The SHEINing – when the smile turns into a grimace
· 19 min read
For roughly three decades, the strategic logic of the Western fashion industry, and many other consumer goods industries for that matter, has been drawn on a single curve. Stan Shih, founder of Taiwanese PC-maker Acer, sketched it in the early 1990s to explain how economic value migrates along a supply chain. High at the two ends, research, design and branding on one side, distribution, marketing and customer relationships on the other and low in the middle, where physical things are actually made. Plot value-added on the vertical axis and the steps of the chain on the horizontal axis, and the curve looks like a smile.
The fashion industry took that smile curve and turned it into a doctrine. Concentrate on the parts that scale through brand equity and consumer attention, and let factories somewhere far away, ideally cheap and obedient, handle the rest. For a long time this looked like very clever capital allocation. It still does, in many boardrooms. But the smile is starting to do something it was not designed to do. Under pressure from a company called SHEIN, it is beginning to twist into a nervous laugh, or rather a grimace.
This is an essay about how that happened, why it all seemed logical each step along the way, and why it still led to a hotel full of haunted ghosts that the legacy industry is now inhabiting and seems unable to check out from.
The hollowing out of Western textile and garment manufacturing is usually narrated as a single event called "globalisation," typically with the tone of someone describing the weather. It was not a single event. It was a sequence and interplay of four distinct economic drivers, each one peeling off a different category of operation, each one rational on its own terms, and each one leaving the smile a little wider and a little emptier in the middle.
The first to go were the most labour-intensive steps, with cut-and-sew at the top of the list. A sewing line is essentially a row of workers on standard machines invented in the mid-19th century, and a supervisor; the capital intensity is low, the skill requirements are real but quickly trainable, and the wage bill dominates the unit cost. Once trade barriers eased and shipping became cheap and reliable, the economic logic for its offshoring became overwhelming.
The Multi-Fibre Arrangement, in place since 1974, had kept this migration partly in check by capping textile and apparel exports from developing countries to OECD markets. When the MFA's quotas were finally completely phased out in 2005, the dam broke. Production that had been distributed across many countries to comply with quotas re-concentrated in the locations with the cheapest hands – China, Vietnam, Bangladesh. Europe and North America retained niches in technical apparel, workwear, very high-end luxury ateliers, and some quick-turn supply for fashion-sensitive segments around the Mediterranean rim or Central America, but the bulk of garment making had left the building.
The second wave was less about wages and more about water, sludge and air. Wet processes such as viscose fibre production, wool scouring, cotton bleaching or complex technical finishing of fabrics are among the most hazardous chemistry-rich industrial processes with correspondingly nasty effluents which became the target of strict environmental regulation in Europe from the mid-1990s.
Due to regulations such as Integrated Pollution Prevention and Control (IPPC, 1996) or the Textile BREF, 2003 rules on effluents, chemical use, emissions and worker exposure tightened significantly and the compliance bill for EU wet processing operators rose sharply. The legislation was, for the most part, reasonable. And while it didn't outlaw textile processing, it made the economics of relocation inescapable. A dyehouse facing a five- or six-figure upgrade to meet new wastewater limits, in a country with rising wages and tightening labour rules, was facing a different cost curve from a competitor in a coastal Chinese industrial park where the same process might be permitted under a much lighter regime. So the dyehouses, the scouring plants, the chemical-heavy fibre lines began to migrate too. Europe, in effect, exported a portion of its environmental footprint along with the work.
It is worth noting that this is the part of the story where Western policy created the very dependence it now complains about. The textile pollution we no longer see in our own rivers did not disappear; it moved beyond the reach of European legislators and out of sight of Western consumers.
The third driver is the one least noticed in popular debate, because it is invisible to the untrained eye and only visible in financial spreadsheets. Fibre extrusion, spinning, weaving and knitting at industrial scale, and modern dyeing and finishing lines, have no problem to meet environmental standards, but they are capital-intensive to build, maintain and operate. A single state-of-the-art spinning mill or polyester filament line is a multi-hundred-million-euro investment, with payback periods that in a generally low-margin industry depend critically on the cost of borrowed money.
Across the 2010s, especially China structurally undercut Western competitors not just on labour or energy but on the price of capital itself. After the housing and infrastructure boom in China started to ebb in the late 2010s, more capital was strategically diverted to energy generation and the build-out of massive production clusters across many industries, including fibres and textiles – the ones that today deliver SHEIN products ultra-fast. State-directed credit based on large pools of domestic savings and policy or state-owned commercial banks willing to finance entire industrial complexes at favourable rates meant that the same machine, often built by the same European supplier, generated a very different return on invested capital depending on where it stood. A European spinner working with single-digit equity returns lending at high cost from private sector banks simply could not match a Chinese counterpart whose effective cost of capital was a multiple lower. So the spinning, the man-made fibre extrusion, the weaving, the knitting and the high-throughput dyeing, printing and finishing lines kept walking east. By the early 2020s, the European share of global man-made fibre and spinning capacity, in particular, had shrunk to a slice.
