April 11, 2008 Letters to the Editor

Updated: June 16, 2008.

QPP Monte Carlo Simulator

I received this letter from Robert.

Have you looked into the Quantext QPP Monte Carlo simulator? I have been using it for the past week and it seems to be a great tool to see how standard deviation affects future returns and the possibility of running out of funds during retirement.

HERE IS MY RESPONSE

I have not used the QPP Monte Carlo simulator. From the promotional material that I read, the QPP has many worthwhile features. It is powerful and advanced.

The QPP Monte Carlo simulator models investments at a lower level than I do. It looks at individual securities. It looks at small slices of the market.

A Monte Carlo simulator is of great value in skilled hands. It is of great value when used to measure sensitivities such as varying standard deviations. It is critically dependent on the quality of its inputs.

I found several interesting articles at the Quantext site.

QPP Description
QPP Articles

Ideally, a Monte Carol simulation generates a variety of possible outcomes consistent with historical behavior. It leads to insights as to what could have been and what might be. The actual historical record is the standard.

The QPP Monte Carlo simulator does not include the effect of valuations directly. They must be introduced through the selection of data inputs.

Thank you for your letter.

Obtaining "Permission" to Ignore Valuations

I received this letter from Rob Bennett.

I liked your article entitled "Diversification" a lot. That's a provocative argument. It is also one that makes a lot of sense.

The two sentences that jump out are: "Just don’t tell me that you are diversifying risk. Risk is tied tightly to valuations." Wow. Those words make me wonder if what is going on is that people are convincing themselves that they have protected themselves by owning lots of assets as a way of giving themselves "permission" to ignore valuations.

This suggests a reason why there is so much confusion over indexing and Passive Investing. People talk of the two concepts as if they are the same thing. I don't see any necessary connection at all. Indexing is a way of achieving large amounts of diversification at low cost. Passive Investing is a "strategy" for putting your head in the sand re the effect of valuations. Why are these two concepts perceived as being related? I'm beginning to think that people are counting on the illusory safety of indexing to make them feel better about the unacknowledged risk they take on by going with Passive Investing. Yikes!

My take is that the problem with the way most people approach diversification is that they seek diversification only among different types of equity investments. True diversification requires consideration of ALL of the various asset types, including the cash-like asset classes (TIPS, IBonds, CDs). These asset classes provide a type of diversification in that they are negatively correlated with the high-emotion asset classes, the ones that become overpriced when people have been talked out of their natural concerns over valuations.

The difference is that this type of diversification is achieved not with buying a fixed percentage of cash-like asset classes. It is achieved by buying whatever amount of the cash-like asset classes needed to keep one's risk profile steady while changes in valuations cause the risk profile of equity investments to rise and diminish.

When thought about this way, there is no need to choose between diversification and concern with valuations. Both are part of the same project. Diversification has to be redefined to include consideration of the need to use diversification for protection from overvaluation of equities.

I agree with your argument. Concern for diversification is greatly overdone. And concern re valuations is greatly underdone. And it's hard not to wonder if there isn't a connection between the two phenomenon -- if the excessive concern re diversification is not being employed as a substitute/distraction for/from feeling a proper concern re valuations.

That was a thought-provoking piece.

HERE IS MY RESPONSE

I believe that the Efficient Market Hypothesis is the connection. It assumes that no one is able to take advantage of price errors. Although thoroughly discredited (refer to David Dreman’s “Contrarian Investment Strategies: The Next Generation”), it continues to have its influence. Its acceptance was a spin off of the longest and strongest bull market ever (1982 to 2000).

Thank you for your insightful letter.

Diversification

DeLong Modifications to PE/10

I received this letter from P/E 10 Wonderer.

Brad DeLong has suggested that P/E 10 gives the earnings 5 years ago, and should be adjusted 6% per year for 5 years to give today's smoothed earnings. Any comments?

P/E 10 Wonderer added this quote and link:

"This is not quite right. This gives you permanent earnings five years ago. To get permanent earnings today you have to grow the average by five years of the trend growth rate of real earnings--which has been 6% per year over the past twenty years. The current value of 28 for the price-to-moving-average-of-lagged-earnings ratio corresponds to a price-to-permanent-earnings ratio today of roughly 21, and to a current r of4.8% per year."

Professor Brad DeLong's Weblog Archive Page

HERE IS MY RESPONSE

Thank you for the link. It is interesting. The site is well worth visiting.

In terms of Professor Brad DeLong's article: his adjustment makes sense. He is trying to tie the numbers down to absolutes, to what would be the “right” price to pay for stocks in a theoretical sense. His research is valuable.

Your question brought to mind the issue of using weighted earnings and/or weighted values of P/E10 for better predictions. I had put this off in the past because I thought that it was difficult to do with Excel. I have now figured out a very simple way to make such calculations. I can introduce weights in the same way that Professor Shiller converts nominal amounts into real dollars in his database. Mathematically, weighting is similar to using the CPI (Consumers Price Index). I will open a new Current Research section to investigate weighted alternatives to P/E10. My first application will be in terms of 10, 20 and 30 year stock return predictions. Another possible application is for me to make some new calculators (similar to the Deluxe Calculator V1.1A08a and/or the Gummy Series of calculators).

Thanks again for your letter. It was helpful.

P/E 10 Wonderer followed up with these gracious words:

Thank you for your response. I am glad that you enjoyed the Brad DeLong link. I look forward to your 10-30 year predictions using weighted P/E 10.

Later, P/E 10 Wonderer continued with this:

John Walter Russell wrote: "Summary

All of these results were close."

After looking over your new research M using Brad DeLong's method, I concluded that the results are fairly close to regular P/E 10 results. Is this a correct conclusion?

Thank you for your time.

HERE IS MY RESPONSE

You are right.

The prediction accuracies with all of the weightings were very close. The confidence limits and R-squared values for each time frame were almost identical regardless of weighting. The “unexplained” randomness is almost identical in each case.

This having been said, the data are suggestive that the DeLong method does the best job for making Year 30 predictions. They are not definitive in terms of the evidence that I have provided.

I HAVE EXAMINED THE DATA FURTHER

I compared predictions with earnings weights of 0.94, 1.00 [i.e., the traditional P/E10] and 1.06 with the actual returns at Years 10, 20 and 30. I calculated and compared the prediction errors.

Nothing caught my attention. I did not see anything obviously different. As far as I have been able to discern, the results are very close.

P/E10 Wonderer has added these kind words:

Thank you, John. The plain English summary was also quite helpful to my limited math skill brain. It’s reassuring to know the results hold up when using updated earnings.

John Walter Russell

I received this letter from Cal.

My first time to your website. Very interesting.

Who are you and what is your background?

By sending this email I am not going to end up on multitudinous mailing/email lists and spams, am I?

HERE IS MY RESPONSE

I am a retired Electronics Engineer.

I do not offer my services for sale. I do not sell or advertise anything. I do not put names on mailing lists. Everything is at my personal expense.

I will never breach your confidentiality.

You may find this background article of interest. I am JWR1945. SWR stands for Safe Withdrawal Rate. FIRE stands for Financial Independence/Retire Early.

Our Roots

Letters to the Editor in 2008

Letters to the Editor in 2007

Letters to the Editor in 2006

Letters to the Editor in 2005