● China-Boosted, Trump-Driven, Power Shift
Did Trump Strengthen China? The Shift in Power Balance After the U.S.-China Summit and the Global Economic Outlook
The core issue in the latest U.S.-China summit was not whether a formal agreement was signed, but that China was presented before the world as a strategic peer of the United States.
On the surface, the headlines focused on tariffs, Taiwan, trade negotiations, and personal rapport between the two leaders. However, the more important issues were China’s recovery in soft power, changes in the U.S. alliance structure, renewed confidence in Chinese AI after DeepSeek, and the shift in U.S.-China competition from a Cold War framework to a “Cool War” dynamic.
Trump-style diplomacy, while advocating America First, has effectively elevated China’s international standing. This is a key development for assessing China’s economic outlook and global investment strategy.
1. Why the U.S.-China Summit Should Not Be Viewed as a “No Result” Meeting
U.S. and Korean media largely characterized the summit as one with significant noise but limited substantive outcomes.
However, Professor An Yu-hwa’s interpretation differs.
The central issues in U.S.-China relations cannot be resolved in a single summit. Tariffs, Taiwan, technology leadership, global supply chains, and security architecture are structural conflicts, not matters that can be settled through a joint statement.
Accordingly, the summit’s significance should be assessed through the image it created and the geopolitical message it conveyed, rather than through formal documentation.
- How the U.S. president treated the Chinese president.
- How global media framed the interaction between the two leaders.
- How third countries interpreted the balance of power between the United States and China.
- What level of confidence Chinese elites and technology talent derived from it.
These four factors are the real significance of the summit.
2. Trump Helped Offset China’s Biggest Weakness: Soft Power
One of China’s main weaknesses has been soft power.
China has strong manufacturing capability, infrastructure investment capacity, and technological development, but it has lacked the global appeal that the United States has cultivated.
The United States has built its international image through Hollywood, Silicon Valley, dollar dominance, its university system, and the narrative of democracy.
China, by contrast, has struggled to project attractiveness despite its strong state capacity.
In this summit, however, Trump treated Xi Jinping with visible respect and projected a sense of parity. That delivered an image benefit that China could not have purchased directly.
This was not merely ceremonial. In international politics, protocol is a language of hierarchy.
The fact that the U.S. president treated the Chinese president as an equal counterpart provided China with a meaningful diplomatic asset.
For China, the message was clear: it is no longer merely a subject to be managed as in the past, but a key player the United States must recognize.
3. The Real Impact of the Summit Lies in the Perception Shift Among Third Countries
The summit was not an event relevant only to the United States and China.
It was watched closely by Europe, the United Kingdom, Canada, Southeast Asia, the Middle East, South Korea, and Japan.
If Trump changes how he engages China, third countries will also recalibrate their positions.
Many may conclude that if the United States is willing to treat China in this way, there may be less need to keep distance from Beijing.
In practice, as U.S. pressure rises, Europe and other advanced economies are more likely to pursue strategic hedging rather than fully decouple from China.
This matters for the global economic outlook.
As U.S.-China tensions intensify, the world is not necessarily splitting into two rigid blocs. Instead, many countries are likely to adopt pragmatic balancing strategies.
This trend has direct implications for global supply chains, commodities, semiconductors, AI infrastructure investment, exchange rates, and capital flows.
4. DeepSeek Gave Chinese AI Talent Something More Important Than Funding: Confidence
Another key point is DeepSeek.
Before DeepSeek, there was reportedly a degree of pessimism among Chinese engineers.
Given U.S. dominance in AI chips, cloud infrastructure, the OpenAI ecosystem, and the global developer network, many believed China would struggle to keep pace at the frontier of AI.
That changed after DeepSeek.
Chinese engineers began to believe that they could also become global leaders.
In technology competition, confidence is not merely an emotion.
It affects talent retention, entrepreneurship, capital formation, and the direction of national policy.
From an AI trend perspective, DeepSeek should be viewed not simply as the success of a model, but as a psychological turning point for China’s technology ecosystem.
U.S.-China competition is no longer defined only by semiconductor export controls.
It has become a comprehensive contest involving AI models, data, power infrastructure, open-source strategy, and the confidence of engineering communities.
