> **来源：[研报客](https://pc.yanbaoke.cn)** # Goldman Sachs # Research # TOPof MIND # THE US-CHINA TECH RACE While the US and China have reached a new trade deal, their fierce race to achieve tech superiority remains as intense as ever, with both pursuing policies to develop self-sufficient tech stacks. So, will such policies prove successful, and what investment opportunities do they provide? The Wahba Initiative for Strategic Competition's Mark Kennedy assesses the tech race's current state (the US leads in a key arena but China is catching up or leading in many others) and GS' Alec Phillips and Hui Shan dig into the why, what, and how of US and Chinese policy efforts to win the race. DGA-Albright Stonebridge Group's Paul Triolo and the Critical Minerals Institute's Jack Lifton then explain the intricacies and chokepoints of the two supply chains in focus here—semiconductors and rare earths—and GS' Daan Struyven digs into power's critical role in determining the winner. Lastly, we explore related investment opportunities, with GS equity analysts and strategists seeing compelling opportunities in rare earths and Asia tech. It is entirely possible that neither the US nor China emerge as the outright victor in the tech race. I can envision a world in which the US leads in developing the most advanced technologies, while China leads in global installations. - Mark Kennedy The desire to reduce our dependence on China [for rare earths] is valid. And there is a viable path to achieving that, even if it won't amount to self-sufficiency. It's just a matter of picking the right companies, technologies, and people. The US expanding or even maintaining its current lead [in advanced AI] is not a foregone conclusion... much will depend on the Chinese semiconductor industry's ability to overcome the hardware chokepoints. - Jack Lifton Allison Nathan | allison.nathan@gs.com Paul Triolo Jenny Grimberg | jenny.grimberg@gs.com # WHAT'S INSIDE INTERVIEWS WITH: Mark Kennedy, Founding Director, Wahba Initiative for Strategic Competition at New York University's Development Research Institute Jack Lifton, Co-chair, Critical Minerals Institute Paul Triolo, Partner, DGA-Albright Stonebridge Group A US POLICY ROADMAP FOR THE TECH RACE Alec Phillips, GS US Economics Research THE WHY, WHAT, & HOW OF CHINA TECH POLICY Hui Shan, GS China Economics Research TECH RACE COMPANY IMPACTS: RARE EARTH'S AND SEMIS GS Equity Research CHINA'S PHYSICAL EDGE IN THE TECH RACE Daan Struyven, GS Commodities Research CHINA'S TECH DRIVE, ASIA'S OPPORTUNITY Tim Moe and Kinger Lau, GS Asia Portfolio Strategy Research ...AND MORE y.grimberg@gs.com Ashley Rhodes | ashley.rhodes@gs.com # Macro news and views # We provide a brief snapshot on the most important economies for the global markets # US # Latest GS proprietary datapoints/major changes in views - No major changes in views. # Datapoints/trends we're focused on US growth, which we expect to modestly reaccelerate to $2.3\%$ in 2026 (on a Q4/Q4 basis) owing to a reduced tariff drag, tax cuts, and easier financial conditions. Fed policy; we expect the Fed to deliver a 25bp rate cut next week followed by two 25bp cuts in 2026 for a terminal rate range of $3 - 3.25\%$ but see downside risks to our forecasts given ongoing labor market weakness. US productivity outperformance vs. other DMs, which structural factors & AI adoption tailwinds should sustain. # US layoffs: on the rise GS Layoff Tracker* and components, Z-score *First principal component of our tracking of WARN notices, initial claims, Challenger layoffs, and earnings calls layoff discussions. Source: Dept. of Labor, Challenger, Gray & Christmas, Federal Reserve, GS GIR. # Europe # Latest GS proprietary datapoints/major changes in views We lowered our 2026/27 Euro area GDP growth forecasts to $1.2\% / 1.3\%$ yoy (from $1.3\% / 1.5\%$ ) to reflect increased export competition with China and lower potential growth. - We raised our 2026 UK GDP growth forecast to $1.1\%$ yoy (from $1.0\%$ ) as the Autumn Budget projections suggest a smaller fiscal drag on demand than we originally assumed. # Datapoints/trends we're focused on - BoE rate cuts, the next one of which we expect this month. - ECB policy; we expect the ECB to remain on hold, though the risks are tilted toward renewed rate cuts next year. # Europe: an export drag from China's second wave Estimated country-level real GDP impact of higher Chinese export growth by end-2029, % Source: Haver Analytics, Goldman Sachs GIR. # Japan # Latest GS proprietary datapoints/major changes in views - No major changes in views. # Datapoints/trends we're focused on - BoJ policy; while Jan 2026 remains our baseline forecast for the next hike, the likelihood of a rate hike this month has risen following a recent hawkish speech by Governor Ueda. - Heightened Japan-China tensions, which we estimate could result in a 0.2pp hit to Japanese GDP growth from reduced Chinese tourism and lower Japanese exports to China. - An expected rise in Japanese potential growth over the next several years, driven by labor productivity growth. # Japan: rising China tensions, falling Chinese tourism Foreign visitors to Japan (Ihs, mn per month) and their inbound spending (rhs, ¥bn per month) Source: JNTO, BoJ, Goldman Sachs GIR. # Emerging Markets (EM) # Latest GS proprietary datapoints/major changes in views We recently raised our CY2025/2026 India real GDP growth forecasts to $7.6\% / 6.6\%$ yoy (from $7.3\% / 6.5\%$ ) following much stronger-than-expected 3Q2025 GDP data. # Datapoints/trends we're focused on China shock 2.0; China's renewed focus on export-led growth will likely benefit its economy and raise its current account surplus to $1\%$ of global GDP by 2029, reducing its trading partners' manufacturing output and employment. - EM growth; we expect growth in China, Brazil, and India to slow in 2026 but pick up in several mid-sized EMs, leaving overall EM growth broadly unchanged. # China shock 2.0 Current account balances as share of world GDP, % Source: IMF, Haver Analytics, Goldman Sachs GIR. # The US-China tech race While the US and China reached a new trade deal following bilateral talks at the October APEC