The CHIPS Act Is Working. The Problem Is It's Working Too Well For The Wrong Factories.

The numbers look reassuring on a spreadsheet. The U.S. Department of Commerce has now committed more than $32 billion in direct grants and conditional loans under the CHIPS and Science Act, with another $11 billion in tax credits deployed through the Investment Tax Credit provisions. Domestic semiconductor manufacturing capacity has begun to materialize; fabs are under construction in Arizona, Ohio, and Texas. The optics of industrial revival are compelling enough. Yet if you are running an advanced node facility in Europe, or managing supply chain strategy for precision component manufacturers, the actual allocation of those funds should trouble you considerably.

The problem is architectural. The CHIPS Act's incentive structure, particularly Section 4732 which governs the direct funding awards, weights subsidies toward production volume and job creation metrics rather than toward technological frontier advancement (what the Act loosely terms "leading edge" manufacturing, defined vaguely as sub-5-nanometer processes or equivalent). This creates a perverse incentive: a manufacturer building a mature-node fab producing 28-nanometer logic chips or specialty semiconductors receives proportionally higher subsidies per job created and per dollar of capital deployment than a facility pushing into advanced packaging or advanced process nodes. The math is straightforward and damaging. A $2 billion fab operation producing lower-complexity semiconductors may receive $500-600 million in grants and favorable tax treatment. An equivalent capital investment in a leading-edge facility, which demands far more specialized labor, engineering overhead, and longer development timelines, often lands closer to $350-400 million in total support.

This is not a minor accounting distinction. It is reshaping where semiconductor capacity gets built globally. Intel's much-publicized Fab 38 expansion in Ohio, which received approximately $3.2 billion in CHIPS Act support over multiple tranches, is anchoring advanced process capability. That is sensible policy and Intel is executing well. But the bulk of committed funding has flowed toward Intel's volume fabs, Samsung and SK Hynix packaging facilities, and legacy players consolidating their U.S. footprint for mature nodes. TSMC's Arizona facility, which does represent genuinely advanced manufacturing, secured roughly $6.6 billion in combined grants and credits, yet finds itself under continued scrutiny regarding technology transfer restrictions and tariff implications under the nascent Indo-Pacific Economic Framework negotiations. Meanwhile, smaller advanced packaging specialists and specialty foundries have received less than 8 percent of total CHIPS Act commitments despite representing some of the most innovation-dense segments of the semiconductor ecosystem.

From a European operations perspective, this matters concretely. Germany's Advanced Semiconductor Manufacturing Europe (ASME) consortium and the EU's European Chips Act counterpart, which committed 43 billion euros through 2030, are attempting to build redundancy and EU technological sovereignty in advanced nodes. They are doing this while watching American policy incentivize capacity at maturity. The result is geographic fragmentation without clear technological differentiation. You cannot compete on cost against American subsidized mature-node capacity; the intervention is too large. You must compete on capability, speed to advanced nodes, and process flexibility. But if the American subsidy architecture is written to discourage leading-edge manufacturing investment, European facilities doing exactly that work end up bearing higher per-unit costs and requiring more aggressive government support to remain viable. It becomes a subsidy race rather than a capability race.

The secondary effect, which operations teams are already experiencing, is supply chain stratification. Companies with diversified fab relationships now face a choice: concentrate advanced node work in Arizona, Taiwan, or South Korea where subsidies and technology partnerships create density, or maintain costlier European capacity for redundancy and geopolitical resilience. Most are doing both, which improves supply chain robustness but increases complexity and working capital requirements. Your logistics team has to manage three-week lead time variance between a subsidized American fab and an unsubsidized European facility producing functionally equivalent advanced components. This is not an abstraction; it is actively reshaping procurement decisions at tier-one manufacturers right now.

The corrective action is specific. The Commerce Department should recalibrate the grant allocation formulas under CHIPS Act Section 4732 to weight technology advancement metrics more heavily than pure capacity metrics. Current rules reward "new U.S. production capacity" uniformly; revised rules should define leading-edge manufacturing (sub-3-nanometer logic, advanced chip-scale packaging with sub-10-micron feature control, heterogeneous integration platforms) and apply multiplier subsidies to facilities demonstrating concrete capability in those areas. This would not reduce total funding; it would redirect existing commitments toward genuine frontier advancement. Intel, Samsung, and TSMC are sophisticated enough to optimize around any incentive structure; the real policy objective should be creating differentiated incentives that prevent subsidy races at mature nodes while accelerating the edge cases where American manufacturing genuinely needs support to compete.

For your capital allocation team, the implication is direct. If you are evaluating fab partnerships or facility expansion decisions, assume the current subsidy architecture persists through the 2028 election cycle and beyond (Congressional politics around semiconductor manufacturing are unusually durable across administrations). Plan your redundancy around the facilities actually receiving substantial CHIPS Act support, which are concentrated in volume production and legacy consolidation. If your advanced node work is Europe-based, you are implicitly accepting a cost and geopolitical hedge premium that may or may not be recoverable in your product pricing. That is a legitimate strategic choice. But it should be made with full clarity about what the American subsidy machine is actually incentivizing, which is less about leading the next generation of semiconductor manufacturing and more about ensuring that the previous generation stays firmly anchored to U.S. soil.