The MATCH Act Is Industrial Policy Dressed as National Security
This month, Congress advanced the Multilateral Alignment of Technology Controls on Hardware (MATCH) Act, a bill that would ban the sale of key semiconductor manufacturing equipment to China and threaten companies in allied nations with expanded U.S. jurisdiction if their governments fail to align with American restrictions within 150 days. Before this legislation goes any further, Congress needs to answer a basic question: Is the existing export control regime producing its intended outcomes? The overwhelming evidence says no, and the MATCH Act’s core mechanism does nothing to succeed where previous efforts have failed.
Export Controls Fund What They Are Designed to Prevent
The most rigorous assessment of the effects of current export controls to American companies shows unambiguous results. Affected U.S. suppliers experienced a $130 billion aggregate decline in market capitalization following export control announcements, along with an 8.6 percent drop in revenue, declining profitability, and a 6.6 percent reduction in employment. The Philadelphia Semiconductor Index—a market cap-weighted index of the 30 largest U.S. semiconductor companies—fell 8 percent following the October 2022 announcement alone. In addition to the initial shock, in the three years following those controls, these U.S. firms have been unable to find customers in domestic and allied markets to replace the customers they lost. The reshoring and friendshoring the policy was designed to encourage failed to materialize.
The results on the Chinese side also paint a very different picture than expected. The same research found that Chinese firms were able to offset the reduction in U.S. suppliers by forming new partnerships with domestic Chinese alternatives. While export controls disrupted the existing relationship between manufacturers in the United States and China, they failed to constrain the progress of Chinese technological development. Instead, these measures have accelerated China’s progress toward strengthening its indigenous technology stack and, perhaps more importantly, they have simultaneously damaged U.S. competitiveness by reducing revenues critical to funding the high levels of research and development (R&D) that characterize the industry. While export controls limit China’s short-term access to frontier technologies, they have clearly accelerated Beijing’s longstanding drive for semiconductor self-reliance.
The equipment market confirms this at scale. China’s use of domestic semiconductor manufacturing equipment rose from 25 percent in 2024 to 35 percent in 2025, surpassing Beijing’s own targets ahead of schedule. Substitution rates for etching and thin-film deposition equipment already exceed 40 percent. In 2022, before the strictest U.S. controls took effect, only one Chinese firm ranked among the global top 20 semiconductor equipment manufacturers. By 2025, three more firms had joined the list—with NAURA Technology Group rising to fifth overall. Quality followed quantity: the Advanced Micro-fabrication Equipment Inc.’s dry etching equipment is now qualified on TSMC’s 7 and 5 nanometer production lines, tested by Intel, Micron, and Samsung.
American export controls did not slow this surge—they created it. By forcing Chinese fabrication facilities to reduce dependence on foreign suppliers, Washington handed Beijing’s domestic equipment industry a captive market, guaranteed policy support, and political urgency. China subsequently required chipmakers seeking state approval to source at least 50 percent of new capacity equipment domestically, with authorities signaling a long-term objective of exclusively domestic equipment. The MATCH Act is a significant escalation of the policy producing these results, and we should not presume it would result in anything but more of the same.
The Allied Coercion Problem
America does not have a monopoly over semiconductor manufacturing equipment. Japan and the Netherlands are China’s primary foreign sources of chipmaking equipment by volume. American firms accounted for roughly $2 billion in direct exports to China in 2025—the lowest level since 2017. A control regime that excludes allied producers does not control the market overall, just the American share of it.
The MATCH Act’s answer is a 150-day deadline backed by threatened expansion of the Foreign Direct Product Rule (FDPR) to cover allied companies’ products. The Biden administration tried the same approach in 2024, signaling FDPR threats against ASML and Tokyo Electron before exempting both after bilateral negotiations concluded. A more assertive administration may not repeat that exemption. That possibility is not an argument for the MATCH Act—in fact, it is the strongest argument against it.
ASML’s manufacturing process incorporates American components, software, and technology at virtually every stage. Invoking the FDPR against them is not a targeted diplomatic pressure campaign; rather, it is a potential disruption of the world’s only supplier of extreme ultraviolet lithography machines—equipment that TSMC, Samsung, Intel, and every other advanced chipmaker depends on. The collateral damage would fall on American firms and American supply chains as severely as on the Dutch. ASML’s China revenue represented 29.1 percent of its total sales in 2025, while Tokyo Electron’s exceeded 40 percent. Neither government can absorb those losses on a congressional timeline, and threatening losses of that scale to governments whose cooperation the United States needs across a range of issues far beyond semiconductor equipment is not costless diplomacy.
Seeking Protection Through Legislation
Congress should be clear about whose interests this bill primarily serves. The MATCH Act targets facilities operated by Yangtze Memory Technologies (YMTC) and ChangXin Memory Technologies (CXMT), two Chinese memory chipmakers growing rapidly despite existing controls. Micron Technology, the sole major U.S. memory chipmaker and direct competitor to YMTC and CXMT, has been a primary driving force behind its passage, with its CEO holding closed-door roundtables with both the House Foreign Affairs Committee and the Senate Banking Committee during the drafting process.
