Heterogeneous public preferences for undergrounding high-voltage power transmission lines: The case of Seoul metropolitan area in South Korea

Local residents generally do not prefer the installation and operation of power infrastructure because of the perceived economic, health, environmental, and landscape risks associated with it. This study investigates public preferences for undergrounding high-voltage power transmission lines, which is a popular option to cope with conflicts related to transmission infrastructure. Undergrounding refers to burying overhead transmission lines underground. However, undergrounding projects are costly, making it difficult for local governments to conduct by themselves. Therefore, national-level support is required to conduct undergrounding of transmission lines. When national funds are used, considering heterogeneous public preferences is important, as the public may have a polarized preference towards the policy. This study utilized the choice experiment and consumer choice model to investigate public preferences for priorities and details of the high-voltage transmission line undergrounding policy. The latent class logit model, which incorporates preference heterogeneity by grouping the samples into a few classes, was utilized to investigate heterogeneous public preferences. Consequently, four distinct classes with different preference structures were identified. The results show that strong heterogeneity was observed, especially in the priority area for undergrounding. Moreover, heterogeneous public preferences were associated with the key demographics of respondents and perceptions towards transmission infrastructure. •We investigated public preferences for transmission line undergrounding policy.•A discrete choice experiment and latent class logit model were used for analysis.•The latent class logit model identified four classes with heterogeneous preferences.•Polarized preferences for undergrounding policy were observed among the classes.•Maximizing the total undergrounded length may not be very effective.