summit, their fierce race to achieve technological superiority—driven by a complex interplay of security, economic, and geopolitical interests—remains as intense as ever, with both countries pursuing policies to develop self-sufficient tech stacks. Whether such policies will prove successful, and the investment opportunities they provide, is Top of Mind. We first ask Mark Kennedy, Founding Director of the Wahba Initiative for Strategic Competition at NYU's Development Research Institute, to assess the current state of the tech race. He explains that while the US leads in many key areas of technological innovation (including semiconductors, AI frameworks, cloud infrastructure, and quantum computing), China is quickly closing the gap or even dominating in three other critical arenas of the race—practical applications like physical AI and robotics, installations of the digital plumbing that underpins technology owing to China's strong presence in the Global South, and technological self-sufficiency as China has worked to reduce its dependence on Western technologies. GS Chief US Political Economist Alec Phillips and Chief China Economist Hui Shan then dig deeper into the why, what, and how of US and Chinese policy efforts to win the tech race. Phillips explains that US policymakers are combining the playbooks deployed during World War II/the Cold War and the US-Japan economic rivalry of the late 20th century by taking an increased role in some areas (like semiconductors and AI) while relying on financial incentives to boost domestic production and purchases in sensitive sectors (like rare earths). And Shan details Chinese policymakers' "holistic and highly coordinated" approach to supporting technology, which involves systematic planning, significant financial backing, and assisting in other areas from land acquisition to talent development. We then explore the two supply chains most in focus amid this race: the global semiconductor supply chain—for which the US has a slight edge—and the rare earth supply chain—which China currently controls (see pgs. 10 and 11). The US accounts for half of the global semiconductor market... Global semiconductor market share by region of company headquarters, % Source: World Semiconductor Trade Statistics (WSTS), Gartner, Omdia, Semiconductor Industry Association, compiled by Goldman Sachs GIR. We first turn to Paul Triolo, Partner at DGA-Albright Stonebridge Group, for an explanation of the intricacies and key chokepoints of the semiconductor supply chain, which he says makes it difficult for any country to achieve self-sufficiency. But he argues that despite the US' current lead in chip innovation, "if any country can do it (i.e., achieve self-sufficiency), it's China, owing to its vast resources and pool of engineering talent", though he sees significant technical and financial challenges. But he also believes that recent US policy efforts to strengthen its domestic semiconductor capabilities don't go far enough, leaving open the possibility that China could close the gap with the US in developing the world's most powerful AI models, while also maintaining or even growing its lead in AI adoption given China's embrace of open-source models. ...while China dominates the rare earths space China vs. US market shares in each category, % Source: IEA, Woodmac, Goldman Sachs GIR. We then speak to Jack Lifton, Co-chair of the Critical Minerals Institute, who breaks down China's grip on the supply chain for rare earths, and especially the heavy rare earths used in defense and other critical industries. While he believes China's chokehold can be broken, he argues that current US efforts to do so will not succeed because US policymakers are not investing in "the right companies, technologies, and people." But he also thinks the need for such efforts is overblown given that US rare earth demand is small and acceptable substitutes exist, at least in the military complex. GS Co-head of Global Commodities Research Daan Struyven nonetheless makes the case that China's dominance in rare earths, together with its abundant power supplies—in sharp contrast to the US' power constraints—give China an edge in the AI and broader tech race. Triolo and Kennedy agree that China's power advantage presents a challenge for the US, which they say sets the stage for the energy-abundant Middle East to play a pivotal role in the US-China tech race. So, how should investors be positioned as US and Chinese policymakers deploy their playbooks for "winning" this race? GS equity analysts find that the stocks of leading semiconductor manufacturing and equipment companies have already largely priced in US-China tech developments. But they think continued efforts by both countries to develop their own tech stacks could remain a key driver of price action for Western rare earth mining and refining companies. And GS senior Asia Pacific strategists Tim Moe and Kinger Lau see significant opportunities for Asia tech companies—which have outperformed this year—amid these ongoing policy efforts. # Allison Nathan, Editor Email: allison.nathan@qs.com Tel: 212-357-7504 Goldman Sachs & Co. LLC # Interview with Mark Kennedy Mark Kennedy is Founding Director of the Wahba Initiative for Strategic Competition at New York University's Development Research Institute. Below, he lays out the current state of the US-China tech race, arguing that the US leads in a key arena, but China is quickly catching up or even leading in many others. The views stated herein are those of the interviewee and do not necessarily reflect those of Goldman Sachs or of NYU. # Ashley Rhodes: Technology seems to be at the center of the US-China strategic rivalry. Is that an accurate characterization, or has the role of technology been overplayed? Mark Kennedy: The role of technology is not being overplayed. Technology isn't just at the center of the US-China rivalry—it's the central switchboard. Whoever controls how technology, data, and computing power are routed will impact every domain—from military might, to economic influence, to the flow of information. So, the significance of technology in the US-China rivalry is anything but overstated. The