The July 24 Cliff: Where Does Taiwan's Tariff Fall Back To From 10%?

In early July, a sales rep at a machine-tool maker in Kaohsiung was staring blankly at a quotation for a shipment bound for Ohio. The customer asked: "That batch you're sending in August — what tariff rate are we actually looking at?"

He couldn't answer. Not for lack of homework — it's that the question simply has no standard answer right now.

What has been propping up Taiwan's exports to the U.S. for the past six months is a temporary tariff called Section 122 — a flat 10% applied globally. It pulled Taiwan's rate down from the original 32% IEEPA baseline all the way to 10% ^[1][5]. But that 10% comes with a built-in self-destruct clock: by law it can run for no more than 150 days. The clock started on February 24, and it runs out on July 24 ^[3][6].

What happens after it expires? No one can promise an answer. This piece sets emotion aside and lays out, item by item: the structure of this cliff, the three possible landing points, which Taiwanese industries stand in the most dangerous position, and what the state, industry and SMEs should each do to catch the fall.

Let's separate three numbers first, because they are often conflated:

Section 122 was designed to let the president impose a temporary surcharge of up to 15%, for no more than 150 days, when "a large and serious U.S. balance-of-payments deficit" arises ^[6]. The Trump administration imposed a 10% rate on almost all imports starting February 24, using it as a bridge to replace the more contentious original IEEPA tariffs ^[3].

The key is those four words: 150 days. This is not a policy choice — it is a statutory ceiling. So the expiration on July 24 is not something anyone switches off — it is a case of time running out and the measure automatically lapsing — unless Congress steps in to extend it ^[6].

Why Taiwan in particular should be nervous: Section 122 is a one-size-fits-all flat 10%, and it does not include the individual exemptions and differentiated rates that various countries negotiated hard to win ^[3]. In other words, the 10% Taiwan has enjoyed for the past six months was a free ride on "everyone gets 10%," not the fruit of Taiwan's own negotiation. Once the free ride stops, Taiwan has to rely on its own agreement to catch the fall — and that agreement hasn't started running yet.

This is the section that most needs honesty: which number it falls back to is currently speculation, not a settled outcome. There are at least three possible landing points, each with its own timing risk —

Path A | Back to the IEEPA baseline (the most painful). If Section 122 lapses and neither the agreement nor any other mechanism is ready to take over in time, Taiwan could in theory revert toward the original IEEPA reciprocal-tariff structure. This is the path with the largest gap on paper, and the break point SMEs fear most. (Inference, medium confidence: the actual reversion mechanism still depends on how U.S. executive action and the courts proceed.)

Path B | Section 301 takes over. Some analysis already points to the U.S. side evaluating Section 301 as a follow-on tool to Section 122 ^[6]. If that happens, the tariff structure would be renegotiated and reclassified by product — Taiwan might not face a single number but an entirely new classification table. (Inference, medium confidence.)

Path C | The 15% agreement takes effect in time (the most stable). The Taiwan-U.S. Agreement on Reciprocal Trade, signed on February 12, locks Taiwan into an "all-inclusive 15%," and grants Taiwan "most-favored" treatment under the Section 232 semiconductor investigation ^[1][7]. The U.S. side has already previewed how the agreement's tariff elements will be implemented in the Federal Register ^[2]. But it has not yet taken effect — it requires approval from Taiwan's Legislative Yuan ^[1]. Once it takes effect, 15% would replace this chaos and become the stable baseline.

These three paths are not mutually exclusive multiple choice — they are a three-way race: whether the U.S. Congress legislates to extend Section 122 before 7/24 (convenient, but low probability), when and how the Court of Appeals for the Federal Circuit rules on the case challenging Section 122's legality ^[4], and whether Taiwan's Legislative Yuan can approve the agreement before the cliff ^[1]. Each clock runs on its own; whichever one arrives first, or gets stuck, determines which box Taiwan lands in.

The most likely short-term outcome may not be a clean A or C at all, but a stretch of "transition-period limbo between the old and new regimes" — the old 10% lapses by law, the new tariff arrangement has not been finalized, and it briefly becomes unclear which number customs should even apply for clearance. For companies whose shipping schedule happens to fall right around late July or early August, this limbo is the most dangerous part: it's not that the tariff is high — it's that there's no way to quote a price at all. (Inference, medium confidence)

Three warning signals to watch: ① whether the U.S. Congress takes any action to extend or replace Section 122 before mid-July; ② the timeline and direction of the Court of Appeals for the Federal Circuit's review of the case challenging Section 122's legality ^[4]; ③ whether the Legislative Yuan's Economics Committee puts the ratification of the Taiwan-U.S. agreement on its agenda ^[1]. If any one of these three signals changes color, it directly rewrites your quotation after 7/24.

