John Bajkowski is AAII’s financial analyst and editor ofComputerized Investing.

and a long time horizon. They are seeking to accumulatewealth, and are willing to accept greater risk for theprospect of greater returns. They might choose to focustheir investment on growth-oriented companies. On theother hand, a retired couple has a shorter time horizonand will be more concerned with preserving their wealth.Therefore, they might choose to focus on stocks that havehigher dividends and lower price volatility.

Clear, focused, narrowly-defined objectives lead to thebest screens. Some common broad screening objectivesinclude seeking growth stocks, value stocks or even stockswith positive price momentum. But before the stockscreening process is started, these objectives should befurther refined to reflect the specific types of stocks thatthe investor is seeking and the best way to measure thesestocks. The young couple in our example may decide toseek growth stocks in the expansion stage of their lifecycle with strong earnings momentum. The retired couplein our example might seek gas and electric utilities withabove-average dividend growth rates.

Primary Screening Criteria

The primary screening criteria should flow naturallyfrom your objective, and should attempt to filter onlythose companies that meet your objective. If you are agrowth investor, it should provide you with a list ofcompanies that are in the growth stage of their life cycle,and not mature cyclical firms.

Screening criteria filter stocks by comparing a company’snumerical or quantitative figure against some base fig-ure. When defining your criteria you will need to decideif you wish to use absolute or relative conditions. Rela-tive conditions compare a company figure against a num-ber adjusted for the current level of a company, industry,or market as a whole. For example, you may be screeningfor companies whose price-earnings ratio is less than thatof the market. As the market’s price-earnings ratio goesup, the screen will get less restrictive, allowing compa-nies with higher price-earnings ratios to pass.

Some criteria should only be used on a relative basis.

Creating Stock ScreensThat Make Practical Sense

By John Bajkowski

Analyst’s Corner

When designing the manual for Stock Investor, AAII’sstock screening and fundamental analysis computer pro-gram, we wanted to include samples of a few basicscreening criteria. We quickly saw that it is easy to comeup with a list of meaningful screening criteria, but apply-ing them in useful ways is challenging.

Screening is the application of quantitative criteria to abroad universe of stocks in order to narrow the list downto a few companies. It allows you to focus your attentionon a smaller but more promising group of stocks.

It is best to look at screening as a multi-stage process:• You must first clearly define the objective of your screen.• Next, you must construct primary criteria that locate

stocks that match your screening objective.• Then, you need to construct a set of secondary criteria

that ensure the companies passing the primary screendid so because they truly meet your objectives and notout of coincidence.

• And last, you must keep in mind that even the bestscreen represents only a starting point for in-depthanalysis.

Defining a Clear Objective

An objective should always be developed before con-structing a screen. The objective should reflect your re-turn objectives, risk tolerance, and investment philoso-phy. Return objectives encompass not only the total re-turn, but the relative contribution from capital gainsversus dividend income. Risk tolerance refers to howeasily the investor copes with volatility in both an abso-lute and relative sense. Relative risk concerns the perfor-mance of a stock in relation to the market; whereas,absolute risk concerns the performance of a stock inde-pendent of the market. Investment philosophy encom-passes the style an investor uses to select stocks.

For example, a young couple usually has little wealth

Ratios such as margins and turnover are very industry-specific and become meaningful only when compared toindustry norms or company trends.

Screens that compare company data to other companyelements or historical averages can also be useful. Be-cause of growth prospects, some companies normallytrade at a higher price-earnings ratio multiple. Screeningfor just companies with price-earnings ratios below thatof the market may lead to a collection of bad stocks thatdeserve low price-earnings ratios and are not really un-derpriced. Using a relative screen that compares acompany’s price-earnings ratio against its historical normor against its expected growth may be a better way topoint out potentially mispriced stocks that warrant acloser look.

Alternatively, you can choose to compare a companyfigure against some constant that does not fluctuate overtime. An example would be a screen for companies witha price-to-book-value ratio below 1.0. As higher overallmarket levels lead to higher valuations, the number ofcompanies passing the price-to-book ratio will decrease.If market levels go to extremes, no suitable investmentsmay pass the screen. Some value investors use absolutescreens such as this to modify their exposure to themarkets. During periods in which the market is pricedhigh, fewer attractive investments appear to replace theovervalued securities that have been sold, which leads toa net reduction in equity assets.

One of the biggest mistakes in criteria construction ishaving a list of criteria that are reasonable individually,but when combined turn out to be contradictory. If youare looking for potential high-growth companies, thenyou shouldn’t team up a requirement for high earningsgrowth with a requirement for a high dividend yield.Companies that are truly growing usually need to usecash for expansion and can’t afford to pay high divi-dends. Combining the criteria in this way will negate theobjective of the screen, probably leaving the investorwith a list of oddball stocks.

By the same token, too many criteria filtering the sametype of companies are often combined. If you are lookingfor contrarian or out-of-favor stocks, screening for lowprice-earnings, price-to-book, price-to-sales, or price-to-cash-flow ratios or high dividend yield will, for the mostpart, list the same type of stocks. Your screening effortswould be better spent focusing on the criteria that bestindicate the type of companies you are seeking and oncriteria that you really understand well.

Table 1 displays sample primary screening criteria us-ing growth objectives and sample criteria using valueobjectives.

Secondary or Conditioning Screens

Even with a clear objective and a well-built primaryscreen, you can expect a number of companies that do

not embody the company type you are seeking to slippast your screen. Your high dividend yield screen willprobably contain some companies ready to cut theirdividend and your growth screen will probably capturesome mature cyclical companies examined during theirnormal cyclical upturn.

While screening is designed to be a preliminary stage inthe security selection process, the screening process shouldinclude a secondary, or conditioning, screen. The condi-tioning screen should help establish that companies pass-ing the primary screen did so because they meet thescreen’s ultimate objective. These differ from the primaryscreening criteria in that if they were used by themselvesto filter companies, they would not locate companies thatmeet your objective.

A primary screen for high-dividend-yielding stocksmay include a criterion for companies whose dividendyields are above that of the company’s five-year average.This will lead to companies with relatively high dividendyields. A conditioning screen would analyze those com-panies to help establish that the dividend is relativelysecure. It might include a criterion that specifies a maxi-mum payout ratio (dividends per share divided by earn-ings per share) of less than 50% to seek companies thatare not paying out more than half of their earnings in theform of dividends. It is a conditioning screen for a high-dividend-yield scan because, by itself, it does not indi-cate whether the dividend yield is high or low.

Table 1 lists sample conditioning criteria for a primaryscreen with a growth objective, and sample conditioningcriteria for a screen with value as a primary objective.

All Screens are Preliminary

When designing stock screens, keep in mind that thereare no miracle screens that produce lists of guaranteedwinners. A well-designed screen, however, should pro-vide you with a preliminary list of stocks that hold somepromise.

In developing a screen keep the following points inmind:• Develop a clear, narrowly defined objective, keeping in

mind the type of stocks that will meet your investmentphilosophy and objectives.

• Construct primary screening criteria that will locatecompanies that match your objective. In constructingprimary screening criteria, avoid using rules that willcancel each other out or that are redundant.

• Develop a set of conditioning criteria that will helpensure that a company passed the primary criteria forfundamental reasons rather than by chance.

• Remember that even the best designed screen is only apreliminary search for investments using a small set ofquantitative factors. A complete in-depth analysis thatexplores both quantitative and qualitative factors shouldfollow any screen.

Table 1.Primary and Secondary Screen Examples

For use with AAII's Stock Investor program (version 2.1)

yield• Current dividend yield above industry average• Current dividend yield above market average• Current dividend yield above an absolute level

Secondary/Conditioning Screen ExamplesFocus on maintenance of dividend payment:• Specify maximum level of payout ratio• Specify minimum level of earnings growthFocus on growth potential for dividends:• Specify minimum level of earnings growthFocus on Liquidity:• Specify minimum quick, or current, ratioFocus on company’s financial strength:• Specify company’s maximum level of debt to total assets,

or• Specify company’s maximum level of total debt to capital,

or• Specify company’s maximum level of debt to equity

Growth ScreeningObjective: Identify companies in stage of rapid and

expanding growth with earnings momentum.

Primary Screen Examples• Specify minimum absolute level of historical earnings

per share growth• Specify minimum absolute level of forecasted earnings

per share growth• Specify growth rate above industry average growth rate• Specify industries that are expanding at rates above that

of the economy• Require positive increase in annual earnings for each

individual period of analysis• Require increasing growth rate in earnings from period

to period

Secondary/Conditioning Screen ExamplesFocus on expanding level of sales to support earnings:• Specify minimum growth rate in revenuesFocus on competitive advantage of company relative toindustry:• Specify ratio of company profit margin to industry profit

margin• Specify increasing profit margin over timeFocus on retention of earnings to support future sales:• Specify maximum level of payout ratios• Specify minimum level of sustainable growth (return-on-

equity x 1.0 – payout ratio)

Value ScreeningObjective: To identify companies whose market price

is low relative to value measures based upon factorssuch as sales, earnings, dividends, cash flow or assets.

Price-Earnings Emphasis

Primary Screen Examples• Current price-earnings ratio below firm’s five-year price-

earnings ratio• Current price-earnings ratio below industry average• Current price-earnings ratio below market average• Current price-earnings ratio below an absolute level

Secondary/Conditioning Screen ExamplesFocus earnings potential of company:• Require minimum level of earnings growth• Require minimum level of revenue growth• Current price-earnings ratio less than half of earnings per

share growth rate• Require increasing or stable profit margin over time• Require profit margin above industry average profit marginFocus on company’s financial strength:• Specify company’s maximum level of debt to total assets,

or• Specify company’s maximum level of total debt to capital,

or• Specify company’s maximum level of debt to equity

Price-to-Book-Value Emphasis

Primary Screen Examples• Current price-to-book-value ratio less than 1.0• Current price-to-book-value ratio below firm’s five-year

price-earnings ratio• Current price-to-book-value ratio below industry average• Current price-to-book-value ratio below market average

Secondary/Conditioning Screen ExamplesFocus on quality of book value:• Low ratio of intangibles/goodwill to book valueFocus on earnings potential of company:• See secondary/conditioning variables for price-earnings

emphasis

Dividend Yield Emphasis

Primary Screen Examples• Current dividend yield above firm’s five-year dividend

A successful stock selection strategy is elusive for manyinvestors. Chasing hot growth stocks often leaves investorsburned, while low-priced, “value” stocks often only becomecheaper. James P. O’Shaughnessy provides a detailed ex-amination of basic investment strategies in his book “WhatWorks on Wall Street: A Guide to the Best-Performing Invest-ment Strategies of All Time” (McGraw-Hill, $29.95).

O’Shaughnessy argues that the majority of investors fail tobeat market averages because they do not follow a disci-plined approach to investing. Instead, investors let theemotions surrounding the market overpower their judgmentand push them off their planned investment course. Thisarticle examines and implements what O’Shaughnessy termsthe “cornerstone” value and growth strategies he developedwhile testing a range of value and growth strategies usingover 40 years of S&P Compustat data.

The Universe

O’Shaughnessy established two base groups of stocksfrom which to pick investments and to serve as performanceand risk benchmarks—all-stocks and those with large capi-talizations. The “all stocks” universe was determined byselecting stocks with a market capitalization (shares out-standing times market price) of $150 million or greater.Rather than use the complete CompuStat database,O’Shaughnessy decided to focus only on stocks that a pro-fessional money manager could buy without too much diffi-culty due to liquidity. O’Shaughnessy based the $150 mil-lion market capitalization cut-off as of December 1994, andadjusted this figure for inflation so that the minimum valuewas approximately $26.7 million in 1952. Limiting the analy-sis to stocks with a market cap above $150 million effectively

cuts out half the stocks currently traded on the New York,American, Nasdaq National Market, and Nasdaq Small Capexchanges.

The large cap group was determined by selecting stockswhose market capitalization was greater than the average forthe overall universe. Typically only about 16% of the compa-nies pass this filter because few very large firms push up theaverage market cap. Testing revealed that the large capgroup had similar return and risk performance to that of theS&P 500.

O’Shaughnessy’s testing revealed that “micro caps” orstocks with market capitalization below $25 million beat allother size groups on a risk-adjusted basis, but he arguesthat it is too difficult to invest in these stocks. A comparisonof the O’Shaughnessy all-stocks universe to the large capuniverse revealed that the all-stocks group had higher per-formance, but they also carried more risk, as measured bystandard deviation of return. However, the return was suffi-ciently higher so that the risk-adjusted return exceeded thelarge cap group.

Value Strategies

O’Shaughnessy tested a number of basic value strategieson both the large cap and all-stocks universe. Value strate-gies use measures such as price-earnings ratios, price-to-book value ratios, price-sales ratios and dividend yields, toidentify out-of-favor investments that are priced attractivelyin relationship to these measures. Stocks whose prices arelow relative to some tangible company factor such as earn-ings are purchased, while companies with high prices rela-tive to these measures are avoided.

Table 1 provides a summary of the single criterion strate-gies tested by O’Shaughnessy. O’Shaughnessy used theCompuStat database to construct a portfolio of the top 50stocks passing a strategy. An equal dollar amount was in-vested in each security, with the portfolio rebalanced annu-

Screening for Growth and ValueBased on “What Works on Wall Street”By John Bajkowski

John Bajkowski is AAII’s senior financial analyst and editor of Comput-erized Investing. Kenneth J. Michal provided research assistance.

Table 1.O‘Shaughnessy’s Single Criterion Test Results

ally. Dividends were included in the return measurements,but transaction costs were ignored. O’Shaughnessy testedthe reverse of a strategy when appropriate. For example,when testing the price-earnings strategy, a portfolio of thetop 50 price-earnings companies was established andtracked, as well as a portfolio of the bottom 50.

The test results confirm the conclusions of many studiesemphasizing the benefit of following a value-oriented ap-proach in selecting stocks.

Growth Strategies

As indicated in Table 1, O’Shaughnessy’s tests of basicgrowth strategies revealed that with the exception of rela-

tive strength, simple growth strategies do not compensateinvestors adequately for the risk possible with these strat-egies. High growth, high margin stocks are often bid up tounrealistic levels. Growth stock investors must watch theirportfolios very carefully, quickly selling off issues with anyhint of disappointing growth. O’Shaughnessy’s tests revealedthat growth stocks often underwent significant periods ofsuperior risk-adjusted performance, especially during strongbull markets. However, these gains were eroded over thelong run during subsequent market periods.

Relative strength stands out as an effective tool for thestock investor, working consistently over the 40-year testperiod. Relative strength compares the price performanceof a security to that of a corresponding index. Stocks outper-

Value Strategies

Price-Earnings: Buy 50 stocks with thelowest price-earnings ratios.

Price-Book Ratios: Buy 50 stocks withthe lowest price-book ratios.

Price-to-Cash Flow Ratios: Buy 50stocks with the lowest price-to-cash flowratios.

Price-to-Sales Ratios: Buy 50 stockswith the lowest price-to-sales ratios.

Dividend Yield: Buy 50 stocks with thehighest dividend yields.

Growth Strategies

One-Year Earnings Increase: Buy 50stocks with the highest one-year earningsper share percentage increases.

Five-Year Earnings Growth: Buy 50stocks with the highest five-year earningsper share increases.

Profit Margins: Buy 50 stocks with thehighest net profit margins.

Return on Equity: Buy 50 stocks with thehighest return on equity (ROE).

Relative Strength: Buy 50 stocks withthe highest one-year price change.

All-Stocks (Market capitalization above $150 million)

While stocks with low price earnings ratios performed betterthan those with high ratios, the low price-earnings stocksunderperformed the all-stocks universe and had higher stan-dard deviation.Low price-to-book value stocks outperformed all-stocks uni-verse on an absolute basis, but also had higher risk leading toa risk-adjusted return equal for both groups.Low price-to-cash flow stocks outperformed all-stocks uni-verse on an absolute basis, but higher risks leads to under-performance on a risk-adjusted basis.

Best performing value measure for the all-stocks universe.Significantly outperforms universe on an absolute basis andrisk-adjusted basis.High-dividend yielding stocks fail to outperform all-stocks uni-verse on an absolute basis and is higher risk.

Leads to both lower return and higher risk. Some shortperiods, however, do show strong performance only to befollowed by very poor performance.

Performance below all-stocks coupled with greater risk levelsresults in terrible performance.

Underperformed all-stocks and exhibited slightly higher risk.

Performance of high ROE stocks was roughly equal to all-stocksindex, but risk was higher, leading to weaker risk-adjustedperformance.Performance of high relative strength stocks exceeds perfor-mance of all-stocks, but at much higher risk. Risk-adjustedreturn below all-stocks.

Large Cap (Market capitalization above average)

Large cap, low price-earnings stocks significantlyoutperformed both large cap universe and highprice-earnings stocks. Although strategy had higherrisk, it was attractive on a risk-adjusted basis.Low price-to-book stocks strongly outperform largecap universe with somewhat higher risk. Superiorstrategy on both a total and risk-adjusted basis.Low price-to-cash flow stocks outperform largecap universe on both an absolute and risk-adjustedbasis. Best performing large cap value strategy.

Strong and consistent performance. Outperformslarge cap universe on both an absolute and riskadjusted basis.Strong and consistent performance. Out per-forms large cap universe with little additional risk.

Results in poor performance and higher risk.Buying stocks with the lowest gains leads toslightly better performance than benchmark, butat higher risk, resulting in risk-adjusted perfor-mance just below benchmark.Performance below large cap index coupled withgreater risk results in terrible performance.

Performed below large cap index with roughly thesame risk level.

Performance below large cap index coupled withgreater risk levels results in poor performance.

Performance of high relative strength stocksexceeds performance of large cap universe, butwith higher risk. Absolute and risk-adjusted re-turn above large cap universe.

Price- 5-Year 5-YearPrice- Book- Price- Earnings Sales 52-Week

Earnings Value Sales Dividend Growth Growth Market RelativeRatio Ratio Ratio Yield Rate Rate Capitalization Strength

Company (Exchange: Ticker) (X) (X) (X) (%) (%) (%) ($ Mill) (%) Description

Cornerstone Value Stocks (Ranked by Dividend Yield)

U.S. West Communications (N:USW) 14.0 4.48 1.7 6.0 na na 17,196.0 -11 Telecommun co. in 14 western states

ARCO Chemical Company (N:RCM) 11.5 2.26 1.2 5.9 5.1 8.6 4,572.4 -26 Intermediate chem. & specl’ty prods

BCE Inc. (N:BCE) 22.5 2.09 0.8 5.6 -11.3 6.0 15,392.0 12 Telecommun servs & equip

RJR Nabisco Holdings Corp (N:RN) 30.1 1.26 0.7 5.1 17.9 2.9 9,885.8 -12 Produces tobacco & snacks

Imperial Oil Limited (A:IMO) 17.3 2.36 1.1 4.9 -2.1 -3.5 7,010.1 -3 Crude oil, natural gas, & petroleum prods

Southern New Eng. Telecom (N:SNG) na 5.37 1.2 4.9 -42.4 2.8 2,379.1 -29 Telecommun prods

Chrysler Corporation (N:C) 7.2 2.19 0.4 4.7 37.5 15.9 23,887.7 -3 Cars, trucks, & accessories

Ford Motor Company (N:F) 10.3 1.50 0.3 4.7 31.0 7.0 36,643.1 -15 Sells cars, trucks & accessories

Deluxe Corporation (N:DLX) 31.4 3.43 1.4 4.6 -12.3 5.6 2,605.0 -21 Check printing & elec’tr funds transfer servs

NYNEX Corp. (N:NYN) 16.7 3.31 1.7 4.6 -30.7 -0.3 22,659.4 -19 Telecommun co. in northeastern U.S. & worldwide

Brascan Limited (A:BRS.A) 7.8 1.00 na 4.4 na na 2,460.5 15 Mining & metals, forest products, oil & gas

MacMillan Bloedel Limited (M:MMBLF) 9.0 0.66 0.5 4.4 42.2 11.8 1,694.4 -14 Forest prods co.

Atlantic Richfield Co. (N:ARC) 12.7 2.78 1.1 4.3 -7.1 -1.6 20,102.5 -8 Petroleum liquids, crude oil, natural gas, & coal

Bankers Trust NY Corp. (N:BT) 14.2 1.75 1.2 4.3 -23.6 1.0 7,418.8 13 Bank holding co.

J.C. Penney Company, Inc. (N:JCP) 14.7 2.02 0.5 4.3 8.7 5.1 11,129.5 -16 Apparel, shoes, home furnish & jewelry retailer

Moore Corporation Ltd. (N:MCL) 20.1 1.45 0.9 4.3 14.7 -1.2 2,212.4 -9 Business forms, printing servs, labels

Bell Atlantic Corp. (N:BEL) 16.4 4.06 2.3 4.2 7.7 1.4 30,260.8 -16 Telecommun co. in Mid-Atlantic region

GTE Corporation (N:GTE) na 6.48 2.2 4.0 -25.8 0.8 45,085.3 -12 Telecommun prods/servs

Dun & Bradstreet Corp. (N:DNB) 19.9 3.94 1.0 4.0 -7.5 2.3 4,171.3 -27 Info. services; credit ratings, investor info

Occidental Petroleum Corp. (N:OXY) 19.4 2.31 0.8 3.9 17.4 -0.8 8,428.3 -11 Crude oil & natural gas

Cornerstone Value Median 14.5 2.19 1.1 4.6 5.1 2.9 9,885.8 -12

Large Cap Median 20.7 2.93 1.6 1.4 13.1 8.0 3,333.0 -3

Cornerstone Growth Stocks (Ranked by Relative Strength)

Brightpoint Inc. (M:CELL) 49.2 5.96 0.9 0.0 55.0 86.3 447.2 86 Distrib wireless telecommun devices

Insight Enterprises, Inc. (M:NSIT) 25.8 2.31 0.5 0.0 37.0 40.9 210.6 86 Direct mrkts microcomputers

Watsco, Incorporated (N:WSO) 33.4 3.15 0.9 0.5 12.3 23.0 337.0 60 Climate control components

PHH Corporation (N:PHH) 18.9 2.62 0.7 1.6 10.8 3.4 1,671.1 54 Integrated management servs

Interface, Inc. (M:IFSIA) 24.8 1.80 0.5 1.1 -13.7 5.2 424.8 42 Interior mrkt prods

Computer Data Systems (M:CDSI) 17.5 2.79 0.7 0.4 21.1 14.1 179.4 37 Info technology servs & prods

Culp, Inc. (N:CFI) 16.8 2.40 0.5 0.7 29.4 15.1 225.7 35 Mfrs & sells upholstery fabrics

Gibraltar Steel Corp. (M:ROCK) 15.5 2.04 0.7 0.0 38.8 21.6 274.2 33 Process value-added steel prods

Alberto-Culver Company (N:ACV) 25.7 3.81 1.0 0.7 15.9 12.7 951.5 29 Health & beauty care prods

Scotsman Industries, Inc. (N:SCT) 14.9 2.09 0.7 0.4 8.6 12.5 292.9 29 Refrig. prods for foodservice industry

Leasing Solutions (M:LSSI) 20.6 3.04 1.4 0.0 na na 181.4 27 Leasing info procss’ng & communs equip

Spartech Corporation (N:SEH) 16.1 2.70 0.8 1.7 16.5 20.2 306.7 23 Thermoplastic materials & molded prods

AMERCO (M:UHAL) 10.9 1.09 0.6 0.0 na na 616.2 22 Holding co. for U-Haul

Avery Dennison Corp. (N:AVY) 25.3 5.04 1.3 1.7 94.8 3.8 4,208.2 21 Specl’ty adhesives for labels

Carlisle Companies, Inc. (N:CSL) 19.2 3.46 1.1 1.5 15.5 10.5 994.0 21 Rubber, plastic, & metal prods

Owens-Illinois, Inc. (N:OI) 15.6 4.32 0.7 0.0 23.9 0.8 2,904.0 20 Plastic pckg prods & glass containers

Coachmen Industries (N:COA) 10.5 2.71 0.5 1.0 28.6 13.6 347.7 19 Recreational vehicles & van conversions

Palm Harbor Homes (M:PHHM) 16.6 3.24 0.7 0.0 80.8 30.0 366.1 16 Produce multi-section mfrd homes

Raymond Corporation (M:RAYM) 14.0 1.77 0.7 0.4 45.4 14.1 193.4 16 Materials handling equip

Personnel Grp. of America (N:PGA) 24.9 1.46 0.8 0.0 16.9 6.3 308.3 15 Personnel staffing servs

Cornerstone Growth Median 18.2 2.71 0.7 0.4 22.5 13.9 342.4 28

All-Stocks Median 19.8 2.62 1.7 0.4 14.2 10.5 562.8 -5

Exchange Key: N =New York Stock Exchange A =American Stock Exchange M = Nasdaq

Table 2.O’Shaughnessy Cornerstone Screening

Statistics are based on figures as of February 28, 1997.Data Source: AAII’s Stock Investor/Market Guide, Inc. and I/B/E/S

forming the index have positive relative strength, whilestocks lagging the performance of the index have negativerelative strength. Relative strength measures help to un-cover pockets of exceptionally strong or weak performance.It seems that stocks in motion tend to maintain their upwardor downward momentum.

Multifactor Strategies

O’Shaughnessy’s studies revealed that it was possible tocombine a number of the factors to establish selectioncriteria, superior in risk and return, to the single factorcriteria. For example, pairing a value measure such as price-to-sales with a growth filter such as strong relative strengthleads to a higher risk-adjusted return than the correspond-ing single factors. The value screen reduces the chance ofpaying too much for a growth stock, while the price momen-tum screen helps to highlight well-priced stocks with marketrecognized value.

O’Shaughnessy tested a number of combinations andsettled on two strategies superior on a risk-adjusted basis:a cornerstone value screen geared towards large cap stocks,and a cornerstone growth screen geared towards the all-stocks universe.

Cornerstone Value

Testing revealed that the value approach was better suitedfor the large cap universe than the all-stocks universe;perhaps smaller-cap stocks with low valuation ratios do nothave the resources to bounce back to glory. Large caps werealso less volatile than the all-stocks universe, which fitsnicely into the more conservative goal of the cornerstonevalue screen. O’Shaughnessy established a screen for largecap market leaders and tested a number of value factors todetermine which factor produced consistent and strongperformance over the long haul.

AAII’s Stock Investor program was used to illustrate thescreen. The large cap universe is defined as stocks whosemarket cap exceeds the average for the complete universe.Stock Investor’s database covers over 7,500 stocks with anaverage market capitalization of $1.3 billion. Establishing aminimum market capitalization of $1.3 billion reduced ouruniverse to 1,024 companies.

The next cornerstone value filter specifies that a companyshould have more common shares outstanding than theaverage stock in the database—a screen for adequate li-quidity. We screened for companies with more than 38.2million outstanding shares, reducing the number of passingcompanies to 858.

The cornerstone value screen ultimately relies upon thedividend yield to highlight attractively-priced stocks. How-ever, a high dividend yield may also signal the market’sexpectation of a dividend payout cut. To help screen outthese potentially weak companies, O’Shaughnessy speci-fied that the companies should have a cash flow per share

which exceeds the database average. One does not nor-mally screen a single period per share item across a crosssection of firms. Per share items are normally studied overtime or converted into ratios that can be meaningfullycompared across firms. O’Shaughnessy does not reveal ifother tests of dividend safety were tested and discarded.An examination of the adequate coverage of dividendpayouts from cash flow or earnings is a common dividedstrength requirement. We followed the O’Shaughnessy screenand looked for firms with cash flow per share greater then$1.60, which reduced the number of passing companiesdown to 646.

O’Shaughnessy also specified that a firm must have totalsales 1.5 times the database average. With sales averaging$930 million in Stock Investor, we screened for companies withsales levels greater than $1.4 billion. This screen cut the listof passing companies to 510. O’Shaughnessy excluded utili-ties to keep them from dominating the dividend yieldscreen, leaving us with 450 companies.

O’Shaughnessy recommends the use of dividend yield tobuild a portfolio for the cornerstone value screen. The topyielding companies significantly outperformed the largecap universe while possessing a similar level of risk. Table2 presents the 20 top yielding stocks of the cornerstonevalue screen. The 4.6% yield for this group is significantlyabove the 1.4% yield of the large cap universe.

Cornerstone Growth

With the exception of relative price strength, the growthstrategies tested by O’Shaughnessy were not very promis-ing. O’Shaughnessy’s single factor test screened for compa-nies with extreme values—factors such as highest earningsgrowth, highest margins, and highest returns on equity. Byreducing these extreme growth requirements and estab-lishing moderate value requirements, O’Shaughnessy wasable to construct a portfolio which had the desired combi-nation of strong price growth and reasonable risk.

The growth stock screen was based upon the all-stocksuniverse because smaller stocks have greater growth poten-tial than their large cap counterparts. The all-stocks uni-verse looks for stocks with a market capitalization above$150 million, providing a starting point of 3,476 companies.

The cornerstone growth strategy focused on earnings con-sistency, rather than relying on high earnings growth levels.A screen for five years of consecutive earnings growth provesto be very restrictive and reduces the number of passingcompanies down to 455 when applied to the all-stocksuniverse.

O’Shaughnessy balances the growth requirement by es-tablishing a maximum price-to-sales ratio ceiling of 1.5.When used independently as a value screen, more restric-tive ceilings of 0.75 or 1.0 are common. The price-to-salesratio was loosened to allow more growth-oriented compa-nies to pass, yet not allow them to have valuations that aretoo extreme. Applying the price-to-sales screen left 132

companies standing.O’Shaughnessy then screened for the 50 companies with

the highest relative price strength over the last year. The top20 stocks ranked by 52-week relative strength are presentedin Table 2.

With the exception of the price-to-sales ratio, the corner-stone value companies tend to have more attractive mul-tiples than the cornerstone growth companies. The results ofthe screen also highlight that certain ratios can benefit fromindustry comparison. Brightpoint passes the cornerstonegrowth screen with a price-to-sales ratio of 0.9, yet it hasrelatively high price-to-book and price-earnings ratios. Re-tail and wholesale firms tend to naturally carry low price-to-sales ratios.

The cornerstone growth companies have much strongerearnings and sales growth rates than their value counter-points. The cornerstone value companies are much larger

Definitions of Screens and TermsThe following is a short description of the terms used in Table 2

Price-Earnings Ratio: Price per share divided by most recent 12months’ earnings per share. Provides a measure of the market’sexpectations regarding the firm’s earnings growth and risk.

Price-to-Book-Value Ratio: Current price per share divided bybook value (common stockholder’s equity) per share. A measure ofstock valuation relative to net assets. A high ratio might imply anovervalued situation; a low ratio might indicate an overlooked stock.

Price-to-Sales Ratio: Current price divided by the sales per sharefor the most recent 12 months. A measure of stock valuationrelative to sales. A high ratio might imply an overvalued situation; alow ratio might indicate an undervalued stock.

Dividend Yield: Current annual dividend per share divided byshare price. A measure of stock valuation relative to dividend payout.A high yield might imply an undervalued situation; a low yield mightindicate an overvalued stock.

EPS Growth Rate-Last Five Years: Annual growth in earningsper share from total operations over the last five fiscal years. Usedas a measure of how successful the firm has been in generating thebottom line, net profit.

Sales Growth Rate-Last Five Years: Annual growth in totalsales over the last five fiscal years. Provides a confirmation of thequality of the historical earnings per share growth rate.

Market Capitalization: Computed by multiplying the last shareprice by average number of shares of common stock outstandingduring the latest quarter.

52-Week Relative Strength: Price performance of a stockrelative to an index, in this instance, the S&P 500. A price momen-tum indicator, it confirms investor expectations and interest bycomparing the performance of a stock relative to the market.

than their benchmark large cap growth, while the growthcompanies are smaller than the all-stocks median. Also, asexpected, the relative strength of the cornerstone group issignificantly higher than the value group. Only three of thecornerstone value stocks have positive relative strength,and their median is significantly below the large cap me-dian—a common characteristic of value screens.

Conclusion

“What Works on Wall Street” provides an interesting ex-amination of investment strategies. O’Shaughnessy arguesthat the market is far from random and that a sound, disci-plined, emotion-free, investment approach is the only wayto beat the market over the long term.

(See AAII’s forum on America Online for a list of screening fields asused in AAII’s Stock Investor program.)

Screening CriteriaFor use with AAII’s Stock Investor program (Version 2.1)

Cornerstone Value Strategy

Found in the Price & Share Data Group

MKTCAP > 1322.9 Market capitalization is greater than the average market capitalization of the overall

database; for example, the overall database average for market capitalization (2/28/

97 data) was 1,322.9 (Note: Average values can be found on the screen after selecting

the field or by running a statistical summary report for all companies.)

SHR_AQ1 > 38.2 Average number of shares outstanding during the latest quarter is greater than the

average number of shares outstanding of the overall database; for example, the overall

database average for number of shares outstanding (2/28/97 data) was 38.2 (Note:

Average values can be found on the screen after selecting the field or by running a

statistical summary report for all companies.)

Found in the Income Statement, Annual Group

CFPS_12M > 1.60 Cash flow per share for the latest 12-month period is greater than the average cash

flow per share of the overall database; for example, the overall database average for

cash flow per share (2/28/97 data) was 1.60 (Note: Average values can be found on

the screen after selecting the field or by running a statistical summary report for all

companies.)

SALES_12M > 1400 Total sales for the last 12-month period is 1.5 times greater than the average total

sales of the overall database; for example, the overall database average (2/28/97 data)

was 930.35, then multiplied by 1.5, was approximately 1,400 (Note: Average values

can be found on the screen after selecting the field or by running a statistical summary

report for all companies.)

Found in the Company Information Group

IND_2_DIG <> 12 Sector is not Utilities

Found in the Multiples Group

Rank by YIELD Rank the resulting companies to seek out the highest yielding stocks

Cornerstone Growth Strategy

Found in the Price & Share Data Group

MKTCAP > 150 Market capitalization is greater than $150 million

Found in the Income Statement, Annual Group

EPS_12M >= EPS_Y1 Total earnings per share for the latest 12-month period is greater than or equal to

total earnings per share for the last complete fiscal year

EPS_Y1 > EPS_Y2 Total earnings per share for the last complete fiscal year is greater than total earnings

per share one year ago

EPS_Y2 > EPS_Y3 Total earnings per share for one year ago is greater than total earnings per share two

years ago

EPS_Y3 > EPS_Y4 Total earnings per share for two years ago is greater than total earnings per share three

years ago

EPS_Y4 > EPS_Y5 Total earnings per share for three years ago is greater than total earnings per share

four years ago

Found in the Multiples Group

PSPS < 1.5 Price to sales per share ratio less than 1.5

Found in the Price & Share Data Group

Rank by RS_52W Rank the passing companies to seek out stocks with the highest relative strength

“A bird in the hand is worth two in the bush.”It’s an old saying, but it’s a sentiment felt by many conser-

vative stock investors who prefer the stocks of stable andestablished companies that provide part of their returnsooner, in the form of dividends, rather than later, in theform of capital gains.

