I get this question on roughly every other discovery call: "We're trying to decide between an FTZ and a bonded warehouse. Which one should we use?" The right answer depends on three numbers about your business. Here's how to think through it.
The short version
FTZ wins when: you have a high-volume, primarily domestic-distribution operation with a non-trivial percentage of dutiable goods. Think: a retailer importing $100M+ and selling through U.S. stores or e-commerce.
Bonded warehouse wins when: you have a high re-export ratio, simpler operations, or you genuinely don't know yet what you'll do with the inventory after it arrives.
That's the elevator pitch. The rest of the post explains why.
What each one actually is
A bonded warehouse is a CBP-supervised storage facility where you can hold imported goods without paying duty for up to 5 years. When you withdraw goods for U.S. consumption, you pay duty at the rate in effect on the withdrawal date. When you re-export, you pay no U.S. duty at all.
An FTZ is a more flexible regime. You can hold goods indefinitely. You can manipulate, kit, or manufacture inside the zone. You pay duty on a "weekly estimated entry" basis, in aggregate, with a single fee per entry rather than per shipment. When you re-export, no U.S. duty. And, importantly, you can elect "non-privileged foreign" or "privileged foreign" status on goods, which lets you choose between duty rates on raw materials vs. finished products.
Both defer duty. FTZ has more operational flexibility. Bonded warehouse has less administrative overhead.
The three numbers
To pick correctly, you need three numbers about your operation:
Number 1: your annual imports value. The higher this is, the better FTZ looks. The reason is the weekly entry fee. CBP charges a flat per-entry fee of $625 (capped). In a bonded warehouse, you pay it on each entry. In an FTZ, you pay it weekly regardless of volume. For a high-volume importer, this alone saves six figures.
Number 2: your re-export ratio. What percentage of your imports leave the U.S. without being sold here? If this is below 10%, the FTZ wins on every dimension that matters. If it's above 40%, the simplicity of a bonded warehouse starts to win. If it's between 10% and 40%, it depends on the other two numbers.
Number 3: your operational complexity inside the storage facility. Do you just store and ship? Or do you kit, label, repackage, or assemble? If you do anything beyond storage, FTZ is the only one of these two regimes that legally allows it. Bonded warehouses are strict about not letting goods be manipulated in ways that change their nature.
You don't have to pick one. Some of our customers run an FTZ at their primary distribution center and a bonded warehouse at a port-of-entry where they receive overflow or transshipped inventory. The numbers below assume you're picking one, but in practice, the bigger you are, the more likely you'll run both.
The side-by-side
Here's how the two compare on the dimensions that matter most:
Duty deferral
FTZ: indefinite. Bonded: up to 5 years. Both effectively work as duty deferral for typical inventory turn cycles.
Re-export duty treatment
FTZ: no U.S. duty paid on re-exports. Bonded: same. Tie.
Manipulation rights
FTZ: extensive. You can kit, label, repackage, assemble, and in some cases manufacture inside the zone. Bonded: limited. You can sort, separate, and clean. You cannot manufacture or substantially change goods. FTZ wins.
Filing burden
FTZ: weekly estimated entry plus monthly recap. More frequent but in aggregate. Bonded: per-shipment entry on withdrawal. More documents but lower coordination overhead. Slight edge to bonded for simple operations.
Tariff timing
Both regimes let you elect "current rate" at the time of withdrawal, which is the duty deferral benefit. FTZ adds the ability to choose Privileged Foreign status, which locks in the duty rate at admission to the zone. For high-volatility tariff categories, this is a meaningful hedge.
Setup cost
FTZ: significant. The application and activation process takes 6 to 18 months and costs $50k to $250k depending on the zone and operator. Bonded: faster. A bonded warehouse application typically clears in 60 to 120 days at much lower cost. Bonded wins here, especially for testing the model.
Ongoing complexity
FTZ: substantial. Quarterly reconciliation, weekly entries, monthly recap, FTZ Board annual reports. Without good software, this burns a multi-person trade compliance team. Bonded: simpler. Standard entry filings on withdrawal. A solo broker can typically handle it. Bonded wins for low-volume operators.
The break-even math
For most of our customer profiles (retailers, brands, manufacturers in the $100M+ import range), the break-even for FTZ vs. bonded happens around $40M of annual dutiable imports. Below that, the setup and operational overhead of an FTZ outweighs the savings. Above that, FTZ pulls ahead and the gap grows.
If your re-export ratio is above 30%, slide the break-even up to about $80M. If your operations include manipulation (kitting, labeling, etc.), slide it down to about $25M because the FTZ regime is the only one that legally accommodates the manipulation.
These are rough thresholds. The right way to do this analysis is on your actual data, with your actual tariff profile, against your actual cost structure.
What we'd actually recommend
If you have an active FTZ and you're considering replacing it with a bonded warehouse: don't. The savings rarely materialize and the operational restrictions on bonded will frustrate you within a year.
If you have a bonded warehouse and you're considering upgrading to an FTZ: model it on your real numbers first. The right time to make the jump is when (a) your import volume is comfortably above $40M, (b) your re-export ratio is below 30%, and (c) your operational complexity is moving toward kitting or manipulation. If any two of those three are true, run the model.
If you have neither and you're picking from scratch: start with a bonded warehouse for the first 12 to 18 months. Use that time to baseline your true import volumes and re-export ratios. Then re-evaluate FTZ once you have real data.
The thing not to do is to pick based on what your supply chain consultant pitched first. Both regimes work. The wrong one for your operation will quietly cost you for a decade.
If you want a model run on your actual numbers, we'll do it. Bring last year's import volumes, your top 20 HTS codes, and your approximate re-export ratio. We'll show you the break-even crossover within 48 hours.