Go to the Globe and Mail homepage

Jump to main navigationJump to main content

When assessing EPS estimates and price targets, don’t forget to check the company’s “moat” (BILL TIERNAN)
When assessing EPS estimates and price targets, don’t forget to check the company’s “moat” (BILL TIERNAN)

The Buy Side

Learning not to trust forecasts - and what to do about it Add to ...

Money managers often hotly debate what's better: growth, value, or perhaps growth-at-a-price.

To add some kindling to the fire, let me say that this distinction is misleading: The market is really divided into "results-oriented investors" and "input-oriented investors." Momentum players fall into the first category, while Warren Buffett and his ilk fall into the second. Most money managers fall somewhere in between.

Let me explain. Results-oriented investors buy visible (or forecast) earnings, dividends, cash flow (or EBITDA). They must therefore assume one or more of the following:

"I am brainier and can forecast results better."

  • "I have access to brainier analysts who do."
  • "I'm only as smart, but can act faster."
  • "I'm neither, but have bigger cojones and so buy what others consider risky."

Input-oriented investors, on the other hand, are humble about their (or anyone's) ability to forecast results. They spend their time figuring and sleuthing which business inputs are worth buying, then buy these cheaply - and wait for results.

But is forecasting earnings really that hard? Yes. Not only are the above four assumptions often wrong, but the economic cycle can make forecasting harder - as, for example, during the past few months.

Since March or so, I have been recommending that you get hedged or short. One of the main reasons was the deeply negative difference between CPI inflation and PPI (producer price index) inflation. This cost-squeeze made it a near-certainty that many companies would miss their numbers - even if management (and the top analysts) were very sure they wouldn't. This indeed has come to pass.

Here's a vignette by way of illustration: Last week I checked the earnings forecast by a top-ranked U.S. analyst of a stock I thought was a short; based on his forecast, the stock actually looked cheap. Out of curiosity I looked at the guy's track record. Well. A year ago he forecasted high earnings for this stock and ranked it a "buy" with a target of $38. Then earnings fell and he cut his target to $31; then he cut it further to $26. The stock is now at $11, and the analyst cut his target overnight to $13 and changed his "buy" to a "hold." And yes, he is still top-ranked.

Now, I don't want to pick on analysts. I view my own forecasts with the same doubt as I view others' - which is why I prefer to buy inputs and wait for them to work out.

Here are two input examples, one tangible, the other intangible:

First, take Lam Research , a big, growing, but cyclical producer of chip-making machinery with 20 years in the business. Results-oriented buyers would forecast earnings and buy if current price doesn't reflect their forecast.

Input-oriented buyers, however, wouldn't like to hang their money on their own (or even top analysts') forecasts. So they would look to buy Lam's economic inputs cheaply, even if analysts are still forecasting low earnings.

What are Lam's inputs? Mainly property, plants and equipment (PP&E) along with research and development (R&D). So let's look at Lam's past record: what level of EBITDA did one dollar of PP&E yield at the cyclical peak and at the trough? If both plant and business are intact, you would expect them to yield at least the average of these levels some time in future. Of course, you must understand the business to know this.

How can you estimate "normalized" earnings from this future EBITDA? Simply subtract its capital expenditures (standing in for depreciation, long term), interest, and corporate tax rate - and voila: long-term earnings! If you can buy these cheaply, go for it, and wait.

But beware: If the company's "moat" was breached, even with the same PP&E and R&D "inputs," all future earnings estimates would fly out the window.

What's a "moat"? It's a term coined by Mr. Buffett, referring to whatever keeps customers tied to the company and prevents competition from barging in. It's an intangible input, but extremely important nevertheless.

Which takes us to the second, intangible example: Research In Motion.

Whereas other smart-phone companies rely on public networks, RIM has a network of servers that makes it difficult for outsiders to peek at its customers' communications. It's a deep moat around RIM's business that has helped it grow and prosper.

However, nosey governments worldwide recently began badgering RIM to give them access to its customers' communications. Once RIM agreed to begin negotiations on the subject (as it did with Saudi Arabia), it lost its "moat." At that moment, all future earnings forecasts became suspect. So if you thought of buying RIM at a low multiple of the top analysts' forecasts, think again. You can both be wrong.

How can you use all this in your investing? First, treat all analysts' EPS forecasts - including your own - with a grain of salt. Second, treat analysts' price targets with an even larger salt grain. Third, aim to buy inputs cheaply, not only results.

But if a company's business moat is infracted, forget about earnings, either current or forecast, and get out.

Report Typo/Error

Follow us on Twitter: @GlobeInvestor

Next story




Most popular videos »

More from The Globe and Mail

Most popular