Social and technical differentiation in smart meter rollout: embedded scalar biases in automating Norwegian and Portuguese energy infrastructure

Energy transitions, defined as socio-material shifts in energy systems, have uneven effects across scales. This study argues that these uneven scalar effects are amplified at the intersection of energy transitions and automation, such as with smart meters. Actors at lower scales (e.g., households) tend to be responsibilised for social aspects but lack control, while higher-scale actors (e.g., utilities) control technical aspects. This research investigates these scalar biases through a comparative analysis of smart meter rollouts in Norway and Portugal. Smart meters, as distributed energy monitors, form the foundation for automating energy infrastructure through real-time data flows, making them ideal for studying scalar biases in automation. The Norwegian case involves an 18-month urban living lab that ran concurrently with a near-universal smart meter rollout. The Portuguese case focuses on the national-scale rollout, considering uncertainties related to distribution grid concessions. The study aims to identify scalar biases characterizing the automation of socio-technical energy infrastructure and understand their implications for shaping post-automation energy futures.

Energy geographies scholarship highlights the uneven scalar effects of energy transitions, often characterized by the spatial concentration of power in urban centers. Increasing automation in energy systems, particularly through smart grids and smart meters, is a key area of concern. Smart city debates often address this at the urban scale, while regional debates focus on network infrastructure implications. Technical scholarship explores spatiotemporal concerns around energy flexibility and the evolution of business models capitalizing on data infrastructures. Smart meter rollouts, enabling improved monitoring and energy flexibility, have become contentious, prompting analyses through urban living labs and comparative national-scale studies. Despite their household-scale installation, smart meters' effects span multiple scales, automating data collection and enabling remote control of energy demand and appliances. Existing research has begun to analyze the uneven scalar effects of smart meter rollouts, but a comprehensive multi-scale analysis is still needed.

This study uses a comparative case study design to maximize differences between the Norwegian and Portuguese smart meter rollouts. **Norwegian Case:** Data is drawn from an 18-month urban living lab (PARENT project) in Bergen, involving 46 households with sub-meter monitors simulating smart meters. Data collection methods included focus group discussions, individual interviews, surveys, and a workshop. This provided insights into household behavior, perceptions, and reflections on the socio-technical and political economy of smart meters. **Portuguese Case:** Data comes from a subset of 80 interviews with energy sector experts, field observations at energy sector meetings and projects conducted over 5 months in 2017–2019. This ethnographic approach, primarily focused on energy transitions and solar energy uptake, included interviews with representatives from DSOs, regulators, government agencies, consultants, and others. The analysis considers the interaction between smart meters and solar PV rollout, and data includes site visits to smart meter pilot projects, like EDP Inovgrid. The contrasting contexts of Norway (predominantly hydropower, abundant electricity) and Portugal (diverse renewable sources, higher energy poverty rates) and the different rollout trajectories (mandatory, rapid rollout in Norway; voluntary, slower rollout in Portugal) provide a rich comparative dataset. The analysis uses a 2x2 matrix (Table 1) to examine control over and responsibility for social and technical aspects of smart meter rollouts at different scales.

Analysis of both cases reveals similar patterns regarding scalar biases in smart meter rollouts. **Norway:** Living lab participants faced challenges with installation, application usability, and data trust issues. Despite access to real-time data, interest in active engagement waned. A widespread feeling of disenfranchisement emerged, with participants perceiving a lack of influence on the rollout's modalities and future implications. The completed rollout, with its national platform Elhub enabling data exchange, raises questions about the extent to which future automation will benefit users. **Portugal:** The Portuguese rollout was viewed as contingent, with public perception of increased costs and regulatory limitations impacting the DSO's investment decisions. Supply-side advantages were evident in the Inovgrid pilot project, but regulatory restrictions limited the DSO's ability to fully capitalize on the potential of smart meters for grid flexibility. National-scale discussions focused primarily on technical aspects and uncertainties surrounding grid concessions, with little consideration for the social aspects of smart meter deployment. EDP's launch of the 'EDP re:dy box,' promoting smart meters with additional Internet-of-Things functions, exemplifies the framing of smart meters as a technical innovation rather than a facilitator of changes in power dynamics within the energy sector. Both cases demonstrate a concentration of control (for both social and technical aspects) at the national scale, while responsibility for social aspects is placed at the household scale, and responsibility for technical aspects at the national scale. Households feel excluded from influencing the rollout's direction and the design of the underlying technology.

The findings demonstrate that control for both social and technical aspects predominantly rests at the national scale in both Norway and Portugal. Responsibility for social aspects is placed upon households, while responsibility for technical aspects remains at the national scale. This uneven distribution of control and responsibility limits household agency in shaping the future of smart grids. The lack of deliberative consensus building across multiple scales and the limited recognition of the socio-technical complexities inherent in smart meter deployments are evident in both cases. The Norwegian case demonstrates a relatively smooth and effective rollout through government-industry coordination; whereas the Portuguese case shows that ambiguities may slow down deployment, yet still proceed without meaningful user involvement. This pattern indicates a prevalent top-down approach, prioritizing technical expertise and national-scale interests over user-centric perspectives. This results in a limited, pre-defined vision that perpetuates existing power dynamics.

This study reveals that smart meter rollouts in Norway and Portugal exhibit scalar biases that limit user agency and shape energy futures in ways that favor centralized control. While both countries have seen successful rollouts, the lack of user-centric design and the absence of deliberative processes across multiple scales result in an undemocratic and potentially inequitable outcome. Future research should explore alternative approaches that promote greater user involvement in the design and implementation of smart grid technologies and examine methods for ensuring equitable distribution of benefits and responsibilities.

The Norwegian case study relies heavily on data from an urban living lab, which may not fully represent the experiences of all Norwegian households. The Portuguese case study, while incorporating national-level data, could benefit from additional data at the household level to provide a more nuanced comparison. The study focuses primarily on the early stages of smart meter deployment; future research would need to consider long-term impacts and adaptation.