Digital sufficiency: conceptual considerations for ICTs on a finite planet [original]

Vol.:(0123456789)

1 3

Annals of Telecommunications

https://doi.org/10.1007/s12243-022-00914-x

Digital sufficiency: conceptual considerations forICTs onafinite

planet

TilmanSantarius1,2,3 · JanC.T.Bieser4,5· VivianFrick2· MattiasHöjer5· MaikeGossen1,3· LorenzM.Hilty4,6·

EvaKern7· JohannaPohl1· FriederikeRohde1,2· SteffenLange1,8

Received: 9 December 2021 / Accepted: 22 April 2022

Abstract

ICT hold significant potential to increase resource and energy efficiencies and contribute to a circular economy. Yet unre-

solved is whether the aggregated net effect of ICT overall mitigates or aggravates environmental burdens. While the savings

potentials have been explored, drivers that prevent these and possible counter measures have not been researched thoroughly.

The concept digital sufficiency constitutes a basis to understand how ICT can become part of the essential environmental

transformation. Digital sufficiency consists of four dimensions, each suggesting a set of strategies and policy proposals: (a)

hardware sufficiency, which aims for fewer devices needing to be produced and their absolute energy demand being kept to

the lowest level possible to perform the desired tasks; (b) software sufficiency, which covers ensuring that data traffic and

hardware utilization during application are kept as low as possible; (c) user sufficiency, which strives for users applying digital

devices frugally and using ICT in a way that promotes sustainable lifestyles; and (d) economic sufficiency, which aspires to

digitalization supporting a transition to an economy characterized not by economic growth as the primary goal but by suf-

ficient production and consumption within planetary boundaries. The policies for hardware and software sufficiency are rela-

tively easily conceivable and executable. Policies for user and economic sufficiency are politically more difficult to implement

and relate strongly to policies for environmental transformation in general. This article argues for comprehensive policies

for digital sufficiency, which are indispensible if ICT are to play a beneficial role in overall environmental transformation.

Keywords Green IT· ICT for sustainability· Sustainable software· Sustainable production and consumption· Rebound

effects· Economic growth· Degrowth

1 Introduction

The discourse on the environmental sustainability of using

information and communication technologies (ICT) has

become increasingly well founded, complex, and interdis-

ciplinary. Nevertheless, large research gaps remain regarding

both empirical and conceptual and theoretical knowledge

(for an overview, see [1, 2]). This paper mainly addresses

the following three research gaps.

(1) Various studies on ICT and environmental sustain-

ability have identified the potential of ICT to reduce

energy and resource inputs but do not consider impor-

tant trends that run counter to that potential and, even-

tually, limit ICT positive contributions (e.g., [3–6]).

For example, ICT-borne efficiency improvements may

cause rebound effects, which countervail parts or all of

the savings potential [7–9]. An increasing number of

* Tilman Santarius

santar[email protected]

1 Department forSocial Transformation andSustainable

Digitalization, Technical University ofBerlin, Berlin,

Germany

2 Institute forEcological Economy Research, Berlin, Germany

3 Einstein Centre Digital Future, Berlin, Germany

4 Department ofInformatics, University ofZurich, Zurich,

Switzerland

5 Department ofSustainable Development, Environmental

Science andEngineering, KTH Royal Institute

ofTechnology, Stockholm, Sweden

6 Technology andSociety Lab, Empa Materials Science

andTechnology, St.Gallen, Switzerland

7 Environmental Campus Birkenfeld, Birkenfeld, Germany

8 Resource Economics Group, Humboldt-Universität Zu

Berlin, Berlin, Germany

Annals of Telecommunications

1 3

publications analyze ICT-borne rebound effects [9–15].

However, an overarching strategy to cope with rebound

effects and with further undesired effects of ICT use

(e.g., induction effects) has yet to be presented.

(2) While many studies on ICT and environmental sustain-

ability provide valuable scientific analyses, they do not

develop comprehensive policies and measures address-

ing all relevant societal actors. For example, several

studies focus on sustainability of software engineering

but do not investigate the combination of both hard-

and software and its implications for policy-making

[16–19]. Other studies focus on a certain analytical

level, e.g., by differentiating between first-, second-,

and third-order impacts, but fail to discuss policy solu-

tions that systematically address all those levels ([1, 20,

21]).

(3) Most studies on ICT and environmental sustainability

do not account for existing sustainability strategies.

