Broker markups in Yunnan — what farmers receive vs. wholesale price

In the spring of 2026, a farmer on the Bulang mountain finishes pan-firing a batch of single-tree shēng pǔ’ěr maocha and wraps it in a paper-lined bamboo basket for the village agent. She will receive about ¥180 per kilogram. That same leaf, after three or four hands, will reach a Kunming wholesaler at ¥1,200 to ¥1,600 — before pressing, wrapping, and brand markup. The spread is not new, but it has widened sharply since 2020 as market demand for micro-lot and gushu teas has intensified. Amgalan Chin, tea.report’s cross-regional tea expert, notes: “The Yunnan broker chain is a classic agricultural ladder — thin margins at the base, fat at the top — yet the farmer bears the harvest risk and quality-control labor.” This article maps that ladder, quantifies what farmers actually receive for four key production zones, and examines whether direct-trade and cooperative models can reshape the economics for the 2026 harvest and beyond.

The farmer’s share — a baseline from three villages

Between February and April 2026, the tea.report supply-chain team interviewed 47 farming households in Manmai (Bulang), Xiao Hu Sai (Lincang), and Zhentianjing (Jinggu). Across the sample, farmers reported average pre-dried leaf prices of ¥130–¥220/kg for first-flush, old-tree material, depending on village reputation and microclimate. After accounting for picking labor (¥40–¥70/kg fresh leaf, roughly four kilos fresh to one kilo dry), fuel for pan-firing, and transport to the village collection point, net retained by the farming household before any broker interaction was ¥60–¥110/kg. For mid-age plantation teas — the workhorse of daily-drinker pǔ’ěr — that net figure falls to ¥18–¥35/kg. “People read about Lao Banzhang hitting ¥7,000 a kilo and assume all farmers are rich,” says Manmai grower Li Chunzhi. “My neighbor sold 18 kilos of single-tree old-aroma to a Kunming team for ¥1,600/kg at the door, but that was one kilo per tree — the rest of his harvest goes through brokers at maybe a quarter of that.”

The broker’s ladder — village collector to Kunming wholesaler

A typical Yunnan pu-erh marketing chain has four tiers. At the village level, a gānzhu (干主, dry-master) works on a commission of 3–8% of the sale to the next tier, sometimes advancing cash to families before the spring pluck. The second tier is the regional aggregator — often based in Menghai town or Lincang city — who consolidates 300–800 kg batches, does a first culling for quality, and sells to Kunming wholesalers at a markup of 20–35% above village gate. The wholesaler adds 40–80% to cover storage, tasting, branding, and distribution to tea merchants in Beijing, Shanghai, or overseas. A final tier, the brand or pressing workshop, adds packaging, provenance documentation, and further margin before reaching the consumer. “Every hand adds a story — or at least a label,” observes Amgalan Chin. “The physical tea might travel 600 kilometers from tree to tea-shop shelf, but the narrative travels much further, and that’s where the premium builds.”

Village-level middlemen — the cash nexus

In remote areas like Mangsong or Bulang’s older hamlets, the village collector is often the only source of liquidity before tea season. They advance seed money for fertilizer and labor, locking in a first-right-of-refusal on the harvest. Tea.report’s survey finds these advances average ¥3,000–¥8,000 per household, equivalent to 30–50% of the eventual seasonal income. The implicit interest is the discounted purchase price versus what the collector can resell to an aggregator — a spread of 12–22% when the harvest is large, widening to 30%+ in drought years like 2023. Farmers rarely receive a transparency statement showing final resale value.

Regional aggregators — quality sorting and lot homogenization

Aggregators in Menghai operate small warehouse-tasting rooms where maocha is re-fired, sorted for broken leaf, and sometimes blended to smooth lot inconsistencies. Their margin depends on geographic reach: an aggregator who makes the 12-hour drive to gather 50 kg from a little-known village like Pasha can earn a markup of 60–80% when presenting it as single-origin to a curious Kunming buyer. By contrast, well-known regions like Yiwu have thinner aggregator margins — 15–25% — because pricing transparency is higher and buyers bring their own cupping teams.

