or even anything) but I do have open eyes.

st
Print the post Back To Top
No. of Recommendations: 3
Galagan

1. To the extent that one introduces deterministic/Monte Carlo methodology, one must make assumptions about the distribution from which random samples are drawn. Many people who favor the intercst approach seem to think that these assumptions can never be relied on, and so they feel safer ignoring the question. But it seems to me that one could at least use a Monte Carlo study to figure out whether a given number (e.g., 4 percent) is anywhere near the right ballpark. In other words, rather than trying to employ the method to come up with its own number, instead use the method to see if you can disprove the hypothesis developed from using pure historical methods.

My comments were not meant to be comprehensive. They are meant to show that there is a lot worth more worth knowing.

I do not consider deterministic approaches, Monte Carlo simulations and historical sequence analysis (the intercst method) to be mutually exclusive. Instead, each element adds a little bit more to our knowledge and understanding. They are all valuable. They are all complementary, not exclusionary. I choose the all of the above option.

I generally prefer not to insist upon distinctions based on semantics. But I have to when I read: instead use the method to see if you can disprove the hypothesis developed from using pure historical methods. A historical sequence analysis does use historical data. The others use the historical record as well. It is just that they extract a different set of information.

The historical sequence approach is always definitive when it shows failure. It is not definitive when it shows success. That is where the other methods help out.

There are a host of sensitivity studies that can be done with each approach. raddr has done some interesting research along those lines. One method is to modify historical sequence information minimally and see what would have happened. The changes were dramatic. raddr swapped the returns from two years and then calculated the historical safe withdrawal rate (using the intercst method). He chose pairs of years randomly. All of the statistics that you would normally calculate remain identically the same when you use this approach. But the safe withdrawal rate dropped considerably for a substantial number of what if cases. I do not remember the numbers. IIRC, a 4% safe withdrawal rate (at 100% safety) ended up being below 3.5% about 20% of the time. Again, do not trust my memory on this.

Similarly, there are a variety of sensitivity studies that you can do with a Monte Carlo simulation that you cannot do from historical sequences alone. You can introduce what if cases related to future stock market returns. You can add other asset classes with a limited historical record. To overcome the uncertainties related to their future performance, you can examine a multitude of test conditions with various underlying statistics.

Let us not forget that intercst included several sensitivity studies as part of the Retire Early Safe Withdrawal Rate study and he has continued to add more. Doing sensitivity studies is a routine engineering procedure.

In recent days we have been hearing the word mosaic quite often. The various methods coupled with sensitivity studies combine to fill in that mosaic. We do not have a full picture yet. But we have much better insights when we use all of the available methods and seek understanding as to why they give the results that they do.

More responses will follow.

Have fun.

John R.
Print the post Back To Top
No. of Recommendations: 2
Galagan

How can one address the granularity of the data without punting? I mean, we only have the historical record that we have. And frankly, I'm skeptical that data from the 19th century is particularly useful - it seems to me the distribution of possible returns would change over time.

IMHO, this is the strongest reason to look at Monte Carlo models and to do a host of sensitivity studies. If your Monte Carlo model predictions match the historical sequence results reasonably well throughout its entire probability distribution, you gain confidence in your predictions. The historical sequence approach has only a few independent data points. It gives you lousy statistics but you trust the results. The Monte Carlo model has a large number of independent data points. You have great statistics but you do not trust the results until you compare it to the historical sequence data. If they are close, you have a useful analysis tool.

You can also do many sensitivity studies by modifying the historical sequence data and seeing what would have happened. I mentioned raddr's swapping pairs of years earlier. Another possibility would be to rearrange a whole sequence of years. How about swapping the 1990s and the 1950s? It might be interesting.

Have fun.

John R.
Print the post Back To Top
No. of Recommendations: 23
John R.:In recent days we have been hearing the word mosaic quite often. The various methods coupled with sensitivity studies combine to fill in that mosaic. We do not have a full picture yet. But we have much better insights when we use all of the available methods and seek understanding as to why they give the results that they do.

I have a question. This is a serious question, and although no one has to answer please don't think that I am being argumentative or anything but sincere. Here goes: At the end of the day, when all of these various methods of estimating fSWR have been examined, do you think that the time and effort spent will have been worth it?

I mean, if I spent 100 hours of my time examining various ways of estimating the fSWR and I came up with a range of values. For example, if the various ways of estimating came up with 3.5-4.5% as a probable range, what would I have gained from undertaking the effort? I would have a considerable amount of understanding about what factors affect the fSWR, but I would not have a fundamentally different view of the fSWR than I do now. It is roughly 4%. If the results come back to be 2%, then I think I would rather retire at 4%, hope the future is no worse than the past, and enjoy myself. And if it turns out that the future is worse than the past, then I will go on with life, living worse than before, along with everyone else, as the future will be worse than the past.