How does one choose among these kinds of stocks?One approach is followed by Geraldine Weiss, editor of

the highly regarded Investment Quality Trends, a La Jolla,California-based newsletter that tracks and recommendsstocks based on her approach.

Weiss melds a conservative, blue-chip investment stylewith a value approach, using dividend yield as a guide tovalue. A high dividend yield signals out-of-favor stocks, butmany such stocks are out-of-favor for good reason—they arefinancially troubled. Weiss’ strategy attempts to weed outtruly financially troubled firms by seeking out-of-favor stockswithin a relatively safe sector of high-quality stocks.

Weiss has outlined her approach in two books, “DividendsDon’t Lie,” with Janet Lowe (Longman Publishing, 1988, out ofprint), and her more recent “The Dividend Connection,” writtenwith son Gregory Weiss (Dearborn Financial Publishing,1995, $24.95, 800/245-2665). These writings are the primarysource for this article.

The Philosophy

Weiss can best be described as a blue-chip value investor—buying quality stocks at a good value.

Quality is Weiss’ first concern—investing in companies thathave withstood the test of time by surviving numerouseconomic cycles without lowering or canceling a dividend.These stocks tend to be more stable and are usually the lastto fall when the economy is declining, making them less

risky than lower-quality or unproved stocks.Weiss also notes that quality companies provide, through

their dividend payments, a reliable source for steady andincreasing income. These payments significantly contributeto an investor’s bottom line return, and in particular providesupport during bear markets by supplying a source of in-come.

Of course, significant profits come from price apprecia-tion. Weiss maintains that all stocks go through cycles ofundervaluation and overvaluation, and that investor profitscan be achieved by taking advantage of these cycles—buying stocks when they are undervalued and subsequentlyselling them when they are overvalued.

But dividends are the star players in the approach. InWeiss’ view, dividends offer the best indication of bothquality and value, besides providing a source of return.

The quality of corporate management can be judged byexamining a firm’s dividend payments and policies. Compa-nies that provide steady dividend payments over the longterm are generating sufficient earnings to first cover allexpenses and debt payments; companies that increasecash dividends year-after-year are able to do so throughincreased earnings and thus are particularly well-managed.

For Weiss, the most important measure of investmentvalue is the dividend yield—a company’s current annualdividend per share divided by share price.

Why? In the long run, the underlying value of the stream ofdividend payments determines the price of a stock. As Weisspoints out, the impact of dividends on price is reflected whenchanges occur—if a dividend is increased, the stock becomesmore valuable and more highly rated; if a dividend is de-creased, stock values decrease.

The dividend yield ratio relates dividends per share to shareprice, similar to other popular valuation measures such as price-earnings ratios (price divided by earnings per share) andprice-to-book-value ratios (price divided by book value pershare). Weiss does not discount these other ratios, but notes

A Blue-Chip Value Investor: SeekingHigh-Quality, Out-of-Favor StocksBy Maria Crawford Scott

Maria Crawford Scott is editor of the AAII Journal.

that distortions in reported earnings and book value moreeasily arise with them. Dividend payments, in contrast, tendto be more predictable and not subject to differing account-ing interpretations; they are real, measurable dollars paidout to shareholders, and offer a more stable measure torelate to share price.

On the other hand, Weiss believes that over the shortterm, prices go to extremes. This causes a company’s divi-dend yield to fluctuate within a range that is unique to eachcompany. When dividend yields are relatively high, theshare price is low relative to dividends per share paid out,indicating that the stock may be undervalued; when divi-dend yields are relatively low, the share price is high rela-tive to dividends per share, indicating that the stock may beovervalued. Weiss suggests buying and selling blue-chipstocks when they reach the extremes of their historicaldividend yield range.

The philosophy combines a high-quality strategy with a valueapproach, and has a goal of minimizing downside risk, maxi-mizing the potential for capital gains, and maximizing thegrowth of dividend income.

The Blue-Chip Universe

Because the safety of dividend payments and their ability torise are central to her approach, Weiss suggests that investorsbegin their selection process by narrowing their initial list downto a universe of blue-chip stocks. In this sense, the pre-selectionof the “universe” of stocks is particularly important.

Weiss seeks companies that are well-known, with goodmanagers, good research and development efforts, goodmarketing skills, well-known products and services, and thatare reluctant to cut a dividend even in times of cyclicalstress. In particular, she lays out six criteria:1) Dividend increases in five of the last 12 years: A measure of

good long-term performance of the company’s ability toincrease net earnings, reflected by a trend of increasingdividends. Weiss terms this the most reliable measure ofgood management.

2) A minimum of five million shares outstanding: An assuranceof liquidity, which in turn prevents manipulation of shareprice. Firms smaller than this may have trouble attractinginstitutional investors, which leads to illiquidity.

3) Shares must be held by at least 80 institutions: Anotherassurance of liquidity, but Weiss views this rule as the leastrigid of her criteria.

4) Improved earnings in at least seven of the last 12 years:Another indication of a well-managed company, indicatingthat a company can survive the tough years and prosper inthe good ones. Weiss notes that she also looks for salesincreases, and profit margins that are under control.

5) It must have paid dividends, with no interruptions, for roughlythe past 25 years: Consistent dividend payments are a signof a profitable, well-managed company; a long record alsoprovides the historical data necessary to evaluate the rangeof dividend yields indicating valuation extremes. Weiss

suggests that if an investor wants to relax the number of years,dividends should have been paid long enough for severalcycles of overvaluation and undervaluation to be estab-lished.

6) The stock must carry a Standard & Poor’s quality ranking nolower than A–. S&P ranks the quality of stocks on a scale of A+to C (a D indicates reorganization) based on past growth andstability of earnings and dividends; Weiss finds the rankingsa useful guide to investment quality. Although stocks musthave a quality ranking of A– to make the initial list, Weissallows stocks to drop to B+ once on the list; however, if theydrop to B, the stock is deleted from her blue-chip list.How many should be included in the universe? Weiss herself

follows 350, a number she says is somewhat arbitrary but is amanageable size, while including a wide variety of industrygroups.

Dividend Yield Criteria

Once a pre-selected list of blue-chip stocks is developed,Weiss suggests the purchase of these stocks based on avalue approach, using historical dividend yields as a guide.

Specifically, Weiss charts a firm’s dividend yield and its priceover a decade or more, and looks for historical high and lowdividend yield “turning points,” when stock prices changedirection. Weiss maintains that most stocks turn at roughly thesame dividend yield each cycle; averaging these dividendyields defines the boundaries.

“If a major rising price trend has ended in a 2% yield area manytimes, and a major declining price trend repeatedly has endedin a 5% yield area, a profile of value has been established forthat stock—a 2% dividend yield identifies a historically over-valued price where the stock should be sold to preserve capitaland protect profits. Conversely, when that stock is priced toyield 5%, it is historically undervalued and a good buyingopportunity is at hand.”

Weiss suggests that investors buy from their blue-chip stocklist when the prices cause dividend yields to be within 10% oftheir historical high dividend yield.

On the other hand, she does not suggest investors purchasean undervalued stock without first examining the reasons whya stock’s price has fallen—in particular, checking for thepossibility of a dividend decrease, which would eliminate itfrom consideration.

Other Factors

Weiss does not discount the use of other value measures whenanalyzing stocks, and in fact suggests several that are useful inconfirming valuations derived in the dividend yield approach.In addition, these valuations can help an investor more fullyunderstand how the market perceives a stock.

In particular, she suggests:1) A price-earnings ratio that is historically low for that particular

stock and other similar stocks, and that is below the marketmultiple. A low price-earnings ratio reflects the market’s low

Universe of stocksHigh-quality ‘blue-chip’ stocks. To get on this initiallist, stocks should have the following characteristics:• The dividend must have increased a minimum of

five times in the past 12 years.• Shares outstanding should number at least five

million.• Shares must be held by at least 80 institutions.• In at least seven of the last 12 years, corporate

earnings should have improved.• It must have paid dividends, with no interruptions,

for the past 25 years.• The stock must carry an S&P quality ranking no

lower than A–.

Criteria for initial considerationBuy from the blue-chip stock list when the pricescause dividend yields to be within 10% of theirhistorical levels of high dividend yield, indicating it ishistorically undervalued.

Other factors• A price-earnings ratio that is historically low for

that particular stock and other similar stocks, andthat is below the market multiple.

• A price-to-book-value ratio that is no higher than1.3, and the closer to 1.0, the better.

• A ratio of current assets to current liabilities of atleast 2.0, and a debt-to-equity ratio of no morethan 50% debt to equity (utility stocks areexcluded because of their unique regulatorystatus). Conservative investors should look for adebt-to-equity ratio or no more than 20%.

• Also be wary of other signs that dividendpayments may be in jeopardy: Payout ratios

approaching 100%, although a rising earnings trendcan support rising dividends; and an indicateddividend higher than reported annual earnings,although strong cash flow can help cover adividend payment when earnings temporarily drop.

• In general, when evaluating a stock for potentialpurchase, examine:

– The company’s financial performance, including itsrecord of earnings, dividends, debt-to-equity ratio,dividend payout ratio, book value, and cash flow.

– The company’s product performance, whether itis manufacturing goods or services that are indemand, its research and development efforts, andits ability to market its products or services.

– The company’s investment performance in theform of capital gains and dividend growth.

Stock monitoring and when to sellAim for a portfolio of about 20 stocks selected froma variety of industry groups.

Continually monitor the portfolio to weed outstocks that no longer satisfy the requirements ofgood value or quality.

Sell stocks when they are within 10% of theirhistorical low dividend yield, indicating overvaluation.If you tend to be reluctant to sell when prices arerising, consider placing a stop-loss order 12% to 15%below the overvalued price.

If dividend yields rise among stocks you own, makesure you understand why, but in general view it asan opportunity to buy more. However, if rise occursbecause a company omits a dividend, wait tempo-rarily for prices to shore up, and then sell, since itno longer meets the blue-chip criteria.

The Geraldine Weiss Approach in Brief

growth-in-earnings outlook for the firm.2) A price-to-book-value ratio that is no higher than 1.3, and

the closer to 1.0, the better. This is a Benjamin Grahamrule-of-thumb that seeks to buy stocks as close to thebare-bones worth of a company as possible.Weiss’ investment approach demands that the cash dividend

be safe. A change in the dividend payment causes the dividendyield to change, so that a share price that was once regarded asundervalued may become fairly valued or overvalued, and viceversa.

How can investors judge the safety of dividend paymentsfor both prospective stocks and stocks that they own? Of

course, stocks on a blue-chip list should have relatively safedividends. But Weiss also provides some guidelines:• Seek stocks with a ratio of current assets to current liabili-

ties of at least 2.0, and a debt-to-equity ratio of no morethan 50% debt to equity (utility stocks are excluded be-cause of their unique regulatory status); conservativeinvestors should look for a debt-to-equity ratio of no morethan 20%. Companies with high levels of debt can run intofinancial problems more quickly during a slowdown, put-ting the dividend in jeopardy.

• Payout ratios approaching 100% are a danger signal. Divi-dends are paid out of earnings, and are relatively safe and

can even be safely increased if the payout ratio (dividendas a percent of earnings) is 50% or less. The closer thepayout ratio is to 100%, the more endangered the divi-dend and the less likely it is to be increased. A risingearnings trend, however, can support rising dividends.

• Similarly, an indicated dividend higher than reported annualearnings is a danger sign. Since firms pay dividends prima-rily from earnings, a company cannot sustain dividend pay-ments above annual earnings. Although strong cash flow canhelp cover a dividend payment when earnings temporarilydrop, an “unprotected dividend is in greater danger than aprotected dividend.”Lastly, Weiss states that investors should examine all funda-

mental factors to fully understand the company before purchas-ing it, and when judging the merits of one prospective bluechip relative to another. She does not go into detail concern-ing what to look for, but she does provide a generalizedguide of what to look at when judging the quality of a firm:1) The company’s financial performance, including its record of

earnings, dividends, debt-to-equity ratio, dividend payoutratio, book value, and cash flow.

2) The company’s product performance, whether it is manufac-turing goods or services that are in demand, its research anddevelopment efforts, and its ability to market its products orservices.

3) The company’s investment performance in the form of capitalgains and dividend growth.

Portfolio Monitoring and When to Sell

Weiss suggests that individuals aim for a portfolio of about20 stocks, enough to assure that the portfolio is diversifiedbut manageable. In addition, selected stocks should be ina variety of industry groups.

On the other hand, Weiss does not believe that an investorshould always be fully invested—if, for example, few stocks areundervalued, or if the overall market appears to be overvalued(based on historical market dividend yields). Keeping moneyin cash during these times, she believes, allows an investor totake advantage when the opportunities eventually arise at theend of a bear market or during a major correction in a bull market.Weiss notes that these times offer investors the best opportu-nities to diversify their holdings because of the large selectionof undervalued stocks.

Portfolios must be continually monitored to weed outstocks that no longer satisfy the requirements of good valueor quality.

Within an investor’s portfolio, Weiss believes stocks should

be sold when prices cause them to be within 10% of theirovervalued extreme range. She views historically low divi-dend yields as “alarm bells,” and although the stock pricemay continue to rise, the upside potential is outweighed bythe downside risk. Of course, dividend yields tend to fallwhen prices are rising—and Weiss notes that investors areoften reluctant to sell these shares, particularly if they haveheld on to them for some time. Her suggestion: Place a stop-loss order 12% to 15% below the overvalued price; themarket will then dictate the sale.

What happens when dividend yields rise—indicating under-valuation—for stocks held within a portfolio? Dividend yieldswill rise if the price drops, and Weiss notes that investors shouldtry to determine the cause of the price drop. In general, however,she views this as a buying opportunity to add shares.

“When bad things happen to good companies, it must beviewed as a buying opportunity rather than a bailout. As longas a stock’s dividend is maintained, even an extremelyundervalued stock merits investment consideration.”

Dividend yields may also rise, and prices drop, if thecompany omits a dividend. In this instance, she suggeststhat investors wait temporarily, since stock prices may moveup shortly after, once the uncertainty has been eliminated.After the “rebound,” however, the stock should be soldbecause it no longer meets the blue-chip criteria.

Conclusion

Weiss’ strategy is an interesting derivative of a classicvalue approach that nonetheless reverses the process: Ratherthan finding undervalued companies first, and then analyz-ing for quality, Weiss first focuses on developing an initiallist of quality companies in which you would be comfortableinvesting, and then purchasing them when they becomeundervalued.

The approach is particularly useful for investors who seekstocks that pay a steady dividend and are likely to continuepaying a steady dividend; it is less useful for investorsprimarily seeking growth and who are trying to limit annualtaxable income. The approach is also well-suited for thosewho want to take advantage of dividend reinvestment plans.

Weiss summarizes her own approach this way:“Successful investing in the stock market is not brain surgery.

Anyone can be a successful investor. The secret is no secret. Itis simply that you confine your selections to blue-chipstocks, you buy them when they are undervalued, and yousell them when they become overvalued. This is the well-litpath of the enlightened investor.”

Screening CriteriaFor use with AAII’s Stock Investor program (Version 2.1)

Found in the Income Statement Annual GroupDPS_12M >= DPS_Y1 Dividends per share for the latest 12-month period is greater than or equal to dividends per share for the last complete

fiscal year

DPS_Y1 > DPS_Y2 Dividends per share for the last complete fiscal year is greater than dividends per share one year ago

DPS_Y2 > DPS_Y3 Dividends per share for one year ago is greater than dividends per share two years ago

DPS_Y3 > DPS_Y4 Dividends per share for two years ago is greater than dividends per share three years agoDPS_Y4 > DPS_Y5 Dividends per share for three years ago is greater than dividends per share four years agoEPS_12M >= EPS_Y1 Total earnings per share for the latest 12-month period is greater than or equal to total earnings per share for the

last complete fiscal yearEPS_Y1 >= EPS_Y2 Total earnings per share for the last complete fiscal year is greater than or equal to total earnings per share one year

agoEPS_Y2 >= EPS_Y3 Total earnings per share for one year ago is greater than total earnings per share two years agoEPS_Y3 >= EPS_Y4 Total earnings per share for two years ago is greater than total earnings per share three years agoEPS_Y4 >= EPS_Y5 Total earnings per share for three years ago is greater than total earnings per share four years ago

Found in the Price & Share Data GroupSHR_AQ1 >= 5 Average number of shares outstanding during the latest quarter is greater than or equal to 5

SHRINST >= 25.4 Percentage of shares outstanding held by institutional investors is greater than or equal to the median value forpercentage of shares outstanding held by institutional investors (Note: The median value can be found on the screenafter selecting the field or by running a statistical summary report for all companies.)

Found in the Earnings Estimates GroupEPSN_EY0 >= 4.0 Number of analysts providing estimates for the current fiscal year is greater than or equal to median value for number

of analysts providing estimates for the current fiscal year (Note: The median value can be found on the screen afterselecting the field or by running a statistical summary report for all companies.)

Found in the User Defined Group (see below)YIELD >= YIELD High A5Y Dividend yield is greater than or equal to the high 5-year average dividend yield

User Defined FieldYIELD High A5Y = ((DPS_Y1/PRICEL_Y1) + (DPS_Y2/PRICEL_Y2) + (DPS_Y3/PRICEL_Y3) + (DPS_Y4/PRICEL_Y4) + (DPS_Y5/

PRICEL_Y5)) /0.5

Found in the Income Statement Annual GroupDPS Dividends per share for the year

Found in the Price & Share Data GroupPRICEL Low price for the year

fell into two camps—“defensive” inves-tors and “enterprising” investors. Thesegroups are distinguished by the amountof “intelligent effort” they are “willingand able to bring to bear on the task.”For the defensive, or passive, investor,his analysis was geared toward avoid-ing serious mistakes or losses. Grahamtried to establish a procedure that pro-vided freedom from great effort and theneed for frequent decision. For the en-terprising, or active and aggressive, in-vestors, Graham established a policyfor those willing to devote the time andcare to the selection of securities that areboth sound and more attractive thanaverage. Table 1 summarizes the Gra-ham investment approach.

Screening: The Defensive InvestorIn establishing a course of action for

the defensive investor, Graham laid outa set of criteria that would help the in-vestor obtain securities which offered aminimum level of quality in terms ofpast performance and current financialposition as well as a minimum level ofquality in terms of earnings and assetsper dollar of share price.

Graham’s analysis for the defensiveinvestor is divided into primary indus-try sectors. Graham presented an in-vestment approach specifically for utili-ties and industrials, but said additionalsectors such as financials should also beselected using these criteria. Our analy-sis was therefore broken down into twosegments—utilities, and the rest of thestock universe. Tables 2 and 4 presentthe approach for these segments.

AAII’s Stock Investor program wasused to perform the screening for thisarticle. As of April 30, its database in-cluded 6,970 companies, including 198utilities. When screening, investors havethe option of screening against relativefactors, such as a profit margin higherthan industry average, or absolute fac-tors, such as a current ratio of 2.0 orhigher. If you are planning on screeningusing criteria versus absolute factors, it

In the May 1996 issue of the AAIIJournal, Maria Crawford Scott analyzedthe investment approach of BenjaminGraham. Graham’s philosophy of try-ing to buy a dollar’s worth of stock forfifty cents has set the tone for the valueinvestor. In this article we will developand apply the stock screens formulatedby Graham in his book, “The Intelli-gent Investor.”

The PhilosophyGraham’s view of security analysis is

founded upon the concept of intrinsicor central value justified by assets, earn-ings, dividends, financial strength andstability, definite company prospects,and quality of management. Without abasis for determining whether a secu-rity is undervalued or overvalued, theinvestor is at risk of being swept awayby the “tides of pessimism and eupho-ria which sweep the security market.”

For investors to have a reasonablechance of better than average results,they must follow policies that are in-herently sound and promising, yet dif-ferent from the policies followed bymost investors or speculators.

Graham warned that buying a ne-glected, and therefore undervalued, se-curity for profit generally proves to bea protracted and patience-trying expe-rience. However, the possibility of ex-traordinary gains only exists when theinvestor disagrees with the market.

While our article covers the analysisand selection of stocks, it is importantto keep in mind that Graham’s analysisof stocks is framed by the constructionand balance of a stock and bond portfo-lio. A neutral portfolio position wouldhave half of the portfolio in stocks andhalf in bonds. This proportion couldfluctuate from a minimum position of25% to a maximum position of 75%depending upon the attractiveness ofstocks relative to bonds.

Types of InvestorsGraham felt that individual investors

FEATURE

is important to break your analysis downinto various market sectors. Factors suchas debt levels, turnover, and margins,can be very industry-specific.

Adequate Size of EnterpriseGraham had a preference for large

companies. He felt that large firms havethe resources in “capital and brainpower” to carry them through adver-sity and back to a level of satisfactoryearnings. This concern came into playfor Graham because he looked at stocksof firms that became unpopular due tounsatisfactory developments of a tem-porary nature. Graham also felt that themarket responds more quickly with aprice increase when an improvement isshown for a large firm than a small firm.

When screening for company size, thethree most popular criteria are marketcapitalization (number of shares out-standing times market price), sales, andtotal assets. Graham focused on salesfor industrials and total assets for utili-ties because they reflect company ac-tivities and size directly, while marketcapitalization is tied to overall marketlevels.

Graham specified that the defensiveinvestor should exclude small compa-nies with less than $100 million of an-nual sales for industrial companies and$50 million in total assets for public utili-ties. Graham specified these levels over20 years ago. Assuming a 5% annualgrowth in prices over the last 25 years,our minimums would increase from$100 million in sales to $340 million forindustrials and from $50 million in totalassets to $170 million for utilities. Thisscreen is much more restrictive for non-utilities than utilities. Except for smallwater and gas utilities, the capital re-quirements for an utility are significant,along with the benefits of economies ofscale.

Strong Financial ConditionGraham used different measures of

financial strength depending upon the

Screening for Value via Benjamin Grahamby John Bajkowski

industry. As a test of short-term liquid-ity, Graham specified a current ratio(current assets divided by current li-abilities) of 2.0 or higher for industrialfirms. No current ratio requirementwas specified for the utility sector.Graham stated that this “working capi-tal [current assets minus current liabili-ties] factor takes care of itself in thisindustry as part of the continuous fi-nancing of its growth by sales of bondsand shares.”

To measure the use of long-term debt,Graham required that for industrialfirms long-term debt not exceed net cur-rent assets, or working capital. This isnot a common ratio for screening pro-grams, but if your screening programallows you to create custom fields itshould be easy to duplicate. Take carethat when using this criteria you specifynot only a upper limit, but a lower limit

as well. If a company’s current liabili-ties exceed its current assets then it willhave a working capital deficit and anegative ratio of long-term debt to netcurrent assets. We specified that thisratio be positive and less than or equalto 1.0.

If your screening program does notallow you to create custom fields youcan use another measure of financialleverage such as debt-to-equity or debtto assets. Graham’s criterion of debt toworking capital screened out about halfthe companies when used by itself, solook for a similar pass rate.

Financing is an important consider-ation for utilities, so Graham specifiedthat investors look at the debt-to-equityratio for this sector. He specified thatdebt should not exceed twice the stockequity (at book value, not market value).This turned out be a much less restric-

tive screen than the financial conditionscreens for non-utilities.

Earnings StabilityGraham liked to look at the historical

company performance over an extendedperiod of time. He had a preference forcompanies that avoided losses duringrecessionary periods. This would pointto industries such as utilities, insurance,food processing, medical supply firms,and pharmaceuticals. Graham recom-mend 10 years of positive earnings inhis screen for the defensive investors.Unfortunately, most screening pro-grams on the market today only coverfive years of income statement data. Ourscreen specified that earnings be posi-tive for only the last five years—a timeperiod without a severe overall eco-nomic downturn. To be true to Graham’soriginal criteria, an investor would want

Table 1.The Benjamin Graham Approach in Brief

Philosophy and styleInvestment in companies whose share prices are near their intrinsic value based on tangible assets, earnings, dividends, financial strength and stability, andquality of management. Buying at or below intrinsic value provides a margin of “protection” that can help absorb unfavorable developments, with subse-quently less risk of a market overreaction on the downside.

Universe of stocksFor “defensive” investors: High-grade dividend-paying common stocks of leading companies; saw utilities as particularly fertile ground. For “enterprising”investors: No restrictions; stocks of unpopular large companies and secondary companies (ones that are not leaders in a fairly important industry) consideredparticularly promising.

Criteria for initial consideration• Exclude small companies with less than $100 million of annual sales for industrial companies and $50 million of total assets for public utilities in 1972

dollars. In today’s market, this would roughly translate into sales of $340 million for industrial companies and total assets of $170 million for public utilities(assuming 5% annual growth).

• Strong financial condition—for industrial companies, current assets (cash, accounts receivable and inventory) should be at least twice current liabilities (short-term debt) and long-term debt should not exceed the net current assets (working capital, or current assets less current liabilities); for public utilities the debtshould not exceed twice the stockholders’ equity (total assets less total liabilities). For “enterprising” investors, he relaxed some of this criteria: current assetsshould be at least 1½ times current liabilities, and debt should not be more than 110% of net current assets.

• Positive earnings for at least the last five years.• Uninterrupted dividend payments for at least past 20 years; for “enterprising” investors, some current dividend.• Minimum increase of at least one-third in per share earnings in the past 10 years (2.9% average annual growth rate).• Current price should not be more than 15 times average earnings for the past three years.• Current price should not be more than 1½ times book value last reported, but price-earnings ratio below 15 could justify higher multiplier of assets. [Rule of

thumb: product of the multiplier times ratio of price to book value should not exceed 22.5]. For “enterprising” investors, he suggested trying to find firmsselling at two-thirds or less of book value.

Secondary factorsSkeptical as to accuracy of subjective judgments concerning growth prospects and management; good management indicated by a good long-term trackrecord.In terms of financial strength, surplus cash and no outstanding issues ahead of the common stock is preferable to firms with large bank loans and seniorsecurities, but modest amount of bonds or preferred stock not necessarily a disadvantage, nor is moderate use of seasonal bank credit.

Stock monitoring and when to sellEmphasized diversification—minimum of 10 different issues and maximum of 30.Buy and hold for the long term and try to ignore market vagaries. Review holdings at least annually in light of intrinsic value and if no shrinkage, continue tohold. Sell if issues rise “excessively” above their intrinsic value and can be replaced by issues much more reasonably priced.

Primary FactorsMarket SectorThe various market sectors have different characteristics in operationsand financing. Ratio analysis is typically more meaningful whenexamined within industrial groups. The first filter excluded utilities.

Adequate Size of the EnterpriseExclude smaller firms, which carry additional risks and tend to trade ata central value below their intrinsic value. Graham specified thatindustrials should have sales above $100 million back in the early1970s. A rough inflation adjustment brings this level up to a minimumlevel of $340 million.

Financial Condition• Current assets should be at least twice current liabilities—two-to-onecurrent ratio.• Long-term debt should not exceed working capital (current assetsless current liabilities). This the type of ratio that is unique to Grahamand will not typically be found in a screening program. If the program

allows for custom fields, it should be easy to create. If you must work with a fixed set of fields, then look for a good measure of financial leverage, such astotal debt to total assets or debt-to-equity. We created long-term debt to working capital, but were also careful to specify that the ratio not drop below zero orcompanies with negative working capital would pass the screen.

Earnings StabilityGraham specified some earnings for the common stock for the past 10 years. Most screening programs will not have data going back that far. Use thegreatest number of years available and check the final results against a printed publication such as Value Line. We looked for positive earnings for each of thelast five years and last 12-month period.

Dividend RecordUninterrupted payments for at least the past 20 years. Most screening programs will not have data going back that far. S&P Stock Guide indicates how longthe dividends have been paid. We screened for positive dividends in each of the last five years, along with a positive indicated dividend (most recent dividendmultiplied by four).

Earnings GrowthGraham specified a 3% annual growth rate over a 10-year period using a three-year average at the beginning and ending periods. This type of growth ratecalculation will not be found in most screening programs. We specified a minimum growth rate of 3% over five years using a more standard growth ratecalculation which only looks at the beginning and ending years’ earnings.

Moderate Price/Earnings RatioGraham specified that the current price to the average of the last three years’ earnings per share should not exceed 15. You should be able to calculate thisfield with a screening program that allows you to create custom fields. The goal of this criterion is to create a portfolio with an average price-earnings ratiothat does not exceed the inverse of the high-grade corporate bond yield.

Moderate Ratio of Price to AssetsGraham felt that current price should be less than or equal to 1½ times book value. However, a low price-earnings ratio could justify a higher price-to-book-value ratio. So he multiplied the price-to-book-value ratio by the price-earnings ratio and came up with a value of 22.5. Most screening programs will not havethe product of the price-earnings ratio and price-to-book ratio calculated, but will allow you to calculate this field. We screened for a value less than or equalto 22.5.

SingleCumulatively Criterion*

Complete Universe 6,970 6,772Market Sector 6,772 6,772Adequate Size of the Enterprise 1,684 1,684Financial Condition: Current Ratio 535 2,705Financial Condition: Long-Term Debt 408 3,287Earnings Stability 244 2,233Dividend Record 127 1,646Earnings Growth 98 3,342Moderate Price-Earnings Ratio 21 3,371Moderate Ratio of Price to Assets 13 1,480

*Of universe of non-utility companies.

to see how the company performedover at least a complete economic cyclewhen selecting the final candidates forthe portfolio.

Dividend RecordA common test for financial strength

over time is a long period of uninter-rupted dividends. In the defensive in-vestor screen, Graham recommendeduninterrupted payments of at least thepast 20 years. In screening for the divi-

dend record we came across the sametime limitation that we encountered withthe earnings screen—only five years ofdata. For the non-utilities, this criterionturn out to be a more restrictive screenthan the earnings screen. But the reasonis not due to companies cutting theirdividends as much as it is a reflectionthat dividend payouts have become lessimportant and are not demanded byinvestors as much. Today, companiesare more inclined to use excess cash to

buy back their shares. To be faithful toGraham’s original criterion, publica-tions such as the S&P Stock Guide andMoody’s Dividend Record report therecord of uninterrupted dividends for awide range of companies.

Earnings GrowthGraham recommended a minimum

increase of at least one-third in per shareearnings in the past 10 years, whichtranslates into about a 3% annual growth

Table 2.Translating Style Into Screening: The Defensive Investor—Non-Utilities

Number of Stocks Passing Screening Criteria

Total 5-Yr. Price-toSales Debt to EPS Price- Book- ReturnLast Current Total Growth Earnings Value on Relative

12 Mos. Ratio Assets Rate Ratio Ratio Equity StrengthCompany (Exchange: Ticker) ($ mil) (x) (%) (%) (x) (x) (%) (%) Description

Bassett Furniture (M: BSET) 490.8 5.8 13.8 36.3 15.3 1.16 7.7 –22 Mfrs furniture

Blair Corp. (A: BL) 562.9 2.7 39.8 6.2 7.4 1.10 14.9 –41 Men’s & women’s apparel, furnishings

Bowne & Co. (A: BNE) 410.3 2.7 22.2 22.3 12.2 1.24 10.1 –12 Document prep and distrib

Excel Industries (N: EXC) 596.0 2.0 50.2 13.6 8.7 1.05 12.1 –17 Vehicle window & door systems

Haverty Furniture Cos. (M: HAVT) 395.5 2.6 62.1 7.0 13.2 1.14 8.6 2 Specialty furniture retailer

Hudson Foods (N: HFI) 1,261.2 2.6 53.5 25.0 11.2 1.26 11.1 –40 Produces poultry & meat prods

Justin Industries (M: JSTN) 461.4 3.6 37.2 27.7 12.5 1.35 10.8 –8 Mfrs bldg materials, footwear; book pub

Oshkosh Truck Corp. (M: OTRKB) 423.0 2.7 30.1 40.1 14.3 1.01 8.1 –7 Heavy-duty vehicles, transport equip

PACCAR Inc. (M: PCAR) 4,848.2 2.2 71.5 32.4 7.7 1.55 20.2 –13 Mfrs heavy-duty on- & off-road trucks

Phelps Dodge Corp. (N: PD) 4,185.4 2.6 42.4 10.2 6.8 1.88 27.9 5 Copper mining co. & engin’d prods mfr

Pitt-Des Moines (A: PDM) 461.3 2.0 45.8 10.3 7.7 0.92 12.0 –7 Custom-engineered prods, structures

Standard Motor Products (N: SMP) 663.5 2.7 58.9 15.8 14.3 1.10 7.7 –31 Mfrs replacement parts for auto systems

Thor Industries (N: THO) 580.4 2.7 24.5 27.2 14.0 1.50 10.7 –27 Recreational vehicles & small busesMedian 562.9 2.7 42.4 22.3 12.2 1.16 10.8 –13Median for All Stocks 96.5 1.9 55.2 10.7 17.5 2.02 9.1 –7

For Use With AAII’s Stock Investor (Version 2.1)

Found in Company Information GroupIND_2_DIG <> 12 Sector is not utility

Found in Income Statement, Annual GroupEPS_12M > 0 Total earnings per share for the latest 12-month period is positiveEPS_Y1 > 0 Total earnings per share last year is positiveEPS_Y2 > 0 Total earnings per share one year ago is positiveEPS_Y3 > 0 Total earnings per share two years ago is positiveEPS_Y4 > 0 Total earnings per share three years ago is positiveEPS_Y5 > 0 Total earnings per share four years ago is positive

SALES_12M >= 340 Total sales for the latest 12-month period greater than or equal to $340 million

Found in Ratios GroupCURR_Q1 >= 2 The current ratio greater than or equal to 2

Found in Growth Rates GroupEPS_G5F > 3 Annual growth in earnings per share over the past five years greater than 3%

Found in User Defined Fields (formulas below)Long-Term Debt to Wkg Cap <= 1 Long-term debt to working capital less than or equal to 1Long-Term Debt to Wkg Cap >= 0 Long-term debt to working capital greater than or equal to 0Price to Avg EPS <= 15 Price to three-year average earnings per share less than or equal to 15P/E Times P/B <= 22.5 Price-earnings ratio times price-to-book-value ratio less than or equal to 22.5

User Defined Fields

Long-Term Debt to Working Capital = LTDEBT_Q1 / ( CA_Q1 - CL_Q1)

Found in Balance Sheet, Quarterly GroupLTDEBT_Q1 Long-term debt at the end of the most recent fiscal quarterCA_Q1 Total current assets at the end the most recent fiscal quarterCL_Q1 Total current liabilities at the end the most recent fiscal quarter

Price-Earnings Ratio Times Price-to-Book-Value Ratio = PE * PBVPS

Found in Multiples GroupPE Price-earnings ratioPBVPS Price-to-book-value ratio

Price to Average Earnings per Share =PRICE / ( ( EPSCON_Y1 + EPSCON_Y2 + EPSCON_Y3) / 3 )

Found in Price & Share Data GroupPRICE The closing market price of the common stock

Found in Income Statement - Annual GroupEPSCON_Y1-Y3 Earnings per share from continuing operations for

the last three fiscal years

Table 3.Graham Defensive Growth, Non-Utilities Screen Results

INDCAT_DIV > 0 Indicated yearly dividend greater than $0.00DPS_Y1 > 0 Dividend paid last yearDPS_Y2 > 0 Dividend paid one year agoDPS_Y3 > 0 Dividend paid two years agoDPS_Y4 > 0 Dividend paid three years agoDPS_Y5 > 0 Dividend paid four years ago

For Use With AAII’s Stock Investor (Version 2.1)

Found in Company Information GroupIND_2_DIG = 12 Sector is utility

Found in Income Statement, Annual GroupEPS_12M > 0 Total earnings per share for the latest 12 month periods is positiveEPS_Y1 > 0 Total earnings per share last year is positiveEPS_Y2 > 0 Total earnings per share one year ago is positiveEPS_Y3 > 0 Total earnings per share two years ago is positiveEPS_Y4 > 0 Total earnings per share three years ago is positiveEPS_Y5 > 0 Total earnings per share four years ago is positive

INDCAT_DIV > 0 Indicated yearly dividend greater than $0.00DPS_Y1 > 0 Dividend paid last yearDPS_Y2 > 0 Dividend paid one year agoDPS_Y3 > 0 Dividend paid two years agoDPS_Y4 > 0 Dividend paid three years agoDPS_Y5 > 0 Dividend paid four years ago

Found in Balance Sheet, Quarterly GroupAssets_Q1 >= 170 Total assets are greater than or equal to $170 million

Found in Growth Rates GroupEPS_G5F > 3 Annual growth in earnings per share over the last five years greater than 3%

Primary FactorsMarket SectorThe various market sectors have different characteristics in operations andfinancing. Ratio analysis is typically more meaningful when examinedwithin industrial groups. The first filter included utilities.

