July COP Report                                                           8/11/08

Investment Gang

Appended below is a note I sent out earlier this week by email.

I have updated the COP Research Model to allow the model to invest in short asset classes and to report right up to the current month including what the model will be invested in for the next month, in this case for August.  I have also attached a spread sheet containing two case studies for this month (the individual investments are only shown for 2007 and 2008YTD).  We will have to wait until the end of the month to see if the model made good choices for August or not.  Click “Trend Report(-1)”  to see this spread sheet.  See the end of this letter for a detailed explanation of the sheet.

As for the performance reports that I actually use for investing there does not seem to be much worth investing in.  Health Care, Turkey and the Mexican Peso look good, however, they are each one-horse shows in their respective arenas.  That strikes me as very risky.  There are several shorts that look good; however, shorts are notoriously volatile and hard to catch.  One reason is explained in the appended note.

Here I am going to insert some comments that I found very interesting.  I clipped these from Prieur du Plessis’s “Postcards from Cape Town” without any permission.  I have recommended his web site before and repeat that recommendation.

David Rosenberg (Merrill Lynch): Up-days of 300 points are typical of bear markets
“The Dow surged 331 points – moves like that typically take place in bear markets. There have been six of these sessions since the bear market began a year ago; there were absolutely none in the 2002-07 bull market. Keep in mind that there have been 10 days this down-cycle when the Dow closed down 300 points or more. In other words, volatility gains momentum in bear markets.

“Go back to the 2000-02 bear market and you will see much the same thing – there were actually nearly twice as many 300 up-days in that down-phase as there were 300-point losses! And this was in the context of a 38% peak-to trough decline in the DJIA. So, think of what happened yesterday as a characteristic of a bear market, even if the shorts got painfully squeezed – that should not be lost on investors as we go through this oversold technical rally (that has seen the financials soar 30% from the mid-July lows).”

Source: David Rosenberg, Merrill Lynch, August 6, 2008.

Richard Russell (Dow Theory Letters): Dow is running at a net loss
“Some items to be noted. According to this week’s Barron’s, earnings on the Dow are now a negative 81.34. The Dow is running at a net loss! This means that the price/earnings ratio for the Dow is infinity. Nothing times nothing = nothing.

GaveKal: Standing in way of improving dollar is likely to be painful
“As more investors realize that the US economy may not be in such dire straits after all, we expect the current positive momentum of the US$ to persist. This is all the more true since the US$ remains very cheap on a purchasing parity basis, especially against the euro, and that the US trade balance is now improving rapidly. Standing in the way of this US$ is likely to be very painful …”

Source: GaveKal – Checking the Boxes, August 8, 2008.

David Fuller (Fullermoney): Chinese slowdown mild compared to Western economies
“Inevitably China’s GDP growth is slowing following measures introduced early last year to curb economic overheating, and to deflate bubbles in the housing and stock markets. Additionally, most of China’s exporters can only experience a downturn because of weak demand from the West.

“I suspect that China’s growth will slow more than most analysts are currently forecasting, due more to the factors just mentioned than a post-Olympics drop in Beijing’s construction projects. However economic activity anywhere near the Olympic village is already on hold and I assume that there will be a pan-China pause during the games, given their importance to the regime.

“A question for investors: Are these factors behind the slowdown reasons for concern or part of China’s success story?

“We have seen China warnings before, and they have been largely wrong. That record will not prevent another wave of bearish articles and reports, now that China’s GDP really is slowing, and they will also partly represent a projection of our own economic problems in OECD countries.

“However there is a vital difference between China’s slowdown and what we are experiencing in the West. China’s economic growth is moderating, largely by government choice, but it is still likely to be a world leader. China has a sound banking system, a massive current account surplus and high personal savings rate.

“Contrast that with the USA, UK and some Continental European economies. Banks are reeling due to reckless speculation and pitiful regulation. Economic growth hovers near recession, both personal and government debt levels are high.

“Taking a long-term view, I know where I would rather invest. China’s superior economic prospects will not prevent the stock market from remaining high beta, but it has already fallen a long way. Moreover the Chinese government is now switching its emphasis from curbing economic overheating, and deflating stock market and property bubbles, to boosting economic growth once again. I think they will succeed.”

 

Research Model Expanded/Updated

 

I have just completed an expansion and update of my COP Research Model (referenced on copstrat.com).  There are two added functionalities. 

