It took almost 4 years for us to get here. We have already seen corrections in the major averages of 5% or more so the 4 year cycle is not the issue.
The issue is how deep will the correction be?
Is it already over?
Sentiment has changed dramatically.
But as negative as sentiment has become it is not yet as negative as it has been at past 4 year presidential cycle lows.
In addition macro economic trends are showing a slowing economy but the FED is fighting inflation rather than easing.
Check out this informative chart:
martincapital.com
COMPOSITE INDICATOR REVISION 26-Jan-01 The latest revision to the composite indicator involves the addition of a new component, the difference between the 30-year T-bond yield and the 91-day T-bill rate. When the difference is negative and the T-bill rate is higher than the T-bond yield, the ends of the Treasury yield curve are said to be "inverted." Although this condition has not occurred many times in the past 40 years, it has been closely associated with deteriorating economic conditions and concurrent stock market declines. This component only has a negative contribution to the composite indicator.
I put it together my model by looking at conditions of inflation, interest rates, Fed moves, breadth, dividends, earnings, etc. over the last 40 years. Those were the things that typically led the stock market in a general way. The model is obviously out of synch at present, particularly with the NASDAQ/high-tech sector. This disconnect implies, to me at least, that things are different this time from most of the past 40 years. Seeing the stock market continue to fall while interest rates are also continuing to fall is not at all typical of the past four decades. Interest rates and inflation have clearly dominated the stock market in the past 40 years, while earnings have had less of an intermediate-term effect. Consequently, my model reacts more to interest rates than to earnings. The reverse is true today: earnings are having much more of an impact on the stock market than inflation and interest rates. What happened this time around was that the stock market, particularly the NASDAQ, was priced for perfection that would continue forever, in accordance with the two assumptions above. We are now seeing an adjustment in stock prices and valuations that is the result of taking earnings for the past and immediate future quarters back into consideration as the primary tool of valuation. As usual, the market is overcompensating and the adjustment is going too far.
Changes in the level of this indicator that forecast major moves in the stock market are:
1. A decline below zero means little chance of a stock market advance.
2. A rise from below zero to plus 3 means a stock market recovery is likely.
3. A drop of 12 points within 8 weeks means that the stock market is vulnerable to steep declines.
4. A rise of 12 points within 8 weeks means a stock market rise is likely.
5. A drop to a level below minus 6 means sell all stocks or hedge and take a vacation.
The three buttons below link to charts that show the trades that would have taken place between 1961 and early 2001, using the present set of parameters. They do not portray trades actually made at the times indicated because I have only been working on this since 1991 and because revisions have been made to the composite indicator since then. Looking at the charts in detail, normal, long entries into the market are marked by green up-arrows with the numeral "1" below them. Short entries into the market are marked by red down-arrows with a "-1" above them. Exits from the market are marked by blue up or down arrows with flat ends and a zero above or below.
A long series of studies has shown me that the business cycle with its associated fluctuations of inflation and interest rates seems to be the primary controlling factor for the stock market. Developing the composite indicator of stock market conditions involved more than five years of work in analyzing 50 years of weekly historical data. The long period of data has been used in order to include a wide variety of economic and market conditions. The indicator is based on 13 separate components that tend to lead the stock market or to reflect in some way on its strength or weakness. The 13 components that make up the indicator use data on short and long interest rates, several commodity price indexes, bond prices, earnings and dividends. I also include components based on market breadth and new highs vs. new lows. I am experimenting with monthly economic data, although my experience with that data so far is that it comes out late and is subject to even later revision. The indicator is in a continuing process of evolution and, hopefully, improvement. The last revision prior to January 2001 was made in May 1997.
I take a different point of view of the stock market from most people. Most private and professional investors seem to view the market as a horse race and try to pick the fastest runners. I view it instead as a big bunch of horses that moves as a herd. As I see it, the forces that most actively move the herd are inflation, interest rates, valuations, dividends, earnings and emotion. I have created my composite indicator of stock market conditions in an attempt to quantify some of these forces and thus to anticipate broad moves in the market. I hasten to acknowledge that no indicator can ever include all of the complex forces that move the stock market up and down, so I do not expect to pick every market top and bottom. However, I believe that this indicator will forecast most of the big moves in the market and thus outperform the average horse in the herd over a long period of time.
Please understand that my comments here about my own particular approach are not meant to discredit those investors, private and professional, who have used other techniques to outperform the market as a whole. Nor is this the primary method used by Martin Capital. There are many ways to excel in the financial markets and I see this particular method of market timing as just one of them. Warren Buffett has made more money in the market than I ever hope to make by using his own completely different methods. Graham and Dodd wrote down ideas on value investing decades ago that have helped many to excel in picking stocks. In contrast, my ideas are new and relatively untested.
The overall performance of the indicator, as far as being in line with major market trends since 1961, is excellent. Two specific periods bother me, however: the market declines of 1962 and 1977. The indicator recommended staying in the market during these substantial declines. These instances inspire caution, as this type of experience could be repeated in the future. I will continue to work on improvements and not expect to be right every time.
