Micron: Secular, Seasonal, Cyclical


I don’t need to tell you Micron’s (MU) stock is cruising down to areas many thought wouldn’t be seen for some time, if ever again. The financial media is unofficially calling a top to the memory cycle. Analysts expect DRAM pricing to destabilize if it hasn’t already and for NAND to continue down further from already weakened readings.

But could it be analysts and Wall Street are calling one thing something else?

It’s easy to see how after a continued rise in memory pricing, any signs of weakening could be considered the beginning of the end – an end to memory good times and a start to margin compression and weakening of profitability.

“But wait,” you might say, “semiconductor equipment suppliers and memory players – most notably Micron – have said demand trends are secular and supply is disciplined.”

And you would be correct. The question really is: What’s going on in the last month as analyst after analyst has retreated revenue and earnings estimates for FY2019? One of these parties aren’t telling the truth, right?

Maybe they are – all of them.

It turns out two things can be true at once. The difference is analysts don’t actually know what they are seeing and Micron management doesn’t believe profitability will grow quarter-over-quarter indefinitely in this new demand era.

Cyclical

I don’t need to explain much of this one to anyone who has taken an interest in either the bull or bear case regarding Micron. This is a term we haven’t been able to get past as this is where the memory market has come from over the course of the last three decades.

Look no further than Micron’s profitability over the last many years.

ChartMU Net Income (TTM) data by YCharts

Those peaks above the zero line are undoubtedly the peaks of the cycles while the troughs below the zero line are quite obviously the bottom of the cycles. And this has happened with predictability and repetition over the last 30 years.

It’s no wonder, then, this term continues to get used as Micron hasn’t proven with enough time this is a thing of the past. There are differences, however, from prior time periods, as there are externalities which haven’t existed before. The easiest of these to point out is the consolidation of the DRAM industry. Over the years DRAM manufacturers have halved, with 23 existing in 1997 to 11 existing in 2012 to today where only six exist by name (Samsung (OTC:SSNLF), SK Hynix, Micron, Nanya, Winbond, and Powerchip) with the top three claiming 95% of the most recent calendar quarter’s revenue.

(Source: A Strategic Analysis of the DRAM Industry After the Year 2000 – Kyung Ho Lee, MIT)

There’s some evidence toward an argument of cyclicality becoming less and less likely. Today, with only three players in nearly complete control of the DRAM market, one can conclude a higher likelihood of supply being maintained and groomed. In a situation with 23 or even 11 vendors at work, there’s less chance of being able to control the market’s supply. When one manufacturer owns 30% of the market and curbs supply to adjust for demand it has a much higher effect than 1 out of 20 with only a 6% share of the market. That 6%’er can’t do much to influence the market, but the 30%’er certainly can have its forces felt.

Coming back to my original point, this is where analysts and the market are predicting we are currently headed – pricing is destabilizing and therefore the cycle must be turning and heading for the inevitable zero-sum endgame. That’s the way it has always happened and nothing market wise has shown it can’t happen again.

But is it possible there’s an alternative explanation?

Seasonality

After two years of steady expansion of net income, with two years plus a quarter expected in just a week, it’s reasonable to expect this trend won’t continue.

Or will it?

With memory pricing on the rise for nearly all of that time, the first crack in the pricing armor sends an immediate worry down analyst and investors’ spines as downward memory prices aren’t good.

Well, wait a minute – that’s not how the DRAM and NAND markets work. I’ll get into more of that in the next section, but for now, I want to discuss why pricing may be fluctuating and why it appears to some analysts why demand is softening.

During the second half of the calendar year, 2017 (the first half of Micron’s fiscal year 2018) memory pricing was very much still on the rise.

(Courtesy of Electric Phred from Micron: We’re In The Chips)

This strength in pricing easily outshined any type of seasonality in the market. Seasonality was smoothed out through the constant upward pressure on pricing, masking any effects. Now, what seasonality am I talking about exactly? The seasonality of most major technology retail brands, everyone from Apple (AAPL) to GoPro (GPRO), to Nvidia (NVDA) to HP (HPQ). All of these consumer brands have a holiday season in the fourth quarter to sell into and all of them require a lot of storage, memory, or both.

You could say, “Well isn’t the seasonality of the fourth quarter good for firming up pricing and an indicator of high demand ahead of this quarter?” Yes, you could. You also could ascertain this is what’s happening in the memory market right now because all analysts have done is worry about expectations of a drop in pricing because of the current trends they are seeing. Taking DRAM, for example, there hasn’t been much belief pricing has weakened substantially in the current quarter. Analysts have said their notes are based on the perceived weakening of demand and a buildup of inventory. This is what I’m seeing seasonality on. Going from the third quarter to the fourth quarter it makes complete sense inventories have risen in order to supply the above-mentioned consumer names – among many more – in order to reduce the likelihood of any supply shortages in components as these companies ramp production of their own.

