Deep dive writeup: $MU <> Micron Technology
For all subscribers
Company: Micron Technology
Tagline: “Our Memory Your Innovation”
IPO date: January 1984
Micron’s historical timeline [click here]
Current stock price: $94.45
Outstanding shares: 1.12 billion
52 week high: $98.45 on January 5, 2022
52 week low: $65.67 on October 12, 2021
Market cap: $105.8 billion
Enterprise value: $101.7 billion
Headquarters: Boise, Idaho
Number of employees: 43,000+
Analysts: 65% have a buy rating
Average price target: $106.81
Investor Relations [click here]
FY2022 Q1 earning report [click here]
FY2022 Q1 earnings presentation [click here]
FY2022 Q1 earnings call prepared remarks [click here]
FY2022 Q1 earnings webcast [click here]
FY2022 Q1 earnings call transcript [click here]
JP Morgan Tech/Auto Forum [click here]
Credit Suisse Annual Tech Conference [click here]
+10% over the past month
+20.7% over the past 6 months
+19.4% over the past 12 months
+328.5% over the past 5 years
+6,582% since the IPO in January 1984
I worked on this $MU writeup with Andrew, you can find him at @AJButton2
Micron Technology ($MU) is a semiconductor company in the memory sub-sector. It supplies memory and storage hardware to the world’s leading tech companies. The bulk of its business consists of selling DRAM and NAND Flash to manufacturers like Apple ($AAPL). It has a smaller business selling similar products (RAM sticks, SD cards and SSDs) to consumers. $MU is unquestionably selling into some of the hottest and highest growth sectors like data centers, automotive and electric vehicles, artificial intelligence and 5G.
Micron Technology’s two main business activities are manufacturing DRAM and NAND Flash.
DRAM is the computer memory used to temporarily store data for quick use. Commonly called ‘RAM’ — it is what your device uses to multitask. NAND Flash is a type of high-speed storage used in modern computers and related devices.
Here’s a short video that provides a nice overview of $MU products:
Micron’s products DRAM products are used in:
The company targets a different client base with its NAND Flash, which is mainly marketed to computer manufacturers and data centers.
Micron sells both its DRAM and NAND Flash products to consumers. Gaming PC enthusiasts will be familiar with Micron’ Crucial RAM brand, which is widely used in gaming computers. Micron also sells SSDs to consumers, both as Crucial and under its corporate name.
Micron is currently the third largest supplier of DRAM in the world, after Samsung and SK Hynix. It has a 23% market share in DRAM globally. It is also a major player in NAND Flash–a more fragmented market–with an 11% market share. Owing to the more consolidated nature of the DRAM industry, profits are higher in that space. Also, Micron does more total revenue in DRAM than NAND–the former accounts for 73% of revenue, the latter 24%. The remaining 3% comes from business activities too small to merit individual mention in Micron’s financial statements.
Micron’s business activities make it one of the few “pure plays” you can find in the memory space. Samsung is the biggest player in the memory industry, with a 42% market share in DRAM and 33% in NAND, but it’s a highly diversified conglomerate. When you buy Samsung you’re not just buying a memory business, you’re also buying a smartphone business, an appliance business, and much more. Not all of this is as “growthy” as memory is right now, so Samsung might not be the best way to take advantage of high DRAM prices and strong demand. There’s also SK Hynix, which is a pure play similar to Micron, but that’s pretty much the only other one. So if you want to invest purely in memory, your options are MU and SK Hynix. That’s a pretty limited selection of stocks, but it’s ultimately a good thing, as this small selection of players means limited competition and high margins.
THE MEMORY INDUSTRY:
Micron Technology is first and foremost a computer memory manufacturer. This is a sub-sector within the semiconductor industry characterized by “commodity” products. When you hear the word “commodity” you probably imagine raw materials like oil, minerals and grain.
Those are commodities, but in fact, manufactured products can be commodities as well. Nevertheless the DRAM market has shown steady growth over the past decade as consumers and enterprises need more/faster computing power as well as more storage. $MU has been able to capitalize on these trends which should continue for many years to come as you can see from the chart below.
Based on this definition, DRAM and NAND Flash are commodities. Or at least, they have been historically. For most of the history of the memory industry, buyers have treated different companies’ offerings as more or less equivalent. You’ll often notice that PCs include stickers advertising which brand of CPU and GPU are in the machine. They will not have such a sticker for the brand of RAM that’s under the hood. There’s a reason for that. As long as they have the same speed and other specs, two sticks of RAM from different manufacturers are basically interchangeable. You can test this out yourself. If you have a laptop you can insert two sticks of RAM (let’s say one stick of Crucial RAM and another stick of Corsair RAM), and you’ll find they will function perfectly together. You can’t do the same test with most modern smartphones as most smartphones use system-on-chip design. But it would work in theory as well.
