Waste Management 49 (2016) 505–515

Contents lists available at ScienceDirect

Waste Management
journal homepage: www.elsevier.com/locate/wasman

A consumption approach to wastes from economic activities

Antoine Beylot a,⇑, Baptiste Boitier b, Nicolas Lancesseur b,c, Jacques Villeneuve a
a
BRGM, BP 6009, 3 av. C. Guillemin, 45060 Orléans Cedex, France
b
SEURECO, 9 Rue de Châteaudun, 75009 Paris, France
c
Université Paris 1 Panthéon-Sorbonne, 106-112 Boulevard de l’Hôpital, 75647 Paris, France

a r t i c l e i n f o a b s t r a c t

Article history: In a context of waste management policies aimed at promoting waste prevention and recycling and,
Received 27 April 2015 conversely, reducing waste landfilling, this study investigates how waste is generated and treated in a
Revised 11 December 2015 consumption perspective. A Waste Input–Output Analysis is implemented that considers 14 waste
Accepted 18 January 2016
fractions and four waste management techniques. Input–Output Tables extended to wastes are initially
Available online 2 February 2016
compiled for the year 2008 considering France and five of its main import suppliers, and further
completed with data on waste treatment. Wastes from economic activities are accordingly reallocated
Keywords:
to the product categories of household consumption that induce their production. In particular, consid-
Input–Output Analysis
Household consumption
ering five waste categories (dry recyclable wastes, mixed wastes, mineral wastes, organic wastes, and
Waste generation total wastes) as an aggregation of the 14 waste fractions studied, the ten product categories with the
Recycling highest contribution account for 64–86% of the total generation of wastes. Waste intensity and volume
Incineration of expenses are analyzed as the drivers for the amounts of wastes induced by each product category.
Landfilling Similarly, the products responsible for the largest amounts of waste landfilling and incineration without
energy recovery, i.e. the management techniques at the bottom of the ‘‘waste management hierarchy”,
are identified. Moreover, this study highlights the relative importance of waste produced abroad as
compared to that produced in France, regarding the total amount of waste induced by French household
consumption. The sensitivity of results to the modeling of import production is subsequently discussed.
Finally, the potential perspectives for this type of consumption approach are considered with respect to
its utility and current limitations in a context of waste policy planning, and more particularly regarding
the way waste policy targets are set.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction following targets in particular are set: a 10% reduction in Municipal

Solid Waste (MSW) generation per inhabitant by 2020 as
Integrated Product Policies (IPP) aim at reducing the impacts compared to 2010; a 30% reduction in non-hazardous non-inert
related to products by considering all the phases of their life- waste as of 2020 (achieving 50% by 2025) as compared to 2010;
cycle and taking action where it is most effective. IPP stand for a and the valorization of 55% of non-hazardous non-inert wastes
combination of a number of policy instruments, including by 2020 (60% by 2025) (Assemblée Nationale, 2014).
extended producer responsibility, eco- and energy labeling, inte- The emissions (in a broad sense, including solid waste genera-
gration of environmental aspects into standardization, directives tion) and primary resources consumption of an economy can be
on products and wastes, etc. (European Commission, 2014). taken into account following two perspectives. In the first of these,
Regarding the latter aspect in the European Union (EU), the Waste the producing economic agent is considered responsible for the
Framework Directive (2008/98/CE) states that waste legislation emissions and consumptions involved during the production of
and policy of the Member States shall apply the ‘‘waste manage- goods and services. This is the perspective used to set part of the
ment hierarchy” as an order of prioritization. Accordingly, in quantified waste policy targets. In the second perspective, emis-
France, specific legislations have set quantified objectives to be sions and resource consumption are considered to be induced by
achieved by 2020 with respect to increased waste prevention and the demand for goods and services, and are allocated accordingly
recycling and, conversely, to reducing waste landfilling. The (Munksgaard and Pedersen, 2001). In the context of IPP planning,
several studies implemented the latter consumption-perspective
in order to calculate the environmental impacts induced by the
⇑ Corresponding author.
final demand and to identify the products to be prioritized from
E-mail address: [email protected] (A. Beylot).