In fact, Micron’s CFO alluded to this lack of concern regarding inventory levels during the Citi Technology Conference by saying, “I’m not sure inventory levels ends up being a big story over time, though.” It’s a simple explanation: Inventory levels at this time are seasonal, not long-term indicators. And if this is what Wall Street analysts are getting their panties in a bunch over, it’s because they forgot what a seasonal effect feels like as last year’s season was overshadowed by domineering pricing.

Secular

This also is where some have worried – the word of this new memory era has gotten lost or never existed in the first place. The worry stems from this memory market trend, which is supposed to be the savior from the cycle, being a smoke screen and not actually coming to fruition. Secular trends can’t have deteriorating pricing, pricing must stay stable for as long as the secular demand stays intact.

Or that’s the conception – or, rather, the misconception.

This is not what’s meant by secular. When Micron talks about secular trends it means trends which are much longer lasting than a PC market or gaming market uplift, trends such as cloud infrastructure or edge device computing or AI deep learning. These trends are slower ramping, longer lasting trends as the world is moving toward computing in this manner. Very flexible and malleable demands of a PC market come and go based on who’s buying desktops and laptops and how much expendable cash consumers of the day have. Demands for data centers and data centers which can handle the increasing needs and continuing capabilities of AI deep learning come from a different source of demand as well as funds. This market is growing exponentially and is not like the current or past PC market. In fact, AI revenue is expected to grow 57 times from 2016 to 2025 – that’s only a little more than six years away, not 10 or 15 years. This means any large company in the world will have to contract out or build its own AI platforms and servers to capitalize. All of it will need DRAM, storage, and new types of memory (think 3D XPoint).

Throughout this growth in major computing trends, there are going to be ups and downs in the memory price market. However, focusing solely on the pricing aspect misses the bigger picture. These secular trends are going to last for years and this means underlying demand will remain as foundational. This is not the same as memory being the same price over this time period; memory will get “cheaper” as time and technology moves on. The key is the supply and cost. Going back to the cyclical discussion above with three major players looking to maintain a happy balance of supply, we can focus on costs which is where the rubber meets the road.

Without going into detail about technology node transitions to ascertain more bits per wafer and thus reducing costs – as we’ve all had our lessons here on Seeking Alpha about relative memory costs – I want to focus on the simple equation of bits costing less in the face of selling for less. Memory pricing coming down does not mean revenue or even profitability comes down or even stops growing. If costs can be reduced faster than average selling prices while the volume of bits shipped grows (a factor in more bits per wafer), Micron is still a very happy company.

If Micron reduces costs of producing bits at a higher rate than the price-per-bit falls, it can still see margin growth. But allow me to go one step further in reminding you it’s not just about commodity (contract) chip prices when the company is producing higher-value, higher-margin products – which allowed for an increase in ASPs last quarter. I’m referring specifically to the company’s NAND ASPs rising while NAND pricing fell.

While on a like-for-like basis, NAND pricing declined modestly sequentially, our overall NAND ASP increased in the mid-to-upper single-digit percentage range, driven by a richer mix of high-value solutions in our NAND portfolio.

– Micron CFO, FQ3 Conference Call

Don’t Confuse The Three

There’s not a whole lot more to say – pricing is not the end-all-be-all. Falling pricing can still produce a very profitable and large free cash flow business. Granted growth quarter-over-quarter won’t be in the 20% and 30% range as it was just a year ago, but Micron doesn’t need crazy growth each quarter for the stock price to rise. The stock simply needs a market perception shift in order for it to see multiple expansion, driving the stock price magnitudes higher. And all of this to say, pricing may not even be falling or will even fall in the next two-to-three quarters like analysts are predicting! But guess what? Even if they do, read this article again because it doesn’t change.

While my opinion and analysis may not make you feel any better about your portfolio’s pain, at least take this away: If predictions of analysts and Wall Street are wrong about what they expect (and they have been in the past, just recently, too) this is one of greatest the buying opportunities we may see from Micron. It won’t take long to prove it’s all due to fears of a return to a cyclical past. It may simply be analysts reading seasonality but claiming cyclicality while ignoring secularity.

If you’d like to be made aware of my opinion and analysis in the future on Micron and other tech companies, then I encourage you to follow me by clicking the “Follow” link at the top of this page next to my name.

Subscribers Got A First Look

I told my subscribers about my thoughts regarding Micron through my service’s chat room as well as my podcast before this article was published. Not only were they aware but they also were given the strategy and the technical chart explanation to capitalize on it, something I don’t share with my public readers in articles like the one you just read. To be made aware of opportunities like this, along with the strategy to profit from it, you need to join me in my service, Tech Cache. My service discusses tech and tech-related companies and the opportunities therein because the growth your portfolio needs is in tech. Right now, you can try it risk-free with a 2-week free trial!

Disclosure: I am/we are long MU.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.



Source link

Leave a Reply

Your email address will not be published.