The fourth driver is the most recent and arrived almost as a single shock. The energy-intensive corners of the chain – viscose and lyocell pulping and spinning, heat-setting, drying, the steam-heavy parts of wet processing, certain finishing operations – depend on cheap, abundant thermal and electrical energy.
Through the late 2010s and early 2020s, European energy prices rose for two reinforcing reasons. The first was a deliberate policy push to decarbonise, which raised the marginal cost of fossil heat while also increasing electricity prices which made some industrial sites economically marginal. The second was the energy shock that followed the Russian invasion of Ukraine in 2022, which removed cheap pipeline gas from the European mix and exposed how dependent industrial baseload had become on a dominant supplier. Industrial gas and electricity prices spiked, then settled at structurally higher levels than the pre-2022 baseline.
China, meanwhile, had spent the previous decade building out generation across every available source – coal at scale, hydro, nuclear, and a renewables build-out so large that by now Chinese electricity production exceeds the combined generation of the EU and US by 44%. Average industrial electricity costs in China stand at 0.08–0.10 €/kWh compared to double these rates on average in the EU. This leads to a competitive disadvantage for energy-intensive European producers that no amount of energy-efficiency measures can bridge. Hence the last remaining energy-hungry steps of the chain began to follow the rest. By the mid-2020s, the migration of the middle was essentially complete.
Stack the four drivers together and you arrive at the present configuration. Western fashion holds the two ends of the smile and very little in between. At the design and branding end sit the trend offices in Paris or London, the licensing portfolios in New York, the heritage brands in Milan or Florence, the design schools, the marketing agencies. At the distribution and consumer-engagement end sit the flagship stores, the e-commerce platforms, the loyalty programmes, the editorial press. In between, a thin layer of regional manufacturing for very fast turnarounds or very protected niches, and a network of agents and sourcing offices managing factories an (increasingly treacherous) ocean away.
This configuration felt safe for a reason that is both true and dangerous. The two ends of the smile are where most of the margin sits. They are also where the moats are supposed to be. Brand equity is hard to replicate; consumer relationships are hard to displace. As long as the lips held, the missing teeth did not seem to matter.
The problem is that SHEIN is now coming for the lips.
SHEIN is regularly described, with a mixture of awe and irritation, as an ultra-fast fashion company. This framing understates what it actually does. SHEIN is a software company that operates a clothing supply chain, and it has built its model precisely around attacking the two ends that the smile curve doctrine reserved for the West.
At the design and trend end, SHEIN does not run a traditional collection cycle. It runs an algorithm. The platform scrapes social media, e-commerce search, and its own customers' behaviour in close to real time, identifies emerging visual cues and demand signals, and pushes a continuous stream of new designs into a network of small, networked factories and workshops that can produce micro-batches of as few as a hundred pieces in a matter of days. Items that sell well are rapidly scaled up; items that flop are killed before they accumulate inventory. The result is something close to a real-time, demand-tested catalogue – tens of thousands of new SKUs per week, each one essentially an A/B test in production.
Compare this to the legacy model. A typical European or American brand designs a collection nine to twelve months ahead of the season, places orders on the basis of buyers' forecasts, ships containers around the world, distributes to wholesale and retail, and discovers what the consumer actually wanted only at the moment markdowns begin. The data flow is slow, the feedback loop is broken, and the system is structurally biased toward overproduction. SHEIN's architecture inverts each of these properties. The trend office is no longer a privileged interpreter of culture; it is an obstruction in the data path.
At the consumer engagement end, SHEIN does something equally unsentimental. It does not build flagship stores. It does not court wholesale buyers. It does not run brand campaigns in glossy magazines. It engages consumers directly on the platforms where they already spend their attention, short-form video, influencer networks, gamified app experiences, and ships from factory warehouses straight to the customer's door. The relationship between consumer and product is mediated by no one except SHEIN itself. The flagship store, the creative director, the magazine editor, the catwalk, the loyalty programme etc. the entire apparatus that legacy brands have spent decades building as a competitive moat is simply routed around.
So the smile is being squeezed from both sides. Design and trend, the left corner, is being eaten by an algorithm that reads consumers faster than designers can guess. Distribution and consumer engagement, the right corner, is being eaten by a direct-to-consumer pipe that does not need the high street, the wholesaler or the magazine. The middle is already gone. What remains of the legacy model is starting to look less like a smile and more like a scare face after too many botched cosmetic interventions.
This is the moment at which the title of this essay stops being a joke.
