Impacts of digitalization on the environment
Digitalization, the process of societal change driven by the increasing use of information and communication technology (ICT), is fundamentally changing existing structures and processes in all economic and social systems, with substantial consequences for the environment. Digitalization impacts the environment in two ways:
— Direct effects refer to the environmental impacts caused throughout the lifecycle of ICT hardware: its production requires resources and energy, it is powered with electricity during use, and it must be disposed of after the use phase.
— Indirect effects refer to the impacts of applying or using ICT, which change existing patterns of production and consumption (e.g. through intelligent heating) and their environmental consequences (e.g. lower energy consumption).
There is a consensus that ICT applications (via indirect effects) have the potential to contribute to environmental protection; however, assessments of indirect environmental effects of ICT and the actions required to exploit those potentials involve substantial uncertainty. For example, results of industry studies often indicate that ICT applications have the potential to avoid more greenhouse gas (GHG) emissions (indirect effect) than the ICT sector causes itself (direct effect). Results of other studies, mostly in the academic field, agree that the increasing effects of ICT on GHG emissions outweigh the reducing effects to date, and that they cancel each other out at best. Such diverging results, driven by inconsistencies in methodological assessment approaches, make it difficult for decision makers to correctly interpret the results and take the environmental impact into account in ICT investment or policy decisions.
A time-use perspective for assessing indirect environmental effects of ICT
A systematic literature review reveals that most assessments of indirect effects focus on ICT impacts on patterns of production (e.g. GHG emissions associated with the production of paper-based books vs. e-book readers). However, ICT also changes patterns of consumption. In particular, ICT use affects how individuals use their time, with manifold consequences for the environment. For example, ICT can reduce transport through virtual mobility or increase transport by creating the desire to travel to places seen on the Internet. Analyzing the indirect environmental effects of ICT from a time-use perspective has significant potential to improve our understanding of these phenomena for several reasons.
First, individual time use, the pattern of activities individuals perform during a day, is crucial for the environmental impacts associated with lifestyles (e.g. taking a walk in the woods requires no electricity, streaming a movie does). At the same time, ICT relaxes time and space constraints of activities (e.g. e- commerce allows consumers to shop for goods from almost anywhere at any time) and thus changes time allocation and the environmental impacts associated with time use. Second, time is a limited resource for everyone due to the hard 24-hour time budget constraint per day. This phenomenon makes time a central link between different activities and their environmental impacts which can be used to model interaction among activities and among ICT use cases impacting time allocation. For example, if a researcher finds that working from home saves 20 minutes of commute time per day, he or she must also answer the question how the time saved is spent. If we add further ICT use cases (e.g. e- commerce, e-banking,) to the assessment, they again change the rules of the game in which all activities compete for the same, naturally limited resource—time. Modeling interaction among ICT use cases is key to investigating systemic ICT impacts such as fundamental changes to lifestyles driven by increasing ICT use. Third, the time-use perspective allows researchers to analyze time rebound effects, which occur when increases in time efficiency lead to an increase in environmental loads (e.g. if time not spent on commuting when working from home is spent on other GHG-intensive activities).
Recognizing these characteristics of the time-use perspective, I develop a conceptual framework for systematically assessing the impact of ICT on time use and environmental impacts using energy use as an exemplary environmental impact category. The core of this framework is that it categorizes ICT impacts on time use into two types:
— ICT impacts on activity planning and execution: for example, parallelization of activities when working while traveling on a train, avoidance of commuting when working from home, or substitution of physical shopping with online shopping.
— Systemic impacts of ICT on time use: effects which only occur through the relationships between variables in the broader system in which the ICT use case takes place. For example, the possibility to work remotely from home can influence families’ decisions regarding where to live because longer commuting distances become more acceptable (because people do not commute as often), which can lead to changes in settlement structures and as well as individual time use, e.g. for leisure or travel.
