Valuation based SWR Question

Research on Safe Withdrawal Rates

Moderator: hocus2004

Post Reply
beachbumz
** Regular
Posts: 79
Joined: Mon Feb 07, 2005 4:00 am
Location: Cocoa Beach, FL

Valuation based SWR Question

Post by beachbumz »

If I understand this 'valuation' based SWR theory, then Hocus and perhaps others are waiting for lower valuations to enter the equities market. (If that's not correct, please let me know.)

What happens to the FSWR if there is a shift in 'normal' valuation? Let's say, pe's get down to 10, but that this is a permanent shift in values (of course, we could only know this years later, but we are trying to predict the FSWR right?). Or perhaps, today's valuations represent a permanent shift in values and that history will prove that a 18pe (or whatever it is) will ultimately turn out to be a 'low' valuation.

I am thinking about this from the same perspective as interest rates. In 1990, mortgage rates dropped below 10% and this was considered very low. In fact, I ran out and bought my first house. Now, 15 years later, people can't even imagine 10% mortgage rates. In fact, when the rates jumped to 7%, people were freakin' out. IMHO, there has been a long-term shift downward in interest rates. I think Paul Terhorst found this out some time back. Anybody remember the book "Wealth Without Risk" by Charles Givens (I think). His saying was "when rates are low, stocks will grow", his dividing line between high and low was something like 9% on the prime rate.

Just wondering...

Beachbumz 8)
Life is Good.
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

The idea you are putting forward reminds me of the argument set forth in the book "DOW 3600." I did not read the book, but I read a number of write-ups.

As I understand the argument, the author maintains that people are in the process of learning that stocks are less risky than they once thought. That discovery is going to push prices dramatically upward, according to the book. Those who get in before the big run-up in prices will do well. Those that get in afterwards will experience returns on their stock investments lower than what has been the historical norm (the risk premium will have been permanently reduced).

I don't personally think this is so, but I view it as a reasonable argument.

All that we have to go by today is the data that is available to us today. The SWR tells you what will work presuming that stocks perform in the future as they have in the past. What you are surmising is that stocks will perform in the future differently than they have behaved in the past.

It could happen. But it is obviously not possible for the historical data to reflect something that hasn't happened yet.

There's nothing wrong with the idea of keeping the scenario you put forward in mind as one possible scenario that may take place. But it is not a scenario that is examined in the typical SWR analysis. The typical SWR analysis assumes that the future will be like the past. It obviously does not assure that the future will turn out to be like the past.
unclemick
*** Veteran
Posts: 231
Joined: Sat Jun 12, 2004 4:00 am
Location: LA till Katrina, now MO

Post by unclemick »

I'm old enough to remember when 6-8% take out without touching principle (so it would grow to keep up with inflation) was considered conservative thinking.

Heck - scratch the little 2000-2003 knit in the market - I could creatively extend the long term to -SWR's where you never had a chance without spending principle.

0% plus 'govt calc inflation rate' for 100% TIPs portfolio and all other combo's being negative.

Lucky I'm an optimist like DeGaul was.
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

I was thinking this over a bit more before I drifted off to sleep last night, and I came up with another way to look at the question you are posing, BeachBumz.

What you are suggesting is that there is a good chance that a worst-case scenario will not pop up in your retirement. That is of course so. If a worst-case scenario (defined as the worst returns sequence we have seen in the historical record, but nothing worse than that) does not pop up, you will be OK with a take-out number higher than the SWR. That doesn't mean that The Data-Based SWR Tool does not have something of value to offer to you.

The 2.4 percent number that we refer to as the SWR that applies at today's valuation levels is the take-out number that is "95 percent safe" for an 80 percent S&P portfolio used to finance a retirement beginning today. There is no law that says that you need to seek 95 percent safety in your plan. If you ask, I believe that JWR1945 can tell you what the number is that is associated with any other level of safety you are interested in knowing about.

