ASSESSING THE POSITIVE AND NEGATIVE IMPACTS
OF ICT ON PEOPLE, PLANET AND PROFIT
Paula van Hoorik, Freek Bomhof and Pieter Meulenhoff
TNO Information and Communication Technology, Brassersplein 2, Delft, The Netherlands
Keywords: ICT, Assessment of the impact of ICT, Sustainability framework.
Abstract: There are numerous methods to measure the impact of ICT on the environment, all of them different in
scope, some of them incomplete and it is difficult to see the connection between them. This study draws an
inventory of methods to assess the impact of ICT which all of them have their pros and cons. Furthermore,
this study defines the building blocks a more complete method to evaluate both the positive and negative
impacts of ICT on People, Planet and Profit.
1 INTRODUCTION
Since the Copenhagen summit (2009, the United
Nations Climate Change Conference), there is the
insight that “nations talk, cities act”. Indeed, the
number of ‘smart cities’ that seek to address
sustainability goals is increasing. However, many
pilots and initiatives in these cities seem to be
inspired mainly by what is possible, and not based
on a careful analysis of the net sustainability effects.
The ICT4EE conference in Brussels (February
2010) concluded that no generally adopted and
available framework exists to assess and compare
the sustainability effects of pilots or measures. Good
starts have been made in various initiatives, though.
Bristol has been working on ecological indicators
using sustainability and quality of life as starting
points (McMahon 2002). The Amsterdam Smart
City initiative (www.amsterdamsmartcity.nl) uses
the concept of ‘value cases’, where a joint financial
and carbon analysis of the projects is executed.
However, only net carbon savings are calculated,
and no attempt has yet been made to incorporate
embodied carbon and second order effects. So, the
‘profit’ side of sustainability is accounted for, and
the ‘planet’ side partially. The ‘people’ aspect of
sustainability remains untouched.
A good initiative to gather and compare energy
and environmental related data is the Urban EcoMap
(urbanecomap.org), which is developed by Cisco as
part of the Connected Urban Development program.
In this article, we describe some of the
sustainability frameworks that exist for areas in
which ICT is the main driving force. These
frameworks have a large variation in scope and
sustainability goals and none of them gives a
complete assessment of the impact of ICT. The end
of the paper defines the requirements for a more
complete method.
We regard sustainability in the light of the Triple
Bottom Line: People, Planet, and Profit.
2 IMPACT OF ICT ON
SUSTAINABILITY
Gartner caused quite some attraction with its
statement in 2007 that the whole ICT sector is
responsible for about 2% of the world’s energy
consumption, comparable to the energy use of the
aviation sector. Building on this, the observation has
been made that if we want the ICT sector to still use
2% of all the world’s energy by 2020, ICT has to
become much more efficient because of the prolific
rise of ICT in virtually any aspect of our society.
(According to (The Climate Group, 2008), the
Business As Usual scenario would mean that ICT
would account for 2,7% of all emissions, compared
to 1,25% in 2002, which is a 216% increase.) At the
same time, it is good to note that ICT itself is a
promoter of energy efficiency: the ACEEE (2008)
states that “For every kilowatt-hour of electricity
that has been demanded by ICT, the U.S. economy
increased its overall energy savings by a factor of
about 10. (…) The extraordinary implication of this
45
van Hoorik P., Bomhof F. and Meulenhoff P. (2010).
ASSESSING THE POSITIVE AND NEGATIVE IMPACTS OF ICT ON PEOPLE, PLANET AND PROFIT.
In Proceedings of the Multi-Conference on Innovative Developments in ICT, pages 45-50
DOI: 10.5220/0003041100450050
Copyright
c
SciTePress
Table 1: Sustainability on different levels.
Scale Challenges, examples, frameworks
ICT / CT
Challenge: Maximum of communication
throughput for a minimum of energy
Framework: ECT
ICT/ IT
Challenge: Maximum of computing
power for a minimum of energy
Example: Intel ‘Westmere’ energy-
saving chip
Framework: Energy Star
ICT / facilities
Challenge: How to maintain the IT and
CT functions with a minimum of energy
for supportive technology?
Framework: Energy Star
Data centre
Challenge: Given the application
landscape, what is the minimum of
energy needed to support this?
Example: virtualisation
Framework: PUE
Software/
application
Challenge: Given the functional
requirements, what is the most efficient
way to accomplish these?
