Market Timing: Should You Attempt It?

» Kirk - Winning on the zigs, losing on the zags

For the last 19 years the average return for the
average investor was 2.6% per year
inflation was 3.1% and
the S&P500 was 12.2%.

Winning on the zigs, losing on the zags
INVESTORS WHO TRY TO TIME THE MARKET USUALLY LEARN THAT IT'S A GAME FOR PROS
By Todd Mason
Knight Ridder

William E. Donoghue is spoiling for a debate on the merits of market timing.

To ignore timing means ``growth funds that don't grow, value funds that don't add value, and managers who don't manage,'' said Donoghue, an investment newsletter writer and money manager in Natick, Mass.

His comments come at a time when New York Attorney General Eliot Spitzer is talking tough about big investors who use market timing to trade mutual fund shares.

Timing involves placing short-term bets on the direction of the stock market, or parts of it. It represents the opposite of a long-term strategy of picking an investment and sticking with it. Fund companies serving buy-and-hold investors often ban market timers because frequent trading translates into higher expenses and taxes.

Citing examples of alleged improprieties at four mutual funds, Spitzer's Sept. 3 complaint rocked an industry built on leveling the playing field for ordinary investors.

Broad accusations aside, market timing isn't unethical or illegal if fund companies publicly embrace the practice, and many do.

Timing is getting a new look from investors after three years of stock-market losses. For example, assets grew 50 percent to $9 billion so far this year at Rydex Funds, a Rockville, Md., fund company catering to timers.

Noted economist and author Peter Bernstein chided institutional money managers earlier this year for not being more opportunistic (read: timing) in an era of stingy investment returns.

But if timing is more acceptable, is it wise? Experts say ``no.''

``The fly in the ointment is execution,'' said Jeffrey Ptak, a Morningstar mutual fund analyst. ``It requires an enormous amount of skill that the vast majority of investors don't have.''

Even the pros lack the smarts and the temperament to beat the market long term, said Mark Hulbert, who rates the performance of market strategists who write investment newsletters.

``Over long, long periods of time, 10 years or more, you're going to find that 80 percent fail and 20 percent succeed,'' said Hulbert, editor of the Hulbert Financial Digest, in Annandale, Va.

Market timers move in and out of the market frequently, betting that they can make money on the market's zigs and zags. These bets stay in place for just days, weeks or even months.

By contrast, buy-and-hold investors, measuring progress in decades, are content to reap the overall gain in the market generated by a growing economy, and increases in corporate earnings and dividends.

Buy-and-hold investors should seek out mutual funds with a low turnover rate, or percentage of its assets bought or sold in a given year. A turnover rate of 30 percent or less indicates a buy-and-hold strategy.

Spitzer is zeroing in on fund companies that say they bar timers in their disclosure documents but do something else in practice.

His complaint alleges that four fund companies, Bank One, Bank of America, Janus and Strong, struck covert deals to aid institutional investors with timing trades that were supposed to be barred.

To be sure, individual investors can be fickle as well. According to an annual survey by Dalbar Inc., a Boston financial services consultancy, the average investor holds a mutual fund for 2.5 years.

``They buy a couple of weeks after the market spikes up and they sell a couple weeks after the market spikes down,'' said Heather Hopkins, a Dalbar vice president. ``They make very poor decisions in timing the market.''

There is an appalling level of self-inflicted damage among investors in Dalbar's estimates of individual investment returns. The study calculates holding periods by extrapolating them from flows of money into and out of mutual funds, and applies them to historical returns.

Dalbar estimates that the average investor earned 2.57 percent a year between 1984 and 2002, or roughly a fifth of the 12.22 percent annual gain of the Standard & Poor's 500 index.

Some timers also have trouble breaking away from the crowd, said Chuck Tennes, portfolio director at Rydex, although timers do better when they focus on trends taking place over months rather than days.

``I don't often see people making money trading rapidly,'' he said. ``Where do you draw the line on timing that's foolish and damaging?''

Donoghue's strategy is to rotate among various industry groups, or market sectors, using mutual funds or exchange-traded funds, which are effectively index funds that can be bought and sold like stocks.

Hulbert, the newsletter analyst, estimates that Donoghue's subscribers would have earned 1.8 percent a year since Jan. 1, 2000, by following the three strategies in the DonoghuePlan ActionGram. In the same years, the Wilshire 5000, a broad measure of the stock market, lost 7.8 percent a year.

Meanwhile, in Morningstar's rankings of separate-account managers, the performance of Donoghue's sector-timing fund at 4.5 percent so far this year trails most of his peers. Separate accounts are a personalized form of mutual fund.

Kirk’s Recommended Books

I STRONGLY recommend everyone read these two books first before you spend ANY money on newsletters, financial plans, mutual funds, etc.

• The Lazy Person's Guide to Investing by Paul Farrell
  “A simple and easy to understand book that offers many well diversified, indexed portfolios and data to show how well they work.”
  ----- Kirk Lindstrom, editor Investing - Personal Finance at Suite101.com
  ----- Read More Critical Praise

• The Four Pillars of Investing : Lessons for Building a Winning Portfolio by William J. Bernstein
  “A Triumph! It is my candidate for the best investment book of 2002.”
  ---- John Bogle, founder of the Vanguard Group.
  