Of the three paths, the only one Taiwan can push on itself is Path C. And the choke point on Path C is not in Washington — it's in Taipei.

This is where the Taiwan-specific angle gets sharpest.

For the 15% agreement to take effect, it first has to clear the Legislative Yuan. But Taiwan right now is in a ruling-party-minority deadlock — the special budget was just heavily cut, and the general budget standoff has only just played out. Getting a trade agreement that touches U.S. relations and the distribution of interests across industries through review within three weeks, inside that kind of political structure, is not an easy bar to clear. (That the agreement must go through the legislative process is noted in ^[1]; the domestic political-structure background is covered in this publication's "After the Great Recall.")

More troublesome still, this agreement is not just a tariff number — it also touches procurement commitments, an investment tilt toward the U.S., agricultural and market-opening issues, a whole basket of sensitive items — each one touching the interests of a different industry and electoral district. Whether the opposition parties choose to let it pass, what conditions they attach, and whether they use it as leverage against the ruling party are all variables. Historical experience is not encouraging either: agreements involving concessions to the U.S. have traditionally been among the easiest things for Taiwan's legislature to hold up and politicize. The difficulty of "completing review within three weeks" is not technical — it is political will.

So, half of where this cliff lands sits in Washington's hands — whether to extend, whether to use Section 301 — and half sits in Taiwan's own legislative efficiency. The latter is one of the few levers we can actively pull ourselves — but only if the ruling and opposition parties are willing to pull this time-limited economic question out of the endless political collision and handle it on its own.

Making things more complicated, this 10% is not even stable within the United States itself. On May 7, 2026, the U.S. Court of International Trade (CIT), in a split three-judge panel, ruled that the 10% Section 122 tariff is unlawful; the government has already appealed to the Court of Appeals for the Federal Circuit ^[4].

This means two things: first, the 10% could change because of the courts even before it expires; second, even if 7/24 arrives and the tariff lapses smoothly, whatever tool comes next — Section 301 or something else — will still have to work its way around the same set of legal disputes. The uncertainty is not a one-time event in July — it is a process that will keep extending.

There is also a thread often overlooked: if the appeal ultimately upholds "122 is unlawful," then the 10% collected over the past six months could, in theory, spawn a dispute over refunds or retroactive claims — importers (including buyers on the U.S. side) who already paid the tariff may or may not get it back, depending on the final ruling and how it is administered. For Taiwanese suppliers, this means even "what rate applied over the past six months" could be reopened and reckoned with. This cliff doesn't just affect future shipments — it could also stir up accounts already settled. (Inference, low confidence, pending the final ruling and U.S. administrative action.)

For exporters, the hardest part has never been "the tariff rate is high" — it's "the tariff rate is unknown." Both 10% and 15% can be built into a cost basis and quoted; what really damages cash flow is that on the day you send out the quote, you have no idea what rate the goods will actually be charged when they clear port.

Talking about "Made in Taiwan" as one bloc when it comes to 7/24 would be badly wrong. The real dividing line comes down to one question: is your product subject to the "reciprocal tariff," or to "Section 232"? The fates of these two tracks are diverging.

The 232 side — relatively capped, relatively stable. Semiconductors and semiconductor manufacturing equipment received "most-favored" treatment under the agreement ^[1][7]; auto parts, lumber and aerospace parts have their rate capped at 15% under the Section 232 framework ^[11]. These industries have an agreement-based ceiling supporting them, so the expiration of the temporary tariff on 7/24 has a relatively manageable impact on them — because both their "floor" and their "ceiling" rest on a separate basis.

The reciprocal-tariff side — most directly exposed to the cliff edge. This is where the real cause for concern lies:

• Machine tools (a hard-hit sector): the United States is Taiwan's second-largest export market for machine tools, averaging about US$400 million a year; a 15% rate has already visibly eroded Taiwan's machine-tool price advantage, and there is a real risk of orders shifting to Japanese and South Korean suppliers ^[10]. This is the textbook case of "a tariff rate translating directly into lost orders."
• Bicycles and e-bikes: the drop from 20% to 15% brought some relief, but a larger structural migration is already underway — since 2025, bicycle and e-bike production lines have kept moving toward Vietnam and Cambodia ^[13]. The tariff is only one of several straws breaking the camel's back.
• Screws, fasteners and other traditional hardware: heavily dependent on the U.S. market, highly substitutable, thin margins — this group is squeezed hard by the reciprocal-tariff structure and has the least room to negotiate ^[12].

