Skip to content

Canada → SE Asia

Library · Tariffs and origin

Finding the duty rate that actually applies, and why the first number is wrong

There is a free government tool that gives the applicable rate, the preferential rate and the phase-out period for a product and a market. Most exporters have never used it, and most of the rates quoted in business cases come from somewhere worse.

Duty rates turn up in business cases with remarkable confidence and remarkably little provenance. Somebody asked a freight forwarder, or a distributor mentioned a figure, or a number from a trade article got typed into a spreadsheet three years ago and has been compounding ever since.

There is a free government tool that answers the question properly, and it is used far less than it should be.

The Canada Tariff Finder

The Canada Tariff Finder lets Canadian businesses check import or export tariffs for specific goods and markets, with a focus on countries Canada has a free trade agreement with. It is a collaboration between BDC, EDC and the Trade Commissioner Service at Global Affairs Canada.

Three things it returns are worth knowing about individually, because the third is the one exporters most need and least expect.

**The rates generally applicable to all nations.** This is the ordinary rate your product faces with no agreement in play, and it is the relevant number for Indonesia, the Philippines and Thailand on this route.

**The preferential rates applicable to Canada where an agreement is in place.** For Malaysia, Singapore and Vietnam under CPTPP, this is the rate a valid preference claim reaches.

**The phasing-out period, where tariffs are being phased.** This is the part that changes how a business case should be written, and it is discussed below.

It will also compare up to three countries or products at once, which is exactly the shape of a market-selection question, and it suggests products matching a keyword or code rather than requiring you to arrive knowing your classification.

The phase-out is the thing nobody models

An agreement rarely takes a rate to zero on the day it enters into force. Sensitive goods, and agricultural goods are usually sensitive, are staged: the rate steps down over a number of years to a final rate, which may or may not be zero.

Two consequences follow, and both are routinely missed.

**"Duty-free under CPTPP" is frequently wrong today and right eventually.** A company reading a summary that the agreement eliminates tariffs, and modelling zero, can be modelling a rate that arrives several years after the shipments in the plan.

**The rate improves on a schedule you can plan against.** A product that is marginal at this year's rate may be comfortable at the rate three years out. That is a genuine strategic input: it can justify entering a market at thin margin to establish a position that improves mechanically, and it is the sort of argument a funding application or an internal business case can be built on.

Both require knowing the actual staging for your line in that market, which is what the tool provides and what guessing does not.

Everything depends on the classification

The tool returns a rate for a code. Give it the wrong code and it returns a perfectly accurate rate for a product you are not shipping.

Worth holding on to how the numbering works, because it explains where the answer stops being portable. Harmonized System codes are developed by the World Customs Organization with a standardised two to six digit nomenclature: the first two digits indicate the product category and the next four to six the subcategories. Countries may add digits to identify products more specifically, taking the code up to ten.

So the first six digits are shared and everything below is national. A rate quoted against a ten-digit code is quoted against somebody's national line, and if that somebody is Canada and not the destination, it is the wrong country's number.

The order that works: establish the classification first, at six digits with the determining facts recorded, then look up the rate, then have the destination's own line confirmed by whoever will file the entry.

Four reasons the landed number differs from the quoted one

The rate is the beginning of the calculation, not the end, and the gap between the two surprises people who have only ever sold domestically.

**Duty is charged on a value that is defined.** Customs value is determined by rules, and it is not automatically your invoice price. Depending on the terms of sale and the destination's rules, freight and insurance may or may not be inside the base the duty is applied to, which changes the amount without changing the rate.

**There is usually more than one charge.** Import duty is one line. Value-added or goods and services tax, excise where applicable, and various processing fees are others, and in several markets the tax is much larger than the duty. Singapore charges essentially no duty on food, which makes the tariff conversation there close to irrelevant while the admission requirements remain as real as anywhere.

**Preference has to be claimed and supported.** A preferential rate is available, not automatic. If the certification is not in order, the entry is made at the ordinary rate and the saving does not happen, whatever the tool says.

**Somebody has to file it correctly.** The importer enters the goods. A rate the exporter established and the importer did not apply is a rate nobody received.

That last one is worth a conversation rather than an assumption. Ask your buyer what rate they expect to enter at and under what basis, before shipping, and reconcile it with what you found. A mismatch discovered in an email costs an email.

Who to believe when two sources disagree

Exporters usually end up with more than one number, and knowing which to trust is most of the work.

**Your own lookup against a classification you established.** Good, and it is the one you can explain. Its weakness is that it stops at the shared six digits unless you also have the destination's national line.

**Your buyer or their broker.** They file the entry and know their own national lines, which makes their number the one that will actually be charged. Its weakness is that they are classifying your product from your description, and they may be doing it quickly.

**A freight forwarder.** Useful for the mechanics of moving goods and not the right party to determine a classification. Forwarders quote from experience with similar shipments, which is exactly the reasoning that puts a product on a neighbouring line.

**Something you were told.** No provenance, and it is the source most business cases actually rest on.

When the first two disagree, that is a useful disagreement, not a problem, and it should be resolved before shipping rather than after. Usually it comes down to the destination's national line beneath a six-digit heading you both agree on, and the resolution is a short exchange with the facts about the product in it.

If it cannot be resolved that way, the product is a candidate for an advance ruling from the destination's customs administration, which settles it with the authority that would otherwise raise it.

Use it for choosing markets, not just for costing one

The comparison feature is the underused part. Because the tool will put up to three markets side by side for the same product, it answers a question most exporters approach far less systematically: where should we go first.

A product facing a high ordinary rate in one market, a phased preferential rate in another and near-zero duty in a third has a sequencing answer visible in about ten minutes, and it is an answer with a source rather than an impression.

That does not decide the question by itself. Duty is one input among several, and on this route it is often not the decisive one: the gates that dominate the work are establishment approval, certification, registration and documentation, none of which a tariff tool sees. A market with an attractive rate and a two-year approval chain is not the easy option.

But it is a real input, it is free, and having it makes the rest of the analysis honest.

Quotas and seasons sit behind the same code

One complication the headline rate does not show, and it bites in agriculture more than anywhere else.

Some goods are subject to tariff rate quotas: a volume enters at one rate and everything beyond it at another, usually much higher. The quota may be allocated, administered by the destination, and already spoken for by the time a new entrant asks. A rate quoted without reference to which side of the quota a shipment falls on can be accurate and useless.

Seasonal rates work similarly, applying different treatment to the same product depending on the time of year, and they exist to protect domestic harvests. A product that is attractive to export in one month can be uncompetitive in another with no change to anything except the calendar.

Neither shows up in a single number. Both are visible if you go looking, and the reason to go looking is that they change the shape of the business rather than the size of a line item: a quota question is about whether volume is available at all, and a seasonal rate is about when shipping makes sense.

For a product where either applies, the rate is the least interesting part of the answer.

What to do with the answer

  • Establish the classification first, and record the facts that determine it
  • Look up the ordinary rate and the preferential rate separately, so you know what preference is actually worth
  • Read the phase-out schedule and model the rate for the years in the plan, not the final rate
  • Check whether the duty saving justifies the cost of qualifying for preference, and be willing to conclude it does not
  • Confirm the destination's national line with the party who will file the entry
  • Model the other charges, especially the destination's consumption tax, because it is frequently larger than the duty

The goal is a landed cost that survives contact with a customs entry. Most do not, and the reason is almost always that the number at the top came from a conversation, not from a source.

Sources

Last reviewed: September 25, 2026

Read next

More on tariffs and origin

All 7 articles on tariffs and origin →