To see the future of wearables, I said look to where jewelry has been worn. The places where people trust metal against skin is where compute inevitably will reside. The categories are easily found. Head/Hair -> AirPods/Beats Arms -> Watch Hands -> Rings Ear -> AirPods Neck -> Beats My prediction has been more or less fulfilled. Today we wear Beats, AirPods, Apple Watch and (though not Apple) rings containing microprocessors. However, what is missing from this categorization is eyewear. In How We Got to Now, Steven Johnson describes the reading lenses of medieval Europe as the first piece of technology “since the invention of clothing in Neolithic times, that ordinary people would regularly wear on their bodies.” →
Stay tuned for more on this topic in the coming weeks.
Nota Bene
Apple and Broadcom extend deal through 2031 → Broadcom and Apple extended their custom-chip agreement through 2031, and Broadcom jumped nearly 4 percent on the news. Apple is around a fifth of Broadcom’s revenue, so this is a lock-in that cuts both ways. Even as Apple designs more of its own silicon, including the C1 modem, it still leans on Broadcom for wireless and RF. The pattern is familiar: secure the supply years ahead, then decide later what to bring in-house. →
Apple destroyed the mid-tier watch market. Now it’s coming for the $200 billion eyewear industry → When Apple entered watches, the mid-tier collapsed. Swatch is down 28 percent and Fossil roughly 70 percent, and Apple now outsells Rolex by revenue. The smart glasses category already surged 167 percent, and Apple is reportedly aiming at late 2027. Eyewear is a $200 billion industry that has never faced a competitor who ships hundreds of millions of wrist computers a year. →
Office Hours
Did a lottery winner become “the company of the future”? → The Wall Street Journal put Nvidia first among companies of the future and Apple twelfth. But Nvidia’s history is a flat line and then a near-vertical one, because the matrix-multiplication chip they already made turned out to be perfect for AI. That is winning the lottery, not building a moat. They have maybe five customers, every one desperate to replace them, while Apple ships hundreds of millions of chips a year; ask where Nvidia is in a decade. →
The portability superpower → Someone asked why MLX lets you write for Apple Silicon and then move to Nvidia, a reverse Trojan horse. It fits a very old pattern. Apple has sweated portability since the Mach kernel and the PowerPC-to-Intel jump, when moving the entire OS looked impossible to everyone but them. Doing it at the level of AI chipsets is nothing new; the same instinct is why a touch-ready Mac, when it finally ships, will feel anticlimactic. →
An Office Hours question asked by Fred S, June 9, 2026.
Q: With MLX, you can write for Apple Silicon and then migrate to Nvidia’s chips — a kind of reverse Trojan horse, since people usually build tools to migrate the other way. What’s going on there?
I’m not an expert on MLX specifically, but it fits a very old pattern. When you look at Apple’s history with portability, they are extraordinarily good at it. Months ago, around the Apple-50 commemoration, I was watching the Computer History Museum interviews with the old hands — Avie Tevanian, Mike Markkula, Bertrand Serlet. Bertrand described the transition from PowerPC to Intel, way before the M series. When it shipped, people were shocked. Nobody could imagine that an OS could be migrated so quickly and so seamlessly
I met Bertrand at WWDC after he gave that talk. He called the old system a skyscraper, and described how they had to move the skyscraper. I told him it was impressive, and he was almost dismissive: this is nothing, we’ve been sweating portability for decades. That came from his background before NeXT, working on the Mach kernel and the portability of Unix, which became the heart of OS X.
So you’d have to go back decades to see how entrenched this is in Apple’s DNA: the instinct to avoid becoming dependent on an architecture they don’t control, and to be clever about the migration path. When Apple bought into Unix, what they got was portability, the ability to scale up and down from server to device, and security, because it was a protected kernel with protected processes. Microsoft never prioritized any of that, so their OS became a kludge that got patched ad infinitum.
