Financial Statistics

Research on Safe Withdrawal Rates

Moderator: hocus2004

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

Financial Statistics

Post by JWR1945 »

I am continually amazed when I look at the statistics associated with economics. There seems to be a total indifference to the power (or sensitivity) of statistical tests.

It is very easy, sometimes trivial, to apply a test that fails to detect a statistical difference. It is quite a bit different to show that any difference that exists is very small or else it would have been detected (with a high level of confidence).

I attribute this to an overreaction to the exaggerated claims of some, especially salesmen, and the intimidating atmosphere faced by academics. For example, I doubt seriously that either Professor Shiller or Professor Campbell really believes that their findings (about P/E10) need as many qualifiers as they provide. It is just that they are conditioned to expect a hostile response whenever they report finding something of value.

A more balanced view would be to acknowledge the existence of routine errors while allowing a greater tolerance when findings are solidly supported by common sense. It is one thing to find obscure relationships associated with great returns in the past and to project their applicability into the future. It is very much another thing to look at something such as P/E10 and conclude that valuations matter.

Whenever we look at data, we are aware of what we cannot do as a practical matter. This much is not limited to economic statistics. But with economic results, we are forced to accept history as it is. We cannot run controlled experiments (with rare exceptions). Very often, we do not know what to look for until we have looked. That puts us in a difficult position since it is best to ask all questions before examining the data. Not having asked the question beforehand, we could easily prevent ourselves from reporting the obvious. There are ways of handling such situations, none of them ideal, but which are sufficiently objective so as to allow us to reach reasonable conclusions. We should not always restrict ourselves to extremely defensive arguments.

I have recently reread The Superinvestors of Graham-and-Doddsville, an edited transcript of Warren Buffett's 1984 talk at Columbia University. Buffett presented overwhelming evidence that there really are people skilled in selecting stocks, that their high returns are not simply a side effect of randomness and that prices really do matter. His evidence is sufficient to satisfy even the most demanding honest statistician and it is audited. Yet, I continue to read assertions that skill does not exist because some insensitive statistical test fails to show significance. Such concepts as efficient markets and random walks are helpful when they are used properly: as first approximations. To insist that they are relevant at all times is folly.

Have fun.

John R.
User avatar
BenSolar
*** Veteran
Posts: 242
Joined: Mon Nov 25, 2002 5:46 am
Location: Western NC

Post by BenSolar »

JWR1945 wrote:Yet, I continue to read assertions that skill does not exist because some insensitive statistical test fails to show significance.
I've rarely or never seen someone say that they thought Buffett's performance was due to luck. The common argument is that picking the next Buffett is very difficult to the point of approaching impossible, until we have the advantage of years and years of outperformance to show us who was the most skilled of the current generation. At which time, of course his fund is awash with money and must come to approach the index.

But, I agree with you that being open to the idea of exploitable long term market inefficiencies is a good thing for the thinking investor to do. I am an admittedly somewhat active investor. Of course building a diversified portfolio of different asset classes and going on autopilot is a fine approach, too, in my opinion. A 100% S&P 500 approach would certainly be ill-advised at this time, IMO.

B.
"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
User avatar
Alec
Admin Board Member
Posts: 31
Joined: Wed Sep 10, 2003 4:00 am
Location: Crofton, MD

buffet, value, and such

Post by Alec »

I think this is the speech John is referring to:

The Superinvestors of Graham-and-Doddsville

I just had a couple of thoughts/questions. First, apparantely the returns are not what investors actually received. There are no adjustments for taxes or commissions. Though, I would imagine that this wouldn't cause very many of these guys to fall below the S&P return, especially that even if an S&P index were available in the 40's, 50's, and 60's, it would surely have had a load attached.

Second, Buffet seems to be saying that value investing is a free lunch.
I would like to say one important thing about risk and reward. Sometimes risk and reward are correlated in a positive fashion. If someone were to say to me, "I have here a six-shooter and I have slipped one cartridge into it. Why don't you just spin it and pull it once? If you survive, I will give you $1 million." I would decline -- perhaps stating that $1 million is not enough. Then he might offer me $5 million to pull the trigger twice -- now that would be a positive correlation between risk and reward!

