by Martin Zweig
Finally, there’s a notation for the month, in this case May, which indicates the latest quarterly report. I keep that record so that, as I go through the cards searching for the stocks, I know whether I have an up-to-date report or a stale one. If we’re already in, say, June, and I find that my latest quarterly report on a company is December, I figure that I probably missed the first-quarter report and will consult financial records to update that. In other words, putting the date of the last report is a check-off method for you. So, based on the day’s earnings reports, Ennis maintains its place in my index file although it’s not an immediate candidate for purchase.
The last stock circled for this particular day is Federated Group, traded over-the-counter. Sales were up sharply from about $40 million to over $74 million, an 83% increase. However, earnings per share, while up 32%, did not keep pace with sales. In some industries the failure of earnings to grow as rapidly as sales could mean heavy competition and price cutting and a subsequent thinning of profit margins. In other cases, it might be a reflection of expenses created by introducing new products, whereby the jump in sales may lay the foundation for much greater future increases in earnings. We simply don’t have enough information to weed the stock out, so it remains a candidate for more investigation on this day.
Let’s recap what we’ve done so far. We have searched through the day’s earnings reports looking for reasonable growth in both revenues and earnings per share. Most stocks failed the test. Four of them were worth at least a second glance and have been circled. Continental Healthcare was quickly eliminated when, through a bit of subtraction, it was seen that the previous quarter’s growth was negative, not consistent with our goal of solid long-term growth. The other three stocks remain with circles at least until the next step.
Once you get the hang of it, it will take only a few minutes a day to filter out the candidates from the also-rans, with the exception of the period roughly from the second through the fourth week of a calendar quarter. That’s when an unusually large number of companies report their earnings, so the task will take you longer. But with experience, even during the heavy reporting seasons, it should not take more than fifteen minutes on a given day to screen the reports.
Please realize there is no substitute for experience. Although it may be slow going at first, your speed will gradually increase and, in a relatively short time, you’ll become an expert at skimming the reports. By the way, always keep a calculator handy so that you can quickly compute the percentage changes in sales or earnings or do the subtraction required to isolate earnings from the previous quarter, six months, or nine months.
We have three stocks left with circles on this day, but so far we don’t even know their prices. The next step, then, is to check the stock tables to get the previous day’s closing prices. Cullinet and Ennis would be found in the listings for the New York Stock Exchange, and Federated Group in the NASDQ over-the-counter listings. For stocks on the New York and AMEX, the stock tables also give the P/E ratios (though they may not include the newest quarterly figures), so you might want to note those down as well. I scribble the prices and the P/E ratios in the margins of the stock tables near these stocks.
In addition, I always look for the change in price on that day. Most of these earnings reports appear either before the market opens or during trading that day. Not too many of the reports come out after 4:00 p.m., when the market is closed. Thus, the stock’s action of that day is often a reflection of how the market absorbed the news of the particular earnings report. The worst thing the stock can do on news of earnings is to drop sharply. That usually indicates the market was disappointed with the earnings report—that it came in less than anticipated. Academic studies have shown conclusively that when earnings are significantly below expectations, such stocks, on average, will underperform the market over the next one to two quarters.
Therefore, if I see a stock down rather sharply percentagewise—say, a point or more on a stock selling above $30, or three-quarters of a point or more on stocks in the $15nmdash;$30 range—I’ll take a rather dim view of that action and probably eliminate the stock right then. Remember that the tape is your best friend. Rotten tape action on the heels of an earnings report—no matter how good that report might seem to you— is often the kiss of death. It’s better to be safe than sorry.
None of the four circled stocks this day had significant changes in price, the largest change being only three-eighths of a point. I do not bother writing down the daily change unless it’s significant. Indeed, you need not even bother with the most negative ones unless you are searching for short sales. I would simply cross out the stock at once. On the other hand, if the stock is up sharply on the heels of the earnings report, say a point and a half or two, it is a very hopeful sign and often means the market was pleasantly surprised by the report. Those same academic studies have found that stocks that report surprisingly good earnings tend to do better than the market as a whole over the next three to six months. So, if you find a big jump in price on the day of an earnings announcement, make a positive notation for yourself.
