Is fabs manufacturing never a dream for the US soon?
US was the first-ever country to manufacture semiconductors. It started with the invention of the transistor by Shockley in the 1940s. This invention gave rise to the early leaders like Fairchild Semiconductor, which produced top companies like Intel. In those days, when every semiconductor company made their own devices and fabs cost a few million dollars, many manufacturing sites evolved in the silicon valley.
In the 80s, the US lost its leadership in the industry after Japan started investing in fabs. Japanese applies their quality and manufacturing segments like memory chips, pushing Intel to move from memory chips to microprocessors in the mid-80s. However, the more significant shift happened in the 90s through the appearance of factories and fabless design companies. Companies like TSMC and UMC lead these fabless companies, and they caused many of the smaller scale fabs to be no longer cost-effective.
China decided to invest in semiconductors in 2014 and began implementing an ambitious “Big Fund” $20B+ initiative, making China a significant player in the global market share. Around 10% market share to a global giant taking on the Koreans, Taiwanese, and Americans. In 2019 they again announced a “Big Fund II” that doubled the size of the previous fund which they received.
In addition to that, local government entities also invested funds in local projects, and state-owned banks make cheap capital available. Now, China’s IC output the same as the US at around 12% of global production.
Chip Manufacturing In The U.S Now
In America, most companies dropped from the manufacturing race after the outsourcing. There are only three well-known companies left: Intel, Micron, and Global Foundries (GF). Each has its niche and do not compete with each other. Intel is an IDM (Integrated Device Manufacturer) and produces microprocessors for PCs and Servers. Micron manufactures memory chips (NAND and DRAM), and Global Fundaries is not even a genuinely American company privately owned by Mubadala, the UAE’s sovereign wealth fund, it is a contract manufacturer for fabless chip companies. But Chip manufacturing is a huge global business, and all the three companies operate global fabs. Intel routinely rotates new fabs between the US, Israel, Ireland, and Micron operate fabs in Japan and Taiwan, and the latest substantial investment was in Singapore.
What is the potential impact of this, and can it change the semiconductor fab landscape meaningfully? How will it affect the economy?
Let’s have a look at the important elements involved in setting up a Semiconductor Fab.
1. Sustained Capital Investment: A “state of the art fab” costs about $10B. The total investment of Semiconductor manufacturers in 2019 totaled $102B. Companies must bring to market new technology nodes every two years, according to Moore’s Law. On this basis, a one-time investment of $23B from the US is a critical decision to face the global competitors. Also, this decision impacts on tax code via credits, tax breaks, etc.
2. Engineering Talent: US Universities produce a vast number of engineering graduates. The US remains a desirable destination for world talent. Staffing new fabs should not be an issue.
3. Infrastructure: Land, water supply, A stable power grid, transportation infrastructure are all essential. From this perspective, the US being a developed country is not lacking.
4. Supply Chain: The supply chain for Semiconductors is as global and crucial as the end product. It is challenging for a country to be fully self-sufficient in manufacturing. The US is well placed for some fundamental building blocks like production equipment with the largest suppliers, LAM, AMAT, and KLA being US companies. But even there, the Dutch company ASML competes on the critical photolithography steps. On the materials side, it is also tricky as the manufacturing process uses hundreds of gases and chemicals, many of which have to source from Asia. So right end to end self-sufficiency for any country is a dream.
Who wants to build the new Fabs in the US?
Intel has continued aggressive technology development, and fab builds from the past centuries. With some additional incentives, it is likely they would choose US sites rather than Ireland or Israel for future investments. Not surprisingly, Intel has been lobbying actively for the US government support of the industry. But earlier, Intel tried the foundry model and not made any advancement. Recent announcements citing more delays in their 7nm process also indicated their willingness to use outside partners for manufacturing. The US must show similar long-term commitment to Federal and State support. But this isn’t easy in our political system.
Since only the top 5 IC makers have the money to make leading generation fabs, any policy must be aligned with the strategic objectives of these companies. Some of that could do through punitive measures like tariffs and trade actions, but this is rarely sustainable in the global economy.
It probably makes more sense to focus on a small subset of critical chips that have real national security implications and continue to fund them as necessary. But a massive scale shift of IC manufacturing back to the US doesn’t seem a likely scenario.
