What Should You Do: Addendum?

Here are some additional details.

Three Good Approaches

We have identified three good approaches:

1) A portfolio consisting entirely of TIPS (and/or ibonds).
2) A portfolio that varies allocations between stocks and TIPS gradually, according to valuations as measured by Professor Robert Shiller's P/E10.
3) A dividend-based strategy.

You can combine these approaches.

TIPS (and/or ibonds)

My response to Greg fills in most of the details that you need right away.

I Bonds versus TIPS

TIPS Ladders for Today opens up more opportunities. Rates had dropped to 1.8% when I wrote it. Today, rates have climbed back above 2.0% at all maturities.

TIPS Ladders for Today

These are your baselines. So long as you can buy TIPS at a 2.0% interest rate, you can withdraw 4.46% plus inflation for 30 years. If you were able to buy TIPS at a 2.2% interest rate, you would be able to withdraw 4.39% plus inflation for 32 years. These rates are with total safety, guaranteed by the United States government. But these baselines leave you with a final balance of zero.

Varying Allocations Gradually with P/E10

We have developed this approach over a long period of time. It was central to our initial research (Current Research A and B). Here are some helpful links:

May 2005 Highlights Summary
Current Research
Fundamental Breakthroughs associated with TIPS
Keeping In Tune with the Human Element

We found that there is a lot of tolerance regarding the details of when you should buy and sell stocks. My main guidance is for you to pay attention to costs. Just because our models buy and sell mechanically does not mean that you should do so. Don’t alternately sell and buy stocks if P/E10 happens to rise and fall just a little bit above and below a threshold for a couple of years. Smooth out the changes.

Choose portfolio SwOptT2 or SwAT2.

Today’s P/E10 Thresholds and Allocations

Professor Robert Shiller provides P/E10 information at his web site. There is always a lag, which means that you need to scale his values.

Professor Shiller’s Web Site

I noticed that his latest set of values with P/E10 was for May 2005, but they did not include the last four quarters of real earnings. Instead, you should scale from the December 2004 data.

S&P500 index: 1199.21 in December 2004.
P/E10: 27.14497997.

Today's S&P500 index level: 1260.53 (during the day).
Today’s Scaled P/E10: 28.53 (nominal, ignoring one year of inflation adjustments and earnings growth).

Scaling for P/E10 equal to 24: the S&P500 index value would be 1060.27.
Scaling for P/E10 equal to 21: the S&P500 index value would be 927.73.
Scaling for P/E10 equal to 13: the S&P500 index value would be 574.31.
Scaling for P/E10 equal to 9: the S&P500 index value would be 397.60.

Scaling for P/E10 equal to 11: the S&P500 index value would be 485.96.

SwOptT2 Stock Allocations:

The stock allocations are 0%-20%-30%-50%-100%.

That is, you would be out of stocks today.

You would stay out until the S&P500 index falls to 1060.
You should hold 20% stocks if the S&P500 were between 930 and 1060.
You should hold 30% stocks if the S&P500 were between 575 and 930.
You should hold 50% stocks if the S&P500 were between 400 and 575.
You should hold 100% stocks if the S&P500 were below 400.

SwAT2 Stock Allocations:

The stock allocations are 25%-40%-75%.

Scaling for P/E10 equal to 21: the S&P500 index value would be 927.73.
Scaling for P/E10 equal to 11: the S&P500 index value would be 485.96.

That is, you would hold 25% stocks until the S&P500 index falls to 930.
You would hold 40% stocks if the S&P500 were between 485 and 930.
You would hold 75% stocks if the S&P500 were below 485.

Portfolio Comparisons:

Our baseline period for calculating Safe Withdrawal Rates is 30 years. There are a variety of ways to extend this to longer intervals.

SwOptT2 has a Safe Withdrawal Rate of 4.4%. It has a Calculated Rate of 5.12%. It is appropriate if you would not mind being out of stocks entirely even if prices were to rise sharply (30% or more) during the next year.

SwAT2 has a Safe Withdrawal Rate of 4.1%. It has a Calculated Rate of 4.84%. It is subject to Benjamin Graham’s constraint that keeps both stock and bond allocations between 25% and 75%. It minimizes regret.

Keep in mind that there is a lot of uncertainty associated with these numbers because they project future events. [Standard engineering practice is to calculate to whatever precision is appropriate for the data. Adjustments come later.]

The Safe Withdrawal Rate corresponds to a high probability of success: roughly 95% of lasting the entire 30 years. The Calculated Rate tells you the most likely result. That is, it is highly likely that your portfolio will grow and allow you to increase your withdrawals above the Safe Withdrawal Rate.

Fixed allocations do not do nearly so well. With 50% stocks and 50% TIPS, the Safe Withdrawal Rate is 3.6% and the Calculated Rate is 4.36% starting at today’s valuations. With 80% stocks and 20% TIPS, the Safe Withdrawal Rate is 3.0% and the Calculated Rate is 4.12% starting at today’s valuations.

More Dramatic Changes in Allocations

We should be able to do even better IF we accept the RISK that stocks will return to their historical valuations within a decade or so.

Notice that our RISK is that we would be stuck with the TIPS BASELINE and a withdrawal rate of 4.46% for 30 years (or 4.39% for 32 years if TIPS yield 2.2%). The biggest drawbacks are the lack of growth and the final balance of zero.

P/E10 is likely to fall as a combination of lower prices and consistently increasing earnings. The market would have to drop sharply to return to normal valuations at today’s level of earnings.

Scaling for P/E10 equal to 13: the S&P500 index value would be 574.31.
Scaling for P/E10 equal to 11: the S&P500 index value would be 485.96.
Scaling for P/E10 equal to 9: the S&P500 index value would be 397.60.

These numbers, although not pleasant for today’s stockowners, are plausible enough after you allow for a decade of earnings growth, general inflation and a sluggish S&P500 index level.

Even those who fear stocks because of price volatility should reconsider when P/E10 falls below its typical historical level of 14 or 15. At these levels, stocks produce ENOUGH INCOME FROM DIVIDENDS ALONE to be compelling.

There is another consideration. If you are able to buy stocks at bargain prices, you are better off with a fixed, high allocation of stocks (than with SwOptT2 or SwAT2).

Dividend-Based Strategies

I have an entire section on dividend-based strategies.

I have found that waiting for valuations to improve helps with dividend-based strategies. You get a double benefit: you get more shares and higher dividend yields.

Have fun.

John Walter Russell
December 5, 2005