In any event, I will have spent 100 hours of my life learning more about something that really won't make a huge difference in my life one way or the other. 100 hours away from golf, away from the family, and 100 hours not sleeping!

Do you think you will find something meaningful enough to justify the time spent?

st
Print the post Back To Top
No. of Recommendations: 10
SloanT:

At the end of the day, when all of these various methods of estimating fSWR have been examined, do you think that the time and effort spent will have been worth it?

I see the exercise as not so much, getting a more accurate idea of what a reasonable fSWR is, but gaining insight into how an individual can structure his investment vehicles and approaches to maximize the probability of success. The most accurately determined probable SWR is always going to be a function of the asset classes selected and the management techniques which can be brought to bear on the portfolio.

Consider this: If we restrict ourselves to the historical sequence method as applied to a portfolio composed of only two asset classes, we will never be able to demonstrate a SWR in excess of what intercst's model currently provides. Because the maximum SWR is calculated based on demonstrated past failures, it can never increase. In the fullness of time, as additional data comes in, it is possible for the SWR to go down, but it will never, ever go up.

I refuse to allow my line of inquiry to be restricted by a self-imposed straight jacket. Once you free yourself from one specific evaluation technique (historical sequences), two asset classes (large cap equities and bonds) and one management approach (annual rebalancing) who knows what may be possible? If by considering other evaluation and management techniques and by expanding the range of asset classes considered, a case can be made higher safe withdrawal rates, will the effort have been worth it? I think so. Maybe the effort has a low probability of success but how will you know if you don't try? If you try you may fail. If you don't try you are guaranteed to fail. There are relatively few folks seriously looking at these issues (most of which enjoy it) and yet there are thousands that could potentially benefit.

Regards,
FMO
Print the post Back To Top
No. of Recommendations: 4
Galagan

3. You'll clearly need to exercise judgment in making some decisions in analysis. But in some cases, one might want to do an analysis both ways to see what happens. For instance, if we have been outside of normal ranges of valuations for an extended period, it might be nice to see not only what happens if valuations revert to normal, but also what happens if you assume the value distribution is permanently changed (the "new paradigm" argument). You wouldn't necessarily believe what that analysis told you, but it would give you a sense of what the extremes look like.

4. Speaking of extremes, it seems to me that Monte Carlo analysis gives the advantage of providing information on extremely improbable "tail" results. This can give the extremely conservative investor whatever confidence they seek. The references in the REHP study to a 100-percent "confidence" level have always seemed misleading to me.


These are two good reasons for looking at deterministic models and Monte Carlo simulations as well as historical sequence results. Notice that I say as well as and not instead of. Use all three and make a host of sensitivity studies. Then you get a better idea of what the future might be.

Looking at valuations highlights the usefulness of a deterministic approach. If you take a simple comparison of P/E10 of the S&P 500 during the recent bubble and in previous eras, especially just before the Great Depression, the valuations during the bubble were far higher than ever before in the historical record. How do you address that issue? The approach that I prefer to use is to assume that the underlying earnings were the key to the safe withdrawal rates of the past. Under that assumption, you can scale your predicted safe withdrawal rate number down from 4% by the ratio of 1920s P/E10 to the current numbers. Using this approach, the safe withdrawal rate estimate at the peak of the bubble would have been scaled down to 2.3%.

(P/E10 is the current price of the S&P 500 index divided by the average of the earnings from the previous ten years. The ten year average is reasonably well behaved. Year to year earnings fluctuate wildly.)

Similarly, you can look at earnings and dividends and use them as inputs to the Dividend Discount Model (suitably modified) to project the stock market's real rate of return. That can be used as an input to a Monte Carlo simulation. In this case, you use both a deterministic approach and a Monte Carlo simulation to estimate a safe withdrawal rate. Include a sensitivity analysis and you can get some interesting results.

Have fun.

John R.
Print the post Back To Top
No. of Recommendations: 3
JWR1945:

Galagan: <<<How can one address the granularity of the data without punting? I mean, we only have the historical record that we have. And frankly, I'm skeptical that data from the 19th century is particularly useful - it seems to me the distribution of possible returns would change over time.>>>

"IMHO, this is the strongest reason to look at Monte Carlo models and to do a host of sensitivity studies. If your Monte Carlo model predictions match the historical sequence results reasonably well throughout its entire probability distribution, you gain confidence in your predictions.

The historical sequence approach has only a few independent data points. It gives you lousy statistics but you trust the results. The Monte Carlo model has a large number of independent data points. You have great statistics but you do not trust the results until you compare it to the historical sequence data. If they are close, you have a useful analysis tool."