profound impact that this technological contest could have on global systems and the resulting geopolitical implications are, if anything, underappreciated. # Ashley Rhodes: Who is currently "winning" the tech race? Mark Kennedy: It's important to understand that there are four key arenas in this race: technological innovation, practical application of the technology, installation of the digital plumbing or infrastructure underpinning the technology, and technological self-sufficiency. The US is currently leading in most advanced technologies, including semiconductors, AI frameworks, cloud infrastructure, and quantum computing, as well as in attracting global talent. However, China is ahead in areas such as quantum communications, hypersonics, and batteries. China is also making rapid strides to catch up to and, in some cases, overtake the US in technological application. For example, China deploys robotics in manufacturing on a scale twelve times greater than the US when adjusted for differences in employee income. And while US regulations often limit applications like drone deliveries to your door, China is proactively testing and deploying advanced physical AI and robotics like uncrewed taxis and vertical takeoff vehicles, accelerating their practical adoption. China is also dominating on the global installations front. It has established a strong presence in the Global South, surpassing the US and other Western nations in building essential digital networks there. And China has made significant strides toward achieving technological self-sufficiency through its dual circulation strategy aimed at reducing its reliance on the West while increasing Western dependence on China. Recent Chinese government measures, such as restricting domestic purchases of Western chips and offering incentives for using domestic alternatives, underscore this push for technological independence. At the same time, China's vast overproduction capacity in batteries and critical minerals has further increased Western dependence on China's supply chains. The US has been ambivalent at best as it relates to this aspect of the tech race and remains reliant on China in many ways. So, on net, while the US leads in the development of the technology itself, China is rapidly closing the gap—or even leading—in application, infrastructure installations, and tech self-sufficiency. # Ashley Rhodes: Have US export controls on advanced chips materially impeded China's technological progress? Mark Kennedy: The impact has not been especially significant. China had already planned to reduce its reliance on American semiconductor technology, and the introduction of US export controls has merely accelerated that transition. China is finding ways to adapt by using less advanced chips and still maintains some access to certain US chip technologies. So, while these measures are slowing China's technological progress, they are not significantly impeding it and are unlikely to be the silver bullet that allows the US to maintain its tech lead indefinitely. # Ashley Rhodes: Conversely, will China's grip on rare earths impede the US' ability to maintain tech leadership? Mark Kennedy: Rare earth elements are a critical input in the tech race, making China's export restrictions a real threat to America's long-term prospects for developing an independent tech ecosystem. While China has granted the US a one-year reprieve on these restrictions, the recent deal between the two countries is fragile and the US will remain vulnerable unless it can rapidly expand its access to rare earths. The US has been aware of this vulnerability for a long time but has admired the problem more than it has taken deliberate steps to address it. America is just now waking up to the seriousness of this threat and beginning to take steps in the right direction, including US government efforts to invest in rare earth companies and take ownership stakes in key firms. But the US is still a long way from achieving self-sufficiency, and true resilience will depend on working closely with allies like Australia and Canada to establish secure, diversified supply chains. Regrettably, US relationships with its key allies have become increasingly fractured, which isn't helpful. # Ashley Rhodes: What else will it take for the US to win? Mark Kennedy: Beyond diversifying supply chains for critical minerals and other essential technological inputs, the US must more broadly move beyond its historically laissez-faire approach toward technology and lean into winning the tech competition across all arenas. This means significantly increasing investment in research and development, reversing the recent trend of stagnant or declining university funding, and easing restrictions that hinder the entry of global talent. The US must also expand its energy infrastructure to support the growing power demands of data and computing. While China has gone far in securing energy resources—particularly in renewables such as wind, solar, and nuclear—the US approach has been inconsistent. The current shift toward nuclear energy is promising, but achieving the necessary scale will take time. And the US must strive to match China's push to dominate installations in the Global South or risk losing influence in these regions. To date, restrictive policies—such as the Biden Administration's Al Diffusion Rule and other restrictions—have limited the reach of US tech. Actions like cutting off Ukraine's satellite access have raised concerns that the US may wield technological access as a form of geopolitical leverage. And where the US withholds its tech, China readily steps in. While recent AI policy discussions have recognized the importance of deploying the domestic tech stack globally, the real challenge is getting multiple government agencies and private companies to act with the same unity and speed as China does. # Ashley Rhodes: The US does seem to be increasing public-private partnerships to make progress toward these goals. How important are such partnerships? Mark Kennedy: Government involvement in the critical minerals sector will play a crucial role given that China's dominance allows it to lower