Promising AI Infrastructure Companies if Trump Loses

Unlike typical big tech companies, AI-related stocks tend to attract concentrated investment from private equity (PE) and hedge funds. These fund managers utilize a '2/20 structure' to maximize performance fees, increasing investment capital through significant leverage. As the high-interest rate environment persists, funding costs have surged, directly impacting these leveraged investments.

Fund managers typically raise short-term capital for long-term investments in AI companies. However, a sharp rise in short-term interest rates can lead to unbearable increases in funding costs, amplifying the risk of margin calls and fund liquidations. Such vulnerable leverage structures are identified as a primary cause of the current decline in AI-related stocks.

In essence, they employ a 'rollover' strategy, continuously borrowing short-term funds (3 months, 6 months, 1 year, 36 months) to sustain long-term investments. The problem arises when short-term interest rates fluctuate sharply, potentially reversing the profit structure of long-term loans and leading to substantial losses for funds.

Recently, surging U.S. Treasury yields have left hedge funds facing a situation where funding rates exceed lending rates. This situation precipitates a fund liquidation crisis and pressure from institutional investors demanding capital retrieval. If a fund is liquidated, AI companies that received investments from that fund become 'stigmatized,' making it difficult for them to secure additional funding.

Hedge fund liquidation not only floods the stock market with sell-offs but also severely impacts the entire AI corporate ecosystem. When a fund liquidates, a large volume of shares from its portfolio AI companies are dumped onto the market, putting downward pressure on stock prices. This leads to a decline in the corporate value of the affected AI companies.

A bigger problem is that in a structure where AI companies receive 'series-based' funding, fund liquidation halts the flow of capital. AI companies that have received venture capital investment find it difficult to secure the next round of funding and may even fail to raise additional capital if they are 'stigmatized.' This results in delaying the development and commercialization of innovative AI technologies.

During his presidency, former President Trump significantly increased tariffs on China and other major trading partners. These tariffs raised import prices and increased production costs for businesses, contributing to overall inflationary pressure.

Furthermore, the Trump administration's Middle East policies and the instigation of international instability fueled rising energy and grain prices. This supply-side cost-push inflation compelled central banks to raise interest rates, which, in turn, led to increased funding costs for hedge funds, impacting the weakness of AI stocks.

Former President Trump's policies also negatively affected the AI industry. Not only did they stimulate cost-push inflation due to tariffs and conflicts, but they also showed attempts to cancel or cut direct subsidies for AI companies. In particular, there were efforts to reverse or reduce policies that had promised subsidy support for AI infrastructure construction, such as the IRA (Inflation Reduction Act) or the CHIPS Act.

These subsidy cuts make it difficult to secure funding necessary for AI research and development and infrastructure construction, hindering the growth of the AI industry. The advancement of AI technology requires massive initial investment and infrastructure building, and a reduction in government support inevitably slows down the pace of innovation.

Given the current situation, there is a high probability that the Democratic Party will gain control of the House in the upcoming midterm elections. Despite the S&P 500 index rising during former President Trump's tenure, there is a widespread perception that the general civilian economy deteriorated. This has led to heightened voter dissatisfaction, which could empower the Democratic Party.

It is highly probable that the Democrats will control the House while the Republicans maintain the Senate, creating a 'divided government' scenario. In such a situation, the Democratic Party can effectively check former President Trump's policies through the legislative power of the House. This suggests that there could be changes to existing tariff policies or subsidy reduction efforts.

The Democratic Party currently identifies power shortages as one of the main causes of civilian inflation and is actively working on solutions. A prime example is their acknowledgment of the potential of Small Modular Reactors (SMRs) while advocating for standardization for safe use and related subsidies.

Concurrently, the Democratic Party is also actively committed to promoting policies for power grid efficiency and standardization. The advancement of the AI industry entails massive electricity consumption, making improvements to power infrastructure essential. Therefore, if the Democratic Party takes control of the House, government funding for AI infrastructure, including small reactors and power grid efficiency, is expected to expand.

To cope with the immense power demand of the AI era, power grid efficiency is urgently needed. Traditional power lines are structured with an aluminum layer wrapped around a steel core to conduct electricity, which is inefficient and has limited transmission capacity. As an alternative, next-generation power line technology using carbon fiber cores is gaining attention. Carbon fiber cores have very strong tensile strength, allowing for wider cables and a dramatic increase in transmission capacity.