The same "Made in Taiwan" label: semiconductors get a moat, machine tools bleed. If the rate snaps back after 7/24, the fates of these two sides will only diverge further — a policy discussion that only talks about "what rate Taiwan is charged" misses "which Taiwanese businesses go under first."

Half the meaning of Taiwan's 15% only becomes clear once placed in the regional competition.

Taiwan's reciprocal tariff was originally set at 20%, higher than Japan's and South Korea's 15%; only in January 2026 did Taiwan negotiate it down to tie Japan and Korea at 15% ^[5][9]. That tie mattered — it stopped the bleeding of "starting the race two strokes behind." But tying is not the same as leading:

• Against Japan and South Korea: 15% pulls all three back to the same starting line; Taiwan now has to compete on technology, delivery times and supply-chain resilience, with the tariff at least no longer a deduction.
• Against Vietnam and Cambodia: these Southeast Asian bases keep pulling away assembly capacity with lower land and labor costs; the relocation of bicycle and some electronics assembly is already underway ^[13]. 15% does not solve this structural problem — it addresses "is the tariff fair," not "is the cost competitive."

This is why the agreement taking effect is urgent (inference, medium confidence): if Taiwan snaps back after 7/24 because the agreement is stuck, while Japan and Korea hold steady at 15%, Taiwan would slide overnight from "tied" back to "behind." The orders that flow away by then may not wait around for the agreement to be ratified.

State level — treat the agreement taking effect as a national-security-grade emergency. First, speed up legislative review: every day the agreement is not in effect is another day Taiwan is exposed to "the 10% has expired, the reversion is undecided"; however deep the ruling-opposition divide, this is an issue with a hard deadline that affects the livelihoods of every industry, and should not be bundled into ordinary political combat. Second, actively track the U.S. timetable: use official and semi-official channels to confirm the arrangement after 7/24 (will Congress extend it? will it switch to Section 301? how will an executive order bridge the gap?) — don't let an information gap become a disadvantage in negotiation and contingency planning. Third, keep a close eye on the Section 232 moat: the "most-favored" treatment for semiconductors and semiconductor equipment ^[1] is the hardest card in Taiwan's hand, and any loosening needs to be caught immediately; at the same time, prepare individual relief and order-transition assistance for hard-hit reciprocal-tariff sectors such as machine tools and fasteners, rather than describing everyone with a single average tariff rate.

Industry-intermediary level (associations, large manufacturers) — translate uncertainty into plannable scenarios. Build a scenario comparison table for members and the supply chain right now: what rate, what effective date, and which HS tariff codes are affected under Path A (back to IEEPA), Path B (301 takes over), and Path C (15% takes effect). Turning "we don't know" into "three known scenarios" lets members actually plan capacity and quotes. Associations for machine tools, bicycles and fasteners should also launch cross-member order and inventory coordination: rather than each firm facing Japanese, Korean and Southeast Asian competition alone, treat "geopolitical tariff risk" as a shared industry issue — speak to the government collectively, and negotiate with buyers collectively.

SME level — three actions you can take today.

• Add a "tariff adjustment clause" to contracts: for shipments going out after 7/24, spell out in black and white who absorbs a tariff change. Sample language: "If the actual U.S. tariff rate applicable at the time of import exceeds the quotation baseline (15%), the excess shall be shared equally 50/50 between buyer and seller" — turn a verbal understanding into something in writing, rather than arguing after the goods reach port.
• Negotiate "scenario-based cost-sharing" with buyers: proactively lay out the price gap from 10%→15% (or a further reversion), and agree in advance on a sharing ratio and trigger conditions, moving the dispute upstream before it happens. A buyer willing to share the risk is often a buyer willing to commit to the long term.
• Keep a cash-flow buffer: until the rate path is clear, price conservatively with a safety margin on items with high tariff exposure; adjust shipping schedules where necessary to avoid the tariff vacuum around 7/24.
• Find out exactly "which tariff you're subject to": first confirm whether your core products fall under Section 232 (semiconductors/auto parts/aerospace, which have a ceiling) or the reciprocal tariff (machine tools/hardware/bicycles, the most exposed) — that determines how worried you should be, and where to put your limited hedging resources.

Editorial note: The factual portions of this article (the 32%→10% tariff context, Section 122 taking effect on 2/24, its statutory lapse on 7/24, the 15% agreement being signed but not yet in effect, and the CIT's 5/7 ruling of unlawfulness) have each been checked against at least two sources; "which number it reverts to after 7/24" is unsettled speculation — confidence levels are marked throughout the text section by section. Readers should read it accordingly and not take it as a foregone conclusion.