On the surface of it, this would appear to be a negative for Micron. Generally, it is better for products to be differentiated rather than interchangeable, as differentiation gives brands more pricing power. However, the nature of the memory industry as an oligopoly serves much the same effect. As mentioned previously, the DRAM market only has three major players. NAND flash has about six. With so few companies selling memory in large quantities, they collectively have some pricing power. Like most commodities, memory is subject to price fluctuations, as I will explore in more detail shortly. The big takeaway is that the small number of players in the space helps keep price pressure to a minimum. It also explains why the comparatively more concentrated DRAM space has better margins than the more fragmented NAND flash space.
We have established the basic features of the computer memory industry: that it’s an oligopoly selling commodities, with three or six major players depending on the sub-sector. With that understood, the next logical place to look is at Micron’s position within the industry. There are two ways we can approach that:
Micron’s overall market share
The specific companies that Micron sells to
The first of these has already been touched on briefly. Micron is currently third in DRAM with a 23% market share, and fifth in NAND Flash, with an 11.1% market share. The chart below shows the market share breakdown in both industries:
As you can see, Micron is a major player–but not the dominant player–in the two sub-sectors it operates in. This gives it solid positioning, though obviously it would be desirable for Micron’s market share to improve. According to South Korea’s Pulse News, Micron’s growth in the memory space has been held back by a lack of specialized equipment.
The publication’s coverage states that SK Hynix and Samsung use expensive EUV– lithography machines from the Dutch company ASML Holdings ($ASML). The story continues to say that Micron has so far been unable to procure this equipment due to its extreme cost–$181 million per machine. With Micron now doing $2.5 billion in quarterly net income, it seems the firm will be able to acquire a few of these devices. At any rate, Micron’s market share has been increasing with or without EUV. According to Pulse, MU has made advances in using the older deep-ultraviolet (DUV) technology. Using this technology, Micron was able to beat SK Hynix and Samsung to several key product milestones, such as the release of 176-layer NAND. So, the lack of EUV hasn’t prevented Micron from being a top player in the space, pushing the needle of innovation forward.
Having established that Micron is a leading memory company, we can now look at which companies it supplies. This is crucial because the bulk of memory revenue is business-to-business (B2B). There is some consumer demand for RAM, SSDs and SD cards, but it’s a comparatively small part of the industry. It’s really the industrial use of DRAM and NAND that cranks out the bulk of Micron’s profit. So, Micron’s buyer relationships are of pivotal importance to any thesis on it.
Among Micron’s largest customers are:
Previously, Micron sold chips to China’s Huawei, a major smartphone and infrastructure company. That lucrative relationship was 10% of Micron’s business at the time. Unfortunately, Micron lost the business when the U.S. banned the sale of semiconductor inputs (such as DRAM) to Chinese firms. The last year in which Huawei was in the picture was in 2018. Micron’s earnings still haven’t climbed back to that level. That is not a strike against the stock today, in 2022, as the company is growing its sales to other customers amid rising DRAM prices. If we look at these two trends (rising sales volumes and DRAM prices) side by side, we can see the potential for Micron’s earnings to get back to 2018 levels in 2022. Micron itself forecasts strong CAGR growth all the way to 2025, which could take the company’s earnings to never-before-seen highs. On top of that, if the U.S. were to ease off with its ban on exporting to Chinese tech companies, then Micron would probably get Huawei’s business back, leading to a quick spike in revenue and earnings.
Having looked at Micron’s position within its industry, we can now look at the company’s business operations, structure and profit by segment.
Micron Technology has a “mass market” business model. That is, a model where the business does not distinguish between different groups of customers. Most of its sales are business-to-business, meaning it sells to other tech companies. These include computer manufacturers, phone manufacturers, and chip makers. It locks in volume-based annual contracts where customers get bulk pricing depending on how much they order.
Micron divides its business into four units:
Compute and networking — data centers, servers, enterprise clients.
Mobile — memory and storage for smartphones and tablets.
Storage — SSDs, both for manufacturers and for consumers.
Embedded — memory products used in smart cars, IoT devices, etc.