That’s why this looks easy for them now. Doing it at the level of AI chipsets is not surprising in the least. By the way, the same engineering shows up in running macOS on an iPad as a touch display — they’ve quietly made the entire Mac interface touch-ready. The day a Mac with a touch UI arrives, it’ll be anticlimactic. They probably built it years ago and are just waiting to turn it on.
Apple plans to disrupt the $200B eyewear market the way it disrupted watches. Swatch fell 28%, Fossil 70%. Smart glasses are targeting late 2027.
When Apple launched the Apple Watch in 2015, the mid-tier wristwatch market had a handful of dominant companies. Swatch Group sold watches under Tissot, Hamilton, and Longines. Fossil Group sold under Michael Kors, Armani, and Kate Spade. Movado sold under Coach, Hugo Boss, and Tommy Hilfiger.
Ten years later, the damage is quantifiable. Swatch’s revenue is 28% lower in 2025 than it was in 2014. Fossil’s sales dropped roughly 70%. Apple became the world’s largest watchmaker by unit volume within a few years and last year overtook Rolex as the number one watch brand by revenue. The Apple Watch now generates an estimated $17 billion annually.1
NB: The global smart glasses category surged 167% from…
…a year earlier in the first quarter, shipping 2.25 million units worldwide, according to consultancy firm IDC. Meta led the pack with nearly 70% market share, followed by augmented reality-equipped glasses maker Shenzhen RayNeo Technology and Chinese consumer electronics maker Xiaomi, according to IDC.
The smart glasses category is tiny compared with the eyewear market ($0.5 billion vs. $200+ billion). Apple's entry will change that.
It's likely that the Apple product line will be called the Apple Vision line, with the Vision Pro holding the high end.
"By the end of the 16th cent. eyeglasses were as common in western and central Europe as desktop computers are in western developed countries today." Eyeglasses served an important technological function at both the intellectual and practical level, not only easing the textual studies of scholars but also easing the work of craftsmen/small bus. During the 15th cent. two crucial developments occurred: the ability to grind convex lenses for various levels of presbyopia and the ability to grind concave lenses for the correction of myopia. As a result, eyeglasses could be made almost to prescription by the early 17th cent. -Renaissance Vision from Spectacles to Telescopes, Vincent Ilardi, American Philosophical Society, 2007.
To see the future of wearables, I said look to where jewelry has been worn. The places where people trust metal against skin is where compute inevitably will reside.
My prediction has been more or less fulfilled. Today we wear Beats, AirPods, Apple Watch and (though not Apple) rings containing microprocessors. However, what is missing from this categorization is eyewear.
In How We Got to Now, Steven Johnson describes the reading lenses of medieval Europe as the first piece of technology “since the invention of clothing in Neolithic times, that ordinary people would regularly wear on their bodies.”
Though worn and labeled as apparel, glasses are treated as jewelry. Made from semi-precious, fine materials, premium branding and sold through bespoke retail, eyewear is granted a special status in our panoply of products.
Before we get into how Apple will fill this space, we should learn a bit about how the eyewear industry emerged and evolved.
Eyeglass frames often have luxury branding: Tiffany & Co., Armani, Brooks Brothers, Ferrari, Burberry, Tory Burch, Chanel, Jimmy Choo, Coach, Swarovski, Dolce & Gabbana, Michael Kors, Versace, Miu Miu, Prada, Ralph Lauren and Burberry are all eyewear brands. There are also eyewear-only brands which define their own luxury, such as Ray-Ban, Oakley, Persol, Vogue, and Oliver Peoples.
They are sold through a carefully segmented set of retailers such as Sunglass Hut, Óticas Carol, LensCrafters, Vision Express, Apollo, Pearle, Générale d’Optique, Target Optical, Pearle Vision, MasVisión, Mujosh, OPSM, Bolon, Oakley, GrandVision.