Regional disparities — Lao Banzhang vs. the rest of the mountain

The classic example of the premium skew remains Lao Banzhang. In 2025, village gate prices for genuine Lǎo Bānzhāng old-tree maocha averaged ¥5,200/kg, according to data collected by the Menghai Tea Industry Bureau. Wholesale in Kunming added roughly 35–50%, landing at ¥7,000–¥7,800/kg. But that 35% broker margin is far smaller than for a village like Naka old-tree, where farmer-gate ¥380/kg becomes ¥850–¥1,100/kg in Kunming — a spread of 120–190%. “The premier names don’t need a broker to persuade a buyer; the broker is just a logistical convenience,” says Amgalan Chin. “For less famous tea, the broker is an essential marketer — and the markup reflects that marketing effort.” In Jingmai, where cooperative structures are stronger (see tea.school for the Jingmai Alliance model), farmer-gate for ecologically-certified old-tree averages ¥520/kg and Kunming wholesale ¥780–¥900/kg, a markup of just 50–73% — still significant, but notably compressed.

The cooperative alternative — direct procurement and farmer-owned brands

A small but growing number of Yunnan tea-growing villages have formed cooperatives that bypass the regional aggregator tier. The Ban Pen Old-Tree Cooperative, established in 2021 with 34 member households, now sells directly to two Kunming tea brands and a Shanghai-based specialty importer. In 2025, members received ¥680/kg for old-tree material, versus an estimated ¥260/kg they would have netted through the broker ladder. The cooperative retains 8% for quality-control labs, cupping certification (using the GB/T 23776-2018 sensory evaluation methodology), and transport. “Co-ops aren’t a silver bullet,” cautions Amgalan Chin. “They require literacy, trust, and a dedicated cupping team — scarce resources in villages where most adults under 40 have already migrated to the cities. But where they work, the farmer’s share can double.” The thetea.app crew visited the cooperative’s spring pressing in April 2026; a short documentary will appear on tea.community later this year.

Transparenzhcc and traceability — does blockchain help the farmer?

Several pilot projects have tried to attach QR codes to pressed cakes that reveal the farmer-gate price, harvest date, and GPS coordinates of the tree cluster. In 2024, an initiative backed by the Lincang county government and a Hangzhou blockchain firm tagged 2,400 cakes from Bangdong old-tree groves. While consumer-side scans increased 300% year-on-year, the actual impact on farmer income was modest: the tagged cakes sold at a 7% premium over untagged lots from the same harvest. “Traceability appeals to the ethical consumer, but it doesn’t automatically recut the financial pie,” says Amgalan Chin. “The cost of the tag — about ¥4 per cake — is still often borne by the farmer or the cooperative, not the broker who extracts the highest margin.” tea.report’s analysis suggests that only when the blockchain record is coupled with a direct-payment mechanism can the middle tiers be trimmed meaningfully.

The quality question — does the markup drive better tea?

There is a counter-argument that broker layers are not entirely parasitic: aggregators perform cupping that grades material for a market that will pay widely different prices for a Bái Háo Yín Zhēn-style budset versus a coarse old-tree leaf. In the absence of cooperative cupping infrastructure, the regional aggregator’s palate is the de facto quality gate. Zhou Xiang, the tea.report green and black tea specialist with deep Hunan sourcing networks, notes the parallel: “In Hunan’s Junshan Yinzhen market, the middleman’s nose for subtle defects — stale pickle, cellar mold — saves the downstream buyer from expensive mistakes. Yunnan’s pu-erh brokers serve the same function, albeit with a heftier commission.” Whether that commission will compress as the next generation of tech-savvy growers begin to sell directly via WeChat and Douyin remains an open question for the 2027 season.

References

1. Menghai Tea Industry Bureau 2025 Harvest Price Survey — Menghai County Tea Industry Office
2. GB/T 23776-2018 — Methodology of sensory evaluation of tea — Standardization Administration of China
3. Ban Pen Cooperative 2025 Annual Report — Ban Pen Old-Tree Pu-erh Cooperative
4. Interview with Li Chunzhi, Manmai village farmer, March 2026 — tea.report field interview
5. C. Wang & Q. Li (2023). Tea supply chain margins in Xishuangbanna, Journal of Chinese Tea Economics, 45(2), 112–128. — Academic journal