Yili Group 伊利集团: How a Hohhot Milk Cooperative Became China's Dairy Champion
I. Introduction & Episode Roadmap
There is a particular kind of Chinese business story that only works if you understand what a refrigerator used to mean. For most of the twentieth century, fresh milk in China was a local product with a shelf life measured in hours, sold in glass bottles to households within a few kilometres of a dairy. A national dairy brand was not merely hard to build — it was physically impossible. Then a company in Hohhot 呼和浩特, the capital of Inner Mongolia, roughly a thousand kilometres from Shanghai and Guangzhou, figured out that if you heated milk hot enough and packed it carefully enough, you could ship it anywhere in the country and it would still be drinkable eight months later.
That company was 内蒙古伊利实业集团 Inner Mongolia Yili Industrial Group, and the shelf-stable carton became the vehicle for one of the largest consumer-goods build-outs in modern Chinese history. Thirty years after listing on the Shanghai Stock Exchange, Yili reported FY2025 revenue of RMB 115.931 billion and net profit attributable to owners of RMB 11.565 billion, up 36.82% year on year — a scale the company describes as roughly a 500-fold revenue increase since its 1996 IPO.1 It sells into more than eighty countries from seventy-seven production bases.2
But the interesting part of the Yili story is not the growth curve. It is the three separate moments when the company should have broken and did not — and the one problem in front of it now that no amount of execution seems able to solve.
The first was a governance implosion in 2004 that removed the founding-era chairman in handcuffs. The second was the 2008 melamine catastrophe, an industry-wide poisoning scandal that killed one of China's largest dairy companies outright and forced a state-owned rescue of another. Yili was caught up in it, survived without a bailout, and emerged into an industry that had been permanently reshaped in favour of large, well-capitalised national players. The third — still unfolding — is demographic. China recorded 7.92 million births in 2025, down roughly 17% from the prior year and the lowest figure in the modern statistical record.3 Infant formula is Yili's fastest-growing and most strategically contested business. It is also a business whose customers are, quite literally, disappearing.
So the questions this piece works through are these. How did Yili out-run 蒙牛乳业 Mengniu Dairy, a company founded by Yili's own former executives, to become the clear number one? Why did Yili survive 2008 when 三鹿 Sanlu did not, and what did that survival buy? And the central investment question: can market-share gains and a near-total payout of profit to shareholders outrun a shrinking domestic consumer base — or is Yili simply a very well-run business harvesting a maturing category?
The answers are not uniformly flattering. Yili's advantages in distribution and scale are real and evidenced. Its answer to "how do we grow in a shrinking market" rests substantially on an acquisition that, four years in, has not yet proven it was worth what was paid. Both things can be true at once, and the gap between them is where the actual analysis lives.
Subscribe on LinkedInII. Origins: From a Hohhot Cooperative to China's First Dairy IPO (1956–1999)
The Hohhot Huimin District dairy cooperative that became Yili was founded in 1956, in a city where the surrounding grasslands made milk abundant and the absence of cold-chain infrastructure made it nearly worthless beyond the city limits.4 For decades this was a purely local operation — a collection point, a small processing plant, and a delivery radius. Inner Mongolia had the cows. What it did not have was a way to reach the people with money.
The restructuring into a joint-stock company came in 1993, part of the broader wave of state-enterprise corporatisation sweeping China at the time. Three years later, in March 1996, Yili listed on the Shanghai Stock Exchange, raising RMB 96.9 million.4 By the standards of what the company would later become, this was a rounding error — less than one-thousandth of a single year's current revenue. But it made Yili the first dairy company listed in China, and that mattered for reasons that had nothing to do with the money raised.
Being listed early gave Yili two structural gifts. It gave the company access to public equity capital at a moment when Chinese consumer companies were overwhelmingly funded by bank debt and retained earnings — and dairy is a capital-hungry business, requiring processing plants, refrigerated logistics, and distribution infrastructure long before the revenue arrives. It also imposed disclosure discipline and an outside-shareholder constituency years before competitors had either. Mengniu, the company that would become Yili's defining rival, did not yet exist in 1996; it was founded in 1999 by former Yili executives and listed in Hong Kong in 2004.
The technology that turned a regional processor into a national brand was ultra-high-temperature processing — UHT. The concept is simple enough to explain at a dinner table: heat milk very briefly to around 140°C, which kills essentially everything living in it, then seal it into a sterile multi-layer carton before anything can get back in. The result sits on an unrefrigerated shelf for months. Yili's adoption of UHT extended shelf life to roughly eight months, which meant a truck leaving Hohhot could reach a corner shop in Guangdong and the product would still be perfectly saleable.4
This is the single most important technical fact in the entire Chinese dairy story, and it is worth dwelling on why. UHT converted milk from a perishable local good into a shippable national commodity. It meant scale economics suddenly applied: build the biggest plants next to the cheapest milk, and distribute nationally. It meant brand advertising suddenly worked, because a brand you saw on television in Shanghai was actually purchasable in Shanghai. And it meant that whoever built the distribution network first would enjoy a lead that competitors would need years and enormous capital to close.
In 1999, Yili created a dedicated Liquid Milk Division and handed it to a manager in his late twenties named 潘刚 Pan Gang. The division became the engine of the company's national expansion, and it became the platform on which Pan built the internal reputation that would carry him, within six years, to the chairmanship. By 2003 Yili was China's largest dairy manufacturer by revenue, and by 2005 the company had crossed RMB 10 billion in sales.4
That is the table-setting. The origin story matters mainly because it explains the physical asset base and the product category that everything else sits on top of. What actually defines the modern company — the governance structure, the ownership question, the concentration of power in one person — begins with a criminal case.
III. The Zheng Junhua Scandal and the Making of Pan Gang (2004–2005)
On the afternoon of 17 December 2004, prosecutors in Hohhot took 郑俊怀 Zheng Junhuai — the man who had run Yili since 1993 and who was known in the Chinese dairy trade as something close to its godfather — out of the company.5 He was subsequently convicted of embezzlement involving more than RMB 16 million and sentenced to six years, of which he served roughly half before his release in 2008.5
The formal charge was embezzlement. The underlying activity, as reported at the time and since, was an attempt to take control of Yili's own state-held shares through vehicles that were not disclosed as connected to management — a covert management buyout, executed through intermediary entities, using company resources. Whatever the legal characterisation, the commercial substance was straightforward: the executives running a state-invested listed company were trying to become its owners without saying so.
It is worth being precise about why this was such a hinge moment, because it is easy to file it away as generic Chinese corporate-governance noise. In the mid-2000s, a very large number of Chinese enterprises were caught in exactly the same bind. They had been built by managers who behaved like founders but owned almost nothing, sitting inside ownership structures held by local governments and state investment vehicles. The economic logic of an MBO was overwhelming. The legal and political framework for doing one openly, at fair value, barely existed. Some managers waited. Some, like Zheng, did not.