Those strategies suggest that a combination of effi-

ciency, consistency, and sufficiency strategies is most

effective in realizing absolute savings [22–24]. Hilty

etal. propose that sufficiency strategies are particularly

suited to addressing ICT-borne rebound and induction

effects [25]. However, sufficiency strategies have not

yet been further specified in the context of ICT.

Recognizing the research gaps on sufficiency strategies

for ICT use and the lack of comprehensive and actor-spe-

cific policy recommendations for environmentally sustain-

able ICT use, including how to address ICT-borne rebound

effects, this paper develops a comprehensive “digital suf-

ficiency” concept. The concept consists of four dimensions:

(a) hardware sufficiency, (b) software sufficiency, (c) user

sufficiency, and (d) economic sufficiency. Along these four

dimensions, we derive and discuss a comprehensive set

of policies for environmentally sustainable ICT use that

involves all major groups of actors: policymakers, individ-

ual users, software developers, the business sector, and civil

society. By doing so, we advance the conceptual knowledge

on environmental sustainability aspects of ICT use and pro-

vide concrete guidance for practitioners wishing to align ICT

use with environmental protection.

The article is structured as follows. In Section2, we

define sufficiency, link it to the “ICT for Sustainability”

debate, and reason why digital sufficiency is an expedient

strategy to ensure that ICT contribute positively to environ-

mental sustainability. In Sections.3 and 4, we define our

approach and the four sub-dimensions of digital sufficiency.

For each dimension, we identify challenges that prevent

resource and energy savings and suggest policy measures to

overcome them. In Section5, we summarize our findings,

highlight interrelationships of the four dimensions of digi-

tal sufficiency, and conclude with a self-critical discussion

on environmental effectiveness and social acceptance of the

suggested policy measures.

2 State ofliterature

2.1 Consistency, efficiency, sufficiency

Over the past decades of environmental sustainability

research, three overarching strategies have been developed

to reduce environmental burdens: consistency, efficiency,

and sufficiency [23, 24]. The strategy of consistency aims

at “doing things better,” namely, closing material and nutri-

ent cycles and bringing cycles of industrial production and

consumption in line with natural cycles, including those of

water, air, climate, or soil recovery. The “circular economy”

[26–28] or the “cradle-to-cradle” concept [29] are types

of consistency strategies. The strategy of efficiency aims

at “doing more with less,” namely, reducing resource and

energy inputs per unit of service or product. Innumerable

articles and studies have been published in favor of increas-

ing energy and resource efficiencies in different sectors and

domains (e.g., [30, 31]). The strategy of sufficiency focuses

on an absolute reduction of resource and energy demand

while maintaining, or even improving, immaterial living

conditions and a “good life” for all [32].

Various definitions of sufficiency can be found. Most of

them understand sufficiency as avoiding overconsumption

while reducing the use of scarce natural resources and fossil

fuel-based energy [1–3]. Traditionally, most literature on suf-

ficiency has focused on individual consumer behavior, e.g.,

rethinking personal needs and avoiding excessive consumer

behavior [35, 36]. Sufficiency-orientated behavior includes

absolute reductions of consumption, modal shifts to more

resource-efficient transport modes, product lifetime exten-

sion, and sharing practices [35, 37]. Related terms and con-

cepts have been discussed as “voluntary simplicity,” “fru-

gality,” “downshifting,” “anti-consumption,” “minimalism,”

or “slow consumption” (e.g., [38–40]). Despite the focus

on individual behavior, several authors have also applied

the concept of sufficiency to the production side (e.g., for

sufficiency-oriented marketing strategies) and to the overall

economic level of activity [41–44]. In addition, much of

what is currently discussed as “postgrowth” or “degrowth”

economics has a strong affinity to sufficiency—both because

politics for degrowth always include sufficiency propos-

als [45–48], and because the term “sufficiency economy”

describes concepts for the macro-level that are similar to

concepts for a postgrowth economy [44, 49].

We define sufficiency as any strategy that directly aims

at decreasing the absolute level of resource and energy use

by reducing the levels of production and consumption.

Sufficiency necessarily involves reflecting upon existing

Annals of Telecommunications

1 3

individual and societal needs, attitudes, and beliefs. And

it requires changes in consumption practices as well as in

production structures, infrastructures, and existing political

incentives that favor consumerism and conventional eco-

nomic growth.