We can eliminate some stocks based on poor tape action the day a report comes out. The next means of elimination is an unreasonably high price/earnings ratio. The P/E ratio is simply the price of the stock (which you have found in your newspaper) divided by the total of the last four quarters of earnings. You won’t have that figure available in the newspaper unless, by chance, the quarterly report was the last of its fiscal year, as was the case with Cullinet. Cullinet showed its earnings for the fiscal year of 81¢. By dividing its price of 24 7/8; by 81¢ of earnings, we see that the P/E ratio is a lofty 31. That’s enough for me to run for the hills. Cross out Cullinet. We don’t know the last four quarters’ earnings for Ennis or Federated, so we’ll have to dig further. At this point, we’ve taken all there is worth getting out of the daily newspaper.
Now we’ll need a new source of data, preferably a good chart service that also maintains decent figures on earnings. Possible services include William O’Neil & Company’s Long-Term Values (12655 Beatrice Street, Los Angeles, CA 90066-0919; 310-448-6800), Standard & Poor’s (25 Broadway, New York, NY 10004; 212-208-8000), Mansfield (27 Wilburn Street, New York, NY 10005; 800-223-3530), or services heavier on financial data and less so on graphics, including Value Line, Inc. (220 East 42nd Street, New York, NY 10017; 800-634-3583).
Once you are in possession of a good chart or financial service you’ll be able to check the earnings for the last twelve months. Take Ennis Business Forms, for example, as illustrated in the following chart. The earnings report in The New York Times showed the quarter through May 31, its first quarter of the fiscal year, at 63¢. Checking the Ennis chart, we can see what the last three quarters were prior to that report. They total 67¢, 66¢, and 89¢. Added to the more recent report, the four quarters total $2.85. A second way to derive that figure is to take the difference between the current quarter and the year-ago quarter, which in this case is 63¢ minus 51¢, or 12¢. In other words, Ennis’s earnings increased by 12¢ above the year-ago quarter. All you have to do is add that 12¢ to the previous four-quarter total of earnings, which was $2.73. That would give you the $2.85 figure.
If you use the Value Line, Standard & Poor’s, the Mansfield, or the O’Neil charts, you can add the quarterly figures to get the last twelve months of earnings. In deriving your calculation you can briefly glance to see how the previous several quarters compared with their respective year-ago numbers. If you find that several earlier quarters were lower or the growth was skimpy, you’ll probably want to ignore the stock. Remember, we’re looking for reasonably steady growth.
PRICE/EARNINGS RATIOS
Earlier, in the case of Cullinet, I commented that we are trying to avoid very high price/earnings ratios. It’s important now to describe in detail what we’re looking for with respect to price/earnings ratios. It would be helpful to examine several decades’ worth of academic studies involving these key rel
ationships. In summary, the data going all the way back to the 1930s show conclusively that stocks with low price/earnings ratios outperform stocks with high price/earnings ratios over the long term. Specifically, then, how can we use the ammunition in our stock-picking method?
Here’s how I go about it. I try to avoid the highest P/E ratios. Of course, the absolute level will vary considerably over the years as the market’s P/E as a whole drifts higher or lower. The Dow Jones Industrial Average itself reached a ghastly 23 times earnings in 1961, an all-time record, barely beating out the prior mark set in 1929 before the crash. But P/E ratios again soared in 1968 and in 1972. It was not unusual in those years to find many stocks selling at 40 or more times earnings, including quite a number that got as high as 80 to 100 times earnings, truly an absurdity.