Why do we want to assume that year over year results of the market are independent? I am not so sure that they are, and assuming that they are in order to Run Monte Carlo analysis so that one had a large number of "independent" data points introduces (and probably increases, IMO) room for error.

It is the same issue that keeps being rephrased in some fashion that with only 130 years of data and rolling 30-year time periods there are only 4-5 independent data points. One the one hand, I am not sure that waiting for 30 years to expire and calling years 31-60 an independnet data point is accurate because it assumes that year 31 is independent of year 30. How long needs to elapse (if ever) before the data is truly independent?

OTOH, the historical data is what it is, so that the only resolution to this issue must introduce extra assumptions into the process. I agree that this can be useful for sensitivity analysis, but I am still undtermined about the degree of reliability be gained thereby. IOW, does the introduction of extra assumptions create a margin of error that is even larger than than the concern about "independent" data points (assuming that arguendo they are actually independent) or does it lessen the margin of error?

"You can also do many sensitivity studies by modifying the historical sequence data and seeing what would have happened. I mentioned raddr's swapping pairs of years earlier. Another possibility would be to rearrange a whole sequence of years. How about swapping the 1990s and the 1950s? It might be interesting."

Ditto regarding assumption of independence based upon calendar year.

Just my $0.02. Regards, JAFO
Print the post Back To Top
No. of Recommendations: 4
Galagan

5. Although I will admit that some outliers challenge the assertion that market timing is not possible, do you believe a study can be reasonably based on the assumption that the participant is an "above-average" investor? Obviously not everyone can be above average, and so results based on such an assumption might give a false sense of security to those who are merely average in their investing.

I have no desire to discuss stock market timing here. I just have recently started two threads on this subject on the http://nofeeboards.com Index Funds discussion board. They are Market Timing References and What the researchers found out.

One thing that is central to any discussion about stock market timing is its definition. It is not easy to define in an objective manner that is suitable for the mindless analysis of a computer.

There is the additional problem that many people react to the words market timing in a mindless, hostile, emotional, knee jerk fashion. That makes reasoned discussion difficult. Fortunately, posting rules are enforced on the http://nofeeboards.com site and the discussions remain civil.

Have fun.

John R.
Print the post Back To Top
No. of Recommendations: 3
FoolMeOnce: "Consider this: If we restrict ourselves to the historical sequence method as applied to a portfolio composed of only two asset classes, we will never be able to demonstrate a SWR in excess of what intercst's model currently provides. Because the maximum SWR is calculated based on demonstrated past failures, it can never increase. In the fullness of time, as additional data comes in, it is possible for the SWR to go down, but it will never, ever go up."

Agreed.

"I refuse to allow my line of inquiry to be restricted by a self-imposed straight jacket. Once you free yourself from one specific evaluation technique (historical sequences), two asset classes (large cap equities and bonds) and one management approach (annual rebalancing) who knows what may be possible? If by considering other evaluation and management techniques and by expanding the range of asset classes considered, a case can be made higher safe withdrawal rates, will the effort have been worth it? I think so. Maybe the effort has a low probability of success but how will you know if you don't try? If you try you may fail. If you don't try you are guaranteed to fail. There are relatively few folks seriously looking at these issues (most of which enjoy it) and yet there are thousands that could potentially benefit."

I also suggest that to be valuable to many people, whatever method and management techniques should be readily reproducible (or "easy" to follow) and not be highly dependment upon personal knowledge unique to the investor; in addition, I suggest that most retirees are looking for something more akin to a "fire and mostly forget" method (annual rebalance to maybe quarterly) and not one that requires 4 hours a day watching for Elliot Waves or the equivalent of 20+ hours/week pouring through 10-ks and other financial statements --- think more like Scott Burns Couch Potato portfolio.

Graham & Dodd wrote their bible decades ago, and while Warren Buffett and Charlie Munger have been very succesful, and so have certain other disciples, there would be far more if it was as easy as some people suggest. For example, CC is a ready proponent of real estate, but she is in the ortage business -- with ready access to lenders and a whole network of real esate agents and brokers. She is in a position to accomplish certain things that can be much harder for people without those same contacts and information.

Regards, JAFO



Print the post Back To Top
No. of Recommendations: 4
This is a serious question, and although no one has to answer please don't think that I am being argumentative or anything but sincere. Here goes: At the end of the day, when all of these various methods of estimating fSWR have been examined, do you think that the time and effort spent will have been worth it?

For me to hear a sincere question at this place is like a man walking in the desert coming across a tall glass of water.

SWR analysis is an extremely powerful tool, SloanT Discussions held at this board to date have not touched the surface of what can be done with a SWR analysis, properly constructed.