prices and squeeze out competitors. So, US companies need to have some guarantee that prices will be at a level that will at least sustain their production. These guarantees will be essential to generating the sources of supply at the mineral level that will allow the US to apply its magic at the digital level. And government involvement will be critical to expanding the reach of US technology to other parts of the world, like the Global South. However, the benefits of other types of government involvement are less clear. For example, it's unclear whether the government taking an ownership stake in Intel will prove helpful beyond the near-term boost in capital and confidence it provided. Such a move could actually be harmful if measures like "golden shares" prevent the company from making the tough decisions that will be required for it to remain globally competitive. Intel's role as the sole domestic foundry of scale is vital, especially as TSMC's investment in the US remains a shadow of the company's efforts to maintain its silicon shield in Taiwan. The trick is providing support without politicized interference. Past US ownership in established commercial enterprises has seldom delivered long-term benefits. # Ashley Rhodes: Conversely, what additional steps must China take to win the tech race? Mark Kennedy: China is already doing many things right. Its most effective strategy has been sustained, targeted investment in key technologies that will determine the future. And China has long understood the importance of securing and leveraging essential resources like rare earths as well as areas like ports. As we've discussed, China has also been ahead in establishing new markets in key regions like the Global South. But China recognizes that it needs to accelerate the advancement of its domestic semiconductor industry. To that end, the Chinese government has begun channelling investments into local chip manufacturing, prioritizing long-term development over short-term performance gains. Another critical priority should be building trust. The Chinese model is characterized by centralized control; their systems are not easily interoperable or replaceable, leading to a degree of lock-in. And China's cybersecurity laws require companies to provide any requested data to government authorities. All this undermines international trust in Chinese tech. The tech competition is in many ways a battle for trust. The US is currently leading in this area given its history of transparency and reliability, while China has more work to do in ensuring that its technology is not only affordable and reliable but also trustworthy. # Ashley Rhodes: How important are other countries in determining who wins the tech race? Mark Kennedy: The technological alignment of other countries will be pivotal in determining the outcome of the tech race. Again, China has been aggressively targeting the Global South, offering comprehensive digitalization packages—often as service contracts—which enable it to build, maintain, and exert substantial control over networks and the data they carry. This access to vast amounts of data, in turn, enables China to train AI models that provide insights into how to tailor products or services specifically for those markets, reinforcing its influence and making it increasingly difficult for competitors to match the value China offers. By contrast, the US has not matched China's government-backed efforts in these markets. Once-prominent programs like USAID, whose grants supporting host country capabilities paired with financing support helped the US prevail in many international infrastructure competitions, have largely been diminished. This has left the US at risk of ceding technological influence while China positions itself to digitally connect up to $70 - 80\%$ of the world's landmass, aligning it away from US interests. The US really needs to consider whether it should be doing a better job of matching China's thumb on the scale in determining through whose networks the world's data flow. The Gulf countries will also play a pivotal role in the tech race given their abundant energy resources. Whoever succeeds in establishing influence within these countries will gain privileged access to vast energy resources they can use to power their technological ambitions and capabilities. That said, with both China and the US likely to be increasingly engaged in the Gulf for this reason, establishing security protocols to ensure that tech provided to the Gulf does not drift to China will be essential. Geopolitically, the US must embrace the Gulf, or China will fill the void. # Ashley Rhodes: Ultimately, how do you envision the US-China tech race playing out? Mark Kennedy: It is entirely possible that neither the US nor China emerge as the outright victor in the tech race. I can envision a world in which the US leads in developing the most advanced technologies, while China leads in global installations, particularly outside of America and its closest allies. China could also prevail in applications as its vast quantity of electric engineering graduates and PhDs could take on a quality of its own. And China could win the "we're independent of you, but you're still dependent on us" race. So, a world could exist in which the US owns the blueprints—leading in software and code—but China owns the buildings—commanding the hardware and circuitry—which would have the potential to significantly reshape economies and geopolitics as we know it. # A US policy roadmap for the tech race # Alec Phillips explores the US policy roadmap for achieving technological self-sufficiency US policymakers have drawn parallels between the tech rivalry with China and the urgent US industrial and scientific efforts during World War II as well as the US-Soviet tech competition during the Cold War. During both periods, intense scientific and technological rivalries led the US government to organize and fund research and development in several key areas of competition (e.g., nuclear fission and aerospace). However, many of the technologies central to the US-China tech rivalry today are primarily commercial. While most have benefitted