THE GREEN ECONOMY PROGRESS

MEASUREMENT FRAMEWORK
METHODOLOGY

EVALUATING NATIONAL PROGRESS TOWARDS POVERTY ERADICATION AND

SHARED PROSPERITY WITHIN PLANETARY BOUNDARIES
Copyright © United Nations Environment Programme, 2017, on behalf of PAGE

The report is published as part of the Partnership for Action on Green Economy (PAGE) – an initiative by the United Nations
Environment Programme (UN Environment), the International Labour Organization (ILO), the United Nations Development
Programme (UNDP), the United Nations Industrial Development Organization (UNIDO) and the United Nations Institute for
Training and Research (UNITAR).

This publication may be reproduced in whole or in part and in any form for educational or non-profit purposes without
special permission from the copyright holder, provided acknowledgement of the source is made. The PAGE Secretariat
would appreciate receiving a copy of any publication that uses this publication as a source.

No use of this publication may be made for resale or for any other commercial purpose whatsoever without prior permission
in writing from the PAGE Secretariat.

Disclaimer
This document was produced with the financial assistance of the European Union. The views expressed herein can in no
way be taken to reflect the official opinion of the European Union.

Citation
PAGE (2017), The Green Economy Progress Measurement Framework – Methodology.

Cover Photos
© Robert Harding and Shutterstock

Acknowledgements
The report and methodology presented here were developed by José Pineda and Gisèle Mueller under the guidance of
Sheng Fulai of UN Environment’s Resources and Markets Branch.

Technical guidance for the theoretical framework was provided by Carmen Herrero, University of Alicante; Antonio Villar,
Universidad Pablo de Olavide; and Eduardo Zambrano, California Polytechnic State University. Ludovica Galotto, James
Rawles, Katharina Bohnenberger, Laura Russo and Nordine Bendou provided excellent research assistance. Akmal
Abdurazakov and Leonardo Ortega provided significant assistance with the construction of the initial database of indicators.

This paper also greatly benefited from two workshops held in April and June 2015 and UN Environment appreciates the
technical inputs from all participants. UN Environment would also like to thank the following people who sent written
comments on a previous version of the paper, which helped to improve the final version: Ulf Narloch (World Bank); Katharina
Stepping (German Development Institute); Kookie Habtegaber (World Wildlife Fund); Satya R. Chakravarty (Indian Statistical
Institute); Martin Halle (Global Footprint Network); Humberto Llavador (Universitat Pompeu Fabra); Andreas Hauser (Swiss
Federal Office for the Environment); and Damien Friot (Shaping Environmental Action). The following colleagues within
UN Environment also submitted written comments: Niklas Hagelberg (DEPI); Hilary Allison (WCMC); Hy Dao (University
of Geneva and DEWA/GRID-Geneva); Pascal Peduzzi (DEWA/GRID-Geneva); Matias Gallardo (Regional Office for Latin
America and the Caribbean); Rowan Palmer, Zhengzheng Qu, Ronal Gainza, Jamal Srouji and Olivia Clink (Resources &
Markets Branch); Llorenç Milài Canals, Julie Godin and James Lomax (Sustainable Lifestyles, Cities and Industry); and
Janet Salem (Regional Office for Asia and the Pacific).

The report was edited by Mark Bloch and Tansy Stobart and designed by Thomas Gianinazzi. Administrative support was
provided by Rahila Somra, Desiree Leon and Fatma Pandey.

PAGE is grateful to the European

Union for providing the funding
support to this project.
THE GREEN ECONOMY PROGRESS
MEASUREMENT FRAMEWORK
METHODOLOGY
EVALUATING NATIONAL PROGRESS TOWARDS POVERTY ERADICATION AND
SHARED PROSPERITY WITHIN PLANETARY BOUNDARIES
 CONTENTS
1
LIST OF FIGURES.......................................................................................................................................1

LIST OF ACRONYMS.................................................................................................................................2

EXECUTIVE SUMMARY............................................................................................................................3

1. INTRODUCTION....................................................................................................................................5

2. PROMOTING THE TRANSITION TO AN INCLUSIVE GREEN ECONOMY.............................6

2.1 OBJECTIVES OF THE GEP MEASUREMENT FRAMEWORK................................................................8
2.2 PROGRESS ON ACHIEVING AN INCLUSIVE GREEN ECONOMY........................................................9
2.2.1 PROGRESS AS MEASURED BY THE GEP MEASUREMENT FRAMEWORK..................... 9

3. THEORETICAL FRAMEWORK ....................................................................................................... 10

3.1 PROGRESS IN THE SINGLE INDICATOR CASE.................................................................................... 11
3.1.1 THE SETTING......................................................................................................................................... 11
3.1.2 TARGETS AND THRESHOLDS........................................................................................................... 11
3.2 THE GEP INDEX (PROGRESS IN THE MULTIDIMENSIONAL CASE)............................................... 13
3.3 THE NECESSITY OF A DASHBOARD....................................................................................................... 14
3.4 AGGREGATING THE INFORMATION FROM THE DASHBOARD AND THE GEP INDEX,
CREATING THE GEP+RANKING............................................................................................................... 16

4. FINAL CONSIDERATIONS................................................................................................................................ 17

5. REFERENCES...................................................................................................................................... 18

ANNEX I – FORMALIZATION OF THE THEORETICAL FRAMEWORK.................................... 21

A. THE BASIC MODEL......................................................................................................................................... 21
A.1 PROPERTIES............................................................................................................................................. 21
A.2 PROOF OF THE CHARACTERIZATION RESULT.............................................................................. 22
B. WEIGHTING OF THE GEP INDEX IN THE CASE OF SEVERAL COUNTRIES................................... 23
C. AGGREGATING THE INFORMATION FROM THE DASHBOARD AND THE GEP INDEX,
CREATING THE GEP+ (THE PROTECTIVE CRITERION)...................................................................... 25

NOTES......................................................................................................................................................... 27

LIST OF FIGURES
FIGURE 1 SETS OF CHALLENGES THAT AN INCLUSIVE GREEN ECONOMY AIMS TO ANSWER...... 7
FIGURE B.1 COMPARING REALIZATION A AND B WITH LINEAR INDIFFERENCE CURVES............. 24
FIGURE B.2 COMPARING REALIZATIONS A AND B ALONE THE 45˚ LINE............................................. 24
 LIST OF ACRONYMS
2

APEC Asia-Pafic Economic Cooperation

GEP Green Economy Progress

GGGI Global Green Growth Institute

GGKP Green Growth Knowledge Platform

HDI Human Development Index

ILO International Labour Organization

IWI Inclusive Wealth Index

OECD Organisation for Economic Co-operation and Development

PAGE Partnership for Action on Green Economy

PPP Purchasing Power Parity

SDGs Sustainable Development Goals

UN United Nations

UNDP United Nations Development Programme

UNIDO United Nations Industrial Development Organization

WDI World Development Indicators

WIPO World Intellectual Property Organization

 EXECUTIVE SUMMARY
3
UN Environment has developed a Green Economy
Progress (GEP) Measurement Framework to help
countries evaluate their overall progress towards
an Inclusive Green Economy and to enable a
cross-country comparison of progress. The GEP
Measurement Framework complements UN
Environment’s previously developed green economy
indicators framework (UNEP, 2012; UNEP, 2014; and
UNEP, 2015), which uses several types of indicators
at different stages of a typical policymaking cycle.
The GEP Measurement Framework has four
objectives. The first is to support the assessment
of progress in achieving a selection of the SDGs
within the 2030 Sustainable Development Agenda
1
and establishing direct links with them. The second
is to help countries monitor progress against
nationally set targets in priority areas and the third
is to introduce greater levels of transparency to
policymaking and provide policymakers with the
tools necessary to set-up policies that support the
transition to an Inclusive Green Economy. The fourth
and final objective is to measure and compare green
economy efforts across countries.
The GEP Measurement Framework is anchored in
an Inclusive Green Economy narrative. An Inclusive
Green Economy is a pathway designed to address
three main global challenges, namely: (a) persistent
poverty; (b) overstepped planetary boundaries;
and (c) inequitable sharing of growing prosperity.
The GEP Index captures these multi-dimensions
of an Inclusive Green Economy. It includes
measurements of accumulation of capital – be it
natural, low carbon and resource efficient, human,
2
or social – which serves as input for producing
goods and services in an environmentally friendly
manner. It also attempts to capture the transition
of consumption, investment, government spending
and trade towards such goods and services. The
GEP Measurement Framework also includes
measurement of the outcome of enabling policies
that are conducive to an Inclusive Green Economy.
Progress in improving these outcomes is then
analysed against specific planetary boundaries,
such as greenhouse gas emissions, water and land
use.
In its initial version, the GEP Measurement
Framework is composed of a GEP Index and a
companion Dashboard of Sustainability indicators.
These components can be both analysed
individually and combined to allow the ranking of
progress by country (GEP+). The first component,
the GEP Index, measures the progress made in
improving the well-being of current generations
in relation to economic opportunities, social
inclusiveness and environmental protection. It is
composed of 13 indicators that capture critical
issues faced in achieving an Inclusive Green
Economy transition, such as material footprint and
inequality. The GEP Index focuses on the progress
achieved by countries with respect to a target set
for each individual indicator. Construction of the
GEP Index utilizes a weighting system that allows
for the assessment of how far off a country is from
the global threshold on a specific component of
an Inclusive Green Economy (an indicator) and
an evaluation of the relative importance of one
component (an indicator) with respect to the others
from the country’s perspective. The Dashboard
of Sustainability includes six indicators that track
the sustainability of any progress that has been
achieved as measured by the GEP Index. Its role
is to monitor the long-term sustainability of the
factors underpinning humanity’s current and future
well-being.
The GEP Measurement Framework, in its current
version, proposes a method for measuring progress
4
that monitors changes in key variables, taking
into account global thresholds that should not be
surpassed and utilizing achievable targets selected
to help countries to move in the right direction
through policy intervention. These components are
critical to obtaining a useful measure of progress,
making the measurement framework a valid
instrument for not only practitioners, but also for
the wider community of researchers and academics
working in the field.
There are important challenges associated with
this line of work and it should be noted that there
is much progress still to be made. Conceptual
challenges remain with respect to the integration
of the GEP Measurement Framework and the
Inclusive Green Economy narrative as a result of
the latter’s complexity and the different implicit and
explicit causal relations that exist. In addition, there
are empirical challenges related to the availability
of indicators. While the focus on progress is a
significant added value of this work, it also imposes
considerable constraints on the potential indicators
that can be used. From a policy perspective, an
additional challenge lies in how to make use of
available national level indicators, which tend to
better capture local realities. The methodology
offered by the GEP measurement framework
is flexible enough to overcome these practical
challenges. A separate publication presents an
application of the methodology at the global level to
Beta test the methodology and to see the different
tradeoffs and challenges of the methodology to
improve its design, and more importantly, to enrich
3
the green economy policy making analysis.
There are at least two ways to expand on the GEP
Measurement Framework for policymaking in the
future. First, the methods used in the framework
are flexible when it comes to selecting indicators,
thereby making inter-country comparison possible
on any particular aspect of an Inclusive Green
Economy as long as the underlying data is available.
Important indicators that are currently unavailable
(such as those that adequately reflect biodiversity
and green jobs) may be incorporated into the
framework whenever they become accessible, thus
expanding the scope of measurement. Second, the
framework can build on UN Environment’s other
related work on indicators (namely Measuring
Progress towards an Inclusive Green Economy
- 2012, Guidance Manual for Green Economy
Indicators - 2014, and Indicators for Green Economy
Policymaking Synthesis Report of Studies in Ghana,
Mauritius and Uruguay – 2015) by adjusting the
choice of indicators to specific country needs and
priorities. This extension would make the framework
more useful in facilitating policymaking in a specific
country. The GEP Measurement Framework is
particularly useful for the monitoring of SDGs at the
country and global level, given its strong linkages
with many of the SGDs. The current Beta test
application of the methodology has 14 direct links
4
to 10 of the 17 SDGs. This will help not only in the
monitoring process, but also in the integration and
articulation of policies by enhancing the linkages
between IGE policies to the overall objectives of
sustainable development.
 1. INTRODUCTION
5
In June 2012, the United Nations Conference on
Sustainable Development (“Rio+20”) endorsed a
series of agreements, two of which stand out with
the ability to alter the way countries approach
sustainability. First, governments agreed to
negotiate a set of Sustainable Development Goals
(SDGs) that would be universal, aspirational and
transformational. Second, they agreed that a green
economy approach could be a tool for achieving
this sustainable development by contributing
to “(…) eradicating poverty as well as sustained
economic growth, enhancing social inclusion,
improving human welfare and creating opportunities
for employment and decent work for all, while
maintaining the healthy functioning of the Earth’s
5
ecosystems.” (Art. 56, “The Future We Want”). In this
context, Rio+20 also called on the United Nations
to provide technical assistance to those countries
wishing to pursue green economy policies, including
through the creation of measures and metrics
that would help track progress of efforts to green
economies and achieve sustainable development.
In September 2015, 193 UN Member States agreed
on a new Sustainable Development Agenda to end
poverty by 2030 and pursue a sustainable future,
supported by a list of 17 SDGs and 169 related
targets.
As the global leader of the Green Economy Initiative,
UN Environment is well positioned to catalyze the
development of green economy indicators with
a view of supporting the implementation of the
green economy concept at the country level. At the
global level, UN Environment conducted a study in
2012 on how to use indicators to develop and track
green economy policies (UNEP, 2012). In 2013, UN
Environment partnered with the OECD, the World
Bank, and the Global Green Growth Institute (GGGI)
via the Green Growth Knowledge Platform (GGKP),
to develop a common green growth indicators
framework (GGKP, 2013).
At the country level, UN Environment, under the
Partnership for Action on Green Economy (PAGE),
developed a framework that combines four types of
indicators into an integrated policymaking process
(UNEP, 2014). Each type is designed to assist at
specific stages of green economy policymaking.
First, indicators for issue identification help identify
and prioritize problems to be resolved through a
green economy approach. Second, indicators for
policy formulation help design solutions by defining
targets and measuring different policy interventions.
Third, indicators for policy assessment provide
critical inputs for estimating the cross-sectoral
impact of policy implementation and for evaluating
the effectiveness of each policy option. Finally,
indicators for policy monitoring and evaluation
assess the real impact of implemented policies in
the medium to long run. This framework was tested
in Ghana, Mauritius, and Uruguay, where green
economy indicators were identified as powerful
instruments to engage stakeholders in shaping the
policymaking process (UNEP, 2015). However, the
country studies also identified challenges in terms
6
of availability and quality of data.
In order to bridge measurement initiatives at the
global level with the indicator work carried out at the
country level, UN Environment has developed a new
GEP Measurement Framework that will facilitate
cross-country comparison of national efforts to
7
transition to greener and more inclusive economies.
 2. PROMOTING THE TRANSITION TO AN INCLUSIVE GREEN ECONOMY8
6
An Inclusive Green Economy is a pathway for
delivering sustainable development and a response
to three sets of global challenges, namely: (a)
persistent poverty; (b) overstepped planetary
boundaries; and (c) inequitable sharing of growing
prosperity. How could this tool be promoted to
address these challenges and ensure a sustainable
future for all people on the planet?
The Inclusive Green Economy contributes to
the overarching goal of poverty eradication and
shared prosperity in an intergenerational context
by safeguarding planetary boundaries, some of
which, e.g. climate, freshwater, ocean and land, are
mirrored by the SDGs. Planetary boundaries should
serve as drivers for innovative solutions that respect
ecological thresholds while improving the livelihoods
of communities around the world. In Africa, for
example, where more than 600 million people lack
access to affordable and reliable energy, off-grid
renewable energy technologies are generating new
opportunities in health care, education and the
wider economy, and contributing to reducing carbon
emissions.
UN Environment’s Inclusive Green Economy
Initiative puts a strong emphasis on the role
of investment and enabling policies to meet
sustainable development targets. In the evolving
narrative of the Inclusive Green Economy such
solutions and outcomes are generated through
the transformation of economic growth, which
is traditionally determined by the market value
of the goods and services produced with little
regard to how goods and services are produced
and/or consumed. An Inclusive Green Economy
emphasizes the accumulation of a new generation
of assets (Rockström et al, 2009) that are
necessary to produce goods and services in an
9
environmentally friendly manner. However, research
on a case-by-case basis is required to identify
the complementarities and trade-offs that exist
between these assets for producing such goods
and services. Moreover, such goods and services
should be produced through decent work and
should contribute to social inclusion. At the same
time, to induce the transformation of production, an
Inclusive Green Economy also promotes the switch
of consumption, investments, public spending and
trade towards goods and services produced with the
new generation of assets.
Any changes in aggregate demand and supply
will ultimately be constrained by global ecological
thresholds. However, in some cases, these
thresholds have already been crossed, as suggested
by the theory of planetary boundaries put forth
by Rockström et al. (2009). Governments have
traditionally regulated economic activities to limit
their environmental impact. Examples include
emission standards for new vehicles and crop
rotation legislation to safeguard soil quality. Under
an Inclusive Green Economy approach, planetary
boundaries should not only be adhered to in a
reactive manner but should also be seized as
opportunities for the introduction of innovative
measures that contribute in particular to “sustained,
inclusive and sustainable growth, full and productive
employment and decent work for all” (SDG 8).
These systemic changes do not typically take
place automatically. Amazonian forests sequester
carbon and host biodiversity for the benefit of
humanity at large. Individuals, businesses, and
even governments of the Amazon region might
not consider it fair for them alone to invest in
Amazonian forest conservation. Owners of
properties might not be interested in investing
in green buildings if the tenants are the primary
beneficiaries from reduced utility bills. Consumers
 might not want to buy energy-saving appliances if
7 the initial costs are significantly higher than regular
appliances. Businesses might find that investing in
the production of durable products may not work to
their advantage.
This is why policies are essential for incentivizing
actions. Particular emphasis should be placed on
policies that can mobilize finance to build the new
generation of all forms of capital and/or reshape
consumption patterns, investment, public spending
and trade. Examples include: (a) fiscal policy
(e.g. reform of fossil fuels can encourage energy
efficiency in both consumption and production);
(b) industrial policy (e.g. government spending on
research & development (R&D) for renewable energy
technologies); (c) sustainable public procurement;
(d) labour training; (e) social safety nets; and (f)
trade liberalization for environmental goods and
services. Other policies, such as rules, regulations
and standards remain important and should
complement green economy policy instruments
that focus more directly on mobilizing finance for
the transition towards an Inclusive Green Economy.
Figure 1 illustrates how the three sets of challenges
that an Inclusive Green Economy aims to address
are interconnected.

Figure 1: Sets of challenges that an Inclusive Green

Economy aims to answer

ERADICATING Within SHARING

POVERTY planetary PROSPERITY
boundaries

ECONOMIC
TRANSFORMATION

Natural Physical Human Social Consumption Investments Public Trade

Capital capital capital capital spending

UN Environment has developed the Green Economy Progress (GEP)

measurement framework, as an integrated policy instrument that both
assesses country efforts towards achieving targets set within planetary
boundaries and allows cross-country comparison of progress10 towards
achieving an Inclusive Green Economy.
 2.1 OBJECTIVES OF THE GEP MEASUREMENT FRAMEWORK
8
The GEP Measurement Framework is intended to
achieve four objectives:
The first objective is to contribute to the
monitoring of progress in implementing the SDGs
through establishing direct links with selected
11
SDGs. This will help monitor progress for specific
SDG targets and support the measurement
and implementation of the 2030 Sustainable
Development Agenda.
The second objective is to assess progress
towards national goals in priority areas. The
framework allows countries to include national
indicators and targets in their “customized-GEP”
measurement framework to track progress in
12
specific areas.
Third, the GEP Measurement Framework
will bring transparency and accountability to
policymaking, draw public attention to sustainable
development challenges, and highlight the
importance of achieving progress in an integrated
manner. If accepted as a basis for comparison
of green economy progress between nations, the
GEP Measurement Framework may also serve to
inspire policymakers and galvanize civil society to
push governments toward the pursuit of ambitious
Inclusive Green Economy agendas. It may also
help policymakers in identifying policy gaps where
more resources are required to increase the speed
and scope of greening their economies and making
them more inclusive.
Finally, the GEP Measurement Framework is
able to compare the efforts made by countries
in achieving the transition towards an Inclusive
Green Economy. The GEP Measurement Framework
helps countries assess where they stand in key
Integrated Green Economy areas while revealing
the challenges that arise from becoming less reliant
on carbon fuels and the opportunities of becoming
resource efficient and socially inclusive. The GEP
Measurement Framework serves as a signal to
countries to change their development path by
designing or reforming national policies to promote
the transition to an Inclusive Green Economy. By
tracking their green economy progress over time,
countries can evaluate how fast they are able to
achieve specific targets and measure the speed of
their transition towards an Inclusive Green Economy.
 2.2 PROGRESS ON ACHIEVING AN INCLUSIVE GREEN ECONOMY
9
There are two important conditions for progress: (a)
progress should aim to generate multidimensional
impacts (i.e. it should reflect the result of an
integrated impact on the economic, social
and environmental dimensions of sustainable
development); and (b) progress should be evaluated
from a medium and long-term perspective. These
conditions are discussed in detail below.
An Inclusive Green Economy could, among other
things, be interpreted as a means of decoupling
economic growth from resource use and
environmental impacts (e.g. reducing material and
environmental footprint). To achieve decoupling, key
factors and policies must be established, including:
(a) private and public investment aimed at greening
the economy; (b) fiscal policies (e.g. ecological
tax reform and phasing out harmful subsidies);
(c) enhanced market access for low carbon
technologies and sustainable technologies in
general; (d) development of green industrial policies;
(e) generation of green jobs; and (f) promotion of
social inclusion and use of trade opportunities
from new markets and technological innovation.
Progress can only be considered to have been
achieved if these improvements in current human
well-being are sustainable, therefore requiring that
the future development path stays within planetary
boundaries.
2.2.1 Progress as measured by the GEP
Measurement Framework
The GEP Measurement Framework uses a
set of green economy indicators to measure
progress against set targets. Any green economy
progress made is measured for each individual
indicator, and is aggregated in a composite index
13
across dimensions where there is a valid policy
substitutability.
Green economy progress on each individual
indicator is measured as the ratio between the
actual change observed and the desired change
with respect to a target for that indicator. Green
economy progress in the multidimensional case is
measured by the aggregation of progress across
indicators for each country into a composite
index. This provides an overall picture of progress
achieved by each country and allows cross-country
comparison of efforts for peer groups of countries
for a same set of indicators. For a meaningful
comparison across indicators within a country and
across countries, a weighting system that allows
for a combination of progress across different
indicators is required: this is in essence the GEP
Index, which will be discussed in the next section.
 3. THEORETICAL FRAMEWORK14
10
Two preliminary considerations must be flagged
before presenting the theoretical framework. First,
it is not possible to quantify everything that should
ideally be measured. Second, not all measurable
variables can be reasonably aggregated into a single
number. This implies that indices will only provide
a partial estimate of the performance that is being
evaluated. Complementing a single number with
a dashboard of indicators might be most useful. A
comprehensive index (e.g. the Human Development
Index) may help to depict a synthetic picture of how
certain aspects of interest are evolving as a whole.
In contrast, a dashboard of indicators not only
provides complementary information to complete
the picture given by the index, but may also help
better understand the nature of the changes in
the indicators of the index, both across time and
between countries.
As described in Section 2, the theoretical framework
looks at two types of progress: green economy
progress in single indicators and multidimensional
green economy progress through the GEP Index.
Progress on a single indicator measures country
achievements for that particular indicator and
informs the country on its performance in one
particular area of development. The GEP Index
measures progress in achieving the transition
towards an Inclusive Green Economy by aggregating
individual progress across dimensions and
weighting the results to make the index comparable
between and within countries. The value of the GEP
Index is that it:
1. Identifies key dimensions to be associated with
an Inclusive Green Economy, each of which may
be approximated by one or several variables (see
previous section);
2. Focuses on the progress made, i.e. the changes
rather than the levels; and
3. Measures the progress made relative to some
standards, i.e. targets and thresholds. Targets refer
to desired changes, whereas thresholds define
some critical levels.
15
The GEP Index refers to the evolution of the green
components in the economic process rather than
to sustainable growth or human development
as measured by the Human Development Index.
This emphasis involves some radical choices, in
particular by putting GDP aside as a reference
variable for the evaluation and substituting it
with green components, such as green trade or
16
green innovation. Two reasons explain why this
approach relies on quantitative measures rather
than on market values as the key inputs for the
GEP Index. First, using market prices to value the
considered elements is not adequate because
market prices reflect demand and supply forces,
which are clearly dominated by developed and large
emerging countries. Second, most green economy-
related variables refer to goods and services for
which there are no well-established markets.
 ures the progress made relative to some standards, i.e. targets and thresholds. Second, most16green economy-related variab
The GEP
than Index refers to the
to sustainable evolution
growth of thedevelopment
or human green components
trade as inno
green the
or measured economic
by reference
the Human
innovation . process
Two
well-established rather
Development
reasons
variable for explain
prices
markets. the
to valuewt
evalua
ets refer to desired changes,
15 whereas thresholds define some critical levels. 16
than Index
to sustainable growth or
. This emphasis humansome
involves development as measured
radical measures
choices, bythan
inrather the by
particular Human
on orDevelopment
putting
market
trade GDP
values
green aside as key
as the a forc
innovation
and supply inpu
. T
Measures the Index progress
15
. 3.
reference made
ThisMeasures emphasis
variable relative tothesome standards, and i.e. targets and thresholds.
ex refers to the evolution of the thefor
green
involves
progress
components
some
evaluation
madein radical
relative choices,
substituting
to some prices in particular
it value
to
standards, with green
the
3.1
i.e. bytargets
putting
components,
considered
Progressmeasuresand GDP
in aside
elements suchSecond,
rather
the
thresholds. as is anot
as
than
single greenon adequa
marke
most
indicato g
argets refer to desired changes, whereas thresholds 16 define some critical levels.rather
the economic process
eainable
relativegrowth reference
to some trade variable
or green
Targets
standards,
or human for the
innovation
refertargets
i.e.
development evaluation
to desired asand . Two
changes, and
thresholds.
measured substituting
reasons
whereas
by the Human explain
and
thresholds it with why
supply green
this
forces, components,
define some critical
Development approach which prices relies
are
levels.such on
clearly
to value the as green
quantitative
dominated
considered de
no well-establish by
16
es, whereas
emphasis trade or green
measures
thresholds define innovation
rather some thancriticalon. market
Two
levels. reasons
valuesbyexplain asputting
the Second,key why inputs thismost approach
for the
asgreena GEP
3.1.1 relies
Index.
economy-related
The and onFirst,
supply
setting quantitative
using
17forces, market
variables whichrefer are to c
Index refersinvolves
to the evolutionsome radical
of thechoices, green componentsin particular in the economic GDP aside
process rather
measuresThe
prices GEP toIndex
rather value refers
than the to
onconsideredtheitevolution
market valueselements ofasthe thegreen
iskey notinputscomponents
adequate
no forasthe
well-established in GEP
because the economic Index.
market
markets. process
First,
prices using rather
reflect market demand
iable for the
ustainable evaluation
growth
than
or human
to
and
sustainable
substituting
development
growth or
with
as measured
human
green components,
Therefore,
development
by the Human
as
such
applying
measured
green
thebyevaluation
Development
the
Second,
Human formula
most
Development in this 3.1
green Progress
economy-re
special case
en
This
of innovation
the green prices
emphasis Index
16
. to
and
components
3.1 Two
involves
value
supply
PROGRESS reasons
some
the
inforces,
theconsidered
IN which
economic
explain
radical THE why
choices,
elements
are
process
SINGLE this
in
clearly is not
rather
INDICATOR
approach
particular
dominated adequate
by relies CASE
putting onbyGDP because
developed
quantitative
aside
market
Suppose,
as
anda
prices
large
no for reflect
emerging
well-established
the time demand countries.
being, markets.
that progres
nher development
15
as measured . forces,
This emphasis
bygreenthekey involves
Human some
Development radical progress: choices, inusingparticular by putting GDPfor aside as athere are
than
11 on andmarket
supply
Second, values most as which
the are clearly
economy-related
inputs for thedominated
GEP variables
Index. by 3.1
First, developed
refer to
Progress goods
market and and
single large
in services
the emerging
indicator. single Focusing countries.
which
indicator on progress case res
e
variable
radical
for the
choices,
evaluation
reference
in particular
and
variable by
substituting
for
puttingthe evaluation
GDP
it with
aside andgreen
as
components,
substituting
a it with suchgreen as green
components, such as green 3.1.1 The set
Second, no most
well-established green economy-related
markets. variables refer to goods and services
in terms 3.1 for which there are
onProgress inభǡ௬the singl
ue the considered
green innovation elements
16
. orTwo is
reasons not adequate
explain because market
16 why this approach relies on quantitative prices reflect demand of changes. Note
௣೒೚೚೏ೞthat the
బ ሻ effect o௬
and substituting trade
it clearly
with green
green innovation
components, .suchTwoasreasons green explain why this approach relies quantitative ሺ௬
ൌ
ௗ௬
ൌ
orces, whichno
s rather than onmeasures are
well-established
3.1.1
market The dominated
valuessetting markets.
as 17 by developed
theonkey inputs and
for the large emerging countries. and 17
“bads” will be different: ሺ௬increasing ‫ כ‬the௬
as GEP Index. 3.1.1
ThatFirst, is, using
The thesetting
progress market
inIndex.
the single indicator case Suppose, for th
‫ כ‬ǡ௬ బ ሻ
asons
tvalue
greenthe explain why this rather approach than relies market
ongoods values
quantitative the key inputs for GEP First,
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ usingൌ ൞market
௦೒೚೚೏ೞ ௗ௬
3.economy-related
3.1
Measures
considered Progress
the
elementsvariables
progress is in refer
notthe
made to
adequatesingle
relative and
to
becauseindicator
some services standards,
market for
case
prices whichi.e.
corresponds there
targets
reflectto are
demand
theand
progress,
ratiothresholds.
whereas
of the actual increasing
and ሺ௬ ǡ௬ the
desired ሻ amount
17 ௗሺି௬ሻ o
3.1.1 The setting ௣್ೌ೏ೞ భ బ
es as the prices to value the
key inputs for the GEP Index. First, using market considered elements is not adequate because market prices reflect demand singleൌindicator ൌ
lished markets.
ply
nts forces, 3.1
which
is not adequate Progress
Targets
and are
Suppose, refer
supply
because
to
fordesired
clearly theindominated
forces,
market the
time which single
changes,
being,
prices by
that
are whereas
reflect indicator
developed
progress
clearly demand is andcase
thresholds
dominated largedefine some critical
emerging
increments
bySuppose,
developed countries.
(for
for andthethe levels.
case
large time of being,
“goods”)countries.
emerging ௦್ೌ೏ೞ ሺ௬ ‫ כ‬ǡ௬ బ ሻ
or reductions
that progress
in terms of cha
ௗሺି௬ ‫כ‬ሻ
is evalu
most green economy-related
evaluated forand variables
only 17referbased
oneeconomy-related
country to goods and services
on avariables
single forgoods
(for thewhichcase there Letare
of “bads”). ‫ ݕ‬ଵ ǡ ‫ݕ‬The stand for there
଴ progress measure
the actual forand the initial
dominated 3.1.1
bySecond,
developed The
most setting
greenlarge emerging countries. refersingle to and
indicator. servicesFocusing for
Suppose,whichon progress for are
the
and time
results
“bads” being,
in the t
ess
stablished in
The the
GEP
variables 3.1.1 single
Index
markets.
referno
refers indicator
well-established
indicator. to
Focusingthe 17oncase
evolution
markets.
progress of the
resultsgreen in components
the That is, in
progressthe
“bads” economic
is in
obtainedthe process
single
thebyconsidered rather
indicator
reversing that ofcase
dimension,the corresponds
and let
“goods”, ݀‫ ݕ‬ൌ ‫ݕ‬will toଵ െth
than to sustainablesetting
toThe goods and services for which there are single indicator. Focusing on p
growth or human development asincrements measured in terms by(for theofHuman changes. Note
Development that thereductions
effect
progress, on the mea
where
15 definition of for
Suppose, achievements
the time and targets
being, in terms
that progress is both in thethe
evaluated
case
numerator value
for only
of
and “goods”)
is defined
in thecountry
one
or
as follows:
denominator. based on
(for
a
the
setting Index . This emphasis involves some radical choices, in particular
17 and for “bads”
by putting will be different:
GDP aside inbyterms
as aincreasingof changes. thethe Note
amount thato
gress reference
inSuppose,
the ofsingle
3.1
single
variable
Progress
changes.
for indicator
theNote that
indicator.
for time
the evaluation
thecase
inFocusing
the
being, single
effect
andthat on indicator
on the measure
progress
progress
substituting
measure
it with
case
is
results evaluated
green in
“bads”
the
components,
is
for onlysuch
definition
obtained
one
of country
achievements
as green
reversing
based
that of
and on a
targets
“goods”
denominator. progress, whereasfunctionincreasing and “bads” the
and amountwill Let be different:
‫ݕ‬ofଵ a଴ “bad”
ǡ ‫ ݕ‬ሺ‫ݕ‬standଵ inc
ǡ ‫ ݕ‬଴ ሻw
dicator trade case single
or17green
of progress in the
in indicator.
innovation
terms of
16 case of “goods” and “bads” will
Focusing
changes.. Two on
reasons
Note progress
that explain
the results
why
effect this
onin the
The progress
approach definition
measure relies of
of on
progress
is increasing
achievements
quantitative in the and
case
linear
of
in‫݌‬௚௢௢ௗ௦
targets “goods”
her the time being,
setting 3.1.1 that
be different: The progress
setting
increasing is17the
evaluated
amount offor only one country
a “good” for the basedcase of on“goods”,
a and progress,
decreasingwhereas and linear increasing td
the considered
measures in rather
terms ofthan on
changes. market Note values that asthe the key
effect inputson for the
the measure GEP Index.
of First, using
progress in market
the case of “goods”
tor. Focusing on and “bads”
willprogress will
results be different:
in the increasing
definition of achievements
The the amount
progress Let ‫ ݕ‬and
ଵ of
ǡfunction
଴ a
stand“good”
targets isreflect
for will
increasing
thedemandincrease
actual and and the
linear measure
in
theofvalue ‫ ݕ‬for
initialଵ of
the
reference
is cast
prices to value increase
the consideredthe measure elements of progress,
is not adequate whereas because in market ‫ ݕ‬the
for pricescase The of “bads”.
progress Thein derivative
this dimension the is defined
simply
changes.
, forand and
Note
thesupply “bads”
progress,
that
time Suppose,
being, thethatwill
effect be
whereas
theprogress different:
ontime the increasing
ofmeasure
is increasing
evaluated the
ofdecrease
progress
for the
amount amount
it. in of the a of
“bad”
case a “good”
ofwill decrease
“goods” will increase
it. the measure of
for the abeing, that by only
progress linear isone inthe country
evaluated for
consideredthebased
for case on
onlyvariable.of a"bads".
one
dimension, country toǡbased
The
and ‫ݕ‬let derivative onൌ for
a‫ ݕ‬of ‫ݕ‬the progre
. Progre
ଵ ଵ ଴
increasing
forces, which amount
are clearly “bad”
dominatedwill developed ‫ݕ‬ progress
and large function
emerging with Let ‫ݕ‬
respect
countries.
ଵ ଴
stand
is݀‫ݕ‬ positive the
െ actual a
will
dicator.
progress be
Second, progress,
different:
Focusing single
is evaluated
most green whereas
increasing
on indicator.
progress
for only
economy-relatedincreasing
the one amount
Focusing
results country of
on the
invariables
the basedamount
aprogress
“good”
definition on
refer will of
results
of
apositive
to a
increase
goods “bad”
in the
achievements andwhen
value will
the
when isdecrease
measure
definition
services
ଵ and
‫ݕ‬defined
൐ ‫ݕ‬for of of
targets
‫כ‬
and
and it.
achievements
which negative
as negative
follows: and targets
areotherwise.
the otherwise.
there considered Trivially,
dimension,
Trivially, the the le
and f
hereas
essof changes. increasing
results
no well-established in
Let
in Note
the the
Letterms‫ ݕ‬ǡ amount
thatଵ
definition ofstand
଴
the
‫ݕ‬markets.changes.
stand of forathe
of achievements
effect for on“bad”
Note
theactual will
theactualthat
measure
and decrease
andthe
and
the effect
ofthe
targets
initial it.initial
on thereference
index
progress measure
isin above
the case
function orof progress
values
isbelow
of of
“goods”
decreasingLet one inin
the
now the
variable
depending
‫ݕ‬value
‫כ‬ case
Theisindex
.denote of
that
on “goods”
defined
the approximates
whether
is desired
above actual
as follows:
orvalue ofprog
the v
Let and
ଵ ଴“bads”
the stand
considered will
for be
the different:
actual
dimension, variable and andincreasing
the initial
let“good” the amount
reference ‫ ݕ‬. below of a
values
Progress “good” of will
the
(p) with increase
variable
respect the
that measure
approximates
to actual of
‫݌‬the initial ଵ reference
ffect on the
s” will be different: measure
‫ݕ‬ ǡ ‫ݕ‬
reference of
increasing progress
values of the in
the amountthe case of “goods”
݀‫ ݕ‬ൌ It‫ݕ‬will
of aapproximates
that ଵ
isെnegative ଴
increase in
the the
one case
measure
depending be given
ofof regress.by:
on whether progress ௗ௬
௚௢௢ௗ௦ ሺ‫ ݕ‬The ǡ ‫ ݕ‬଴is ሻ ൌ బ , ‫݌‬௕௔
ing
nd 3.1 Progress
the
for
, whereas amount
the increasinginthe
the
actualprogress,
considered
value
of
the a
and is
“good”
the
considered thewhereas
defined single
dimension,
will
initial
amount as of indicator
increase
reference
dimension, a “bad” will case
increasingand
follows: and thelet
values
let the
measure
݀‫ݕ‬ amount
of
ൌ ‫ݕ‬the
decrease it. െ of
ଵ ofvariable
‫ݕ‬ ଴a “bad”
. Progress that will
above decrease
(p)
approximates
or with
below it.
respect
the target. to It the
is initial
negative reference
in the caseprogress ௬ in
variable. ‫݌‬௚௢ ௗ
ed dimension,
mount of avalue“bad” and is
willlet
defined
Progress decrease
݀‫( ݕ‬p) ൌ ‫ݕ‬as ଵ
with െ ‫ ݕ‬. Progress
଴
follows:
it.respect to the initial (p) with referencerespect Remark: to theof initial
Whenregress. reference
there are several countries, international ‫ݏ‬௚௢௢ௗ௦ ሺ‫ ݕ‬ǡ ‫ݕ‬compa‫כ‬ ଴ ሻൌ
ed as
stand follows:
3.1.1 for The setting
Let
value
the actual ‫ ݕ‬ଵis ǡ and଴17
stand
‫ݕ‬defined the as forfollows:
initial the actual and
reference valuesthe initial
of‫ ݕ‬ଵthe reference
performed
variable The
ௗ௬ values progress
directly
that of by ଵ in
the
approximates ଴this
variable
confronting dimension
that their
ௗሺି௬ሻ approximates is simply
rates the corresp௬
of achievement
‫݌‬ ሺ ǡ ‫ݕ‬ ଴ ሻ ൌ , ‫݌‬ ሺ ‫ݕ‬ ǡ ‫ݕ‬ ሻ ൌ
edered
initialdimension,
reference thevaluesconsidered
and letof ݀‫ݕ‬the ൌdimension,
‫ݕ‬variable
ଵ
െ ‫ ݕ‬଴ . that Progress let௚௢௢ௗ௦
andapproximates ݀‫ݕ‬ (p) ൌwith ‫ ݕ‬଴ .ௗ௬Progress
‫ ݕ‬ଵ଴ െindicator.
respect variable.
to
బ
the ሺ(p)
௬Remark: ௕௔ௗ௦
ଵwith
When
initial ଴ ሻrespect
there
reference are
ௗሺି௬ሻ The
toseveral
the progress
௬ బ initial reference
countries, inLet thisnow dimension
‫ כ ݕ‬deno
value istimedefined asൌfollows:
‫݌‬ ሺ ‫ݕ‬
௚௢௢ௗ௦ଵ is଴evaluated
ଵ
ǡ ‫ݕ‬ ሻ ൌ , ‫݌‬ ௕௔ௗ௦ ‫ݕ‬ ǡ ‫ݕ‬ ൌ ௬ బ variable.
ൌ ‫ ݕ‬ଵSuppose,
efined െ‫ݕ‬ as.follows:
଴ for
Progress the(p)
‫݌‬௚௢௢ௗ௦ with
ሺ‫ݕ‬ ଵ being,
ǡ ‫ݕ‬respect
଴ሻ that
ௗ௬to, the progress
initial ǡ ‫ ݕ‬ሻ ൌ ௬ బ ௬ for international
ሺ‫ ݕ‬reference
ௗሺି௬ሻ బ only one country comparisons based foron eacha indicator be cangiven by:
బ ‫݌‬௕௔ௗ௦
The progress in this௬ dimension
single indicator. Focusing on progress results in the definitionbeofperformed is simply the Letcorresponding
now ‫ݕ‬
achievements‫כ‬
denote
directly
15
growth
See:andthe
by targets their ratestheof
or
desired
confronting reduction
value ofrate
http://hdr.undp.org/sites/default/files/hdr2015_techn the
variable, an
Theof variable.
progress in this ଴dimension is simply 3.1.2
ǡ ‫ݕ‬଴ ሻ଴the
ௗ௬ Targets
corresponding
be given and
by: growththresholds
16
ௗሺି௬ሻ
In line
orwith the
reduction “Beyond
Let(Stiglitz
now
GDP”
rate
denote
approach
of the the
(See Report by
thedesired v
‫כ‬
in terms changes. Note ଵ that the ௗ௬ effect ‫݌‬௚௢௢ௗ௦ on ሺ‫ݕ‬ the
ሺ‫ݕ‬ଵ ǡଵmeasure ሻൌൌௗሺି௬ሻ of
బ , ‫݌‬of
progress ሺ‫ݕ‬ଵ ǡin
achievement ‫ ݕ‬଴theሻ ൌor case the of “goods”
progress made ‫ݕ‬
et on each
s ଵin this dimension
ௗ௬ ‫݌‬ is௚௢௢ௗ௦simply ሺ‫ݕ‬ ǡ ‫ݕ‬
the
ௗሺି௬ሻ
ሻ ൌ ,
corresponding ‫݌‬ ௕௔ௗ௦ growth‫ݕ‬ or ௬reduction ௕௔ௗ௦ rate of Social
the ௬ బ Progress al., 2010)) and recent critici
ൌ variable.
బ బ
‫ ݕ‬ǡ ‫ݕ‬and ଴ ሻ “bads” will
, ‫݌‬௕௔ௗ௦ be
ሺ‫ݕ‬ଵdifferent:
The progress ǡ ‫ ݕ‬଴ ሻinൌthis௬increasing
dimension the
௬ is simplyamount the of a “good” ௬ will increase measurement
indicator. the measure beframework
givenof by: has deliberately decided not to u
௬బ
Let now ‫ ݕ‬denote
బ
the desired value ofwillProgress
the variable, is synonymous
it.and call with
for ‫כ‬this ‫כ‬moving
decision is଴ that in
െ ‫ ݕ‬. Then,
GDP the not“right”
only measures
thethe direction;
‫ כ‬green but th
଴ ሻtarget will of ind
‫כ‬
progress, whereas corresponding increasing growth the or amount
reduction ofratea simply
“bad”
of the thedecrease ݀‫ݕ‬ ൌ‫ݕ‬ ௗ௬
ress in this
The
dimension
progress
is
in this
simply the
dimension
corresponding
is
growth
corresponding
an indicator or reduction will be
growth
rate
towards
assessed
of
allows
or
the
an
the
Inclusive
reduction
against
GEP
‫ݏ‬ ௚௢௢ௗ௦
index
Green
rate
athe
to
ሺ ‫ݕ‬
target
‫כ‬of
Economy.
capture
ǡ ‫ݕ‬ and ൌThe
(although బ
,
inclusion
a௬threshold.‫ݕ‬
‫ݏ‬ ௕௔ௗ௦
partially)
ሺ
the
imply the desired
enote
Let now
be given
‫ݕ‬
variable.
variable.
corresponding
‫כ‬
denote
value ofgrowth
by: the
the variable,
desired
or reduction value
and call rate of
݀‫ ݕ‬ofൌthe
‫כ‬ the variable,
‫ ݕ‬െ ‫ ݕ‬. Then,denoted
‫כ‬ ଴ and call
the target
݀‫ݕ‬ ‫כ‬
ൌwill
‫ݕ‬ ‫כ‬
െ ‫ݕ‬ ଴
. Then, target will
measuring the access to basic services, health and ‫ݏ‬progre
educa
for the actual and the initial reference threshold, bythat t, plays no apparent role soanthat ௚௢௢ௗ௦
Let ‫ ݕ‬ଵ ǡbe ‫ ݕ‬଴ given
stand by: values of 3.1.2 the variable Targets 17 and thresholds
approximates
economic growth will bring to promoting
15
Inclusive Green
the considered Let dimension,
Let now ‫ ݕ‬denote denote andthe letdesired
the ݀‫ݕ‬ ൌ ‫ݕ‬value
desired ଵ
െ ‫ݕ‬of. the
value ଴
Progress ratio (p) between
ௗ௬with ‫כ‬
respect actual‫ כ‬to ൌ
and See
the‫ ݕ‬initialdesired
െௗሺି௬
Annex ‫כ‬
‫ ݕ‬.reference
ሻ
I.Achange.
for a discussionThe choice
of of
See: http://hdr.undp
properties in‫ݕ‬ the is,
‫ כ‬basic
th
ሺ‫ݕ‬of ‫ כ‬the ‫כ‬଴ ሻvariable, and call Then, the target will
‫כ‬ ‫כ‬ ‫כ‬ ଴
variable, ݀‫ݕ‬
, ሺ ଴ ሻ
16
‫ כ ݕ‬denote the desired value of the variable, ‫ݏ‬ and call ݀‫ݕ‬ǡ ‫ݕ‬ ൌ ൌ
‫ݕ‬ ‫כ‬
െ ‫ݕ‬ ଴‫ݏ‬
. Then, ‫ݕ‬
the ǡ ‫ݕ‬target ൌ will In line with the “Bey
‫כ‬the ‫ כ‬evaluation protocol. There
‫ כ‬ሻ ௬ బ is a natural way of combining the
value is defined as follows: ௚௢௢ௗ௦ ௬ బ Progress ௕௔ௗ௦
of the variable, be
andgiven
and call by:‫ כ‬ൌ ‫ݕ‬ௗ௬
݀‫ݕ‬ െ ‫ݏ ݕ‬Then, ‫כ‬the
‫ כ כ‬଴.
the଴‫כ‬target
target willwill ௗ௬
be 15 ሺ‫ כ ݕ‬ǡ ‫ ݕ‬଴ ሻisൌ
ௗሺି௬
synonymous with moving in the “right” Social Progress (Stig
బ , ‫ݏ‬௕௔ௗ௦
଴ ௗሺି௬ ሻ
by: ‫ݏ‬௚௢௢ௗ௦ ሺ‫ כ ݕ‬ǡ ‫ ݕ‬଴ ሻ ൌ బ , ‫ݏ‬௕௔ௗ௦ ௚௢௢ௗ௦ሺ‫ ݕ‬ሺǡ ‫ݕ‬ ‫ ݕ‬ǡሻ‫ݕ‬ൌ ሻ ൌబ target ௬ and making the threshold play a role. It is the following in
See: http://hdr.undp.org/sites/default/files/hdr2015_technical_notes.pdf
௬ బ measurement framew
given by: ௬ ௬ 16 direction;
In line withtherefore,the “Beyond anyGDP” observed change of an
15 approach (See Report for thisby the Commissio
decision is th
ௗ௬଴ ௗ௬ ‫ כ‬ଵ Social ௗሺି௬ሻ ௗሺି௬ ‫כ‬ሻ See: http://hdr.undp.org/sites/default/file
‫݌‬ ௚௢௢ௗ௦
ௗ௬ ‫ כ‬ሺ‫ݕ‬ ‫ݏ‬
ଵ
ǡ ‫ݕ‬
௚௢௢ௗ௦ ‫ כ‬௬
଴ ሻ
ሺ‫ݕ‬ ൌ ‫כ‬
ǡ ‫ݕ‬ , ሻ ‫݌‬ൌ ௕௔ௗ௦
ௗሺି௬ ሺ‫ݕ‬ ଴ ሻ
‫ כ‬ሻ‫ ݏ‬ǡ ‫ ݕ‬indicator
, ሺ‫ݕ‬ ൌ ‫כ‬
ǡ Progress
‫ݕ‬ ଴ ሻ will
ൌ (Stiglitz
be et
assessed al., 2010))
against
16
In line
anda
with
the
target
the
recentand
“Beyond
criticisms
towards
GDP”
an of using G
Inclusive
approach
ௗ௬ ‫כ‬ ‫ כ‬଴ ሻሺ‫ ݕ‬ǡௗሺି௬
‫ݏ‬௚௢௢ௗ௦ ‫ כ‬଴ሻ‫כ‬
‫ ݕ‬ൌ ሻ , ‫ݏ‬௕௔ௗ௦ ሺ‫ ݕ‬ǡ ‫ ݕ‬଴బሻ ൌ ௬బబ ௕௔ௗ௦ measurement ௬ బ framework ௬బ has deliberately
Social ‫ כݕ‬ൌ decided
Progress ƒš ሼnot to use
allows
‫ݐ‬ǡ
(Stiglitz ߣ‫ݕ‬et
଴ ሽthemeasures
al.,
GEP
ǡ ߣ2010))
൐ ͳ andindeo
଴ሻ
ൌ బ , ‫ݏ‬௕௔ௗ௦15ሺ‫ݕ‬See: ǡ ‫ݕ‬ ൌ ௬బ
http://hdr.undp.org/sites/default/files/hdr2015_technical_notes.pdf
బ
௬ fora this
threshold.
decision Note
is thatthat
GDP in this
not formulation,
only measures the
green but
measuring also ‘brown’
the accese
௬ ௬
towards an Inclusive Green measurement
Economy. framework
The inclusion of has deliberately
The progress
15
16
Ininlinethiswithdimension
the “Beyond is GDP” simplyapproach the (See corresponding
Report by the growth threshold,
Commission orGEP denoted
reduction
on by rate
the Measurement t, plays no
forofthis
the
of apparent
Economic
decision role soindicators
economic
isPerformance
that17 GDP not
growth
and
only
such willa
mea
undp.org/sites/default/files/hdr2015_technical_notes.pdf
See: http://hdr.undp.org/sites/default/files/hdr2015_technical_notes.pdf allows the index to capture (although partially) the green compo
Social Progress (Stiglitz et al., 2010)) and the recent and the
criticisms following
of using GDP in the
as a case
(or the of
only)“bads”:
measure of well-being,
See
the
Annex
GEP
I.A for a
“Beyond variable.
GDP” 16 approach
In lineTherefore,
with(Seethe Report applying
“Beyond thethe
by GDP” evaluation
Commission
approach on
(Seeformula
theReport inbythis
Measurement of Economic
the Commission that Performance
measuring progress
on the for
theMeasurement
access andanyto basic of towards
indicator
Economic
services, anhealth
is simply Inclusive
the
Performance Green
andratio andEconomy.
education allows Thethe
pplying
Stiglitz et the evaluation
al.,Social
2010)) measurement
and
Progress
15 the formula
recent
(Stiglitz
framework
criticisms
et
in 2010))
al.,
this
of has
using special
deliberately
andGDP the
case
asrecent
a decided
(or the gives not to
only)
criticisms
the
use following
measure
of
measures
using of GDP
well-being,
economic as
of expression
GDP
a
growth thethe
(or
in
GEP
will only)
bring
of
the calculation allows
measure
to promoting
of the
of
progress.
GEP index
well-being,
an Inclusive
The main
the to
GEP
Green
reason (althoug
capture
Economy.
special
for
See:
this case
decision gives
is thatthe GDP following
not only expression
http://hdr.undp.org/sites/default/files/hdr2015_technical_notes.pdf
measures of
green but also between
‘brown’ actual
economic and desired
activities, whichchange.
do not The
supportchoice the transition
mework has measurement
deliberately decided not“Beyond
to use
hasmeasures of GDP in the calculation of progress.
17 The main thereason ofmeasuring the access to basic services,a hp
16
In line framework
with the
/hdr.undp.org/sites/default/files/hdr2015_technical_notes.pdf deliberately
GDP” approach decided
(See not
Report to by usethemeasures ‫ כ‬See
Commission of GDP
onAnnex
the inMeasurement
I.A calculation
for a discussion progress.
of Economic ‫ כ‬The
of properties
Performance
‫ݕ‬decision
ൌand ‹main
in the reason
ሼand
‫ݐ‬ǡ basic ሽmodel
ߚ‫ ݕ‬଴toenergy ߚ ൏and
ǡpromoting ͳ an
thatLet
s the GDP now not
h15_technical_notes.pdf
“Beyond for thisdenote
GDP”
‫ כ‬towards
‫ݕ‬only progress:
measures
Social
decision
approach the
an
Progress isgreen
(See desired
Inclusive
that but
(Stiglitz
GDP
Report also value
Green
et
not
by ‘brown’
al.,
the 2010))
only of
Economy. the
and variable,
economic
measures
Commission
Therecent
the inclusion
activities,
ongreen and
but
the Measurement
ofcall
which
criticisms
also
indicators
do݀‫ݕ‬
of not
using
‘brown’
of ൌ
of ‫ כ ݕ‬as
such
support
GDP
economic
Economic െas
is,the
a (or. Then,
଴ transition
green
activities,
Performance only)the
the trade,
‫ݕ‬therefore, angreentarget
measure
which
and do of will
innovation
important
economic growth
well-being,
not support the
renewable
of
will
GEPthe
bring
transition
eportGreen
sive by theEconomy.
Commission allows
The on the GEP
the Measurement
inclusion of index
indicatorsto of capture
Economic
such as (although
Performance
green trade, partially)
and
green the
innovation green andcomponents
renewable of
energy GDP. In 17
addition,
See the
Annex inclusion
I.A for a of indicators
discussion 15of proper
ent beto given
ess (Stiglitz
criticisms
et by:2010))
al.,
towards
of using
measurement
and the recent
an Inclusive
GDP
measuring as a the
framework
(or the
Green
accessonly)
has
criticisms
Economy.
measure
deliberately
of using decided
GDP as aof
The inclusion
of బwell-being,
not
the andGEP
to use
(orindicators measures
the only) measure of GDP
suchevaluation
as green in the
of well-being,
trade, calculation
protocol. the GEP
green of progress.
innovation
toThere is the The main
and renewable
asupport
natural way reason
of energy
index
nt framework capture
has (although
allows for this
deliberately
the GEP
partially)
decision
decided
index is the
that ௣to
green
GDP
notcapture
to to use basic భservices,
components
not ሺ௬only
measures ǡ௬ measures
(although
of health
ሻ ofpartially)
GDP GDP.green
in ௬ In
భ ି௬
the
the
education
addition,
but బalso ‘brown’
calculation
green ofallows
the inclusion
economic
progress.
components
the of
of
GEP
The index
indicators
activities,
GDP.main which
Inreason
capture
do not
addition, the
positive aspects
the transition
inclusion of indicators
that higher
not totouse
cess
sion is that GDP
measures
basic services,
not
of
towards
only
GDP
health
economic in
measures and
an growththe
Inclusive calculation
education
will
green Green
but
೒೚೚೏ೞ
bring of
allows progress.
Economy.
also the
to promoting
‘brown’ GEPൌ
The
The
ௗ௬
anmain
index
economic ൌ reason
to capture
Inclusive
inclusion of This
Green the
indicators
activities,
formulation
whichpositive
Economy.
suchdo aspects
combining
as
notgreen
support
indicates
that
trade,these
thehigher
green two that
aspects,
innovation
transition
countries
and the choice
renewable must of
energy have a desirable
the
measuring anthe access to basic services,
௦discussion
doሺ௬ బhealth
‫ כ‬ǡ௬support and education
‫ כ‬transition‫ି כ‬௬ బ allows the GEP index to capture the positive aspects that higher
reen but also
will bring ‘brown’
to promoting 17 economic
See
allows Inclusive
Annex
the GEPactivities,
I.AGreen
index which
for൞of Economy.
ato capture not ofሻ properties
(although ௗ௬
the
‫כ‬partially)in௬ the
thethebasic
green model
“right” and
components side
‫כ‬aሻ proof
of
ofrenewable
GDP. of
the the characterization
threshold
In energy
addition, the inclusion result.
(or,play atofathat
arole. minimum, a target
indicators
thethe[1]
೒೚೚೏ೞ
asௗ௬ ௗሺି௬
asܲ‫ݏݏ݁ݎ݃݋ݎ‬ ൌ
nclusive Green Economy. The inclusion indicators such green trade, green innovation and
on of
for
GEP a indicators
discussion
index to
economic
such
of growth
greenintrade,
properties
measuring the
17capture (although partially)
will
the bring
‫ݏ‬
green
basic
access model
௚௢௢ௗ௦to
௣the ሺ‫ݕ‬
toinnovation
basic
‫ כ‬భ a଴
promoting
andǡ ‫ݕ‬ and
services,
బ ሻ an
ሺ௬ ǡ௬ components
green ሻ ൌ
proof Inclusive
renewable
of
health
ௗሺି௬ሻబ
the, ‫ݏ‬ Green
energy ሺ‫ݕ‬
characterization
and௕௔ௗ௦
of education
௬
GDP. బ ି௬
‫כ‬Economy.
Inభǡ ‫ݕ‬ ଴ ሻ ൌ target
result.
allows
addition, 15the
the GEP
బ
and
index
inclusion
making
to
of capture
indicators
threshold
positive aspects higher
ally) the green See Annex I.A for a discussion ್ೌ೏ೞ of propertiesൌ ௬ in the ൌ basic model and a ௬
proof of the characterization result.
he education
access to allows
components
economichealth
basic services,
of GDP.
growthand In addition,
bring15
will education
௦್ೌ೏ೞ
the
to promoting inclusion
‫כ‬allows an GEP
the
ǡ௬ బ ሻ aspects
of indicators
Inclusive Green
ௗሺି௬ ‫כ‬ሻindex ௬to
“goods”,
బ ି௬ ‫כ‬Economy.
capture countries
the positive aspects should
that higher never be below the threshold in the
nd the
17
SeeGEP AnnexindexI.A to for
capture
aGreen theሺ௬
discussion positive
of properties that in thehigher
basic model 15 and a proof of the characterization result.
owth will bring
ve Green Economy.
xheI.A for amodel
to
discussion
promoting an Inclusive Economy. they15should never be above the threshold in the final period. Ev
basic and a of properties
proof in the basic model
of the characterization result.and a proof of the characterization result.
ress in the 15 single indicator case corresponds 15 “right”
to the ratio of side of theand
the actual threshold,
desired they should still be making pr
15
for 16the
See: case of “goods”) or reductions (for the case of “bads”). The progress
http://hdr.undp.org/sites/default/files/hdr2015_technical_notes.pdf determined (see Section 4.5 below). This formulation provides a
In line with the “Beyond GDP” approach (See Report by the Commission on the Measurement of Economic Performance and
“bads”
SocialisProgress
obtained byetreversing
(Stiglitz al., 2010)) andthat of the
the recent “goods”,
criticisms of using threshold
both
GDP in
as athe is only)
a relevant
numerator
(or the measureandelement.
of in the The
well-being, the GEPrationale of this method of
measurement framework has deliberately decided not to use measures of
. for this decision is that GDP not only measures green but also ‘brown’ economic case ofactivities,
GDP in “goods”,
the calculation any sensible target can be expressed as ‫ כ ݕ‬ൌ
of progress. The main reason
which do not support the transition
towards an Inclusive Green Economy. The inclusion of indicators such as the initial
green valueinnovation
trade, green of theand variable). This is admissible as long a
renewable energy
allows the GEP index to capture (although partially) the green components of GDP. In addition, the inclusion of indicators
s function
measuringisthe increasing
access to basicand linear
services, in and
health ‫ ݕ‬for
ଵ
the allows
education casethe ofabove
"goods",
GEP theto threshold.
index and decreasing
capture Otherwise,
the positive andthatone
aspects should require ‫ כ ݕ‬ൌ ‫ݐ‬, as t
higher
economic growth will bring to promoting an Inclusive Green Economy. In other words, for “goods”,଴if a country is initially below the thr
or the17 case
See Annexof I.A"bads". The ofderivative
for a discussion of basic
properties in the the model
progress function
and a proof with respect
of the characterization result.to ‫ ݕ‬is
 of gative
pending
f regress.
mparisons
mparisons regress. otherwise.
onforforwhethereachTrivially,
each actual progress
indicator
indicator the canfunction
can be is above
be is decreasing or below the in ‫ݕ‬ . The
target.
epending
ess.
re
ntnt several
or
or the theon whethermade
countries,
progress
progress actualon
international
made on progress each
each comparisonsis above orfor below eachthe indicatortarget. can be
synonymous
ress.
several
several countries,
confronting countries, with
their rates moving
international
international in
of achievement the “right”
comparisons
comparisons direction;
by
or the more
for
forprogress therefore,
each than
each indicator one
indicator made any can
still observed
results
can
on each be in change
be a value of that is lower than the value of th
al inwill
nfronting the be“right”
countries,
nfronting assessed
their
their direction;
international
rates
rates against ofoftherefore, a targetany
comparisons
achievement
achievement andobserved orafor
or threshold.
target
the
the each progresschange
for
progress indicator Note
this made of that
made can in
country onthis
be
on should
each
each formulation, be to atthe least reach the threshold. In the
enoted
aing
raltarget theirand
countries, by t,
rates plays
ainternational
threshold. no
of achievement apparent Note comparisons that role
or in so
thethis thatfor progress
formulation,
reasoning
progress each indicator made forthe
is symmetric. any
on can each indicator is simply
beA target takes the form ‫ ݕ‬ൌ ‫ ݕ‬, with  ൏ ͳ (i.e. a the ‫כ‬ ଴