You are looking at the tool from a narrow perspective. You are thinking that it can only do one thing, tell you the take-out percentage from your portfolio. That's like people who used to say that everyone would want a home conputer because they could use it to store their receipes. It's technically true, but it understates the potential of the tool by so much that it is sort of silly..

You can do scores of things with a SWR analysis, properly done, that you cannot imagine today. There is no way to even list the things until the board reaches a point where it is willing to allow a small amount of valid information on the subject to be posted and discussed. But if we ever get to that point, you won't be asking afterwards whether it was all worth it or not.

If we ever get the the point where we permit informed discussions of the SWR concept at this board, we won't have to worry about having enough on-topic threads at this place for a long time to come. We will be working on something that matters instead of the trash we devote all of our energies to today.
Print the post Back To Top
No. of Recommendations: 5
Galagan

6. I see some value in challenging and questioning the assumptions and limitations of the study. Many will not. Many want an answer, and they will accept a flawed answer over a correct non-answer. Then given that flawed answer, if things don't work out the way they hope, then they will make modifications.

That last has been my major problem with the whole SWR question. If someone is comfortable with a 4 percent withdrawal and then their portfolio drops 50 percent, will they really keep taking withdrawals based on the original portfolio value? It seems to me that nearly everyone would look for ways to cut back. Similarly, if their portfolio does extremely well, they would spend more. If a study doesn't take that into account, how much practical good does it do?
[Emphasis added.]

These are the questions that got hocus in trouble in the first place. It is not so much a matter of questioning any answer per se. It is more a matter of adding a whole lot more information to get a better picture...to fill in the mosaic. That is what you get from doing sensitivity studies and using additional analysis approaches.

I have observed that real retirees do cut back, just as you have suggested. They may have a lot of confidence in the Retire Early Safe Withdrawal Rate study. But they do not have enough confidence to withdraw 7% or 8% in today's bear market. That is a critical factor that hocus has wanted to consider...without being told that anybody who makes such a change is (insert bad word here).

Have fun.

John R.
Print the post Back To Top
No. of Recommendations: 1
You have no recommendations left for today.

Damn! You are asking the right question, st, and I whish I could rec it.
Print the post Back To Top
No. of Recommendations: 4
FMO:The most accurately determined probable SWR is always going to be a function of the asset classes selected and the management techniques which can be brought to bear on the portfolio.

Ok, I buy that. Doesn't it make more sense to frame your discussion on asset allocations, then, and not SWR? Personally, I think the SWR is far, far less important than most major personal finance decisions. SWR is the decision you make at the end of the accumulation phase, but asset allocation is something you decide before, after, and during the accumulation phase. Ditto for risk tolerance.

Now you might say that many here on the board will not allow a discussion outside of the 'efficient frontier' or some such nonsense, but why is the discussion limited by their involvement? Isn't it enhanced by their involvement just the way stock allocation discussions are enhanced when you effectively argue that real estate has a place in every investor's portfolio?

FMO, if you accumulate a portfolio of properties that will give you $X is annual income along with a stock and bond portfolio that (according to the 4% rule) will give you $Y in annual income, won't you retire when X+Y is high enough? What can a SWR study do for that decision?

Actually, here is a better question:
If the outcome of the SWR indicates that you are better off without real estate (a hypothetical), would you abandon it? No, because you enjoy it.
If sure-fire studies indicated that buying was better than renting would Intercst buy? No, because he doesn't want to.
If I could get a guarenteed market-beating return by purchasing a condo and renting it out would I do so? No, because real estate scares me. I don't want the commitment, I don't like working with my hands, and I would have a lower 'sleeping threshold' carrying real estate than I would buying stocks on margin. It's just the way I am.

I agree with your point on asset allocations, but I still don't grasp what the point is in SWR calculations?

Perhaps we should educate newbies in the variety of acceptable investment opportunites out there and tell people to invest in what they love...

st
Print the post Back To Top
No. of Recommendations: 4
That is a critical factor that hocus has wanted to consider...without being told that anybody who makes such a change is (insert bad word here).

Yes, what I want to do is to return to talking about the subject matter of the board, instead of all the junk that goes with a never-ending smear campaign.

Imagine the possibilities.
Print the post Back To Top
No. of Recommendations: 5
You [referring to SloanT, who asked whether there is enough to be learned from knowing the correct safe withdrawal rate (SWR) to make it worth the trouble we are going through] are asking the right question, st, and I whish I could rec it.

It is an excellent practical question. The sort of people asking this question are the sort of people that I am trying to have a conversation with.

I have been mocked many times on this board for the fact that I have compiled 40 binders of information on the subject matter of this board. I collected that material over a period of 12 years. I think it is fair to say that I know a lot more about the subject matter than the current board general.