Though eyewear includes non-prescription glasses, prescription lenses are also branded. Lens brands include Crizal, Essilor, Eyezen, Kodak, Optifog, Transitions, Varilux, Xperio and Shamir.
It may seem like a diverse and broad value chain but all these brands, retailers and lenses are property of one company, EssilorLuxottica.
It is the result of the 2018 merger two major conglomerates: Essilor (French) focused on lenses and acquirer of 250 companies and Luxottica (Italian) owning a combination of frame factories, designer labels and thousands of retail outlets.
At time of merger, Essilor had around 45% of the prescription lenses market, and Luxottica 25% of the frames. Combined they sell close to a billion pairs of lenses and frames every year, through 17,740 stores with a workforce of more than 140,000. They have around 1.5 billion customers, mostly in the premium or high-end of the market.
Since the merger, EssilorLuxottica continued its acquisitions with a set of prescription laboratories in South America and Europe, optician groups in Europe and US and Heidelberg Engineering, a German maker of optical coherence tomography (OCT) diagnostic instruments, confocal microscopy, and scanning lasers.
In July 2024, EssilorLuxottica bought US streetwear brand Supreme, expanding its brand portfolio beyond eyewear and lenses for the first time.
It also acquired Israeli startup Nuance Hearing indicating that it may introduce hearing aid technology into its lenses.
It also owns EyeMed, one of the largest vision insurance plans in the US.
There is something here more than meets the eye.
A lenticular species
Almost everyone wears glasses at some point in their lives. In developed countries, the rule of thumb is that around 70% of adults need corrective lenses. In Japan it's 75%. Including sunglasses, the penetration of eyewear is probably close to 90%.
We are at an unprecedented moment in the story of human vision – namely, the accelerating degradation of our eyes. For several thousand years, human beings have been operating mostly with unaided vision. But that is coming to an end. No one is sure why but across the world, we are becoming a species that wears lenses.
There are two things happening. First, there’s a largely unreported global epidemic of myopia, or shortsightedness, which has doubled among young people in a single generation. Scientists once thought myopia was mainly genetic, but about 10 years ago, it became clear that children’s upbringing was also harming their eyesight. The effect is most severe in east Asia, where myopia has always been more common, but the rate of increase has been worldwide. In the 1950s, between 10% and 20% of Chinese people were shortsighted, but now, among teenagers and young adults, the proportion is more like 90%. In Seoul, 95% of 19-year-old men are myopic, many of them severely, and at risk of blindness later in life.
Across the developing world, a slower process is underway as populations age, urbanize, and move indoors for work.
In 2013 a paper forecast that half the world’s population will be myopic by 2050 – almost 5 billion people. In the course of a single generation, worldwide, researchers have recorded a rough doubling in the number of people who become short-sighted as children.
The main reason, as far as anyone can agree, is a reduction in the time that they spend outdoors. Sunlight helps to moderate levels of dopamine, which in turn influence the development of the eye. Too much dopamine means eyeballs grow too long, and become oblong in shape, focusing light in front of the retina, rather than on it.
Eyewear history shows people don’t start wearing glasses when everything seems out of focus. Instead, they wear them to participate in new forms of entertainment and labor. The mass market for spectacles didn’t emerge when they were invented in 13th-century Italy, but 200 years later in Germany, alongside the exploding availability of printed works, because people wanted to read. The printing press created a new job-to-be-done, reading, for which glasses were ready to be hired.
Scale and Style
In 2012, Essilor first publicized the 2.5 billion statistic, highlighting the shift from outdoor living to indoor usage and the resulting demand for eyewear.
EssilorLuxottica, with its own designers, engineers, factories, supply depots, and retail outlets, currently has almost 18,000 stores and contracts with 100,000 opticians worldwide. This allows it to bring products to market faster and in greater quantities than its rivals, while also retaining a larger proportion of its profits.
Considering various colors and face shapes, EssilorLuxottica produces about 27,000 models. Its plants produce 400,000 frames daily.
