Nongfu Spring: The Loneliest Wolf and China's Water Wars

I. Introduction & Episode Roadmap

On a Wednesday in late February 2024, a 69-year-old billionaire walked into a memorial hall in Hangzhou to pay respects to a man he had spent three decades fighting. The dead man was 宗庆后 Zong Qinghou, founder of 娃哈哈 Wahaha, the beverage company that had defined China's mass-market drinks industry for a generation. The mourner was 钟睒睒 Zhong Shanshan, founder of 农夫山泉 Nongfu Spring, once a Wahaha distributor, later Zong's most bitter rival, and by then the richest person in China.

He came to pay respects. Within days, the internet decided he had come to gloat.

What followed was not a product recall, not an accounting restatement, not a competitive loss. It was a nationalist social-media pile-on that accused Nongfu's packaging of looking too Japanese, resurfaced a thirty-year-old commercial grudge as a moral failing, and seized on a fact disclosed in the company's own IPO prospectus — that Zhong's son and board member held American citizenship. Roughly HK$30 billion, close to US$3.8 billion, came off Nongfu Spring's market value inside two weeks.¹²

Then the damage kept compounding. By the end of 2024, revenue from packaged drinking water — the business the entire company was built on — had fallen 21.3% for the full year.³ Not a few points. A fifth of the franchise, gone, from a category with no meaningful technological change, no new entrant, and no product defect.

That is the puzzle this story is about.

The mechanism deserves to be stated precisely, because "boycott" undersells it. Nothing about Nongfu Spring's product changed. Nothing about its cost position changed. No competitor launched anything. The water in the bottle in March 2024 came from the same lakes and springs it had come from in January. What changed was a shared belief among enough consumers, distributors and shopkeepers that carrying this particular brand had become socially expensive — and that belief propagated through a retail system built on millions of small, independent stocking decisions.

Nongfu Spring is, on the numbers, one of the strongest consumer businesses in Asia. In 2025 it recorded revenue of RMB52,553 million and profit attributable to owners of RMB15,868 million, up 22.5% and 30.9% respectively, at a gross margin of 60.5% — a margin structure that would be enviable for a luxury goods house, let alone a company that sells water in plastic bottles.⁴ It has spent thirty years building a physical asset base — permitted natural water sources, bottling plants built at the source — that a competitor cannot simply copy with capital.

And yet the single largest value-destroying event in its listed history had nothing to do with any of that.

The framing question is therefore not the usual one. It is not "does this company have a moat." It plainly has something. The question is what happens to a structural advantage when the threat is not a competitor but the internet itself — and whether an investor can underwrite a business whose worst drawdown was triggered by a rival founder's funeral.

It is worth being clear at the outset about the scale of the thing, because the numbers are easy to under-feel. Nongfu Spring employs more than 28,000 people and spent RMB5,100 million on employee benefits in 2025.⁵ It operates seventeen permitted natural water sources across mainland China, from a volcanic cluster in Hainan to glacier meltwater in Xizang. It sells its products through a distributor network that reaches into township and village retail, and it derives more than 99% of its revenue from a single country.⁴ This is a very large, very concentrated, very physical business.

The story runs through the founder's improbable origin, the water-source bet that defined the model, the 2020 listing that briefly made Zhong the wealthiest man in China, the tea business that quietly became larger than water, the 2024 crisis, the 2025 and 2026 recovery, and the succession question nobody at the company has yet answered in public.

Throughout, the discipline is the same: separate what management says from what the filings prove, and test every claim about durability against the strongest disconfirming evidence in the company's own thirty-year record.

↑ Back to Top

II. Zhong Shanshan 钟睒睒: The "Loneliest Wolf" (1954–1996)

Zhong Shanshan was born in 1954 and did not finish formal schooling. The Cultural Revolution took care of that. He worked construction. He grew mushrooms. He sold curtains. By his own account and the accounts of Chinese business press over decades, he failed at most of it.

His first real break came as a journalist for the Zhejiang Daily, a job that gave him something more useful than a salary: a professional habit of asking who benefits from a story, and how a story gets made. It is impossible to read the next forty years of his career — the media counterattacks, the lawsuits against newspapers, the studied refusal to court reporters — without noticing that the man who became China's most press-hostile executive learned the trade from the inside.

The reporting years also gave him a map. A journalist in reform-era Zhejiang spent his days inside factories, county governments and agricultural collectives at exactly the moment those institutions were being rewired for private commerce. He saw where the raw materials were, who controlled the permits, and how a provincial official decided which outsider got to build something in their county. Thirty years later, the core operational skill of Nongfu Spring is still the ability to negotiate long-term extraction rights and factory sites with local governments in remote mountain districts. That is not a marketing skill. It is a reporter's skill, industrialised.

Before that, there was the failure period, which he has never particularly hidden. He went south to Hainan during the island's boom years and tried several ventures, including a mushroom-growing operation that collapsed when the climate turned out not to cooperate. The details vary by telling, but the shape does not: a man in his thirties with no degree, no capital and no network, losing money on a string of small businesses in a country that was inventing capitalism in real time.

In March 1993 he founded 养生堂 Yangshengtang, a health-products company, and served successively as its director, general manager and chairman.⁵ The early product line was the stuff of 1990s Chinese consumer capitalism at its most exuberant: health tonics, including a turtle-blood preparation marketed on the promise of vitality. It sold. It also taught him the single most valuable lesson of his commercial life — that in a market with low institutional trust, the story a consumer believes about where a product comes from can be worth more than the product's measurable properties.

Yangshengtang still exists, still privately held, and still wholly owned by Zhong. It matters to this story because it is the vehicle through which he controls Nongfu Spring, and because it remains a counterparty to the listed company today.⁵

In September 1996 he set up Zhejiang Thousand-Island Lake Yangshengtang Drinking Water Company, the predecessor of Nongfu Spring.⁵ He was 42, twice-failed, and entering a market that already had a dominant incumbent.

The contrarian bet was simple and, at the time, commercially insane. 娃哈哈 Wahaha, 康师傅 Master Kong and the rest were selling purified water: municipal supply run through reverse osmosis, cheap, consistent, and made anywhere. Zhong decided to sell natural water pulled from a specific, named, geographically fixed place — starting with 千岛湖 Qiandao Lake in Zhejiang — and to build the factory at the source and bottle it there rather than ship bulk water to a plant near the customer.

That decision made his cost structure worse. It made his logistics harder. It constrained his growth to the pace at which he could find and permit new sources. And it gave him the only thing that mattered: a claim no purified-water competitor could make.

Think about the freight arithmetic for a moment, because it explains why nobody else did this. Water is heavy, cheap, and mostly worthless per kilogram. The entire economic logic of purified water is that you make it close to where you sell it, so you never pay to move the water itself very far. Nongfu inverted that. It committed to producing in whichever remote mountain valley happened to contain the aquifer, and then paying to truck finished product across a continent-sized country to the consumer. Every competitor looking at that plan in 1996 would have concluded, correctly, that the unit economics were worse. What they would have missed is that the extra freight cost buys a permanent product difference, and a permanent product difference eventually buys a price premium that more than covers the freight.

The advertising line that carried it was five characters — 农夫山泉有点甜, "Nongfu Spring tastes a little sweet." It is a strange claim. Water does not taste sweet. That was the point. The slogan implied that this water was different in kind, and it invited the consumer to test the proposition personally, which is the most durable form of brand building there is.

