Toward a Circular Economy Scan: Measuring Circular Practices among Retailers in the Brussels Capital Region

Full text

TOWARD A CIRCULAR
ECONOMY SCAN
MEASURING CIRCULAR PRACTICES AMONG
RETAILERS IN THE BRUSSELS CAPITAL REGION
written by
David Bassens
Sarah De Boeck
Wojciech Kębłowski
Deborah Lambert
Hunter Reinhardt
Research report of the Circular City research project, funded by Innoviris.
October 2020

Cosmopolis – Centre for Urban Research 2
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ....................................................................................................... 3
LIST OF FIGURES ................................................................................................................. 4
1. INTRODUCTION ............................................................................................................. 5
1.1. Context and urgency of the research ........................................................................ 5
1.2. The evaluation and creation of data to measure the circular economy ..................... 6
2. THEORETICAL APPROACH: MEASURING CIRCULAR PRACTICES ........................... 8
2.1. Definition of the circular economy? ........................................................................... 8
2.2. The scale of the circular economy? .........................................................................10
2.3. Micro-level indicators ...............................................................................................11
2.4. Where does the circular economy meet the retail sector? .......................................13
3. METHODOLOGICAL APPROACH .............................................................................14
3.1. Research context of the project: existing data on the circular economy in the BCR .14
3.2. Selection of the neighbourhoods .............................................................................15
3.3. Development of the circular economy scan .............................................................16
4. RESULTS ...................................................................................................................26
4.1. Results of the questionnaire ...................................... Error! Bookmark not defined.
4.2. Insights about the circular practices of Brussels retailers .........................................27
5. DISCUSSION & CONCLUSION ..................................................................................35
5.1. A circular economy scan ‘in the making’ ..................................................................35
5.2. A pilot for a larger study: further research ................................................................36
5.3. Conclusion ..............................................................................................................37
6. REFERENCES ...........................................................................................................39

Cosmopolis – Centre for Urban Research 3
ACKNOWLEDGEMENTS
Cosmopolis – Centre for Urban Research of the Vrije Universiteit Brussel gratefully
acknowledges hub.brussels for co-organizing the data collection and performing a
descriptive analysis of the survey results.
This research is generously supported by the Anticipate – Prospective Research 2017
program of Innoviris, the institution of the Brussels Capital Region encouraging scientific
research and innovation.
COSMOPOLIS
CENTRE FOR URBAN
RESEARCH
Cosmopolis is based at the Vrije Universiteit Brussel and
is dedicated to engaged research and teaching in
geography, spatial planning and urban design.
Cosmopolis is concerned to better understand our
contemporary and increasingly global urban condition
through theoretical and empirical research.

Cosmopolis – Centre for Urban Research 4
LIST OF TABLES AND FIGURES
Figure 1
Order of priority for circularity strategies: the R-ladder
Source: Delahaye et al. 2018
Figure 2
Question: ‘How do you choose your products when you place your order?’
Source: Composed by the authors
Figure 3
Matrix extract focusing on the neighbourhood of Anneessens regarding survey
question: ‘How do you choose your products when you place your order?’
Source: Composed by the authors
Figure 4
Question: ‘Are you trying to reduce packaging?’
Source: Composed by the authors
Figure 5
Matrix extract focusing on the neighbourhood of Anneessens regarding survey
question: ‘Are you trying to reduce packaging?’
Source: Composed by the authors
Figure 6
Matrix extract with open-ended responses regarding survey question:
‘Give examples of other criteria that you favour to order products’
Source: Composed by the authors
Figure 7
The geography of survey responses
Source: hub.Brussels
Figure 8
Survey results for particular Rs obtained across sectors mixes
Source: Composed by the authors
Figure 9
Survey results for particular Rs obtained across business types
Source: Composed by the authors
Figure 10
Survey results for particular R’s (circular practices) obtained across
neighbourhoods
Source: Composed by the authors
Table 1
Turning the 10 R ladder into operational strategies and into a questionnaire
Source: Delahaye et al. (2018), methodological approach by the authors
Table 2
Timetable of the feedback loops to refine the questionnaire
Source: Composed by the authors

Cosmopolis – Centre for Urban Research 5
1. INTRODUCTION
To what extent have existing economic activities adopted ‘circular practices’ and how can
we measure these practices? This is the main question we would like to answer is this
research report. We do so by describing and evaluating the methodological approach of
the creation of a tool to measure such circular practices: the circular economy scan.
1.1. Context and urgency of the research
The focus of this study is the ground-level, embedded and seemingly mundane circular
economy practices of small and medium enterprises (SMEs) in the retail sector. The
setting of the research is the Brussels Capital Region (BCR) where a regional program
for circular economy
1
(RPCE) is adopted in 2016. This program aims to revitalise the
BCR’s economy and brings together a sustainable turn and the specific challenge of job
creation. In a next step, Innoviris, the Brussels’ regional agency for research and
innovation, launched a research call on the circular economy in 2017 and part of the
budgets were allocated to Cosmopolis – Centre for Urban Research. This report is one
of the main research outputs of the allocated Circular City research project.
The Innoviris call centres on the question of achieving an equilibrium between, on the one
hand, the socio-economic diversity of the BCR and its inhabitants, and, on the other hand,
the requirements of the BCR’s potential transition towards the circular economy
paradigm. We argue that the framing of the call is somewhat less sensitive to the more
fine-grained economic geographies of small and medium enterprises (SMEs). Therefore,
instead of conceptualising the policy challenge as a question of raising awareness about
the circular economy, lowering its barriers, and increasing its accessibility to Brussels
households, we posit that it may be equally valuable to scrutinise to what extent smaller-
scale bottom-up activities that may qualify as ‘circular practices’ are already taking place
throughout Brussels.
2
Our starting hypothesis is that long-held mundane re-using and
11
Regional Program for Circular Economy (RPCE) – Programme Régional d’Économie Circulaire (PREC) –
Gewestelijk Programma voor Circulaire Economie (GPCE).
2
Both the Brussels Capital Region (BCR) and Brussels refer to the Brussels Capital Region. The terms are
interchangeable throughout this report.

Cosmopolis – Centre for Urban Research 6
valorisation practices are still taking place and hence participate to the circularity of
Brussels.
This research aims to complement existing Brussels-centred studies about material flows
of waste, water, energy, and construction materials (cfr. Christis et al., 2019; Stephan &
Athanassiadis, 2018; Zeller et al. 2019) by taking a first step towards examining how the
circular economy – its practices, activities and products – is rooted in the socio-spatial
context of Brussels. Next to the existing analyses of material flows on the macro-level,
we offer a micro-level analysis of the ‘circular’ economic activities already taking place in
Brussels, often without carrying the ‘circular’ label.
To further finetune our research subject, it is important to state that our purpose is not to
measure the efficiency of circular economy policies, strategies and measures, nor to
research how circular economy could be further developed in Brussels. Instead, the focus
is on how to measure the circular economy in the BCR. Therefore, this is mainly a
research report on the methodology of defining the circular economy, of measuring the
circular economy, and of describing and evaluating the first steps in creating a data tool
to scan the circular economy. More specific, we aim to identify and assess the adoption
of circular principles by retailers, identified as locally embedded intermediaries of the
circular economy, and operating between broad policies and individual consumers.
1.2. The evaluation and creation of data to measure the circular
economy
While administrations rely on existing datasets to fathom employment or corporate
turnover, there is little systematic information collected on the degree to which firms in the
Region have adopted circular practices. As we will see in section 3.1.1., the absence of
data is a well-known issue. To this end, we have set up a retail survey that aims to shed
light on potential circular economy practices that exist in various retail sectors in Brussels.
We focus on the retail sector as it forms an important part of the local economy by
providing goods and services to urban dwellers, as well as a substantial amount of jobs
(9% of the total employment in the BCR). In addition, retail is locally anchored, forming a
central part of the local economy and urban life. Moreover, as in many other urban

Cosmopolis – Centre for Urban Research 7
contexts in advanced capitalism, in the past decades Brussels has become a major centre
of consumption, making retail a crucial lever in a potential transition to a more circular
urban economy and society.
The objectives of the retail survey are fourfold: (1) to identify and analyse the circularity
of existing economic retail practices in Brussels, (2) to investigate where and what
differences might exist between businesses, neighbourhoods, and sectors within the city-
region, (3) to generate new data, and (4) to propose a roadmap for an alternative to the
existing datasets which currently do not allow to identify circular economy activities. The
survey is set up as a pilot study to check to measure circularity at the regional level. If
scaled up, refined, and standardised, the survey can be the basis of a circular economy
scan – a data collection tool – to be implemented at the Brussels regional level.
Overall, we find that the scores on the different circular strategies are very low. The most
explicit result is that 50% of the retailers use led lighting to spare electricity, indicating that
we are very far from any radical shift toward the circular economy in the Brussels retail
sector. Our circular scan in the making provides in a workable data collection tool to
measure circular practices of retailers. In order to compare key circular strategies, retail
sectors, as well as neighbourhoods, further refining is necessary because the samples
our too small and unevenly distributed across strategies, sectors, and neighbourhoods.
As a result, when some neighbourhoods or sectors do score better, it does not necessarily
mean that they are closer to circularity than other neighbourhoods, but still they are doing
better in this very ‘uncircular’ retail landscape.
The report continues as follows: section two is about the theoretical approach and
explicates the academic literature that underlays the making of the circular economy
scan; section three covers the methodological approach and demonstrates how the
academic literature, and the 10 R ladder with 10 key circular strategies in particular, is
translated into an operational definition to direct the formulation of survey questions;
section four discusses the survey results and evaluates the methodological approach;
section five concludes and gives directions for further research.