Into the vacuum stepped Pan Gang, then thirty-five, elevated to chairman and president in 2005 after roughly a decade inside the company and six years running its most important division. He had the operational credibility — the liquid milk business he built was the reason Yili was the largest dairy company in the country — and he had the advantage of not being implicated.
The resolution Yili arrived at is the part that deserves scrutiny. Rather than leaving management with no equity, the company went the other way, using post-2005 equity incentive plans to give executives large, legal, disclosed stakes. Over the following two decades this transformed Pan Gang from a salaried manager into the company's largest individual shareholder, holding roughly 286.7 million shares, or about 4.53% of the equity.6
The charitable reading is that this was textbook governance repair: the illegal path was closed, and a legal, transparent, shareholder-approved path was opened in its place, aligning the people running the business with the people who owned it. The sceptical reading is that management ended up with a large ownership stake in the company either way — the difference being paperwork, sequencing, and the consent of the board.
Both readings survive the evidence, and an honest assessment has to hold them together. What is not in dispute is the structural outcome. Yili has had no controlling shareholder since 2011. Its largest single holder, Hohhot Investment Co., held 8.51% as of late December 2025 — a meaningful stake but nowhere near control.6 The consequence is a company with a genuinely dispersed register, no state parent standing behind it, and an executive chairman who is simultaneously its largest individual owner. The upside of that structure is that Yili behaves commercially rather than politically. The risk is that there is no single large shareholder positioned to check the chief executive.
That question — how much of Yili is Pan Gang, and what happens if that ever changes — comes back later, and it comes back with a specific, dateable stress test attached. But before any of it could matter, the entire Chinese dairy industry had to nearly destroy itself.
IV. 2008: The Melamine Crisis and the Flight to Scale
In the summer of 2008, hospitals across several Chinese provinces began seeing an unusual pattern: infants, many of them only a few months old, presenting with kidney stones. Kidney stones in babies are rare. Clusters of them are not a coincidence.
The cause was melamine, an industrial chemical used in plastics and fertiliser, which had been added to watered-down raw milk at collection stations. The reason was almost mundane in its cynicism. Milk quality was, and largely still is, assessed by protein content, and protein content is inferred by measuring nitrogen. Melamine is nitrogen-rich. Add it to diluted milk and the milk tests as though it were undiluted. It is the food-safety equivalent of loading a cargo container with bricks so it passes a weight check.
The epicentre was 三鹿集团 Sanlu Group, then one of China's largest infant-formula producers. But the investigation that followed established the more disturbing fact: this was not one bad company. A nationwide sweep found melamine in products from 22 of 109 tested dairy companies — including Yili and Mengniu, at concentrations far below Sanlu's, but present nonetheless.7 Roughly 300,000 infants were affected nationally, with a small number of deaths confirmed.78
The commercial consequences separated the industry into survivors and casualties with brutal speed. Sanlu collapsed entirely. Mengniu, listed in Hong Kong and therefore priced in real time by international investors, saw its shares fall around 60% in a single day when trading resumed, and ultimately required a rescue by state-owned COFCO 中粮集团, which took a controlling position. Yili, listed in Shanghai and less exposed to infant formula at the time, was damaged but did not require a bailout.
Here is where the analysis gets genuinely counter-intuitive, and it is the most important structural insight in this entire story. The melamine scandal, which by any reasonable accounting was a catastrophe for Chinese dairy, turned out to be the single largest structural gift the two national leaders ever received.
Three mechanisms drove this. First, regulation. National dairy safety and testing standards were tightened substantially, and a formal infant formula registration regime — 配方注册制 — was subsequently introduced, sharply limiting the number of formula recipes any company could register and requiring documentation, traceability, and testing infrastructure that small and regional dairies simply could not afford. Compliance cost is a fixed cost. Fixed costs favour scale. Hundreds of small producers exited.
Second, supply chain. The adulteration had happened at scattered independent collection stations, so the industry response was to bring milk sourcing under direct control — large-scale farms, owned or contractually locked, with testing at every stage. This is capital-intensive in exactly the way that advantages companies with access to public equity markets.
Third, and most durably, consumer psychology. Chinese parents concluded — rationally — that a large national brand with a reputation to protect and money to spend on testing was safer than a cheap local one. That preference did not fade. It became the default. Nearly two decades later, the premium consumers pay for perceived safety in Chinese dairy remains a live, monetisable phenomenon.
Put those together and you get the modern structure: a national market where, by 2023, Yili held roughly 24.4% of dairy product sales and Mengniu roughly 17.9%, with everyone else fighting over the remainder.[^9] That duopoly is a direct descendant of 2008. It was not competed into existence. It was, to a substantial degree, regulated and traumatised into existence.
The uncomfortable corollary, which any long-term holder should sit with, is that the same mechanism runs in reverse. If the trust that consolidated the industry around Yili was created by a food-safety disaster, it can be destroyed by one. This is the classic low-probability, high-consequence tail risk: it may never happen, and if it does, the damage is not modelled by any normal earnings sensitivity. Sanlu's equity did not decline. It went to zero.
Survival, though, is not the same as winning. Coming out of 2008, Yili was not yet the clear leader. Mengniu was.
V. Overtaking Mengniu: The Race to #1 (2005–2013)
The Mengniu story is one of the great sprints in Chinese business. Founded in 1999 by 牛根生 Niu Gensheng and other former Yili executives, it grew at a pace that made Yili look sedate, taking the revenue lead in the mid-2000s through aggressive marketing, rapid distribution expansion, and a willingness to outsource capacity and outsource milk sourcing in order to move faster. For a period, the student had comprehensively beaten the school.
Then two things happened. Mengniu absorbed the harder blow in 2008 and spent the following years under COFCO's ownership, with the strategic consequences that state control tends to bring: more caution, more executive churn, more attention paid to considerations other than the P&L. Yili, meanwhile, had no controlling shareholder to answer to and a chairman with a growing personal equity stake and a decade-long operating record in exactly the segment that mattered most.
Yili drew level and passed Mengniu around 2013, and the gap has widened more or less continuously since. By FY2024, Yili's revenue of RMB 115.78 billion compared with Mengniu's RMB 88.7 billion.910 In FY2025, Mengniu's revenue fell a further 7.3% to RMB 82.2 billion while Yili's held essentially flat at RMB 115.9 billion — meaning that in a year when the whole Chinese dairy market was under pressure, the gap between number one and number two widened by roughly RMB 6 billion.111
The mechanism of the overtake was not a single brilliant strategic move. It was the accumulation of ordinary execution over a decade: liquid milk product development, relentless expansion of the distributor network into lower-tier cities and rural counties, brand investment sustained through downturns, and a willingness to build owned infrastructure rather than rent it. None of this is glamorous. All of it compounds.
It is worth being fair to Mengniu, because a foil that is simply written off as incompetent teaches nothing. Mengniu's FY2025 results actually showed real operational repair: gross margin improved to 39.9%, net operating cash flow reached a record RMB 8.75 billion, and profit attributable to shareholders rose almost fourteen-fold to RMB 1.545 billion — though that last number says more about how impairment-battered the prior year was than about underlying earnings power.11 Management guided to mid-single-digit revenue growth for 2026 and launched a three-year shareholder return plan.12 Mengniu is not dying. It is a competent number two that has spent several years cleaning up acquisitions that did not work.