The number that JWR1945 calls the "Calculated Rate" is the take-out number that is "50 percent safe." Using the same methodology that he uses to determine the 95 percent safe number and the 50 percent safe number, he can also determine the 90 percent safe number or the 80 percent safe number or the 70 percent number.

This is why I often make the point that The Data-Based SWR Tool is a descriptive tool rather than a prescriptive tool. It does NOT tell you how to invest. It just provides you with insights you can use to inform your investing decisions. It takes the historical data and wraps it into one nice little package of a number so that you can perform useful comparisons without having to spend hours and hours rooting through all the data yourself.

We of course do not expect stocks to perform in the future exactly as they have in the past. There are always going to be things that change as time marches on. The idea behind SWR analysis is that you need some sort of guide to form a reasonable expectation as to how your investments may perform and looking at how they have performed historically gives the best guidance available.

The wise aspiring early retiree does not mechanically take the SWR number and make that the take-out number in his plan, no further questions asked. He determines what the SWR is, and then he also takes into consideration his thoughts as to how stocks may perform differently in the future, his particular life goals, his particular financial circumstances, and all sorts of other things in determining what to invest in, what take-out number to use, and so on.

Knowing the SWR does not mean that you have no further work to do in developing your Retire Early plan. Still, I have been making use of this tool for nine years now and I have found it to be the most powerful tool for developing investing strategies that I have ever come across. It is not a one-stop solution. But it is indeed a tool of great value for those who put it to use for the purposes for which it was intended.
unclemick
*** Veteran
Posts: 231
Joined: Sat Jun 12, 2004 4:00 am
Location: LA till Katrina, now MO

Post by unclemick »

Being left handed - I would turn it around to say a methodogy of analysis techniques to provide insight into possible retirement investing options.

SWR tool may be convient for posting efficiently - but in my view it is way too limiting to convey the scope of the methodogy/analytical approaches being developed. Implies single point and static. Whereas - as pointed out - it is evolving and provides insight to able to consider a variety of strategies.

SWR tool ???? - and you wonder why so much razzing about 100% TIPs.

Heh, heh, heh, heh.

SWR Research - you got it right that time.
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

it is evolving and provides insight to be able to consider a variety of strategies.
I very much agree with what you are saying here, UncleMick.

At the beginning, The Data-Based SWR Tool was the combination of two analytical approaches: (1) the idea of identifying the SWR, which at the time I developed the new tool was being done primarily through use of the conventional SWR methodology; and (2) the value investing concept, the idea that all assets provide differing long-term value propositions when purchased a low prices or moderate prices rather than at high prices.

What I did in the mid-1990s, when I was putting together my Retire Early plan, was to say "Assuming that this value investing stuff is all true (and the historical data does indeed back up the claim that value investing works), what sort of SWR would you come up with then?" The answer is, you would come up with very different answers that what you come up with when you ignore valuation, as is the case with the conventional SWR methodology.

It is the combination of the two approaches that provides the Data-Based SWR tool its great power. Value investing by itself provides many insights, but they are often not actionable insights because they are often not quantified insights. What I like about SWR analysis is that things are reduced to a number. It helps to reduce things to numbers when you are trying to compare Option A to Option B; when one number is significantly higher than another, that tells you something. The flaw with the conventional methodlogy, of course, is that it gets the number wrong. There's some value in knowing the conventional methodology number (properly referred to as the Historical Surviving Withdrawal Rate) but there is obviously far more value in knowing the correct number. To get the correct number, you have to factor in the effect of changes in valuation, which is the single biggest factor in any analytically valid SWR analysis.

The first number that we wanted to calculate when we started exploring this stuff was the SWR as traditionally defined--the highest take-out number that is guaranteed to work in the event that the future is no worse than the past. There had been a lot of discussion in our community of the number generated by REHP study and we wanted to know the number that compares to that one when the calculations are done properly. We determined that the number you get from taking an informed look what the historical data says is nowhere even remotely in the same ballpark as the REHP number for the years since our various boards have been in existence.