Cosmopolis – Centre for Urban Research 8
2. MEASURING CIRCULAR PRACTICES
The quantification of the circular economy and the work on developing indicators is an
important aspect of the academic literature on the topic in the last decade. The concept
of circular economy embraces all stages of the product life cycle, from design to waste
management. Developing accurate indicators is seen as an important step towards
circular economy implementation. Numerous indicators have been developed to assess,
measure and quantify progress. Research on indicators has focused on the product
design and production stages to reduce the environmental and social impacts of
production.
While existing studies (Zhang, 2013; Dinarès, 2014) measure the systemic metabolism
of cities (cfr. urban material and energy flows), a holistic understanding of the circular
economy should also include ground-level praxis. Yet, in parallel to studying what aspects
to consider in understanding the circularity at the scale of a small business (and at the
larger scale of the neighbourhood), this project is also concerned with its quantification.
In other words, our objective was to measure the existing circular level of a firm, a sector,
or a neighborhood.
We use four steps to come to an analytical approach that forms the basis for what we
call the circular economy scan. These four steps consider: (1) the definition of the circular
economy, and (2) the scale of the circular economy. Then, we zoom in on the scale of the
micro-level and look at (3) micro-level indicators. Finally, we focus on the literature where
(4) the circular economy meets the retail sector.
2.1. What is the circular economy?
2.1.1. The need to include ‘mundane’ circular practices
Defining the circular economy is not all but straightforward, because ‘what to be measured
in the sense of circular economy is subject for debate as the definition is ambiguous, and
indicators might lead to different or even incoherent conclusions.’ (Moraga et al. 2019:
453). Consequently, it is not surprising that Saidini et al. (2019: 7) found as many as fifty-
five different sets of indicators focused on different scopes, purposes, and applications
such as for benchmarking and comparison, for labelling products, as well as for
advocating a regulatory change.
For every definition of the circular economy — of which Kirchherr et al. (2017) identified
as many as 114 — there is an equally disparate metric that can be used to measure the
circularity. Therefore, it is crucial to delimit what the circular economy actually entails in

Cosmopolis – Centre for Urban Research 9
this study. We study ground-level, embedded and seemingly mundane practices in SMEs
in the retail sector, rather than circulation of material flows, and larger-scale initiatives
taken inter alia by larger corporate actors.
2.1.2. The 10 ‘R’s: the theoretical framework for an operational definition
We use a circularity ladder, or the so-called R-ladder, as the analytical starting point for
the circularity scan (Delahaye et al. 2018). The R-ladder is a tool to classify strategies
according to the level of circularity.
Figure 1: Order of priority for circularity strategies: the R-ladder
Source: Delahaye et al. 2018
Circularity implies reusing materials in a way that strives for the highest quality. Ideally,
the entire product is reused. If that is not possible, then parts of the product are reused
and finally the raw materials or materials that come from a product. As we see in Figure

Cosmopolis – Centre for Urban Research 10
1, it is a priority ladder, where 1 scores the highest and 10 the lowest. The highest
strategies on the ladder contribute the most to a reduced use of new resources. By
contrast, recycling and energy recovering are at the lowest of the ladder and should only
be used when other strategies are not possible.
2.2. The scale of the circular economy?
Next to the delineation of the circular economy, the academic literature also centers on
the appropriate scale of analysis. Territorial definitions will affect the levels of material
consumption and production, and impact the measure of the circularity of the given space.
In addition, circular economy principles can be applied at different scales: from material
to product, from city to the global economy.
We follow the approach of the Circular Economy Policy Research Center (Vercalsteren
et al., 2018), where three scales (not mutually exclusive) – macro, meso and micro – can
be ascribed to circular economy indicators:
• Macro-level: (material) exchanges between the economy and the environment
such as international trade and material accumulations in national economies. The
first assessments to measure circular economy, based on material flows analysis,
were developed for the macro-level.
For example, the French Observation and Statistics Department (Ministère de l'environnement, de
l'énergie et de la mer, en charge des relations internationales sur le climat 2017) has worked on
defining a set of ten indicators to monitor the transition to a circular economy at the national level
and allow European comparison. They follow the product development cycle: five indicators at the
level production (e.g. the number of ecolabel holders), two for action (e.g. consumption expenditure
per capita for maintenance and repairing) and two for back-end (e.g. amount of stored non-
hazardous waste). A last indicator counts the number of circular economy-related jobs.
• Meso-level: industry, consumption activity or at the level of specific materials. The
economic, environmental or social performance of a region, a product group or an
industry.
For example, the Chinese government indicator system includes indicators for industrial parks.
Potting et al. (2018) developed a framework to measure circular economy transition of individual
product chains.
• Micro-level: specific decision processes at a business or local level or concerning
the specific substance or individual products.
For example, the Ellen MacArthur Foundation (EMF) developed in 2015 a methodology (Material
Circularity Indicator) at the product level that can be used at the company level by aggregating the
results of all the products produced by a company (EMF 2015). Again, these indicators focused
only on technical cycles and materials. In 2019 EMF developed a new framework, Circulytics, to
go beyond and measure a company’s circularity across its entire operations (EMF 2020).

Cosmopolis – Centre for Urban Research 11
Each scale aims to measure a different aspect of the circular economy. Measuring gross
indicators can be more easily accomplished on larger scales, as aggregates in sectoral
proficiencies can be used as surrogates for measuring the impact of circular economy
strategies. The main focus of our research is the micro-level, since we focus on circular
economy practices at the scale of the firm and the scale of the neighbourhood. In addition,
while we scrutinise micro-level practices, our intention is to develop a set of micro-level
indicators that could be aggregated to the meso-level and provide information on the retail
sector in different neighbourhoods.
Vercalsteren et al. (2018) classify their indicators review not only according an axis of
scale, but also according an axis of circular economy strategies covered (see 2.1.2., the
10 R-ladder) and an axis on the technology or the socio-institutional focus. Especially this
last axis of socio-institutional indicators is important for our research because it refers to
governance and infrastructure aspects. Therefore, the axis focusses on the degree to
which businesses are involved in collaboration schemes, information sharing or waste
management procedures (ibid, 10). The authors find that most of the indicators in existing
dashboards and frameworks are macro-level indicators (p.33) and that physical
parameters (product production, resource consumption, and other weight-based metrics)
are much more common than socio-institutional aspects. Other sources confirm or
reinforce these challenges, and literature reviews see the state-of-the-art as often
reductionistic (Gasparatos et al., 2008; Elia et al., 2017), unilateral (Moraga et al., 2019),
or insufficient in measuring all the dimensions of the circular economy (Parchimenko et
al, 2019; EASAC, 2016).
Based on a review of the state of the art, we conclude that new indicators are needed for
identifying whether, how, and where particular economic activities engage in different
kinds of ‘circular’ strategies, behaviours, and attitudes. This is particularly evident in terms
of capturing the contribution of inhabitants for whom local SMEs may act as potentially
important channels through which one can engage in circular economy on the
neighbourhood level. To conclude, our own work on developing a circular economy scan
is focused on the micro-scale and tries to cover the 10 Rs.
2.3. Micro-level indicators
Indicators that focus on the micro-scale — as well as indicators operating across scales
— focus on narrow and economic aspects of the circular economy, to the detriment of
environmental and social dimensions (Kristensen & Mosgaard, 2019, Sassanelli et al.,
2019). Moraga et al. (2019) found that while smaller scales may more easily capture life
cycle analyses than European-level macro indicators, they still focus on the preservation
of materials, rather than on the development and reinforcement of functional or systemic
frameworks.

Cosmopolis – Centre for Urban Research 12
The European Environmental Agency (Rood & Kishna, 2019) reports that micro-level
indicators might be useful to monitor niche developments that would remain otherwise
invisible in macro-level indicators (p.43). It is exactly this gap we try to respond to with the
development of a ground-level circular economy scan. Instead of using conventional
indicators that measure practices such as end-of-life strategies, our intention is to capture
social or economic changes in a circular system (Rossi et al., 2020) by uncovering
existing sharing practices or refusal behaviours. Our purpose is not to measure the
efficiency of circular economy policies, strategies and measures, but to identify and
assess the adoption of circular principles by retailers, identified as locally embedded
intermediaries of the circular economy, and operating between broad policies and
individual consumers.
Nevertheless, we recognise that a multitude of circular economy indicators exist,
developed primarily to measure the efficiency of strategic planning or the circularity of a
given system. Index-based metrics for measuring the circular economy include those
based especially on material and energy flows, land use and consumption, or other life
cycle-based indicators (Elia et al., 2017), as well as less direct methods that aim to move
past purely material metrics. These relate to the literature debating on how to categorize
the disparate methods of measuring the circular economy (Herva et al., 2011; EASAC,
2016; Saidani et al., 2019), to critically analyse the appropriateness of various indicators
(Elia et al., 2017), and to outline a rough decision tree for selecting appropriate indicator
methodologies.
For businesses specifically, organizational indicators and strategies have been
developed both academically and by organizations involved in the field. In France,
Entreprises pour l’Environnement and the Institut National de l’Economie Circulaire have
assembled a working group to analyse existing tools and to develop indicators for
businesses, finding that a plurality of measurement strategies are already in use (Moesch
et al., 2018). Janik & Ryszko (2019) found nineteen circular economy indicators available
at the micro level. Most of these indicators focus on product life cycle and hence on
measuring product or material circularity. They found that only one of them focuses on
environmental, economic and social aspects together.
More recently, the Ellen MacArthur Foundation (2020) developed an online tool to
conduct a comprehensive analysis of a given organization’s circularity, and to provide
feedback for strategic improvement. Their method generates a single score based on
twenty-one to thirty indicators (depending on the industry) and tries to look beyond the
sole resources and material flows quantification by adding an evaluation of the company’s
circular economy strategies such as targets, implementation plans or internal learning
programmes.