from publicly funded research and have military or strategic uses, the markets for these technologies are much broader and, in many cases, the technology has already been commercialized. Indeed, private sector funding for AI or semiconductors dwarfs government programs. This presents a challenge that goes beyond scaling up public (i.e., military) spending, with arguably greater parallels to the US-Japan economic rivalry in several sectors during the 1970s and 1980s. US policy appears to be following approaches along both lines, with a larger government role in some areas while primarily relying on financial incentives for production or purchase of domestically-produced products in sensitive sectors. # Focusing investment in strategic sectors While no formal definition of the sectors of greatest strategic importance exists, the Trump Administration appears to prioritize AI, quantum computing, nuclear technology, and biotechnology, along with related inputs like rare earths/critical minerals and semiconductors. Robotics and drones, shipbuilding, aerospace, and power technology also appear to be priorities, though not as central a focus. Technologies targeted by several government initiatives <table><tr><td></td><td>Genesis Mission (Nov. 2025)</td><td>White House memo on R&D priorities (Sep. 2025)</td><td>Trump memo to OSTR Dir. (Mar. 2025)</td><td>DoD "Critical Technology Areas" (Nov. 2025)</td><td>Sec. 232 Invest-igation</td><td>China-focused Sec. 301 Invest-igation</td><td>Outbound invest-ment rules (Nov. 2024)</td><td>Subsidy program (CHIPS, IRA, OBBBA)</td><td>Loan or equity stake (2025)</td><td>Trade partner investment goals</td></tr><tr><td>AI</td><td>●</td><td>●</td><td>●</td><td>●</td><td></td><td></td><td>●</td><td></td><td></td><td>●</td></tr><tr><td>Quantum</td><td>●</td><td>●</td><td>●</td><td>●</td><td></td><td></td><td>●</td><td></td><td></td><td></td></tr><tr><td>Nuclear</td><td>●</td><td></td><td>●</td><td></td><td></td><td></td><td></td><td>●</td><td>●</td><td>●</td></tr><tr><td>Semis</td><td>●</td><td></td><td></td><td></td><td>●</td><td>●</td><td>●</td><td>●</td><td>●</td><td>●</td></tr><tr><td>Biotech</td><td>●</td><td>●</td><td></td><td>●</td><td></td><td></td><td></td><td></td><td></td><td>●</td></tr><tr><td>Critical Minerals</td><td>●</td><td></td><td></td><td></td><td>●</td><td></td><td></td><td>●</td><td>●</td><td>●</td></tr><tr><td>Space</td><td>●</td><td>●</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr><tr><td>Shipbuilding</td><td></td><td></td><td></td><td></td><td></td><td>●</td><td></td><td>●</td><td></td><td>●</td></tr><tr><td>Aerospace</td><td></td><td></td><td></td><td></td><td>●</td><td></td><td></td><td>●</td><td></td><td>●</td></tr><tr><td>Steel</td><td></td><td></td><td></td><td></td><td>●</td><td></td><td></td><td></td><td>●</td><td></td></tr><tr><td>Robotics</td><td></td><td></td><td></td><td></td><td>●</td><td></td><td></td><td></td><td></td><td></td></tr><tr><td>Energy trans.</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr><tr><td>/storage</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td>●</td><td></td><td>●</td></tr></table> Source: White House, Department of Defense, USTR, Goldman Sachs GIR. The Trump Administration has several tools to invest in strategic sectors: - CHIPS Act grants: While around $43bn of the$ 50bn in direct funding in the CHIPS Act (2022) has been allocated, the Commerce Department continues to deploy funding from the Act, including for financing equity stakes in key firms. - Advanced manufacturing incentives: The One Big Beautiful Bill Act (OBBBA) renewed and expanded the Advanced Manufacturing Investment Credit (48D) from $25 \%$ to $35 \%$ , which subsidizes domestic investment in semiconductor manufacturing capacity and related tech. While 48D was originally enacted in the CHIPS Act, funding is open-ended and separate from the $50bn grant authority noted earlier. The OBBBA also preserved the Inflation Reduction Act's (IRA) Advanced Manufacturing Production Credit (45X), which subsidizes domestic production of batteries as well as wind and solar components, but the new legislation reduces the credit for costs attributable to products from Chinese companies. The OBBBA also included broader manufacturing incentives, including immediate expensing of manufacturing structures and equipment. - Government financial stakes: The Trump Administration has announced stakes in firms in the critical minerals, nuclear energy, semiconductor, and steel sectors totaling $10bn (Intel accounts for nearly$ 9bn of this). While some of these financial commitments involve equity, others consist primarily of loans. OBBBA expanded the scale and reach of these programs via a $1.5bn appropriation allowing up to $200bn in lending through the Dept. of Defense Office of Strategic Capital, which will likely lead to lower-cost, longer-term loans than these firms would otherwise receive. OBBBA also gives $2bn to the Defense Innovation Unit, which focuses on scaling commercial tech for military use, and $10bn to the Industrial Base Fund, for investments in critical minerals supply chains and to bolster the US critical minerals stockpile. - Trade partner investment pledges: The Trump Administration announced agreements with the European Union ( $600bn), Japan ($ 550bn), South Korea ($350bn), and Switzerland ($200bn) to invest in the US. The structure of these pledges is not entirely clear but the most defined of them, the Japanese investment pledge, suggests that these pledges might function primarily as low-cost loans with a small equity component, targeting a combination of strategic sectors (e.g., semiconductors, energy storage, and data centers) and traditional investments (e.g., natural gas and electric utility infrastructure, fertilizer production, etc.). - Lighter domestic regulation: The Trump Administration has pledged to ease regulation in general, targeting permitting new projects and energy generation, with Trump signing an executive order in July to accelerate permitting of data center infrastructure. Congress is debating permitting reform, with proponents making another push for enactment in the next few months. If enacted, this