Compute and networking is by far the largest of Micron’s business units — it did $3.4 billion in revenue in Q1, or 44% of the total. It also had the second-best growth of the four segments, at 34% year-over-year.
Next up is mobile, the second largest of Micron’s business units. This is the business unit that gets the most publicity as it’s the one where Micron sells to “hot” smartphone and tablet manufacturers like Apple and (formerly) Huawei. Hot discussion topic it may be, but mobile is nowhere near CMBU in terms of size. It did $1.9 billion in revenue in the most recent quarter, or 24% of the total.
Third on the list is storage. This is probably the weakest of all of Micron’s business units right now. It was last on the list by revenue in Q1, doing $1.1 billion in sales. It also had by far the weakest revenue growth of the four, at 26% year-over-year.
Last on the list is embedded. This is the second smallest of Micron’s business units, but also the fastest-growing. It did $1.2 billion in revenue in Q1, or 15.5% of the total, and it grew revenue by 51% year-over-year. This makes perfect sense, as the types of products that use Micron’s embedded products (EVs, IoT etc) are all growth products that are seeing huge increases in adoption. The use of computer chips in everyday items is rapidly picking up steam, and with that comes the need for memory, the kind that Micron sells. As long as the popularity of these products continues inching upward, then demand for Micron’s memory will continue to grow.
Having thoroughly reviewed Micron’s industry and business, we can now turn the page to its financials. Micron technology had a year of solid growth in 2021, though as you’re about to see, is still down compared to where it was in 2018. It operates in an extremely volatile “boom and bust” industry where prices are going up and down all the time. For this section, I’ll take a look at $MU most recent quarter first, then the long term averages.
In its most recent quarter, Micron Technology delivered the following:
$7.69 billion in revenue, up 33%
$2.31 in GAAP net income, up 187%
$2.16 in adjusted EPS, up 177%
$3.9 billion in cash from operations, up 100%
Incredible numbers. We’ve got solid triple digit growth in virtually all profit metrics but most impressive might be the free cash flow which gives them capital for buybacks, dividends, acquisitions, capital improvements, etc.
Everything above looks great, right? Indeed it is, for the most recent quarter but if we turn to the long term historical averages we start to identify a potential problem. For the trailing three year period, the CAGR growth rates in the metrics above were:
-2% in revenue.
-20% in net income.
-19% in adjusted EPS.
-25% in cash flow
So what I’m saying is that $MU’s business is very strong right now but there’s no guarantee it will continue in perpetuity. We need to keep an eye on these fundamentals and make sure they don’t start falling apart.
This fact, of course, stems from Micron’s blowout performance in 2018. That year, Micron was firing on all cylinders with RAM prices high and Huawei’s business still intact. As the chart below shows, DRAM reached a huge peak in 2018, fell in 2019, and then recovered abruptly in 2020, with the modest uptrend more or less continuing to this day.
As you can see from the chart, DRAM prices still have yet to hit their 2018 highs. This fact is also reflected in Micron’s earnings trend, which you can see in the chart below:
Micron’s earnings are highly correlated with RAM prices. Micron’s revenue, EPS and cash flow charts all follow almost the exact same trend as RAM prices from 2018 to the present. The question then becomes whether RAM prices will continue going up, level off or start to decline? To answer that question we need to take a deep look at what drives RAM price volatility.
RAM PRICE VOLATILITY:
The price of RAM, like that of all commodities, depends on supply and demand. Technically this is true of every product but it is most obviously felt in commodities, for which demand isn’t easily propped up through marketing or differentiation.
Historically, the prices of DRAM and NAND flash have been volatile. The chart above showed the trend in various types of RAM since 2018. That wasn’t just a recent trend, it reflects the entire history of the industry. John MacCallum’s website has a history of RAM prices going all the way back to 1957. It’s a fascinating document to look at. It shows prices as high as $392 for memory with only 0.00098 of capacity back in 1957. Obviously, the computer industry was totally different back then, with only a handful of customers buying RAM for mainframe computers from an even smaller number of suppliers. Still, that first point in 1957 is the start of a history of RAM prices that have seen many ups and downs.
A few select data points from more recent years include:
$88 for two sticks of 8gb DDR3-1600 in February 2013.
$129 for the exact same two sticks in December of 2013.
$92 for the exact same two sticks in 2015.
$49 for two sticks of 8gb DDR4-2133 in early 2016.
As you can see, the price of the DDR3 rises in 2013 then falls in 2015 — then we have a much faster form of the same RAM going for far less in early 2016.
