The Room Where It Happens
Discerning chip policy through RFIs and 5-Year Plans
🎶 Why do you write like you’re running out of time? 🔶
🎶 Write day and night like you’re running out of time? 🔶
This week’s article breaks from Chip Capitols’ typical coverage to take a peek behind my desk.
Friends and colleagues have been asking how I write my deep-dive pieces.
It goes without saying that I find the semiconductor industry to be both fascinating in its own right and a lens through which to examine the issues of our times. A barrier for people outside the chip world, however, is that the industry’s dynamism lies in its immense complexity:
The technology is impossibly complicated. (I am not an engineer.)
The supply chain and business strategies are unparalleled. (I do not have an MBA or PhD in industrial organization.)
The political issues cover diverse legal areas and societies. (Unlike Hamilton, I am not –yet– a lawyer. Nor am I China expert 中国通.)
Despite not being any of these things, I know where to look for the knowledge I lack. Many thanks are due to my former colleagues in DC for teaching me how to teach myself, and Chip Capitols is an exercise in further developing these skills.
In this article, I will share some of the lessons I have learned so far in hopes of helping others doing this sort of research.
A Young but Crescendoing Ride
Before diving into research methods, let’s review what Chip Capitols has been up to in the four months since founding. My articles generally fit into one of four buckets:
Global Chip Subsidy Comparisons: These popular deep-dives compare the US’s CHIPS and Science Act to subsidies in other major semiconductor-producing regions. Knowing where my expertise is strong and where it is not, I write these articles in collaboration with friends of mine who are from the respective regions.
Chip Capitols has so far done a China-US comparison and an EU-US comparison. Next up are South Korea, Japan, and Taiwan.
Public-Private Chip Consortia Studies: Public-private research consortia have played critical roles in the chip industry’s development, and each nation takes a different approach to structuring their consortia.
Chip Capitols has analyzed how different consortia address issues ranging from infrastructure geography to IP management.
Chinese Media Analysis: The Chinese Communist Party’s official newspaper, the People’s Daily 人民日报, often features articles by senior officials laying out official views on semiconductor related topics. I translate these on alternating weeks and offer a short analysis.
Chip Capitols has covered topics including a Chip Promotion Law proposal at China’s Two Sessions, the pillars of Chinese science policy, and Chinese critique of the quadrilateral chips alliance.
Other Themes: Countries do not spend billions of dollars on the chip industry willy-nilly. I enjoy examining the external policy goals shaping chip programs in order to better understand the political context of each country.
Chip Capitols has published articles examining export restrictions countries place on chipmakers in return for subsidies and the role of chips in countries’ green transition plans.
Now that we’re on the same page regarding what Chip Capitols is all about, the next sections will dive into where this content comes from.
Learning About Chips
Requests for Information (RFIs) are a veritable treasure trove of information. It is a shame that more young researchers are not aware of this resource.
To quote a high school AP US Government professor’s description of the federal rule-making process:
Publication in the federal government’s newspaper [the Federal Register] marks the beginning of what is known as the “notice and comment” period in which all parties affected by the proposed regulation are encouraged to make their own opinions known to the agency.
When a US government agency prepares to act on a responsibility granted to it by Congress, it publishes a notice (often in the form of an RFI) where all interested stakeholders can submit public comments. For any given RFI, nearly every relevant trade association, company, and individual will submit a public response.
Critically, these responses summarize every resource the entity believes supports their worldview. By looking at just a few of these responses, I can quickly understand the different sides on a given policy issue, the technical facts, and the relevant sources.
For example, let’s look at my first Chip Capitols article, NSTC Geography and the Proxy Battle for Federal Research. This piece demonstrates the power of RFI research very well. In one section, I say:
In theory, a program like the NSTC could be fully centralized, fully distributed, or follow a hybrid model.
Academics, federal labs, and the most R&D intensive semiconductor companies (namely, fabless design firms) advocate the most centralized version of the “hybrid” NSTC model…
Industry stakeholders, namely semiconductor manufacturing companies and their affiliated organizations, advocate a very different approach… The manufacturers’ philosophy is captured by TSMC’s position, which calls for the NSTC’s Centers of Excellence to feature both specific focus areas and economically self-sustainable business models.
My citations in this passage link to responses by chipmakers and coalitions to a January 2022 Request for Information from the US Department of Commerce. After identifying divergences between the submissions, I developed the article’s narrative arch and cited the RFI responses as my sources. Critically, the RFIs explain both the relevant business points (research “stacks”) and technical points (chip industry “verticals”).
Learning About the World
Perhaps the most valuable thing Chip Capitols puts out is the rapidly growing collection of articles comparing global semiconductor subsidies. These pieces highlight deeper themes about politics and economics, using chip subsidies as a conduit.
Drawing such themes requires deep understandings of both countries discussed in a given article. To that end, I work with coauthors who can challenge assumptions I make in developing each piece’s overarching theme. They know their countries, I know mine; together, we can say smart things.
Beyond the ongoing comparison series, I regularly research global semiconductor policies for Chip Capitols’ other articles. Not having a local coauthor for each piece, I have developed different strategies for each region.