Example: Green coding
Framework: Microsoft Joulemeter
IT Functions/
processes
Challenge: What is the most energy
efficient way to reach a business goal?
Example: e-billing
Framework: OpenDCME
Companies
Challenge: Sustainable operation
Framework: GRI, ISO14064
Cities/ Regions
Challenge: How to balance sustainable
working, living and travelling?
Example: Smart Cities
Framework: none
Economies
Challenge: Combine maximum welfare
with maximum sustainability
Example:
Framework: ISO14064
finding is that ICT provide a net savings of energy
across our economy”.
The Climate Group (2008) expects that usage of
ICT will increase substantially in the coming years,
therefore paying attention to the direct energy
consumption of ICT itself is certainly very
important. However, besides the "greening of ICT"
activities, it is recognized that the smart application
of ICT can have a substantial impact on reducing our
energy consumption in other fields. This is called
'greening by ICT' and includes for instance a shift
from delivering physical products to delivering
services: the 'dematerialisation' effect of ICT (Romm
1999). An example is less need for printing because
more information will be available electronically
(Boston 1999), although predictions concerning
paperless offices do not easily come true (Kohl
2004). With the advent of e-paper solutions,
however, the environmental aspects of
dematerialisation in offices (Schmidt 2009) are
receiving more attention.
Additionally, the effects of ICT that seem to
have good advantages in the short term could be
undone in the long run. That can happen due to so
called rebound effects. If, for example, ICT enables
cheaper production, the demand for products will
raise thus increasing pollution. These rebound
effects make it difficult to evaluate the effects of
ICT on sustainability. What is needed is a method to
look at direct, second order and third order effects in
a uniform and structured way. An attempt is made
by introducing the RAP method (Bomhof, 2009),
that tries to assess the net effects of interlinked
positive and negative side effects of ICT. The
method is qualitative by nature, however, so it is not
easy to compare measures and pick out the best.
A more elaborate way to assess the net impact of
ICT-driven sustainability innovations is presented in
(Hilty, 2006). This analysis is quite rigorous and has
a good eye on so-called revenge (or rebound)
effects, but is only carried out on a macro level
(country or continent) and should be adapted to
assess the net effect of a set of ICT innovations.
Knowing all this raises new questions: how can
we estimate, or even calculate the effects of ICT?
How can we be sure that ICT indeed has this
greening effect? And if we knew that, how could we
design ICT applications that enhance sustainability?
The ACEEE analysis was made on a macro scale:
the economy of a whole country is compared to ICT
developments and investments. It enables to draw
conclusions on a statistical basis. The ACEEE has
formulated regression equations between factors as
Total Electricity Use and ICT Capital Stock, or Total
Energy Use and Economic Growth, and ICT
Investments. The impact of ICT on a small scale
(micro level: one worker, or one department in a
company) could be determined by careful analysis of
how processes and procedures change as a result of
ICT. The energy impact could also be measured or
estimated.
2.1 Sustainability on Different Levels
Looking at the various frameworks and methods,
one can put them into a hierarchical order.
Within each level, initiatives and frameworks
exist to achieve maximum energy efficiency for that
level. The challenges at each level can be formulated
as: “given the requirements of the next level, how
can we achieve these with a maximum of
sustainability?”. Sometimes the challenge for a level
is treated on that level alone. A pitfall that is
commonly seen is that levels are skipped without
even noticing them: a company that wants to
INNOV 2010 - International Multi-Conference on Innovative Developments in ICT
46
‘greenwash’ its image, will invest in energy efficient
office lighting without paying attention to inefficient
business processes.
Our starting point has been with energy
efficiency, so not surprisingly the ‘planet’ aspect of
each level is covered. An observation can then be
made that the two other sustainability aspects,
‘people’ and ‘profit’ are usually not incorporated in
the frameworks that are used on the lower levels.
In the next sections, some of these frameworks
will be presented in a little more detail.
