After the Global Factory
How Apple mastered global manufacturing—and why that era is ending
The iPhone in your pocket embodies a manufacturing miracle that's about to die. In his essential book Apple in China, Patrick McGee reveals how Apple achieved what seemed impossible: outsourcing production whilst maintaining absolute control over its value chain. Through meticulous reporting from factory floors and boardrooms, Patrick shows how Apple kept design, knowledge, and brand—the parts that appreciate—whilst leaving factories and workers—the parts that depreciate—to others. The model generated what long appeared as limitless returns, explaining how Apple once became the most valuable company in the world.
Now it's unravelling. Not because Apple failed, but because the conditions Patrick documents—China's unique manufacturing ecosystem, unfettered global trade, and Apple's relentless design innovation—are all disappearing at once.
This unravelling fits what I call 'late-cycle investment theory'—the pattern whereby techno-economic paradigms reach maturity and existing arrangements break down.
Apple built its manufacturing miracle during the maturity phase of oil, automobiles, and mass production, cleverly using computing and networks to optimise a system rooted in the previous paradigm. Now, as computing and networks itself reaches maturity, the global institutions that enabled Apple's model—unfettered trade, stable supply chains, China's unique ecosystem—are collapsing. The conditions that made the iPhone possible emerged from one paradigm's twilight; they're disappearing as another reaches its own.
Now, the rush to “bring manufacturing home” promises resilience but delivers something darker: the permanent destruction of living standards as we shatter the very systems that made modern prosperity possible. Understanding why requires grasping how manufacturing really works—not through full ownership, but through controlling the right pressure points at the right time. Patrick’s reporting provides the foundation; what follows explores where this story leads next.
1/ Henry Ford once believed that owning every step of production was the key to industrial success.
The Ford Motor Company’s River Rouge Complex was a monument to industrial ambition. Raw materials entered one end—iron ore, coal, rubber—and finished Model T automobiles rolled out the other. Ford owned the mines, the plantations, the ships, the steel mills, even the power plant. It was the ultimate expression of vertical integration: control everything, depend on no one.
But by 1927, this marvel of integration was crippling the company. The problem wasn't efficiency—River Rouge was stunningly efficient. The problem was that it could only produce one thing, namely the Ford Model T. When consumers began to demand variety—different colours, models, features—Ford's tightly integrated system became a straitjacket. Every change meant reconfiguring the entire chain.
Alfred Sloan, General Motors's CEO, saw the opening. Instead of owning everything, Sloan decided GM should focus on what mattered most: brands and final assembly. Let suppliers compete to provide components. Let dealers handle sales. Keep only the parts that give real advantage—design, marketing, and putting it all together.
The transformation was methodical. GM sold off its steel mills but kept design studios. It spun off parts suppliers but retained final assembly. It franchised dealerships but controlled the dealer network. By 1935, GM had shed most of its value chain whilst keeping final assembly and key strategic functions—brand development, product planning, and quality standards.
It worked brilliantly. While Ford struggled to adapt its monolithic system, GM offered “a car for every purse and purpose.” More importantly, GM's return on equity soared. Without the capital burden of steel mills and mines, the company could invest in what generated the highest returns: brands that commanded premium pricing and assembly operations that ensured quality control.
This battle revealed a fundamental truth about manufacturing: owning everything is often a fool's game. The smartest players control just the right parts. But which parts? And how do you maintain control without owning them?
2/ To thrive, manufacturers don’t need to own everything—just the points that matter most.
From studying how industries evolve, I've developed what could be called the strategic control theory (otherwise known as ‘the grip’): successful companies need command positions at critical junctures in the value chain—but not the whole thing. Think of it like controlling a river. You don't need to own every tributary. You need the high ground where streams converge.
GM discovered this by accident. By gradually shedding everything except brands (which create customer pull) and assembly (which determines final quality), they could influence the entire chain without owning it. Suppliers had to meet GM's specifications or lose access to GM's volume. Dealers had to follow GM's rules or forfeit their franchises. The company exercised control through strategic command positions, not comprehensive ownership.
This control creates what Microsoft's Tren Griffin, drawing from cable mogul John Malone, calls “wholesale transfer pricing power” (WTPP)—the ability to extract profit from others in the value chain by controlling either demand or access to something they need.
Malone made himself rich by controlling downstream cable distribution. He would tell new channels, “I'll give you access to my subscribers, but I want equity in your company.” Upstream, he insisted on multiple suppliers for set-top boxes, never letting any one vendor gain pricing power over him. The rule: never depend on a single source for anything crucial.
Manufacturers face the same dynamic. With control at key junctures, they can capture most of the profit pool. Without it, critical suppliers or distributors shift the balance of power and claim a larger share of industry profits.
But maintaining strategic control becomes meaningless if the underlying economics don't work—and manufacturing presents a fundamental challenge that has destroyed countless companies who thought they had found the right control positions. The question becomes: which positions grant real control, and which merely create the illusion of it?
3/ Making things is hard enough—but making money from making things is even harder.
Manufacturing presents a fundamental mathematical challenge. Factory operations always exhibit diminishing returns. Build a plant, and you're making a large bet on future demand. If demand falls short, you're stuck with costly idle capacity. During the Great Depression, Ford's integrated factories became financial tombstones.
Successful manufacturers overcome this by pairing activities that show diminishing returns with others that generate increasing returns. Think of this as the Northern Side versus the Southern Side of manufacturing economics.