Cosmopolis – Centre for Urban Research 13
To conclude, this review shows that the vast majority indicators are developed to measure
circular policies and practices in large companies already involved in developing circular
economy communication and practices. Quite differently, our work aims to uncover
everyday activities of small businesses participating at developing a circular economy
while not engaged ‘officially’ in circular economy strategies.
2.4. Where does the circular economy meet the retail sector?
The academic literature on circular economy in the retail sector is still quite sparse.
Existing studies typically analyse retailers’ annual reports or their communication about
programs and actions to reach circular economy (Jones & Comfort, 2017; Jones &
Comfort, 2018; Istudor & Suciu, 2020). Another set of studies focuses on barriers and
opportunities in the retail sector (Dangana et al., 2012). In particular, this literature tends
to focus on very large-scale retailers such as H&M and food-waste production in the
mass-retail sector (Bernon et al., 2018; Mondello et al., 2017).
More aligned with our approach, Adam et al. (2017) define four circular economy business
models for retailers: (1) the repair café, and (2) the second-hand shop contribute to
product life expansion, (3) reverse retailing to resource recovery, and (4) the leasing
model to the service economy. While their work does not offer any proposal for how to
measure retailers’ involvement in these models, we incorporated these different aspects
in our questionnaire.

Cosmopolis – Centre for Urban Research 14
3. METHODOLOGY
The methodological chapter is the main section of this report on the creation and
evaluation of a tool to scan the circular economy. This section continues with (1) the
research context of the project with an overview of the existing data on the circular
economy in the BCR, and (2) the selection of the neighbourhoods where we conducted
the circular economy survey. Then, we explain (3) how we developed the tool of the
circular economy scan. First, by transforming the 10R ladder into an operational tool to
create survey questions. Second, by improving the questions through several feedback
loops. And third, by explaining how we measured circularity. This section concludes with
the limits of the circularity scan.
3.1. Research context of the project: existing data on the
circular economy in the BCR
Currently, there is no comprehensive method of tracing and mapping of the Brussels’
regional public authorities to analyse local circular economy activities. As a result, there
is no publicly available circular economy information, neither at the level of companies,
nor at the level of a neighbourhood, a municipality or the region. Yet, a few initiatives that
could be used as a starting point still exist:
• Quickscan to assess neighbourhood projects: developed by the BCR
(https://besustainable.brussels/charte/vision/), under the Be Sustainable initiative,
to enable an assessment of neighborhood projects, based on fifty questions
covering issues from environment, water, materials to governance.
• Circular economy certification: the Brussels circular economy certification
(Bruxelles Economie et Emploi, n.d.) that can be allocated to SMEs based on a
questionnaire (thirty-two questions, eleven criteria within three categories: good
practices, business model, material flows). While the questions cover similar
aspects to our survey items, their questionnaire targets companies that are
explicitly involved in circular economy. This certification checklist could act as a
first database of Brussels-based companies actively involved in circular economy.
• Evaluation of the level of knowledge about circular economy by Brussels
companies: a survey commissioned by Hub (Sonecom 2019) to evaluate the level
of knowledge about the circular economy by Brussels companies and also existing
circular economy practices. Four hundred SMEs were surveyed. One of the results
of this evaluation was that while 75% of the Brussels companies did not know

Cosmopolis – Centre for Urban Research 15
about circular economy, 45% of them engaged with circular economy related
practices. Our survey aims to unpack further this finding.
These three former initiatives are not only an important starting point for this research,
they also inspire the development of a circularity scan. We position this research within
the existing context and develop a tool that can help fill the gap in the existing data as
well as in the methods of evaluation.
3.2. Selection of the neighbourhoods
The research focuses on five selected neighbourhoods: (1) Anneessens, (2) Oud Laeken
Est, (3) Dailly, (4) Haut Saint-Gilles, and (5) Rodebeek Constellations. These five
neighbourhoods exemplify the socio-economic and ethnic diversity of BCR and its spatial
division as they are situated both west and east of the Brussels canal. We categorized
the chosen neighbourhoods according to their socio-economic profile. While circularity is
likely to be part of a wider set of trendy consuming practices, we wanted to explore more
mundane circular practices as well in relation with more embedded practices of mutual
aid that could in theory be related with more diverse socio-economic profile.
• Working-class neighbourhoods: both Anneessens and Oud Laeken Est
exemplify the so-called ‘croissant pauvre’ of Brussels that comprises the most
precarious neighbourhood of the region: high density, high rates of unemployment,
and higher proportion of old housing, social housing and foreigner population
compared to the rest of Brussels.
• Neighbourhoods with gentrification dynamics: Dailly and Haut Saint Gilles
exemplify the gentrification dynamics operating at the regional scale. While Saint-
Gilles is overall a diverse municipality, the ‘Haut’ neighbourhood hosts a wealthier
population, with higher income levels, less unemployment and enjoying a better
housing quality (more ‘bourgeois’ houses and more renovation). It has become an
attractive area for a Western European population with a high concentration of
French population. As for the neighbourhood of Dailly, it is caught between wealthy
areas on the east and more impoverished ones on the west.
• Wealthy neighbourhoods: Quite differently, Rodebeek Constellations in Woluwé
Saint Lambert, is a wealthier neighbourhood in our sample, with higher income
levels and smaller rates of unemployment, less density and better housing
compared to the rest of the region.
The research targets nearly all retail activities (food, mobility, IT, households, textile,
furniture, gift, toy and hardware shops) present in each of the five neighbourhoods. We

Cosmopolis – Centre for Urban Research 16
excluded surveying franchised retailers as part of retail chains as they do not in general
have internal policies of their own.
3.3. Development of the circular economy scan
3.3.1. Partner of the survey: hub. Brussels
The survey is conducted in collaboration with hub.brussels, the Agency for Business
Support. This regional administration offers a wide range of free advice, services and
tools to help developing enterprises. The kind of research collaboration was twofold: (1)
Hub provided twenty surveyors working in pairs during ten working days to run the survey
with small retail shops, and (2) Hub gave feedback about the questionnaire that was used
for the survey. The type of actors with whom Hub works are mainly small retail enterprises
whose managers have diverse backgrounds. Therefore, the survey needed to be short,
straightforward and simple. As a result, the questionnaire covers a series of basic circular
economy aspects, and looks at how they are adopted by various retailers.
3.3.2. Operationalization of the 10 R ladder into a methodological survey tool: the
questions
The questionnaire is largely concerned with the micro-scale of the circular economy. The
goal of the questionnaire is to measure the particular circular behaviour and the
operational business strategies in the BCR, and to compare them across sectors and
neighbourhoods as components and microcosms of the region’s circularity. Because a
variety of sectors are investigated, it is important to detect operational strategies or to
construct indicators that encapsulate the full breadth of circular practices, instead of
unilateral or mono-dimensional aspects. The 10 R framework (see section 2.1.2 for an
explanation) provides an exemplary hierarchy for assigning value to such strategies and
practices.
To compile the questionnaire, we therefore operationalized the 10 R ladder (Table 1) and
adapted it to the retail sector. The key circular strategies (Refuse, Rethink, Reuse, Repair,
Refurbish-Remanufacture-Repurpose, and Recycle) are translated into operational
strategies, who are turned into retail-related actions, and finally, these actions are
transformed into survey questions.
Some of the key circular economy strategies were more difficult to adapt to the retail
sector, in particular those related to the production aspect (e.g. the design of products).
While some of the items are sector specific, a large part of the questions applies across
sectors. As production activities typically do not occur in the retail sector, we translated a
circular economy scenario to retailers who sell services instead of goods, and who inform
customer about maintenance and repair services, the environmental impact (reducing