Also, some readers may ask: does language ability matter in a world with automatic translators and generative AI? Yes. Yes, it does.
When researching a topic as specialized and fast changing as semiconductors, generative AI does not have enough data for each country. You need to know where to search, and knowing the relevant language makes searching significantly faster. Perfect fluency is not strictly necessary, but not knowing a single word in the relevant languages makes my coverage of Korea, Japan, and Germany significantly poorer.
Researching China🇨🇳
The PRC’s completely different political and economic systems make Chinese chip policies a wonderful foil for US policies.
Knowing the local language is critical for conducting high-quality research on Mainland China’s chip policy. Government documents are often only available in Mandarin, and think tank papers use terms of art that do not translate well into English. A recent article of mine, ”Chip Promotion Law” Proposal at China’s Two Sessions, is illustrative:
[In proposing that the PRC adopt a Chip Promotion Law, a prominent CCP member] suggests applying a relatively new concept in Chinese science policy to national semiconductor research goals, jiēbǎng guàshuài (揭榜挂帅). This roughly translates to “revealing the lists and taking command.” Local governments and the central government have embraced this strategy of publishing lists of technology needing to be developed and awarding innovators who bring solutions to market.
Reading an auto-translated version of this article, one would miss the deeper meaning of jiēbǎng guàshuài. Fortunately, while (very slowly) reading the original Chinese, I realized this term was unusual, looked it up on a Chinese search engine, and learned its context.
To be clear, I am far from proficient in Mandarin. My Chinese language skills are at the lower end of HSK 5 (upper-intermediate), but I am devoting a lot of time to improving. Despite not yet being fluent, my novice Mandarin is still an enormous help in researching the PRC’s chip policy. It allows me to:
Do keyword searching in Chinese to find primary source material,
Skim articles and statutes to pick the most important parts to translate fully,
Not rely on often-shoddy English-language sources.
Lastly on the China front, what sources do I use? I find it most helpful to go top-down by 1) starting with central government and Party plans, 2) moving on to ministerial-level implementation plans, and 3) ending with announcements of projects funded by those plans. Last week’s article, Chip Policy for the Green Transition, is illustrative:
CENTRAL GOVERNMENT LEVEL—
After years of focus on shrinking the footprint of compute, the 14th Five-Year Plan’s National Informatization Plan took a major step in calling for the expansion of chips’ handprint.
MINISTERIAL LEVEL—
Last year featured the first notable implementation of China’s 14th Five-Year Plan’s National Informatization Plan… Six administrative departments issued the Industrial Energy Efficiency Improvement Action Plan (2022).
PROJECT LEVEL—
The National High Technology Development Program (the 863 Program) has played a critical role awarding funds for developing semiconductor technologies in a range of energy efficiency applications. One 863 Program funding opportunity offered ¥80 million (11.6 million USD) for projects researching high voltage, silicon carbide semiconductors…
Researching the European Union🇪🇺
Researching EU chip policies poses different challenges as they cannot be understood in a vacuum. They function hand-in-hand with member-state policies, so researching national semiconductor programs is critical to fully understanding what is going on in Europe.
I’ve studied French for years and am fairly proficient, so I can research France’s chip programs easily. Government documents and French think tanks’ reporting explain the relationship between national programs and EU-level frameworks. This passage from last week’s article, Chip Policy for the Green Transition, is again illustrative:
EU member states’ public research programs for semiconductor applications in the energy sector function primarily as Important Projects of Common European Interest (IPCEIs)… A 2018 IPCEI proposed by France, Germany, Italy, and the UK (then an EU member) demonstrates how European governments conduct research into green chip technology…
Having contributed nearly €900 million to the joint IPCEI with Germany, Italy, and the UK, France launched the Plan Nano 2022 to support semiconductor research that could feed into automotive, IoT, and other increasingly digitized downstream applications.
Unlike France’s permanent national research agency, the Agence nationale de recherche (ANR), Plan Nano works closely with industry to fund joint research. This is legally possible due to the state-aid exemptions provided by the EU’s IPCEI framework…
Unfortunately, European countries outside the French lingo-sphere are harder for me to examine. For the same green chip policy article above, I originally also wanted to analyze how Germany implemented the IPCEI framework, but I ran into a wall when critical sources were only available in German and as PDFs. I ran the doc through a translator, but the English was still confusing and referred to concepts that I am simply unfamiliar with.
Chip Capitols’ EU-US comparison article was possible because I collaborated with a German classmate of mine. He could not only read German sources but could also explain the EU’s and member states’ domestic contexts. Such collaborations are enlightening but not available for every project.
Why do you write like you’re running out of time?
My goal with Chip Capitols is learning for learning’s sake. It is a way to share knowledge with other smart people around the world about an industry I find fascinating. There are few greater joys than when I receive comments from you responding intelligently to what I write.
Next week, Chip Capitols will resume its regular programing as we begin to analyze South Korea’s semiconductor incentives.🇰🇷 Stay tuned for an exciting collaboration with another friend of mine as we continue to cross borders and languages in search of chips!
Bravo! Many thanks for your work. I hope you will dive into the photonic fab when it begins production this year in Beijing.