Fool.com: Building a Better Mousetrap? (Workshop) September 11, 1998

More on the Margins Screen

Atlanta, GA. (Sept. 11, 1998) -- Today, I'm continuing to pursue the margins theorem I proposed on Tuesday. The goals are to see if and when the theorem would have caught the mechanical models' big winners, and whether it would have helped us avoid the models' losers. (We're also testing the general usefulness of applying this one purely fundamental screen).

The theorem worked quite well for Dell <% if gsSubBrand = "aolsnapshot" then Response.Write("(Nasdaq: DELL)") else Response.Write("(Nasdaq: DELL)") end if %>. Today we'll test it against Best Buy <% if gsSubBrand = "aolsnapshot" then Response.Write("(NYSE: BBY)") else Response.Write("(NYSE: BBY)") end if %>, which was up 176% year-to-date as of my August 26 study date, providing a huge spark to the relative strength portfolios, as we saw last Friday.

Best Buy's annual net margins tell part of the story:

 
 FY91     0.70% 
 FY92     1.00% 
 FY93     1.20% 
 FY94     1.40% 
 FY95     1.10% 
 FY96     0.70% 
 FY97     0.02% 
 FY98     1.10% 
 1Q99     0.80% (vs. -0.20%)

The rising margins (along with booming sales) pushed the stock from less than $2 in April 1991 to nearly $19 a share in April 1994. (The fiscal year ends in February, with year-end results announced in April.) The shares then spent several years in the wilderness, hitting a low around $4 in February 1997 as profits disappeared and short-sellers were betting the company might face bankruptcy due to lame PC sales, generous financing plans, poor inventory management, and a massive debt caused by fast expansion. Since crashing, though, the stock has soared to the August high of $54.81.

The good news is that by employing the theorem, you would have definitively sold the stock in September 1994 at around $20 (just 12% shy of the high set the next month) and stayed away from it during its painful descent over the next three years. During the early part of this decade, the company's earnings per share occasionally didn't keep up with net margins growth, and net margins sometimes didn't keep up with sales. In other words, my theorem would have had you trading in and out of the stock during that trip from $2 to $19, sometimes missing good chunks of pleasant scenery.

For example, as sales growth outpaced earnings during the winter of '92-'93, you would have sold at around $6, only to be stuck buying back in during September 1993 at $10. Of course, you also would have sold in April 1994 around $18 and bought back in during June 1994 at $14, saving some of the bucks you had missed out on. This second trade is interesting because it shows that keying off the EPS figure rather than net margins can work. That's because 4Q94 results reported in April 1994 showed sales up 94%, net income up 99%, but EPS up just 61%. Strange as it may seem, that was a sell signal according to my theorem.

However, the more substantial problem is highlighted by Best Buy's current turnaround run. While margins and earnings per share improved massively beginning in 2Q98, sales growth remained pathetic until 4Q98 when EPS rose 550% on 22% higher sales. You could have bought the stock then at around $35 and enjoyed a healthy gain since then, but you would have come to the party quite late -- indeed, long after the relative strength models said, "Go for it."

Thus, Best Buy suggests that my proposed use of the Fool's current rising margins screen won't always work well to find you the best buy at the earliest date. (Recall that with Dell, it had worked beautifully). The most important shortcoming is that the 15% minimum sales growth threshold can cause you to stay away from a stock despite obvious improvements in the fundamental story. Indeed, Best Buy's awesome run over the last 18 months has really depended on a conscious decision to slow the sales growth in order to focus on enhancing operations and margins.

In my view, the massive 75% gain in net margins and EPS recorded in the second quarter of FY98 (reported last September) was enough in the context of the developing story to see Best Buy as a promising investment, despite the tiny 1% revenue gain. Indeed, the possibilities were apparent even earlier (see my July 1997 Daily Double on the company). Buying after the 2Q98 report, you would have paid about $12 a share and, from what I can tell, beat the mechanical models to the punch.

Despite my mental lapse while examining the FY98 results in April, I believe, as I said in this Wednesday's Evening News column, that the numbers, the story, and the economy all pointed to continued good times for Best Buy -- and still do. Preliminary 2Q99 results suggest you still want to be holding.

As I mentioned last week in my column on American Woodmark <% if gsSubBrand = "aolsnapshot" then Response.Write("(Nasdaq: AMWD)") else Response.Write("(Nasdaq: AMWD)") end if %> -- my ideal investment -- I think the 15% sales growth threshold for our margins screen may be too high. Best Buy confirms that suspicion. At the same time, though, Best Buy suggests (as Dell did) that keying off of EPS rather than net margins makes good sense.

Next Tuesday, we'll look at the Gap <% if gsSubBrand = "aolsnapshot" then Response.Write("(NYSE: GPS)") else Response.Write("(NYSE: GPS)") end if %>, since it was the third best performer from any of the portfolios, with a 71% year-to-date gain as of August 26.

Check out the latest file updates for the Workshop:
New Rankings | 1998 Returns | New Database

What Happened to Robert Sheard?