The Northern Side exhibits diminishing marginal returns. Each additional factory costs more to build as you compete for prime locations. Labour costs rise not just because wages increase, but because managing larger workforces brings higher marginal costs—greater complexity, higher turnover, increased propensity to unionise. Beyond a certain point, each new facility adds less value than the last, especially when production capacity overlaps with existing plants.
The Southern Side, on the other hand, can show increasing returns. Brand-building scales beautifully—the stronger your brand, the more impact each marketing pound has. Process knowledge compounds—each manufacturing innovation makes the next breakthrough more likely. Distribution networks create positive feedback loops—more retailers want to carry popular products, whilst popular products become easier to distribute.
To judge overall returns, you subtract the Northern Side’s costs from the Southern Side’s benefits. If gains from brand strength, accumulated knowledge, or market reach grow faster than the costs of facilities and operational complexity, total returns keep rising. The Southern Side compensates for the Northern Side when increasing-return activities outweigh the drag of capital and overhead.
Consider how this works across industries:
Luxury brands like Louis Vuitton operate flagship stores in high-rent locations that obviously show diminishing returns—each store costs more whilst reaching fewer new customers. But exclusive distribution strengthens the brand, creating increasing returns through growing desirability. Each store doesn't just sell handbags; it reinforces scarcity and exclusivity, enabling premium pricing worldwide. The brand appreciation more than offsets the distribution costs.
Pharmaceutical companies follow a similar logic. Manufacturing facilities are costly burdens with strict regulatory requirements and high fixed costs. But each successful drug creates increasing returns through patent protection, accumulated clinical knowledge, and physician relationships. The intellectual property and market position justify the manufacturing overhead.
Intel dominated semiconductors for decades by pairing the Northern Side of fabs (which cost billions and depreciate rapidly) with the Southern Side of design expertise and industry standards. Each generation of chips reinforced Intel's position as the essential supplier, generating returns that justified the massive capital requirements.
This approach fails, however, when brands lack sufficient increasing returns to compensate. Many direct-to-consumer startups discovered this by copying luxury distribution models without the brand strength to support them. Nike made a similar error recently by shifting to direct sales and pulling back from retail partners, only to learn that those retail relationships delivered better overall returns than owning distribution.
This dynamic—finding the right economic balance whilst securing strategic control—becomes even more critical during late-cycle phases. As I've argued in my work on late-cycle investment theory, when technological paradigms mature, innovation loses its protective power and strategy decides winners.
The key insight is that sustainable manufacturing requires more than strategic control: it requires pairing diminishing-return production with increasing-return activities that justify the costs. This dynamic (finding the right economic balance in addition to securing strategic control) determines who has WTPP in the value chain. When your increasing-return activities (brands, knowledge) grow faster than your diminishing-return costs, you gain leverage to extract profits from suppliers and distributors. Without this advantage, they extract profits from you.
The difficulty of finding the right pairing, one that comes with both strategic control and the right return trajectory, explains why some companies thrive whilst others, making identical products in identical factories, struggle to survive. In particular, it explains why the PC industry's evolution would prove so catastrophic for Western manufacturers.
4/ Western PC makers once thought they could outsource the hard parts and keep the profits—but they were wrong.
The PC industry offers the perfect cautionary tale of what happens when manufacturers lose their strategic control positions. As production moved to Taiwan and China in the 1980s and 1990s, Western brands like Dell, HP, and Compaq thought they were being clever—letting Asian partners handle the messy, capital-heavy work whilst they focused on design and marketing. Initially, this worked. Dell's direct-sales model and just-in-time manufacturing made it a Wall Street darling.
But over time, Asian suppliers commoditised everything. Taiwan's electronics manufacturers developed standard motherboard designs that any brand could use, supported by government-backed industrial clusters where suppliers could achieve economies of scale impossible for scattered Western manufacturers. Chinese assembly operations offered identical capabilities to all buyers, treating standardisation as a competitive advantage rather than a compromise.
Memory, processors, and graphics cards became interchangeable components available to anyone with a purchase order. Western PC brands ended up selling near-identical boxes because no supplier would customise for just one buyer when standardisation offered such compelling economies of scale.
In other words, the PC brands had lost their strategic control. Without command over key differentiation points, competition became purely about price and distribution. Dell's direct model became a temporary advantage rather than a sustainable moat. HP and Compaq merged in desperation. Gateway disappeared entirely. Only those who maintained control positions—Intel with processors, Microsoft with operating systems—captured lasting value from the PC boom.
This commoditisation trap revealed a brutal truth: in electronics manufacturing, standardisation is a one-way street. Once suppliers develop common platforms, brands lose their leverage permanently. The industry structure shifts from differentiated products to commodity boxes, and no amount of marketing can reverse it.
This commoditisation trap perfectly illustrates late-cycle dynamics. During maturity phases, as innovation plateaus, suppliers gain leverage and standardisation becomes inevitable—unless firms maintain strategic control at critical junctures.
Apple watched this disaster unfold and learned the essential lesson: never let suppliers commoditise your differentiation. But executing this insight would prove far more difficult than recognising it.
5/ Apple’s survival—and later dominance—depended on finding a way to outsource without losing control.
Apple's journey to China shows how strategic control theory works in practice—and how the company avoided the commoditisation trap that destroyed PC manufacturers.