Import tariffs have changed multiple times since January, but the cost of imported goods has barely moved. With tariff policy in flux and shipping costs soaring from the Strait of Hormuz crisis, U.S. manufacturers must redesign their products to build resilience against relentless disruptions.

Importers celebrated on Feb. 20, when a 6-3 Supreme Court ruling declared $164.7 billion in tariffs collected under the International Emergency Economic Powers Act (IEEPA) unlawful, putting massive refunds in play. The victory lasted exactly four days.

On Feb. 24, the tariffs returned as President Donald Trump declared a 10% global surcharge under a different tariff rule, Section 122. When that expired on July 24, the Trump administration immediately imposed Section 301 duties covering roughly 99% of American imports, followed by a 50% tariff on certain Canadian goods in August.

The tariff policy spanned three separate statutes but produced the exact same outcome. Landed costs sit near where they did in January, and nobody can say where they will sit next January. The number that matters is not the rate; it is the uncertainty.

Besides import tariffs, there is another shock. Second-quarter crude-oil and petroleum-liquid flows through the Strait of Hormuz collapsed from a prewar 21.6 million barrels daily to just 4.9 million, with the Energy Information Administration forecasting no return to normal before early 2027. Brent crude hovered near $90 in August, up from $65 a year earlier, driving up bunker costs that carriers quickly passed along. Diverting around Africa through the Cape of Good Hope doubled spot container rates on affected routes.

Reshoring Is a 10-year Answer to a One-Week Problem

Tariffs are meant to provide incentives for U.S. manufacturers to reshore by bringing production home. In reality, reshoring remains formidable, and meaningful headway has been scarce.

Indeed, manufacturing payrolls stood at 12.64 million in August, still below the February 2020 level of 12.74 million. The bottleneck is not capital. It is skilled labor. Deloitte projects 1.9 million manufacturing jobs could go unfilled by 2033 if skills gaps are unaddressed.

The push to reshore is a decade-long risky bet that does nothing for this fiscal year. Firms still need to keep importing from China, Vietnam and India. They need levers that work now.

The Cost U.S. Manufacturers Can Actually Control

Duties and freight scale with units shipped. Carrying excess inventory generates steep losses through markdowns, liquidation and disposal, scaling in tandem with forecast error.

Forecast error itself is compounded by two controllable variables: SKU breadth and lead-time commitment. When transit times lengthen, firms must commit further in advance based on less accurate forecasts while carrying more in-transit inventory. As forecast uncertainty grows, safety stock mounts, and the risk of unsold goods increases.

The upfront hit from freight and tariff spikes stings, but the downstream inventory penalty cuts far deeper.

There are three levers that U.S. manufacturers can pull to ease the pain.

Cut variety. Focus on producing popular items. Retailers are already doing it. BJ’s Wholesale told analysts in August it will drop about 20% of its SKUs, reducing from roughly 7,500 items toward 6,000. The trade-off is real, however: pruning tail demand concedes niche buyers and shelf presence.

Postpone differentiation. Design products so they stay generic as long as possible. Hewlett-Packard proved this.  In the 90s, HP shipped generic DeskJets out of Singapore and localized them at a European distribution center. Manufacturing cost rose slightly. Total manufacturing, shipping and inventory cost fell by about a quarter. One generic printer serving two customer segments halved distribution-center inventory. Crocs runs the same play today: product variety lives in the Jibbitz charm, not the shoe. The charms are high-margin, roughly 8% of brand revenue, and they sell personalization without multiplying molds or unsold clogs.

Build to order where the margin allows. A shirt cut to a customer’s measurements is rarely returned and never marked down. Nike and Nordstrom both run this at scale.

The Tariff Twist Executives are Missing

Postponement now does something it did not do in 1997. Ad valorem duties apply to customs value at the border. Import a generic module, add the differentiating content in the U.S., and the dutiable base falls. That matters more since April, when a proclamation shifted Section 232 assessment on derivative articles from metal content to the full value of the article.

Late-stage differentiation is a tariff strategy, not merely an inventory strategy. And unlike a lobbying campaign, it survives the next change of statute.

The obstacle is organizational, not technical. Marketing is paid on revenue and wants more options. Engineering is paid on bill-of-materials cost and resists the pricier universal part. Operations wants stable volume. Postponement requires someone to arbitrate with a shared cost model. HP’s real achievement was that negotiation.

Manufacturing executives should ask a single question this quarter: What share of our landed cost is locked in before we know who the customer is? A firm that cannot answer is running a book of bets it has never priced. Tariff refunds, if they ever arrive, are merely a windfall.

Product architecture is a deliberate strategy against tariffs shocks and transit volatility.