could facilitate a buildup of the power grid, where undercapacity threatens to become an obstacle to AI infrastructure investment. # Creating a demand signal for certain industries While demand is clear in some segments like AI and semiconductors, it is less reliable in others. In those areas, US policy provides assurances/guarantees for certain technologies produced in the US or pursuant to public-private agreements: # - Offtake agreements, floor prices, and other market interventions: Recent agreements that the Trump Administration has reached with companies in targeted sectors include offtake agreements and floor prices to limit vulnerability to market and non-market forces (see pg. 17). This is particularly relevant in areas like rare earths, where the periodic lifting of restrictions could lead to substantial price volatility and make domestic production uneconomic, facilitating continued Chinese dominance in the sector. # - Government procurement and demand subsidies: Defense spending has been an important contributor to US innovation and will likely play a role in some sectors. OBBBA provides $\sim$ 30bn for domestic shipbuilding, for example. And the IRA continues to subsidize renewable energy investment and production and electric vehicle purchases, though the OBBBA both tightened and phased these down to place more emphasis on domestically-produced technology. # Reducing demand for Chinese technology In addition to boosting investment in and demand for US tech, the US has in place or is contemplating several policies that aim to reduce demand for competing Chinese products: - Tariffs on Chinese products: The Trump Administration reduced the $30\%$ tariff imposed on imports from China earlier this year to $20\%$ as part of the October deal, which also paused US export controls in return for China's easing of rare earth export restrictions. The remaining tariff consists of $10\%$ on all imports plus an additional $10\%$ "reciprocal" tariff that exempts several categories, discussed below. These tariffs apply on top of the $\sim 10\%$ weighted average tariff applied on Chinese imports during Trump's first term. We expect tariffs on China to remain at the current level for the foreseeable future, with risks toward lower tariffs if the Administration does not fully replace them should the Supreme Court block them in its upcoming ruling. - The threat of tariffs on key sectors: The Trump Administration launched national security investigations under Sec. 232 into imports of semiconductors, pharmaceuticals, critical minerals, drones, polysilicon, wind turbines, robotics and industrial machinery, as well as medical consumables and devices. While we have assumed that the Administration will follow through with threatened sectoral tariffs, we no longer assume the imposition of any pharmaceutical tariffs in light of the political challenge this could create and the Administration's signaling that most major firms in the space would be exempt. Similarly, tariffs on semiconductors/electronics could be controversial ahead of the midterm elections and would raise the cost of consumer electronics. That said, the Administration will likely continue to use tariffs to protect sensitive sectors in the US, particularly from competition from lower-cost Chinese firms. - Regulation of Chinese components in US systems: During the final days of the Biden Administration, the Office of Information and Communications Technology and Services (OICTS) and the Bureau of Industry and Security (BIS) at the Commerce Department issued rules limiting the use of tech from China or other "foreign adversaries" in connected and self-driving vehicles. Some lawmakers have urged the Trump Administration to take similar actions in other areas, including AI and cloud infrastructure, robotics, energy generation and storage, network hardware, and chip manufacturing equipment, among other sectors. Separately, the Commerce Department has issued guidance that the use of advanced Chinese chips would likely violate US export control policies. - BIOSECURE Act: In October, the Senate passed an amendment to the National Defense Authorization Act (NDAA) that would bar federal agencies from contracting with "companies of concern", which could limit US firms dealing with the federal government from using inputs from Chinese entities. While this provision looks unlikely to be included in the final Act, it reflects US policymakers' clear intent to limit reliance on the Chinese health supply chain. # Limiting diffusion of certain US technology US policy has also focused on limiting the supply of technology to China, particularly in the areas of AI and semiconductors: - Export controls: The US continues to impose export controls on sensitive technologies, particularly advanced semiconductors and chip manufacturing technology. The US "entity list" has grown this year, and the Commerce Dept. expanded the definition of an affiliated entity under a September 2025 rule that would have had a big impact on the ability of Chinese companies to procure US technology. However, implementation of this rule was delayed as part of the October deal and it appears unlikely that the US will impose further export controls on China for the next year. - Outbound investment rules: Restrictions on US investment in Chinese entities that operate in AI, chips, and quantum technologies, which the Biden Administration finalized in late 2024, are now in effect. Congress is also finalizing its NDAA legislation for 2025 and is considering codifying similar but potentially broader rules, though outbound restrictions look unlikely to be included in the final legislation. While investment in Chinese biotech, aerospace, and other key sectors are not covered under current rules, the White House has signaled an intent to consider these transactions. - GAIN Act and the Diffusion Rule: The Biden Administration finalized rules at the start of 2025 that would have required US chip companies to retain a sufficient portion of Al-relevant chips for domestic users and strictly limited exports to several countries, including China. The Trump Administration rescinded that rule in May, but