http://dx.doi.org/10.1016/j.wasman.2016.01.023
0956-053X/Ó 2016 Elsevier Ltd. All rights reserved.
506 A. Beylot et al. / Waste Management 49 (2016) 505–515

an environmental perspective (Tukker et al., 2004; Palm et al., Z d : matrix of domestic intermediate consumptions (n categories
2006; Weidema et al., 2006). These studies focused on environ- of products
n industries). The term ‘‘industry” here refers to
mental impacts, including global warming or human toxicity, for ‘‘a group of establishments engaged in the same, or similar,
example. When included, wastes were only taken into account at kinds of production activity” (Eurostat, 2008)
a relatively aggregated level. Some additional studies specifically Z m : matrix of imported intermediate consumptions (n cate-
assessed the wastes embodied in products distinguishing between gories of products
n industries)
several waste fractions, respectively in Portugal (Barata, 2002), yd : vector of final demand for domestic products (n categories of
Japan (Nakamura and Kondo, 2009), the USA (Court, 2012) and products
1)1
Taïwan (Liao et al., 2015). In France, at a time where debates are
ym : vector of final demand for imported products (n categories
underway concerning how policies should be designed in order
of products
1)
to support ‘‘sustainable consumption” (CAS, 2011), several recent
studies scrutinized environmental impacts in a consumption per- Ad : technology matrix of domestic input coefficients (n cate-
spective focusing on household consumption and disregarding gories of products
n industries)
other final demand categories (CGDD, 2011; BIO Intelligence Am : technology matrix of imported input coefficients (n cate-
Service, 2012; ADEME, 2014). gories of products
n industries)
In this context, this study aims at assessing waste generation Ac : technology matrix of country C – total coefficients (i.e.
and treatment induced by French household consumption. It dif- domestic products + imports; n categories of products
n
fers mainly from the available literature according to four aspects: industries)
(1°) the implementation of a more disaggregated framework, in wd : matrix of domestic multipliers of waste generation (f waste
terms of products, activities (including waste treatment activities) fractions
n industries)
and waste fractions; (2°) the use of a relatively large range of wc : matrix of multipliers of waste generation associated with
activity-specific data both in terms of waste generation and treat- country C (f waste fractions
n industries)
ment; (3°) its specific application to France; and (4°) the distinction W d : matrix of domestic waste generation induced by the
between waste produced domestically and waste produced domestic final demand (f waste fractions
n categories of
abroad. This study firstly makes it possible to determine ‘‘priority products)
products” to be targeted by French waste policies. These corre- W c : matrix of waste generation induced in country C by the
spond to the products inducing the largest amounts of waste in final demand (f waste fractions
n categories of products)
economic activities (both in France and abroad) as well as to those W t : matrix of total waste generation induced (domestically and
inducing the largest amounts of waste landfilling and incineration abroad) by the final demand (f waste fractions
n categories of
without energy recovery (i.e. those treatments at the bottom of the products)
‘‘waste management hierarchy”). Secondly, this study addresses ai : column vector of the share of imports from country i over
the sensitivity of results with respect to hypotheses on the produc- the total of French imports (n categories of products
1)
tion of imports. Finally, the potential perspectives for such a con-
X
sumption approach are discussed regarding its utility and current with ai ¼ 1 ð1Þ
limitations in a context of waste policy planning. i