Which is the actual lesson of the race. In a category where the product is close to a commodity and the technology is available to everyone, the winner is not decided by insight. It is decided by consistency of execution, ownership structures that permit long-horizon decisions, and the discipline to avoid destroying capital on deals. Yili won the 2005–2013 race largely by making fewer unforced errors. Whether that record holds is a question the next sections put under real pressure — starting with the business that generates well over half of everything Yili sells.
VI. The Core Business: Liquid Milk — Industry Structure, Competition, Economics
Walk into any Chinese supermarket and the liquid milk aisle tells you the whole competitive story in about fifteen seconds. There are two dominant brand blocks — Yili's 金典 Satine and its base white milk range on one side, Mengniu's 特仑苏 Milk Deluxe on the other — occupying the majority of shelf space, priced within a few jiao of each other, promoted with near-identical mechanics, and physically almost indistinguishable in the glass. Around them sit regional players, a fresh-milk section, and a wall of yoghurt.
That aisle generated RMB 36.13 billion of revenue for Yili in the first half of 2025, roughly 58% of the group total — and it declined 2.06% year on year, in a half where group revenue grew 3.37%.13 This is the central operating fact about Yili today: the biggest business is the slowest one, and it has been for several years.
FY2024 was the year the problem became undeniable. Group revenue fell 8.2% — the first sustained revenue contraction in the company's modern history — and net profit attributable to owners fell about 19% to RMB 8.45 billion.9 Management attributed the top-line decline substantially to channel destocking: distributors and retailers holding too much inventory, which Yili deliberately worked down rather than continuing to push product into a saturating market. That is the right decision, and it is also the kind of decision that is only necessary because too much was pushed in the first place.
But destocking is a one-year explanation, and the softness has outlasted it. The more durable driver is consumer downgrade — Chinese households trading from premium organic UHT to standard UHT, from branded to promotional, from three cartons a week to two. Liquid milk in China has moved from a penetration story to a maturity story, and the shift happened faster than most forecasts assumed.
The five forces, honestly applied
Run the category through Porter's framework and the picture is not flattering.
Rivalry is intense and structurally so, because the products are close substitutes and both leaders have the balance sheet to fund a price war indefinitely. Buyer power — which in consumer goods means channel power — has risen sharply. The rise of e-commerce and livestream commerce on 抖音 Douyin, 京东 JD.com and 天猫 Tmall gave platforms leverage that traditional supermarkets never had: transparent pricing, instant comparison, and the ability to demand promotional funding as the price of visibility. Supplier power is currently weak, because China has a raw milk surplus, but it is cyclical and will not stay weak forever. Substitutes are a genuine and growing pressure: plant-based drinks, ready-to-drink coffee and tea, and functional beverages all compete for the same occasions. Barriers to entry are the one favourable force — national cold-chain and distribution reach takes a decade and billions to replicate, and post-2008 regulatory compliance makes small-scale entry uneconomic.
You can see the rivalry in the cost structure. Yili's selling expense ratio ran at roughly 19% of revenue in FY2024, up about 1.1 percentage points year on year.14 Roughly one yuan in five of everything the company sells goes to advertising, promotion, channel rebates and trade spend. That is what maintaining share in a commoditised category with powerful channels actually costs. Encouragingly, that ratio came down to about 16.9% in the first quarter of 2025 as the company pulled back promotional intensity — evidence that at least some of the spend was discretionary and that management is willing to trade volume for margin when it chooses.14
Who Yili is actually fighting
Mengniu is the direct national competitor and the only company with comparable reach. 光明乳业 Bright Dairy carries genuine heritage in Shanghai and real strength in chilled fresh milk, but never converted regional credibility into national distribution — it is a strong local brand in a national market. New Hope Dairy has quietly run the best gross margins in the peer group by concentrating on chilled and fresh products in its home regions, which is a useful demonstration that the fresh-milk sub-category is structurally more profitable than shelf-stable UHT, because it is harder to ship and therefore harder to commoditise. 君乐宝 Junlebao has grown fast in the second tier, particularly in formula, and represents the kind of competitor that erodes share at the edges rather than attacking the centre.
Yili's advantage over all of them is not a product moat and it is not pricing power. It is distribution density: more distributors, deeper penetration into county-level and rural markets, more cold-chain nodes, and better in-store execution accumulated over two decades. That is a real advantage with real economics — it lowers the cost of reaching each incremental consumer and it makes national product launches faster and cheaper than any competitor can manage.
It is also, and this is the honest bear point, capital-intensive and imitable. Mengniu has spent years building the same thing. Distribution reach is a scale advantage, not a structural one — it protects against small entrants but not against an equally large rival. And a distribution advantage only converts into growth if the category is growing.
Which is precisely the problem. Yili's answer to "why do we win" in liquid milk is well-evidenced. Its answer to "why do we grow" is not. There were signs of stabilisation — company disclosures pointed to low-temperature white milk growing more than 20% in the first half of 2025, and management said liquid milk returned to positive growth in the first quarter of 2026, describing it as the segment's first positive quarter after a prolonged adjustment.1315 One quarter is a data point, not a trend, and the comparison base was itself depressed. Whether the 2024–2025 contraction was cyclical or structural is not yet settled by the evidence, and investors should resist the temptation to declare it settled.
If liquid milk is the mature core, the growth has to come from somewhere else. Yili's answer was to spend roughly US$800 million on it.
VII. Betting on Nutrition: Infant Formula, Ausnutria, and the Fight Against Demographics
In October 2021, Yili filed a plan to acquire a 34.33% controlling stake in 澳优乳业 Ausnutria Dairy, a Hong Kong-listed infant formula company with Dutch manufacturing assets and, critically, the Kabrita brand — the dominant name in China's imported goat-milk infant formula, which had accounted for more than 60% of that category for three consecutive years from 2018 to 2020.16
The terms were not cheap. Yili's Hong Kong subsidiary agreed to buy 531 million existing shares at HK$10.06 each — roughly HK$5.3 billion, a premium of about 33% to the thirty-day average before Ausnutria's trading halt — plus 90 million newly issued shares, bringing total consideration to approximately HK$6.25 billion, or around US$800 million.16 The deal completed in early 2022, and Yili chose to leave Ausnutria listed and operating independently rather than absorbing it.
The strategic logic was clear and defensible at the time. Yili's revenue was overwhelmingly liquid milk, a category already showing signs of maturity. Infant formula carried structurally higher gross margins, higher brand loyalty, and — thanks to the post-2008 registration regime — meaningful protection from new entrants. Yili had cow-milk formula. It did not have goat, which was the fastest-premiumising sub-category in Chinese infant nutrition, commanding higher prices on claims of digestibility. Buying the category leader rather than building a challenger brand was, in principle, the efficient move.
Four years on, the results demand a harder look.