5. Trump Pressures Allies, but That Also Reflects Greater U.S. Dependence on Them
Trump has consistently demanded more from allies in defense spending, trade terms, and investment burden-sharing.
On the surface, this appears to be an attempt to pressure allies.
However, Professor An interprets this differently.
The U.S. can no longer bear the full burden alone, which makes allies more necessary than before.
In other words, the former U.S. stance was closer to “I will handle it,” whereas the current stance is “I am under pressure too, so you must contribute.”
This reflects changes in the U.S. economic position.
Large fiscal deficits, debt service burdens, industrial restructuring, rising military expenditure, and inflationary pressure are pushing the United States to demand greater contributions from allies.
Trump’s pressure strategy is therefore better understood as alliance re-negotiation rather than alliance dismantlement.
For South Korea, this shift is highly significant.
Defense cost-sharing, semiconductor investment, battery supply chains, shipbuilding cooperation, nuclear power exports, and AI data center infrastructure may all be linked to the U.S. demand for greater burden-sharing from allies.
6. U.S.-China Relations Are Moving Toward a “Cool War,” Not a Cold War or Hot War
Professor An describes U.S.-China relations as a “Cool War.”
This is neither full separation as in the Cold War, nor military escalation into a hot war.
It is a condition of intense competition without total destruction of the other side.
Xi Jinping’s key phrase was “strategic stability.”
This implies that the two countries should avoid a Thucydides Trap scenario and compete without turning the relationship into a zero-sum confrontation.
China’s message to the United States is essentially this:
“We will not immediately challenge your status.”
“But do not treat us as you did in the past.”
“A new relationship must reflect our influence.”
This message may have had some traction with Trump.
If the United States and China enter direct confrontation, both sides would incur substantial losses.
The United States would face inflation and supply-chain shocks, while China would suffer damage to exports and access to technology.
Accordingly, the two countries are likely to maintain a complex relationship of competition, negotiation, pressure, and transaction.
7. Trump’s Display of Personal Rapport Is a Political Device, Not a Weakness
Trump often emphasizes personal rapport in summit diplomacy.
Phrases such as “we are very close” are characteristic of his political style.
Xi Jinping, by contrast, prefers language centered on bilateral prosperity, long-term vision, and strategic relations.
This also reflects differences in political culture between the United States and China.
U.S. politics emphasizes the individual, leadership, performance, and personal image.
Chinese politics emphasizes the state, the collective, history, face, and long-term strategy.
Trump’s emphasis on his relationship with Xi is not simply boastful behavior. It reinforces his image as a leader capable of direct dealings with the world’s most important figures.
He designs diplomacy around scenes and public signaling rather than policy documents.
From this perspective, Trump’s protocol, remarks, expressions, and on-camera behavior are all part of negotiation.
8. Taiwan Remains Both the Core of U.S.-China Conflict and a Negotiation Tool for Trump
Taiwan is the most sensitive point in U.S.-China geopolitics.
The fact that Taiwan was discussed in the summit is highly significant.
It suggests that Xi Jinping was able to raise the issue directly with the U.S. president at the highest level.
At the same time, it places Trump in a position where he must hear China’s demands on Taiwan directly.
There is, however, a more important interpretation.
If China were truly overwhelmingly strong, it would not need Trump to state opposition to Taiwan independence.
As with Hong Kong and Macau, if China had full control, external validation would not be necessary.
In other words, China’s need for U.S. statements indicates that its power is not yet complete.
Trump understands this well.
That is why he uses Taiwan as a bargaining chip in negotiations with China.
For Trump, Taiwan is not primarily an ideological issue but a transactional instrument.
He can tighten or relax pressure depending on what maximizes U.S. gains.
That said, a direct U.S. trade of Taiwan to China is not realistic.
Taiwan is deeply tied to semiconductor supply chains, U.S. Indo-Pacific strategy, Japanese security, and South Korean security.
Accordingly, Taiwan is both a short-term bargaining chip and a long-term flashpoint in the great-power competition.
9. Iran Negotiations, China’s Mediation Role, and the Possibility of a U.S.-China Grand Bargain
An important point raised after the summit was the reference to Iran-related peace negotiations.
This suggests that the United States may have viewed China as having a potential mediating role in Middle East issues.
China has deep economic, energy, and diplomatic ties with Iran.
