Source: Content compiled from Fortune.com
The vultures are circling-the U.S. could lose one of its most important manufacturing assets. Qualcomm, ARM, Apollo, and others have been looking for ways to squeeze profits out of Intel after last quarter's disastrous earnings report.
Intel, the world's most valuable semiconductor company for 30 years in a row, now trades below book value. At current prices, it's a tractable proposition for many tech companies, even much smaller ones, to buy Intel. But where Intel's factories, designers, and intellectual property go is critical.
Intel is the only major U.S. manufacturer of advanced logic semiconductors, even if it's no longer a leader. During the pandemic, we learned that semiconductor shortages can be devastating to the economy. Moreover, we need to develop cutting-edge capacity to stay ahead in many advanced computing and defense-related technologies, including artificial intelligence. Most potential buyers of Intel will likely focus on cost-cutting and see little value in money-losing manufacturing subsidiaries, known in the industry as foundries. In other words, profit-seeking buyers can't be counted on to maintain U.S. manufacturing capabilities.
If Intel had world-class capacity, the whole world would benefit. TSMC and Intel have been competing for chip leadership for 30 years. Until seven or eight years ago, Intel was winning this battle. While this never-ending competition has made the world richer and must continue, dependence on a strategic manufacturer with a state-of-the-art factory in Taiwan is a global risk. In semiconductors, process R&D and cutting-edge manufacturing must be co-located. Therefore, TSMC will never move its most advanced technology to the United States because the cost is too high, too much infrastructure is required, and too many key employees must relocate. America needs Intel.
Intel management
Intel management and the board must take responsibility and act decisively to stop the bleeding. The actions they have taken so far are not enough. For years, many Intel watchers (including ourselves) have argued that Intel must separate its foundry/manufacturing business from its design business. Within Intel's corporate structure, Intel's foundry business has little chance of success.
Nvidia, Qualcomm, Broadcom, and others are eager to acquire a second foundry other than TSMC, but they hesitate as long as Intel competes directly with them. Samsung, the only advanced chip maker, has similarly found that many chip design companies, such as Apple and Nvidia, tend to avoid its foundry as a potential competitor. Intel's management has also failed to prove that it can effectively operate a foundry. Intel has provided foundry services to the industry for 20 years but has never built a successful business. Failure to meet goals and deadlines and management turnover do not inspire confidence.
Intel CEO Pat Gelsinger is a true technologist who has played a major role in the company's illustrious history. Now he faces a difficult decision: whether to break up this iconic company. He has announced a plan to establish Intel Foundry as an independent subsidiary within Intel. But that is not enough. Sentiments aside, the path for the US and Intel should be clear.
Since Intel's design business is still profitable, it needs to sign a long-term supply contract with a newly formed foundry independent of Intel. Just as AMD separated its manufacturing business and formed Global Foundries in 2009 and signed a long-term supply agreement, the new Intel Design Company needs to partially cover the losses of the foundry and guarantee sales for a few years.
Intel's design company alone cannot support an independent foundry. However, Intel's manufacturing business is the only hope to maintain the most advanced nodes on US soil. An independent foundry will be open to all US, Korean, Japanese and European companies to accumulate sufficient production and ensure its commercial viability.
Since this is a public good (all of Intel's current competitors and customers, as well as US and global consumers, will benefit), the US government (in cooperation with allies) can and should play a key role.
The CHIPS Act provides the US government with $39 billion in grants to revitalize US semiconductor manufacturing. The government has promised (but not yet paid) up to $8.5 billion in grants and $11.5 billion in low-cost loans to Intel. Now Intel risks becoming the Solyndra of this administration. This would be disastrous for both the government and Intel. The government has the power to force Intel onto a better path—it must now use that advantage.
Without industry, there can be no industrial policy: the government should be very clear about what it is willing to fund, including Intel's corporate structure. This means the government should insist on separating Intel's design and manufacturing operations into two truly independent companies.
Time is not on our side. Intel lost its lead in less than a decade-and it will take at least five years to get back on track. The pace of change in the chip industry requires management and government to act quickly. TSMC is not slowing down. The longer we wait, the less competitive we will be.
View more at EASELINK