the Senate passed an amendment to the NDAA in October that would resurrect some aspects of it, giving US customers priority before allowing export. The final legislation looks unlikely to include these restrictions, but the outcome is uncertain. # A tailored policy roadmap US policy approaches to the tech race will vary by sector. The AI sector, for example, has little need for financial incentives, but could benefit from regulatory easing and public-private coordination, particularly related to the mobilization of federal research programs and diffusion across the economy (note the Administration's recent "Genesis Mission" to coordinate research and make federal scientific datasets broadly available). By contrast, small amounts of funding can have a much bigger impact in other sectors, where financial incentives are likely to play a larger role. The total annual market for rare earths, for example, is smaller than several of the individual manufacturing incentives in the IRA, CHIPS Act, or OBBBA. Quantum computing investment is similarly small scale, and federal financial resources seem likely to play a larger role there. In the near term, US efforts in the tech race with China will likely focus more on domestic incentives rather than on restricting growth of China tech. Following the recent trade deal, it seems unlikely that the White House will take steps that could destabilize the relationship in coming months, but frictions in the relationship are bound to return eventually. # Alec Phillips, Chief US Political Economist Email: alec.phillips@gs.com Tel: 202-637-3746 Goldman Sachs & Co. LLC # The why, what, & how of China tech policy # Hui Shan assesses the why, what, and how of China's policy support for technology The Chinese government has long emphasized the importance of science and technology, with this focus intensifying dramatically in recent years as the US-China tech race has heated up. Indeed, China has implemented several policies to advance technological innovation, including establishing a new Central Science and Technology Commission in the "Party and State Institution Reform" in 2023 to coordinate the construction of the national innovation system, as well as restructuring the Ministry of Science and Technology to optimize the management of the science and technology innovation system. With the US-China tech rivalry showing no signs of slowing, China's drive for technological self-sufficiency will likely remain central to China's policy agenda. # The why Driving this increased emphasis on science and technology has likely been the Chinese leadership's view that the US and China are locked in long-term strategic competition, making technological self-reliance vital for China's development and national security. Notably, while the 13th Five-Year Plan (FYP) in 2015 contained no mention of the "international situation" (国际形势), the 2018-19 US-China trade war prompted its inclusion in the 14th FYP along with the phrase "a great change unseen in a century" (百年未有之大变局). By 2025, in the wake of the Covid pandemic, Russia's invasion of Ukraine and subsequent Western sanctions on Russia, stricter US semiconductor export controls, and sharply higher US tariffs on China, the phrase "international situation" was cited three times in the 15th FYP proposal, with the language "high winds, rough seas, even raging storms" (风高浪急甚至惊涛骇浪) used to describe the seriousness of the geopolitical challenges that China faces. Against this backdrop, China is determined to strengthen its industrial base, maintain its manufacturing competitiveness, and enhance its domestic innovation capability. # The what The Chinese government's approach to technological progress demonstrates both continuity and evolution. In 2015, policymakers introduced the "Made in China 2025" plan, which included primary development priorities of information technology, robotics, green energy and electric vehicles (EVs), aerospace equipment, ocean engineering and high-tech ships, railway equipment, power equipment, new materials, pharmaceuticals and medical devices, and agricultural machinery. Since then, China has made remarkable progress in these sectors, securing a dominant position in the global markets for EVs, renewable energy, and shipbuilding. Looking ahead, the 15th FYP proposal outlines three key areas for future development. The first centers on "chokehold technologies," including integrated circuits, industrial machine tools, high-end instruments, basic software, advanced materials, and biomanufacturing. In these sectors, policymakers will likely spare no effort or expense to achieve breakthroughs and attain self-reliance given their importance to China's national security and resilience against external pressure. China has the top number of patent applications in the world Share in global patent applications, % Source: World Bank, compiled by Goldman Sachs GIR. The second area focuses on "emerging industries," including new energy technologies, advanced materials, aerospace, and the low-altitude economy (i.e., drones and urban air mobility). These sectors will likely receive substantial investment as the government aims to make them important drivers of China's economic growth. The third area targets "future industries," including quantum technology, biomanufacturing, hydrogen energy and nuclear fusion, brain-computer interfaces, embodied intelligence (integrated AI and robotics), and sixth-generation (6G) mobile communications. While these sectors are largely in the early stages of development, policymakers view them as potential game changers that could redefine global technological leadership in the coming decades. The government's ongoing support ensures that China remains at the forefront of research and innovation in these cutting-edge fields, even if widespread commercialization may still be years away. # The how China's government support for science and technology—and their adoption in industrial sectors—is holistic and highly coordinated. First, it involves systematic planning, including through the FYPs as well as detailed catalogs published by the National Development and Reform Commission (NDRC) identifying