Adequate Size of the EnterpriseExclude smaller firms, which carry additional risks and tend to trade at acentral value below their intrinsic value. Graham specified that utiltiesshould have total assets above $50 million back in the early 1970s. Arough inflation adjustment brings this level up to a minimum level of $170million.

Financial ConditionDebt should not exceed twice the stock equity at book value. Wescreened for a debt-to-equity ratio equal to or below 2.

Earnings StabilityGraham specified some earnings for the common stock for the past 10 years Most screening programs will not have data going back that far. Use the greatestnumber of years available and check the final results against printed publication such as Value Line. We looked for positive earnings for each of the last fiveyears and last 12-month period.

Dividend RecordUninterrupted payments for at least the past 20 years. Most screening programs will not have data going back that far. S&P Stock Guide indicates how long thedividends have been paid. We screened for positive dividends in each of the last five years along with a positive indicated dividend (most recent dividendmultiplied by four).

Earnings GrowthGraham specified a 3% annual growth rate over a 10-year period using a three-year average at the beginning and ending periods. This type of growth ratecalculation will not be found in most screening programs. We specified a minimum growth rate of 3% over five years using a more standard growth ratecalculation that only looks at the beginning and ending years’ earnings.

Moderate Price/Earnings RatioGraham specified that the current price to the average of the last three years earnings per share should not exceed 15. You should be able to calculate this fieldwith a screening program that allows you to create custom fields. The goal of this criterion is to create a portfolio with an average price-earnings ratio that doesnot exceed the inverse of the high-grade corporate bond yield.

Moderate Ratio of Price to AssetsGraham felt that current price should be less than or equal to 1½ times book value. However, a low price-earnings ratio could justify a higher price-to-book-valueratio. So he multiplied the price-to-book-value ratio by the price-earnings ratio for a value of 22.5. Most screening programs will not have the product of theprice-earnings ratio and price-to-book ratio calculated, but will allow you to calculate this field. We screened for a value less than or equal to 22.5.

SingleCumulatively Criterion*

Complete Universe 6,970 198Market Sector 198 198Adequate Size of the Enterprise 171 171Financial Condition 163 185Earnings Stability 135 158Dividend Record 127 172Earnings Growth 52 82Moderate Price-Earnings Ratio 29 116Moderate Ratio of Price to Assets 25 110

* Among utility stocks only.

Table 4.Translating Style Into Screening: The Defensive Investor—Utilities

Number of Stocks Passing Screening Criteria

Price to Average Earnings per Share= PRICE / ( ( EPSCON_Y1 + EPSCON_Y2 + EPSCON_Y3) / 3 )

Found in Price & Share Data GroupPRICE The closing market price of the common stock

Found in Income Statement - Annual GroupEPSCON_Y1-Y3 Earnings per share from continuing operations for the

last three fiscal years

Total 5-Yr. Price-toSales Debt to EPS Price- Book- ReturnLast Current Total Growth Earnings Value on Relative

12 Mos. Ratio Assets Rate Ratio Ratio Equity StrengthCompany (Exchange: Ticker) ($ mil) (x) (%) (%) (x) (x) (%) (%) Description

Baltimore Gas & Electric (N: BGE) 2,934.8 1.0 60.0 7.6 12.9 1.40 10.6 –13 Electricity & natural gas in MD

Brooklyn Union Gas Co. (N: BU) 1,256.0 1.5 61.5 5.9 13.2 1.45 10.8 –16 Distribs natural gas at retail

Connecticut Energy Corp. (N: CNE) 236.3 1.0 66.1 3.8 12.2 1.30 10.5 –20 Holding co. for natural gas distrib

Connecticut Water Serv (M: CTWS) 39.4 0.6 70.2 10.8 12.0 1.46 12.2 –14 Holding co. for water supplier

Consolidated Edison Co. (N: ED) 6,536.9 1.1 55.8 4.6 10.1 1.25 12.5 –16 Supplies electric serv to NYC

Consumers Water Co. (M: CONW) 101.8 0.7 75.2 43.3 13.2 1.41 10.6 –15 Holding co. for water utilites

DQE, Inc. (N: DQE) 1,231.3 1.0 68.5 8.0 12.0 1.54 12.9 –7 Holding co. for electric prod & distrib

E’Town Corp. (N: ETW) 108.4 0.5 68.2 4.5 12.9 1.14 8.6 –19 Holding co. for water treat & real est

Energen Corp. (N: EGN) 327.2 0.9 62.8 9.0 13.2 1.42 10.8 –20 Holding co. for prod & distrib of gas

Florida Progress Corp. (N: FPC) 3,055.6 1.1 61.7 3.1 13.0 1.50 11.5 –15 Diversified electric holding co.

General Public Utilities (N: GPU) 3,804.7 0.7 64.2 8.6 8.4 1.25 14.8 –13 Holding co. for electric util in NJ & PA

Houston Industries (N: HOU) 3,680.3 0.3 61.7 27.1 4.7 1.26 9.6 –15 Holidng co for elec util & power gener

IWC Resources Corp. (M: IWCR) 147.1 0.6 74.0 8.1 12.5 1.41 11.5 –17 Holding co. for water systems in Indy.

Montana Power Co. (N: MTP) 974.8 0.9 57.9 6.7 12.2 1.04 10.0 –24 Electric & natural gas systems in MT

Nevada Power Co. (N: NVP) 750.0 1.0 60.6 15.1 12.7 1.37 9.6 –20 Electric serv for Las Vegas and area

North Carolina Nat. Gas (N: NCG) 158.1 1.0 59.3 5.6 12.7 1.70 13.4 –11 Distribs natural gas in NC

Northern States Power (N: NSP) 2,568.6 0.7 63.6 6.6 11.9 1.58 13.0 –17 Distribs electricity & natural gas

Pacific Gas & Electric (N: PCG) 9,621.8 1.0 66.0 7.4 7.6 1.11 14.8 –34 Supplies electricity & gas in CA

PECO Energy Co. (N: PE) 4,186.2 0.5 65.7 35.5 9.4 1.23 12.9 –24 Electric & gas serv to southeast PA

Public Service Co. of NC (N: PGS) 256.0 0.4 64.3 6.6 13.3 1.66 12.3 –15 Distribs natural gas in NC

Public Service Ent. Group (N: PEG) 6,164.2 0.8 64.3 7.3 9.7 1.19 12.2 –26 Holding co. for electric & gas utilities

SIGCORP, Inc. (N: SIG) 338.7 0.8 63.8 4.7 11.7 1.67 12.2 –13 Generates & distribs electricity & gas

SJW Corp. (A: SJW) 97.4 2.0 61.2 3.4 10.9 1.15 10.6 –13 Holding co. for water util & real estate

UNITIL Corp. (A: UTL) 156.7 0.9 67.9 8.3 11.9 1.51 12.7 4 Holding co. for electric & gas in East

Western Resources (N: WR) 1,572.1 0.8 66.7 3.7 11.0 1.21 10.8 –24 Electric & natural gas utility

Median 974.8 0.9 64.2 7.3 12.0 1.40 11.5 –16

Median for All Utilities 806.9 0.9 66.5 2.1 13.2 1.51 11.0 –13

Median for All Stocks 96.5 1.9 55.2 10.7 17.5 2.02 9.1 –7

Found in User Defined Fields (formulas below)Debt to equity <= 2 Debt to equity ratio less than or equal to 2Debt to equity >= 0 Debt to equity ratio greater than or equal to 0Price to Avg EPS <= 15 Price to three-year average earnings per share less than or equal to 15P/E Times P/B <= 22.5 Price-earnings ratio times price-to-book-value ratio less than or equal to 22.5

User Defined Fields

Debt to Equity = LTDEBT_Q1 / (Equity_Q1 + Pref_Q1)

Found in Balance Sheet, Quarterly GroupLTDEBT_Q1 Long-term debt at the end of the most recent fiscal quarterEquity_Q1 Common stock equity at the end the most recent fiscal quarterPref_Q1 Preferred stock equity at the end the most recent fiscal quarter

Price-Earnings Ratio Times Price-to-Book-Value Ratio = PE * PBVPS

Found in Multiples GroupPE Price-earnings ratioPBVPS Price-to-book-value ratio

Table 5.Graham Defensive Growth Utilities Screen Results

Low Price-Earnings RatioA primay factor to indicate that the stock is cheap on a relative basis is tolook only at firms whose price-earnings ratios are in the lowest 10% of theoverall market. We kept screening for a lower and lower price-earningsratio until a maximum value of 10.2 filtered out all but 10% of the stocks.

Financial Condition• Current assets should be at least 1½ times current liabilities—currentratio greater than or equal to 1.5.• Long-term debt no more than 110% of net working capital (currentassets less current liabilities). This is the type of ratio that is unique toGraham and will not typically be found in a screening program. If theprogram allows for custom fields, it should be easy to create. If you mustwork with a fixed set of fields, look for a good measure of financialleverage such as total-debt-to-total-assets or debt-to-equity. We created a

long-term debt-to-working-capital ratio, but were also careful to specify that the ratio not drop below zero or companies with negative working capital would passthe screen.

Earnings StabilityGraham specified some earnings for the common stock for the past five years.

Dividend RecordGraham specified some current dividend payout.

Earnings GrowthGraham specified that last year’s earnings should be higher than that of five years ago.

Moderate Ratio of Price to AssetsGraham felt that current price should be less than or equal to 120% of tangible book value. Not all screening programs will provide a tangible book valuemeasure and may provide book value that includes intangibles such as purchased goodwill.

SingleCumulatively Criterion*

Complete Universe 6,970 6,970Low Price-Earnings Ratio 696 696Financial Condition: Current Ratio 336 3,708Financial Condition: Long-Term Debt 253 3,366Earnings Stability 100 2,391Dividend Record 46 2,599Earnings Growth 38 4,150Moderate Ratio of Price to Assets 16 1,477

rate—a rate that roughly keeps pacewith inflation over the long term. With-out such a criterion, a screen looking forcompanies with low multiples may listcompanies with poor prospects. WhileGraham felt that even companies in astate of “retrogression” could be of in-terest if they could be purchased at alow enough price, this was not the do-main of the defensive investor.

Our filter specified a five-year growthrate in earnings greater than 3%. Whenexamined as single screening criterion,this requirement turned out to be one ofthe most restrictive filters for both mar-ket groups—only half the companiespassed this filter.

Moderate Price/Earnings RatioGraham seemed to express frustra-

tion with the impact of special chargeson the earnings per share calculation.He felt that management’s discretionin establishing reserve accounts makesit difficult for the investor to determinewhether they truly reflect the operation

of the firm for a specific time period. Tohelp get around this problem and tosmooth the impact of the business cycle,Graham often averaged earnings over aperiod of several years. In specifyingthe price-earnings filter, Graham re-quired that the price to average earn-ings over the last three years be no morethan 15. His goal in establishing the cut-off was to produce a portfolio with anaverage multiplier of 12 to 13. Grahamwanted to establish a portfolio that waspriced reasonably compared to the yieldavailable to the AA bond yield. At thetime he wrote the book, investment-grade bonds were yielding 7.5%. Theinverse of this yield (1 divided by 0.075),13.3, determines the overall portfolioprice-earnings ratio objective. With cur-rent investment-grade bond yields at7.6%, there was no need to change thecut-off in today’s market environment.But if bond yields were to go up, aninvestor would require that the price-earnings ratio be lower to consider pur-chase of stocks. Conversely, lower yields

result in a higher price-earnings cut-off,which makes more stocks available forconsideration.

Moderate Ratio of Price to AssetsGraham was a believer in using low

price-to-book-value ratios to selectstocks and normally required a ratiobelow 1.5 for the defensive investor.However, he also felt that a low price-earnings ratio could justify a higherprice-to-book-value ratio. Therefore, herecommended that investors multiplythe price-earnings ratio by the price-to-book-value ratio and not let that valueexceed 22.5—the product of a currentprice-earnings ratio of 15 and a price-to-book-value ratio of 1.5.

For our non-utility segment this filterturned out to be the most restrictiveindividual criterion.

The securities passing all of the filtersare presented in Tables 3 and 5.

Screening: The Enterprising InvestorGraham had a number of recommen-

Table 6.Translating Style Into Screening: The Enterprising Investor—Secondary Companies

Number of Stocks Passing Screening Criteria

Total 5-Yr. Price-toSales Debt to EPS Price- Book- ReturnLast Current Total Growth Earnings Value on Relative

12 Mos. Ratio Assets Rate Ratio Ratio Equity StrengthCompany (Exchange: Ticker) ($ mil) (x) (%) (%) (x) (x) (%) (%) Description

Chicago Rivet & Machine (A: CVR) 24.4 4.4 20.3 9.0 8.2 1.11 13.5 –9 Rivets & rivet setting machinery

Excel Industries (N: EXC) 596.0 2.0 50.2 13.6 8.7 1.05 12.1 –17 Vehicle window & door systems

Federal Screw Works (M: FSCR) 92.3 1.9 49.6 16.1 6.1 0.82 13.3 –4 Mfrs industrial component parts

Howard B. Wolf (A: HBW) 14.8 4.1 20.0 20.5 7.8 0.99 12.6 –21 Women’s fashion apparel

Liberty Homes (M: LIBHA) 164.8 2.9 26.3 38.6 7.7 0.97 12.5 10 Manufactured homes

Maine Public Service Co. (A: MAP) 55.3 2.1 62.5 2.0 7.3 0.59 8.0 –27 Prod & distrib electric in Maine

Martin Industries (M: MTIN) 105.5 2.5 46.3 NA 6.3 1.15 15.1 NA Mfrs heating, office, & leisure prods

Northstar Computer Forms (M: NSCF) 24.2 3.3 33.9 10.7 9.0 1.04 11.8 –34 Custom business & financial forms

Pitt-Des Moines (A: PDM) 461.3 2.0 45.8 10.3 7.7 0.92 12.0 –7 Custom-engineered prods, structures

Seaboard Corp. (A: SEB) 1,076.0 3.3 57.7 13.5 9.7 0.86 8.9 –38 Meat process’g; commodity merch

Sun Television & Applian (M: SNTV) 809.1 2.5 58.9 31.2 6.8 0.46 6.8 –63 Electronics & appliance retailer

The Smithfield Cos. (M: HAMS) 24.1 4.0 14.8 77.9 5.7 1.01 6.1 6 Produces branded food products

The Somerset Group (M: SOMR) 11.2 12.4 23.8 5.1 9.3 1.07 11.4 19 Holds bank holding company

The Tranzonic Cos. (A: TNZ.A) 158.3 2.9 35.3 2.4 8.4 0.68 8.0 –46 Paper, cloth & vinyl products

Treadco, Inc. (M: TRED) 147.7 2.8 29.4 29.2 7.8 0.56 7.1 –64 Tire retreader; truck tire dealer

Westwood Corp. (M: WNMP) 31.2 1.7 38.8 35.4 6.6 0.92 13.9 –25 Marine elecrical distribution prods

Median 98.9 2.8 37.0 13.6 7.7 0.95 11.9 –21

Median for All Stocks 96.5 1.9 55.2 10.7 17.5 2.02 9.1 –7

Found in the Multiples GroupPE <= 10.2 Set maximum price-earnings ratio so that only 10% of the companies pass filter

PBVPS <= 1.2 Price-to-book-value per share less than or equal to 1.2

Found in Ratios GroupCURR_Q1 >= 1.5 The current ratio greater than or equal to 1.5

Found in Income Statement, Annual GroupEPS_12M > 0 Total earnings per share for the latest 12-month periods is positiveEPS_Y1 > 0 Total earnings per share last year is positiveEPS_Y2 > 0 Total earnings per share one year ago is positiveEPS_Y3 > 0 Total earnings per share two years ago is positiveEPS_Y4 > 0 Total earnings per share three years ago is positiveEPS_Y5 > 0 Total earnings per share four years ago is positive

INDCAT_DIV > 0 Indicated yearly dividend greater than $0.00

EPS_Y1 > EPS_Y5 Earnings per share last year greater than earnings per share four year ago

Found in User Defined Field (formula below)Long-Term Debt to Wkg Cap <= 1.1 Long-term debt to working capital less than or equal to 1Long-Term Debt to Wkg Cap >= 0 Long-term debt to working capital greater than or equal to 0

Found in Balance Sheet, Quarterly GroupLTDEBT_Q1 Long-term debt at the end of the most recent fiscal quarterCA_Q1 Total current assets at the end of the most recent fiscal quarterCL_Q1 Total current liabilities at the end of the most recent fiscal quarter

Table 7.Graham Enterprising Growth Screen Results

For Use With AAII’s Stock Investor (Version 2.1)

User Defined Field

Long-Term Debt to Working Capital = LTDEBT_Q1 / ( CA_Q1 - CL_Q1)

dations for how the enterprising inves-tor could hope to profit in the market.His procedure for developing a list ofcandidates among the secondary issuesare of particular interest to the investorwith a screening program. Grahamwould never recommend secondary is-sues for the passive, defensive investor,however, it is an area of interest for theaggressive investor with the time to care-fully study the market and buy thesesecurities at a significant discount.

Graham defined a secondary com-pany as a firm that is not a leader in afairly important industry—however, afirm with an established record ofgrowth is not considered secondary.

Table 6 describes Graham’s approachfor the enterprising investor.

Low Price-Earnings RatioOne of the keys to selecting secondary

issues is to purchase them at a signifi-cant discount—Graham felt these stockstended to almost always trade belowtheir intrinsic value. It was only during

NA = not available

a bull market when little distinction wasmade between primary and secondaryissues that the prices of these secondaryissues approached or exceeded theirintrinsic value.

Graham’s first screen for the enter-prising investor was to look for compa-nies trading with price-earnings ratiosin the lower 10% of all stocks. Someprograms may allow you to specify apercentile ranking for firms when es-tablishing your criteria. If yours doesnot, simply specify a lower and lowerprice-earnings maximum until only 10%of the firms pass the filter. For our ex-ample, a maximum price-earnings ratioof 10.2 filtered exactly 10% of the firmsin Stock Investor.

One warning that Graham gave of thelow price-earnings filter was for cycli-cal firms with widely fluctuating earn-ings. These firms often trade at highprices and low price-earnings ratios ingood years when they should be sold,and low prices and high or non-existentprice-earnings ratios in bad years whenthey should be considered for purchase.For these firms, Graham recommends atest of price low to past average earn-ings as suggested for the defensive in-vestor.

Strong Financial ConditionFor the enterprising investor, Graham

relaxed his tests of the firm’s financialposition. For the current ratio the mini-mum of 1.5 was specified, while long-term debt was not to be higher than110% of net current assets. These re-quirements cut our list of 696 firms downto 253.

Earnings StabilityGraham also loosened the require-

ment of earnings stability, specifyingthat earnings be positive for each of thelast five years; cutting the list further toonly 100 companies.

Dividend RecordFor the enterprising investor, Graham

only specified that firms pay some levelof current dividends—a simple filterthat screens out nearly two thirds of thefirms in Stock Investor when used as astand-alone criterion. As a rule, only

more mature companies past their stageof strong, capital-intensive growth canafford to pay a cash dividend. This filterwill also cut out some low price-earn-ings stocks that are troubled and underfinancial duress. Applying the dividendfilter cut our list down to 46 stocks.

Earnings GrowthWhen it came to earnings growth,

Graham again was less restrictive forthe enterprising investor, requiring onlythat earnings for the latest year be higherthan earnings five years ago. This filtercut our list down to 38 stocks.

Ratio of Price to AssetsThe final criterion specified that the

current price be less than or equal to120% of tangible book value. This re-quirement is more restrictive than forthe defensive investor and makes noadjustment for the level of the price-earnings ratio. Since Graham felt thatsecondary firms normally trade at a dis-count to their intrinsic value, it is notsurprising that a tougher filter was speci-fied.

Applying this final criterion cuts ourfinal list down to 16 firms which arelisted in Table 7.

We kept the comparative statistics thesame for all three screens so that com-parison between the results of the screenscould be made. One of the most tellingfigures is the 52-week relative strengthratio, which compares the performanceof this group to the performance of theS&P 500. A figure of zero would indi-cate performance equal to the S&P whilea negative figure indicates relativeunderperformance. As a group, thesestocks have fallen out of favor with themarket.

In structuring a portfolio of thesecontrarian stocks, Graham stressed theneed for a broad diversified portfolio—a minimum of 10 holdings, but prefer-ably a larger group consisting of about30 of the best prospects selling at signifi-cant discounts from their intrinsic value.

John Bajkowski is editor of Computer-ized Investing and senior financial ana-lyst at AAII.

Graham developed a simple valuationtechnique based on his observations of thestock market and its relationship to the cur-rent earnings level, the expected earningsgrowth rate, and the earnings multiplier. Thehigher the current earnings multiplier, orgrowth rate, the higher the valuation.

His original formulation was based onobservation when the prevailing Aaa-ratedcorporate bond yield averaged 4.4%. Asinterest rates rise, price-earnings ratios fallbecause future earnings are worth less to-day when discount rates rise. The formulabelow adjusts for the current interest rates:

Value = [EPS(8.5 + 2g)]4.4/Aaa

Where g is the expected annual growthrate for the next seven to 10 years, EPS is theearnings per share for the most recent 12months, and Aaa is the current rate on Aaa-rated corporate bonds. The formula (8.5 +2g) × (4.4/Aaa) could be broken out toform the adjusted earnings multiplier.

The table to the right uses this valuationmodel to establish valuation levels for thestocks that passed the three Graham filters.The first three columns examine the price-earnings ratios, comparing the multiplesbased upon the last 12-month earnings pershare, an average earnings per share of thelast three fiscal years, and the Graham cal-culated multiple. In general, the Grahamcalculated multiples indicate that the stockscould support a much higher price thanthese stocks currently have. The primaryfactor affecting this formula is the growthrate assumed for the company. We used thehistorical five-year growth rate. While this isa reasonable starting point, the growth rateused in the formula should be a forecastedgrowth rate over the next seven to 10 years.A number of the companies have historicalgrowth rates above 20%, which cannot besustained for a long period. Graham wasreluctant to use growth rates above 8% in hisanalysis and preferred to lower his estimateto provide a margin of safety. The Grahamimplied growth rate uses the current priceand earnings to back out the growth rateimplied by the market. The final two columnscompare the current price to the valuationprice using the historical growth rate. Whilethe majority of these stocks would seem tobe bargains, remember the analysis is onlyas good as the input. —John Bajkowski

Price-Earnings Ratio (x) Growth Rates (%)Trailing 3-Yr. Last I/B/E/S Grhm Price ($)12-Mo. Average Grhm Five Est’d Imp’d Current Grhm’s

EPS EPS Adj Years Grth Grth (4/30/96) Value DescriptionThe Defensive Investor—Non-UtilitiesBassett Furniture (M: BSET) 15.3 14.5 46.9 36.3 8.0 9.0 25.00 76.53 Mfrs furnitureBlair Corp. (A: BL) 7.4 6.3 12.1 6.2 NA 2.1 23.88 39.20 Men’s & women’s apparel, furnishingsBowne & Co. (A: BNE) 12.2 10.1 30.7 22.3 NA 6.3 17.50 43.96 Document prep and distribExcel Industries (N: EXC) 8.7 9.1 20.6 13.6 10.0 3.2 13.25 31.42 Vehicle window & door systemsHaverty Furniture Cos. (M: HAVT) 13.2 13.6 13.0 7.0 15.0 7.1 13.88 13.68 Specialty furniture retailerHudson Foods (N: HFI) 11.2 13.1 33.8 25.0 12.5 5.4 13.00 39.29 Produces poultry & meat prodsJustin Industries (M: JSTN) 12.5 9.9 36.9 27.7 10.5 6.5 11.75 34.78 Mfrs bldg materials, footwear; book publrOshkosh Truck Corp. (M: OTRKB) 14.3 12.6 51.3 40.1 NA 8.1 14.88 53.41 Mfrs heavy-duty vehicles, transport equipPACCAR Inc. (M: PCAR) 7.7 9.7 42.4 32.4 8.0 2.3 50.00 275.84 Mfrs heavy-duty on- & off-road trucksPhelps Dodge Corp. (N: PD) 6.8 12.7 16.7 10.2 10.0 1.6 72.50 178.19 Operates copper mining co.& prods mfrPitt-Des Moines (A: PDM) 7.7 11.6 16.8 10.3 NA 2.3 42.75 94.01 Custom-engineered prods, structuresStandard Motor Products (N: SMP) 14.3 11.9 23.2 15.8 8.5 8.1 17.63 28.56 Mfrs replacement parts for auto systemsThor Industries (N: THO) 14.0 12.4 36.4 27.2 NA 7.8 19.13 49.89 Recreational vehicles & small busesMedlian 12.2 11.9 30.7 22.3 10.0 6.3 17.63 43.96The Defensive Investor—UtilitiesBaltimore Gas & Electric (N: BGE) 12.9 13.4 13.7 7.6 3.4 6.8 26.00 27.72 Electricity & natural gas in MDBrooklyn Union Gas Co. (N: BU) 13.2 13.9 11.7 5.9 4.0 7.1 25.50 22.68 Distribs natural gas at retailConnecticut Energy Corp. (N: CNE) 12.2 12.6 9.3 3.8 3.5 6.3 19.75 15.01 Holding co. for natural gas distribConnecticut Water Serv (M: CTWS) 12.0 12.4 17.4 10.8 3.0 6.1 26.00 37.82 Holding co. for water supplierConsolidated Edison Co. (N: ED) 10.1 10.3 10.2 4.6 1.0 4.4 29.63 30.02 Supplies electric serv to New York CityConsumers Water Co. (M: CONW) 13.2 12.6 55.0 43.3 3.0 7.1 17.63 73.78 Holding co. for water utilitesDQE, Inc. (N: DQE) 12.0 14.1 14.1 8.0 4.8 6.1 26.13 30.92 Holding co. for electricity prod & distribE’Town Corp. (N: ETW) 12.9 12.4 10.1 4.5 2.0 6.8 27.88 21.99 Holding co. for water treat & real estateEnergen Corp. (N: EGN) 13.2 11.9 15.3 9.0 6.0 7.1 22.75 26.54 Holding co. for prod & distrib of gasFlorida Progress Corp. (N: FPC) 13.0 13.9 8.5 3.1 3.0 6.9 32.50 21.28 Diversified electric holding co.General Public Utilities (N: GPU) 8.4 12.0 14.8 8.6 3.0 2.9 31.63 56.24 Holding co. for electric utilities in NJ & PAHouston Industries (N: HOU) 4.7 12.5 36.3 27.1 3.5 –0.1 21.00 161.90 Holidng co for elec utility & power genr’nIWC Resources Corp. (M: IWCR) 12.5 13.6 14.3 8.1 3.5 6.5 20.50 23.45 Holding co. for water systems in Indy.Montana Power Co. (N: MTP) 12.2 10.9 12.6 6.7 3.0 6.3 21.88 22.70 Operates elec & gas systems in MTNevada Power Co. (N: NVP) 12.7 11.6 22.4 15.1 3.5 6.7 20.00 35.18 Electric serv for Las Vegas and areaNorth Carolina Nat. Gas (N: NCG) 12.7 13.9 11.4 5.6 6.5 6.7 25.25 22.70 Distribs natural gas in NCNorthern States Power (N: NSP) 11.9 13.4 12.5 6.6 3.0 6.0 46.63 49.12 Distribs electricity & natural gasPacific Gas & Electric (N: PCG) 7.6 9.0 13.4 7.4 1.0 2.2 22.63 40.33 Supplies electricity & natural gas in CAPECO Energy Co. (N: PE) 9.4 10.9 46.0 35.5 4.0 3.8 24.88 121.51 Electric & gas serv to southeast PAPublic Service Co. of NC (N: PGS) 13.3 14.7 12.5 6.6 7.0 7.2 15.75 14.82 Distribs natural gas in NCPublic Service Ent. Group (N: PEG) 9.7 9.8 13.3 7.3 2.0 4.0 26.13 36.24 Holding co. for electric & gas utilitiesSIGCORP, Inc. (N: SIG) 11.7 13.5 10.3 4.7 3.0 5.8 33.38 29.54 Generates & distribs electricity & gasSJW Corp. (A: SJW) 10.9 11.3 8.8 3.4 NA 5.1 38.63 31.45 Holding co. for water utility & real estateUNITIL Corp. (A: UTL) 11.9 12.3 14.5 8.3 NA 6.0 22.38 27.32 Holding co. for electric & gas in EastWestern Resources (N: WR) 11.0 10.7 9.2 3.7 3 5.2 29.63 24.85 Electric & natural gas utilityMedian 12.0 12.4 13.3 7.3 3.0 6.1 25.50 29.54The Enterprising InvestorChicago Rivet & Machine (A: CVR) 8.2 12.8 15.3 9.0 NA 2.8 31.00 58.30 Rivets & rivet setting machineryExcel Industries (N: EXC) 8.7 9.1 20.6 13.6 10.0 3.2 13.25 31.42 Vehicle window & door systemsFederal Screw Works (M: FSCR) 6.1 8.5 23.5 16.1 NA 1.0 23.50 90.25 Mfrs industrial component partsHoward B. Wolf (A: HBW) 7.8 9.3 28.6 20.5 NA 2.5 6.75 24.65 Women’s fashion apparelLiberty Homes (M: LIBHA) 7.7 10.3 49.6 38.6 NA 2.4 11.00 70.45 Manufactured homesMaine Public Service Co. (A: MAP) 7.3 5.6 7.2 2.0 NA 2.0 17.25 17.01 Prod & distrib electric energy in MaineMartin Industries (M: MTIN) 6.3 7.7 NA NA 15.0 1.2 9.25 NA Mfrs heating, office, & leisure prodsNorthstar Computer Forms (M: NSCF) 9.0 9.6 17.3 10.7 NA 3.5 7.00 13.50 Custom business & financial formsPitt-Des Moines (A: PDM) 7.7 11.6 16.8 10.3 NA 2.3 42.75 94.01 Custom-engineered prods, structuresSeaboard Corp. (A: SEB) 9.7 10.4 20.5 13.5 NA 4.0 213.00 453.39 Meat process’g; commodity merchan’gSun Television & Applian (M: SNTV) 6.8 4.3 41.0 31.2 10.0 1.6 4.00 24.22 Electronics and home appliance retailerThe Smithfield Cos. (M: HAMS) 5.7 13.6 95.1 77.9 NA 0.6 11.13 187.39 Produces branded food productsThe Somerset Group (M: SOMR) 9.3 11.3 10.8 5.1 NA 3.8 15.00 17.43 Holds bank holding companyThe Tranzonic Cos. (A: TNZ.A) 8.4 9.0 7.7 2.4 NA 3.0 10.50 9.55 Paper, cloth & vinyl productsTreadco, Inc. (M: TRED) 7.8 6.2 38.7 29.2 NA 2.4 7.25 36.02 Tire retreader; truck tire dealerWestwood Corp. (M: WNMP) 6.6 5.0 45.9 35.4 NA 1.3 1.63 11.48 Marine elecrical distribution prodsMedian 7.7 9.2 20.6 13.6 10.0 2.4 11.07 31.42Median for All Stocks 17.5 14.7 17.3 10.7 15.0 6.8 14.25 16.80

Putting a Value on a Stock

NA = not available

would never reveal.However, even with its quantitative

focus, screening is an art as much as ascience. Designing effective screens re-quires a clear objective, basic knowl-edge of finance, and a clear understand-ing of the strengths and weaknesses ofthe screening program.

Defining a Clear ObjectiveThe first step in constructing a screen

is establishing a clear, focused, narrowly-defined objective. Common objectivesinclude seeking out small, emerginggrowth companies, out-of-favorcontrarian plays, and high-yielding blue-chip firms.

On the surface, it would seem that afilter that combines a mix of unrelatedobjectives would be the ultimate screen.In reality, screens that look for extremevalues defined by a combination of con-trasting objectives typically turn up com-panies in odd situations, not necessarily

In the January 1996 issue of the AAIIJournal, Maria Crawford Scott analyzedthe investment approach of T. RowePrice. T. Rowe Price, who died in 1983,developed his long-term buy- and-holdgrowth philosophy in the 1930s whenthe prevailing approach was to try totime the market and greater emphasiswas placed on investing in income-producing stocks. In this article, we willdevelop and apply the T. Rowe Pricestock screen while discussing the over-all process of developing screening fil-ters.