There are others here who visit occasionally or who once posted regularly who also know a lot more about it than he does. Look at the posts that JWR1945 put up last night and I hope it is clear to you that he knows a lot more. Look at Wanderer's old posts and Raddr's old posts and you will see that they know a lot more too. These four posters (including me) have different views on SWRs, but I think you will find that all four agree that the SWR concept is a valuable one. Perhaps Wanderer not so much, I am not sure. But the other three will certainly tell you that it is valuable, and I don't think that Wanderer will entirely disagree.

This is what it means to be a leader, RonBass. A leader is not someone who creates deliberate provocations to send a board into flames just for the smirks he gets out of it. A leader is someone who has studied the subject matter enough to be able to say to the community, "this is worth knowing, this is the direction in which we should channel our energies."

I have a long history of informed and honest posting at this board, and I am telling you that there are few concepts as important in a long-term sense to the growth of the Retire Early movement than the SWR concept. It is a powerful tool. It is hard to see that from your perspective here because in the discussions here the tool has been twisted to make things that cannot possibly be true sound plausible. You are looking at a knarled hammer, and it doesn't look like much.

But use the tool for the purpose for which it was intended, and it will work miracles for you. No, you can't safely take a 4 percent withdrawal from a 74 percent S&P stock portfolio. That's a joke. But wouldn't you like to know what sort of stock allocation will permit you that withdrawal rate?

Does taking the stock allocation down to 60 percent do the trick? 50 percent? What?

SWR analysis tells you the answer. It does not tell you perfectly. It does not predict the future. What it does is tell you what the answer was in earlier historical periods. It is the best way that I know of to quantify the analysis you need to do in setting a portfolio allocation. The only alternative I know of is guesswork, and that is not so satisfying.

One of the deceptions that intercst has employed to con the board on this question is the idea that, if 4 percent does not work for his extreme allocation, it doesn't work for any allocation. This is a pure 100 percent garbage assertion, and anyone the least bit familiar with the SWR concept knows it. Just because you cannot take 4 percent from a 74 percent stock allocation does not mean that you cannot take it from other allocations.

Wouldn't you like to know what the historical data says on this question of how far down you need to go in your stock allocation to get to the withdrawal percentage you seek? In my view, it's information worth knowing to anyone even half serious about early retirement.

It is information that each of us on this board could discover for ourselves if we were willing to ask the Disruptors to knock off the funny business and allow us to use this board for the purpose for which it was created. They agreed to post in good faith when they signed up at this place, so all we are asking them to do is something that they have already promised to do. I think we should ask. That's the hard part. The good stuff comes when we start learning again and leave all this other nonsense behind us.
Print the post Back To Top
No. of Recommendations: 4
Because the maximum SWR is calculated based on demonstrated past failures, it can never increase. In the fullness of time, as additional data comes in, it is possible for the SWR to go down, but it will never, ever go up.

This is precisely why intercst does not want us to discuss this issue. Each time we learn something new, the number for his preferred allocation goes down.

This does not mean that the number for other allocations always goes down. There are many asset classes outside the two examined in the intercst study and many mixes of asset classes that provide SWRs higher than the 74/26 mix intercst refers to as "optimal."

It's optimal only so long as you refuse to look at the data. When you look at the data, you see that it is far from any sane definition of the word "optimal" in most circumstances.

Intercst uses the SWR concept as a club to win little debating points whenever his preferred investment approach is questioned. Those wanting to learn about the subject matter use the SWR as a learning tool. Use it that way, and its value increases by leaps and bounds.

Once you free yourself from one specific evaluation technique (historical sequences), two asset classes (large cap equities and bonds) and one management approach (annual rebalancing) who knows what may be possible?

That's exactly right, FoolMeOnce. Those are wonderful words.

It is not 1999 anymore. We can no longer afford to live in this fantasy world where the way to invest for early retirement is all stocks, all the time. It doesn't work. We need to calculate SWRs for a wide range of asset classes, a wide range of allocation percentages, and a wide range of financial circumstances. Then the tool pays off.

The way that SWR analysis is used today on this board, it is more trouble than it is worth. If the community is not willing to discuss SWRs honestly, we would be better to all just reach a consensus that the issue be taken permently off the table. I have proposed this before. I know the power of this tool, and I think it would be a shame for us not to exploit that power. But it is better not to discuss it at all than to continue to discuss it in the deceptive way in which we have discussed it up until now.

Will the effort have been worth it? I think so. Maybe the effort has a low probability of success but how will you know if you don't try? If you try you may fail. If you don't try you are guaranteed to fail. There are relatively few folks seriously looking at these issues (most of which enjoy it) and yet there are thousands that could potentially benefit.

Again, wise words. The great benefit to the board from talking about SWRs in a serious way is that it would give us something to talk about that relates to the subject matter of the board. We are all here to learn about early retirement, so that subject matter pulls us together. We do not all agree on politics, so discussions of that subject tend to pull us apart.