We have gone far beyond that at this point. JWR1945 has examined lots of questions far removed from the question of what is the highest take-out number that is guaranteed to work presuming that the future is no worse than the past. I think it is fair to refer to this other stuff as "SWR Research" because the analyses being done are making use of the same general analytical techniques as was earlier use to determine the actual SWR number. In most cases, JWR1945 is using the calculator that he developed for purposes of asking the initial question.

We now use the term "SWR Research" loosely, and my guess is that our use of it will become more loose over time. What makes it "SWR Research" as we now use the term is that we are making use of historical data to generate insights about how to best invest to win sustainable financial freedom in the present. That is a statement of the ground rules. Anything that fits under those ground rules fits at this board, so far as I am concerned.

It may be that somewhere down the road we will want to come up with a different name than "SWR Research" for the businsess we conduct here. A name that I use in my head is "Passion Investing." That one will strike people as counter-intuitive because the usual thought is that numbers analysis is not too passionate an activity. I can make a case that it is when the numbers are being used for the purposes to which we are putting them here, but that would be another of those long posts, so I am inclined to skip it for the time-being.

Another option that has some appeal to me is "Probability Investing." I think that phrase sums up what we are doing pretty well. We are calculating the probabilities of various potential outcomes and using those probabilities to inform our investing decisions. I like that one because it sounds broader than "SWR Research" and it might for that reason pull in some people who think of a board limited to determining the SWR as being too narrow to be worth their time.

My sense, though, is that JWR1945 prefers "SWR Research." You put forward a post making a similar point to the one you put forward above at an earlier time, and my sense was that he expressed a preference for the current term. JWR1945 is the one that does the most important work here, so it is obviously his call as to what we call this stuff.

I am OK with any of those three terms and am open to consideration of others. I think that down the line we might want to consider options. I don't see that there is any harm with sticking with the term "SWR Research" indefinitely, though. It does describe what we do adequately. Determining the SWR was our starting-point and we have just sort of taken it from there, following the implications of our findings where they led us.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

Beachbumz, look at the posts related to A New Tool. That methodology strips off the starting price.

The New Tool requires that you estimate future stock market returns. People usually apply some form of the Gordon equation for such a purpose. John Bogle of Vanguard uses the following (except that he prefers to replace the dividend growth rate term with the earnings growth rate):

investment return = dividend yield + dividend growth rate (per year)

speculative return = change of P/E during the period being examined

Total return = investment return + speculative return

You can choose your own speculative return when using The New Tool. That leaves you with looking up the current dividend yield and looking at history to come up with the dividend growth rate. Unfortunately, the dividend growth rate has varied sporadically. Fortunately, the earnings growth rate has been steady, around 1.1% to 1.5% per year (plus inflation).

Of course, dividends come out of earnings. If used wisely, money saved by reducing the dividend payout ratio is recouped in growth. Many dispute that it is ever used wisely. Others point out accounting issues (such as the assumptions related to pension funding) as a reason for an apparent inability to increase growth.

In spite of this, IMHO the other posters have answered your question better than I have. Their information is much more useful.

Have fun.

John R.
beachbumz
** Regular
Posts: 79
Joined: Mon Feb 07, 2005 4:00 am
Location: Cocoa Beach, FL

Post by beachbumz »

Thanks for the replies, but I'm not sure I got the answer to my question (I'm not even sure there is one). Let me put it another way:

My understanding is that you (Hocus) believe that we are in a period of high valuations. It's not my intention to argue that point, but let's just suppose that, in 30 years, when we look back at things that today's pe turns out to be the lowest pe during that 30 year period of time. Assuming that you are not in equities now, or have a low equity allocation based on the 'advice' of the 'data based swr tool', at what point would the 'advice' change, thereby allowing you to take advantage of the rise in equities.