Cosmopolis – Centre for Urban Research 17
overconsumption by decreasing sales promotional practices), materials and further use
in the final phase of the life cycle.
Table 1: Turning the 10 R ladder into operational strategies and into a questionnaire
Other specific circular operational strategies and actions related to retailers are reducing
transport time, by developing partnerships with other retailers (e.g. in the same street);
reuse waste, by diminishing food and/or packaging waste, or by taking back unused
Key circular
strategies
Operational circular strategies Actions Questions
0 REFUSE Avoidance of toxic or dangerous products Do you avoid toxic or dangerous products?
Reduce packaging
Choice of suppliers with little or no
packaging
Are you trying to reduce packaging by changing suppliers?
Selling unpackaged products Are you trying to reduce packaging by selling products without packaging?
Selling high quantity of products Do you offer your customers products sold in bulk?
Do you offer your custormers products sold by lot?
1 RETHINK Reduce transportation time Sharing or grouping transport of goods
Are you trying to reduce the cost or transportation time of your goods by
sharing or grouping transport with other companies?
Filling trucks to the maximum
Are you trying to reduce the cost or transportation time of your goods by
filling the delivery trucks to the maximum?
Long-term use of products
Talking with stakeholders about long-
term use of products
Do you discuss with your customers about long-term use of products?
Are you trying to reduce packaging by talking to your suppliers?
Product leasing Do you offer your customers product rental?
Use of unsold or expired products Giving away unsold or expired products
Do you donate your products when they are unsold or expired? To clients,
family members or non-profit organizations?
2 REDUCE Local production Promotion of local production Do you promote products made or produced in Belgium?
Talking with stakeholders about local
production
Do you discuss with your customers about the origin of the products?
Local distribution and wholeselling Working with local distributors
Are you trying to reduce the cost or transportation time of your goods by
choosing wholesalers in Brussels?
Reduce packaging Buying a high quantity of products Are you trying to reduce packaging by buying larger packaging?
Reduce operational consumption of the
shop
Use of mechanical and/or technical
devices
Do you do things so save the store's operating costs? For example, do you
use closed refrigerators?
Do you do things so save the store's operating costs? For example, do you
have devices to reduce consumption?
3 REUSE Reduce packaging Reutilizing supplier packaging
Are you trying to reduce packaging by reutilizing supplier packaging a
second time?
Long-term use of products Selling of second-hand products Do you offer your customers second-hand products?
Reuse waste Valorization of organic waste Is organic waste from your business valorized with food valorization?
Use of unsold or expired products
Putting unsold or expired products on
sale
Do you put your products on sale when they are unsold or expired?
4 REPAIR Long-term use of products Buying repairable products You favor products that are easily repairable by the customer?
Talking with stakeholders about long-
term use of products
Do you discuss with your customers about repair of the products?
Repairing broken products
For warranty products, do you repair broken products reported by your
customer?
5 - 7
REFURBISH -
REMANUFACTURE -
REPURPOSE
Reuse waste
Recovering spare parts of products to
reuse
For warranty products, do you recover spare parts to reuse of products
reported by your customer?
For out-of-warranty products, do you recover spare parts to reuse of
products reported by your customer?
8 RECYCLE Reuse waste
Taking back packaging Do suppliers take back packaging?
Recovering spare parts of products to
resell
For warranty products, do you recover spare parts to sell them of products
reported by your customer?
For out-of-warranty products, do you recover spare parts to sell them of
products reported by your customer?
Valorization of organic waste Is organic waste from your business valorized by composting ?

Cosmopolis – Centre for Urban Research 18
products; and the long-term use of products, through the maintenance and repairing of
products.
Furthermore, Table 1 demonstrates how some operational strategies relate to more than
one R, such as reduce packaging (Refuse, Reduce, Reuse), the long-term use of
products (Rethink, Reuse, Repair), the use of unsold or expired products (Rethink,
Reuse), and reuse waste (Reuse, Refurbish, Remanufacture, Repurpose, Recycle).
3.3.3. Improving the questionnaire with the help of feedback loops
In addition, the questionnaire included open questions to capture practices in a more
inductive way, meaning that we use some broad open questions to collect extra data. We
see what these data bring as new knowledge to refine the questions in a follow up
research.
The questionnaire is further improved with the help of three feedback loops: (1) an expert
group, (2) two pilot tests conducted by the researchers at the scale of the neighbourhood,
and (3) two pilot test conducted by two teams of 10 surveyors of Hub at the scale of the
neighbourhood. Table 2 provides a timetable of the feedback loops used.
The development of the questionnaire was supported by an expert focus group
3
with
specialists on retail, survey methodology and circular economy. This session enabled us
to gather feedback and comments on the first draft of the survey. As a result, we
streamlined and shorten the survey to make it more accessible to retailers.
TIMETABLE
When?
What?
Why?
21 March 2019
Expert focus
group
To insert expertise and receive feedback on the first draft
of the questions.
20 May 2019
Pilot test 1
In the neighbourhood La Chasse, done by the researchers
to check the accessibility of the questions.
22 May 2019
Pilot test 2
In the neighbourhood Madou, done by the researchers to
check the accessibility of the questions.
3
With representatives from Hub, from ULB and from the Chair on Circular economy

Cosmopolis – Centre for Urban Research 19
8 July 2019
Pilot test 3
One team of 10 surveyors executed a first pilot tes in the
neighbourhoods of Matonge, Flagey and Louise to further
polish the questions.
9 July 2019
Pilot test 4
A second team of 10 surveyors executed the second pilot
test in the neighbourhoods of Matonge, Flagey and Louise
to further polish the questions.
Table 2: Timetable of the feedback loops to refine the questionnaire.
With the draft further revised, we conducted ourselves two pilot test around the
neighbourhood of La Chasse and Madou to adjust the survey items, and verify their
clarity. In the last stage and in coordination with hub.brussels, we prepared a two-days
session with the surveyors who carried out the survey. Two teams of 10 surveyors
executed two pilot tests in the neighbourhoods of Matonge, Flagey and Louise. The
feedback received was very positive and allowed us to further polish the survey content.
Twenty-three surveys were completed during this final test and based on these results,
we concluded that the survey was consistent for gaining a good insight into existing retail
practices underpinned by circular economic principles.
3.3.4. The development of the circular scan: measuring circularity
The circular scan that we built for this study, based on the survey’s items, is meant to be
descriptive, qualitative and multi-dimensional. By using the 10 R ladder we intent to
include the ten key circular economy strategies from Refusing to Recovering into the
scan. The questions have been divided according to their corresponding R, and these Rs
have been measured individually across datasets and by isolating the three main
variables: (1) neighbourhood, (2) sectoral mix, and (3) business type. Due to the limited
nature of the survey and its resultant data, some categories (business types and mixes)
have been merged into meaningfully compare scores (otherwise we might have a
category with three retailers and another with thirty retailers, for instance). The results
has been developed through four steps.
3.3.4.1. Step 1: formatting ‘raw’ data
The ‘raw’ data has been formatted as two matrices: simple yes/no questions (filtering
questions), or those that led to an open-ended question were excluded from this matrix,
and thus the calculation of the circular scan.
Likert questions have been graded on a scale from 0-4 (never, sometimes, often,
always). For example, as we demonstrate in Figure 2, when we ask how retailers order
products, more than 50% ‘always’ avoids toxic or dangerous products and only 10%
‘always’ favours the cheapest products.

Cosmopolis – Centre for Urban Research 20
Figure 2: Question: ‘How do you choose your products when you place your order?’.
The question is then turned into a matrix, where we can focus on the level of the
neighbourhood (Anneessens in Figure 3), the level of a sector, and the level of a
subsector. We also added the key circular economy strategies above the circular
economy actions that we ask the retailers. Non-responses are left blank and are not
figured in the calculation.
How do you choose your products when you place your order?
NCA
Reduce
Repair
Rethink
Neighborhood
Name
25. Mix
29. Type
35. You
favour the
cheapest
products
36. You
promote
products
made or
produced in
Belgium
37. You favour
products that
are easily
repairable by
the customer
38. You avoid
toxic or
dangerous
products
ANNEESSENS
EQUIPEMENT DE LA
PERSONNE
Textiles, Chaussure
1
2
2
4
ANNEESSENS
HORECA
Alimentation -
Restaurant - Café
4
0
4
yayANNEESSENS
EQUIPEMENT DE LA
PERSONNE
Alimentation -
Restaurant - Café
4
4
4
4
ANNEESSENS
HORECA
Alimentation -
Restaurant - Café
4
4
4
ANNEESSENS
HORECA
Alimentation -
Restaurant - Café
2
2
4
ANNEESSENS
BIENS DE
QUOTIDIENNETE
Alimentation -
Restaurant - Café
2
4
0
ANNEESSENS
BIENS DE
QUOTIDIENNETE
Alimentation -
Restaurant - Café
ANNEESSENS
SERVICES
Electroménagers /
Electroniques
3
4
3
4
ANNEESSENS
EQUIPEMENT DE LA
MAISON
Electroménagers /
Electroniques
2
2
3
3
ANNEESSENS
EQUIPEMENT DE LA
MAISON
Electroménagers /
Electroniques
2
2
3
3

Cosmopolis – Centre for Urban Research 21
ANNEESSENS
BIENS DE
QUOTIDIENNETE
Alimentation -
Restaurant - Café
ANNEESSENS
BIENS DE
QUOTIDIENNETE
Alimentation -
Restaurant - Café
0
4
4
ANNEESSENS
EQUIPEMENT DE LA
MAISON
Electroménagers /
Electroniques
2
4
4
4
ANNEESSENS
BIENS DE
QUOTIDIENNETE
Alimentation -
Restaurant - Café
0
1
3
4
Figure 3: Matrix extract focusing on the neighbourhood of Anneessens regarding survey
question: ‘How do you choose your products when you place your order?’
Multiple choice questions have been graded on a scale from 0-1 (yes or no)
4
. For
example, in Figure 4, we ask whether the retailers try to reduce packaging. Almost 50%
of the retailers does not take any steps in that direction.
Figure 4: Question: ‘Are you trying to reduce packaging?’.
4
With the exception of Question 68, as respondents could respond with two answers to this question (they could
have both (2), either (1), or neither (0) a device to reduce water consumption or a device to reduce gas).