Zhong's personal style hardened into the shape it still holds. He gave almost no interviews. He did not attend industry gatherings. He was described in the Chinese press as 独行侠 — a lone operator — and eventually as the "loneliest wolf." He ran his company as a nearly closed system, making the brand decisions and the human-resources decisions himself, a practice the 2025 annual report still describes in the present tense.⁵

The rivalry with 宗庆后 Zong Qinghou dates from this period and is worth understanding, because it was later retold as a morality tale. Zhong began as a Wahaha distributor in Hainan and Guangxi before going independent. When Nongfu subsequently built its entire marketing platform on the proposition that natural water is healthier than purified water, it was, by direct implication, telling consumers that Wahaha's product was inferior. In April 2000 Zhong publicly declared that purified water contained no minerals and was not good for people, and announced Nongfu would stop making it altogether. The industry's response was collective and furious: dozens of purified-water producers, with Wahaha prominent among them, issued a joint statement accusing Nongfu of unfair competition, and the company was fined RMB200,000.⁶ That grievance sat dormant in Chinese business memory for twenty-four years and then reappeared, fully formed, in February 2024.

Keep that personality in mind. In 2024 it would determine, for months, how a listed company with tens of billions of dollars of market value responded to the worst reputational crisis in its history.

↑ Back to Top

III. The Water-Source Bet as a Business Model (1996–2013)

To understand why Nongfu Spring's water business is structurally different from a soft-drink business, start with a distinction that most Western consumers never have to think about, because their tap water is drinkable.

China's packaged water market splits into two economically different products that sit next to each other in the same cooler.

Purified water is manufactured. You take municipal or ground water, strip everything out of it with reverse osmosis, and bottle it. The input is available anywhere, the process is standard, and the resulting product is essentially identical across brands. Competition therefore runs on price, shelf placement, and distribution density. It is a commodity conversion business.

Natural water and mineral water are extracted. The product is defined by a specific aquifer, spring or lake, and the right to draw from it requires a local extraction permit and, in practice, a long relationship with a local government that has environmental obligations of its own. You cannot manufacture a new Changbai Mountain spring. The number of good sites is finite, and each one, once permitted and built out, is unavailable to everyone else.

That is as close to what Hamilton Helmer calls a cornered resource as a beverage business gets: preferential access to a coveted asset, obtained on terms that independently enhance value. Nongfu spent three decades accumulating them one at a time. By the end of 2024 it had twelve major sources.³ By the March 2026 results it had sixteen, having added 八大公山 Badagong Mountain in Hunan, 龙门山 Longmen Mountain in Sichuan and the 念青唐古拉山 Nyainqêntanglha range in Xizang during 2025, then 轿子雪山 Jiaozi Snow Mountain in Yunnan.⁴ By the August 2026 interim report it had seventeen, with the Leiqiong Haikou volcanic cluster in Hainan added in April.⁷

Note the pace. Roughly one new source every eight months, over thirty years. This is not a business you can accelerate with a funding round.

The company has also spent heavily on making the sources visible, which is the part most investors skip and which is arguably the whole strategy. Nongfu runs a campaign it calls 天然水源，透明工厂 — "Natural Water Sources, Transparent Factories" — that turns its bottling plants into destinations. In 2025 its industrial-tourism programme drew more than 700,000 visits, of which nearly 40% were participants in structured water-source research tours, mostly young people.⁴ It sent employees and their families on more than a hundred source-exploration trips reaching roughly 5,000 people between May and November. It brought in the Olympic swimming champion 潘展乐 Pan Zhanle as brand ambassador and filmed him at the sources.⁴

Read cynically, that is a marketing budget with a tour bus attached. Read structurally, it is something more interesting: an attempt to convert an abstract quality claim into a physical memory. A consumer who has walked through the plant at Changbai Mountain has a defence against an internet rumour that a consumer who has only seen an advertisement does not. After 2024, that distinction stopped being theoretical.

The source portfolio also generates product optionality that a purified-water competitor structurally cannot match. Nongfu has built out a water matrix aimed at specific use cases — infant formula preparation, tea brewing, meetings, dining, outdoor sport — and in June 2025 launched a transparent edible ice product sold through Sam's Club, on the logic that good water makes good ice.⁴ None of these are large. Collectively they demonstrate that owning the input lets you segment a commodity into a portfolio, which is the mechanism by which a low-value liquid becomes a 60%-gross-margin business.

It is also worth being precise about what the resource protects. It protects the premium claim. It does not protect the price. Any bottle of water on the shelf substitutes physically for any other, and a thirsty consumer at a convenience store faces essentially zero switching cost. What the sources buy Nongfu is the right to charge a few jiao more and to be believed when it says its water is healthier — a differentiation that lives in the consumer's head and is therefore vulnerable to anything that changes what is in the consumer's head.

Which is exactly what happened for the first time in 2013.

On April 10 that year, the Beijing Times published the first of what became a sustained campaign of reporting alleging that Nongfu's bottled water was held to a provincial Zhejiang standard, DB33/383-2005, rather than the national standard — and that on certain measures its barrelled water was, in the paper's framing, of lower quality than tap water.⁸ The company's response was not a press conference. It was a lawsuit. Nongfu sued for RMB200 million, said the coverage had cost it more than RMB2 billion in sales, and told the court the paper had run 67 pages of material against it inside a month.⁹ The paper counterclaimed, demanded an apology, and asked for one yuan in damages.

Two things about that episode matter for everything that follows.

First, it established the shape of the threat. Nongfu's premium is a trust premium. A trust premium can be attacked directly through the media in a way that a cost advantage cannot. A rival with a cheaper plant has to build the plant; a rival — or an unaffiliated columnist, or an anonymous account — who can make consumers doubt the water needs only a keyboard.

Second, it established the shape of the response. Zhong's instinct under attack was, and remains, legal and combative rather than explanatory. He did not spend 2013 on television reassuring consumers. He sued, and he publicly suggested the campaign had been orchestrated by a state-owned competitor — 华润饮料 China Resources Beverage, owner of the 怡宝 C'estbon brand — an accusation never established in court.

The company survived. It kept its category leadership. By the 2010s Nongfu was at or near the top of China's packaged water market by value, and its distribution had pushed into township- and village-level retail in a way that few competitors could match. The 2013 attack cost real money and left no permanent scar.

For an investor, that is the correct and uncomfortable read: the moat held, and the vulnerability was real. Both. The 2013 episode did not falsify the source-based advantage. It bounded it — showing that the advantage is a quality advantage, not an immunity, and that its cash flows can be interrupted by events that no competitive analysis would ever forecast.

Eleven years later, the same mechanism fired again, at roughly ten times the scale. But before that, there was a party.

↑ Back to Top

IV. The 2020 IPO: Briefly the Richest Man in China

For twenty-four years Nongfu Spring did not need public money. It was profitable, cash-generative, and wholly controlled by a man with no interest in explaining himself to strangers. Then, on September 8, 2020, it listed on the Hong Kong Stock Exchange, and the market lost its composure.

The retail tranche was subscribed 1,148 times over — at the time the most oversubscribed offering in the exchange's history — with institutional demand covered roughly 60 times.¹⁰ The deal raised about US$1.08 billion at HK$21.50 a share.¹⁰¹¹ On debut the stock opened near HK$39.40 and traded as much as 85% above the offer price.¹²

That single day's move did something arithmetically dramatic. Zhong Shanshan owned the overwhelming majority of the company. Marking that stake to a public price, plus his separate holding in 北京万泰生物 Beijing Wantai Biological — a Shanghai-listed vaccine and diagnostics maker that had been a direct beneficiary of pandemic testing demand — briefly put his net worth around US$59 billion and made him the richest person in China, ahead of 马云 Jack Ma and 马化腾 Pony Ma.¹³

Step back from the fireworks for a second and ask the obvious question: why list at all?

Nongfu did not need the capital. It was throwing off cash, it had no acquisition programme, and its founder had spent twenty-four years demonstrating a preference for answering to nobody. The listing brought disclosure obligations, a Hong Kong regulator, an audit committee, independent directors, and — crucially — a daily public price on his life's work. In hindsight, the most consequential thing the IPO produced was not money. It was a mark-to-market on Zhong Shanshan's net worth, published every day, which is precisely the number that would later be used against him.

Two features of that listing deserve more scrutiny than they usually get.