Stephen King's The Shining is, among other things, a story about a system that has gone wrong and the people who are unable to see it. The Overlook Hotel is beautiful, expensive, and patiently murderous. It is haunted by the ghosts of every bad decision ever made within its walls, and it works on the minds of its caretakers until they become instruments of its own destructive logic. Jack Torrance arrives intending to write, recover, and look after his family. The hotel, gently and then not so gently, persuades him that the way out is to repeat the past.
This is, with very little stretching, the position of the legacy fashion supply chain.
Walk the corridors of any large Western fashion group and you will meet the ghosts. There is the ghost of the fragmented supply chain; fibre in one country, yarn in another, fabric in a third, cutting and sew in a fourth, and visibility ending at tier 2 on a good day. There is the ghost of overproduction, the structural surplus that every season produces and that no one wants to publish, because the volume itself is the point. There is the ghost of ineffective forecasting, the spreadsheets confidently predicting six months out what the consumer is going to want, by colour, size and style, with an accuracy that the consumer's own browsing history would humiliate. There is the ghost of markdown dependency, where headline margins are sustained only by a discount cadence so predictable that no consumer believes the full price anyway. And there is the ghost of inventory write-downs, the quarterly ritual of declaring that several truckloads of perfectly serviceable clothes are worth less than the boxes they sit in.
Management, in this story, plays Jack Torrance. Each season, executives walk into the same room with the same flipchart. Each season, they promise the system will be brought under control. Each season, the system whispers back that the problem is volume, the answer is volume, and the lever is volume. So they source cheaper, ship more, discount harder, write down faster, and assure analysts that the digitisation programme, three years in, is finally about to deliver. The hotel keeps its caretaker. The caretaker keeps the hotel.
The trick of The Shining, and the reason it is one of King's most uncomfortable novels, is that the hotel is not separate from Jack. It works through him. The supply chain ghosts of legacy fashion are not external constraints either. They are the accumulated consequence of choices the industry kept making, about how far to outsource, how cheap to push, how broad to make the collection, how often to refresh, how to compensate buyers, how to report inventory, what to tell investors. Each individual decision was defensible. The system they produced is not.
The other character in King's novel is Danny, the small boy with the shining – the ability to see what others cannot. Danny sees the ghosts. He sees the bath in room 237. He sees what Jack is becoming. None of the adults around him want to look, because what Danny sees would oblige them to act, and acting would oblige them to leave.
SHEIN, in this essay's reading, is Danny.
This is uncomfortable to write, and I want to be careful with it. SHEIN is not a hero. It is a fast fashion company with real and legitimate criticisms attached to its name; on labour conditions in its supplier network, on the environmental footprint of an ultra-fast model, on intellectual property, on tax and trade arbitrage through de-minimis loopholes, on the sheer volume of clothes it is pushing into the world. None of those criticisms is dismissed here. They deserve their own essays, and many, many have been written.
But on the specific question this essay is concerned with, the structural condition of the legacy fashion supply chain, SHEIN has done something that no NGO report, no consulting deck, no regulatory hearing has been able to do. It has held up an ugly mirror and forced the industry to look. The mirror says, your collection cycle is a joke, your forecasting is theatre, your store network is an expensive way of displaying products nobody wants, your write-downs are not bad luck, your moat at the consumer end is thinner than you thought, and the entire model you built on the smile curve assumed that no one would ever come for the lips. The mirror is not flattering. It is also not wrong.
If anything, the appropriate posture toward SHEIN is not adoration and not demonisation, but the kind of grudging acknowledgement one gives to a competitor who has correctly identified one's own weaknesses. The job of the legacy industry now is not to keep finding reasons to dismiss the mirror. It is to look into it.
The dominant Western response to SHEIN so far has been moral. Trade press articles read like obituaries for journalistic balance. Politicians announce fees for small shipments and more inspections. Trade associations describe SHEIN as a threat to civilisation, occasionally to fashion. Consumers are scolded for buying from it. The implied logic is that if we can make SHEIN seem bad enough, the legacy model will look good by comparison and we can keep doing what we are doing.
This is the wrong strategy for three reasons, each of them practical rather than moral.
First, the legacy model has its own large and well-documented sustainability bill, the faster cycles, the overproduction and markdown losses that the consumer still pays, the long-haul logistics, the declining product quality, the inventory write-downs that quietly end up incinerated or in the secondary export markets. Pretending this bill does not exist makes the comparison embarrassing for whoever is making it.
Second, SHEIN's worst characteristics are not unique to SHEIN. They are characteristics of the model the industry has been pursuing for thirty years, just executed more efficiently. Outsourced labour with weak oversight, environmental costs displaced to other jurisdictions, opaque supplier networks, very low unit prices propped up by very high volumes – none of this is invented by SHEIN. SHEIN industrialised what was already there. Attacking the symptom rather than the model means losing the argument the moment a critical journalist looks at the footnotes of any other major fashion brand's sustainability report.