Cosmopolis – Centre for Urban Research 22
When represented in a Matrix, we can again focus on the level of the neighbourhood
(Anneessens in Figure 5), the level of a sector, and the level of a subsector.
Rethink
Refuse
Reduce
Reuse
Refuse
N/A
Neighborhood
16. Name of
Business
25. Mix
29. Type
59. By
talking to
your
suppliers
?
60. By
changing
suppliers?
61. By
buying
larger
packagin
g?
62. By
reutilizing
supplier
packaging a
second
time?
63. By
selling
products
without
packaging?
64. I do
not take
any
steps in
this
direction
ANNEESSENS
EQUIPEMENT
DE LA
PERSONNE
Textiles,
Chaussure
0
0
0
0
0
1
ANNEESSENS
HORECA
Alimentation -
Restaurant -
Café
1
0
0
0
0
0
ANNEESSENS
EQUIPEMENT
DE LA
PERSONNE
Alimentation -
Restaurant -
Café
1
0
0
0
0
0
ANNEESSENS
HORECA
Alimentation -
Restaurant -
Café
0
0
0
0
0
0
ANNEESSENS
HORECA
Alimentation -
Restaurant -
Café
0
0
1
0
0
0
ANNEESSENS
BIENS DE
QUOTIDIENNETE
Alimentation -
Restaurant -
Café
1
0
0
0
0
0
ANNEESSENS
BIENS DE
QUOTIDIENNETE
Alimentation -
Restaurant -
Café
1
0
0
0
0
0
ANNEESSENS
SERVICES
Electroménagers
/ Electroniques
0
0
0
0
0
0
ANNEESSENS
EQUIPEMENT
DE LA MAISON
Electroménagers
/ Electroniques
0
0
0
0
0
0
ANNEESSENS
EQUIPEMENT
DE LA MAISON
Electroménagers
/ Electroniques
0
0
0
0
0
1
ANNEESSENS
BIENS DE
QUOTIDIENNETE
Alimentation -
Restaurant -
Café
0
0
0
0
0
1
ANNEESSENS
BIENS DE
QUOTIDIENNETE
Alimentation -
Restaurant -
Café
0
0
0
0
0
1
ANNEESSENS
EQUIPEMENT
DE LA MAISON
Electroménagers
/ Electroniques
0
0
1
1
1
0
ANNEESSENS
BIENS DE
QUOTIDIENNETE
Alimentation -
Restaurant -
Café
0
0
0
0
0
1
Figure 5: Matrix extract focusing on the neighbourhood of Anneessens regarding survey
question: ‘Are you trying to reduce packaging?’.
An auxiliary tab containing the open-ended responses per neighborhood has been
included in the spreadsheet (Figure 6). This data is not included in the calculation.

Cosmopolis – Centre for Urban Research 23
Neighborhood
40. Give examples of other criteria
that you favour to order products
DAILLY
Ecological environnement ; fair-
trade
Fraicheur
Le choix du client
Les envie des clients
Peinture
GEORGES HENRI
La facilité de livraison
La proximité
la qualité x5
La sympathie
Les garanties des fournisseurs
Seconde main
HAUT SAINT-GILLES
Artisans locaux
Bio
Ethique
Fraîcheur, qualité/prix
L’origine des produits
La proximité
La qualité des produits (x6)
Le contact, provenance, qualité
Les connaissances
Les tendances
Figure 6: Matrix extract with open-ended responses regarding survey question:
‘Give examples of other criteria that you favour to order products’
3.3.4.2. Step 2: Calculating each R
Two pivot tables have been constructed in the Index Calculation tab: one for the Likert
items, and the other for multiple choice. These values are based on the averages found
in the Likert Matrix and Multiple Choice Matrix tabs, respectively. From here, one can
sort by neighborhood, mix, type, or any combination of the three. Multiple items can be
selected per variable (for instance, Anneessens and Dailly, or Services and Horeca). The

Cosmopolis – Centre for Urban Research 24
values that each table provides is the average of variables per R. For instance, selecting
‘Haut St. Gilles’ in the Likert pivot table will provide the averages of each R category (from
the Likert questions) across respondents in St. Gilles.
3.3.4.3. Step 3: Calculating averages
The averages generated by these pivot tables have been separately compiled as Likert
and Multiple Choice averages below the tables. They are then combined (averaged) to
incorporate both question types into one value. Because the Likert questions from the
survey only reflect some R-values, the averages of NCA (which stands for ‘no circular
activities’), Recycle, Repair, Reduce, and Rethink are significantly higher than the
remaining R-values (Refuse, Reuse, Refurbish, Remanufacture, Repurpose), which are
simply taken from the corresponding Multiple Choice scores.
3.3.4.4. Step 4: Standardization
In order to balance these scales (0-4 and 0-1), this combined average was standardized
using the z-score formula. The z-score is a technic to compare results (of a sector or a
neighborhood) to the average results. The resulting z-score is thus a standardized,
normal distribution
5
of each R-category on a scale of -3 to 3: a positive value means a
neighborhood or mix is more likely to employ that R as a strategy than other
neighborhoods/mixes, and a negative value means it is less likely to use that R than other
neighborhoods/mixes.
Finally, the score calculated in step 4 is transferred to the Synthetic R Index tab.
3.3.5. Methodological conclusion: limits of the circular economy scan
When establishing the circular economy scan, we have identified four main limits:
1. The circularity scan is not self-referential, it is standardized as a normal
distribution. As a consequence, each value is a comparison to the other values in
each R-category. It means that the scan does not show whether or not each
5
This means that each R will have an average of 0 and a standard deviation of 1. In other words, the values from
the combined average table are converted to represent how far each value is from the average. If one
neighborhood has a higher combined score than another, it will have a proportionally higher z-score. This
effectively levels the scales by which each R is measured.

Cosmopolis – Centre for Urban Research 25
variable is, by itself, circular. However, one can see if a neighbourhood, mix, or
type consistently does better in comparison to others. To this end, the values per
variable have been averaged (in the Synthetic R Index tab).
2. The range of questions proposed in the survey is limited. While much work
has gone into the wording and selection of each question, and although the 10 R
framework guides the principles behind each question, there is a disparity between
the frequency of each R-category. To give an example: there are eleven Rethink
questions, but only three questions in the Remanufacture/Repurpose/Refurbish
category. Yet this can be explained by the fact that the Rethink strategy is more
complex and nuanced than the other set of strategies that can be captured more
straightforwardly. It can also partly be explained by the fact that we sampled these
questions in the retail sector, a sector that is by nature more oriented versus
consumption instead of production.
3. As a result, because of the limited range of questions, one can note an
overrepresentation or skewing of the data. For example, if there is only one R
related Likert question, a business that ‘always’ does so – according to the way
the index is calculated – is thus represented as twice as good than one that only
does so ‘sometimes’.
4. There is a disparity between number of respondents in each category which may
skew the data. For example, there are 14 respondents in Anneessens and 41 in
St-Gilles. However, this is managed by weighing the results. Since the current
survey is but a pilot, however, the results are mostly to be treated as a proof of
concept for a circular economy scan. Statistically robust conclusions about
neighbourhood or sectoral differences (and the interaction of both) cannot be
drawn from the current sample.

Cosmopolis – Centre for Urban Research 26
4. RESULTS
4.1. Composition, geography, and validity of the sample
The survey conducted in co-operation with hub.brussels focused on small and medium
enterprises operating in retail in the BCR. We embraced diverse retail sectors following
categories developed by Hub, such as ‘Biens de loisirs’ (or leisure products, among which
are tobacconists, bookshops, photoshops, and toy stores), ‘Biens de Quotidiennete’ (or
daily products, which include grocery stores, bakeries, newsagents, wine shops, fruit
shops, coffee and tea shops, and trattorias), ‘Equipment’ (including first- and second-
hand clothing, textiles and shoes, opticians, florists, furniture, electronics and household
goods, ‘Horeca’ (or hotels, restaurants and cafes), ‘Services’ (including hairdressers and
tattoo parlors, locksmiths, vehicle repair, money transfer agents, phone stores, and travel
agencies), and ‘Transports’ (car salons, tyre service).
While approximately 500 retailers were expected to respond to the survey, which would
have constituted a representative sample of the retail landscape in Brussels, in the final
survey, conducted in August 2019, 20 surveyors working in pairs operated for 10 working
days. 177 workable responses across a variety of sectors and neighborhoods could be
used for the analysis. While 662 retailers were visited, 485 were absent or did not want
to answer the survey. Figure 7 shows the geography of the survey responses.
Figure 7: The geography of survey responses (red dots).
Source: Hub.Brussels