This is the kind of volatility you see in the memory industry. Prices aren’t always perfectly comparable because RAM gets better over time, but you’ll often see drops in demand that result in better RAM being cheaper in year 2 than worse RAM was in year 1. That’s price volatility at work.
And what causes this volatility?
There are many factors, but the biggest is how end-users budget for and use memory. Seeing RAM and NAND flash as commodities, they usually buy it in bulk, acquiring the amount they think they need for a quarter or year. Often, they will be buying it up furiously when demand for their products is strong. Then, when their sales slow down, they’ll be sitting on too much of it, and suddenly slow down their purchases. This results in lower RAM prices, as the suppliers now have less orders coming in at the previous price point. To keep selling it, they need to lower the price enough that it makes sense for buyers to stockpile it.
The above pretty well explains why RAM prices have historically been volatile. The question is whether they will always be volatile in this way, and if so, whether the next year will be a good one for RAM prices.
As for whether RAM prices will always be volatile, the answer to that would seem at first glance to be an obvious ‘yes,’ but it actually might not be. The classification of RAM as a commodity seems to imply it is subject to the boom and bust cycles that all commodities are, but it may eventually cease to be a commodity. Computer chips are getting faster all the time, and with that increase in processing power comes the need for better RAM. With faster and more spacious RAM comes more simultaneous processes for a fast CPU to handle. Put simply, you need good, fast RAM for the CPU to reach its full potential (and vice versa). So, chips with high clock speeds need RAM to keep up with them. Over time, as competition in the chip space increases, there may be more differentiation in RAM required for different tasks. Already, we’re seeing Micron tout its advances in RAM for 5G smartphone chips and advanced graphics. Will buyers see these Micron products as superior and different from those of Samsung and SK Hynix? Only time will tell, but it’s a real possibility.
As for whether next year will be a good one… we’re seeing different predictions on that front. Micron is forecasting 20% to 30% growth in demand through 2025. TrendForce expects prices to dip 0.3% but revenue to rise due to more shipments. On a more optimistic note, Citi’s Christopher Danely expects prices to keep rising through 2022. Apple might be slowing down iPhone production which could cause a dip is prices however Facebook ($FB) and others are investing heavily in the metaverse, an ecosystem of VR and AR apps that will require enormous amounts of server power. That could be a positive tailwind for Micron’s CNBU unit for the next decade.
Hopefully I haven’t lost you yet because the valuation metrics are where $MU gets me the most excited. You’ve probably noticed in 4 of my last 5 writeups ($DLO $GLBE $OLPX $BRO) that I’m focusing on “profitable growth” companies and even though $MU has been around for 40+ years they still check this box.
Just below is next quarter’s guidance from the company but that’s just a short term snapshot, for the sake of talking valuation let’s skip down to the next graphic which is the full year consensus estimates.
$MU is not a hyper growth, emerging tech company but the company is very profitable, has incredible margins and trades at a ridiculously cheap multiple. Since $MU is not on a normal fiscal year I thought it would make more sense to show the quarterly estimates instead so these numbers are the next 4 quarters…
Revenues: $34.5 billion
Revenue growth: 16.6%
EBITDA: $19.4 billion
Net Income: $11.1 billion
Free Cash Flow: $4.8 billion
Now if we start to crunch these numbers using the enterprise value of $101.7 billion…
$MU is now trading at 9x NTM (next twelve month) earnings with 16.6% revenue growth and 39.5% earnings growth. If $MU wasn’t in a “commodity” or “cyclical” industry they’d probably be trading at 20-25x earnings with these numbers but they’re clearly getting penalized because of the historical ups and downs of the DRAM and NAND sectors. I don’t think this is necessarily fair because $MU has done a nice job of evolving their product portfolio, customer base, sector exposure etc
With that said I do think it’s very possible that we see $MU benefit from some multiple expansion over the next 6-12 months if they can continue to put up impressive quarterly results.
In my opinion we could see $MU trading at 14x earnings in the next 6-12 months which would get the stock price to ~$138 per share or ~45% higher from current prices.
I rarely post the compensation of executives but I rarely writeup about $100+ billion companies where the average pay in the C-suite is $11.2 million per year — but I’m not going to make a big deal out of this because it’s pretty common these days.
Analysts are definitely bullish on $MU
TipRanks shows the average price target from analysts at $107.73
Yahoo Finance shows the average price target from analysts at $106.81
Of the 23 analysts who cover Micron… 18 have a “buy” rating, 4 have a “hold” rating, and just 1 has a “sell” rating.
