June 15, 2009 Letters to the Editor

Updated: July 22, 2009.

Setting up a General Retirement Portfolio to use as a Guide

I received this letter from Dennis.

Hi John..

Q. In your Leisure time, would you put together a Gen. Portfolio for, as follows:

1. Retired age 65

2. Has SS

3. And Using at a per $1k /yr needed for income

A. How much is needed to provide that $1k/yr for the next say 20 yrs? till age 85 or 25 yrs = age 90?

B. And Where would you put this $ in?

And of course anything else you feel pertinent to put in.

Thank you


Thank you for an interesting question.

Just about everything that I come up with delivers 6%+ of the original balance plus adjustments to match inflation. That is, plan on $16000 to $17000 for every $1000 generated. If you can tolerate the risk, there is at least one reasonable investment that generates $1000 plus growth for $13200.

Social security income is secure and it has an inflation escalator. I would add a low cost Single Premium Immediate Annuity to bring the balance high enough to meet your minimal needs.

This guarantees that your needs are covered even if you start losing your mental faculties, an issue for older citizens.

If you check the TIPS table, you will find that you can withdraw 6% (plus inflation) for 20 years if you can maintain a TIPS ladder at 2% interest. This is $16700 per $1000 of income. You can cover 26 years at 5% (plus inflation). This is $20000 for every $1000 of income.

This covers investments with the highest levels of safety. They match inflation as well.

I do not cover individual investments as a rule. I will make a partial exception to point out what you should be looking for in additional investments.

I have a practice portfolio (Something Simple) that consists of two ETFs (Exchange Traded index Funds). It is a dividend blend portfolio. I combine common stock dividend oriented DVY with preferred stock ETF PFF in a 50%-50% mix. DVY took a hit because of the banking crisis. PFF did not even though it is heavily in financial instruments.

I put this into my Simplified Automatic Allocator. It is a spreadsheet.

I assumed (in accordance with MarketWatch) that DVY has a yield of 4.74% and PFF has a yield of 9.82%. I assigned DVY a 5.5% per year growth rate of the nominal dividend amount. This matches the S&P500. I assigned PFF a zero percent dividend growth rate. I assumed 3% per year inflation. I assumed 2% TIPS.

The net result is a continuing 6.7% (plus inflation) continuing withdrawal rate. It would never run out of money. It could suffer a temporary setback up to 25% if inflation takes off. If so, the minimum withdrawal rate would be 5.0% (plus inflation). This would last only a handful of years. It would recover fully.

The required investment is $15000 for each $1000 generated. But the withdrawal amount could fall by 25% temporarily.

Another investment comes to mind.

I have seen the REIT Realty Income Corp “O” recommended by several sources. They understand the needs of income investors. This REIT has a dividend of 7.58%. They have a long history of growing dividend payments.

Assuming that you can match the yield of “O” with a safe combination of investments and that they all continue to grow their income streams, it would take $13200 per $1000 of income.

Something Simple


I received this letter from Rainy.

I have found your website to be a fascinating read, though the math is a bit daunting.

I will like to ask what would you recommend for a young investor in his/her early 20s seeking to accumulate funds for retirement?

I was thinking of investing in dividend-paying stocks. In addition, should we sell the stock if it is over-valued even though it continues to pay good dividends?



And I thank you.

From my Book Section:

“If I had to select only two investment books, my first choice would be David Dreman’s The Contrarian Investment Strategies: The Next Generation. My other choice would be Lowell Miller’s Single Best Investment.”

I agree with your choice. Dividend paying stocks are an excellent choice. Lowell Miller’s book focuses on dividend investing. It is outstanding.

You should sell any stock when it gets far ahead of itself. How far is a judgment call. Deciding when to sell is difficult.

There is one thing, however. You need to see how smoothed earnings support dividends, not single year earnings. Many times, a company will have an earnings shortfall that lasts only one or two years. Avoid overactive trading. Do not be overly concerned so long as dividends seem secure.

Once you make a decision, stick by it. Don’t second guess yourself. This is the hardest part of investing.

You will do well to read “Passion Savings” as well. It can accelerate your financial independence dramatically.

Books Section

Permission Granted

I received a letter from Elizabeth asking for reprint permission for Dividend related information.


I grant full reprint permission for anything at this site as long as you mention your source.

Thank you for making an official request. I appreciate that.

Asset allocation and management with a dividend/yield focus

I received a letter from Robert.

Perhaps you have already done this. If so please point me to the place on your website or elsewhere.

I'm a yield/dividend, in retirement, investor - using a portion of the dividend stream as income and another portion for increasing my equity interests.

There does not seem to be much that looks at allocation with a dividend orientation. An algorithmic look backwards modeling would seem to be possible. One could create security allocations on an annual %$ portfolio yield basis. Then annually selecting equities from some sort of yield-oriented (and yield growth)(and perhaps equity value) criteria/list...and see what happens over the years.

