Modern Portfolio Theory
There are two explanations which preserve the integrity of earlier and more theoretical discussions of risk and return. First, as was explained in the last article, according to modern portfolio theory, the risk premium of an individual asset is not measured by its own variability considered in isolation but rather by the contribution which it makes to the variability or riskiness of a diversified portfolio to which it is added. An individual stock with very great variability would be expected to have a rate of return not much greater than that of bonds of high quality if the returns on the stock were not highly correlated with returns on the market as a whole and it therefore did not add much to the riskiness of a diversified portfolio. In fact, a common stock whose returns were not correlated at all with returns on the market would be expected to have returns equal to those on a riskless asset.
It is plausible to assert that stocks of the lowest quality or with the greatest historic variability have returns which are less highly correlated with the market than stocks of higher quality and less historic variability. The stocks of the highest quality are often stocks of very large and widely diversified corporations—stocks such as American Telephone, General Electric, General Foods, Standard Oil Company (New Jersey) and so on. These great corporations are often deemed to be “blue chips” or of “investment grade,” the implication being that they are of high quality and have rates of return which can be predicted with greater confidence than can those of stocks of lesser quality. Since each of these corporations is very large, and since the profitability of each depends upon levels of demand in almost all parts of the country and in many different industries, it is not surprising that rates of return on these stocks are highly correlated with movements in the general economy and in the market as a whole.
By contrast, stocks of the lowest quality or with the greatest variability are often stocks of relatively small and immature companies whose profitability depends to a greater degree upon things other than the great tides in the movement of the general economy and the general market. It is easy to think of exceptions, but the generalization seems valid. If that is so, the stocks of lowest quality and with greatest historic variability may well have returns that are not highly correlated with the market and which consequently do not contribute as much to the riskiness of diversified portfolios as might be surmised on the basis of their total variability. This explanation amounts to saying that historic variability does not determine the risk premium of an individual asset, and that the deficiency in the measure has important consequences for the stocks with the greatest variability.