Screening’s AppealStock screening’s appeal to investors

is the speed with which a well-designedfilter can highlight stocks that meritfurther analysis. A screening systemusing a broad database can identifyinteresting companies that a haphaz-ard system of relying on tips fromfriends and investment professionals

FEATURE

investment gems. The purpose of screen-ing is to come up with a list of compa-nies sharing a common characteristicthat you believe makes them worthy offurther analysis.

Table 1 presents the summary of theT. Rowe Price investment approach. Theobjective of our screen can be foundwithin the philosophy and style sec-tion—locate stocks of companies withlong-term earnings growth prospects inthe early growth stages of their life cycle.

Primary FactorsThe primary screening criteria should

flow naturally from your objective andshould attempt to filter only those com-panies that meet your objective. In thecase of the T. Rowe Price growth screen,our primary criteria should provide uswith a list of companies that are in thegrowth stage of their life cycle, not ma-ture cyclical firms.

The criteria for initial consideration

Screening on Growth: The T. Rowe Price Approachby John Bajkowski

Table 1.The T. Rowe Price Approach in Brief

Secondary factors:• Management, which is key. Look for management with substantial

interest in the firm and that appears to be planning intelligently for thefuture

• Strong finances: increase in capital from retained earnings; availabilityof additional financing

• Relatively high return on invested capital• Favorable profit margins relative to industry; favorable trend in profit

margins• Absence of cut-throat competition in the industry• Little government interference• Good labor relations, but total payroll not large relative to gross

revenue

Stock monitoring and when to sell:• Monitor for changes in trends of sales, profit margin and return on

invested capital• Monitor for changes in management, competitive stance, regulatory

changes, and changes in industry• Sell when company no longer fits definition of “growth” company,

when the outlook does not continue to be favorable, when there is abreak in trend of sales, profit margin, and return on invested capital

• Trim when prices reach excessively high levels, as measured by price-earnings ratios relative to historical levels

Philosophy and styleInvestment in stocks of companies with long-term earnings growthprospects, under the theory that growth in earnings will ultimately bereflected in growth of market values and dividends; and investment instocks in the early, growth stages of their life cycle before they become“glamorized.”

Universe of stocksNo restrictions, but smaller capitalization stocks viewed as particularlyfertile ground.

Criteria for initial consideration• Earnings per share increasing at the peak of each succeeding major

business cycle, as well as increased projected earnings per share atpeak of future business cycles. Earnings are examined relative toindustry and overall market to find distortions caused by business cyclesto determine the real trend in earnings, and to rule out non-growth“cyclicals.” To ensure long-term growth of earnings, companies shouldbe in growing industries. Areas of possibility include: new industries,divisions of old industries experiencing vigorous growth as a result ofnew products or new uses for old products, and specialty industries withexpanding products and markets.

• Long-term earnings per share growth greater than inflation.• Price-earnings ratio not high relative to historical average.

Translating Style Into Screening

Primary Factors:Earnings per share increasing at the peak of each succeeding major business cycle

Few screening programs maintain enough data to establish such a perspective. This type of filter would be best applied manually after thecomputerized screening process.

To ensure long-term growth of earnings, companies should be in growing industriesIn screening, you can include only perceived growth industries or exclude industries with few growth prospects. However, there are often segmentsof growth within a mature industry. Alternatively, companies with few growth opportunities can be excluded manually after the screen is performed.

Long-term earnings per share growth greater than inflation.The current low inflationary environment (3%–4%) makes this a rather loose filter in screening for growth stocks. A requirement of 8.1% in past growthin earnings per share, which was the mid-point for the companies in Stock Investor, is used.

Price-earnings ratio not high relative to historical averageUse a filter in which the price-earnings ratio based upon trailing 12 months’ earnings is below the average price-earnings ratio over the last five years.The hidden screen behind this filter is positive earnings for each of the last five years and a requirement that the company have five years of pricehistory. The calculation of the five-year price-earnings average required positive earnings over each of the last five years (price-earnings ratios arenot meaningful for negative numbers) as well as price data for each of those years. The historical average can be skewed upward by one year ofweak earnings, so a screen excluding companies whose price-earnings ratio was above 40 for any of the last five years was added.

Secondary Factors:Substantial management interest in the firm

Screen for companies with insiders holding at least 20% of outstanding shares.Strong finances—increase in capital from retained earnings

Given the other filters, not a very restrictive screen. However add a requirement for payout ratio to be below 100%. Modern interpretations maycall for positive cash flow or free cash flow, which would be more restrictive.

Strong finances—availability of additional financingMore of a qualitative examination than a quantitative screen. No screen added; however, if you are interested in considering the type of factors alender would require, a focus on financial leverage, cash positions, cash flow, and coverage ratios would be appropriate.

Relatively high return on invested capitalT. Rowe price examined return on invested capital (earnings before interest, taxes and dividends divided by equity and long-term debt). A screenfor company return on assets higher than industry norms is substituted.

Favorable profit margins relative to industryScreen for gross and net profit margins to exceed industry norms.

Favorable trend in profit marginsScreen for gross and net profit margins below levels last year and five years ago.

Absence of cut-throat competition in the industryScreen for above-industry average margins and improving profit margins.

Little government interferenceDifficult screen to implement except by excluding industries with known regulatory issues such as the utility group.

Good labor relations, but total payroll not large relative to gross revenuesMore of a qualitative evaluation than a quantitative analysis. Reference to payroll relative to revenues may also be better suited toward industrialfirms versus those in the service sector.

presented in Table 1 conceptually de-fine our primary screening factors. Con-verting these concepts into quantitativescreening criteria can be difficult due tothe inherit limitations of screening pro-grams.

T. Rowe Price suggests identifyinggrowth companies by looking for in-creasing earnings per share at the peakof each succeeding major business cycle,as well as increased projected earningsper share at the peak of future businesscycles. I am not aware of any screeningprograms geared toward individual in-

vestors that provide data going back farenough in time to screen for this occur-rence. We are currently in an economicexpansion, which began in March of1991. The peak of the previous expan-sion occurred in July of 1991, while youwould have to go back nearly 15 yearsfor the peak of a prior expansion, whichoccurred in July of 1981. (Note: TheSurvey of Current Business publishedby the U.S. Department of Commerceis a good source of data on measures ofeconomic activity.)

The filter make sense financially—it

is trying to exclude cyclical companiesthat currently may look like growthstocks—however, it can not be reason-ably applied automatically. After a pro-longed economic expansion, the growthrates of cyclical companies may lookextremely high because you are com-paring earnings against a very low baselevel. Current five- and three-yeargrowth rates are well into the doubledigits for many cyclical firms. Chrysler’sfive-year annual growth rate in earn-ings from continuing operations is48.7%, and its three-year annual growth

Rowe Price also expressed interest inlooking at divisions of old industriesexperiencing vigorous growth as a re-sult of new products or new uses for oldproducts. In the end, when applyingthis screen it may be better to screen forreasonable levels of growth and manu-ally exclude firms that do not have stronglong-term growth potential.

Growth Greater Than InflationT. Rowe Price was an early proponent

of buying a good company at reason-able price and holding it for a long time.He recognized the potential for inflationto erode the value of his holdings andspecified that earnings per share shouldbe increasing faster than the rate of infla-tion.

In the last five years, inflation as mea-sured by the consumer price index hasgrown at about 3.5% per year—roughlyin line with the long-term inflation rate.Specifying a minimum historical growthrate of 3.5% or 4.0% would lead to a veryunrestrictive screening filter. The me-

dian five-year earnings per share growthrate for the 6,600 companies in AAII’sStock Investor program is 8.1%. Since weare trying to identify companies withabove-average growth rates and havenot specified any other minimumgrowth levels, a minimum growth rateof 8.1% was specified. This filter, whenapplied to AAII’s Stock Investor data-base, reduced the number of compa-nies to 2,693 from 6,636. More than halfthe number of companies were excludedby this filter because not all firms havebeen around long enough to calculatethe growth rate.

Growth at a Reasonable PriceWhile T. Rowe Price concentrated on

growth, he would not pay any price forthat growth. When looking at price-earn-ings ratios, he would prefer not to buystocks selling at levels high relative totheir historical average. A current price-earnings ratio lower than its historicalaverage would be a potential sign that astock is undervalued, while a current

rate comes in even higher at 87.2%. Salesgrowth over the same time periodsclocks in at 8.2% annually over fiveyears and 21.1% over three years. In-dustries that make heavy use of fixedassets and financial leverage can showtremendous earnings growth once theyget past their breakeven point.

Growth firms represent companiesgrowing well beyond the level of theoverall economy. A mature firm suchas Chrysler, in a cyclical industry suchas autos, will roughly grow and con-tract at the rate of the overall economy.While it may be possible to expandmarket share and grow at a slightlyhigher rate than the industry, there arestill severe restraints on the firm’s long-term growth potential.

In screening for growth stocks, it maybe possible to focus on industries in theearly stages of their life cycle, which canbe fairly easy to accomplish with screen-ing programs. Alternatively, you canexplicitly exclude industries known tobe at their mature stage; however, T.

Definitions of Screens and Terms

EPS Growth Rate—Last Five Years: Annual growth in earnings per share from total operations over the last five fiscal years. A measure of how successfulthe firm has been in generating the bottom line, net profit.

Sales Growth Rate—Last Five Years: Annual growth in total sales over the last five fiscal years. Provides a confirmation of the quality of the historical earningsper share growth rate.

Current Price-Earnings Ratio: Market price per share divided by the most recent 12 months’ earnings per share from total operations. A measure of themarket’s expectation regarding the firm’s earnings growth potential and risk. Firms with very high price-earnings ratios are being valued by the market on the basisof high expected growth potential.

Price-Earnings Ratio—Five-Year Average: An average of the high and low price-earnings ratio for the past five years. Provides a base from which to comparethe current level of the price-earnings ratio. Trailing 12-month ratios below the five-year average may point to an undervalued stock.

% of Shares Held by Insiders: The percent of common stock held by all the officers and directors as a group plus beneficial owners who own more than5% of company’s stock as disclosed in the most recent proxy statement. A high percentage should indicate that management is as concerned about theperformance of the stock as are other shareholders.

% of Shares Held by Institutions: The percent of common stock held by all reporting institutions. Provides an indication of the institutional interest in thecompany.

Return on Assets—Trailing 12 Months: Net income for the most recent 12 months divided by the total assets reported last quarter. Provides a measure of themanagement’s efficient use of assets.

Net Profit Margin—Trailing 12 Months: Net income divided by total sales over the last 12 months. When compared against industry norms and over timeprovides a indication of the competitive nature of the industry and the company’s competitive position.

Gross Profit Margin—Trailing 12 Months: Gross income divide by total sales for the last 12 months. When compared against industry norms and over timeprovides a indication of the competitive nature of the industry and the company’s competitive position.

5-Year 5-Year % of % of Trailing 12 MonthsGrowth Rate Current Avg. Shares Shares Return Net Grossin in P/E P/E Held by Held by on Profit Profit

EPS Sales Ratio Ratio Insiders Institutions Assets Margin MarginCompany (%) (%) (X) (X) (%) (%) (%) (%) (%) Description

Illinois Central Corp. 96.1 8.3 13.7 14.5 22.1 94.1 9.4 20.1 56.3 Operates railroad track

The Smithfield Cos. 77.9 5.3 5.3 10.9 64.9 14.7 6.5 3.9 37.0 Brand food products

HMI Industries 39.9 27.7 8.0 12.6 60.3 16.9 8.6 5.2 32.7 Mfrs vacuum cleaners

Xtra Corporation 35.1 11.3 12.6 14.1 41.2 na 3.8 15.2 76.8 Transport equip leasing serv

Dawson Geophysical Co. 34.9 29.5 15.0 17.3 30.1 37.4 6.7 7.7 28.8 Seismic data for locating oil & gas

IHOP Corp. 31.1 11.5 12.9 23.5 44.4 95.0 6.8 10.1 49.0 Int’l House of Pancake restaurants

Century Telephone Ent. 29.9 20.2 15.9 25.0 46.0 83.7 6.6 20.1 50.9 Regional telecomm company

Cohu 29.7 23.9 11.6 13.7 21.2 33.6 20.3 11.6 40.0 Closed circuit TV cameras and sys

Wendy’s International 24.1 5.5 19.1 21.5 21.0 58.3 10.2 7.7 47.7 Fast-food restaurants

Max & Erma’s Restaurants 22.6 16.2 11.4 18.8 53.9 18.5 4.4 3.3 74.0 Family restaurants

Astronics Corp. 21.4 2.4 9.6 13.2 30.0 5.4 7.4 6.0 30.8 Prods/process in electro sys

Howard B. Wolf 20.5 7.7 7.9 11.5 43.2 na 10.1 6.1 35.2 Women’s fashion apparel

UNR Industries 19.4 –0.7 10.7 11.1 72.0 21.3 16.8 10.3 24.1 Mfrs welded steel tubing

Drew Industries 19.0 –2.1 9.3 12.9 41.5 20.7 26.7 7.5 27.1 Aluminum & vinyl windows & doors

TriMas Corp. 18.0 19.3 12.3 20.7 47.1 36.3 9.1 10.0 32.7 Mfrs industrial prods

Albertson’s 16.6 9.9 18.9 22.6 28.0 46.1 11.4 3.7 25.6 Superstores & supermarkets combo

Comair Holdings 15.3 18.0 17.1 18.5 26.0 55.5 10.9 10.4 62.6 Scheduled air transport

Regal-Beloit Corp. 14.5 7.7 12.0 20.6 25.0 62.8 17.9 10.9 29.8 Power transmission prods

Benetton Group S.P.A. 14.1 11.0 13.8 15.8 80.0 na 6.7 8.5 41.2 Casual apparel

Varlen Corp. 14.0 8.5 6.9 16.5 44.5 81.1 8.3 4.9 24.6 Engineered industrial prods

Baldor Electric 13.1 8.2 18.7 21.8 43.0 37.8 10.3 6.7 29.3 Electric motors & drives

Tuscarora Incorporated 12.5 13.8 14.2 18.3 32.7 19.9 8.2 5.6 24.3 Prods for protective pkg

Int’l Flavors & Fragrance 10.8 8.6 22.1 23.6 25.0 54.3 16.6 17.6 48.5 Mfrs flavors & fragrances

Landauer 9.1 6.6 17.5 18.2 45.7 55.0 26.0 29.6 70.9 Analyzes hazard exposure

Chicago Rivet & Machine 9.0 4.7 7.9 13.7 31.9 19.8 10.8 9.1 35.6 Rivets & rivet setting machines

Table 2.T. Rowe Price Screen Results

na= not available

price-earnings ratio that is high com-pared to its historical average mightindicate an overvalued firm. This as-sumes that the growth prospects of thefirm have not changed fundamentallyover the period of study and that basedupon the historical relationship of priceto earnings, the market is not correctlydiscounting the future earnings poten-tial of the firm.

Applying the filter with a require-ment that the current price-earningsratio be below the average price-earn-ings ratio reduced the number of pass-ing companies from 2,693 to 595. This isa restrictive screen for a number of rea-sons. With the market at record highs,many firms are trading with price-earn-ings ratios well above their normal lev-els.

Many firms also do not have a five-year average price-earnings ratio. Tocalculate this figure a company needsfive years of price data and five years ofpositive earnings. Negative earningslead to non-meaningful price-earningsratios, so positive earnings for each ofthe last five years are required. Thisillustrates that a seemingly simple singlefilter can in fact have many built-in hid-

Note: We compared the actual holdings of T. Rowe Price funds with the stocks that appear on this list. Only five of the stocks in this list are held in T. RowePrice mutual funds. Three stocks are held in T. Rowe Price Equity Index, which attempts to emulate the S&P 500. TriMas Corp. was the most widely heldstock in the list, appearing in five T. Rowe Price mutual fund portfolios. It should be noted that the founder sold his mutual fund company prior to his death,so there may be no actual correspondence between his original investment practices and the manner in which the funds are managed today.

den requirements.Beyond negative earnings, unusually

low earnings may also throw off stan-dard price-earnings ratio screens. Short-term drops in earnings due to eventssuch as special charges, extraordinaryevents, or even general slowdowns canlead to unusually high price-earningsratios. As long as the market perceivesthe earnings decline as temporary, thehigh price-earnings ratio will be sup-ported. Because the average price-earn-ings screen relies on a normal situation,these outlier ratios must be excluded.When performing a hands-on evalua-tion you can manually exclude yearswith negative or unusually low earn-ings. However, when screening a largeuniverse of stocks, it is best to establishconditioning criteria that eliminate com-panies with extreme price-earnings ra-tios.

Five filters were added that excludedcompanies whose price-earnings ratioexceeded 40 for any of the last five years.This reduced the number of passingcompanies to 300 from 595.

Secondary Factors: ManagementT. Rowe Price felt that good manage-

ment was an important considerationwhen selecting firms, believing thatmanagers should have substantial in-terest in the firm and not receive com-pensation primarily from big salariesand pensions. A filter requiring thatinsiders own at least 20% of outstand-ing shares dropped the number of com-panies down to 211 from 300. The pri-mary weakness with this filter is thatthe SEC defines an insider as a “benefi-cial owner” who maintains at least a 5%position in the firm. Under this defini-tion, an investment adviser who ownsat least 5% of a firm’s shares will beincluded as an insider. To make the bestuse of this screen, it would be a goodidea to look at the proxy statements ofthe companies that pass the completescreen and see how many shares theofficers and directors own.

Much is made out of owning compa-nies with good management. Certainly,a good management team is essentialfor the long-term viability and growthof the firm. However, many investors

in effect double count the benefit of goodmanagement. With a good managementteam the company should have above-average growth, profitability better thanother firms in the industry, improvingprofit margins, etc. In other words, theeffect of good management should beapparent in the results.

Strong FinancesT. Rowe Price thought that a com-

pany should have strong finances to beconsidered for purchase. The failing ofmany growth firms is the inability toobtain further capital for future expan-sion. T. Rowe Price focused in on theability to increase capital from retainedearnings and on the ability to obtainfurther financing. These two criteria dealwith the primary sources of companyfunds—internal cash flow from opera-tions and additional outside sources ofcapital. At its most basic level, an in-crease in capital from retained earningssimply means the company is profitableand is not paying out all of its earningsin the form of dividends. Since we havealready screened for positive earnings,we simply need to screen for a payoutratio (dividend per share divided byearnings per share) below 100%. Ap-plying this filter had no effect on theoverall screen. Since earnings can bereduced by non-cash charges such asdepreciation, an alternative screen couldlook at actual cash flow measures andrequire that this be positive.

Availability of additional external fi-nancing is more of a qualitative exami-nation than a quantitative screen. Itwould consider the ability of the com-pany to raise capital from the equitymarkets by floating additional stock.Here the market environment for initialpublic offerings would play a role. Manytechnology firms had to scrap or delayplans for floating common stock latelast year when the technology stocks asa whole retreated. Beyond equity financ-ing, companies can try to float debt orlook toward bank lending. If you wereinterested in considering the type of fac-tors a lender would require, you couldadd filters that look at the current levelsof debt (financial leverage), cash posi-tions, cash flow, and coverage ratios.

Return on Invested CapitalT. Rowe Price looked for high and

expanding return on invested capital(earnings before interest, taxes and divi-dends divided by common equity, pre-ferred stock and long-term debt) whenselecting stocks. He was trying to cap-ture firms that made effective use oftheir capital by keeping a lid on produc-tion costs and expanding into new mar-kets. Stock Investor does not include thiscalculation, so return on assets was sub-stituted. Return on assets is calculatedby dividing net income by total assets,and using it represents the type of ad-justments often required when apply-ing a computer screen. Return on assetswas used instead of return on equitybecause it includes preferred stock, long-term debt, common equity, and short-term liabilities, which more closelymatch the return on invested capitalcalculation. Our screen looked for com-pany return on assets higher than in-dustry norms, cutting the number ofcompanies down to 174 from 211.

Favorable Profit MarginsFavorable profit margins may provide

an indication that a company has a com-petitive advantage over its peers. How-ever, because profit margins are veryindustry specific, they must be com-pared against industry norms. Twoscreens, one that looked for gross mar-gins above industry medians and an-other that compared net profit marginsagainst industry medians, were in-cluded.

Beyond margins that are currentlyhigh, it is important that the companybe able to maintain its profit margins. Adeclining profit margin can point to in-creased competition and lower prices orweak cost control. Therefore, screenswere added that looked for current netand gross margins above levels five yearsago. These screens reduced the numberof companies down to the 25 firms dis-played in Table 2.

Absence of Cut-Throat CompetitionThe screens above that compared gross

and net profit margins against industrynorms and required improving mar-gins should help to pinpoint the level of

competition. There really is no need toadd additional screens for this factor.After the screen, however, it would be agood idea to study the industries of thepassing companies to gain a feel forhow competitive the industry might beand whether there are any changes onthe horizon that might effect the firmdown the line. For example, Wendy’s,which passed all of the filters and isoperating in the fast-food industry,would seem to be operating in a rathercompetitive field.

Little Government InterferenceThe level of government interference

is also a difficult screen to implement.The best mechanical approach wouldbe to exclude industries with knownregulatory issues, such as the utilitygroup. Alternatively, companies thatpass the screen can be examined one byone and the impact of government in-terference could be assessed individu-

ally. This is the type of discussion thatwould be covered in the company’s an-nual 10-K filings with the SEC. Thesecan be obtained directly from the com-pany or can be downloaded from theInternet for a large number of publiccompanies at the SEC Web site (http://www.sec.gov).

Good Labor RelationsMuch of T. Rowe Price’s work was

written when industrial companies werethe strength and growth of our economy.With the shift toward the service sectorin the U.S., some of the rules tend to dealmore with the analysis of the typicalunion-based work force—although eventoday major league baseball has its shareof labor problems. This screen is alsomore of a qualitative evaluation than aquantitative analysis. While it is an im-portant concern, it would be best ad-dressed when analyzing the passingcompanies individually.

The ResultsTable 2 presents the results of our

interpretation of the T. Rowe Pricescreen. The companies are ranked bygrowth in earnings per share. Whilethe T. Rowe Price style is associatedwith growth, many of the rules weused also focused on value. Many ofthe final companies appear to be in-dustrial firms in cyclical industries. Fur-ther analysis would be required to seeif they possess qualities that providethe potential for long-term growth.

Screening can be a very rewardingprocess, but looking at the results of ascreen such as this highlights thatscreening is the first stage in the invest-ment process. In-depth analysis of thecompany and its industry must be per-formed before your money is invested.

John Bajkowski is editor of Comput-erized Investing and senior financialanalyst at AAII.

For the Computerized Investor Using AAII’s Stock Investor

Screening Criteria:Found in Growth Rates GroupEPS_G5F > 8.10 Five-year earnings per share growth rate is greater than the median

growth rate for all stocks

Found in Multiples GroupPE < PE_A5Y Price-earnings ratio is greater than five-year average price-earnings ratio

Found in User Defined Fields (formulas below)PE Y1 < 40 Price-earnings ratio last year is less than 40PE Y2 < 40 Price-earnings ratio one year ago is less than 40PE Y3 < 40 Price-earnings ratio two years ago is less than 40PE Y4 < 40 Price-earnings ratio three years ago is less than 40PE Y5 < 40 Price-earnings ratio four years ago is less than 40

Found in Price & Share Data GroupSHRINSD > 20 Percentage of insider ownership is greater than 20%

Found in Ratios GroupPAYOUT_12M < 100 12-month payout ratio is less than 100%ROA_12M > INDUSTRY.ROA_12M (N) 12-month return on assets is greater than that of narrow industry groupNPM_12M > INDUSTRY.NPM_12M (N) 12-month net margin is greater than that of narrow industry groupNPM_12M >= NPM_Y1 12-month net margin is greater than that of net margin last yearNPM_12M > NPM_Y5 12-month net margin is greater than that of net margin four years agoGPM_12M > INDUSTRY.GPM_12M (N) 12-month gross margin is greater than that of narrow industry groupGPM_12M >= GPM_Y1 12-month gross margin is greater than that of gross margin last yearGPM_12M > GPM_Y5 12-month gross margin is greater than that of gross margin four years ago

User Defined Fields:PE Y1 = PRICE_Y1/EPS_Y1PE Y2 = PRICE_Y2/EPS_Y2PE Y3 = PRICE_Y3/EPS_Y3PE Y4 = PRICE_Y4/EPS_Y4PE Y5 = PRICE_Y5/EPS_Y5

Price fields found in Price & Share Data GroupEPS fields found in Income Statement, Annual Group

Investors may be tempted to scan for defensive securitiesconsidering the recent market volatility and rich valuationlevels of the market compared to historical benchmarks (acurrent price-earnings ratio of 18.9 and dividend yield of 2.2%versus a 15.1 average price-earnings ratio and 3.7% averagedividend yield since the 1960s). Many consider the ultimatesafety net for a firm to be a high cash position. With a large cashhoard, a firm should be able to weather an economic storm, orso the reasoning goes. However, there are difficulties in identi-fying firms that have truly strong financial positions and gaininga feel for the long-term prospects of these cash-rich firms.

Strengths and Drawbacks of a High Cash Position

A healthy cash position provides important flexibility andsafety to a firm. Cash-rich firms should be able to more easilymeet their debt obligations, decreasing the probability of acreditor weakening the position of the equity investors or evengaining control of the firm. During an economic slow-down, cashallows a cyclical firm to continue its research and developmentefforts, as well as undertake capital expansion or productivityimprovements, in anticipation of an economic rebound.

Firms with excess cash positions can also elect to distributethe cash to shareholders in the form of dividends—althoughdouble taxation is a weakness to the high payout strategy. Thefirms pay corporate taxes when they earn the money andshareholders must pay taxes at their marginal tax rate when theyreceive the dividend. To avoid the double tax, many firms havechosen to use excess cash to repurchase shares on the openmarket. This helps to boost the share price in the short term byproviding demand for shares. And with fewer outstandingshares, the same level of net income boosts earnings per share.

Firms with excess cash can also attempt to use the cash

strategically to broaden their product lines or diversify into newareas. This can be accomplished either through direct capitalinvestment or the outright purchase of another firm.

Cash-rich firms can also be attractive acquisition candidates.While much more common in the leveraged buyout craze of the1980s, the cash prize reduces the actual purchase price of thefirm and the cash flow that allowed the cash hoard to be builthelps to service the debt incurred acquiring the firm.

A high cash position can also be a disadvantage. Cash isgenerally defined as cash plus marketable securities that arereadily convertible into cash. This would consist of bank depos-its and short-term instruments such as Treasury bills. The cashposition may reduce profitability if it earns a lower rate of returnthan other assets in the company. One would expect anycorporate investment to earn more than the money market ratein the long run.

When finding firms with large cash balances, the criticalquestion becomes: Why are they holding on to the cash? Oftenit points to a firm in a mature industry with little growth pros-pects. The firm may have reasonable profit margins, but littleneed for additional capital. For such a firm, the need for a goodmanagement team is especially important. Unlike a start-up,which must pass the tests of the capital markets to raise cash inan effort to expand into a new market, management has carteblanche to spend the firm’s cash as it sees fit. Wall Street is filledwith stories of firms divesting or spinning off unprofitabledivisions that were acquired a few years back at rich premiums.

Measuring Cash Levels

The amount of cash per share to the market price per shareprovides a useful indication of the cash level of the firm. GeneralElectric may have $3 billion in cash, but this works out to a cashlevel equal to about 2.5% of the stock value (cash divided bymarket capitalization).

Screening for firms with high proportions of cash to share pricerepresents a reasonable starting point in tracking down cash-

Screening for Cash-Rich Firms ThatWill Put Their Money to Good UseBy John Bajkowski

John Bajkowski is AAII’s senior financial analyst and editor of Com-puterized Investing.

Growth Rate 52-Net Cash Free Cash Price- I/B/E/S Est. Week

Cash to to Flow to Dividend Earnings Historical Long- Historical RelativePrice Price Price Yield Ratio EPS Term Sales Strength(%) (%) (%) (%) (X) (%) (%) (%) (%) Description

Large-Cap Stocks (above $1.5 billion)Temple-Inland (N) 179.2 na –5.6 3.0 8.6 –9.0 9.0 8.6 –35 Holding co. in paper, pkg, & build’g prod

UAL Corp. (N) 75.4 –152.3 19.2 0.0 7.4 –44.7 10.0 7.3 49 Holding co. of United Airlines

Northwest Airlines (M) 52.1 –55.8 21.5 0.0 11.5 na 9.0 na 60 Holding co. of Northwest Airlines Inc.

Delta Air Lines (N) 34.7 –48.9 3.5 0.3 9.5 –6.8 9.0 7.3 1 Air transport for passengers, mail, freight

General Motors Corp. (N) 33.9 na 9.3 3.1 7.3 –0.5 6.0 4.1 –1 Autos, trucks, & related parts; defense prods

King World Produc’ns (N) 33.1 25.7 9.0 0.0 11.8 7.9 11.0 4.8 –11 Distribs TV programs, feature films

Foster Wheeler Corp. (N) 32.5 –35.6 –9.3 1.8 21.7 13.9 15.0 12.4 6 Design, engineer’g, construct’n, mfg operations

Apple Computer (M) 32.4 –47.3 –21.6 1.7 20.4 –1.8 15.0 14.8 –50 Personal computers & related products

PACCAR (M) 32.0 –54.9 8.5 2.1 7.5 –2.6 8.0 5.0 –17 Multinational design, mfr heavy-duty trucks

Moore Corporation Ltd. (N) 31.5 5.4 0.6 4.8 7.1 –10.7 12.0 –2.4 –19 Manufactures business forms

Canon (M) 29.8 –21.3 –0.6 0.6 49.0 –6.5 15.0 7.4 –9 Electronics; fine chemicals; & engineer’g

Novell (M) 28.9 18.8 8.0 0.0 13.5 21.5 15.0 32.6 –56 Networking and application software

National Semiconductor (N) 28.3 –4.8 –7.9 0.0 7.0 48.7 12.0 7.3 –33 Designs, manufactures semiconductor prods

General Dynamics (N) 26.3 3.6 9.6 2.5 13.0 1.8 8.0 –20.2 –2 Armored vehicles; submarines; coal mining

Liz Claiborne (N) 26.2 3.0 5.0 1.4 26.2 –10.8 12.0 8.9 57 Designs & mkts apparel, fragrances, & cosmetics

Mid-Cap Stocks ($250 million to $1.5 billion)

Chris-Craft Indus. (N) 121.2 101.5 14.4 0.0 40.5 –33.1 na 12.5 –5 TV broadcasting; polyvinyl alcohol film

WHX Corporation (N) 105.9 32.7 16.4 0.0 5.4 –36.0 10.0 0.8 –16 Integrated steel manufacturer

Amdahl Corp. (A) 77.6 22.2 9.5 0.0 9.4 –14.7 5.0 –4.9 –39 High-perform general-purpose computer systems

VLSI Technology (M) 76.4 43.8 3.5 0.0 13.4 109.4 21.0 15.3 –40 Custom & semi-custom integrated circuits

BHC Communications (A) 66.0 56.8 6.8 0.0 44.3 –32.8 na 13.1 –5 Holding co. in TV broadcasting

LTV Corporation (N) 65.9 –9.7 37.1 0.0 7.2 –8.9 7.5 –6.6 –35 Steel & energy industries

National Presto Indus (N) 65.3 51.3 –2.4 4.8 13.7 –5.6 6.0 0.1 –26 Electrical appliances and housewares

Seaboard Corp. (A) 61.9 20.1 –44.7 0.4 10.8 13.5 na 13.7 10 Poultry & pork process’g;

Navistar International (N) 57.6 na 33.1 0.0 5.5 25.8 5.5 10.5 –46 Holding co. of diesel truck manfacturer

Alaska Air Group (N) 48.2 –100.3 13.1 0.0 23.7 –9.1 9.0 7.5 13 Holding co. for Alaska Airlines & Horizon Air

Mercer International (M) 47.8 7.1 3.9 0.0 4.7 43.1 20.0 55.3 0 Pulp & paper indus; financial services

Silicon Valley Group (M) 45.9 22.3 –3.7 0.0 13.4 30.5 22.0 20.2 –31 Automated wafer process’g equip

Continental Airlines (N) 45.4 –75.1 16.2 0.0 7.2 16.2 10.0 –1.3 338 Air transport of passengers, cargo, mail

FSI International (M) 41.0 19.2 –8.9 0.0 10.4 48.3 25.0 28.5 –43 Equip used in fabrication of integrated circuits

Farmer Brothers Co. (M) 38.3 30.5 8.4 1.6 12.6 –0.3 na 3.5 –14 Coffee, spices, & allied food prod for food serv

Shadow Stocks (small-cap firms with low institutional interest)

Astrosystems (M) 97.9 93.0 14.6 0.0 55.8 33.4 na 4.1 5 Electronic measure & control devices

Daxor Corp. (A) 97.5 80.2 1.4 0.0 27.2 43.8 na 17.4 –13 Cryopreservation tech for human semen & blood

Wilshire Oil Co. of Texas (N) 62.4 51.1 –7.7 1.2 19.0 73.3 na 9.9 –29 Explore & prod oil & gas in U.S. and Canada

Salem Corp. (A) 48.8 –36.8 –7.7 2.6 13.1 13.8 na 1.2 –21 Industrial equip for furnaces, metal process’g

Sevenson Environmental (M) 47.6 30.5 –4.3 1.3 9.7 5.3 15.0 8.3 –22 Field serv for remediation of hazard materials

Intrav (M) 45.4 –50.3 –4.8 6.3 9.6 na na na na Escorted international travel programs

Thermo Voltek Corp. (A) 42.5 29.7 0.1 0.0 38.6 19.8 20.0 27.7 49 Instruments to test electronic & elec’l systems

National Beverage Corp. (A) 39.3 –46.6 5.4 0.0 8.0 72.6 na 3.0 –22 Holding co. for branded soft drinks

Detroit & Canada Tunnel (M) 32.3 25.4 –10.4 1.6 10.4 0.1 na –8.7 –8 Operates tunnel connecting Detriot & Windsor

Align-Rite International (M) 32.2 17.9 0.8 0.0 11.4 na na na na Manufactures photomasks for integrated circuits

Lynch Corp. (A) 32.0 –57.3 –8.8 0.0 19.1 8.2 na 16.4 20 Holding co. for multimedia, serv & manufac’g

American Filtrona Corp. (M) 27.5 8.5 2.6 3.0 13.2 0.0 na 0.4 –7 Fiber filters for tobacco prods

New Mexico & Arizona Land (A) 26.6 23.8 7.1 0.0 6.3 21.9 na 15.9 49 Real estate sales, exchanges, & devlp

FDP Corporation (M) 25.9 16.7 0.9 0.0 23.6 52.7 na 5.0 –8 Application software for life insurance indus

Industrial Scientific (M) 24.7 18.6 5.7 0.0 15.6 29.3 11.0 24.9 –26 Portable instruments for measuring gases

Table 1.Cash-Rich Firms

Source: AAII's Stock Investor and Market Guide. Data as of 2/29/96.