During the 20th century, the eyewear business transformed a physical impairment into an opportunity to display style. In doing so, optical retailers learned that for something that costs only about $15 to make, customers are happy to pay 10 or 20 times that amount. The gross margins are 60% to 67% for high-volume, online, or budget retail, 67% to 75% for mid-range/acetate) and 75% to 85%+ for branded and premium materials.
History
The conditions that have allowed for the rise of Essilor and Luxottica are rooted in the way spectacles were sold. Until the end of the 19th century, you could buy a cheap pair of glasses – for reading or for distance – off the rack in any department store, or from a jewelry shop, or a guy in the street. Eyewear was a craft and unregulated. It was buyer beware. Eyeglass sellers in the 18th and 19th centuries were notorious for scams and faulty lenses.
The birth of the optometry profession around 1900 changed this. A new profession of respectable spectacle-sellers, similar to pharmacists, sought to standardize eye tests and restrict the sale of glasses to licensed professionals only. Their goal was to raise standards. However, making cheap, widely available product accessible only to a few authorized sellers increased prices, and the potential for new form of corruption.
The first opticians were disdained by ophthalmologists – proper doctors of the eye, who trained in hospitals and considered selling glasses as an unskilled trade. The first optometry course in the US was taught at Columbia University’s physics department because it was not allowed inside the medical school.
But the new professionals persevered and found new ways to protect their jobs. Across Europe and in the US, optometry laws and regulations were passed to control the prescription and selling of eyewear. Many of these had a quality of health assurance aspect, but they also had the effect of creating a highly protected marketplace. For instance, opticians fought against advertising which might reveal prices and allow comparison shopping.
The scams and bribes became legend. Lens and frame makers bribed practitioners to prescribe their products and investigations and convictions forced divestitures and value chain break-ups.
Into this space the components of EssilorLuxottica stepped in and began to consolidate the market. Luxottica has spent the last 30 years acquiring prominent elements of the optical business, including frames, brands, and retail chains. Essilor, has focused on acquiring lens manufacturers, instrument makers, prescription labs, and the science of sight.
The strategy for EssilorLuxottica is to obtain the power to decide how billions of people will see and what they pay for it. The company can screen populations and provide eyewear across the globe all the while controlling supply, prices, and margin.
New technologies such as automated optometry kiosks and online retailing that threaten independent opticians are sustaining to EssilorLuxottica and thus they are investing and acquiring.
Today humanity is watching the world through screens, in front of them, but not on their eyes. It seems only a matter of time before those screens become lenses. The same lenses that are now an essential product for almost everyone.
One more thing: Exactly one year ago, in July 2025, Meta acquired a minority stake in EssilorLuxottica with about 3% of the equity as part of its partnership with Ray-Ban. Bloomberg said that Meta had bought a stake worth around 3 billion euros ($3.5 billion) in EssilorLuxottica and is considering further investments that could build its holding to around 5% over time.
Meta CEO Mark Zuckerberg called it a "symbolic" gesture to cement their long-term partnership.
Broadcom, Apple extend chip supply deal through 2031
July 6, 2026 8:07 AM
July 6 (Reuters) - Broadcom said on Monday it has agreed to expand its partnership with Apple through 2031 to develop and supply a range of custom chips, sending the shares of the chipmaker up nearly 4% in premarket trading.
Broadcom has been a supplying key components to Apple for a very long time, including custom radio frequency chips used in iPhones, Wi-Fi and Bluetooth connectivity chips and other networking semiconductors.
Apple is one of Broadcom's largest customers. The iPhone maker accounts for about 20% of its annual revenue, according to analysts, making it a critical contributor to the chipmaker's business.
The extended partnership reinforces Apple's strategy of locking in long-term supply agreements with key chipmakers to bolster the resilience of its supply chain.
In 2023, the companies announced a multibillion-dollar agreement for Broadcom to develop and manufacture 5G radio frequency components.
