Is the Raddr Methodology Analytically Valid?

Research on Safe Withdrawal Rates

Moderator: hocus2004

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

Post by JWR1945 »

When making comparisons of long-term returns and Safe Withdrawal Rates among portfolios, remember that the traditional high stock portfolio includes commercial paper.

These calculations are for 80% stocks and 20% commercial paper and 0.00% in expenses.

I have made a graph for 1923-1972 30-year returns versus the percentage earnings yield 100E10/P. I had Excel fit the graph with a straight line.

With 30-year periods and excluding all sequences with dummy data, results from 1923-1972 are available for making a curve fit if we stay within the modern era. The formula is y = 0.2976x+3.6738 and the variation in the data is very close to and slightly greater than plus and minus 1.5%. R-squared is 0.3112. In terms of P/E10, the equation is y = [29.76 / (P/E10)]+3.6738.

I looked up the values of P/E10 from the post Calculated Rates of the Last Decade dated Wednesday, Jun 23, 2004.
http://nofeeboards.com/boards/viewtopic.php?t=2657

Code: Select all

1995    20.219819
1996    24.763281
1997    28.333753
1998    32.860928
1999    40.578255
2000    43.774387
2001    36.98056
2002    30.277409
2003    22.894158
The last entry in Professor Shiller's list is for November 2003. The S&P500 index was at 1054.87 and P/E10 was 25.898702. [To help with scaling: today's the S&P500 index started at 1134.41. If ten-year earnings were the same as in November 2003, today's P/E10 would be 25.898702*(1134.41/1054.87) = 27.851533.]
I used the 1923-1972 results to make the following table. The formula is y = [29.76 / (P/E10)]+3.6738 and the confidence limits are plus and minus 1.5%. Recent S&P500 index levels have been close to the referenced level of 1134.11.

Year, 1923-1972 Calculated 30-Year Return, Upper Confidence Limit, Lower Confidence Limit

Code: Select all

1995    5.15   6.65   3.65
1996    4.88   6.38   3.38
1997    4.72   6.22   3.22
1998    4.58   6.08   3.08
1999    4.41   5.91   2.91
2000    4.35   5.85   2.85
2001    4.48   5.98   2.98
2002    4.66   6.16   3.16
2003    4.97   6.47   3.47
Today's Calculated 30-Year Return would be only slightly more than that of 1997. It would be less than that of 1996.

Summary

These results apply when a portfolio is left untouched for 30 years assuming that all dividends are reinvested. Expenses have been set at 0.00% in these calculations to make them compatible with previous calculations.

Starting from today's valuations, this portfolio [80% stocks and 20% commercial paper] can be expected to grow at an annualized real rate of 4.7% plus and minus 1.5% in 30 years.

Compare this with a portfolio of 100% stocks. It will grow at 5.3% plus and minus 2%.

With the [80%] high stock portfolio, one can depend upon a real return of 3.2% (minimum). With an all-stock portfolio, one can depend upon a real return of 3.3% (minimum). That is, 30 years is long enough for the upside of a 100% stock portfolio to overcome its downside risk as compared to an 80% stock / 20% commercial paper portfolio.

It turns out that both the high (80%) stock portfolio and the 100% stock portfolio produce almost identical minimum returns at 30 years. Discussions about the penalty [at high levels of safety] for selling stocks early, when based upon a 30-year period, apply equally well to both portfolios.

Have fun.

John R.
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

"Discussions about the penalty [at high levels of safety] for selling stocks early, when based upon a 30-year period, apply equally well to both portfolios. "

Do you agree that the penalty is a most severe one? It sure seems so to me.

With a penalty for selling stocks that is that severe, it seems to me that aspring early retirees need to be made aware of the importance of bringing their stock allocation down to a level where they are absolutely confident that they can avoid stock sales to cover living expenses. That's the lesson that I am picking up in these recent analyses of the historical data.

Am I on solid ground putting forward such advice or is there something that I am not taking fully into account?
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

hocus2004 wrote:With a penalty for selling stocks that is that severe, it seems to me that aspiring early retirees need to be made aware of the importance of bringing their stock allocation down to a level where they are absolutely confident that they can avoid stock sales to cover living expenses. That's the lesson that I am picking up in these recent analyses of the historical data.

