Landed Cost in Brazil: Why Your Buyer’s Numbers Never Match Yours

Landed cost numbers rarely match between a US exporter and a Brazilian buyer because Brazil stacks several federal and state taxes on top of the FOB price, and one of those taxes varies by the buyer’s own state and tax regime, so two honest calculations can land on two different totals for the identical shipment.

The scenario: a strong FOB quote that stalls on landed cost

Picture a US manufacturer of industrial equipment, or a consumer goods company, that has just closed weeks of back-and-forth with a promising Brazilian buyer. The FOB quote landed well below what the buyer was paying a regional supplier, samples were approved, and the buyer seemed ready to move. Then the conversation stalls. Not because the buyer lost interest, but because the moment they tried to translate your FOB price into a real, all-in cost to receive the goods in Brazil, the number they got did not match the number you sent them.

That gap is one of the most common, and most misread, reasons a promising US-Brazil deal goes quiet. The US side usually assumes the buyer is negotiating, stalling, or losing interest. The Brazilian side usually assumes the exporter does not understand how their own country’s import costs actually work, and hesitates to keep pushing a number that will not survive contact with their finance team. Neither side is wrong to be confused. Brazil’s landed cost math is genuinely more layered than most FOB-to-door estimates a US exporter would build for any other market, and it does not resolve with a single conversion rate or a flat percentage rule of thumb.

What taxes actually stack on top of your FOB price into Brazil?

Brazil applies several distinct taxes on top of the FOB or CIF value of an imported shipment, and they are not interchangeable line items you can average into one number. There is a federal import duty (II), a federal industrialized products tax (IPI), a pair of federal social contribution taxes generally referred to together as PIS/COFINS, and a state-level value-added tax (ICMS) that is administered separately by each of Brazil’s states. Each layer has its own base, its own rules, and in some cases its own exceptions depending on the product’s classification. We are intentionally not quoting specific rates or percentages here, because the correct figures depend on the product, the buyer’s state, and the buyer’s tax regime, and a wrong number is worse than no number.

On top of that tax stack sits everything else that makes up a real landed cost: international freight, cargo insurance, port and terminal handling charges, and the customs broker’s fee for clearing the shipment through Brazilian customs. All of those vary by shipment, by port, and by the broker involved, which is one more reason a single generic markup applied to your FOB price will not hold up.

Why does the same product land at a different cost in different Brazilian states?

This is the detail that surprises most US exporters the first time it comes up. ICMS, the state-level tax in the stack, is not a single national rate. It is administered by each Brazilian state individually, and the rate that applies depends on which state the importing company is registered in, along with the specific interstate and product rules that state applies. That means the exact same shipment, at the exact same FOB price, from the exact same supplier, can land at a different total cost for a buyer registered in one state than for a buyer registered in another.

Layered on top of that, Brazilian companies also elect into different corporate tax regimes, and that election changes how some of these taxes are credited, offset, or calculated for that specific company. So even two buyers in the same state, in the same industry, buying the same product from you, are not guaranteed to land on the same total cost. This is not a loophole or an inconsistency in the system; it is simply how a federated tax structure with state-level administration behaves, and it is a normal, well-understood feature of importing into Brazil for anyone who does it professionally.

Why your back-of-envelope number and your buyer’s accountant disagree

Here is the failure mode we see most often. A US exporter, trying to be helpful and move the deal forward, builds a rough landed cost estimate for the buyer using a generic “Brazil import tax rate” found on a forum, a trade publication, or pulled from an AI tool. It looks reasonable, it is offered in good faith, and it is often wrong for that specific buyer, because it does not, and cannot, account for the buyer’s state or tax regime without that information.

“Two honest landed cost estimates, built without the same state and tax regime inputs, are not going to agree, and that disagreement is not a sign either side is negotiating in bad faith.”

Guelcos, from client conversations with US exporters on stalled Brazilian deals

When the buyer’s own accountant or customs broker comes back with a different total, the exporter often reads it as the buyer padding the number or losing enthusiasm, and the buyer often reads the exporter’s estimate as naive or uninformed about their own market. Neither read is accurate. What actually happened is that two different people, both acting in good faith, built two different estimates from an inherently variable, state-and-regime-dependent system, using different assumptions neither one stated out loud. The deal does not stall because the numbers do not match. It stalls because nobody trusts either number enough to commit to it.

The fix: a real landed cost model, not a forum tax rate

At Guelcos, we build that landed cost model directly against the buyer’s actual state and tax regime rather than a generic rate. The fix is not for a US exporter to become a Brazilian tax expert, and it should not be. That is not a reasonable ask of a sales or export team, and trying to self-teach it from public sources is exactly how the generic-rate problem happens in the first place. The fix is to get a real landed cost model built for the buyer’s specific state and tax regime, using the buyer’s actual registration details rather than a national average, before the deal stalls on a number nobody can defend internally.

Done properly, that model gives both sides the same starting point: a total cost figure the buyer’s own finance team can check against their own math and agree with, instead of a number the exporter produced alone and hoped would hold up. That single change, moving from an estimate to a model built on the buyer’s actual state and regime, is often the difference between a deal that stalls indefinitely on an unresolved number and one that moves to a purchase order.

What are the other four blockers worth checking?

Landed cost is the second most common reason a promising US-Brazil deal goes quiet, but it is not the only one. If your buyer’s landed cost math checks out and the deal is still stuck, it is worth checking four other things. First, whether the buyer currently holds an active Radar import licence through Siscomex, since that authorisation can lapse after a period of inactivity and is separate from any commercial agreement between you. Second, whether the buyer has an independent way to verify your company, since negotiating with a US supplier they have never met, based only on a catalogue and a call, sometimes creates hesitation on its own. Third, whether your product needs certification under INMETRO, ANATEL, or ANVISA before it can be legally sold in Brazil. Fourth, whether the buyer, or their logistics partner, actually knows how to run the shipping and customs clearance operation end to end, from container loading decisions through port free time and customs release. Any one of these can produce the same silence a landed cost mismatch produces, which is exactly why a quiet buyer needs a real diagnosis rather than a guess.