This question uses the assumption that you are mistaken about this being a period of high valuations and should therefore avoid or reduce exposure to the equity markets. (I could just as easily reverse the scenario if you believed that we are in a period of low valuations).

Beachbumz 8)
Life is Good.
unclemick
*** Veteran
Posts: 231
Joined: Sat Jun 12, 2004 4:00 am
Location: LA till Katrina, now MO

Post by unclemick »

Unclemick's iron clad theorm ( with some minor credit to Ben Graham).

When the Dow Jones yield is at 4% - you should own about 50% stocks.
At around 2% - 25% stocks and north of 6% about 75% max. Never 100%.

If I'm wrong - I'll buy you a cup of coffee in twenty years.

Of course I'm not following my own rule - being about 55% stocks, even if 10%(of total portfolio) is REIT Index.
beachbumz
** Regular
Posts: 79
Joined: Mon Feb 07, 2005 4:00 am
Location: Cocoa Beach, FL

Post by beachbumz »

Hi Unclemick!
unclemick wrote:Unclemick's iron clad theorm ( with some minor credit to Ben Graham).

When the Dow Jones yield is at 4% - you should own about 50% stocks.
At around 2% - 25% stocks and north of 6% about 75% max. Never 100%.

If I'm wrong - I'll buy you a cup of coffee in twenty years.

Of course I'm not following my own rule - being about 55% stocks, even if 10%(of total portfolio) is REIT Index.
That doesn't sound like such a bad system! Is the Dow a broad enough measure? (rhetorical ? I guess since you follow it :) ) I'm not sure I would count the REITs in the stock %, I look at it as a separate asset class. FTR, I will most likely never have 50% in equities. You know I'm kinda fond of real estate, hehe.

Problem is, I don't drink coffee! :lol:

Beachbumz
Life is Good.
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

at what point would the 'advice' change, thereby allowing you to take advantage of the rise in equities.
I make an effort to avoid putting forward specific suggestions on how to invest, BeachBumz. I see it as my role to provide solid general insights and then to let those taking in the information decide for themselves what to do with it. That's why I am always saying that the SWR Tool is a descriptive tool rather than a prescriptive tool. It doesn't tell you what to do. It just reports what the historical data says re SWRs.

Personally, I would not be inclined to allow a jump to even higher levels of overvaluation to cause me to increase my equity exposure. If valuations climb higher still, the SWR for stocks drops lower still. The SWR that applies today permits me some equity exposure, but not a great deal. Lower SWRs for stocks would make it that much more problemmatic for me to go with a high equity exposure.

It's possible, though, that my writing business could bring in a good bit of money and thereby change my financial circumstances enough to permit me a higher equity exposure than I hold today (I have a zero percent allocation to stocks today). If that were to happen, and if the SWR for stocks had not dropped too much lower, I could see going with a 30 percent stake in stocks or something along those lines. In the event that stocks perform in the future somewhat in the way in which they have always performed in the past, I expect to increase my equity exposure to at least 50 percent, and perhaps a bit more than that if I am able to bring in enough money from my writing business to permit me to take on more risk.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

beachbumz wrote:Thanks for the replies, but I'm not sure I got the answer to my question (I'm not even sure there is one). Let me put it another way:

My understanding is that you (Hocus) believe that we are in a period of high valuations. It's not my intention to argue that point, but let's just suppose that, in 30 years, when we look back at things that today's pe turns out to be the lowest pe during that 30 year period of time. Assuming that you are not in equities now, or have a low equity allocation based on the 'advice' of the 'data based swr tool', at what point would the 'advice' change, thereby allowing you to take advantage of the rise in equities.

This question uses the assumption that you are mistaken about this being a period of high valuations and should therefore avoid or reduce exposure to the equity markets. (I could just as easily reverse the scenario if you believed that we are in a period of low valuations).