Current political efforts to reduce resource demand and

emissions to sustainable levels often pursue technology-

based approaches that focus on efficiency and consistency,

and often neglect sufficiency [50, 51]. For instance, with

Sustainable Development Goal No. 12, the global commu-

nity agrees on the principle of “doing more and better with

less” [52], which refers to efficiency strategies to achieve a

win–win situation for economy and ecology. Yet, both effi-

ciency and consistency strategies ignore rebound, induction,

and growth effects and, hence, may fail to deliver significant

savings, as any relative reduction of inputs can be outplayed

by an absolute increase of output [53, 54]. Therefore, with-

out neglecting the valuable contributions of efficiency and

consistency strategies, several authors suggest that those

strategies should be accompanied—if not guided—by a suf-

ficiency strategy that ensures absolute reductions in resource

and emissions intensities [55–57]. Furthermore, several

authors overcome the ill fortune that sufficiency is some-

times mistakenly associated with attitudes such as absti-

nence or renouncement by suggesting that the sufficiency

strategy can benefit both the environment and the quality of

life [32, 33, 58, 59].

3 Environmental impacts ofICT andreasons

fordigital sufficiency

ICT environmental effects have been under discussion for

more than a decade [1, 21, 60–64]. A few studies have

attempted to calculate ICT aggregate environmental impacts

throughout all sectors [5, 6, 65]. However, those studies can-

not claim to exhaustively cover all global impacts of ICT,

and methodological developments for a comprehensive and

consistent synopsis are still in their infancy [6]. For certain

sectors or fields of applications, a large number of studies

have been conducted—documented, for instance, by the

growing number of publications in the discourse “ICT for

Sustainability” [66]. However, sector-based, micro-level, or

case study analyses obviously do not allow for conclusions

on whether the aggregated net effect of introducing ICT into

society is positive or negative [1, 67, 68].

Over the years, a taxonomy of first-order and higher-

order environmental effects has emerged [21]. First-order

or “direct effects” relate to producing ICT devices and

infrastructures and to the electricity demand from using

those digital devices and services. Higher-order or “indi-

rect effects” result from social change associated with apply-

ing and using ICT. Hilty [1], Horner etal. [21], and others

distinguish between positive higher-order environmental

effects, such as substitution or optimization effects, and

negative environmental effects, such as rebound or induc-

tion effects (see also [11, 69]). Yet, because the introduction

of ICT devices and services can be associated with a broad

spectrum of behavioral and structural changes in economy

and society, the net sum of all higher-order effects are dif-

ficult to determine.

The literature identifies four mechanisms that inhibit the

realization of positive effects and build the basis for devoting

greater attention to sufficiency. First, “direct effects” (first-

order) come with manifold environmental problems, includ-

ing extraction of scarce resources and insufficient environ-

mental standards within the ICT production process [70–72].

Although ICT can contribute to enhancing recycling and

circular material flows in other sectors [26], ICT hardware

production itself is far from living up to the premise of envi-

ronmental consistency, which would imply that devices be

made of renewable resources or fully recycled materials in

a circular economy.

Second, the volume of data storage, processing, and trans-

mission is currently increasing exponentially. For example,

data volumes are being particularly driven by big data ana-

lytics, artificial intelligence, and online video streaming

[73]. Notwithstanding the fact that energy and resource effi-

ciencies of data processing and transmission are constantly

increasing (e.g., by the shift from 3 to 4G and eventually 5G

mobile networks) and that there is a non-linear relationship

between data traffic and environmental footprint [74], past

saving potentials have so far been either fully or to a large

part outpaced by vastly increasing data volumes [75].

Third, not only hardware, software, and ICT system man-

agement have become more efficient, but also the applica-

tion of ICT in many areas of economic production and con-

sumption has improved various efficiencies, including cost

efficiency, energy efficiency, time efficiency, and behavioral

efficiency. Such efficiency improvements may help reduce

existing environmental burdens [5, 76–78], but at the same

time, they also generate rebound effects, which countervail

parts of the savings potential, if not all [7, 15, 53]. Even in

cases where rebound effects at micro-level are small, many

studies have shown that taking into account higher-order

rebound effects can be decisive for the aggregate environ-

mental impact [13, 61, 79–81].

Fourth, the introduction of ICT applications opens up new

opportunities for production and consumption and, hence,

leads to induction effects, which occur if ICT use generates

additional energy and resource demand [2, 11, 82]. Overall,

ICT-borne rebound and induction effects foster economic

growth [83], which eats up potential energy and resource

savings from optimization and substitution effects.

Note that these four mechanisms do not necessarily imply

a negative aggregate environmental impact of ICT, but they

Annals of Telecommunications

1 3

do curtail the contributions derived from efficiency and con-

sistency strategies. Hence, in the following sections, this

paper elaborates the concept of digital sufficiency in a way

that addresses each one of these four countervailing forces.