From Washington’s perspective, managing Middle East risk may require Chinese cooperation.
This illustrates the complexity of U.S.-China relations.
The United States must contain China, but it also needs China’s cooperation on many issues.
The same applies to Russia, Iran, North Korea, commodity markets, global financial stability, climate issues, and supply chain management.
U.S.-China relations are therefore not simply adversarial. They combine competition and cooperation.
10. Why the “China Collapse Theory” Is Difficult to Accept at Face Value
The latter part of the discussion addressed the so-called China collapse theory.
In South Korea and the West, repeated forecasts cite China’s property crisis, local government debt, youth unemployment, demographic decline, and weak consumption as evidence that China is nearing collapse.
Professor An does not accept this view.
China does face serious economic problems.
The property sector is under restructuring, private consumption is weak, and youth employment is a major concern.
However, it is overly simplistic to conclude that these issues will lead directly to state collapse.
China manages crises differently from Western market economies.
Because the state exerts strong control over the financial system, banks, land, industrial policy, and capital flows, the pace and form of crisis transmission differ.
For China’s economic outlook, the key issue is not whether it will “collapse,” but how its growth model will change.
China previously relied on property, infrastructure, exports, and low-cost manufacturing.
Going forward, it is shifting toward advanced manufacturing, electric vehicles, batteries, solar power, AI, robotics, semiconductor self-sufficiency, and digital currency initiatives.
The challenge is that this transition is highly painful.
China should therefore be assessed not in terms of collapse, but in terms of prolonged slow growth, industrial restructuring, weak domestic demand, and technological self-reliance competition.
11. The Most Important Point Rarely Highlighted in Mainstream Coverage
The key issue in this summit was not tariff percentages, the presence or absence of a joint statement, or the tone of Taiwan remarks.
The real significance was that the United States publicly treated China not as a managed subordinate, but as an equal variable that must be negotiated with.
This granted China part of the international legitimacy it has been seeking.
China gained something more important than growth rates.
It gained the image that the United States can no longer dismiss it casually.
This image also strengthens internal cohesion within China.
Chinese citizens, entrepreneurs, engineers, and officials may interpret the U.S. president’s treatment of the Chinese leader as recognition of national standing.
The confidence gained by Chinese engineers after DeepSeek and the diplomatic confidence generated by this summit are linked.
Ultimately, U.S.-China competition is not only a contest of military power and GDP.
It is also a contest of confidence, narrative, image, face, and third-country perception.
This is the dimension most frequently omitted in mainstream reporting.
12. Implications for Investors and the South Korean Economy
First, U.S.-China tensions are not ending.
However, they are more likely to continue in a managed competition framework than to lead to full decoupling.
Second, an outright exclusion of China-related assets may be a risky strategy.
China’s economy is slowing, but it remains highly competitive in AI, electric vehicles, batteries, robotics, solar power, and advanced manufacturing.
Third, while the U.S. economy appears dominant, it is facing fiscal pressure and the growing burden of alliance costs.
Fourth, South Korea is not simply choosing between the United States and China. It must manage supply-chain exposure and security simultaneously.
Fifth, the AI trend may expand beyond U.S. big tech to include Chinese open-source and low-cost high-performance competition.
Sixth, geopolitical risk increases market volatility, but it can also create opportunities in defense, power infrastructure, semiconductor equipment, data centers, and energy security-related sectors.
13. Key Variables to Monitor Going Forward
- How Trump’s China rhetoric changes during the U.S. presidential and midterm election cycles.
- Whether China increases military pressure on Taiwan or focuses on managing its diplomatic messaging.
- The pace of global expansion by Chinese AI firms following DeepSeek.
- Whether U.S. semiconductor export controls further accelerate China’s domestic technology development.
- The extent to which Europe, the United Kingdom, and Canada expand economic cooperation with China.
- Whether China’s property crisis becomes a systemic financial risk or is absorbed through long-term restructuring.
- How U.S. Treasury yields and the dollar respond to changes in U.S.-China relations.
< Summary >
The latest U.S.-China summit appeared to produce few formal agreements, but it materially improved China’s international image.
Trump’s America First approach nevertheless made China appear as a strategic peer through protocol and negotiation style.
China gained soft power and internal confidence, while DeepSeek marked a psychological turning point in AI competition.