The overpayment question
Ausnutria's revenue and profit fell sharply in the years after the deal closed. Then, on 24 July 2026, the company issued a profit alert for the first half of 2026: revenue of approximately RMB 3.065 billion to RMB 3.165 billion, down roughly 20% from RMB 3.89 billion a year earlier, and a net loss of RMB 685 million to RMB 785 million, against a profit of RMB 180.9 million in the prior-year period.17
Management's framing deserves attention, because how a company explains a bad number is itself evidence. Ausnutria attributed the loss primarily to one-off inventory adjustments and non-cash asset impairments, stating that excluding these items, core operating net profit would have been approximately RMB 155 million to RMB 255 million, and that core business fundamentals remained stable. It also cited external pressures: changes in the international logistics environment, volatile shipping costs, and tightening regulatory oversight disrupting supply and fulfilment for certain products.17
Take that explanation at face value and the operating business is roughly breakeven-to-modestly-profitable at a scale far below what it was when Yili bought it. Do not take it at face value, and the pattern is more concerning: inventory write-downs and asset impairments are not weather events. They are the accounting recognition that products previously expected to sell at a certain price will not, and that assets carried at a certain value are not worth it. When those charges recur, the "one-off" label wears thin.
Either way, the benchmark question is unavoidable. Yili paid roughly US$800 million and a 33% control premium for an asset that, four and a half years later, was guiding to a nine-figure renminbi loss in a single half-year. On the current evidence, the acquisition has not earned its price. That is not a verdict on whether it eventually will — Kabrita remains a genuine category leader and goat formula remains a real premium niche — but it is the fair reading of the record so far, and it is the single largest open question in Yili's capital allocation history.
The demographic problem, stated plainly
The macro backdrop makes the execution bar considerably higher. China recorded 7.92 million births in 2025 against 11.31 million deaths, with the total population falling by 3.39 million to 1.4049 billion.3 The 2025 birth cohort was down roughly 17% year on year and was the lowest in the modern record.18
The arithmetic for infant formula is unforgiving. Formula consumption is driven almost entirely by the size of the zero-to-three population, and that population is a rolling sum of recent birth cohorts. When the cohorts fall by double digits, the addressable market contracts with a lag of about a year and keeps contracting for three. Government pro-natalist measures — including a nationwide childcare subsidy of RMB 3,600 per child per year up to age three — provide some support to spending per baby, but no policy tested anywhere in East Asia has yet reversed a birth-rate decline of this magnitude.19
This is the risk that sits underneath everything else in the segment. A company can win share brilliantly and still see revenue fall, if the pie shrinks faster than the slice grows.
The scramble inside the shrinking pie
Which is exactly what the competitive dynamic has looked like. 中国飞鹤 China Feihe had long been the leading infant formula brand in China, at roughly 17.5% share against Danone at about 14% as of early 2025.19 In March 2025 Feihe announced an unusually aggressive move: RMB 1.2 billion — around 6% of its turnover — allocated to a nationwide birth subsidy programme providing at least RMB 1,500 per eligible pregnant family in free formula, equivalent to roughly six free cans.19 Jefferies analyst David Hayes characterised the scale bluntly, and analysts flagged "margin reset risks" for Danone, Yili, Friso and a2 Milk as competitors were forced to respond.19
The results, so far, have been instructive. Feihe's own first-half 2025 revenue fell 9.4% to RMB 9.15 billion and profit fell 46% to RMB 1.03 billion.20 Yili, meanwhile, reported its goat-milk infant formula share rising three percentage points to 34.4% and e-commerce sales growing over 30%.20 Yili's milk powder and dairy products segment grew 14.26% to RMB 16.58 billion in the first half of 2025 — the fastest-growing part of the company — and the company stated it had reached 18.1% share in infant formula, the number one position in China.13
That claim was repeated for the full year: Yili said FY2025 marked the first time it took the top infant formula market share position in China, with milk powder overall ranking first nationally.12 Two caveats belong alongside it. First, this is company-stated language drawn from commissioned retail measurement, not an independently audited figure, and the definitional boundaries of "infant formula share" vary meaningfully between data providers. Second, taking share leadership in a category where the leader's own revenue is declining is a genuinely ambiguous achievement — it may reflect superior execution, or it may reflect a willingness to spend more aggressively than a rival that has decided the volume is not worth the margin.
The honest framing is that this segment is simultaneously Yili's best evidence of competitive strength and its most exposed strategic bet. The share gains appear real. The category is shrinking. And the acquisition purchased to win it has, to date, destroyed rather than created reported earnings. An investor who believes the Ausnutria deal works needs the losses to stop compounding first — and there is not yet evidence in the disclosed record that they have.
Fortunately, the segment that quietly generates a great deal of Yili's value has almost nothing to do with babies.
VIII. The Quiet Compounder: Ice Cream and the Upstream Supply Chain
Every Chinese summer, a small commercial war is fought inside the freezer cabinets of roughly six million small retail outlets. It is won on placement, on turnover speed, on the reliability of the person who restocks the cabinet, and on whether the ice cream in it is still frozen. It is a business that rewards precisely the capability Yili spent twenty years building — cold-chain logistics and distribution density — and Yili has held the number one position in Chinese cold drinks for thirty-one consecutive years.2
The segment generated RMB 8.23 billion in the first half of 2025, up 12.39%, roughly 13% of group revenue.13 For the full year, revenue approached RMB 10 billion with double-digit growth.1 It is the smallest of Yili's three core segments and, on the available evidence, the healthiest: it grows, it is defensible in a way liquid milk is not — because a competitor must replicate an entire frozen distribution network, not merely a shelf-stable one — and it is decoupled from the birth rate entirely. Overseas cold-drink revenue grew 10.2% in 2025, giving it a second growth vector.1
Yili does not disclose segment profitability in enough detail to size ice cream's contribution to earnings precisely. But strategically it functions as ballast: a durable, high-return sub-business whose competitive logic is the same distribution advantage that liquid milk relies on, applied to a category where that advantage is harder to copy.
Buying milk at the source
The second thread of Yili's strategy over the past decade has been control of raw material, pursued both abroad and at home.
Abroad, the landmark deal was Westland Milk Products. In March 2019, Yili agreed to acquire the New Zealand co-operative — the country's third-largest dairy producer — for NZ$588 million, about US$404 million, at NZ$3.41 per share via a scheme of arrangement.21 The terms included an unusual commitment: guaranteed milk collection for existing supplier farmers and a payout of at least the Fonterra farm gate milk price for ten seasons, backed by a guarantee from Yili's acquiring subsidiary.21 That structure was as much political as commercial — foreign acquisitions of New Zealand dairy assets require Overseas Investment Act consent and are politically sensitive, and Yili bought goodwill along with the plant.
What Yili actually acquired was access to grass-fed New Zealand milk, a premium export platform, and a manufacturing base outside China. This followed the 2013 purchase of Oceania Dairy, also in New Zealand, and sits alongside overseas production bases that now number among the seventy-seven the company operates globally.42 The strategic value is real but bounded: New Zealand dairy assets are chiefly a supply-security and premium-positioning play, not a growth engine, and Fonterra-adjacent assets have historically traded on modest multiples reflecting their commodity exposure.