To date, only few publications address sufficiency strate-

gies in the context of ICT. Back in 2008, Hilty mentioned

the need for sufficiency strategies with regard to ICT [88].

Santarius and Lange were the first to coin the term “digital

sufficiency” and develop the concept in their book “Smart

Green World” [89]. They present three dimensions: technical

sufficiency, data sufficiency, and user sufficiency [2]. This

article is based on that work but elaborates on the concept

and adds a fourth dimension—economic sufficiency.

The “Shift Project” report has introduced the terms “lean

ICT” and “digital sobriety” [90]. Given the overall messages

of the report, these terms appear close to the concept of digi-

tal sufficiency presented here. However, the report neither

clearly defines the terms nor elaborates on what dimensions

or elements the concept includes.

4 Methodology

The conceptual considerations on digital sufficiency pre-

sented in this article have been developed in two specific

contexts. First, in 2016 the 6-year research group Digitaliza-

tion and Sustainability (www. susta inable- digit aliza tion. org)

was established, aiming “to analyze rebound risks and suffi-

ciency opportunities from digitalization.” Six authors of this

paper are members of that group, stemming from different

disciplines (psychology, marketing, engineering, sociology/

social sciences, economics). This article is based on internal

colloquia conducted between 2016 and 2019. The colloquia

applied techniques of interdisciplinary co-creation [91, 92]

to synthesize conceptual considerations of digital sufficiency

that were developed and explored in greater detail within

smaller scientific projects and doctoral theses. In addition,

three transdisciplinary workshops [93] with representatives

from industry, civil society, labor unions, and federal gov-

ernment bodies were conducted to develop “transformation

knowledge” and “target knowledge” [93] regarding digital

sufficiency. The policy proposals presented here were devel-

oped and critically examined in those workshops.

Second, the digital sufficiency concept was further

advanced in a workshop consisting of 12 researchers and

practitioners at the 7th International Conference on ICT for

Sustainability in Finland in 2019. This workshop integrated

perspectives from two epistemic research communities:

social sciences addressing issues of sustainable development

and technical sciences addressing challenges of digitaliza-

tion/ICT. The multidisciplinary co-authorship of this paper

was set up as an upshot of that workshop and integrates

knowledge from the disciplines of the research group with

knowledge from the disciplines of informatics, computer sci-

ences, and future studies.

5 Digital sufficiency andits four dimensions

Building on our general definition of sufficiency (Sec-

tion2.1.), we define “digital sufficiency” as any strategy

aimed at directly or indirectly decreasing the absolute level

of resource and energy demand from the production or appli-

cation of ICT.

Based on the abovementioned four mechanisms related

to why ICT currently fall short in delivering environmental

gains (see Section2.2.), we distinguish four dimensions of

digital sufficiency: (a) hardware sufficiency, (b) software

sufficiency, (c) user sufficiency, (d) economic sufficiency.

To elucidate the difference between these four dimensions

and corresponding strategies for efficiency and consistency,

each of the following subchapters starts by explaining the

related efficiency and consistency terms and contrasting

those terms with what is meant by sufficiency for the respec-

tive dimension.

5.1 Hardware sufficiency

Increasing hardware efficiency would aim either at reduc-

ing the energy and material per production unit of hardware

(e.g., per fabricated smartphone) or at increasing energy

efficiency in the use phase to keep relative energy demand

per unit of computing power at a minimum. Hardware con-

sistency would aim at eliminating toxic materials in the ICT

production process, ideally achieving production cycles with

fully renewable or recycled materials, powered by renewable

energy.

In contrast, hardware sufficiency aims at being able to

produce fewer devices, designing devices last for a long

time, ensuring that their complexity and resource use do

not surpass the purpose they are designed for (“not cracking

a nut with a sledge-hammer”), and keeping their absolute

energy demand at the lowest level possible to perform the

desired tasks.

5.1.1 Challenges throughoutthelifecycle ofICT hardware

Environmental impacts throughout the life cycle of ICT

hardware (production, operation, disposal) are caused by

mining raw materials for production, causing production

waste and emissions, providing the energy needed in all

phases of the life cycle, and by the processes of end-of-life

treatment [88, 94]. Production of ICT hardware requires

more than 50 chemical elements, including many scarce

and toxic metals, the mining of which often has toxic

impacts on humans and ecosystems [95]. For end-user

Annals of Telecommunications

1 3

devices, the production phase usually accounts for the

highest share of lifecycle-wide energy demand and CO2

emissions [72, 94]. Device collection remains a main chal-

lenge at the end of product life. Devices from data centers,

base stations, etc. are more often recirculated while PCs,

tablets, smartphones, and other end-user devices are often

stored at home. If collected, removing toxic components

and recovering scarce metals remain major issues [96].