Am I on solid ground putting forward such advice or is there something that I am not taking fully into account?
Yes, you are on very solid ground. Portfolios started in the mid-1960s were hit by a double whammy: stock prices collapsed (as multiples collapsed) and real dividend amounts fell!

There is something that you may not be taking fully into account although you are fully aware of it. You can adjust the composition of your portfolio. This is the most important consideration behind dividend-based strategies. They get a double benefit (the opposite of a double whammy). Not only can people increase their stock holdings comfortably, their portfolios are highly likely to grow at least as much as the overall stock market and they are likely to outperform the overall market.

Have fun.

John R.
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

"Yes, you are on very solid ground."

Thanks for confirming that, JWR1945.

"They get a double benefit (the opposite of a double whammy)"

I understand. I see this as a separate question, however. To see the extent of the Stock Selling Penalty, you need to compare apples to apples. Since we calculate the SWR for 80 percent S&P stocks, to see the size of the Stock Selling Penalty, we need to look at S&P stocks when calculating the comparable 30-year return number for stocks where there are no sales and where dividends are reinvested.

There are gaps in my understanding of the benefits of owning high-dividend stocks, as there are gaps in my understanding of so many other aspects of the SWR project. But I do possess a limited understanding of some of the benefits, and the argument in favor of owning high-dividend stocks makes a good bit of sense to me. If it is possible to obtain a not-so-bad 30-return from owning a moderate amount of S&P stocks even at today's valuations, my guess is that it is possible to obtain something at least a little better than a not-so-bad 30-year return from owning a moderate amount of carefully selected high-dividend stocks at today's valuations.

If your portfolio is big enough to finance a safe retirement with a not-so-bad return, a not-so-bad return can be pretty darn good! It's better than a stick in the eye. And even a stick in the eye is better than a busted retirement. So, for some, discovering a strategy that the historical data indicates will provide a not-so-bad return could be pretty darn exciting stuff.

I don't think that the strategy being suggested by the data examined in this thread is a strategy that many would have come up with without looking at the data. The Data Knows All. But it doesn't tell all unless you ask the right questions. The secret of investing effectively for early retirement is constructing analytically valid methodologies for uncovering the mysteries until now hidden within the data.
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

Are we studying numbers? Or are we writing poetry?

The only true answer is: "We're doing both!"
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

hocus2004 wrote:Are we studying numbers? Or are we writing poetry?

The only true answer is: "We're doing both!"
This sounds like a quote from a mathematician who is talking about an elegant theorem in advanced mathematics.

Have fun.

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

Post by JWR1945 »

hocus2004 wrote:I don't think that the strategy being suggested by the data examined in this thread is a strategy that many would have come up with without looking at the data. The Data Knows All. But it doesn't tell all unless you ask the right questions. The secret of investing effectively for early retirement is constructing analytically valid methodologies for uncovering the mysteries until now hidden within the data.
I have mentioned this before.

I consider hocus2004's questions to be helpful and important in my research. People should never downplay his contribution.

Have fun.

John R.
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

"People should never downplay his contribution."

I am a Numbers Dunce. That needs to be kept in mind at all times. For a Numbers Dunce, though, I don't think I do so bad at this stuff!

Also, there is at least one benefit that follows from me being a Number Dunce. There aren't too many community members who possess the numbers skills of JWR1945. So SWR analysis has the potential to become intimidating to a high percentage of the board community. My numbers duncery makes it possible for me to pull the discussion down to the level of grunt talk any time that JWR1945 tries to get too fancy with his theorems and equations and such.

The community needs Numbers Guys. It also needs Grunts and Dunces. We all have our own special role to play. The way I see it is, God made me a Numbers Dunce for a reason, and my job is just to make some good come of it.