U.S.-China relations are likely to develop as a “Cool War,” combining competition and cooperation rather than escalating into cold separation or open conflict.
Taiwan remains the most important geopolitical risk, but it is also a key bargaining tool in Trump’s negotiations with China.
The China collapse narrative is too simplistic; the more relevant framework is long-term slow growth and industrial restructuring.
For South Korea and investors, the critical variables are U.S.-China competition, China’s economic trajectory, changes in the U.S. economy, AI trends, and global supply-chain realignment.
[Related Articles…]
- U.S.-China Power Competition and the Global Economic Order
- AI Trends and the New Variables in China’s Technology Rise
*Source: [ 경제 읽어주는 남자(김광석TV) ]
– 트럼프가 중국을 키워줬다? 미중 정상회담에서 바뀐 ‘힘의 균형’ | 경읽남과 토론합시다 | 안유화 교수 [1편]
● DRAM, AI, Bottleneck, Surge
At the End of the Bottleneck Is DRAM: Investment Themes From the SOL Global DRAM Semiconductor Plus ETF
The key point in this discussion is not simply that “HBM is strong.”
The real issue is that the bottleneck in the AI semiconductor market is shifting from GPUs to HBM, and then to DRAM and storage.
In particular, demand is growing by more than 20% annually, while supply growth for commodity DRAM may remain close to 5%, creating a significant supply-demand gap.
If this structure persists, it could support higher memory prices, improve earnings for Samsung Electronics and SK Hynix, and expand demand for global semiconductor ETFs.
The newly launched SOL Global DRAM Semiconductor Plus ETF is notable as a Korea-listed ETF designed to capture this DRAM-centered trend.
1. The AI Semiconductor Bottleneck Is Shifting From GPUs to Memory
The first main beneficiary of the AI investment cycle was the GPU.
AI models require massive parallel processing, which made GPUs, rather than CPUs, a key focus.
This elevated Nvidia to a central position in global equity markets, while Google TPU and other big tech in-house accelerators also expanded rapidly.
However, as AI accelerators improved, a new constraint emerged.
Processing power increased, but the bandwidth for moving data into GPUs became limited.
HBM, or high bandwidth memory, was introduced to address this issue.
HBM uses a vertically stacked DRAM structure to enable faster data transfer.
As a result, higher HBM demand also increases DRAM demand.
The investment theme in AI semiconductors is therefore expanding from GPUs to HBM and now to DRAM as a whole.
2. Why DRAM Is the Most Important Bottleneck Now
As the AI market shifts from training to inference and agentic AI, memory demand is increasing further.
In the training phase, the priority was building large models.
In the inference phase, the priority becomes supporting a large number of concurrent user requests and responses.
Agentic AI goes further by retaining context, retrieving data, and continuing tasks.
This requires large-capacity server DRAM.
To handle long context windows and concurrent workloads, memory requirements rise sharply.
Demand extends beyond HBM to server DRAM, enterprise SSDs, high-capacity HDDs, and NAND storage.
In other words, as AI infrastructure investment expands, memory semiconductors become a bottleneck across the stack.
This is the starting point for the DRAM investment case.
3. Key Numbers: 22% Demand vs. 5% Supply
The most important figures in the original analysis are 22 versus 5.
22 refers to the long-term DRAM demand growth rate.
According to Samsung Securities, long-term DRAM demand is expected to grow by approximately 22% per year.
Forecasts vary by brokerage, but the common view is that demand expectations are being revised upward over time.
By contrast, 5 refers to the supply growth rate for commodity DRAM.
Even if the memory trio expands wafer capacity by about 15% annually, roughly 70% may be allocated to HBM.
Under that scenario, effective supply growth for commodity DRAM may be limited to around 5% per year.
Demand is rising by more than 20%, while supply is increasing by only about 5%.
This imbalance supports the case for higher DRAM prices.
It also explains why the memory cycle may be shifting from a standard cyclical pattern toward a structurally tight supply environment.
4. Why Supply Cannot Be Expanded Quickly
Many investors may ask a simple question: why not just build more fabs?
In semiconductors, the answer is more complex.
First, new fab expansion takes time.
Equipment must be installed, cleanrooms completed, testing performed, and yields stabilized.