None of this delegitimizes the bill’s national security rationale, as commercial interest and national security can point in the same direction. But Congress should be honest about when legislation is doing both jobs simultaneously, because the overlap creates a specific risk: that the mechanism chosen reflects benefits to the domestic industry rather than what actually constrains the adversary. The history documented above suggests those are not the same thing. YMTC has been on a restricted trade list since 2022. CXMT operated for years in the shadow of potential U.S. restrictions before finally being added to the entity list in 2025. Yet both companies are still growing, and the mechanism proposed is an escalation of an approach that has not worked in the past.
A market-based critique transcends the national security debate. Micron is the third-largest memory chipmaker in the world behind Samsung and SK Hynix. It is not losing ground to YMTC and CXMT because of unfair trade practices or technology theft; it is happening because Chinese firms are producing competitive products at competitive prices and moving up the value chain toward markets in which Micron is a direct competitor. The instinct to address that competitive pressure through export restrictions rather than through investment, innovation, and market performance is precisely the kind of regulatory protectionism that distorts markets and insulates incumbents from the discipline of competition. It also has a poor track record. Restricting Chinese memory chipmakers has been U.S. policy for years; however, YMTC and CXMT are still growing. The MATCH Act proposes to do more of the same, at greater diplomatic cost, on behalf of a domestic industry that has not demonstrated it can close the gap through its own commercial efforts.
The Case Against Escalation
The case for the MATCH Act rests on the theory that sufficiently comprehensive restrictions applied with sufficient allied coordination can meaningfully constrain China’s ability to produce advanced semiconductors, with revenue losses to U.S. firms as an acceptable side effect. However, because of the stakes involved in the technology competition with China, we must scrutinize that theory. This argument has a prerequisite: the controls must actually be constraining China. The evidence says they are not.
The most common metric offered as proof is America’s commanding share of global artificial intelligence (AI) compute. That statistic has two problems. First, it is a snapshot of chips already sold and installed rather than a forecast of where the competition is heading. Second, it has limited explanatory power, even as a snapshot. As of March 2026, Chinese AI models closed a performance gap within 2.7 percent of the leading American model while operating at a fraction of U.S. compute capacity. If compute denial were a limiting factor on Chinese AI progress, that convergence should not have been possible. China’s domestic chip market share reached 41 percent in 2025, and its AI compute capacity is projected to grow 46 percent annually. America’s current lead reflects the effects of past investment—not the trajectory going forward.
The revenue losses that export controls impose on American firms compound this problem rather than offsetting it. U.S. semiconductor companies reinvest 17.7 percent of revenue into R&D. This means that every dollar of revenue lost removes just under $0.18 from the innovation pipeline required to sustain American leadership. Firms can only stay ahead by reinvesting current profits into the chips that will define the following competitive cycle. A policy that accepts permanent revenue losses as the price of denying China today simultaneously accepts a slower rate of American advancement toward tomorrow. China has shown it does not need to match the American frontier; it simply needs the pace of frontier development to slow. China’s rate of progress toward self-sufficiency is best predicted by the market and technological complexities of each semiconductor segment, not by the extent to which export controls were applied. The compounding dynamic does not automatically work in America’s favor, and the burden of proof belongs to those proposing to accelerate those losses through escalation.
Conclusion
Existing controls have imposed significant costs on American firms and have not been replaced with reshoring or friendshoring. When the U.S. restricted Micron’s access to Chinese markets, Beijing retaliated with a dubious cybersecurity review that contributed to a 49 percent year-on-year revenue drop for Micron in fiscal year 2023. China has and continues to gain market share across every major equipment category in which U.S. controls have been tightest. The coercion mechanism required to get semiconductor manufacturers in allied nations to follow our lead at significant cost to themselves has already been attempted once and walked back when confronted with economic reality.
At the same time, China’s domestic equipment industry is progressing unimpeded. NAURA’s order backlog runs through early 2027. Beijing has institutionalized domestic sourcing requirements that will insulate Chinese chipmakers further from the effects of Washington’s next set of export controls. There are clear indications that the window in which export controls could have meaningfully altered China’s technological trajectory has closed.
The MATCH Act and broader export controls on semiconductors would impose real costs on American firms and real diplomatic friction with allied governments in exchange for theoretical strategic gains that are increasingly difficult to demonstrate.
The question Congress must ask is not whether to contest China’s semiconductor ambitions; it is whether this particular mechanism with this particular record is the right instrument for reaching that objective. Available evidence does not support the conclusion that escalating a failing policy will produce a different result. Before voting to advance the MATCH Act any further, Congress should be required to explain why this time would be different.