Vietnam cinnamon exports: what the EU import record actually shows

Vietnam cinnamon exports to the European Union have roughly doubled in six years, and the record that says so is public. Eurostat puts EU27 imports of cinnamon from Viet Nam at 2,662 tonnes in 2019 and 6,145 tonnes in 2025. That is the sort of line a supplier puts on a slide, and on its own it is close to useless to a buyer. What follows is what the series does say, what it cannot say, and the three places a reader of trade statistics goes wrong before finishing the first column.

The reason to start here rather than with an industry estimate is that this record is not ours and not Viet Nam's. It is what the importing side wrote down at the point the goods entered, collected by the customs administrations of the member states and published by the statistical office of the European Union. A buyer sitting in Rotterdam or Hamburg can open the same query and get the same numbers, which is a different kind of claim from a figure quoted in a brochure.

What the series actually says

Eurostat Comext publishes EU trade at eight-digit detail back to 1988, in dataset DS-045409. Filtered to the EU27 as reporter, Viet Nam as partner, HS 0906 (cinnamon and cinnamon-tree flowers) and the import flow, the annual net quantity runs 2,662 tonnes in 2019, 3,484 in 2020, 4,917 in 2021, 5,474 in 2022, 4,691 in 2023, 5,481 in 2024 and 6,145 in 2025. Seven points, one dataset, one extract, stamped 16 July 2026. Nothing in that run is smoothed, projected or averaged across sources.

Two things stand out. The first is the size of the move: the 2025 figure is roughly 2.3 times the 2019 one, which over six years is a growth rate that does not happen to a mature origin by accident. Somebody on the European side spent those years qualifying Vietnamese material, writing it into specifications and getting it through quality departments. The second is that the run is not monotonic. 2023 broke it, and the break is where most of the useful information is.

The 2023 fall was the market, not the origin

Read on its own, 2023 looks like a bad year for Vietnamese cassia: 5,474 tonnes down to 4,691, a fall of about 14 per cent. Read against the same table's totals, it looks like something else entirely. Extra-EU imports of cinnamon from every origin fell from 16,207 tonnes in 2022 to 13,346 tonnes in 2023, a fall of about 18 per cent. The European market took less cinnamon from everywhere that year, and it took proportionally more of what it did take from Viet Nam: 33.8 per cent of extra-EU arrivals in 2022 against 35.1 per cent in 2023.

This is the single most useful habit in reading a trade series, and it costs one extra query. A level tells you almost nothing without the denominator that goes with it. An origin can grow while losing position, and it can shrink while gaining it, and a buyer who reacts to the level alone will read a destocking year across the whole category as a supply problem at one origin. The 2023 numbers say the opposite of what the volume line says on its own.

It is worth being honest about the limit of that reading too. Knowing that Viet Nam held its position through a down year does not tell you why. Buyers switching origin, buyers running down stock, a currency move, a freight rate, a poor crop somewhere else: the table is compatible with all of them and settles none of them. What it does settle is that the fall was not specific to Viet Nam, which is the thing a one-origin chart would have led you to believe.

Volume and value are different lines

The same query with the value indicator instead of the quantity one returns euros, and the two do not move together. EU27 imports of cinnamon from Viet Nam were worth about 6.7 million euros in 2019, peaked at about 19.9 million in 2022, fell to about 14.7 million in 2023, and reached about 17.7 million in 2025. Divide value by quantity and the picture sharpens: roughly 2,524 euros per tonne in 2019, rising to about 3,628 in 2022, then easing to about 2,881 in 2025.

So 2025 was the largest volume the EU has ever taken from Viet Nam under this heading, at the lowest euro per tonne since 2020. A supplier presenting only the volume line is telling the truth and creating a false impression, because the natural inference from a record import figure is a tight market, and the value line says the opposite. Anybody negotiating against a chart should ask which indicator it was drawn from before they concede anything to it.