The first is what the company did with the money, which is: not very much, for a very long time. The listing and over-allotment raised net proceeds of about HK$9,377 million. As of December 31, 2025 — more than five years later — Nongfu had deployed HK$5,080 million of it, or 54.2%, leaving HK$4,297 million unspent.⁴ The board reallocated HK$726 million from "strengthening fundamental capabilities" to buying production facilities and building factories on March 24, 2026, and extended the utilisation deadline to December 31, 2027. That deadline had already been extended once before, from August 2024 to the end of 2026.⁴

Read that generously and it is a company that refuses to spend money badly. Read it skeptically and it is a company that raised over a billion dollars it did not have a concrete use for, and has now revised its own spending timetable twice. Both readings are defensible. What is not defensible is calling it disciplined without noting the repeated extensions.

The second feature is the governance fact the prospectus disclosed and nobody priced: Zhong's son and non-executive director held foreign citizenship. It sat in the offering document, publicly available, entirely lawful, and utterly inert — for three and a half years.¹¹

This is a genuinely useful lesson about how information behaves in this market. The fact was never hidden. Analysts read the prospectus. Cornerstone and institutional investors did diligence. Everybody who cared to know, knew. The fact simply had no price, because nothing in the environment made it salient. Then one funeral made it salient, and it repriced instantly. Risk that sits dormant in a disclosure document is not the same as risk that has been assessed and dismissed — and the difference only becomes visible after the fact.

There is also a structural point buried in the 84% stake. When one holder owns that much, the free float is small, and a small float amplifies price moves in both directions. The 1,148-times retail subscription and the 85% debut pop were partly a function of genuine demand and partly a function of there simply not being many shares. The same arithmetic runs in reverse when sentiment turns.

Since the January 2021 high, the shares have gone through a long derating. Much of the 2020–21 valuation was pandemic-era multiple expansion, not durable value: the business grew, and the multiple contracted faster. As of the most recent data available in September 2026, the shares had returned roughly 10% in total over three years and were down around 9% year-to-date, trading near 25 times earnings against a peer group closer to 17.¹⁴ An investor who bought the story at the top has spent five years being right about the company and wrong about the stock.

That is worth sitting with before the next section, because the growth engine that carried the business through the derating was not water at all. It was a tea nobody wanted.

↑ Back to Top

V. Building the Second Engine: Beyond Water

In 2011 Nongfu Spring launched an unsweetened ready-to-drink tea called 东方树叶 Oriental Leaf — sold in international markets and the company's English filings as East Leaf. Chinese consumers hated it. It was regularly voted onto internet lists of the worst-tasting beverages in China. Sweet tea was the market; unsweetened brewed tea in a bottle tasted, to a palate raised on sugar, like something that had gone wrong.

The obvious commercial decision was to kill it. Nongfu kept it on the shelf for more than a decade.

Here is the deepest structural fact about the company today, and it surprises almost everyone who still thinks of Nongfu as a water business: packaged water is no longer its largest segment.

In 2025, tea beverages generated RMB21,596 million, up 29.0%, and accounted for 41.1% of group revenue. Packaged water generated RMB18,709 million, 35.6% of revenue.⁴ In the first half of 2026 the gap widened further: tea RMB13,122 million, up 30.1% and 44.2% of the mix, against water at RMB9,641 million, up 2.1% and 32.4% of the mix.⁷

The profit picture is starker still. Nongfu discloses segment results, and in 2025 the tea business threw off RMB10,374 million of segment profit against RMB7,001 million for water.⁴ Tea is not merely bigger. It is the larger profit pool, and it earns a higher margin on revenue than the water business it overtook.

That inversion is the single most important thing to understand about this company's current economics, and it changes the investment question. The cornered-resource argument — the thing that makes Nongfu genuinely hard to copy — now protects a minority of revenue and a minority of profit.

So what is protecting the tea business?

Three things, in descending order of durability. Distribution, which is real: Nongfu reaches into township and village retail with a density built over three decades of selling water, and a new tea brand cannot rent that. Upstream integration, which is emerging: the company has been building and donating modern tea-processing plants in growing regions such as 普洱 Pu'er and 临沧 Lincang in Yunnan, and running blockchain-based traceability from fresh leaf through processing to finished product — the same "we control the source" logic transplanted from water to tea leaves.⁴ And brand, which is the least durable of the three, because brand is what 2024 proved can be taken away.

There is a quieter piece of evidence for where the company is actually putting its money, and it sits in a table most readers skip. Segment depreciation and amortisation in 2025 was RMB1,113 million for tea against RMB931 million for water.⁴ Depreciation is the shadow that capital expenditure casts a few years later. The company has already been building more tea capacity than water capacity, which means the mix shift now visible in revenue was decided in the capital budget several years ago. Management did not discover the tea business in 2025. It was building for it while the market was still calling Nongfu a water company.

The product cadence tells the same story. Oriental Leaf runs an annual ritual it calls "one spring, one autumn" — a limited 龙井 Longjing spring tea made from pre-Qingming leaves, launched for the fifth consecutive year in April 2026, and an osmanthus oolong in autumn — which builds a seasonal purchase habit of the kind normally associated with agricultural luxury rather than packaged beverages.⁴⁷ Around that spine, the company added a 1.5-litre family bottle for Chinese New Year, chenpi white tea and chrysanthemum pu-erh flavours in 2025, and 335ml and 900ml white tea formats in 2026.⁴⁷ In April 2026 it launched a cold-extracted Longjing sold through cold-chain distribution at Sam's Club, which held the top position on that retailer's sugar-free beverage bestseller list for several consecutive months.⁷

The 茶π Tea π line — original tea extract blended with fruit juice, aimed at younger consumers through music festivals — went through a reformulation in the first half of 2026 that cut sugar across every product in the series by more than 25%.⁷ That is a revealing move. A company confident in its franchise reformulates a working product toward where it thinks the consumer is going. A company managing decline does not touch the recipe.

The category position is, for now, dominant. Industry research puts Oriental Leaf at roughly three-quarters of China's sugar-free tea category, well ahead of サントリー Suntory, the Japanese early mover.¹⁵ Across the whole ready-to-drink tea market, including sweetened products, Nongfu's share of sales rose from 23.7% in the first quarter of 2025 to 31.1% in the first quarter of 2026, while 康师傅 Master Kong slipped from 27.4% to 26.8%.¹⁶

Now the disconfirming evidence, in the same passage rather than buried in a risk list. The same industry research that credits Oriental Leaf with category dominance also records that sugar-free tea sales in China posted their first-ever negative growth in 2025, with penetration in first- and second-tier cities approaching saturation — and that a challenger brand, 果子熟了 Guozi Shule, grew more than 800% to take third place in the category.¹⁵ Nongfu's tea revenue grew 29% in a year the category itself stopped growing. That is share capture, not market growth.

Share capture is a fine thing to own. It is also, mechanically, a finite thing. A company taking share in a flat category is running down a resource. The tea growth rate and the category growth rate are two different numbers, and only one of them has been compounding.

The rest of the portfolio is smaller but consistent. Functional drinks, led by 尖叫 Scream and the Victory Vitamin Water line, reached RMB5,762 million in 2025, up 16.8%, and grew a further 15.5% to RMB3,348 million in the first half of 2026 on the launch of a new electrolyte range advertised around the World Cup.⁴⁷ Juice — the "17.5°" and NFC lines, sold on a not-from-concentrate, no-additives platform — reached RMB5,176 million, up 26.7%.⁴ The "other" bucket holds soda water, the 炭仌 TANBING chilled coffee line, plant beverages including a birch juice and a herbal 百合麦冬汤 lily-and-ophiopogon drink, and actual agricultural produce: oranges from Jiangxi, apples from Xinjiang, jasmine from Guangxi.⁴⁷

Two things are worth noticing about that long tail. The first is the channel. A striking share of Nongfu's premium launches — the cold-pressed 900ml orange juice, the cold-brew black coffee, the transparent ice, the cold-extracted Longjing — go to market through Sam's Club rather than through the traditional distributor network. That is a deliberate second channel with different economics: higher price points, membership-warehouse volumes, and a consumer who is already paying for curation. It is where Nongfu tests premiumisation without disturbing the mass-market price architecture it protects so carefully everywhere else.