Price-Earnings Ratio: Market price per share divided by the firm’searnings per share. A measure of how the market currently values thefirm’s earnings growth and risk prospects. High price-earnings ratiosindicate high expectations of future earnings growth.

Historical EPS Growth Rate: Annual growth in earnings per shareover the last five years. A measure of how successful the firm has beenin generating the bottom line, net profit.

I/B/E/S Est. Long-Term Growth Rate: The median growth rate inearnings per share in continuing operations expected over the next fiveyears that is being forecasted by analysts tracked by I/B/E/S. Anindication of the growth expectations for the firm.

Historical Sales Growth Rate: Annual growth in sales over the lastfive fiscal years. Used to provide a confirmation of the quality ofhistorical earnings per share growth rate.

52-Week Relative Strength: Price performance of the stock duringthe last year relative to the price performance of the S&P 500 index. Afigure of 0 indicates the stock had the same percentage priceperformance as the market. A figure of 5 indicates that the stockoutperformed the market by 5%.

Definitions of Screens and Terms

Cash to Price per Share: Cash and cash equivalents per sharedivided by the market price per share. Indicates what percentage ofthe stock price is equal to cash per share. The higher the percentage,the more cash on hand relative to share price.

Net Cash to Price per Share: Cash and cash equivalents pershare less current liabilities per share divided by market price.Indicates percentage of stock price equal to net cash per share, amore conservative measure than cash to price per share. A higherpercentage indicates a higher level of cash per share relative to shareprice.

Free Cash Flow to Price per Share: Free cash flow per sharedivided by price per share. Free cash flow per share is defined as cashfrom operations minus capital expenditures and dividends paid.Positive ratios indicate that the firm has positive free cash flow and isable to meet the needs of its capital expenditures and dividendpayouts from internal operations. The higher the percentage, thegreater the free cash flow relative to share price.

Dividend Yield: Annual dividends per share divided by price pershare. An indication of the income generated by a share of stock.

The following is a short description of the screens and terms used for this month’s Shadow Stock listing.

rich firms. When performing such a screen,it is important to exclude financials, be-cause by the nature of their business theyare required to hold large cash positions.Utilities were also excluded because oftheir regulated nature and overall lowgrowth potential.

To perform this screen, AAII’s Stock Inves-tor program was used. In addition to ex-cluding financials and utilities, a filterrequiring positive earnings from continu-ing operations for the last 12 months wasspecified as a minimum current profit-ability requirement, along with a mini-mum share price requirement of five dol-lars. Without the minimum share pricerequirements, bankrupt firms with a shareprice of a few pennies would dominate.The securities were broken into threegroups—large-capitalization, mid-cap,and Shadow Stocks. (Shadow Stocks aredefined by AAII as stocks of small non-financial companies with low institutionalinterest, and that have had positive earn-ings for the last two years.) The 15 stockswith the highest percentage of cash toprice are listed for each group in Table 1.

As important as it is to look at cash, it isequally important to look at the financialobligations of the firm. In the mid-capgroup, WHX Corporation’s cash positionis equal to 105.9% of its share price, com-

pared to 121.2% for Chris-Craft Indus-tries. A quick look at balance sheets ofthe firms reveals large differences. WHXis the ninth largest domestic integratedsteel manufacturer and has emerged froma chapter 11 reorganization. Its $12.32per share in cash is quickly reducedwhen one considers the firm’s short-termliabilities and long-term debt. Somefirms build up a cash reserve to ensurethat they can meet the required pay-ments of their short-term debt and cur-rent portion of long-term debt. When bor-rowing money, they may be required tohold compensating balances as terms ofthe loan. In the case of WHX, its annual 10-K filing indicates that it intends to retainany future earnings for working capitalneeds and to finance capital improve-ments, and it does not intend to pay cashdividends on the common stock for theforeseeable future. In addition, the termsof the company’s senior debt places cer-tain limitations on WHX’s ability to paycash dividends.

A useful modification to the gross cashto price per share ratio is to subtract theshort-term liabilities from cash to estab-lish a net cash per share figure, whichprovides a better measure of the excesscash on hand. Dividing the net cash pershare by the share price indicates how

much of this “excess cash” is availableon a per share basis. In the case of WHX,considering short-term liabilities dropsthe ratio of cash to price from 105.9% to32.7%. While this is still a significantnumber, it paints a different picture thanthat for Chris-Craft Industries, whichmanages to maintain a net cash to shareprice per share ratio of 101.5%.

While the mention of the name Chris-Craft may bring the image of boats tomind, the company has reinvented itselfinto primarily a television broadcasterthrough its majority owned subsidiary,BHC Communications, which also madethe screen as a separate entity—high-lighting the problem of double countingwhen screening.

Many of the firms with positive ratiosof cash to price per share have negativeratios once short-term liabilities are con-sidered. The effect is especially pro-nounced with the many airline stocksthat passed the initial screen. These firmshave heavy capital expenditure require-ments and are heavy users of cash inoperations. While these companies havemade a recent turnaround and the indus-try is turning profitable after a number ofyears of running in the red, the high grosscash positions are not significant whenscreening for cash-rich firms.

Three firms, Temple-Inland, GeneralMotors, and Navistar, do not show a netcash to price per share ratio calculationbecause they do not maintain clear-cutdivisions between current and long-termassets and liabilities.

Beyond looking at the static cash posi-tions of these firms, an examination of theactual cash the firm is providing is evenmore important for a long-term investor.Free cash flow is calculated by taking thecash flow from operations as reported onthe firm’s statement of cash flow andsubtracting capital expenditures anddividends. This measure attempts tocapture whether the firm is generatingenough cash to help fund any necessaryinternal capital expenditures. Apple,which has a positive cash to price ratio,has a negative ratio of free cash flow pershare to market price. It is not generatingenough cash from its operations to coverits currently required capital expendi-tures. Its current position points out theneed for the company to determinewhether it should jettison some of itsproduct development ideas and focuson those that show the greatest potential.

In contrast, another technology firm,Novell, has a lower cash to price ratio, yethas positive net cash to price per shareand positive free cash flow to price pershare. Novell dominates the network op-erating systems marketplace for personalcomputers, but is facing increased com-petition from vendors such as Microsoft.In the past, Novell has used its built-upcash hoard and excess cash flow to ac-quire products in an attempt to diversifyits product line. It acquired UNIX SystemLaboratories in 1993, merged withWordPerfect in 1994 and purchasedQuattro Pro from Borland in 1994. As is toooften the case, things did not go as wellas planned. Novell just sold off itsWordPerfect and Quattro product lines toCorel and its UnixWare product line to theSanta Cruz Operation, so that it couldfocus on its primary business unit.

The price-earnings ratio is the tradi-tional measure of value and market ex-pectations. The price-earnings ratio iscomputed by dividing a stock’s price byits most recent 12 months’ earnings pershare. The price-earnings ratio is closelyfollowed because it embodies the

market’s expectation of future companyperformance; companies with high priceearnings ratios have greater expectationof future performance than those with lowratios. A value investor would typicallysearch for stocks with low price-earnings,with the belief that the market hasmispriced the stock.

The price-earnings ratios for this groupranges from a low of 4.7 for Mercer Inter-national to a high of 55.8 for Astrosystems.Many of the cyclical companies such asNavistar are currently carrying low price-earnings ratios because the market feelsthat the economy’s growth potential islimited and these firms would suffer if theeconomy topped out. So while Navistarcurrently has a positive free cash flow pershare and has a very high earnings growthrate of 25.8% annually over the past fiveyears, its growth potential is limited. TheI/B/E/S consensus long-term earningsgrowth estimate of 5.5% confirms thismarket view. It is common for cyclicals tocarry low price-earnings ratios late intothe economic cycle.

Many of the companies that passedour initial screen are capital-intensivefirms that benefited from operating le-

verage, in which a small boost in salestranslates to a larger boost in earnings.Sales growth rates are often used to con-firm the strength of earnings per share.

The final data element presented inTable 1 shows the 52-week relativestrength versus the S&P 500 index; posi-tive figures indicate the percentage thatthe stock outperformed the S&P 500, andnegative numbers indicate the percent-age that the stock underperformed theS&P 500. As a whole, this group has per-formed poorly on a relative basis, withthe airline group being the notable ex-ception. The few firms with an “na” desig-nation have gone public in the last year.

Conclusion

Screening for cash-rich stocks is not asimple process. Preliminary filters shouldscreen for companies that have not onlya high level of cash per share, but also astrong balance sheet, the potential forfuture earnings growth, and positive freecash flow per share. Much of the finalanalysis rests on your belief inmanagement’s ability to use any cashhoard wisely.

Screening CriteriaFor use with AAII's Stock Investor program (version 2.1)

Found in the Company Information Group:

IND_2_DIG <> 07 Sector is not equal to Financial

IND_2_DIG <> 12 Sector is not equal to Utility

IND_3_DIG <> 0933 Industry is not equal to Real Estate

Operations

Found in the Price and Share Data Group:

MKTCAP > 20 Market capitalization in the last

quarter is greater than $20 million

PRICE > 5 Price is greater than $5 per share

Found in the Income Statement—Annual Group:

EPSCON_12M > 0 Earnings per share from continuing

operations for the last 12 months

is greater than 0.

User Defined

Cash to Price per Share or Net Cash to Price per Share is Greater than 20%

Cash per Share:

Cash_Q1 / Shr_AQ1

Cash to Price per Share:

((Cash_Q1 / Shr_AQ1) / Price) * 100

Net Cash per Share:

(Cash_Q1 - CL_Q1)/SHR_AQ1

Net Cash to Price per Share:

(((Cash_Q1 - CL_Q1)/SHR_AQ1) / Price) * 100

Found in Balance Sheet—Quarterly:

Cash Q1

Current libilities Q1

Found in Price & Share Data:

Shares average Q1

Price

John Bajkowski is AAII’s financial analyst and editor of Computerized Investing.

Examining the investment tech-niques of successful money managerscan prove insightful when trying to es-tablish or refine your personal tech-niques. For this issue’s Analyst’s Cor-ner, we explore the techniques of in-vestment manager Peter Lynch andapply a basic stock screen using someof his principles. In his two books,“One Up on Wall Street” and “Beatingthe Street,” Peter Lynch describes theanalytical process that led to his suc-cess at the helm of Fidelity’s Magellanfund.

Lynch’s Laws

• Stick with industries and companies youknow and understand. Getting ideas is acritical starting point for a bottom-upstockpicker like Peter Lynch. He advo-cates looking around at companies andproducts for stock ideas and then per-forming in-depth analysis to determineif the stock is a buy. This would becontrasted with a top-down approachthat starts the analysis with an overalleconomic forecast that leads to sec-

tors, industries, and finally companiesexpected to perform well.

In looking for ideas, Lynch favors ar-eas that you understand and whereyour intimate knowledge is a competi-tive advantage. For example, if you area pharmacist, use your knowledge toanalyze the drug industry. Also noticetrends around you—is a new store inthe mall a hit? If so, investigate it.Leverage your knowledge as a con-sumer, hobbyist, and professional inyour investments.• Do your research before investing. Lynchobserves that many people follow ahunch or a tip and invest in a companywithout doing any research. Very oftenthese are the same individuals whospend many hours researching whichcoffee maker is the best on the marketand scouring the papers to discoverwhich store offers the best price.• Is it a buy? Finding a good company isonly half of the battle in making asuccessful investment. Buying at a rea-sonable price is the other. Lynch lookstoward both earnings and assets whenit comes to valuing stocks. An earnings

examination focuses on the ability ofthe company to earn future income.The greater the earnings prospects,the more valuable the company. In-creasing earnings will translate to in-creasing prices. Assets are importantin determining the base value of acompany should it be split up andsold off in pieces.• Carefully consider the price-earnings ratio.The earnings potential of a company isa primary determinant of companyvalue. At times, the market may getahead of itself and overprice even astock with great prospects. The price-earnings ratio helps to keep your per-spective in check. The ratio comparesthe current price to the most recentlyreported earnings. Stocks with goodprospects should sell with higher price-earnings ratios than stocks with poorprospects.• How does the price-earnings ratio compareto its historical average? By studying thepattern of price-earnings ratios over aperiod of several years, you can de-velop a sense of the normal level forthe company. This knowledge shouldhelp you avoid buying into a stock ifthe price gets ahead of the earnings orsend an early warning that it may betime to take some profits in a stock youown. If a company does everythingwell, you may not make any money onthe stock if you paid too much for it.• How does the price-earnings ratio compareto the industry average? Comparing acompany’s price-earnings ratio to thatof the industry may help reveal if thecompany is a bargain. At a minimum, itleads to questions as to why the com-pany is priced differently. Is it a poorperformer in the industry, or is it justneglected? Lynch’s ideal investmentis a neglected niche company, control-ling a market segment, in anunglamorous industry that would bedifficult and time-consuming for a com-petitor to break into.• How does the price-earnings ratio compareto its earnings growth rate? Companieswith better prospects should sell withhigher price-earnings ratios. A usefulvaluation technique is to compare theprice-earnings ratio to the earningsgrowth rate. A price-earnings ratio of

Finding a “good” company is only half the battle.Buying at a reasonable price is the other.

Investment Techniques:Stock Screeningà la Peter LynchBy John Bajkowski

Modified Long-P/E to EPS Term Shares Net Price asEPS P/E Growth Dividend Debt to Held by Cash per % of 52-

Growth Ratio Rate Yield Capital Institutions Share Price Wk. HighCompany (Exchange) (×) (×) (%) (%) (%) (%) ($) ($) (%) Description

Large Cap (above $1.5 billion)Intel Corp (NM) 0.22 11.1 49.5 0.4 5.4 73.7 4.88 61.50 83 Semicond components & systems

General Dynamics (NY) 0.25 11.2 41.3 3.2 14.6 55.9 12.63 43.75 88 Military marine & aircraft

Sun Co (NY) 0.37 13.7 30.3 6.3 26.8 53.6 (5.73) 28.75 82 Energy resources integrated petro coal

NFC Plc ADR (AM) 0.43 14.8 31.1 3.6 36.6 0.2 (0.24) 15.38 72 Transport, logistics, home delivery serv

Magna Int’l A (NY) 0.47 12.3 23.9 2.0 10.1 75.3 0.20 36.88 68 Automotive components & systems

Medium Cap ($250 million to $1.5 billion)IP Timberlands (NY) 0.14 4.1 18.1 11.6 0.2 1.2 0.20 24.88 80 Forest resource management

Bowne & Co (AM) 0.16 7.8 45.3 2.0 4.3 53.3 0.53 18.00 63 Finance & corporation printing

Family Dollar Stores (NY) 0.28 9.6 31.0 3.0 0.0 64.7 0.11 11.38 62 Operates discount stores

Scitex Corp (NM) 0.28 14.0 48.0 2.3 0.1 29.7 6.39 22.50 84 Computer imaging systems

Briggs & Stratton (NY) 0.30 9.9 30.4 2.6 15.7 71.6 10.09 70.25 78 Mfrs air cooled engines & auto locks

Standard Microsystems (NM) 0.30 11.8 38.8 0.0 6.0 40.6 1.96 19.88 74 Mfrs semiconductor circuits

Lattice Semicon (NM) 0.32 15.3 47.2 0.0 0.0 68.2 5.65 18.50 75 Programmable logic devices

Clarcor Inc (NY) 0.33 15.2 42.7 3.1 19.0 52.2 (1.18) 20.25 90 Mobile & environ’l filtration prods

Instrument Systems (NY) 0.35 10.6 30.4 0.0 12.6 45.7 0.77 7.88 81 Mfrs diversified instruments

Rayonier Timberlands (NY) 0.35 6.5 4.3 14.3 32.8 0.2 0.00 35.25 88 Marketing, sale of timber

Ballard Medical Prods (NY) 0.37 15.3 41.4 0.5 0.0 38.1 0.94 10.38 56 Disposable medical products

XTRA Corp (NY) 0.37 17.8 46.4 1.1 51.1 63.1 na 51.25 96 Tractor-trailer, cargo contain leasing

Lancaster Colony (NM) 0.43 17.8 40.2 1.4 12.2 73.1 (0.07) 35.00 89 Mfrs food and auto products

Church & Dwight (NY) 0.47 18.0 36.5 1.9 4.3 47.0 (0.04) 23.00 79 Mfrs, mkts soap & cleaning prods

Weatherford Int’l (AM) 0.47 20.0 42.8 0.0 6.3 65.0 (0.83) 12.38 84 Equip for petroleum industry

Champion Enterprises (AM) 0.48 14.1 29.1 0.0 5.7 36.7 1.07 39.25 99.04 Holding co, Champion Home Builders

Shadow Stocks (small firms with lower institutional interest)Norex America (AM) 0.08 1.8 21.7 0.0 14.6 1.4 (6.41) 9.25 82 Mkts, provides leisure cruises

Huffman Koos (NM) 0.17 6.4 37.2 0.0 18.7 12.8 na 7.75 86 Specialty furniture retailer

SBE Inc (NM) 0.22 9.1 40.8 0.0 0.0 15.9 na 7.50 60 Designs, sells computer hardware

Diodes Inc (AM) 0.26 11.7 45.1 0.0 3.6 5.4 0.22 5.50 55 Semiconductor devices

CCA Industries (NM) 0.27 11.0 41.4 0.0 7.1 11.2 0.84 4.38 48 Makes specialty cosmetic products

Shelter Components (AM) 0.28 9.2 32.2 0.4 28.4 58.4 (2.14) 12.00 73 Mfrs, sells, distrib prods for houses

Genovese Drug (AM) 0.35 11.6 30.9 2.2 0.0 14.2 (3.44) 11.00 79 Drug store chain in NYC

Prima Energy (NM) 0.35 11.8 33.8 0.0 6.0 8.3 na 13.75 81 Explor, devlp, prod oil & gas

Comcoa Inc (NM) 0.37 12.0 32.8 0.0 0.0 6.4 0.28 14.00 89 Rent-A-Center

Craftmade Int’l Inc (NM) 0.42 15.1 36.1 0.2 0.0 22.7 0.03 10.44 84 Designs, distrib, mkts ceiling fans

Jennifer Convertible (NM) 0.42 13.6 32.5 0.0 0.4 7.2 2.19 7.75 48 Sells sofabeds

Methode Electronics B (NM) 0.48 20.0 41.0 0.5 0.0 11.6 1.13 20.00 100 Mfrs electronic component devices

Refac Technology Dev’lp (AM) 0.48 9.9 20.5 0.0 0.0 17.6 2.75 7.13 60 Administers int’l technology licenses

Xscribe Corp (NM) 0.48 10.7 22.1 0.0 7.2 8.0 na 2.56 44 Hard-, software for computer-aided systems

Western Beef (NM) 0.49 10.6 21.6 0.0 17.4 9.4 (0.80) 7.50 52 Wholesale meat and poultry

Jones Medical Indus (NM) 0.50 11.6 22.1 1.3 1.6 27.6 (0.08) 7.75 44 Prod branded & generic ethical drugs

Prime Medical Corp (NM) 0.50 12.0 24.1 0.0 7.4 6.9 na 3.13 86 Estab & mgmt of cardiac rehab centers

na = not available Exchanges:NY= New York Stock ExchangeAM= American Stock ExchangeNM= Nasdaq National MarketNS= Nasdaq Small Cap

Sources: Stock Investor/Media General; S&P Stock Guide. All data as of September 30, 1994.

Table 1.Low P/E to Earnings Growth Stocks

Definitions of Screens and TermsThe following is a short description of the screens and terms used in the table.

Modified P/E to EPS Growth: Current price-earnings ratio di-

vided by the sum of the historical earnings growth rate and dividend

yield. Ratios below 0.5 are considered attractive. Ratios above 1 are

considered poor.

P/E Ratio: Market price per share divided by most recent 12

months’ earnings per share. A measure of the market’s expectations

regarding the firm’s earnings growth and risk.

EPS Growth Rate: Annual growth in earnings per share from total

operations over the last five fiscal years. A measure of how successful

the firm has been in generating the bottom line, net profit.

Dividend Yield: Indicated annual dividend divided by market price.

Long-Term Debt to Capital: Long-term debt divided by the total

of long-term debt and preferred and common equity. Provides a measure

of the financial strength of the company. The lower the figure, the stronger

the balance sheet.

Shares Held by Institutions: Percentage of shares outstanding that are

held by institutions. Provides an indication of the level of Wall Street

interest in the stock.

Net Cash per Share: The total of all cash and cash equivalents less long-

term debt, divided by the number of shares outstanding. Provides an

indication of the financial strength of the company and a support for the

stock price.

Price: Most recent market price as of September 30, 1994.

Price as % of 52-Wk. High: Most recent market price divided by highest

market price over the last 52 weeks.

half the level of historical earningsgrowth is considered attractive, whileratios above two are considered unat-tractive.

Lynch refines this measure by add-ing the dividend yield to the earningsgrowth. This adjustment acknowledgesthe contribution that dividends maketo an investor’s return. The ratio is cal-culated by dividing the price-earningsratio by the sum of the earnings growthrate and the dividend yield. With thismodified technique, ratios above oneare considered poor, while ratios be-low 0.5 are considered attractive.• How stable and consistent are the earnings?It is important to examine the histori-cal record of earnings. Stock pricescannot deviate very long from the levelof earnings, so the pattern of earningsgrowth will help to reveal the stabilityand strength of the company. Ideally,earnings should move up consistently.• Avoid hot companies in hot industries. Lynchprefers to invest in companies withearnings expanding at moderately fastrates (20% to 25%) in non-growth indus-tries. Extremely high levels of earningsgrowth rates are not sustainable, butcontinued high growth may be factoredinto the price. A high level of growth fora company and industry will attract agreat deal of attention from both in-vestors, who bid up the stock price,

and competitors, who provide a moredifficult business environment.• What is the level of institutional holdings?Lynch feels that the bargains are lo-cated among the stocks neglected byWall Street. The lower the percentageof shares held by institutions and thelower the number of analysts followingthe stock, the better.• How large is the firm? Small firms havemore upside potential than large firms.Small firms can easily expand in sizewhile large firms are limited. A smallfirm like Starbucks can double in sizemuch more easily than a large firmsuch as General Electric.• How strong is the balance sheet? A strongbalance sheet provides maneuveringroom as the company expands or ex-periences trouble. Lynch is especiallywary of bank debt, which can usuallybe called in by the bank on demand.• What is the level of net cash per share?Lynch likes to look at the net cash pershare to help discover both a supportfor the stock price and the financialstrength of the company. The net cashper share is calculated by adding thelevel of cash and cash equivalents,subtracting the long-term debt, anddividing the result by the number ofshares outstanding.• Are insiders buying the stock? Insiderbuying of shares is a positive sign,

especially if it is spread among a num-ber of individuals. While insiders mayhave many reasons for selling hold-ings, they generally buy stock whenthey feel it is an attractive investment.• Is the company buying back shares? Lynchfavors companies that buy back theirshares over companies that choose toexpand into unrelated businesses.Share buybacks become an issue oncecompanies start to mature and havecash flow that exceeds their capitalneeds. The share buyback will help tosupport the stock price and is usuallyperformed when management feels thatthe current stock price is favorable.

Applying the Lynch Screen

While Peter Lynch is a bottom-up,kick-the-tires type of stockpicker, someof his principles are useful screeningcriteria. Our first screen excluded fi-nancial firms. Peter Lynch is a big fan offinancial stocks and presents a seriesof screens he uses to help select banksand savings and loans in “Beating theStreet.” We had to exclude them fromthe general screen, however, becausetheir financial statements cannot bedirectly compared to other firms. AAII’sStock Investor program was used to per-form the screen and we were left with5,751 non-financial companies out of a

Stock-Picking Checklist

Your analysis should center on the factors that will move the stockprice.

Look for low P/Es compared to earnings growth and dividend yield

Look for P/E in lower range of historical average

Look for P/E below industry average

Study the pattern of earnings, especially how they reacted during arecession.

Look for a low level of debt, especially bank debt

Net cash per share should be high relative to stock price

Be wary of earnings growth rates above 50%

Small companies should be favored in your search, they have moreupside potential

Look for low percentage of shares held by institutions and number ofanalysts following stocks

Insider buying by a number of insiders is a positive sign

If so, this will support the stock price and probably indicates thecompany has been ignored, but management is confident

total of 8,123 stocks.Price-earnings ratios are an impor-

tant aspect to Lynch’s analysis andmake an excellent primary screen. Ourscreen used the ratio of price-earningsto the earnings growth rate plus thedividend yield. A ratio less than orequal to 0.50 was specified as a cut-off,leaving 441 companies.

Lynch is wary of companies growingtoo quickly, so the next filter specifieda maximum earnings per share growthrate of 50%. The number of firms pass-ing this filter was 258.

The final filter required that the long-term debt-to-capital ratio for each com-pany be less than its industry average,leaving 125 firms.

The screens we applied seemed tofavor smaller firms. Of the 125 firmsthat passed the screen, only five werelarge-cap stocks—16 were medium-cap, and 104 were small-cap. All of thelarge- and medium-cap firms thatpassed the filter are shown in Table 1,ranked in ascending order by the modi-fied price-earnings to growth ratio.Within the small-cap area, 17 of thestocks passing the screens were

Shadow Stocks, and these are pre-sented in Table 1.

Most of the companies passing thescreens had relatively low price-earn-ings ratios when compared to the cur-rent market level of 18.0. The historicalgrowth rates, however, were well aboveaverage, with many ranging between30% and 40% per year. These are his-torical growth rates and are not sus-tainable for a long period of time; someare due to special situations. GeneralDynamic’s growth rate of 41.3% wasdue to a one-time earnings boost in1993. Looking only at earnings fromcontinuing operations and the year-by-year earnings figures would presenta better picture of earnings growth andconsistency.

The dividend yield for two of thesecurities—IP Timberlands andRayonier Timberlands—should clue aninvestor in to a special situation. Bothof the securities are limited partner-ships created to manage the lumberresources of International Paper andRayonier, respectively. These are part-nerships structured with finite lives andthey highlight the importance of in-

• Invest only in industries and companies you understand and know thespecific reason that you are buying the stock

• How does the price-earnings ratio compare to the growth rate in earningsand dividends?

• How does the price-earnings ratio compare to its historical average?

• How does the price-earnings ratio compare to the industry?

• How stable and consistent are the earnings?

• How strong is the balance sheet?

• What is the cash position?

• Avoid hot companies in hot industries

• Big companies have small moves, small companies have big moves.

• What is the level of institutional holdings?

• Are insiders buying stock?

• Is the company buying back shares?

depth analysis after any basic screen.The ratio of long-term debt to capital

is generally lower for the Shadow Stocksthan for the medium- and large-capstocks. Small-cap stocks have a moredifficult time raising capital throughthe bond market than larger stocksand often turn to banks for capital. Aclose examination of the financial state-ments, especially in the notes to thefinancial statement, should help toreveal the use of bank debt.

A large number of the listed compa-nies in Table 1 have very high percent-age of shares held by institutions, anegative in Lynch’s opinion.

Net cash per share highlights a po-tential hidden asset for a number ofcompanies and is of greater interestfor companies that are distressed, turn-around potentials, or asset plays.

Only a basic level of screening wasperformed when presenting the Lynch-based screens. Much of the analysisadvocated by Peter Lynch is subjec-tive in nature, requiring hands-on analy-sis. As Peter Lynch stresses, it is pos-sible to succeed at investing, but youmust be willing to do your work.

Rules

Exclude financial stocks

Found in the Company Information Group

Ind_2_Dig <> 07 Sector is not equal to Financial

Ind_3_Dig <> 0933 Industry is not equal to Real Estate Operations

Low price-earnings ratio in comparison to the growth rate in earnings plus the dividend yield

Found in User-Defined Group

Yield Adjusted PE to Growth Ratio < 0.5 Yield Adjusted PE to Growth

Yield Adjusted PE to Growth Ratio > 0.0 Ratio between 0.0 and 0.5

PE / (EPSCon_G5F + Yield) Calculation of the Yield Adjusted PE to Growth Ratio PE divided by the sum

of EPS Cont-Growth 5yr and the current Dividend Yield

Be wary of earnings growth rates above 50%

Found in Growth Rates Group

EPSCon_G5F < 50 Five-year earnings per share growth from continuing operations is less than

50%

Look for a low level of debt, especially bank debt

Found in Ratios Group

LTD_TC_Q1 < INDUSTRY.LTD_TC_Q1 (N) LT Debt / Tot Cap Q1 is less than that of the narrow industry group median

or

TL_TA_Q1 < INDUSTRY.TL_TA_Q1 (N) Tot Liab / Assets Q1 is less than that of the narrow industry group median

Optional Criteria Not Used in Screen

Look for PE in lower range of historical average

Found in Multiples Group

PE < PE_A5Y PE less than PE-Avg 5yr

Look for PE below industry average

Found in Multiples Group

PE < INDUSTRY.PE (N) PE is less than that of the narrow industry group median

Look for low percentage of shares held by institutions

Found in Price & Share Data Group

ShrInst < 28.3 Inst’l Ownership is low compared to average or median displayed when

selecting field

Screening CriteriaFor use with AAII's Stock Investor program (version 2.1)

Using all the rules can be very restrictive and sometimescontradictory. But it is hard to argue with screens that tryto find neglected stocks on the verge of a turnaround.

Stock Market Winners:Applying the Rules inToday’s EnvironmentBy John Bajkowski

John Bajkowski is AAII’s financial analyst and editor of Computerized Investing.

In September 1989 the AAII Journalpublished an article by MarcReinganum, titled “Investment Charac-teristics of Stock Market Winners,” whichexamined the common traits of a groupof winning stocks. These types of ex-aminations can be noteworthy if theyhelp to establish financial relationshipslikely to hold true over time.

The study examined 222 stocks high-lighted as winners in a publication byWilliam O’Neil & Co. titled “The Great-est Stock Market Winners: 1970-1983.”The goal was to establish the character-istics common to these stocks prior totheir rise to super stock status. Ninetrading rules were developed thathelped to identify the winners:• Price-to-book-value ratio less than

1.00;• Accelerating quarterly earnings;• Positive five-year growth rate;• Positive pretax profit margin;• Relative strength rank of at least 70;• Relative strength rank of the stock in

the current quarter is greater thanthe rank in the previous quarter;

• O’Neil Datagraph rating of at least 70;

show that buying into neglected, out offavor stocks leads to investment suc-cess. While the market does a good jobof valuing securities in the long-run, inthe short-run it can overreact to infor-mation and push the prices away fromtheir true value. Measures such as price-to-book-value ratio, price-earnings ra-tio, and dividend yield help to identifywhich stocks may be truly underval-ued.

The price-to-book value ratio is de-termined by dividing market price pershare by book value per share. Bookvalue is generally determined by sub-tracting total liabilities from total as-sets and then dividing by the numberof shares outstanding. It represents thevalue of the owners’ equity based uponthe historical accounting activities. Ifaccounting truly captured the currentvalues of the firm, then one would ex-pect the current stock price to sell nearthis accounting book value. Over thehistory of a firm, many events occurwhich can distort the book value figure.For example, inflation may leave thereplacement cost of capital goods withinthe firm way above their stated bookvalue, or the purchase of a firm maylead to the establishment of goodwill,which is an intangible asset boostingthe level of book value. Some servicesare more conservative in reporting bookvalue and may subtract out the value ofintangibles such as patents, copyrights,trademarks, or goodwill. Of course itmakes these values incomparable withservices that include intangibles. Dif-ferent accounting policies among in-dustries may also come into play whenscreening for low price-to-book stocks.

AAII’s Stock Investor program was usedto screen for the potential stock marketwinners. The first screen specified aprice-to-book ratio below 1.5, leaving2,501 companies out of a complete da-tabase of 8,145 NYSE, Amex, NasdaqNational Market, and Nasdaq Small Capstocks. The maximum price-to-bookratio level is higher than the originalstudy so that a slightly larger group ofcompanies would pass the completeset of filters. As a primary screen, speci-fying a price-to-book ratio below 1.0led to about 14% of the companies pass-

• Stock selling within 15% of its maxi-mum price during the previous twoyears; and

• Fewer than 20 million common sharesoutstanding.The use of these rules or screening

criteria produced returns significantlyhigher than the market. While the groupof stocks passing the filter were slightlymore risky than the market, the addi-tional risk did not account for the ex-traordinary returns of the winners.

There can be problems encounteredwhen trying to apply trading rules de-termined during a specific point in timewith a specific group of stocks. Thisarticle discusses some of the difficul-ties in trying to interpret and applythese rules in the real market environ-ment.

Low Price-to-Book-Value

When examined independently, thefirst screen requiring that the price-to-book-value ratio be below one pro-duced the highest rate of return. Thisfinding coincides with many studies that

ing the filter, versus about 30% passingthe 1.5 price-to-book-value limit.

Valuation levels of stocks vary overtime, often dramatically from bear mar-ket bottoms to bull market tops. Dur-ing the depths of a bear market, manyfirms can be found selling for a price-to-book ratio less than one. In the latterstages of a bull market, few companiesother than troubled firms sell for lessthan book value per share.

Earnings Momentum

The low price-to-book screen is verygood at identifying neglected firms, butsecondary, or conditioning, screens arealso needed to help identify whichstocks may be poised for a turnaround.Quarterly earnings per share for win-ners rose on average 46% in the originalstudy and exhibited an increase fromthe previous quarter’s growth rate. Ac-celerating earnings attract attention,and may be one of the first signs that acompany is poised for an upturn.

The earnings measure used in theoriginal study was somewhat crude in

that it did not consider the seasonalityof quarterly earnings. A more usefultechnique is to compare one quarter tothe same quarter last year—i.e., thisyear’s second quarter is compared tolast year’s second quarter. Many firmshave annual seasonal cycles, either inproduction or sales. Comparing similarquarters is one way of taking these sea-sonal changes into consideration.

In our screen, earnings from continu-ing operations for the most recent twoquarters were required to be abovetheir respective quarters last year. Thisdropped the number of companiesdown to 824 from the 2,501 passing thelow price-to-book-value screen.

To emphasize momentum or accel-eration in quarterly earnings, the nextcriterion specified that the rate ofchange between the recent quarter andits counterpart last year be greater thanthe increase between the previousquarter and its counterpart. This mo-mentum screen dropped the numberof passing companies down to 383.