The boom in inference - the process by which models respond to user queries - has made custom chips crucial, increasing the orders for advanced processors and intensifying competition.
While Apple has increasingly designed its own processors and, more recently, its C1 cellular modem, it continues to rely on Broadcom for key wireless connectivity and radio frequency components.
Q: The Wall Street Journal just ranked Nvidia the top “company of the future,” with Apple down at number 12. What’s going on with Apple’s brand image — and is that ranking right?
Look at Nvidia’s history. For most of it, revenue was flat. Then, in the last few years, it wasn’t. We usually picture growth as a hockey stick — really a smooth, almost exponential curve. Nvidia isn’t that. It’s a flat line, and then another line at a very steep angle. Zero to one. It isn’t even a curve; it’s two lines.
Compare it to everyone else. Meta has a nice curve. Microsoft and Google look brilliant. Apple, apart for seasonal spikes, is similar, and so is Amazon. But Nvidia is a step function. That's because they got lucky. The chip they’d been making for years — a graphics accelerator, a matrix-multiplication machine that does one thing, scaled up — turned out to be useful first for crypto, and then, just as suddenly, perfect for AI, which is nothing more than linear algebra. They won the lottery.
Now, they’re smart, and they didn’t squander it. They’re building ecosystems and platforms on top. I’m not against Nvidia. But to call a lottery winner the “best company of the future” says nothing about sustainability, processes, priorities, or scale. How many customers does Nvidia have? Five, maybe — the big hyperscalers, every one of which is desperate to fire them. How many chips does it ship? Tens of millions, each costing tens of thousands. Very profitable. But Apple ships hundreds of millions of A-series chips a year, and more. The silicon Apple puts into people’s hands is orders of magnitude more prolific.
If Apple’s roadmap bakes-in exactly what AI needs then where is Nvidia in a decade? Maybe in the PC business. Maybe the next Intel. The premise that they have an unassailable advantage, what Buffett calls a moat and the strategy people call sustainable competitive advantage, is exactly what I’m not convinced of. IBM in its heyday had far more diversity. Nvidia, to me, is a very good company, but a very lucky one, comparable to Qualcomm. And building chips was never a foundation for becoming the greatest company in the world forever.
A commodity is a product whose price does not rise when performance improves, so businesses trading in commodities compete on production and distribution rather than design. That is why Apple does not build memory: it would lock itself into its own cost structure and give up the flexibility to negotiate across vendors. Memory makers can only make money by lowering cost per bit, which they do by shrinking process nodes. Beijing spent hundreds of billions over more than a decade building strategic memory suppliers; twelve years in, that industry may only be reaching profitability because of the current imbalance. →
Nota Bene
Apple's Q1 iPhone sales in Africa grew 44% → Counterpoint Research reports Africa's smartphone shipments fell 12% in Q1 2026, hit by fuel prices, memory-driven price rises, and extreme price sensitivity. iPhone bucked the trend and grew 44%. Apple does not adapt to markets; markets evolve to become Apple markets. →
Counterpoint: Apple to Hit Record Market Shares in Smartphones, Laptops, Tablets; Watch Share Also Set to Grow → Counterpoint expects Apple to reach its highest-ever share in smartphones near 25%, plus record share in laptops and tablets in 2026, while most OEMs face double-digit declines. Stable pricing, high margins, and supply chain control let Apple weather the memory crisis best. The article predates the June 25 price increase, but the conclusion still holds. →
Office Hours Q&A
Earnings are an opinion, cash flow is a fact → Dave E. asked whether the hyperscalers face an AI bust and how Apple would adjust. There is revenue but little profit or free cash flow because everything is being reinvested. The bubble bursts only if debt gets mispriced, and so far AI has not become a debt story. →
Why is betting against Apple the extraordinary claim? → An Office Hours question about Apple's forward prospects given 26.6% CAGR over twenty years. Analysts make one-year calls swamped by short-term noise, while the CAGR reflects a decades-long trend. The extraordinary claim is that a flywheel with this much momentum suddenly stops. →
Will agents kill the app economy? → Question was: whether agentic AI would consolidate revenue to Apple and agent providers. Apple keeps trusted permissions inside apps: apps become callable APIs, Apple still hands off credentials, and the app economy lives on alongside intent, touch, and vision. →
An Office Hours question asked by Dave Emery, June 9, 2026.