The second is the agricultural integration, which is more than corporate-responsibility decoration. Nongfu grows or contracts its own oranges, jasmine and increasingly its own tea leaf, runs standardised contract-farming models, and has built blockchain traceability across the tea chain from fresh leaf to finished bottle.⁴ The strategic logic is a straight copy of the water playbook: own the source, make the source visible, and convert provenance into premium. Whether it earns a return commensurate with the capital is not separately disclosed.

One genuine adjacency worth a sentence and not a section: in June 2025, IFBH — the Singapore-incorporated parent of the Thai "if" coconut water brand, which derives about 92% of its revenue from mainland China — listed in Hong Kong at HK$27.80 and closed its first day up 42%, past a HK$10 billion valuation. Zhong's private fund, Daohe Chengguang, participated in the international placement.¹⁷¹⁸ It is optionality and a signal about where he sees the next category, not a driver of Nongfu's earnings.

The honest conclusion on the platform thesis: the claim that Nongfu's water DNA transfers into adjacent categories is supported, not merely asserted — the segment profit numbers prove a real business was built, not a line extension. But the version that survives scrutiny is narrower than the promotional version. What transferred was distribution reach and a manufacturing philosophy. What did not transfer was the cornered resource. Tea is defended by scale and brand, and brand has already been shown to be breakable.

Which brings us to the year it broke.

↑ Back to Top

VI. 2024: The Water War

宗庆后 Zong Qinghou died in late February 2024. He was 79, and in the Chinese popular imagination he was the anti-billionaire: a man who reportedly lived on RMB50,000 a year, flew economy, ran Wahaha from a cluttered office, and had built a national brand on drinks for ordinary families. His death produced a genuine and enormous wave of public grief.

It also produced a vacuum, and the internet filled it with a comparison.

The online campaign against Nongfu Spring began on or around February 28, 2024, after Zhong attended Zong's memorial.¹⁹ It moved fast and it moved on three fronts at once, which is why it was so hard to answer.

The first front was aesthetic. Users claimed that design elements on Nongfu's packaging — architectural motifs on the Oriental Leaf tea bottles, the red bottle cap — evoked Japanese temples and the Japanese flag.¹⁹ These claims were, on any serious reading, tendentious. That did not matter. The accusation was legible in three seconds on a phone screen, and the rebuttal required a paragraph.

The second front was historical. The decades-old feud between Zhong and Zong was retold as a morality play in which Zhong, a former Wahaha distributor, had built his fortune by attacking a national champion.

The third front was personal, and it was the one that stuck, because it was true. Zhong's son, a non-executive director of the listed company, held American citizenship — a fact disclosed in the IPO prospectus. From there the argument wrote itself: is this really a Chinese company?²⁰

The damage was immediate and then it was slow. In the first two weeks the market took off roughly HK$30 billion, close to US$3.8 billion, of value.¹ Retailers pulled stock; convenience stores in Jiangsu were reported removing Nongfu products from shelves.² But the sharper lesson is what happened after the news cycle ended. On August 28, 2024, when the company reported a sharp slowdown in first-half profit growth, the shares fell more than 12% in a single session — nearly six months after the initial flare-up.²¹ The stock made its post-listing lows in September 2024.

And the operating damage, when the full year closed, was worse than the market drawdown had implied. Packaged water revenue fell 21.3% to RMB15,952 million, from RMB20,262 million the year before.³ Group revenue grew 0.5%. Profit was essentially flat.³ A business whose water franchise had compounded for three decades lost a fifth of it in twelve months to a cause that appears in no competitive framework ever devised.

Management's response is the part an investor should study hardest, because it is the only part that was a choice.

For roughly five months, the company said very little of consequence in public. Zhong did not go on television. He did not hold a shareholder call to lay out a recovery plan — Nongfu does not hold conventional quarterly earnings calls, and no transcripts of management Q&A for this period are available through standard transcript providers, which means the market spent the crisis reading press releases rather than questioning executives.

What the company did do in that window was fight on a different battlefield. In July 2024 it sent a legal demand to the Hong Kong Consumer Council over a bottled-water quality report, and the watchdog apologised and revised its assessment of Nongfu's water from 4.5 stars to 5.²²²³ The company won. It also spent its most visible crisis-period capital attacking a consumer watchdog in a different jurisdiction over a rating, while its core mainland business was down a fifth.

Then, later in the year, the posture changed. By the company's own account in the FY2024 results, Zhong gave exclusive interviews across two episodes of CCTV's Dialogue in August and December 2024, appeared at a media conference in Ganzhou in November, and directly addressed the controversies — while the company pursued administrative and legal action against accounts it characterised as rumour-mongers.³ The chairman's statement for that year described what had happened as "sustained and organized smearing attacks" and noted that water market share had declined for three straight months.³

So the fair characterisation is not that management never engaged. It is that management engaged on its own schedule, roughly five to nine months after the crisis began, through state broadcast media rather than investor channels, and that the sequencing looks like a founder deciding when to speak rather than a company executing a communications plan. Whether that was stubbornness or shrewd patience is a judgement call. What is not a judgement call is that shareholders had no articulated recovery plan to hold management to during the worst of it.

The channel damage is the part that explains why the decline lasted a year rather than a news cycle. Bottled water in China does not sell through consumer loyalty; it sells through presence. A distributor decides how many cases to take. A convenience-store manager decides how many facings to give a brand in a cooler with finite space. Once a brand becomes politically awkward, both of those people have a rational, entirely non-ideological reason to reduce their exposure — and once the facings are gone, they are refilled by a competitor and are slow to come back. That is the transmission mechanism between an internet argument and a 21.3% revenue decline, and it is why the company's own account describes market share falling for three consecutive months rather than collapsing in a week.³

There was one substantive strategic response, and it was clever. In April 2024 Nongfu relaunched a purified water product — the green bottle — in a single 550ml size, drawn from the same natural sources as its premium red-bottle water, under the slogan "the green bottle is a bit sweet, the red bottle is healthier."³ It was simultaneously a price-fighting SKU aimed straight at C'estbon's purified-water stronghold and a way to keep volume in the system while the red bottle absorbed the reputational hit.

Now the falsification test, applied where it belongs rather than in a risk appendix.

The bull claim is that Nongfu's source-based advantage makes its water franchise durable. The strongest disconfirming evidence in the company's own record is not competitive at all. It is 2013 and 2024: two occasions, eleven years apart, on which a non-economic attack on the trust component of the brand produced a material revenue and share impact without any change in product, cost position, or competitive set. The second event was roughly an order of magnitude larger than the first in economic consequence.

Weighing it properly: same business, same brand, same founder, same failure mechanism, escalating scale, and — critically — no evidence that either episode permanently dislodged category leadership. That history does not reject the moat claim. It narrows it. The defensible version is: Nongfu's water advantage is real and recoverable but not continuous, and the business should be underwritten as one that will periodically surrender a year of water revenue to a reputational shock it did not cause and cannot forecast. The frequency observed is roughly once a decade; the sample size is two, which is not enough to call it a rate.

The KPI that would confirm or falsify the narrowed version is not revenue growth. It is whether the next such episode, whenever it comes, produces a recovery of similar speed — or a permanently lower water share.

↑ Back to Top

VII. The Recovery: What 2025 Proves and Doesn't Prove

The recovery, when it came, was fast and complete on the metrics management chose to report.

