Tabell’s Market Letter – February 13, 1981

TABELL'S MARK.ET. LETTER 909 STATE ROAD, PRINCETON. NEW JERSEY 08540 DIVISION OF MEMBER NEW YORK STOCK EXCHANO. INC MEMBER AMERICAN STOCI( eXCHANGE February 13, 1981 Anyone who has spent some time in the securities industry is used to being greeted with the ques- – – tion,. Whatith!nn.ark-!ioing–BaQk illthe d!!rl5 'ages ;hen. we-Jirst, enterectthe 'profession-, the- question had a very specific meaning. It basically asked for the change in the Dow-Jones Industrial Average. In those simpler days, before the invention of the computer, and prior to the invasion of Wall Street by refugees from mathematics departments, the Dow-Jones Industrial Average was just about all there was. It dated back to 1884, when Charles Dow first compiled a list of 12 (later 20, then 30) stocks and utilized their average price as a market indicator. Mr. Dow's original average has subsequently been severely criticized on a number of grounds involving, among others, the inclusion of only 30 stocks, the choice of those stocks, the relative weighting method, and the unscientific method of adjusting for splits. Other market indicators have emerged, including the capitalization-weighted SliP 500 and NYSE indices, along with a host of others. The advent of the computer, and the fact that almost all securities trades are now on line to EDP equipment, has made possible the calculation of averages and indices in limitless variation. No longer restricted to the relatively few stocks that Dow and his staff could calculate by hand, we now have averages comprising individual prices of in excess of 5000 securities. What Dow was doing when he first calculated his average, of course, was following a statistical procedure that goes back into the dimmest reaches of mathematical antiquity. The creation of an average was then, and remains, an attempt to describe a disparate group of numbers bymeans of a single number. This remains true today, regardless of how many stocks an average may include or how sophisticated its calculation may be. It remains a reference to a of quantities and is a form of mathematical shorthand, an attempt to pinpoint one property oTUiiit particular group. Only a single property, however, is described. Consider these two groups 102, 96, 100, 104, 98, and; 10, 350, 5, 132, 3. One does not need to be a mathemetician to realize that the two groups are very different. Yet they have one attribute in common. The average value of each is 100. –fn-OIder-toquantify-the-diHerenee-bet-ween-the-secGndgroupandthefirst.we-needLto-examine– -. another property of groups of numbers — variability. What obviously distinguishes the second group . above is that it is a good deal more variable. This same concept of variability applies to changes in averages over time. Consider an average of three stocks, each selling at 100. If one moves to 104, another to 102, and the third stays the same, their average would have moved to 102. If one doubles and another declines to 6, with the third remaining the same, the new average will also be 102. The dimension within which the component stocks have varied over time, however, is quite different. Which brings us back to Dow. Even in the 1880's, Mr. Dow was smart enough to realize that if stocks varied widely in disparate directions, the concept of an average would be meaningless. He knew intuitively that such was not the case. Stocks tend to vary in harmony with each other, and, therefore, as Dow well knew, an average would be a useful tool. In mathematical terms, this property is called covariance, and the fact that stocks possess it was rediscovered in the 1960's and 1970's by mathematicians armed with computers. The fact that the rediscovery was not particularly earth- shaking was unfortunately lost on many of them. All of the above rambling is intended to lead us in the direction of a sermon on the current stock market. The recognition that stocks tend to move together goes back to Dow, and this tendency certainly continues, in some degree, to be a fact. There exists some evidence, in our view, however, that such may be less the case in recent years than has been true in the past. The extent of co- variance in common stock prices, in other words, may be diminishing, slightly and imperceptibly . perhaps, but to a degree significant enough to affect portfolio behavior. A recent instance that can be documented is the period 1976-1978, during which time, in general, large-capitalization stocks tended to decline significantly while smaller stocks advanced sharply. Intuitively, based on our own experience, there has been no recorded instance of comparable behavior. We are unable to quantify this suspicion, but we strongly believe it to be true. In the most recent past, the last couple of months, we have seen the averages do eSSentially noth- ing. This would lead one to believe, given a high degree of covariance, that most stocks were doing nothing at all. Such has not been the case. Energy stocks have been declining sharply, while other issues have been moving ahead. This is a lack of covariance which, again intuitively, we suspect may not often have been present during past periods, such as those enumerated last week, when the averages moved in s narrow trading range. One of the things that makes the stock market so fascinating is the fact that it possesses many properties which remain unchanged over time. Others, however, subtly become altered. In the case of variability, we may presently be witnessing such an alteration. Dow-Jones Industrials (1200 PM) 934.73 S & P Composite (1200 PM) 127.19 Cumulative Index (2/12/81) 1025.41 ANTHONY W. TABELL DELAFIELD, HARVEY, TAB ELL AWT sla No statement or e)(preSSlon of opinion or any other moIler herein contolned i, or IS 10 be deemed to be, directly or indirectly, on offer or the OII(llollon of on offer to buy or sell any secunty referred to or mentIOned The motter IS preented merely for the CO!'lvel'lenC of the subSCriber While oNe believe the sources of our Informa1,0 to be reluJble, we If'I no way represenl or guarantee the accuracy thereof nor of the tatement mude herein Any action to be token by Ihe subSCriber hould be based 0 hiS own investigation and Informollon Janney Montgomery Scott, Inc, as a corpo'atlon, and lIS of/'cers or employees, may now have, or may later toke, poslhons or trades In respect to any secufltles mentioned If'I Ihl or any future Issue, and SiJch pOSition may be ddferenl from any views now or hereafter expressed In llIs or any olher Issue Janney Montgomery Scolt, Inc. whICh IS registered With the SEC as on Investmenl adVisor, may give odvke 10 ,ts ,veslment adVISOry and olhe! customers Independently of any statements mode, thiS or In any olher Issue Further mformOI,on on any security men honed herein IS avatlable on request