Where US Candle Imports Come From in 2026 (HS 3406)

If you run a candle brand, you have probably spent 2026 bracing for "the China tariff." Here is the problem with that: for candles, you are mostly fighting the last war. In full-year 2025 the US imported $1.16B of candles, tapers and the like (the HS 3406 customs line), and China supplied just 3.4% of it. Vietnam supplied more than half. The tariff exposure that actually shows up on your landed cost is a Vietnam-and-Canada story, and the sooner you model it that way, the sooner you stop hedging a risk that barely exists and start managing the one that does.

Where US candles actually come from

Start with the total. US imports of HS 3406 (the Harmonized System line for "candles, tapers and the like") came to $1,159,994,894 in full-year 2025, up 5.0% from $1.11B in 2024. That growth matters because the candle category is structurally import-fed. Domestic candle manufacturing in the US is flat at roughly $2.7B (IBISWorld), while retail demand runs $3.6B to $3.9B (Grand View Research). Imports fill the gap between what the US makes and what it buys, and that gap is widening.

Now the concentration. Vietnam alone supplied 53.7% of all US candle imports in 2025, $622.6M of the $1.16B total. That is more than the next four origins combined. Canada was second at 21.4% ($248.7M), and then it drops off a cliff: Mexico at 4.7%, India at 4.5%, China at 3.4%. The top two origins, Vietnam and Canada, are 75.1% of all US candle import value between them. Add Mexico, India, and China and the top five reach 87.8%.

When I talk to founders running a candle or home-fragrance brand at $5M to $30M, the instinct is almost always to frame sourcing risk as a China question, because that is the headline they have been reading for two years. The data says otherwise. The structural center of gravity for US candle sourcing is Vietnam, and your tariff exposure follows the dollars. Here is the full top-10 origin table for 2025.

Canada surged, China faded

The year-over-year shift is the part most operators have not internalized yet. Canada's US candle imports jumped from $157.8M in 2024 to $248.7M in 2025, up 57.6%. That single move took Canada from a 14.3% share to 21.4% and made it the second-largest origin in the category. Nothing else in the top six grew like that. Vietnam edged up 2.8%, and Mexico, India, China, and France all shrank.

China shrank the most. Its candle imports fell from $63.0M in 2024 to $39.7M in 2025, down 37.0%, dropping its share from 5.7% to 3.4%. That is not a one-year wobble. It is the continuation of a long structural retreat, and the reason is specific: candles from China carry the A-570-504 antidumping order on petroleum wax candles. This is not Section 301 and it is not the reciprocal-tariff news cycle. It is a decades-old, punitive antidumping duty that sits on top of any tariff, and it has quietly designed China out of the US candle market over many years.

The practical read for a brand: if your sourcing risk model still treats China as your candle exposure, you are anchored to a 2018 mental model. The category has already moved. When I talk to founders this size, the ones who get caught flat-footed are the ones who spent a year planning a China contingency and never built the Vietnam one. Here is the full trend table.

The tariff map for HS 3406 in 2026

Here is the fact that reframes the entire cost conversation: the HS 3406 base most-favored-nation rate is 0%, duty-free. There is no base duty on a candle entering the US. That means your tariff cost is built entirely from the country-specific add-on layers stacked on top of zero, and those layers differ sharply by origin.

For most origins, the only add-on is the country-specific reciprocal-tariff layer. As of the most recent trade reporting we could verify (The Vision Council reciprocal-tariff chart), all reciprocal tariffs were "deemed invalid and are not being implemented as of February 24, 2026." Treat that as fast-moving trade reporting, not statute. The point for planning is that the reciprocal layer is the swing variable, it is volatile, and it is where your scenario modeling should live. Re-verify it at the moment you commit a PO.

For Canada and Mexico, the USMCA changes the math. Candles from those two origins enter duty-free if they qualify as originating under USMCA rules of origin and you actually claim the preference. That "if" is the whole game. The rate is zero, but only if your paperwork establishes origin and you file the claim. For China, the stack is different again: the Section 301 layer (verify whether your exact 3406 subheading is on a current USTR list) plus the A-570-504 antidumping order on petroleum wax candles. That antidumping duty is the load-bearing cost, and it is punitive.

What this means for your tariff math

The headline takeaway is simple: your candle tariff exposure is a Vietnam-plus-USMCA question, not a China question. More than half your category's import dollars sit on the Vietnam line, and the cost on that line is whatever the reciprocal-tariff layer ends up being on top of a zero base rate. That is the number that moves your gross margin, and it is the one worth building a real scenario around.

When we have struggled with this on the modeling side, the thing that worked was forcing the order of operations: model the biggest-dollar origin first, not the scariest-sounding one. A brand sitting on 60% of its candle COGS in Vietnam and 3% in China should spend 95% of its tariff-planning energy on Vietnam. It feels counterintuitive because China is the one in the news, but the math does not care about the news. The pattern we see again and again is operators who model a 25% China scenario in obsessive detail and never run a single Vietnam scenario, then get surprised when a reciprocal change on Vietnam moves their landed cost by real money.