Some of these challenges can be addressed by strate-

gies towards greater hardware consistency. But given the

limited knowledge on raw material substitutes, basing all

hardware production on 100% renewable resources within

a reasonable period of time does not seem feasible [97,

98]. Moreover, even under optimum industrial conditions,

only a small subset of the materials can be recovered [95]

and much ICT hardware enters informal recycling chan-

nels from which few elements are recovered, often under

hazardous conditions [99].

At the same time, the absolute number of ICT devices

is constantly increasing: An increase from 18 billion

ICT devices in 2017 to over 27 billion devices in 2022

is expected [100]. Moreover, the acceleration of product

cycles increases demand for new hardware and stimulates

resource depletion. Functioning hardware is often rendered

obsolete by software evolution or even by planned obsoles-

cence [101]. Furthermore, the trend towards the Internet

of Things, in which everyday objects are enhanced with

ICT hardware, could lead to increased software-induced

obsolescence [18], which will affect not only the electron-

ics part but the whole “smart thing” in which it is embed-

ded [88].

During the use phase, impacts are mainly caused by the

electricity required to power ICT hardware [1]. The environ-

mental impacts depend on the electricity mix used for oper-

ating the hardware, a mix still dominated in many countries

by fossil fuels. Especially for data centers, which operate

permanently, electricity consumption causes major environ-

mental impacts throughout the whole lifecycle. Besides the

amount of data processed (see software sufficiency below),

decisive factors for the energy demand of data centers are

waste heat recovery, cooling technology, and server utiliza-

tion [102]. Data center infrastructure and associated envi-

ronmental impacts are expected to increase significantly in

the future [103, 104].

Altogether, such challenges are likely to countervail large

parts of the saving potentials achieved by strategies that

improve efficiency in hardware production. Undoubtedly,

improving the material and energy efficiency of ICT and

using electricity from renewable sources can be effective

levers to reduce environmental impacts. However, efficiency

and consistency strategies alone may not lead to absolute

savings [105]—which is why sufficiency strategies are

essential.

5.1.2 Elements ofhardware sufficiency

Extending the useful life of devices reduces the demand for

new devices and thus slows down the flow of resources from

extraction to waste. If manufacturers design repairable and

upgradable devices (e.g., through modular design), these

will be able to match demand for computing power for a

longer period. Also, purchasing smaller devices (e.g., lap-

tops instead of desktop computers) often reduces impacts as

these require less material resources and energy during both

production and operation [94].

With regard to end-of-life treatment, improving the recy-

cling systems in terms of collection and recovery rates is

also important. To improve recovery, devices need to be

designed for it (e.g., through use of screws instead of glue)

[95]. However, the goals of reparability and recovery often

conflict with the demand for compact and light devices.

5.1.3 Policies thatpromote hardware sufficiency

Policies targeting hardware sufficiency mainly address

manufacturers, retailers, purchase departments, and infra-

structure providers (e.g., network, data center) and relate to

changes in hardware design, purchase, use, and end-of-life

treatment. With respect to infrastructures, policies should

incentivize shared use among providers to increase utiliza-

tion, reduce hardware use, and create synergies in building

and operation. Beyond long-lasting design and efficiency,

network hardware in base stations can also be mutualized

in many ways.

Policies for hardware design, which significantly influ-

ences the environmental impacts during production, opera-

tion, and end-of-life treatment, should focus on four aspects.

(1) Developing standards that ensure low environmental

impacts during production, e.g., a design directive can

require manufactures not only to avoid critical and hazardous

materials (consistency) but also to constantly increase the

share of recycled materials and reused parts. (2) Legislating

design principles for long-lasting hardware, e.g., extending

warranty periods and setting minimums (e.g., components

need to be replaceable, use of standard interfaces). (3) Enact-

ing energy consumption standards for hardware, e.g., setting

standards not for relative energy consumption (efficiency)

but for absolute energy consumption of hardware. (4) Devel-

oping policies to encourage reuse and recycling, e.g., by

setting minimum standards for the rates in recovering met-

als, which would also benefit manufacturers as soon as the

recovery of materials from used devices is cheaper than the

mining of raw materials.

To enhance sufficiency in purchasing hardware, first of

all, public and private organizations can adopt policies to

reduce the number of devices, e.g., through bring-your-own-

device (BYOD) strategies or private use of company-owned

Loading more pages...