Samsung Profits When the iPhone Wins. It Built Its Whole Company to Make Sure of It.

Hold up the phone Samsung wants you not to buy, and look at the screen. If it's a premium iPhone, there's a good chance the glass lighting up in your hand was made by Samsung — the same company whose Galaxy phone sits across the aisle in the store, trying to beat it.⁷ This is not an accident, and it is not a truce. It is the entire design. Samsung built a company that gets paid when its rivals win, so that losing the phone war and winning the phone war pay almost the same.

The official story is that Samsung is Apple's bitter handset rival that also, awkwardly, happens to supply it — an uneasy 'frenemy' arrangement forced by circumstance. That gets it backwards. The supplying came first. Samsung was making the parts inside other people's devices for decades before it cared who won the phone, and the rivalry is the recent, optional layer on top of a much older, deliberate machine.

The factory was built before the rivalry existed

Samsung Electronics — incorporated in 1969, a separate legal entity from the old trading house, headquartered in Suwon⁴ — did not back into components. It chose them, early and hard. It entered semiconductor production in 1974 by acquiring Hankook Semiconductor, and by 1992 it was the world's leading manufacturer of memory chips.⁵ By 2004 it had developed the world's first 8 GB NAND flash chip.⁵ None of that was about smartphones, because smartphones did not exist. It was about owning the layers underneath whatever device came next.

And the structure underneath it was architectural, not improvised. Samsung's vertical-integration network was first created in Korea in the early 1970s and then replicated abroad in Mexico, Malaysia, and China — a three-tier stack with Samsung Electronics as the final assembler at the top, component makers in the middle, and material suppliers at the bottom.⁶ This is the part the 'frenemy' framing misses entirely. The same hands that bolt together a Galaxy also pour the materials, etch the chips, and grow the panels — and they will sell every one of those layers to anyone, including the company building the phone Samsung is trying to outsell.

How a rival's win becomes Samsung's revenue

Here is the mechanism, worked down to the cash. A smartphone is, to a component supplier, a bundle of high-margin parts that has to be bought from someone. Memory and a display are two of the most expensive, and Samsung sits at or near the top of both. So when a competitor sells a flagship phone, it is not purely a loss for Samsung's handset division — it is a purchase order for Samsung's components division. The rival has done the marketing, fought for the shelf space, and absorbed the risk of the launch; Samsung simply collects on the parts inside. The whole point of owning the layer beneath the product is that you get paid no matter whose logo goes on the front.

This is why the components business behaves like a structural subsidy rather than a sideline. Samsung Display still holds the largest share of Apple's OLED orders, even after Apple deliberately diversified to LG Display and BOE to avoid depending on a rival.⁷ The diversification is the tell: Apple tried to route around Samsung and still couldn't fully do it. And on the first foldable iPhone, Samsung has secured a reported three-year exclusive panel supply agreement — Apple accepting the terms because, with BOE's foldable panels deemed inadequate and LG Display lacking a proven foldable smartphone track record, no other viable supplier exists.⁹ When your rival cannot escape buying from you even when it actively wants to, you are not a supplier. You are a toll.

Look at where the money goes and the priorities stop being ambiguous. Of the KRW 53.6 trillion Samsung Electronics spent on capital in 2024, KRW 46.3 trillion went to the semiconductor division and KRW 4.8 trillion to displays.² The overwhelming share of the firm's investment flows into the layers it sells to everyone — not the phone it sells against everyone. A company spending like that is not a phone-maker with a component arm. It is a component empire with a phone attached, and the phone is partly there to be a guaranteed first customer for the factories.

Wasn't it just the Apple deal that launched all this?

The romantic version says Samsung's whole strategy crystallized in one handshake: a secret 2005 chip deal in which Samsung agreed to supply Apple and quietly powered the first iPhone. It's a tidy origin myth, and the primary record doesn't support its tidiness. Apple's own November 2005 press release describes a multi-vendor flash-memory agreement with five suppliers at once — Hynix, Intel, Micron, Samsung, and Toshiba — with Apple prepaying $1.25 billion across all of them.¹ Samsung was one of five, not the chosen one.

There is a kernel of truth underneath. Samsung's ex-CEO Hwang Chang-gyu recounts in his memoir proposing to bundle Samsung's mobile application processors into the Apple flash deal in exchange for lower flash prices — and the original 2007 iPhone did launch on a Samsung processor — a fact independently confirmed by hardware teardowns.⁸¹⁰ But notice what that story actually proves. The clever move wasn't winning an exclusive; it was using a commodity contract Samsung shared with four competitors as a doorway to sell a second, higher-value layer. That is the integration working exactly as designed. The deal didn't create the strategy. The strategy is what let Samsung turn an ordinary multi-vendor memory order into a foothold inside its future rival's flagship.

The honest objection: doesn't arming your rival end badly?

The fair counter is that this looks dangerously like feeding the enemy: every panel sold to Apple sharpens the weapon aimed at Galaxy, and a supplier funding its own competitor is a textbook strategic blunder. It's a real risk, and it cuts both ways — the customer hates the dependency too, which is precisely why Apple spent years cultivating LG Display and BOE to dilute Samsung's grip.⁷ So the position is not unconditional. It holds only as long as Samsung's layers stay genuinely the best available, which is what that KRW 46.3 trillion of annual semiconductor investment is really buying: not just capacity, but the lead that makes a rival keep ordering against its own interest.²

And the moat is not eternal. Gartner's final 2024 rankings show Samsung retaining No. 2 in semiconductor revenue, with NVIDIA vaulting from third to first for the first time as AI demand reshaped the chip market.³ Being the parts-maker only pays while you make the parts everyone needs — and the parts everyone needs can change underneath you. The structural subsidy is powerful, but it is a lead that must be continuously repurchased, not a rent that collects itself.

Samsung made a quiet bet decades before the smartphone existed: that it was better to own the road every device drives on than to bet everything on a single device. So it built the factories, the panels, the chips, and the three-tier stack underneath them all — and then, almost as an afterthought, built a phone to compete at the surface. The phone might win or lose; the surface is volatile, the brand fragile, the war never-ending. But the layer beneath it gets sold either way. Samsung didn't set out to beat the iPhone. It set out to get paid for it — and most years, that is the better trade.