Table 1 presents the quarterly earn-ings per share figures used for the

screening. Looking at the raw data canassist in judging the significance of anydecisions based upon percentagechanges. For example, very small earn-ings figures can lead to distorted growthrates. Elmer’s Restaurant experienceda 400% increase in its latest quarterover the same quarter a year ago, butthis increase was based upon a changein earnings per share from one cent tofive cents.

Minimum Fundamentals

As further proof that a company’s for-tunes have turned around, the next twocriteria specified a minimum level ofcompany fundamental performance.

The first criterion required that thefive-year annual growth rate in earn-ings per share be positive. Applyingthis criterion cut the number of firmsdown to 140. In the original study, thefive-year growth rate was based uponthe most recent five years of quarterlydata. The growth rates used in thescreening are based upon the firm’sfiscal-year data, which will not show

Price-to AnnualBook- Quarterly Earnings Per Share EPS Relative Price as No. ofValue Second Quarter* First Quarter* Growth Strength % of 52- SharesRatio 1994 1993 1994 1993 Rate Rank Wk High Outstanding

Company (Exchange) (×) ($) ($) ($) ($) (%) (%) (%) (mil) Description

Concord Fabrics A (AM) 0.47 0.55 0.36 0.60 0.48 6.5 89 87 2.1 Devlps, designs fabrics

Canandaigua Wine B (NM) 0.53 0.41 0.29 0.35 0.25 54.2 86 98 3.4 Makes dessert, table, & sparkling wines

Dynamics Corp of Amer (NY) 0.76 0.45 0.18 0.24 0.16 4.1 83 99 3.9 Mfgs & sells electronic devices

TSR Inc (NM) 0.78 0.10 0.04 0.07 0.00 39.6 87 89 1.5 Operates IBM computer systems

Bindley Western (NM) 0.90 0.34 0.30 0.33 0.30 15.8 86 97 10.8 Wholesales ethical pharmaceuticals

Oglebay Norton (NM) 0.99 2.75 1.12 (0.06) (0.83) 10.5 90 100 2.5 Lakes shipping: mining & manufacturing

Kings Road Entertainment (NS) 1.05 0.07 0.03 0.03 (0.05) 20.8 81 86 5.3 Devlp, produc, finance motion pictures

FDP Corp (NM) 1.18 0.04 0.01 0.03 (0.03) 35.1 92 94 3.4 Software to market life insurance prods

Schultz Sav-O-Stores (NM) 1.18 0.54 0.45 0.41 0.35 12.6 77 100 2.6 Regional food retailer and wholesaler

FRP Properties (NM) 1.20 0.35 0.17 0.16 0.12 8.4 90 89 4.0 Regulated carrier, real estate

Elmer’s Restaurants (NS) 1.21 0.05 0.01 0.02 (0.01) 12.5 93 94 1.8 Owns, operates, sells restaurant franchises

Bell Industries (NY) 1.33 0.57 0.33 0.33 0.22 3.0 83 98 6.3 Distribs, mfrs electronic components

Windmere Corp (NY) 1.35 0.57 0.08 0.03 0.02 5.2 76 94 17.9 Imports & markets personal care prods

Nortek Inc (NY) 1.36 0.44 0.12 0.06 (0.11) 24.6 93 87 12.5 Mfrs commercial & residential bldg prods

Ropak Corp (NM) 1.37 0.53 0.41 (0.06) (0.11) 39.3 92 93 4.3 Prod plastic shipping containers

Univ Health Services B (NY) 1.44 0.57 0.46 0.72 0.60 28.9 92 97 13.1 Owns, operates acute care hospitals

Table 1.Stock Market Winners Screen

Exchanges:NY = New York Stock Exchange; AM = American Stock Exchange; NM = Nasdaq National Market; NS = Nasdaq Small Cap

*calendar quarters

Source: Stock Investor/Media General. All data as of August 31, 1994.

intra-year turnaround as quickly as thetrailing quarterly data.

The second criterion required thatthe pretax profit margin be positive.This is determined by taking sales andsubtracting cost of goods sold, operat-ing expenses, interest expenses, anddepreciation and amortization, and di-viding the result by sales. The grossprofit margin was substituted and iscalculated by subtracting cost of goodsand operating expenses from sales anddividing the result by sales. The grossoperating margin is not as stringent asthe pretax profit margin used in theoriginal study, but it captures much ofthe same effect. It represents the typeof compromise often necessary whenimplementing a screening strategy.Requiring a positive gross profit marginbrought the number of companies downto 87.

Requiring a positive five-year earn-ings growth rate or a positive grossoperating profit margin by themselvesare not very restrictive criteria. How-ever, they help to screen out some ofthe very weak firms that have sometime to go before turning around, ifever. It is interesting to note that funda-mental measures such as profit mar-gins rose substantially during the major

price moves. Requiring a high profitmargin as a screening criterion wouldmean missing at least part of this majorprice advancement.

Price Momentum

The next set of criteria help to iden-tify stocks that have already shownupward price movement. Patience isrequired when selecting stocks usingpurely contrarian rules. It often takestime for the market to recognize valuein the firm. The study indicated thattechnical measures such as strong andimproving relative strength, a currentstock price near its high, and a highDatagraphs ranking, point to stockslikely to advance further.

The weighted relative strength rank-ing was the primary price momentumindicator used in the study. The weight-ing required that the most recent quar-terly price change be given a weight of40% and the previous three quarterseach weighted 20%. The weighted pricechanges were then ranked for all thestocks and the relative position indi-cated via percentage rank. Stocks witha 90% relative strength rank had aweighted price change better than 90%of all the firms.

The study indicated that the winnershad relative strength rank of 70% orbetter before their main price move.When this criterion was applied it re-duced the number of companies to 33.The study also indicated that it is bestif relative strength rank also increasedfrom the previous quarter. While thisfurther screen was not applied to nar-row the data set, this is the type ofvariable that could be investigated af-ter the screening process.

The other price-based screening cri-terion developed in the study requiredthat the current price be within 15% ofthe high price for the last two years.This rule reinforces the requirement ofprice strength. In applying this screen,the 52-week high was used in place ofthe two-year high. The 52-week highscreen is not as strict but the 52-weekhigh is more readily available for moststocks. When applied to our list of stocksit cut the number of firms down to 16, asshown in Table 1.

The original study was performedusing the Datagraph books (publishedby William O’Neil and sold primarily toinstitutional investors), which includeboth fundamental and technical data. Itwas found that the winners usually hada high Datagraph rating in the buy quar

Definitions of Screens and TermsThe following is a short description of the screens and terms used in the table.

Price-to-Book-Value Ratio: Market price per share di-vided by book value (assets less liabilities) per share. Ameasure of stock valuation relative to net assets. A high ratiomight imply an overvalued situation; a low ratio might indi-cate an overlooked stock.

Quarterly Earnings per Share: Net income from con-tinuing operations of a firm divided by the number ofcommon stock shares outstanding. Second Quarter 1994:Earnings per share for the most recent quarter. SecondQuarter 1993: Earnings per share four quarters ago. Pro-vides a comparison for the latest quarter’s earnings pershare. First Quarter 1994: Earnings per share one quarterago. First Quarter 1993: Earnings per share five quartersago. Provides a comparison figure for this year’s first quarterearnings per share. Comparison of quarterly earnings helpsto provide an indication of accelerating earnings.

Annual EPS Growth Rate: Annual growth in earningsper share from continuing operations over the last five fiscalyears. A measure of how successful the firm has been ingenerating the bottom line, net profit.

Relative Strength Rank: Relative price change, com-puted here with the most recent quarterly price changegiven a weight of 40% and the three previous quarters eachweighted 20%. The weighted price change is then comparedto other stocks over the same time period.

Price as % of 52-Wk High: Most recent market pricedivided by highest market price over the last 52 weeks.

No. of Shares Outstanding: Total number of shares ofstock held by shareholders. Provides an indication of thetrading liquidity of the firm.

ter. The Datagraph rating uses a pro-prietary weighting mix for “reportedearnings, capitalization, sponsorship,relative strength, price-volume charac-teristics, group rank and other factors.”Since most individuals will not haveaccess to this charting and data ser-vice, it was not used in the screen. It isworth noting, however, that the otherrules seem to reflect much of the infor-mation contained within the Datagraphrating.

Limited Float

This criterion examines the numberof shares outstanding, often termed thefloat. The study found that 90% of thefirms had fewer than 20 million sharesoutstanding before their main price rise.

The midpoint or median figure was 5.7million shares, which doubled duringthe two years that each “winning” stockwas held. This probably indicates manyof the firms split their shares duringtheir big price increase.

Some investors look for a stock tohave a limited float with the belief thatthe price move on positive informationwill be magnified by a limited numberof shares available. Applying thisscreening criterion did not cause anyfirms to drop out. It seems that in today’smarket the other screening criteria dida good job of filtering out the largerfirms that have more shares outstand-ing.

The list of stocks passing all thescreens is presented in Table 1. Theyare ranked on the price-to-book-value

ratio. The list represents a diverse setof relatively small firms. Like all screens,this list provides a beginning point forfurther in-depth analysis.

Conclusion

Examining the characteristics of paststock market winners might prove in-teresting. But using all of the rules maylead to a very restrictive screen in thepresent. Screens can also be in someways contradictory. In this case, thereis a strong price-to-book neglect screentied to look-at-me earnings and stockprice movements.

But it is hard to argue with screensthat try to find neglected stocks on theverge of a fundamental and technicalturnaround.

Weighted relative strength rank of at least 70

Found in User-Defined Group

Weighted Rel Strenth > X Weighted Relative Strength greater than value which returns only 30% of companies

Note: Try different values for X until 30% of companies pass. For example with a database size of 7,000 companies X should be varied until about 2,100

companies pass screen.

Calculation of the Weighted Relative Strength

Weighted Relative Strength = (Price Change Q1 * 40%) + (Price Change Q2 * 20%) + (Price Change Q3 * 20%) + (Price Change Q4 * 20%)

((Price/Price_Q2 - 1)*40) + ((Price_Q2/Price_Q3 - 1)*20) + ((Price_Q3/Price_Q4 - 1)*20) + ((Price_Q4/Price_Q5 - 1)*20)

Low price-to-book-value ratio

Found in Multiples Group

PBVPS < 1.5 Price/Book is less than 1.5 (Note: original study required a more restrictive 1.0)

Accelerating quarterly earnings

Found in Income Statement - Quarterly Group:

EPSCon_Q1 > EPSCon_Q5 EPS Cont Q1 greater than EPS Cont Q5

EPSCon _Q2 > EPSCon _Q6 EPS Cont Q2 greater than EPS Cont Q6

Found in User-Defined Group

Percent change between the recent quarter and its counterpart last year be greater than the percent change between the previous quarter and its

counterpart

Q1/Q5 EPS Change > Q2/Q6 EPS Change EPS Change greater than Q2/Q6 EPS Change

Q1/Q5 EPS Change = (EPS Cont Q1 / EPS Cont Q5 - 1) * 100

(EPSCon_Q1/EPSCon_Q5 - 1)*100 Calculation of the User-Defined Field—Q1/Q5 EPS Change

Q2/Q6 EPS Change = (EPS Cont Q2 / EPS Cont Q6 - 1) * 100

(EPSCon_Q2/EPSCon_Q6 - 1)*100 Calculation of the User-Defined Field—Q2/Q6 EPS Change

Positive five-year growth rate

Found in Growth Rates Group

EPSCon_G5F > 0 Five-year earnings per share growth from continuing operations is greater than 0%

Positive profit margins (Note: This was changed from the pre-tax margin in the original study)

Found in Ratios Group

NPM_12M > 0 Net Margin 12m greater than zero

or

GPM_12M > 0 Gross Margin 12m greater than zero

Stock Selling within 15% of its maximum price

Found in User-Defined Group

Price as % of 52wk high => 85 Price as % of 52wk high greater than or equal to 85%

Calculation of the Price as % of 52wk high

(Price/PriceH_52W)*100 Price as % of 52wk high = (Price / Price high 52W) * 100

Fewer than 20 million common shares outstanding

Found in Price & Share Data Group

Shr_AQ1 <= 20 Shares Average Q1 less than or equal to 20 million

Screening CriteriaFor use with AAII's Stock Investor program (version 2.1)

Earnings estimates embody the expectations that are builtinto a stock price. Tracking these expectations and theirchanges can be a rewarding strategy.

Analyzing the Analysts:Stock Valuation andEarnings RevisionsBy John Bajkowski

John Bajkowski is AAII’s financial analyst and editor of Computerized Investing.

Investors quickly learn that the mar-ket is forward-looking. Security pricesare established through expectations,and prices change as these expecta-tions change or are proven incorrect. Inthe last 10 years, we have seen a sig-nificant increase in the services thattrack and analyze expected earningsper share estimates.

Services such as I/B/E/S, Nelson’s,Standard & Poor’s, and Zacks provideconsensus earnings estimates by track-ing the estimates of thousands of in-vestment analysts. Tracking these ex-pectations and their changes is an im-portant and rewarding strategy for stockinvestors. [See the Stock Investing Ba-sics column starting on page 27 forsources of earnings estimates.]

In using earnings estimates, the firstrule to keep in mind is that the currentprice usually reflects the consensusearnings estimate. It is common to seeprice declines for stocks that reportearnings increases from the previousreporting period because in manycases, while the actual earnings repre-sent an increase, the increase is not as

great as the market had expected. Earn-ings surprises occur when a companyreports actual earnings that differ fromconsensus earnings estimates.

During the earnings reporting sea-son, financial newspapers such as theWall Street Journal provide daily re-ports on earnings announcements.Firms with significant earnings surprisesare often highlighted.

Positive earnings surprises occurwhen actual reported earnings are sig-nificantly above the forecasted earn-ings per share. Negative earnings sur-prises occur when reported earningsper share are significantly below theearnings expectations. The stock pricesof firms with significant positive earn-ings surprises show above averageperformance, while those with nega-tive surprises have below average per-formance.

Changes in stock price resulting froman earnings surprise can be felt imme-diately, and the surprise also has along-term effect. Studies indicate thatthe effect can persist for as long as ayear after the announcement. This

means that it does not make sense tobuy a stock after the initial price de-cline of a negative earnings surprise.There is a good chance that the stockwill continue to underperform the mar-ket for some time. It also indicates thatit may not be too late to buy into anattractive company after a better thanexpected earnings report is released.

Not surprisingly, large firms tend toadjust to surprises more quickly thansmall firms. Larger firms are tracked bymore analysts and portfolio managers,who tend to act quickly.

Firms with a significant quarterly earn-ings surprise also often have earningssurprises in subsequent quarters. Whena firm has a surprise, it often is a signthat other similar surprises will follow.This is sometimes referred to as thecockroach effect—like cockroaches, yourarely see just one earnings surprise.

Revisions to earnings estimates leadto price adjustments similar to earn-ings surprises. When earnings estimatesare revised significantly upward—5%or more—stocks tend to show aboveaverage performance. Stock prices offirms with downward revisions showbelow average performance.

Changes in estimates reflect changesin expectations of future performance.Perhaps the economic outlook is bet-ter than previously expected, or maybea new product is selling better thananticipated.

Revisions are often precursors toearnings surprises. As the reportingperiod approaches, estimates normallyconverge toward the consensus. A flurryof revisions near the reporting periodcan indicate that analysts missed themark and are scrambling to improvetheir estimates.

Table 1 represents the results of ascreen for firms with significant earn-ings estimate revisions. AAII’s Stock In-vestor, which contains earnings estimatesfrom I/B/E/S, was used to perform thescreening. About half of the 8,000 secu-rities in Stock Investor include earningsestimates. The first screen filtered outthose firms with less than four esti-mates for the current fiscal year. Thisfilter helps to ensure that revisions ac-tually reflect a change in general con-

Price ChangeEPS Estimate Range

Table 1.Firms with Upward Revisions in Earnings Estimates

I/B/E/S EPS Estimate

na = not available Exchanges:NY= New York Stock Exchange; AM= American Stock Exchange; NM= Nasdaq National Market; NS= Nasdaq Small Cap

Sources: Stock Investor/Media General; I/B/E/S; and S&P StockGuide. Data as of July 29, 1994.

No. No. of7/1 7/29 Change of Revisions High Low Last Qtr. Last Yr.

Company (Exchange) ($) ($) (%) Estimates Up in Mo. ($) ($) (%) (%) Description

Large Cap (above $1.5 billion)Alcan Aluminum Ltd (NY) 0.05 0.08 60 31 5 0.55 (0.35) 17 24 Produces aluminum

U.S. Cellular (NY) 0.08 0.12 50 9 2 0.30 (0.28) 9 (0) Own, oper cellular phone systems

Int’l Business Machines (NY) 3.03 4.00 32 23 20 4.80 3.45 8 39 Mfrs business machines & computers

Southern Pacific Rail (NY) 0.85 1.00 18 9 5 1.15 0.80 (10) na Oper freight railroad network in the West

Tandem Computer (NY) 1.00 1.15 15 15 7 1.33 0.95 17 34 Disk drive & microcomputer system

General Instrument (NY) 2.61 3.00 15 13 7 3.20 2.50 12 45 Systems, equip to cable & satellite TV indus

Inland Steel (NY) 1.60 1.82 14 20 8 2.25 1.19 13 48 Integ steel & steel serv

Coca Cola Enterprises (NY) 0.31 0.35 13 13 5 0.42 0.20 (6) 15 Soft drink distributer

Georgia Gulf (NY) 1.70 1.90 12 15 9 2.50 1.50 24 95 Prod commodity chemicals

Apple Computer (NM) 1.45 1.60 10 33 15 1.80 1.11 12 21 Computer systems, tech application

Dana Corp (NY) 2.00 2.20 10 13 9 2.30 1.55 6 4 Manufactures parts for auto industry

Aluminum Co of Amer (NY) 1.37 1.50 9 24 4 2.10 1.00 15 10 Aluminum production; finished aluminum prod

Genentech Inc (NY) 1.00 1.09 9 16 6 1.18 0.90 1 16 Devlp health care prods using gene splicing

Lyondell Petroch (NY) 1.13 1.23 9 12 4 1.62 0.45 2 38 Integrated petrochemical & petroleum processor

Eastman Chemical (NY) 2.78 3.00 8 13 4 3.30 2.55 16 na Chemical co spin-off of Eastman Kodak

Medium Cap ($250 million to $1.5 billion)ASARCO Inc (NY) 0.05 0.15 200 13 3 1.30 (1.50) 21 53 Prod silver, copper, lead

Read-Rite Corp (NM) 0.07 0.12 71 5 2 0.15 0.06 10 38 Magnetic recording head for disk drive

Amdahl Corp (AM) 0.30 0.40 33 15 2 0.65 0.00 2 31 Computers, software, communic’n systems

Cobra Golf (NM) 1.48 1.80 22 4 4 1.85 1.52 23 na Designs, mfrs, mkts golf clubs

Tencor Instruments (NM) 1.10 1.33 21 4 2 1.60 1.00 21 108 Water defect inspection systems

NACCO Industries (NY) 3.55 4.23 19 4 2 4.60 3.50 10 8 Mines/mkts lignite & bituminous coal

Cyrix Corp (NM) 1.50 1.78 19 7 5 1.85 1.50 44 43 IBM-comp microprocess/coprocess

VLSI Technology (NM) 0.89 1.04 17 15 10 1.10 0.88 3 23 Designs, mkts integrated circuits

Novellus Systems (NM) 2.00 2.33 17 12 8 2.35 2.00 1 44 Chemical vapor deposition equip

Quantum Corp (NM) 2.73 3.18 16 12 6 4.00 2.60 (7) 47 Rigid disk drives

Magma Copper (NY) 0.75 0.87 16 16 3 1.45 (0.28) 13 44 Owns/oper copper mines

Modine Mfg (NM) 1.65 1.90 15 8 5 1.95 1.55 5 36 Mfrs, sells heat transfer equip

Oak Industries (NY) 1.48 1.70 15 5 4 1.79 1.50 18 (5) Mfrs electronic controls & circuits

Robert Half (NY) 1.50 1.70 13 4 3 1.73 1.60 22 84 Personnel placement services

Regal-Beloit Corp (AM) 1.83 2.06 13 4 4 2.10 2.05 2 35 Makes cutting tools, taps, dies, reamers

Small Cap (below $250 million)Coho Energy (NM) 0.07 0.10 43 4 1 0.18 0.03 22 (28) Explor, devlp oil & natural gas

Arkansas Best (NM) 0.50 0.60 20 5 2 0.68 0.40 (4) 39 Ship general commodities

Alaska Air Group (NY) 0.85 1.00 18 16 6 1.85 0.47 4 28 U.S. & Alaska airline service

Network Equip (NY) 0.40 0.45 13 7 3 0.65 0.40 14 25 Advanced communications prods

Tultex Corp (NY) 0.25 0.28 12 9 1 0.40 0.25 (5) (41) Mfrs sportswear & yarn

Redman Industries (NM) 2.05 2.28 11 4 2 2.50 2.02 8 na Producer of manufactured homes

Hauser Chemical Research (NM) 0.20 0.22 10 4 1 0.25 0.03 (12) (56) Petrochemical analytical serv

Dura Pharmaceuticals (NM) 0.11 0.12 9 4 2 0.14 0.10 29 121 Specialty pharmaceutical concern

Conmed Corp (NM) 1.02 1.10 8 5 2 1.20 0.75 0 10 Mfg of disposable medical devices

SPX Corp (NY) 1.10 1.18 7 10 4 1.60 1.05 15 5 Automotive components & serv prods

Nuevo Energy (NY) 0.70 0.75 7 9 1 1.25 0.57 6 (16) Explor, mkt oil & gas

Loyola Capital (NM) 1.53 1.63 7 4 2 1.70 1.55 12 59 Federal stock savings & loan assoc

Sullivan Dental Products (NM) 0.80 0.85 6 5 2 0.93 0.80 10 (23) Distrib dental prods & equip

Foothill Group A (NY) 1.56 1.64 5 4 4 1.69 1.60 5 17 Financial co specializing in lending

Transmedia Network (NM) 0.43 0.45 5 4 1 0.48 0.36 (11) 78 Restaurant charge cards

Table 2.Using Consensus Earnings Estimates

Earnings EstimatesFirms with high expected earnings growth tend tounderperform the market because it is difficult to meet themarket’s high expectations. Companies with low earningsexpectations tend to do better than expected.

Prices embody current earnings estimates.

Earnings Estimate RevisionsStock prices of firms with significant upward revisions (5% ormore) generally outperform the market. Firms with signifi-cant downward revisions underperform the market.

Earnings revisions are often a precursor to earnings sur-prises. Stock prices react positively to upward revisions.

Earnings SurpriseStock prices of firms that significantly exceed their earningsexpectations (positive earnings surprise) outperform themarket, while those with negative earnings surprises

underperform.

The earnings surprise effect is long-lasting. The greatest effectof the surprise can be felt immediately, but the effect of theearnings surprise can be seen for as long as a year. The effectof the surprise tends to be longer lasting for negative earningssurprises.

The stock prices of large firms adjust to surprises more quicklythan those of small firms.

Earnings surprises often follow in groups—the cockroach ef-fect.

The chance of an earnings surprise is greater if the range ofestimates for a company is wide.

The price move can be more dramatic, however, if an earningssurprise occurs for a firm with a very tight range of earningsestimates.

sensus, not just a change by one or twoanalysts. However, requiring that astock have at least four analysts report-ing earnings estimates will knock outmost of the very small-capitalizationstocks.

The next filter required that the firmhave a change in its current fiscal-yearestimate over the course of the lastmonth, leaving about 400 firms. Thestocks were divided into three marketcapitalization groups. The percentagechange in the estimate from one monthto the next was calculated. The 15 firmswithin each group with the greatestpercentage increase in estimate arelisted in Table 1.

The screen includes a number of cy-clical firms experiencing better thanexpected demand for their products.For example, producers of metals suchas Alcan Aluminum, Inland Steel, Alu-minum Company of America, andASARCO are seeing strong enough de-mand for products that they are able toraise their prices for the first time inyears.

Whenever your filter involves the per-centage change of a variable, there is a

risk that firms with very small basenumbers will dominate. With an in-crease of 200%, ASARCO had the high-est percentage revision in the me-dium-cap group. This jump in percent-age change is caused by an increase inearnings estimate from $0.05 to $0.15.When working with percentagechanges, it is helpful to use an addi-tional screen to confirm the signifi-cance of the change.

The number of estimates for eachfirm is provided to help gauge theinterest in the firm and the meaning-fulness of the overall estimates. Thelarger the firm, the greater the numberof analysts that will track it.

The number of revisions upward in-dicates how many analysts have re-vised their estimates upward in thelast month. When compared to thenumber of analysts making estimates,this is a confirmation of the signifi-cance of the percentage change in es-timates. You can put more faith in arevision if a large percentage of theanalysts tracking a firm have revisedtheir estimates. VLSI Technology, lo-cated in the medium-cap group, had

10 of the 15 analysts tracking the firmrevise their estimates upward. This com-pares with Amdahl, which had only twoof its 15 estimates revised upward.

Some investors screen for earningsestimate revisions by looking at thenumber of revisions. Stock prices offirms with more upward revisions thandownward revisions have shown aboveaverage returns, while those with moredownward revisions tend tounderperform.

Examining the range of estimates pro-vides an indication of the consensuswithin the group of estimates. A widerange of estimates would point to greatdisagreement among analysts, indicat-ing greater uncertainty and greaterchance for an earnings surprise.

The 13 estimates for ASARCO rangefrom a low of –$1.50 to a high of $1.30—a range of $2.80. This is a wide rangecompared to the Novellus Systemsearnings estimates, which range from$2.00 to $2.35.

The price move can be more dra-matic, however, if an earnings surpriseoccurs for a firm with a very tight rangeof earnings estimates.

I/B/E/S EPS Estimate: The consensus of analysts’ esti-mates for earning per share. Reflects the earnings level builtinto the stock price.

EPS Estimate % Change: The percentage change in theconsensus earnings estimate over a one-month period. Pricesof firms with positive revisions tend to perform better thanaverage.

No. of Estimates: Number of analysts providing earningsper share estimates. Indicates how widely a firm is followed.Widely followed firms tend to react more quickly to estimaterevisions.

No. of Revisions Up in Mo.: The number of analystswho revised their earnings per share estimate for thestock upward during the month. Provides an indication ofstrength of the earnings revision.

EPS Estimate Range: The highest and lowest earningsper share estimates given by analysts for most recentmonth-end. Indicates the level of consensus among ana-lysts. The wider the range, the greater the divergence inopinion and the greater the chance for an earnings sur-prise.

Price Change: The percentage change in the stock pricefor the last quarter and the last year.

Definitions of Screens and TermsThe following is a short description of the screens and terms used in the table.

The last two columns of data help toillustrate the recent price moves as themarket has adjusted to changes in ex-pectations. Most of these stocks haveexperienced fairly strong price moveson both a short-term (three-month) and

intermediate-term (one-year) basis. Alarger price change for the last quarterthan for the last year would indicate apositive change in price trend. NACCOIndustries and Oak Industries are twosuch examples.

Earnings estimates are important.They embody the expectations builtinto a stock price. Table 2 summarizesthe main points to keep in mind whendealing with consensus earnings esti-mates.

Screening CriteriaFor use with AAII's Stock Investor program (version 2.1)

Significant analyst following

Found in Earnings Estimates Group

EPSN_EY0 > 4 EPS Est Y0-Est no is greater than 4

Increase in earnings estimate over the course of the last month

Found in Earnings Estimates Group

EPS_EY0 > EPSPM_EY0 EPS Est Y0 is greater than EPS Est Y0-Last Month

and/or

EPS_EY1 > EPSPM_EY EPS Est Y1 is greater than EPS Est Y1-Last Month

Upward revisions in last month with no downward revisions in last month

Found in Earnings Estimates Group

EPSUM_EY0 > 0 EPS Est Y0-Rev Up is greater than zero

EPSDM_EY0 = 0 EPS Est Y0-Rev Down is zero

and/or

EPSUM_EY1 > 0 EPS Est Y1-Rev Up is greater than zero

EPSDM_EY1 = 0 EPS Est Y1-Rev Down is zero

While price-earnings ratios are indicators of value, anapproach that simply looks for low ratios can be misleading.

The Basic Techniques forUsing Low P/Es to FindUndervalued StocksBy John Bajkowski

John Bajkowski is AAII’s financial analyst and editor of Computerized Investing.

ings ratios while those with low price-earnings ratios are expected to havelow growth. Screening just for stockswith a low price-earnings ratio mayleave you with a list of companies withlittle or no growth prospects.

Table 1 illustrates the problems as-sociated in screening for low price-earn-ings stocks. AAII’s Stock Investor data-base of 8,000 stocks was used to per-form the screening. The first screeningfilter excluded financial firms becausetheir non-standard financial statementsdo not allow for direct comparison withfirms in other industries. The next set offilters required that the firms have fiveyears of data and that earnings pershare must be positive for the last 12months. Table 1 presents the 20 secu-rities with the lowest price-earningsratio that passed the screens. The price-earnings ratios for the these securitiesrange from 1.3 to 4.2 compared with theS&P 500’s current price-earnings ratioof 19.2.

Any screen requires that detailedanalysis be performed on the resultinglist of companies, and for good reason.

This list is loaded with troubled firms.Studies indicating the advantages ofinvesting in low price-earnings stocksuse large portfolios of stocks to reducethe higher risk of investing in any singlestock. Littlefield, Adams, for example,is burdened by the uncertainty of anSEC investigation examining a widerange of corporate activities. RaytechCorp. has been operating under Chap-ter 11 bankruptcy protection since 1989and has a cloud of over 3,000 asbestos-related lawsuits hanging over its head.Also, some of the securities that passedthe screen are not corporations. Forexample, Samson Energy is a masterlimited partnership involved in the pro-duction, development, and acquisitionof oil and gas drilling operations, a rela-tively high-risk and uncertain activitythat often leads to lower price mul-tiples.

In looking at Table 1 you may noticethat only a few of the securities haveinformation entered for the average andrelative price-earnings valuation mod-els. A weakness of using the price-earn-ings ratio for analysis is that dividingprice by a negative earnings per sharefigure produces a meaningless num-ber.

Using Historical Averages

One useful way to use price-earningsratios is to compare current multiplesagainst historical averages. A currentprice-earnings ratio lower than its his-torical average would be a potentialsign that a stock is undervalued, whilea current price-earnings ratio that ishigh compared to its historical averagemight indicate an overvalued firm.Models that examine historical aver-ages assume that the growth prospectsof the firm have not changed funda-mentally over time and, based on his-torical relationships of price to earn-ings, the market is not correctly dis-counting the future earnings potentialof the firm.

To help illustrate the valuation as-pect of the average price-earningsmodel, it is common to multiply thefive-year average price-earnings ratioby the most recent 12 months’ earnings

The price-earnings ratio is one of themost basic measures of value for inves-tors. The price-earnings ratio, or mul-tiple, is computed by dividing a stock’sprice by its most recent 12 months’earnings per share. The price-earningsratio is followed so closely because itembodies the market’s expectations offuture company performance throughthe price component of the ratio andrelates it to historical company perfor-mance as measured by earnings pershare. This article will explore some ofthe basic price-earnings ratio tech-niques employed in searching for un-dervalued stocks.

