AnalysisTower Semiconductor caught in tug-of-war between U.S. and China over chip supremacy
Tower Semiconductor caught in tug-of-war between U.S. and China over chip supremacy
Intel’s $5.4 billion acquisition of the Israeli company faltered as geopolitical tensions reshape the chip landscape
When two giants engage in a direct confrontation, it is often the smaller entities that bear the initial brunt. In the ongoing struggle between the U.S. and China for dominance in the chip industry, the local chip manufacturer from Migdal Ha'Emek finds itself squarely caught in the crossfire. The Chinese regulatory authority's refusal to greenlight the multi-billion-dollar deal selling Tower Semiconductor to Intel isn't primarily due to the inherent nature of the deal or its potential competitive implications within China. Instead, it is a peripheral casualty in the larger chip conflict that has been escalating between the U.S. and China over the past several months.
About a year and a half ago, Tower Semiconductor and Intel reached an agreement for Intel to acquire Tower for a cash sum of $5.4 billion. At the time, this amount represented a considerable premium of more than 60% over Tower's stock price. Tower Semiconductor specializes in producing relatively straightforward chips for other companies and doesn't operate on the same scale as manufacturers like TSMC, which are responsible for crafting advanced, high-performance chips used by companies like Apple. Nevertheless, Tower has significant clients such as Broadcom and holds a notable presence in critical markets like electric vehicle chips and camera sensors.
For Intel, purchasing Tower was viewed as an opportunity to leverage the strategic transformation spearheaded by CEO Pat Gelsinger since he assumed office in February 2021. Historically, Intel has focused solely on producing chips developed and designed in-house. This approach had proven successful during the boom years of personal computing, spanning from the 1980s through the early 2000s. Intel managed to establish its brand in households, with the Pentium name becoming synonymous with computing and the "Intel Inside" label becoming a sought-after feature for new computers.
However, Intel failed to effectively integrate itself into the smartphone revolution and missed the emergence of the foundry model, which involves companies providing chip manufacturing services for those lacking their own fabrication facilities. Prominent players like Nvidia, the world's most valuable chip company with a market capitalization surpassing one trillion dollars, don't possess independent manufacturing capabilities. Qualcomm, responsible for processors in advanced Android devices, also outsources chip production. Even Apple, which designs most of its chips for iPhones, Macs, and other devices, relies on external manufacturers like Taiwanese giant TSMC for fabrication.
Up until Gelsinger's arrival, Intel had adhered to the outdated strategy of exclusively producing its own chips. This caused Intel to overlook a substantial opportunity, despite having the knowledge and expertise necessary to compete against TSMC, Samsung, and others. Instead of securing lucrative manufacturing contracts from major names like Apple, Nvidia, and Qualcomm, these deals went to competitors willing to manufacture chips for them.
Recognizing this lapse, Gelsinger made early moves to pivot Intel toward the foundry sector. While this transformation is a long-term process demanding investments in the tens of billions, it offers the potential to diversify Intel's revenue streams significantly. As part of this strategy, Intel embarked on several initiatives, including a projected $50 billion investment in constructing new manufacturing facilities in Ohio and Arizona, an anticipated $25 billion outlay for a new factory in Kiryat Gat, and the acquisition of Tower. This strategic shift was intended to grant Intel entry into the less prestigious but high-demand realm of less advanced chips and to potentially transition Tower's existing customers to contracts for producing high-performance chips.
Gelsinger's approach aligned with the Biden administration's broader vision to establish the U.S. as a dominant force in both chip development and production. A prominent aspect of this vision was the passage of the CHIPS Act by Congress last summer, allocating $50 billion for enhancing and fortifying the domestic chip industry. Of this amount, $39 billion would be distributed in the form of subsidies to companies establishing manufacturing facilities within the U.S. Notably, Intel and TSMC, the latter of which is also constructing a plant in Arizona, were poised to receive substantial portions of these subsidies.
Regrettably, the flip side of the Biden administration's strategy has now led to the collapse of the Tower acquisition deal. This is because, in tandem with fostering local industry growth, the government also intends to curb China's chip sector and, above all, prevent its access to advanced chip technology. In October, the administration enacted a prohibition on exporting advanced chips and chip-making equipment to China. More recently, there have been moves to restrict the export of AI chips (Nvidia developed a specialized AI chip that adhered to these restrictions) and to limit Chinese companies' access to cloud services leveraging such chips. Moreover, this month saw a prohibition on American investments in chips, AI, and quantum computing.
In response, China imposed its own export restrictions on critical materials like germanium and gallium, essential for producing advanced chips and other technological goods, such as LCD screens. Although the Chinese regulatory authority hasn't explicitly commented on the Tower deal, it is reasonably inferred that the non-approval is a consequence of U.S.-China tensions and a desire to curb the rise of an American competitor, with the potential strengthening of Tower coming perhaps at China's expense.
China's chip industry predominantly focuses on manufacturing simple chips more akin to Tower's products than TSMC's. The link to Intel could furnish Tower with a considerable advantage, enabling it to secure contracts at the expense of Chinese manufacturers and potentially impacting local production to some extent. This potential setback might have been more palatable for China if not for the American restrictions. Conceivably, China might have even sought conditions for completing the deal that better suited its interests. Nevertheless, the ongoing dramatic U.S.-China clash over chip supremacy has created a situation where neither side, especially China, is overly concerned about collateral damage involving a relatively small chip producer in Migdal Ha'emek.
The Chinese regulatos approval was required as China is a significant market for both companies. The Chinese regulator is renowned for its cautious approach toward transactions involving foreign and Chinese companies. This wariness stems from concerns over potential technology transfers to foreign entities and the possible impact on competition within the Chinese market. The regulator has previously declined approval for numerous transactions, including one involving the American firm Tesla and the Chinese company Neo.
By virtue of its policy, the Chinese regulator is required to sanction any merger or acquisition involving a company generating revenue from China. In the case of Intel, 27% of its earnings (equivalent to approximately $17 billion) were derived from China in 2022.
Yuval Weinreb, founding partner at Krystal Eight and host of the "Understanding China" podcast, remarked, "China has stacked up difficulties under the guise of antitrust concerns, but it's quite evident that this is actually a ploy to exert pressure on the U.S. Intel, a leading U.S. chip manufacturer, benefits from government subsidies via the CHIPS Act. China seeks to use its leverage, particularly its vast electronics manufacturing market, which is crucial for companies like Intel. China has been Intel's top revenue source for nearly a decade, affording it substantial antitrust influence. Despite the Israeli-American nature of this acquisition, Intel cannot afford to relinquish the Chinese market or engage in disputes with Chinese regulators, making it probable that they abandoned the purchase."
While Tower can take solace in the $353 million cancellation fee from Intel, the loss to Intel is far greater than the payout to Tower. The acquisition was anticipated to bolster Intel's portfolio, positioning it as a quality contender capable of providing customers a comprehensive array of chip solutions.
“Our foundry efforts are critical to unlocking the full potential of IDM 2.0, and we continue to drive forward on all facets of our strategy,” said Intel CEO Pat Gelsinger. “We are executing well on our roadmap to regain transistor performance and power performance leadership by 2025, building momentum with customers and the broader ecosystem and investing to deliver the geographically diverse and resilient manufacturing footprint the world needs.”
However, the reality is that without Tower, its capabilities, manufacturing facilities, and the customers it brings, Gelsinger's task of transforming Intel into a leading foundry player will become more challenging, costly, intricate, and time-consuming.