I have said it before and I will say it again. We need to open up some space for informed and honest posting on the subject matter of the board if the board is to survive (except in some new form as a sort of Political Asylum II). If there are some who don't like the idea of us having on-topic discussions, they can do what we are told to do when we complain of the off-topic stuff. They can put all those who like to discuss on-topic stuff in their p-box and return to their discussions of the latest exploits of Joe Millionaire.
Print the post Back To Top
No. of Recommendations: 3
SloanT

I have a question. This is a serious question, and although no one has to answer please don't think that I am being argumentative or anything but sincere. Here goes: At the end of the day, when all of these various methods of estimating fSWR have been examined, do you think that the time and effort spent will have been worth it? [Emphasis added.]

I am taking you at your word. I am answering you in the narrowest sense possible...for that is how you are looking at fSWR analysis.

For example, if the various ways of estimating came up with 3.5-4.5% as a probable range, what would I have gained from undertaking the effort?

You would know whether to make contingency plans. You would know how much of a shortfall that you might have to cover.

Do you think you will find something meaningful enough to justify the time spent?

Again answering in the narrowest sense possible, you would know enough of what you might have to do in the future so as to be able to sleep now.

From a broader perspective there is a wealth of useful information that comes from studying safe withdrawal rates and developing tools and theories based on safe withdrawal rate analysis. We know, for example, that selling large amounts of stocks during market downturns is what causes the failure of retirement portfolios. Volatility is the killer. Now look at an alternative investment with the same long-term return as the S&P 500 but with less volatility. Its safe withdrawal rate (i.e., same degree of safety) is higher. You could retire earlier and/or you could retire safer.

There are many asset classes with long-term returns and volatilities different from the S&P 500. Stocks with higher dividends often have lower long-term returns and lower volatility. A dividend strategy has a different safe withdrawal rate than an S&P 500 strategy. Assuming that you take reasonable care (e.g., avoid chasing dividend yield by itself), you might be able to retire earlier and/or safer. Combining asset classes may also allow you to retire earlier and/or safer.

The 4% safe withdrawal rate number applies only to a straight S&P 500 strategy. Other strategies have different safe withdrawal rates. Those rates are worth knowing. They will tell you how much earlier and/or how much safer you can retire.

There are many other considerations. In general, you should analyze your investment approaches differently depending upon what stage you are in (accumulation, maintenance or distribution) and how much time is involved (e.g., 5 years before retirement or 20 years to go). Understanding safe withdrawal rate theory and developing tools assists you in developing your own personal strategy.

Have fun.

John R.






Print the post Back To Top
No. of Recommendations: 4
A historical sequence analysis does use historical data. The others use the historical record as well. It is just that they extract a different set of information.

These words from a JWR1945 post put up last night are so important that I want to highlight them for the benefit of any Information Seekers trying to make sense of this mess.

A study that looks at one type of historical data and ignores all others does not provide the final answer to a question. At best it provides a tentative guess. When other data shows that as a matter of mathematical certitude that the tentative guess was wrong, you abandon it and begin the effort of coming up with something more in accord with reality.

That is what people trying to learn something about a question they are studying do, amyway.

The following words from that same post are also more important than the last 100,000 words posted to this board that preceded them.

The historical sequence approach is always definitive when it shows failure. It is not definitive when it shows success. That is where the other methods help out.

My hope in winning some of these all-important polls in the future is that there are people out there in the real world who possess a basic understanding of the subject matter of this board. Don't be fooled because you don't see many of them posting to this board today. They have their reasons. But I am going to try to get some people with some basic understanding of the issues to visit here in days to come, and then we will take some new polls and we will see how the Wheel of Fortune turns then.
Print the post Back To Top
No. of Recommendations: 5
I agree with your point [referring to FoolMeOnce] on asset allocations, but I still don't grasp what the point is in SWR calculations?

You are asking an important question, SloanT.

SWR analysis does not tell you what to do. It is important to accept that as a starting point.

If I did a SWR analysis and it showed that real estate has the highest SWR, I would not invest in real estate just for the reasons you cite. I don't have any background in real estate, and I would not feel comfortable moving to that asset class in any significant way without developing some.

But that doesn't mean that SWR analysis does not have value for me. It helps me set the allocations between the asset classes with which I do feel comfortable. Are you better going with 50 percent stocks or 60 percent or 40 percent? SWR analysis, properly done, provides insights based on what happened in the past.

You might even decide if the SWR for real estate is high enough that it's worth going to the trouble to learn a little about that investment class. Perhaps SWR analysis leads you to put a foot in the water, and then you get to like it and years later you thank your lucky stars for that SWR analysis that got you started.