At home, the more consequential structure is 优然牧业 Youran Dairy. Spun out in 2015 as Yili's dedicated raw milk platform and listed in Hong Kong in 2021, Youran operates large-scale farms supplying Yili's processing plants. Yili has held roughly a third of the equity — approximately 33.93% as of mid-2024 through two holding entities — with funds managed by PAG collectively holding around 29.77%.22
The commercial relationship is far tighter than the ownership percentage suggests. Under a long-term framework agreement, Youran must supply 70% of its raw milk output to Yili, and Yili has agreed to purchase the remaining 30% if Youran is willing and able to supply it.22 In practice the concentration is near-total: in the first half of 2024, Youran sold RMB 7.32 billion of raw milk, of which RMB 6.90 billion — 94.28% — went to Yili.22
This is the mirror image of Mengniu's relationship with China Modern Dairy, and it embodies the same theory: after 2008, controlling your milk supply was a safety imperative and a quality differentiator. In a rising milk price environment, it is also a hedge — when input costs go up, your equity stake in the supplier appreciates.
The theory has one obvious flaw. It works in both directions.
IX. When the Supply Chain You Built Turns Against You: The 2023–2025 Raw Milk Glut
Around 2019, Chinese dairy farms began importing heifers at scale, betting on a domestic milk market that had grown reliably for two decades and on government targets encouraging self-sufficiency. Cows take roughly two years to reach milking maturity. The arithmetic was therefore fixed years in advance: the milk from that wave of imports would arrive in 2021 and 2022, whether or not anyone still wanted it.
They did not. Chinese milk output climbed while demand softened, and prices fell for an extended and painful stretch. By the last week of June 2023, the national raw milk price had reached about RMB 3.77 per kilogram — down 13.5% from the August 2021 peak of RMB 4.36 — against a production cost of roughly RMB 3.80 per kilogram even for large, debt-free ranches.23 In other words, the most efficient farms in the country were selling milk below cost. More than 60% of Chinese dairy farms were operating at a deficit, with only the top-performing fifth to third still profitable, and an industry veteran quoted at the time described it as the sector's darkest period in over fifteen years.23
Processors handled the surplus the only way they could: converting it to powder and warehousing it. At the worst point, in February 2023, some 10,800 tonnes of milk were being turned into powder daily — about 17% of all milk purchased.23 That inventory did not disappear. It sat on balance sheets, and eventually it had to be written down or sold at a loss.
Herd liquidation followed, as it always does. Farms culled inefficient cows, average herd sizes contracted, and smaller operations — those under 3,000 head — shrank fastest.23 By 2025 the correction had largely done its work on the supply side: national milk output reached 40.91 million tonnes, up just 0.3% on the year, essentially flat after years of expansion.3
The two-sided effect on Yili
For Yili's processing business, cheap raw milk is straightforwardly good. Milk is the dominant input cost in liquid dairy, and a multi-year decline in its price is a gross margin tailwind that flows through with a short lag.
For Yili's balance sheet, it was the opposite. The company's stake in Youran meant it participated directly in the losses of the upstream sector — through equity-method investment losses, through impairments on the carrying value of that stake, and through the write-downs on inventory and biological assets that farms across China were forced to take. Youran's reported losses in this period were substantially driven by revaluation of biological assets — accounting language for the fact that a dairy cow's value depends on the milk and beef prices she is expected to generate, and both fell.
The result showed up starkly in FY2024. Revenue fell 8.2%, but net profit attributable to owners fell about 19% — the profit decline running more than twice the revenue decline.9 A processor benefiting from cheaper inputs should, all else equal, see profit fall by less than revenue, not more. The gap is the upstream drag, and it is the clearest available evidence that vertical integration converted from moat to earnings-volatility source.
This is the analytical point worth carrying forward. Vertical integration is often described as though it were unambiguously defensive. It is not. Owning your supplier means owning your supplier's cycle. Yili built the upstream position for food-safety and supply-security reasons that were entirely rational after 2008, and the cost of that insurance turned out to be a multi-year drag on reported earnings during a downcycle it could not control.
Interrogating the rebound
Which brings us to FY2025 and the number that requires the most scepticism in this entire story. Yili's net profit rose 36.82% to RMB 11.565 billion on revenue that was essentially flat at RMB 115.931 billion.1
A 37% profit increase on flat revenue is arithmetically only possible through some combination of margin expansion, mix shift, cost reduction, and — crucially — the absence of prior-year charges. All four were plausibly present. Mix shifted favourably as milk powder grew 14% and cold drinks 12% while liquid milk shrank. Raw milk costs stayed low. Selling expense discipline improved. And the comparison base was a year depressed by impairment.
How much of the rebound is durable margin improvement and how much is a base effect? The disclosed record does not permit a precise split, and any confident answer is guesswork. But the direction can be sanity-checked against a peer: Mengniu's profit rose almost fourteen-fold in FY2025, explicitly because the prior year's large impairments did not recur.11 When both duopolists post enormous profit growth in the same year on declining or flat revenue, the most parsimonious explanation is that both are lapping a bad base, not that both simultaneously discovered operating leverage.
That does not make Yili's FY2025 result fake — the underlying margin improvement was real, and first-quarter 2026 results extended it with revenue up 5.47% to RMB 34.83 billion and net profit up 10.68% to RMB 5.40 billion.15 It does mean the growth rate is not the right number to extrapolate. The level of margin is more informative than the change, and one clean year is what investors have.
The person who has to make that margin stick has now been running the company for twenty-one years.
X. Management Today: Pan Gang — Power, Ownership, and the 2018 Test of Trust
Pan Gang holds four titles at Yili simultaneously: chairman, president, chief executive, and party secretary. In a company with no controlling shareholder, that is an extraordinary concentration of authority. There is no state parent to overrule him, no founding family with a blocking stake, and no single institution on the register large enough to force a change. The largest holder sits at 8.51%; Pan himself is the largest individual owner at roughly 4.53%.6
For a company approaching RMB 116 billion in revenue and three decades of listed history, this structure is unusual anywhere in the world. It is worth assessing on behaviour rather than on theory.
The capital allocation record
The strongest evidence in Pan's favour is what Yili has done with its cash, consistently, over a long period.