3 FRAMEWORKS TO ASSESS
THE “GREEN-NESS” OF ICT
3.1 Energy Efficiency of Network
Equipment: ECR
Today, the energy efficiency of network equipment
is seldom taken into account in the purchase process
of network equipment. One of the underlying
reasons is that among vendors there is no generally
accepted standard that defines the energy efficiency
of network equipment. The Energy Consumption
Rating (Cueppens, 2008) (ECR) Initiative is an
attempt to provide an efficiency metric that provides
this insight. The ECR initiative defines a framework
for measuring the energy efficiency of packet based
network equipment. The goal of ECR is to define a
standard rating that can be used for comparing the
energy efficiency of network equipment.
The ECR Initiative defines public available test
procedures and methodology to asses the energy
efficiency of various types of network equipment.
The outcome of the testing procedure is two energy
efficiency ratings, expressed in terms of Watts per
Gbps. The basic ECR rating is defined as the energy
efficiency at the maximum achievable network load
without packet loss. Also defined is a variable load
metric ECR-VL, which is a weighted average of the
energy efficiency at network loads between idle and
100%.
The success of the ECR standard is dependent on
the adoption by the vendors of network equipment.
At this moment the ECR is only supported by Ixia
and Juniper Networks.
3.2 Energy Efficiency of Computer
Systems
A well known standard to rate the efficiency of
consumer products is the EnergyStar standard
(EnergyStar, 2009). Energy Star includes a
specification to rate the efficiency for desktop and
notebook computers. This includes an efficiency
rating based on separate categories for computer
systems (desktops, notebooks, thin-clients and small
servers) as well as a minimal required PSU
efficiency.
3.3 Energy Efficiency of Data
Centres: PUE
The green grid (GreenGrid, 2007) is a global
consortium of IT companies and professionals with
the goal to improve energy efficiency in data
centres. The green grid works on the development of
standardized and accepted set of metrics,
methodologies and processes to achieve this goal.
The Green Grid has defined the efficiency metric
PUE, and its inverse DCiE, as a standard to express
the energy efficiency of data centres. Today, the
Green Grid, and the PUE metric are widely adopted
by the ICT industry. The Power Usage Effectiveness
metric PUE is defined as:
PUE = Total Facility Power / IT Equipment power
(1)
The Total Facility Power is the measured power
dedicated to the data centre and the IT Equipment
power is the power measured of all ICT equipment,
mainly computers, network components and storage.
The calculation of PUE is relatively straightforward
and most complexity finds its origin in accurate
bookkeeping: the allocation of energy measurements
to IT equipment and facility power. The major
advantage of PUE is its wide acceptance, and
straightforward implementation.
At the same time, PUE methodology has some
limitations. The most important limitation is that
energy efficiency of hosted ICT infrastructure in
data centres is beyond the scope, and therefore not
known. PUE does not provide insight in the
effectiveness of individual ICT systems. As an
example: Computer systems which are in a
continuous idle state could possibly be switched off
so that energy is saved. From a mere PUE
perspective switch of a system could have a negative
impact on efficiency because the total IT equipment
power consumption is reduced, but probably the
total cooling capacity of the data center is not: this
means that the ratio “Total Facility Power” versus
“IT Equipment Power” becomes less favourable.
One criticism of measures like PUE is that it
only focuses on the hardware, and not the logic
behind it. An example is easily demonstrated: in an
ASSESSING THE POSITIVE AND NEGATIVE IMPACTS OF ICT ON PEOPLE, PLANET AND PROFIT
47
overdimensioned data center, where lots of cooling
equipment is running but where only part of the
floor is occupied by working IT equipment, it is easy
to increase the power efficiency by just adding a
bunch of old, unused computers. In that case, hardly
extra cooling is needed, but these old computers do
use power. This means that the ratio of ‘total energy
used’ versus ‘energy used for IT only’ becomes
more favourable. The main problem is that PUE is a
ratio, that can be made ‘better’ by decreasing the
numinator, but also by increasing the denominator.
Another problem is the scope of the PUE itself: it
does not give an idea of the usefulness of the IT
equipment.
Another limitation is that some sustainability
measures, like reusing generated heat though heat
pumps, do not have an impact on the efficiency of
the data centre itself and therefore have no impact on
PUE.
3.4 OpenDCME
The Data Centre Measure of Efficiency
(openDCME.org) is positioned as an extension to
more ‘technical’ measures like PUE, EUE and
DCiE. These technical measures give an indication
of the efficiency of the datacenter hardware,
including infrastructure needed for cooling and other
operating conditions.