Many studies point to profitability ofinvesting in out-of-favor stocks. Valueinvestors seek out firms with low price-earnings ratios with the belief that themarket may have overreacted to nega-tive news and is not correctly discount-ing their future earnings potential. Asimple search for low price-earningsratios, however, can be misleading as aguide to undervalued stocks. Typically,firms with high growth potential tradewith correspondingly high price-earn-

5-Year Average Ratio of P/E Ratio ofPrice EPS Average P/E Avg. P/E 5-Year Relative P/E Rel.as of Last P/E P/E Share Valuation P/E Share Valuation

6/3094 12 Mo. Ratio Ratio Valuation to Price Relative Valuation to PriceCompany (Exchange) ($) ($) (x) (x) ($) (x) (x) ($) (x) Description

Littlefield, Adams (AM) 4.81 3.58 1.3 nmf nmf nmf nmf nmf nmf Engages in the imprinting of athletic wear

Advanced Semicon (NM) 2.63 1.60 1.6 nmf nmf nmf nmf nmf nmf Mfrs equipment to produce semiconductor devices

Samson Energy LP (AM) 8.63 5.26 1.6 nmf nmf nmf nmf nmf nmf Prod, devlp, acquisition of oil & gas prop

Raytech Corp. (NY) 4.50 2.67 1.7 4.5 12.02 2.67 0.22 11.51 2.56 Mfrs asbestos & other friction prods

Castle Energy (NM) 14.50 8.00 1.8 nmf nmf nmf nmf nmf nmf Explor’n, devlp, produc’n oil & gas

Norex America (AM) 9.00 5.11 1.8 nmf nmf nmf nmf nmf nmf Mkts, provides leisure cruises

America West Airlines (NM) 3.75 1.88 2.0 nmf nmf nmf nmf nmf nmf Regional airline service

Audiovox Corp. (AM) 6.88 3.47 2.0 nmf nmf nmf nmf nmf nmf Mkts, distribs automotive aftermarket prods

WPP Group PLC (NM) 2.94 1.45 2.0 10.5 15.23 5.18 0.57 16.20 5.51 Multi-national marketing services

LVI Group Inc. (NY) 0.63 0.29 2.2 nmf nmf nmf nmf nmf nmf Electrical equipment distributors

Fairchild Corp. A (NY) 3.88 1.58 2.5 nmf nmf nmf nmf nmf nmf Motor freight carrier & aircraft replacement parts

Buffton Corp. (AM) 1.19 0.42 2.8 nmf nmf nmf nmf nmf nmf Mfrs avionic, electronic, plastic products

Highwood Restaurant (NM) 0.75 0.26 2.9 nmf nmf nmf nmf nmf nmf Explores for strategic, rare-earths, precious metals

Hathaway Corp. (NM) 4.00 1.19 3.4 nmf nmf nmf nmf nmf nmf Mfrs, sells elec power record equip & elec motors

Salant Corp. (NY) 7.00 1.89 3.7 nmf nmf nmf nmf nmf nmf Mfrs, mkts apparel prods for entire family

Allied Research Corp. (AM) 4.19 1.09 3.8 nmf nmf nmf nmf nmf nmf Devlps, prods weapons systems, ammo

Instrumentarium (NS) 11.25 2.89 3.9 nmf nmf nmf nmf nmf nmf Distribs hospital equip and supplies

IP Timberlands LP (NY) 24.13 6.02 4.0 6.1 36.72 1.52 0.35 41.30 1.71 Forest resource management

Maxco Inc. (NM) 8.25 2.06 4.0 12.1 24.93 3.02 0.75 30.28 3.67 Engineers, fabricates special machine for welding

Flanigan Enterprises (AM) 4.00 0.96 4.2 nmf nmf nmf nmf nmf nmf Pkg liquor stores, cocktail liquor

5-Year Average Ratio of P/E Ratio ofPrice EPS Average P/E Avg. P/E 5-Year Relative P/E Rel.as of Last P/E P/E Share Valuation P/E Share Valuation

6/3094 12 Mo. Ratio Ratio Valuation to Price Relative Valuation to PriceCompany (Exchange) ($) ($) (x) (x) ($) (x) (x) ($) (x) Description

Georgia Bonded (NM) 5.00 1.04 4.8 32.5 33.80 6.76 1.88 38.32 7.66 Mfrs elastomeric impregnated paper prods

WPP Group PLC (NM) 2.94 1.45 2.0 10.5 15.23 5.18 0.57 16.20 5.51 Multi-national marketing services

Adac Laboratories (NM) 8.38 1.23 6.8 34.4 42.31 5.05 1.71 41.23 4.92 Nuclear medicine computer systems

LaBarge Inc. (AM) 1.19 0.16 7.4 35.8 5.73 4.82 2.18 6.84 5.75 Mfrs tubular, modular, electrical prods

Merisel Inc. (NM) 8.75 1.07 8.2 37.0 39.59 4.52 2.15 45.09 5.15 Wholesale microcomputer hardware & software

Holly Corp. (AM) 28.50 3.87 7.4 25.8 99.85 3.50 1.35 102.40 3.59 Refining, explor’g, produc’g petroleum

Research Indus (NM) 8.00 0.60 13.3 44.4 26.64 3.33 2.41 28.34 3.54 Sales health care prods & real estate development

Dresser Indus (NY) 20.50 2.17 9.4 31.3 67.92 3.31 1.68 71.46 3.49 Supplier of prods & serv to oil & gas cos

ACC Corp. (NM) 13.75 1.70 8.1 26.7 45.39 3.30 1.37 45.65 3.32 Diversified telecommunications company

Teltronics Inc. (NS) 9.00 0.75 12.0 36.5 27.38 3.04 1.92 28.22 3.14 Mfrs telecommun equip

Anacomp Inc. (NY) 3.00 0.22 13.6 38.1 8.38 2.79 2.38 10.26 3.42 Provides computer & micrograph serv

Times Mirror (NY) 30.13 2.41 12.5 34.5 83.15 2.76 1.83 86.45 2.87 News & book publish, info serv, TV

USA Waste Service (NY) 11.63 0.79 14.7 39.8 31.44 2.70 2.34 36.23 3.12 Provides solid waste management and serv

Raytech Corp. (NY) 4.50 2.67 1.7 4.5 12.02 2.67 0.22 11.51 2.56 Mfrs asbestos & other friction prods

Dorchester Hugoton (NM) 13.25 0.94 14.1 36.8 34.59 2.61 2.04 37.58 2.84 Partnership in oil & gas properties

Providence & Worchester Rail (NM) 7.50 0.54 13.9 35.6 19.22 2.56 2.15 22.76 3.03 Operates interstate freight system

General Microwave (AM) 7.88 0.76 10.4 26.4 20.06 2.55 1.43 21.30 2.70 Electronic measurement and control equip

Pennzoil Co. (NY) 51.25 3.38 15.2 37.8 127.76 2.49 2.10 139.12 2.71 Explor, produc oil & gas; mfrs refined petro prods

Pacific Telecom (NM) 21.75 3.06 7.1 17.7 54.16 2.49 0.96 57.59 2.65 Telephone communication serv

Surgical Care (NY) 13.25 0.95 13.9 34.2 32.49 2.45 1.80 33.52 2.53 Devlps, owns outpatient surgical care facilities

Table 2.Price-Earnings Average Screen Ranked by the Ratio of Valuation Based on Average P/E to Current Price

Table 1.Low Price-Earnings Ratio Screen Ranked by Price-Earnings Ratio

nmf = no meaningful figure Exchanges:NY= New York Stock Exchange; AM= American Stock Exchange; NM= Nasdaq National Market; NS= Nasdaq Small Cap

Sources: Stock Investor/Media General; I/B/E/S; and S&P StockGuide. All data as of June 30, 1994.

5-Year Average Ratio of P/E Ratio ofPrice EPS Average P/E Avg. P/E 5-Year Relative P/E Rel.as of Last P/E P/E Share Valuation P/E Share Valuation

6/3094 12 Mo. Ratio Ratio Valuation to Price Relative Valuation to PriceCompany (Exchange) ($) ($) (x) (x) ($) (x) (x) ($) (x) Description

WPP Group PLC (NM) 2.94 1.45 2.0 10.5 15.23 5.18 0.57 16.20 5.51 Multi-national marketing services

Seneca Foods (NM) 22.50 3.01 7.5 18.0 54.18 2.41 1.00 59.00 2.62 Food processing; distrib services

Raytech Corp. (NY) 4.50 2.67 1.7 4.5 12.02 2.67 0.22 11.51 2.56 Mfrs asbestos & other friction prods

Merrimac Indus (AM) 8.38 1.16 7.2 15.5 17.98 2.15 0.86 19.56 2.33 Component/subsystem in signal process

Bowne & Co. (AM) 20.75 2.39 8.7 17.0 40.63 1.96 1.01 47.32 2.28 Finance & corporation printing

Lynch Corp. (AM) 26.38 3.56 7.4 14.9 53.04 2.01 0.85 59.31 2.25 Diversified manuf’g activities

Boston Celtic s NY) 20.00 2.81 7.1 14.8 41.59 2.08 0.81 44.62 2.23 Professional basketball team

Napco Security Sys (NM) 3.13 0.47 6.7 13.7 6.44 2.06 0.74 6.82 2.18 Mfrs security alarm prods

Park Electrochemical (NY) 28.75 2.58 11.1 22.0 56.76 1.97 1.23 62.20 2.16 Prints circuit materials

Robinson Nugent (NM) 5.75 0.47 12.2 24.3 11.42 1.99 1.33 12.25 2.13 Sockets & custom electromechan’l prods

Sun Microsystems (NM) 20.63 1.94 10.6 19.9 38.61 1.87 1.12 42.58 2.06 Supplies network-based computing systems

PSICOR Inc. (NM) 9.25 1.01 9.2 17.1 17.27 1.87 0.96 19.01 2.06 Provides perfusionist equip to hospitals

B.A.T. Indus PLC (AM) 12.31 1.19 10.3 19.3 22.97 1.87 1.07 24.95 2.03 Mfrs tobacco prods

REFAC Tech Devlp (AM) 7.00 0.79 8.9 15.5 12.25 1.75 0.91 14.09 2.01 Admins interna’l technology licenses

Plasti-Line (NM) 7.50 0.72 10.4 19.0 13.68 1.82 1.06 14.96 1.99 Designs, mkts illuminated outdoor signs

GTI Corp. (NM) 10.25 1.15 8.9 16.9 19.44 1.90 0.90 20.29 1.98 Mfrs electronic sealer components

Top Air Mfg (NS) 0.88 0.07 12.6 24.9 1.74 1.98 1.26 1.73 1.97 Designs, mfrs, sells agricultural equip

Bridgford Foods (NM) 8.31 0.62 13.4 25.0 15.50 1.87 1.33 16.16 1.94 Makes, distributes frozen foods

Utah Medical Prods (NM) 6.88 0.60 11.5 20.6 12.36 1.80 1.11 13.06 1.90 Dev/mfrs/mkts medical devices

United Stationers (NM) 9.75 1.15 8.5 15.0 17.25 1.77 0.82 18.48 1.90 Distribs wholesale office prods

Table 3.Price-Earnings Relative Screen Ranked by the Ratio of Valuation Based on P/E Relative to Current Price

S&P 500 Boston Celtics

Index Value Price-Earnings Ratio Stock Price Price-Earnings Ratio P/E Relative to MarketHigh Low EPS High Low High Low EPS High Low High Low($) ($) ($) (×) (×) ($) ($) ($) (×) (×) (×) (×)

1989 $359.80 $275.31 $22.87 15.7 12.0 $19.500 $13.500 $1.88 10.4 7.2 0.66 0.60

1990 $368.95 $295.46 $21.34 17.3 13.9 $19.125 $14.500 $1.23 15.6 11.8 0.90 0.85

1991 $417.09 $311.49 $15.97 26.1 19.5 $20.875 $16.125 $1.69 12.4 9.5 0.47 0.49

1992 $441.28 $394.50 $21.95 20.1 18.0 $23.875 $16.250 $1.19 20.1 13.7 1.00 0.76

1993 $470.94 $429.05 $21.88 21.5 19.6 $21.625 $16.375 $0.80 27.0 20.5 1.26 1.04

5-year average 20.1 16.6 17.1 12.5 0.86 0.75

5-year average (combined high & low) 18.4 14.8 0.81

Table 4.Sample Calculations Using Boston Celtics LP as an Example

Recent S&P Price-Earnings Ratio 19.6

Recent Stock Price $20.00Earnings per Share (last 12 mos.) $2.81Price-Earnings Ratio 7.1

Average P/E ModelShare Valuation = Average P/E Ratio × EPS

= 14.8 × $2.81= $41.59

Ratio of Share Valuation to Current Price= Valuation ÷ Current Price= $41.59 ÷ $20.00= 2.08

P/E Relative ModelAdjusted P/E = Current Market P/E × Average P/E Relative

= 19.6 × 0.81= 15.9

Share Valuation = Adjusted P/E × EPS= 15.9 × $2.81= $44.68

Ratio of Share Valuation to Current Price= Valuation ÷ Current Price= $44.68 ÷ $20.00= 2.23

ton Celtics has trended up as the pricelargely stayed flat and earnings havedecreased. The average price-earningsratio over the last five years is 14.8.Multiplying the earnings per share forthe last 12 months by the average price-earnings ratio leads to a valuation of$41.59. This is 2.08 times the currentprice. Boston Celtics just missed mak-ing the listing in Table 2, but made thelisting in Table 3, which is based uponthe price-earnings relative model.

One weakness with average price-earnings approach is that it looks purelyat historical relationships, while the cur-rent market price is driven by futureexpectations. The trailing 12-monthearnings per share figure may be un-usually high or low due to a one-timeevent, or the historical average maynot reflect a change in the company,industry, or economic environment. Forthe Boston Celtics Limited Partnershipthe trailing earnings per share figure of$2.81 represents a tremendous increaseover the $0.80 for the last fiscal year.Much of this earnings increase can beattributed to the selling of an option toFox Television providing the right topurchase a 26% ownership interest inthe television station owned by thepartnership. The earnings figure alsoincludes an insurance settlement in-volving the death of a Celtics basket-ball player.

To get around the limitation of his-torical earnings per share, estimatedearnings can be used. Consensus earn-ings estimates, however, are usuallyonly available for larger, more activelyfollowed companies. A screen requir-ing consensus earnings estimates willexclude a number of interesting ne-glected stocks.

Price-Earnings Relative Models

Another way to use price-earningsratios is to compare them with the in-dustry price-earnings ratio, or even theoverall market ratio. Based upon eco-nomic conditions and factors, the fairvalue of the market can change. Forexample, the market can support higherprice-earnings ratios under the condi-tion of low interest rates than it can

under high interest rate conditions. Theprice-earnings relative is determinedby dividing a company’s price-earn-ings ratio by that of the market. Basedon relative growth and risk expecta-tions, companies trade at market mul-tiples greater or smaller than that ofthe market. One would expect a com-pany with prospects better than themarket or with lower risk to have ahigher price-earnings ratio than themarket. Comparing a firm to its indus-try is an equally useful technique thathas the benefit of isolating interestingcandidates within a specific industry.

A price-earnings relative above 1.00would indicate that a company’s price-earnings ratio is typically above themarket’s price-earnings ratio, while aprice-earnings relative below 1.00would indicate that a company’s price-earnings ratios tends to be lower thanthe market’s. By averaging the price-earnings relative over time, you canestimate a price-earnings relative thata stock tends to follow.

Table 4 also illustrates the calcula-tions used to determine the price-earn-ings relative. Over the last five yearsthe Boston Celtics have averaged aprice-earnings multiple below that ofthe S&P 500 leading to an average price-earnings relative of 0.81. Multiplyingthe price-earnings relative by themarket’s current price-earnings ratioprovides an adjusted stock price-earn-ings ratio. The assumption is that themarket is fairly valued and that thecompany’s relationship to the markethas not changed. A stock price valua-tion can be determined by multiplyingthis adjusted stock price-earnings ratioby the earnings per share figure. Theratio of the valuation to the currentprice provides a useful screening mea-sure. Table 3 is based upon this ratio ofvaluation determined by the price-earnings relative valuation divided bythe current price.

To eliminate outliers, the high limitfor the price-earnings ratio was tied tothe market’s price-earnings level. Com-panies whose price-earnings ratios weremore than double the market’s in anyyear were excluded. This filter was morestringent than the one used with the

per share to arrive at a price estimate.Comparing this valuation price to thestock’s current price provides a usefulratio for screening stocks. Table 2 pro-vides the results of a stock screen basedupon the average price-earnings ratio.The information pertaining to the aver-age price-earnings model is highlighted.

Because the screen relies on the five-year average price-earnings ratio, thisscreening required that the firms havefive years of positive earnings per share.This is why so many of the firms listedin Table 1 did not make it into Table 2.Beyond negative earnings, which leadto meaningless price-earnings ratios,unusually low earnings may also throwoff standard price-earnings ratioscreens. Short-term drops in earningsdue to events such as special charges,extraordinary events, or in some caseseven recessions may lead to unusuallyhigh price-earnings ratios. As long asthe market interprets the earnings de-crease as temporary, the high price-earnings ratio will be supported. Be-cause the average price-earnings modelrelies on a normal situation, these “out-lier” price-earnings ratios must be ex-cluded. When performing a hands-onevaluation you can manually excludeyears with negative earnings or unusu-ally high price-earnings ratios. However,when screening a large universe ofstocks, it is best to establish criteriathat eliminate companies with extremeprice-earnings ratios. For the averageprice-earnings screen, companies withratios above 100 for any of the last fivefiscal years were excluded. If you wantto be more conservative, a tighter re-quirement, such as ratios above 40 or50, might be specified.

The top 20 firms ranked on the ratioof average price-earnings valuation tocurrent price are shown in Table 2. Toarrive at the valuation, the earnings pershare for the last 12 months was multi-plied by average price-earnings ratio.

Table 4 illustrates the calculationsinvolved in the average price-earningsmodel. The Boston Celtics Limited Part-nership owns and operates the BostonCeltics as well as a Boston radio andtelevision station. Over the last fiveyears, the price-earnings ratio of Bos-

EPS Last 12 Mo.: Earnings from continuing operations forthe most recent 12 months divided by the number of com-mon shares outstanding.

P/E Ratio: Market price per share divided by most recent 12months’ earnings per share. A measure of the market’sexpectations regarding the firm’s earnings growth and risk.

5-Year Average P/E Ratio: An average of the high and lowprice-earnings ratios for the past five years. Provides a baselevel from which to compare the current price-earnings ratio.

Average P/E Share Valuation: Five-year average price-earnings ratio multiplied by earnings per share for the mostrecent 12 months. Gives an estimate of price supported byhistorical price-earnings average. Can also be computed withexpected future earnings per share.

Ratio of Avg. P/E Valuation to Current Price: Estimatedaverage price-earnings share valuation divided by currentprice. A ratio of 1.00 indicates that the valuation estimate is

equal to the current price. A ratio above 1.00 indicates anundervalued security while a ratio below 1.00 indicates anovervalued security.

5-Year P/E Relative: Ratio of historical company price-earn-ings levels relative to those of the S&P 500 index. Provides anindication as to whether the company traditionally trades at apremium or discount to the market.

P/E Relative Share Valuation: Price-earnings relative multi-plied by the company earnings per share. Gives an estimate ofstock price value supported by the historical relationship of theprice-earnings ratio to the market’s, and the current market andcompany situation. Can also be computed with expected com-pany earnings per share.

Ratio of P/E Rel. Valuation to Current Price: Estimatedprice-earnings relative share valuation divided by current price.A ratio of 1.00 indicates that the valuation estimate is equal tothe current price. A ratio above 1.00 indicates an undervaluedsecurity; a ratio below 1.00 indicates an overvalued security.

Definitions of Screens and TermsThe following is a short description of the screens and terms used in the tables.

average price-earnings ratio screen, cre-ating a more conservative screen andleading to a largely different list of com-panies in Table 3 than in Table 2.

The companies passing these screenswere sorted based upon their ratio ofprice-earnings relative share valuationto current price, with the top 20 firmslisted in Table 3. The fields pertainingto the price-earnings relative model arehighlighted. As with the previousscreens, the list is made up of compa-nies that have fallen out-of favor, suchas B.A.T. Industries and SunMicrosystems. B.A.T. is a tobacco com-pany caught in the midst of a stormsurrounding the tobacco industry. SunMicrosystem, once a high-flying com-puter workstation manufacturer , hasseen its earnings decline in recent years.Both firms find their current price-earn-ings ratios hovering at about half themarket level of 19.6, not at their histori-cal slight premium to the market.

Conclusion

Screening for stocks by looking at

price-earnings ratios can help highlightfirms that have fallen out of favor. As inany technique, there are many ways totake a simple rule and expand upon itto meet your objectives. Looking atjust low price-earnings stocks will high-light companies the market is neglect-ing, but this screen also tends to high-light the most troubled issues that re-quire very detailed analysis. It is alsocommon for low price-earnings screensto be dominated by a few industriesthat are currently out of favor.

Screening based upon an examina-tion of the average price-earnings ratioallows you to seek out companies thathave fallen out of favor, but may not beas troubled as firms from the pure lowprice-earnings screen. The averageprice-earnings model looks to pastearnings valuation to help set a bench-mark comparison. It identifies firms thathave deviated from their normal valu-ation level, with the expectation thatthey will move back toward their typi-cal levels. This model also has its weak-nesses: It assumes that nothing funda-mental to the company, industry, or

market has significantly changed.The primary benefit of the price-earn-

ings relative model over the averageprice-earnings ratio is that it allows foradjustments to broad economicchanges affecting the market. It identi-fies those stocks that have deviatedfrom their long-term relationships to abenchmark index, while still assuminga stable company relationship to themarket over time. As with all of thetechniques presented, industries thathave fallen out of favor may dominatethe analysis.

This article has focused purely onidentifying primary screening criteria.Screening for low price-earnings firmsturns up companies that have fallenout of favor, some for good reasons. Inconstructing screening criteria, you maywish to include a number of condition-ing criteria that help indicate items suchas the future earnings potential of thefirm, the financial strength of the firm,as well as the strength of the firm withinits industry. Investing in low price-earn-ings stocks can be rewarding, but cau-tion is required.

Low PE Versus Historical Average PE Ratio

Found in Company Information Group:

IND_2_DIG <> 07 Sector is not equal to Financial

IND_3_DIG <> 0933 Industry is not equal to Real Estate Operations

Found in User Defined Group:

PE Y1 < 50 User Defined PE Y1 = Price_Y1 / EPS_Y1

(Note: This can be varied to make screen more or less restrictive)

PE Y2 < 50 User Defined PE Y2 = Price_Y2 / EPS_Y2

(Note: This can be varied to make screen more or less restrictive)

PE Y3 < 50 User Defined PE Y3 = Price_Y3 / EPS_Y3

(Note: This can be varied to make screen more or less restrictive)

PE Y4 < 50 User Defined PE Y4 = Price_Y4 / EPS_Y4

(Note: This can be varied to make screen more or less restrictive)

PE Y5 < 50 User Defined PE Y5 = Price_Y5 / EPS_Y5

(Note: This can be varied to make screen more or less restrictive)

Found in the Income Statement - Annual Group:

EPS_Y1 > 0 EPS Y1 greater than 0

EPS_Y2 > 0 EPS Y2 greater than 0

EPS_Y3 > 0 EPS Y3 greater than 0

EPS_Y4 > 0 EPS Y4 greater than 0

EPS_Y5 > 0 EPS Y5 greater than 0

Found in the Multiples Group

PE < PE_A5Y PE less than PE-Avg 5y

Found in the Growth Rates Group

Sales_G5F > 5 Sales-Growth 5yr is greater than 5%

(Note: This can be varied to make screen more restrictive)

EPS_G5F > 5 EPS-Growth 5yr is greater than 5%

(Note: This can be varied to make screen more restrictive)

Found in the Ratios Group

TL_TA_Q1 < Industry.TL_TA_Q1 (N) Tot Liab/assets Q1 greater than narrow industry median

Found in User-Defined Group

Price PE Avg Val / Price > 1.5 PE Avg Val / Price = (EPS_12M * PE_A5Y)/Price

Ratio of PE Average Valuation to Current Price greater than 1.5 times

(Note: This number can be varied to make screen more or less restrictive)

Note: To see calculated PE Average Valuation create the following User-Defined Field:

PE Avg Valuation = EPS_12M * PE_A5Y

Screening CriteriaPrice Earnings Average Screen Ranked by Ratio of Valuation Based on Average P/E to Current Price

For use with AAII's Stock Investor program (version 2.1)

Screening CriteriaPrice-Earnings Relative Screen Ranked by the Ratio of Valuation Based on P/E Relative to Current Price

For use with AAII's Stock Investor program (version 2.1)

Found in Company Information Group:

IND_2_DIG <> 07 Sector is not equal to Financial

IND_3_DIG <> 0933 Industry is not equal to Real Estate Operations

Found in User Defined Group:

PE Y1 < 50 User Defined PE Y1 = Price_Y1 / EPS_Y1

(Note: This can be varied to make screen more or less restrictive)

PE Y2 < 50 User Defined PE Y2 = Price_Y2 / EPS_Y2

(Note: This can be varied to make screen more or less restrictive)

PE Y3 < 50 User Defined PE Y3 = Price_Y3 / EPS_Y3

(Note: This can be varied to make screen more or less restrictive)

PE Y4 < 50 User Defined PE Y4 = Price_Y4 / EPS_Y4

(Note: This can be varied to make screen more or less restrictive)

PE Y5 < 50 User Defined PE Y5 = Price_Y5 / EPS_Y5

(Note: This can be varied to make screen more or less restrictive)

Found in the Income Statement - Annual Group:

EPS_Y1 > 0 EPS Y1 greater than 0

EPS_Y2 > 0 EPS Y2 greater than 0

EPS_Y3 > 0 EPS Y3 greater than 0

EPS_Y4 > 0 EPS Y4 greater than 0

EPS_Y5 > 0 EPS Y5 greater than 0

Found in the Multiples Group

PE < PE_A5Y PE less than PE-Avg 5y

Found in the Growth Rates Group

Sales_G5F > 5 Sales-Growth 5yr is greater than 5%

(Note: This can be varied to make screen more restrictive)

EPS_G5F > 5 EPS-Growth 5yr is greater than 5%

(Note: This can be varied to make screen more restrictive)

Found in the Ratios Group

TL_TA_Q1 < Industry.TL_TA_Q1 (N) Tot Liab/assets Q1 greater than narrow industry median

Found in User-Defined Group

Ratio of PE Relative Valuation to Current Price greater than 1.5 times

(Note: This number can be varied to make screen more or less restrictive)

To perform this screen the PE relative, PE relative valuation, and the ratio of PE relative valuation to Price must be calculated

To calculate PE relative S&P 500:

• First, calculate PE Rel—High (UDEF10) as ((Price—High Y1/EPS Y1)/ S&P High PE Y1 + (Price—High Y2/EPS Y2) / S&P High PE Y2 + (Price—High

Y3/EPS Y3) / S&P High PE Y3 + (Price—High Y4/EPS Y4) / S&P High PE Y4 + (Price—High Y5/EPS Y5) / S&P High PE Y5) / 5

Where S&P PE High is 18.3 for ’95, 15.7 for ’94, 21.5 for ’93, 20.1 for ’92, and 26.1 for ‘91

Screening criteria continued on next page

Screening Criteria contin.

UDEF10 =

(PriceH_Y1/EPS_Y1)/18.3 + (PriceH_Y2/EPS_Y2)/15.7 + (PriceH_Y3/EPS_Y3)/21.5 + (PriceH_Y4/EPS_Y4)/20.1 + (PriceH_Y5/EPS_Y5)/26.1

• Second, calculate PE Rel—Low (UDEF11) as (Price—Low Y1/EPS Y1)/ S&P Low PE Y1 + (Price— Low Y2/EPS Y2) / S&P Low PE Y2 + (Price—

Low Y3/EPS Y3) / S&P Low PE Y3 + (Price— Low Y4/EPS Y4) / S&P Low PE Y4 + (Price— Low Y5/EPS Y5) / S&P Low PE Y5) / 5

Where S&P PE Low is 13.5 for ’95, 14.3 for ’94, 19.6 for ’93, 18.0 for ’92, and 19.5 for ‘91

UDEF11 =

(PriceL_Y1/EPS_Y1)/13.5 + (PriceL_Y2/EPS_Y2)/14.3 + (PriceL_Y3/EPS_Y3)/19.6 + (PriceL_Y4/EPS_Y4)/18.0 + (PriceL_Y5/EPS_Y5)/19.5

• Third, calculate PE Relative S&P 500 (UDEF12) as average of PE Rel—High and PE Rel—Low

UDEF12 = (UDEF10 + UDEF11) / 2

• Fourth, calculate ratio of PE Rel Val to Price (UDEF14) as (PE Relative S&P 500 * Current S&P PE *EPS_12M)/Price

UDEF14 = (UDEF12 * 19.0 * EPS_12M)/Price

In the screen you can then use the user-defined PE Rel Val/Price ratio and specify a value greater than 1.5 times

(Note: This number can be varied to make screen more or less restrictive)

UDEF14 > 1.5 PE Avg Val / Price > 1.5

Note: To see calculated PE Relative Valuation create the following User-Defined Field (UDEF13):

UDEF13 = UDEF12 * 19.0 * EPS_12M PE Rel Valuation = PE Relative S&P 500 * Current S&P PE *EPS_12M

Growth stock investing has appeal because of the potentialfor high returns, but it comes at the price of high risk.

A Screening Strategy forInvesting in Stocks WithHigh Growth PotentialBy John Bajkowski

John Bajkowski is AAII’s financial analyst and editor of Computerized Investing.

The mention of growth investingbrings a sparkle to many an investor’seye. Looking for rapidly growing firmsin hot industries seems much sexierthan picking through the value playsthat other investors have cast off. Butsearching for growth stocks is like allinvestment techniques—it has its upsand downs. The allure of buying into astock with the potential for a 10-foldincrease in price must be balancedwith the potential for substantial pricedeclines if the firm fails to meet themarket’s growth expectations.

Screening for Growth

Screening can be used as a first stepin identifying growth stock prospects.Screening is the process of applying aset of criteria to a set of stocks to filterout those companies that merit moredetailed examination. The table onpage 29 of this issue lists a few publica-tions that provide preliminary screensof growth companies. Readers with ac-cess to computers can use a number ofsoftware programs or dial-up services

to screen for growth stocks. AAII’s StockInvestor program was used to performthe screening for this article.

While there are many ways to mea-sure company growth, most investorsfocus on earnings growth, with an em-phasis on a high and expanding rate ofgrowth. A common first screen forgrowth stocks is to specify an absoluteminimum growth level. A minimumannual growth rate of 15% in earningsfrom continuous operations over thelast five years was the first screen ap-plied in our example. While 15% maynot seem very restrictive, it will knockout most of the cyclical firms just com-ing out of the recession. In selecting atime period for historical analysis, theeconomic environment should be keptin mind. True growth companies areexpanding throughout the economiccycle.

Screening based upon earnings re-quires a careful analysis of a firm’sreports, which can highlight how thegrowth was achieved. Was the growthdue to acquisition or internal expan-sion? Did currency translation impact

on earnings? How are same store sales?To help buffer the impact of extraordi-nary items on earnings, earnings fromcontinuing operations were usedthroughout this article. Looking at salesgrowth will also provide a confirmationof how earnings were achieved.

Secondary Growth Screens

The next filter applied examinedyear-to-year earnings per share changesfrom continuing operations. In screen-ing for growth companies it is impor-tant to examine the year-to-year fig-ures for steady and increasing earn-ings. A screen requiring increased earn-ings for each of the last five years wasspecified. If you wish to be more strin-gent in your screening, you might re-quire an increase in the year-to-yeargrowth rate for each of the last fiveyears. This more stringent screen fo-cuses on the momentum of earnings.

Investors examining growth stockslook toward any signs that a trend ingrowth may be broken. Quarterly earn-ings are closely examined and devia-tions from the expected norm arequickly rewarded or punished. The sea-sonality of sales and earnings for mostfirms, however, do not allow investorsto compare one quarter to the preced-ing quarter in a meaningful manner. Todeal with seasonality, it is best to com-pare one quarterly figure to the samequarter one year prior. A decrease fromthe same quarter one year ago is awarning flag that merits investigation.In our screening process, higher quar-terly earnings than the same quarterone year ago for each of the last fourquarters was required.

Beyond examining overall growth ormomentum, many investors examinehow a company compares to its indus-try peers. The ability to expand withinan industry group may point to a firmthat has a competitive edge.Therefore,our final growth screen specified thecompany’s earnings growth to be abovethat of its industry’s average.

One other screen we applied ex-cluded financial firms because of thenon-compatibility of the financial state-ments.

EPS HistoricalEPS Current Sales 52-Week

Historical I/B/E/S Last Fiscal Yr. Trailing Fiscal Growth RelativeAverage Est. 12 Mo. Forecast 12 Mo. Yr. Est. Rate Strength

Company (Exchange) (%) (%) ($) ($) (X) (X) (%) (%) Description

Large Cap (above $1.5 billion)U.S. Healthcare (NM) 180.0 17.5 2.00 2.25 18.8 16.7 31.2 114 Health maintenance organization

Cisco Systems (NM) 140.6 35.0 0.92 1.22 32.9 24.8 120.0 138 Mfrs computer network products

Parametric Technology (NM) 96.8 40.0 0.95 1.12 30.0 25.4 96.2 97 Devlps, mkts integrated software prod

United Healthcare (NY) 67.2 28.0 1.25 1.65 33.2 25.2 58.4 145 HMO administra’n serv on contract basis

EMC Corp. (NY) 62.7 30.0 0.77 0.96 23.2 18.6 56.0 212 Designs, mkts enhancemet prod for computers

Amer Power Conversion (NM) 57.8 35.0 0.60 0.75 37.1 29.7 73.5 154 Devlp, mkts power supply products

Cabletron Systems (NY) 57.7 25.0 4.20 5.30 24.5 19.4 66.2 111 Hardware & software prod supporting LANs

CUC International (NY) 52.1 28.0 0.77 0.98 38.2 30.0 28.8 134 Member-based consumer services

Intel Corp (NM) 49.5 18.0 5.38 5.95 11.3 10.3 29.5 121 Designs, mfrs semicond components, systems

BMC Software (NM) 47.4 25.0 3.10 4.00 19.4 15.0 41.0 112 Devlps standard systems software products

Microsoft Corp (NM) 47.3 25.0 3.72 3.92 24.9 23.6 47.0 102 Designs, mfrs, mkts software packages

Blockbuster Entertain (NY) 39.9 25.0 1.18 1.35 23.0 20.1 53.4 144 Owning and licensing videotape stores

Linear Technology (NM) 38.4 25.0 1.36 1.49 34.9 31.9 23.6 180 Designs, mfrs linear integrated circuits

Sysco Corp (NY) 37.7 16.5 1.16 1.21 22.4 21.5 10.0 98 Distrib consumer foods & food serv

Home Depot (NY) 37.4 30.0 1.01 1.32 41.6 31.8 37.5 93 Retail building material stores

S&P 500 –1.1 8.0 21.98 30.42 20.5 14.8 na na

Medium Cap ($250 million to $1.5 billion)Vencor Inc (NY) 93.8 26.5 1.37 1.60 23.8 20.4 51.4 126 Acute-care servs to complex patients

Westcott Communic’ns (NM) 77.1 35.0 0.55 0.74 30.0 22.3 44.7 99 Prod training, educational programs

CML Group (NY) 75.1 20.0 1.26 1.30 10.7 10.4 28.8 50 Specialty retail

Snyder Oil Corp. (NY) 64.2 15.0 0.80 0.59 24.4 33.1 41.1 103 Oil & gas producing prop/gas processing

Owens & Minor (NY) 62.2 16.0 0.94 1.05 24.7 22.1 10.0 142 Wholesale drug, hospital & surgical supplies

Xilinx Inc (NM) 61.3 30.0 1.71 2.30 32.5 24.1 55.6 164 Mfrs semiconductors, system develop software

Invacare Corp. (NM) 59.8 16.0 1.54 1.75 17.4 15.3 18.4 104 Designs, mfrs durable medical equip

Tech Data Corp. (NM) 59.0 25.0 0.83 1.05 21.5 17.0 44.9 146 Distrib computer-related prods

Briggs & Stratton (NY) 53.1 10.0 6.14 6.22 13.7 13.5 6.8 126 Manufactures air-cooled engines & auto locks

KCS Energy Inc (NY) 49.5 na 1.20 2.41 20.2 10.1 38.8 118 Holding co.: propane distrib, oil & gas serv

Applebee’s International (NM) 47.2 30.0 0.45 0.60 47.8 35.8 27.0 189 Franchises/opers national restaurant chain

Bowne & Co. (AM) 45.3 na 2.20 2.18 10.5 10.6 15.1 112 Finance & corporation printing

Horizon Healthcare (NY) 44.5 25.0 0.94 0.90 24.9 26.0 20.6 174 Opers long-term care facilities

Oakwood Homes (NY) 44.4 18.0 1.40 1.45 15.3 14.7 31.9 110 Mfrs mobile homes, oper mobile home parks

Progress Software (NM) 44.4 28.0 2.02 2.40 22.0 18.5 44.8 109 Devlp, support integrated application software

S&P MidCap 400 na na 7.97 8.75 21.7 19.7 na na

Shadow Stocks (small firms with low institutional interest)Ashworth Inc* (NM) 144.9 35.0 0.36 0.47 30.6 23.4 116.1 133 Mfrs golf apparel

Homecare Mgmt Inc (NM) 95.7 na 0.54 0.54 30.8 30.8 80.3 153 Home care service to elderly

Methode Elec B (NM) 59.7 17.5 0.77 0.85 22.7 20.6 8.9 113 Mfrs electro component devices

Marten Transport (NM) 55.5 15.0 1.58 1.62 10.8 10.5 11.2 123 Long-haul truckload carrier

Gates/F.A. Distrib (NM) 55.4 30.0 1.20 1.27 16.5 15.6 32.9 140 Distribs microcomputers & periph

Rotech Medical Corp. (NM) 45.9 na 0.85 0.98 23.2 20.2 39.4 158 Mkts, distribs home healthcare prod

Medical Technology Sys (NM) 39.7 na 0.68 na 12.3 na 66.1 89 Mfrs, sells nursing home equip

Turf Paradise Inc (NS) 31.6 na 0.67 na 16.8 na –0.7 163 Oper horse racing track in Arizona

Cosmetic Center B (NM) 30.8 20.0 0.93 1.02 18.2 16.5 14.0 141 Retail & distrib’n of cosmetics

Volunteer Capital (NY) 22.6 35.0 0.93 0.45 12.6 26.1 0.8 138 Own, franchisor of fast food restaurants

Diodes Inc (AM) 20.2 na 0.42 na 21.7 na 7.5 491 Semiconductor devices

BGS Systems (NM) 19.8 na 2.23 2.55 9.3 8.1 13.9 53 Design, devlp software prod

Reflectone Inc (NM) 19.8 na 0.86 na 9.9 na 10.4 93 Flight simulators & training devices

Stocks With High Earnings Growth

EPS Growth Rate P/E Ratio

Sources: Stock Investor/Media General data as of April 30, 1994;I/B/E/S data as of May 13, 1994.