Packaged water revenue grew 17.3% in 2025 to RMB18,709 million.⁴ Group revenue crossed RMB50 billion for the first time in the company's history. Gross margin expanded 2.4 percentage points to 60.5%, helped by falling PET resin and packaging costs and cheaper sugar — and, notably, by the company's own decision to restrict the share of sales going through e-commerce channels in order to hold price discipline in its distributor network.⁴ Selling and distribution expenses fell from 21.4% of revenue to 18.6%, partly because 2024 had been an Olympic advertising year.⁴

Profit rose 30.9%, and the board proposed a final dividend of RMB0.99 per share — about RMB11,134 million in total, against the RMB8,547 million paid on the prior year.⁴

The first half of 2026 extended it: revenue up 16.0% to RMB29,718 million, profit up 16.6% to RMB8,887 million, gross margin up another 0.6 points to 60.9%.⁷

Now separate what this proves from what it does not.

What it proves. The 2024 damage was cyclical, not structural. Distribution held. Shelf space came back. Consumers who stopped buying red-bottle water resumed buying it. This is consistent with the broader pattern of Chinese consumer boycotts, which have historically been intense and short relative to underlying purchase habit — and it is direct evidence that Nongfu's advantage in water was never really about consumer affection. It was about being the water that is physically present in every cooler in the country, at a quality claim nobody can match, at a price the consumer barely notices.

What it does not prove. Three things, and they matter.

First, the water business is not compounding. Look past the 17.3% headline, which is measured against a collapsed base, and look at the first half of 2026: water revenue grew 2.1%.⁷ Two years on from the shock, water is running roughly flat against a normal comparison. The 2025 number was recovery, not growth. An investor who reads 17.3% as the run rate will be wrong.

Second, nothing about the recovery demonstrates any capability to prevent a recurrence. The 2024 trigger was exogenous — a rival founder's death — and entirely unrelated to Nongfu's execution. Management's own language in the FY2024 outlook was that the company would "enhance our senses of crisis and danger" and "boost our risk resistance capabilities."³ That is an intention, not a mechanism. There is no disclosed change to board composition, to the son's citizenship status, to the ownership structure, or to communications governance that would make the next episode smaller.

Third, the recovery was materially helped by input costs. Roughly a third of the 2025 margin expansion story is PET resin and sugar, which are cyclical and have already turned: in the first half of 2026 the company noted that PET procurement prices increased even as juice and sugar fell.⁷ Margin at 60.9% is a peak-ish number resting partly on a commodity cycle, and it should not be extrapolated as a permanent step-change.

There is also a working-capital signal worth flagging, because it is the kind of thing that gets lost in a good-news year. Inventory turnover slowed from 82.3 days at the end of 2024 to 95.5 days at the end of 2025, which the company attributed to production stocking and raw-material build for the tea supply chain.⁴ That explanation is plausible and consistent with the upstream integration story. It is also a thirteen-day deterioration in a year when revenue grew 22.5%, and it is worth watching whether it reverses. At June 30, 2026 it stood at 94.7 days.⁷ Meanwhile other income fell 19.2% in 2025 on lower deposit interest as rates came down — a reminder that a chunk of this company's reported profit historically came from sitting on a large cash pile, and that source of income is shrinking.⁴

Myth versus reality, briefly. The consensus story about 2025 is "Nongfu recovered from the boycott." The filings support a more precise version. Water recovered its lost volume and then stopped growing. Tea, which was already the largest segment before the recovery began, did the heavy lifting for the group. And the profit reacceleration owed as much to a 2.8-point drop in selling expense ratio and a favourable commodity year as it did to any restoration of pricing power. The recovery was real. It was also, in composition, mostly not about water.

Management's own narrative across the two chairman's statements is, to its credit, internally consistent. The 2024 letter framed the year as a "historical test" of the brand; the 2025 letter opened with the RMB50 billion milestone and then said something quieter and more interesting — congratulating "our new management team, which has withstood the test of defamation and has been growing increasingly mature."⁵ For a founder who has run brand and human resources personally for thirty years, that is the closest thing to a public statement about delegation he has made.

Which is the natural bridge to the question that hangs over everything.

↑ Back to Top

VIII. Current Management: Ownership, Incentives, and Succession

Zhong Shanshan is 71. He is Chairman of the Board, an executive director, and — unusually for a company of this scale — the General Manager, meaning he holds both the chair and the top executive operating role. He is also a director of every subsidiary. The 2025 annual report states plainly that he is responsible for overall development strategy, business plans, major operational decisions, and the direct management of brand and human resources.⁵

As of December 31, 2025, he held approximately 84.04% of the company's total share capital: about 17.15% directly and about 66.88% indirectly through Yangshengtang, which he owns outright.⁵

Eighty-four percent is not "high founder ownership." It is control so complete that the public float is a rounding error in any governance calculation. Minority shareholders have economic exposure and essentially no voice.

The case for that structure is the capital allocation record, and it is genuinely unusual. Over the year ended December 31, 2025, the group reported no significant investments held and no material acquisition or disposal of subsidiaries, associates or joint ventures.⁴ Growth has come from building things: bottling plants at water sources, tea-processing capacity, category extensions developed in-house over a decade. The Oriental Leaf story is the clearest evidence — a product kept alive through years of consumer ridicule because the founder believed the category would eventually arrive, and which is now the largest profit pool in the company.

But the phrase "no value-destroying M&A" needs bounding rather than celebrating, and there are two specific pieces of counterevidence an investor should weigh.

The first is that the IPO proceeds have sat largely unspent for six years, with the utilisation deadline extended twice and HK$726 million reallocated between purposes in March 2026.⁴ A company that raised money it could not deploy is not the same as a company that deployed money brilliantly.

The second, and more instructive, concerns the other public company in the family. Zhong founded and chaired 北京万泰生物 Beijing Wantai Biological from November 2001 until January 2021, and Yangshengtang remains its largest shareholder.⁵ Wantai was the second engine of his 2020–21 wealth, riding COVID diagnostics and an HPV vaccine franchise. In 2025 it reported revenue of about RMB1.82 billion, down roughly 19%, and a net loss of about RMB398 million — its first annual loss since listing — as centralised procurement price cuts and competition hit the vaccine business even in the year its domestically developed nine-valent HPV vaccine was approved.²⁴[^25]

Wantai is not a Nongfu subsidiary and its losses do not touch Nongfu's accounts. But it is directly relevant to two claims investors make about this founder. It shows that his record outside beverages includes a business that converted a genuine technical milestone — a landmark vaccine approval — into a loss-making year, which is a useful check on the reflex that a marquee approval equals revenue. And it is evidence against treating "Zhong Shanshan touches it, it works" as a general proposition. His edge appears to be specific to consumer products with source-based quality stories, not universal.

On related-party dealings, the disclosure is clean but not zero. In 2025 Nongfu purchased RMB344.7 million of goods from Yangshengtang and its associates against an annual cap of RMB346.0 million, and paid RMB81.2 million for basic R&D and testing services against a cap of RMB100.0 million.⁵ Against RMB52.5 billion of revenue these are immaterial in size. The detail worth noting is that the goods-purchase line ran at 99.6% of its cap — a number an activist would flag not as self-dealing but as a structure worth watching, since caps that bind tend to get raised.

On incentives below the founder, something changed in 2025 that has not been widely noticed. Equity-settled share-based payment expense jumped from RMB5.5 million in 2024 to RMB76.8 million in 2025 — a fourteen-fold increase off a trivial base.⁴ For most of its listed life, Nongfu has run on cash compensation and founder authority rather than equity. A meaningful award programme, extended to company employees and to a small number of Yangshengtang staff, is the first structural evidence that the organisation is being built to function with more people holding ownership stakes.⁵ Against employee benefit expenses of RMB5,100 million it is still tiny. But the direction matters more than the size, and it lines up with the chairman's reference to a "new management team."

There is also a standing non-compete undertaking from Zhong and Yangshengtang in favour of the listed company, which the independent directors reviewed for the 2025 year and confirmed had been complied with.⁵ That is the correct structure for a founder who runs a large private group alongside a public one, and it is the kind of arrangement that only matters when it is tested.