Another thing that a valid SWR analysis can do is get you to a safe early retirement a lot sooner. If all you know is the intercst approach, you would think that the SWR for 74 percent stocks is always 4 percent. Does that make sense to you? Do you believe that someone who retired with a 74 percent stock allocation in 1982 had the same SWR as someone who did so in 2000? The idea defies common sense.

When your common sense tells you that something doesn't add up, usually the reason is because it doesn't. Of course the SWR for stocks was not the same in 1982 as it was in 2000. It was either higher than 4 percent in 1982, or lower than 4 percent in 2000, or, most likely of all, both.

I don't get any pleasure in being the one to tell people that the SWR for 74 percent stocks was nowhere close to 4 percent in the year 2000. But a valid SWR analysis sometimes works the other way too. In a few years, we might be back to valuation levels where you can get a 5 percent SWR for stocks. If you have not yet started your retirement, that SWR might be available to you in your future. Don't you want to know that?

Knowing that the SWR is 5 percent rather than the 4 percent number that intercst throws around means that you can safely begin your retirement years earlier. There is never a requirement that you base your plan on the insights provided by a valid SWR analysis. But I cannot imagine any scenario in which you are better off not having those insights available to you to follow or reject. Knowing what the historical data says is always a positive, never a negative.
Print the post Back To Top
No. of Recommendations: 6
I think I would rather retire at 4%, hope the future is no worse than the past, and enjoy myself. And if it turns out that the future is worse than the past, then I will go on with life, living worse than before, along with everyone else, as the future will be worse than the past.

I'm with you. What are the chances that the future will be *as bad* or worse than the WORST of the past? Then, what risks do we take today that are greater than that? Some of those are even controllable...diet, (lack of) exercize, seat belt usage. Why bite nails over the 0.1% risks while ignoring the 1% and 10% risks?

If someone is *paranoid* that the worse-than-past-worst-case scenario is on the horizon, he can buy an annuity at 4.5% to 6%. He/she can even get one with inflation adjustment, one that continues to pay a reduced amount to surviving spouse, etc. One downside is that when reality is not so bad, the annuity passes no money to any heirs or charity, as the remainder of the stash from which the 4% is withdrawn would.

If the worse-than-past-worst-case scenario includes an asteroid destroying the world's economies or even mankind, then the withdrawal rate doesn't really matter anyway.
Print the post Back To Top
No. of Recommendations: 2
JAFO31

Ditto regarding assumption of independence based upon calendar year.

Just my $0.02. Regards, JAFO


I do not make that assumption.

Have fun.

John R.
Print the post Back To Top
No. of Recommendations: 7
Yes, what I want to do is to return to talking about the subject matter of the board, instead of all the junk that goes with a never-ending smear campaign.

You're doing it right now and don't even realize it. You never will "get it."
Print the post Back To Top
No. of Recommendations: 10
This is precisely why intercst does not want us to discuss this issue. Each time we learn something new, the number for his preferred allocation goes down...

Intercst uses the SWR concept as a club to win little debating points whenever his preferred investment approach is questioned. Those wanting to learn about the subject matter use the SWR as a learning tool. Use it that way, and its value increases by leaps and bounds.


I told you so! I told you so! I told you so! I told you so!

See, the discussion just got started and already Hocus is distracted and debating how to have the debate.

Guess we can safely put A.D.D. on the list as well.

neener-neener-neener

Golfwaymore
Print the post Back To Top
No. of Recommendations: 25
he sort of people asking this question are the sort of people that I am trying to have a conversation with.

You are having a conversation with them. Two fine posters have stepped forward to try to mediate your ridiculous rantings. Yet instead of expressing gratitude for an end to your siege, you keep blubbering about how you're TRYING to have a conversation, you're trying, trying, TRYING to express your views. If you HAD a view, this would be the place to express it but, so far, all you've expressed is that you can't express your views. What a farce.
Print the post Back To Top
No. of Recommendations: 13
hocus hallucinates,

One of the deceptions that intercst has employed to con the board on this question is the idea that, if 4 percent does not work for his extreme allocation, it doesn't work for any allocation. This is a pure 100 percent garbage assertion, and anyone the least bit familiar with the SWR concept knows it. Just because you cannot take 4 percent from a 74 percent stock allocation does not mean that you cannot take it from other allocations.


You can, in fact, go to www.retireearlyhomepage.com, download a free Excel spreadsheet, and calculate the historical safe withdrawal rate for any asset allocation you like. I know you've never done that, yet you continue to charge me and others with deception when we call attention to your questionable claims.