Exchanges:*Formerly Charter Golf Inc. NY= New York Stock Exchange; AM= American Stock Exchange; NM= Nasdaq National Market; NS= Nasdaq Small Cap

for companies with above-average pros-pects. To better judge price-earningsratios, many analysts look at price com-pared to forecasted earnings. For theS&P 500, this brings its high trailingprice-earnings ratio of 20.5 back to amore reasonable price-earnings ratiofiscal year estimate of 14.8.

The 52-week relative strength figurespoint to a collection of companies thathave largely outperformed the mar-ket—quite a contrast to the value-based stocks presented in last month’sAnalyst Corner. The relative strengthfigures near 50% for a few of thesecompanies, however, reveal the vola-tility of this group.

Investing in growth stocks can be anextremely rewarding experience. Suc-cess, however, requires careful analy-sis and constant monitoring of the port-folio.

Viewing the Results

To highlight a cross section of compa-nies, the firms passing the screens weredivided up into groups based uponmarket capitalization. The top 15 large-cap and mid-cap companies ranked byhistorical growth in earnings from con-tinuing operations are shown. Alsoshown are 13 Shadow Stock companiesthat passed all the screening require-ments. (Shadow Stocks are defined byAAII as stocks of non-financial compa-nies that are small, with low institu-tional interest, and that have had posi-tive annual earnings for the two previ-ous years. The complete 1994 ShadowStock listing appeared in the FebruaryAAII Journal.)

Our screen focused on historical per-formance, but growth stock investorsreally focus on expected performance.

Definitions of Screens and TermsThe following is a short description of the screens and terms used in the table.

EPS Growth Rate—Historical Average: Annualgrowth in earnings per share from continuing opera-tions over the last five fiscal years. A measure of howsuccessful the firm has been in generating the bottomline, net profit.

EPS Growth Rate—I/B/E/S Est.: The median growthrate in earnings per share from continuing operationsover the next five years that is being forcasted byanalysts as reported by I/B/E/S (345 Hudson St., NewYork, N.Y. 10014). An indication of the consensus inearnings growth expectations for the firm.

EPS Last 12 Mo.: Earnings from continuing opera-tions for the most recent 12 months divided by thenumber of common shares outstanding.

EPS Current Fiscal Yr. Forecast: Earnings fromcontinuing operations for the current fiscal year of thecompany that is being forcasted by analysts as reportedby I/B/E/S.

Price-Earnings Ratio—Trailing 12 Mo.: Marketprice per share divided by most recent 12 months’

earnings per share from continuing operations. A mea-sure of the market’s expectations regarding the firm’searnings growth and risk. Firms with very high price-earnings ratios are being valued by the market on thebasis of high expected growth potential.

Price-Earnings Ratio—Fiscal Yr. Est.: Market priceper share divided by earnings per share from continuingoperations that is being forcasted by analysts as re-ported by I/B/E/S.

Historical Sales Growth Rate: Annual growth intotal sales per share over the last five fiscal years for thefirm. Used to provide a confirmation of the quality ofthe historical earnings per share growth rate.

52-Week Relative Strength: The price performanceof a stock during the last year relative to the perfor-mance of the overall stock market. The market isdefined as the Media General Composite Market ValueIndex of over 7,000 stocks. A figure of 100% indicatesthe stock had the same percentage price performanceas the market. A figure of 105% indicates that the stock

Looking at market consensus forecastfigures helps to provide an indicationof the expectations surrounding thevaluation. It is the company’s ability tomeet and, more importantly, exceedthese expectations that lead to greatgains. For example, Cisco Systems hasthe second highest historical earningsgrowth of the large-cap stocks. Just af-ter the screen was performed for thistable, Cisco came out with a quarterlyearnings announcement that met themarket’s consensus estimate. However,over the last few years Cisco has con-sistently exceeded the consensus esti-mates. The market greeted the earn-ings announcement with a big pricedecline for Cisco and many of its tech-nology competitors.

The price-earnings ratios for stockson our list tended to be above themarket average, as would be expected

Screening Criteria:Found in Growth Rates Group

EPSCon_G5F > 15 Five-year earnings per share growth from continuing operations is greater than 15

(Note: This can be varied up to look for stronger growth)

EPSCon_G5F > INDUSTRY.EPSCon_G5F (N)

Five-year earnings per share growth from continuing operations is greater than that of

the narrow industry group median

Found in Income Statement - Annual Group

EPSCon_Y5 > 0 EPS Cont Y5 greater than 0

EPSCon_12M >= EPSCon_Y1 EPS Cont 12m greater or equal to EPS Cont Y1

EPSCon _Y1 > EPSCon _Y2 EPS Cont Y1 greater than EPS Cont Y2

EPSCon _Y2 > EPSCon _Y3 EPS Cont Y2 greater than EPS Cont Y3

EPSCon _Y3 > EPSCon _Y4 EPS Cont Y3 greater than EPS Cont Y4

EPSCon _Y4 > EPSCon _Y5 EPS Cont Y4 greater than EPS Cont Y5

Found in Income Statement - Quarterly Group:

EPSCon_Q1 >= EPSCon_Q5 EPS Cont Q1 greater than EPS Cont Q5

EPSCon _Q2 > EPSCon _Q6 EPS Cont Q2 greater than EPS Cont Q6

EPSCon _Q3 > EPSCon _Q7 EPS Cont Q3 greater than EPS Cont Q7

EPSCon _Q4 > EPSCon _Q8 EPS Cont Q4 greater than EPS Cont Q8

Found in Price & Share Data Group

RS_52W > 0 Rel strength 52W is greater than 0

Optional Elements Include:Minimum Level of Sales Growth

Found in Growth Rates Group

Sales_G5F > 15 Five-year sales growth greater than 15

(Note: This can be varied up to look for stronger growth)

Profit Margins Equal to or better than Industry

Found in Ratios Group

NPM_12M > INDUSTRY.NPM_12M (N) 12-month net margin is greater than that of narrow industry group

or

GPM_12M > INDUSTRY.GPM_12M (N) 12-month gross margin is greater than that of narrow industry group

Strong Forecasted Earnings

Found in Earnings Estimates Group

EPS_EG5 > 15 EPS Grth Est is greater than 15 (Note: This can be varied up to look for stronger

growth)

Increasing Earnings Estimates or No Decreasing Earnings Estimates

Found in Earnings Estimates Group

EPSUM_EY0 > 0 EPS Est Y0-Rev up is greater than 0

EPSUM_EY1 > 0 EPS Est Y1-Rev up is greater than 0

EPSUM_EG5 > 0 EPS Grth Est-Rev up is greater than 0

and/or

EPSDM_EY0 = 0 EPS Est Y0-Rev down is 0

EPSDM_EY1 = 0 EPS Est Y1-Rev down is 0

EPSDM_EG5 = 0 EPS Grth Est-Rev down is 0

Criteria for the Computerized InvestorFor use with AAII's Stock Investor (version 2.1)

A dividend-yield strategy can help you find potentiallyundervalued stocks with low downside risk, provided thedividend is secure.

Screening for StocksUsing a Dividend-YieldApproachBy John Bajkowski

Screens based on relative levels com-pare the yield against a benchmarkthat may fluctuate, such as the currentdividend yield for the S&P 500. In thiscase the investor does not require thatthe yield meet some minimum level,but instead that it maintain its histori-cal relationship with the benchmark fig-ure. Common screens examining rela-tive yields include comparisons againstsome overall market level, industrylevel, historical average or even someinterest rate benchmark. The screensfor this article were performed using ahistorical average as the benchmark.

Investors looking to perform customscreens can use one of a number ofsoftware programs or information ser-vices that provide fundamental dataon companies. (See the January 1994Analyst’s Corner for information on com-puterized investing tools and a listingof available screening tools.)

For those without a computer, a num-ber of publications provide regular list-ings of high-dividend stocks and rap-idly expanding dividend stocks. (Seethe table on page 33 of this issue formore information on these sources.)

Applying the Screens

AAII’s Stock Investor program was usedto perform the screening for this article.

The first filter excluded utilities, realestate investment trusts, closed-endmutual funds, and financial firms. Eachof these groups has unique financialcharacteristics requiring that they beanalyzed separately.

The next screen required that a com-pany have five years of both price anddividend records. When screeningagainst a historical level, remember toinclude a historical period that coversboth the up and down periods of amarket and economic cycle.

Selecting a time period is a balancebetween using one that is too short andonly captures a segment of the marketcycle and one that is too long and in-cludes a time period that is no longerrepresentative of the current company,industry, or market. Periods of betweenfive and 10 years are most common forthese types of comparisons.John Bajkowski is AAII’s financial analyst and editor of Computerized Investing.

as this are based on the premise thatmarkets tend to overreact to good andbad news and push the price of a secu-rity away from its intrinsic value. Valueinvestors hope to identify thesemispriced securities through the use ofa consistent set of rules called a valua-tion model.

Screening is the first stage in thisprocess and it involves scanning a groupof securities to find those that meritfurther in-depth analysis. Absolute orrelative levels may be used in screen-ing for high-yield stocks. A screen re-quiring an absolute level might look fora minimum dividend yield of 4% beforean investment would be considered.Absolute screens can lead to passivemarket timing—cash levels tend tobuild up when investors cannot findsuitable investments that meet theminimum requirement during times ofmarket extremes. Also, screens thatonly look at absolute levels can beweak because they may turn up com-panies from a single industry that tradi-tionally has higher dividend yields, suchas utilities.

Many investors turn toward the divi-dend yield as a measure of value intheir quest for selecting underpricedsecurities. A stock’s dividend yield iscomputed by taking the indicated divi-dend—the most recent quarterly divi-dend multiplied by four—and dividingit by the share price. If a stock’s pricerises faster than its dividend, the divi-dend yield will fall, indicating that theprice may have been bid up too far andmay be ready for a decline. Conversely,if the dividend yield rises to a highlevel, the stock may be poised for anincrease in price, if the dividend can besustained.

This article will focus on strategiesused to screen for high dividend yield-ing stocks, while the Stock InvestingBasics article that follows covers thedividend valuation process.

The Dividend-Yield Strategy

Like all basic value-oriented tech-niques, the dividend-yield strategy at-tempts to identify investments that areout-of-favor. Contrarian techniques such

5-Year Dividend Cash EPS 52-WeekDividend Average Indicated Growth Dividend Payout Growth Relative

Yield Yield Dividend Rate Paid Ratio Rate StrengthCompany (Exchange) (%) (%) ($) (%) Since (%) (%) (%) Description

Large Cap (above $1.5 billion)Clorox Co.* (NY) 3.6 3.2 1.80 11.9 1968 47.2 8.0 94 Mfrs, mkts non-durable consumer prod

Cooper Indus* (NY) 3.6 2.4 1.32 7.3 1947 47.3 2.3 73 Compression & drill equip; electric, electron prod

Times Mirror* (NY) 3.5 3.3 1.08 4.1 1892 43.9 –35.7 89 Newspaper & book pub; newsprint; info serv; TV

Quaker Oats* (NY) 3.4 2.3 2.12 12.9 1906 45.6 11.3 89 Prod grocery prods, cereals, snacks, pet food

Brown-Forman Corp. B* (NY) 3.3 2.5 2.84 14.1 1960 46.2 2.3 97 Produces distilled spirits & wines

Dresser Indus* (NY) 3.2 2.7 0.68 7.5 1948 31.7 –6.6 98 Supplier of prods & serv to oil & gas devlp cos

Schering-Plough Corp.* (NY) 3.2 2.2 1.80 18.2 1957 41.1 19.3 92 Mfrs drugs & beauty aids

Norfolk Southern* (NY) 3.0 2.8 1.92 9.3 1901 47.2 2.9 97 Holding co.: Norfolk; Western & Southern rail

General Electric* (NY) 2.9 2.7 2.88 11.3 1899 41.6 8.6 106 Power systems; consumer & indus’l prod

CPC Int’l* (NY) 2.9 2.4 1.36 10.3 1920 42.7 8.7 103 Branded grocery products business

Honeywell Inc.* (NY) 2.9 2.3 0.96 12.2 1928 37.9 33.0 98 Mfrs heat & air condition’g control

Abbott Labs* (NY) 2.9 1.5 0.76 18.9 1926 39.1 15.0 100 Pharmaceuticals; hospital, laboratory prod

Union Pacific* (NY) 2.8 2.4 1.60 7.8 1900 44.0 6.0 88 Transportion; energy & natural resources

Gerber Products* (NY) 2.8 2.0 0.86 21.3 1941 46.5 14.7 95 Consumer prod for infant care; trucking

Johnson & Johnson* (NY) 2.8 1.5 1.04 16.7 1905 36.9 14.5 89 Healthcare, pharmaceutical, industrial prod

Mid Cap ($250 million to $1.5 billion)Pacific Telecom (NM) 5.7 4.5 1.32 8.1 1976 43.9 –22.8 94 Telephone communication serv

Goodrich, B. F.* (NY) 5.2 4.8 2.20 6.3 1938 47.0 –20.3 82 Mfrs, sells chemical intermed & polymeric prod

Handleman Co.* (NY) 4.0 2.6 0.44 4.0 1963 44.0 0.8 69 Sell/distrib prerecorded music & books

GATX Corp. (NY) 3.6 3.4 1.50 9.6 1919 46.8 –24.5 115 Mfrs railcars, specific steel & indus’l equip

Ennis Business Forms (NY) 3.6 2.9 0.56 13.1 1973 47.9 6.8 87 Mfrs business forms & paper items

Longs Drug Stores* (NY) 3.4 2.6 1.12 6.6 1961 46.5 –1.6 89 Oper drugstore chain

EG&G Inc.* (NY) 3.4 2.0 0.56 13.0 1965 36.9 7.9 69 Mfrs, mkts electronic instruments

American Business Prod (NY) 3.3 3.1 0.80 9.6 1941 48.1 5.3 86 Business forms & supplies

Crawford & Co. B (NY) 3.3 1.8 0.50 14.1 1965 41.5 7.6 66 Insurance claims adjusters

Block Drugs Co. (NM) 3.2 1.6 1.04 15.8 1971 35.7 9.5 64 Mfr denture, dental care products; drugs

Kimball Int’l B (NM) 3.1 2.6 0.84 12.2 1954 45.5 –2.7 93 Mfrs pianos, organs, office & home furn

Gallagher (Arthur J.) (NY) 3.1 2.3 0.88 8.2 1985 34.7 9.2 77 Insurance broker; risk management

Giant Food A* (AM) 3.1 2.1 0.70 15.7 1960 45.8 –4.3 85 Supermarket chain

Universal Foods (NY) 3.0 2.3 0.92 13.5 1934 40.1 7.7 82 Mfrs, produces foods & beverages

Mercantile Stores* (NY) 2.7 2.6 1.02 6.9 1940 43.4 –11.9 103 Dept stores; beauty salons; shopping centers

Small Cap (below $250 million)Nash Finch Co. (NM) 4.4 3.6 0.72 2.2 1965 49.3 2.6 81 Distrib wholesale & retail supermarket prod

Marsh Supermarkets A (NM) 4.2 2.7 0.44 5.9 1960 45.8 3.5 71 Oper supermarkets & convenience stores

Dibrell Brothers (NM) 3.8 2.2 0.72 20.4 1925 34.5 26.8 62 Process, mkts tobacco

United Stationers (NM) 3.3 3.2 0.40 1.3 1972 34.8 –7.9 62 Distrib wholesale office prod to retailers

MMI Medical (NM) 3.2 2.6 0.16 2.4 1988 24.6 -4.9 75 Medical diagnostic imaging serv

Stepan Co. (AM) 3.2 2.3 0.84 11.4 1967 41.3 –5.5 77 Produces basic & intermediate chemicals

Sealright Inc. (NM) 3.2 1.5 0.46 28.6 1986 45.0 3.5 65 Mfr, mkts round paperboard & plastic containers

Hako Minuteman (NM) 3.1 3.0 0.32 8.8 1988 42.4 –4.1 117 Makes, distrib floor & carpet care equip

Fay’s Inc. (NY) 3.1 2.2 0.20 7.9 1975 40.0 1.4 84 Oper chain of retail drug stores

Flamemaster Corp. (NM) 3.1 2.1 0.12 18.9 1989 42.9 15.6 83 Makes flame retardant & heat resistant prod

Zero Corp (NY) 3.0 2.9 0.40 4.1 1974 49.4 –7.7 89 Mfrs electron packaging & special enclosures

Analysts Int’l Corp. (NM) 2.7 2.6 0.48 9.6 1988 36.2 12.0 84 Furnishes computer programming services

Super Food Services (NY) 2.7 2.3 0.36 4.1 1971 40.9 –12.8 124 Wholesale food distribution & support serv

Insteel Indus (NY) 2.7 2.2 0.24 8.3 1986 31.4 –7.0 77 Mfsr welded wire prod

General Binding (NM) 2.7 1.6 0.40 18.1 1975 42.1 3.3 86 Mfrs business machines & supplies

Table 1.Top Dividend Yielders

Source: Stock Investor/Media General; data as of March 31, 1994.*S&P 500 stock NY= New York Stock Exchange; AM= American Stock Exchange; NM= Nasdaq National MarketExchanges:

Measures exist that help to identifythe safety of the dividend. The payoutratio is perhaps the most common ofthese and is calculated by dividing thedividend per share by earnings pershare. Generally the lower the number,the more secure the dividend. Any ra-tio above 50% is considered a warningflag. However, for some industries, suchas utilities, ratios around 80% are com-mon. The current payout ratio for theDow Jones utility group is 82%, versus55% for the Dow Jones industrial group.A 100% payout ratio indicates that acompany is paying out all of its earn-ings in the form of dividends. A nega-tive payout ratio indicates that a firm ispaying a dividend even though earn-ings are negative. Firms cannot affordto payout more than they earn in thelong term. For the final criterion werequired a payout ratio between 0%and 50%, leaving 86 companies.

To highlight a cross section of stocks,along with any differences due to com-pany size, the stocks were divided upinto three groups based upon marketcapitalization. The top 15 dividend-yielding stocks for each group are listedin Table 1. Market capitalization is de-termined by multiplying the number ofshares outstanding times a firm’s stockprice. The largest firms have a marketcapitalization above $1.5 billion dol-lars, the middle capitalization firmsrange in size between $250 million and$1.5 billion, and the small capitaliza-tion firms are those below $250 millionin market capitalization.

Definitions of Screens The following is a short description of the screens and terms used in the table.

Dividend Yield: Indicated dividend divided by current price. Provides a relative valuationmeasure when compared against historical average dividend yield.

Five-Year Average Yield: Average company dividend yield during the last five years.

Indicated Dividend: Expected per share dividend payment for the next year.

Dividend Growth Rate: Annual dividend growth rate in dividends per share overthe last five years. An indication of the past company strength and dividend payment policy.

Cash Dividends Paid Since: The number of consecutive years that a cash dividendhas been paid. An indication of the stability of the company’s past dividend payment policy.

Payout Ratio: Dividends per share for the last 12 months divided by earnings per sharefor the last 12 months. Provides an indication of the safety of the dividend. Figures between0% and 50% are considered safe. Figures ranging between 50% and 100% are consideredearly warning flags. Negative values and values above 100% are considered red flags for adividend cut if the levels persist. Beyond examining a single year, look for trends.

EPS Growth Rate: Annual growth rate in earnings per share over the past five years.

52-Week Relative Strength: The price performance of a stock during the last yearrelative to the performance of the overall stock market. A figure of 0% indicates the stockhad the same percentage price performance as the market. A figure of 5% indicates that thestock outperformed the market by 5%.

Dividend analysis is geared towardestablished firms that are past theirexplosive growth and capital-intensivestage. To help filter out companies pay-ing just token dividends, a minimumdividend yield of 2% was specified.

Beyond a minimum level of divi-dends, we screened for companies thathave paid a dividend for each of thelast five years and never reduced theirdividend. Dividend levels are set bythe board of directors based on con-sideration of the current company, in-dustry, and economic conditions. Be-cause dividend cuts are tantamount toan announcement that the firm is finan-cially distressed, dividends are set atlevels that the company should be ableto afford throughout the economic cycle.

The next filter required that thecompany’s current dividend yield behigher than its five-year average divi-dend yield. This filter seeks out com-panies whose dividends have increasedfaster than increases in share price, orwhose current share price has dippedrecently.

While it might seem that the screen-ing process should be over with thislast screening filter, before a companycan be considered a buy the security ofthe dividend must be examined. A highdividend yield may be a signal that themarket expects the dividend to be cutshortly and has pushed down the priceaccordingly. A high relative dividendyield is a buy signal only if the divi-dend level is expected to be sustainedand hopefully increased.

The Results

The three groups, while similar in fun-damentals differ slightly. The mid-capstocks have the highest dividend yieldamong the three groups. It was alsosurprising to find a number of S&P 500stocks among these mid-cap securi-ties. This, along with the 52-week rela-tive strength measure points to a col-lection of companies that havestumbled slightly, and the relative per-formance of their stock prices indicatesthis. The 52-week relative strength mea-sures the relative stock price perfor-mance versus the market as a whole.Figures above 100% indicate that a stockhas outperformed the market, whilefigures below 100% indicateunderperformance. The vast majorityof the companies that passed thescreens have underperformed the mar-ket over the last year. The small caps,however, showed the weakest perfor-mance, both fundamentally in terms ofearnings growth and technically in termsof price strength.

Screening for relative high dividendyield is based upon the time-honoredrule of buying low and selling high.Examining a stock’s dividend yield pro-vides a useful framework to identifypotential candidates.

To succeed at this strategy, you needto develop a set of tools to not onlyidentify which stocks have relativelyhigh dividend yields, but also which ofthese stocks have the strength tobounce back.

Found in the Company Information Group:

IND_2_DIG <> ‘12’ Sector <> Utility

Found in the Income Statement - Annual Group

DPS_Y5 > 0 Dividend Y5 > 0

DPS_IND >= DPS_12M Dividend, indicated >=

Dividend 12m

DPS_12M >= DPS_Y1 Dividend 12m >= Dividend Y1

DPS_Y1 >= DPS_Y2 Dividend Y1 >= Dividend Y2

DPS_Y2 >= DPS_Y3 Dividend Y2 >= Dividend Y3

DPS_Y3 >= DPS_Y4 Dividend Y3 >= Dividend Y4

Screening CriteriaFor use with AAII's Stock Investor program (version 2.1)

DPS_Y4 >= DPS_Y5 Dividend Y4 >= Dividend Y5

Found in the Growth Rates Group

DPS_G5F > 0 Dividend-growth 5yr > 0

Found in the Ratios Group

PAYOUT_12M > 0 Payout ratio 12m > 0

PAYOUT_12M <= 50 Payout ratio 12m <= 50

Found in the Multiples Group

YIELD >= 2 Yield >= 2 (Note: This number can be

varied based upon the market situation)

YIELD > YIELD_A5Y Yield > Yield-Avg 5y

Shadow Stocks

The stock market may conjure up images of the oldcattle ranges to many investors, who certainly ride herdon their own stock.

And like the colorful lingo that is exchanged amongranch hands, colorful phrases abound in the stock mar-ket.

One phrase that has been making the rounds in themedia recently is: “Buy ’em at half their growth and sell’em at 1.5 times their growth.”

Actually, this phrase is an old saying that has beenaround for some time. The idea behind it is simple, butlike many sayings, it is well worth remembering.

From Saying to Practice

What does the saying mean from a practical view-point?

The best way to tackle it is to examine a stock’s price-earnings ratio relative to its earnings growth rate. Whena stock is selling at a price such that its price-earnings

The Shadow Stock series was started by AAII in January1985. The list is updated each year and consists of nonfinancialstocks that are small, with low institutional interest, and thathave had positive annual earnings for the two previous years.Non-Shadow Stocks are all stocks other than Shadow Stocks.

Shadow Stocks are presented to illustrate screening tech-niques that individual investors may find useful in theirinvestment programs. They are presented as illustrations andare not recommendations. Further analysis of individual Shad-ow Stocks is necessary for any investment decision. A completeexplanation of the Shadow Stock series, along with a list of allShadow Stocks, appeared most recently in the March 1990issue; new members receive explanations in the new memberpacket.

John Markese is executive vice president and director ofresearch at AAII. The series is researched by John Bajkowski.

ratio is less than half of its earnings growth, it is a buy.When a stock is selling at a price-earnings ratio that is 1.5times earnings growth or greater, it should be sold.Everything in the middle of these two extremes is pre-sumably a hold.

The idea is to purchase a stock with some demon-strated earnings growth before the market recognizesthe potential and bids up the price-earnings ratio. Even-tually, the market will recognize the potential, and it willoverprice that potential, raising the price-earnings ratiorelative to growth above the 1.5 threshold. That will thenbe a signal to the investor to jump from what is assumedto be a runaway value train.

As has been noted often, high growth rates of earn-ings—above 25%—are difficult to maintain for extendedtime periods. That’s because the firm and the market forits product mature, and competitors invade the highlyprofitable product market territory. If earnings growthhas been extraordinarily high for a small firm over 5years, the next five years may not be as productive;price-earnings ratios, however, are often run-up by ex-pectations of even higher growth.

The potential for earnings growth disappointments isgreat.

Refining the Rough Edges

Concentrating on the buy side, simply looking for aprice-earnings ratio relative to earnings growth below0.5 is not enough. A firm that has deteriorating or un-stable earnings can still produce high historical earningsgrowth, but the earnings stream is so risky that the price-earnings ratio remains low in compensation for the risk.

There are a few easy screens that individual investorscan employ to eliminate firms with unstable earningsgrowth. First, look back at five fiscal years of earningsdata, and drop from contention any firm with negativeearnings. Second, for every year over the last five yearsand including the most recent 12 months, insist that the

Stock-Picking Roundup:Buy ’Em Low, Sell ’Em High

By John Markese

earnings per share has increased, even if the increase wasslight. Earnings could be flattening out, a condition youwould notice as you look at the data, but the earningstrend would still have recently been up.

The Shadow Stock Range

The Shadow Stocks listed this month passed these twoearnings trend screens, and had price-earnings ratiosrelative to earnings growth of 0.5 or lower. The price-earnings ratios of these firms averaged just above themarket, with an average of 14, but their annual growthin earnings was far above the overall market, with anaverage of 65%. The earnings growth rates of these firmsare impressive, although some had earnings five yearsago of just a few pennies.

The table presents several earnings reference points:the earnings for the fiscal year five years ago, two yearsago and last year, as well as the earnings for the mostrecent 12 months. Note that the earnings for the mostrecent 12 months will overlap the last fiscal year to agreater or lesser extent depending on the time period thefirm’s fiscal year covers. For example, Uno RestaurantCorp., the originator of Chicago deep dish pizza, nowdishing it out throughout the country, went from $0.06a share five years ago to $0.50 in the last full fiscal year,to most recent 12 months of earnings of $0.55. BecauseUno’s fiscal year-end is October 1, and the data for themost recent 12 months is as of June 30 (the end of thefirm’s most recent reported quarter), the two earningsfigures overlap by three months—July through Septem-ber.

The Shadow Stock data provides an example of why itis important to examine earnings in the individual years;just looking at five-year annual earnings can be deceiv-ing. Uno’s phenomenal 70% compound annual growthin earnings over five years would be difficult to main-tain. The firm had earnings per share of $0.06 in fiscalyear five, $0.14 in fiscal year four (not shown), and $0.30in fiscal year three (not shown), resulting in a doublingof earnings from fiscal year five to fiscal year four, andthen another doubling in fiscal year three. Rapid growthin new restaurants can double earnings and double themagain. But the momentum of new restaurant openings islikely to slow, slowing earnings growth. Nonetheless,looking for firms with a low price-earnings ratio relativeto growth is a worthwhile initial screen.

Finally, the price as a percentage of the 52-week stockprice high is presented as a glimpse of recent pricemovement, which means market recognition. These firmsaverage 72% of their recent highs, but given the marketdirection of late, this statistic probably understates theirprice movement. Many have had substantial stock pricemoves up, and although the average price-earnings ratiois a reasonable 14, a few such as Uno’s are above 20 timesearnings, indicating that the market has fairly high ex-pectations of future growth.

With a few additional simple screens, and a cautiousapproach that includes a close look at how these firmsachieved their growth from a business and productviewpoint, “buy ’em low at half their earnings growth”is a solid bit of financial wisdom.

While it won’t guarantee the pick of the herd, it will atleast steer you in the right direction.

Definitions of Screens and Terms

The following is a short description of screens and terms used for this month’s Shadow Stock listing.

P/E to Annual Five-Year Growth Rate: Price-earn-ings ratio (price per share divided by earnings pershare) divided by the annual growth rate in earningsover the last five fiscal years. An indication of how themarket values the firm relative to its historical earn-ings growth rate. A low relative value—below 0.5—may indicate that the market has not yet recognized afirm’s potential.

P/E Ratio: Market price per share divided by earn-ings per share. Provides a measure of the market’sexpectations regarding the firm’s earnings growthand risk.

Annual Five-Year EPS Growth Rate: The compoundannual growth rate in earnings over the last five fiscalyears.

Earnings Per Share: Net income after all expensesand taxes, divided by the number of common sharesoutstanding. An indication of earnings stability andtrend when examining individual years over the lastfive-year period.

EPS Last 12 Months: Earnings per share for a fullyear through the firm’s most recent reported quarter.

EPS One Year Ago, Two Years Ago and FiveYears Ago: Earnings per share for the years indicatedbased on the firm’s fiscal year.

Price as Percent of 52-Week High: Current marketprice divided by the highest price over the past 52weeks. A price momentum indicator that shows therecent movement of share price.

P/E to AnnualAnnual 5-year Price as5-year EPS Last One Two Five Percent of

Growth P/E Growth 12 Year Years Years 52-weekRate Ratio Rate Months Ago Ago Ago High

Stock (Exchange) Ticker (x) (x) (%) ($) ($) ($) ($) (%) Description of Firm

Intelligent Electronics (O) INEL 0.1 8.1 95 1.82 1.02 0.59 0.07 62 Franchise; computer stores

Pentech International (O) PNTK 0.1 15.0 124 0.30 0.25 0.17 0.01 74 Mkts writing/drawing prod

Swift Energy (A) SFY 0.1 7.4 61 1.48 1.35 1.16 0.20 85 Oil & gas explor'n, devlp, produc'n

Odetics Inc. 'B' (A) O.B 0.1 18.0 130 0.32 0.28 0.25 0.01 77 Mfrs electro equip to record

Bridgford Foods (O) BRID 0.2 20.1 102 0.91 0.83 0.63 0.05 76 Frozen & snack food products

Code Alarm (O) CODL 0.2 11.3 57 0.95 0.86 0.72 0.14 45 Mfrs automobile security systems

Lillian Vernon (A) LVC 0.2 9.0 43 1.18 1.17 0.88 0.28 61 Direct mail catalog sales

T2 Medical Inc. (O) TSQM 0.2 25.0 111 0.99 0.80 0.62 0.04 80 Operates infusion therapy ctrs

Group 1 Software (O) GSOF 0.2 9.9 42 0.71 0.70 0.66 0.17 74 Mfrs IBM-compatible software

Lynch Corp. (A) LGL 0.2 12.4 51 1.53 1.31 0.98 0.25 61 Glass machin'y; test equip; quartz

Arkansas Freightways (O) AFWY 0.3 15.7 54 1.16 1.00 0.83 0.18 78 Motor carrier, general freight

Uno Restaurant Corp. (A) UNO 0.3 22.7 70 0.55 0.50 0.35 0.06 77 Operates pizza restaurants

Mestek Inc. (N) MCC 0.5 7.3 16 0.94 0.89 0.79 0.50 73 Climate control equip; engin'g

Ben & Jerry's Homemade 'A' (O) BJICA 0.5 15.9 31 0.91 0.79 0.63 0.27 69 Mfrs ice cream/franchise stores

Bowl America 'A' (A) BWL.A 0.5 13.4 26 1.46 1.42 1.19 0.57 81 Bowling centers

Hipotronics Inc. (A) HIP 0.5 12.6 23 1.51 1.50 1.16 0.65 77 Hi volt test/power supply equip

Shadow Stocks: Low P/E to Growth Rates

*Based on firm’s fiscal yearExchanges: N = New York Stock Exchange

A = American Stock ExchangeO = Over the counter

Statistics are based on figures as of August 31, 1990. Data provided by MPSoftware’s MarketBase, Needham Hts., Mass. Descriptions provided byStandard & Poor’s Stock Guide.

Earnings per Share*

Screening CriteriaFor use with AAII's Stock Investor program (version 2.1)

Found in the Mutliples Group:PE_TO_G5F <= 0.5 PE to EPS growth 5yr less

than or equal to 0.5

Found in the Income Statement—Annual Group:EPSCon_Y5 > 0 EPS Cont Y5 greater than 0

EPSCon_12M >= EPSCon_Y1 EPS Cont 12M greater orequal to EPS Cont Y1

EPSCon_Y1 > EPSCon_Y2 EPS Cont Y1 greaterthan EPS Cont 2

EPSCon_Y2 > EPSCon_Y3 EPS Cont Y2 greaterthan EPS Cont3

EPSCon_Y3 > EPSCon_Y4 EPS Cont Y3 greaterthan EPS Cont4

EPSCon_Y4 > EPSCon_Y5 EPS Cont Y4 greaterthan EPS Cont5