en
arent
nting actual by and
role
their more
so
rates desired
that than
ofprogressonechange.
achievement stillforresults The anyor choice in a value
indicator
the of ‫ݕ‬is is,
value
progress
‫ככ‬
that
ofsimply
by therefore,
the is variable).
more
made lower
theon than an
than
each important
one
This the isvalue
still results
admissible decision
of the of
in athreshold, value that
provided then
‫ݕ‬is ൏
଴ lowerthe
‫ݐ‬, i.e.,than a country the value is
hresholds
on
ge.re protocol.
Thetarget
than one There
choice forofresults
still ‫ݕ‬is is,
this ‫כ‬ acountry
natural
in a value
therefore, way
should an ofimportant
that combining
beis Otherwise,
tolower
by at
more leastthese
thantarget
decision than onereach
the two for
one
of value
should aspects,
the
stillof
this thethe
threshold.
country
results
require ‫ݕ‬choice
threshold,
inshould
‫כ‬ aInvalue
ൌ ‫ݐ‬,the asofthen tthe
becase istothe
that at ofmaximum
is “bads”,
least
lower reach
than the thevalue
the
admissible threshold. of
value. the
resholds
;
making
uralesholds
therefore,
therefore,
for way this reasoning
theof any
any
threshold
country
combining observed
observed
It isis symmetric.
should
the playthese
following change
bechange
a role. to
twoin A at
the of
of
target
Itaspects,
iscase
leasttheof takes
following
reachtarget
the
“goods”: the the
choice form
inthreshold.
reasoning
for the
this
of‫ݕ‬the ‫כ‬
case ൌ
country is‫ݕ‬of ଴
, “goods”:
with
symmetric.
InOnce the
should the case ൏target ͳbeA(i.e.
of target aatdecrease
to“bads”,
is formulated takes
least thethe reach
this inform thethe
way, initial
the
‫כ‬
ൌ ‫ ݕ‬, withIn the
threshold.
‫ ݕ‬following ଴
൏ͳ
us d.
ning lds
d. with
Note
Note
role.is12 moving
that
that
value
Itsymmetric. in
in
of
is the following in
this
this
the the
A target “right”
formulation,
formulation,
variable). in the bytakes direction;
This
more of
case the
the is form
than therefore,
admissible
“goods”: one ‫ݕ‬reasoning
Once ‫כ‬
ൌstill the by any
provided , more
value
଴
‫ ݕ‬results with
target observed
is symmetric.
of in
than
is
൏the
‫ݕ‬ ଴
(i.e. change
variable).
one
aformulated
ͳvalue
൏ ‫ݐ‬, i.e.,
expression aAstill
that a
decrease
target of
country
This
is
thisis by
results takes
lower inin
ismore
way,
obtained: is thethe
admissible below
a value
than
the initial
thanform theone
following the
that
‫ݕ‬ ‫ כ‬threshold.
provided
value ൌstill of‫ݕ‬ ,the
results
isexpression
lower
଴
with
‫ ݕ‬than
଴
൏in‫ݐ‬, isathe
threshold,
൏ i.e., (i.e.
value
obtained:
ͳ a then
value athd
cou
olds
with
of with
essed
ress
ress themovingmoving
for
for against
any
Otherwise,
variable). in
inathe
any indicator
indicator the
target
Thisone “right”
“right”is
is
isshouldand
admissible direction;
‫ככ‬direction;
simply
simply
target a threshold.
require
‫ ݕ‬ൌ ƒš ‫ݐ‬ǡ ߣ‫ ݕ‬ǡ ߣ ൐ ͳby by the
the
more
for more
ሼ ‫ݕ‬therefore,
therefore,
provided ‫ כ‬଴଴Note that
this ൌthanthan
ሽ valueas
country
‫ݐ‬,  one
‫ݕ‬ t
one଴any
any
is
of
target
൏ the
still observed
in
theobserved
Otherwise,
still
should
‫ݐ‬, this
maximum
i.e., a
variable).
for
results
results be this one
country change
formulation,
in
tochange
in aadmissible
should
This
country
a
at is
value
value
least of
isof
below therequire
admissible
should
thatthat
reach value.
target
the
is is be
lower
‫ݕ‬ ‫כ‬
lower
the for
threshold.ൌ this
provided
to as
at
thanthan
threshold.
‫ݐ‬, t country
leastis
the
the the
‫ݕ‬ reach
଴
value
value
In ൏ should
maximum
the ‫ݐ‬, i.e.,
of the
of the
case a
the becountry
threshold. to
admissible
threshold,
threshold
of at
“bads”, islea Inb
moving
ssed‫כ‬sed
, plays
, therefore,
wise, therefore,
ൌ ƒš against
againstin
no
oneሼ‫ݐ‬ǡshould the
ߣ‫ݕ‬ a a
apparent
an ଴ “right”
target
target
an ሽimportant
important
ǡߣrequire direction;
and
and
role
൐ ͳ ‫ݕ‬decision
‫כ‬ a
soa
decision threshold.
threshold.
that
ൌ ‫ݐ‬, target
reasoning therefore,
as progress
targettof
ofis is the
for Note
Note
for any
maximum
symmetric. forthat
that
Otherwise,
thisthiscountryobserved
any inin
reasoning
country this
this
indicator
Aadmissible
one
target change
formulation,
shouldformulation,
should should
istakes is
symmetric.
be beof
simply
value. require
totoat
the the
the the
atleast
form ‫ݕ‬A ‫כ‬
least ‫ݕ‬reasoning
target
‫כ‬
ൌ ‫ݐ‬, ൌ as
reach
reach
ௗ௬ t ,isthe
‫ݕ‬takes
଴ is
the
the
with symmetric.
the maximum
form
threshold.
௬threshold.
భ
൏ି௬ͳ (i.e.
బ
‫ ݕ‬A ‫כ‬
ൌ target
the
admissible
aIn In ଴
‫ݕ‬the
decrease , with
case takes
case value.
in of the
of
൏the ͳ“ba f
(iin
“b
plays moving
gainst
plays no
no a in
targetthe
apparent
apparent “right”
and a
role
role direction;
threshold.
so
so that
that therefore,
Note
progress
progress that for
for any
in anythis
any observed formulation,
indicator
indicator change is
is simplythe
simply of the
the ‫ כ‬ൌ ௠௔௫ሼ௧ǡఒ௬ బ ሽି௬ బ ǡ ߣ ଴൐ ͳ
owing d desired
hese two
these two
inOnce
the change.
aspects,
aspects,
the
case and target
of The
the
the
the “bads”:ischoice
choice
choice
following formulated
value of
in of
of
reasoning ‫ݕ‬
the
the
reasoning
of
the
‫כ‬
the
caseis, therefore,
thisvariable).
ofway,
is“bads”: thevalue
issymmetric.
symmetric. an
following
ThisOnceis important
of AtheAthe expression
targettarget
target
admissible decision
variable).
takes is formulated
takes is
the
provided This
the of
obtained:
form value ‫ݕݕ‬this
‫ݕ‬of
isቐadmissible
form ଴ ‫כ‬
ௗ௬ ൏‫כ‬
ൌൌway,the
‫ݕ‬i.e.,
‫ݐ‬, ଴
, the
‫ݕ‬variable).
଴
, provided
with
awith following
൏൏This
country ͳͳ (i.e. expression
‫(ݕ‬i.e.
is is൏aadmissible
below i.e.,
decrease
a‫ݐ‬, theathresh
decrease is
coun obt
in pro
int
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ ൌ
noagainst
desired
: desired
in apparent
There
in
he the
the
target a
is
case
case atarget
change.
change.
is role
natural
of
of
formulated and
“goods”: so
The
The
“goods”: way a
that
choicethreshold.
choice of
this progressway,of
combining
Otherwise,of ‫ݕ‬‫ݕ‬
value
‫ ככ‬Note
the
valueofis,for
is, following
ofthe
one that
any
therefore,
therefore,
these the Once in
indicator
two this
anan
expression
variable).
should variable). the
Otherwise,
require formulation,
important
aspects, is
important
target
ThisThis simply
‫ ݕ‬isthe
isis
‫כ‬ decision
obtained:
one
ൌ the
decision
choice
formulated
admissible the
should
admissible
‫ݐ‬, as of
t is the of
of the
this
require
provided
provided
maximum way,
Otherwise,‫ ݕ‬ൌ
ௗሺି௬ሻ
‫כ‬ the
‫ݐݕ‬, following
one
as ൏‫ݐ‬,tబ‫ݐ‬,is
‫ݕ‬admissible
଴ ଴
ൌ൏
௬ బ ି௬ భ
should
i.e., the
i.e., expression
aamaximum require
country
country
value. is
‫ כ‬obtained:
ǡ ߚ ൏is‫ݕ‬isͳadmissible ൌ ‫ݐ‬, as
below
below the t isth
the vt
ௗሺି௬ ‫ כ‬ሻ ௬ ି௠௜௡ሼ௧ǡఉ௬భబ ሽ బ
here nochange.
dhere
hreshold apparent
is
is aa natural
playTherole
natural achoice way
way
role. soItof that
ofof ‫ݕݕ‬progress
iscombining
the ‫כככ‬
combining is, therefore,
following
Otherwise,
Otherwise,
ൌ ‹ሼ‫ݐ‬ǡ ߚ‫ ݕ‬ሽ ǡ ߚ ൏‫ͳ כ‬ൌ for
these
these in any
଴଴ an
thetwo
two
one one indicator
important
aspects,
case
ௗ௬ aspects,
should
should of ௬is
“goods”:the
the
భ ି௬ బ
require
require simply
decision choice
choice
‫ݕ‬ ‫ݕ‬‫ כ כ‬of
ൌ the
ൌ ‫ݐ‬,of
of
‫ݐ‬, as the
the
as t is
t is the the maximum
maximum ௗ௬admissible
admissible ௬ ି௬value. value.
ǡ ߣ൐ͳ ൌబ ǡ ߣ൐ͳ
‫כ‬ed change. The choice of is, therefore, an important decision of
‫כ‬ బ ሽି௬ బ
seshold a natural way of combining these two aspects, the choice of the ௗ௬௬‫כ‬భ ି௬௠௔௫ሼ௧ǡఒ௬ బ ሽି௬ బ
eshold play
play
ൌ ‹ሼ‫ݐ‬ǡ ߚ‫ ݕ‬ሽ ǡ ߚ ൏ ͳ aa role.
଴role. It
It is
is the
the following
‫ݕ‬ following
Once ௗ௬
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ the ൌ in
in
target the
the
ൌ ௬ భቐି௬ case
iscase ௗ௬
బ
formulated ofof
Once “goods”:
“goods”:
௠௔௫ሼ௧ǡఒ௬
ǡ ߣ ൐ ௬ͳబ ି௬ the this target way, is the formulated
following ܲ‫ݏݏ݁ݎ݃݋ݎ‬ Once this the
expression
ௗ௬ way,
ൌ target
ቐ
ൌ ௗሺି௬ሻ బand the isis formulated
following
obtained:[1’] ǡ బߣି௬൐భ ͳ in ‫ݕ‬way,
expression this ଵ is the obta fo
The ௗሺି௬ሻ function ofభ progress is alwaysௗ௬increasing ௠௔௫ሼ௧ǡఒ௬ൌሽି௬ బ ௬linear in the c
isplay a natural way ‫ כ‬of combining these two aspects, the choice of the
‫כ‬ బ ሽି௬ బ ‫כ‬
dation a role.
indicates‫ݕ‬This It is thatthe following
ൌ formulation
ƒš
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ ሼ
countries ‫ݐ‬ǡ ߣ‫ݕ‬ ଴ ሽin
ൌ ቐindicates
ௗ௬the
Once
ǡ ߣOnce
must ൐ ͳhave case
௠௔௫ሼ௧ǡఒ௬
the thattarget
the of
target “goods”:
adecreasing
desirable is‫ כ‬is
ൌ
formulated
formulated change by more thisor ǡ
ሽ way,
this ߚ ൏
thanthe
way,
target
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ ͳ the
one ‫כ‬‫כ‬following
following
set ൌstill results
ቐtoௗሺି௬ሻ be[1’] expression
expression
on in ‫כ‬aሻబ value is
௬ బis obtained:
obtained:
that isబሽlower than th ǡ ߚ ൏ ͳ
countries
ି௬ భ ሻ must and linear
௬ బ ି௠௜௡ሼ௧ǡఉ௬ in ‫ ݕ‬ଵ in‫ ݕ‬௬the భ ି௬case of “bads”. ௬ ି௬The భ ି௬derivative of progre
బௗሺି௬ బ ௗሺି௬ భ ି௠௜௡ሼ௧ǡఉ௬
d play a role. ‫ככ‬It is theሼሼfollowing ଴଴ሽሽ ௗሺି௬ሻ in the ൌ case
௬
of “goods”: ǡ ߚ ൏ ͳ ൌ ǡబ ߚ ൏and ͳ
side es must of the have ‫ݕ‬‫ݕ‬ ൌൌ
threshold ƒš
ƒš
a desirable
have ‫ݐ‬ǡ‫ݐ‬ǡ
a desirable ߣ‫ݕ‬
ߣ‫ݕ‬
(or, change ǡ ǡߣ
atchangeߣ ൐൐
a minimum, ͳͳ‫ כ‬ሻor ortarget
target a‫ ݕ‬target
‫כ‬
set
set బ ሽ to to beofbe on target
reaching
on for
The
ௗ௬
‫ כ‬the this
function
ൌ‫ݐ‬௠௔௫ሼ௧ǡఒ௬ country
threshold).
బ
‫כ‬of progress should
଴ǡ ‫כ‬Forሻߣ is
൐
ௗ௬
బbe
௬always
ͳ ൌ toincreasing
atబሽleast
௬
ǡreach
ߣ ൐ ͳthe
ௗ௬thre
ൌ
‫כ‬ ଴
ௗሺି௬ ௬ బ ି௠௜௡ሼ௧ǡఉ௬depends on whether ‫ݕ‬ ‫כ‬ ൌ or ‫ݕ‬ ൌ బ 
ሽି௬
ௗሺି௬
‫ݕ‬ బ , but ௗ௬ in‫כ‬ either
ି௠௜௡ሼ௧ǡఉ௬
௠௔௫ሼ௧ǡఒ௬ case బ ሽି௬ it
బ is negative ௗ௬ ‫ כ‬in௠
case
untries
at a minimum, ‫ݕ‬of ൌ“bads”:
should
The ƒšfunction ሼ ‫ݐ‬ǡ
never ߣ‫ݕ‬
a“right” ሽ ǡ ߣ
beside
of
target ൐
below
progress ͳ
ofof the reachingthe isthreshold
always the in
increasing
threshold). the Thefinal reasoning
functionand period,
ൌ ቐlinear
ௗ௬
ofwhereasis ௗ௬ symmetric.
ௗ௬ܲ‫ݏݏ݁ݎ݃݋ݎ‬
in“bads”, ଵ
in బforthe
భ
௬ ௬ି௬ భ ି௬
“bads”,
ൌAబ
case
బ
target
ቐofௗሺି௬ሻ ǡ of takes
‫כ‬൐“goods” the form
బܲ‫ݏݏ݁ݎ݃݋ݎ‬
and ‫ݕ‬ ‫כ‬
ൌ 
ൌ ቐ ௗሺି௬ሻ ‫ݕ‬ ଴
,
in ‫[ݕ‬1’]in with
inFor progress isభcase always ߣincreasing and linear
the threshold (or, at ܲ‫ݏݏ݁ݎ݃݋ݎ‬
a minimum, linear ‫ݕ‬ൌ in the “goods” and ௬decreasing
ଵ
and positive the case of ‫ כ כ‬ൌ௠௔௫ሼ௧ǡఒ௬ whenever బ బ ǡ‫് ߣ ݕ‬ ൐ͳ‫ݐ‬, ͳ (i.e. lower initial଴ͳvalues
ି௬ భ
‫כ‬ ଴ሽ ௗሺି௬ሻ ௗ௬ ௗ௬ ௬௠௔௫ሼ௧ǡఒ௬
ି௬ బ ሽି௬బ ሽି௬
ase
ase neverthe
unction of
of “bads”:
“bads”:
‫ݕ‬
be ൌabove
decreasing
of ƒš
progress ሼ ‫ݐ‬ǡ
the ߣ‫ݕ‬inthreshold
and is ǡ ߣ
linear
always ൐
finalinin ͳ the in
increasing
ଵ
final
the case period.
The and of Even
“bads”.
linear
function ܲ‫ݏݏ݁ݎ݃݋ݎ‬ifof
ܲ‫ݏݏ݁ݎ݃݋ݎ‬
in value
countries
The of
derivative
inand
progress ൌ theൌ the
ቐ ቐ ሻare
ൌ
casevariable).
௬already
of
isprogress of
always ‫ݕ‬progress This
“goods”on ǡthe
బincreasing
ߚ
is with ൏
and ͳ
admissible ൌ
respect and provided
to
linear ଴బ ǡ ߚଵ൏
‫ ݕ‬ሽinderivative ‫ݕ‬ ൏
inௗሺି௬ the ‫ݐ‬,‫כ‬ofi.e.ൌp
decreasing ‫ݕ‬ሻ‫כ‬caൌ
ଵ
elow threshold a target the
of reaching period,
‫ݕ‬the threshold). whereas “Progress”For for
“goods”, “bads”,
for ‫ݕ‬each and
ௗሺି௬ given linear
‫כ‬ linear
ௗሺି௬ሻ
ௗሺି௬ሻ
బin
inି௠௜௡ሼ௧ǡఉ௬
ଵ
in௬ బ௬the
the
ି௬
బ
in
ି௬ሽభcaseభcase
the
ௗሺି௬ ofሻof
‫כ‬
case“bads”. “bads”.
బ
ofThe
௬ ି௠௜௡ሼ௧ǡఉ௬ The
derivative
“goods”). ‫ ݕ‬When
“bads”: depends final‫ݕ‬countries
‫כ‬
ൌ ‫‹ݕ‬
on inሼEven
whether ‫ݐ‬ǡthe ߚ‫ݕݕ‬ ଴
ifሽnever
ǡcountries
ߚ still‫ݐ‬൏ orͳ“bads”. the ,threshold
but inon either Otherwise, case itௗሺି௬one
is negativeൌ
should ൌ require
in the బ‫ݕ‬
ǡ ǡ
‫כ‬ߚ
ଵሽ‫ ݕ‬,ൌbut
ߚ ൏ ൏ ͳ
‫ݐ‬,The ͳ
as teither
is the maximum adm
‫ݕ‬ሽcase ‫ݐ‬,of “goods”,
‫כ‬ ‫כ‬ ଴
asingoldof in the and
the linear
threshold, in
period. ଵ
they should
should case ൌ ofbe ‫ݕ‬beൌare
below making ‫ݕ‬The
decreasing
already
“Progress” derivative
depends
progress and
withinthe of
on
linear
and
respect progress
whether
ofin progress
toand
‫ݕ‬ ଵ
ௗሺି௬
‫ݕ‬in ‫ݕ‬ሻwith
‫כ‬଴ ‫כ‬
‫כ‬ሻthe
is ൌneed
with ௬ బ‫ݐ‬ respect
బor
௬case
positive respect
ି௠௜௡ሼ௧ǡఉ௬ ‫ݕ‬toof
ି௠௜௡ሼ௧ǡఉ௬
‫כ‬
ൌifbe
బ
to
“bads”.
to ‫ݕ‬ ଴ ଴
൐ depends zeroin derivative ଵ case
ifon‫ݕ‬whether ൌ ‫ݐ‬, and ofit progres
isnegativnega
f
ble
ble
ds “bads”:
change
change
on and be
whether ‫ݕ‬or ‫ככ‬
‫ݕ‬or4.5
positive ൌtarget
ൌ ‫‹ ݕ‬
target‫כ‬‹
ൌin‫ݕݕݐ‬ ሼ
ሼ‫ݐ‬ǡ‫ݐ‬ǡ
or
‫כ‬
‫כ‬
the ߚ‫ݕ‬set
ߚ‫ݕ‬set ଴଴ሽሽ
case
‫כ‬
Theൌto ǡwhereas
to ǡߚ be
ߚformulation
൏
be
of൏,and
଴
functionon
ͳͳbut
on
“bads”, forin whenever
either
of dependsprogress caseThe and ‫ݕ‬on ‫כ‬
positive
aitfunction
is is
whether (i.e. of
negative
always lower
in
of the
‫כ‬
ininitial
progress
increasing case
‫ݐ‬theor so‫ݕ‬case of
‫כ‬values
Theis“bads”,
and of
always ଴ yield
‫ݕ‬function but
,linear inain
whenever
,“goods”, higher
of
increasing
in ‫ ݕ‬progress
either in‫ݕ‬value
‫כ‬
്itand
case
the of
is caseitislinear
(i.e. is always
lower
negative ‫ݕ‬incre
ininitial ଵ
in vtt
a of “goods”
ଵ
hould (see Section
still making
the below).
final period,‫ݕ‬This
progress ‫ݕ‬ and
“bads”, provides
words, 
they need should
crossing method
്
to ‫ݐ‬,be
the
‫ݕ‬doing
threshold
ൌ in
in
ൌ which
the right butthe either
direction case
involves
‫ݐ‬, negative
premium. T
get
get tes
ositive
his a relevantthat
of
of in‫ݕ‬
formulation
‫כ‬
countries
ൌ
reaching
reaching ‹
“Progress”
theelement.
case ሼ ‫ݐ‬ǡ
never ߚ‫ݕ‬
the
theof
providesmust
for
଴ሽ
The
be ǡ
“bads”,ߚ
threshold).
above have
threshold).
each ൏ ͳ
a decreasing
rationale
methodthegiven
whenever a The desirable
TheFor
For
threshold of progress
offunction
function
‫ݕ‬and
this
doing
‫כ‬
in് change
and
method
thelinear
‫ݐ‬,so in
(i.e.
of of
final inthe
positive or
decreasing
progress
progress
inof
period.
which target
case
“Progress”
lower ଵ
‫ݕ‬defininginin Even
theOnce
initial the
the is ‫ݕ‬
ofisand
‫ כ‬the
set
“goods”).
for
values
case
always
case
‫ݕ‬ always
‫כ‬‫כ‬ inis to
target
each
linear
theofyield
the be on
increasing
case is
When
ingiven
“bads”,
increasing
“bads”.
following.‫ݕ‬
of formulated
a decreasing
ଵ higher
in ‫ݕ‬the
The
“goods”, ‫כ‬
progress
whenever and
In value
andcase
‫ݐ‬, linear
ൌderivative
the
and this
theand
‫ݕ‬ ‫כ‬
linear
positive way,
of്derivative
in
of the
linear
“bads”.
inin
‫ݐ‬,of in‫ݕ‬ the
case
(i.e. ଵ ଵ
progress
‫ݕ‬
the following
lower
in The
in
case of
‫ݕ‬theof
ଵ
the “goods”).
initial
inderivative
of the
case
withcase expressio
values
caseofofof
respect Whe
of
“goo
“go yi“t
pr
‫כ‬ decreasing in y* for the case of "goods", and increasing for the case of “bad
sods”,
hesess”
he
tionalethat
that
threshold
final
final countries
anycountries
‫“ݕ‬Progress”
for of ൌ
period,
period,
each(or,
‹
sensible
this methodmust
ሼatmust
‫ݐ‬ǡ a ଴minimum,
ߚ‫ݕ‬
whereas
whereas
ifgiven with
target
countries have
ሽhave
ofǡ ߚare
for
respect
progress for
can ൏ aalready
aͳ“bads”,
depends
defining desirable
bedesirable
“bads”,
to aexpressed
in
decreasing target
decreasing
‫ݕݕ‬on ଴‫כ‬
the is
on isthe change
change
positive
case
whether
the of
“right”andand
as reaching
“Progress”ofside
following. ifor
‫ݕ‬or
‫כ‬depends
‫ݕ‬linear
linearൌ target
ଵtarget
‫“כ‬goods”).
‫“ݕכ‬Progress”
ൌ
of ൐ ‫ݐ‬in
the for
‫ݕ‬In ଴the
in
or
‫ݐ‬,଴‫ݕ‬zero
,the‫ݕ‬ଵ‫ݕ‬
‫ݕݕ‬on
withଵ ‫ככ‬
When
each
‫כ‬in set
in
ൌset
threshold).
if
with
whether
the
the ‫ݕ‬totoଵ ‫כ‬be
given
case
଴
“bads”,
൐ be
respect
case
‫ݕ‬,ͳ(i.e.
ൌ but ൌ‫ݐ‬,‫ݕ‬ on
on ‫כ‬For
and
progress
‫ݐ‬,of
inofൌ
an
whenever the to negative
‫ݕ‬derivative
‫ݐ‬depends
“bads”. or
“bads”.
either increase ‫ݕ‬is
଴ ‫כ‬
inThe
case ൌ The otherwise.
positive
theon of
case
଴ whether
in‫ݕ‬itderivative
,is but
derivative
(i.e. if lower
‫ݕ‬inଵof൐
negative In
either
of
‫ݕ‬ ‫כ‬other
of
‫ݐ‬, zero
“goods”).
progress
ൌ
initial in orif‫ݕ‬case
‫ݐ‬case
progress
the
values ‫כݕ‬When
itଵൌ is
with
ൌ of
with ‫ݐ‬, ଴and
‫ݕ‬,“goo
‫ݕ‬negativ ‫כ‬
but
respe
respൌn
uld
an countries
hreshold
hreshold
Even
alueEven
ess” never
beof if
if
with must
(or,
(or,
be
countries
countries
words, respect
the variable).
expressed at
at
below have
a
crossinga
threshold, minimum,
are
are
to
as ‫ ݕ‬Thisthe
‫ݕ‬‫כ‬ a
minimum,଴ desirable
already
alreadythreshold
the
is
ൌthey positive
‫ݕ‬and should
଴ on
threshold
is, with a
a
on depends change
target
target
the
the
depends
positive
admissible in
if
still ‫ݕ‬ the
ଵin
 ൐beͳ(i.e. of
of
inon
൐ theon or
final
the
‫ݐ‬,
making as zerotarget
reaching
reaching
right
“Progress”
whether period,
whether
case
anlong if
and
progress ‫ݕ‬
increaseof‫ݕ‬
as
ଵ ‫כ ככ‬
the
direction
words, ‫ݕݕ‬the set
whereas
with
positive
“bads”,
ൌ ൌ ‫ݕ‬ൌ‫ݐ‬in‫ݐ‬or
‫ݐ‬,
and ଴
଴ to
threshold).
threshold).
൐ be
involves
crossing
andrespect in
orwhenever
‫כ‬
‫ݕݕ‬i.e.
‫ݐ‬, on
for
negative
the
‫כ‬
ൌൌwhen
yield the
to
case aFor
For
“bads”,
‫ݕ‬a‫ݕ‬higher
଴‫ݕ‬ ଴premium.
଴
threshold
otherwise.
, ‫ݕ‬,butis
of
but
‫כ‬
a in
് positive
and
“bads”,
ineither
‫ݐ‬,
value
country either
(i.e. Trivially,
in
positive
oflowerInifthe other
whenever
case
‫ݕ‬
“Progress” ଵ
iscase right
in
൐ “Progress”
the
ititis
initial
‫ݐ‬, ‫ݕ‬
for ‫כ‬ௗ௬
direction
zero case
negative
isvalues if
negative
്
each ‫ݐ‬, of
‫ݕ‬
ൌ given ଵ is
(i.e.
௬
“bads”,
yield
ൌ
భ ି௬ బ
involves
lower
ininthe
‫ݐ‬, and
aబthe a
negative
whenever
initial
case
higher case prem
ǡ ߣ valu ൐ val
ofof ͳ
tprogress
countries must have a desirable change Theor index
target takes
‫כ‬
set on
to bea value on above or below one, depending ௗ௬ ‫כ‬ ௠௔௫ሼ௧ǡఒ௬ on whether ሽି௬ బ act
bove ldis
dhreshold.
d never
,progress (or,
never
crossing the
admissible at
be
be abelow
andbelow
threshold
decreasing
the
and minimum,
Otherwise,  the
threshold
 andthe
in
and
and
as long in y*the