OpenDCME strives to overcome the drawbacks
of for example the PUE, by focusing on efficiency-
inducing policies rather than mere hardware
measurements. The model identifies 16 Key
Performance Indicators that are grouped into 4
quadrants.
Table 2: OpenDCME model.
The data center:
DCiE
Floor usage
Bypass
Recirculation
The IT Assets:
Network architecture
efficiency
Storage architecture
efficiency
X86 architecture
efficiency
RISC architecture
efficiency
The tooling:
Network
management
Storage
management
Compute
management
Electrical and
mechanical
management
The processes:
Change and
configuration
management
Product lifecycle
management
Capacity management
Service Level
management
This approach enables datacenter management to
assess more areas that have impact on energy use, to
incorporate supporting processes, like product
lifecycle management, as well. The scope is limited
to datacenters, so this approach is only suitable for
‘green it’ initiatives. Besides, it cannot be easily
used to compare different scenarios.
3.5 Global Reporting Initiative
The Global Reporting Initiative (gri.org) has
introduced the Sustainability Reporting Guidelines
that should be applicable to any organisation
regardless of size or business. The guidelines are
meant to provide transparent information on the
sustainability performance of the organisation.
The ‘people’, ‘planet’ and ‘profit’ aspects of
sustainability can be recognized from the indicators
that are proposed for:
• Economic
• Environmental
Social performance (labour practices & decent
work; human rights; society; product responsibility)
Each indicator is supplied with a protocol that
describes how the indicator should be measured. The
focus of GRI is to encourage organisations to report
on their sustainability indicators, and to make this
kind of reporting just as normal as financial reports
for the organisation’s profit are today.
The indicators in the GRI form a very good basis
for judging the net sustainability performance of an
organisation, yet it does not make an attempt to
bring the indicators on a level that enable to compare
organisations.
3.6 ISO 14064
ISO 14064 is an international standard that specifies
and guides the quantification and reporting of
greenhouse gas (GHG) emissions. ISO 14064 is
compatible with the GHG protocol
(www.ghgprotocol.org). It provides governments,
businesses, regions and other organisations with a
set of tools for programs aimed at measuring,
quantifying and reducing greenhouse gas emissions.
These standards allow organisations take part in
emissions trading schemes using a globally
recognised standard. ISO 14064 provides a
comparable method of reporting that is
internationally considered as “good practice”. The
standard provides a methodology to quantify GHG
emissions at the organisation level and the project
level. Also includes are guidelines and requirements
for validation, verification and certification.
INNOV 2010 - International Multi-Conference on Innovative Developments in ICT
48
Assessing GHG emission at the organization
level is separated into three tiers:
Direct GHG emissions occur from an
organization’s own resources, like furnaces,
machines or owned vehicles;
GHG emissions related to purchased electricity;
Other GHG emissions i.e. related to purchased
products and services like transportation.
The Carbon Disclosure project
(www.cdproject.net, CDP) provides a platform,
where organisations can disclose their measured
GHG emissions like ISO14064. This includes
climate change strategies and reduction targets. The
CDP data is available to investors, policy makers,
governments, researchers and the public.
4 REQUIREMENTS FOR A
COMPLETE ASSESSMENT OF
THE IMPACT OF ICT
We found a lot of initiatives of communities and
cities on doing something sustainable and often,
some form of ICT is used. However, not one of them
seemed to be able to answer questions such as
‘which are to most sustainable activities?’, and
‘which of these systems is able to reduce the most
CO2? What about the effects on people and what
about the business case? In short, in order to
compare ICT and ICT applications and advise on
which are the most sustainable choices to make,
none of the frameworks described above did fit our
purpose.
With one of the statements made in Copenhagen
(“nations talk, cities act”) at the background, it
seems appropriate to aim our efforts to the scale of
cities and/or regions. It is the scale at which concrete
measures can be designed, implemented, and
evaluated, often with the help of local companies.
And, to reduce the carbon footprint of a city or
region, it is not enough to focus on ‘Green IT’ alone:
the ‘Greening by IT’ effects have to be accounted
for as well.