I must give credit and thanks to those whose ideas have been the most valuable in the development of the composite indicator and the evolution of my newsletters. Martin Pring on the subject of the business cycle, John Murphy on intermarket technical analysis and Norm Freeburg's newsletter, Formula Research, have all been major contributors. If you as a reader get anything useful out of reading this material, you owe thanks to all of them just as I do.
COMMENTS from my newsletter of 29-Oct-01 on why the composite indicator was out of step:
The indicator rose from below zero in June 2000 to plus 29 just before the terrorist attacks, a large increase based on improvements in conditions like those listed above that have historically been good for the stock market. At no other time in the past 40 years has such an increase in the value of the indicator been accompanied by a steep decline in the stock market. Likewise, at no other time in that period have a set of conditions including falling interest rates, low inflation, easy monetary policy, etc., all largely beneficial to stock prices, accompanied any severe stock market decline.
All these beneficial conditions being the case, how did we get such a severe decline in the stock market? What was different this time around? Hindsight is a wonderful thing, so we can look back now and see several elements developing into a stealth perfect storm. Near the end of the longest economic expansion in history in late 1999, the Fed had already raised rates in three quarter-point hikes between July and November 1999. Many have forgotten now, but most people were worried about the possibility of a worldwide crisis occurring as 1999 rolled over into 2000, fearing that many computers wouldn’t be able to handle the change in dates and would lock up. People stockpiled water, food and cash in order to be prepared for a shutdown of many systems. The Fed, in order to help avoid any financial disaster, boosted the money supply. This took place while GDP growth was hitting 8.3% in the last quarter of 1999. After all that, Y2K turned out to be a non-event. Having briefly hit the accelerator hard to dodge the Y2K bullet, and having seen the overly rapid growth rate, the Fed put on the brakes with a vengeance. The culmination of Fed tightening was the half-point rate hike in May 2000, done after it was clear that the economy had already begun to slow. Evidence for the slowing can be seen in the drop in the North American Purchasing Managers Index in the bottom chart above. Other evidence was also already available by April 2000 that GDP growth had slowed substantially.
The Fed and its actions were by no means the whole cause of the debacle. During late 1999, in a phenomenon directly related to Y2K, production and sales of computers and software surged, as many businesses and individuals bought new equipment out of fear of being shut down after the first of the year 2000. The high rate of sales pushed the total business inventories to sales ratio down to 1.36 in January 2001, the lowest since 1979. With less on the shelf to sell relative to what was going out the door, even in the face of what appeared to be a minor slowdown, manufacturers were stimulated to produce more. At the same time, other typical late-cycle phenomena were taking place. Inventory build-up and overproduction occurred in many industries, particularly communications. For instance, excessive amounts of fiber optic cable were put in place in anticipation of rapid growth in broadband communication.
It became apparent several months into the year 2000 that sales of high-tech or information technology equipment were not continuing to be as strong as had been expected. It was not clear at first whether this slowdown in sales was just part of a general slowdown in the economy engineered by the Fed or whether it was due to the extraordinarily high rate of sales in late 1999 having captured sales from the next year. In time it became clear that it was due to both causes and the latter had far more of an impact.
Then we had the presidential election fiasco. Following the national elections in early November 2000, we had a 37-day hiatus when we didn’t know who was going to be our president. By December, consumer confidence had dropped significantly, further undermining a weakening economy. The financial markets hate uncertainty, so the stock market dropped further.
The Fed put the icing on this bitter cake by waiting until January 2001 to begin cutting interest rates. The NAPM Index had already showed December to be the fifth straight month of contraction in the manufacturing sector -- not just slower growth, but five months of actual contraction. Though more rate cuts followed quickly, the economic storm was under way, generated by all of the elements cited above. The terrorist attacks of September 11th will hopefully be the last important event in this negative series.
Looking back over the sequence of events and conditions that led to the stock marker decline, some were normal or typical for an economic cycle, some “normal” conditions were not present and some unique events and conditions were present. It was normal for the Fed to begin raising rates late in an economic cycle when the economy began to grow at faster than a sustainable rate and the unemployment rate was at a historic low. Compared to all other cycles since 1960, it was not normal for there to be so little inflation at that point, other than in oil prices. It was not normal for long interest rates to stay relatively low at that stage, but that tends to go along with the low inflation rate. The Y2K, high-tech sales issues and presidential election confusion were not at all typical. The terrorist attacks were certainly not normal or typical.
The failure of the composite indicator to anticipate the stock market’s decline since mid-2000 occurred because it was not sensitive to some of the conditions listed above that had a big impact on the stock market. I have always said that a huge number of factors impact the stock market as a whole and that the best one can do is to take some of the most influential ones into consideration. This time, the indicator missed some. Furthermore, it is likely that it will fail to sense other types of unusual, negative conditions in the future. Nevertheless, when this indicator turns negative with increasing inflation and rising interest rates near the end of normal economic cycles, the chances are strong that it will anticipate a falling stock market.
email Alston Boyd, Economic Director |
