June 3, 2008 Letters to the Editor

Updated: August 1, 2008.

Dollar Cost Averaging at Year 15

I received this letter from Rob Bennett.

I love the articles that you write that use Scenario Surfer runs to compare different strategies. They almost always cause me to think about angles that I had not previously considered.

You find that "Valuation-Informed Indexing (VII) narrows the spread of the data considerably.... The potential upside dollar amount is not as high as with 100 percent stocks." I have four observations to offer in response.

One, early-year results have a disproportionate effect because of the compounding returns phenomenon. A portfolio of 100 percent stocks certainly has a greater upside potential than VII in the early years. In those cases in which 100 percent stocks goes ahead it can stay ahead for some time because the investor obtains the benefits of compounding returns on those early good years unless a really bad year comes along to wipe them out.

Two, the downside potential of stocks is also much bigger at today's valuations, of course. In those cases in which the 100 percent stock investor gets hit with something bad in the early years, it is the VII investor who obtains the benefits of compounding returns. So, yes, VII serves to narrow the spread. It protects you from worst-case scenarios and makes it less likely that you will see best-possible scenarios.

Three, my strong hunch is that VII would compare better to 100 percent stocks if the test were a 30-year test rather than a 15-year test. In the runs that I have done, it is often one bad year that knocks out the high-stock-allocation strategy. Those can come up in 15 years. But sometimes it takes 30 years for one of them to come up. The knock-out blows deliver a harder hit if the come up early. But knock-out years can deliver a hard hit even if they come after the completion of 15 years.

Four, I don't fault you for setting things up in the way you did but I do believe that there is one aspect of the set-up that is more than a little bit unrealistic. You have the 100 percent stock investor staying 100 percent in stocks even after experiencing a big hit. How many would do that in real life? In real life, most 100 percent stock investors would lower their stock allocations dramatically after suffering a big hit. That would bring the 15-year results for 100 percent stocks down dramatically.

I would love to see you do some research focusing on that last factor -- the extent to which a lowering of one's stock allocation after a big hit would hurt long-term results for those who start out with high stock allocations. My view is that it is excessively high stock allocations that cause huge hits (VII investors are to a large extent protected from them). One of the risks of going with excessive stock allocations is that you will pull out of stocks (in part or in whole) after suffering the big hit. Few high-stock investors will acknowledge this before suffering the big hit. Still, the reality is that this is one of the big risks of strategies of going with high stock allocations at times of high valuations (when the risk of big price drops is greatest).

Thanks for another high-insight article, John.

HERE IS MY RESPONSE

Thank you for another insightful letter.

On your third point, I will have to disagree. I wrote this summary about Year 30 results in Current Research L: When Price Drops Occur. In fact, this is what led me to look at Year 15.

The Dollar Cost Averaging Trade Off

This is what happens with dollar cost averaging and Valuation Informed Indexing combined:

Largest two values: 139,683 and 119,653.
Middle two values: 113,740 and 110,540.
Smallest two values: 88,710 and 88,387.

This is what happens with dollar cost averaging with 100% stocks:

Largest two values: 248,310 and 181,564.
Middle two values: 144,957 and 114,324.
Smallest two values: 70,087 and 62,875.

Valuation Informed Indexing narrows the spread at a slight reduction in the median balance (the middle values). It gives up the impressive upside of 100% stocks. But it assures a higher income under worst case conditions.

Dollar cost averaging with 100% stocks succeeds best when price drops occur early and when they are deep. Valuation Informed Indexing removes the sensitivity as to when price drops occur.

Current Research L: When Price Drops Occur

The reason behind the Year 30 results is easy enough to understand.

By Year 15, the worst is over. Market valuations have become favorable once again. Owning 100% stocks can make sense under such circumstances.

Valuation Informed Indexing often had high stock allocations after Year 15.

I like your suggestion for new research. I will make some sensitivity studies.

Rob Bennett followed up with these playful remarks.

You say: "On your third point, I will have to disagree."

I say: "It's so totally unfair when this sort of thing can be checked out by looking at the numbers!"

I'm just joking around.

