South Korea Reduces New Transmission Towers, Shifts to Underground Lines for Grid Expansion
The South Korean government plans to significantly adjust its approach to grid expansion: the number of new transmission towers will be reduced by up to 40%, with more new transmission lines being laid underground to accelerate the overall speed of grid expansion.
The direct trigger for this adjustment is the continuous rise in electricity demand from data centers and semiconductor factories—these two types of major electricity consumers have much higher requirements for power access speed and stability compared to ordinary residents and commercial users. Traditional overhead transmission lines have long been constrained by land acquisition and resident opposition along the route, resulting in lengthy approval cycles. Underground lines can bypass many ground land acquisition coordination steps, becoming an alternative solution to speed up access.
The unit cost of underground lines is typically significantly higher than that of overhead lines, which is a well-recognized rule in the transmission industry. South Korea is clearly opting for "higher unit costs" in exchange for "faster construction speed and lower ground resistance," representing a proactive concession in cost structure rather than a simple technical upgrade choice.
Data centers and semiconductor manufacturing are the direct beneficiaries of this round of grid acceleration—South Korea is a core production hub for global memory chips, with major foundries like Samsung Electronics and SK Hynix located domestically. Coupled with the recent surge in AI computing power demand driving the construction of new data centers, these facilities have a much more urgent need for power access than what previous grid planning cycles could accommodate.
Reducing the number of transmission towers means less land occupation and visual resistance along the route in the future, but underground lines involve more complex geological exploration, municipal coordination, and longer construction periods for each segment. Whether the goal of "accelerating expansion" can truly be achieved depends on whether the capacity for underground engineering can keep pace with the expansion rhythm.
Mechanically, the direct beneficiaries of the grid acceleration are data center operators and semiconductor manufacturing companies, as the reduction in waiting time for power access directly impacts the production rhythm of their new capacity. The pressure falls on the entities responsible for grid construction and operation, which need to absorb the much higher construction costs of underground lines compared to overhead lines. This portion of the cost is likely to be passed on to electricity consumers and residents through electricity prices or special power infrastructure fees in the long term. For the transmission equipment supply chain, this represents a structural shift in demand—from traditional transmission towers and overhead wire suppliers to underground cable and utility tunnel engineering equipment suppliers.
Source: Public Information
ABAB AI Insight
This is not the first time South Korea has exposed the contradiction of "manufacturing expansion outpacing grid access speed" in its national grid planning—over the past decade, there have been multiple complaints in the South Korean industry regarding major semiconductor expansion projects such as Samsung Electronics' Pyeongtaek foundry and SK Hynix's Yongin industrial park, where factories were completed but grid access lagged behind. The lagging grid planning behind manufacturing expansion is a long-standing structural issue in South Korea's power system.
This adjustment shifts marginal funding for grid investment from "overhead tower lines" to "underground cable lines," essentially converting the time costs originally consumed in land coordination and resident negotiations directly into capital expenditure—spending more money for faster approvals and construction cycles. The implicit flow of funds is directed towards power engineering contractors and underground cable manufacturers, rather than traditional transmission tower manufacturers.
Similar logic has also appeared in the grid construction of Singapore and the Tokyo metropolitan area: high-density, high land value areas generally prefer to use underground utility tunnels instead of overhead lines to reduce land acquisition friction, even though the unit cost is several times higher. However, the shortened approval cycle is more valuable for time-sensitive industries (semiconductors, data centers). South Korea is currently in a transitional phase where grid infrastructure is shifting from "coverage-driven" to "industry-timeliness-driven".
Structurally, this represents a reconstruction of the industrial chain: semiconductors and data centers are evolving from mere "electricity consumers" to forces that "drive the priority and technical route choices of grid construction." The mechanism is that when the electricity consumption growth rate of a single type of major consumer far exceeds the conventional planning rhythm of the grid, planning departments are forced to shift from "designing at the lowest cost uniformly" to "customized design according to industry timeliness requirements." The logic of power infrastructure construction begins to obey the rhythm of industrial expansion, rather than the other way around.
ABAB News · Cognitive Law
Whoever cannot afford to wait for capacity will rewrite the infrastructure rules.
Slow approvals are more expensive than high costs.
The grid follows computing power, not the other way around.