The Challenge of Cyclicals

Louis Corrigan (TMF [email protected])

Atlanta, GA. (Oct. 13, 1998) -- In the last few weeks, I noted that my proposed margins theorem pretty much falls on its face when grappling with stocks of cyclical businesses. In fact, as we saw with oil services outfit Input/Output <% if gsSubBrand = "aolsnapshot" then Response.Write("(NYSE: IO)") else Response.Write("(NYSE: IO)") end if %>, the market may begin discounting a future downturn nearly as soon as a cyclical business starts delivering rising margins. As Rinus pointed out in a post to our AOL message board, cyclicals present a challenge not just to the margins screen but to models which rely on Value Line timeliness rankings or IBD's relative strength screens.

That's because these models tend to pick up on earnings and price momentum, so a cyclical company like Input/Output may hit the model at or near its peak. That's a problem because cyclicals live in the land of peaks and valleys. As Peter Lynch has said, charts of cyclicals look like the "polygraphs of liars or the maps of the Alps." Of course, what looks like a peak can always just end up as a day camp halfway up Everest. But from what I can tell, the Workshop models don't provide any obvious means for distinguishing a cyclical with further momentum from a cyclical about to nosedive.

What to do? For those inclined to tinker with the models, it seems like one option might be to note which stocks are cyclicals and to develop some separate discipline for dealing with them. While even fundamental analysts who follow a cyclical industry can make mistakes, detailed knowledge of an industry and its patterns can make a huge difference in knowing when you're in the valley and when you've reached a peak. I'd prefer to tackle the issue by studying the industry.

But those using a more mechanical approach could at least create a system for protecting themselves from serious downturns. For example, folks pursuing a one-year holding period might want to deploy a month-by-month or quarterly holding discipline for cyclicals. During Input/Output's plunge over the last nine months, the stock crashed through timeliness and high relative strength ratings. I don't know what metric would be most useful as a stop-loss threshold (RS falling below 85? Timeliness rating falling?), but I would encourage those interested in the idea to check recent cyclical losers from the models and do a little backtesting. It's simply preferable to take a 25% loss on a cyclical than watch it plummet even more.

Of course, to do this, you need to know what constitutes a cyclical stock. This becomes a bit tricky, because cyclicals come in different stripes.

The first order cyclicals are usually called deep cyclicals. They include firms heavily dependent on the economy. They tend to fall on the first signs of an economic slowdown and rise very quickly and very early as economic demand begins to pick up. These include commodity businesses like paper, steel, chemicals, and construction equipment.

Certain high-tech stocks also function a bit like deep cyclicals, though they also have something of their own internal logic, especially since they often key off of technological innovation and may have generally more robust long-term growth prospects than classic deep cyclicals. In the last year, for example, companies that make semiconductor capital equipment, chips, disk drives, and other PC components have pretty much been slaughtered. While the international economic troubles played a significant role in this downturn, these businesses tend to be relatively capital intensive yet highly profitable at their peaks. So they're continually attracting competitors who eventually create excess capacity, high inventories, and brutal price competition. That then drives the weaker players out of the game until the next upturn starts the cycle over again. The leading firms usually rise from the ashes by being first to market with the next technology.

Durable goods manufacturers (autos, refrigerators, etc.), home builders, and airlines are also quite economically sensitive. So, too, are many banks and brokerages. Economic turbulence often creates turmoil in financial markets, leading to trading losses as well as a drop-off in investment banking revenues. Loan defaults and weak demand for capital also hurt lenders.

In still other industries, though, product cycles may trump even industry-specific cycles, at least for a time. You see this often in software, where a hot new title overcomes other obstacles. All of these industries are typically contrasted to noncyclical businesses such as food, pharmaceuticals, and insurance, which provide stuff people tend to buy in good times and bad.

All of this is important for my margins theorem, too, since consumer-oriented companies include many that are sensitive to the economy as well as to specific product cycles. For example, while I love the Gap's <% if gsSubBrand = "aolsnapshot" then Response.Write("(NYSE: GPS)") else Response.Write("(NYSE: GPS)") end if %> swinging khaki ads, even "no-fashion" fashion tends to have its ups and downs. And as we've seen over the last few weeks, many retailers -- including model faves the Gap and Best Buy <% if gsSubBrand = "aolsnapshot" then Response.Write("(NYSE: BBY)") else Response.Write("(NYSE: BBY)") end if %> -- have been hit despite delivering decent to strong gains in same-store sales. That's because investors have started factoring in a potential recession, which stands to impact virtually every retailer.

While we've seen that the rising margins screen can help you spot retailing successes relatively early on, it will almost surely be late in spotting an economic slowdown. Investors are likely to see fears confirmed in weaker monthly sales figures (and thus a falling stock price) months before that weakness shows up in earnings. In that sense, relative strength rankings can themselves serve as a downside check on the margins screen. The point, though, is that the more cyclical a business is, the more closely its stock must be monitored.

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