oneas a
threshold
threshold
for in
need
need
need target
final
the the
“Progress”
should to
‫ ݕ‬൐require ଴period.
case
to
to
andbe in
rightin
and of
bebe the
the
of reaching
positive
positive
determined
‫ݐ‬, i.e. ‫ݕ‬ final
forfinal
Even
"goods",
direction when words,
in
‫ככ‬each period,
ifin
(see the
period,
countries
theand
involves
the threshold).
“Progress”
case
given
‫ݐ‬,aascountry case
Section whereas
t thewhereas
‫ݕ‬
increasing
decreasing
crossing
is the ofa are
of
progress
4.5 premium.
“bads”,the
for
“bads”,
isminimum for
for
already
in For
y* “bads”,
each“bads”,
the
threshold for
whenever
in
progress wheneveron
thecase
the
Trivially, the
given
admissible case
in case
in of
the “bads”.
‫ כ‬of
“Progress”
‫ݕ‬ ܲ‫ݏݏ݁ݎ݃݋ݎ‬
“Progress”
the
‫כ‬progress
‫ݕ‬ ്of
case ് ‫ݐ‬, "goods",
right (i.e.
(i.e.
“goods”).
‫ݐ‬,
of
value. direction
“goods”). for
in
lower is
lower ൌ and
the ቐ
each
When
When increasing
involves
case
initial
initial given
‫ݕ‬valuesof
‫כ‬values
ൌ a for
progress
“goods”).
‫ݐ‬, yield
theyield the
premium. a a case
in
When
higher
derivative highe Tr
theo
orൌbelow
బ భ
target. ItFor is positive in
଴ case of progress ௗሺି௬ሻ and ൌ negative௬ ି௬ in ǡthe ߚIn c
veold
sr
ve
eshold,
asings
rds,
hould be
aathe
the (or,
below
methodin
method
ds, forrequire y*at
threshold
threshold
they for
“goods”, athe
of
of minimum,
should
the
‫ ݕ‬ൌ threshold
doing
doingin
below).
‫כ‬ in
case
if ‫ݐ‬, the
thestill
so
so of
This
a country final
final
in
in
as t is the
a bein
"goods", target
“Progress”
which
which
formulation the
period.
period.
making final
“Progress”
is initially andof
“Progress”
the
minimumthe Even
Even reaching
period,
progress
increasing
with
provides below ifif
for countries
countries
decreasing
for
respect
a
admissible each
method
the the
whereas and
each for
“Progress”
to threshold).
theare
given
thresholdof‫ݕ‬ in ଴for
are
given
value. and
case
y*
doingis “bads”,
already
already
for 
withof
progress the
progress
positive
and need
, the on
“bads”.on
case
respect if
multiplying the
the
to
in
‫ݕ‬ ଵof
in
derivative be
to
the
൐ "goods",
the ‫ݐ‬, “Progress”
‫ݕ‬ case
zero
of is
case
this if
“Progress”
valueand
positive
of ‫ݕ‬of increasing
with
ଵ“goods”).
ൌ if
“goods”).
‫ݐ‬,
with respect
and
‫ݕ‬ ଵ
൐
respect
ௗሺି௬ When
‫ݐ‬, for
‫כ‬When
negativeሻ to
zero
to the
௬ ‫ݕ‬‫ݕ‬ ଴‫כ כ‬case
if
‫ݕ‬ is ଵpositive
otherwise.
ൌ‫ݕ‬ൌ
బ ି௠௜௡ሼ௧ǡఉ௬ ‫ݐ‬,ൌ of
‫ݐ‬,the‫ݐ‬,the “bads”
బ ሽand if൏
deriv
der ne‫ݕ‬o
er be
threshold below in the
the threshold
final period. in the
Even finalif period,
countries GEP is whereas
are equal already to for one “bads”,
on this
the meansଵthe country isଵ ଵequivalent to a position in w
hold,
hold,
n
of
of y 4.5 defining
defining
is they
they
below).
The ‫ݕݕ‬
initially should
should
index
‫כ‬‫כ‬ This
belowis
issothethe
in still
still
formulation
takes be
following.
following.
which
the be
on thewords,
threshold making
making
a threshold
value provides
“Progress”
In
In above
“Progress”
the
the
and progress
progress
crossing
16 aa
ismultiplying or method
with
relevant below
with
the and
and words,
respect
respect
threshold
element. of
this  
one,
The and
and
doing
index
to
value to‫ݕ‬in‫ݕ‬so
depending
crossing ଴ ଴need
need
isisin
takes
the which
positivethe
ispositive
right to
to
on threshold
positive abe
be the
whether
value
if
directionif‫ݕݕ‬words, ଵ
൐൐ actual
abovein
‫ݐ‬,,zero
‫ݐ‬,involves
zerothe
zero or
crossing progress
ifbelow
ifright
‫ݕ‬a‫ ݕ‬ൌ ‫ݐ‬,one,
the
direction
premium.
ൌ is
and
,‫ݐ‬,and andabove
depending
threshold
negative involves
negative
negative
Trivially, on
in otherwise.
thea premiu
otherwise
“Progress wheth
right d
e threshold in the final period. Even if countries when itare is‫ כ‬greater
already than
on the one this means the country is equivalent to a positi
4.5
4.5
ൌ hey
element.
dex
ൌ ‫ ݕݕ‬,, with ଴଴should
below).
below).
takes or
withThe This
below
on This
16  ൐ still
rationale
a
൐The the
value be
formulation
formulation making
rationale
ͳ(i.e.
ͳ(i.e. anof
target.
above this
an increase It or
decreasing
increase is
of this method progress
provides
provides
method positive
below
words,
words, in a
a
of
one,
in and
method
method
indefining
y*
crossing
in of definingcrossing case
depending
The
for  the and
indexof
theof ordoing
decreasing
the‫ݕ‬ case 
doing
is
progress
below
on
takes
threshold need
the
threshold
is the of so
Theso
whether the
on in
"goods", to
in
following.
and a
in which
y*
function
in be
which negative
target.actual
value
for
thethe and the
thethe
In
right It
abovethe
of is
case
increasing
right in
progressprogress
direction the
positive
or
decreasing
of
direction
otherwise. In other words, crossing the threshold case
below is
"goods",
for inabove
isthe
involvesof
case
one,
in
involves regress.
y*
and
always
case of
depending
for
a aprogress
ofthe
increasing When
case
increasing
“bads”.
premium.
premium. on and
of
for whether negative
"goods",
the
and
Trivially,
Trivially, caselinear actua
and
“Prog ofini
“Pro
its targets, so on and so forth. If a single indicator is considered, any pr
gow ment.they
ow).
ment.
asthe
nsible
as ‫ݕݕ‬should
This ଴ formulation
GEP
The
The
target
଴
൐ ‫ݐ‬,isi.e.
൐ rationale
target. rationale
‫ݐ‬, still
equal
Itfollowing.
can
i.e. iswhen
bebe
when to
of
of
positive making
provides
one
this
this
expressed
aa the
In countrythis
method
method
in
country case
case progress
a
means
as
decreasingmethod
decreasing
of isof
‫ݕ‬
is of
ofprogress
‫כ‬
ൌ
“goods”, the and
defining
defining ‫ݕ‬of
or
in଴
in
any doing
country
,y* sensible
below
with
y* and
for
‫ݕ‬ and
‫ݕ‬for‫כ‬‫כ‬
is
is
GEP so
is
the the
the
the
൐ equivalent
the
negativein
case need
which
following.
iscasefollowing.
equal
target.
ͳ(i.e.
target decreasingofan into
of Itthe
to
the
in be
to
is
increase
"goods",
"goods",
the In
one a
In
case position
the
positive
and the
right this
and andof
in
linear means
direction in in
regress.
increasing inwhich
case
increasing ‫ݕ‬the ଵ
involves ofWhen
infor it
country amet
progress
for
the the the
case
premium. its
caseisand
case target,
equivalent
of of of negative
“bads”.
“bads”.
Trivially,“bads”. toina derivat
The positio
the ca
compared across countries in this indicator by‫ כ‬simply comparing the progre
elow).
ssible The
ttequal
ible
variable).
is
is the targetThis
target
the when
rationale
tominimum
oneformulation
can
can
This
minimum itof isbe
this
can be greater
isthismeans method
expressed
beexpressed provides
expressed
admissible
admissible than the
Theas ofone
country
as
as index
as
value.
value. ‫ݕ‬a‫כݕ‬long
‫כ‬method
this
defining ൌൌtakes means
is ‫ݕ‬equivalent
‫ݕ‬as ‫ݕ‬,,GEP
଴଴ ‫כ‬of଴doing
withis the
with
on‫ݕ‬
with a൐ is The
value൐
‫ݐ‬,country
equal
൐ toso
following.
when index
i.e.
ͳ(i.e.
ͳ(i.e.ain
above
(i.e. to
when which
is
itposition
is
anone
an
depends takesequivalent
In
greater the
athis
increase
orincrease the
in
on
below on
country
“Progress” than
which
means
awhether
value
in
in
one,tois one
is itathe
Theposition
met
abovethis
depending
decreasing‫ݕ‬countryൌmeans
its
‫כ‬index or where
‫ݐ‬target, is
takes
inbelow
or on the
‫ ݕ‬equivalent
whether
ൌfor it
on
one,
theexceeded
country
‫ݕ‬a଴,case value
depending
butactual toisabove
in
of aequivalent
position
progress
either on orcase below
whethe toitwh
inis ai
ab
in଴଴‫כ‬single
mind that different countries may have different targets. This means re
it
arget .
ariable). The
ariable).is
herwise,
threshold
threshold rationale
greater its
canThis be
This
onetargets,
and
and than
expressed
is
is
should
anof onethis
so
admissible
admissible
increase
multiplying
multiplying onmethod
this
require as and
means
inthisor
‫ݕ‬the
this
‫ כ‬so
as
‫ݕ‬asbelow
ൌ ‫כ‬ of
The
ൌ long
The
long
initial
value
value defining
forth.
the ଴
‫ݐݕ‬,index,as country
with
index
the
value If
asttarget.
as a
takes
of
is when
‫ݕ‬‫ݕ‬takes
‫ݕ‬ the
the൐൐ is the
‫ݐ‬,or
Iton
minimum
ͳ(i.e.
൐variable).
‫ݐ‬, it
onis its
equivalentis
i.e.
i.e. following.
indicator
abelow
aan targets,
greater
value
value
positive
when
when
This increase
and to
the
above
isabove
admissible is
ain a
thanIn
so
target.
case
apositive
country
countrythe
considered,
on
position
in one
ororbelow
“goods”, and
of It this
below
inis
value. is so
where
progress
the
and one,any
means
positiveor forth.
one,
case
increasing it
below progress
exceeded
depending If
the
ofin“bads”,
depending
and a
the
case
negative
for single
country
the made
target.
ofon
caseon
wheneverwhethercan
indicator
is
progress equivalent
Itthe
inwhether
of is positive
“bads”. be
‫ݕ‬and
case is
actual
‫כ‬ actual
് of considered,
to in
‫ݐ‬,negative a
progresspositio
case
progress
regress.
(i.e. lower inofWti
gets, so compared
on and across
so forth. countries
If a ‫ כ‬single in ଴ this
଴indicator always
indicator
its is
targets,
compared
by
compared
considered,simply
so on
with
comparing
across
and any
a
so
common
countries
progressforth. the If
element
progress
made
ain this
single can
given
functions,
indicator be
indicator
byby the
keeping
simply
is
threshold.
considered, comparing any the
pro
target
ods”, ). This
erwise,
rwise, can
if one
a be
iscountry
one admissible
should
should expressed isrequire
admissible require as
asaslong
initially GEP
‫ݕݕ‬long‫כ‬‫כ‬
‫ݕ‬
below ൌൌas or
ൌ is
as or
‫ݐ‬,‫ݐ‬, asbelow
below
equal
‫ݕ‬tt is
as
‫ݕ‬
the , with
is
൐the the
to the
the
threshold 
one
‫ݐ‬,, i.e. target.
i.e. target.
minimum
minimum
൐ whenthis
whenGEP
ͳ(i.e.
and It It
meansan
is is increase
equal
positive
aadmissiblepositive
country is
admissible
multiplying
a country the
“Progress” to one
in
country inin
this case
value.
value. this
case
forvalue is of means
of
each GEP
progress
equivalentprogress is
given the equal country
and
to and a
progress to
negativeone
negative
position is inthis
equivalent
in
in
thein means
the the
whichcase case to
caseitthe a
met
of of country
position
of regres
“goods regre
its ta
ared
s”,
s”,e).one across
ififThis in
aashould mind
is
country
country countries
admissible
require that
is
isisabove different
‫ݕ‬in
initially
initially ‫כ‬the
ൌ this
as
16 ‫ݐ‬,when
below
below long
as
threshold. countries
indicator GEP
t the it the
GEP
as
is
the isis is
‫ݕ‬
Otherwise,by଴may
greater
equal simply
equal
minimum
threshold
threshold ൐ ‫ݐ‬, have
comparedthan
to
one to
i.e.andone
and when
one different
comparinginmultiplying
one this
admissible
should mind
across
this
multiplying itameans
this is
means targets.
that
the
greater
means
country
“Progress” value. different
progress
countriesthe
this
this
Thethe is This
than
the country
value inone
country
country
with
value
index means
countries
functions,
this
respect
takes when
isthis
is relative
indicator
equivalent
on a to may
keeping
itvalue
means
isequivalent is
equivalent by
‫ ݕ‬଴greater
is realizations
have
the
above tosimply
toa
positiveto adifferent
orthan
country
positioncomparing
aposition
below ifone
position ‫ݕ‬is are
ଵ targets.
one,in
൐this
equivalent
where
inwhich
‫ݐ‬, whichtheThis
means
zero itprogresto
itifitmet
‫ݕ‬metଵmea
the
excee aൌitpc
e,dcountrythat should always
different compared
countries with
mayits athreshold
common
have
twhentwhen
targets, different
it element
isminimum
itso is greater
ongreater targets.
in multiplying
mind
and its given
thanalways
that
than
so This
targets,
one
forth. by
different
one the
compared
words, means
this so
this
If athreshold.
on
meansmeans
single
crossing with
relative
countries
and the so
the aon
indicator the common
realizations
may
its
forth.
country
country have
targets,
isIf
threshold aelement are
different
so
single
isconsidered,
is equivalent
equivalent on
inprogress
the giventargets.
and
indicatortoto
any
right isa by
so the
isThis
forth.
position
aprogress
position
direction threshold.
means
Ifwhere
considered, a single
where
made
involves rela
itan
itcanea
one require as is
thethe 3.2 The GEP
admissible Index value. (progress in the multidimensional cas
‫כ‬
is initially require below 16ൌ
‫ݕ‬16 the ‫ݐ‬,as is minimum and admissible this value depending whether actual above or
sa compared
country is initially with value.a
16 In common
belowothercompared the element
words, its itsfor
threshold given
targets, across
targets,
“goods”, and by
always
so so the
ifcountries
on onand
amultiplying
countrythreshold.
compared
compared
andso in
is this
forth.
sodecreasingforth.
this with
across
indicator
valueIf a common
countries
Ifaasingle
below by
insingle
the y*target.simply
for element
compared
in
indicator this
indicator
theIt is comparing
case positive given
indicator
isofacross
isconsidered, by
considered,
"goods",
in case by the
countries
simply
the progress threshold.
ofand any
progress any in this
comparing
functions,
progress
increasingprogress indicator
for the
madekee
mad
the pr
initially in mind
16 below the threshold and multiplying this compared
compared that across
different across in mind
countries
countries that in
mayin different
thisthis have indicator
indicator countries
different by inby
and negative in the case of regress. When GEP is mind
simply
simply may
targets. that have
comparing different
comparing
This different
means the countries
the targets.
progress
progress
relative may This
functions, have
functions
realizations mean d
3.2 The GEP
value by morealways
Index thanin
(progress
oneinmind compared
mind
still resultsthat For
thatdifferent ina
in a always
with
different the
value
3.2
composite
a common multidimensional
The
countries compared
countries
that The element
GEP
indexindex may may
equal
to Index
with be
have
takes togiven
have well
a common
one on alwayscase)
(progress
connected
bya the
different
different
this value
means compared
element
threshold.
targets.
targets.
above to
the country
in the the
given
This
or This multidimensiona
policymaking
with by
ismeans
below a common
means the
equivalent one,threshold.
relative
relative
depending process,
to element realizat
realiza on it
g
The GEP Index (progress
is lower than the value alwaysin
always the
of the
multidimensional
compared a3.2
threshold, with
compared normativeThe thenwiththe GEP
aacommon weighting
common
or below
Index case)
a element
(progress
system
element
the target.
position ingiven given
which
that
It is by in
itby
helps
the the
theits
positive
met threshold.tomultidimensional
threshold. understand
in case
target, whenofit is progress and
greater and neg
guide case p
by more than one For a composite
target stillforresultsindex
this country to be
in a should well
value be that weighting
connectedto at is least lower For to system
reach the
athan GEP must
policymaking
composite theisvalue equal
than
not
indexone of only
process,
tothis to
one
the recognize
be well
this means
threshold,
means it is
the country
that
preferable
connected then theis allthe indicators
to
country to the
equivalent have are
policymaking
is to equivalent
a
potentially proc
to a
composite
hatby
arget more
is lower a normative
than
for this index
than one
thetheto
country be
stillweighting
well
value
threshold. results
should
3.2
connected
ofthe
In in
the
The
system
becase a value
threshold,
to of
GEPto
at“bads”,that
the but
thatFor
leastthen
Index also
helps
policymaking
a
is
reach
the
3.2 a
composite
lower
reasoning toThe
(progress
take
normative
than
thethe threshold. into
understand
when GEP
process,
index
the account
it isIndex
in
weighting
valueto
position itthe
and
be
greater is the
of
In where well guide
preferable
the
the case
3.2
(progress
multidimensional
local
system
connected
than The
and
policy.
one
threshold,
it exceeded tothat
of “bads”, thisGEP
global
have in
to The
helps
then
the
the
means
its targets,
Index
contexts.multidimensional
normative
to case)
policymaking
the the
the so on and
(progress
understand The
country requiremen
process,
is and
equivale in th
gu
it i
mative
target
ast by reach
easoning more weighting
weighting
for this
the
isthansymmetric. system
country system
isthreshold.
one stillA
symmetric. must
should InAthat
target
results the
target
3.2
not
be
3.2
helps
case
takes
in a
The
only The
tovalue
takes at to
of
thethe
GEP
recognize
least
GEP
both
aunderstand
“bads”,
form
that
form
Index
reach ‫כ‬Index
locally that
‫ݕ‬is weighting
normative ൌ the
the
lower ‫ݕ‬all ଴ (progress
and
and (progress
its
with
than globally
indicators
,weighting system
guide
targets,
threshold.  so
the ൏ value ͳ
forth. are
must
policy.
system
so in
adds in
In Ifof
(i.e. onthe
the
the
potentially
aanot The
and
decrease
the
single
multidimensional
complexity
that
case
multidimensional
only normative
so helps
threshold, of
indicator of
recognize
forth.
in“bads”, to
equal
the to the
Ifinitial
isthen a the
considered, weights
importance
thatthe all any
understand
single case)
case)
but
indicators
indicatorand guideincreasesisare considpote
pol
po t
ting
reasoning
form target
alue system
of ‫כ‬but
‫ ݕ‬the foris
ൌ also
must
symmetric.
variable).
‫ݕ‬this଴ take
, , with
with not
country into
only
൏ ͳAshould
 This account
recognize
Foradmissible
target
(i.e.
is
(i.e. aaadecrease the
composite
takes
be to the
decrease that local
at index
all and
indicators
weighting
index
form
inleast
inprovided
the theinitial for
global
For to
‫כ‬
initial
‫ݕ‬reach but setting
ൌbe
value
‫ݕ‬ ଴a also
system
൏‫ݕ‬of contexts.
are
composite
well
the‫ݐ‬,compared
଴
, with
i.e.,policy
take
potentially
must
connected
threshold. into priorities.
Thenot
index
൏ ͳacross
aprogress
country (i.e.requirement
account
of
only
In to
to equal
is
made be
the
athe ForThe
countries
decrease
below the
recognize
well complexity
importance
a
policymaking
case
can be local
the for
composite
connected
of thatpolicy
and
inthreshold.
compared this
the initial
“bads”, all comes
globalrelevance
indicators
index
to
process,
indicator the to
the countries
across from
contexts.
be
policymaking
it are
well
byissimply the
preferable combinat
The
potentially
connected
compari requir
proce to h
so
value
rovided
Otherwise, take
reasoning both
into
ofthe ‫ݕ‬one଴ locally
account
variable).
is
൏ symmetric.
should
‫ݐ‬,the i.e., and the
ThisA
a require
variable). globally
country local
a
This istarget and
For
normative
‫ݕ‬admissible
isൌ
‫כ‬
is Foradds
below a global
a
takes
‫ݐ‬,
admissible as tthe complexity
composite
composite forces:
contexts.
but
weighting
isprovided
the the
providedform
threshold. also
maximum a
indexindexthe
to
both
‫ݕ‬‫ݕ‬ൌ ൏
‫כ‬ ଴the
The
take
normative
system to need
locally
to be
in weights
requirement
into
be mind
‫ݐ‬, , i.e.,
‫ݕ‬admissible
଴ for
well
well
that and
account
withaincountry flexibility
weighting but
connected globally
connected
helps
that൏ value.
this increases
for
the to
different
(i.e.isabelow
ͳindicator of
policy
local
system a
to adds
to weights
the
understand
decreasethe the
countries
by simply complexity
relevance
and
normative that global
policymaking
policymaking
the comparing in
usefulness
helps
and
threshold.
in order
may
the initial to
contexts.
weighting to
guide
have for
of
the
understand
process,
process,
the progress them
the
weights
The
system
policy.
different it itto
requirement
is is
Thethatbe
but
and
preferable
preferabl
targets. differ
increa
helps
norma guid Th
ocally
Otherwise,
the
value maximum index
and
of the one for
globally setting
should
,admissible
variable).
i.e., aadds policy
require
country This complexity
weighting
value.
isis ‫ݕ‬priorities.
below aadmissible
‫כ‬ aൌnormative
normative
‫ݐ‬, as
the to
system tThe isdepending
the
threshold. both
the complexity
weights
must
weighting
provided locally
weighting
maximum
Otherwise,index
weighting
not ‫ݕ‬but on and
only
଴ comes
for
system
൏ local
system
always i.e.,from
setting
increases
globally
‫ݐ‬,system
recognize
admissible characteristics
that
athat
compared
functions, policy
must
country the
the
addsthat
helps
helps
value. combination
notpriorities.
usefulness
keepingcomplexity
all weighting
with
is only
to and
indicators
to a
below understand
in understand
common
mind ofthe
The
recognize the ofto
system two
are need
complexity
the the
threshold.
that competing
that
potentially
and
element
differentand for
weights
must allguide comparisons
comes
not
guide
given
countries but
indicators only
of from
increases
equal
policy.
policy.
by recognize
the are the
importa
The tocom
poten
The
thresho thbtn
no
Oncefor Otherwise,theforces:
setting targetpolicy
one the
isshould
one needrequire
priorities.
formulated
should for
require The
but flexibility
this complexity
weighting
also
‫ݕ‬ weighting
‫כ‬way,ൌtake ‫ݐ‬,, theoftinto
as
as different
weights
system comes
index
system
tfollowing
is is the account
the for
but
must in
from
must
maximum
maximum dimensions
forces: order
setting
also
expression the
not
the not the
takeonly
localfor
only need
combination
policy
admissible into
is themof
recognize
may recognize
andobtained: the
for
priorities.
account to
global
have value.data
ofbe
flexibility
that
different two
The
but
thatthe across
different
contexts. competing
allall of
complexity
also
local
indicators
targets. take
indicators indicators
for
weights
and
The This each
into comes
global
are in
account
are
requirement
means and
country
order fromthe
contexts.
potentially
potentially
relative across
for them
thepolicy
The
of of countrie
combinatio
local equal
equal to
and
require
releva be
impgl
im
:Once the
ollowing depending
need
the expression for
targetadmissible on
flexibility
is formulated local of both characteristics
weights but locally
but also
value.this way, the following expression is
is obtained: also in and order
taketakeacrossand
forces:
globally
into for
into the
both countries,
depending
them
the
account need
adds
account need
locally to for
the be
theacross
foron
and
complexity comparisons
local local
different
flexibility
local globally indicators
and andto characteristics
for
the
global of both
adds
global to
each
weights
obtained:are always compared with a common
realizations within
be
locally
complexity
contexts.
contexts. possible
country in
but a
and and country
order
TheThethe along
globally
to
increases theneed
for
requirementand
requirementthemthe
weights for
adds
the the combine
comparisons
to be
complexity
but
usefulness
for for differen
differe
policyincreas
policy co
of
re
ding Onceondifferent the local
target dimensions
characteristics
is formulated index ofbothand
this the
both
forway, data
the
locally
locally
setting
ௗ௬ need
the Index
across
depending
and and
policy ௬for
followingభ ି௬resolves
indexindicators
different
comparisons
globally
globally
priorities.
బ on
foradds
expression
3.2this
local
setting
adds The The
complexity
and
dimensions to across
policy
complexity
complexity
complexity
is
element
GEP
be
characteristics
obtained: ofwith
possible Index
countries
priorities.
given to the
index
bythe
to
comes athe
thedata
along
and for(progress
weighting
The
weights (for
weights
from
threshold.across
the
setting the each
need
complexity
the system,
but policy
but in
indicator
indicators
for
increases
combination thewhich
comparisons
priorities.
comes
increases multidimens
and allows
from
the
of the across
The
two the
usefulnes
usefulne progre
to
compl
compecomb beco
‫ כ‬ൌ ௠௔௫ሼ௧ǡఒ௬ బǡ ߣ ൐ ͳ
ent dimensions భ
௬ ି௬ across countries,
బ of theܲ‫ݏݏ݁ݎ݃݋ݎ‬ data across
forces: across index
ൌ indicators
index ቐthe indicators
for
ௗ௬ ௗ௬ need
for settingeach
setting within
different particular
and
for௬ policy
భ
forces:
ି௬ adimensions
across
flexibility
policy
బ ሽି௬
బ country
across the
priorities.
priorities.indicator
countries, and
ofcountries
need
weights of
The Thethe
for across
across
the combine
(for data
flexibility
in
complexity
complexity order countries,
indicators
each
forces: acrosscomparison).
of
for
comesindicator
comes the
weights
them across
within
indicators
need
from
from
[1’] to in for
be
the aThe
indicators
the country
and
order GEP
flexibility
different
combination
combination for and
across within
them
for the
ofcountries
each
of oftoatwo
weights coun
comb
two becouco cd
బ ሽି௬ బ ǡ ߣ ൐ ͳ ௗሺି௬ሻ ൌ ௠௔௫ሼ௧ǡఒ௬ ௬ బ ି௬ భ ǡ ߣ൐ͳ
s௠௔௫ሼ௧ǡఒ௬ countries, Index resolves
across this
indicators
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ complexity
depending within
forces:
forces:
ൌ ቐௗሺି௬ with
ௗ௬a the
on
‫כ‬
ௗ௬country
the a
across
ൌlocal needweighting
progress
need
‫ כ‬ሻ ൌ௬ బ ି௠௜௡ሼ௧ǡఉ௬
and
for Index
௬depending
for
[1’]௬characteristics
భ బ ሽି௬
ି௬ బ the
countries,
across system,
flexibilityresolves
flexibility
బ
ǡ For combine
ߚ൏ which
across
countries.
on ofand
aͳcomposite this
local
ofweights weights allows
complexity
comparison).
indicators
the characteristics
need in
index progress
depending
inorder order
for with
to beThe
within for to
a
for
comparisonsGEP
anda
well on be
weighting
themcountry
them
[1’] analyzed
local
the to
connected need besystem,
and
tocharacteristics
tobebe for
forthe
different
to which
combine
comparisons
different
possible
the forfor
policymakin allows
and each
along com
each th
బ ି௬ భ బ ሽ ǡ ߣ ൐ ͳ
௬ బ ି௬ భ ௗሺି௬ሻ
resolves
ൌ ௬ బି௠௜௡ሼ௧ǡఉ௬ బሽ ǡ ߚ ൏ ͳ eachthis particular
complexity indicator
with different
ܲ‫ ݏݏ݁ݎ݃݋ݎ‬ൌ ௗሺି௬ a across
weighting
depending
depending dimensions
ቐ ௗሺି௬ሻ
ௗ௬countries,
system,
‫כ‬
ൌIndex
on on local
௠௔௫ሼ௧ǡఒ௬ local of across
each
which
resolves
different the
బ ሽି௬ బ
data
characteristics
characteristics
ǡ a indicators
particular
allows this
ߚdimensions progress
complexity
across
normative
൏ ͳ indicatorwithin
and of to
indicators
and with
the
weighting the a
be country,
across analyzed
a
different
needdata
needweighting
and countries,
for
system for as
across for
dimensions
across wellsystem,
comparisons
comparisons as
across
indicators
that
[1’] countriesoverall
helpsof which indicators
the
to and
to allows
data
(for
be be across
each
possible
understand within
progres
across
possible cou
indic alaia
‫כ‬ሻ
Let ௬ బnow:
ି௠௜௡ሼ௧ǡఉ௬ ௬ బ ି௬ భ ሽ
బ
particular progress
indicator across across countries.countries, different
different across dimensions ൌindicators
each
dimensions acrossprogress
particular ofwithin
oflinear countries,
the the across
ǡindicator
dataaͳ
data country, countries.
ଵ across across
across a as well as
countries,
across
indicators
indicators
indicators overallacross
countries,
andwithin
and across aindicators
across across
country within
indicators
and the a combin
countr
within
The function of progress across is always countries,increasing
ௗሺି௬ ‫כ‬ሻ
across andindicators
௬ బ ି௠௜௡ሼ௧ǡఉ௬ బሽ
weightingߚ ൏
in ‫ݕ‬within insystem the country
case must ofnot and “goods”onlythe recognize and countries
combine countries
comparison).
that (for(for
all indicators each The ai
each
ess across countries. ଵ ଵ Index across resolves
across countries,
countries,progress
this Index
complexity across across
across resolves with countries.
indicators
but indicators a this
also weighting
ଵtake complexity
within
withinintosystem, aIndex
acountry
account with
country resolves
whichathe weighting
and and this
allows
localthe theand complexity
system, comparison).
progress
combine
combine with
which
to be aallows
comparison). weight
analyzed pT
The function
easing
decreasing Let
andnow: andlinear oflinear progress
in ‫ݕ‬in in ‫ ݕ‬in is the
the always
case caseofincreasing of“goods” “bads”. and and
The
Let now: linear
derivative in ‫ݕ‬of inprogressthe case with ofrespect “goods” toand
17 ‫ ݕ‬଴ global contexts. The
IndexIndex resolves
resolves this each
thiscomplexity particular
complexity with with indicator
aaweighting
weighting across each system, particular
countries,
system, whichwhich indicator
across across
indicators countries,
within ovac
a
ow: decreasing
“bads”.
depends The function on and
The derivative
whether linear
of ‫כ ݕ‬in
progress of ‫ݐ‬ଵor
ൌ‫ݕ‬progress
eachinis‫ݕ‬the particular
ൌcase
‫ כ‬always
with
‫ ݕ‬଴, but of indicator
increasing
respect “bads”.
Let in eithernow: to The
across
‫ݕ‬and଴ derivative
caselinear both countries,
it is locally in of
negative inand
‫ ݕ‬ଵ progress
across
in globally
the the case
indicators
with
case ofadds
respect
of“goods”
within
“goods”, toallows
complexity allows
a଴ country,
‫ݕ‬and progress
toprogress
the asweights totobe
well asanal
be ana
bu
progress across countries. progress across countries.
depends
tnd eitheroncase
inpositive
decreasing inwhether
andtheit linear
case ‫כ ݕ‬of
is negative ‫ݕݐ‬progress
in “bads”,
ൌ ଵor in each
in‫כ ݕ‬the
each
the
whenever
ൌ across
particular
‫ݕ‬particular
case
case ଴
, but countries.
ofof‫“ݕ‬bads”. ‫ כ‬indicator
in indicator
്either
“goods”, ‫ݐ‬, (i.e. Thecase
17 lower
across
across
index isfor
it initial
derivative
countries,
countries,
negative setting
values
of progress
across
inpolicy
yield
across
the awith case indicators
higher
indicators
priorities. of value
respect The
“goods”, within
to
within
ofcomplexity
‫ ݕ‬଴17
aacountry,country, comes as aswell wella
from
and ‫ כ‬positive in the case of progress
progress ଴across
across ‫ כ‬countries.
countries. forces: thevalues need for flexibility of weights in order for them
r‫ݕ‬
Progress”
depends ് ‫ݐ‬, (i.e. forlower
on eachinitial
whether given‫כ ݕ‬valuesൌ“bads”,
Letprogress
‫ ݐ‬ornow: ‫כ ݕ‬whenever
yield ൌ ain‫ݕ‬higher the
,17 but‫ݕ‬case value
in്Let ‫ݐ‬,of
either (i.e.
ofnow: lower
“goods”).
case itinitial
isWhen negative ‫ כ ݕ‬ൌin yield‫ݐ‬,the
Let the a case
now: higher
derivative ofvalue 17 of
“goods”, of
“Progress”
eand
Progress”case ofwith
positive for
“goods”). each
inrespect
the case given
When
to of progress
‫ ݕ‬଴“bads”, ‫ݕ‬is‫ כ‬Let positive
ൌ Let ‫ݐ‬,now: the
whenever
now: inifderivative
the
‫ ݕ‬ଵ ൐case ‫ݐݕ‬, ് ‫ݐ‬,of(i.e.
‫ כ‬zero ‫ݕ‬depending
if“goods”). lower
ଵ
ൌ ‫ݐ‬, initial and When on ‫ݕ‬local
negative
values ‫כ‬
ൌ yield characteristics
thea higher
‫ݐ‬,otherwise. derivative In valueother and
ofof the need for compa
ଵ ଵ ଴ ଵ different
ଵ dimensions ‫כ‬ of the data across indicators and acr
 ݀‫ݕ‬௝ ‫ݕ‬௝ଵ െ ‫ݕ‬௝଴ ݀ሺെ‫ݕ‬௝ ሻ ‫ݕ‬௝଴ െ ‫ݕ‬௝ଵ
‫ݕ݀݌‬ ሺǤ
‫ݕ‬௝ଵሻെൌ‫ݕ‬௝଴ ଴ ൌ ǡ ݀ሺെ‫ݕ‬ ‫݌‬௕௔ௗ௦ ሺǤ‫ݕ‬௝ ௝ሻ଴ െ
ൌ ଵ௝ଵെଵ ଴ ଴ൌ
௚௢௢ௗ௦
‫ݕ‬௝ǡ ‫ݕ‬ ଴ ௝ ሻ ݀‫ݕ݀ݕ‬ ௝ ‫ݕ‬௝‫ݕݕ‬௝ ‫ݕ‬െ‫ݕ‬଴௝‫ݕ‬௝ ଴
‫ݕ݌‬௕௔ௗ௦
௝
‫݌‬௚௢௢ௗ௦ ሺǤ ሻ ൌ ଴ ൌ ଴ ‫݌‬ ௝
௕௔ௗ௦ ሺǤ‫݌‬ሻ ൌ
‫݌‬
௚௢௢ௗ௦ ሺǤ
௚௢௢ௗ௦଴ ሻ
ሺǤ ൌ
ሻ ൌൌ ଴ ଴ ଴ ൌ ൌ ଴ ௝଴ ǡ ǡ ‫݌‬௕௔ௗ௦
௝ ሺǤ ሻሺǤൌ
ሻ
‫ݕ‬௝ ‫ݕ‬௝ ‫ݕ‬௝ ‫ݕ‬௝‫ݕ‬௝ ‫ݕ‬௝ ‫ݕ‬௝‫ݕ‬௝

where sub-index j refers to a particular indicator, ݆ ‫ܬ א‬, where ‫ ܬ‬ൌ ‫ܤ ׫ ܩ‬

where sub-index j refers to a particular whereindicator,
where sub-index
sub-index where
݆ j‫ א‬jrefers
‫ܬ‬,refers ‫ ܤ ׫‬is the
toto‫ܬ‬aൌaparticular
‫ܩ‬particular set of݆ ‫݆א‬
indicator,
indicator,
indicators, consisting of “goods”, G, and “bads”, B (in the understanding that ‫ܩ‬
indicators, consisting of “goods”, G, and “bads”, Bconsisting
indicators,
indicators,(in the understanding
consisting ଵofof
“goods”,
“goods”,
଴ that
G, and
G, ‫“ ܩ‬bads”,
and BB
‫“ת‬bads”,
‫ ܤ‬ൌ ‫)׎‬.(in(in
the u
the
݀‫ݕ‬௝ ‫ ݕ‬െ‫ݕ‬
݀ሺെ‫ݕ‬௝ ሻ ‫ݕ‬
3.2 THE GEP INDEX (PROGRESS IN THE MULTIDIMENSIONAL
‫݌‬௚௢௢ௗ௦ ሺǤ ሻ ൌ ଴ ൌ CASE)
௝
଴
௝
ǡ ‫݌‬௕௔ௗ௦ ሺǤ ሻ ൌ ଴ ൌ
13 The weighting methodology
The weighting methodology to construct The to
theconstruct
Theweighting
weighting
‫ݕ‬
GEP Index௝ the ‫ݕ‬ GEP
is performed
methodology
methodology ௝ Index is performed
in two steps.
totoconstruct
construct thetheGEP
‫ݕ‬ in
A first
GEP
௝ two
Index
Inde
ௗ௬ೕ ௗ௬ ௗሺି௬ ሻ
weightingweighting
is applied to is Progress
applied to (ௗ௬Progress
‫ כ‬in the case
weighting (ௗ௬isೕof in“goods”
the case
applied toto and ofௗሺି௬
Progress
ೕ ሻ ௗ௬ௗ௬
“goods”
‫ כ‬ሻ (in(the
ೕ ೕ
‫ כ‬in
and
case
the of
case
ೕ
in“good
‫“כ‬bads”) the c
weighting ‫כ‬is applied Progress
ௗሺି௬ ௗ௬ ‫ כ‬in theௗሺି௬case ሻofof “goo
where sub-index “wrong”
For a composite index to be well connected ೕ j refers side to ೕ
a threshold
of the particular indicator,
(belowೕ
ଵ
ௗ௬
theೕ threshold
଴
ೕ ݆ ‫א‬ ‫ܬ‬, ೕ
where ‫ܬ‬
to give
to the policymaking to give
greater
process, it is greater
weight to the
preferable
indicators, weight to the
to progress toof
to
for
consisting progress
those
give
give
goods
of greater
and
“goods”, of
countries
greater abovethose
weight that
weight
G, the
and to countries
݀‫ݕ‬are
the
to ௝the
threshold ‫ݕ‬progress
on the
progress
“bads”, ௝for
B
‫ݕ‬that
െ“bads”).
“wrong”
௝ ofof
(in are
those oncountries
side
those the
of “wro
the݀ሺെ
countrie
ଵ ଴ ‫݌‬௚௢௢ௗ௦଴ ሺǤ ሻ ଵൌ ଴ ൌ ǡ the ‫݌‬understandin
௕௔ௗ௦ ሺǤ ሻ ൌ
threshold ݀‫ݕ‬
(below ௝ the‫ݕ‬ െ ‫ݕ‬
threshold for goods ݀ሺെ‫ݕ‬
and above
threshold
threshold ሻ
(below
௝ (below‫ݕ‬
theെ
௝ the ‫ݕ‬
threshold
the ‫ݕ‬௝ for
௝threshold
threshold for“bads”).
‫ݕ‬for଴goods
goods Consequently,
and and above
above the ‫ݕ‬th
the
ሺǤ ሻthreshold
௝(below
have a normative weighting
ൌ system ൌ that ǡ the‫݌‬௕௔ௗ௦
helps threshold
ሻ ൌ for଴ goods
ሺǤConsequently, and
ൌfor each above
indicator, the
the ௝threshold
corresponding for “bads”).
௝ ௝
‫݌‬௚௢௢ௗ௦
to understandfor
andeach
guideindicator,
௝ Thethe ‫ݕ‬௝corresponding weight is set as the ratio between the initial level of
଴ ଴ ଴
for ‫ݕ‬each
policy. indicator,
normative the for each
for
corresponding
weight ‫ߨݕ‬ො௝௝indicator,
each indicator,
is weight
set as the
‫ݕ‬௝the the corresponding
ratio
ߨො corresponding
is set
betweenas the
the weight
weight
ratio
initial ߨො௝ߨොis
between
level ௝ set
is setasas
thet
The weighting methodology to construct ௝ the GEP Index is performe
weighting system the variables,
must not only ‫ݕ‬௝଴ ,recognize
the variables, and threshold ‫ݕ‬that
଴ ‫ݐ‬௝where
ǣ the
ofthe the variables,‫ݕ‬j ௝଴‫ݕ‬refers
variables,
sub-index
variables, , ௝଴and and
, and threshold
to a particular
threshold
threshold ‫ݐ‬௝ ǣ:‫ݐ‬௝ ǣ indicator, ݆ ‫א‬ௗሺି௬ ‫ܬ‬, w
௝ , and threshold ‫ݐ‬௝ ǣ ௗ௬ೕ ೕ
where sub-index are jpotentially
refers to weighting
a particular is applied
indicator,
indicators, to‫ݐ‬ Progress
where ‫ ܬ‬ൌof‫ܩ‬
݆ ‫ܬ א‬,௝consisting ( in
‫ ܤ כ׫‬is thethe case of
set“bads”, of “goods” and
‫ݐ‬௝“goods”, G, and ௝B ௝(in the under
all indicators of equal importance ‫ݐ‬ ‫ݐ‬ ‫כ‬
‫ ۓ‬଴ǡ ݂݆݅ ‫ܩ א‬ௗ௬ ೕ ‫ۓ ۓ‬଴ ǡ ଴ ǡ ݂݆݅ ௗሺି௬
݂݆݅‫אא‬ ‫ ܩ‬ೕ‫ܩ‬
indicators,
but also consisting
take into account of “goods”,the local G,andandglobal“bads”, B (in the understandingۖ‫ݕ‬௝ ‫ۓ‬that ǡ ‫ܩ‬ ‫ת‬ ݂݆݅
‫ܤ‬ ൌ ‫א‬‫)׎‬. ‫ܩ‬ ۖ ‫ݕ‬ۖ ‫ݕ‬
to give greater weight ߨො ൌ to the progress ۖ‫ݕ‬௝଴ of those countries ߨො௝ ߨොൌ௝ ൌ GEP
௝ ௝
that are on
contexts. The requirement for policy relevance The ௝weighting ‫۔‬ ‫ݕ‬ ଴
methodology
ߨො ൌ to construct the ‫۔‬ ‫ݕ‬‫۔‬
଴ ଴ Index is pe
‫ݕ‬
threshold (below the threshold
the GEP Indexۖ
௝
ǡ ௝ for
݂݆݅ ‫ݕܤ۔א‬௝goods ଴ and above the threshold
ௗ௬ೕ A first ۖ ۖ ǡ ǡ ݂݆݅
௝ ௝
݂݆݅‫א‬for ‫ܤܤ‬
‫א‬
Theboth locally and
weighting globally addstocomplexity
methodology constructto the weighting is
‫ݐ‬ performed
‫ ە‬௝is applied to in two
Progress
ǡ steps.
݂݆݅ ( ‫א‬ in
‫ܤ‬ the case‫ە‬ ‫ݐ‬
‫ە‬ ‫ݐ‬
of “goods” an
for each indicator, the corresponding
ௗሺି௬ೕ ሻ ‫ݐ ە‬௝
ۖ weight ௗ௬ ߨ‫כ‬ො is set as the ratio bet ௝ ௝
weights but increases the usefulness ௗ௬ೕ of the index ೕ௝
weighting is applied to Progress (ௗ௬‫ כ‬in the case of ଴“goods” and ௗሺି௬‫ כ‬ሻ in the case of “bads”)
for setting policy This priorities.
formulation The complexitythe
gives variables,
more comes to‫ݕ‬௝give
weight ,toand greater
threshold
progress weight
on ೕthose ǣtogives
‫ݐ‬௝gives the progressinofwhich
indicators thosecountries countriesare that
ೕ
This This formulation
formulation more moreweight weighttotoprogress progressonon thosetho
to give greater
from the combination weight toof the
two
This progress
competing of those
forces: countries
threshold that
(below are on
the the “wrong”
threshold for‫ݐ‬sidegoods of the
and above the thresh
starting at an formulation
initially disadvantaged gives more starting weight
position
starting atat with
an to progress
respect
aninitially to
initiallydisadvantagedon the௝ those
disadvantaged threshold, indicators
positionbut that inwith
with which
are resp
threshold (below the threshold for goods and above
for the
each threshold
indicator, forthe “bads”).
corresponding ‫ ۓ‬଴ ǡ weight
Consequently, ݂݆݅ ‫ߨא‬position
ො ‫ܩ‬is18set 18 as the ra
res
the need for flexibility
makingofefforts weights to in order
overcome for them
starting at an initially disadvantagedsuch a This
situation
making
makingformulation
18
. It
efforts also
efforts gives
position
totoprovides more
overcome
overcome weight
an ‫ݕ‬
withsuch
ۖ initial to
respect progress
idea
a situation to on
to the. It.threshold
countries also on pro
௝ such a situation It also p
௝
for each indicator,
to be different forthe
where each corresponding
to country
set depending
priorities weight
(the ߨො௝ isthe
onweight
local set
is as
variables,
those
higher thethe ratio
indicators଴ betweenߨ
‫ݕ‬set
more, and in
the threshold
which ො18the
country initial
ൌ(the
countries is ଴inlevel
‫ݐ‬weight
ǣweight are
a of
starting
disadvantaged at an initial
making efforts to overcome such a situation . It ‫۔‬alsowherewhere to to ௝ set priorities
priorities ௝(the
‫ݕ‬௝ provides an initial idea ct
௝ is higher
is higher the the moremore the the
variables, ‫ݕ‬௝଴position
the characteristics , and
and the threshold
need
vis-à-visfor‫ݐ‬ ǣ
comparisons
theset relevant to be
threshold), initially
and can disadvantaged
be interpreted position ۖaswith an ǡ threshold),
respect
incentive݂݆݅ ‫ݐ‬௝ ‫א‬ ‫ܤ‬
to tothe
where௝to priorities (theposition
position
weight vis-à-vis
vis-à-vis
is higher the the relevant
the relevant
more‫ە‬ ‫ݐ‬ ௝
threshold),
the ‫ ۓ‬଴ ǡ and
country and isimprove
can
݂݆݅incan
‫א‬ be ininte
abe‫ܩ‬disadv
interp
possible alongthose the different dimensions
indicators in which of ‫ݐ‬
the data threshold, but that are making efforts to ۖ ‫ݕ‬
overcome
‫ ۓ‬଴atheǡ country thoseis relatively
those indicators
indicators worse in inwhich off
which (i.e. further
a interpreted
acountry
country away
௝ is isrelativelyfrom the
relatively worswo
௝
across indicators and
position
across countries
vis-à-vis
(forۖ ‫ݕ‬
each
relevant
݂݆݅ ‫א‬ ‫ܩ‬
such
threshold),
a situation.
18 and
It also
can be
provides
ߨො௝ ൌ
an initial ଴ideaas to
an incentiv
threshold). ௝ threshold).
threshold). ‫ ۔‬௝ ‫ݕ‬
indicator across countries, thoseacrossindicators
This ଴ in which
ߨො௝ ൌ formulation
indicators gives a countrymore is to
weight relatively
topriorities
progress worse (the on ۖ off ǡ (i.e.
those ݂݆݅further
is indicators
‫ܤא‬ a
‫ݕ۔‬௝within countries on where set weight
‫ە‬ ‫ݐ‬௝
a country and Applying the combined threshold). starting ۖTheat ǡ an݂݆݅initially
‫ܤא‬ disadvantaged position with respect to the
the comparison).
former model ‫ݐ ە‬௝ GEP in the case higher
Applying ofthe
Applying more
different
the theformerthe
former country
weights model
modelisforinin adifferent
in
18
disadvantaged
the thecase case indicators,
ofofdifferentdifferentthe wew
Index resolvesfollowing this complexity with making
a weighting
expression for the (not This efforts to overcome
initial
formulation
yet normalized)
following position
following GEPsuch
vis-à-vis
gives
expression
expression Index a
more situation
the
forisforrelevant
weight
obtained:
the the (not(not .
to It
yetyet also
threshold),
progress
normalized)provides
normalized) on those GEP an
GEP init
ind
Inde In
Thissystem,
formulationwhich allows gives progress
moreApplying to be
weight the
where
analyzed former
to progressto for model
set starting
priorities
on and at
those in
can(thethe casein disadvantaged
weight
be initially
indicators
an interpreted of
is as
which different
higher
ancountries the weights
incentive moretoare thefor
improve
position country
with different
respect is int
each particular indicator across countries,
following expression
position across
vis-à-visfor ௗ௬thein
ೕ those
(not indicators
yet ೕ ሻ in which
normalized) aGEPcountry Index isbe relatively
ௗ௬is obtained:
σ௝‫ߨ ீא‬ො௝ the relevant
σ௝‫א‬஻ ߨො௝ threshold), and can ೕinterpreted
18
starting at an initially disadvantaged position
‫ ܲܧܩ‬ൌ 18
with
making respect
൅
efforts to to the
ௗሺି௬ threshold,
overcome
‫ܲܧܩ‬
such but
ൌ
a that
σ௝‫ீא‬
situation
σ ߨො
are ௗ௬
൅ೕ . It also ௗሺି௬
σ௝‫א‬஻ σ௝‫א‬஻
as
provides
ߨො௝ߨොௗሺି௬
ௗሺି௬ ೕ ሻ an
[2] ೕሻ
‫ܲܧܩ‬ ൌ ௝‫ீא‬ ௝ߨො ௝ ‫ כ‬൅ ௝ ௗሺି௬
indicators within a
making efforts to overcome such athosecountry, as well as overall
situation . Itwhere
indicators ௗ௬ worse
‫כ‬
also provides
ೕin to whichoff (i.e. further
anainitial
set priorities ௗሺି௬ ‫כ‬
ೕ (the
ሻ
country away
ideaweight tofrom
iscountries the
is higher
relatively threshold). ‫כ‬
onೕthe
ௗ௬ ௗ௬ ೕ more
worse the(i.e.
off
‫כ‬ሻ ‫כ‬
count
ೕ ೕሻ f

