The Illusion of American Leverage
Material realities come home to roost
In response to China’s announced export controls on rare earth elements and other materials (9 October 2025), the US administration, from President Trump down, spent a week ramping up its rhetoric on tariffs against China. The President himself threatened the application of 100% tariffs, and Treasury Secretary Scott Bessent came out swinging. He basked in erroneous trumped-up reports of the dismissal of China’s trade negotiator, Li Chenggang, only to be humiliated when Li arrived in Kuala Lumpur with a bounce in his stride for trade talks.
The presidents of China and the US are slated to meet on the sidelines of APAC in Seoul later in October, even though for a day or so, Trump seemed to indicate that he wasn’t willing to meet with Xi. That braggadocio moment seems to have passed, and a meeting of sorts is back on the cards.
The US presented a strong rhetorical front, seeking to create the impression that it holds “the cards”. But a closer examination reveals a stark asymmetry, one where U.S. tariffs inflict minimal damage on China’s resilient economy while exposing America’s vulnerabilities in critical sectors like AI and clean energy. Far from bending China to its will, such measures risk boomeranging, amplifying U.S. dependence on Chinese supply chains and opening the US to the risks of potent financial countermeasures. The truth is, in 2025, the U.S. has negligible trade-related leverage over China, and escalating tariffs could accelerate a de-dollarisation trend that undermines American global primacy.
To understand this, let’s start with the numbers. Those who read my April 2 piece Never Look a Gift Horse in the Mouth, will be familiar with the overall tenor of the argument here. The asymmetric significance of real materials - such as rare earths - over dollars was amply on display over first the dysprosium controls (see my Memo to Capital Hill: You can keep your dollars - we’ve got the dysprosium) and more recently with the wider range of export licensing measures announced on October 9, 2025 by China’s MOFCOM, the implications of which I explored here.
So to recap:
China’s exports to the U.S. account for approximately 2.8% of its nominal GDP, roughly $500-560 billion annually based on a $18-20 trillion economy. This figure, while significant in absolute terms, represents only about 15-20% of China’s total exports, which reached a record $328.6 billion in September 2025 alone, surging 8.3% year-over-year despite global headwinds. If the U.S. were to impose blanket 100% tariffs, effectively severing this trade link, the initial shock to China might seem daunting. Yet, some quick economic modelling suggests the gross impact on China’s annual GDP growth rate would be a mere 0.144%, a blip in an economy projected to grow ~5% nominally this year.
Why so small? Exports aren’t pure GDP contributors; their value-added portion (after subtracting imported inputs) is typically 50-70%, bringing the real hit closer to 1.7% of GDP. For the sake of this discussion, let’s use the standard expenditure approach to GDP, whereby GDP = Consumption (C) + Investment (I) + Government Spending (G) + Net Exports (X - M). In this case, the loss of U.S.-bound exports (reducing X) is partially offset by the cessation of U.S. imports (reducing M). China’s imports from the U.S. total around $150-180 billion annually, or 0.8-1% of GDP, including agricultural goods and semiconductors. Halting these improves China’s trade balance by about 1% of GDP, narrowing the net export loss to roughly 1.8%.
Critics might argue this static view ignores ripple effects such as job losses in export hubs like Guangdong, supply chain disruptions or reduced investor confidence. Fair enough, but China’s economy is no longer the export-dependent system of the early 2000s. Domestic consumption now comprises ~50% of GDP, investment another 40%, and government spending almost 10%. Net exports contribute a negligible amount to overall GDP. China, thus, has ample tools to absorb U.S.-related trade shocks. Fiscal policy loosening, through infrastructure stimulus, tax cuts or consumption vouchers, could inject 1-2% of GDP ($180-360 billion) to boost C, I, and G. Historical precedents abound: China’s $586 billion stimulus in 2008 stabilised growth amid global recession, and similar measures in 2020 countered COVID-19 fallout. A range of stimulus measures have also been introduced in 2025, which continue to work their way through the economic system.
Moreover, organic trade redirection to the rest of the world would further mitigate the loss of the U.S. market. The U.S. is just one market; ASEAN, the EU and Japan absorb 45-50% of China’s exports, with ASEAN alone rivaling the U.S. at $500-600 billion. Global trade growth, projected at 3-4% in 2025 by the WTO and IMF, could add $80-120 billion in demand for Chinese goods annually. Even conservatively assuming a 12-month adjustment period - accounting for retooling supply chains and forging new contracts - China could see 20-30% of lost U.S. exports ($100-150 billion) being in effect absorbed by these partners or by its own vast domestic market. Combined with fiscal stimulus, this offsets the 0.144% growth dip entirely, rendering the aggregate impact negligible.
The U.S. market is simply not as important in relative terms as it once was.
In essence, U.S. tariffs amount to a self-fulfilling prophecy of diminished leverage. As trade flows dry up, China’s need to accumulate and recycle U.S. dollars evaporates. Historically, Beijing’s trade surplus with America has fuelled its purchase of U.S. Treasuries, stabilising the yuan and “funding” U.S. deficits. But with no imports to finance in dollars, China faces a surplus of USD inflows from maturing bonds and reserves. As of July 2025, China’s Treasury holdings stand at $730.7 billion - the lowest since December 2008 - down from over $1 trillion in recent years, signalling ongoing diversification amid tensions.