The unit value is not a price, and it should not be treated as one. It is a customs value divided by a customs weight, across every grade, every form and every contract in the year, including intercompany transfers. It cannot tell you what a container of a particular specification cost. What it can do is tell you the direction the average moved, which is enough to know whether a quote is being anchored to a year that no longer resembles the market.

Three traps in the raw figures

The first trap is the unit. Eurostat returns this series as QUANTITY_IN_100KG, so the 2025 value in the response is 61,454.76, not 6,145. Anyone who copies the raw figure and calls it tonnes has overstated the trade by a factor of ten. This is not a hypothetical error; it is the single easiest one to make with Comext, because the number looks like a plausible tonnage on its own.

The second trap is the part year. The extract used here also carries monthly data, and 2025 is complete: twelve monthly returns that sum to exactly the annual figure. 2026 is not. At the time of writing it holds five months, January to May. Comparing a five-month run against a twelve-month one produces a collapse that exists only in the calendar, and it is astonishing how often a chart does exactly that at the right-hand edge.

The third trap is assuming two customs records of the same trade agree. They do not, and they are not supposed to. Eurostat Comext records what the EU says arrived. UN Comtrade carries what Viet Nam reported leaving under the same HS 0906 heading, to the whole world rather than to the EU, and for 2023 that line is about 255 million US dollars. Its quantity field for the same year reads 58,134,300 kg, and every row in the run carries an estimation flag; the 2020 row reports a net weight of zero against a quantity of 33,263,968 kg, which is a field that has failed rather than a year with no weight in it. The values are usable. The tonnages from that source are not, and this article quotes none of them as fact.

What the record cannot tell you

Heading 0906 is cinnamon and cinnamon-tree flowers. It does not separate cassia from Ceylon cinnamon at the four-digit level, it does not distinguish quills from broken from ground, it does not mark organic from conventional, and it says nothing whatever about coumarin, moisture, ash, pesticide residues or microbiology. A rising volume is not evidence that the average lot got better or worse. It is evidence that more kilograms crossed a border.

That matters because the two questions a buyer actually has are neither of them answered by a trade table. Whether an origin can supply at all is a question the record answers well, and for Viet Nam it answers yes at a scale that has been growing for six years. Whether a particular lot meets a particular specification is a laboratory question and a paperwork question, and it is the one that decides whether a shipment is usable. Our catalog of whole and milled botanicals lists the forms we supply, and the trade desk quotes against a specification rather than against a description, because the specification is where the answerable questions live.

The honest summary of a trade statistic is that it is a good instrument pointed at the wrong question. It tells you about the market you are buying in. It tells you nothing about the goods you are buying. Both are worth knowing, and confusing one for the other is how a buyer ends up with a well-researched view of a category and a container they cannot use.

How to check every figure here yourself

Open Easy Comext, linked above, and select dataset DS-045409. The settings that produce every figure here are these: reporter EU27_2020, partner VN, product 0906, flow 1 for imports, indicator QUANTITY_IN_100KG for the tonnage or VALUE_IN_EUROS for the euros, frequency annual, and the years 2019 to 2025. Change the partner to EXT_EU27_2020 for the extra-EU total, or to Indonesia, China or Sri Lanka to put another origin beside this one. The same dataset is available as a bulk download and behind a query API for anyone who would rather script it than click through it.

That is a deliberate choice about how this piece is written rather than a courtesy. A figure a reader cannot reproduce is a figure they have to trust, and there is no good reason to ask for trust about something a URL settles in a second. If a number in this article is wrong, the query above is how you find out, and we would rather be corrected than quoted.

One practical note on doing it. Eurostat restates figures as member states file corrections, so an extract taken next month may differ slightly from the one here, particularly for the most recent year. Record the extract date whenever you quote a trade figure, exactly as the chart below does. A number without a vintage is not reproducible, and in a contract discussion a number nobody can reproduce is worth rather less than no number at all.