The exact opposite is true with value investing. If you buy a dollar bill for 60 cents, it's riskier than if you buy a dollar bill for 40 cents, but the expectation of reward is greater in the latter case. The greater the potential for reward in the value portfolio, the less risk there is.
I'm not sure I totally agree. If someone is willing to sell me something, that is worth $1 for only $0.60 or $0.40, I have to ask why? Is it b/c he is stupidly valuing it? Possibly. Or is it b/c this something is more risky to own (for whatever reasons) than something that is worth $1? Or is it a bit of both?

It's also interesting that all of these guys were/are value investors (duh). So, it is b/c of their shear genius (or knowledge they learned from Graham) that they outperformed the market? Or is it b/c they simply owned value stocks? That's all that the statistical regressions test. Comparing them to the S&P 500 is just bad benchmarking.

So, does this article actually help us out at all. Well, not really. Will there be people who do better than the market? Absolutely. So, who are they? And will anyone give me a money back guarantee (not likely) if they choose the wrong person for me? People can show me as many patterns as they like, and as many "reasons" why these patterns will continue. Unfortunately, we will never, ever, be able to tell with certainty whether or not these are actual patterns or noise?

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

Post by JWR1945 »

BenSolar wrote:I've rarely or never seen someone say that they thought Buffett's performance was due to luck. The common argument is that picking the next Buffett is very difficult to the point of approaching impossible
Actually, what Warren Buffett did was to demonstrate that there already were many Warren Buffetts out there. He had pre-identified several outstanding investors 15 or more years before he made his speech and presented his data. He had selected them so as to be able to meet the most demanding standards of academics and statisticians. There is no doubt that he satisfied all of the demands of statistics. However, he did not present his results in the form of reporting statistics.

His outstanding investors had very few things in common. They bought different companies. Some diversified to an extreme. Others concentrated their portfolios. What they did share in common was having taken an investment course by Benjamin Graham and they all got the idea of value investing.

Quoting Buffett:
One sidelight here: it is extraordinary to me that the idea of buying dollar bills for 40 cents takes immediately with people or it doesn't take at all. It's like inoculation. If it doesn't grab a person right away, I find that you can talk to him for years and show him records and it doesn't make any difference....It doesn't seem to be a matter of IQ or academic training. It's instant recognition or it is nothing.
Another relevant quote:
...Adding many converts to the value approach will perforce narrow the spreads between price and value. I can only tell you that the secret has been out for 50 years, ever since Ben Graham and Dave Dodd wrote Security Analysis, yet I have seen no trend toward value investing in the 35 years I've practiced it.
Yes, Alec, it is a free lunch. But only a few are willing to take advantage of it.

IMHO, Buffett was not trying to tell you how to select a fund manager. He was telling you how to go about investing on your own.

Alec provided the right internet link. My own reference is an Appendix in The Intelligent Investor by Benjamin Graham (copyright 1973) as printed in 1985.

Have fun.

John R.
RobBennett
* Rookie
Posts: 7
Joined: Thu May 06, 2004 12:46 pm

Post by RobBennett »

Yes, Alec, it is a free lunch. But only a few are willing to take advantage of it.

My take on this is that the thought processes that go into the formation of investing strategies almost always involve an emotional component. Investors often purport to be using data to form their strategies. In many cases, however, the data analysis is more a rationalization of a decision made for other emotion-oriented reasons. The data is not being used to determine the investment strategy, but to justify a strategy decided prior to examination of the data.

I do not mean to pick on Alec by using the argument he has put forward as an example of the phenomenon I am getting at, but I think his argument above provides a good illustration of it. Alec notes that Buffett seems to be offering a "free lunch" to investors willing to take valuation into account. He is quite right about that; taking valuation into account does indeed provide enhanced return without requiring that one take on added risk, according to the historical data.

The suggestion being made by Alec is that there is something suspicious about this claim of a free lunch; it can't really by that easy to obtain an enhanced return from your investing dollar, can it? But advocates of indexing argue all the time in favor of a different sort of free lunch, the free lunch that comes with indexing (you presumably obtain an acceptable return without having to pay fees to fund managers or to devote the effort to select promising individual stocks or whatever). I don't know whether <b>Alec</b> favors indexing or not, but, presuming he does, I question whether he would be as suspicious of the free lunch claimed for indexing as he is of the free lunch claimed for value investing.

