1. By controlling the proportion of our assets committed to cash (CDs, money market funds, U. S. Treasury bills, etc.) and the proportion committed to common stocks (recognizing that the erosion of purchasing power wrought by inflation is among the very real risks of investing).
2. By controlling the degree of diversification versus concentration that characterizes our portfolio with respect to both industries and individual companies. (Incidentally, modern portfolio theory contends that capitalism bestows no incremental return for not being diversified.)
3. By controlling the quality of the portfolio in terms of the quality of the companies we acquire and continue to own (a form of risk management comparable to carrying fire insurance on our house – not needed unless we have a fire).
4. By controlling the degree of market risk to which we expose ourselves in terms of the relative price levels of the securities we hold in our portfolio.

1. Our starting point in the analysis of securities currently owned consists of a review of the Standard & Poor’s “quality” ratings and the Value Line “timeliness” ratings for each company in the portfolio. If both ratings are “above-average” for a particular company, and if neither that company nor its industry constitutes an inordinately heavy concentration, relative to the portfolio as a whole, we shall probably not give the position much further scrutiny on such occasion.
2. If the security does not now carry a Standard & Poor’s quality rating that is above-average (A+, A, or A-), we undertake to evaluate the degree of risk associated with its ownership, as indicated by an analysis of the underlying company. We next ascertain the significance of this risk, relative to the size and quality of the portfolio as a whole, and make a judgment about the appropriateness of continued assumption of this risk by the client, relative to our perception of the client’s “tolerance for risk.”
3. If the security does not now carry a Value Line timeliness rating that is above-average (#1 or #2), it may be for either one or a combination of two reasons:
   1. There may be a reversal or deceleration in the operating parameters (most notably, what Value Line calls “earnings momentum”) of the underlying company that indicates a less-than-likely prospect for above-average performance in the year ahead. Our own analysis undertakes to determine to what extent this may be the case and, in particular, attempts to ascertain if and when the company might reasonably again be expected to exhibit an above-average performance. We then attempt to relate this “turnaround-time” to our perception of what we call our client’s “turnaround-time tolerance.”
      1. In addition to the risk inherent in a common stock associated with the quality of the underlying company (which we seek in part to measure with the S&P rating as explained above), there is also the risk associated with the price-level of the stock. If the risk associated with a stock’s price is not commensurate with what appears to be its prospects for growth, this element of risk, too, is reflected in the Value Line timeliness rating.
         
         There are several ways of measuring the price-level of a common stock, relative to its historic norms. The most common yardstick is the “price-earnings ratio” – the stock’s price per share divided by the company’s earnings per share. Because of the volatility of corporate earnings, as well as a lack of uniformity of opinion as to what measures of earnings should be used, other measures of a stock’s price level are frequently employed in conjunction with, or in lieu of, the price-earnings ratio. Among the more popular of these other ratios are price-to-sales, price-to-cash flow, and price-to-book value. Price-to-dividend (the reciprocal of dividend yield), price-to-capital spending, and price-to-retained earnings are also useful ratios.
         
         Such ratios tend to be of greater value for comparing the price of a company’s stock with its own historical past than for comparing companies in different industries, of different quality, or with different growth rates.
         
         
4. Finally, if a security is in a taxable account, the desirability of its sale on the basis of the foregoing factors must be evaluated in terms of the tax consequences of its sale. If the security is held at a loss, there are unlikely to be any tax obstacles to its sale; whereas, if there is a taxable gain, we must weigh this factor, too, in making our recommendation.

The rankings are produced primarily by a computer program using as input the earnings and price history. The system tends to assign high ranks to stocks with low price-earnings ratios relative to historic norms and to the current price-earnings ratio on the market. The system also tends to assign high ranks to stocks whose quarterly earnings reports show an upward momentum, relative to the quarterly earnings on the market as a whole, and to stocks that have upward price momentum. These factors are weighted by the computer program. The weights used on the different factors are chosen by doing a cross-sectional regression on past data. The set of weights that seems to give the best predictive ability is then chosen. In sum, the one-year rankings are based on growth in earnings, price momentum, and the price-earnings ratio of each stock relative to the market and to historical standards for that stock.

Standard & Poor’s says of its “quality” ratings the following:

The point of departure in arriving at these rankings is a computerized scoring system based on per-share earnings and dividend records of the most recent ten years – a period deemed long enough to measure significant time segments of secular growth, to capture indications of basic change in trend as they develop, and to encompass the full peak-to-peak range of the business cycle. Basic scores are computed for earnings and dividends, then adjusted as indicated by a set of predetermined modifiers for growth, stability within long-term trend, and cyclicality. Adjusted scores for earnings and dividends are then combined to yield a final score.

A ranking is not a forecast of future market price performance, but is basically an appraisal of past performance of earnings and dividends, and relative current standing. These rankings must not be used as market recommendations; a high-score stock may at times be so overpriced as to justify its sale, while a low-score stock may be attractively priced for purchase. Rankings based upon earnings and dividend records are no substitute for complete analysis. They cannot take into account potential effects of management changes, internal company policies not yet fully reflected in the earnings and dividend record, public relations standing, recent competitive shifts, and a host of other factors that may be relevant to investment status and decision.

With respect to these two rating systems, it is useful to note the following:

1. These are largely objective and statistical rating systems. They involve almost no subjective input by individual analysts.
2. They are relative rating systems. While a stock with a good rating is expected to deliver a good performance in a good market, it may be expected to deliver no more than a “less bad” performance in a bad market.
3. While the Standard & Poor’s quality rating is of only marginal value as a prognosticator of potential future appreciation, the Value Line timeliness rating is of only marginal value as a measure of quality. Together, however, the two ratings make a good pair. While the Value Line rating provides an indication of how a stock might perform in the marketplace if things go as expected, the Standard & Poor’s rating provides us with a measure of our potential for misery if the Value Line rating fails us.
4. The interpretation of why the current ratings are what they are, and the probable duration of the circumstances responsible for the current ratings are, for the most part, left to the resources of the rest of us.

Clifford G. Dow, CFA, CFP®, ChFC
Chartered Financial Analyst
Certified Financial Planner™
Chartered Financial Consultant