First, the model can now be updated at the end of every month with the most current market results and it will, therefore tell you how the model would invest next month.  Now, I do not use the model for investments as it has only 50 asset classes and makes selections for me.  I work off of a larger report each month and make selections that suit my risk tolerance.  However, I thought it will be interesting to compare since the model has a very good track record – better than mine in a several years.

Second, since there are now a great number of short selling ETFs, I thought it would be interesting to see how the model would have behaved if it also had short asset classes to chose from over the last 19 years.  So, from the 50 asset classes in the model I created 50 short asset classes.  Now the model has 100 asset classes to choose from.  Surely, given more choices and the ability to have sold short during the many down markets over the last 19 years the model could improve its performance.

Guess again.  I have more test cases to run, however, on the base cases the performance results are poorer if you give the model the chance to short.  There are four years in which it does better (1990, 1998, 2001 and 2002) however, over all 19 years the performance is reduced.  Using short classes, it has not done better in 2007 or so far in 2008.   I am not surprised since I believe market downdrafts to be abrupt and hard to catch.  Looking at the details that is exactly what happened.  The model saw a short class do well for a month and jumped in only to be caught by a rebound month.

In the research model there is a graph in which each of the asset class months are color coded.  Orange means that the asset class was a high performer last month, the basis on which you chose to invest, and a bummer this month.  In down markets there are a lot of orange cells, especially in the short asset classes.  If you are following the performance of hedge funds during the last nine to ten months, you have seen that this story has applied to them as well.  The Wall Street Journal reported this week (I did not retain the article) that hedge funds have experience their worst ever quarter – or was it two quarters.

So far in 2008 (to the end of July), the model is down about 5% or so (depending on the study case you look at) which beats the S&P which was down about 12% at the end of July.  In April the model invested in eight shorts – all went orange.

I am glad that I only follow the strategy when asset classes are performing within my risk tolerance.  I am currently positive for the year (just barely) since I am mostly in cash and low risk assets such as currency funds.  Actually, my winnings have had to compensate for one fund that I kept, for tax reasons and I had confidence in the fund, yet it is down 20%.  Never trust – stay with the program. 

When I get more time to work with the model, maybe there will be more gems.  In a few days I will report what the model says to invest in for August.

Meanwhile my strategy paper on the web site is much in need of an update.  Probably the most important update is that I have learned a lot more about the benefits and methods of timing the market.  In the paper I acknowledged all of the street wisdom against timing and just said that I did it anyway.  Now I am surer about why and how.

Best, Mal Williams

Explanation of “COP Model to Date” spread sheet.

The multicolor table shows with color coding (see color legend on the right) the investments made in 2007 and 2008 YTD and for August. The table below that shows a year by year summary of the last 18 years with the 18 year overall average on the extreme left.  Note that the model not only beats the S&P 500 (16.63% versus 9.91%) but also does so at a lower standard deviation (11.51% to 13.53%).  Here you can see what years it did better and where it did poorer.  Note that the model did poorer in 1994 through 1998.  The reason is that the S&P was on a roll as the best asset class in the world for that period so it was hard to beat.  If you were an individual investor in that period I am sure you would have over-weighted the S&P rather than allowing only 1/15^th to be invested in the S&P as the model does. 

Note the large number of orange cells, good classes turning bad, starting in November of 2007 and the large number of yellow, new winners emerging.  This clearly shows the sea change the markets went through in this period and the high level of volatility and uncertainty.  Note the high concentrations again in April and June of 2008.  Any doubt why you had a hard time making money? Now note the large lengths of stable green in the earlier part of 2007. 

Suppose you had the ability to know ahead of time to invest in the yellow.  For that look at the returns in the fourth line down (in lavender) for Annual Return for Y&G.  It’s fun to dream.  The line with “H=1.5” in it is for the case that you took until the middle of the month to change your investment selection.  For these 50 asset classes the Morningstar data is available (for free on their web site) late the night of the last day of each month, so H=1 investing is not impossible. The line with “H=1 Minus S&P) shows the difference in the models return and the S&P.  I chose to use H=1.5 investing because (see the chart) the standard deviation is lower and it gives me more choices of investment classes and diversity.

Now look at the second set of two charts in which the research model was allowed to choose among the same 50 asset classes as above plus 50 shorts of the same asset classes.  You will see that overall performance is degraded and only a few down market years are marginally improved as noted above.