I am not here trying`to say that indexing is good or bad, or to say that Alec is right or wrong to be suspicious of Buffett's promise of a free lunch. What I am trying to suggest is that emotions almost always play a role in the formulation of investing strategies. I believe that the reason is that our success with investing determines what success we achieve with pursuit of our life goals. We all all emotional about pursuit of our life goals, so it is all but impossible not to get emotional about investing issues.

What makes it tricky is that so many investing arguments are framed in non-emotional terms. People cite numbers and charts and data, and give the impression of being objective about their claims. They often genuinely are not able to see that the same arguments that they use against investing strategies they do not favor apply just as strongly to those they do favor.

My ultimate goal with the data-based SWR tool is to make the tool a customized tool, one that personalizes the investing decision for each investor making use of it. I hope though that personalization effort to find a new focus for the emotional energy arising in the investment decision-making process. My hope is to provide a tool that allows people to be up front about the emotions that come up when making investment decisions and thereby to channel those emotions into more constructive directions.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

..emotions almost always play a role in the formulation of investing strategies. I believe that the reason is that our success with investing determines what success we achieve with pursuit of our life goals. We are all emotional about pursuit of our life goals, so it is all but impossible not to get emotional about investing issues.

What makes it tricky is that so many investing arguments are framed in non-emotional terms. People cite numbers and charts and data, and give the impression of being objective about their claims. They often genuinely are not able to see that the same arguments that they use against investing strategies they do not favor apply just as strongly to those they do favor.
Emphasis added.

IMHO, this is a major insight. It think that it is true. It would explain much of what I have seen in the last couple of years.

Have fun.

John R.
User avatar
BenSolar
*** Veteran
Posts: 242
Joined: Mon Nov 25, 2002 5:46 am
Location: Western NC

Post by BenSolar »

Alec wrote:I Comparing them to the S&P 500 is just bad benchmarking.
This is certainly correct. I suspect that not only would they need to be compared to a value index, but to a small-mid cap value index. I can guarantee the results of such a comparison won't be so striking as when compared to the S&P 500.

Some (most?) of them would still show outperformance. :)
"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
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

To BenSolar and Alec:

I believe that you are engaging in a logical inconsistency common among those who slice and dice their allocations. You are comparing predictions from before the fact with market behavior that occurred after the fact. Slice and dice comparisons are quite useful in terms of explaining what has happened. They are less useful in predicting what will happen. This point is acknowledged, in fact, when you read about recency and how it causes Mean Variance Optimizers to err.

I think that there is merit in the slice and dice methodology. I think that it has a degree of predictive power in addition to its historical explanatory power. It has only limited value for advancing an argument that skilled investors do not exist or that they cannot be identified in a reasonable amount of time.

Have fun.

John R.
User avatar
BenSolar
*** Veteran
Posts: 242
Joined: Mon Nov 25, 2002 5:46 am
Location: Western NC

Post by BenSolar »

JWR1945 wrote:I believe that you are engaging in a logical inconsistency common among those who slice and dice their allocations. You are comparing predictions from before the fact with market behavior that occurred after the fact. Slice and dice comparisons are quite useful in terms of explaining what has happened. They are less useful in predicting what will happen.
I dont' see the logical inconsistency in saying: these investors are all trained in and practice value investing. They tend to look through mid and lower cap stocks for their picks because that is where the bargains are. Their investing style is small-mid cap/value. Fama and French style 3 factor analysis predicts that value will outperform, and it predicts that small will outperform. An index investor can capture this outperformance by investing in a small cap value index. If we look at the superinvestors results without comparing them to an index with similar Fama/French characteristics, then it is possible that any outperformance we see over the general market is only that outperformance predicted by Fama and French.

Maybe the period they excelled in was just a lucky one for small/value investing above and beyond the Fama/French predictions. In an alternate universe there might have been a small growth guru who trained a legion of analysts to invest in small growth and they went forth and rode an unprecedented wave of performance in small growth over a couple of decades. Are we to conclude skill in this case? What if the small growth indexes did just as well? Do we still consider it skill?

I maintain that until you compare their results to an index with comparable size and growth factors, you can not say it was skill. And I'm still certain that some of those investors spanked the relevant index and one would be hard pressed to claim they don't have skill. But the ones who outperformed the S&P by a few percentage a year, well that could be explained by the value and size factors.