Even mosquitos have an important role to play in God's cosmic plan. I'm the mosquito of the SWR Research Group and proud of it!
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

There's been confusion from the early days as to what it is that SWR analysis does. It is my sense, for example, that many community members are not even clear as to what sort of return assumption is implicit in the REHP study claim that a 4 percent withdrawal is "100 percent safe" for those with a portfolio of 74 percent S&P stocks. Even if the methodology used to generate the 4 percent number were analytically valid (it is not), the 4 percent number would signify a not-good 30-year return. The REHP methodology assumes that the portfolio will be depleted over the course of the 30 years in a worst-case scenario. In the case in which a 4 percent take-out is the highest take-out that permits the portfolio to survive 30 years, the return obtained is only slightly in positive territory. A zero percent return would permit a take-out of 3.3 percent to work. I think we need to begin thinking about developing a terminology that makes this point, and other related points, more clear.

I've been thinking this for some time. But the numbers put forward in this thread lead me to believe that the need is more pressing than I previously realized. I do not possess complete confidence in the numbers put forward in this thread. I need to study them more and come to understand them better. For the time-being, however, I am working on the assumption that the numbers discussed in this thread accurately portray the realities. In the event that is indeed so, we have an important story to bring to the world of investors interested in winning financial freedom early in life.

One side of the story is that putting a high percentage of your assets in an S&P index at today's valuation levels offers a highly dubious value proposition. The other side of the story is that putting a moderate percentage of your assets in an S&P index at today's valuation levels offers a not-altogether-bad value proposition. If the numbers in this thread stand up to scrutiny, we have discovered something of immense value. We have discovered that it is possible to obtain a strong value propositon from investing in a stock index even at times of extreme overvaluation so long as you take into account the possible effect of the severe Stock Selling Penalty and limit your allocation percentage accordingly.

What makes this confusing is that SWR analysis has generally been designed to be applicable only to retirees. It is not appropriate for non-retirees to draw conclusions from SWR studies as to how to invest during the asset accumulation stage. Everyone understands this is so as a technical matter. The reality, however, is that many have improperly drawn inferences from SWR analyses as to how to invest during the accumulation stage. There are many posts in the archives showing this to be the case.

We have now come to a strange pass in the road. We have shown beyond any reasonable doubt that the conventional methodology studies greatly overstate the SWR at times of high valuation. That finding has been the focus of controversy for some time now. In the past two weeks, we have explored data that suggests a very different sort of guidance for investors. This thread suggests that, in the right circumstances, S&P stocks are a moderately attractive investment choice even at times of extremely high valuations. I see this new finding as offering us a possible means of defusing much of the tension that has revealed itself in most SWR threads that have been put forward since the Great SWR Debate kick-off post of May 13, 2002.

Much of the controversy is over the fact that the Data-Based SWR Tool generates such shockingly low numbers. We have all recently lived through the greatest bull market in U.S. history. Many community members find it hard to relate to the idea that the SWR for TIPS in the year 2000 was more than three times the SWR for an 80-percent S&P portfolio. That's what the numbers say, but it is a finding that people find hard to believe. I am thinking that our more recent finding may help us win back the confidence of community members who reacted to shock to the numbers we put forward at earlier times re the SWR for high-stock portfolios at the sorts of valuation levels that have applied since the late 1990s.

It appears that the reality may be that, while the SWR for an 80-percent stock portfolio is indeed shockingly low, the SWR for a stock investment that represents a smaller percentage of one's portfolio may be a good bit higher. It is beginning to look like the key is not whether you are invested in stocks or not. The key is whether you are using reasonable allocations in your stock investments or not.

Go with an 80 percent stock allocation, and you leave yourself vulnerable to the severe Stock Selling Penalty. Go with a more modest stock allocation, and you can by taking on a not unreasonable amount of risk take a 4 percent withdrawal rate from the portion of your assets invested in S&P stocks. Go with a strategy of careful selection of high-dividend stocks and you might be able to justify a take-out number of even a little more than that. We are well on the way to the formulation of investment recommendations that are both true to the data (unlike the conventional methodology recommendations) and not so shocking to investors who have become accustomed over the past 20 years to hearing that stocks can be counted on to provide healthy returns over the long term. Putting our two big insights together may permit us to build a package insight more powerful and appealing than either of the two insights is standing on its own.