Changes in time use affect direct energy requirements through the energy used while performing activities (e.g. in the form of electricity for cooking or fuels for transport). Indirect energy requirements, the energy embedded in goods and services used to perform activities, only change if production of goods and services can be avoided (e.g. if working from home leads to fewer cars being purchased or less office space being built). From a time-use perspective, the energy impacts of ICT use depend on the direct and indirect energy requirements of the activities before and after adoption of an ICT use case.
Demonstrating the time-use approach with the ICT use case telecommuting
I demonstrate the time-use approach by showing how time-use data can be analyzed and linked with data on the energy requirements of activities to assess the energy impacts of a change in time allocation using the example use case telecommuting. Telecommuting means substituting physical presence in the employer’s office with virtual presence and remote access to data, e.g. by working from home or from a local co-working space and thereby reducing commute time and the related energy consumption. Telecommuting is subject to time rebound effects. That is, reducing commuting allows telecommuters to spend the commute time saved on travel for other purposes and non-travel activities such as leisure, which are associated with their own energy requirements.
I apply the time-use approach using time-use and travel data collected in an actual co-working living laboratory in Stockholm, Sweden. I find that people spend the commute time they saved mainly on non-travel activities (e.g. leisure or chores) and only to a small extent on ‘private travel’. This substitution can lead to a reduction in net energy requirements because travel (especially individual motorized travel) is associated with higher energy requirements than most non-travel activities.
However, the size of the time rebound effect of telecommuting depends on the marginal energy requirements of the substitute activities, i.e. the energy impacts of a change in time use, which are difficult to predict. For example, spending more time on car travel directly increases fuel consumption; however, spending more time on house cleaning only increases direct energy requirements if energy- consuming appliances are used longer (e.g. vacuum cleaners, stoves). Plus, the time rebound effects of telecommuting depend on the transport modes because transport modes differ significantly in their energy requirements. For example, car commuters can realize high energy savings through telecommuting because car travel is highly energy-intensive. In contrast, for bikers or pedestrians, the direct energy requirements of travel (and telecommuting-induced energy savings) are zero, and thus the effect of any additional energy required for substitute activities is to increase net direct energy requirements.
A second case study of the co-working living lab in Stockholm broadens the scope by investigating environmental impacts of telecommuting beyond impacts due to changes in time allocation. It shows that besides time rebound effects, telecommuting can cause further environmental effects. For example, working from a co-working space can lead to an increase in office space (e.g. due to the co-working space in addition to the employer’s office space) and energy required for heating, cooling, and lighting the space.
Whether telecommuting brings about energy savings depends largely on telecommuting-induced changes to:
(1) telecommuters’ time spent in transport and use of transport modes,
(2) spacerequirementsatallworklocations(employeroffice,co-workingspace,andhomeoffice
space),
(3) substitutetravelandnon-travelactivities,goods,andservicesandtheirenergyimpacts(time
and income rebound effects).
Thus, telecommuting does not necessarily lead to energy savings, but should be accompanied by additional energy savings measures. Organizations adopting telecommuting or providing telecommuting services (in particular co-working space providers) should advise telecommuters concerning their preferences regarding work location and transport modes. All stakeholders should work together to find strategies to reduce the total office space required. If all actors adopt such measures, telecommuting can be a viable ICT application to reduce the environmental impacts of work, relieve pressure on transport systems, and increase the well-being of workers. However, if organizations and telecommuters do not address these energy-saving measures, additional energy required for space heating and cooling, a possible change in transport modes used, and time and income rebound effects can compensate or even overcompensate for commute-related energy savings. The fact that a large number of employees can work from home during the COVID-19 pandemic is an impressive example of the social benefits of flexible work models.
This dissertation shows that the time-use approach is a useful—if not key—element of methods for assessing the environmental effects of ICT (using energy impacts of telecommuting as an example). I encourage researchers and ICT companies to apply the time-use approach in combination with other production- and consumption-focused approaches to shed light on indirect environmental effects of ICT from various perspectives and to identify pathways for aligning digitalization with environmental protection.