Yesterday I wrote about the life cycle of a growth stock, using Walmart (WMT) as an example. Key thoughts: 

–a typical growth stock has a five-year life;

–the best of the best are able to reinvent themselves and extend this super-growth period.

when the music stops

Things can get ugly when the market senses a fast-growing company is beginning to mature and the rate of profit gains is slowing. Not only does the stock fall because of less-than-expected earnings figures; the PE contracts, as well, as the market revises down its thoughts about future growth. 

An example: Say a stock is trading at $30, with $.75 in earnings per share over the past four quarters and the expectation of $1 in eps during the year ahead. Say, also, that the forward multiple is 30. If the actual turns out to be $.85, or 15% growth instead of 30%+, then the stock could end up trading at 15x $.85, or about $13, with a likely target of $15 for the following year. In short, the stock is more than cut in half. The reality may not be quite as bad as this, and the pain probably doesn’t come all at once (although a big chunk does), but the ultimate bad move for a growth stock investor is to be holding a stock when the market realizes the eps growth machine is running in a lower gear.

NVDA

The NVDA story: 

(1) the company started in 1993 as a designer of graphics chips for PCs to supplement the work done by CPUs. 

(2) It became the provider of choice for graphics cards for online gamers, as that business exploded. 

(3) It then became the go-to company for crypto miners. 

(4) Now it’s the first choice for AI, with demand clearly, for now anyway, significantly outpacing supply.

NVDA earnings

I haven’t done my own estimates for NVDA. I’m relying on the IBES consensus for the fiscal year ending 1/31/25 is eps of $21/share, meaning a forward PE for the stock of 33x. The trailing PE, which has less relevance, is 92x. The difference between the two does, however, illustrate the enormous jump in earnings expected for the coming year.

My judgment is that the $21 is what’s driving the stock, and is pretty much already baked into today’s price. If so, further upward movement will be predicated on actual earnings from NVDA exceeding the consensus estimate. 

the most important issue

The most important qualitative question, to my mind, is whether the 1/25 numbers represent a peak in earnings– meaning 1/26 earnings will be lower than 1/25 results (bad/very bad for the stock at the point when the market begins to work this out)–or whether it’s thinkable they’ll be significantly higher (good/very good). A flat 1/26 would likely not be good, either.

I don’t have an answer for any of this right now. The only relevant observation I can come up with is that Super Micro Computer (SMCI), a company that I’ve epically underestimated and whose main virtue seems to me to be that it has strong access to NVDA AI chips which it installs into the computers it sells, has more than doubled so far this year and has very recently reported that demand for NVDA chips still handily outstrips supply. A canary for our coal mine?

We’ll presumably learn more from NVDA when it reports in two weeks.

more on Monday