I am thinking that it may be time to begin moving away from the narrow vision of SWR analysis that limits the tool to use only for retirees. Setting things up that way causes a lot of confusion. A good number of community members are not retirees, but they obviously want to participate in investing discussions and many of our investing discussions are SWR discussions. So they tend to listen to what is said about SWRs and presume that it applies in some indirect sense to their own investing strategies. That can cause trouble because the extent to which realities that apply in the asset distribution stage (retirement) also apply in the asset accumulation stage (pre-retirement) is not generally well understood.

The key distinction, it seems to me, is not whether you are retired or not retired. The key distinction is are you in a position where you will sell stocks if there is a big price drop. If a big price drop will cause you to sell stocks, stocks generally do not offer a strong value proposition. If you are in a position where you will not sell stocks and will reinvest dividends regardless of the size of any price drops, you can reasonably expect decent long-term returns from stocks. My proposal is that we switch the focus from the question of "Are stocks any good when valuations are high?" to "How can an investor protect himself from the severe Stock Selling Penalty when stock prices are high?"

The stock selling penalty does not apply only to retirees. An investor in the asset accumulation phase does not need to sell stocks to cover his living expenses. But how much practical difference does this make in the real world? An investor in the accumulation phase who has 80 percent of his portfolio in stocks is going to sell stocks if prices fall by 50 percent or so, is he not? My understanding of the historical data is that this is what has generally happened in the past. So I think that it is reasonable to believe that this is what is going to happen in the future too.

The real risk is not in buying high-priced stocks. It appears that stocks perform well enough to provide reasonable returns to the investor over the long term so long as he limits his allocation to high-priced stocks. The real risk is putting too high a percentage of one's accumulated assets in high-priced stocks. Keep your stock allocation to 30 percent or so, and you are not going feel strong pressure to sell stocks if prices drop severely. Invest so much in stocks that all of your hopes are riding on them never having a severe price drop, and you make yourself vulnerable to the Stock Selling Penalty, the greatest destroyer of Retire Early dreams that we have come across in our travels thus far.

The ideas put forward in this post are tentative ones. I have a lot more study and thinking to do in connection with the data put forward in this thread. My tentative thought, however, is that it may be time to consider a shift in focus. I am thinking that we should be advising all aspiring early retirees, both those in the asset accumulation stage and those in the asset distribution stage, to be sure to avoid the Stock Selling Penalty. That is the real killer of Retire Early dreams. Avoid the Stock Selling Penalty by keeping your investment in high-priced stocks to a moderate allocation percentage, and you don't need to worry so much about what sort of return scenario is going to pop up in the next few years. Invest for the long term, and the long-term growth potential of stocks will bail you out of most possible jams. Go too high with your stock allocation percentage, however, and the Stock Selling Penalty will eat you alive. It is the Stock Selling Penalty that is the monster of our investing nightmares. This is a monster that attacks not only retirees, but all investors pursuing the rewards of a life of financial freedom.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

hocus2004
My proposal is that we switch the focus from the question of "Are stocks any good when valuations are high?" to "How can an investor protect himself from the severe Stock Selling Penalty when stock prices are high?"
Such discussions merit new threads.

In the mean time, I have posted the results of what happens at 30 years with combined portfolios that are rebalanced annually. In all cases, an investor would have been better off by keeping his stock portfolio separate. My calculations were for valuations of P/E10 = 27.0, which is just barely above those of today.

Our retirement calculators can examine dollar cost averaging. We just make the withdrawals negative. [That is, a positive withdrawal is a negative deposit. A positive deposit is a negative withdrawal.] The only difficulty is describing results in terms of an annualized gain. The calculators would use the initial deposit, which would be very low. OTOH, the calculators still tell us what the portfolio balance is each and every year.

There is an issue of how portfolio balances change with time. There is a wide spread of possible returns, including some serious losses, at year 15. Would an investor who has a real total return of 0% to 1% (annualized) remain committed to stocks? It might be difficult.

Have fun.

John R.
Post Reply