Many people now work partly or permanently from home. This saves petrol because they no longer have to drive to the office—which is good for the energy transition. Or not. Because at home, they consume even more energy for heating and cooking, and video calls also consume extra energy in the data centers of Microsoft, Zoom and the like. Does digitalization now save energy or are we just blowing more CO₂ into the atmosphere?

Evangelos Panos' answer gives us hope. Together with former Paul Scherrer Institute (PSI) doctoral student Lidia Stermieri and researchers of ETH Zurich, the researcher from the Energy Economics Group in the Lab for Energy Systems Analysis at PSI has shown in an elaborate calculation model that "digital lifestyles" will actually save 10% to 20% in energy in 2050 compared to 2020.

Rebound effects—savings in one place that lead to more consumption elsewhere—are more than compensated for by more efficient technologies and changes in behavior. "Digitalization does not solve all problems," says Panos, "but it supports the transformation of the energy system on the way to net zero emissions." The researchers report on their findings in the journal Energy Policy.

Socio-economic decisions taken into account

For her calculations, Stemieri used STEM (Swiss TIMES Energy Systems Model), which has been developed over years of painstaking work at Energy Economics Group into a powerful model for simulating an energy system. Six million equations with six million variables represent the Swiss energy system and the interactions of technologies, energy and emissions carriers, and sectors.

The variables are linked to each other in many dimensions, also across time and space, via the equations. This allows various scenarios to be calculated up to 2050, i.e., up to the year in which Switzerland aims to be carbon-neutral, or even beyond.

"It is the only model that can model development paths for Switzerland's entire energy system over long periods of time and with a very good temporal resolution and a very detailed representation of all energy system sectors," says Panos.

However, STEM only models the techno-economic side of the energy system and does not provide any information on socio-economic factors. For example, how people make decisions about their energy consumption. People who only work from home may no longer buy a new car. Though they might buy a heat pump at home, but only if they are the owner of the property and not a tenant. Many interwoven considerations play a role here, which together can have a significant impact on the energy system in the long term.

This is why Stermieri has docked another model onto STEM: SEED (Socio-Economic Energy model for Digitalization). SEED maps the heterogeneity of the decision-making processes of actors in households, parts of the service sectors and industry in order to analyze synergies and interactions in the introduction of energy-saving digital services and practices.

These socio-economic decisions are exchanged via an interface with STEM. In combination, this has resulted in one of the most advanced models for energy systems with a national scope in Europe.

Different assumptions and their scenarios

Stermieri used her model to calculate and analyze two possible scenarios. In the "Frozen" scenario, she assumes that society shows no interest in advancing digitalization and that it therefore stagnates. In contrast, digitalization accelerates in the "Accelerated" scenario. Society shows a positive attitude towards digitalization and technological development, supporting new digital social practices, progresses accordingly. This also shows the positive influence on energy consumption.

Stermieri needed to run her model 20 to 30 times for each scenario, which takes a fast PC around 60–90 hours of intense computations. During each of these runs, the two energy systems interacted with each other several times so that the interplay between society and technology could be realistically modeled. "We're not predicting the future," emphasizes Stermieri, "we're exploring it by calculating scenarios based on the what-if principle."

Even if none of these scenarios will materialize in exactly the same way, they are very valuable as a decision-making aid for those responsible in politics and business. "The scenarios show possible courses of action and their consequences. This is particularly interesting when comparing different scenarios," says Stermieri.

Future work to integrate psychological effects

Stermieri's model covers all sectors—households, industry and politics—and not just individual technologies. "This makes it unique in Europe," says Panos. It also shows the development from the simple models from the 1980s, which initially only calculated the energy sector in a simple way, to models that also simulate the interaction of different technologies.

The current model also includes socio-economic aspects and is therefore more interdisciplinary than ever before. However, even this is only an intermediate step towards even more realistic models, which will also consider findings from psychology in the future.

Panos is working on this with doctoral student Shadi Firoozyalizadeh in a sub-project of the CoSi research program (Co-Evolution and Coordinated Simulation of the Swiss Energy System and Swiss Society). The project aims to investigate the social change processes that are necessary for the transformation of the energy system.

In the project, psychologists are investigating, among other things, how people make decisions about their energy consumption and technology choices. Firoozyalizadeh will then integrate this decision-making psychology into the SEED model.

Panos is certain that this generation of models will lead to even more informed decisions when designing a new energy and climate policy. "Policymakers will have a unique tool to support them in their decisions that takes greater account of social aspects, because technology alone is not the solution."