Q: Will agentic AI replace the vibrant app economy, consolidating revenue to Apple and the agent providers? Or will it enable a new set of apps?
The premise of intent as an operating system is that you assemble the solution to a complex problem out of many modules: models, private data, awareness of the user’s context, and many kinds of input. The orchestrator assembles that solution and decides whether to hire a module provided by an app. Most people who assume intent computing also assume a post-app future. Apple doesn’t. On the outer ring of their "orchestrator" model of computing, they put either Siri or apps to deliver the experience.
Here’s the key. Within apps lie the permissions we entrust. Your banking credentials live in your banking app, because you trust the institution and it has earned that trust. The "Claw" theory says: just harvest all of it. Find the bank code, the social-security number, the mother’s maiden name, hand it to an orchestrator, and let it act as your agent. Apple says no. They don’t want the things that have earned your trust to be disintermediated, especially when something goes wrong. If you get hacked, it wasn’t Apple’s OS that lost your keys.
So what apps used to provide was an interface, plus data, plus a trusted relationship with the provider. What intent computing may take away is the interface — which was the costliest part of building an app anyway, with millions of developers recreating the same drop-downs and authentication steps. The app becomes a callable API. That looks like a demotion, but Apple will still invoke the app, hand off the credentials, and often let the app deliver the final experience.
This is disruption theory: as a value chain modularizes, profit capture migrates — what Clay called evaporation and condensation at different points in the chain. But branching isn’t extinction. Just because something evolves and forks doesn’t mean the old form dies; that’s why there are still apes and humans even though one evolved from the other. We’re going to have intent, and touch, and vision, all in parallel. The old way of invoking an app directly can live on for fifty years. Yes, there’s still money in the app economy. There’s much more than that, too.
While most of the other OEMs are struggling, Apple, with its stable pricing, high margins and strong supply chain control, is in the best position to weather the memory crisis and win market share…
While most of the other OEMs are struggling, Apple, with its stable pricing, high margins and strong supply chain control, is in the best position to weather the memory crisis and win market share across key segments.
Apple’s share increases in multiple categories also reflect the increasing strength of its ecosystem, with more premium users skewing towards Apple.
“This year defines Apple’s unique resilience within the industry. The multi-segment market share gains also mean a compounding impact on the company’s ecosystem strength. Apple is still gaining new users across categories, and it will become increasingly difficult for competitors to make an Apple user switch to their operating system.”
Apple’s iPhone shipments are likely to remain flat in 2026, while most of the other key OEMs will see double-digit declines. OEMs have to resort to price increases to sustain margins, but Apple has not raised iPhone prices yet. Premium segments have higher margins and can absorb more price shocks. At the same time, affluent consumers are less affected by price increases. Moreover, the iPhone 18 Pro series is expected to perform better than the iPhone 17 Pro series, and the iPhone 17 model is likely to continue its strong run in the absence of the iPhone 18 in 2026. As a result, Apple will reach its highest-ever share in the smartphone market in 2026 at around 25%.
The overall tablet market will also decline in 2026 due to inventory correction and memory shortages, but Apple is expected to grow, driven by an ongoing refresh cycle across its iPad portfolio and continued demand from emerging markets. Following the launch of the M5-powered iPad Pro in late 2025 and the iPad Air refresh earlier this year, Apple is expected to further strengthen its lineup with new base iPad and iPad mini models during the remainder of 2026. This broad refresh cycle should help Apple maintain consumer interest across multiple price tiers and better withstand the demand pressures affecting the wider tablet market, supporting further market-share gains.