Carbon fiber cores are lighter and stronger than traditional steel cores, reducing cable sag and enabling more power transmission. This critical carbon fiber core is currently produced exclusively by CTC Global. If this technology is commercialized, it is expected to enable stable power supply to AI data centers through upgrades to existing transmission grids.

Sumitomo Electric possesses unique alloy technology that solves the softening phenomenon of aluminum wires, one of the challenges in power grid efficiency. By alloying aluminum with zirconium, they enhance wire strength and prevent sagging, enabling stable power transmission.

Beyond wire technology, this company also boasts a wide range of cutting-edge material technologies essential for the AI era, including next-generation semiconductor materials, glass substrates, and optical fibers. With a technology portfolio that can play a crucial role across the entire AI industry, such as optical fibers for high-speed communication in data centers and next-generation materials for high-performance semiconductors, it is proposed as an investment candidate.

The U.S. can maintain a strong dollar by keeping interest rates relatively high, but most other countries are not in a position to raise rates like the U.S. This exacerbates dollar strength, increasing the burden of import prices for these countries.

Rising import prices eventually lead to domestic inflation, and currency depreciation creates pressure to deplete foreign exchange reserves for exchange rate defense. Ultimately, countries face a situation where they must either sell U.S. Treasury bonds to protect their currency's value or reluctantly join the interest rate hike competition.

China's early release of open models is driving market standardization competition, which in turn prompts U.S. companies to responsively release their own open models. This competition increases AI technology accessibility, laying the groundwork for more developers to utilize AI.

If AI technology becomes more widely and rapidly adopted, it could reignite debates about increased semiconductor demand. The spread of open models fosters the diversification of AI-driven services and applications, potentially leading to a greater demand for semiconductors needed for AI training and inference.

As seen in the 19th-century increase in coal consumption and steel demand, when technology advances, unit costs fall and efficiency improves, the overall demand for that resource often explodes. The decrease in GPU prices and performance optimization could be key to generating demand for AI semiconductors.

Even if GPU functionality is simplified or efficiency increases, as more calculations can be performed at a lower cost, the spread of AI services accelerates. This can ultimately lead to an increase in overall GPU demand. Therefore, the AI semiconductor market is likely to enter a virtuous cycle where technological efficiency leads to price reductions, and these price reductions, in turn, drive increased demand.

Due to these structural characteristics, memory semiconductors are difficult to simplify, which means it is hard to reduce their prices. Questions arise as to whether major memory semiconductor companies like Samsung Electronics and SK hynix have resolved this issue.

Analysis also suggests that structural limitations may restrict demand expansion. Furthermore, the difference in the price-to-earnings ratio (PER) between Micron and Hynix is attributed to shareholder return policies. Micron's PER is 15x, while Hynix's is only 6x, which is seen as reflecting the market's perception that Hynix is passive in shareholder returns. This is interpreted as a reflection of concerns that the company's profitability is not adequately translated into shareholder value.

Micron's mentioned facility investment plan is also interpreted as a strategy to secure the company's long-term growth momentum.

A company's shareholder return policy is part of a strategic choice for long-term corporate value appreciation rather than short-term profit distribution. Especially in the rapidly changing semiconductor industry, maintaining technological leadership through continuous investment is crucial, and this is ultimately judged to lead to long-term shareholder value growth.

The market casts a suspicious eye, fearing that even if SK hynix performs well, its funds could be diverted to other struggling affiliates within the group. These concerns affect SK hynix's shareholder return policy and act as a negative factor in its corporate valuation.

Investors evaluate SK hynix's investment burden and management transparency comprehensively, believing that these uncertainties will remain a discount factor for corporate value until resolved. Ultimately, the market assesses a company's growth potential alongside its risk management capabilities, paying close attention to the potential for affiliate risks within the group to transfer to the parent company, SK hynix.

If the Democratic Party gains control of Congress in the midterm elections, interest in certain policy-benefiting stocks is expected to rise. The Democratic Party will likely show policy interest in Small Modular Reactor (SMR) technology to solve the power shortage problem, and related companies could benefit. Concurrently, active support for technological development to improve transmission efficiency is also expected. In particular, companies like Sumitomo Electric, a key materials company for transmission grid efficiency, should be noted.

Furthermore, the hydrogen fuel cell industry also has a high probability of turning around with budget support, aligning with the Democratic Party's eco-friendly policy stance. Government investment in power supply, crucial for AI infrastructure construction, and sustainable energy solutions is expected to expand, serving as a growth opportunity for related companies.