-- posted by Kirk

» Normxxx - Market Timing on the QT

1. I swear by Portfolio Management Theory (although, at the current state of the art, the more complex forms need very active management).

2. I am a market timer who adjusts class allocations dynamically, but not on "gut feelings," but according to my models which attempt to identify which classes are performing best and which classes can be expected to perform best in the near future (months to years).

3. The very active dynamic modelling I am now using requires quarterly reallocation, although some of the reallocations do not occur on neat quarterly bounderies.

I think that people should understand that if they change allocation percentages on the fly according to no model, then they are doing the "bad" kind of timing that leads to 2.6% returns or less. At the least, they are at partially defeating the strengths of Portfolio management.

-- posted by Normxxx

» Kirk - Re: Market Timing on the QT

I am a market timer who adjusts class allocations dynamically, but not on "gut feelings," but according to my models which attempt to identify which classes are performing best and which classes can be expected to perform best in the near future (months to years).

I've sure enjoyed your contributions here Normxxx. I just want to say thanks!

I think one of the simpler models of dynamically adjusting allocation is the one Dr. Ed Yardeni uses based on the Fed Model.

The Fed Model

I actually track this Fed Model myself as the data from others doesn't seem to be so easy to find. I'd sure like to find an updated chart from Yardeni as I just have raw data and processed numbers for my work whereas a chart is best for showing others.

I also haven't allowed my asset allocation to change much more than 15% on a monthly basis from the peak of the bull to the depths of the bear. The key for me is the asset allocation model for me has me adding stocks when the market is down and selling some as it goes up. I have some "hysteresis" in my allocation which allows winners to run and bear to wreck some damage before I get too busy buying. You might say this 15% range is "the art part" whereas my target asset allocation is based on Modern Portfolio Theory (MPT) which suggests the added return for being over 80% in equities doesn't justify the added volatility.

I haven't tried to identify which asset classes will outperform as you've done. I find it too hard to become "expert" in all the market sectors. I prefer to pick a market sector I have training and experience in that is far too difficult for most to really understand the science behind: semiconductors and communications.
The division of HP that I worked at in R&D designing new products and businesses for 20 yrs was called "Optical Communication Division" which was later changed to "Communication Semiconductor Solutions Division" when they started to grow.

So far, this has all helped me pick tech companies that have greatly outperformed the Nasdaq, S&P500 and DJIA. They are also highly volatile which means I get many opportunites to sell some high and buy some when low. Over the years, it has made for impressive returns in both my personal and newsletter portfolios. What I find really interesting is the compound annual return I measure for my personal portfolio since 1991 is fairly close to the 5 yr number for my newsletter. It makes me wonder if it is coincidence or if my style and sector yield that return?

  Date       Kirk    S&P500   Delta
2003 YTD    +58.4%   20.7%    37.7%  as of 10/14/2003 
 

                                Kirk   S&P500+  NASDAQ 
4.75+ Yrs 12/31/98 to 10/10/03 141.0%  ( 9.2%) (11.8%)
Annualized Annual Return        20.2%  ( 2.0%) ( 2.6%) 
 

9/30/98 Inception Value:   $100,000
9/30/03 5 yrs later value: $345,395 up 245.4%
S&P500 between 9/30/98 and 9/30/03: up   4.6%
With dividends reinvested. All Numbers unaudited. 
Click for a free issue of my newsletter .

-- posted by Kirk

» Normxxx - Re: Re: Market Timing on the QT

Thanks for the thanks.

I can't say that a good market timing model improves the results of a well balanced portfolio over a long period of time, but it does seem better at letting you sleep nights. As I said elsewhere, I have been out of the market (except for a small number of shares) since May 19, and so missed out on much of the summer fun. But on the other hand, I was out all last summer until October, and so missed out on that peak adrenaline time. Also, my REITs, gold, and emerging markets have done very well, so I can't complain.

You are right about having to become (at least semi-)expert in each of the classes. You have to know enough to find several advisors for each class who are not feeding you some version of their private truth or worse, and you must constantly monitor that these gurus do not develop some fixed idea, so that they warp while the asset class woofs. It seems to be only natural, from time to time, that all gurus and systems fail. You have to be conversant enough with the class (and have some "circuit breakers") so that you are not dragged down with them, but not so sensitive that you get out just before the system kicks in again. That's why you need several gurus. Don't expect them to agree, but when they all don't agree with each other or they all do agree with each other, that should alert you to examine what is happening closer and look for other info sources.

I am familiar with Ed Yardeni's Fed model and, in fact, use a version of it to determine overall stock and bond risks. However, there are problems with the model.

1. There has been a lot of discussion of whose or which earnings to use. Before about 1996, operational earnings were about the same as GAAP earnings, so do we use estimated GAAP or estimated operational earnings going forward?. Maybe we should use trailing earnings, since Wall Street's estimated earnings seems to have about a 15% growth bias built in (which, like the stopped clock, is sometimes, but rarely correct).