Cosmopolis – Centre for Urban Research 27
4.2. Circular practices of Brussels retailers
In Annex 1 we have put the ‘raw data’ that are collected by hub.brussels. These are the
answers to the questionnaire of the survey of the retailers that correspond with the red
dots on Figure 7 about the geography of the survey responses. The questions are in
French and give a good overview of the circularity of practices of Brussels retailers.
4.3. Insights about the circular practices of Brussels retailers
The results of the survey are analysed from four perspectives: (1) a general presentation
of the results, and a more detailed analysis of the circular economy scan across (2) sector
mixes, (3) business types, and (4) neighbourhoods.
4.3.1. General presentation of the results
These are the key insights and main results obtained from survey, including some of the
open questions of the questionnaire. They are presented according to three tiers
gathering the strategies into three different clusters: the top tier includes strategies that
are higher up on the ladder, the second tier includes strategies related to expanding
lifespan of products, the lowest tier includes en-of-life strategies.
TIER 1: Refuse – Reduce – Rethink
• Most Tier 1 circular economy practices are limited to technical solutions. Energy
reduction strategies are focused on technological rather than behavioural
solutions.
• Using LED lights is the most common way of reducing energy consumption.
• Energy reduction technologies are common across all sectors and
neighbourhoods.
• Rethink strategies in the form of the donation of unsold or expired products are
strongly present in Brussels, and account for half of all end-of-life disposal reported
by retailers.
• 32% of retailers have been using shared transportation with other businesses in
order to reduce packaging and transport.
• Yet, organizing it at the level of suppliers and producers is not a common strategy
to reduce packaging and transport.
• 23% of businesses organize delivery for their customers. Similarly, 75.5% of
businesses’ suppliers package most orders.

Cosmopolis – Centre for Urban Research 28
• More than half of all the businesses approached in the survey take steps to reduce
packaging.
• Locality is a strong factor in both product placement (15% ‘usually’; 18% ‘always’)
and in transport cost reduction (20% and 32.8%, respectively).
• Although 54% of retailers ‘always’ avoid toxic and dangerous products, and many
of them valorise local production, focus on product/service price and quality is
more important.
• 12,4% of retailers inform their customers about social and environmental aspects
of their purchases.
• In specific cases, there are clear signs of refusal behavior, for instance ensuring
minimal waste by calculating demand upstream.
• Very few businesses lease their products.
TIER 2: Re-use – Repair – Remanufacture – Repurpose
• 41% of retailers report sending warranty products back to the supplier; 27% repair
them onsite, and as many as 41% scrap them to resell, recycle, or reuse spare
parts.
• Very few businesses offer recycled or refurbished products.
• Repairability is rarely a key factor for retailers.
TIER 3: Recycle - Recover
• End-of-life strategies for both warranty and out-of-warranty products are in place
across sectors and neighbourhoods.
• Across sectors, product suppliers often take responsibility for taking back
packaging.
• 50% of respondents do not throw out products when they are expired or unsold.
• 40% of businesses do valorise organic waste.
• 77% of retailers across all sectors do not have access to on-site processing
facilities.
These first insights show that retailers have indeed been engaging with some practices
that support the transition to a circular economy such as reducing packaging, repairing
products or valorising organic waste. Yet, the reduction strategies seem to be still

Cosmopolis – Centre for Urban Research 29
prominently focused on technological rather than behavioural solutions and taking place
largely under the third tier, which represent practices contributing to a reduced use of new
resources. Nevertheless, we see positive signs that a larger engagement with circular
practices is possible, for instance through retailers’ interest for favouring local supply
chains, or taking the time to discuss social and environmental impacts of the products
with their clients.
4.3.2. Sector mixes
Looking at the retail mixes scrutinized in the survey, some observations can be made
regarding their performance in terms of transitioning to the circular economy. Figure 10
gives an overview of how much key circular strategies are adopted by the retailers.
Remanufacture, Repurpose and Refurbish have been combined into one category. The
survey demonstrates that the adoptions of circular strategies is low, though there are
variations regarding the sector and the circular strategies. The colour coding of the cells
represents the scores for particular a R from the lowest to highest (resp. red and green).
For example: The sector of ‘personal equipment’ scores badly on repair (-2,04), while the
sector of ‘household equipment’ scores high on Rethink (1,75).
Sector mix
Recover
Recycle
Remanufacture /
Repurpose /
Refurbish
Repair
Reuse
Reduce
Rethink
Refuse
Average
# of respondents in
this category
Leisure goods
1,67
0,96
1,20
-0,30
1,46
-1,33
-0,73
-1,48
-0,03
11
Daily goods
-1,19
-1,38
-0,95
0,70
0,59
1,11
0,03
1,38
0,21
56
Household
equipment
0,95
0,41
1,32
0,51
0,40
0,03
1,75
0,93
0,77
25
Personal
equipment
0,89
0,39
-0,95
-2,04
0,00
-0,86
0,76
0,44
-0,32
31
Horeca
0,08
-1,37
-0,95
0,13
-0,78
1,45
-1,19
-0,46
-0,45
39
Transport and
Services
0,33
0,99
0,31
1,00
-1,66
-0,39
-0,62
-0,80
-0,17
15
The colour coding of cells organizes the scores for particular R from the lowest (in red) to highest (in green).
Remanufacture, Repurpose and Refurbish have been combined into one category.
Figure 10: Survey results for Rs across sectors mixes.

Cosmopolis – Centre for Urban Research 30
‘Household equipment’ does particularly well, scoring high on Rethink (1,75) and
Remanufacture
6
(1,32), and noting a good score on Refuse (0,93), with few values that
dip below the average. These positive results likely derive from the character of goods
and services offered in this sector, which could have high potential for remanufacturing.
This sector might also provide retailers with diverse alternative supply options that allow
them to ‘rethink’ what kind of good and services they want to offer in their store and avoid
fewer circular items. There is a lower score on Repair (0,51) and Reuse (0,40) that are
placed higher on the ladder, and this may be explained by the fact that is easier to recover
spare parts than repairing..
A series of above-average scores can be noted in ‘Leisure goods’, on Recover (1,67),
Reuse (1,46), Remanufacture (1,20) and Recycle (0,96).
‘Daily goods’, which embraces primarily grocery stores, obtains the highest score among
all mixes in Refuse (1,38), yet otherwise notes the lowest scores among all mixes in
Recycle (-1,38), and Recover (-1,19). It also has a below-average result in
Remanufacture (-0,95). These results can be explained by the fact that on one hand it
might be easier in groceries to implement Refuse strategies. On the other hand, Recover
and Remanufacture are less relevant for this mix.
Similarly, ‘Horeca’ has the very same results in this category, further scoring low on
Rethink (-1,19) which suggest the challenge of ‘rethinking’ supply strategies by avoiding
non-circular products and practices, and, less expectedly, the difficulty of engaging in
recycling practices (Recycle: -1,37). Even though this retail mix scores high on Reduce
(1,45).
‘Personal equipment’ obtains a particularly low score in Repair (-2,04). This outstanding
result may be interpreted by saying that this reflects the very high level of consumption of
clothes and shoes with a very short use time expectancy that one see in our society. In
parallel, we see in this retail mix below-average results in Remanufacture (-0,95) and
Reduce (-0,86), and only slightly above-average ones in Refuse (0,44), Rethink (0,76)
and Recover (0,89).
6
For brevity, in the text we will refer to ‘Remanufacture / Repurpose / Refurbish’ as ‘Remanufacture’.

Cosmopolis – Centre for Urban Research 31
Quite logically, ‘Transport and Services’ obtains the highest score among all mixes in
Repair (1,00), as most a reparability of transport and mobility vehicles and instruments is
an important factor for consumers. It is nonetheless surprising to note that the score on
Recycle (0,99) is higher in this mix than in ‘Horeca’ (-1,37) and ‘Daily Goods’ (-1,38).
‘Transport and Services’ holds the lowest results in Reuse (-1,66) and below-average
ones in Reduce (-0,39), Rethink (-0,62) and Refuse (-0,80).
4.3.3. Business types
The analysis of particular business types against the Rs grid of circular practices reveals
a certain advantage of ‘Household Goods and Materials’ in terms of engagement in
circular practices, as noted in Figure 11.
Business types
Recover
Recycle
Remanufacture /
Repurpose /
Refurbish
Repair
Reuse
Reduce
Rethink
Refuse
Average
# of respondents in
this category
Groceries–
Restaurants–
Cafes–Florists
-0,25
-0,88
-1,10
0,49
0,33
1,69
-0,94
1,56
0,11
102
Mobility
1,10
-0,59
0,33
0,43
-1,48
-0,55
-0,95
-1,23
-0,37
7
Household Goods
and Materials
0,66
1,68
1,48
0,79
1,30
-0,25
1,42
-0,13
0,87
31
Accessories
-1,51
-0,21
-0,70
-1,72
-0,15
-0,89
0,47
-0,19
-0,61
37
The colour coding of cells organizes the scores for particular R from the lowest (in red) to highest (in green).
Figure 11: Survey results for particular Rs obtained across business types.
This business types notes the highest scores of all in 5 out of 8Rs: Recycle (1,68),
Remanufacture (1,48), Rethink (1,42), Reuse (1,30) and Repair (0,79). It does not come
as a surprise, given the sectors that the sectors it combines—’Bricolage, construction’;
‘Electroménagers, Electroniques’; ‘Jouets, Articles cadeaux’; and ‘Meubles’ —
traditionally offer consumers wide options of product reparability and reuse. It is, however,
interesting to note that this business type also scores high on Rethink, apparently allowing
retailers to consider relatively circular supply options. In Recover this business type is
surpassed by Mobility (1,10), a business type that otherwise notes rather surprisingly low
scores in Reuse (-1,48), Refuse (-1,23) and Rethink (-0,95). ‘Groceries–Restaurants–
Cafes–Florists’ is a category of retailers that are likely to produce predominantly organic