At the other end of the spectrum, the Dow Jones P/E ratio was not much over 6 at the 1974 and 1982 bear market bottoms. It wasn’t easy at those points to find many stocks with P/E ratios of 20 or more. But whatever the high end of the range, I try to avoid those P/E’s. As for, say, Cullinet, at 31 times earnings, the stock may go higher, but that’s of no interest to me. What is of interest is an ultrahigh price/earnings ratio, nearly triple that of the market as a whole as of mid-1985. That tells me the situation is much too risky.
If Cullinet’s future earnings work out as expected by Wall Street in general, or perhaps better than expected, the stock might do okay. But a disappointment could be devastating. Once P/E’s get to ultrahigh levels, there’s no room for error. If the growth is expected to be, say, 30% and the earnings come in up 20–25%, the stock might be hit hard. If the growth comes in at, say 10%, the stock could easily be mauled and, if the worst happens, with earnings lumping, a stock with a high P/E could easily fall 70–80%. Thus, stocks with very high P/E’s can on occasion do well, but from a risk-reward standpoint they’re poison.
Shifting to the opposite end of the field, the academic studies, as previously noted, have shown that very low P/E’s perform the best. However, there are a few technical biases with the studies that can call into question some of the results, particularly the fact that companies that went bankrupt were often excluded from the studies, therefore giving the results somewhat of an upward bias.
There are generally two types of stocks that you’ll find with extremely low P/E’s, say in the 3, 4, or 5 range, as of 1988. The first is the company in extreme financial difficulty. It still has earnings but investors are not willing to put much of a multiple on them because of the danger that the company could go bust. You are not getting a bargain buying a stock at 3 times earnings if its balance sheet is terrible and if the company itself is going to go under in the next year or two. Of course, there is no certainty what will happen. I suggest, though, that if you are going to buy a stock with an extremely low P/E ratio, you first carefully investigate the balance sheet and other financial facts about the firm. I generally ignore this type of stock because it doesn’t fit my shotgun-type approach.
The other type of stock with a very low P/E ratio is a firm in an industry suffering from general neglect, probably because of some bad news overhanging the group. For example, in recent years, public-utility stocks that have nuclear plant construction under way have sold at very low P/E’s because of the risk that construction might be halted or that the companies might run out of funds to finance these ventures. Sometimes, a company’s low P/E is the result of record earnings that Wall Street does not expect to continue perhaps because of the cyclical nature of the businesses or foreign competition. These are the types of low-P/E stocks that academic studies indicate perform the best. A group of such stocks, held for a year or two, probably would outperform the market, especially if one’s portfolio were reevaluated once or so a year to weed out the stocks that have gone up to higher P/E’s and replace them with reasonably solid stocks that have drifted down to low P/E’s.
Since my approach emphasizes stable and reasonable growth, I am very unlikely to find an extremely low P/E associated with that. For if a company does have stable and solid growth, that is likely to be recognized at least to some degree, and the P/E ratio is more prone to be in the normal or higher end. Indeed, if I come across a stock with excellent growth that has a very low P/E, I immediately get suspicious. I will check further to see if there is a problem on the balance sheet or if the backlog of orders has dropped off, or if there is some other outstanding negative: there often is. Thus, I usually don’t go for the very low P/E’s because they generally don’t mesh with my approach. However, with caution, there’s nothing wrong with using the low-P/E method of investing exclusively.
I also avoid the very-high-P/E stocks. Most of the stocks I select have P/E’s near the average for the market or somewhat above it. If the market’s average P/E is, say, 10, and I find stocks with better growth and more stable growth than the market as a whole, I would expect that the fair value of these stocks would warrant a higher-than-market P/E, say an arbitrary 14 or 15. If I can find such a stock with a P/E of 11 or 12, other things equal, it would be a bargain. If the P/E were up to, say, 16 or 17, it would be less of a bargain, although if the growth rate were high enough and if the company had a significant competitive advantage, giving the higher probability of stable growth, it still might be worth the price. This is a judgment I base on my experience and instinct.