Within the laptop segment, the MacBook Neo’s release will drive growth momentum in favor of Apple in 2026. Apple’s Mac shipments in 2026 will grow 23% YoY, while the overall laptop market will decline 11%. Compared to other OEMs, Apple will gain the highest percentage points of market share in the laptop segment in 2026. Within iPad and Mac as well, Apple’s share will reach its highest ever in 2026.
Smartwatches are less impacted by the memory crisis. The segment is expected to show 1% growth in 2026, with Apple growing faster than the overall segment, driven by full-year availability of the SE 3 and Watch Ultra 3 models. 2025 saw Apple’s first comprehensive portfolio refresh after 2022.
Another factor is the lowering of entry points in Apple’s ecosystem. Apple has already introduced affordable iPhone (e), watch (SE 3) and now Mac with the MacBook Neo. The company clearly wants to make its products accessible to the larger consumer base, and this is working more in its favor now as other OEMs increase prices in some segments driven by the surge in memory prices.
Apple’s share increases in multiple categories also reflect the increasing strength of its ecosystem, with more premium users skewing towards Apple. This will also mean a strong upgrade cycle for all these product categories in the future. Added to this is Apple Intelligence. At WWDC 2026, Apple showcased some strong use cases, especially for Apple Intelligence-backed Siri AI. If Apple delivers and evolves its AI landscape, it will act as a strong stickiness factor for the entire ecosystem.
Commenting on the role of Apple Intelligence, Principal Analyst Varun Mishra noted, “Apple holds the most important key, the user touchpoint across the most used devices on the planet. If delivered, a personalized Apple Intelligence will act as a strong ecosystem differentiator and unlock a new multi-device upgrade cycle for Apple.”
NB: This article was written before the price increase announced on the 25th of June. So pricing is not as "stable" as the article implies. Nevertheless, the conclusions are likely correct.
Share will grow as the effect of memory constraints will be far greater on competitors. To wit:
GoPro warned this month that it might go out of business after memory costs shot up between 80% and 115% at the end of the first quarter. And shares of speaker maker Sonos are down 23% this year as memory prices pressure margins."
Footnote: I remember when GoPro and Sonos were paraded as fantastic innovations that Apple could not hope to achieve.
More notable is that even with these high unit market shares, the share of users is significantly higher. Increasing the unit share will gradually translate in a more resilient user share in the long term. Users lead to increasing ecosystem, services and loyalty.
Q: You recently calculated Apple’s CAGR at an impressive 26.6% over the last twenty years. Without giving stock advice, what do you make of Apple’s forward prospects?
The point of my article was that more than half of analysts have a hold, a sell, or no opinion on Apple, and I found that surprising. I aggregated the ratings myself, and it settled to about fifty firms actively covering, split roughly half Buy, half Don't Buy. A “hold” says the stock is fairly priced, and if a company is growing twenty percent a year, it doubles its revenue in three to four years. So if that’s the growth, why would you not buy?
Computing the CAGR from each year forward to the present, over twenty years shows the curves are volatile early but then settle around twenty to twenty-five percent. So why are these analysts so cautious? Because they’re making one-year calls, swamped by near-term noise, while the CAGR is a multi-year return: lots of short-term volatility, lots of steady long-term growth.
I’m only dumb enough to assume that twenty years of history is long enough to see a pattern. I know history doesn’t guarantee the future. But we could have had this same conversation every day for the last fifteen years, and the flywheel just kept spinning and it also kept getting more massive. It’s actually extraordinary to assume that a massive flywheel with this much momentum suddenly stops. That’s, for me, the harder thing to argue than that it won't stop.
