The FDA Doesn't Refuse Chinese Tea for the Reason You ThinkFDA拒收中国茶叶，问题大多出在标签，不在农药

Ask a Chinese tea exporter, or the buyer vetting one, what the biggest FDA risk is, and pesticide residue comes up almost immediately. It’s the headline story: illegal pesticide residue, FDA’s Import Alert 99-08, a shipment turned back because someone upstream used the wrong chemical.

For tea leaf specifically, that’s mostly not what’s happening.

We went through FDA’s own import-refusal data for China-origin tea and China-origin herbal and botanical supplements: more than 300 tea refusals and over 2,000 supplement refusals filed since 2001. The pattern is more specific, and more fixable, than “FDA is worried about pesticides.”

What we actually looked at

FDA logs every refused shipment with a specific charge code behind it: missing English labeling, undeclared allergens, illegal pesticide residue, and so on. We pulled every China-origin refusal filed under FDA’s Industry Code 31 (Coffee and Tea) and Industry Code 54 (the broader category that covers herbal and botanical “teas” like chrysanthemum or goji, ginseng, and other extracts) from 2001 through mid-2026. Then we grouped the charge codes into categories: mislabeling, pesticide residue, chemical or heavy-metal contamination, and so on.

That grouping is our own reading of FDA’s charge codes, not an official FDA breakdown, so treat the percentages below as an analysis, not a government-published statistic. It’s also a reason to check FDA’s own records directly, not a secondary summary. Even the basic product code is easy to get wrong: tea’s actual FDA code is Industry Code 31, and a nearby number, 24, is vegetables.

This isn’t stale data either. 2025 was the second-highest year on record for tea refusals, and the annual count has stayed in roughly the same 10-to-30 band since 2008, with no clear downward trend.

Tea: it’s mostly the label, not the leaf

Of 305 China-origin tea refusals in FDA’s data, the single biggest cause is mislabeling: 122 refusals, 40.0 percent. The specific gaps that show up again and again are missing English-language labeling, missing or incorrect Nutrition Facts, no ingredient list, and no importer or firm name on the package.

Pesticide residue, the thing everyone worries about, accounts for 21 refusals: 6.9 percent. A chemical or heavy-metal finding, mostly melamine or a “poisonous” charge, shows up more often than pesticide does: 42 refusals, 13.8 percent.

If you supply tea: the fastest, cheapest fix on this list isn’t a lab test. It’s a proper US-format label: English text, a Nutrition Facts panel built to US rounding and serving-size rules, a complete ingredient list, and your firm’s name on the package. None of that requires a certifier or a chemist.

If you buy tea: ask to see the actual label artwork a supplier plans to ship with, not a translated spec sheet. Correct English on a spec sheet and a US-format Nutrition Facts panel are two different documents, and that gap is where a surprising share of shipments get stopped.

The other fixable gap is a filing, not a contamination finding

The second-largest cause is narrower, and once you know it exists, straightforward to close. 62 of the 305 tea refusals, 20.3 percent, were cited for both a missing Food Canning Establishment (FCE) registration and a missing scheduled-process filing. Both trace to the registration and process-filing requirements for low-acid canned and acidified foods under 21 CFR Parts 108 and 113.

This is almost certainly a canned or bottled tea problem, not a loose-leaf or bagged one. 35 of the 305 tea refusals in the dataset were specifically coded as liquid tea.

If you make or plan to ship ready-to-drink or bottled tea into the US, this registration and process filing is a known, specific step, not a discretionary one, and it has nothing to do with whether the tea itself is contaminated. FDA happens to file it under its adulteration rules, but it’s a paperwork requirement, not a food-safety finding.

Where the pesticide fear is earned

The picture changes once you leave tea leaf and move into herbal and botanical supplements: ginseng, chrysanthemum and goji blends, and other botanical extracts. FDA files these under a separate industry code, 54, and the refusal pattern looks nothing like tea’s.

	Tea (IC31)	Herbal & botanical supplements (IC54)
China-origin refusals, 2001 to mid-2026	305	2,167
Top cause: mislabeling	40.0%	23.4%
Pesticide residue specifically	6.9%	16.5%

Pesticide residue accounts for 358 of 2,167 supplement refusals, 16.5 percent, more than double tea’s rate. Broader adulteration (pesticide residue, microbial contamination, chemical or heavy-metal issues, and a cluster of supplement-specific problems combined) is the majority cause at 64.6 percent. Mislabeling is still common, 23.4 percent, but it’s no longer the dominant story the way it is for tea.

If you supply or buy herbal or botanical blends, the compliance conversation should look different from tea’s. Contaminant testing and sourcing controls deserve real budget here, not just a clean label. A chrysanthemum or goji “tea” gets classified with ginseng and herbal extracts for FDA’s purposes, not with green or black tea, and the risk profile follows the classification, not the word “tea” on the package.

The claims problem hiding inside “other”

One more pattern is worth naming, because it doesn’t show up as its own line item. It hides inside a catch-all “other” category.

For herbal and botanical supplements, 223 refusals (10.3 percent) fall into a bucket dominated by a single charge: UNAPPROVED. That means FDA decided a product’s own marketing made a disease or treatment claim strong enough to treat the product as an unapproved new drug rather than a food. Counting every instance of that charge across the full dataset, not just cases where it was the primary reason for refusal, it appears 379 times.

This isn’t a theoretical risk. It’s the same failure mode we wrote about in a claims-language case study. Language that reads as ordinary marketing in one market, “detox,” “lowers blood pressure,” “boosts immunity,” crosses a real line in the US. A structure-and-function claim becomes a disease claim, and the product becomes, legally, an unapproved drug. FDA’s own refusal data shows this happening at scale, not as an edge case.

Easterlies view — The mistake we see most often is treating “FDA compliance” as one undifferentiated risk to insure against, instead of a short, sequenced list of specific fixes. For loose-leaf or bagged tea, the label is worth fixing first: it’s cheap, entirely within a supplier’s control, and the single most common reason a shipment gets turned back. For canned or bottled tea, the FCE registration and process filing are a compliance step, not a manufacturing overhaul. Skip it, and you lose a shipment to paperwork instead of the product. For herbal and botanical blends, the numbers argue for spending more, earlier, on testing rather than on the label. We’re not the certifier, the lab, or the labeling specialist in any of this. What we help with is figuring out which of those a given product actually needs first, instead of guessing.

A rough order of operations

• Loose-leaf or bagged tea: fix the label first. It’s the single most common reason a shipment gets refused, and it’s entirely in a supplier’s control.
• Canned or bottled tea: confirm the FCE registration and process filing are actually filed, not just planned.
• Herbal or botanical blends: budget for contaminant testing and sourcing verification before anything else. This is the category where the risk looks like what people assume tea’s risk looks like.
• Any product with marketing copy attached: put a claims review on the list too, next to the label and the lab report, not instead of them.

None of this replaces a lab, a labeling specialist, or counsel once a claim gets close to the line, and we’re not any of those things. What we can do is help you work out which of the above actually applies to your product before you spend money on the wrong fix. If that’s useful, get in touch, and if it turns out you’re not ready yet, we’ll tell you that too.