Discussion with Ed Easterling of Crestmont Research

Research on Safe Withdrawal Rates

Moderator: hocus2004

hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Discussion with Ed Easterling of Crestmont Research

Post by hocus » Wed Aug 06, 2003 1:41 am

This thread is for the Special Event Discussion with Ed Easterling of Crestmont Research scheduled to begin at 10:00 am Eastern time today and to continue until noon.

I am extremely pleased that Ed has agreed to meet with us today to discuss the research he has done regarding long-term stock returns. I became aware of the research he has done about a month ago, and found it most illuminating re a number of issues that have been explored during the 14-month Great Debate on SWRs. I know that those who participate in the event today will learn from Ed and I hope that perhaps this board community will be able to give back a little to him by sharing what we have learned as a result of our own research on some issues of mutual interest.

Here's a link to a John Mauldin article where the Easterling research is discussed, which includes links to the research.

http://frontlinethoughts.com/article.asp?id=mwo062703

Here is a link to a post that Ed (who posts here under the screen-name Crestmont) recently put to this board.

http://www.nofeeboards.com/boards/viewtopic.php?t=1210

I have asked JWR1945 to ask the first question. I expect that JWR1945 will be putting up his post roughly at 10:00 am Eastern time.

All visitors to the site are welcome to participate in the discussion that will follow from the JWR1945 post and the Crestmont response that proceeds from it.

hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus » Wed Aug 06, 2003 5:03 am

I have not heard from Ed Easterling this morning re the Special Event discussion. My sense is that he may have gotten pulled away on other business.

If I do not hear anything for another 10 minutes, I will temporarily cancel the event and see if we can reschedule for another time later today or later this week. Since the only known participants are Crestmont, JWR1945 and me, I will just choose a time that works for the three of us. But I will put up a notice here as soon as I know of a new time, and all others are of course welcome to participate.

Crestmont
* Rookie
Posts: 20
Joined: Thu Jul 24, 2003 4:00 am

Post by Crestmont » Wed Aug 06, 2003 5:05 am

Though my e-mails may not have gotten through to hocus, I am here and looking forward to JWR1945's first question.

hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus » Wed Aug 06, 2003 5:06 am

OK. I see that both Crestmont and JWR1945 are browsing the forum.

A warm welcome, Crestmont!

Why don't you shoot with your first question, JWR?

JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 » Wed Aug 06, 2003 5:08 am

Good morning and thanks. This is my question for Mr. Easterling:

Mr. Easterling, what if you're wrong?

You may choose how broadly or how narrowly to answer this question. In addition, you may choose: wrong about what?

I will leave my question as a fragment. Words that might go along with it include: what should I do? and how will I know that you are wrong? what is a good backup plan? Such additions are all optional.

Have fun.

John R.

JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 » Wed Aug 06, 2003 5:10 am

BTW, feel free to break this down into several smaller questions.

Have fun.

John R.

Crestmont
* Rookie
Posts: 20
Joined: Thu Jul 24, 2003 4:00 am

What if wrong?

Post by Crestmont » Wed Aug 06, 2003 5:17 am

JWR,

Great question and one that I reflect upon frequently. My research is historically based and creates a series of observations for your perspective. Generally it's not intended to be a prediction, rather it provides perspectives for the reader's interpretation. If I'm wrong in some areas, it's in reflecting a relationship between P/E's and inflation for example. Most of the relationships that are included are generally accepted. Actually, I hope the observations are wrong and that we can have another 20 years like the past 20.

But to you point about what if and what to do...seems that the data and current situation (relatively low interest rates and high stock valuations) encourage different approaches to investing than in generally directional markets.

In a recent posting, I included: "8. Techniques: seek advice from an independent financial advisor about techniques to deal with non-trending markets. The stock and bond bull markets of the 1980's and 1990's were trending markets and benefited from "buy-and-hold"Â￾ and "buy-on-dips"Â￾ strategies. The next phase could be another secular bear cycle (see http://www.crestmontresearch.com/pdfs/S ... 0Chart.pdf). Note the difference in secular bull and secular bear periods. Ask your advisor about the effect of more frequent rebalancing, covered call option writing, higher yielding preferred stocks and other securities with higher current return, TIPS, hedge funds, and other risk-controlled and value-added investment strategies. "

Pausing to send and check for other questions...more to follow

hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus » Wed Aug 06, 2003 5:20 am

Crestmont:

If JWR has follow-ups on his question, please respond to those first.

If you get to a point where you are free to handle a question from me, here's my first one.

Do you think it is possible to time the market and experience financial gain from doing so?

Crestmont
* Rookie
Posts: 20
Joined: Thu Jul 24, 2003 4:00 am

More to JWR

Post by Crestmont » Wed Aug 06, 2003 5:21 am

Continuing...

Let's dig for an example of "wrong"...does anyone have particular aspect that they see as more at risk of being wrong?

JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 » Wed Aug 06, 2003 5:23 am

Thanks for this initial response.

I admit that I have assumed that somebody would ask the question what if you are right?

I also realized that you had already given us some good leads along those lines. It is good that you have put them into this thread.

Don't let me restrict you in any way whatsoever.

Have fun.

John R.

JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 » Wed Aug 06, 2003 5:26 am

Crestmont

Let's dig for an example of "wrong"...does anyone have particular aspect that they see as more at risk of being wrong?

I have nothing in mind. I think that you are right.

It is just that there are always smart things to do if you are wrong and there are stupid things as well.

Thanks.

John R.

JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 » Wed Aug 06, 2003 5:29 am

Maybe it is good to focus on stupid things to avoid. You do not have to restrict yourself to my initial question. Just things to avoid in general.

Have fun.

John R.

hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus » Wed Aug 06, 2003 5:29 am

Does anyone have particular aspect that they see as more at risk of being wrong?