Regards,
"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
RobBennett
* Rookie
Posts: 7
Joined: Thu May 06, 2004 12:46 pm

Post by RobBennett »

Here's a link to a thread now going on at the Berkshire Hathaway board at the Motley Fool site.

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

There are a number of good points made in this thread re the psychological stresses that must be endured to succeed at value investing. I thought that these two were particularly relevant to consideration of the SWR matter.

Juicy Quote #1: Value investing almost requires periods of underperformance....With a value investor, if the market is going against him, if his positions are dead money; they will be dead today, they will be dead tomorrow morning; they will be dead next week, next month. Next year? The outperformance that the value investor believes will surely be his often comes with a lot of honest suffering in the short term. And who wants that?

Juicy Quote #2:The key to me is how to confirm that the strategy is working. The bottom line is long-term actual returns, but you can't really use that as a regular confirmation. Instead, I have to step back one (big) step and take on faith that in the long term the market is efficient, i.e. that market price eventually meets intrinsic value, so what I need to do is keep tabs on intrinsic values.

I got out of stocks in 1996. My recollection is that the DOW was at about 5600 on the day I got out. I believe that this was a few months before Greenspan's Irrational Exhberance speech. The DOW skyrocketed in 1997, 1998, and 1999. So I had three years when my data-based SWR theory was put to a pretty darn difficult real-world test.

I had a friend who used to mock me in a good-natured way from time to time for my investing approach. There were times when I experienced doubts about whether I had done the right thing. What I would do in those moments is turn to my 40 black binders of Passion Saving information and insights, and look again at the data that had caused me to get out of stocks in the first place. The data always said the same thing, that I had done the right thing given my investing goals (not necessarily the right thing for investors with other sorts of goals). So I stuck it out.

Today I see the benefits. My retirement began on August 1, 2000. I have little slack in my plan. Had I followed the conventional SWR studies and taken a 4 percent withdrawal on a high stock allocation portfolio, I would be back in the work force today. So I think of the data-based SWR as The Number That Saved My Life. That doesn't mean that I didn't feel emotional strain sticking with the program during the times when it didn't appear in the short term to be such a hot idea.

One of the major benefits of SWR analysis is that it provides the investor using it with the emotional fortitude he needs to stick with a strategy during the time-periods in which the strategy is not providing a current-day payoff. The One Who May Not Be Named (TOWMNBN) is very much aware of this (to his credit). He has many times made the point that it is an unusual investor for whom his SWR calculations will actually apply; he describes the type of investors for whom his approach will work as "MasterMinds" and says that they make up about 10 percent of the population. I don't agree with everything that TOWMNBN says on this point, but I believe that the core point is a very important one. For the SWR tool to work, you must be willing to stick with the allocation you elect through thick and thin.

If you change allocation percentages when stocks go down in price, the math behind the SWR numbers doesn't apply. It is absolutely essential that the investor making use of SWR analysis plan in his retirement to be a real long-term buy and hold investor. There have been a good number of comments made in recent months by defenders of the conventional methodology indicating that they do not see the importance of the SWR investor possessing this true buy-and-hold mentality.

There was a great thread at the Early Retirement Forum (the one that had 235 posts and over 5800 reads) that explored a host of interesting angles. One thing that shocked me in that thread was how a number of smart posters (people like Dory36, but by no means only him) made the argument that, in the event the 4 percent number does not work out in the real world, the investor can always make changes in his stock allocation to "adjust" for this reality. I view this as a very dangerous way to analyze the SWR question.

If investors make adjustments after stocks go down in price, they are going to lose lots of money. They are going to imperil their retirements. The whole idea of SWR analysis is to pick a take-out number so safe (100 percent safe presuming that the future is like the past) that the odds of the number not working are sufficiently low that the investor will not be tempted to change his allocation. To adopt a take-out number you do not have confidence in with the thought that you can always bail out of stocks if prices drop too low is madness, in my view. It is the worst of all the possible investing strategies that I can imagine. If you are thinking that there is a serious chance that you are going to feel a need to bail out, you are almost certainly not using a valid SWR analysis to come up with the take-out number and allocation percentage you are employing for your plan.

