HS 9506 sporting goods imports: China falls below 50%

If you import anything under HS heading 9506 (sporting goods and physical-exercise equipment), 2025 was the year the ground moved under your sourcing model. US imports of the category totaled $7.03B, down 7.3% from 2024, and inside that shrinking pie China dropped below half the market for the first time. The orders did not vanish. They moved to Vietnam, Taiwan, and Thailand. This post charts where US sporting goods actually come from now, and then does the part most coverage skips: the by-origin tariff math, so you can model your own landed cost instead of guessing.

The 2025 origin mix: China below half for the first time

US imports of HS 9506 goods were $7.03B in 2025 on an imports-for-consumption basis, down 7.3% from $7.59B in 2024. That is the Census Bureau's full-year cumulative figure, measured at month 12. The category contracted in dollar terms, which matters: this is not a story of a growing market reshuffling its suppliers. It is a shrinking category whose biggest supplier is shrinking faster than the whole.

China supplied 48.9% of the category in 2025, or about $3.44B. A year earlier it was 56.4%. That 7.5-percentage-point drop is the first time China has sat below half of US sporting-goods imports in the modern series. The next two origins, Taiwan and Vietnam, are now both within striking distance of each other near the 14% mark, and everything past them is a long tail of single-digit-percent suppliers.

When I talk to founders importing at this size, the first instinct after a number like that is relief: "good, we are diversified, China is under half." That read is a trap. Half of a $7B category is still an enormous single-country dependency, and the concentration in your specific SKUs is almost always higher than the category average, because most brands buy a narrow slice of 9506 from one or two factories. The category being 48.9% China does not tell you what your purchase order is.

Where the orders went: Vietnam, Taiwan, and the share shift

The orders that left China did not evaporate, they relocated, and the destinations are legible. Vietnam was the standout: up 31.2% year-over-year to $921M, lifting its share from 9.3% to 13.1%. Taiwan grew a steadier 3.5% to $976M and now holds 13.9%, making it the #2 origin for the category and a more components-and-technical-goods alternative to China. Thailand grew 5.4% to $221M. Mexico and Canada were roughly flat to slightly down.

The table below carries the full per-origin detail behind both charts.

The pattern we see again and again is that the share shift looks tidier on a chart than it feels in an operating account. A brand does not move 30% of its volume to Vietnam in a quarter. It qualifies one factory, runs a pilot PO, fights through longer lead times and a quality ramp, and only then shifts a line. The Census numbers are the sum of thousands of those slow, expensive migrations, not a switch anyone flipped.

The tariff bill by origin (and why "just move to Vietnam" isn't a free lunch)

Here is the part that changes the decision. Tariff exposure on 9506 is no longer a single-country line item, and the origin you move to carries its own regime. Using exercise equipment (HS 9506.91) as a representative subheading, the mid-2026 stack looks like this: China-origin goods carry roughly 27% all-in, built from a 4.6% MFN base, the 7.5% Section 301 List 4A duty, and a 15% Section 122 global surcharge that is in force through July 24, 2026. Vietnam carries about 25%: the same 4.6% base plus a flat 20% reciprocal tariff that took effect July 2, 2025 and was maintained under the October 2025 US-Vietnam framework. Taiwan sits around 20% (base plus the 15% surcharge, with no China 301). USMCA-qualified Mexico can be 0% if it meets the rules of origin.

Read the China and Vietnam rows next to each other and the "just move to Vietnam" reflex falls apart. The gap is about two points of duty, and Vietnam's 20% reciprocal tariff has no scheduled sunset the way the China surcharge does. A two-point duty difference is the kind of thing a single quarter of FOB price movement or one freight-rate swing erases entirely. The honest version of the diversification case is not "Vietnam is cheaper." It is "Vietnam reduces single-country concentration risk, at a landed cost that is roughly comparable, and you should size that move on risk, not on a tariff saving that may not exist."

Two caveats keep this honest. First, base MFN varies across the 9506 heading: balls, golf gear, and nets do not all carry the 4.6% rate that 9506.91 does, so map your actual 10-digit HTS lines. Second, the Section 122 surcharge is a moving target. If it expires on July 24, 2026 and is not extended, China drops to roughly 12% and Taiwan to about 4.6%, which reorders the whole table. Date-stamp every rate you put in a model.

Freight is stable; tariffs are the swing factor

It is tempting to blame freight for landed-cost pain, because freight is the line operators watched obsessively in 2021 and 2022. The data says that is the wrong thing to obsess over right now. The Cass Freight Index (Shipments) read 1.041 in May 2026, essentially flat against roughly 1.054 a year earlier. Inbound Asia air freight (BLS series IC1312) was 179.5 in April 2026 versus 170.8 in April 2024, a modest rise. Freight is a stable-to-slightly-rising cost layer, not the variable swinging your landed cost.

When we have worked through a landed-cost model with a brand at this stage, the freight line usually moves the answer by a point or two, while the duty line can move it by ten or more. That is the whole argument for spending your modeling time on the tariff stack. A 2-point freight wobble is noise next to a 27% versus 0% duty spread across origins. Put your attention where the dollars are.