whereprogress
to setacross priorities countries.
(the weight is higher the more positionthe country
vis-à-visis thein a disadvantaged
ೕ relevant ௗሺି௬ೕ ሻ initial
threshold), and can be interprete
threshold). ‫ܲܧܩ‬The ൌ σ
ௗ௬
ߨො௝ former ൅is σ ߨොnormalize
position vis-à-vis thesecondThe relevantstep is to normalize
threshold), and canthose (or
be re-weigh)The
Applying
interpreted second
second
௝‫ீא‬
indicators the asௗ௬ step
ߨො௝an‫כ‬to
instep
ೕ
obtain
to
is
model
incentive
which ௝‫א‬஻ to a the
normalize
into the
௝ ௗሺି௬ weights,
‫ כ‬ሻ (or
case
improve
country ೕ
(or
of
is ߨin , which the
re-weigh)
௝re-weigh)
different
relatively take
the ߨො௝into
worse ߨොto
௝ to ob
ofo
thoseLet now: consideration
indicators in which a country is relatively the relevance progress in one
consideration
weights
worse
threshold). indicator
consideration
off for(i.e. the
different vis-à-vis
the
furtherrelevance
relevance
indicators,the
away the others.
progress
progress
from This
following in
the in reweighting
one one indicator
indicator will
vis-à-
vis
indicate the relative Applying
importance the former
of indicate
one model
indicator
indicate
expression the the
for
in
relative
the
the
compared
relative
(not
case
importance
yet toofthe
importance
normalized)
differentothers
of ofone
GEP one
Index
weights
and is enables
indicator
indicator forcomp com d
threshold).
ଵ ଴ ଴ The second ଴ ଵstep is଴ toଵ normalize (or re-weigh) the ߨ ො to obtain the weights, ߨ , w
݀‫ݕ‬௝ ‫ݕ‬௝ െ ‫ݕ‬௝ ‫ݕ‬ ଵ
݀‫ݕ‬௝ aggregation
௝ െ ‫ݕ‬݀ሺെ‫ݕ‬
௝ ௝‫݌‬ ‫ݕ‬௝ following
ሻ indicators
of െ݀ሺെ‫ݕ‬‫ݕ‬௝ ௝within
ሻ ‫ݕ‬ ௝expression
a country
െ ‫ݕ‬௝ asfor
aggregation well
aggregation
obtained: the asof (not
comparison yet normalized)
indicators
of indicators within ௝
ofwithin
results a country GEP as
across
a country Index
countries
as well wellis
asasobtaine
and
compar
comp
௝
‫ݕ‬௝଴ Applying the former model in thevis-à-vis case 19 19ofthe different weight
‫݌‬ ሺǤ ሻ ൌ ൌ ǡ
ሺǤ଴ ଴ሻ ‫ݕ‬ൌ଴ ଵ ଵ consideration ሺǤ ሻ ൌ ൌ
଴ the relevance
ሻ ଴ൌ ଵ ଴ ൌ଴
଴ ǡ
௚௢௢ௗ௦ ‫݌‬௕௔ௗ௦
‫ݕ‬଴ ‫ݕ‬across ଴ ௕௔ௗ௦ൌଵ 19 progress inthe one indicator ߨ௝ ߨothers. This r
௝‫ݕ‬௝ ‫ݕ‬௝‫ݕ‬௝െ ‫ݕ‬௝ ‫ݕ‬௝଴ ݀ሺെ‫ݕ‬ ݀ሺെ‫ݕ‬ ௝‫ݕ‬௝െെ
௝ ሻ௝ ሻ௝ ݀ሺെ‫ݕ‬ ௝௝‫ݕ‬ሻ௝‫ݕ‬௝indicators
‫ݕݕ‬௝଴௝ െ ‫ݕ‬௝ within ‫ݕ‬௝଴ ‫ݕ‬௝ a country . across
Let ߨ௝ denote
across indicators
indicators weight
within
within aattached
acountry
country to. Let
.indicator
Let denote
௝ denotej in the thethewe
ൌǡ ǡ ǡApplying
‫݌ ݌‬௕௔ௗ௦ the
ሺǤሺǤሻ ሻൌ‫݌‬ൌ ሺǤ ൌformer
ሻൌ ଴ൌ ଵ model in the the
ൌ indicate
௕௔ௗ௦଴ ଴ ‫ ݕ‬aggregate caserelative
of different weights
following
importance for one
expression
of
composite
different indicators,
forindicator
the (not the
yetcompared
normalized)
toGEP
theIndex
GEP Index, with σaggregate
is
others ߨ௝ ൌ ͳ. composite GEP
aggregate composite ೕ Index,
GEP Index, with
with ͳ. ͳ.
௕௔ௗ௦ σ௝‫א‬௃
‫ݕ‬௝଴ sub-index݀ሺെ‫ݕ‬ ‫ݕݕ‬௝଴௝௝଴‫ݕ‬௝଴
‫ݕ‬௝௝ሻ‫ݕ‬௝ to ௝ aെparticular ‫ݕ‬௝଴ ௝‫א‬௃ ௗ௬ ௗሺି௬σ௝‫א‬௃ ೕߨሻ௝ ߨൌ௝ ൌ
where
‫݌‬௕௔ௗ௦ ሺǤ ሻfollowing
ൌ j refers expression
ൌ Normalized
for the (not
indicator,
aggregation yet
݆ ‫ܬ א‬, normalized)
where
of ‫ ܬ‬ൌ ‫׫ ܩ‬GEP
indicators Index
‫ ܤ‬iswithinthe set is
‫ܲܧܩ‬
a obtained:
of ൌ
country σ as
௝‫ீא‬ ߨොwell
௝ ൅
as σcomparison
௝‫א‬஻ ߨ
ො ௝ of results acros
toweights are then
‫ ܤ‬isdefined ‫ ܩ‬as follows:
Normalized
Normalized weights
weights are ௗ௬then
are ೕ then defined
defined as follows:
as ೕfollows:
‫כ‬ ‫כ‬
rs indicators,
to a particular ‫ݕ‬௝଴ofindicator,
where
consisting “goods”,‫ݕ‬
sub-index
଴
݆‫א‬
௝G, and ‫ܬ‬, where
refers
“bads”, Ba (in ‫ ܬ‬ൌunderstanding
particular
the ‫׫ ܩ‬indicator, the
that set ‫ܤת‬ ofൌ ‫)׎‬. 19 ௗ௬ೕ
ௗሺି௬ ሻ
ௗሺି௬ೕ ሻ
across indicators within଴ a ଵcountry . Let ‫ܲܧܩ‬ ߨ௝ denoteൌ σ௝‫ீא‬the ߨො௝ weight ൅ σ௝‫א‬஻attached ߨො௝ ௗሺି௬‫ כ‬to ind
ular
“goods”,
alar indicator,
particular G, and
indicator, ݆ ݆‫“א‬bads”,
indicator, where
‫ܬא‬,‫ܬ‬,, where
where ݆B‫(א‬in ‫ܬ‬ൌthe
‫ܬ‬, ൌ‫ܩ‬௝understanding
‫ܬ‬ൌwhere
݀‫ݕ‬ ‫ܤݕ׫׫ܩ‬௝ଵ‫ܤ‬ isൌis
‫ܬ‬is
െ ଴
‫ݕ‬the‫ܩ‬ǡ the
௝the ‫ܤ ׫‬set
set ofset
that
is ௗ௬of of ‫ת‬
the
ೕሻ ‫ܩ‬
indicators, set
݀ሺെ‫ݕ‬ ‫ ܤ‬௝ൌሻof‫)׎‬. ‫ݕ‬௝ െ ‫ݕ‬
ௗሺି௬ ೕሻ ௝
ௗ௬ೕ‫כ‬ ೕሻ
The weighting methodology ‫݌‬
଴ ௚௢௢ௗ௦
ሺǤ ሻ
ଵ to construct
ൌ
଴ ‫ܲܧܩ‬ theaggregate
଴ൌGEP σ௝‫ீא‬Index ‫݌‬ iscomposite
ߨො௕௔ௗ௦
The
ሺǤ ൌ
൅ σ௝‫א‬஻
performed
‫ כ‬second ଴ ߨ GEP
ൌ
ො௝step
in ଴‫כ‬Index,
two ‫ݕ‬steps.
is to with σ௝‫א‬௃ ߨ(or
A first
normalize ͳ.
௝ ൌ re-weigh) the ߨ[2]to obtain the weig
ො
”,“bads”,
bads”, B B(in
indicator,
G, and (in
“bads”,
݀ሺെ‫ݕ‬ ‫א‬the
݆the understanding
understanding
B where
(in
௝ the
consisting
௝‫ܬ‬,
ሻ ‫ݕ‬ െ ‫ݕ‬௝ ‫ܬ‬ understanding
of ൌ“goods”, ‫ݕ‬
‫ܩ‬ௗ௬ೕ‫׫‬ ௝ that
that ‫ ܤ‬is ‫ݕ‬ the
‫ܤת‬that
‫ܩܩ‬,‫ת‬௝and ‫ܤ‬ൌ set
ൌ‫)׎‬. ‫)׎ܩ‬.
“bads”, ௝
‫ת‬of‫ܤ‬ೕൌ
ௗ௬ (inௗሺି௬ ‫ݕ‬
theೕ௝ሻ
‫)׎‬. ௗሺି௬ ௝ೕ ሻ ௝
‫݌‬weighting
௕௔ௗ௦ ሺǤ ሻ ൌ is applied ൌ to Progress (ௗ௬‫ כ‬in the case Normalized
of “goods” and weights
consideration ‫ כ‬ሻ inare then
the case
Thethe second
ofdefined
“bads”) as follows:
relevance step is to
progress normalize
in one (or re-weigh)
indicator the
vis-à-vis intotheobtain
othet
଴ ଴
logy to construct
‫ݕ‬
ds”, B (in the understanding
௝ the ‫ݕ‬ ௝ GEP
understanding that Index
18 that is
It is important
ೕ performed
‫ ܤ ת ܩ‬ൌto ‫)׎‬. note in two steps.
). that for a country experiencing
ௗሺି௬ ೕ A first
18The 18
It second
regress
It is step
in an
important
is important is that
tothat
toindicator
note
to note normalize
in which
for (or
a itcountry
a country
for re-weigh)
is initially disadvantaged
experiencing
experiencing the
regress ߨො௝with
regress an
inrespect
indicator
an indicat
to givewhere
greater
The
sub-index
weight
second to jtherefers
progress
step
to
is
aof particular
those
tointwo
normalize
indicator,
countries ೕ(or
that ‫ܬ א‬,onwhere
݆are
re-weigh) the ‫ܬ‬ൌ‫ܩ‬
“wrong”
theconsideration
‫׫‬
side
ො௝ totoobtain
‫ ܤ‬theis
of the
the set of will this
theweights,
the relevance ߨ௝ ,,progress
whichindicator ininto
take one
intoindicator vis-à-vis t
tct theGEP
the
construct
Progress GEPௗ௬ ೕIndex
Index
the
( (belowin GEP
theisiscaseperformed
performed
Index ofisis
to the
in
performed
“goods”
relevant
two
and
threshold,
steps.
steps.
in
ௗሺି௬ two Aindicate
ሻ Athis weighting
first
first
steps. the
Apolicy
first relative
system
ߨ will
to toimply
the
obtain importance
that
relevant regress
relevant
the threshold, of
threshold,
weights, have
thisone
a significant
weighting
weighting
which takesystemweight.
system will compared
will In
imply other
implythat words,
that to willth
regress the
regress w
indicators,
threshold
ndicator, ݆‫א‬ ‫ܬ‬, ‫ כ‬where consisting
the‫ ܬ‬threshold
ൌ ‫ܤ׫ܩ‬ of “goods”,
weighting
for goods
the G,and
set and
system “bads”,
of above provides‫ כ‬Bin
the (in the
threshold case
the understanding
signals onfor ofpriorities.
“bads”)
“bads”). that ‫ ܤ ת ܩ‬system
weighting
weighting
Consequently, ൌ system
‫)׎‬. provides
providessignals
signalson onpolicy priorities.
policy priorities.
ௗ௬ೕconsideration the19ೕmethodology
ௗሺି௬ ೕrelevance
ሻ first ௗሺି௬ೕ ሻ
ೕ ሻ progress in the
one
aggregation indicate
indicator
of 19 19 the relative importance of one indicator compared
vis-à-vis
ofindicators the others.
within This
aindicates
country reweighting
as well as will
comparison of resu
she GEP Index is the
The weightingperformed in two
to constructthesteps.
therelevance
GEP A first
consideration the relevance progress in of
one
ௗ௬ೕ ௗሺି௬ ሻThe ௗሺି௬ weighting indicates the progress
The made
firstfirst
The in each
weighting
weighting of thethe
indicates areas, as captured
relevance
the of
relevance the byprogress
the indicators;
progress
the made
made the
in in
each o
each
hethe
s”, (Bௗ௬
forcase
case
(in theof
‫ כ‬in
each theof“goods”
“goods”
case
indicator,
understanding ofand and‫ܩ‬ௗሺି௬
“goods”
corresponding
that in
‫ כ‬ሻand
‫ ת‬second
‫ܤ‬ ‫כ‬ൌ inthe
weight
‫)׎‬. casecase
ߨො is of of“bads”)
‫ כ‬ሻ௝in however,
the
set as “bads”)
case of “bads”)
the ratio between the initial level of
the ೕprogress of
indicate those the countries
relative
ௗሺି௬ ሻ that
weighting,
importanceare on theof “wrong”
makes
one
across side
it possible
isindicator
indicators ofestablish
to the
isaggregation
second
second
compared
within of indicators
comparisons
weighting,
weighting,
to within
however,
however,
the within
and
makes
others
19 across
makes ita
and country
it countries
possible
possible to as
enables wellthat
(given
establish
to as
establish thecomparison
sum of within
comparisons
comparisons all with
isa
the The
variables,weighting
଴ Index
, and is performed
methodology
ௗሺି௬
threshold ሻ ೕ ೕ
ೕ ‫ݐ‬weights to inௗሺି௬
two
construct ೕ steps.
the A
GEPfirst weighting
Index performed inindicator
two steps. Acountry
vis-à-vis
tofirst the .Annex
Let
others. ߨfor
This denote
reweighting the weight
ofwill attach
case of “goods” ‫ݕ‬ ௝ and ௗሺି௬‫ כ‬ሻ in ௝ the case It
ǣ is equal18
of
to
is “bads”)
1). See
important Annexto I.B
note for a discussion
that for a
across
of
weights
country weighting
is equal
weights
indicators
for
equal
experiencing the
1). GEP
See
to 1).
within
Seeindex
Annex
regress
a ininI.B
the
an
country
௝multiple
I.B for country
a19discussion
a discussion
indicator
. Let in of case.
whichweighting
weighting
it
denote is forfor
the
initially
the
GEP
the
disadva
weight
GEi
of
esholdthose
those
ogress ofcountries
for goods
countries aggregation
those that
and
that
countries are are
above
of on on
thatthe
thethe
indicatorsare “wrong”
threshold
ௗ௬ೕwithin
“wrong”
on the side
for
sidea
“wrong” ofof the
“bads”).
countrythesideas Consequently,
well
of theas
ௗሺି௬ ೕcomparison
ሻ
18 of results across countries ߨ
and
GEP Index is
weighting applied to Progress ( ௗ௬
performed
is applied in two
to ೕ steps.
Progress A( first
‫ݐ‬ in
in
‫ ۓ‬ೕ‫כ‬଴ to
௝ the
the
ǡ the
case
case
݂݆݅relevant
‫ܩ א‬19aggregate
ofof “goods”
“goods”
threshold, and composite
and in the
this ‫ כ‬ሻweighting
ௗሺି௬ GEP
indicate
case ofIndex,
the
system will with
relative
“bads”) importance
imply σ
that ߨ௝ ൌ
௝‫א‬௃regress . indicator
of ͳone
will ௝ 18 18
have a significant weight.
sand
ose
for and
orresponding above
above
countries
goods andacross
the
theweight
that indicators
threshold
threshold
above
ௗሺି௬ ೕare
ሻ ߨ
the
ො is
on set
for for
the
threshold within
as“bads”).
“bads”).the
“wrong” ۖfor aweighting
‫ݕ‬௝ratio country
Consequently,
between
Consequently,
side
“bads”). of the. Let theߨ denote
௝ initial
Consequently,
system provides aggregate
level
ೕ
signals the onofweightcomposite
policy attached
priorities. GEP toIndex,
indicator with σ j ௝‫א‬௃
in the
ߨ ௝ ൌ ͳ. of
ase of “goods” and
to give greater in
in ௝ the
the case
case
ௗሺି௬ೕ‫ כ‬ሻweight to the progress
of of “bads”)
“bads”)
ߨො௝ ൌ to give greater
Normalized
19of those countries weight
that to
areweights
on the compared
are then
“wrong” side to the
defined
of the others and
asmade enables
follows: aggregation
aggregate composite GEP The
௝ Index,
଴ first
with weighting
σthe indicates the
ൌ ͳ. Normalized relevance weights are then defined as follows:areas, as captured b
of the progress in each of the
eight
ight
ndnding
eshold
se
above
countries ‫ݐ‬௝is
௝is set
ǣthe
ߨො௝ߨොweight set
threshold ߨොas
as
that are thethe
thresholdisthe
set ratio
ratio
onprogress
௝(below thethe
for
as
“wrong”
between
between
“bads”).
the ratio
of side
threshold those for ‫۔‬ ‫ݕ‬
the the
countries
goods
the
initial
ǡ initial
Consequently,
ofۖbetweensecondand that‫ܤ‬level
‫א‬level
the
above
݂݆݅weighting,areofon
initial
௝‫א‬௃ ߨ௝level
ofthreshold
the offor
however, makes indicators
“bads”). within
Consequently,
it possible to establisha country as wellwithin
comparisons as comparison
and across of countries (give
‫ݐ‬
ht
‫ݐ‬ǣ௝ ߨ
aboveො is set
thefor asNormalized
each
thresholdthe ratio
indicator,‫ݐ‬
for௝ “bads”).weights
between
the correspondingare
the
Consequently,
‫ە‬then
initial
௝ defined
weight
weights level
ߨis
ො௝ is as
of
set
equal follows:
as
to the
1). ratio
See between
Annex I.B the
for ainitial level
discussion of
of weighting for the GEP index in the multiple country
௝ ‫ݕ ۓ‬଴଴, ǡand threshold
݂݆݅ ‫ ݐܩ א‬ǣ
ߨො௝ ‫ݐ‬is thethe
variables,
‫ݐ‬௝set formulation
as ‫ݐ‬௝ratio gives
ۖbetween
‫ݕ‬௝௝ moretheweight
initialto
level of 18
௝This ௝ progress on those indicators in which countries are
‫ۓۓ‬starting ǡ ǡ ‫ۓ‬ ݂݆݅
݂݆݅
at‫ߨݕ‬
‫א‬ǡ
ො଴௝ ൌ‫א‬‫ܩ‬ ‫ܩ‬ ݂݆݅ ‫א‬ ‫ܩ‬ ‫ݐ‬
‫ݕ‬௝଴‫ݕ‬௝଴
௝
an initially
଴ disadvantaged position ‫ۓ‬withǡ respect
isto the threshold, but
thatthat are
௝ ۖ
ۖ ۖ ݂݆݅
It ‫א‬ ‫ܩ‬important
18
௝ ‫ݕ۔‬௝ 18 ۖ‫ݕ‬௝
଴ to18 note
It is importantfor a tocountry
note thatexperiencing
for a country regress in anregress
experiencing indicator in indicator
in an which it inis whic
init
ൌ଴ ǡ making ଴ߨ
଴‫݆݂݅ݕ‬
݂݆݅ ‫א‬
ො௝ ‫א‬ൌ‫ ܩ‬18଴ ‫ܩ‬
efforts to ۖ ǡ
overcome ݂݆݅
such ‫א‬
a ‫ܤ‬
situationߨ
ො ൌ . It also provides
to the an initial
relevant idea to
threshold,
to the countries
this
relevant on
weighting
threshold, system
this will
weighting imply
system that regress
will imply will
that have
regress a significa
will have
௝ ‫۔‬
‫۔‬ ‫ݕ‬௝where௝ ‫ݕ‬௝ priorities
to‫۔‬set It is ‫ە‬ ‫ݐ‬௝ (the weight
important to note isthat
higher
௝
for a country
the more
଴
‫ݕ۔‬௝ experiencing regress
in a in an indicatorsystem
ininitial
which onit policy
is initially disadvantaged with respect
ǡ the19 country
weighting issystem disadvantaged
provides signals priorities.
weighting provides signals on policy priorities.
ۖ଴ ۖ ‫ݐ‬ǡ ǡ ݂݆ۖ݅
݂݆݅ ‫ܤܤא‬relevant
to‫א‬ǡ the ݂݆݅ ‫ א‬threshold,
‫ܤ‬ this weighting ۖ system ݂݆݅ ‫ܤ א‬imply
will that regress
19
The will weighting
first have a significant weight. In other words, the made in each of the
௝ ‫ە‬
‫ە‬ ‫ݐ‬௝position
௝ ‫ە‬ ‫ݐ‬weighting
vis-à-vis
௝ the relevant threshold),
system provides andon
signals can ‫ݐ‬௝ interpreted
‫ ە‬be
policy
The first weighting indicates
priorities. as an incentive to improve thein relevance of the progress the
indicates the relevance of madeprogress
in each of the areas, as
ǡ those
݂݆݅ ݂݆݅
‫ א ܤ א‬19 ‫ܤ‬ second weighting, second
however, weighting,
makes however,
it possiblemakes to it possiblecomparisons
establish to establish comparisons
within and within
acrossand coua
more
௝
indicatorsThe in which
first a
weighting country
weight to progress on those indicators inweights
indicatesis relatively
the relevance worse
of theoff (i.e.
whichis equal
progress further
countries
made
to
away
in
weights
1). are
each
See
from
of
is Annexthethe
equal toI.Bareas,
1). for as captured
Seea Annex
discussion
by
I.B forof the indicators;
a weighting
discussion for the
of weighting
the GEP for the GEP
index in theindex
multi
This formulation gives more weight to progress on those indicators in which countries are
threshold). second weighting, however, makes it possible to establish comparisons within and across countries (given that the sum of all
18
 ng expression for the (not yet normalized) GEP Index is obtained:

ase of different weights for different

ௗ௬ indicators,
ௗሺି௬ೕ ሻ the
σ௝‫ߨ ீא‬ො௝ ೕ‫ כ‬൅ σ௝‫א‬஻ ߨො௝ [2]
ormalized) GEP ‫ܲܧܩ‬
Indexൌis obtained:
ௗ௬ೕ ‫כ‬
ௗሺି௬ೕ ሻ

cond
ೕ stepො is toೕሻnormalize (or re-weigh) the ߨො௝ to obtain
ௗሺି௬
[2] the weights, ߨ௝ , which take into
‫ כ‬൅ σ௝‫א‬஻ ߨ ௝ ௗሺି௬ ‫ כ‬ሻ
eration the relevance progress in one indicator vis-à-vis the others. This reweighting will
ೕ ೕ

e the relative results across of

importance countries
19
and acrosscompared
one indicator indicators to the The others double
andweighting
enablessystem allows the GEP Index
-weigh) the ߨො to obtain
within a the
country weights,
. Let ߨ , which
denote take
the into
weight ߨෝఫ both assess how far off a country is from the
to
ation 14of indicators within a country as well as comparison of results
௝ ௝ ߨ௝ ൌ across countries and
one indicator
indicators withinvis-à-vis the
a country
attached others.
19
. LetjThis
to indicator reweighting
denote
ߨin௝ the will attached
the weight
aggregate composite σ௝‫א‬௃
toߨෝఫindicator
threshold j inevaluate
and to the the relative importance of
one indicator ߨෝఫcompared to the others and enables
ate composite
ߨ௝ ൌ GEP
GEPIndex,
Index, with σ௝‫א‬௃ ߨ௝ ൌ ͳ.. one area (indicator) with respect to the others from
σ௝‫א‬௃ Finally,ofthe following expression
countriesforand
the GEP Index is obtained:
ry
lizedas weights
well as areߨෝఫthen defined
comparison results across
as follows: the country’s perspective. This is a real advantage of
Let ߨ௝ denote the weight attached to indicator j in the
Normalized weights are then defined as follows: the GEP methodology, because it informs national
e . ݀‫ݕ‬௝ ݀ሺെ‫ݕ‬௝ ሻ
‫א‬௃ GEP
ߨ௝ ൌ ͳIndex is obtained: ‫ ܲܧܩ‬ൌ ෍ ߨ௝ ‫ כ‬൅ and෍global ߨ௝action.‫כ‬As time passes and the country’s
follows: ߨෝఫ ௝‫ீא‬ ݀‫ݕ‬௝ ௝‫א‬஻ ݀ሺെ‫ݕ‬௝ ሻ
ߨ௝ ൌ situation evolves, weights in the GEP Measurement
portant to note that for a country experiencing
݀‫ݕ‬ ݀ሺെ‫ݕ‬ ሻ σ௝‫א‬௃inߨෝan
regress
ఫ
indicator in which it is initially disadvantaged with respect
௝ ௝
levant threshold, this weighting system will imply that regress will have a significant weight. Framework
In otherwill adjust
words, theto reflect the new set of
෍ systemߨprovides
௝ ‫כ‬൅ ෍ which
signals ߨ௝ ispriorities.
on policy equivalent
‫כ‬ to the expression (after substituting the expression for the normalized
ߨෝ
௝‫ீא‬ ݀‫ݕ‬ ௝
st weighting indicates ௝‫א‬஻
theweights):
݀ሺെ‫ݕ‬
relevance of the
ሻ ൌ
௝ߨ progress made in each of the areas, as captured by the indicators; themakes the GEP Index (relative
ఫ priorities. This feature
llowing
weighting, expression for the GEP ௝Index
σ is
௝‫א‬௃ ߨ ෝobtained: to(given
indexesthat with fixed/common weighting for all
ng regresshowever, makesinit which
in an indicator possible
it istoinitially
establish comparisons
ఫ
disadvantaged within
with and across countries
respect the sum of all
sll equal to 1). See Annex I.B for a discussion
imply that regress will have a significant weight.ߨof weighting
In for
otherthewords,
GEP indexthe in the multiple country case.
on Finally, thethe
(after substituting followingexpression ෝఫ
expression forthe
the GEP Index is countries)
௧ೕ ௗ௬ೕ a well-suited
௬ೕ ௗሺି௬ೕapproach for policy design
ൌ ௝ 18 for normalized బ
. ଵ ሻ
ߨ௝ ݀‫ݕ‬ ‫ܲܧܩ‬ ൌ ݀ሺെ‫ݕ‬ ሻ బ ൈ ൤σ௝‫ ீא‬బ ‫ כ‬൅ σ௝‫א‬஻ ‫ כ‬൨ [2’]
lowing expression
progress made in each of
‫ܲܧܩ‬and for
obtained:the the
areas,
ൌacross GEPas
෍ countries Index
captured
ߨ௝ (givenσ ൅is
by
௝‫א‬௃ ߨ
෍obtained:
ෝ
the
ఫ indicators;
ߨ σ
the
೟ ೕ
௝ ೤ೕ and monitoring.
௬ ೕ ௗ௬ ೕ ௧ ೕ ௗሺି௬ ೕ ሻ
ablish comparisons within that the sum௝ of all ାσ
of weighting for the GEP index in the௝‫ீא‬
݀‫ݕ‬௝‫ כ‬case.௝‫א‬஻ ݀ሺെ‫ݕ‬
multiple country
ೕ‫א‬ಸ ೤ೕ ‫ כ‬ሻ ೟ೕ
బ
௝
ೕ‫א‬ಳ

following
ଵ 18 expression
௧ೕ ௗ௬ೕfor the GEP ݀‫ݕ‬௝ೕሻ is obtained:݀ሺെ‫ݕ‬௝ ሻ
Index
௬ೕబ ௗሺି௬ Finally, to assess GEP within planetary boundaries,
ൈ ൤σ ‫ܲܧܩ‬The ൌ double
൅ ෍σ weighting
ߨ ௝ ൅
൨ system
෍ allows
ߨ ௝ the GEP
[2’]Index to both assess how far off a country is from
ೕivalent೤ೕ to the ௬ expression ௝‫(ீא‬after substituting the expression for the normalized
బ ௝‫ீא‬ ௝‫א‬஻ ‫כ‬ ‫כ‬ the progress achieved in the GEP Index indicators
బ
ೕ ௗ௬ೕ
‫כ‬
the threshold ݀‫ݕ‬௝to
௧ೕ ௗሺି௬
and
‫כ‬
ೕሻ ௝‫א‬஻ ݀ሺെ‫ݕ‬ ௝ ሻ importance
బ ାσೕ‫א‬ಳ ೟
ೕ ೕ ݀‫ݕ‬௝ evaluate the݀ሺെ‫ݕ‬ relative
௝ሻ
of one area (indicator) with respect to the
are compared to the progress made in the indicators
others
‫ܲܧܩ‬ ൌ෍ from the ߨ௝ country’s
൅ ෍perspective.
ߨ௝ This is a real advantage of the GEP methodology, because
௝‫ீא‬ ݀‫ݕ‬௝‫כ‬ ௝‫א‬஻ ݀ሺെ‫ݕ‬௝‫ כ‬ሻ of the Dashboard of Sustainability with theweights
goal
ivalent to the expression (after substituting
it informs national and global action. the Asexpression
time passes forandthe
thenormalized
country’s situation evolves,
the GEP Index which to both
in the assess
is equivalent
GEP how
to thefar
ଵ Measurement off aௗ௬country
expression
௧ೕFramework is
ೕ (after will௬ೕ from
బ
adjust
ௗሺି௬ ೕ ሻ toof reflect
highlighting whether
the new planetary
set[2’]
of boundaries
priorities. This feature
‫ܲܧܩ‬of ൌ one బ ൈ ൤σ௝‫ ீא‬బ ൅ σ௝‫א‬஻ to ൨
ative importance
equivalent to the expression
substituting
makes ೟area
ೕ theGEP
the (indicator)
(after
expression
೤ forwith
substituting
௬the
ೕ Index (relative
ೕ
‫כ‬
ೕ respect
tothe
ௗ௬normalized
indexes the
expression ‫כ‬ for
have the normalized
been
௧ೕ withೕ ሻfixed/common
ௗሺି௬ overstepped
weighting or not.
for all It should
countries)be a
noted
well-
σೕ‫א‬ಸ బ ାσೕ‫א‬ಳ
. This is a real advantage
weights): ೤ೕ
of
suited approach the೟ೕ
GEP methodology,
for policy௧ design because
and monitoring.
௬ బ ௗሺି௬ ሻ that the thresholds of indicators in the dashboard
ଵ ௗ௬
[2’]in the GEP Index are
ೕ ೕ ೕ ೕ
As time passes ‫ܲܧܩ‬andൌ the ೟country’s ೤బ
൤σ௝‫ ீא‬௬ బ ௗ௬
ൈ situation ‫ כ‬൅ σ௝‫א‬஻weights
evolves, ‫ כ‬൨ and of some indicators
Finally, ೕ to assess GEP within
ೕ ೕ ೕ planetary ௧ ೕ boundaries,
ௗሺି௬ ೕሻ the progress achieved in the GEP Index
eighting system
will adjust to reflectallows
σೕ‫א‬ಸ the
ൌ the
‫ܲܧܩ‬indicators
ାσ GEP
೤ೕ new ೟set
బ ଵ ೕ‫א‬ಳ Index
ൈ of
to both
priorities.
௧ ೕ ௗ௬ ೕ assess
This
σ
௬ బ
how far off a country
feature൨ determined on the[2’]
ೕ ௗሺି௬ ೕ ሻ is from
basis of scientific literature, while
൤σ௝‫ீא‬
are೤ೕ బ compared ௬ೕబ ௗ௬ to‫ כ‬൅the ௝‫א‬஻progress
௧ೕ ௗሺି௬ೕ‫ כ‬ሻ made in the indicators of the Dashboard of
and to evaluate
exes with fixed/common the relative
σ
೟ೕ
ೕ‫א‬ಸ೤weighting
Sustainability బ ାσ importance
ೕ‫א‬ಳwith
೟ೕ
ೕ
forgoal
the
of ೕone area
all countries)
of highlighting
(indicator)
a well-whetherother
with respect
thresholds
planetary
to the
in the
boundaries GEP Index
have areoverstepped
been empirically
he country’s perspective. This is a real advantage of the GEP methodology, because
ೕ
monitoring.
eighting system allows or not.the GEP Index
It should to both
be noted thatassess how far determined.
the thresholds off a countryinis the
of indicators from dashboard and of some
ional
and and global
to evaluate action.
systemthe As
relative time passes
importance
Index to of and the country’s situation evolves, weights
in one area
GEP (indicator) with respect to the literature, while other
eary
weighting indicators
allows in
the GEP the GEP Index both are determined
assess how on
far off a the basis
country of scientific
is from
boundaries, the progress achieved the Index
easurement
he country’s Framework
thresholds
perspective. will
This adjust
in importance
the
is aGEP to
real reflect
Index of are the new
empirically set of
determined.priorities. This feature
old and tomade
rogress evaluatein thetherelative
indicators ofadvantage
the one of the
area
Dashboard GEP
of methodology,
(indicator) with respect to because
the
EP Index
ional and (relative
global to indexes
action. As with
timeispasses fixed/common weighting for all countries) a well-
haveand theoverstepped
country’s
m the country’s perspective. This a real advantage of the GEPsituationmethodology, evolves, weights
because
hting whether planetary boundaries been
ach for policy
national
easurement 3.33.4
design
and global THEand
action.
Framework The Asnecessity
monitoring.
NECESSITY time passes toOF ofA aDASHBOARD
and dashboard
the country’s situation evolves, weights
thresholds of indicatorswill in adjust
the dashboard reflect the
andnew of some set of priorities. This feature
EP Measurement
Index Framework
(relative to indexes will adjust
with to reflect
fixed/common the new set
weighting of priorities.
for all in This
countries)feature
sess
ermined GEP within
on(relative planetary
the basis boundaries,
of scientific the progress
literature, while achieved the GEPa Index well-
GEP Index It to
is important
indexes to remember
with fixed/common that the GEPother
weighting Indexfor isallnot intended atowell-
countries) be a "sustainable development
ch fordetermined.
policy design
e compared
ically to the andprogress
monitoring. made in the indicators of the Dashboard 20 of
index",
roach for policy design andnor an index of "progress" adjusted for sustainability . Assessing sustainability is an
monitoring.
It is important to remember that the GEP Index degree of substitutability of the different forms of
with GEP
sess the goal withinof highlighting
planetary
exercise that whether
involves planetary
boundaries, thethe boundaries
progress
future and isachieved have
therefore, been
in theoverstepped
primarily GEP andIndexunavoidably, a forecasting
assess GEP is not intended
within planetary to boundaries,
be a “sustainable the development
progress achieved human,
in the social,
GEP of economic and environmental capital
Index
ould be noted that the thresholds of indicators in the dashboard and some
board
e compared index”,
are GEP compared
to exercise.
to
thenorprogress
the an
A correct
progress
made
index of “progress”
made
in adjusted
and complete the indicators
inbasis
the20 of
assessmentofof the
for
indicators
Dashboard
sustainability
in generating
of literature,
the Dashboard
would
well-being;
of
of and
require: (a) a correctly
(e) a determination
the
with the goal of Index are determined
specified
highlighting dynamic on
whether the
stochastic
planetary model scientific
boundaries of the economy
have been while
and the
overstepped other
environment; (b) a correct
lity with the goal ofassessment
highlightingofwhether
sustainability.Assessing planetary
sustainability boundaries
is an have ofbeen
how overstepped
stringent sustainability tests should be
GEPthe GEP
ould be be
Index Index
noted
isnoted
not are
that empirically
intended the bedetermined.
thresholds
to
present
a the of and
"sustainable
future
indicators preferences
in the
development thedashboardfor
dashboard the inhabitants
and of of all
some countries; (c) a procedure
should that
exercise the
that thresholds
involves of
future indicators
and is in
therefore,
to rank 20social states within generations; (d) a correct assessment of the degree (e.g. and of
sustainabilitysome as “future welfare above currentof
the
djusted
in the GEP forIndex
GEP Index are
sustainability
primarilyare determined.
determined Assessing
and unavoidably, on
on the
the basis
sustainability
basis ofof scientific
is
scientific an literature,
literature, while
while other other
substitutability of theadifferent
forecasting formsexercise.
of human,A welfare”economic
social, vs. “sustainability as non-decreasing
and environmental capital in
ecessity
ndthe isGEP
in the GEPIndex
therefore,of
Indexacorrect
are dashboard
are empirically
primarily
empirically and
and complete
generating
determined.
unavoidably,
determined.
assessment
well-being; and (e) ofaasustainability
forecasting
determination ofwelfare”). With thissustainability
how stringent objective in mind, testsand underbe
should
ssessment of would sustainability
(e.g.require:
sustainabilitywould
(a) a correctly require:
specified
as "future (a)
welfare a
dynamiccorrectly
above currentsome regularity
welfare" conditions, oneascan
vs. "sustainability deem the
non-decreasing
t to
ecessity
el remember
necessity
of the economy that
ofofastochastic the
adashboard
dashboard
and
welfare"). GEP the
modelWithIndex is not
environment;
of this intended
objectiveand
the economy (b)
in mind, ato be
correct
the and under somea "sustainable development
currentregularity
socioeconomic conditions,
path asone can deem
sustainable the
if and
20
nferences
index of for"progress"
the inhabitants
current adjusted of
socioeconomic for
all sustainability
countries;
environment; (b) a correct assessment of presentpath as(c) a . Assessing
procedure
sustainable if and onlysustainability
if
if ܸ݀ሺ‫ݐ‬ሻ ൒ Ͳǡ is an
where:
, where:
ʹͳ
21

tant
t to involves
erations; remember (d)
to remember the aand
future
that correct
that the
future and
theGEP isIndex
assessment
GEP
preferences therefore,
Index the of
is not
is
for primarily
the
intended
intended ܸ݀ሺ‫ݐ‬ሻǣ
inhabitants to
toofand
degree bebe
ൌallσa௞aunavoidably,
‫݌‬of
"sustainable
"sustainable
௞ ή ݀‫ݏ‬௞ ሺ‫ݐ‬ሻ
adevelopment
forecasting
development (3)
rnorrect
an
of index
human,
index and of complete
"progress"
social,
of "progress"
countries; assessment
economicadjusted
adjusted
(c) a procedureand for of sustainability
sustainability
forenvironmental
sustainability
to rank social states
20
20.
capitalwould
Assessing
. Assessing in require: (a)
sustainability a correctly
is
ܸ݀ሺ‫ݐ‬ሻǣ
sustainability is an ௞ ௞ an ൌ σ ‫݌‬ ή ݀‫ݏ‬ ௞ ሺ‫ݐ‬ሻ
namic
hat
ermination
involves thestochastic
involves the
of how future
within model
future and
stringent of is the
and is sustainability
generations; economy
therefore,
therefore, primarily
(d) a correctprimarily and
teststhe
and
assessment the
and
should
and
൫of environment;
unavoidably,
be the normatively
൯unavoidably,
‫݇݌‬the are a (b) a
forecastingcorrect
a forecasting determined "shadow prices" associate
A
of
above correct
present and
current
orrect and completeand complete
future
welfare"
20
assessment
Forpreferences
a method
vs. for
to combine
"sustainability ofthe
an sustainability
inhabitants
assessment
as would
of
of development
non-decreasing all require:
adjusted by(a)
countries; (c)a acorrectly
procedure
sustainability, see Pineda (2012).
21
Thisassessment of sustainability
implies a non-decreasing economic would
and
discounted utilitarian require:
environmental
sum (a) capital
of generational a correctly
utility, stocks ሺ‫ ݇ݏ‬ሻ.paths along which
i.e. economic
,dynamic
ial
namicandstatesunder stochastic
stochasticwithin model
generations;
someintergenerational
regularity
model
of well-being
the economy
the (d)
of conditions, economya one
does not and
correct
can
and For
decline. the
deem
the
environment;
assessment
more the
environment;
information, of (b)
the
see Marc (b)
a degree
correct
a correct
Fleurbaey of Blanchet
and Didier (2013).
nt y of
of present
the and future
different forms preferences
of human, for the
social, inhabitants
economic of all
andcountries;
19
environmental(c) a procedure
capital in
ofnablepresent if andand future
only if ܸ݀ሺ‫ݐ‬ሻ preferences
൒ Ͳǡ ʹͳ where: for the inhabitants
This measurement of all countries; exercise (c)would
a procedure be
social
ell-being; states
and within
(e) a generations; (d)
determination of a correct
how stringent assessment
sustainability of the testsdegree
should of demanding
be
in practice and questions
ial
ൌ σ states
‫݌‬ ή ݀‫ݏ‬
bility ௞of௞the different
within
௞ ሺ‫ݐ‬ሻ generations; (d)
forms of human, social, iteconomic
a correctwould be assessment (3)
andpossible
environmental
of the
to produce degree
capital the
of
in "correct" sustainability assessme
ybility
of as "future
the different welfare
forms above
of
well-being; and (e) a determination of how system. human,current welfare"
social, economicvs. "sustainability
stringentAdditionally, and as non-decreasing
environmental
sustainabilityittests is highly capital
should be in that this exercise would be widel
unlikely
h this objective in mind, and under some regularity conditions, one can deem the
 ܸ݀ሺ‫ݐ‬ሻǣ ൌ σ௞ ‫݌‬௞ ή ݀‫ݏ‬௞ ሺ‫ݐ‬ሻ (3)
and the ൫‫ ݇݌‬൯ are
and are the ܸ݀ሺ‫ݐ‬ሻǣ
thenormatively
normatively σ
determined
determined ௞ ሺ‫ݐ‬ሻ
ൌ ௞ ‫݌‬௞ ή ݀‫"ݏ‬shadow thatprices" associated
are taking place withwith human,
these (3) one
stocks.social,
Thus,
15 and the ൫‫݌‬prices”
“shadow
economic ൯ are associated
the normatively
݇and environmental with determined
human,
capital ሺ‫"ݏ‬shadow
social,
stocks ሻ. prices"
could associated
present relevant with human,
thresholds thatsocial,
(according
݇
economic
economic and and environmental
environmentalcapital capital stocks stocks ሺ‫ ݇ݏ‬ሻ. to the relevant scientific literature) are desirable/
This measurement exercise would be demanding in practice and questions
not desirable to cross and, remain as to whether
in addition, targets that
This
it Thismeasurement
would be exercise
possible to
measurement exercise would be demanding would
produce be demanding
the "correct" in practice
sustainability and questions remain
will serve to measure the extent to whichprice
assessment and as to
shadow whether progress
itsystem.
would
in practice be possible
Additionally,
and questions to produce
it is remain
highly as the
unlikely "correct"
to whether sustainability
that this exercise is being would assessment
made betowards
widely meeting and shadow
accepted as aprice
certain guide
social,
system. Additionally, itܸ݀ሺ‫ݐ‬ሻǣ
isthat ൌisσat௞unlikely
highly ௞ ή ݀‫ݏ‬
‫݌‬the ሺ‫ݐ‬ሻ this exercise would be widely accepted(3)
௞ that asThis
a guide
for policy
it would – possible
be a problem to produce the core
“correct” of how to take into account
economic not just thegoals.
or environmental well-being, but
is the
and the ൫for
‫݌‬also൯ are
policy the
– a normatively
the preferences problem that determined
of future is at the core"shadow of how prices"
to take associated
into account with not human,
just the social,
well-being, but
݇sustainability assessment and individuals
shadow price yet to be born. objective of the dashboard, in combination with the
economicalso and the preferencescapital
environmental of future individuals
stocks
system. Additionally, it is highly unlikely that this
ሺ ‫ݏ‬ ݇ ሻ . yet to be born.
GEP Index.
One way to proceed in light of these challenges is to remain agnostic about prices and keep
exercise would be widely accepted as a guide for
One
track
This measurement way of totheproceed
exercise changes
would in in light
be the ofstocks,
demandingthese ݀‫ݏ‬ challenges
in௞ , practice
and present isandto questions
remain
those changesagnostic
remaininabout prices and
asatodashboard
whether forkeep
each
trackpolicyof –
thea problem
changes that in istheat the
stocks,core ݀‫ݏ‬ of how, and to take
present Can wechanges
those dispensein with
a the index-dashboard
dashboard for each
it would be country.
possible This to more
produce modest approach௞sustainability
the "correct" is compatibleassessment with: (a) anand outright
shadow acceptance
price of the
into account
country. This not just
more the well-being,
modest but alsocompatiblethe dichotomy bycapital
simplyunder adding the dashboardor,
intrinsically
system. Additionally, itlimited
is highly unlikelyapproach
substitutability this is
that between exercise wouldwith:
the different beforms (a) of
widely an outright
accepted as acceptance
guide of the
a consideration
preferences
for policyintrinsically
–even
a if it wasn’t
problem of future
limited
that at individuals
issubstitutability
limited; the(b) core theofyet to be
between
extraordinary
how toborn. the into
take different
difficulty,
accountvariables
forms
both ofinto
notethical
just the
capitaland
the index?
under We
technical,
well-being, dobut not
consideration believeor,
in identifying that
even if
the properof
also the preferences it wasn’t limited;
"trade-offs"
future individuals (b)
betweenyetthe extraordinary
formsto beofborn. difficulty, this
both
capital, as discussed above. approach
ethical is
and appropriate.
technical, To
in see this,
identifying imagine
theOne properway"trade-offs"
to proceed inbetween light of these formschallenges
of capital,isas discussed that weabove. have an index of ‘sustainability adjusted
One way toNotwithstanding
to in lightthe
remain agnostic
proceed about
of fact
these that
prices those
and keep
challenges prices toare
istrack of hard agnostic
remain to pin down,
well-being’ thatwe
about acts must
prices likeand do GEP
the what
keepinwe can to the
regards
Notwithstanding
track of the assess
changes the in the
importance
the fact
stocks, of
the changes in the stocks, ௞ and present thosethat
the
݀‫ݏ‬ those
magnitude
, and prices
present of are
the
thosehard
changes to variables that matter for present well-beingtobut that
changes pin thatdown,
inarea we
taking must
dashboard place do what
with
for each we
these can
stocks.
assess
country. ThisThus, more
changes the
onein importance
could
modest present
a dashboard approachof the
for magnitude
relevant
each thresholds
is compatible
country. of the
This with: changes
that (according
more anthat
(a) penalizes are taking
tothe
outright theacceptanceplace
relevant
growth withof these
scientific
in variables thethat stocks.
literature)
threaten the
Thus,
intrinsicallyare
limited one could
desirable/not present
substitutabilitydesirable relevant
to
between
modest approach is compatible with: (a) an outright cross thresholds
the and,
different in that
addition,
forms(according
targets
of capital to
that the
under relevant
will serve to
consideration scientific
sustainability of that well-being. Any such measure or, literature)
the extent
index
are
even if it wasn’t desirable/not
toacceptance
which progress
limited; (b) desirable
theis being to
extraordinary cross
made and,
towards in
difficulty, addition,
meeting
both targets
certain
ethical that
and will
social, serve
technical, economic to
in measure
or
identifying the extent
environmental
of the intrinsically limited substitutability may end up classifying countries having, say, low
the propertogoals.
whichThis
"trade-offs" progress
is
between is being
the objectiveforms
between the different forms of capital under
made
of
of the towards
dashboard,
capital, as meeting
discussed certain social,
in combination
above. with the economic
GEP Index. or environmental
life expectancy and low greenhouse gas emissions
goals. This is the objective of the dashboard, in combination with the GEP Index.
consideration or, even if it wasn’t limited; (b) the and a high value for both in a similar way, while their
Can wethe
Notwithstanding dispense
fact that with those the index-dashboard
prices are hard to dichotomy
pin down,bywe simply
mustadding do what thewe dashboard
can to variables 22

Can extraordinary
we dispense difficulty,
with the both ethical and technical,
index-dashboard dichotomy in bypositions
simply clearly
adding need
the to be differentiated.
dashboard variables This
assess theinto the index?
importance of We
the do not believe
magnitude of the thatchanges
this approach that areistakingappropriate.place with To see these this, imagine that we
stocks.
Thus, oneintoanidentifying
could the
index of the
index?
present Weproper
do not
‘sustainability
relevant
“trade-offs”
believe
adjusted
thresholds
between
that this
well-being’
that
forms
approach
(according
of is appropriate.
that toacts
follows as long
the like the GEP
relevant Toasin see
scientific
wethis,adoptimagine
regards the principlethat we
to the variables
literature)
that “we
an capital,
that
are desirable/notindexmatter as‘sustainability
of discussed
desirable for present
to cross above. adjusted
well-being
and, in addition, well-being’
but that thatthat
penalizes
targets consider
actswillthe theitGEP
likeserve
growth ourmeasure
to inmoral duty
invariables
regards thenotto
that to
the
extent impose
variables
threaten onthe future
that matter
sustainability
to which progress for
is being present
of that
made well-being
well-being.
towardsAny but that
such index
meeting penalizes
certain may generations
the
end up
social, growth
classifying
economic any
in form
variables
or countriesof sacrifice
that
environmental having, say, lownot
that
threaten we do
the
23
goals. Thissustainability
isNotwithstanding
life expectancy
the objective of that
ofthe
and well-being.
the fact
low that those
greenhouse
dashboard, Anyinprices
such gasindex
combination are
emissionsmaywith
hard end upaGEP
accept
and
the classifying valuecountries
forIndex.
high ourselves”. for both having, in a similar say, low
way,
life
whileexpectancy
to pintheirdown, and
positions low
we must clearly greenhouse
do whatneed we can gas
to tobeassessemissions and 22a high value for both in a similar way,
differentiated. This follows as long as we adopt the
22
while
principle
Can we dispense their positions
that “we
with the index-dashboard
the importance clearly
consider
of the magnitude of need
it our to be
moral
dichotomy
the changesdifferentiated.
duty bynot simplyto impose This follows
adding on
thefuture asgenerations
dashboard longvariables
as weany adoptform theof
principle
sacrifice
into the index? We do that
thatnot“we consider
webelieve
do notthat acceptit our
thisfor moral
ourselves.”
approach duty not
23 to impose on future generations any form of
is appropriate. To see this, imagine that we
23
an index ofsacrifice that we do
‘sustainability not accept
adjusted for ourselves.”
well-being’ that acts like the GEP in regards to the variables
that matter3.5 for Aggregating
present well-being thebutinformation
that penalizesfrom the growth the dashboard
in variables that and the GEP
threaten the Index,
3.5 Aggregating
creating
sustainability the GEP+
of that well-being. theAnyrankinginformation
such index may end up classifying countries having, say, lowIndex,
from the dashboard and the GEP
creating
life expectancy and low thegreenhouse
GEP+ ranking gas emissions and a high value for both in a similar way,
22
while their Let the convention
positions clearly need be that to be thedifferentiated.
variables in the GEPfollows
This Index contribute,
as long asinwea adopt comprehensive
the way,
Let the
towards
principle that convention
the measurement
“we consider be that
it our moral the
of dutyvariables
the welfare in
not to impose the GEP
or development Index contribute,
on futureofgenerations
the presentany in a comprehensive
generation,
form of and carry way,
towards
some
sacrifice that we do thenotmeasurement
limited information
accept of the
on
for ourselves.” welfare
23
its sustainability. or development
Variables that of the arepresent
related generation,
to the sustainability and carryof
some
development are placed in the dashboard. Just as progress was calculated for each indicator of
limited information on its sustainability. Variables that are related to the sustainability y in
3.5 Aggregating
development theare information
placed in the from
dashboard.
the GEP Index, it is calculated for each indicator K in the dashboard as the Just dashboard
as progress and
was the
calculated GEP for Index,
each
ௗ௄ ೕ indicator
, for all relevant y in
‫כ‬
creatingthethe GEP+
GEP Index,ranking
it is calculated for each indicator K in the dashboard as
ௗ௄ೕ‫כ‬
, for all relevant
ௗ௄
ௗ௄
ೕ ೕ

indicators ݆ ൌ ͳǡ ǥ ǡ ‫ܬ‬.
indicators ݆ ൌ ͳǡ ǥ ǡ ‫ܬ‬.
Let the convention be that the variables in the GEP Index contribute, in a comprehensive way,
towards the measurement of the welfare or development of the present generation, and carry
some limited22
23
information
See on its
more on this from sustainability.
Fleurbaey and BlanchetVariables that are related to the sustainability of
(2013, p. 21).
22 Fleurbaey and Blanchet (2013, p. 50).
development Seeare placed
more on this in the
from dashboard.
Fleurbaey and Just
Blanchet as progress
(2013, p. 21). was calculated for each indicator y in
23
Fleurbaey and Blanchet (2013, p. 50). 20 ௗ௄
the GEP Index, it is calculated for each indicator K in20the dashboard as ೕ‫ כ‬, for all relevant
ௗ௄ ೕ

indicators ݆ ൌ ͳǡ ǥ ǡ ‫ܬ‬.

