Charles J. Collins (1965) - Solar phenomena have been a source of scientific interest and investigation since Sir William Herschel, in 1801,found a correlation between sunspot activity and terrestrial phenomena [...] Modern science is giving considerable attention to solar phenomena in relation to disruption of the earth's magnetic field, to human health, and to weather, including rainfall, temperature, and cyclone frequency. The security analyst's interest is more directly concerned with the directly concerned with the effect of solar phenomena on business, and on speculation as evidenced by the ebb and flow of prices over our stock exchanges [This paper points] out one simple correlation of solar-stock market movements that will, fortunately, come to another test within the two or three years ahead. This is an apparent relationship between a recurrent phase of each sunspot cycle and an important stock market peak. The matter is of interest at this time for the reason that considerable attention is being given by students of the stock market as to when the broad advance that has been under way for a number of years is to reach a terminal point. This sunspot correlation, as discussed below, may throw some light on the subject. Briefly stated: It appears that an important market peak has been witnessed or directly anticipated when, in the course of each new sunspot cycle, the yearly mean of observed sunspot numbers has climbed above 50.
[...] Over the 94-year period under review, there were seven completed sunspot cycles, and it appears that an eighth was completed and a new cycle was started in 1964.During these eight cycles, not onlywas an important stock market peak concurrently witnessed (1881, 1892, 1916, 1936,1946, 1956)or directly anticipated (1906, 1929) by the above-50 count in sunspots, but, in four instances (1881, 1916, 1929, 1936), the designated peaks also marked the extreme or secular peaks for the entire sunspot cycle. The year 1890 seems an exception. In May of that year, the stock index reached its high of 5.62. In August 1892, the 5.62 level was again attained and, as concerns the yearly mean of the monthly stock indexes, the year 1892 peaked at 5.55, as compared with 5.27 for the year 1890 [...] In other words, in six instances, important stock market peaks and the sunspot climb above 50 came the same year, the two exceptions being 1906 and 1929. As to the 1906 exception, it will be noted, from the monthly range stock market chart, that the market peaked in January of that year, with December 1905 not far behind the January 1906 peak.
From a study of stock market history in relation to solar phenomena, a second theorem may be adduced: In each solar cycle, the largest stock market decline, in terms of percentage drop, comes after the sunspot number, on an annual basis, has climbed above 50. In the light of the foregoing observation, the 94 years of sunspot activity under review seems to occupy a rather narrow latitude for dogmatism. Thus, the preceding remarks should not betaken as a definitive prognosis of pending stock market behavior. Instead, they present a rather interesting correlation that has existed for a period of years between sunspot activity and major market peaks. Ergo, since the solar cycle is now at a point germane to this correlation, it seems worthwhile to present the previous relationship and await events, not without interest, of course, but mostly in the spirit of an enquiring attitude.
Originally printed in Financial Analysts Journal, November-December 1965; reprinted in Cycles Magazine in March 1966, and again in Cycles Magazine, Vol. 40, No. 3, September/October 1989]; editor's postscript of the 1989 reprint: "It is interesting to note the relation between above-50 crossingsand the stock market since 1965. In July 1966, the mean sunspot number moved above 50. The stock market shortly thereafter plunged in a major correction. In January 1978, the mean sunspot number again went above 50. The stock market, which had been in a downtrend, continued into a bottom after this date. In October 1987, the mean sunspot number went well above 50 to 60.~ and the 1987 crash followed. The mean sunspot number will next rise above 50 in about 1998."