2. You certainly can't use it for timing (which is why I use it in determining asset class risk); it has large periods when it is out of step with the market. It is really a warning of when stocks or bonds, as a class, are getting overbought or oversold (relative to each other), in the long term. But you already knew that.

3. Models do not necessarily work if some of their input parameters are artificially constrained or if they are at the extremes. Currently, you can argue that our interest rates are being held artificially low (not just by the Fed, but by China, Japan, etc.) and, in fact, are abnormally low on a real basis (currently, the real short term interest rate is about a negative 1% or so (shades of Japan).

4.1 The Fed model works well in a world where recessions are caused by the Fed shutting down demand through raising interest rates because of the economy overheatng. When rates are subsequently lowered, pent up demand in housing, autos, and CAPEX zoom. It is hard to see much pent up demand in housing and autos, which have gone gangbusters throughout this downturn. Or in CAPEX, where capacity utilization is still hovering around 75%. INTEL just came out with a relatively glowing report (although, if you read the fine print, you will see they are not so certain of 2004). Yet INTEL just cut their CAPEX budget. And several of the other biggies have spent less than 50% of their 2003 IT budgets. (Unlike the government, I don't see them scrambling to spend the rest in the last 3 months of the year.)

4.2 Currently, we are in a(n over)supply side recessionary environment (which is why deflation is a threat and the job market is so poor). Our last experience with such an environment was in the '30s, and we didn't cope very well. (Nor has Japan in the '90s.) Although currently we seem to be doing much better, if the economy doesn't kick start here, we are in serious trouble. We are out of both fiscal and monetary stimuli (regardless of what Fed governor Bernanke says). The acid test will be in 1Q 2004 when the echos of the current stimuli have faded away.

5. Earnings are being (artificially?) maintained by drastic cost cutting; so far, there has been little improvement in revenues.

6. The model may be saying that bonds are way overpriced (which they probably are, since interest rates are being held artificially low).

I believe that the market is doing a repeat of late 1999 and for the same reason: a vast (over?) supply of money and credit which has to find a home. It will tank again in 2004 if it looks like a double dip, or if the financial system otherwise locks up. At this point, I think a repeat of 9/11 would be much less of a stock market event, except to let off a lot of the pressure. If it happens before next year, or if next year looks good (we muddle through once more), then I think it would be a great buying opportunity. But if it looks like a double dip, look out below, and I don't think it will wait for May to crash (as it didn't in 2000).

I have much more to say on this, but this is more than enough for one message!

WHAT'S WRONG WITH THIS PICTURE?

-- posted by Normxxx

» Normxxx - Re: Re: Market Timing on the QT

I, too, am a retreaded "techie." I started in electronics engineering (analog computers) in the '50s, switched to digital and software in the '60s (and was a "hybrid" computer engineer for a time), and retired in February of this year (finally ran out of steam). So, actually, I am an ex? software guru (which is to say, I knew most of the software luminaries from before they lit up). I also took about 4 years timeout along the way to get a Ph.D. in Psychology (don't ask me why, but the whole world, and especially the world of software engineering, seemed increasingly nuts). Since it was in Mathematical Psychology, it has proved of some value subsequently.

Anyways, for the tech sector (since I am no longer current), my chief guru is Fred Hickey. Do you know anything about him and/or what is your opinion? He seems to be universally admired, if not much listened to (he is really bearish!) I have been getting his letter for about 2 years and he sounds like he knows whereof he speaks.

-- posted by Normxxx

» Kirk - Re: Re: Re: Market Timing on the QT

I don't follow Fred Hickey and have never read his newsletter. What is it called? If he is large, then he should be followed by Hulbert which will give you the real truth about his abilities.

-- posted by Kirk

» Normxxx - Re: Re: Re: Re: Market Timing on the QT

His newsletter is not his primary source of income. His primary source of income is investing his own money (isn't that refreshing!) So he lives in NH and is semi-incommunicado (no advertising; no website or eMail address; no telephone, except to accept subscriptions. The newsletter is called, The High-Tech Strategist.

Since he has been solidly bearish from at least 1998 (I believe; I've only been getting his newsletter for 2 years), he is often quoted by Alan Abelson in Barron's. But others know of him as well, and he seems to have a good record over the last 10-15 years or more.

Business Week
As you can see, his bearish overall position doesn't stop him from playing the rallies.

Marc Fabor

-- posted by Normxxx

» Q_out - Re: Re: Market Timing on the QT

Perhaps this would help.
<img src="/files/mysites/qout/svm-1.gif" width=450 height=294>

<img src="/files/mysites/qout/bhoestarts.gif" width=53 height=34 align="left">
Q_out
DISCLAIMER: My words and observations are general in nature, and are not meant as specific investment advice. Individuals should consult with their own advisors for specific investment advice.

-- posted by Q_out

» Normxxx - Re: Re: Re: Market Timing on the QT

Where does he hide it? I couldn't find it on his site.

-- posted by Normxxx

» Kirk - Re: Re: Re: Market Timing on the QT

Great! Thanks!

The Fed Model

-- posted by Kirk