22
See more on this from Fleurbaey and Blanchet (2013, p. 21).
23
 goals. This is the objective of the dashboard,
sustainability in combination
threshold), progress with
for the second the GEPrelates
country
this progress Index.
shouldtobethe
considered relatively
sustainability more This gi
thresholds.
social, economic or environmentalimportant for the overall progress of this country, and the
importance planet,
of this towards
progress. IGE.
For To capture
example, if twoboth
countries ex
with the GEP Index. aspects, we multiply progress on each dashboard indicator by
Can we dispense with the index-dashboard dichotomy by simply adding the dashboard variables
country was already on a
a weight relating
sustainable path the
whileinitial
another coun
into the index? We do condition
not believe to the threshold, ߨො௝ , (explained previously
that this approach is appropriate. in section
sustainability
To see 3.2). progress
threshold),
this, This weighting
imagine forwe
that therequires
second an
country sh
mply adding theandashboard variables additional modification to the GEP Index important in order forto the
allow for comparability
overall progress of thisbetween
country,the
and the pla
index of ‘sustainability adjusted well-being’ that acts like the GEP in regards to the variables
opriate. To seethat
this,matter
imagine measures of progress of the GEP Index and the dashboard indicators (now multiplied
aspects, we multiply progress on each dashboard indicat by the
forthat we well-being but that penalizes the growth
presentweight). Therefore, in order to facilitate the comparison
in variablesforthat threaten theof the GEP+,
the GEP in regards to the variables condition toneeded
the threshold,the construction
ߨො௝ , (explained previously in sectio
sustainability of that well-being. Any such
we must multiply the index may end
GEP Index up average
by the classifying
additional
countries
weights,having,
of modification
the ௝ , of
ߨොto say,
theitsGEP
low
indicators.
Index inThese
order to allo
life3.4
owth in variables AGGREGATING
that threaten
expectancy andthe
low THE INFORMATION
greenhouse gas emissions FROM andTHE
a
modifications allow for a comparable achievement
DASHBOARD
high value
measures for
profile both ANDin
of each country
of progress a THE
similar
of the in
GEP way,
the Index
GEP
INDEX,
sample. and the dashboard
CREATING
lassifying countries
while having,
their THElowGEP+
say,
positions clearlyRANKING
need to be differentiated.22 This follows
weight). as long
Therefore, as weto adopt
in order facilitatethethe comparison neede
igh value
16 for both in a similar way,
principle that “we consider it our
For a given moral“x”duty
country, is annot to impose
achievement onmust
profile
we future
vector ofgenerations
dimension
multiply the GEP any
(J+1), Indexform
which
by isofgiven
the by theof the w
average
s follows as long as we
sacrifice thatadopt
we dothe
notGEP Index for
accept multiplied by the
ourselves.” 23 average of ߨ
ො௝ across its indicators
modifications allow foranda acomparable
set of weighted progress
achievement profile of e
on future generations any form of
Let the convention bemeasures
that the variables
of in thetheGEP J of the
ௗ௄భ
J dashboard ‫ߨ ۃ‬ො ǡ ǥ ǡ
dashboard
ௗ௄಻
ߨො ‫ ۄ‬. . Therefore
ௗ௄భ‫כ‬ ଵ ௗ௄಻‫כ‬ ௃
3.5 Aggregating
Index the information
contribute, in a comprehensive from
way, towards
ௗ௄
theௗ௄ dashboard
For a givenand the
country, “x” GEP ೣIndex, profile vector of dim
is an achievement ೣ
š ൌ ൬‫ߨ݁݃ܽݎ݁ݒܽ כ ܲܧܩ‬ො௝ ௫ ǡ ‫ۃ‬ௗ௄ ߨොଵ ǡ ǥ ǡ ௗ௄ ߨො௃ ‫ۄ‬൰.GEP Index multiplied by the average of ߨො௝ across its indicato
భ ಻

hboard andcreating
GEPthe GEP+
of the ranking
ೣ‫כ‬ ೣ‫כ‬
the measurement welfare or development Therefore
the Index,
భ ಻
ௗ௄
measures of the J dashboard ‫ۃ‬
of the present generation, and carry some limited ௗ௄
Although, for the reasons explained above, the GEP Index should ௗ௄ notೣ be combined with the
Let the convention
information be that the variables
on its sustainability. Variables in the
that are GEP Index contribute, š ൌ ൬‫ܲܧܩ‬of‫ כ‬sustainable in a comprehensive
ܽ‫ߨ݁݃ܽݎ݁ݒ‬ ‫ ۃ‬ೣభ‫ߨ כ‬ොଵ ǡ ǥ ǡ way,
ො௝ ௫ ǡdevelopment,
ௗ௄ ೣ
಻
‫ߨכ‬ ‫ۄ‬൰..
ො௃information
Dashboard of Sustainability in a composite measure ௗ௄భ the
ೣ
towards the measurement of the welfare or development of the present generation, and ௗ௄ carry
಻

ontribute, in asome related to the sustainability

comprehensive way,from on of development
both instrumentsare can nonetheless inform which countries are in a comparably more
limited information favorable its sustainability.
position thanwas others. Variables that are related to the sustainability of
placed
f the present development
generation, and in the dashboard. Just as progress Although, Although,for thefor reasons
the reasons explained above,above,
explained the GEP the GEP Index
are carry
placed in the dashboard. Just as progress was Dashboardcalculated of for each indicator
Sustainability in a y in measure of sustai
composite
t are related to calculated
the sustainability for each of indicator y in the GEP Index, it is
The GEP+ is for a final ranking by comparing
Index should not be combined
progress on instruments
the as ௗ௄ೕ
indicators
with the Dashboard
in all
the dashboardinform with any
the GEP Index, it is calculated each indicator K in the fromdashboard
both , for
can relevant
nonetheless which cou
was calculated for calculated
each indicator for eachy indicator
ingreen economy K in the progressdashboard made,as as measured of Sustainability in a index.
by the position
GEP composite ‫כ‬
measure of allows us to
ௗ௄ೕ This methodology
favorable than others.
dashboard as indicators
ௗ௄ೕ
, , for
for ݆all ൌ relevant
ͳǡ ǥ ǡ ‫ܬ‬. rank all index-dashboard profiles but not sustainable
to combine development,
the index and thedashboard
information from into a
information
ௗ௄ೕ‫כ‬
synthetic index24. When comparing progress based
both instruments
The GEP+on the iscanaGEP Index
nonetheless
final rankingand by the dashboard,
inform
comparing which countrieson the in
progress
are ranked according to their least-performinggreen type of progressprogress
economy based onmade, the principle
as measuredof Priority to GEP in
by the
However, for the dashboard indicators it is countries are in a comparably more favorable
the Worst Achievement. This methodology sends
rank the
all policy
index-dashboardmessage that
profiles a country
but not that
to is only
combine the inde
22
critical
See moreto onunderstand
this from Fleurbaey not only
making and progress
Blancheton
progress but
(2013,
a few also
p.aspects
21). of anposition
Inclusive than
Green others.
Economy24 will not necessarily be doing better
23
Fleurbaey and Blanchet (2013, p. 50). synthetic index . When comparing progress based on the GEP
how this progress relates than to onethe thatsustainability
is moving forward in all areas. Ranking countries based on the area
are ranked according to their least-performing type of progress in which they are
20
thresholds. This givesmaking specific theinformation
least progress about gives the policy Theincentive
GEP+
the Worst is to countries
a final ranking
Achievement. to implement
byThiscomparing a more
methodology balanced
progress sendsand the policy m
integrated
the importance of this progress. For example, if policy approach that is aimed at
on themaking moving
indicators forward
progress across
in theondashboard the
a few aspects broad
with ofanyspectrum
green of
an Inclusive an Econom
Green
Inclusive Green Economy. This methodologythan serves a
one that isdouble movingpurpose for countries undertaking
two countries experienced similar progress but one economy progress made, asforward
measured in allbyareas.
the GEP Ranking countries
Inclusive Green Economy action: it allows24making them to thelearn
least about
progress theirgivesrelative green
the policy economy
incentive to countries
country was already on a sustainable
performance whilepath alsowhileinforming them indexon how . This methodology allows
integrated policy approach that is aimed atcompare
their least-performing areas us to
of rank
progress all moving forward
25
another country was not withsustainable
the achievements (above of the
other countries index-dashboard
. Inclusive Green profiles
Economy.but notThis to combine
methodology the serves a double
However, for the dashboard indicators it is critical to understand not only progress but also how
sustainability threshold), progress for the second index and Inclusivedashboard Green information
Economy action: into aitsynthetic
allows them to learn a
this progress relates to the sustainability thresholds. This gives specific information about the
performance while also informing themon on thehow their least-per
importance country should be considered
of this progress. For example, relatively
if twomore countries experienced index . When
with similar comparing
progressprogress
the achievements but onebased
of other countries25.
country wasimportant
already on forathe overall progress
sustainable path while of thisanother
country,countryGEP wasIndex and the dashboard,
not sustainable (above countriesthe are
and the planet, towards IGE. To capture both
sustainability threshold), progress for the second country should be considered relatively more ranked according to their least-performing type of
or the dashboard indicators
aspects, it
we multiply is critical
progress to understand
oncountry,
each dashboard not only progress but also how
important for the overall progress of this and the planet,progress towards based
IGE. on Tothe principle
capture bothof Priority to the
ss relates to the sustainability thresholds. This is so Thisbecausegives specific information
(see Annex I.C),about the does not admit a real-valued representation due
24
the Protective Criterion like the leximin,
aspects, weindicator
multiply by progress
a weight relating toon each
the lack the
of dashboard
initial
continuity ofcondition indicator
the preference ordering.by For a
Worst weight
Achievement.
proof relating
see the example This
of the
Moulin initial
methodology
(1998, page 34). sends the
e of this progress. For example, if two 25
countries
This experienced
method of creating the rankingsimilar limits the progress
incentives forbut one across equally important aspects of an Inclusive
substitution
condition totothe threshold, ߨ
the threshold, ௝ , (explainedො , (explained
Green Economy; previously
previously in section
it also givesinincentives to progress 3.2).
policy This
in allmessage weighting
aspects and that requires
a country
penalizes any partialthat an
viewis thatonly
onlymaking
concentrates on a
as already on a sustainable path while another
policy areas.country was not sustainable24 (above the
additional modification
section 3.2). This thefewGEP
to weighting Index inan order
requires additional to allow progress for comparability on a few between
aspects of an the
Inclusive Green
This is so because the Protective Criterion (see Annex I.C), like the leximin, do
ity threshold), progress for the second country should be considered relatively 21lackmore
measures of progress of
modification to thetheGEP GEPIndex Index and the
in order to allowdashboardfor indicators
Economy
to the(now
Thiswill notmultiplied
necessarily bybethe
of continuity of the preference
doing
ordering. For proof see the example of M
limits better than for substitution acr
for the overall progress of this country, and the planet, towards IGE. To capture both
25
method of creating the ranking the incentives
weight). Therefore,
comparability in order to
between facilitate
the measures the comparison
of progress needed for the
one that construction
Green
is Economy; it also
moving of the
gives GEP+,
incentives to progress in all aspects and penalize
we multiply progress on each dashboard indicator by a weight relating fewthepolicy areas. forward in all areas. Ranking
initial
we must multiply the GEP Index by the average of the weights, ߨ
ො , of basedindicators.
its on the area These
o the threshold,ofߨො௝the GEP Index previously
, (explained and the dashboard in section indicators
3.2). This weighting countries ௝
requires an in which they are 21
modifications allow for a comparable achievement profile of each country in the sample.
modification to(now the multiplied
GEP Index by the in weight).
order to Therefore,
allow for in comparability making the least
between the progress gives the policy incentive
of progress of order the GEP to facilitate
Index and the comparison
the dashboard needed for the (now to
indicators countriesby
multiplied to implement
the a more balanced and
For a given country, “x” is an achievement profile vector of dimension (J+1), which is given by the
herefore, in orderconstruction
to facilitateofthe thecomparison
GEP+ ranking, needed
we must for multiply
the construction integrated of thepolicy GEP+, approach that is aimed at moving
GEP Index multiplied by the average of ߨො௝ across its indicators and a set of weighted progress
multiply the GEP Index by the average of
the GEP Index by the average of the weights, ௝ , ௗ௄ forward the weights, ߨො , of its indicators. These
across the broad spectrum of an Inclusive
of However, for the dashboard indicators‫ ۃ‬it భisߨො critical ௗ௄಻ to
measures
ons allow for a comparable
of its indicators.
theThese
achievement J profile
modifications
dashboard
of eachallowcountry
for a ௗ௄
ǡ sample.
ǥ ǡ ௗ௄Economy.
in భ‫כ‬theଵGreen ‫ߨכ‬ ො௃ ‫ۄ‬understand not
.This methodology only progress but also how
Therefore serves a double
this progress relates
ೣ
to the sustainability thresholds. This gives specific information
಻ about the
comparable
š ൌ ൬‫ߨ݁݃ܽݎ݁ݒܽ כ ܲܧܩ‬ ௫ achievement
ොimportance
ǡ ‫ۃ‬
ௗ௄భ ೣ
ߨො ǡ ǥ ofǡ profile
ௗ௄
this
಻
ߨො of each country
progress.
‫ۄ‬൰. For in
example, ifpurpose
two for countries
countries experiencedundertaking similarInclusive
progress Green but one
௝ ‫כ‬ ଵ ௗ௄಻ೣ ௃of dimension (J+1), which is given by the
‫כ‬
n country, “x” is an theachievement
sample.
ௗ௄భೣprofile vector Economy action: it allows them
country was already on a sustainable path while another country was not to learn about(above
sustainable their the
multiplied by the average of ߨො௝ across its indicators and a set of weighted relative country progress
green economy
sustainability threshold), progress for the second should performance
be considered while also
relatively more
Although, for the reasons explained above, ‫ ۃ‬the
ௗ௄ GEP Indexௗ௄ should not be combined with the
of the
For a given Jimportant
country, dashboard
foristhe
“x” overall
an achievement
భ
ߨ
ො
progress
ଵ ǡ ǥ ǡ
profile
಻
of ߨ
ො ௃ ‫ۄ‬
this .
country,
informing Therefore
and themthe planet,
on how towards
their IGE.
least-performing To capture
areas both
Dashboard of Sustainability in a composite measure ௗ௄భ‫כ‬ ‫כ‬
ofௗ௄sustainable
಻ development, the information
vector
ೣ aspects,
of dimension
ௗ௄ ೣ we
(J+1), multiply
which isprogress given by the on each dashboard of progressindicator compare with by atheweight achievements relatingofthe other initial
from ොboth
ܽ‫ߨ݁݃ܽݎ݁ݒ‬ ௫ ௗ௄ instruments
௝ ǡ ‫ۃ‬ௗ௄ೣ ‫ߨ כ‬
భ
ො ଵ ǡ ǥ ǡ ௗ௄ ಻ can nonetheless inform which countries are in a comparably more
ೣ‫ߨ כ‬ ො௃ ‫ۄ‬൰. to the threshold, ߨො , (explained previously 25
GEP Index condition
multiplied in section 3.2). This weighting requires an
favorable position
భ
than ಻
others. by the average of ௝ across its countries.
indicatorsadditionaland a set ofmodification weighted progress to the GEP Index in order to allow for comparability between the
measures
for the reasons explained measures
above, theof progress
GEP Index of should
the GEPnot Index and the dashboard
be combined with the indicators (now multiplied by the
The GEP+ is a final ranking by comparing progress on the indicators in the dashboard with any
weight).measure
of Sustainability in a composite Therefore,
of in order to facilitate
sustainable the comparison
development, needed for the construction of the GEP+,
the information
green economy progress made, as measured by the GEP index. This methodology allows us to
we must
instruments can nonetheless multiply
inform the countries
which GEP Index arebyinthe average of the
a comparably moreweights, ߨො , of its indicators. These
rank all index-dashboard profiles but not to combine the index and dashboard information ௝into a
position than others.24 modifications allow for based
a comparable achievement
synthetic index . When comparing progress on the GEP Index andprofile of each country
the dashboard, countriesin the sample.
are ranked according to their least-performing type of progress based on the principle of Priority to
 4. FINAL CONSIDERATIONS
17
The Green Economy Progress measurement
framework highlights certain critical aspects of
the development challenges that policymakers
must address, in an integrated manner, to ensure
societies are able to transition towards an Inclusive
Green Economy. These challenges include, among
others, controlling an increasing material footprint,
rising emissions and increased freshwater
withdrawal, while at the same time ensuring that
further development is not compromised and that
economic opportunities are created, ecosystem
services are preserved and social inclusiveness
is promoted. The GEP Measurement Framework
addresses these challenges by providing the
advantage of having a double lens through which
one can examine progress towards an Inclusive
Green Economy.
The Green Economy Progress Measurement
Framework, in its current version, proposes a
method for measuring progress that monitors
changes in key variables, taking into account
global thresholds that should not be surpassed
and utilizing achievable targets selected to help
countries on moving in the right direction through
policy intervention. These components are critical to
obtaining a useful measure of progress, making the
measurement framework a valid instrument for not
only practitioners, but also the wider community of
researchers and academics working in the field.
There are important challenges associated with
this line of work and it should be noted that there
is still much progress to be made. Conceptual
challenges remain with respect to the integration of
the GEP Measurement Framework and the Inclusive
Green Economy narrative, because of the latter’s
complexity and the different implicit and explicit
causal relationships that exist. In addition, there are
empirical challenges related to the availability of
indicators. In fact, one important limitation of the
GEP Measurement Framework is the lack of data
for a large group of countries and for a long period
of time with which to measure progress. While the
focus on progress is a significant added value of
this work, it also imposes considerable constraints
on the potential indicators that can be used.
For example, some available indicators are only
approximate proxies of what we are attempting to
measure, while other indicators are of better quality,
but are limited in time and country coverage. From
a policy perspective, an additional challenge lies in
how to make use of available national indicators,
which tend to better capture local realities.
The methodology offered by the GEP measurement
framework is flexible enough that it can overcome
these practical challenges. A separate publication
presents an application of the methodology at the
global level, where progress was measured for a
sample of 105 countries between 2004 and 2014.
The purpose behind the application is to Beta
test the methodology and learn about its different
tradeoffs and challenges. This, in turn, can help
improve the methodology’s design, and most
importantly, enrich the overall process of green
26
economy policy making analysis.
 5. REFERENCES
18

Barberà, S. and Jackson, M. (1988). Maximin, Leximin, and the Protective Criterion: Characterizations
and Comparisons. Journal of Economic Theory, 46:3444 (1988). Available at: http://citeseerx.ist.psu.edu/
viewdoc/download?doi=10.1.1.297.1477&rep=rep1&type=pdf

Barro, R. and Lee J.-W. (2013). A New Data Set of Educational Attainment in the World. Journal of
Development Economics, 104:184-198.

Bringezu, S. (2015). Possible Target Corridor for Sustainable Use of Global Material Resources. Resources,
4(1), 25-54.

Dasgupta, P., Mäler, K.-G. (2001). Wealth and sustainable development. World Economics, 2 (3).

Fleurbaey, M. and Blanchet D. (2013). Beyond GDP Measuring Welfare and Assessing Sustainability.
Oxford: Oxford University Press.

Green Growth Knowledge Platform (GGKP). (2013). Moving towards a Common Approach on Green
Growth Indicators. Seoul, Paris, Nairobi and Washington DC: Global Green Growth Institute, Organisation for
Economic Co-operation and Development, United Nations Environment Programme and World Bank

Green Growth Knowledge Platform (GGKP). (2016). Measuring Inclusive Green Growth at the country
level. Seoul, Paris, Nairobi and Washington DC: Global Green Growth Institute, Organisation for Economic
Co-operation and Development, United Nations Environment Programme and World Bank

Herrero, C., Villar A. and Zambrano E. (2016). Technical notes for the construction of the Green Economy
Progress Measurement Framework. Mimeo.

Leadley, P.W. et al. (2014). Progress towards the Aichi Biodiversity Targets: An Assessment of Biodiversity
Trends, Policy Scenarios and Key Actions. Montreal: Secretariat of the Convention on Biological Diversity.

Moulin, H. (1998). Axioms of Cooperative Decision Making. Cambridge: Cambridge Univ. Press.

Nykvist, B., et al. (2013). National Environmental Performance on Planetary Boundaries: A study for the
Swedish Environmental Protection Agency. SEPA: Stockholm.

Statistical Office of the European Communities. (1999). The Environmental Goods and Services Industry:
Manual for Data Collection and Analysis. Paris: OECD Publishing.

Palma, J.G. (2011). Homogeneous middles vs. heterogeneous tails, and the end of the ‘Inverted-U’: the
share of the rich is what it’s all about. Cambridge Working Papers in Economics (CWPE) 1111. Cambridge:
Cambridge University.
19

Pineda, J. (2012). Sustainability and human development: a proposal for a sustainability adjusted HDI
(SHDI). 2012. ASERS. Theoretical and Practical Research in Economic Fields. Vol. III: 2(6), Winter 2012.

Pineda, J. and Galotto, L. (2015). Review of existing Green Economy Indicators. Mimeo

Rockstrom, J. et al. (2009). Planetary boundaries: exploring the safe operating space for humanity. Ecology
and Society.14(2): 32.

Schmoch, U. (2008). Concept of a Technology Classification for Country Comparisons. Final Report to the
World Intellectual Property Organization (WIPO).

Sheng, F. (2016). “An Emerging Theory of an Inclusive Green Economy.” Web log post: http://web.unep.org/
greeneconomy/ 26 Jan. 2016. Web: http://web.unep.org/greeneconomy/blogs/emerging-theory-inclusive-
green-economy

Stiglitz, J. E., Sen, A., and Fitoussi, J. P. (2010). “Report by the commission on the measurement
of economic performance and social progress”, Paris: Commission on the Measurement of Economic
Performance and Social Progress.

Swedish Environmental Protection Agency (SEPA). (2013). National Environmental Performance on

Planetary Boundaries. Report 6576.

United Kingdom Committee on Climate Change (UK CCC). (2008). Building a low-carbon economy – the
UK’s contribution to tackling climate change. Available at: http://archive.theccc.org.uk/archive/pdf/TSO-
ClimateChange.pdf

United Nations (2011). Working towards a Balanced and Inclusive Green Economy: A United Nations
System-wide Perspective. New York: United Nations.

United Nations General Assembly resolution A/RES/66/288. The Future We Want. New York: United
Nations.

United Nations Development Programme (UNDP). (2014). Human Development Report 2014. Sustaining
Human Progress Reducing Vulnerabilities and Building Resilience. Technical Notes. New York: UNDP.

United Nations Development Programme (UNDP). (2015). Human Development Data. Available at: http://
hdr.undp.org/en/data

UNEP. (2011). Towards a Green Economy: Pathways to Sustainable Development and Poverty Eradication.
Geneva: UN Environment.
20

UNEP. (2012). Measuring Progress towards an Inclusive Green Economy. Geneva: UN Environment.

UNEP. (2014). Using Indicators for Green Economy Policymaking. Geneva: UN Environment.

UNEP. (2015). Indicators for Green Economy Policymaking – A Synthesis Report of Studies in Ghana,
Mauritius and Uruguay. Geneva: UN Environment.

University of the United Nations / International Human Dimensions Programme on Global

Environmental Change, United Nations Environment Programme (UNU-IHDP/UN Environment).
(2014). Inclusive Wealth Report 2014. Measuring progress toward sustainability. Cambridge: Cambridge
University Press.

Villar, A. (2011). Who Meets the Standards: A Multidimensional Approach, Modern Economy, 2(4): 614-
624.

World Bank. World Development indicators. Available at: http://databank.worldbank.org/data/home.aspx

World Health Organization (WHO). (2014). WHO indoor air quality guidelines: household fuel combustion.
Geneva: WHO.
 ANNEX I – FORMALIZATION OF THE THEORETICAL FRAMEWORK
21

A. THE BASIC MODEL

ANNEX I – Formalization of the theoretical framework
zationANNEX
of the theoretical
I–
Consider a society
framework
Formalization
whose performanceof
is tothe
be theoretical framework
let p(q), p(s) denote a permutation applied to the
A. The basic model
EX I – Formalization of the theoretical framework
evaluated across several dimensions, with respect
A. The basic model
to a vector aof society targets or reference values that is have
elements of q and s. Then, F(q, s) = F(p(q), p(s)).
Consider whose performance to be evaluated across several dimensions, with
been previously set. that The second property,dimensions,
normalization, determines a
basic model
rmance is to berespect
evaluated to a vector
across of
Consider a society whose performance is to be evaluated targets
several or reference
dimensions, values
with haveacrossbeen several
previously set. with
eference valuesrespectthat to ahave vector been of previously
targets or reference set. scale
values that have been previously set. for the evaluation function. It says that the
r a society whose Let ‫ܭ‬performance
Let ൌ ሼͳǡ ʹǡ ǥ ǡ ݇ሽ is denote
to beaaset
denote set of of
evaluated dimensions,
dimensions, across several withvalue A realization
of the
݇dimensions,
൒ ʹ. function
withis zero for thiswhensociety
all outcomesis a
of adimensions,
to vector Let vector
with
ൌwith
of‫ܭ‬targets ሼͳǡ ʹǡ݇orǥ‫ܙ‬ ൌ ݇ሽ. AA
ሼ‫ݍ‬
൒reference
ǡ ʹ. ‫ݍ‬realization
ǡ ǥ
realization
denote
ଵǡ ǡ
ଶ values ‫ݍ‬ ሽ.
a ௞set The
for for entry
this
thishave
of
that society
dimensions, q is
society a real
j is apreviously
been vector
with number
is a ݇ ൒ set. ʹ.arethat
A zero describes
and the for
realization the
value value
is equal
this of
societyvariable
to one a Itall
is when j. is
y qj is a real number
vector ‫ܙ‬ ൌ ሼ‫ݍ‬
implicitlythat
ଵǡ ‫ݍ‬ଶ ǡdescribes
ǥ ǡ ‫ݍ‬௞ ሽ.. The
assumed The the value
entry
entry
that j of
is isa variable
allqjqdimensionsareal realnumber j. It is
number
can bethat describes
outcomes the
approximated value
exactly
quantitativelyof variable
match thebytargets.
realj. Itnumbers.
is
ሼͳǡ ʹǡ ǥ ǡ can
nsions denote
݇ሽimplicitly There
be that aassumed
set
is aofparameter
describes
approximated dimensions,
the
that all vector
value
quantitatively with of
of variable
dimensions by ݇reference
൒real Aimplicitly
ʹ.isnumbers.
j. can
It berealization
values ‫ ܛ‬ൌfor
approximated ሼ‫ݏ‬ଵǡ ‫ݏ‬thisǡ ǥ ǡsociety
‫ݏ‬௞ ሽ that isdescribes
ଶquantitatively aby real the targets fixed
numbers.
ൌ ሼ‫ݍ‬ଵǡ ‫ݍ‬ଶvalues
ference ǡ There
ǥ ǡ ‫ݍ‬௞ ሽ.‫ܛ‬isThe
for the
assumed
ൌaሼ‫ݏ‬ entry ǥ qǡ ‫ݏ‬j ௞is
‫ݏ‬different
ଶ ǡ that
parameter
ଵǡ all adimensions
that
ሽvectorreal
dimensions. number
describes
of canHow
reference that
be those
the describes
targets
approximated
values thresholds
‫ܛ‬fixed
ൌ ሼ‫ݏ‬the value
are setof Normalization:
variable
ଵǡ ‫ݏ‬ଶ ǡ ǥ ǡ ‫ݏ‬௞ ሽ that
will describes
depend j. It on
is the
each
Let specific
0 denote
targets theproblem
fixed vector whose
y those
assumedthresholds
for that (this question
aredimensions
the quantitatively
different
all set dimensions.
willby will
depend
real be
can discussed
on
How
numbers.
be each those
There
approximatedlater,
specific when
is a problem
thresholds parameteraddressing
are set will
quantitatively the construction
depend
components
by onareeach
real numbers. of
equal the
specific GEP
to zero.problemIndex).
Then, F(0, s) = 0, F(s,
ter, when(this
a parameter addressing
question
vector ofthe will construction
referencebe
vector of reference values discussed
values of the
‫ܛ‬later,
ൌ GEP
ሼ‫ݏ‬ when
‫ݏ‬
ଵǡ ଶ
Index).
ǡ ǥ addressing
ǡ ‫ݏ‬௞ ሽ that the
describes construction
the
s) = 1. targets of the
fixedGEP Index).
ifferent dimensions. An evaluation
that How those
describes problem,
the targetsorfixed
thresholds simply
are forset a problem,
the will depend
different canon beeachidentified
specific withproblem
a pair of vectors (q, s). In order
a problem,
estion willAn can
be betoidentified
evaluation
discussed evaluate
dimensions. problem, society’s
later,with
How when a
or pair
those outcomes
simply of vectors
addressing a problem,
thresholds theqare
(q, relative
s).can In
construction
set willbetoidentified
order a vector
of theThe of last
with
GEP targets
a pair
Index). s,
of avectors
property, continuous
additivity,(q, s). function F that
In order conditions
establishes
q relative to evaluate associates
a vector of targets
society’s to each
s, a problem
continuous (q, s)
function a real F number,
that F(q, s), which provides information on the
depend on eachoutcomes specific problem q relative (this to a vector of targets
question on thes,behaviour
a continuous of thefunction
evaluation F function
that when the
s) a real
ation number,
associates
problem, orfulfilment
F(q,
to
simply s),
each
aof the
which
problem, objectives,
problem provides
can (q,be s) must
information
a
identifiedreal be number,
identified.
with ona the This
F(q,
pair of s), function
which
vectors (q, iss).obtained
providesIn from an intuitive
information
order on the set of
will be discussed later, when addressing the vector of achievements changes from q to q’ = q + c
be identified.
ate fulfilment
society’s properties
This
outcomes function described
of theq objectives,is obtained
relative to below.
must from
a Index).
vectorbean ofintuitive
identified.
targets This set
s, aof function
continuous is obtained
function from
F thatan intuitive set of
construction of the GEP for some k-dimensional vector c (which may include
es to each properties A.1described
problem (q, s) a below.
Properties real number, F(q, s), which provides positive information on the
and negative numbers). The property
nt of the objectives, must be identified. This function is obtained from an intuitive set of
A.1 Properties
An evaluation problem, or simply a problem, can requires the change of the index to correspond to
es described below. The first property, neutrality, says that all objectives are equally important (a more
be identified with a pair of vectors (q, s). In order the evaluation of that change by general
the very same case
operties
that all objectives
The first is evaluate
to referred
are equally
property, to later
neutrality,
society’s on).says
important
outcomes This
(athat implies
more allgeneral
q relative that
to a rearranging
objectives caseare equally the important
evaluationordering of
(a weighted
function. more
This is a realizations
general and
case property
very natural
ies that rearranging
is referredtargets
vector the
to does
ordering
oflater
targets not affect
on).s,This the
aofcontinuous
weighted evaluation.
implies that realizations
function Formally:
F thatand the ordering
rearranging that is most of weighted
useful when realizations
the data are andcollected from
tion. Formally:
targets
property, neutrality, does not
says that
associates affect
to each the evaluation.
all objectives
problem (q, are Formally:
s) a equally
real number, important (aseveral more generalsources, case or across different time periods, or
ed to later on). F(q, Thiss),impliesNeutrality: that Let (q,
rearranging s) be
which provides information on the fulfilment a
the givenordering problem of and
weighted let p(q), p(s)
realizations denote
when there are mistakes anda permutation
to be corrected. applied Thetonew
given problem
does not affect and
 the let
ofNeutrality: the
p(q),
evaluation.
the objectives, elements
p(s) denote
LetFormally:
(q,
must of
s) be a q and
permutation
be aidentified. s.
given problemThen, F(q,
applied
This function s) =toF(p(q),
and let p(q), data p(s)).
p(s) denote a permutation
can be integrated by simply applied to
computing the
hen, F(q, s) = F(p(q), the p(s)).
elements of q and
is obtained from an intuitive set of propertiess. Then, F(q, s) = F(p(q), p(s)). value of that change and adding the result to the
The
Neutrality: Let (q,described s) besecond
a given
below. property,
problem normalization,
and let p(q), determines
p(s) denote a a scale
permutation for the
original value evaluation
applied
of thetoindex. function.
Formally: It says that
on, determines
he elements Theof athe
second value
scale
q and property, for ofthe
s. Then, F(q, the function
evaluation
normalization, is zero
function.
s) = F(p(q), p(s)). when It all
says outcomes
that are zero
determines a scale for the evaluation function. It says that and the value is equal to one when
when all outcomes
the value all outcomes
arethe
of zero and
function exactly match
theisvalue
zero isthe
when equal targets.
all to one when
outcomes are zero and Additivity:
the value is Letequal
(q, s)to beone when and let q’ =
a problem,
rgets.
ond property, all outcomes
normalization, exactlydetermines match the targets. a scale for the evaluation function. It says
A.1 Properties q + c. Then, F[(q +thatc),s] = F(q,s) + F(c,s)
 Normalization: Let 0 denote
e of the function is zero when all outcomes are zero and the value is equal to one when the vector whose components are equal to zero. Then, F(0,
ote
mes the vectormatch
exactly  whose s) = 0, F(s,
components
Normalization: Let s)
are0= denote
1.
equal tothe zero. vector Then, whose F(0, components are equal to zero. Then, F(0,
The thefirsttargets.
property, neutrality, says that all objectives The following result is obtained:
s) = 0, F(s, s) = 1.
are equally important (a more general case is
Normalization: TheLet following
The0 lastdenote result
property,
theon). isadditivity,
obtained:
The
vector following
whose establishes result conditions
components is obtained:
are equal ontothe zero.behaviour
Then, of the evaluation function
F(0,
referred
when the to vector
later of This implies
achievements that rearranging
changes from q to Theorem:
q’ = q + c Afor
continuous
some function F satisfies
k-dimensional vector c
blishes conditions
s) = 0, F(s,The s) =last on
1.theproperty, the behaviour of the evaluation function
additivity, establishes conditions on the behaviour of the evaluation function
q(whichordering may of weighted
+ cinclude realizations
positive and negativeand targets the properties of neutrality, normalization
changec of and
changeswhen from the
Theorem: tovector
q’ =A qof for some
achievements
continuous Theorem: k-dimensional
changes
function A F from vector
satisfies
continuous q numbers).
tocq’the
function + The
= q properties
c for F
property
some requires
ofk-dimensional
satisfies neutrality,
the
the normalization
vector
properties
the and
of neutrality, normal
negative(whichnumbers).does
index not
to
The affect
correspond
property the evaluation.
to
requiresthe Formally:
evaluation
the change of
may include positive and negative numbers). The property requires the change of the that
of thechange additivity,
by the veryif and
same only if:
evaluation function. This
additivity,establishes
property, additivity, if and onlyconditions
if:additivity,onif the
andbehaviour
only of the evaluation function
if: when
ation of that is correspond
change
index to a very
by the natural
very property
tosame
thefrom that is most
evaluation
evaluation useful
function. Thisby the the very
data are௤ collected from several This
sources, or
e vector of achievements changes q to q’of =q that
+ cchange
for ‫ܨ‬ሺࢗǡ
some࢙ሻk-dimensional
ൌ
ଵ
σ
same evaluation
ೕ vector c
function.
ଵ ௤ೕ
[A.1]
most useful when across

the
is a positive different
Neutrality:
data
very natural are time
Let
collected (q,periods,
s) be
from a or
given
severalwhen there
problem
sources, andare
or mistakes to be corrected.
‫ܨ‬ሺࢗǡ ࢙ሻ The
ൌ σ
new
௝‫א‬௄ ௦ orcan be
data
and property
negativethat is most useful when the data are௞the
collected
௦ೕ from several ௞sources,
௝‫א‬௄
may include numbers). The property requires change of the ೕ
when thereacross integrated
are mistakes
different to by
besimply
time corrected.
periods, computing
or The
when new the value
dataarecanofbethat change and adding the result to the original
correspond to the evaluation of that change bythere
the very mistakes
same to be corrected.
evaluation function. TheThisnew data can be
of thatvalue
the valueintegrated ofsimply
change theandindex.
addingFormally:
the result to the original
natural property thatbyisall
Moreover, most
these
computing
useful when the
properties
Moreover,
the value
aredata of that
are
independent.
all these
change and adding the result to the original
collected
properties fromareseveral sources, or
independent.
value of the index. Formally:
different time periods, or when there are mistakes to be corrected. The new data can be
 Additivity: Let (q, s) be a problem, and let q’ = q + c. Then, F[(q + c),s] = F(q,s) + F(c,s)
ed by simply computing the value of that change and adding the result to the original
This
roblem, and let theorem
 q’Additivity:
= q + c. Then,shows F[(q
Let (q, s) that
This
+be
c),s] assuming
atheorem
= F(q,s) +and
problem, the
shows
F(c,s) principles
let q’ that of neutrality,
= q + assuming
c. Then, F[(qthe
+ c),s] =normalization
principles
F(q,s) + of and additivity
neutrality,
F(c,s) normalization an
the index. Formally:
amounts to choosing amounts a precise to and very intuitive
choosing a precise 26
evaluation
and very intuitive evaluation function: a(q)
function: a vector of outcomes is of outc
vector
evaluated with respect to a vector
evaluated of targets
with respect to a(s)vector
as theof arithmetic
targets (s)mean of arithmetic
as the the relativemean of
 Theorem: A continuous function F satisfies the properties of weighted neutrality, norm
and additivity, if and only if:
ଵ ௤ೕ
‫ܨ‬ሺࢗǡ ࢙ሻ ൌ σ௝‫א‬௄
௞ ௦ೕ

Moreover, all those properties are independent.