I'll offer a personal perspective here. I don't think it is possible to be "wrong" in what you are doing.

You are just putting forward information on what the data says. It could be that a conclusion you draw from the data turns out to be wrong, but that is not being "wrong" in the sense of doing something wrong.

You put forward the data, people grapple with it and hopefully learn from it. People need to understand that no one can predict the future. But information is power, and you are giving people information they did not possess before. You are empowering people. That's my take.

Crestmont
* Rookie
Posts: 20
Joined: Thu Jul 24, 2003 4:00 am

Market timing

Post by Crestmont » Wed Aug 06, 2003 5:30 am

hocus,

Most will say that it's not possible to 'time' the markets and I agree. There are several ways to view timing though. If timing intends to be for the weeks and months that it goes up and out when it goes down, I've yet to see anyone that can consistently do that. However, some reasonable techniques (more frequent rebalancing, for example) can be successful in using the volatility or directional trends of the market to an investor's favor.

The recent book by Ben Stein is an interesting read, however has a critical fallacy. Their strategy says that it works well to buy with the market is below the moving averages (yet stay invested otherwise). Not only is this somewhat of a compromise to the integrity of the arguement (i.e. wouldn't it be even better to exit the market when the signals are reversed?), but also it works because it reflects a fairly simple concept: in upwardly sloped cycles, it will always mathematicaly produce positive results to buy in the dip and ride the cycle wave up. It does not do a good job of addressing the impact of executing the same strategy over the past few years or though an extended flat choppy secular bear cycle.

hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus » Wed Aug 06, 2003 5:37 am

The recent book by Ben Stein is an interesting read, however has a critical fallacy.

That's a great response because that gives us something to sink our teeth into at some later date. Thanks for offering that observation.

Here' s my next one for the next time the decks are otherwise clear.

How long have you been doing research into long-term stock returns? What has been the biggest surprise finding for you? What sort of feedback have you gotten from people responding to the research?

Crestmont
* Rookie
Posts: 20
Joined: Thu Jul 24, 2003 4:00 am

Not so smart things...

Post by Crestmont » Wed Aug 06, 2003 5:45 am

One not so smart thing is to ignore the facts and the data. Or to create a financial plan with internal fallacies or inconsistencies.

It seems ironic that many investors seem to understand the realities of bonds, yet hold out irrational hope for stocks??

For example, with interest rates have been on a 20 year decline and near historic lows. As a result, bond investors have made solid capital gains in additon to the coupon. The total return on bonds for the past 20 years and longer display solid high single digit returns (depending upon the period). I find very few willing to assume that bonds will generate their historical average return from current levels. Even though bonds have fallen significantly over the past few months, there's still a low chance of historically average returns even ffrom these levels.

However, for stocks, similar conditions are true. The past 20 years have provided strong gains as P/E have risen to relatively high historical levels. Long-term returns are still benefiting from the past 20 years despite the papst few years (why? because P/E's are still above average levels). So how can we assume historically average returns from stocks from these levels for the relatively near term (next 10 years or more)?

Examples of assumption problems would be to include an inflatioon assumtion that was not consistent with interest rates or transaction costs that are too low.

Examples of structural problems would be to set up an investment structure that did not enable the return stream to adjust for inflation (since withdrawals and expenses are likely to track inflation).

JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 » Wed Aug 06, 2003 5:50 am

You have made an excellent point just before I have posted this. (We are all composing and responding out of order.) Going backwards a little bit:

From hocus:

You are just putting forward information on what the data says. It could be that a conclusion you draw from the data turns out to be wrong, but that is not being "wrong" in the sense of doing something wrong.

How would you recognize that something has changed, that the data is now telling us a different story?

For example, the market began to tell a different story when the bubble began.

Have fun.

John R.

hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus » Wed Aug 06, 2003 5:54 am

It seems ironic that many investors seem to understand the realities of bonds, yet hold out irrational hope for stocks

I think that's a great observation. I think that the problem with stocks is that people have had the idea beaten into their heads that stocks always provide the best return over the long term.

I don't think it is so, but I think that many people believe this. So there is a sense in which people just don't care to perform a rational analysis when it comes to stocks. If there is some magical rule that requires that stocks always provide the best return in the long-term, why bother? If that were so, you would not need to look at data to know what investment choice were best.

hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus » Wed Aug 06, 2003 5:57 am

We are all composing and responding out of order.

Yes, but we're all doing great. It's a trip. We're getting lots of good stuff out on the table in s ahort amount of time.

It's alive!

Crestmont
* Rookie
Posts: 20
Joined: Thu Jul 24, 2003 4:00 am

How long, surprises, and feedback

Post by Crestmont » Wed Aug 06, 2003 5:58 am

Although I've done some limited work over the past 20 years, the concentrated effort started 2 years ago. At first, it focused on assessing the stock market and then expanded from there. The biggest surprise (insight) has been to see the historical cycles and to find that some commonly accepted axioms are not valid in all conditions.

For example, interest rates and inflation....a generally non-connected relationship before the 1960's.

The feedback has been very positive and has also provided additional avenues to pursue. The 100x100 year stock market matrix can from comment from one of our investor clients. We were talking about the fallacy of using Ibottson's long term return number because it has entry point risk (the starting point significanlty affects the long term return). So I generated a series of analyses showing returns from 1990, 1980, 1970, 1960, 1950, 1940, 1930, an 1920 and also showed aggregated results.

To which the client responded, "what if there's something unique about the start of decades...don't you still have starting point risk?!" As a result, the 100x100 chart was developed to display every starting year and every subsequent ending year since 1900. Colors and other data were added to elevate it from data to information. That chart has generated a lot of positive feedback. It's available in several versions and formats at www.CrestmontResearch.com.

Post Reply