What might China do with this hoard? The “nuclear option” - a massive dump of $100-300 billion in Treasuries - has been debated ad nauseam. Such a sell-off could flood the market, spiking 10-year yields by 0.5-1.5% (from 4% to 5.5-6.5%), inflating U.S. borrowing costs by $50-150 billion annually, depreciating the dollar by 5-15%, and fuelling CPI inflation by 1-3%. The Federal Reserve might intervene, but at the risk of recession, potentially shaving 0.5-1% off U.S. GDP growth in 2026. Yet, this would devalue China’s remaining holdings (a 10% price drop costs $70 billion) and provoke retaliation, like asset freezes or sanctions on Chinese banks.
All that’s true, but there’s a twist. In a no-trade scenario, these deterrents fade. With tariffs already severing economic ties, devaluation losses become “meaningless, by and large,” as the assets lose strategic value for trade recycling. Bank sanctions? Irrelevant when U.S. exposure is minimal. That said, China is unlikely to pull the trigger. The existence of a tacit threat alone is far more potent. Subtle signals - state media hints, small sell-offs of $10-20 billion - could unsettle markets, raising yields 0.2-0.5% and depreciating the dollar 2-5% without incurring losses. This brinkmanship forces U.S. policymakers to blink, deterring escalation while preserving China’s reserves for diversification into gold, euros or commodities, further accelerating de-dollarisation via BRICS yuan trade. In any case, China has little reason to be seen as to be oiling the market.
Compounding this financial leverage is America’s Achilles’ heel, namely, supply chain dependence. A trade cutoff wouldn’t just halt $462 billion in U.S. imports from China (2024 figures, with 2025 on track to match); it would cripple strategic industries. Take rare earth elements, essential for semiconductors and magnets in AI hardware as a case in point, and which sparked off the latest round of tete-a-tete. China controls 91% of global refining and 85% of supply, with the U.S. importing 70-80% directly or indirectly from China. A ban or disruption - brought about by China’s tightened export controls on rare earths and lithium batteries and other key industrial materials - could delay U.S. chip production by 6-18 months, stalling AI growth projected at 30-40% this year. Firms like Nvidia and AMD would face shortages. What would this cost them in output? Perhaps up to $50-150 billion in foregone output. And how would the market react? Would tech stocks - many of which are increasingly seen to be chronically over-valued - tank? Are impairments of 20% realistic?
Jitters around the AI-fuelled tech stock bubble are likely to shape Washington’s risk tolerance, just as they have done in the past.
The energy sector fares no better. China dominates 75-80% of solar panel production and 70-80% of battery minerals like lithium, cobalt and graphite. The U.S., to the extent that it pursues clean energy, relies on China for nearly three-quarters of its lithium-ion batteries and 75% of EV battery imports. A supply cutoff could slash solar installations 20-40%, hike battery costs 30-50% and stall $100-200 billion in projects. This could see electricity prices increase 5-15% and hinder AI data centres’ 50% power demand surge. Combined with a Treasury threat-induced yield spike, this “double whammy” could shave 0.7-1.5% off U.S. GDP ($200-400 billion).
This asymmetry of possible impacts described here underscores another observation. America’s trade war strategy is outdated. China’s economy, once export-reliant particularly vis-a-vis the mature advanced markets (basically in the 2000s), now thrives on an ever-growing domestic market, domestic innovation and global partnerships like the Belt and Road Initiative. Beijing can weather a 12-month adjustment with fiscal buffers and diversified trade, emerging stronger by reducing USD dependence.
The U.S., meanwhile, grapples with material constraints to AI and energy development, both of which are symbiotically fundamental to America’s own development narrative. Limited material resources in the face of competing demands for electricity supplies from AI and manufacturing ultimately creates the risk of an electricity-induced Dutch Disease moment, in which costs across all economic and social sectors rise.
Add to this mix the disruptions to low-end chips occasioned by the hamfisted expropriate from Chinese ownership of Nexperia by the Dutch government under pressure from the United States, and it is clear that many downstream industries could be temporarily disrupted or even permanently crippled through upstream supply chain blockages. You don’t get to re-industrialise without more and cheaper electricity and robotics or automation; and much of today’s “day to day” industrial outputs like automotives are dependent on the steady supply of high volumes of mid- to low-range chips, a large proportion of which are actually made in China. Bosch, VW, Nissan, Honda (Japan) and even U.S. based car makers, are warning of temporary production suspensions. These are self-inflicted wounds.
Tariffs may rally domestic political bases but ultimately will only exacerbate inflation, slow growth and cede technological ground.
As trade negotiators from the US and Chinese side convene in Kuala Lumpur, and preparations are underway for a presidential summit in Seoul, it is now increasingly clear that the U.S. speaks loudly but in fundamental material terms, it’s China that holds the cards. After all, the cards are made in China.