The following result is obtained:
sult is obtained:The following result is obtained:
Proof
Moreover,
Theorem: all these properties are
A continuous independent.
function F satisfies the Forproperties
a given problem (q, s), let dj(a)
of neutrality, denote a vector
normalization and
(i) It is easy to see that function [A,1] satisfies all properties. Let us verify the rec
ntinuous 22 function additivity,
Theorem:F satisfies A ifcontinuous
andtheonly properties
if: function Fofsatisfies neutrality, whose
normalization
the properties elements are
and all
of neutrality, zero except entry
normalization and j that
௤ೕ
only if:
F satisfies This theorem
theadditivity,
properties if and shows
of only that
if:
neutrality, assuming the principles
normalization and ‫ܨ‬ሺࢗǡ ࢙ሻ ൌ ௞ σ௝‫א‬௄ ௦ is equal
ଵ to a. Applying repeatedly the property[A.1] of
The following result is obtained: For a given problem (q, s),ೕ let dj(a) denote a vector whose elements are all ze
of neutrality, normalizationଵand additivity ௤ೕ amounts ࢙ሻ ൌ additivity,
ଵ ௤
σ Applyingೕ we have: [A.1] we have:
d: ‫ܨ‬ሺࢗǡ ࢙ሻ ൌ ௞ σ௝‫א‬௄ ௦ entry j that‫ܨ‬ሺࢗǡ is equal to௞ a. ௝‫א‬௄ ௦ೕ [A.1] the property of additivity,
repeatedly
to choosing
ଵ
σ
‫ܨ‬ሺࢗǡ ࢙ሻ ൌ ௞Moreover,
௤ ೕ a precise and very intuitive
ೕ evaluation
[A.1]
Theorem: A continuous ௝‫א‬௄ ௦ function all these F properties
satisfies the are independent.
properties of neutrality,
‫ܨ‬൫ሺ‫ݍ‬ଵ ǡ Ͳǡ ǥ normalization
ǡ Ͳሻǡ ࢙ሻ ൌ ‫ܨ‬ሺࢊଵ ሺ‫ݍ‬ଵand ሻǡ ࢙ሻ ൅‫ܨ‬ሺ૙ǡ ࢙ሻ ൌ ‫ܨ‬ሺ݀ଵሺ‫ݍ‬ଵ ሻǡ ࢙ሻ
function: aೕvector of outcomes (q) is evaluated with
ction
tisfies F the
additivity, satisfies
if and onlythe
propertiesif: properties
Moreover,
respect to of
all these
a vector
of neutrality,
properties
neutrality,
of targets
normalization
are independent.
normalization
(s) as the arithmetic and and ‫ܨ‬൫ሺ‫ݍ‬ଵ ǡ ‫ݍ‬ଶ ǡ Ͳǡ ǥ ǡ Ͳሻǡ ࢙ሻ ൌ ‫ܨ‬ሺࢊଶሺ‫ݍ‬ଶ ሻǡ ࢙ሻ ൅‫ܨ‬ሺ݀ଵ ሺ‫ݍ‬ଵ ሻǡ ࢙ሻ
se properties are independent.
This theorem shows that assuming ଵ the ௤ೕ principles of neutrality, normalization and additivity
dependent. mean of the relative achievements ‫ܨ‬ሺࢗǡ (the
࢙ሻ ൌratiosσ௝‫א‬௄ [A.1]
…
ଵ Thisamounts
theorem
௤ೕ ଵ toshows choosing
௤ೕ thata precise
assuming and௞the ௦ೕ
veryprinciples
intuitive evaluation
of neutrality, function: a vector of
normalization outcomes
and additivity (q) is
hows ‫ܨ‬ሺࢗǡ ‫ܨ‬ሺࢗǡ
σ
࢙ሻ ൌassuming
that ࢙ሻ
between ൌ
evaluated outcomes
the σ and
principles
௝‫א‬௄with targets).
௦ೕ respect of neutrality,
to anda vector [A.1]
normalization
of targets
[A.1] and ‫ܨ‬ሺࢗǡ ࢙ሻ ൌ ෍ ‫ ܨ‬൫ࢊ௝ ൫‫ݍ‬௝ ൯ǡ ࢙ሻ
additivity
(s) asfunction:
the arithmetic mean of the(q)relative
ng the principles ௞ amounts
௝‫א‬௄
of to௞ choosing
௦neutrality,
ೕ a precise
normalization and very intuitive
additivity evaluation a vector௝‫א‬௄ of outcomes is
Moreover,
sing a all
precise theseandproperties
very
achievements
evaluatedfunction:
very intuitive evaluation are
intuitive
with respect independent.
evaluation
(the ratios
a vector function:
between
to ofa outcomes outcomes
vector of(q) a vector
targets
is and of outcomes
targets). (q) is
(s) as the arithmetic mean of the relative
respect to a The case of
vector in whichtargets different
(s) objectives
as the have
arithmetic mean of a the
tor
are achievements (the ratios
of targets (s) as the arithmetic mean of the relative
independent. between outcomes Let and
now targets).
1a denote vectorrelative
with all entries equal to a. In this particular case, we
dent.
This theorem different
showsThe thatimportance
case assuming
in can
which be
the readily
principles
different handled.of
objectives neutrality,
have Let now
normalization
different 1a denote
importance andacan vector
be with
additivity all entries
readily handled.equal Letto
he ratios between
outcomes and targets). outcomes and targets).
amounts to choosing a precise
Let case ଵin which
The ߨ ൌ ሺߨ ǡ ǥ ǡ ߨ and very
denote
௞ denote
ሻ different intuitive
a vector
a vector evaluation
of
of relative
objectives function:
relative
have different a
weights
a. Invector
that of outcomes
describe
this particular
importance cancase, the (q) is
importance
we have:
be readily of
handled. Let each
suming the principles of neutrality, normalization and additivity 27
eevaluated
principles
tivesdifferent withof
have different ߨ neutrality,
objective,
respect
weights
ൌ ሺߨ ǡ to
that
ǥ
importanceǡ ߨ aߨnormalization
describe
ሻ vector
൒ Ͳ,‫݆׊‬,
denote the of
σ
a
can be readily ߨ
importance
vector and
targetsൌ ͳ.
of additivity
Then
(s)
of as
each by
relative
handled.can the making
weights the
arithmetic following
that mean
Let be readily handled. Letdescribe change
of the
the of variable:
relative
importance ‫ݕ‬௝ of
ൌ ݇ߨ ௝ ‫ݍ‬௝ , the
each
ch objectives ଵ have௞ ௝
different ௝‫א‬௄
importance
௝
eintuitive
and very
achievements intuitive
evaluation
f relative weights (the evaluation
function:
ratios
following
objective, between
that ߨdescribe
objective, a
௝ ൒extension
function:
vector
outcomes
Ͳ,‫݆׊‬, σ ௝‫א‬௄of
the ߨ of a
௝the
vector
outcomes
and
former
ൌ ͳ.
importance
of
targets).
Thenresult
. Then outcomes
(q)
by making
by
of each is
is obtained:
making (q) is
the following change ‫ܨ‬ሺ૚ܽǡ ૚‫ݐ‬ሻ of ൌvariable:
෍ ‫ݕܨ‬൫ࢊ ௝ ൌ௝ ሺܽሻǡ ௝ , the
૚‫ݐ‬ሻ
݇ߨ௝ ‫ݍ‬
enote a vector of relative weights that describe the importance of each
vector
targets
en by of(s)targets
making as
the the(s)
following
thefollowing
followingas the of
arithmetic
extension
change
change arithmetic
ofmean
the former
variable:
variable: mean
of‫ݕ‬௝the
result ‫ݍ‬௝of
ൌ ݇ߨrelative
௝ is thethe relative
,obtained:
the
௝‫א‬௄
‫ܨ‬ሺ૚ܽǡ ૚‫ݐ‬ሻ ൌ ෍ ‫ ܨ‬൫ࢊ௝ ሺܽሻǡ ૚‫ݐ‬ሻ
‫݆׊‬,
The
σ case ߨ ൌ ͳ.
௝‫א‬௄ ௝in which Then by making the following change
haveMoreover, of variable:
neutrality ‫ݕ‬
implies: ൌ ݇ߨ ‫ݍ‬ , the
‫ݍ‬௝ ௝ ௝ handled. ௝‫א‬௄
een and
mes
esult outcomes
targets).
is obtained: anddifferent
targets).
following objectives
extension of the former different
result importance
is obtained: can be
௝ readily
Moreover, Let
‫ܨ‬ሺ૚ܽǡ ‫ܨ‬ሺࢗǡൌ‫ܨ‬ሺ૚ܽǡ
૚‫ݐ‬ሻ ࢙Ǣ෍
ߨሻ ൌ૚‫ݐ‬ሻ
෍‫ܨ‬ ൌ
൫ࢊ ෍ ௝ neutrality
ߨሺܽሻǡ ‫ܨ‬
૚‫ݐ‬ሻ ൫ࢊ
‫ܨ‬ሺ૚ܽǡ implies:
ሺܽሻǡ
૚‫ݐ‬ሻ ෍ ‫ ܨ‬൫ࢊ௝ ሺܽሻǡ ૚‫ݐ‬ሻ
ൌ૚‫ݐ‬ሻ
on of the former
ଵ
result is
௞
obtained:
ߨ ൌ ሺߨ ǡ ǥ ǡ ߨ ሻ denote a vector of relative Moreover, weights neutrality implies:
that describe the
௝‫א‬௄ ‫ݏ‬
‫ ݍ‬importance
௝ ௝ ‫ܨ‬൫ࢊ ሺܽሻǡ of
௝ ૚‫ݐ‬ሻ each
‫ܨ‬ሺࢊ ሺܽሻǡ ૚‫ݐ‬ሻ
ൌ௝‫א‬௄ ௝ ௝ ௛
‫ܨ‬ሺࢗǡ ࢙Ǣ ߨሻ ൌ ෍ ௝‫א‬௄ ߨ ௝‫א‬௄
objective,
objectives ߨ ൒ Ͳ,‫݆׊‬,
௝have different σ ‫ݍ‬ ߨ ൌ
௝ ௝ importanceͳ. Then by making
can implies: the following
Moreover,
be readilyLet change
neutrality
handled. implies:
௝‫א‬௄Let ௝ of௝
‫ݏ‬variable: ‫ݕ‬ ൌ
‫ܨ‬൫ࢊ௝௝ ሺܽሻǡ ૚‫ݐ‬ሻ ݇ߨ ‫ݍ‬ , the
௝ ௝ൌ ‫ܨ‬ሺࢊ௛ ሺܽሻǡ ૚‫ݐ‬ሻ
have
‫ܨ‬ሺࢗǡ different
࢙Ǣ ߨሻ ൌ ෍importance
Moreover, ߨ௝ Moreover,
௝‫א‬௄
neutrality can implies: be readily
neutrality ‫ݍ‬௝ handled. ‫ܨ‬ሺ૚ܽǡ૚‫ݐ‬ሻ
‫ܨ‬ሺ૚ܽǡ ૚‫ݐ‬ሻൌൌ෍ ෍ ‫ܨܨ‬any ൫ࢊ൫ࢊ௝௝ሺܽሻǡሺܽሻǡ૚‫ݐ‬ሻ ૚‫ݐ‬ሻ
following extension Now
of the
‫ݏ‬ the formerproperty result of is neutrality
obtained: is to be understood as “weighted ‫ܨ‬൫ࢊ ሺܽሻǡ neutrality”,
૚‫ݐ‬ሻ ൌ ‫ܨ‬ሺࢊ௛saying ሺܽሻǡ ૚‫ݐ‬ሻ that
tor of relative
ative weights thatpermutation weights
௝‫א‬௄ ‫ܨ‬ሺࢗǡ that
࢙Ǣ
describe ofthe
௝ ߨሻ ൌ describe
෍ ߨ
importance the
௝ ‫ܨ‬ሺ૚ܽǡof
Therefore, importance each of each ሺܽሻǡ
௝ ௝‫א‬௄
௝‫א‬௄
Now the property the variables
of neutrality
௝‫א‬௄ ‫ݏ‬௝ ‫ݕ‬isଵ ǡ to ‫ܨ‬൫ࢊ
ǥ ǡ૚‫ݐ‬ሻ
‫ݕ‬
be
ൌ෍
௞௝ ሺܽሻǡ
leaves ૚‫ݐ‬ሻ
understood ‫ܨ‬൫ࢊ
the ൌ‫ܨ‬௝൫ࢊ ሺܽሻǡ
‫ܨ‬ሺࢊ
index ௝as ௛૚‫ݐ‬ሻሺܽሻǡ૚‫ݐ‬ሻ ൌ૚‫ݐ‬ሻ
unchanged.
“weighted ‫ܨ‬ሺ૚ܽǡ
ሺܽሻǡ ෍ saying
‫ܨ‬ሺࢊ௛neutrality”,
૚‫ݐ‬ሻ ൌ૚‫ݐ‬ሻ ‫ ܨ‬൫ࢊ௝ ሺܽሻǡ that ૚‫ݐ‬ሻany
ͳ. Then by
making themaking
following thechange
following change
of variable: of‫ݕ‬௝variable:
ൌ ݇ߨ௝ ‫ݍ‬௝ ,‫ݍݕ‬the
Therefore, Moreover,
௝Moreover,
ൌ௝‫א‬௄ ݇ߨ௝ ‫ݍ‬௝Therefore, neutrality
,neutrality
the implies:
implies:  ‫ܨ‬൫૚‫ݍ‬௝ ǡ ૚‫ݏ‬௝ ൯
to be understood permutation
as “weighted of theneutrality”,
variables ǡൌǥ ǡ෍
‫ݕ‬ଵsaying ௞ leaves
‫ݕ‬Therefore,
that ߨ௝ any
௝the index unchanged.
‫ܨ‬ሺ૚ܽǡ ૚‫ݐ‬ሻ ൌ ݇‫ܨ‬൫ࢊ௝ ሺܽሻǡ ૚‫ݐ‬ሻ ֜ ‫ܨ‬൫ࢊ
௝‫א‬௄
‫ܨ‬൫ࢊ ൫‫ݍ‬ ‫ܨ‬ሺ૚ܽǡ ൯ൌ૚‫ݐ‬ሻ ൌ෍ ‫ ܨ‬൫ࢊ௝ ሺܽሻ
mer Moreover,
result is obtained: neutrality implies:
‫ܨ‬ሺࢗǡ ࢙Ǣ ߨሻ ‫ܨ‬൫ࢊ ሺܽሻǡ
௝௝ሺܽሻǡ ௝ ൯ǡ૚‫ݐ‬ሻ૚‫ݏ‬௝ൌ
૚‫ݐ‬ሻ ൌ ‫ܨ‬ሺࢊ ሺܽሻǡ ૚‫ݐ‬ሻ
s obtained: Moreover, neutrality
‫ݏ‬ implies: ‫ܨ‬ሺ૚ܽǡ ૚‫ݐ‬ሻ ൌ ෍ ‫ܨ‬ ௝ ൫ࢊ ሺܽሻǡ ૚‫ݐ‬ሻ ‫ܨ‬ሺࢊ ‫ܨ‬൫૚‫ݍ‬ ௛௛ሺܽሻǡ ݇ ǡ ૚‫ݐ‬ሻ
૚‫ݏ‬
௝‫א‬௄ ൯
ty௞ leaves
‫ݕ‬ of neutrality indexisunchanged.
theTherefore, A.2to Proofbe understood
of the characterization as “weighted resultneutrality”,
௝‫א‬௄ ௝ saying that௝ ሺܽሻǡ any૚‫ݐ‬ሻ ֜ ௝‫א‬௄  ௝ ௝ ௝
Now the Therefore,property of neutrality is to be ‫ܨ‬൫ࢊ ௝ ሺܽሻǡ ૚‫ݐ‬ሻ ൌas
understood ‫ܨ‬ሺࢊ ‫ܨ‬ሺ૚ܽǡ௛ ሺܽሻǡ
૚‫ݐ‬ሻ ૚‫ݐ‬ሻ ൌ ݇‫ܨ‬൫ࢊ ‫ܨ‬൫ࢊ௝ ൫‫ݍ‬௝ ൯ǡ ૚‫ݏ‬௝ ൯ ൌ‫ܨ‬൫૚‫ݍ‬௝ ǡ ૚‫ݏ‬௝ ൯
Moreover,
‫ܨ‬ሺ૚ܽǡ ૚‫ݐ‬ሻ ൌneutrality
݇‫ܨ‬൫ࢊ ሺܽሻǡ implies:
૚‫ݐ‬ሻ
‫ܨ‬൫ࢊ௝௝ሺܽሻǡ ૚‫ݐ‬ሻ ൌ ‫ܨ‬ሺࢊ௝ ௛ ሺܽሻǡ ֜ ‫ܨ‬൫ࢊ ൫‫ݍ‬ ൯ǡ ૚‫ݏ‬ ൯ ൌ ݇
he variables ‫ݕ‬ଵA.2 ‫ݕ‬௞ leaves
ǡ ǥ ‫ݍ‬ǡ Proof of the‫ݍ‬ the index unchanged.
characterization result neutrality implies:
Moreover,
௝ ૚‫ݐ‬ሻ
௝
݇
Now the property Theorem:
“weighted
of
௝ neutrality A
neutrality”, ௝continuous
is saying
to be function
that
understoodany F
permutation satisfies
as
Substituting “weighted of
Therefore,
Therefore, the
 in the properties
neutrality”,
equation  of saying
above, ‫ܨ‬൫૚‫ݍ‬
weighted wethat ǡ ૚‫ݏ‬
neutrality,
get: any ൯‫ܨ‬൫૚‫ݍ‬ ǡ ૚‫ݏ‬
‫ܨ‬൫ࢊ௝ ൯
normalization ሺܽሻǡ ૚‫ݐ‬ሻ ൌ ‫ܨ‬ሺࢊ௛ ሺܽሻǡ ૚
࢙Ǣ ߨሻ ൌ‫ܨ‬ሺࢗǡ
sult ෍ ࢙Ǣ ߨሻ ߨ௝ ൌ ෍ ߨ௝ ‫ܨ ݏ‬ሺ૚ܽǡ ૚‫ݐ‬ሻ ൌ‫ܨ‬ሺ૚ܽǡ ݇‫ܨ‬൫ࢊ ૚‫ݐ‬ሻ ሺܽሻǡ ൌ૚‫ݐ‬ሻ ݇‫ܨ‬൫ࢊ ֜ ௝ ሺܽሻǡ௝૚‫ݐ‬ሻ
‫ܨ‬൫ࢊ ൫‫ݍ‬ ֜૚‫ݏ‬
൯ǡ ‫ܨ‬൫ࢊ ൯
‫ܨ‬൫ࢊൌ
௝ ൫‫ݍ‬ ൯ǡ ૚‫ݏ‬
ሺܽሻǡ
௝ ૚‫ݐ‬ሻ
௝
௝ ൯ൌൌ‫ܨ‬ሺࢊ
௝
ሺܽሻǡ
௝
૚‫ݐ‬ሻ
௝
permutation௝‫א‬௄ ofTherefore,
the ‫ݏ‬
variables
the and
Theorem:variables
௝ additivity,
௝‫א‬௄ ‫ݕ‬ ଵ ǡ
A continuous ǥ
௝ ǡ ‫ݕ‬
if and leaves
௞ leavesonly the
the index
function
if: index
F satisfies unchanged.
௝
unchanged.
Substituting the properties
in the equation
௝
of weighted௝
above, ௝
‫ܨ‬ሺࢗǡ neutrality,
we ࢙݇ሻ get: ൌ ௞ σnormalization
ଵ ௛ ݇   ௝ ǡ ૚‫ݏ‬௝ ൯ ‫ܨ‬൫૚
‫ܨ‬൫૚‫ݍ‬
Therefore, Substituting in the equation above, we get: ௝‫א‬௄ ‫ ܨ‬൫૚‫ݍ‬
characterization
satisfies the properties result of weighted neutrality, normalization ‫ܨ‬ሺ૚ܽǡ
‫ܨ‬ሺ૚ܽǡ ௤ ૚‫ݐ‬ሻ
૚‫ݐ‬ሻ ൌൌ ݇‫ܨ‬൫ࢊ
݇‫ܨ‬൫ࢊ ሺܽሻǡ
ሺܽሻǡ ૚‫ݐ‬ሻ
૚‫ݐ‬ሻ ֜ ֜ ‫ܨ‬൫ࢊ
‫ܨ‬൫ࢊ ൫‫ݍ‬൫‫ݍ‬ ൯ǡ
൯ǡ ૚‫ݏ‬
૚‫ݏ‬ ൯൯ൌ ൌ
and additivity, if and only if: ‫ܨ‬ሺࢗǡ ࢙ሻ ൌ σ‫ܨ‬൫૚‫ݍ‬
ଵ ೕ
௝ ǡ ૚‫ݏ‬
‫ܨ‬ሺࢗǡ ௝ ൯ ࢙ሻ ൌ ଵσ
ଵ ௝ ௝
σ௝‫א‬௄ ‫ܨܨ‬൫૚‫ݍ‬ ൫૚‫ݍ‬௝ ǡ௝૚‫ݏ‬ ǡ ૚‫ݏ‬
௝௝
൯[A.2]
௝௝ ௝௝
Therefore, ௞ ௝‫א‬௄ ௦ೕ ‫ܨ‬ሺࢗǡ ࢙ሻ
 ൌ ௞௞ ௝‫א‬௄ ௝൯
௝‫ܨ‬൫૚‫ݍ‬
ty
A.2 isProof
to be
understood understood
of the as “weighted
characterization as “weighted ‫ܨ‬ሺ૚ܽǡ ૚‫ݐ‬ሻneutrality”,
neutrality”,
result ൌTherefore,
saying݇‫ܨ‬൫ࢊ௝that ሺܽሻǡsaying ૚‫ݐ‬ሻ any ֜ ‫ܨ‬൫ࢊ that any
௝ ൫‫ݍ‬Substituting
௝ ൯ǡ ૚‫ݏ‬ ଵ ௝ ൯ ൌ ௤ೕ in the equation above, we૚‫ݏ‬get: ௝ ǡ ૚‫ݏ‬௝ ൯
tinuous function F Substituting
satisfies Substituting
the in the
properties equation in ofthe equation
above,
weighted we
Now observe that our
above,
get:
neutrality,
‫ܨ‬ሺࢗǡ ‫ܨ‬ሺ૚ܽǡ࢙ሻ ൌ we૚‫ݐ‬ሻ σ get:
ൌ ݇‫ܨ‬൫ࢊ
normalization ௝ ݇
ሺܽሻǡ ૚‫ݐ‬ሻ ֜ ‫ܨ‬൫ࢊ ௝ ௝൫‫ݍ‬ ൯ǡ ൯
௝ [A.2] ൌ
A.2 Moreover,
Proof ௤all
of those properties are independent. ௞ assumptions
௝‫א‬௄ imply that function F is homogeneous ݇ o
ଵ ǡ ǥ the
ves leavesunchanged.
ǡ ‫ݕ‬௞ index the index ೕ the characterization result
unchanged. ௦ೕ
Substituting
Substituting in
in
ଵ thethe equation
equation ‫ܨ‬ሺ૚ܽǡ  above, ૚‫ݐ‬ሻ ൌ we
above, ݇‫ܨ‬൫ࢊ
we ௝ ሺܽሻǡ‫ܨ‬൫૚‫ݍ‬
get:
get: ૚‫ݐ‬ሻ ֜ ௝ ǡ ૚‫ݏ‬‫ܨ‬൫ࢊ௝ ൯௝ ൫‫ݍ‬௝ ൯ǡ ૚
ଵ
and
Theorem: ‫ܨ‬ሺࢗǡ
only if:A continuous ࢙ሻ ൌ σ [A.2]
Now
‫ܨ‬ሺࢗǡ observe
࢙ሻ ൌ
ଵ
that
‫ܨ‬ሺࢗǡ
σ our
࢙ሻ ‫ܨ‬ ൌ
൫૚‫ݍ‬ σ
assumptions
ǡ ૚‫ݏ‬ ൯ ‫ܨ‬ imply
൫૚‫ݍ‬ ǡ ૚‫ݏ‬that ൯ function F is [A.3]
homogeneous [A.3
of d
Moreover, ௞ function
௝‫א‬௄
all those Fproperties
satisfies are thethat
q, properties
Now
is,
independent. ሺߣࢗǡobserve ࢙ሻofൌweighted
ߣ‫ܨ‬ሺࢗǡ that
‫ܨ‬ሺ૚ܽǡ ࢙ሻour , neutrality,
૚‫ݐ‬ሻ forassumptions
ൌallߣ
݇‫ܨ‬൫ࢊ ൐ normalization
Ͳ . Define
ሺܽሻǡ ૚‫ݐ‬ሻimply ֜௝a‫ܨ‬൫ࢊ that
new ௝௝ ൫‫ݍ‬ function
function
௝ ൯ǡ ૚‫ݏ‬௝ ൯ ൌ Ff ଵis: »homogeneous
݇» +  » aso
Substituting
௦ೕ
in the equation above, we get:࢙ሻ ൌ ߣ‫ܨ‬ሺࢗǡ ௞ ௝‫א‬௄ ௞ ௝ ௝௝‫א‬௄௝ ଵ
‫ܨ‬ሺࢗǡ
‫ܨ‬ሺࢗǡ ࢙ሻ࢙ሻ ൌ ൌ σ σ »‫ܨ‬ ‫ܨ‬൫૚‫ݍ‬൫૚‫ݍ‬ »௝௝ǡ ǡ૚‫ݏ‬
૚‫ ݏ‬as ൯
and
on
additivity, if Theorem:
dependent.
result
andProof only if:A continuous ‫ܨ‬ሺࢗǡfunction࢙ሻ ൌ q,σFthat
ଵ q, that
is, ሺߣࢗǡ
௤
satisfies ೕ is,
Substituting the
ሺߣࢗǡ
࢙ሻ ൌ ߣ‫ܨ‬ሺࢗǡ ଵ
in the ࢙ሻ
equation ,
࢙ሻ , for allߣ ൐݂൫‫ݍ‬ for allߣ ൐
above, Ͳ
Ͳ . Define
[A.2] ௝
. Define
we get:
ǡ ‫ݏ‬௝ ൯ ‫׷‬ൌ a‫ܨ‬൫૚‫ݍ‬
a new
new ǡfunction
௝
function
૚‫ݏ‬௝ ൯
f : »
f௞௞: »  ௝‫א‬௄
௝‫א‬௄ »+ 
+ »as௝௝൯fol
௞ ௝‫א‬௄ ௦ೕ ‫ܨ‬ሺࢗǡ ଵ ࢙ሻ ൌ௤௞ೕ σ௝‫א‬௄ ‫ ܨ‬൫૚‫ݍ‬௝Substituting ǡ ૚‫ݏ‬௝ ൯ ݂൫‫ ݍ‬ǡ ‫ݏ‬in൯ the ‫׷‬ൌ equation
‫ܨ‬൫૚‫ݍ‬ଵ ǡ ૚‫ܨݏ‬ [A.3] we get:
above,
൯൫૚‫ݍ‬
properties
Proof Nowofobserve (i)weighted
It isNow easy neutrality,
that to our
observesee that normalization
‫ܨ‬ሺࢗǡ
that function
assumptions ࢙ሻ
our
Substituting σand
[A,1]
ൌ assumptions imply
௝‫א‬௄ satisfies
in theNow
that allobserve
imply
equation
function properties.
that that
above,
F݂൫‫ݍ‬ our
function
is ௝ ǡ௝ ௝assumptions
Let
‫ܨ‬ሺࢗǡ we
‫ݏ‬ ൯ us
࢙ሻ
homogeneous
௝ get:
‫׷‬ൌ verify
ൌ
[A.2]F
‫ܨ‬൫૚‫ݍ‬ is σhomogeneous
௝the
௝‫א‬௄
௝ ǡ ૚‫ݏ‬ imply
௝reciprocal.
௝ ൯
of ௝that
ǡ ૚‫ݏ‬௝ ൯ 1
degree ofindegree 1
௞
se properties
ion Fthe satisfies are independent.
the properties
additivity, ifItand only ofto if:weighted neutrality,
௞
normalization
௦ೕ
Now
Now function observe
observe that
that
F is homogeneous our
our assumptions
assumptions ofଵ degree imply
imply
1 in q,that that function
thatfunction
‫ܨ‬ሺࢗǡ
ଵ
is, ൌ ௞ σF௝‫א‬௄
࢙ሻ F isis‫ܨ‬hoh൫
fies
Moreover, allq,
propertiesthatpropertiesof
is,Now
(i)weighted
q, that is easy
is, neutrality,
ሺߣࢗǡ ࢙ሻsee that
,
ൌ for
ߣ‫ܨ‬ሺࢗǡ normalization
function
allߣ ࢙ሻ As , [A,1]
forthis
.
satisfies
allߣ
Define function
൐ aͲ .all
new Define properties.
inheritsfunction a the
new Let us
properties
function verify
f
the
of : »
reciprocal.
homogeneity
 as
» follows:
 » and as normalizatio
follows:
those ሺߣࢗǡ ࢙ሻ are
observe ൌ ߣ‫ܨ‬ሺࢗǡ
independent.
that ࢙ሻ
our assumptions ൐ Ͳ imply
As q,thisthat thatfunctionfunction
is,ሺߣࢗǡ ൌൌFߣ‫ܨ‬ሺࢗǡ
inherits is the f :
homogeneous» ‫ܨ‬ሺࢗǡ
࢙ሻproperties

, ,elements
forall
»
࢙ሻ
allߣ
ൌ  ௞of
of
σ »..degree
homogeneity
௝‫א‬௄
.Define
‫ܨ‬
Define
൫૚‫ݍ‬
+ a1a ௝ ǡ
in
૚‫ݏ‬ ൯
andfunction
new
௝ normalization,
function
tion [A,1] satisfies all properties. For a Let given us problem
verify ‫ݍ‬the reciprocal.
(q, s),ߣAslet
q,
‫ߣݍ‬dobserve
j(a)
thatdenote is, ሺߣࢗǡvector
thatainherits ࢙ሻ࢙ሻ ߣ‫ܨ‬ሺࢗǡ
whose ࢙ሻ for
for allߣ + ൐ ൐ ͲͲ
are Define
all zero anew new
Fexcept ff ::»» o
௝ ൌ ‫ݍݏ‬௝ ǡ Now ൌ
ൌ‫ݏ‬
this
ǡ ௝Τ ൌ
‫ݏ‬௝‫ݍ‬function
, we
Τ ‫ݏ‬ ,
have:
we have:
our assumptionsthe properties imply that
of function
homogeneity isand homogeneousnormalizatio
q, thatentry is, For
ሺߣࢗǡ ࢙ሻ ൌ ߣ‫ܨ‬ሺࢗǡ ࢙ሻ , for allߣ ൐ ௝ Ͳ݂൫‫ݍ‬ . Define
௝ ௝ ǡ ‫ݏ‬௝ ൯௝a‫׷‬ൌ ௝݂൫‫ݍ‬
new ‫ܨ‬൫૚‫ݍ‬ ǡ
function
function‫ݏ‬ ൯
ǡ ૚‫ݏ‬‫׷‬ൌ ൯ ‫ܨ‬൫૚‫ݍ‬
f : »  ǡ ૚‫ݏ‬
» ൯ » as
as follows:
follows:
j௤thata givenଵis equal problem (q,
௤ೕto a. Applying
q, s),
‫ݍ‬௝ thatൌ ‫ݏ‬let dൌ j(a)
ߣrepeatedly denote the aproperty
vector
࢙ሻ , forwhose
௝ ௝௝ Now
௝of observe
‫ݏ‬elements
௝
additivity, +௝ we that are
have:‫ݏݏ‬௝our all
݂൫‫ݍ‬
݂൫‫ݍ‬ assumptions
௝zero
௝ǡ ǡ‫ݏݏ‬௝௝൯൯‫׷‬ൌexcept
‫׷‬ൌ‫ܨ‬൫૚‫ݍ‬ ‫ܨ‬൫૚‫ݍ‬ : »௝imply
௝ǡ ǡ૚‫ݏ‬
૚‫ݏ‬ »௝௝൯൯+that funct
௝ ǡis, ሺߣࢗǡ , we have: allߣ . Define a new function f is » as
ଵ ‫ݍ‬௝ ࢙ሻ ‫ݏ‬௝ൌ ߣ‫ܨ‬ሺࢗǡ ൐
Proof ‫ܨ‬ሺࢗǡ ࢙ሻ ൌ ‫ܨ‬ሺࢗǡ σ௝‫א‬௄࢙ሻೕ ൌ ௞ σ௝‫א‬௄ ௦ Now [A.2] Τ
observe that[A.2] our assumptions ௝Ͳ‫ݍݏ‬ imply that function F homogeneous o
sy
let to dj(a)see that entry
denote function
a ௞vector j that[A,1]
whose
௦ೕis equal satisfies
elements all
to a.ೕ Applying properties.
are all zero݂൫‫ݍ‬
repeatedly
‫ܨ‬൫ሺ‫ݍ‬ଵ ǡ Ͳǡ ǥ ǡ Ͳሻǡ ࢙ሻ Let
except
ǡ ‫ݏ‬ ൯ ‫׷‬ൌus
the
௝ ௝ൌ ‫ܨ‬ሺࢊଵ ሺ‫ݍ‬ଵ ሻǡ verify
‫ܨ‬൫૚‫ݍ‬
property ǡq,૚‫ݏ‬ the that൯
of
௝ ࢙ሻ ௝൅‫ܨ‬ሺ૙ǡ ࢙ሻ ൌ‫ݍ‬ reciprocal.
is,
additivity, ݂
ሺߣࢗǡ ቆ
݂ ቆ
࢙ሻ
‫ܨ‬ሺ݀
௝
we
ൌ௝‫ݍ‬ ǡ ‫ݏ‬ǡ ቇ
have:
‫ݏ‬
ߣ‫ܨ‬ሺࢗǡ ቇ
௝ሺ‫ݍ‬௝ଵ ሻǡ ࢙ሻ
௝ ൌ ൌ ࢙ሻ
௝ ݂൫‫ݍ‬
,
݂൫‫ݍ‬for ௝ ǡ
ǡ ‫ݏ‬
allߣ
‫ݏ‬ ௝൯൯ ൌൌ ൐ ͳ ͳ Ͳ . Define a new func
(i) It is easy to see that function [A,1] satisfies ‫ݏ‬ ‫ݍ‬
‫ݏ‬ ௝ ௝
are independent. q, that is,all ሺߣࢗǡ properties.
࢙ሻ ൌ ߣ‫ܨ‬ሺࢗǡ Let ࢙ሻus , for verifyallߣ ݂
the݂൫‫ݍ‬ ‫ݍ‬reciprocal.
௝௝Ͳ ଵǡ ‫ݏ‬௝ ൯ ‫׷‬ൌ a
.௝ǡDefine
ቆ൐homogeneity
‫ݍ‬௝inherits‫ݏ‬௝࢙ሻ ቇൌ
‫ܨ‬൫૚‫ݍ‬
‫ݍ‬௝௝௝new௝ ǡfunction
݂൫‫ݍ‬
૚‫ݏ‬௝ ൯ f : »  » +  » as
dent.
epeatedly the propertyAs this function
of additivity, Aswe‫ܨ‬൫ሺ‫ݍ‬ this
have: function
inherits the inheritsproperties the properties
Asof
As homogeneity
this
this of
function
function ௝and
inherits normalization,
the
the and ௝‫ݍ‬ǡ ‫ݏ‬௝ ൯normalization,
properties
properties
ൌ ͳ by
ofof letting
homogeneity by lettin an
௝ ǡ ‫ݏ‬homogeneity an
ଵ ǡ Ͳǡ ǥ ǡ Ͳሻǡ
‫ܨ‬൫ሺ‫ݍ‬ ǡ
ଵ ଶ ‫ݍ‬ ࢙ሻ ǡ Ͳǡൌ ǥ ‫ܨ‬ሺࢊǡ Ͳሻǡ ଵ ሺ‫ݍ‬
࢙ሻ ଵൌሻǡ ࢙ሻ
‫ܨ‬ሺࢊ ൅‫ܨ‬ሺ૙ǡ
ଶ ଶ ሺ‫ݍ‬ ࢙ሻ
ሻǡ ࢙ሻ ൌ ‫ܨ‬ሺ݀
൅‫ܨ‬ሺ݀ ‫ݍ‬ ଵ ሺ‫ݍ‬ ሺ‫ݍ‬ ሻǡሻǡ ࢙ሻ
௝ଵ ଵ֜ ݂൫‫ ݍ‬ǡ ‫ݏ‬௝ ൯ ൌ ௝
ଵ ‫ݍ‬ ‫ݍ‬ ௝ ݂൫‫ݍ‬ ௝ ൯ ‫׷‬ൌ ‫ܨ‬൫૚‫ݍ‬௝ ǡ ૚‫ݏ‬ ௝൯
݂൫‫ݍ‬ ֜ ǡ ݂൫‫ݍ‬
‫ݏ‬ ൯ ‫׷‬ൌ ௝ ǡ
௝ ‫ݏ‬‫ܨ‬൫૚‫ݍ‬
௝௝൯ ൌ ǡ ૚‫ݏ‬ ൯
࢙ሻ ൌ
en ሺ‫ݍ‬
‫ܨ‬ሺࢊଵ ଵ‫(ݍ‬q,
problem ሻǡ
௝ ࢙ሻ
ൌ ൅‫ܨ‬ሺ૙ǡ
s), As
ߣ ൌ࢙ሻ
‫ݏ‬௝ problem
ǡ let ‫ݍ‬ this
ൌ௝ൌ
d௝‫ݍ‬jall
(a) function
௝௝,ǡ we
‫ݏݏ‬denote
Τthose
‫ܨ‬ሺ݀ ߣଵ ൌଵhave:
ሺ‫ݍ‬ inherits
‫ݏ‬௝ , we
‫࢙ ݍ‬ሻΤvector
ሻǡ the
ǡ ‫ݍ‬ଶhave: properties of homogeneity ௝ΤΤ‫ݏ‬ ,we
weଵhave:
௝,function and
have: normalization,
௝ ௝ ௝ ‫ݏ‬
‫ݏ‬
‫ݍ‬௝௝ by ௝ letting
let ௝ad‫ܨ‬൫ሺ‫ݍ‬ ǡwhose Ͳሻǡelements
Ͳǡ ǥaAs ǡ vector ࢙ሻ‫ݍ‬ൌ ‫ݍ‬௝௝ ൌ ൌ‫ݏݏ‬௝௝ଶǡ ǡߣሺ‫ݍ‬
‫ܨ‬ሺࢊ …ߣare
ൌ ൌሻǡ‫࢙ݍݍ‬ሻ ௝all ‫ݏ‬௝zero
൅‫ܨ‬ሺ݀ except
ሺ‫ݍ‬ ሻǡ ࢙ሻ
For a given Moreover, (q, s), properties j(a) denote
are
ଵ independent. this whose
function elements
ଶAs inherits
this arethe all zero
ଵproperties
inherits
֜ ݂൫‫ݍ‬ except
the of homogeneity
properties
௝ ǡ ‫ݏ‬௝ ൯ ൌ
of and normalizatio
ǥ ǡ Ͳሻǡ ࢙ሻ ൌ ‫ܨ‬ሺࢊ‫ݍ‬௝ଶሺ‫ݍ‬ൌ ଶ‫ݏ‬ሻǡ ǡ ࢙ሻ
ߣൌ ൅‫ܨ‬ሺ݀‫ݍ‬௝ ‫ݏ‬ଵ௝ ,ሺ‫ݍ‬we
Τ ሻǡ have:
࢙ሻ … ‫ݏ‬ ‫ݏ‬ ‫ݏ‬ ‫ݏ‬ ‫ݏ‬ ‫ݏ‬ ‫ݏ‬
al to ja.that Applying
is equalrepeatedly the property of ‫ ݏ‬ǡ ߣ‫ܨݏ‬ሺࢗǡ
‫ݍ‬௝ additivity, we have: ௝‫ݏ‬௝ homogeneity ൫‫ݍ‬൯௝As this
andfunction inherits the properties of homog
௝ ௝
entry to a. Applying ଵ repeatedly the ൌproperty ௝ൌ ‫ݍ‬of Τadditivity, ‫ܨ‬we ൫ࢊቇǡ௝have: normalization, ௝௝ by letting
݂, ‫ݏ‬we
ቆ have:
‫ݏ‬௝ൌ ൯ǡ௝࢙ሻ ௝௝
௝࢙ሻ ෍ ݂ of ‫ݍݍ‬௝௝ǡ ǡ‫ݏݏ‬௝௝ቇቇൌൌ ݂൫‫ݍ‬
ቆ homogeneity ݂൫‫ݍ‬ ௝௝ǡ ǡ‫ݏݏ‬ ൯ൌൌͳͳ
function [A,1]
… satisfies all properties. Let us ݂௝‫ݏ‬ቆAs
verify ‫ݍ‬this
the function
௝ ǡ ‫ݏ‬reciprocal.
௝ቇ ൌ ‫݂ݍ‬൫‫ݍ‬
௝ǡ ‫ݏ‬ ௝௝ ‫ݏ‬ൌ
inherits ௝ ൌthe ݂൫‫ݍ‬ ௝ ǡ ‫ݏ‬௝ ൯ ൌ ͳ݂ ቆ
ͳ properties ‫ݍ‬ ‫ݍ‬ ‫ݍ‬ ‫ݍ‬ and ௝௝൯normalizati
A,1] satisfies
‫ܨ‬൫ሺ‫ݍ‬ଵ ǡ Ͳǡ all ǡ properties.
ǥProof Let us࢙ሻ࢙ሻ verify ‫ܨ‬ሺࢊthe reciprocal.
௝ ‫࢙ݍݍ‬ሻ ௝ ‫ݍܨݍ‬௝௝൫ࢊሺ‫ݍ‬
௝‫א‬௄
‫ݍ‬ǡ௝‫ݏ‬௝൫‫ݍ‬൯ൌ ‫ݏ‬௝࢙ሻ ǡ ߣ‫ݍ‬ൌ ௝ ‫ݍ‬௝‫ݏ‬Τ௝ ‫ݏ‬௝ , we we have: have: ௝ ௝
ଵ ሺ‫ݍ‬ ଵ ሻǡ ሺ‫ݍ‬ ଵൌሻǡൌ࢙ሻ
௝ ௝
Ͳሻǡ ࢙ሻ ‫ܨ‬൫ሺ‫ݍ‬ ൌ ଵ‫ܨ‬ሺࢊ ǡ Ͳǡ ǥ ǡ Ͳሻǡ ൌ൅‫ܨ‬ሺ૙ǡ ሺ‫࢙ݍ‬ሻ ଵ݂ሻǡቆൌ ࢙ሻǡ‫ܨ‬ሺ݀
‫ܨ‬ሺࢗǡ ൅‫ܨ‬ሺ૙ǡ௝ǡ ଵ ‫ݍݏ‬௝ൌቇΤ࢙ሻ ෍
Substituting ‫ܨ‬ሺ݀
݂൫‫ݍ‬ ଵ௝into ൯ǡ
௝ଵ ሻǡ௝equation
ൌ࢙ሻ ͳ [A.2] for all j, we ‫ݏ‬௝ get:
௝ ‫ݏ‬௝ ,‫ݍ‬ we have:
ଵ‫ݍ‬
௝ ൌ ‫ݍݏ‬ ௝ ௝ߣ
௝ൌ
Substituting ௝‫א‬௄ into equation ‫ݍ‬௝ [A.2]
݂ ቆ ‫ݍ‬for
‫ݍ‬ ǡ ‫ݏ‬all ቇ j,
ൌ we get:
݂൫‫ݍ‬ ǡ ‫ݏ‬௝ ൯ ൌ ͳ ‫ݍݍ‬
‫ܨ‬ሺࢗǡ ࢙ሻ ൌ‫ܨ‬൫ሺ‫ݍ‬ ෍ ǡ‫ܨ‬ ൫ࢊ ൫‫ݍ‬ ൯ǡ ࢙ሻ
‫ܨ‬൫ሺ‫ݍ‬ ǡ ‫ݍ‬ ǡ Ͳǡ ǥ ǡ Ͳሻǡ ࢙ሻ ൌ ‫ܨ‬ሺࢊ ሺ‫ݍ‬ ሻǡ ࢙ሻ