Beachbumz 8)
Speaking for myself:

I would need a credible reason. Something supported by cause and effect. That is all.

There are lots of ways of looking at this. Here is an example:

Suppose you expect stocks to go up because they have always gone up. They will outperform because they have always outperformed.

Consider the following.

The long-term annualized, real return of stocks with dividends reinvested has been 6.5% to 7.0%. By long-term, I am referring to 50 years because that is what John Bogle used in a book.

Suppose that you insist that the long-term annualized, real return is going to stay around 6.5% to 7.0%. What happens if you include the run-up of the bull market, say from the mid-1980s? Starting from then and going forward 50 years, we end up with a bleak forecast for the next few years.

I put up a post along those lines and I included some numbers. I got some emotional, negative reactions. But if you start from the mid-1980s and insist that the 50-year long-term return of the stock market is going to stay inside of its historical range, then the next 30 to 35 years don't look so hot.

Have fun.

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

Post by JWR1945 »

For beachbumz:

In terms of your original line of thinking, I believe that our use of (inflation-adjusted) real returns instead of nominal returns is sufficient to get us very close to the right numbers.

In addition, remember that price-to-earnings ratios automatically cancel most of the effects of inflation. That is, nominal earnings, which rise because of inflation, cause prices to rise as well.

Have fun.

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

Post by JWR1945 »

This question uses the assumption that you are mistaken about this being a period of high valuations and should therefore avoid or reduce exposure to the equity markets. (I could just as easily reverse the scenario if you believed that we are in a period of low valuations).

Beachbumz
Another thing to consider are your objectives and alternatives.

unclemick touched on this ever so lightly.

These days, we have TIPS and Ibonds, which are better than inflation-matched cash. If we wish for a portfolio to last 30 years, we can always get 3.3% without risk. We can draw out 2.5% for 40 years. We can draw out 2.0% for 50 years.

These numbers give us a baseline for our thinking.

An alternative is to consider a dividend-based strategy. This is attractive to many because TIPS interest rates have fallen below 2%.

With a dividends-based strategy, your emphasis is not so much on prices as they are on yields. Yields automatically reflect information about prices.

[Don't forget: Prices still matter! They matter a lot!]

Have fun.

John R.
Mike
*** Veteran
Posts: 278
Joined: Sun Jul 06, 2003 4:00 am

Post by Mike »

We can draw out 2.5% for 40 years.
About the same as for portfolios that include 50% S&P.
We can draw out 2.0% for 50 years.
All bonds don't seem to work too well for youngsters.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

Mike wrote:
We can draw out 2.5% for 40 years.
About the same as for portfolios that include 50% S&P.
We can draw out 2.0% for 50 years.
All bonds don't seem to work too well for youngsters.
Those numbers are with a zero percent interest rate.

If you can depend on getting a 2% interest rate, the numbers are much better:
1) You can withdraw 4.46% for 30 years.
2) You can withdraw 3.66% for 40 years.
3) You can withdraw 3.15% for 50 years.

I believe that you can always get the equivalent of TIPS at 2% interest by using a combination of high dividend stocks, TIPS and/or Ibonds and commercial paper. Coming up with such a combination is far from trivial. You need the ability to draw cash immediately in emergencies, which eliminates a stock-only portfolio, but you may need a higher yield than you can get in the market, which eliminates a 100% TIPS portfolio.

[The yield to maturity of long-term TIPS is currently below 2%.]

Have fun.

John R.
Mike
*** Veteran
Posts: 278
Joined: Sun Jul 06, 2003 4:00 am

Post by Mike »

. Coming up with such a combination is far from trivial.
Indeed. The equations are even more complex for taxable accounts. Taxes affect TIPS' return if they are not in an IRA. I bonds have low annual contribution limits, and lower yield than TIPS. Rebalancing with I bonds can be a protracted affair, and equities may become the best bet again before you can even put the optimal amount in the I bonds.
Post Reply