The cheapest line in your cross-border budget is the one most likely to disappear. Section 321 — the de minimis channel that lets sub-$800 entries cross duty-free — has carried a growing share of e-commerce import economics for years, and reforming it has been the live policy risk on every serious planning board since 2025 (S24). Mid-year review season is when next year's budget assumptions get set, so here's the outlook position, stated plainly: I don't know what happens to de minimis, you don't either, and anyone who claims to is selling something. What a CFO can do — what this whole post is — is replace the single assumption with a costed scenario set, so that whatever lands, your budget already contains it.
First, the demand side of the outlook, because it frames everything: cross-border isn't a niche you can wait out. DHL's Trade Atlas projects global trade growing at a 3.1 percent compound rate through 2029 (S12, cited in ProShip's diversification analysis), and the regional expression of that number rolls past DFW every morning — the trailer counts coming up I-35 from Laredo, the Mexico-origin first touches in the buildings along the Alliance corridor. Your cross-border exposure is growing whether or not your budget admits it. The question is only whether the duty math underneath it is an assumption or a model.
So build the model. Three scenarios, each fully costed against your actual entry data — not industry averages, yours.
Scenario one: status quo. De minimis works next year the way it works today. Cost basis: your current effective duty avoidance, measured. This requires knowing — actually knowing — what share of your inbound units and landed-cost advantage rides the sub-$800 channel. [S-cite: share of US e-commerce import entries using Section 321, most recent CBP figures]. If your team can't produce your own number inside a week, that's finding number one, and it's free.
Scenario two: narrowed eligibility. The channel survives but tightens — category exclusions, country-of-origin restrictions, data requirements that raise the effective cost of using it. Model it as a partial repricing: which SKU cohorts lose eligibility, what formal entry costs them, and what the added brokerage and compliance overhead does to per-unit landed cost. This is the scenario that punishes operational unreadiness most, because it converts a pricing question into a paperwork question mid-year.
Scenario three: full duty exposure. The conservative budget case: every current de minimis unit pays its classified rate plus entry costs. Run it SKU by SKU — duty rates vary too much by classification for an average to mean anything. The output isn't a prediction; it's a ceiling. A budget that survives scenario three doesn't fear the news cycle. That's the whole point of the exercise: buying back your attention.
Notice what the three scenarios share: they're only as good as your entry data and your classification hygiene. That's the implementation finding hiding inside the finance exercise. If last summer's invoice audit told you your surcharge model drifts — published increases understating realized by 100 to 200 basis points (S2, CNBC on surcharge persistence) — the de minimis model has the same failure mode, with bigger units. Garbage classifications in, confident garbage out.
A word on the vendor conversation, because cross-border tooling demos well in June. MetaPack and the Auctane family market cross-border duty and tax capability; Centiro goes to market on coverage breadth — 175 countries is the claim. Take all of it at face value and notice it answers a different question. Coverage tells you a system can compute today's rules in many places. A policy-shift scenario asks what your freight costs under rules that don't exist yet, weighted by your SKU mix, your entry history, your broker fees. No vendor's coverage map contains that answer — it has to be built from your data, on a model you control. Tooling executes the plan. It isn't the plan.
Since it's June, the operational footnote, from someone who spent enough summers on a Fort Worth dock to plan by the forecast: the same discipline applies to the heat. ERCOT conservation appeals and triple-digit afternoons are an annual operating condition here, not a surprise — dawn yard moves, protected first-wave doors for meltable and battery-spec freight, charging schedules that respect the grid's worst hours. Budget season is when next summer's heat plan gets funded or doesn't. Scenario planning for policy and scenario planning for weather are the same muscle: stop treating the recurring thing as news.
The tradeoff, named: building the three-scenario model costs real analyst time in a season when finance teams are buried in mid-year review, and scenario three may produce a number leadership doesn't want in the deck. Put it in anyway. The alternative is a budget with a silent bet on Washington baked into the cheapest line — and bets you don't know you've made are the only truly unhedgeable kind.
Concrete next step: the de minimis exposure review. Send us ninety days of import entry data — entry types, values, classifications, broker statements. We'll return your duty-exposure-by-scenario model inside a week: current avoidance measured, narrowed-eligibility cost by SKU cohort, and the full-exposure ceiling, formatted to drop straight into your budget pack. Next year's budget is going to take a position on Section 321 either way. Take it on purpose.