൅‫ܨ‬ሺ݀
௝
ሺ‫ݍ‬ ሻǡ

࢙ሻ ‫ݍ‬ ௝ ௝ ௝
‫ݍ‬ ͳ
௝
֜ ֜ ݂൫‫ݍ‬
݂൫‫ݍ‬ ‫ݍ‬ ‫ݏ‬௝ǡ௝ǡ‫ݏݏ‬௝௝൯൯ൌൌ ௝௝ ‫ݏ‬௝
ଵ ‫(ݍ‬i)ଶ ǡ It
Ͳǡ ǥ ǡ ͲሻǡLet
௝ ௝ ࢙ሻଵ ൌ nowଶ ‫ܨ‬ሺࢊ ሺ‫ݍ‬
1aଶfunctionଶ ሻǡ ࢙ሻ[A,1]
denote a൅‫ܨ‬ሺ݀
vector ଶ ଵଶሺ‫ݍ‬  ଵ֜
with ሻǡall ࢙ሻ
݂൫‫ݍ‬ entries ‫ݏ‬௝ ൯ଵ‫֜ݍ‬
௝ଵǡinto ௝ൌ
equal ݂൫‫ݍ‬to ௝ ǡ ‫ݏ‬௝ ൯Inൌ‫ݏ‬
௝
‫ܨ‬ሺࢗǡ ࢙ሻ‫ݏ‬௝ൌ
௝ ௝௝
݂ ቆ‫ݍ‬have: ‫ݍ‬௝ ǡ ‫ݏ‬௝ ቇ‫ݏݏ‬௝௝ൌ ݂൫‫ݍ‬௝ ǡ ‫ݏ‬௝ ൯ ൌ
q,
a)s), let dj(a)
denote
௝‫א‬௄
denote
a vector is easy a vector
whose to see
elements that
whose elements
are all…zero
Substituting
satisfies
are except
֜ ݂൫‫ݍ‬ allallzero௝ ǡ ‫ݏ‬௝ ൯ ൌ except equation ‫ݏ‬௝ a.[A.2] this
݂ ቆ ‫ݏ‬for
௝ particular
all
‫ݏ‬௝ ቇj,ൌwe
‫ݍ‬௝ ௝ǡ ‫ܨ‬ሺࢗǡ ࢙ሻ ൌ
case,
get:
݂൫‫ݍ‬
ͳ෍
݇෍ ‫ݍ‬௝ ǡ௝‫ݏ‬௝‫א‬௄
we
௝൯ ൌ ‫ݏ‬௝‫ݍ‬௝ͳ ௝ ‫ݍ‬௝
properties. Let Let nowus1a …denote
verify a vector with all entries equal
the reciprocal. ‫ݏ‬௝ to a. In this‫ݍ‬particular ௝ ֜ ݂൫‫ݍ‬௝ ǡcase, ‫ݏݍ‬௝௝൯ ൌ ݇ we
ͳ have:
௝‫א‬௄ ‫ݍ‬‫ݏ‬ ௝ ‫ݍ‬௝
ing ‫ݏ‬௝ 
dly repeatedly
with the property
all entries
the
equal ofproperty
additivity,
to a. In thiswe
of additivity, have:
particular
we have:
case, we have:  ‫ܨ‬ሺࢗǡ ࢙ሻ ൌ ෍ ‫ݍ‬
݇ൌ consider ֜ ݂൫‫ݍ‬ ǡ ‫ݏ‬ ൯ ൌ
‫ܨ‬ሺࢗǡ ࢙ሻ ൌ ෍ ‫ ܨ‬൫ࢊ௝ ൫‫(ݍ‬ii) ௝ ൯ǡ ࢙ሻ ௝‫א‬௄ ‫ݏ‬௝ the following indices:
௝ ௝ ௝
‫ܨ‬ሺࢗǡ ࢙ሻ ൌ෍ ‫ܨܨ‬ሺ݀ ൫ࢊ௝ ൫‫ݍ‬ ௝ ൯ǡ ࢙ሻ In order27to separate those properties, ֜ ݂൫‫ݍ‬௝ ǡ ‫ݏ‬let ௝ ൯ us
‫ݏ‬௝
ǥ
‫ܨ‬ሺࢊ ǡ Ͳሻǡ ࢙ሻ
ሺ‫ݍ‬ ሻǡൌ ࢙ሻ ‫ܨ‬ሺࢊ
൅‫ܨ‬ሺ૙ǡଵ ሺ‫ݍ‬ ଵ ሻǡ
࢙ሻ ࢙ሻൌ ൅‫ܨ‬ሺ૙ǡ
‫ܨ‬ሺ݀ ሺ‫ݍ‬ ࢙ሻ ሻǡൌ ࢙ሻ ଵ ሺ‫ݍ‬ ଵ ሻǡ ࢙ሻ
௝‫א‬௄
Substituting into equation [A.2] for
‫ݏ‬ ௝ all‫ݍ‬j,j,௝wewethe get:
ଵ ଵ ଵ ଵ ௝‫א‬௄ (ii) In order 27to Substituting
separate those into equation
properties, [A.2]
let usfor all
consider get: following indices
Substituting Substituting
into equation into equation
[A.2] for all [A.2]
j, we for
get: all j, we get: ‫ ܨ‬஺ ሺࢗǡ ࢙ሻ ൌ ෍
‫࢙ݍ‬ሻ ǡ Ͳǡ ǥ
ଶ ൌ ‫ܨ‬ሺࢊଶ ሺ‫ݍ‬ ǡ Ͳሻǡ ࢙ሻ ൌ ‫ܨ‬ሺࢊ
27ଶ ሻǡ ࢙ሻ ൅‫ܨ‬ሺ݀ ሺ‫ݍ‬
Substitutingሻǡ
ଶ ଶ ሺ‫ݍ‬ଵ ሻǡ ࢙ሻ ࢙ሻ ൅‫ܨ‬ሺ݀ into ሺ‫ݍ‬
equation
ଵ ଵ ሻǡ ࢙ሻ [A.2] for all j, we get:
(ii) In order to separate those properties, ஺ ሺࢗǡ let us consider
‫ݏ‬‫ݍ‬
௝‫א‬௄ ௝ ௝ the ͳ
ͳ following ‫ݍݍ‬௝௝ indices
Let now 1a denote ଵa vector with all entries equal Substituting
to a. In this into equationcase,
particular [A.2]
‫ݍ‬௝ ͳ for
we all
‫ܨ‬
have: ‫ݍ‬j, we ࢙ሻ get:
ൌ ෍ ‫ܨ‬ሺࢗǡ
‫ܨ‬ሺࢗǡ ࢙ሻ࢙ሻ ൌ ൌ ෍෍
ͳ case, ͳ we ‫ ݍ‬have: ௝‫א‬௄ ‫ݍ‬ ‫ݏ‬௝ ݇݇ we ௝‫א‬௄‫ݏݏ‬
1a denote … a vector…with all entries equal to a. In this ‫ܨ‬ሺࢗǡ particular ௝
‫ܨ‬ሺࢗǡ ࢙ሻ ൌ࢙ሻ ൌ ෍ ‫ܨ‬ሺࢗǡ ෍ ௝࢙ሻ Substituting
ൌ ෍ ‫ ܨ‬஺into ሺࢗǡ ࢙ሻ equation
ൌͳ ෍ [A.2] ‫ݍ‬௝ for all j, ௝‫א‬௄ get:
௝௝
Substituting It satisfies into
݇all ݇ equation
properties
௝‫א‬௄ ‫ݏ‬௝‫א‬௄ ௝ ௝ ݇ for
‫[ݏ‬A.2]
except all ‫ݏ‬j,௝࢙ሻ
normalization.
௝‫א‬௄ ‫ܨ‬ሺࢗǡ weൌget:෍௝‫א‬௄ ‫ݏ‬௝
ͳ
‫ܨ‬ሺࢗǡ ࢙ሻ ൌ௝෍ ‫ ܨ‬൫ࢊ௝ ൫‫ݍ‬௝ ൯ǡ ࢙ሻ ݇ ௝‫א‬௄ ‫ݏ‬௝
ൌ෍ ‫ ܨ‬൫ࢊ ൫‫ݍ‬௝ ൯ǡ ࢙ሻ 27 It satisfies all properties except normalization.
(ii)In Inorder orderto toseparate
separate those ͳ
properties, ‫ݍݍ‬௝௝
letus ‫ܨ‬ሺࢗǡ
usconsider ࢙ሻ
consider ൌ ෍ the௝‫א‬௄ fo
௝‫א‬௄ (ii) ‫ܨ‬ ஻
‫ܨ‬ሺࢗǡ those
ሺࢗǡ ࢙ሻ࢙ሻ ൌ ൌproperties,
݉݅݊ ෍ ௝‫א‬௄ ቊ ቋ
let ݇ the fo
௝‫א‬௄ 27 ‫ݏ‬
ࢗǡ
‫ݍ‬௝࢙ሻ
Τ‫ݏ‬௝ൌ ߣ‫ܨ‬ሺࢗǡ
, we ࢙ሻ , for allߣ ൐ Ͳ . ௝Define
have:
௝
a new௝ function ௝
B. Weighting f : » of»the +  GEP » Indexas follows: in the case of several countries
observe that our assumptions imply that function F is homogeneous of degree 1 in
‫ݏ‬௝ ௝ ǡ ‫ݏ‬௝ ൯ ‫׷‬ൌ ‫ܨݏ‬൫૚‫ݍ‬
݂൫‫ݍ‬ ௝ ௝ ǡ ૚‫ݏ‬௝ ൯
ࢗǡhis
࢙ሻ ൌfunction
ߣ‫ܨ‬ሺࢗǡ ࢙ሻinherits
, for allߣthe
݂൐ ቆ Ͳproperties
‫ ݍ‬ǡ ‫ݏ‬௝ ቇ ൌa new
‫ݍ‬௝ . ௝Define ‫ݍ‬௝
of݂൫‫ݍ‬ homogeneity
௝ ǡ ‫ݏ‬The ͳ f : »and
௝ ൯ ൌ objective
function  »of normalization,
this section
+ »
as follows: bytoletting
is discuss how to make a comparative evaluatio
‫ݍ‬௝ Τ‫ݏ‬௝ , we have: countries C. Now the variables are ‫ݕ‬௝ଵ ሺ݅ሻǡ ‫ݕ‬௝଴ ሺ݅ሻǡ ‫ݕ‬௝‫ כ‬ሺ݅ሻǡ ‫ݐ‬௝ ሺ݅ሻǡ ߨ௝ ሺ݅ሻ where the term (i)
his function inherits the݂൫‫ݍ‬ ௝ ǡ ‫ݏ‬௝ ൯ ‫׷‬ൌ ‫ܨ‬൫૚‫ݍ‬
properties of ௝ ǡ‫ݍ‬૚‫ݏ‬௝ ௝൯
homogeneity and normalization, by letting
‫ݏ‬௝ ֜ ݂൫‫ݍ‬௝ ǡ ‫ݏݏ‬௝௝ ൯ ൌ ‫ݏ‬௝ country in the corresponding set C.
‫ݍ‬௝ Τ‫ݏ‬௝ , we have: ݂ ቆ ‫ݍ‬௝ ǡ ‫ݏ‬௝ ቇ ൌ ݂൫‫ݍ‬௝ ǡ ‫ݏ‬௝ ൯ ൌ ͳ
‫ݍ‬௝ ‫ݍ‬௝
his function inherits the ‫ݏ‬properties ‫ݏ‬of homogeneity
Two types and of normalization,
comparisons by among letting countries can be considered. One refers t
௝ 
݂ ቆ ‫ݍ‬௝֜ǡ ‫݂ݏ‬൫‫ݍ‬
௝ ‫ݍ‬
௝ ቇ ൌ௝ ǡ ‫ݏ‬௝ ൯݂൫‫ݍ‬ ௝ ǡ ‫ݏ‬countries
௝൯ ൌ ͳ
௝
‫ݍ‬௝ Τ‫ݏ‬௝ , we have: Substituting ൌ
‫ݍ‬௝ into equation ‫ݍ‬௝ [A.2] ‫ݏ‬௝ for all j, wewith get: respect tosimply a single take indicator.
the values: In that case, we simply take the values:
stituting into equation [A.2] ‫ݏ‬௝ for all j, we‫ݏ‬௝ get: ‫ݍ‬௝
23 ݂ ቆ ‫ݍ‬௝֜ ǡ ‫ݏ‬௝݂൫‫ݍ‬
ቇ ൌ௝ ǡ ‫ݏ‬௝ ݂൫‫ݍ‬൯ൌ ͳ ௝ ǡ ‫ݏ‬௝ ൯ ൌ‫ͳݍ‬ ݀‫ݕ‬௝ ሺ݅ሻ ݀ሺെ‫ݕ‬௝ ሻሺ݅ሻ
‫ݍ‬௝ ‫ݍ‬
‫ܨ‬ሺࢗǡ ࢙ሻ௝ ൌ ෍ ‫ݏ‬௝ ௝
‫ כ‬ሺ݅ሻ ǡ ݂‫ܩ א ݆ݎ݋‬ǡ ǡ ݂‫ܤ א ݆ݎ݋‬
 ݇ ‫ݍ‬௝ ௝‫א‬௄ ‫ݏ‬௝ ݀‫ݕ‬௝ ݀ሺെ‫ݕ‬௝‫ כ‬ሻሺ݅ሻ
֜ ݂൫‫ݍ‬ B. Weighting
ǡ ‫ݏ‬௝ ൯get:
ൌ of B. the GEP
Weighting Index of in
the theGEP case Index of several
in the countries
case of several countries
stituting into equation [A.2] for all j,௝we ‫ݏ‬௝
n order tointo separate those The ࢙ሻ ͳ
objective and of ‫ݍ‬this
The see how
section
௝objective those values
ofisthistoindices:
discuss
section are inisthe
and seehow to to corresponding
make
discuss a comparative
how to countries.
inmake aevaluation
comparative of evaluatio
a set of
stituting equation [A.2]properties,
for all‫ܨ‬ሺࢗǡ j, welet us෍
ൌget:
݇
consider
‫ݏ‬
the following how those values are the corresponding
(ii) In order to separate countries C. B.
Now
௝‫א‬௄ Weighting
countries
‫ݍ‬ ௝the variables C. of
Now the GEP
are
the ‫ݕ‬ Index
ଵ
variables
ሺ݅ሻǡ ‫ݕ‬ ଴
inሺ݅ሻǡ the
are ‫ݕ‬ ‫כ‬
caseଵ
ሺ݅ሻǡ
‫ݕ‬ ‫ݐ‬
ሺ݅ሻǡ of
ሺ݅ሻǡ‫ݕ‬ ଴several
ߨ
ሺ݅ሻǡ ሺ݅ሻ ‫ݕ‬ ‫כ‬ countries
where
ሺ݅ሻǡ ‫ݐ‬ the
ሺ݅ሻǡ ߨ term
ሺ݅ሻ (i)
where refers
the to the
term (i)
‫ ܨ‬஺ ሺࢗǡ ࢙ሻthose properties, ‫ݍ‬௝௝otherlettype us countries.
௝ ௝ ௝ ௝ ௝ ௝ ௝ ௝ ௝ ௝
ൌͳ ෍ The of comparison refers to the corresponding GEP indices. In this cas
allcountry
consider thefor ‫ܨ‬ሺࢗǡ ࢙ሻindices:
following ൌin the෍௝‫א‬௄ corresponding
country‫ݏ‬௝ set C.
in the corresponding set C.
tituting into equation [A.2] j, we get: B. Weighting of
inthe theGEP
݇
n order to separate those properties, let us consider the following indices: decide
The
௝‫א‬௄ ‫ݏ‬௝objective
how to weigh
of this each
section dimensionis to discuss howIndex
different to make in theacase
countries. Thereof several
comparative are two countrie
possi
evaluatio
ͳ ‫ݍ‬ The other type of comparison refers to the
ൌ ෍ofregard: choosing a the
common weighting ‫ݕ‬௝ system or‫ݕ‬keeping each country’s individual
term (i)w
ଵ ଴ ‫כ‬
Two ࢙ሻ types countries
Two ‫ݍ‬௝௝ types
comparisons C. of Now among
comparisons variables
countries among arecan ሺ݅ሻǡbe‫ݕ‬
countries ሺ݅ሻǡ
௝considered. ௝canሺ݅ሻǡ ‫ݐ‬be ௝ ሺ݅ሻǡ
One ߨ௝ ሺ݅ሻ referswhere to the
comparing
tisfies all properties except normalization.
‫ܨ‬ሺࢗǡ
‫ܨ‬ ஺ ሺࢗǡ
࢙ሻ ݇
ൌ ෍ ‫ݏ‬ The objective
corresponding of
GEP this
indices. section In this toconsidered.
is case, discuss
one hashow toOne to refers
make a
n order to separate those properties, countries let uswith consider
country
௝‫א‬௄
‫ݏ‬௝ the
௝
respect
countries intofollowing
the
with respect indices:
corresponding
a single indicator.
to a single set
In that C.indicator.
case, we In simply
that case, takewe thesimplyଵvalues: take
଴ the‫ כ‬values:
௝‫א‬௄
A common ‫ݍ‬௝ weighting system countries
decide makes howC.toNow weigh theeach
the international variables
dimension arein‫ݕ‬the
comparison ௝ ሺ݅ሻǡ different
‫ݕ‬௝ more
ሺ݅ሻǡ ‫ݕ‬௝ intuitive
ሺ݅ሻǡ ‫ݐ‬௝ ሺ݅ሻǡ a ߨ
‫ܨܨ‬ ஻ ሺࢗǡ
஺ ሺࢗǡ ࢙ሻ࢙ሻൌൌ݉݅݊ ෍௝‫א‬௄ ቊ‫ݍ‬௝ ቋ country
countries. inThere
the corresponding
are two possibilities set C. in this regard:
order to separate those properties, let us consider losing
Two the following
‫ݏݏ‬௝௝ types
the individual
of݀‫ݕ‬comparisons traits (thatamong is, possibility
countries of
can using be this
considered.formula to
One individually
refers t
௝‫א‬௄ ௝ ሺ݅ሻindices: ݀‫ݕ‬ ௝ ሺ݅ሻ ݀ሺെ‫ݕ‬ ௝ ሻሺ݅ሻ ݀ሺെ‫ݕ‬ ௝ ሻሺ݅ሻ
tisfies all properties except normalization. achievement of the ǡ ݂‫݆ݎ݋‬
objectives). ‫א‬
choosing ‫ܩ‬ǡ The most‫ כ‬natural
aindicator.
common
ǡ ݂‫݆ݎ݋‬ ‫א‬ ‫ܩ‬ǡ ǡ ݂‫݆ݎ݋‬
weighting ‫א‬
choice ‫ܤ‬
system would ǡ be defining
or keeping
݂‫݆ݎ݋‬ ‫א‬ ‫ܤ‬ those co
countries
‫ݍ‬ ௝ with respect
‫כ‬
݀‫ݕ‬௝ ሺ݅ሻ to a single ‫כ‬
݀‫ݕ‬௝ ሺ݅ሻ ݀ሺെ‫ݕ‬௝ ሻሺ݅ሻ In that case,
݀ሺെ‫ݕ‬௝ ሻሺ݅ሻ ‫כ‬ we simply take the values:
‫ ܨ‬஺ ሺࢗǡ ࢙ሻ ൌ
It satisfies all properties ෍ anormalization.
except weighted
‫ݍ‬௝ average of the Two
each types ofcoefficients
individual
country’s comparisons
individual of all
weights. amonginvolved countries
countries can using be theco
tisfies all
tisfies all properties
properties exceptexcept normalization.
additivity.
஻
‫ ܨ‬ሺࢗǡ ࢙ሻ ൌ ݉݅݊௝‫א‬௄ ௝‫א‬௄ ቊ௝ ቋ ‫ݏ‬
shares. ‫ݏ‬௝ That will produce the݀‫ݕ‬ countries following with respect
ሺ݅ሻthe correspondingformula for݀ሺെ‫ݕ‬ to a single
each indicator.
country i in the set C: In that case, we
and ஼see how those ‫ݍ‬and௝ values
see howare thosein the valuescorresponding
are ௝in countries. ௝ ሻሺ݅ሻ
countries.
‫ܨ‬ ሺࢗǡ ࢙ሻ ൌ ‫ݍ‬ ǡ ݂‫݆ݎ݋‬ ‫א‬ ‫ܩ‬ǡ ǡ ݂‫ܤ א ݆ݎ݋‬
‫ܨ‬ ஻ ሺࢗǡ
࢙ሻ ൌ ݉݅݊ ‫ݏ‬ ቊ
௝
ቋ A common݀‫ݕ‬௝ ሺ݅ሻ weighting system
‫כ‬
݀ሺെ‫ݕ‬ makes ‫כ‬
ሻሺ݅ሻthe
isfies all properties except normalization. ௝‫א‬௄ ௝
‫ݏ‬௝ ௗ௬ೕ ሺ௜ሻ ݀‫ݕ‬௝ ሺ݅ሻ௝ ௗሺି௬ೕሻ ݀ሺെ‫ݕ‬௝ ሻሺ݅ሻ
tisfies The other typeThe of comparison refers international
‫ܲܧܩ‬ሺ݅ሻ
to the corresponding σ௝‫ீא‬
ൌrefers comparison
ߨ ൅
GEP σ௝‫א‬஻
more ߨintuitive
indices. ‫כ‬In at
thisthe cost one
case, has to(Bǡ
tisfies allall properties
properties exceptexcept additivity.
neutrality. q.e.d. ‫ݍ‬ ௝
other type of comparison to ௝ ௗ௬the ‫ כ‬corresponding
ೕ ሺ௜ሻ
݀‫ݕ‬
௝‫ כ‬ௗሺି௬
ሺ݅ሻ
ǡ ݂‫݆ݎ݋‬
ೕ ሻGEP ‫א‬indices.
‫ܩ‬ǡ In this
݀ሺെ‫ݕ‬ ‫ כ‬cas
ሻሺ݅ሻ
‫ ܨ‬஻ decide
ሺࢗǡ ࢙ሻ஼ ൌhow ݉݅݊௝‫א‬௄ to‫ݍ‬ and ቊ see
௝weigh
decide
where: ቋ each howto
how those
dimension
weigh valueseachof
in losing
are
the in the
the individual
different
dimension corresponding
incountries. traitsThere
the different (that ௝ is, the possibility
countries. are two There
countries. possibilities
are twoinposs
௝
this
tisfies all properties except additivity. ‫ܨ‬ ሺࢗǡ ࢙ሻ ൌ ‫ݏ‬ ௝
regard: choosing ‫ݏ‬regard:
௝ a common
choosing weighting
a common of using
system this
weighting or formula
keeping system to
each individually
or country’s
keeping
݊ሺ݅ሻ evaluate
individual
each the
country’s weights. individual
and seerefers howߨ௝those ൌthe values
෍objectives). are in the corresponding
indices. Incountries
It satisfies all
tisfies all properties except neutrality. properties ஼ except
q.e.d.
ሺࢗǡ
‫ݍ‬The other
additivity.
௝ type of comparison achievement ofto the corresponding
σ
௜‫א‬஼ ௛‫א‬஼ ݊ሺ݄ሻ
Theߨ௝most ሺ݅ሻGEP natural this cas
‫ ܨ‬28 ࢙ሻ ൌ
isfies all properties except additivity. A common weighting ‫ݏ‬decide
A
௝ common howsystem to weigh
weightingmakes each thedimension
system
choice international
makesbe
would inthe the different
comparison
international
defining those countries. morecomparison
coefficients There
intuitive as amoreatarethe two cost possi
intuitive of
‫ݍ‬ regard: choosing a common The other weighting type system of comparison or keeping refers each tocountry’s
the correspondingindividual G
w
tisfies all properties except neutrality. losing the individual ௝losing traits the individual(that is, the traits possibility
of(that is, of using
the possibility thisindividual
formula
of using tothisindividually
formula evaluate
to individuall the
‫ ܨ‬஼ ሺࢗǡq.e.d.
࢙ሻ ൌ where ߨ௝ ሺ݅ሻ is the weightdecide weighteddimension average j inofcountry
the i,
how to weigh each dimension in the different countr n(i) the population
coefficients of of country i a
achievement ‫ݏ‬of the objectives).
௝achievement
population of of the
country The most
objectives).
h, natural
alla involved
country The choice
most is
which
countries would
natural
in the
using be
choice
theirsame defining would
sub-group
population those be coefficients
defining
shares.asorcountry those(i) asc
28 A common weighting regard:
system choosing
makes the ainternational
common weighting
comparison system more keeping
intuitive ea
isfies all properties except neutrality. a weighted q.e.d.average a weighted
countries of the that individual
averageshareofthe coefficients
the
That individual
same level
will produce ofcoefficients
allthe
of involved
development
following ofcountries
all involved
formula according using
for their
countries
each to the using tha
population
Human
shares. That will losing
shares.
Index). producethe That individual
the will following
producetraitsformula
(that
the is,for
following theeach possibility
formulacountry ofi each
for using
in thecountry this
set C: formula i in the toset individually
C:
It satisfies all properties 28 exceptachievement neutrality. q.e.d. country
A common i in the set C: system makes the international com
of the objectives). Theweightingmost natural choice would be defining those co
a weighted average losing the individual traits of (that is, the possibility of using
using the th
ൌ σof the individual coefficients all involved countries
ௗ௬ ೕ ሺ௜ሻ ௗሺି௬
ௗ௬ ೕ ሺ௜ሻೕሻ ௗሺି௬
The alternative
‫ܲܧܩ‬ሺ݅ሻ ߨ௝ ௗ௬
‫ܲܧܩ‬ሺ݅ሻ
௝‫ீא‬achievement is
‫ כ‬ൌ ൅σσ௝‫ீא‬ keeping
ߨ ߨ
offormula
the ൅
‫כ‬objectives). σ the
ߨ
௝‫א‬஻ ௝ The ‫כ‬most
ೕሻ
individual
(B.1)
natural we (
the setchoice wo
௝‫א‬஻ ௝
௝ ௗ௬
28 shares.గෞണThat will produce theೕfollowing ሺ௜ሻ ௗሺି௬ ‫כ‬
ೕ ሺ௜ሻೕ for each ௗሺି௬
ሻ country ೕሻ i in C:
B. Weightingwhere: of the GEPߨwhere: ௝Index
ൌ σ గෞ
ೕ‫ ಻א‬ണ
inሻ, the which is the approach
a weighted averageadopted of the for individual
the GEP coefficients Index. This ofoption all invo p
case of several countries individual characteristics, shares. Thatfor will produce
݊ሺ݅ሻ the
݊ሺ݅ሻfollowing formula for each butcou
ൌallowing ߨ௝ ߨindividual ൅ σevaluation
ௗ௬ೕ ሺ௜ሻ ௗሺି௬ೕሻof achievement, ma
ߨ௝‫ܲܧܩ‬ሺ݅ሻ ෍ ൌ σ௝‫ீא‬ ൌ௝ ௗ௬ ෍ ‫ߨכ‬௝ ሺ݅ሻ ௝‫א‬஻ ߨ௝ ௗሺି௬ ߨ௝ ሺ݅ሻ‫כ‬ (B
to interpret the international comparison. However, where: ௜‫א‬஼ σ ௛‫א‬஼ ݊ሺ݄ሻ ೕ ሺ௜ሻ
௜‫א‬஼ σ ݊ሺ݄ሻ ሻ
௛‫א‬஼the country-specific weighting sy
ೕ
ௗ௬ೕ ሺ௜ሻ ௗሺି௬ೕ
where:
given a sensiblehow to interpretation in terms of egalitarian ‫ܲܧܩ‬ሺ݅ሻ ൌ σ௝‫ߨ ீא‬௝values.
equivalent ൅ σ௝‫א‬஻ ߨ௝
GEP Index in the The caseobjective
of several of this section is to
countries discuss
݊ሺ݅ሻ ௗ௬ೕ‫כ‬ሺ௜ሻ ௗሺି௬ೕ‫כ‬
make a comparative where ߨ௝ ሺ݅ሻ is the
evaluation of aweight
where setߨof ௝ ሺ݅ሻ of isdimension
countries the weight jof in dimension
country i, n(i) j inthe country population i, n(i) the of country
population i and ofn(h)country is thei a
where: ߨ௝ ൌ ෍ ߨ௝ ሺ݅ሻ
population of country
population h, a country
of country which is in the same σ
sub-group ݊ሺ݄ሻ as country (i) (for example
s section is to discuss C. Now the how to make
variables area comparative Note thatevaluation
each country’s of h, a seta country
individual
of which isgive
weights ௜‫א‬஼ in ௛‫א‬஼thethesame (constant) sub-group slopeas of country
the indiffe
݊ሺ݅ሻ (i)
‫ כ‬countries that for share
countries
the the that
corresponding same share level the of
variables, samedevelopment
as level
they of
define according
development
the to
associated the
according
ߨ Human
ൌ
rates ෍ ofto Development
the
substitution.Human ߨ ሺ
݊ሺ݄ሻ i aA
ଵ ଴
he variables are ‫ݕ‬௝ ሺ݅ሻǡ ‫ݕ‬௝ ሺ݅ሻǡ ‫ݕ‬௝ ሺ݅ሻǡ ‫ݐ‬௝ ሺ݅ሻǡ ߨ௝ ሺ݅ሻ where where the theterm term (i) (i) refers to the ௝
௜‫א‬஼ σ
௝
Index). where
Index).
determine ߨ ௝ ሺ݅ሻ is the weight of dimension j in
which indifference curve is considered for each country, from the spe country i, n(i) the population of௛‫א‬஼ country
ponding set C.
population ofFigure
realizations. country B.1 h, a country
illustrates the which is of in thecountries,
same sub-group as country (i)
refers to the country in the corresponding set C. share the where ߨ௝ ሺ݅ሻ is case
the weight
the weight two
ofofdimension
dimension redj in and
j in country
blue,
country with
i, n(i)realiz
the
The countries
alternative
The
B, respectively, that is
alternative same
keeping is level ofthe
keeping development individual
the according to
weights
individual the Human we (
parisons among countries can be considered. One refersregarding to comparing two indicators. The figure makes it clear that the proble
population
i, n(i) the population of country h, a country
of country i and n(h) which is theis in the same su
గෞണ Index).
comparing గෞണ
ct to a single indicator. In that ߨ௝ ൌ we
case, σೕ‫಻א‬simply
ሻ, which ߨ௝take ൌ σis the theሻ,those
ෞണvalues:
which realizations
approach is the adopted
countries
which
approach forcorrespond
the
adopted GEPtheto Index.
forsameindifference
theThis leveloption
GEP curves
Index. that intersec
preserves
This option the
accp
Two types of comparisons ෞണamong
గ countries
ೕ‫ ಻א‬గ can be population that of country share h, a country which isofin development
the
is above the other in one part and below in another.
considered. One individual refers to comparingcharacteristics,
individual
The countries with forIndex).
allowing
characteristics,
alternative individual
same allowing
issub-group evaluation
forkeeping
individual
as country of achievement,
evaluation
(i) (for
the example of but makes it harder
achievement,
individual but we m
݀‫ݕ‬௝ ሺ݅ሻ ݀ሺെ‫ݕ‬ to ௝ ሻሺ݅ሻ
interpret the to international
interpret the comparison.
international However,
comparison. the However, 29 the country-specific
country-specific weighting system weighting can be s
respect
ǡ ݂‫݆ݎ݋‬ ‫ܩ א‬ǡto a single indicator. ǡ ݂‫݆ݎ݋‬In that
ߨ‫א‬௝ ‫ܤ‬ ൌcase,
గෞ ണ weሻ, which is the countries
approach that share adopted the same for the levelGEP of development
Index. This option p
݀‫ݕ‬௝‫ כ‬ሺ݅ሻ ݀ሺെ‫ݕ‬ given
‫כ‬
ሻሺ݅ሻa sensible interpretation
given a sensible interpretation
σ ෞ
గ in terms The of egalitarian
in terms alternative
equivalent
of egalitarian values.isequivalent keeping
values. the
௝ ೕ‫಻א‬ ണ
according to the Human Development Index).
individual characteristics, ߨ௝ allowing
ൌ σ గෞ ሻ,for
ෞണ
గ individual
which is theevaluation approachofadopted achievement, for the butGEP ma
alues are in the corresponding countries. Note that each to interpret
country’s the international
individual weights comparison.
give
ೕ‫ ಻א‬ണ
the
Note that each country’s individual weights give the (constant) slope of the indiffe However,
(constant) the
slope country-specific
of the indifference weighting curves s
given individual characteristics, allowing for individual evaluation
for the corresponding for thea corresponding
sensible
variables, interpretation
as they define
variables, in as terms
thethey ofdefine
associated egalitarian therates equivalent
of substitution.
associated values.
rates ofActual
substitution.values
to interpret the international comparison. However, the coun
omparison refers to the corresponding determine GEP whichindices.
determine In this
indifference which case,curve one is has
indifference tocurve is
considered forconsidered
each country, for each from country, the specific frompoint the spe of
Note that each country’s given a
individual sensible weights interpretation in terms of egalitarian equival
h each dimension in the differentrealizations. countries. There Figure are
realizations.B.1two possibilities
illustrates
Figure B.1 in
the illustrates
case this
of two the case ofgive
countries, redthe
two and(constant)
countries, blue, with red slope of the
realizations
and blue, with Aindiffe
and
reali
ommon weighting system or keeping each country’s
B, respectively, for the corresponding
B,regarding individual
respectively, weights.
two regarding
indicators.variables,two as
Theindicators. they define
figure makes The the it clear
figure associated that the
makes itrates ofthat
problem
clear substitution.
isthe that of
prob
determine which indifference Note that curve each is country’s
considered individual
for each weights
country, give
from thethe (constspe
comparing those comparingrealizations those which correspond
realizations which to correspond
indifferencetocurves indifference that intersect, curves that so that one
intersec
realizations. Figure for the corresponding variables, as they reddefine the with associate
g system makes the international is abovecomparison
the other more
is above in one intuitive
the part otherand atinB.1
the
below
one illustrates
cost and the
of
in another.
part below case in of two
another. countries, and blue, realiz
B, respectively, regarding determine
two which indifference curve is it considered forprobl
each
traits (that is, the possibility of using this formula to individually evaluate theindicators.
29 The figure makes
29
clear that the
comparing those realizations realizations.which Figure correspond B.1 illustrates
to indifference the case curvesof two that countries,
intersec
 ௝‫ ீא‬௝ ௗ௬ ‫כ‬ሺ௜ሻ ௝‫א‬஻ ௝ ௗሺି௬ ‫ כ‬ሻ
ೕ ೕ

where:
݊ሺ݅ሻ
ߨ௝ ൌ ෍ ߨ௝ ሺ݅ሻ
௜‫א‬஼ σ௛‫א‬஼ ݊ሺ݄ሻ

where ߨ௝ ሺ݅ሻ is the weight of dimension j in country i, n(i) the population of country i and n(h) is the
population of country h, a country which is in the same sub-group as country (i) (for example
countries that share the same level of development according to the Human Development
Index).

T
The
he alternative is keeping
alternative is the individual
keeping weights the The egalitarian
individual weights equivalent( approach consists of
ෞ
గ
24 ߨ௝ ൌ σ ണ ෞ ሻ,, which
which is
is the
the approach
approach adopted
adopted for for
the the
GEP Index. selecting
This option a ray that
preserves intersects
the both indifference
ೕ‫ ಻א‬గണ

GEP Index.
individual This option
characteristics, preserves
allowing the individual
for individual curves
evaluation of achievement, butand allows
makes one to compare the points in
it harder
to interpret the international comparison. However, the country-specific weighting system can be
characteristics, allowing for individual evaluation of
given a sensible interpretation in terms of egalitarian equivalent values.
the indifference curves along this ray. Note that
achievement, but makes it harder to interpret the those points will always be comparable as they are
Note that each country’s individual weights give the (constant) slope of the indifference curves
international comparison. However, the country- situated over the same ray and that each point in
for the corresponding variables, as they define the associated rates of substitution. Actual values
specific
determineweighting systemcurve
which indifference canisbeconsidered
given a sensible that the
for each country, from intersection is equivalent
specific point of (i.e. yields the same
realizations. Figure B.1 illustrates the case
interpretation in terms of egalitarian equivalent of two countries, red and blue, with realizations A and
evaluation) to the corresponding original value.
B, respectively, regarding two indicators. The figure makes it clear that the problem is that of
values.
comparing those realizations which correspond to indifference curves that intersect, so that one
is above the other in one part and below in another. Interestingly enough, it can be shown that keeping
29
Note that each country’s individual weights give independent weights for the different countries (the
the (constant) slope of the indifference curves for approach adopted here) amounts to selecting the
the corresponding variables, as they define the 45o line as the appropriate ray, as shown in Figure
associated rates of substitution. Actual values B.2. What is the meaning of taking the diagonal
determine which indifference curve is considered for ray? The diagonal ray represents, in the GEP Index
each country, from the specific point of realizations. context, the case in which the achievements meet
Figure B.1 illustrates the case of two countries, red the targets in all dimensions. Therefore, using
and blue, with realizations A and B, respectively, country-specific weights to aggregate dimensions
regarding two indicators. The figure makes it amounts to using the egalitarian equivalent
clear that the problem is that of comparing those approach, in which the 45o line is associated with
realizations which correspond to indifference curves full achievement of targets.
that intersect, so that one is above the other in one
part and below in another.

Figure B.1: Comparing realization A and B with Figure B.2: Comparing realizations A and B alone
linear indifference curves the 45˚ line (when targets are met in all
dimensions)

A A
A’
Dimension 2

Dimension 2

B B
B’

Dimension 1 Dimension 1
 ࣞሺ‫ݔ‬ǡ ‫ݖ‬ሻ ൌ ሼ݇following
The ‫ א‬ሼͳǡ ǥ ǡ ‫ ܬ‬is
൅a ‫ݔ‬௞ ് ‫ݖ‬௞ ሽ
ͳሽǣconsequence of Theorem 3 in Barberà an

The following is a consequence of Theorem 3 in Barberà and Jackson (JET, 1988).

Theorem (The Protective Criterion)

Theorem (The Protective Criterion)