To make some sense out of what might be a reasonable P/E, you’ll have to go back and study the earnings trend for the past several years. Look at the little box in the Lankford chart, for Ennis Business Forms, on page 206. You’ll see that for its 1980 fiscal year, earnings were $1.56. They then trended up in the subsequent years to $1.72, $1.80, $2.21, and $2.73. Using my trusty calculator, I find that that works out to a four-year compounded growth rate of 15% per annum. By the way, it is very useful if you have a calculator that can easily do compounded rates of return. I find the Hewlett-Packard 12C calculator to be excellent in that respect. A great financial calculator, it’s small, easy to carry around, and runs on batteries. Texas Instruments also makes a good one, and there are several other satisfactory ones on the market.
Back to Ennis. A 15% per year growth is not a bad rate. In addition, Ennis’s earnings were up every single year, and the latest quarter showed a 23% gain, which is even higher than the longer-term trend. I liked that. At a price of 37½Ennis’s P/E of 13 was somewhat higher than the 10 to 11 P/E for the market as a whole. But that seemed very reasonable given Ennis’s steady returns over the years. That’s why the stock remained on my potential buy list. Two reasons why I was cool on it at the time included the gain of only 8% in sales for the quarter, which I previously noted, and the fact that four insiders had sold stock in the preceding few months.
In the Lankford charts or the other services you might be using, you can see a brief description of the company’s business, which in Ennis’s case involves the production of business forms. Use Value Line or Standard & Poor’s if you want a somewhat more detailed description of the company’s business. Ennis’s industry is interesting because companies consistently need its products even during business downturns. Sales might ease off somewhat in a recession, but it’s not a highly cyclical business nor is it burdened by heavy fixed costs. Moreover, its growth rate is not spectacularly high, which can actually be a blessing in some cases. Ultrahigh rates of growth attract competition, and competition eventually destroys profit margins and the growth rates. If Ennis and the rest of that industry can “plod” along at 15% a year growth, that’s quite good, especially if you can buy the stock at roughly a market P/E. If Ennis’s P/E were to drift back a couple of points toward the average for all stocks, it could get very attractive.
Let’s move on to the last stock in the Times listings of company earnings we’ve been considering, Federated Group. A perusal through the Lankford Quote OTC charts shows that Federated earned $1.16 for the twelve months ending February 1985. Earnings were up 6¢ for the first quarter of the 1986 fiscal year (found by subtracting 19¢
of the year-ago quarter from the current 25¢). Simply add that 6¢ incremental increase to the trailing $1.16 figure, and you’ll get $1.22 of earnings for the last twelve months. The day that report came out Federated was 22½ giving it a P/E of 18. That’s not exactly cheap for a stock that a few years ago lost 7¢ a share, down from a tiny profit in 1980. While earnings had grown since then, the erratic nature of the net in 1980–81 bothered me.
A second dubious factor here was that the quarterly earnings were up 32%, which was fine, but they had jumped more than six-fold in the previous quarter, so I couldn’t be certain whether growth was slowing or not. Federated might have proven to be a good stock, but I preferred to pass on it for the moment. So, my work on the earnings reports for June 18, 1985, produced only one stock for my potential buy list, Ennis Business Forms, and that had been on the list for quite some time.
I just want to add another thought or two about earnings trends. I try to look for some stability in the direction. If the earnings are increasing at a steady 15% each and every year, that’s great. If earnings are up 30% one year, up 5% the next, up 40% the next, down 2% the next, and up 20% the next, the long-run growth rate might be fine but the lack of stability is a negative. Second, reported earnings are not always what they seem. A variety of accounting methods can “smooth” earnings in a direction the company’s management desires. If you use my approach, you don’t need to dig further to see whether the reported earnings are overstated or understated. In some cases we will buy bummers, but in the long run you will find a number of excellent stocks based on my criteria. If you have the time and inclination to dig further, go right ahead and do so, but it may wind up being a full-time job.
