It has been emphasized that the main vogue for being a formula trader began in the
late thirties, and was primarily a reaction to the market declines of 1929-32 and 1937-38. Naturally, the market analysts who first worked with formulas were more interested in building protection against declines than profiting from advances, and they understandably assumed that the severity of future drops in market prices would match these two earlier periods.

As a formula trader, it is almost impossible to resist the temptation to forecast stock prices, and it is difficult for a formula investigator to know at any particular time whether he is making a forecast on the basis of available facts or whether he is allowing his optimism or pessimism of the moment to dominate his efforts. In 1949, for example, one investigator wrote about the original Keystone plan—whose weakness has turned out to be too low a secular growth rate—"more recent stock-price fluctuations gives us some cause to question the assumption that the trend will be as strongly upward in the future as it appears to have been over the entire period from 1897 to 1946."x If this commentator had been writing either three years earlier or three years later, it is doubtful that he would have made such a criticism.1

As we have seen, a number of bank trust departments and investment counselors use formulas, but it is likely that many more use them as a guide than are willing to publicize the fact. One trust officer has said he doubted if very many trust departments used formulas, because "they wouldn't want to admit their judgment was that bad." One large investment advisory organization has spent a great deal of effort drawing up a median based on a wide range of economic and market data, but has consistently refused to make the fact public, since it doesn't want to be identified with formulas, even though it uses this median on some portfolios it manages, with the clients' approval.

A large number of investors, particularly institutions, still use formulas and are quite happy with them. Although some of the pioneering endowment funds, such as those of Yale, Vassar and Oberlin, have given them up, many others, including those of Syracuse and Kenyon, still use formulas and are satisfied with the results.

Are formulas worth using?
Whether or not any particular investor should be a formula trader is, of course, a matter of individual judgment. Some formulas, such as the Genstein Plan, require a fair amount of calculation, and many people are unwilling to discipline themselves to set aside time to manage their investments.

At the high points of big bull markets, many investors are ready to scoff at formulas. It is true that any portfolio containing bonds is at a disadvantage during bull markets, but how many individual portfolios perform as well as the Dow-Jones during a bull market? Besides, who is to predict that the market will always be one big bull market after another? A formula is powerless to take maximum advantage of a straight-up price rise, but the more normal pattern of stock prices is to undergo frequent periods of decline also. The "ideal formula timing plan," as summed up by one authority "is not that which secures the greatest gain for a given assumed pattern of security-price fluctuations but one which achieves the greatest gain for a degree of risk appropriate to the circumstances of the investor."2

Undoubtedly, the prestige of formula investing is at its lowest ebb during periods of steadily rising prices, but after every decline a new revival of interest occurs, simultaneous with the discovery by many investors that they are not the analytical geniuses they had previously thought themselves to be. In 1949, for example, formula investing had proved itself superior to the average investor's judgment, and Business Week reported: "Despite the steep hills and valleys on market price charts, formula-investing during the past two decades would have produced far better results than those achieved by most individual money managers."3

Security Selection
One subject that has not been touched on so far is the question of how to select the common stocks for a formula trader portfolio. In the examples examined in this book, one or another of the popular stock averages has been used to indicate movement of the stock portion of the account. Obviously, investors do not buy stock averages.

Most commentators on formula investing suggest that investors buy stocks of above-average volatility. This would include most of the relatively high quality "growth stocks." It is not necessary to concentrate only on such stocks, however. What is important is to buy only those stocks which are actively traded, of good quality, and subject to at least average fluctuation.

As noted previously, there is no necessity for the investor to give up his prerogative of selecting the stocks he feels suit his requirements best. Whether he concentrates on conservative blue-chip stocks, growth stocks or wildly speculative issues, he can get the same benefits from a formula. The purpose of a formula—even if the investor using it does not always follow its dictates with precision—is to provide a touchstone for adjusting one's portfolio against probable market moves and maintaining a strong financial position under all circumstances. And this purpose will be fulfilled no matter what stocks the investor selects.

In speaking of the indications given by formulas, it is not intended that the investor necessarily retain exactly the same stocks at all times, even though the formula specifies that a certain proportion of stocks be held. The formula investor should pay careful attention to his portfolio and switch his stocks around somewhat as their outlooks change.

How To Build Your Own Formula
The investor who concludes that he might profit by being a formula trader is faced with the problem of what type he wants to use, and exactly what rules he is going to set up for himself. A basic consideration, of course, is the amount of risk he is willing to assume.

