Someone stepped in something. At least leave your shoes by the door.
MS’ Smittipon Srethapramote’s recent note following up on AMSC was actually quite helpful. It reminded me to look at AMSC from the short side (vs his “base case” target 20% higher at $28), something I have not done in a long time. I see Smitty as an indication of how Consensus thinks, and my suspicion is that he doesn’t know what’s really happening. And no, I have no affiliation with the Citronellas who claim AMSC is a fraud. A company and its stock are frequently dissociated.
A little background. AMSC is a serial stock seller. It has this super-sexy story about “an option call on superconductors” that is always just a year or two away and never seems to pan out, and it constantly sells shares in an effort to build out capacity to support this segment. Check out the number of times it has come to market over the past fifteen years. It’s sick.
Up until these past few years, the company never made money and always sold shares. It recently lucked into a segment that led the company to make money, so it has broken out…in a sense…
But the recent report suggests AMSC might head back into the gutter again soon. After Citi’s stupid “Hold” rating was maintained on the back of a massive target price cut, I had to see what was up with my favorite short. Lo and behold! Smitty to the rescue! He gave me everything I needed to know.
First off, if some analyst is stupid enough to pop off with a DCF analysis, it begs to be ridiculed. Almost without exception, a sober DCF always leads to a target price lower than where the stock is currently trading. The meat of a DCF is, after all, in the assumptions, and if you are trying to win banking business your assumptions will always be positive.
So Smitty pops off with a DCF on AMSC. The inputs look solid (assuming his model is correct), the assumptions look OK…wait a second! Sobriety?!? No, his DCF calls for $23 and his “base case target” is $28 (“base case” is what MS uses when it has a Hold on a stock and does not want to publish a specific target).
Next, we start looking for signs of cash flow strain. This is what will tell us whether the company is looking for cash sometime soon. A company’s currency is its stock multiple, and the best time for a company to sell shares is when it has a decent multiple since it leads to less dilution. Check out the chart on AMSC. It does not take a genius to figure out that its currency is going down and it better get hot if it thinks it needs cash. A serial stock seller like AMSC is keen to this sort of thing.
Look at Smitty’s share count vs stock based comp. He shows a near double in stock based comp but no change to share count. This doesn’t really work. Something smells already.
Next, cash flow. In a strange twist, his working capital assumptions lead to zero contribution or drag even though sales grow significantly. It is a rare capital goods company that can grow sales without investing in working capital. In 2012E, cash from ops is 81mln, but we should assume something less than 81mln and higher than net income of 46mln. Let’s split the difference and call it 60mln. His capex estimate is set to 44mln, correctly capturing increasing sales and the continuing drag from the wires venture. Even with the weird working capital assumptions, no flags so far, even with a cynical eye.
Let’s take a look at the model to make sure nothing wacky is going on here. The balance sheet and cash flow statements are derivatives of the income statement, so any slip-up here and the balancing act mentioned above shifts quickly.
AMSC recently acquired “The Switch,” a company whose margins are in the 20s vs AMSC’s current high-30s/low-40s, operating margin in the low teens vs legacy AMSC near 20%. This new company’s main customer is China’s Goldwind, meaning there is virtually no way this new segment’s product margin is ever going to rise. With prior main customer Sinovel flattening, there is virtually no way AMSC’s original wind business margins are going to rise, either. “The Switch” had ~$180mln in 2010 sales vs AMSC’s $430mln, meaning this margin impact is material. Yet, Smitty’s model does not call for stiff margin declines, and incremental margins are not represented at all. With operating margin barely declining – and in fact increasing in out-years on declining product margin – the cash flow model is at risk of unbalancing negative.
Which circles us back to his $23 DCF. His current optimistic model assumes
- minimal decline to product margins despite the recent acquisition
- opex benefit from stock based comp but no increase to share count
- questionable working capital assumptions
If margins do turn out to be lower, this DCF unravels quickly. This story becomes one about declining margins on higher sales, and the higher sales need to be enough to offset. We see how that is working out for the solar names which are successfully building sales faster than margin decline, and the things are trading 5x-10x. I wonder how an unsuccessful company will fare?
AMSC recently issued its Q4 report, and all estimates are in. At $24, the stock is trading 21x 2011 EPS and 17x 2012 EPS. Consensus gross margin declines minimally through 2011 and 2012, shows no estimates for operating margin or EBIT, but does show declining and then rising estimates for EBITDA.
Thanks to Smitty’s Consensus thinking, I now have a decent short idea. I love short ideas. Brings me back to the day. I will be dusting off the old AMSC model to make sure I am not missing anything here, but I suspect I will wiggle into a small short position on AMSC this week, more as a pre-emptive measure than anything spiteful. No driving angry on this one…yet.