On governance direction, two changes point the right way. Zhong stepped down as chairman of the Nomination Committee on May 20, 2025, handing it to independent non-executive director Yang Lei Bob.⁵ And at the May 2026 AGM the company amended its articles to cancel the supervisory committee, consolidating oversight into the board structure familiar to international investors.⁵ Ernst & Young audits the accounts, and the FY2025 results carried no qualification, no pledged assets, and no significant contingent liabilities.⁴

On shareholder returns, the proposed FY2025 dividend of RMB11,134 million represents roughly 70% of the year's attributable profit — a high payout for a business still adding water sources and building tea capacity, with capital commitments of RMB4,906 million outstanding at year-end.⁴ Nongfu pays only a final dividend; the board resolved not to declare an interim for the first half of 2026.⁷ The company also states explicitly that it has no predetermined payout ratio.⁵ So the generous distribution is a discretionary choice by a controlling shareholder who receives 84% of it, not a policy commitment minorities can rely on.

Now the succession question, which is where the outline's conventional wisdom needs updating against the filings.

Zhong Shu Zi, 38, has been a non-executive director since June 2017 and is Zhong Shanshan's son. He holds a bachelor's degree in English from the University of California, Irvine, earned in 2011, and a master's in international business from Zhejiang University, earned in 2021. He joined the group in January 2014 and has been General Manager of Yangshengtang's Brand Center since January 2020.⁵

What is new is the operating track. Between September 2023 and November 2025 he moved through three line roles in Nongfu's Sales & Marketing Center — Hangzhou Regional General Manager, Zhebei Regional General Manager, and Regional General Manager for East China Zone 1. And since March 2026 he has served as Assistant to the General Manager of the company — that is, assistant to his father in the top executive role.⁵

That is a succession sequence. It is not a succession plan: no timetable has been disclosed, no transition milestones have been articulated, and he remains a non-executive director rather than an executive one. But an investor who says "the son does not run daily operations and nothing is happening" is now behind the disclosure. Something is visibly happening, and it is being done in the most conservative way available — regional P&L rotations followed by a staff role next to the chairman.

The risk in that path is precise and uncomfortable: the successor being groomed is the same individual whose citizenship was the sharpest weapon used against the company in 2024. The governance question and the reputational question are, in this case, the same person.

↑ Back to Top

IX. Competitive Landscape & Industry Structure

If you want to understand how strange this industry is, consider that in the eighteen months after the 2024 boycott, the rival that benefited most from it nearly stopped functioning.

Wahaha 娃哈哈 was the emotional winner of March 2024. It was also, by then, a company whose founder had just died without a settled succession. What followed was not a comeback but a slow-motion structural crisis. Zong Qinghou's daughter 宗馥莉 Zong Fuli moved to consolidate control, securing her claim to her father's stake through a notarised inheritance certificate while a dispute over roughly US$1.8–2.1 billion of offshore assets involving alleged half-siblings ran through the Hong Kong courts, with asset preservation orders upheld in September 2025.²⁵²⁶ She agreed to acquire a 46% state-owned stake for more than RMB2 billion.²⁷ Because trademark use required unanimous shareholder approval under the existing structure, the operating business moved toward a new brand, 娃小宗 Wa Xiao Zong, for the 2026 sales year. And in late March 2026, nearly all Wahaha plants except those making red-labelled bottled water halted production for about a week during the control reshuffle.²⁷

The irony is exact. The nationalist wave of 2024 punished Nongfu for having a foreign-passport heir and rewarded Wahaha as the authentic national champion. Two years later, Nongfu had recovered and Wahaha was fighting over offshore trusts in a Hong Kong court, changing its brand name, and idling factories. The market's emotional judgement and its commercial judgement pointed in opposite directions, and the commercial one won.

China Resources Beverage 华润饮料, owner of 怡宝 C'estbon, is the more serious competitor and a very different animal. It listed in Hong Kong in October 2024 at a valuation of roughly US$4.4 billion, with a heavily oversubscribed retail tranche.²⁸ Its position is genuinely dominant — but in a different sub-category. C'estbon held about 32.7% of China's purified drinking water market, roughly four times the second-place player, selling more than 14.6 billion bottles in a year.²⁹

Purified and natural water are adjacent but not identical competitive sets, which is why Nongfu's April 2024 green-bottle relaunch mattered strategically. Nongfu entered C'estbon's category directly, using its own natural sources as the input, and framed the two products as complements in a single brand architecture. It is a rare offensive move from a company that mostly builds rather than attacks.

Master Kong 康师傅 is the cautionary tale. In 2025 its beverage business generated RMB50,123 million, down 2.9%, with its core tea business down 5.1% to about RMB20.6 billion, and its dealer network shrinking by 9,606 outlets to 57,609.¹⁶ That is the year Nongfu passed it to become China's largest beverage company by revenue. A dealer network contracting by 14% while a rival's tea business grows 29% tells you where shelf space is moving. Uni-President 统一 remains a broad-portfolio Taiwanese-heritage competitor across tea, water and instant noodles.

In sugar-free tea, the challenger set is where the pressure builds. サントリー Suntory is the established number two with a long-standing oolong franchise. 元气森林 Genki Forest pioneered sugar-free positioning in sparkling water and has pushed into tea. And 果子熟了 Guozi Shule went from a rounding error to third place in the category in roughly a year.¹⁵ Concentration at the top of sugar-free tea is high, but it has begun to leak at the edges precisely as the category's own growth stalled.

Run the Five Forces on this properly and it is not the uniformly attractive picture the margins imply.

Buyer power at the point of purchase is high and switching costs are nil — a consumer at a cooler substitutes freely. What offsets it is retailer and distributor economics: Nongfu's stated strategic goal, repeated in the 2025 filings, is to "generate profits for distributors and retail partners," and its deliberate throttling of e-commerce sales to protect distributor price discipline is a real, disclosed mechanism for keeping the channel loyal.⁴ That is buyer power managed, not absent.

Supplier power is low. Nongfu owns its sources, builds its own plants, roasts its own coffee beans, and is integrating upstream into tea leaf and jasmine cultivation. Its main exposure is commodity PET resin, which it does not hedge.

Threat of new entrants is bifurcated, and this is the crux. In natural-source water it is genuinely low — permits are scarce, sites are fixed, and it took Nongfu thirty years to assemble seventeen. In tea, functional drinks and juice it is moderate-to-high; Guozi Shule's rise proves a well-executed new brand can take a category position in about eighteen months.

Substitutes are everywhere, since every beverage substitutes for every other, and the health trend that made sugar-free tea is the same trend that will make whatever displaces it.

Rivalry is intense across every category, in a market where the leading player has under 15% of total non-alcoholic beverage revenue.

A word on the distribution machine itself, because it is the least glamorous and most important asset in the story. China's beverage retail is not a handful of national chains. It is millions of small outlets — corner shops, roadside kiosks, township grocers, canteens — served through layers of regional distributors, each of whom must be given a reason to carry a case of your product instead of somebody else's. Nongfu's stated operating principle, repeated in its 2025 filings, is to "generate profits for distributors and retail partners," and it backs that with concrete choices: throttling its own e-commerce volume to stop online discounting from cannibalising channel margins, and running "one product, one code" promotional mechanics that let it push consumer incentives through the channel rather than around it.⁴

That is worth contrasting directly with Master Kong's shrinking dealer base. A distributor network is not a fixed asset; it is a set of relationships that decay if the economics stop working. Nongfu spends real money keeping those economics intact, and the 2024 episode demonstrated exactly why: when the brand came under attack, the thing that determined the depth and duration of the damage was whether distributors and shopkeepers kept stocking it.