There is also a cost-pressure point worth keeping straight. The pressure on candle landed costs in 2026 is a policy story, not a factory-price story. Underlying import input prices from China have been flat-to-deflating, not spiking. So when your landed cost moves, assume it is the tariff layer doing the work, and model the policy, not the factory.

What to do about it this quarter

Five concrete moves, in order.

First, pull country of origin from your actual customs entries, not from your supplier list. A single supplier can ship the same SKU from two countries in the same year, and your duty follows the country, not the vendor relationship. Group your candle COGS by origin.

Second, map landed cost per SKU: FOB unit cost plus freight plus the duty stack for that SKU's origin. Do this at the SKU level for your top sellers, because that is where the dollars concentrate and where a tariff change actually bites.

Third, if you source any Chinese candles, confirm whether they fall within the A-570-504 scope. The order targets petroleum and paraffin wax candles. Soy, beeswax, and some novelty candles may sit outside scope per Commerce scope rulings. This is the single most important compliance question for any brand still touching Chinese candle supply, and the wrong assumption either over-reserves or leaves you exposed.

Fourth, if you import from Canada or Mexico, make sure you are actually claiming USMCA preference and that your origin documentation holds up. The duty-free rate is worthless if the paperwork does not establish origin. This is found money for a lot of brands that assume "Canada" automatically means "zero."

Fifth, run two reciprocal-tariff scenarios on your top-10 candle SKUs: a zero-reciprocal case (consistent with the Feb 24, 2026 reporting that reciprocal tariffs were not being implemented) and a stress case at whatever rate your Vietnam line would plausibly carry if that changed. The gap between those two numbers is your real planning range, and it is almost entirely a Vietnam number.

Founders bracing for the China tariff on their candle line are mostly fighting the last war. China is 3.4% of US candle imports and shrinking; Vietnam is 53.7% and is where your dollars actually sit. The base duty is zero, so your whole cost story is the country-specific layer on top, and the only line big enough to move your margin is Vietnam. Model that one first.

For more on how tariff and trade costs flow through DTC unit economics, see our interim CFO services overview, and compare the origin-and-tariff picture in adjacent categories in our beauty import origins breakdown and supplements import origins breakdown.

Sources and methodology

The origin and value data is from the US Census Bureau, foreign trade imports timeseries, endpoint api.census.gov/data/timeseries/intltrade/imports/hs, pulled through the Census MCP because the public endpoint returns "Missing Key" without authentication. The HS line is 3406, where I_COMMODITY_LDESC reads "CANDLES, TAPERS AND THE LIKE," verified on every pull.

The value variable is CON_VAL_YR, the year-to-date imports-for-consumption total value in USD. The December reading is the full-year cumulative total, so we used time=2025-12 and time=2024-12. Each country's share is its CON_VAL_YR divided by the "TOTAL FOR ALL COUNTRIES" line on the same HS code and period. Exact 2025 figures: total $1,159,994,894; Vietnam $622,587,070 (53.67%); Canada $248,719,157 (21.44%); Mexico $54,671,146 (4.71%); India $52,462,200 (4.52%); China $39,685,053 (3.42%); the USMCA bloc combined $303,390,303 (26.16%). Exact 2024 figures: total $1,105,066,874; Vietnam $605,879,485; Canada $157,817,008; China $62,973,925.

The tariff and trade-remedy structure was assembled from USITC HTS, USTR Section 301 notices, and Commerce / Federal Register sources, triangulated with Perplexity and Parallel.ai. The HS 3406 base most-favored-nation rate is 0%, duty-free, so total landed duty for any origin is the base of zero plus the country-specific reciprocal layer plus, for China only, Section 301 and the antidumping order. The A-570-504 antidumping order on petroleum wax candles from China was continued per Federal Register notice 2021-23560 (October 29, 2021). A PRC-wide rate around the 108% range has been cited historically, but it must be verified against current Commerce cash-deposit instructions before you state a number, so we deliberately do not assert a specific rate here.

The reciprocal-tariff status reflects trade reporting (The Vision Council reciprocal-tariff chart) that all reciprocal tariffs were "deemed invalid and are not being implemented as of February 24, 2026." This is directional reporting, not statute, and it moves fast, so every reciprocal-rate reference in this post is conditional and should be re-verified at the moment you commit an order.

External corroboration: OEC independently confirms the 2025 HS 3406 totals (US imports $1.16B, Vietnam $623M, Canada $249M, Mexico $54.7M, India $52.5M, China $39.9M), matching the Census pull within rounding. We use the Census CON_VAL_YR figures as primary throughout. Market sizing context (US retail candle market $3.6-3.9B, domestic manufacturing roughly $2.7-2.8B) is from Grand View Research and IBISWorld and is used only to explain why the category is import-fed, not as part of the trade math.

A limitation worth stating: HS 3406 covers candles, tapers, and the like as one customs line, so it does not separate paraffin from soy or beeswax, and the import data cannot by itself tell you which of your Chinese SKUs fall inside the A-570-504 antidumping scope. That scope question has to be answered SKU by SKU against the order, not inferred from the trade totals.

Frequently asked questions