Earlier this month, I reassessed my cautious stance on A-Power Energy Generation Systems (Nasdaq: APWR) following a roughly 30% drop in the shares. The stock suddenly looked pretty cheap, according to some analyst earnings estimates, but I couldn't get a handle on the 2010 outlook with any great confidence. That was enough to keep me on the sidelines.

Today, we learned that those Wall Street estimates -- and there are just a handful of them -- were extremely wide of the mark. A-Power's guidance came in at roughly $380 million in revenue and $45 million in net income. That represents pretty modest growth over 2009, while analysts were looking for a really big change. One individual, perhaps overly excited about A-Power's new wind turbine manufacturing capabilities, forecasted revenue at more than $600 million.

At least I proclaimed ignorance.

Like solar compatriot ReneSola (NYSE: SOL), A-Power is full of surprises. Sometimes, as with the announcement of the West Texas wind farm, these surprises are pleasant. Other times, as with the out-of-the-blue equity offering in January, not so much.

Those institutional investors who bought into January's private placement at $14.37 per share are now down a quick 25% or so. Once bitten, they may be twice shy the next time A-Power passes the hat around. That's a big problem, given that the company needs to keep financing new manufacturing plants abroad to compete with giants like General Electric (NYSE: GE) and Siemens (NYSE: SI) in the wind- power realm. The proposed Nevada plant is a clear example of this imperative.

Beyond mismatched expectations and their impact on A-Power's cost of capital, another issue is that A-Power is now looking at three consecutive years of approximately flat earnings per share, on a non-GAAP basis. This was supposed to be a growth business! A-Power has certainly grown into all sorts of new alternative energy businesses, like solar panel manufacturing, and its average share count will be up by around 50% this year compared with 2008. The good kind of growth is so far proving elusive, though.

The good news in all of this is that $45 million in net income would still represent a nearly 18% return on beginning shareholders' equity. While A-Power has disappointed investors today, it does have a shot at generating satisfactory returns this year. At today's share price, A-Power looks cheap on a forward earnings basis, but you should also bear in mind that a company's value derives from a long-term stream of future cash flows -- not one year's earnings.

The key question, then, is whether A-Power can increase intrinsic value per share for many years to come. I think the company's scattershot growth model is not doing it any favors. Had the company stuck to its distributed generation business, we would likely be having a very different conversation today.