The SWR is supposed to be a safe number. That's why the first word in the SWR phrase is "safe." What is happening is that people are lookng at the number you get when you do the analysis in an analytically valid way, and saying "that number would require me to go lower with my stock allocation than I want to go, so I am going to adopt the assumption that for the first time in recorded history changes in valuation levels will have zero effect on long-term returns. If I do that, I can come up with a SWR at least two full percentage points higher than the one you get by looking at the historical data, and that permits me to go with the stock allocation I want to go with and still retire on schedule."

The logic works if you buy into the premise, but the premise is nuts. Changes in valuation affect long-term returns as a matter of "mathematical certainty" (William Bernstein's phrase), so it is unreasonable to stake all of the capital you have accumulated from a lifetime of work on the naked hope that for some magical reason this will not continue to be the case once your particular retirement begins. It's like putting everything down on a number on the roulette wheel that has only a small chance of turning up.

I believe that the quotes above suggest some of the emotions that are at play in this matter. Value investing is always hard because it always requires that the investor making use of it remain deliberately out of touch with what is going on in the market during long stretches of time. The valuation levels we have experienced in the recent bubble are the greatest ever seen in the U.S. markets, so value investing is harder today than it has ever been before. This is why it is so hard to get people to acknowledge what the historical data says about SWRs at this time-period, I believe.

Human beings have emotions. Emotions often trump logic. The logic for the data-based SWR is unassailable. But the numbers generated by the analysis are hard to accept emotionally. I believe that the solution to the disconnect is to develop alternate sources of emotional solace and thereby to overcome the emotional pull that many feel to invest too high a percentage of their savings in stocks at times of high valuations.

Why is it that I, a self-acknowledged numbers dunce, was able to come up with the concept of the data-based SWR? I believe that it is because of the intense emotional desire I felt for early retirement. I loved the feeling that came with the stock gains I experienced in the mid-90s. I wanted those feelings to continue. Like everyone else, I was shocked by the numbers that are generated by a data-based SWR analysis. For me, though, the emotional appeal of early retirement was so great that I ultimately was able to do what the numbers said regardless of the emotional distress that resulted from doing so.

I experienced emotional distress, but I also experienced emotional relief, the relief that came from knowing that my FIRE plan (or Passion Saving plan, or whatever you want to call it) was rooted in something solid. The numbers trumped the emotions in my case. Not because I am not subject to the same emotional pulls as all others. Because I had something else to get emotional about, early retirement, and the emotions I felt re early retirement were stronger.

On the surface, the Great SWR Debate is about a number. It seems strange that discussions of a number would cause so much emotion to display itself on the various boards. The explanation of the apparent riddle lies in understanding that the SWR number is a number that packs a powerful emotional punch. When we are talking about SWRs, we are talking about a number with great significance to people's hopes of achieving their most important life goals
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

BenSolar wrote:I don't see the logical inconsistency in saying: these investors are all trained in and practice value investing. They tend to look through mid and lower cap stocks for their picks because that is where the bargains are.
Actually, Benjamin Graham argued against investing in anything except large cap stocks. He pointed out that the lower prices of secondary issues persist in spite of their relative attractiveness. Large capitalization companies are much more likely to get through bad times than secondary stocks. That is what gives them a premium.

The logical inconsistency that I see is that the indexes that people select for comparison are put together after the fact to judge the merit of what others decided to do before the fact.

Read Buffett's speech again. His investors differed in almost all respects except for their emphasis on value.

Have fun.

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

Post by JWR1945 »

One of the major benefits of SWR analysis is that it provides the investor using it with the emotional fortitude he needs to stick with a strategy during the time-periods in which the strategy is not providing a current-day payoff.
This is one of the reasons that I have put so much emphasis on monitoring the health of one's portfolio during retirement.

There is always the possibility of an analytical error no matter how carefully a study is put together. I want to be able to identify such flaws early enough to make corrections and adjustments before very much damage is done. I don't like the idea of discovering that something might be wrong after two or three years of serious losses.

We now understand that the conditions that supported the original 4% withdrawal rate with a portfolio heavily invested in the S&P500 no longer exist: valuations are well outside of the historical range and dividend yields are minuscule. There is no comparable historical condition similar to this in the first decade of retirement. It is sad to think of those who depended on the original number and who have not taken advantage of the latest rally to shore up their portfolio's safety.

Have fun.

John R.
Post Reply