Third, and most importantly, the moral framing distracts the legacy industry from the work it actually needs to do. Demonising SHEIN feels like a strategy. It is not a strategy. It is a substitute for one.
I would put it more bluntly. Instead of demonising SHEIN, the industry should applaud it for holding up a very large stop sign in front of a business model that has been broken for some time, and that no amount of internal change management seemed able to stop. SHEIN did not cause the haunting. It just made it impossible to keep pretending the hotel was fine.
If the smile curve was the strategy of the last thirty years, what is the strategy of the next thirty?
I think it has three pillars, and none of them sound especially radical. They sound, in fact, like things the industry has been told for years, but never implemented. The difference is that the SHEIN mirror has made the cost of not doing them impossible to ignore.
The first pillar is regional reintegration of supply chains. Not full reshoring, that is neither realistic nor desirable for every category. But a deliberate rebuilding of regional capabilities for the part of the assortment where speed, responsiveness, customisation and traceability create more value than the last unit of FOB cost saving. For Europe, this means re-anchoring some spinning, weaving, knitting, dyeing and finishing capacity within reach of the design and consumer end of the chain, supported by industrial policy that takes energy, capital and skills seriously, not as slogans, but as three necessary legs of a solid industrial stool. The current EU debate about strategic industries does not yet treat textiles as one. It should.
The second pillar is deep digitisation and data-driven operations. This is the part where the legacy industry has the most to learn from SHEIN, and the most to gain from doing it on its own terms. Real-time demand sensing, micro-batch testing, integrated PLM and ERP, digital product passports, traceable supplier data, AI-supported design and merchandising: none of these are exotic science fiction concepts. The technology exists, mature suppliers exist, the EU's Digital Product Passport regulation is going to require much of the data infrastructure anyway. The remaining barriers are organisational, cultural, and a long tail of legacy IT. The companies that solve those barriers in this decade will be the ones that survive into the next.
The third pillar is what I would call demand-responsive, well-curated quality. The legacy industry's instinct, faced with SHEIN, is often to argue for quality as a moral category: "we make better clothes." This argument (arguably incorrect anyway) does not move customers who are buying for a Saturday night or click the purchase button based on a TikTok video. And this market segment on a continent with a rapidly ageing population is much smaller than many assume anyway. The more useful framing is that quality and curation work as a strategy when they are paired with a very precise understanding of the customer segment being served, and with an operating model that delivers the right product, in the right quantity, at the right moment, with the right story. Mass-market, undifferentiated, slow-cycle assortments are not a refuge from SHEIN. They are SHEIN's natural prey. A well-defined segment, a curated offer, a responsive supply chain and a meaningful story together can be defended. The smile, in this version, gets put back on a face that actually knows whose it is.
Return for a moment to the title. The SHEINing – when the smile turns into a grimace.
The smile, I think, belongs to the customer. After two decades in which the fashion industry has lectured the customer about her sustainability footprint, dangling labels of organic cotton or recycled polyester in front of her while shipping her thirteen mediocre seasons a year, designed by buyers guessing six months ahead and rescued by markdowns. This customer may be allowed a small, dark laugh at the spectacle of the same industry now wringing its hands about a competitor who simply asked her, in real time, what she wanted. This dark laugh will turn into a broad smile when the product she desires is put reliably in front of her, without sifting through endless volumes, without confusing price signals and frustrating out-of-stock experiences, also by the rest of the fashion industry.
The grimace belongs to the legacy industry. Not because SHEIN is unstoppable, it is not, and its own model has serious internal contradictions, but because the easy strategy of moral superiority is bound to fail. The only remaining strategy is to actually fix the business: rebuild the middle of the chain regionally for flexible on-demand production, digitise the entire operation properly, and stop selling undifferentiated volume to a customer who never asked for it.
The smile curve was a useful description of where value sat in a world where the consumer waited patiently for the industry's decisions. That world is over. In the world that is arriving, value sits wherever the loop between consumer signal and physical product is shortest, cleanest and most trustworthy. SHEIN built that loop crudely, cheaply, and at scale. The legacy industry can build a better one. It will not get to do so while it is still arguing that the hotel is fine.
Time to leave the Overlook.
This article is also published on LinkedIn. illuminem Voices is a democratic space presenting the thoughts and opinions of leading Sustainability & Energy Thought Leaders, their opinions do not necessarily represent those of illuminem.
Track the real-world impact behind the sustainability headlines. illuminem's Data Hub™ offers transparent performance data and climate targets of companies driving the transition.
Jonathan Lishawa
AI · Energy
Luca Miggiano
Public Governance · Corporate Sustainability
illuminem briefings
Geothermal · Renewables
Utility Dive
Green Tech · Sustainable Business
energynews
Green Tech · Sustainable Business
Energy Storage News
Green Tech · Sustainable Business