This is not a process that can be completed in one or two years.
Second, yield improvement is the key variable.
Samsung Electronics’ earlier difficulty in the HBM market was largely related to yield and customer qualification.
In semiconductors, installed capacity does not immediately translate into usable output.
Third, cleanroom capacity is limited.
Because semiconductors have long been viewed as a cyclical industry, companies have historically avoided aggressive expansion, even during upcycles.
Investment was even more restrained during downturns.
As a result, when the AI semiconductor supercycle accelerated after 2023, supply tightened rapidly.
Samsung Electronics and SK Hynix are announcing new expansion plans, but meaningful benefits are expected only from 2027 onward.
Accordingly, the DRAM supply shortage is unlikely to ease quickly in the near term.
5. As HBM Expands, Commodity DRAM Becomes Tighter
A critical point in this discussion is the relationship between HBM and commodity DRAM.
HBM is a structure that stacks DRAM dies vertically.
On a comparable capacity basis, HBM uses roughly 3 to 5 times more wafer consumption than standard DRAM.
At the same time, HBM demand cannot be reduced in the AI accelerator market.
Nvidia GPUs, AMD AI accelerators, and big tech custom chips all require more HBM.
As wafer allocation shifts toward HBM production, capacity available for commodity DRAM declines.
This matters for a simple reason.
Strong HBM demand benefits memory manufacturers.
At the same time, it tightens commodity DRAM supply further.
That increases the likelihood of higher prices for server DRAM and standard DRAM as well.
In other words, the HBM cycle is not only an HBM story. It can also lift the entire DRAM pricing cycle.
6. This Memory Cycle May Differ From Past Cycles
The semiconductor industry has traditionally been highly cyclical.
Investment concentrated in upcycles, supply increased, and prices later weakened, leading to downturns.
This is why Samsung Electronics and SK Hynix have often traded in long-term ranges.
However, this cycle may be different.
First, AI semiconductors are becoming infrastructure rather than consumer products.
In the past, demand was driven largely by B2B markets such as PCs, smartphones, and servers.
Now, governments and big tech are expanding AI infrastructure spending as a strategic priority.
Second, long-term supply agreements have changed in nature.
In the past, customers could often cancel orders, leaving memory companies to absorb the burden.
Today, suppliers have stronger bargaining power, with advance payments, cancellation penalties, and legally binding contract terms becoming more common.
Some agreements are also reported to include clauses that allow pricing adjustments.
Third, the memory industry has consolidated into a three-player structure centered on Samsung Electronics, SK Hynix, and Micron.
This reduces the likelihood of destructive competition.
These three companies now have greater ability to manage profitability through supply discipline.
As a result, the memory cycle may last longer and earnings stability may improve relative to the past.
7. Core Structure of the SOL Global DRAM Semiconductor Plus ETF
The product introduced here is the SOL Global DRAM Semiconductor Plus ETF.
It is described as a Korea-listed global DRAM-focused ETF scheduled to list on October 7.
The original video was produced with support from Shinhan Asset Management.
The ETF’s main feature is its concentration on the global memory trio.
Samsung Electronics, SK Hynix, and Micron each account for roughly 25%, bringing the combined weight of the three names to about 75%.
The remaining 25% is allocated to NAND, storage, memory equipment, and high-capacity HDD-related companies.
The portfolio contains about 10 names in total, resulting in a relatively concentrated structure.
A more concentrated portfolio makes it easier to maintain meaningful exposure to the sector leaders.
This ETF is designed to capture the memory semiconductor cycle through the companies most directly tied to the industry.
8. How It Differs From Existing HBM ETFs
There are already several HBM-related ETFs listed in Korea.
HBM is clearly important in the AI semiconductor market.
However, a pure HBM strategy may overlook commodity DRAM, server DRAM, NAND storage, and HDD exposure.
The SOL Global DRAM Semiconductor Plus ETF is closer to a product that captures the full memory bottleneck, not just HBM.
As AI infrastructure investment expands, demand is rising not only for HBM but also for DRAM, SSDs, HDDs, and NAND across the data processing and storage ecosystem.
Geographic diversification is another feature.
The ETF includes Korean leaders Samsung Electronics and SK Hynix, U.S. player Micron, Japanese NAND leaders, and U.S. high-capacity HDD companies.