For FY2024 — a year in which profit fell sharply — Yili proposed dividends of RMB 7.726 billion, a payout ratio of 91.4%, the highest in company history.24 Alongside it the company ran a buy-back-and-cancel programme of up to RMB 2 billion, having repurchased 36.2 million shares for RMB 877 million by the end of April 2025.24 Combined, total shareholder returns reached 100.4% of profit for the year — Yili returned slightly more than it earned.24 Cumulative dividends since the 1996 listing had reached RMB 50.859 billion, the largest of any Chinese dairy company.24
The FY2025 distribution was larger in absolute terms and lower as a ratio: a final dividend of RMB 0.90 per share totalling approximately RMB 5.693 billion, on top of an interim distribution of about RMB 3.036 billion, for roughly RMB 8.729 billion — around three-quarters of net profit, with the record date set for 4 June 2026.2526 Alongside it, Yili published a 2025–2027 shareholder return plan committing to an annual payout ratio of no less than 75% and a dividend of no less than RMB 1.22 per share.25
Read those two years together and the picture is more interesting than either alone. The 91.4% payout in FY2024 was not a new policy; it was the mechanical result of holding the dividend roughly stable through an impairment-depressed profit year. The FY2025 policy floor of 75% is the actual commitment — generous by Chinese A-share standards, and notably a floor rather than an aspiration. What it also implies is that the very high FY2024 ratio should not be extrapolated. Investors sizing the income case should anchor on the stated floor and the per-share commitment, not on the peak ratio.
Still, the behavioural evidence is genuinely strong: a company that maintained distributions through its worst year in two decades, cancelled repurchased shares rather than parking them in treasury, and published a multi-year, numerically specific commitment. That is not the behaviour of management empire-building with shareholder money. It is close to the opposite.
The 2018 stress test
Late in March 2018, a piece of fiction began circulating on WeChat. It described a dairy company that closely resembled Yili and a chairman who closely resembled Pan Gang, and in the story the chairman is detained by police. It went viral.27
Yili's shares fell sharply, and roughly RMB 6.1 billion — close to US$960 million — of market value evaporated.28 The company's explanation was that Pan was abroad receiving treatment for a heart condition. He reappeared publicly on 31 May 2018 at the annual shareholders' meeting, and the shares rose about 10% that day to their highest close in a month.28 Two individuals were subsequently prosecuted for spreading the story on "picking quarrels and provoking trouble" charges: a blogger who received a one-year sentence suspended for eighteen months, and a journalist sentenced to eight months.27
Yili also went on the offensive, publicly accusing Zheng Junhuai — released from prison in 2008 — of orchestrating defamatory attacks on the company over more than a decade, and asserting that he had been permitted to serve his sentence at home.27 Thirteen years after the fact, the founding-era scandal was still generating public conflict.
What should an investor take from this? Not that Pan did anything wrong; the record does not support that, and the rumour was legally established to be fabricated. The signal is about fragility. A single unverified story about one man's whereabouts removed roughly RMB 6 billion of value in days and did not fully recover until he physically appeared in a room. That is a precise, dateable measurement of key-person risk, and it is unusually high for a company of this size in a category as unglamorous as milk.
The share pledge
A more recent disclosure sharpens the governance question. On 9 January 2026, Yili announced that Pan Gang planned to reduce his holding by up to 62 million shares between 29 January and 14 April 2026, raising approximately RMB 1.762 billion at the then-prevailing price. The stated purpose was to repay matured stock-pledge financing loans taken against shares acquired through equity incentive subscriptions and market purchases.6 His 2024 compensation was RMB 19.74 million, among the highest for A-share company chairmen.6
The mechanics are worth understanding. Equity incentive plans require executives to pay for shares, and executives frequently borrow against the shares to fund the purchase. When those loans mature, they must be repaid — often by selling some of the stock. This is legal, disclosed, and common. It is also a reminder that the alignment created by the post-2005 incentive structure is partly financed by leverage, and that a large personal stake built with borrowed money creates its own pressures. It does not tell you management has lost conviction; a sale of 62 million shares against a holding of 286.7 million leaves the great majority intact. But it is the kind of detail a sceptical investor would want in front of them alongside the equity-alignment argument.
The net read on management credibility is genuinely mixed and should be held as such. Capital allocation discipline is strong, consistent, numerically committed, and maintained through a downturn — that is the highest-quality evidence available and it points clearly favourable. Set against it: an unusual concentration of executive power with no controlling shareholder to check it, a demonstrated share-price sensitivity to the fate of one individual, and a major acquisition that has not yet worked. Investors should weigh the Ausnutria record specifically when assessing whether the discipline shown in dividends extends to M&A, because the evidence on those two dimensions currently points in different directions.
XI. Competitive & Structural Analysis: Why Yili Wins (and Where It Might Not)
Strip away the narrative and ask the question a sceptical analyst would ask: what, mechanically, prevents a competitor from taking Yili's business?
Running the company through Hamilton Helmer's 7 Powers framework produces a short list of genuine advantages and a longer list of things that look like advantages but are not.
Scale economies are the real one, and they operate through distribution rather than manufacturing. Yili's distributor network, cold-chain nodes, and in-store execution capability reach further into China's lower-tier cities and rural counties than anyone except Mengniu. The economics are simple: fixed network cost spread over the largest revenue base in the industry produces the lowest cost per point of distribution. That is why Yili can launch a national product faster and cheaper than a regional player, and why its roughly 19% selling expense ratio buys more shelf presence than a smaller competitor's would.14
Counter-positioning through regulation is the second, and it is unusual. The post-2008 infant formula registration regime, national testing standards, and traceability requirements impose fixed compliance costs that a large player absorbs comfortably and a small one cannot. This is a moat the incumbents did not build — the regulator built it for them, in response to a disaster the incumbents were partly implicated in. It is durable so long as the regulatory framework holds, which on current evidence it does.
Branding is where the analysis has to get honest. Yili has genuine brand equity, particularly the safety association that Chinese consumers attach to large national dairy names, and particularly in premium sub-brands like Satine. But in the base liquid milk category, brand loyalty is thin. Consumers switch on price and promotion, which is exactly why both duopolists spend so heavily on trade support. If brand were a strong power here, the selling expense ratio would be falling, not sitting near a fifth of revenue.
Switching costs, network effects, cornered resources, and process power are all essentially absent. There is no lock-in on a milk carton. There is no network effect in dairy. Yili's upstream milk access through Youran is close to a cornered resource in theory, but China currently has a milk surplus, which means secure supply is worth considerably less than it was five years ago.
So the honest structural verdict is: two real powers, one weak one, four absent. That is a solid, defensible, but not impregnable competitive position — consistent with a company that holds roughly a quarter of a large national market and cannot easily take much more of it.
The risk radar, sized to what actually moves the stock
Structural demand decline ranks first, and it has two distinct legs. The birth-rate collapse directly shrinks the infant nutrition addressable market on a multi-year lag. Consumer downgrade, meanwhile, pressures the mix in liquid milk — households buying standard rather than premium UHT. These are not cyclical fluctuations awaiting mean reversion; the demographic leg in particular is close to arithmetically locked for the next several years.
Raw milk cycle volatility ranks second, transmitted through the Youran stake and through inventory and powder valuations. The cycle appears to be bottoming — output growth has flattened and herd liquidation has run its course — which would turn this from a drag into a modest tailwind. But the same mechanism that produced FY2024's outsized profit decline will operate in reverse if milk prices overshoot upward, compressing processing margins even as the equity stake recovers.
