This Just In: Upgrades and Downgrades

Shares of semiconducting superstar Intel (NASDAQ: INTC) are jumping this morning, rising more than 2% on the back of a new upgrade from Wall Street banker JMP Securities. That sounds like good news, except for one nagging detail ...

And that's that we have no details on why JMP upgraded the stock. Make no mistake, multiple news outlets have confirmed that the upgrade happened. StreetInsider.com for example, notes that although Intel currently costs less than $35 a share, JMP is saying the stock's worth closer to $45 per share. It's just that no one seems to know why.

Speak up, JMP. We can't hear you!
There are situations more frustrating to the individual Intel investor than the one we face today. We know that JMP likes Intel. We know the stock is now 2% more expensive because JMP has publicly said it likes it. But we don't know whether we should pay the higher price, because we don't know whether JMP's opinion is worth following.

So let's fix that.

Let's go to the tape
At Motley Fool CAPS, we've been tracking how well JMP Securities' stock picks perform for nearly a decade now. And by and large, it seems they've performed pretty well. According to our stats, JMP ranks in the top 15% of investors we follow.

The bad news is that JMP's performance has been pretty spotty. Although it's had some fantastic winners (see below), in fact, only about 45% of JMP's recommendations actually beat the market at all. The worse news is that JMP has been particularly bad at picking semiconductor stocks ...

As you can see from the sampling above, JMP made a fabulous -- and prescient -- call on smartphone specialist Skyworks Solutions, a stock it first recommended more than six years ago. JMP was right about Skyworks then, and it's probably still right about it today. With nearly $600 million in trailing free cash flow, Skyworks sells for less than 24 times FCF, true. But Skyworks is also growing at a 21%rate, and paying a modest 1.4% dividend. It's not "cheap," exactly, but it's not terribly expensive, either.

Elsewhere in the semiconductor sector, however, JMP has racked up a record of just 30% of its stocks outperforming the S&P 500 -- and this gives cause for worry.

Parsing the news
You see, while the media haven't done a great job of reporting on JMP's reason for recommending Intel, we can make an educated guess that it had something to do with what Intel told investors at its Investors' Day presentation on Thursday.

There, Intel gave guidance for the year ahead, predicting that it will:

• grow revenue "in the mid-single digits,"
• earn a gross profit margin on that revenue of "62%, plus or minus a couple points,"
• and invest "$10 billion, plus or minus $500 million" in capital spending. 

Now let's put those points in context. According to data from S&P Capital IQ, Intel generated $12.2 billion in positive free cash flow over the past 12 months. Its market capitalization, ex-cash (i.e. its enterprise value) is $156.5 billion. And so its enterprise value-to-free cash flow ratio works out to 12.8.

Valuating Intel
That would be fine for a company growing earnings in the low teens, percentage-wise. But according to Yahoo! Finance, most analysts are looking for just 8% annualized earnings growth out of Intel over the next five years -- and Intel might not even achieve that much.

Consider: Take Intel's forecast "mid-single-digits" revenue growth as a starting point. Then subtract a fraction of a percent for the forecast 62% gross margin (because Intel today is grossing closer to 63%). This implies that earnings will grow slower than revenue, and could even fall short of analysts' hoped-for 8% growth.

Even more concerning, in my view, is Intel's planned growth in capital spending to $10 billion "plus or minus." That's a big increase over the $7.1 billion Intel spent over the past 12 months, and threatens to crater free cash flow -- perhaps dropping cash profits below $10 billion, and increasing Intel's EV/FCF ratio up to 15 -- or more.

The upshot for investors
Long story short, it's not hard to see how, a year from now, we could be looking at an Intel costing 15 times cash profits, but growing those profits at 7% or less. That stock won't look like much of a bargain, and I'd be willing to be it's not going to cost anywhere near the $45 target price JMP has assigned it. I'd be even more willing to make that bet after seeing how poorly JMP has fared with its previous semiconductor picks.

Simply put, Intel is a stock that looks cheap today, but probably isn't as cheap as it seems -- and JMP Securities is wrong to recommend it.