The highest price target on the stock is $165, the lowest is $58.
Analyst estimates of Micron’s earnings have been trending upward over the past month. One month ago, the consensus EPS for Q2 was $1.85 and now it’s $1.94.
These are the analysts price targets according to StockTargetAdvisor.com and when they were last updated.
If we look at who owns $MU stock we see a heavy amount of institutional interest. Mutual funds own ~48% with other institutions holding ~32% and the rest is made up of insiders and undisclosed holders. Only 0.2% of the float is held by insiders which is a relatively low level of insider ownership and sometimes a red flag or bad sign but in $MU’s case it’s very understandable. Micron is a 43-year old company, mature and established. No founders are still involved in running the company so insiders mainly get shares through stock awards and individual purchases. This simply does not produce the same level of insider ownership that you can see in young, founder-led companies.
However the benefit of being an “older” company is that $MU no longer has forced sellers ie venture capital or private equity firms on the cap table so we don’t have to worry about lockups and secondary offerings. Most of my recent writeups are young companies where early investors still own 30-40% of the outstanding shares and over the next few years they’ll need to sell down those stakes… that’s not the case with $MU which is one less thing for us to worry about.
This list of top shareholders is basically a who’s who of fund companies and asset managers — some of these holdings are in active funds and some are probably in ETFs but that’s not really important. The top 25 shareholders/funds currently own 50% of the outstanding shares led by the 800-pound giants like Vanguard, Blackrock, American Funds (ie Capital Research & Management) and State Street.
$MU was down 1.25% on Friday but was still well above its 50-day moving average at $83 and the 21-day moving average at $90. As you can see, $MU started breaking out to all time highs last week but ended up pulling back to the 10d EMA.
Even though I’m not showing it on this chart (gets too crowded), $MU bounced off the 10d EMA on Friday at $93.93 before closing slightly higher. Obviously I’d like to see $MU stay above the 10d EMA but that might depend on the overall broad markets, more specifically the $QQQ which is the Nasdaq index.
$MU was last below the 50d moving average in early November, given the recent earnings results and the relative strength over the past couple weeks I’d be surprised if $MU retested the 50d.
This top chart is $MU with the daily candles but lets skip to the next one which looks more interesting.
From this chart below, not only can you see that $MU is on the verge of breaking out to new all time highs but if you draw a high from the highs in 2018 and connect that line to the highs in early 2021, it looks quite obvious that $MU is not only in a very bullish uptrend but that $106+ level looks very achievable — once $MU gets to that HH trendline we’ll need to see it power through and close above otherwise the sellers will come in and $MU might need to correct and take a breather before challenging those ATHs again.
The bottom line on $MU is that business is currently very strong thanks to high demand and pricing power. The fundamentals look great and show no signs of letting up anytime soon. Basically with $MU you’re getting a company growing revenues by 30% and earnings by 30-40% yet the stock is trading at less than 10x NTM earnings — this is very cheap and I believe $MU will benefit from some multiple expansion (aka rerating) in the coming quarters and they continue to impress investors with their results.
Not only is demand for their products strong but we’re seeing $MU come out with more advanced products to meet their customer’s needs/demands. 5G smartphones are certainly promising and advanced graphics cards are only becoming more and more popular as shown by NVIDIA’s ($NVDA) recent success.
I currently have a 4% position in $MU — for the past month I’ve been transitioning my portfolio to “profitable growth” stocks given what’s going on with interest rates, the Fed, quantitative easing (QE) and the overall rotation from growth to value. I will likely keep my $MU position unless the stock was unable to stay above the 50d SMA, in that case I’d probably be a seller and then would buy back my shares once the technicals improved.
I hope you enjoyed this deep dive writeup on $MU — if you have any thoughts, comments or questions please feel free to reach out. In terms of semiconductor companies, I also have positions in $AMD and $NVDA so there are many other ways to invest in this growing industry. Another option would be the semiconductor ETFs such as $SMH or $SOXX.
Disclaimer: The stocks mentioned in this newsletter are not intended to be construed as buy recommendations and should not be interpreted as investment advice. Many of the stocks mentioned in my newsletter have smaller market capitalizations and therefore can be more volatile and should be considered more risky. I encourage everyone to do their own research and due diligence before buying any stocks mentioned in my newsletters. Please manage your portfolio and position sizes in accordance with your own risk tolerance and investment objectives.