Promotional intensity ranks third. Feihe's subsidy programme demonstrated that a competitor willing to spend 6% of revenue to defend share can force the whole category into a margin reset.19 Yili's share gains in formula came during precisely this period, which raises the fair question of how much was won with product and how much was bought with promotion.
Food-safety tail risk is low-probability and extreme-consequence, for reasons the 2008 section established. It is not a risk that shows up in a model. It is a risk that shows up in whether the equity exists.
Governance and key-person risk is real but secondary — material enough to price, not material enough to be the thesis.
Two risks that get airtime elsewhere deserve to be downgraded here. Technology disruption is largely irrelevant to the mechanics of processing and shipping milk; the interesting technology question in Chinese dairy is channel evolution, which is a distribution issue already covered. Geopolitical and supply-chain risk exists — Ausnutria's own profit alert cited international logistics volatility and tightening regulatory oversight — but Yili's revenue remains overwhelmingly domestic, which limits the exposure.17
Where an activist would push
A sceptical investor with a stake and a willingness to write letters would focus on three things.
First, carrying value. Given Ausnutria's guided half-year loss and its history of impairments, is the stake carried at a realistic value on Yili's balance sheet, and what assumptions underpin it? The company has not disclosed a detailed impairment sensitivity, and this is the single most consequential accounting judgment in the accounts.
Second, payout sustainability. If a meaningful share of FY2025's 37% profit growth reflects a favourable base rather than structural margin improvement, then the payout ratio is being assessed against an earnings number that may not repeat. The 75% floor commitment is a genuine constraint on management, but constraints made in good years are tested in bad ones.
Third, portfolio complexity and accountability. Yili holds a minority position in a listed upstream supplier that sells it nearly all of its output, a minority-controlled listed infant formula subsidiary in Hong Kong, and wholly owned overseas manufacturing. Each was individually justifiable. Collectively they produce an earnings stream with several moving parts that shareholders cannot easily model, and a structure where poor performance in an affiliate flows to the parent's P&L without the parent having full operational control. An activist would ask what management's specific plan is if Ausnutria's losses continue, and whether there is a point at which the company would sell.
XII. Bull vs. Bear: The Investment Case
The bull case
The bull case starts from the observation that Yili is winning share in a consolidating industry and returning nearly all of its profit to shareholders while doing it.
The franchise is the clear number one in Chinese dairy — roughly 24.4% of the market against Mengniu's 17.9% as of 2023, with the revenue gap having widened further since as Mengniu's top line contracted faster.[^9]11 Yili holds the top position in liquid milk and in milk powder overall, has led cold drinks for thirty-one consecutive years, and claims the leading infant formula share.12
The growth engines are diversifying away from the mature core. Milk powder and dairy products grew 14.26% and cold drinks 12.39% in the first half of 2025 while liquid milk declined — meaning two-fifths of the company is growing at low-to-mid double digits and is structurally higher-margin than the part that is not.13 Overseas, infant goat formula revenue grew 50.7% and ice cream 10.2% in FY2025, from a small base but in the right direction.1 On the FY2025 results call, Pan Gang identified deep-processed dairy, adult nutrition, and international as the three growth priorities for the next five years — a strategic articulation that is at least consistent with where the disclosed growth is actually coming from.15
The capital returns are the strongest single pillar. A committed payout floor of 75%, a per-share dividend floor of RMB 1.22, share buybacks with cancellation, and RMB 50.9 billion of cumulative distributions constitute a shareholder-return record that few Chinese consumer companies match.2425 For an investor whose thesis is "consolidation winner in a mature category, harvested responsibly," that is the core of the case.
And the raw milk cycle should turn from headwind to tailwind. With national output growth flat at 0.3% in 2025 and herd liquidation complete, upstream losses should normalise, removing the impairment drag from reported earnings while input costs remain historically low.3
The bear case
The bear case is not that Yili is a bad company. It is that Yili is a very good company in a category that has stopped growing, and that the strategy chosen to address that has not yet worked.
Liquid milk's contraction may be structural rather than cyclical. China's per-capita dairy consumption penetration story has largely played out in the urban markets that matter most economically, and the substitutes competing for the same consumption occasion — plant-based drinks, RTD coffee and tea — are proliferating. One quarter of positive growth in Q1 2026 against a weak base is not sufficient evidence of a turn.15
Ausnutria increasingly reads as an expensive, poorly timed deal. Roughly US$800 million and a 33% control premium purchased a business now guiding to a substantial half-year loss, with recurring impairments and inventory adjustments, bought immediately before the steepest phase of China's birth-rate decline.1617 Management has not disclosed a specific remediation plan or a point at which it would reassess the holding.
Vertical integration has become a source of earnings volatility rather than protection, as FY2024's profit decline running at more than twice the revenue decline demonstrated.9 Investors are being asked to own a processing business and a leveraged bet on the raw milk cycle in the same security.
Promotional intensity may compress margins even in the scenarios where Yili wins. Share gains achieved during an industry-wide subsidy war are share gains of uncertain quality. And the FY2025 margin improvement is not yet distinguishable, from public disclosure alone, from a favourable base effect.
Finally, governance. Concentrated executive power, no controlling shareholder, a demonstrated share-price sensitivity to rumours about one individual, and leveraged personal shareholdings add a layer of key-person risk uncommon for a company of this scale.27286
The synthesis
Yili has a credible, evidence-backed answer to "why do we win": distribution density that competitors cannot cheaply replicate, scale economics in a fragmenting-to-consolidated market, regulatory barriers that favour incumbents, and a capital allocation record that has survived a genuine downturn intact.
It has a much thinner answer to "why do we grow in a shrinking market." The infant formula share gains are real but were achieved in a contracting category during a subsidy war. The overseas businesses are growing fast from bases too small to move the group. The ice cream franchise is excellent but is roughly 13% of revenue. And the single largest capital commitment made in service of the growth question — Ausnutria — has, four and a half years in, produced losses rather than earnings.
The bet embedded in Yili's equity is therefore quite specific. It is not a bet on Chinese dairy growing. It is a bet that Yili can take enough share, hold enough margin, and return enough cash that the shrinking of the market matters less than the consolidation of it. The evidence for the first and third parts of that bet is good. The evidence for the second is one year old.
XIII. Durable Lessons and What to Watch
Three lessons generalise well beyond dairy.
The first is that a catastrophe can be the best thing that ever happens to the survivors of an industry. The melamine scandal was a genuine human tragedy and it wrecked several companies. It also created, in the space of about two years, a regulatory regime and a consumer preference structure that handed the two largest national players a duopoly they had not been able to win competitively. Investors analysing any regulated consumer industry after a scandal should ask not only who was hurt, but who ended up structurally better positioned once the rules were rewritten. The answer is very often the largest incumbent that survived.
The second is that vertical integration is a position, not a hedge. Yili's control of upstream milk supply was rational, defensible, and correctly motivated by food safety. It also meant that when the raw milk cycle turned down, Yili owned both the benefit — cheaper inputs — and the cost — equity losses and impairments in the supplier. The net effect was to amplify rather than dampen earnings volatility. Any company that owns a meaningful stake in its own supply chain has converted a purchasing relationship into a cyclical exposure, and the accounts will show it.
