Letters to the Editor

P/E10 Graph

Ralph wishes to see a chart showing the current P/E10.

Visit Professor Shiller’s web page. Select his data. Then click the “html file” or download it as an Excel file (or as a zip file). Figure 1.3 shows P/E10 versus the year. Note: the price-to-earnings ratio is P/E10, not the P/E ratio for a single year.
Professor Shiller's web site
Professor Shiller's Online Data page
Professor Shiller's data

Worry-Free Investing by Zvi Bodie and Michael Clowes

Greg mentioned this in an email.

May I also recommend another book to you? Your work goes hand in hand with Zvi Bodie's book, Worry-Free Investing. Bodie is a finance professor at Harvard and he recommends a tips, i bonds approach with never more that 10% in equity index funds or call options (leaps) for those willing to take some risk. His work is a little short on depth and implementation. Your work seems to take this to the next level. William Bernstein is quite critical of Bodie's thesis, but Bernstein assumes that most of us can handle market volatility. I have lots of respect for Bernstein but believe that he is wrong about this for the vast majority of savers/investors - I doubt that most of us can handle risk as expressed by volatility:
Bernstein about Zvi Bodie's approach

HERE IS MY RESPONSE:

I purchased Zvi Bodie’s book, Worry Free Investing. It has good breadth. It is an excellent introduction. As with any book with specific details, the details change with time and some have changed already.

I found names for some of my formulas. What I call the TIPS Equivalent Safe Withdrawal Rate (TESWR) is the same as the formula for a Single Premium Immediate Annuity (SPIA). In addition, I often use the formula for the future value of an annuity except that I make the first deposit at the end of the first year. My formula is simpler.

Chapter 6, Stocks are Risky, Even in the Long Run, does an excellent job of explaining why you cannot make withdrawals based simply on the long-term annualized return of a portfolio (6.5% to 7.0% plus inflation in the case of an all-stock portfolio). The sequence of returns can cause you to run out of money too soon. [If you are lucky, it can extend the duration of withdrawals and/or allow you to increase your withdrawals.]

I am less than enthusiastic about using call options since they are centered on price by itself (instead of total return) and because options are tied tightly to a calendar.

I disagree sharply with William Bernstein’s critique of Zvi Bodie’s ideas. Bernstein points out that prices would rise and interest rates would fall if everybody started buying TIPS. He ignores what would happen next. He assumes that people would be stuck in an unsatisfactory investment.

There is no such thing as overloading a free market [except on a very short-term basis and not with TIPS, which can be printed at will]. Individuals in the market react to price changes. Many people would start looking elsewhere to invest. Bernstein’s criticism applies only if everybody is forced to invest in TIPS. It is much different when people have choices and TIPS is one of several.

Being able to buy TIPS in a 401k plan is not the same as being compelled to buy them. It’s a great idea to be able to buy them.

In fact, we have been seeing the popularity of TIPS increase consistently. Yields have fallen. I checked. Zvi Bodie’s book has a 2003 copyright. His examples assume a TIPS interest rate of 3.0%. Today’s rates are close to 1.8% (i.e., yields-to-maturity for 10, 20 and 30 year bonds sold on the secondary market).

I view stocks differently from Zvi Bodie and William Bernstein. I am satisfied that we can discern the likely behavior of stocks. [Bubbles fall into the unlikely category except to note that they happen more often than most people imagine.] We cannot discern the short-term behavior. It is very close to being random. We can discern what is likely in the intermediate-term (5 to 20 years) based on valuations.

In other words, we can tell when stocks are expensive and when they are cheap. We can take advantage of that better than simply rebalancing our portfolios. Our time frame is in terms of (several) years, not a matter of months.

One thing that we can discern easily is the dividend yield of a stock. Dividend yield defines one measure of valuation. Provided that a company is likely to stay in business, a high dividend yield is an indication of a favorable price. When stocks are cheap, you have your choice of many top quality companies offering high dividend yields.

Today’s prices are not cheap. Today’s stocks are expensive. For that reason, I start out aligned with Zvi Bodie.

Suppose that you could buy most stocks between 25% and 50% of their current prices. My guess is that even Greg would be tempted. Especially, when he sees high quality stocks at prices cut to 25% of today’s levels but still paying today’s dividend amounts. Typically, this means dividend yields of 6% from top-notch companies.

I expect something like that to happen in the next decade or two (5 to 20 years).

Regarding the Faster and Easier Approach

Rob Bennett asked me this question.

John: This question pertains to your article titled "Faster and Easier SWR Calculations" and the Postscript to that article.

You say: "Comparing the HSWR50/HDBR50 results from Calculated Rates of the Last Decade with an earnings yield 100E10/P of 3.5% with Faster and Easier SWR Calculations, the newer approach is more optimistic. Its Safe, Calculated and High Risk Withdrawal Rates are 3.72%, 4.53% and 5.16%. The older, more accurate method calculates these rates as 3.03%, 4.04% and 5.05%."

Is it possible to say WHY the results from the faster and easier approach turn out to be more optimistic results?

Is it reasonable to believe that over time, as more data points are entered into the historical record, there will be less divergence between the results obtained from use of the slower and harder methodology and from use of the faster and easier methodology?

Intuitively, it seems to me that this would be the case. But I don't have enough confidence in my understanding of the statistics-related questions at play to feel comfortable saying whether it is or not.

Rob

HERE IS MY RESPONSE:

Taking more data points usually solves statistical issues. This time, however, I don't think that it will.

I am pretty sure that the correct answer is NO. I do not expect results from the two different methodologies to converge.

I have looked into the nature of errors. The problem seems to be that Safe Withdrawal Rates are dependent on the spread of the data. [The Safe Withdrawal Rate is the lower confidence limit of the Calculated Rate.] Apparently, the spread of dollar values at year 30, which is what the new procedure uses, does not translate well into the spread of Historical Surviving Withdrawal Rates about the Calculated Rate.

I have reported my findings in the General Topics section.
Faster and Easier SWR Calculations: Excursion

Earlier Letters

Here is a link to a recent letter from Greg about TIPS and taxable (non-qualified) accounts
TIPS and taxable (non-qualified) accounts

Here is a link to an earlier discussion with Rob Bennett about SAFE and HAZARDOUS REGIONS
SAFE and HAZARDOUS REGIONS

Here is a link to an early discussion with Rob Bennett about Safe Withdrawal Rates and Historical Surviving Withdrawal Rates
Safe Withdrawal Rates and Historical Surviving Withdrawal Rates

Have fun.

John Walter Russell
Posted July 24, 2005.