Here's a recent poll we did on the question which is listed in the FAQ.

http://boards.fool.com/Message.asp?mid=18771318

It seems the vast majority of the board, many of who have much more knowledge of math and arithmetic than you do, disagree.

intercst
Print the post Back To Top
No. of Recommendations: 2
FMO, if you accumulate a portfolio of properties that will give you $X is annual income along with a stock and bond portfolio that (according to the 4% rule) will give you $Y in annual income, won't you retire when X+Y is high enough? What can a SWR study do for that decision?

What an SWR study can do in this case is use historical data to determine what might happen if the future is similar to, or at least no worse than, the past. See, lets say you retire when X+Y hits your target, and then a few months later, lets also say that most of your properties are in a hi-tech region, then suddenly all sorts of bad things happen, like a tach-bust, corporate malfeasance, massive tech layoffs, and declining tech wages. After a year or two, the demand in your area declines somewhat and rents decline by 10% (things like 1 free month for signing a year contract), meanwhile your state is hurting for money and raises property taxes, now X is 0.88*X. Lets say that at the same time, the market dramatically declines by 30-40% and you aren't comfortable anymore with taking 4% from your portfolio, so Y declines as well. Now what do you do ?

A good hSWR study can tell you what happened in the past (to X and Y)under those circumstances, if you believe that the future will be similar, and not worse, then you can make certain decisions based on that information.
Print the post Back To Top
No. of Recommendations: 4
<If you HAD a view, this would be the place to express it but, so far, all you've expressed is that you can't express your views. What a farce.>


Catherine:

I can remeber one of your very early posts on the board when you asked: "Who is Hocus?" I guess that question has been answered clearly enough for you by now? <grin>


BRG

Print the post Back To Top
No. of Recommendations: 2
John -

Thanks for all your excellent responses.

I do not consider deterministic approaches, Monte Carlo simulations and historical sequence analysis (the intercst method) to be mutually exclusive.

Thanks for clarifying that. I agree.

I have no desire to discuss stock market timing here. I just have recently started two threads on this subject on the http://nofeeboards.com Index Funds discussion board. They are Market Timing References and What the researchers found out.

I respect your choice. I will inquire there.

It is not so much a matter of questioning any answer per se. It is more a matter of adding a whole lot more information to get a better picture...to fill in the mosaic. That is what you get from doing sensitivity studies and using additional analysis approaches.

Agreed.

I will see if I can free up any time to help. Thanks again.

dan
Print the post Back To Top
No. of Recommendations: 1
I would like to see someone come up with a model which shows 99.9% certainty that they are gonna be around in 30 years.
Print the post Back To Top
No. of Recommendations: 3
At the end of the day, when all of these various methods of estimating fSWR have been examined, do you think that the time and effort spent will have been worth it?

It is if you think it's fun. Which I do. That's one reason I'm here.

phantomdiver
Print the post Back To Top
No. of Recommendations: 0
"I would like to see someone come up with a model which shows 99.9% certainty that they are gonna be around in 30 years."

I feel pretty certain that I won't be around in thirty years. I recently had my fiftieth birthday. Most of my family has the good sense to kick off in their mid seventies, although my mother only made it to 51, and my dad to 65. - Art
Print the post Back To Top
No. of Recommendations: 3
I raise this question in ernest. My father lived to 85, my mother 77.
I am retired for about a year now, age 56. But I have had my first angioplasty already. Diet, excercise, statins? Sure, but I don't see me living 40 years.

Funny how we ponder this stuff. Like I think about buying an annuity and letting some insurance company figure it out. Nah,,,, buying an annuity for life when interest rates are at 41 year lows is, well, duh!!!!

Gee, if I wait a year, rates may well be better, so I get more income for the buck. Also, my estimated age on the actuarial table goes down a hair, so there's a little more money right there. But mostly, if I die tommorrow, my family gets the money, not some insurance guy.

This is a complicated problem we are dealing with, but we should all remember, it's a nice problem to have. When I look at the savings rates in this country, I think it's a problem a lot of folks aren't going to have... and that's a problem for all of us.
Print the post Back To Top
No. of Recommendations: 0
I would like to see someone come up with a model which shows 99.9% certainty that they are gonna be around in 30 years.

I don't know how one would define THAT high a chance...

But in DW's family back at least three generations, women either die in accidents, die of environmental diseases (lung cancer for smokers, for example), die of uterine cancer, or live past age 85. DW has had a hysterectomy and has no notably carcinogenic habits.

Since DW won't be 55 when we retire, the 30-year outlook is a serious consideration.

There is another factor. Medical research continues to advance life expectancy - I think I read that it has gone up two years in the last ten years. In the last 5 years some things have happened in medical and biological research that haven't YET had a notable impact on life expectancy but plausibly COULD do so within a couple years. Sometime in the current century will be a 10-year period where the life expectancy at birth in the industrialized world increases by not less than 30 years. (Although this change may not be recognized until after the fact.)
Print the post Back To Top