The element of risk is present in all investment schemes, and it is the purpose of a formula to minimize it, while working toward some growth of capital. The amount of risk the investor wants to build into his formula is up to him—the more risk the greater the profit possibilities. The constant-ratio plans examined earlier assumed a 50-50 stock-bond relationship, but an investor willing to take on more risk would be perfectly justified in setting the stock percentage substantially higher—at 80 percent, say.

Although, as discussed in detail earlier, there is no "best" formula that will suit all investors, the two which are widely used, practical, easy to operate and profitable over periods of time are dollar averaging and the constant-ratio plan. Most investors can undoubtedly profit by using one or another of these, adjusting them to suit individual needs. As noted previously, the two can quite easily be combined to take advantage of the merits of each. If it were possible to predict what type of market is in the cards for next year, it would be easy to construct exactly the right formula—but then it would be even easier and more profitable to throw out formulas altogether.

Other Factors Influential
Aside from the question of risk, convenience and psychological satisfaction also must influence the choice of a formula. For example, an investor who finds even the relatively simple mathematics involved in the Graham or Genstein intrinsic-value plans irksome would be foolish to try to follow such a system. And the investor who has little faith in the infallibility of the Keystone channels would undoubtedly not for long follow one of them even if he started. The formula must "feel" right; the investor must be convinced of the value of the formula, so that he will not be tempted to discard it when it happens to be performing poorly.

The writer has not attempted to present every formula ever invented. The formulas described are felt to offer a representative sampling of the most practical or widely publicized types to enable the investor to select whatever features he may prefer. There are numerous possibilities for other plans. A good source of future formula methods might well be in market "timing" techniques such as the "confidence index," one or another of the breadth indexes or advance-decline indicators, or the strength measurements issued by Lowry's Reports. These technical approaches to the market could undoubtedly be adapted easily to the formula principle. A formula based on current stock yields would probably have given good results over past years.

It is possible that a good formula could be built around the loan-deposit ratios of commercial banks, the underlying assumption being that the Federal Reserve Board—which influences this ratio by its actions in the money market—is a major influence on the stock market. Although the relationship of business cycles to stock market cycles was distant during the war and postwar years, it may be that a formula based on business indicators might again be profitable in the future. All these areas, of course, would require detailed investigation, but the important point is that the reader need not feel restricted to the formulas that have been devised in the past.

In order to show how an original formula may be constructed with little trouble, this writer has devised and tested a modification of the constant ratio formula, based on odd lot indexes, that would have given adequate results over a number of years. Like the "compromise plan" described on page 77, it is simply the constant ratio plan with an added feature, based on odd lot trading, to improve results.

As used by specialists in the study of odd lot statistics, the significance of these extends far beyond that indicated by the use made of them in this formula.4 For the present purpose, it will be sufficient to point out that studies conducted by Garfield A. Drew have shown that enthusiasm of odd lot investors for stocks at certain periods may signal danger in the market, while an indifferent or bearish attitude may signal a market low point. Each day, the number of shares bought and sold in odd lots on the New York Stock Exchange during the preceding trading day are released by the Exchange and published in leading newspapers. The index used in this formula is the Odd Lot Balance Index originated by Garfield A. Drew and regularly published by Drew Investment Associates of Boston. It is a three-month moving average of the ratio of odd lot sales to purchases, multiplied by 100. An index, for example, shows that sales exactly equalled purchases; a figure of 90 means that sales were 90 percent of purchases. The higher the index goes, the more stock odd lot investors are selling relative to their purchases, and the lower it goes, the more stock they are buying relative to their sales. For this formula, the significance of the figures is that when odd lotters are heavy sellers, it is time to increase stock holdings, and when they are buying heavily, it is time to sell some stocks.

TABLE 10
STOCK-BONDRATIOCALLEDFOR
 

As drawn up, the formula is a 50-50 stock-bond constant ratio, with the proviso that when the Odd Lot Balance Index rises above 90, the stock portion of the account is raised to 75 percent, and the bond portion accordingly reduced to 25 percent. The account is reviewed every six months, and changes made as indicated. If no change in the stock-bond relationship is dictated, then the account is readjusted to the previous percentages. (Other percentages could be adopted—a 40 percent stock ratio when the Index is below 90, for example, and 90