How to model your own landed cost

The math is simple enough to do in a spreadsheet, and you should, because the category average will mislead you. For each candidate origin, compute landed cost as FOB order value, plus duty, plus freight, plus broker and Merchandise Processing Fee (MPF). Duty is FOB multiplied by the all-in duty rate for that origin from the table above. The effective uplift over FOB is (landed cost minus FOB) divided by FOB.

Worked example. Take a $100,000 FOB order of exercise equipment with $8,000 of freight and $400 of broker/MPF. From China at 27.1% all-in, duty is $27,100, so landed cost is $135,500, a 35.5% uplift over FOB. From Vietnam at 24.6%, duty is $24,600 and landed cost is $133,000, a 33.0% uplift. From USMCA-qualified Mexico at 0%, duty is $0 and landed cost is $108,400, an 8.4% uplift. The China-to-Vietnam move saves $2,500 on this order; the China-to-Mexico move saves $27,100, if you can actually source it there and meet the rules of origin.

The point of running it on your own numbers is that the inputs that matter (your FOB price by factory, your freight mode, your real HTS lines) are specific to you, and the duty rate is volatile enough in 2026 that any model needs editable, date-stamped rate fields. If you want help building that model against your actual COGS mix rather than a category illustration, that is exactly the kind of work our fractional CFO services cover.

What to do with this if you import sporting goods

Three moves, in order. First, map your SKUs to their actual 10-digit HTS lines and pull the real base rate for each, because 9506.91's 4.6% is not the universal base across the heading. Second, quantify your single-country concentration at the SKU level, not the category level. The category is 48.9% China; your top three SKUs might be 90% China, and that is the number that should drive the conversation. Third, pressure-test any diversification scenario against the by-origin tariff table, not the headline, and re-run it with the Section 122 surcharge both on and off so you can see the post-July-2026 sensitivity before you commit capital.

For the broader picture on how tariff exposure flows through DTC unit economics, see our footwear import tariff tracker and electronics import tariff map, which run the same origin-and-duty analysis on adjacent consumer categories.

China under half the category reads like diversification, but it is not the number that runs your business. Your purchase order is. The category being 48.9% China tells you nothing about whether your top SKUs are 90% China, and the by-origin tariff math says moving off China is a risk decision, not a guaranteed saving. Model your own landed cost, date-stamp the rates, and decide on the dollars in front of you.

Sources and methodology

Census imports (primary). Import values come from the Census Bureau's USA Trade International Trade Imports HS timeseries, endpoint https://api.census.gov/data/timeseries/intltrade/imports/hs, variable CON_VAL_YR (imports for consumption, cumulative year-to-date) for I_COMMODITY=9506 at time=2025-12 and time=2024-12. Per-country values were pulled via CTY_CODE and the API's returned CTY_NAME was treated as authoritative where a code-to-country mapping was ambiguous. "Rest of world" is the total minus named origins (about $661M). All figures are full-year and measured at month 12.

Tariff rates. Rates were verified via Perplexity regulatory research against primary sources as of 2026-06-13. HS 9506.91 base MFN is approximately 4.6% (Column 1 General, USITC HTS). The Section 301 List 4A duty on China-origin goods is 7.5%. The IEEPA reciprocal and fentanyl tariffs were struck down by the Supreme Court on February 20, 2026 and replaced by a Section 122 global surcharge (10%, then 15% from February 22, 2026), time-limited through July 24, 2026 unless extended. The US-Vietnam reciprocal tariff is 20%, effective July 2, 2025 and maintained under the October 2025 framework. Primary sources cited include USTR Section 301 materials, White House presidential actions, the Yale Budget Lab state-of-US-tariffs analysis, and USITC's HTS. The 2026 tariff regime is volatile, so re-verify the live HTS line and Section 122 status before acting.

Freight context. Freight indicators come from FRED: the Cass Freight Index Shipments series FRGSHPUSM649NCIS (index Jan 1990 = 1; May 2026 = 1.041) and the Inbound Price Index for Air Freight, Asia, BLS series IC1312 (index 2000 = 100; April 2026 = 179.5, April 2024 = 170.8). Cass measures US domestic freight shipment volume, not ocean container spot rates, so it is used as a general freight-cost backdrop rather than a China-import spot rate.

Limitations. Census HS 9506 is the 4-digit heading; the subheading mix (9506.91 exercise equipment versus balls, golf, and nets) is not broken out here, and base MFN varies by 10-digit line, so the tariff-stack chart is illustrative for 9506.91. The tariff figures are stacked statutory rates, not effective-rate averages, and actual entries vary by exclusions, USMCA claims, and any AD/CVD orders. No ocean-container spot-rate series was available in FRED for this run, so Cass stands in as a proxy backdrop.

Update cadence. This analysis is refreshed when full-year Census import data and any material tariff change land. The next likely trigger is the July 24, 2026 Section 122 surcharge expiry date, which, if it lapses, reorders the by-origin tariff table.

Frequently asked questions