In 7 Powers terms, the honest allocation is: cornered resource protects roughly a third of revenue; scale economies in distribution protect most of the rest; branding protects all of it, and is the power that 2024 demonstrated can be attacked directly. There is no network effect, no switching cost, no counter-positioning, and no process power in evidence. A company deriving its defence primarily from distribution scale and brand is a strong consumer business. It is not an unassailable one.

↑ Back to Top

X. Business & Investing Lessons

Thirty years is long enough for a company to teach you things. Nongfu Spring's record contains at least seven lessons that generalise beyond bottled water, and several of them are uncomfortable.

The best moat in a commodity business is a physical constraint you got to first. Everything Nongfu has built rests on a single insight from 1996: in a product where the physical output is nearly identical across producers, the only durable differentiation is the input. Coffee, spirits, mineral water, certain foods — the same structure appears wherever provenance is verifiable and the source is finite. The corollary is that this kind of advantage is accumulated slowly and cannot be bought quickly, which is why it is durable and why it is a bad fit for investors who want compounding to be legible quarter to quarter.

A structural advantage and a reputational vulnerability are separate risk categories, and owning one does not insure the other. Nongfu's water sources are as close to a physically uncopyable asset as consumer staples offers. They did nothing at all in 2024, because the asset was never what was under attack. Underwriting a cost or resource advantage tells you what a competitor can do to you. It tells you nothing about what a crowd can do to you.

Patience in category building is measurable, and it looks like failure for a long time. Oriental Leaf spent roughly a decade as a punchline before becoming the company's largest profit pool. The lesson is not "keep everything forever" — most products voted worst-tasting in the country really are bad. It is that a business with a controlling owner and no quarterly pressure can hold a position for ten years, and that this is a genuine, if rare, structural advantage of founder control. The flip side is that no external mechanism exists to force the abandonment of a bad ten-year bet, and Nongfu's record contains only the bet that worked.

High founder ownership delivers capital discipline and key-person risk as a single trait, not two. The same concentration that produced an organic growth record with no significant acquisitions also means one 71-year-old makes the brand decisions, sets strategy, holds both the chair and the general manager roles, and decides personally when the company speaks during a crisis. Investors who praise the first should price the second.

Consumer nationalism in China is a distinct risk category with different physics from Western boycott dynamics. It is faster — days, not months. It attaches to identity markers that are typically already public and previously ignored. It is largely unhedgeable at the company level, because the trigger can be an unrelated event. And, on the two-observation record available here, it is recoverable. What it is not is manageable in advance.

Certification and category leadership are not the same as durable earnings. Wantai's landmark vaccine approval accompanied a loss-making year. Read that across to Nongfu's tea position: a 75% category share in a category that stopped growing is a different asset from a 75% share in a growing one.

Raising capital you do not need creates obligations you did not price. Nongfu's listing brought a daily mark on the founder's wealth, a public prospectus containing a family detail nobody cared about for three and a half years, and a use-of-proceeds schedule that has now been extended twice. None of those costs appeared in the underwriting deck. For a business already generating more cash than it could deploy, the honest question is whether the listing added anything beyond liquidity for a shareholder who has never sold.

Where an advantage lives determines what can take it away. Nongfu's water advantage lives in the ground — in permits and geology — and geology is indifferent to public opinion. Its tea advantage lives in shelf space and consumer preference, both of which are rented rather than owned, and both of which can move within a year. When a business shifts its centre of gravity from one kind of advantage to the other, the risk profile changes even if the growth rate improves. That shift has already happened at Nongfu, and it is visible in the segment tables long before it will be visible in the results.

Recovery evidence and prevention evidence are different things, and markets routinely confuse them. A company that bounces back from a shock has demonstrated resilience, which is genuinely valuable. It has not demonstrated that the shock will not recur, and it has certainly not demonstrated that the next one will be smaller. The distinction matters most when the shock is exogenous, because the recovery tells you about the business while the trigger tells you about the world. Investors who read Nongfu's 2025 numbers as closing the 2024 question are answering a different question than the one that was asked.

A rival's collapse is not your competitive advantage, but it does buy you time. Wahaha's post-succession unravelling handed Nongfu two years in which its most emotionally resonant competitor was distracted by courts and trademarks rather than by shelves. That is luck, and it should be labelled as luck in any assessment of the 2025 recovery. The durable question is what Nongfu built with the time.

↑ Back to Top

XI. Strategic Position, Risk Radar & What to Watch

Nongfu Spring enters the second half of 2026 in a strong operating position and a contested narrative one. It is the largest beverage company in China by revenue, it has restored its water business to roughly flat growth on a normalised base, and its tea franchise is compounding at close to 30% while taking share from a shrinking incumbent. Its balance sheet carries RMB26,534 million of cash, deposits and bank balances against RMB5,186 million of interest-bearing borrowings, with a gearing ratio of 14.2% at June 30, 2026.⁷ More than 99% of revenue comes from mainland China, with Hong Kong and Singapore launches in 2025 representing a genuine but immaterial start on internationalisation.⁴³⁰

The material risks, each tied to a mechanism rather than a label:

Reputational and nationalist-sentiment risk. Demonstrated twice, in 2013 and 2024, at escalating scale. The mechanism is that Nongfu's price premium rests on a trust claim, and trust claims can be attacked by parties with no capital and no product. Consequence: a double-digit percentage revenue decline in a single segment inside one year. This is the standout, company-specific risk and nothing disclosed suggests it has been mitigated.
Key-person and succession risk. One person holds 84% of the equity, chairs the board, serves as general manager, and personally directs brand and HR. The named successor candidate is progressing through operating roles but is not an executive director, has no disclosed timetable, and is the individual whose citizenship was weaponised in 2024.
Competitive intensity in the growth engine. Sugar-free tea recorded its first negative category growth in 2025 while a new entrant took third place in about a year.¹⁵ Nongfu's tea growth is currently share capture in a flat pool, which is a finite mechanism.
Water-source and regulatory risk. Extraction depends on local environmental permits, and the company's own filings track a tightening food-safety regime — the amended Food Safety Law effective December 1, 2025, and the entrusted-production and food-recall measures taking effect December 1, 2026.⁴⁷ The Hong Kong Consumer Council dispute showed that watchdog friction is live rather than hypothetical. Nongfu reported no material non-compliance with food safety or environmental laws in 2025.⁵
Input costs and consumer softness. PET resin prices rose in the first half of 2026 even as sugar and juice fell, and the company does not hedge interest rates or systematically hedge FX.⁷ Chinese retail sales of consumer goods grew just 1.3% year-on-year in the first half of 2026, with beverage sales at large retailers up only 1.0% in 2025 — a low-growth backdrop against which Nongfu's 16% growth is entirely share-driven.⁴⁷
Tax and subsidy dependency. A slice of Nongfu's reported profitability rests on policy. Certain subsidiaries in western China qualify for a preferential 15% enterprise income tax rate running to December 31, 2030; subsidiaries in Guangxi and Xizang benefit from local exemptions that bring their effective rate to 9%, with the Xizang measure currently scheduled through 2027 and the Guangxi programme through 2030.⁴ The group also recognised RMB575.7 million of income-related government grants in 2025.⁴ None of this is improper or unusual for a Chinese manufacturer with plants in remote provinces. It does mean the effective tax rate and a small but real part of pre-tax profit are policy-dependent on defined expiry dates, which is exactly the sort of thing that quietly reprices when it lapses.
A second-layer note worth one sentence: R&D costs were RMB310.6 million in 2025, roughly 0.6% of revenue and essentially flat on 2024, which sits awkwardly beside management's stated ambition to "evolve toward a R&D oriented enterprise in the beverage industry."³⁴ The innovation record is real but it is formulation and packaging innovation, not research intensity.
Treasury and FX, briefly. The group holds meaningful foreign currency balances — HK$1,115 million and US$617 million at June 30, 2026 — and recorded exchange losses of RMB199 million in 2025 and RMB180 million in the first half of 2026. It states that it has no interest rate hedging policy and hedges FX only opportunistically.⁴⁷ For a company with over 99% of revenue in renminbi, this is a self-inflicted and easily fixable line item rather than a structural exposure, but it has now cost real money two periods running.