For retail investors, this provides access to the global memory semiconductor value chain without selecting individual countries or stocks.
9. What It Means to Access DRAM Through Retirement Accounts
The SOL Global DRAM Semiconductor Plus ETF is a Korea-listed ETF.
That means it can be used in retirement accounts such as IRP and pension savings accounts.
Until now, direct access to a global DRAM theme through retirement accounts has been limited.
Using retirement accounts may provide tax deferral benefits.
Compared with direct investment in overseas ETFs, the tax treatment may be more favorable.
The original material noted that tax burdens could fall to the 5% range under tax-deferral structures.
However, taxes vary depending on account type, holding period, and withdrawal method.
Investors should confirm the applicable rules with their broker and tax advisor before investing.
10. The Key Point That Other Coverage Often Misses
First, the key issue is not simply that HBM is strong; it is that HBM production tightens commodity DRAM supply.
Most coverage focuses only on rising HBM demand.
From an investment perspective, the more important point may be that HBM expansion absorbs capacity that would otherwise support commodity DRAM.
Second, memory manufacturers now have much stronger pricing power than in the past.
Historically, long-term supply contracts often left memory companies bearing the burden when customers canceled orders.
Today, advance payments, cancellation penalties, and price adjustment clauses are improving supplier leverage.
Third, AI semiconductors are moving from a consumer cycle to an infrastructure cycle.
Unlike the past, when demand depended on PC and smartphone replacement cycles, big tech and governments are now making long-term investments in AI data centers.
This may reduce earnings volatility and extend the cycle.
Fourth, a DRAM ETF should be viewed as a way to invest in supply-demand imbalance, not merely as a thematic product.
The investment case is based on the widening gap between demand and supply, not on theme exposure alone.
11. Risks Investors Should Monitor
Even with a strong DRAM investment case, risks remain.
First, if AI infrastructure spending slows more than expected, memory demand forecasts may weaken.
Second, differences in HBM yield improvement between Samsung Electronics and SK Hynix may lead to stock-specific divergence.
Third, U.S. semiconductor restrictions and China-related export controls may affect global semiconductor ETF returns.
Fourth, if DRAM prices rise too quickly, customers may adjust inventory levels.
Fifth, for Korea-listed overseas ETFs, investors should also monitor currency movements and index construction methodology.
Accordingly, this ETF is more suitable for investors taking a medium- to long-term view on AI infrastructure investment and the memory semiconductor supercycle rather than those seeking short-term price swings.
Investment decisions should be based on portfolio allocation, risk tolerance, and investment horizon.
12. Investment View in One Line
The first bottleneck in the AI semiconductor market was GPUs.
The second was HBM.
The third is now DRAM and storage.
Even if GPUs continue to improve, AI performance remains constrained without sufficient memory to store, retrieve, and process data.
DRAM may therefore be re-rated not as a simple component, but as a core bottleneck asset in AI infrastructure.
The SOL Global DRAM Semiconductor Plus ETF is designed to capture this trend through concentrated exposure to Samsung Electronics, SK Hynix, and Micron.
Investors considering semiconductor ETFs should look not only at HBM, but also at DRAM supply constraints and the memory pricing cycle.
< Summary >
The bottleneck in AI semiconductors is shifting from GPUs to HBM, and then to DRAM and storage.
DRAM demand may grow at around 22% annually, while commodity DRAM supply growth could remain near 5%.
As HBM production expands, wafer capacity may shift toward HBM, intensifying the shortage of commodity DRAM.
The memory trio of Samsung Electronics, SK Hynix, and Micron now has stronger pricing power than in the past.
The SOL Global DRAM Semiconductor Plus ETF allocates roughly 75% to these three companies and the remainder to NAND, storage, and equipment-related names.
The ETF may also be usable in retirement accounts with tax deferral benefits, but investors should carefully assess currency, cycle, and regulatory risks.
[Related Articles…]
- AI Semiconductor Supercycle and Memory Investment Strategy
- Global ETF Opportunities in Next-Generation AI Infrastructure
*Source: [ Jun’s economy lab ]
– 병목의 끝에는 DRAM이 있다! SOL글로벌DRAM반도체플러스가 나왔습니다(SOL글로벌DRAM반도체플러스 ETF)