The third is that capital-return discipline matters more, not less, when growth stalls. The temptation for a company facing a maturing core business is to spend its way into a new one. Yili has done some of that, with mixed results. But it has simultaneously maintained one of the most consistent distribution records in Chinese equities, including through its worst year in two decades. For a business whose growth question is genuinely unresolved, the cash returned is the part of the return that does not depend on the strategy working.
Three KPIs worth tracking
Milk powder and dairy products segment revenue growth and margin. This is the cleanest read on whether the Ausnutria bet and the broader nutrition strategy are working. If this segment can sustain double-digit growth with stable or improving margin while births continue to fall, it means Yili is genuinely taking share profitably. If growth slows toward flat, or if growth persists but margin deteriorates, it means the share is being bought with promotion — which is a fundamentally different and much less valuable outcome.
Youran-linked investment income and impairment. The line where equity-method results and impairments on the upstream stake land is the single cleanest signal of whether the raw milk cycle is a drag or a tailwind. Watch for the point at which it stops subtracting from profit. That transition, more than any operating improvement, will determine how much of a future profit increase is real versus base effect.
Dividend payout ratio and buyback pace against the stated 75% floor and RMB 1.22 per share commitment. This is the test of whether capital allocation discipline holds through a growth slowdown. Management has published specific numbers; investors should hold them to those numbers, and treat any drift below the floor, or any large acquisition funded at the expense of it, as material new information about priorities.
What would change the thesis
Three developments would meaningfully shift the picture. A sustained recovery in raw milk farm economics — farm-gate prices durably above production cost — would remove the upstream drag and validate the vertical integration strategy, at the cost of some processing margin. Evidence that Ausnutria's losses are stabilising rather than deepening, with the "one-off" charges genuinely ceasing to recur, would begin to rehabilitate what is currently the weakest item in Yili's capital allocation record. And a sustained reacceleration in liquid milk volumes across several quarters — not one — would suggest the 2024 contraction was a cyclical destocking episode after all, rather than the beginning of a structural decline in China's largest packaged beverage category.
XIV. Epilogue & Outro
Thirty years after a Hohhot dairy raised RMB 96.9 million on the Shanghai Stock Exchange, Yili is a company of roughly RMB 116 billion in annual revenue, seventy-seven production bases, and distribution into more than eighty countries.412 It is the largest dairy company in Asia and the most disciplined capital returner in its industry, with more than RMB 50 billion of cumulative dividends behind it and a numerically specific commitment to keep going.2425
It is also a company still defined by two things it did not choose. The first is a scandal it survived — the melamine crisis that consolidated an entire industry around the companies large enough to absorb the new rules, and that left behind a tail risk which will never fully go away. The second is a demographic decline it cannot outspend, running through the exact segment on which the next decade of value creation was supposed to depend.
The competitive question that animated Yili's first three decades — could it beat Mengniu — has been answered, and answered decisively enough that it is no longer the interesting question. Mengniu is a competent number two working through its own repairs, and the gap has widened rather than closed.
The question for the next decade is different and considerably harder. It is whether a company can keep growing profit in a country that is having fewer babies and, for now at least, drinking less milk. Yili's answer so far has been to take share, hold margin, control costs, and return nearly everything it earns. That is a coherent strategy and, on the evidence, a well-executed one. Whether it is sufficient is not yet knowable — and any investor who tells you otherwise is reading a story rather than the accounts.
References
-
Yili Group Reports FY2025 Dual Growth in Revenue and Profit, Marking 500-Fold Revenue Surge in 30 Years Since Its Listing — GlobeNewswire, 2026-05-15 ↩↩↩↩↩↩↩↩↩
-
Yili Group Reports FY2025 Dual Growth in Revenue and Profit — Yili Group official news, 2026-05-15 ↩↩↩↩↩↩
-
Statistical Communiqué of the People's Republic of China on the 2025 National Economic and Social Development — National Bureau of Statistics of China, 2026-02-28 ↩↩↩↩
-
Yili's Dairy Dominance: From China to Global Markets — CKGSB Knowledge ↩↩↩↩↩↩
-
'Dairy Godfather' Zheng Junhuai has returned to the industry after a spell in jail for embezzlement at Yili Dairy — Asian Extractor, 2015-01-11 ↩↩
-
The 2008 Milk Scandal Revisited — Yanzhong Huang, Forbes, 2014-07-16 ↩
-
Yili Reports FY2024 Revenue of 115.8 Billion Yuan, Reinforcing Its Position as Asia's Leading Dairy Company — GlobeNewswire, 2025-05-08 ↩↩↩↩
-
China Mengniu Dairy FY2024 Annual Results presentation — Mengniu Investor Relations ↩
-
Mengniu Dairy 2025 presentation: revenue dips but cash flow soars — Investing.com, 2026-03-26 ↩↩↩↩
-
Earnings call transcript: Mengniu Dairy sees revenue dip, cash flow highs in H2 2025 — Investing.com, 2026-03-26 ↩
-
伊利股份上半年扣非净利润大增31.78% 乳业龙头重新定义国奶发展新高度 — 证券时报 (STCN), 2025-08 ↩↩↩↩↩
-
诚通证券-伊利股份-600887-2024年年报2025年一季报点评 — Sina Finance, 2025-05-23 ↩↩↩
-
伊利股份年度及一季报实现营收利润双增 全品类全面发力开启新增长周期 — Sina Finance, 2026-04-30 ↩↩↩↩
-
Chinese Dairy Giant Yili Files Plan to Buy One-Third of Rival Ausnutria — Caixin Global, 2021-10-28 ↩↩↩
-
Ausnutria 2026 Interim Profit Alert: Revenue Expected to Reach Approximately RMB3.065 billion to RMB3.165 billion — ACN Newswire, 2026-07-24 ↩↩↩↩
-
Baby bust: China births plunge 17% in 2025 to historic low as population shrinks — South China Morning Post, 2026-01 ↩
-
What Feihe's infant formula subsidy plans mean for Danone, Yili — DairyReporter, 2025-04-02 ↩↩↩↩↩
-
China's infant formula sales up for major players in first half of 2025 — NutraIngredients, 2025-09-01 ↩↩
-
Yili set to buy NZ's Westland for $404m — DairyReporter, 2019-03-18 ↩↩
-
Chinese Dairy Farmers Face Dark Times as Supply Glut Shows No Sign of Abating — Yicai Global ↩↩↩↩
-
Something sour at Yili milk — The China Project, 2018-10-24 ↩↩↩↩
-
China's top dairy maker Yili Group sees shares rocket 10pc on reappearance of chairman Pan Gang — South China Morning Post, 2018-05-31 ↩↩↩
LinkedIn
WhatsApp
Reddit
Hacker News
Email
Kindle
RSS Feed
Spotify
Apple Podcasts
Amazon Music
Audible
YouTube