Cosmopolis – Centre for Urban Research 32
waste. Therefore, it is not surprising that they note the lowest scores among all business
types in Recycle (0,88) and Remanufacture (-1,1), while noting the highest scores in
Reduce (1,69) and Refuse (1,56). ‘Accessories —a type that combines ‘Opticiens,
bijouteries, cordonniers’, and ‘Textiles, Chaussure’—notes the lowest average score
among all categories (-0,61), obtaining the lowest ones in Repair (-1,72) and Recover (-
1,51). However, it should be noted that the number of responses is uneven across
business types: from 102 in ‘Food and Florists’ through 37 in ‘Accessories and 31 in
‘Household Goods and Materials’ to only 7 in ‘Mobility’).
Strengths within business types seem to correlate strongly with their parent mixes, where
retail mixes are associated with business types (perhaps unsurprisingly, as types are
largely just a subset of mixes). For instance, the ‘Household Goods and Materials’ type-
group and ‘Equipment de la Maison’ mix do particularly well, with few values that dip
below the average (see Table 2 below). While the general profile may be comparable
across types and their parent mixes, the specific strengths and weaknesses differ. While
both the ‘Horeca’ mix and ‘Alimentation, etc.’ type both excel in the Refuse category, for
instance, the latter eclipses the former in Refusal.
4.3.4. Neighbourhoods
What emerges from the survey is that no neighbourhood stands out as particularly visible
hub for circular retail activities or, inversely, an area where hardly any circular retail
activities are present, as noted in Figure 12. However, across neighbourhoods within the
given dataset, Anneessens — arguably the most working-class, ethnically diverse, and
precarious neighbourhood explored in the survey — consistently ranks higher than all
others in nearly every R-category.
Neighborhoods
Recover
Recycle
Remanufacture /
Repurpose /
Refurbish
Repair
Reuse
Reduce
Rethink
Refuse
Average
# of respondents in
this category
ANNEESSENS
1,3
0,7
2,3
2,0
0,0
1,8
1,3
-0,7
1,10
14
DAILLY
0,3
-0,6
-0,2
-0,4
-1,6
0,2
-1,2
0,2
-
0,43
28
GEORGES HENRI
-0,5
-0,5
0,2
0,4
0,1
-0,9
0,0
-1,1
-
0,28
38
HAUT SAINT-
GILLES
0,8
-0,4
-0,6
0,6
1,6
0,3
-1,1
0,0
0,14
41

Cosmopolis – Centre for Urban Research 33
QUARTIER
BRABANT
0,8
-1,1
-0,6
-1,1
-1,0
-1,6
-0,3
2,2
-
0,35
20
ROODEBEEK -
CONSTELLATIONS
-1,2
2,1
-1,0
-0,9
0,8
0,3
1,6
-0,6
0,14
13
VIEUX LAEKEN
EST
-1,5
-0,2
0,0
-0,5
0,2
-0,1
-0,3
0,1
-
0,31
23
The colour coding of cells organizes the scores for particular R from the lowest (in red) to highest (in green).
Figure 12: Survey results for particular R’s (circular practices) obtained across
neighbourhoods.
The neighbourhood of Anneessens scores particularly high for Remanufacture (2,3),
Repair (2,0) and Reduce (1,8). This score sits at around two standard deviations above
the norm, whereby Anneessens obtains the best results of all neighbourhoods examined.
Anneessens also leads in Recover (1,3). The only negative score obtained in the
neighbourhood is in Refuse (0,7). As a result, Anneessens has an overall good result
because it scores very high on both Repair and Remanufacture strategies. But at the
same time, it has a very high score of non-circular activities so this is at odds and is
difficult to explain. This is a reason why we want to explore further the survey result by
organizing feedback sessions with retailers.
However, Quartier Brabant, a neighbourhood that has a similar social composition in
terms of social class, notes a high score in only one particular category, Refuse (2,2), and
an average one in Recover (0,8), while noting negative scores in all other categories. The
high score in Refuse strategy is explained in the dataset by the fact that there are many
shops selling in bulk. However, the shops in question are in the mix ‘textile/shoes’ which
does not matching (one would have expected Groceries instead). This score is therefore
quite difficult to interpret and indicate that we should be cautious with the results overall.
Another working-class and ethnically diverse neighbourhood, Vieux Laeken-Est, obtains
even lower scores, getting the lowest score of all neighbourhoods in Recover (-1,5), and
near-0 scores in all other categories. The highest score for Recycle is was obtained in
Roodebeek (2,1), a middle-class neighbourhood in which besides Rethink (1,6) and
Reuse (0,8), all scores are negative. Similarly, the predominantly upper middle-class Haut
Saint-Gilles leads in Reuse (1,6), yet besides Recover (0,8) notes negative scores. Dailly
stands out due to its particularly low scores in Reuse (-1,6) and Rethink (-1,2), and scores
oscillating around 0 in the remaining categories.
Although these results could be caused by an overrepresentation in particular
neighborhoods of certain retail mixes which themselves excel more in these categories,
this does not hold up to scrutiny. For instance, in Anneessens, the sectors such as

Cosmopolis – Centre for Urban Research 34
household equipment and leisure goods, which outperform generally in Remanufacture
and Repair, in which where the neighbourhood scored quite high– namely, household
equipment and leisure goods – respectively make up only 20% and 0% of Anneessens’
respondents. This lack of correlation can be seen in other neighborhoods as well. In
Quartier Brabant, for instance, a neighborhood in which half of respondents fall under the
‘personal equipment’ retail mix, Refusal is the only R in which the neighborhood tops the
others. Respondents in this area fall behind the other neighborhoods in nearly every other
R category, often by more than one standard deviation. However, the personal equipment
category does not reflect these trends in comparison to others: across R-categories, the
mix averages around only 0.06 standard deviations below the mean – compared to
Quartier Brabant’s 0.35 standard deviations below – and its Refusal score is middling at
around 0.47. In general, the presentation of the scores have to be taken with caution as
there are very few significant results (which are those outside -1.96 and + 1.96).
Further, because the index calculation does not include open-ended auxiliary questions,
we may be missing important pieces of the circular puzzle. For instance, when looking at
scores obtained in Haut St. Gilles, it does not rank particularly well among its peers.
However, the open responses offered by St Gilles respondents are both more numerous
(perhaps on account of its large number of respondents relative to other neighborhoods)
and more circular, and offer great hints at what is happening on the ground. Similarly, it
is difficult to specify to what extent the scores obtained in specific business categories
and neighbourhoods depend on the amount of responses obtained, which is uneven.
While the data obtained in the survey allows to compare different neighbourhoods and
retail categories and establish those engaging in circular activities to a larger or smaller
extent, it is difficult to measure this extent objectively. Despite obtaining high scores,
specific neighbourhoods and retails categories are not where circular economy is indeed
being advanced, for they are simply less bad than others.

Cosmopolis – Centre for Urban Research 35
5. DISCUSSION
5.1. A circular economy scan ‘in the making’
Although there are differences between variables for particular sector mixes, business
types and neighbourhoods, it is difficult to say whether these variables independently
cause differences in scores or merely correlate with them. It can be said, however, that a
larger or smaller number of respondents does not, according to this dataset, correlate
with a higher or lower score. Therefore, differences of scores across sector mixes,
business types and neighborhoods do not result from large differences in the number of
respondents. With few exceptions, there also does not appear to be a strong correlation
within neighborhoods or sectors across R-values. These results could thus reflect a
relatively homogenous noncircular economy situation across the BCR; alternatively, it
could stem from a small sample of respondents.
The sort of analysis made possible by this deconstruction of the data hints at the state of
the circular economy as it currently exists in Brussels. Looking at detailed and parsed
results, it is possible to recognize which variables have stronger effects on the circularity
of a given subset. However, one must critically recognize what the data actually
represents. Where a neighborhood like Anneessens appears to shine bright, it is rather
that the businesses within this neighborhood have responded more favourably towards
questions that fall under each category. It is not sufficient to look at the results and assert
that this neighborhood is much better at repairing than another, or that this mix practices
remanufacturing more than another, because the data is easily skewed by a number of
factors. This can be explained among others by the necessary reductionism that
accompany the construction of the circular economy scan. Even if each value accurately
reflects the R-category it represents, the categorization of the survey’s questions
(necessarily) represents only parts of the circular economy as we conceptualize it.
Therefore, understanding what the values represent is imperative to analysing the data
critically. The framework by which we construct this circular economy scan – as well as
the foundation behind the survey – is a step in the right direction. The survey as it exists
now was constructed in a way to highlight existing circular economy practices, and to
investigate the viability of metrics alternative to existing datasets, which currently do not
allow to identify circular economic activities. To this extent, it has largely served its
purpose. Many respondents have submitted responses that hint at a range of circular or
at least sustainable practices in their businesses. Analyzing the data allowed us to identify
where strengths and weaknesses exist, and sheds light on the behavior, circular or
otherwise, of businesses.