Let ‫ظ‬௉ be a symmetric and convex order on Թ௃ାଵ that satis
Consequences and Independence of Duplicated Conseque
C. Aggregating Let ‫ظ‬thebe a symmetric
information and
fromconvex order on Թ௃ାଵ that satisfies Pareto,ofIndependence of Identical
Figure B.2: Comparing realizations௉
A and B alone the 45˚ line (when Consequences
targets are metand Independence
in all Duplicated
the dashboard Consequences
andin the and Independence of Duplicated Consequences. Then
one 2545˚ line (when
dimensions)
the targets are met all GEP Index, if and
Consequences. Then ‫ظ ݔ‬௉ ‫ ݖ‬if and onlyonly
if if ‹௞‫ࣞא‬ሺ௫ǡ௭ሻ
creating the GEP+ (The Protective
C. Aggregating FigureB.2:
Figure B.2:the
Criterion) information
Comparing
Comparing realizations
realizations from AAand ‫ ݔ‬the
and ‫ظ‬B௉Balone dashboard
‫ݖ‬alone ifthe
the and 45˚
45˚ lineand
line
only (when
if(when the GEP
targets
targets
‹ ௞‫ࣞא‬ሺ௫ǡ௭ሻ
are‫ݔ‬
are Index,
metin
met inall
all
௞ ൐ ‹௞‫ࣞא‬ሺ௫ǡ௭ሻ ‫ݖ‬௞
32
Aom dimensions)
dimensions)
and the
B alone
creating dashboard
the
the45˚GEP+ and
line (when (The the
targets GEP
Protectiveare metIndex, inCriterion)
all
32
e Criterion) C.AAggregating
C. Aggregating
comparison between thethe
the information
information
GEP Index and from from the
the the dashboard
dashboard
This ordering, like and
and the thetheordering,
min GEP Index,
GEP Index,
compares
onsons
ation
ns AAAandand
and B
B B alone
fromcreating
alone
alone the the
the
the 45˚
dashboard
45˚
45˚
Figure line
line
line
B.2: (when
(when
(whenComparingandtargets
targets
targets are
theProtective
are
are GEP met
met
met
realizations in
in
in all
allAall
Index, and B alone the 45˚ line (when targets are met in all but only
A comparison creating between
information the
the the
from GEP+
GEP+ GEP
the (The
(The
Index
dashboard Protective
and
is made the Criterion)
Criterion)
information
according from
alternativesthe dashboard
in terms of their is made
worst case
dimensions)
otective
ndaccording Criterion)
the information to atoB.2:
Figure symmetric
Comparing
from the
a symmetric and and convex
convexorder
realizations
dashboard order A‫ظ‬
is andonBthe
made the
alone spaceԹ
space the 45˚ oftaking
achievements,
௃ାଵline (when targets are met in all
into account which satisfies on which they
the dimensions
ormation dimensions)
from the dashboard and the GEP Index,
rmation from
nd
and
‫ظ‬andthe
rmationB
onB following
alone
B alone
alone
the from
the
the
the
spaceԹ
Figure45˚
45˚
45˚ the
principles:
the
௃ାଵ
C.
line
line
B.2:
AA comparison line
of
comparison dashboard
dashboard
Aggregating
(when
(when
(when targets
achievements,
ofComparing
achievements,targets
targets
between
between areareand
and
are theGEP
met
which
which
realizations the
the met
met the
the
in
GEPinin
all
all
all
satisfies
satisfies
A and GEP
GEP
information Bthe
Index
Index aloneIndex,
Index,
and
and thefrom the
differ.
45˚information
the
the line (when
information dashboard
targets
from are
from themet
the and inthe
all GEP
dashboard
dashboard Index,
is made
is made
2: Comparing dimensions)
realizations A and B alone the 45˚ line (when targets are met in all
eProtective
Protective
Protective
alone
Index the and 45˚ lineCriterion)
Criterion)
Criterion)
following
C.according
according
the creating
Aggregating
(when to
to
information principles:
targets the
aa symmetricthe
symmetricare met
from GEP+
information
the and
and all(The
convexProtective
in dashboard
convex from
order
order is made‫ظ‬‫ظ‬on the
on the
the Criterion)
dashboard
spaceԹ of
spaceԹ ௃ାଵ
௃ାଵ
and the GEPwhich
of achievements,
achievements, Index,
which satisfies
satisfies
ns)1. Pareto This
The ordering,
application oflikethe the min
Protective ordering,
Criterion compares
for the alternati
ation
tion
on
vex from
Given from
from
order
two the
C.
onthe
the
the
the
‫ظ‬creatingthedashboard
dashboard
achievement
Figure
Figure
dashboard
following
following
Aggregating
spaceԹ
B.2: the
B.2:
principles:
principles:
௃ାଵ
GEP+
profiles
Comparing
Comparing
and
ofthe and
and
x, (The
z the
inthe
the
information
achievements,
realizationsԹ GEP
GEP
GEP
Protective
௃ାଵ
realizations , ‫ݔ‬A ൒ Index,
which
and
A
Index,
‫ݖ‬Index,
fromsatisfies
implies
B
and alone
B the
Criterion)
alone‫ݔ‬the‫ظ‬ 45˚
‫ݖ‬.
the
This ordering,
dashboard
line
45˚ (when
line (when and
targets
like
targets the
are
theGEP
met
are
min ordering,
in
met all
inIndex,
all
compares alternat
regating the 1. Pareto
information from the dashboard and only taking
construction
Index,from the dashboard is made they
into
of theaccount the
GEP+ requires dimensions
us to define on the which theydd
otective
ective
‫ݔ‬tective
from
GEP
GEP
GEP Index
൒ ‫ݖ‬Index Criterion)
the
Index
implies Criterion)
Criterion)
dashboard
creating
and
and
and the
the
the
dimensions)
‫ݔ‬ the
information
A information
dimensions)
‫ظ‬ ‫ݖ‬. comparison and
GEP+
information the (The
from
from
from
between GEP the
the
the Index,
Protective
dashboard
dashboard
dashboard
the GEP Index Criterion)
isismade
is and the
made
made the GEP
only taking
information into account the dimensions on which
1.‫ظ‬Pareto
1. Pareto achievement profile for each country as follows:
ng
convex
ive the
convex
convex
2. GEP+
order
order
order
Criterion)
Independence ‫ظ‬
A Given (The
on
‫ ظ‬on onthe
comparisonthe Protective
theIdentical
spaceԹ
spaceԹ
spaceԹ
according
of betweento
௃ାଵ
௃ାଵ
௃ାଵ
of
aof Criterion)
ofachievements,
achievements,
achievements,
symmetric
Consequences
the GEPx, andIndex which
which
which
convex and satisfies
satisfies
satisfies
order
the ‫ ظ‬on This
the spaceԹ
information ordering, ௃ାଵ
from the like
of the min ordering,
achievements,
dashboard madecompares
is which satisfies alternat
sndex Let and
xand
C.
Given
andaccording
zthebe
Aggregating
C.
Given
two
Aggregating
two
two two
theto
achievement
achievement
achievement
following
achievement
a symmetric
the the
principles:
profiles
and
information
profiles
profiles information
profiles x,
x, z
withIndex
convex
z in
in
‫ݔ‬௞order
Թ
ൌ
Թ from
௃ାଵ
௃ାଵ
,
‫ݖ‬isand
,
for
‫ݔ‬
‫ݔ‬from
൒
onthe
൒ ‫ݖ‬
some
‫ݖ‬the thedashboard
implies
implies Thedashboard
The
݇ ‫א‬only
the information
spaceԹ
‫ݔ‬‫ݔ‬ ‫ظ‬
ሼͳǡ௃ାଵ
‫ظ‬ ‫ݖ‬.
‫ݖ‬.
application
application
ǥtaking
ǡof
‫ܬ‬൅
and
ͳሽ.into
Then
and
of
achievements,of the
the
the
account theGEP GEP
Protective
Protective
ifthe
‫ݖ ظ ݔ‬which
Index,
Index,
Criterion
Criterion
anddimensions
satisfies
for
forthe
on theconstruc
which constru
they d
zIndex
Index and
, ‫ ݔ‬the
൒the information
information
information fromfrom
from thethe
the dashboard
dashboard
dashboard isis made
A comparison between the GEP ௞‫ظ‬made
made This from
ordering, the
like dashboard
the min iscountry
ordering, made compares alternativ
in Թ௃ାଵ ‫ݖ‬ implies
implies
creating
creating ‫ݔ‬ ‫ظ‬
the ‫ݖ‬.
the. GEP+ GEP+ (The (The Protective
Protective Criterion) 1.
the
Criterion)theThe country’s
achievement
achievement achievement
profile
profile for
for for
each
eacheach stock
country j
as
asin the
follows:
follows:
hex
xex ‫ݔ‬only
order
arison ൌ ‫ݖ‬if
order
௞order ‫ظ‬ ‫ظݔ‬
onthe
for
on
between
௞‫ظ‬ on ‫ظ‬
some
thethe
the
according
and the information
ି௞ ି௞ . ݇ ‫א‬to
‫ݖ‬spaceԹ
following
spaceԹ
spaceԹ
the
2. GEP principles:
ሼͳǡ
௃ାଵ ǥof
௃ାଵ
௃ାଵ ǡof
of
Index‫ܬ‬achievements,
a symmetric
2. Independence
Independence ൅ ͳሽ.
and Then
achievements,
achievements,
and
from of which
the‫ݔ‬convex
‫ظ‬ if order
‫ݖ‬which
which
information and
of Identical
the satisfies
Identical satisfies
satisfies
dashboard ‫ظ‬from on thethe
Consequences dashboard
spaceԹ
is made
Consequences
௃ାଵ is made
of achievements, which satisfies
1.௃ାଵ Pareto only
dashboard takingisinto given accountby the expression the dimensions on which they di
1.The The
1.someThe country’s
country’s achievement
achievement for each stock j jinin the dashbo
g to a symmetric and convex order on the spaceԹ of ‫ݔ‬ achievements, which satisfies Then ‫ݔ‬for ‫ظ‬each stock the dashb
௃ାଵ
der equences
‫ظ‬ on the the LetLet
spaceԹ following
2.x x and
and
Independencezz
of principles:
be
be two
two
achievements, achievement‫ظ‬
achievement
of Identicalprofiles which profiles
profiles
satisfies
Consequences with
with ‫ݔ‬ ൌ ൌ ‫ݖ‬ ‫ݖ‬ for
for some application
݇ ݇ ‫א‬ ‫א‬ ሼͳǡ
ሼͳǡ ǥ ǥ of
ǡ ǡ‫ܬ‬ ‫ܬ‬ ൅the
൅ ͳሽ.
ͳሽ. Protective
Then ‫ظݔ‬ Criterion
‫ ݖݖ‬ifif andand for the constru
essx,x,x, zz z in
in in ԹԹ Թ ௃ାଵ
௃ାଵ ௃ାଵ
, , ‫ݔ‬,‫ݔ‬ ‫ݔ‬ ൒൒ ൒ ‫ݖ‬ ‫ݖ‬‫ݖ‬ implies
implies
implies
Given two‫ݔ‬‫ݔ‬ ‫ݔ‬ ‫ظ‬ ‫ظ‬ ‫ظ‬ ‫ݖ‬.
achievement
‫ݖ‬.
‫ݖ‬. x, z in Թ ௞௞
௃ାଵ
, ‫ݔ‬ ௞
൒௞
‫ݖ‬
This implies ordering, ‫ݔ‬ ‫ظ‬ ‫ݖ‬. like
wing 3. Independence
ofilesprinciples:
with ‫ݔ‬௞only
1. A Pareto
only comparison
A
ൌ ‫ݖ‬௞ififfor
of
comparison
‫ିݔݔ‬௞
Duplicated
ି௞some ‫ظ‬ ‫ିݖݖ ظ‬௞ ି௞݇
between Consequences
between the the GEP
.. ‫ א‬ሼͳǡ ǥ ǡ ‫ ܬ‬൅ ͳሽ. Then ‫ ݖ ظ ݔ‬if and GEP Index Index and and the the information
information
the achievement ௗ௄ from
ௗ௄ೕ ೕ ௧ೕ௧ೕ from the profile for each country as follows: alter
the the
dashboard min
dashboard ordering,
is made
is made compares
Let x, z be such that, for some k, m, ‫ݔ‬ ൌ ‫ݔ‬ and ௃ାଵ‫ ݖ‬ൌ ‫ ݖ‬. ThenThe ‫ݖ‬ ௃ାଵ
‫ظ ݔ‬application if and ή ή only, ,if,ififwe
of we
the
we
if ‫ݔ‬wishwish
wish ‫ ظ‬for
Protective‫ݖ‬for
for stock
stock
stock
ifdimensions ‫ܭܭ‬
Criterion ௝௝to grow,
togrow, for the
grow, and
and construct
es Given
1.according two achievement
according
Pareto to atosymmetrica symmetric profiles and x,convex
௞and z convex
in
௠ Թ order ,order
௞‫ ݔ‬൒ ‫ݖ ظ‬on ௠‫ظ‬impliesthethe
on spaceԹ
onlyThis
spaceԹ taking
‫ݖ‬. ௗ௄ of
ordering,
௃ାଵ
ௗ௄ ೕ ೕ‫ כ‬achievements,
‫כ‬
of
௄ ௄ like
ೕ ೕబachievements,
into
బ
account ି௞ the which
minି௞ which
the satisfies
ordering, satisfies compares
on which alte
the
Consequences
onsequences
onsequences
and only if Let x
2. and z
Independence
. be two achievement
of Identical profiles with
Consequences the 1. The
achievement country’s achievement
profile for each for each
country stock
as j
follows: in the dashb
oin
‫ݔ‬ in and
ԹԹԹ , ‫ݔ‬, ,‫ݔ‬௞൒
௃ାଵ ௃ାଵ
௃ାଵ ‫ݖ‬ ൌ
‫ ݔ‬൒൒ ‫ݖ‬
‫ݖ‬Giventhe‫ݔ‬
implies
‫ݖ‬3.
3. .
ି௠ Then‫ظ‬
following
the
implies
implies ‫ݖ‬
following
Independence
Independence ି௠ ‫ݔ‬
two‫ݔ‬achievement
‫ظݔ‬‫ظظ ݔ‬ ‫ظ‬‫ݖ‬.‫ݖ‬.‫ݖ‬ if
principles:
‫ݖ‬. of and
principles: only
profiles ି௞
of Duplicated
Duplicated if ‫ݔ‬ ‫ظ‬ ‫ݖ‬
x,Consequences
z in ି௞ if
Թ , ‫ ݔ‬൒ ‫ ݖ‬implies only
Consequences
௃ାଵ This ordering,
‫ظ ݔ‬ ‫ݖ‬.taking ௗ௄ௗ௄ೕ into ௄ ௄like
బ బ
the
account min ordering,
the dimensions compares alterna
on which th
‫ כ‬ή ή ,݇,ifif we wish
wishfor stock ௝ to toshrink
௠ ௠ ೕ ೕ
nt tntprofiles
profileswith
profiles with
with ‫ݔݔ‬௞‫ݔ‬௞௞ൌ ൌൌ ‫ݖݖ‬௞‫ݖ‬௞௞for
Let for
xfor
for some
and some
some
some ݇݇݇two
zԹIdentical
be ‫אאא‬ሼͳǡ ሼͳǡ
ሼͳǡ ǥǡǡ‫ܬ‬ǡ‫ܬܬ‬൅
achievement ൅൅ͳሽ. ͳሽ. Then
Then
. Then
Then ‫ظݔݔݔ‬
profiles ‫ݖݖظظ‬with‫ݖ‬ifififand and
and and ‫ݖ‬௞ for ௗ௄ some
ೕ
we for stock ‫ܭܭ‬ shrink
2. Independence of ǥǥConsequences ͳሽ. ‫ݔ‬௞ ൌ ௗ௄ ೕ ೕೕ‫ כ‬௧ೕ௧ ೕ ‫ א‬ሼͳǡ ǥ ǡ ‫ ܬ‬൅ ͳሽ. Then ‫ݔ‬௝ ‫ ݖ ظ‬if and
o achievement profiles x, z in , implies and‫ݖݖ‬only taking into ‫ظ‬బaccount the forifdimensions on for which the they
௃ାଵ
ାଵ
, ‫ ݔ‬൒ ‫ ݖ‬implies
nsequences LetLet ‫ݔ‬x,x,‫ظ‬zz‫ݖ‬. be such
be such that, that, for ‫ݔ‬ ൒
for some‫ݖ‬
some k,k, m, ‫ݔ‬ ‫ظ‬
m,‫ݔݔ‬௞௞ ൌ ‫ݖ‬. ൌ ‫ݔݔ‬௠௠ and ௞௞ ൌ ൌ1. The
‫ݖݖ‬௠ Thencountry’s
௠..application
Then ௗ௄
‫ݔכݔ‬ή ‫ظ‬ ௧ ೕ achievement
‫ݖ‬,‫ݖ‬ifififwe and
and wish only
only for
ifstock
‫ିݔݔ‬௞ each
ି௞ ‫ظ‬
stock
‫ݖݖܭ‬௝ି௞
‫ظ‬Criterion toififgrow, j in the dashbo
and
Define the
Let x and
difference
and only
only
z be if ‫ݔ‬
set
two ‫ظ‬
to ‫ݖ‬ be
achievement . equal to the
profiles set
with of dimensions for
The
some ࣞሺ‫ݔ‬ǡ ‫ݖ‬ሻ ௗ௄on ೕ ǡ ‫ ܬ‬of
௄which thethe Protective
Then index- if and
ି௞ cons
equences
uences
quences 2.
m, ‫ݔ‬௞ ൌ ‫ݔ‬௠ and ‫ݖ‬௞ ൌ ‫ݖ‬௠ .ି௠ 1.
andand Pareto
1.
IndependencePareto
only
only ifif ‫ݔ‬‫ݔ‬ ି௞
Then ‫ିݔ‬௠
‫ظ‬of
‫ظ‬ ‫ݖ‬‫ݖ‬Identical
ି௞
. . Consequences
‫ ݖ ظ‬if and only if ‫ିݔ‬௞ ‫ିݖ ظ‬௞ if ‫ݔ‬ ௞ ൌ ‫ݖ‬ ௞ ݇ ‫א‬ ሼͳǡ
ೕ ǥ ൅ ͳሽ. ‫ݔ‬ ‫ظ‬ ‫ݖ‬
the
The
‫ݔ‬achievement
first ௗ௄
term ௧ೕ௄
,ೕin theprofile computation for each ofcountry
aCriterion
acountry’s as and follows:
achieveme
బ
esdashboard
the set of only ‫ݖ‬profiles xtwo and
ି௠
.݇z‫א‬ differ,
ି௠
ǡ that is: The
some ‫ݔ‬application
first
ௗ௄ೕ
term
ήೕ ή బǥ if,,in ‫ܬ‬we
ǡ ifif ofͳሽ.
the
we the
wishcomputation
wish Protective
for
forforstock
stock of௝‫ܭ‬
‫ܭ‬ to grow,
country’s
shrink. for the con
achievem
ofiles
files
endence
ces withwith
with ‫ݔ‬of௞‫ݔݔ‬ ൌ௞dimensions
Given
Identical
௞Let ൌ ൌ ‫ݖ‬if
௞x Given
‫ݖ‬௞for
and
௞ two
‫ݔ‬for
for
ି௞some
zsome
‫ظ‬ someachievement
‫ࣞݖ‬ሺ‫ݔ‬ǡ
Consequences
be ି௞݇achievement
two ݇‫א‬ ‫ݖ‬ሻ
ሼͳǡ on
ሼͳǡ
ሼͳǡ
‫א‬achievement
ǥǥǥ ‫ܬ‬which
ǡprofiles
ǡ൅
‫ܬܬ‬൅൅profiles
ͳሽ.ͳሽ.
ͳሽ.the
Then x, zindex-
Then
Then
profilesx,‫ݔ‬inz‫ظݔ‬‫ݔ‬Թin‫ݖ‬௃ାଵ
with
‫ظ‬‫ظ‬ Թ ,and
‫ݖ‬if‫ݖ‬௃ାଵ
ifif‫ݔ‬and,൒‫ݔ‬ൌ‫ݖ‬൒
௞and ‫ݖ‬implies
௞‫ݖ‬for implies ‫ظ‬ ݇‫ݖ‬.
‫אظ‬
ௗ௄ ‫ݖכ‬. ሼͳǡ
ೕ ೕ‫ כ‬௄ೕ௧ೕ
൅we wish Then ‫ݔ‬ stock
‫ ݖ ظ‬if ௝ to shrink
and
dConsequences
Consequences3. Independence
Consequences 3. Independence of
of Duplicated Duplicated Consequences
Consequences The application
progress
the
progress
ௗ௄
achievement as as of
calculated
calculated the Protective
profile for in in each the
the Criterion
dashboard.
country
dashboard. for the
The
as follows:
The constru
second
secon
dwith z be‫ݔ‬௞two ൌ ‫ݖ‬achievement
only Define
Define
for if ‫ିݔ‬௞
some the
the ‫ظ‬݇ profiles
‫א‬ ି௞ሼͳǡ. ǥ ǡ ‫ܬ‬with
‫ݖ‬difference
difference ൅ set
ͳሽ. ‫ݔ‬௞Then
set ൌ
toto ‫ݖ‬௞bebe
‫ݔ‬for‫ظ‬ some
equal
equal
‫ݖ‬ if andto ݇ ‫א‬the
to theሼͳǡ set ǥset ǡ ‫ ܬ‬of
൅of 1.ͳሽ. The Then
dimensions
dimensions country’s
ௗ௄‫ݔ‬ ೕ ‫ظ‬ ௄ೕబ‫ݖ‬ ࣞሺ‫ݔ‬ǡ
ࣞሺ‫ݔ‬ǡ if and
achievement
‫ݖ‬ሻ
‫ݖ‬ሻ on
on which
which for
the
the each
index-
index- stock j in the das
meee k,k,k,m, m,m,‫ݔݔ‬௞‫ݔ‬௞௞ൌ ൌ௞ൌ
3. ‫ݔ‬ ‫ݔ‬ ‫ݔ‬ and
and
and
Independence Let
‫ݖ‬‫ݖ‬ ‫ݖ‬ ൌx,
ൌ ൌ z
‫ݖ‬‫ݖ‬ ‫ݖ‬ be..
of . Then
Then
Then such ࣞሺ‫ݔ‬ǡ
Duplicated ‫ݔ‬‫ݔ‬ ‫ݔ‬ ‫ظ‬
that,
‫ظ‬ ‫ظ‬ ‫ݖ‬‫ݖ‬ ‫ݖ‬ if
if if
for
ሼ݇ and
and
and
‫ݖ‬ሻ ൌConsequencessome
‫א‬ ሼͳǡonly
only
only k, if
if if
m,
‫ݔ‬
ǥ ǡ ‫ ܬ‬൅ ͳሽǣ‫ݔ‬ ‫ݔ‬ ‫ݔ‬ ‫ظ‬‫ظ‬ ‫ظ‬
ൌ
௞ ‫ݔ‬௞ ് ‫ݖ‬ ‫ݖ‬ ‫ݖ‬‫ݔ‬ the
௠ ‫ݖ‬௞if
if if
and ሽachievement
achievement
‫ݖ‬ 
achievement ൌ ‫ݖ‬ . ή Then ,
is if
is we
the
profile ‫ݔ‬
the ‫ظ‬ wish
importance
‫ݖ‬ for
if
importance for
each
and stock
only of
country
of ‫ܭ‬
ifthat
‫ݔ‬ to
that shrink
progress,
as‫ظ‬ ‫ݖ‬follows:
progress, if which
which isis
equal
௞ ‫ظ‬ ‫ݖ‬ ି௞ .
to thedashboard 2. ௠ ௠ ௠Independence
2.
dashboard Independence
set of dimensions
௞௞ ௞ ௠௠ ௠
profilesxxand
profiles of Identical
of Identical
andzz‫ݖ‬ሻdiffer,
ࣞሺ‫ݔ‬ǡ differ,
on whichConsequences
that is:
that is:the index-
ି௞
ି௞ ି௞ ି௞ି௞ି௞
The
1.
௞
The firstௗ௄ ‫כ‬
௠ ௧ೕ
ೕterm in the
country’s computation
achievement for
௝
ofeach
ି௞
aweights
country’s
ି௞
stock achievem
j in the da
and employed ௗ௄ in the
௧ determination of the of the GEP.
zonly ifof‫ݔ‬two ‫ظ‬for‫ݖ‬some . Consequences employed in ‫ݖ‬inthe ,the ifandǡ determination
‫ܬ‬we wish ifinfor of
stock the ifweights
௝if to grow, of andthe GEP.
ೕ ೕ
equences
‫א‬ ሼͳǡ ǥ ǡ ‫ ܬ‬൅ Let
sequences
sequences 3.Let
ͳሽǣ ‫ݔ‬x,௞Let
xz്and
Independence be
x, ሽsuch
x‫ݖ‬௞zand be be z be
such that,
two that, for
achievement
achievement
Duplicated
ି௠ some
ି௠ k, profiles
m,profiles
m, ‫ݔ‬௞ ൌ with ‫ݔ‬௠with and
and
‫ݔ‬௞ ‫ݔ‬ൌ ‫ݖ‬௞௞‫ݖ‬ൌ 1.for
ൌ ௞The
‫ݖݖ‬The .some
for Then
first
some
country’s ݇term
‫ظכא ݔ‬ ήሼͳǡ
݇as ‫א‬ ifሼͳǡ
ǥ ǥ ǡ ‫ܬ‬only
computation
൅ ͳሽ.
൅ ͳሽ. Then Then
‫ݔ‬for ‫ݔ‬of ‫ظ‬ if‫ܭݖ‬
‫ݔ‬a dashboard.
‫ݖ‬country’s
‫ିظ‬௞ and and
hat is: progress ௄ೕబ௧achievement
calculated the each stock j inThe the dashb
secon
௞ ௠ ି௞
The following and onlyis a ifconsequence . of Theorem 3 in Barberà and Jackson for
The for each
each
(JET,
first
ௗ௄ stock
stock
term
ௗ௄ 1988).
ೕ
in j jin
the in the
the dashboard
dashboard
computation is
ofis given
a given
country’s byby the
the expression
expressio
achieveme
m, m,
endence
‫ݔ‬
nces ‫ݔ‬‫ݔ‬ ൌ ൌ ൌ
‫ݔ‬ ‫ݔ‬ ‫ݔ‬ ofand and only
Duplicated
Let ‫ݖ‬ x,
‫ݖ‬only
‫ݖ‬ൌ if
z ൌൌ
‫ݖ‬ ‫ݔ‬if
be‫ݖ‬‫ݖ‬.
‫ݔ‬ ‫ݔ‬ Then
. . Then
Then
‫ظ‬
Consequences
‫ظ‬
such Then
‫ݖ‬ ‫ݖ‬
‫ݔ‬ .
that, ‫ݔ‬‫ظ‬
‫ݔ‬ .‫ظ‬‫ظ‬‫ݖ‬ ‫ݖ‬if
for‫ݖ‬ if
ifif
andand
and
and
some onlyonly
only
onlyk, if
if
m, if
if
‫ݔ‬ ‫ݔ‬‫ݔ‬ ‫ݔ‬ ‫ظ‬
ൌ
ି௞ሼ݇ ‫ظ‬ ‫ظ‬
‫ݔ‬
ሼ݇௠‫ݖ‬ ‫ݖ‬ ‫ݖ‬ andif if
ሼͳǡǥ
ሼͳǡ if
and ‫ݖ‬ ൌ ‫ݖ‬ achievement
. Then
‫ݔݔ‬௞௞ௗ௄
achievement ‫ݔ‬ in
ሽ ή is, the
‫ظ‬ೕ
‫ݖ‬ stock
if
ೕ
and
if we j isonlythe
wish for
importance country’s
if ‫ݔ‬ ି௞ stock
‫ظ‬ ‫ݖ‬degree
of that if
ି௞ ‫ܭ‬௝progress,to grow, and which is
௞ ௞௞ ௠௠ ௠ ௞ ௞௞ ௠ ି௞௠ ௠ି௠ ି௞ ି௞ ି௠ ି௞ ࣞሺ‫ݔ‬ǡ
ࣞሺ‫ݔ‬ǡ ‫ݖ‬ሻ‫ݖ‬ሻ
ି௞ ൌൌ
௞ି௞ ‫א‬
ି௞ ‫א‬the
ି௞
ି௞ ǥǡ௞ǡ‫ܬܬ‬൅ ൅ͳሽǣ ͳሽǣ ്്ೕௗ௄‫ݖݖ‬ௗ௄௞௞ሽ
௠ బ
be bebeinequal
3 equal
equal
Barberà to
to toand the theJackson
the set
set
setDefine ofofofdimensions
dimensions
dimensions
the‫ ݔݖ‬difference
(JET, 1988). ࣞሺ‫ݔ‬ǡ‫ݖ‬ሻ
ࣞሺ‫ݔ‬ǡ
ࣞሺ‫ݔ‬ǡ ‫ݖ‬ሻ‫ݖ‬ሻ
set on
on
on which
which
to‫ݖ‬whichbe equal the
the to index-
index-
index- the set
progress  of ifdimensionsasೕ௧ೕೕ‫ כ‬௄ೕ௄ೕబ
ή calculated
, if we ࣞሺ‫ݔ‬ǡ wish ‫ݖ‬ሻ in on
forthe whichdashboard.
stock theto
‫ܭ‬ index-
shrink The second
beൌ such
ൌ and
ǡ ‫ݖ‬ሻ‫ݔ‬that ሼ݇ that, and foronly
only
‫ݖא‬௞ሼͳǡൌǥ‫ݖ‬ǡ௠‫ܬ‬.൅dashboard some if
if
Then ‫ݔ‬
ͳሽǣ ି௠ ‫ݔݔ‬ k, m,
‫ظ‬
‫ݖ ظ‬௞ି௠ ൌ. ‫ݔ‬ and
ifሽ and only if ‫ିݔ‬௞ ‫ିݖ ظ‬௞ if ‫ݖ‬ ൌ . Then ‫ݔ‬ ‫ظ‬ ‫ݖ‬ if and of only 
progress
employed ௗ௄ೕௗ௄ ή
‫ݔ‬
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ሺ݆ሻas ,
in the
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ሺ݆ሻ ‫ظ‬if we
‫ݖ‬
calculated wish
if
బ determination ή in forthe
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݆ሻ.
ή ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݆ሻ. stock
dashboard. ‫ܭ‬ to grow,
of the weights of the GEP.
௝ and
௞ ് ௞ ௠ ௞ ௠ ‫כ‬ ‫כ‬
ି௞ బ ି௞ ௝
er,fer,
er, that
that
௠ is:
is:
is: profiles x and z differ, that is: achievement ௗ௄ ௄ೕ ೕೕ ௧ೕ௄ೕ
is (JET, the importance theof that
if ‫ିݔ‬௠ ‫ିݖ ظ‬௠ Define3.
.The
The the difference
Independence
3.following
Independence
following isisaaof set
Duplicated
of
consequence
consequence to be Consequences
Duplicated equal
ofConsequences
of Theorem
Theorem to the33in set
in Barberà of dimensions
Barberà The and
and
for ௗ௄
second
each Jackson
Jackson stock ࣞሺ‫ݔ‬ǡ
‫ כ‬ή ௧(JET,
௄ೕబterm j‫ݖ‬ሻ
,inin on
ifthe we
the which
wish for
computation
1988).
1988). dashboard index-
stock
of such
is given ‫ܭ‬progress,
௝ to byshrink
which is
the expressio
ೕ ௗ௄
ual
qual
qual Theorem
toto tothe the
the (The
dashboard
set Let
Defineset
set Let
of Protective
x,
ofofthe zprofiles
zx,dimensions
be
dimensions be
dimensions such
difference Criterion)
such xࣞሺ‫ݔ‬ǡ
and
that, that,
ࣞሺ‫ݔ‬ǡset
ࣞሺ‫ݔ‬ǡ z ‫ݖ‬ሻ
for
‫ݖ‬ሻ differ,
some
for
on
to
‫ݖ‬ሻ onon somethat
which
be k, m,
which
which
equal is:
k, the m,
the
‫ݔ‬the
to‫ݔ‬ൌindex-
index-
index-
‫ݔ‬ൌ௠of
the ‫ݔ‬and
௠ and
set ‫ݖ‬of
௞ ൌ
employed
‫ݖ‬ൌ௠ .‫ݖ‬first
‫ݖ‬dimensions Then. ‫כ‬Then in
ήterm ೕ the
,‫ظ‬if‫ݖ ݔ‬in
‫ࣞݔ‬ሺ‫ݔ‬ǡ ೕ
we
‫ݖظ‬ሻdetermination
wish
if‫ݖ‬importance
and
if and
on whichfor
only only stock of
ifofthe
‫ݔ‬if the‫ظ‬
ି௞‫ݔ‬index-
weights
toa‫ିݖ‬௞
‫ܭ‬progress,
௝‫ݖ‬given
‫ିظ‬௞ shrink
if if of the GEP. P
Theorem 3 in Define the difference set to be equal to the ௞set ௞achievement ௧ೕ is the that
‫א‬Barberà and Jackson (JET, 1988). The
2. The ௗ௄country’s GEP the computation
achievement is of country’s
by the achiev
expressio
௞ ௠ ି௞
ࣞሺ‫ݔ‬ǡࣞሺ‫ݔ‬ǡ‫ݖ‬ሻ
ࣞሺ‫ݔ‬ǡ ‫ݖ‬ሻ‫ݖ‬ሻൌ ൌ ൌሼ݇ሼ݇ሼ݇ ‫א‬ ‫א‬ ሼͳǡ
ሼͳǡ
ሼͳǡ ǥǥ ǥǡǡ‫ܬ‬ǡ‫ܬܬ‬൅ ൅ ൅ͳሽǣ ͳሽǣ
ͳሽǣ ‫ݔ‬‫ݔ‬ ‫ݔ‬ ് ് ് ‫ݖ‬ ‫ݖ‬ ‫ݖ‬ ሽ ሽ ሽ ࣞሺ‫ݔ‬ǡ ‫ݖ‬ሻ ൌ ሼ݇ ‫א‬ ሼͳǡ ǥ 2.ǡ
for each‫ܬ‬ The൅ ͳሽǣcountry’s
ೕ ‫ݔ‬
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ሺ݆ሻ
stock ് ‫ݖ‬
jindex-
in ሽ GEP ή achievement
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݆ሻ.
௞ the dashboard is given by the expression is given by the expressi
at atis:is:
is: ofand and onlytoonly if ‫ݔ‬ifି௠ ‫ݖ‬௞.ି௠ z.which
௞ ௞
he
to difference
the set dashboard set
dimensions
dimensions be
profiles equal
‫ࣞݔ‬ሺ‫ݔ‬ǡ‫ظ‬
௞ ௞ ‫ظݖ‬to
xି௠ and௞
on the set the
differ,
which of
that
the dimensions
is:
index-
index-dashboard ࣞሺ‫ݔ‬ǡ ‫ݖ‬ሻ on which which the
is calculated
௞
Let ‫ظ‬ be a symmetric and
ି௠
convex
‫ݖ‬ሻ
order ࣞሺ‫ݔ‬ǡon Թ
‫ݖ‬ሻ
௃ାଵ
ൌ that
ሼ݇ ‫א‬ satisfies
ሼͳǡ ǥ ǡ ‫ܬ‬ ൅
progress
Pareto,
ͳሽǣ
The
‫ݔ‬ ് first
Independence
‫ݖ‬ ሽ
‫ܲܧܩ‬
as തതതതതതതതതതതതതതതത
term ή ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ inusing
calculated the the same
of computation
Identical
,
in the
where
principles
dashboard.
of a country’s Theachie sec
rd profiles ௉ x and Theorem
Theorem z differ, (The
(The that is:
Protective
Protective Criterion)
Criterion)  ‫ܲܧܩ‬ ή തതതതതതതതതതതതതതതത
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ , where
eceeԹof ofofTheorem
Theorem
Theorem 333in profiles
ininBarberà
Barberà
Barberà
The x and z
andJackson
and
and
following differ,
Jackson
Jackson that is:
(JET,
(JET,
(JET, 1988).
1988).
1988). The first
௞
employed
achievement
progress term
௞ in
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ሺ݆ሻ in
the
asࣞሺ‫ݔ‬ǡ the
determination
iscalculated computation
ή
the (JET, ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݆ሻ.
importance of
in thetheof a
weights country’s
of dashboard. of
that progress, the achievem
The whicse
‫ݖ‬ሽ௞௞ሽሽ is aset consequence of Theorem 3setin Barberà and
ሽ Jackson
ଵଵ ூ 1988).
‫ݖ‬ሻ‫ݖ‬ሻൌConsequences
௃ାଵ
ൌൌሼ݇ ሼ݇ that
ሼ݇
‫אא‬ሼͳǡ‫א‬ሼͳǡ satisfies
ሼͳǡ
ǥǥ Define ‫ܬ‬Define
ǡǥ‫ܬ‬ǡ ǡ൅ ‫ܬ‬൅൅ and
ͳሽǣͳሽǣ‫ݔ‬௞Independence
Pareto,
the
ͳሽǣ ‫ݔݔ‬the
௞് ്്‫ݖ‬௞Independence
௞difference
‫ݖ‬difference of Duplicated
set of
toࣞሺ‫ݔ‬ǡ be
to Identical
beequal
‫ݖ‬ሻ ൌ Consequences.
equalሼ݇to ‫ א‬to theǥ
ሼͳǡ the set
ǡ‫ ܬ‬൅ of Then
ͳሽǣ dimensions
of
2.‫ݔ‬௞The dimensions
്Put തതതതതതതതതതതതതതതത
௞‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬
‫ݖ‬country’s
തതതതതതതതതതതതതതതത
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ ࣞሺ‫ݔ‬ǡ GEP‫ݖ‬ሻ ൌ on
‫ݖ‬ሻ on
σwhich
ൌachievement
σ ூ which the
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݅ሻ the
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݅ሻ index-
is index-by the expressi
given
progressGEP.
employed
achievement as
simply, calculated
in(JET, the the country’s
is determination
the in௜ୀଵ
importance
ூ ூ ௜ୀଵ the
achievement dashboard.
ofof thethat for each
weights The
progress, of the secon G
whi
ated݇ ‫ א‬ሼͳǡ ǥ ǡ ‫ܬ‬The
Consequences. ൅Let Let
ͳሽǣfollowing
dashboarddashboard
‫ظ‬‫ظݔ‬௞௉௉ ്beThen
be ‫ࣞݖ‬ሺ‫ݔ‬ǡ is
a௞aprofilesa consequence
profiles
‫ݖ‬ሻ
symmetric
ሽsymmetric ൌ ሼ݇ x and xand
‫א‬ and
ሼͳǡ
and zǥ ‫ܬ‬of൅Theorem
differ,
zǡ differ,
convex
convex that ‫ݔ‬௞is:്is:
that
ͳሽǣorder
order on ‫ݖ‬3௞ ሽin
on ԹԹ௃ାଵ Barberà
௃ାଵ
that
that and
satisfies
satisfies
achievement
stock
Jackson
 j in Pareto,
Pareto,
the is the
1988).
Independence
Independence
തതതതതതതതതതതതതതതത
dashboard importance is given
, where by of
ofoftheIdentical
Identical
that
expression progress, which i
Theorem
eorem
heorem 3 3in3Թ inin
Barberà
The Barberà
Barberà
following andand
and‫ ݔ‬Jackson
‫ظ‬Jackson ‫ ݖ‬consequence
Jackson
is a if(JET,
(JET, and1988).
(JET, only
1988).
1988). if Theorem
of ‹௞‫ࣞא‬ሺ௫ǡ௭ሻ 3 in Barberà ‫ݔ‬௞ ൐ ‹ for
2.and The each
employed Jackson stock
‫ܲܧܩ‬
country’s
‫ݖ‬௞ in(JET, ή j
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬
GEP
the 1988). in the dashboard
achievement
determination of the weights of the G is is given
given by bythe the expres
expressio
order
rion) on ௃ାଵ Consequences
Consequences
that satisfies ௉ and
and
Pareto, Independence
IndependenceIndependence ofofDuplicated
Duplicated
of Identical Consequences.
Consequences. The
௞‫ࣞא‬ሺ௫ǡ௭ሻ
achievement Then profile
Then ଵ ூ
for each country is then the (J+1)
‫ܬ‬employed in ሽthe ሽ determination of the isweights byofthe theexpre GEP
on) ‹
on) Theorem (The തതതതതതതതതതതതതതതത
wing is Barberà
a consequence ‫ݔ‬and
௞ ൐ Jackson
‹of Theorem ‫ݖ‬௞ Protective
3 in Barberà Criterion)and The
for achievement
‫ݔ‬௞each ‫݁ܿ݊ܽݐݎ݋݌݉ܫݖ‬
്௞stock j in profile
the σfor
ൌ dashboard ௜ୀଵeach country
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݅ሻ given is then (J+1
32‫ݖ‬ሻJackson (JET, ǡ 1988).
emof 3 in
Duplicated ௞‫ࣞא‬ሺ௫ǡ௭ሻ Consequences. ௞‫ࣞא‬ሺ௫ǡ௭ሻ (JET,
Then 1988). ࣞሺ‫ݔ‬ǡ ࣞሺ‫ݔ‬ǡ ‫ݖ‬ሻ ൌ ሼ݇
ൌ ሼ݇ ‫ א‬ሼͳǡ ‫ א‬ሼͳǡ ǥ ǡǥ ൅ ‫ͳܬ‬ሽǣ൅ ͳሽǣ ‫്ݔ‬௞‫ܲܧܩ‬ ‫ݖ‬ή௞തതതതതതതതതതതതതതതത
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ሺ݆ሻ ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ ήfor ூ , where
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݆ሻ.
Theorem (The Protective for eachand
and the the achievement
stock achievement j in the dashboard each
ଵforூ each ofis of thethe stocks
given stocks in
by the the
in the dashboa
expressio
dashboa
nvex 32 orderon onԹ ௃ାଵ
that‫ظ‬ satisfies ‫ظ‬‫ظ‬Criterion)
‫ݔݔ‬Pareto, ififand
௉௉ ‫ ݖݖ‬Independence andonly
Independence only ififof Identical ‹௞‫ࣞא‬ሺ௫ǡ௭ሻ
‹ ‫ݔݔ‬௞௞ ൐ തതതതതതതതതതതതതതതത
൐ ‹௞‫ࣞא‬ሺ௫ǡ௭ሻ
‹
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ሺ݆ሻ ௞‫ࣞא‬ሺ௫ǡ௭ሻൌ ‫ݖݖ‬௞௞ ήσ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݆ሻ.
௜ୀଵ ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݅ሻ
nvexvex order
order on The Թ Թ ௃ାଵ
௃ାଵ
The that
that
Let
following
following satisfies
satisfies
be is
is a‹ a a Pareto,
Pareto,
symmetric
consequence
isconsequence Independence
and of convex
Theorem ofof
order 3 Identical
Identical
in on Թ ௞‫ࣞא‬ሺ௫ǡ௭ሻ
௃ାଵ
that satisfies Pareto, ூIndependence of Identical
if Theorem The following
(The‫ ݔ‬Protective a consequence
Criterion) ‫ݖ‬௞ of Theorem of Theorem 3 in3 Barberàin Barberà and and JacksonJackson (JET, (JET, 1988). 1988).
௉
nd The achievement profile for each country is then the (J+1
ence
nceceonly of
of ofDuplicated
Duplicated
Duplicated Let
‹
‫ظ‬ Barberà
beConsequences.
Consequences.
Consequences.
Consequences
௞‫ࣞא‬ሺ௫ǡ௭ሻ
a and ௞ ൐
symmetric Jackson Then
Then
Then
and
௞‫ࣞא‬ሺ௫ǡ௭ሻ
and (JET, Independence
1988).
convex order on of Թ Duplicated
௃ାଵ 32 32
that Consequences.
2.2.
satisfies The
The
ܲ‫ݏݏ݁ݎ݃݋ݎ‬ሺ݆ሻ
country’s
country’s
Pareto, GEPThen
Independence GEP
ή achievement
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݆ሻ.
achievement of of is given
Identical is given
by the by the expre
(The Protective ௉Criterion) and the achievement for each the stocks in the dashboa
To
The To understand
expression understand
achievement howhow profile one
one could
for could each use
usecountrythe
the information
information
is thenbythe ininthe
theaa
(J+1)-
order
der
order onon on ԹԹԹConsequences
௃ାଵ ௃ାଵ
௃ାଵ
Let 32
that that
that
‫ظ‬௉satisfies
satisfies
besatisfies
a symmetric and
Pareto,
Pareto,
Pareto, Independence
Independence
and convexoforder
Independence
Independence Duplicated
ofof of
Identical Թ Consequences.
Identical
onIdentical ௃ାଵ
that satisfies 2. ThePareto, country’s
Then Independence
തതതതതതതതതതതതതതതത GEP achievement of Identical is given the expr
of fif
if andandonly
and
eDuplicated
fon Duplicated
aDuplicated
symmetric
onlyififif Theorem
only Theorem
Consequences.
and
Consequences
‹
‹
‹
Theorem
Consequences.
Consequences. convex (The
௞‫ࣞא‬ሺ௫ǡ௭ሻ
௞‫ࣞא‬ሺ௫ǡ௭ሻ
௞‫ࣞא‬ሺ௫ǡ௭ሻ (The
(The
‫ݔ‬Protective
‫ݔݔ‬Then
Protective
Then
order
and ௞ ௞൐൐൐‹
௞Protective
Then
‹௞‫ࣞא‬ሺ௫ǡ௭ሻ
‹
on Թ௞‫ࣞא‬ሺ௫ǡ௭ሻ
Independence ௉ ‫ݖݖ‬௞‫ݖ‬
‫ ݔ‬Criterion)
‫ظ‬
Criterion)
Criterion)
௞‫ࣞא‬ሺ௫ǡ௭ሻ
௃ାଵ
that ௞ ௞
satisfies
of if andPareto,
Duplicated onlyConsequences.
if Independence
2. ‹
calculate
and
The
 achievement
calculate
the
country’s
௞‫ࣞא‬ሺ௫ǡ௭ሻ
‫ܲܧܩ‬
Thenof‫ݔ‬the the
௞ ൐
Identical
GEP+,
ή ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬
GEP‹GEP+, for
௞‫ࣞא‬ሺ௫ǡ௭ሻ
consider
consider
achievement each ‫ݖ‬, where of thethe
theisstocks
case
case by
given
where
where
in the ‫ݔݔ‬and
thedashboar and‫ݖݖ‬a
express
Թ ௃ାଵ
that satisfies Pareto, Independence of Identical ଵ ூ௞
countries.  ‹ Then
തതതതതതതതതതതതതതതത it can
തതതതതതതതതതതതതതതത be argued
,, where
where that
that if ‫ ݔ‬is in a more fa
if ‫ݔ‬ is in a more
ences
icated and Independence
Consequences. 32
32 32 Then of Duplicated ‫ظ ݔ‬௉ ‫ ݖ‬Consequences. if and only if Then ‹௞‫ࣞא‬ሺ௫ǡ௭ሻ To
32case ‫ݔ‬௞understand
countries.
൐ ‫ܲܧܩ‬
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ Then ή ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬
how
it‫ ݖ‬can ൌ be
௞ oneூ ଵcould
σ௜ୀଵ argued ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݅ሻ
use the‫ ݔ‬information in the
fa
a

case that
‫ܲܧܩ‬ that ή തതതതതതതതതതതതതതതത
௞‫ࣞא‬ሺ௫ǡ௭ሻ
the
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬
തതതതതതതതതതതതതതതത
the worst
worst , where
achievement
achievement ூ inin ‫ݔ‬ isis greater
greater than
than the
th
donly
donly
only if ifif ‹ ‹Let
‹ Let
‫ظ‬ ‫ظ‬be
௉ ௉‫ݔ‬௞‫ݔ‬ a
be
be
‫ݔ‬௞൐ symmetric
௞൐a ൐ symmetric
symmetric
‹ ‹
‹ and
‫ظ‬௉ ‫ݖ ݖ‬and
‫ݔ‬௞‫ࣞא‬ሺ௫ǡ௭ሻ and convexconvex
௞‫ݖݖ‬௞௞convex
if and only order order
orderif on on
Թ ௃ାଵ௃ାଵ
‹Թ that thatsatisfies
calculate
‫ݔ‬௞ ൐ ‹
satisfies ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬
Pareto, Pareto,
the GEP+, Independence
‫ݖ‬௞toଵthe ൌ
Independenceconsider σ ௜ୀଵ ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݅ሻ
of Identical
of Identical
the case where ‫ ݔ‬howand ‫ݖ‬th
௞‫ࣞא‬ሺ௫ǡ௭ሻ
௞‫ࣞא‬ሺ௫ǡ௭ሻ
௞‫ࣞא‬ሺ௫ǡ௭ሻ ௞‫ࣞא‬ሺ௫ǡ௭ሻ
௞‫ࣞא‬ሺ௫ǡ௭ሻ
principle of ூ ூWorst Achievement
32 ௞‫ࣞא‬ሺ௫ǡ௭ሻ ofPriority
തതതതതതതതതതതതതതതത and
andis
௞‫ࣞא‬ሺ௫ǡ௭ሻ
Consequences
Consequences if and and
only and Independence
if Independence of Duplicated
of Duplicated Consequences. To  understand
principle
Consequences. ‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ Then Priority
Then howൌone to σ the ௜ୀଵWorst
could useAchievement
‫݁ܿ݊ܽݐݎ݋݌݉ܫ‬ሺ݅ሻ the information is the
in howac t
if ‹32 ‫ݔ‬ ‫ظ‬ that
‫ݖ‬ satisfies Pareto, Independence
‹ of ‫ݔ‬ Identical
൐ ‹ ‫ݖ‬ The countries.
achievement Then it can
profile ூ be for arguedeach that
country if ‫ݔ‬ is
is inthen a more
the fa
(J
32 ௉ ‫ݔ‬௞ ൐ ‹௞‫ࣞא‬ሺ௫ǡ௭ሻ ‫ݖ‬௞
32
௞‫ࣞא‬ሺ௫ǡ௭ሻ ௞‫ࣞא‬ሺ௫ǡ௭ሻ ௞ ௞‫ࣞא‬ሺ௫ǡ௭ሻ ௞
32 constructed.
constructed.
calculate the GEP+, consider the case where ‫ ݔ‬and ‫ ݖ‬a
and case the thatachievementthe worst achievement for each ofthat the in ‫ݔ‬stocks is greater than th
32 32 The
countries. achievement Then it can profile be argued for each country
if ‫ ݔ‬is in isinathenthe
more dash
thefav(
‫ݔظ ݔ‬௉‫ݖظ‬௉ ‫ ݖ‬if and if and only only if if ‹‹ ‫ݔ‬
principle ‫ݔ‬൐ ‹
൐
௞ ௞ of ௞‫ࣞא‬ሺ௫ǡ௭ሻ ‹Priority ‫ݖ‬ to
௞ the‫ݖ‬ ௞ Worst Achievement andthe is how t
The
௞‫ࣞא‬ሺ௫ǡ௭ሻ
case achievement
௞‫ࣞא‬ሺ௫ǡ௭ሻ
and the achievement
that the worst profile
௞‫ࣞא‬ሺ௫ǡ௭ሻ for
for each
achievement each country ofinthe ‫ ݔ‬isstocks is then
greater inthanthe (J+1das
the
and constructed.
the achievement for each of the stocks in the dashbo
32 32 principle of Priority to the Worst Achievement and is how th
To understand how one could use the information in th
constructed.
calculate the GEP+,
To understand how oneconsider the the
could use case where ‫ ݔ‬and
information in
Tocountries.
understand how
calculate Then one could use
it can consider
the GEP+, the
be arguedthe information
thatcase
if ‫ ݔ‬iswherein the
in a more
‫ ݔ‬an
calculate
case the
that GEP+,
the worstconsider the
achievement case
in where
is ‫ݔ‬
greater
countries. Then it can be argued that if ‫ ݔ‬is in a mo
‫ݔ‬ and
than‫ݖ‬
GEP+ (The
e GEP+ Protective
(The Protective
1. Pareto
the following
according Criterion)
Criterion)
principles:
to a symmetric and convex order ‫ ظ‬on the spaceԹ௃ାଵof achievements, which satisfies
Given two
1. Pareto the following achievement profiles x, z in Թ௃ାଵ, ‫ ݔ‬൒ ‫ ݖ‬implies ‫ݖ ظ ݔ‬.
principles:
etweenGiventhe
between two achievement
theGEP 1.Index
GEP Pareto
Index andprofiles
and thethe x, zinformation
in Թ௃ାଵ, ‫ ݔ‬from
information ‫ ݖ‬implies
൒ from thethe ‫ݔ‬dashboard
‫ݖ ظ‬.
dashboard is ismade made
mmetric and 1. 2.
Given
Pareto
convex Independence
ordertwo achievement
‫ ظ‬on thethe of Identical
spaceԹ profiles Consequences
x, z in Թ ௃ାଵ
, ‫ݔ‬ ൒ ‫ݖ‬ implies ‫ݔ‬ ‫ظ‬ ‫ݖ‬.
symmetric and convex order ‫ ظ‬on spaceԹ of of௃ାଵ௃ାଵ
achievements,
achievements, which
which satisfies
satisfies
2.
ciples: Independence Let of x and
Identical z be two achievement
Consequences
Given two achievement profiles x, z in Թ , ‫ ݔ‬൒ ‫ ݖ‬implies profiles
௃ାଵ with ‫ݔ‬ ൌ ‫ݖ‬ for
௞‫ݖ ظ ݔ‬. some ݇ ‫ א‬ሼͳǡ ǥ ǡ ‫ ܬ‬൅ ͳሽ. Then ‫ ݖ ظ ݔ‬if and
rinciples: ௞
Let x and z be2.two only if ‫ିݔ‬௞ ‫ିݖ ظ‬௞ .ofprofiles
achievement
Independence Identical with ‫ݔ‬௞ ൌ ‫ݖ‬௞ for some ݇ ‫ א‬ሼͳǡ ǥ ǡ ‫ ܬ‬൅ ͳሽ. Then ‫ ݖ ظ ݔ‬if and
Consequences
only if ‫ିݔ‬௞2.‫ظ‬Independence
‫ݖ‬Let .
ି௞ x and z beoftwo achievement
Identical Consequencesprofiles with ‫ݔ‬௞ ൌ ‫ݖ‬௞ for some ݇ ‫ א‬ሼͳǡ ǥ ǡ ‫ ܬ‬൅ ͳሽ. Then ‫ ݖ ظ ݔ‬if and
ement profilesLet The
x 3.
only
and Independence
achievement
x, zx,inz Թ if z ‫ݔ‬be two
‫ظ‬ ‫ݖ‬
, ‫ ݔ‬,൒‫ିݖ ݔ‬௞
profile
. of
achievement Duplicated
for eachprofiles Consequences
country is then
with ‫ݔ‬௞ ൌ ‫ݖ‬௞ forachievement
some ݇ ‫ א‬ሼͳǡ profiles
ǥǡ‫ ܬ‬൅ ofͳሽ.twoThen
countries.
‫ ݖ ظ ݔ‬Then
if andit can
evement profiles ௃ାଵ
in Թି௞ ௃ାଵ
൒implies
‫ ݖ‬implies ‫ݖ ݔظ ݔ‬. ‫ݖ ظ‬.
3. Independence
26 theLet of x,
only if ‫ିݔ‬௞ ‫ିݖ ظ‬௞ . z
(J+1)-vector be
Duplicated such
given that,
by
Consequences for
the some
GEP k,
achievementm, ‫ݔ‬ ௞ ൌ ‫ݔ‬௠ and be‫ݖ‬௞argued
ൌ ‫ݖ‬ ௠ . Then
that if ‫ݔ‬ ‫ظ‬is ‫ݖ‬in if
a and
more only if
favorable‫ݔ‬ ‫ିݖ ظ‬௞ if
ି௞position

of of
ce LetIdentical
x, z be
Identical 3.and
such
and
Consequences that,
the only forif ‫ݔ‬some
achievement
Independence
Consequences ି௠ ‫ظ‬ ‫ݖ‬k,ି௠
offor m, . ‫ݔ‬௞ of
each
Duplicated ൌ the and ‫ݖ‬௞ inൌthe
stocks
‫ݔ‬௠Consequences ‫ݖ‬௠ . Thenthan ‫ ݖ ظ ݔ‬ifit must
and only be the if case
‫ିݔ‬௞ ‫ظ‬that if worst
‫ିݖ‬௞the
owo and only if3.‫ିݔ‬௠
achievement
achievement Let‫ݖ ظ‬x,ି௠
dashboard,
Independence
profiles
profiles .be
zwith as‫ݔ‬such
with ‫ݖ‬ൌthat,
calculated
of௞Duplicatedfor for
above.
some some݇ ‫ ݇א‬k, ‫ א‬m,
Consequences ǥ ǡ‫ݔ‬ǥ ǡൌ ൅‫ݔ‬௠ͳሽ. and
Then if‫ݖ‬.and
‫ݖ‬achievement
‫ݔ‬௞ ‫ظ‬ൌ ‫ݖ‬௠ ifThen in ‫ظ ݔ‬ if andthan
is ‫ݖ‬greater onlythe
if worst
‫ିݔ‬௞ ‫ିݖ ظ‬௞ if
௞ ‫ݔ‬ൌ ௞‫ݖ‬ ௞ for some ሼͳǡ ሼͳǡ ‫ܬ‬௞൅ ‫ͳ ܬ‬ሽ. Then ‫ظݖ ݔ‬ and
Let Define
and
x, z only
be the
suchif ‫ݔ‬ difference
that,
‫ظ‬ ‫ݖ‬ for . set
some to
k, be
m, ‫ݔ‬ equal
ൌ ‫ݔ‬ toand the‫ݖ‬ set
ൌ ‫ݖ‬ of.
௞ achievement
dimensions
Then ‫ݔ‬ if ࣞሺ‫ݔ‬ǡ
and ‫ݖ‬ሻ on
only
in . This is the principle
‫ظ‬ ‫ݖ‬ ifwhich
‫ݔ‬ି௞ ‫ظ‬of
the ifindex-
Priority
‫ିݖ‬௞ to
௞. ି௠ ି௠ ௞ ௠ ௠
Define theand To dashboard
difference
only if ‫ିݔ‬௠
understand profiles
set‫ظ‬how to
‫ିݖ‬௠be . equal
one x and
could use z differ,
to the that is:
set of dimensions
the information in the ࣞሺ‫ݔ‬ǡ on which the
Worst‫ݖ‬ሻAchievement and isindex-
how the ranking of
of of
ce dashboard
Duplicated
Duplicated profiles
Define
Consequences
the x and
achievement
Consequences the zdifference
differ,
profile that
of setis: to
two be equal
countries to the setoverall
to calculate of dimensions
progress, GEP+, ࣞሺ‫ݔ‬ǡis on which the index-
‫ݖ‬ሻconstructed.
dashboard ࣞሺ‫ݔ‬ǡ ‫ݖ‬ሻ ൌ ሼ݇ ‫א‬ ሼͳǡ ǥ ǡ ‫ܬ‬ ൅ ͳሽǣ ‫ݔ‬ ് ‫ݖ‬ ሽ
that,
h that, forfor
some Define
some k,
them, the
k,GEP+,
m, ‫ݔ‬ൌ௠profiles
‫ݔ‬௞ ‫ݔ‬ൌ௞difference ‫ݔ‬and
consider ௠ and ‫ݖ‬the
௞ ‫ݖ‬ൌ
xcase
set and
௞‫ݖ‬ ൌ௠ to‫ݖ‬. zwhere
differ,
be
Then
௠ . Then ‫ ݔ‬that
equal ‫ظݖ ݔ‬to
and
‫ظ‬ ifis: the
‫ ݖ‬and
if
areand setonly
only
the of
if ‫ݔ‬ifdimensions
ି௞‫ିݔ‬௞ ‫ିظݖ ظ‬௞‫ݖ‬௞ି௞ if ifࣞሺ‫ݔ‬ǡ
௞ ‫ݖ‬ሻ on which the index-
‫ିظݖ‬௠ . .
‫ିݖ‬௠ ࣞሺ‫ݔ‬ǡ
dashboard profiles x and z differ, that is: ‫ݖ‬ሻ ൌ ሼ݇ ‫א‬ ሼͳǡ ǥ ǡ ‫ܬ‬ ൅ ͳሽǣ ‫ݔ‬ ௞ ് ‫ݖ‬ ௞ ሽ
The following is a consequence ࣞሺ‫ݔ‬ǡof ‫ݖ‬ሻTheoremൌ ሼ݇ ‫ א‬ሼͳǡ 3 inǥBarberà
ǡ ‫ ܬ‬൅ ͳሽǣ and ‫ݔ‬௞ ്Jackson
‫ݖ‬௞ ሽ (JET, 1988).
ence Theset
erence following
to to
set bebe is a
equal consequence
equal to to thethe setset ofof Theorem
ofdimensionsࣞሺ‫ݔ‬ǡ3‫ݖ‬ሻ
dimensions in ൌBarberà
ࣞሺ‫ݔ‬ǡሼ݇
ࣞሺ‫ݔ‬ǡ ‫ א‬ሼͳǡ
‫ݖ‬ሻ on
‫ݖ‬ሻandon ǡ Jackson
ǥwhich
‫ܬ‬which
൅ ͳሽǣ ‫ݔ‬the
the (JET,
௞ index- ௞ሽ
് ‫ݖ‬index- 1988).
siles
x and
x and z differ, The
that
z differ, that following
is: is: is a consequence of Theorem 3 in Barberà and Jackson (JET, 1988).
Theorem (The
The following is a consequence Protective Criterion) of Theorem 3 in Barberà and Jackson (JET, 1988).
Theorem (The ࣞሺ‫ݔ‬ǡ Protective
ࣞሺ‫ݔ‬ǡ‫ݖ‬ሻ ‫ݖ‬ሻ ൌ ሼ݇ ൌ ሼ݇ Criterion)
‫ א‬ሼͳǡ‫ א‬ሼͳǡ
ǥ ǡǥ ‫ܬ‬൅ ǡ ‫ͳ ܬ‬ሽǣ
൅ ͳሽǣ ‫ݔ‬௞ ‫്ݔ‬௞ ‫്ݖ‬௞ ሽ‫ݖ‬௞ ሽ
Let ‫ظ‬௉ be
Theorem a symmetric
(The Protectiveand convex order on Թ௃ାଵ that satisfies Pareto, Independence of Identical
Criterion)
Consequences and Independence Թ௃ାଵ of Duplicated Consequences. Then
a Let
sconsequence ‫ظ‬௉ be Theorem
consequence a of
symmetric
Theorem
of (Theand
Theorem in3convex
3Protective
Barberà
in Barberà order
Criterion)
and and onJackson
Jackson that
(JET, satisfies
(JET, 1988).Pareto, Independence of Identical
1988).
ConsequencesLet and ‫ظ‬௉Independence
be a symmetric of Duplicated
and convexConsequences.
order on Թ௃ାଵ that Then satisfies Pareto, Independence of Identical
Let ‫ظ‬ Consequences
௉ be a symmetric and and‫ݔ‬ ‫ظ‬
Independence
convex
௉ ‫ݖ‬ if and
orderof on only
Duplicated
Թ if
௃ାଵ
that ‹
Consequences.
satisfies
௞‫ࣞא‬ሺ௫ǡ௭ሻ ‫ݔ‬௞ ൐Then
Pareto, ‹ ௞‫ࣞא‬ሺ௫ǡ௭ሻ ‫ݖ‬௞
Independence of Identical
Consequences ‫ݔ‬ ‫ظ‬ ‫ݖ‬ if and only
௉ and Independence of Duplicated if ‹ ‫ݔ‬
௞‫ࣞא‬ሺ௫ǡ௭ሻConsequences.൐ ‹ ‫ݖ‬
tective Criterion)
Protective Criterion) ௞
32 ௞‫ࣞא‬ሺ௫ǡ௭ሻThen ௞
‫ظ ݔ‬௉ ‫ݖ‬ if and only if ‹௞‫ࣞא‬ሺ௫ǡ௭ሻ ‫ݔ‬௞ ൐ ‹௞‫ࣞא‬ሺ௫ǡ௭ሻ ‫ݖ‬௞
32
metric and convex order onon ‫ظݔ‬
that ‫ ݖ‬satisfies
௉satisfies if and
Pareto,onlyIndependence
if Independence
‹௞‫ࣞא‬ሺ௫ǡ௭ሻ ‫ݔ‬௞Identical
൐ ‹௞‫ࣞא‬ሺ௫ǡ௭ሻ ‫ݖ‬௞
mmetric and convex order Թ௃ାଵ Թ௃ାଵ that Pareto, 32 of of Identical
d Independence
and Independence of ofDuplicated
Duplicated Consequences.
Consequences. Then Then 32

‫ݔظ ݔ‬௉‫ݖظ‬௉ ‫ ݖ‬if and only

if and if if ‹
only ‹
௞‫ࣞא‬ሺ௫ǡ௭ሻ ‫ݔ‬௞ ‫ݔ‬൐௞ ‹
௞‫ࣞא‬ሺ௫ǡ௭ሻ ൐ ‹
௞‫ࣞא‬ሺ௫ǡ௭ሻ ‫ݖ‬௞ ‫ݖ‬௞
௞‫ࣞא‬ሺ௫ǡ௭ሻ

32 32
 NOTES
27
1
The current beta test application of the
methodology has 14 direct links to 10 of the 17
SDGs. For an overview of the links between the
SDGs and the GEP measurement framework,
see PAGE (2017), The Green Economy Progress
Framework- Application (2017).
2
See Section 2 for a definition of this “new
generation of capital”.
3
See PAGE (2017). The Green Economy Progress
Framework – Application.
4
There is a potential for the GEP Measurement
Framework to cover more selected SDGs and
targets. The flexibility of the framework allows for
the inclusion of many of the SDGs indicators related
to green economy. This is particularly important for
applications of the framework at the country level
where a richer set of indicators may be available.
5
United Nations (2012).
6
See GGKP (2016) for a complete review of
main approaches and indicators as well as the
identification of research gaps.
7
A significant literature review was conducted prior
to this project to assess the landscape of existing
indices related to sustainable development. A
gap was found in the area of the measurement of
green economy progress, which was one of the
motivations for developing the GEP Measurement
Framework. See Pineda and Galotto (2015) for more
information.
8
This section is largely based on “An Emerging
theory of an Inclusive Green Economy” (Sheng,
2016).
9 These assets include: (a) renewable natural capital
(e.g. freshwater, forests, and fisheries); (b) low
carbon, resource efficient physical capital (e.g. solar
panels, wind turbines, public transport systems,
and waste treatment/recycling/re-manufacturing
facilities for a circular economy); (c) human capital
with green job skills (e.g. installation, operation and
maintenance of energy efficient equipment); (d) and
social capital (e.g. equitable access to justice, social
services, and opportunities, social safety nets, and
social protection floors).
10
See Section 2.2 for a definition of progress.
11
The current beta test application of the
methodology has 14 direct links to 10 of the 17
SDGs. For an overview of the links between the
SDGs and the GEP measurement framework, please
see The Green Economy Progress Framework:
Application (2017).
12
To ensure consistency throughout UN
Environment’s work, the country-tailored version of
the GEP measurement framework will be integrated
into UN Environment’s framework on green
economy indicators under PAGE. This framework
has been designed to link indicators with the
integrated green economy policymaking process.
13
In this context, “valid” is to be understood as
“reasonable” or “acceptable”.
14
Based on technical documents prepared by
Carmen Herrero, Antonio Villar and Eduardo
Zambrano (2016), Antonio Villar (2011), and on
technical discussions with José Pineda and Gisèle
Mueller.
15
See: http://hdr.undp.org/sites/default/files/
hdr2015_technical_notes.pdf
16
In line with the “Beyond GDP” approach (See
Report by the Commission on the Measurement of
Economic Performance and Social Progress (Stiglitz
et al., 2010)) and the recent criticisms of using GDP
as a (or the only) measure of well-being, the GEP
measurement framework has deliberately decided
not to use measures of GDP in the calculation of
28
progress. The main reason for this decision is that
GDP not only measures green but also ‘brown’
economic activities, which do not support the
transition towards an Inclusive Green Economy.
The inclusion of indicators such as green trade,
green innovation and renewable energy allows the
GEP index to capture (although partially) the green
components of GDP. In addition, the inclusion of
indicators measuring the access to basic services,
health and education allows the GEP index to
capture the positive aspects that higher economic
growth will bring to promoting an Inclusive Green
Economy.
17
See Annex I.A for a discussion of properties in
the basic model and a proof of the characterization
result.
18
It is important to note that for a country
experiencing regress in an indicator in which it is
initially disadvantaged with respect to the relevant
threshold, this weighting system will imply that
regress will have a significant weight. In other
words, the weighting system provides signals on
policy priorities.
19
The first weighting indicates the relevance of the
progress made in each of the areas, as captured
by the indicators; the second weighting, however,
makes it possible to establish comparisons within
and across countries (given that the sum of all
weights is equal to 1). See Annex I.B for a discussion
of weighting for the GEP index in the multiple
country case.
20
For a method to combine an assessment of
development adjusted by sustainability, see Pineda
(2012).
21
This implies a non-decreasing discounted
utilitarian sum of generational utility, i.e. economic
paths along which intergenerational well-being
does not decline. For more information, see Marc
Fleurbaey and Didier Blanchet (2013).
22
See more on this from Fleurbaey and Blanchet
(2013, p. 21).
23
Fleurbaey and Blanchet (2013, p. 50).
24
This is so because the Protective Criterion (see
Annex I.C), like the leximin, does not admit a real-
valued representation due to the lack of continuity of
the preference ordering. For proof see the example
of Moulin (1998, p. 34).
25
This method of creating the ranking limits the
incentives for substitution across equally important
aspects of an Inclusive Green Economy; it also gives
incentives to progress in all aspects and penalizes
any partial view that only concentrates on a few
policy areas.
26
See PAGE (2017). The Green Economy Progress
Framework – Application for an application of the
methodology at the global level.
PAGE gratefully acknowledges the support of all its funding partners:

European Union

Federal Ministry for the Environment, Nature Conservation,

Building and Nuclear Safety, Germany

Ministry for Foreign Affairs of Finland

Norwegian Ministry of Climate and Environment

Ministry of Environment, Republic of Korea

Government Offices of Sweden

Swiss Confederation, State Secretariat for Economic Affairs SECO

 For further information:

PAGE brings together the expertise, convening power PAGE Secretariat

and networks of five UN agencies – UN Environment, UN Environment
International Labour Organization, UN Development Economics and Trade Branch
Programme, UN Industrial Development Organization, 11-13 Chemin des Anémones
and UN Institute for Training and Research – to support CH-1219 Chatelaine-Geneva
countries in addressing one of the most pressing challenges Switzerland
of the 21st century: transforming their economies and page@unep.org
financial systems into drivers of sustainability and social
equity.

The Partnership supports nations and regions in reframing

economic policies and practices around sustainability
to foster economic growth, create income and jobs, www.un-page.org
reduce poverty and inequality, strengthen the ecological
foundations of their economies and achieve the Sustainable
Development Goals.

@PAGExchange

@GreenEconomyUNEP
www.un-page.org

un-page.org/newsletter