What is needed is a method to make a complete
assessment the sustainability of ICT (applications)
that is able to tell whether solutions such as
videoconferencing, e-invoicing and working at home
really do make a difference. The method should
include not only the direct effects of ICT but also the
indirect and system effects. Direct effects are the
effects of the production and use of ICT; indirect
effects are the effects of the use of ICT on the use of
energy; and system effects are the effect of
behavioural changes or completely different
industrial processes and usage patterns by ICT
(Bomhof, 2009). Sustainable development is defined
by the UN (Brundtland, 1987) as a process of
change in which the exploitation of resources, the
direction of investments, the orientation of
technological development; and institutional change
are all in harmony and enhance both current and
future potential to meet human needs and
aspirations. That implies that all aspects of
sustainability, People, Planet and Profit should be
part of the method since all three are equally
important to create a true sustainable solution.
The method should focus on long term
maximum added value; however that should be the
maximum added value for all aspects of society and
not for individuals and individual companies. That
means that prices should reflect added value and lost
added value to society and that all societal values
and cost should be calculated to the investor. If the
method is able to show the added value and lost
added value of different solutions and options, it will
enable cities, companies and other organisations to
make the most sustainable choices. Research is
currently done to achieve such a method. It can
already be observed that this encourages to bring
together knowledge from various fields (such as,
technological knowledge, labour efficiency &
quality, business modelling, life cycle analysis) and
leads to a better mutual understanding of
representatives from these fields.
5 CONCLUSIONS
A lot of excellent work has been done on assessing
the sustainability effects of ICT-driven initiatives.
However, it is not yet possible to make a sound
comparison between these initiatives in terms of
sustainability effects. We propose building a
framework that takes into account all relevant
aspects of sustainability, people, planet and profit,
and quantify them as much as possible. This
framework also makes a clear distinction between
direct, indirect and system effects.
REFERENCES
ACEEE, 2008, Information and Communication
Technologies: The Power of Productivity, How ICT
sectors are Transforming the Economy While Driving
Gains in Energy Productivity (Report number E081)
American Council for an Energy-Efficient Economy.
ASSESSING THE POSITIVE AND NEGATIVE IMPACTS OF ICT ON PEOPLE, PLANET AND PROFIT
49
Bomhof, F., Van Hoorik, P., Donkers, M., 2009,
Systematic Analysis of Rebound Effects for "Greening
by ICT" Initiatives. In Communications&Strategies
76, 4th quarter 2009, pp77-96.
Boston Consulting Group, 1999. Paper and the Electronic
Media, BCG.
Brundtland, 1987, The Report of the Brundtland
Commission, Our Common Future, Oxford University
Press
Ceuppens, Luc, Kharitonov, Daniel, and Sardella, Alan,
2008. Power Saving Strategies and Technologies in
Network Equipment Opportunities and Challenges, in
Risk and Rewards. Proceedings of IEEE Symposium
on Applications and the Internet.
The Green Grid, 2007, “The Green Grid Power Efficiency
metrics; PUE and DCiE”, In www.thegreengrid.org.
Hilty et al, 2006. "The Relevance of ICTs for
Environmental Sustainability – A Prospective
Simulation Study", in Environmental Modeling &
software 21, 1618-1629.
A. Kansal, F. Zhao, J. Liu, N. Kothari, A. A.
Bhattacharya, 2010. “Virtual Machine Power Metering
and Provisioning”. http://research.microsoft.com/
pubs/120435/JoulemeterVM.pdf, Microsoft Research,
Kohl, D. F. (2004): "From the Editor… the Paperless
Society… not Quite Yet", The Journal of Academic
Librarianship, Vol. 30, Issue 3, May, pp. 177-178.
McMahon, S. K., 2002. The development of quality of life
indicators—a case study from the City of Bristol UK,
Ecological Indicators, Volume 2, Issues 1-2,
November 2002, Pages 177-185.
Romm, J, A. Rosenfield & S. Hermann, 1999. The internet
economy and global warming, Centre for Energy and
Climate solutions/global environment and Technology
Foundation, Washington DC.
Energy Star, 2009, Program Requirements for Computers
Version 5.0, US Environmental Protection Agency.
The Climate Group, 2008, Smart 2020, Enabling the Low
Carbon Economy in the Information Age, The Climate
Group.
Schmidt A. and Hedal Kløverpris, N., 2009,
Environmental impacts from digital solutions as an
alternative to conventional paper-based solutions,
FORCE Technology Applied Environmental
Assessment Lyngby Denmark
INNOV 2010 - International Multi-Conference on Innovative Developments in ICT
50