Cosmopolis – Centre for Urban Research 36
5.2. A pilot for a larger study: further research
To further our understanding of BCR’s circular reality, what will be needed in subsequent
investigations is a more robust dataset and a broader range of questions. The takeaways
from this survey and its data include not only the quantitative representation of practices
as they currently exist, but also the many responses to open-ended questions, which
illustrate that there are certain practices which might be replicated or encouraged across
the region. Metrically expressing the circularity of businesses, and isolating variables that
shape these realities, will continue to develop as we move forward with our existing
dataset.
Some ways to improve the analysis of a next survey results include alternative technics
for standardization or for calculation. A complementary and necessary approach is to
elaborate a larger scale study, covering more broadly the different strategies within the
survey questions to allow better comparison between strategies. In addition, a future
survey should be more equally distributed in terms of respondents per neighborhood, per
type and mix. Neighborhood distribution is particularly important for the results to be
interpreted spatially.
Overall, the survey constitutes a step towards developing a CE scan based on indicators
that could be included, after some improvements, in future data collection methodology
of regional administrative bodies responsible for retail and commerce (hub.brussels),
transport (Bruxelles Mobilité) and food production and consumption (Bruxelles
Environnement), and consequently incorporate in existing databases on Brussels’
businesses. These indicators will aim at detecting and measuring retailers’ practices, to
consequently assess economic activities vis-à-vis several circular practices or ‘Rs’
(Rethink, Refuse, Reduce, Reuse, Recycle, Repair, Refurbish, Remanufacture,
Repurpose).

Cosmopolis – Centre for Urban Research 37
6. Conclusion
This is a research report on the methodology of defining the circular economy, of
measuring the circular economy, and of describing and evaluating the first steps in
creating a data tool to scan the circular economy. More specifically, we aimed to identify
and assess the adoption of circular principles by retailers, identified as locally embedded
intermediaries of the circular economy, and operating between broad policies and
individual consumers.
The first objective of this report was to identify and analyse the circularity of existing
economic retail practices in Brussels. To this end, we transformed the theoretical
framework of the 10R ladder of circular economy practices into an operational definition,
that functioned as an underlayer to create the survey questions for the retailers. The
questions were improved with the help of several feedback loops. Twenty surveyors of
hub.brussels working in pairs operated for 10 working days and 177 workable responses
across a variety of sectors and neighborhoods could be used for the analysis.
The second objective was to investigate where and what differences might exist between
businesses, neighbourhoods, and sectors within the city-region. Although the circularity
scan in its present form shows differences between sectors and neighbourhoods, the
samples are too still small and unevenly distributed across neighbourhoods and retail
sectors to be able to talk of robust results. Statistically robust conclusions about
neighbourhood or sectoral differences (and the interaction of both) cannot be drawn from
the current sample. More extensive data collections would lead to more robust patterns
that can be analysed meaningfully based on the circular economy scan.
The third objective of this report was to generate new data on circular economy practices
in the BCR that focused on smaller-scale activities. The starting hypothesis of this report,
therefore, was that long-held mundane re-using and valorisation practices were already
taking place and hence participate to the circularity of Brussels. To test this hypothesis,
we worked with the main research question: to what extent smaller-scale bottom-up
activities that may qualify as ‘circular practices’ are already taking place throughout
Brussels? Although the sample of 177 retailers is but a fraction of the Brussels retail
sector, the results are somewhat disappointing. While our results show that retailers
indeed have practices that support the transition to a circular economy, to a large extent
circular practices have not entered retail yet. The policy conclusions following these new
data of the circularity scan of the sample of 177 respondents in the retail sector
demonstrate that this sector still has large potential to become more circular.
The fourth and last objective was to propose a roadmap for an alternative to the existing
datasets which currently do not allow to identify circular economy activities. This is a

Cosmopolis – Centre for Urban Research 38
methodological conclusion on the development of the tool of the circular economy scan
A further elaboration of this scan should consider the following main limits and try to
overcome them: (1) the fact that the circularity scan can only be used to compare
neighbourhoods, businesses, and sectors, since the scan is not self-referential and does
not show whether a variable is by itself circular. A further elaboration of the scan should
also go deeper into (2) a more balanced disparity of questions related to the key circular
strategies (cf. the 10 Rs). Now some key strategies are over- or underrepresented due to
the presence of more or less survey questions, and because of the uneven number of
respondents per neighbourhood. Further research could also focus more on qualitative
research, where ethnographic observations of retailer-consumer-relations and practices,
in combination with short semi-structured depth-interviews of retailers, could extend the
list of possible circular operational strategies and actions. As such, the circular economy
scan, can become more accurate.

Cosmopolis – Centre for Urban Research 39
7. REFERENCES
Adam, S., Bucker, C., Desguin, S., Vaage, N. and Saebi, T. (2017). Taking Part in the
Circular Economy: Four Ways to Designing Circular Business Models. Available at SSRN:
https://ssrn.com/abstract=2908107.
Bernon, M., Tjahjono, B. and Ripanti, E. F. (2018). Aligning retail reverse logistics practice
with circular economy values: an exploratory framework. Production Planning & Control,
29(6), 483–497.
Bruxelles Economie et Emploi. (n.d.). Questionnaire relatif  l'adquation de votre activit
aux critres de reconnaissance des entreprises impliques dans l'conomie circulaire.
retrieved from/ http://werk-economie-
emploi.brussels/documents/16195/2574667/190322+-+Formulaire+-
+ANNEXE+Obligatoire+-
+Questionnaire+critère+de+reconnaissance+Economie+Circulaire.pdf/e8ab4a23-ccdf-
4954-8cf5-2697500e3750
Christis, M., Athanassiadis, A. & Vercalsteren, A. (2019). Implementation at a city level of
circular economy strategies and climate change mitigation — the case of Brussels.
Economy and Society Journal of Cleaner Production, 218, 511–520.
Dangana, Z., Pan, W. and Goodhew, S. (2012). Delivering sustainable buildings in retail
construction In: Smith, S.D. (Ed.). Procs 28th Annual ARCOM Conference, 3-5
September 2012, Edinburgh, UK, Association of Researchers in Construction
Management, 1455-1465.
De Schoenmakere, M., Hoogeveen, Y., Gillabel, J., Manshoven, S. and Dils, E. (2019).
Paving the way for a circular economy: insights on status and potentials. European
Environmental Agency. Retrieved from: https://www.eea.europa.eu/publications/circular-
economy-in-europe-insights#tab-news-and-articles
Dinarès, M. (2014). Urban Metabolism: A review of recent literature on the subject.
Documents d'Anàlisi Geogràfica, vol: 60 pp: 551-571.
EASAC: European Academies Science Advisory Council. (2016). Indicators for a Circular
Economy.Retrieved from
https://www.easac.eu/fileadmin/PDF_s/reports_statements/Circular_Economy/EASAC_I
ndicators_web_complete.pdf

Cosmopolis – Centre for Urban Research 40
Elia, V., Gnoni, M. G., & Tornese, F. (2017). Measuring circular economy strategies
through index methods: A critical analysis. Journal of Cleaner Production, 142, 2741–
2751.
EMF. (2015). Circularity Indicators: An approach to measuring Circular Economy.
Retrieved from: https://www.clmsostenible.es/wp-content/uploads/2019/02/Circularity-
Indicators_Project-Overview_May2015.pdf
EMF. (2020). Circulytics Question Indicator List. Retrieved from:
https://www.ellenmacarthurfoundation.org/assets/downloads/Circulytics-question-
indicator-list.pdf
Gasparatos, A., El-Haram, M., & Horner, M. (2008). A critical review of reductionist
approaches for assessing the progress towards sustainability. Environmental Impact
Assessment Review, 28(4-5), 286–311.
Herva, M., Franco, A., Carrasco, E. F., & Roca, E. (2011). Review of corporate
environmental indicators. Journal of Cleaner Production, 19(15), 1687–1699.
Istudor, L.-G. and Suciu, M.-C. (2020). Bioeconomy and Circular Economy in the
European Food Retail Sector. European Journal of Sustainable Development, 9(2), 501.
Janik, A. & Ryszko, A. (2019). Circular economy in companies: an analysis of selected
indicators from a managerial perspective. Multidisciplinary Aspects of Production
Engineering, 2(1), 523-535.
Jones, P. & Comfort, D. (2017). Towards the circular economy: A commentary on
corporate approaches and challenges. Journal of Public Affairs, 17(4).
Jones, P. & Comfort, D. (2018) The Circular Economy and the Leading European
Retailers: A Research Note. European journal of Sustainable Development Research, 2
(2). pp. 1- 8.
Kirchherr, J., Reike, D. and Hekkert, M. (2017). Conceptualizing the circular economy: An
analysis of 114 definitions. Resources, Conservation and Recycling, 127, 221-232.
Kristensen, H. S., & Mosgaard, M. A. (2020). A review of micro level indicators for a
circular economy – moving away from the three dimensions of sustainability? Journal of
Cleaner Production, 243, 118531.
Longaray Moraga, G., Huysveld, S., Mathieux, F., Blengini, G. A., Alaerts, L., Van Acker,
K., De Meester, S., et al. (2019). Circular economy indicators : what do they measure?
Resources conservation and recycling, 146, 452–461.