industries to encourage, restrict, or eliminate. For example, the latest catalog (released in 2024) encourages certain specialty chemicals, while targeting some small-scale, oil- and gas-based, and polluting chemical productions for restriction or elimination. Second, this support includes significant financial backing from various levels of government, including through government-guided funds for tech investments. As of 2024, China had established over 2,000 of these funds, with a targeted total of RMB12.8tn<sup>1</sup>. And at the March 2025 Two Sessions, the NDRC announced a RMB1tn government-guided fund for early-stage tech investments in areas like AI, quantum technology, and hydrogen storage. Beyond direct investments, grants, and subsidies, the Ministry of Finance offers tax benefits to tech companies and the PBOC runs a relending program dedicated to technology and innovation. Consequently, Chinese industrial sectors receive far more financial subsidies than their global counterparts<sup>2</sup>. China splashes out on industrial subsidies Industrial subsidies for 14 key industrial sectors, averages over 2005-22, % of annual firm revenue Source: OECD, compiled by Goldman Sachs GIR. Third, China's government support extends far beyond traditional tax, credit, and financing assistance. In strategic priority areas, the government often offers cheap land, expedited permits and approvals, and abundant energy supplies while facilitating connections between upstream and downstream industries to foster a complete supply chain. It also helps source and develop needed talent, and direct government procurement toward domestic innovations. For example, local governments often sell land to raise revenues, with industrial land priced at only a tenth of residential land. This allows the government to collect implicit taxes from households and then use those funds to subsidize industrial production. And in 2025, the Ministry of Education added 25 undergraduate majors in higher education, including AI, decarbonization, and digital governance. Such a coordinated and systematic approach is key to China's success in achieving significant advancements in technology and high-tech manufacturing over relatively short periods of time. China's system for supporting tech innovation and emerging industries allows it to mobilize resources from all directions and rapidly scale up production following breakthroughs. However, this system also tends to generate wasteful investment and overcapacity as Chinese local governments respond to the same signals from the central government. In recent years, top leadership has become increasingly concerned about these drawbacks, with President Xi recently stating that not all local governments should develop AI, build data centers, and make EVs. Official outlets now emphasize the importance of building a "unified national market" (全国统一大市场) and growing "new-quality productive forces in alignment with local conditions" (因地制宜发展新质生产力). And, since July, the government has launched "anti-involution" initiatives across many industries to curb investment and prevent excessive price cutting. However, China's repetitive investment and industrial overcapacity issues are deeply rooted in its political system and cannot be easily solved. Local officials, eager to demonstrate political loyalty and competence, often focus policy efforts on sectors prioritized by the central government. Unless the incentive structure changes—enabling local officials to pursue independent development strategies and achieve promotion without pursuing central directives—accelerated investment in industries explicitly mentioned in the 15th FYP proposal will likely continue. As a result, overcapacity will likely remain a feature of China's economy, even as it shifts from steel and cement in the past to solar panels and EVs today, and potentially to semiconductors and AI in the future. China's capacity exceeds global demand for several key products China's capacity as a share of global demand in 2024, % Source: Goldman Sachs GIR. # Leveraging strengths in the US-China tech competition China has increasingly relied on two key advantages in its tech competition with the US: its dominance in global supply chains and strong economic ties with the Global South. A notable example of supply chain leverage is China's recent tightening of export controls on critical minerals—including those announced on October 9 covering rare earth minerals, related products like alloys and magnets, as well as refining and processing technology and equipment. Given the importance of rare earths in high-tech manufacturing and China's dominance in refining and processing, such restrictions could significantly disrupt global manufacturing supply chains, prompting the US to ease some of its semiconductor export controls in exchange for China suspending its proposed rare earth export controls, demonstrating the effectiveness of this leverage. China is also working to establish its technology standards abroad through trade and investment in the Global South. For example, Huawei operates in over 170 countries, encouraging these countries to adopt Chinese telecom technology standards. As more countries follow suit, the resulting network effect increases the attractiveness of China's standards for others. And in 2024, officials called for deeper participation in international standards organizations—including the International Organization for Standardization (ISO), the International Electrotechnical Commission (IEC), and the International Telecommunication Union (ITU)—and emphasized strengthening standard connectivity with Belt and Road Initiative participants. With China now a larger trading partner than the US for over 140 countries, the Chinese government seeks to use these ties to expand markets where Chinese technology prevails. # Hui Shan, Chief China Economist Email: hui.shan@q.com Tel: 852-2978-6634 Goldman Sachs (Asia) LLC # Semiconductors, at a glance # What are semiconductors, and who makes them? Semiconductors are also known as microchips or chips. A microchip is a set of electronic circuits layered on a thin wafer of semiconductor m