THE HILL REPORT

The Rule Lands Years Before the Effect Does

Connor Hill · InsightfulWord · August 17, 2026

There is a specific piece of machinery that the U.S. government now uses to remove foreign components from American systems, and almost nobody outside the trade bar can describe how it works. It is worth understanding, because it has become the default instrument, because it operates on a timetable that bears no relation to how such actions are reported, and because the gap between those two things is where a great deal of confident commentary goes wrong.

The authority is the Information and Communications Technology and Services regime, administered by the Department of Commerce. Its distinguishing feature is that it does not work like a tariff or a sanction. A tariff applies at a border on a date. A sanctions listing prohibits dealings with a named party. The ICTS mechanism does something different and slower: it prohibits classes of transactions involving technology with a defined connection to a designated foreign adversary, and it does so through notice-and-comment rulemaking with phased compliance dates written into the rule itself.

The worked example is the connected vehicles rule, and its timeline is the most useful thing in this entire subject. The final rule was issued on January 16, 2025, with an effective date of March 17, 2025. The prohibitions on covered software apply to model year 2027 vehicles and later. The prohibitions on covered hardware apply to model year 2030 and later.

Read those dates together. A rule finalized in January 2025 does not remove a single covered hardware component from a single vehicle sold in the United States until model year 2030 — five years later. The announcement, the effective date, and the operative date are three different things separated by years, and coverage of such actions almost invariably reports the first as though it were the third.

The reason for the gap is not indecision. It is that supply chains cannot be re-engineered on the schedule of a news cycle. A component embedded in a product design has to be replaced by a qualified alternative, which must be identified, tested, certified and produced at volume. Give an industry eighteen months and the rule breaks the industry. Give it five years and the rule reshapes it.

That structural fact — the mandatory lag between a restriction and its effect — is the single most useful thing an ordinary observer can hold about this policy area, and it survives changes of administration, party and adversary.

What follows is how the mechanism actually operates, what compliance requires of a company, and which parts of it are genuinely uncertain.

What the Regime Prohibits and How

The ICTS authority reaches transactions rather than goods, which is a more expansive and more complicated design than it first appears.

A rule under this authority identifies a class of technology, defines the connection to a designated country that triggers coverage — ownership, control, jurisdiction, or direction by a person subject to that country's jurisdiction — and prohibits specified transactions involving it. The connected vehicles rule covered the People's Republic of China, including Hong Kong, and Russia.

The prohibition operates on importers, manufacturers and sellers rather than on the foreign supplier, which is the crucial architectural choice. A foreign firm outside U.S. jurisdiction cannot easily be compelled to do anything. An American importer can, and the rule works by making that importer responsible for knowing and certifying what is inside the product it brings in.

That certification requirement is the operative burden. Importers and manufacturers must file annual declarations of conformity, certifying that covered items are not designed, developed, manufactured or supplied by persons connected to the restricted countries. Records supporting those declarations must be retained for ten years, and material changes must be notified within sixty days.

The rule also provides for general authorizations covering lower-risk transactions — initially small businesses, non-public-road vehicles, display and testing vehicles, and repair imports — which is the pressure-release valve that keeps a broad prohibition from producing absurd outcomes at the margins.

The Part Companies Find Hardest

The compliance problem is not legal. It is informational, and it is the reason these rules take years.

A modern electronic product contains components from suppliers who themselves bought from suppliers. The firm assembling the final product has a contractual relationship with its direct vendors and frequently no visibility past them. Asked to certify that nothing in a product was designed, developed, manufactured or supplied by an entity connected to a particular country, most manufacturers discover they do not have the information to answer.

Building that visibility means constructing a supply chain map that did not previously exist, obtaining attestations from vendors who may themselves be guessing, and doing it for every component in every covered product. That work is expensive, slow, and never entirely complete, because supply chains change continuously.

The second difficulty is qualification. Identifying a compliant alternative supplier is the beginning rather than the end. The alternative component must be tested in the system, validated for performance and reliability, certified where certification applies, and produced at the volume and price the product requires. In regulated or safety-critical products, that process alone can consume years.

This is why phase-in periods are measured in model years rather than months, and why a rule that appears permissive on its face can be extraordinarily demanding in practice.

What Genuinely Remains Uncertain

Several features of this regime are unsettled, and honest treatment requires distinguishing them from the parts that are now routine.

Scope creep is the first. The authority is drafted broadly, and each successive rule has addressed a different technology category. Whether it extends to any given class of equipment is a policy decision that has not been made until a proposed rule appears, and predicting which category comes next has a poor track record.

Reciprocity is the second. Restrictions of this kind invite responses, and the targeted country has its own instruments — export licensing on materials, market access conditions, regulatory approvals. The net effect on any American industry depends on both directions, and analysis that models only the outbound restriction is modeling half the problem.

Substitution capacity is the third and most concrete. A rule removing a component from American systems only works if a compliant alternative can be produced at scale. Where alternative production exists, the rule reshapes purchasing. Where it does not, the rule either gets extended, narrowed, or waived — and which of those happens is usually determined by the capacity that exists when the compliance date arrives rather than by anything decided at issuance.

The Response Is Part of the Policy

Any analysis of a restriction that models only the American side is modeling one participant in a two-sided exchange, and the record of the past several years is that the second side responds.

The instruments available to a targeted country are not symmetrical with the ones used against it, which is what makes the interaction hard to forecast. Where the United States restricts what its firms may buy, a counterpart may restrict what its firms may sell — most effectively in materials and processing steps where its share of world supply is high and substitution is slow. Export licensing on critical minerals and refined inputs has been used precisely this way, and it bites at a different point in the chain than a component prohibition does.

Regulatory approval is the second lever. Multinational firms operating in a large market depend on approvals — for products, for acquisitions, for joint ventures — and those processes can slow without any announcement being made. That form of response is deniable, difficult to measure, and effective.

Market access is the third. A country that is a major customer for an industry can shift procurement without prohibiting anything, simply by directing state-linked purchasers elsewhere.

The relevant point is not that restriction is futile. It is that the net effect on any particular industry is the combination of the outbound measure and the inbound response, that the response typically arrives on its own timetable rather than immediately, and that the sequence has repeatedly produced outcomes different from those forecast at announcement. Anyone assessing consequence needs both sides, and only one of them is ever in the initial coverage.

The Investment Question Nobody Can Answer Yet

There is a reason this policy area generates so much confident commentary and so little useful analysis, and it is worth stating directly rather than implying it.

The commercial consequences of a restriction depend on facts that are not knowable at announcement: the final scope after comment, the phase-in schedule, which general authorizations are granted, whether alternative supply materializes at the required volume, and what the targeted country does in response. Each of those can move the effect on any given industry by a wide margin, and several are decided years after the initial coverage.

What can be said in advance is structural and modest. Restrictions of this type advantage suppliers already outside the covered jurisdiction who have qualified capacity, disadvantage firms with deep exposure and thin visibility into their own supply chains, and impose a compliance cost that scales with product complexity rather than with company size — which falls hardest on the smallest covered firms.

Beyond that, the honest position is that the schedule is public, the scope is public, and anyone claiming to know the consequence before the compliance dates arrive is forecasting rather than reading.

The connected vehicles rule remains the clearest available demonstration. Issued in January 2025. Software prohibitions from model year 2027. Hardware prohibitions from model year 2030. Three dates, one rule, and only the first of them was ever a headline.