You can do the same with GDP. Countries compound at two to three percent for two hundred years, and the most likely future is more of the same, not utopia or catastrophe. Extrapolating a long-term trend a few years out is not an extraordinary claim. The extraordinary claim is that it stops: that Apple loses its place as the interface to the most valuable devices in the world, that customer satisfaction evaporates, that all the threats we’ve heard a hundred times finally land. I’ve heard them before. They never happened. So why should I believe them now?
"I’ve never seen a shortage without a subsequent glut, but I’ve seen gluts never followed by a shortage." -Nassim Taleb
A commodity is a product or service which does not increase in price if its performance increases.
In a typical commodity exchange (coffee, pork bellies, iron ore, oil, etc) an offer of a "better" batch would not result in a higher price. The market treats all instances as equivalent with no regard to who produced them or how they were produced.
That does not mean that commodities prices cannot rise. Although not determined by quality or performance or sourcing, commodities respond to supply/demand imbalances. These are unpredictable and are affected by externalities and multitudes of factors.
Because performance is not differentiable, businesses trading in commodities are not responsive to innovation in the product. They may benefit from production and distribution innovation however. This is why scale, consolidation, regulation and legislative action are primary drivers of value.
That is also why Apple doesn’t design or build memory. Pricing will oscillate but, as a producer, Apple would be locked into their own cost structure and wouldn’t be able to negotiate with multiple vendors, nor have flexible responses to surges or drops in demand.
As a commodity, memory chips from any manufacturer are effectively interchangeable and pricing is not determined by producer. Therefore, the only way memory makers can make money is by lowering their own cost, which they do by increasing density (bits per wafer.)
The design isn’t the hard part or the limiting function for density and thus cost for DRAM or NAND chips. As Joe Lion pointed out in a Mastodon thread:
By far, the single most important thing for increasing bits per wafer is shrinking the process nodes which decreases cell size, where a DRAM cell is "1 transistor, 1 capacitor."
The design of the DRAM cell is the same as laid out in Robert Dennard's original DRAM patent from 1968. The design of the interface and periphery circuits, while important for performance, are nearly insignificant for decreasing overall die-size, which is, again, the critical factor in $ per bit.
Lion continues,
This is different than CPU/GPU/logic chips, which are not a commodity, and therefore pricing can be set by the manufacturer based on whatever the manufacturing cost happens to be, plus their desired gross margin.
To put another way, TSMC and Intel have never sold a CPU chip for a loss because they set the pricing themselves, and they will never price their own product below their own manufacturing cost, whereas memory makers frequently go through years-long cycles of selling chips for a loss because memory prices are set by global supply and demand.
Into this emerged a political dimension. The Chinese government declared memory as a strategic commodity and spent hundreds of billions of dollars developing its own memory suppliers over more than a decade. These investments take many years to bear fruit. A new fab would take five years before the first wafer, 10 years before high volume yield, and 15 years or more before cost competitiveness with incumbent DRAM vendors. It may not ever become profitable. China is 12 years in and its state-sponsored memory industry may only be reaching profitability due to the market imbalance just experienced.
The current situation is a demand shock. There isn't a shortage due to a loss of capacity but a surge in demand fed by a rush to spend on AI infrastructure. The buyers of memory are in a race to secure capacity in the belief that it's either first to reach some critical mass or they assume a logic of winner take all. [The fact that there are open source alternatives that are nearly the same performance does not seem to deter investors.]
Price insensitive demand is causing a windfall for producers who are effectively feeling as if they are in control and will remain so forever. But at the Taleb quote in the heading suggests, all this will pass. There has never been a situation where a commodity shortage has resulted in a perpetual undersupply (unless a cartel forms--and even that is hard to keep going indefinitely.]
The reason commodities balance in price is by their definition. Unless memories can be differentiated on performance, their price will float. If their price will float, high prices will attract investment and thus production and thus a glut and prices will fall.
Will Apple and indeed the entire set of industries using memories endure this shock? You would have to believe that unlike every other time, this time it's different.
