Industrial policy must be grounded in facts and expertise, not armchair opinions

A sudden proposal to relocate the “Yongin Semiconductor Cluster,” which holds the fate of Korea’s semiconductor industry, to the Honam region has swept through the political arena. Even though it has already been finalized as a national strategy, with tens of trillions of won in capital committed and site preparation in full swing, revisionism driven by political calculations is rearing its head. At the center of this dangerous claim is Lee Bong-ryeol, who calls himself a semiconductor expert based in Singapore. He insists that “technology can be implemented anywhere, and Honam, with its abundant renewable energy, is the optimal location,” misleading the public and politicians alike.

But to be clear, semiconductors do not belong to the realm of politics, where everything is run through a vote counter. They belong to the unforgiving world of physics and engineering, where even a nanometer of deviation or 0.001 seconds of power instability is unacceptable. At a time when plausible rhetoric from non-experts is threatening to shackle a hundred-year national agenda, we must confront cold, emotionless facts and industrial logic.

The biggest flaw in Lee Bong-ryeol’s argument is that he tries to lump all technologies together under the single word “semiconductor.” The STMicroelectronics Singapore fab where he works mainly handles mature or legacy nodes in the 90nm to 180nm range, such as power management ICs (PMICs) and microcontrollers (MCUs). These are technologies that were already fully established 20 years ago, where the core task is to operate equipment and manage productivity according to fixed manuals.

By contrast, the plants that Samsung Electronics and SK hynix are building in Yongin will run 3nm-class, ultra-fine advanced nodes. These are not mere production facilities. They are massive R&D convergence hubs that push up against the limits of physics, such as quantum tunneling effects, and where thousands of PhD-level engineers must constantly tweak and redesign hundreds of chemical-material recipes in real time to tame tens of thousands of new variables that arise every single day.

What is even more concerning is the speaker’s actual expertise. According to internal sources at the company, Lee Bong-ryeol is not a process architect who leads process design, but production staff who operate established processes. The company may grant the title of “Engineer” even to production roles, but researching the “why and how” of a process is fundamentally different from carrying out the “what.” Using experience in “following manuals” on mature nodes as a basis to argue over “site selection” for cutting-edge nodes is like using experience baking bread at a neighborhood bakery to lecture on where to locate and how to design a state-of-the-art molecular gastronomy lab.

The claim that “if the conditions are good enough, talent will go even to hell” is the product of sheer ignorance about how semiconductor innovation actually happens. Innovation does not spring from explicit, documented knowledge; it emerges from tacit knowledge that is shared and refined as people work and struggle together.

The “K-Semiconductor Belt” that runs from Pangyo (design) through Giheung and Hwaseong (memory/foundry) to Pyeongtaek is not just a string of factory sites. It is a vast intellectual ecosystem where fabless designs are implemented in fabs, equipment from materials and component suppliers is installed on the lines, and university research is deployed on the shop floor. The single reason Nvidia and Apple refuse to leave Silicon Valley despite punishing living costs and taxes is that an irreplaceable density of talent has formed there.

What would happen if the Yongin cluster were forcibly relocated to Honam for political reasons? Top-tier talent would not be packing for Honam; they would be booking flights to the United States or Japan. Yongin is the geographic red line our companies can realistically offer to keep global talent anchored here. Ignoring this is effectively pushing companies to “forfeit the global war for talent,” a self-destructive act in all but name.

An executive at an AI memory startup, responding to our questions, summed it up this way:

The argument that “there are many solar farms in Honam, so let’s move the fabs there” ignores the basic principles of the power grid, namely frequency synchronization and transmission efficiency. A semiconductor fab is a precision facility that requires voltage and frequency to remain razor-sharp and constant 24 hours a day, 365 days a year. Trying to connect solar power—whose output fluctuates wildly with the weather—directly to a plant without routing it through the national grid, which acts as a massive buffer, is a technical act of suicide.

The RE100 policy Apple is trying to achieve is not about putting wind turbines or solar panels right next to its factories. It is achieved by purchasing renewable energy generated in remote locations through an advanced power grid via power purchase agreements (PPAs), or offsetting through certificates. In other words, the key is not physical location but the network.

The scientific solution for using Honam’s surplus renewable energy is not to relocate factories that cost hundreds of trillions of won, but to expand an ultra-high-voltage direct current (HVDC) transmission network that can carry that electricity to the Seoul metropolitan area where the demand is. Blocking the construction of transmission towers, whose safety has been proven, with unscientific “electromagnetic wave scare stories,” and then proposing factory relocation as an alternative, is not just the tail wagging the dog—it is the tail trying to break the dog’s back.

Fact-Check Sheet

The semiconductor cluster is not a tool for advancing the political agenda of balanced regional development. It is the only ark of survival for the Republic of Korea amid the US–China tech hegemony war.

What we need now is not to take dictation from half-baked arguments made by non-experts. We must listen to the engineers on the ground who stay up all night fighting the limits of physics and pushing for yield improvements, and to the companies struggling in global markets. The government and political circles must not be swayed by the agitation of so-called “expert spokespersons,” but instead concentrate the nation’s capabilities on the essential tasks of successfully establishing the Yongin cluster and expanding the transmission network in a timely manner.

History warns us in no uncertain terms that when politics contaminates the realms of science and industry, the cost ultimately falls squarely on the public.