The three KPIs that matter most. Not a dashboard — three numbers that carry the thesis:

Packaged water revenue growth on a normalised base. Not the 2025 recovery rate. The 2.1% first-half 2026 print is the real signal, and whether water can grow mid-single-digits from here determines whether the cornered resource is still producing or merely defending.
Tea segment revenue growth relative to the sugar-free tea category's own growth. If Nongfu grows 25% while the category grows 0%, that is share capture with a ceiling. If the category resumes growth, the same 25% means something entirely different. Watch both numbers, not one.
Gross margin in packaged water specifically, and group gross margin against PET resin prices. This is the pricing-power test. Margin holding while resin costs rise would be genuine evidence of price realisation; margin falling with resin would confirm that the recent expansion was a commodity gift.

↑ Back to Top

XII. Bull vs. Bear

The bull case rests on four evidenced mechanisms rather than on management rhetoric.

The company owns seventeen permitted natural water sources assembled over thirty years — an asset base that a competitor with unlimited capital could not replicate on any relevant timescale, and which underpins a gross margin near 61%.⁴⁷ It has demonstrated, with segment profit numbers rather than narrative, that its distribution reach and quality-sourcing philosophy transferred into a second category large enough to overtake the first. It has an organic-growth record with no significant acquisitions or disposals, no impairments of consequence, no pledged assets, and a clean audit.⁴ And the 2025–26 numbers established that the worst reputational shock in its history was reversible in about eighteen months.

Add to that a genuine competitive tailwind that arrived by accident: the two rivals best positioned to take share during Nongfu's weakest period were a state-linked purified-water specialist in an adjacent sub-category and a family-controlled national champion that spent the following two years in inheritance litigation and production halts.

The bear case is not primarily about execution, which is the point.

It rests first on the fact that the cornered resource now protects a shrinking share of the business. Water is roughly a third of revenue and less than a third of segment profit. The larger, faster-growing engine is defended by distribution scale and brand — the exact combination the 2024 episode proved can be shaken by a non-economic shock while unit economics stay intact.

It rests second on the demonstrated, twice-observed vulnerability to reputational attack, with no disclosed mitigation, no change to the underlying facts that were weaponised, and a management crisis playbook that resolved through state broadcast interviews five to nine months after the fact rather than through any process shareholders could observe or hold to account.

It rests third on succession, where the concentration of ownership, strategy, brand voice and crisis judgement in a 71-year-old, and the grooming of the specific heir whose passport was the flashpoint, together constitute a risk that no amount of operating performance addresses.

It rests fourth on the growth engine's own category dynamics: sugar-free tea has stopped growing as a category, challengers are multiplying, and Nongfu's tea growth is currently a share transfer from Master Kong that cannot continue indefinitely.

And it rests fifth on valuation. The shares have derated hard from their 2021 peak, and an investor buying today is not paying the pandemic-era multiple. But they still trade at a premium to the sector, and the market has spent five years compressing the multiple while earnings grew.¹⁴ The question is not whether the derating has happened. It is whether the market is now pricing a business with a once-a-decade, unhedgeable, double-digit-revenue-impact tail risk — or a business with a permanent 61% gross margin and 30% earnings growth.

The activist stress test. Imagine a skeptical fund manager with a large position and a willingness to be unpopular. What would they actually put in the letter?

Four things. First, that a company holding RMB26.5 billion in cash and deposits against RMB5.2 billion of borrowings, with more than HK$4 billion of six-year-old IPO proceeds still unspent and a utilisation deadline extended twice, is running a lazy balance sheet — and that the correct response is either a defined capital return programme or a defined investment programme, not a discretionary annual dividend set by the holder of 84% of the shares.⁴⁷ Second, that combining the chair and general manager roles in one 71-year-old, with no disclosed succession timetable, is an unforced governance choice that a single announcement could improve. Third, that the crisis playbook of 2024 — months of public silence from the company while the market marked the shares down, followed by a legal campaign against a Hong Kong consumer watchdog and eventually two state-television interviews — was optimised for the founder's preferences rather than for shareholders' information needs. Fourth, that the company's disclosure of the metrics investors most need is thin: it does not publish red-bottle versus green-bottle volumes, distributor counts, or category-level market share, which means outsiders track the most important dynamics in this business through third-party scan data rather than through the company.

The counterargument to all four is the same and it is not trivial: this structure produced a business that grew revenue 22.5% and profit 30.9% in 2025, with no acquisitions, no impairments, no pledged assets and a clean audit.⁴ An activist would be arguing for better governance around a machine that is currently working. That is a harder case to make than the usual one, and it is why no activist has made it.

Net framing. The affirmative case here is unusually well evidenced by consumer-staples standards: real physical assets, real category leadership, a real second engine with real segment profit, a real recovery. Testing it against the company's own history over thirty years does not reject the thesis. It narrows it in two specific places — the resource advantage covers a minority of the business, and the brand advantage has a documented failure mode that recurs. What remains genuinely unproven, in the sense that no evidence either way exists yet, is whether the enterprise works without Zhong Shanshan. That is not a caveat. On an 84%-owned company where one person holds both the chair and the general manager role, it is the central open question.

↑ Back to Top

XIII. Epilogue

Thirty years after Zhong Shanshan started bottling water at the edge of a lake in Zhejiang, the anniversary framing the company chose for 2026 was not "bigger" or "faster." It was 稳一点、慢一点、远一点 — steadier, slower, further.⁷ For a company that just posted 22.5% revenue growth and passed Master Kong to become the largest beverage business in China, that is a deliberately anticlimactic slogan, and probably an honest one.

Entering the second half of the year, the picture is clear on two of three fronts. The recovery is intact and the water business has stabilised at low growth. Tea is now structurally the larger business by both revenue and profit, and the company that most people still describe as a bottled-water company derives more of its earnings from a drink that was once voted the worst-tasting beverage in China.

The third front remains open. There is a succession sequence — regional general manager rotations, then assistant to the general manager since March 2026 — and there is no succession plan, no timetable, and no public discussion of what happens to a business where one man has personally made the brand decisions for three decades.

Two things will make the next few years legible. The first is whether water can grow at all from a normalised base, or whether seventeen sources and thirty years of brand building now support a business that simply holds its position while tea does the compounding. The second is whether the sugar-free tea category resumes growth, or whether Nongfu spends the rest of the decade taking share in a pool that has stopped filling. Neither question is about crisis management. Both are about whether the ordinary business is as good as the last two years of headline growth implied.

And the question the 2024 episode raised has not been answered, because it cannot be answered by operating performance. Was that a tail event — a rival founder's death, a coincidence of timing, a crowd that moved on — or the first properly-sized data point in a recurring risk that Chinese consumer brands with founder-nationalist optics will have to keep pricing in?

The company's own record offers a partial answer and refuses to offer the rest. Twice in thirteen years, the most valuable thing Nongfu Spring owns turned out to be the thing it controls least: what several hundred million people believe about where its water comes from, and about the man who sells it.

Nongfu Spring Co., Ltd.

Table of Contents

Nongfu Spring: The Loneliest Wolf and China's Water Wars

I. Introduction & Episode Roadmap

II. Zhong Shanshan 钟睒睒: The "Loneliest Wolf" (1954–1996)

III. The Water-Source Bet as a Business Model (1996–2013)

IV. The 2020 IPO: Briefly the Richest Man in China

V. Building the Second Engine: Beyond Water

VI. 2024: The Water War

VII. The Recovery: What 2025 Proves and Doesn't Prove

VIII. Current Management: Ownership, Incentives, and Succession

IX. Competitive Landscape & Industry Structure

X. Business & Investing Lessons

XI. Strategic Position, Risk Radar & What to Watch

XII. Bull vs. Bear

XIII. Epilogue

References