Thanks very much for clearing that up, John. I continue to marvel at how much we learn about what may happen in the future just by devoting some attention to how things have turned out in the past.

Swiss Annuities

I received this letter from Don.

I have found your web pages and research of great interest and value in planning for my retirement especially as regards designing a portfolio and planning a withdrawal strategy. Thank you for you thorough and scholarly approach.

As I look into the future, I am increasingly concerned at both the domestic and foreign political climate and their implications for the dollar, taxation, and inflation. In an effort to avoid the financial negatives of these trends on the USD [United States Dollar], I am intrigued by the concept of "Swiss annuities". Fixed or variable, immediate or deferred the idea of investing in a gold based currency via an annuity for a potentially more secure future payout (in CHF) [Swiss Franc] seems a good strategy. Of course, like other investments "timing" would be of significant importance for exchange rates, age and interest rates!

I would be interested in your "take" on this idea and how it might be worked into a general dividend-bond-annuity strategy.

Thank you for your time and your web page!

HERE IS MY RESPONSE

Thank you, Don, for your very kind words.

I am unfamiliar with Swiss annuities. I limit my comments to what I have picked up via a brief search of the web.

My understanding is that a United States citizen is NOT allowed to buy a fixed Swiss annuity. He can only buy a variable annuity. It cannot be self directed.

The most common complaint against annuities (and Variable annuities in particular) is high cost. If I understand correctly, this is also true of Swiss annuities. They have annual fees totaling close to 2%. [I have also read that some have annual fees as low as 0.8%.] Unlike their United States counterparts, however, these are annual fees as opposed to an upfront load.

Swiss annuities are free of withholding, which may make them more attractive than owning a Swiss bank account.

Swiss annuities have a guaranteed minimum return.

I did not ask for a quote to find out how much income a Swiss annuity pays out. I do not know how large the income stream might be in terms of the original investment.

I appreciate your concern about the United States dollar.

I do not claim any expertise on currencies. I will mention Steve Forbes’s observation that weakening the dollar has never improved the economy. I will add that I am beginning to hear words about strengthening the dollar. The question remains as to whether we will actually do it. Dr. John Hussman (of the Hussman funds) believes that we will not. He doubts that the Federal Reserve will lift interest rates to strengthen the dollar. He believes that the Federal Reserve is more likely to keep interest rates low because of subprime mortgages.

June 16, 2008 Weekly Market Comment by John Hussman

P/E10

I received this letter from Paul.

I have been reading your site with great interest, especially intrigued with concept of P/E10.

Today is 8/1/2008.

What is the current P/E10?
How do I find a current number on your site?
How often do you update it?

HERE IS MY RESPONSE

Thank you, Paul.

P/E10 is the invention of Professor Robert Shiller, who wrote “Irrational Exuberance.” It is the current (inflation adjusted) S&P500 price divided by the average of the previous decade of (inflation adjusted) earnings. Professor Robert Shiller maintains the S&P500 database in the Online Data section of his web site. He updates his data about twice a year. Professor Shiller’s database is the standard for studies.

Professor Shiller’s Web Site

You can determine today’s value of P/E10 by using the Stock Returns Predictor. Use the “Stock Returns” button on the left. If you know the current level of the S&P500, you can determine the current value of P/E10.

Unfortunately, the display does not let you enter the S&P500 value directly. You have to select a value of P/E10 to try (you can use the slider or you can type in a value) and then press calculate. Display will then show the current S&P500 value for that particular value of P/E10.

Today (August 1, 2008), P/E10 is between 22 and 23.

Each week, I update the reference value of P/E10. This is what you see when you first bring up the Stock Returns Predictor.

The actual value of P/E10 drifts slowly between updates from Professor Shiller’s site. This is because smoothed earnings (the E10 portion of the equation) generally rise over time. I generally check his site for updates about twice a month.

A search of P/E10 finds far too many references to be helpful. If you need just one, I recommend “The Logical Sequence.”

The Logical Sequence

Letters to the Editor in 2008

Letters to the Editor in 2007

Letters to the Editor in 2006

Letters to the Editor in 2005