I've not seen anyone attempt to do any kind of modeling for true investment portfolios. (Where the intent is to retain ownership and spend the yield/dividend from the investment - as opposed to the intent to sell portions of the equity-ownership - correctly named speculation) Also all the modeling is "market wide", which of necessity, is a highly speculative assessment - as the vast majority of equities pay no yield.

This all appears to be pretty virgin territory for research. But, then again, it may be done and I've just not come across it.


Thank you. You are right. This is an exciting area of research. And, yes, I have made some progress along these lines.

Use the SEARCH feature to find articles about “Income Stream Allocators.” You will find a series of articles investigating the income stream produced by an initial purchase of dividend generating stocks (or interest bearing securities).

I started out using a series of investments with different initial dividend yields and dividend growth rates. I quickly found that the mathematics favor a combination of an investment portfolio which grows its payout rapidly with an investment that starts out with a very high yield. I call this a “Dividend Blend.” You will also find a wealth of articles centered around a Dividend Blend.

The Dividend Blend starts out with an excess of income from the high yielding investment. You use the excess later to fill in the middle years. Later, the fast growing dividend payers take over, increasing income much faster than inflation.

One of the problems with this analysis is the reinvestment of dividend income. It is not clear what price to use for future purchases. As a result, I used a side account in my spreadsheets. I invested in TIPS to manage excess cash. Real life investors can expect to do better.

While you are at it, consider a high yield, low dividend growth portfolio on its own merit. The most important issue is compensating for inflation. If you reinvest part of the yield, you can grow principal enough to keep up. I refer to this as a straight Income approach. I like what Ben Stein and Phil DeMuth wrote along these lines. I like what ElLobo at the Morningstar Income & Dividend Investing discussion board has to say even better.

The biggest problem with dividend investment research has been the need to make assumptions. What is the initial dividend yield? What is the dividend growth rate? What is the inflation rate? Except for today’s initial dividend yield, the answers are not very reliable.

We can relate to history and the long term behavior of the S&P500 index to get an idea of what is reasonable. The nominal dividend growth of the S&P500 index has been remarkably stable at 5.5% per year (annualized). The long term inflation rate has been around 3% per year (annualized). Spikes in inflation have affected S&P500 real dividend amounts greatly. The biggest drop in dividend buying power was close to 25% (to 75% of the original amount in real dollars). I have a graph called “S&P500 Dividend Theory” that you may find helpful.

In spite of these difficulties, we can draw meaningful information from this research. As a practical matter, the individual investor needs to make spreadsheets and follow his investments closely. I imagine that making a detailed update every five years would be sufficient.

Do not be afraid of buying equities yielding a little bit more than the S&P500. Conceptually, just buying the index and then removing non-payers would do as much. There are a lot of quality companies that pay much better than the S&P500 index as a whole. Just be sure to apply other measures of value as well (low P/E, low P/B and low Price/sales). Take advantage of Mergent’s Dividend Achievers to identify companies that grow dividends consistently. But be careful not to pay too much. Pay attention to the initial dividend yield.

Subject: P/E 10

I received a letter from Paul.

Hi all,

Having trouble confirming current P/E10.

Shiller's Online Data website ends in 2007 at 26.7.

Early Retirement shows 28.

Updated Shiller by month which shows 16.24.

Dshort.com shows 16.7.


Thanks for a great website.


Thank you. The exact level of P/E10 is not critical. A change of 1 or 2 seldom makes a big difference. However, I do keep it up to date at this site. If you know the current S&P500 value, you can determine P/E10. [You use the Stock Returns Predictor. I explain how at the P/E10 button on the left.] What I report is actually a linear extrapolation, but it works very well.

Check Professor Shiller’s Online Data again. He keeps his data up to date. The latest entry is 2009.07, which is year 2009 and month 07 (July). His P/E10 values are accurate through April 2009, which is exactly one month later than the last month with an entry under “Real Earnings.”

[Later values use zeroes in the calculations for April, May and June. Professor Shiller’s Figure 1.3 uses his July 2009 calculation. He reports 16.24. This has three months with zero earnings.

I exclude the months with zero earnings.

Again, a higher level of precision is not critical.

I cannot confirm the data from other web sites except for Rob Bennett’s. He uses the same Stock Returns Predictor as I do.

Professor Shiller’s Online Data

Letters to the Editor in 2009

Letters to the Editor in 2009

Letters to the Editor in 2008

Letters to the Editor in 2008

Letters to the Editor in 2007

Letters to the Editor in 2007

Letters to the Editor in 2006

Letters to the Editor in 2006

Letters to the Editor in 2005

Letters to the Editor in 2005

Search this site powered by FreeFind