2. Method S: activity-specific matrix of allocation of waste to treatments

(of dimensions: in line, f waste fractions
t treatment meth-
2.1. Input–Output Analysis ods; in column, n industries)
T: matrix of waste treatment induced by the final demand (of
2.1.1. Calculation framework for waste generation embodied in the dimensions: in line, f waste fractions
t treatment methods;
final demand in column, n categories of products)
Economic Input–Output Analysis has long been used to calcu- 0 1
1
late the supply chain impacts of producing goods and services, B .. C
In
1 ¼ @ . A ðn categories of products
1Þ
both in economic and environmental terms (Leontief, 1970; Lave
et al., 1995; Joshi, 1999). The framework for Input–Output calcula- 1
tions, including the distinction between domestic production and
imports, is recalled below (Cadarso et al., 2012; BIO Intelligence hi: operation of diagonalization of a column vector
Service, 2012). The following set of equations aims at expressing ⁄: element-by-element multiplication of two matrices
the production (respectively domestic and abroad) as a function
of the final demand in domestic and imported products (Eqs. (5) Input–Output Analysis is based on the economic identity
and (11)). The corresponding effects of final demand on waste gen- between the total domestic output of an industry and the sum of
eration can then be deduced (Eqs. (6) and (12)). Wastes from eco- the demand for that industry’s output from other domestic indus-
nomic activities are accordingly reallocated to the product tries plus the final demand (Eq. (2)):
categories of the final demand that induce their production. A
graphical representation of the calculation framework is presented xd ¼ Z d In
1 þ yd ð2Þ
in Fig. 1, as a simplified version of that depicted by Court et al.
Eq. (2) can be rewritten as:
(2015) in the case of hazardous waste. The following notations
are considered:
xd ¼ Ad xd þ yd ð3Þ
d
x : column vector of domestic production (of dimensions n cat-
egories of products
1)
xm : column vector of imports (n categories of products
1)
X m : column vector of production abroad required to supply 1
We consider in the following equations that final demand is only of one kind
imports (n categories of products
1) (household consumption).
 A. Beylot et al. / Waste Management 49 (2016) 505–515 507

Fig. 1. Generation of wastes from economic activities as a response to household consumption: a graphical representation of the Input–Output framework.

h 1
i
Considering: W c ¼ wc ðI  Ac Þ1 Am ðI  Ad Þ hyd i þ hym i ð12Þ
!
Z d ði; jÞ
Ad ¼ Z d hxd i1 i:e: Ad ði; jÞ ¼ ð4Þ The total waste production in response to final demand can
xdj subsequently be calculated, distinguishing domestic wastes from
those generated abroad:
Hence the calculation of domestic production induced by the
final demand for domestic products: Wt ¼ Wd þ Wc ð13Þ
d 1 d In a first approach in this study it is assumed that production
xd ¼ ðI  A Þ y ð5Þ
abroad is identical to domestic production, both in terms of input
The corresponding effects in terms of domestic waste genera- coefficients and waste multipliers:
tion can finally be calculated as:
Ac ¼ Ad þ Am ð14Þ
d d 1
W ¼ w ðI  A Þ hy i
d d
ð6Þ wc ¼ wd ð15Þ
Eq. (6) provides the amount of waste generated domestically by
However in that case the specificities related to the production
industries as a response to final demand for domestic products. It
of imports in terms of intermediary consumptions and waste gen-
does not account for the amount of wastes originating directly
eration are not accounted for. Consequently, a second modeling
from household consumption as post-consumer waste (their calcu-
approach is implemented, considering specific input coefficients
lation would require additional specific modeling) nor for waste
and waste multipliers as to several countries (r in Eq. (16)) supply-
treatment (that are accounted for later via Eq. (17)). As in Eq. (2),
ing imports to the country under study (France in the following).
Eq. (7) represents the identity between supply and consumption
The generation of waste abroad as induced by the total final
in the specific case of imports:
demand in domestic products and imports is accordingly recalcu-
xm ¼ Z m In
1 þ ym ð7Þ lated in a multiregional ‘‘unilateral” approach by use of Eq. (16):

Considering: X
r h 1
i
! Wm ¼ wi ðI  Ai Þ1 h/i i Am ðI  Ad Þ hyd i þ hym i ð16Þ
m m d 1 m Z m ði; jÞ i¼1
A ¼ Z hx i i:e: A ði; jÞ ¼ ð8Þ
xdj In the following the corresponding results are presented as a
sensitivity analysis to the hypothesis on imports. It should be
Eq. (7) can be rewritten as: noted that this study is limited to the multiregional unilateral cal-
culation approach only, and thus does not account for feedback
xm ¼ Am xd þ ym ð9Þ
effects in a multiregional ‘‘multilateral” approach (e.g. supply of
d French imports to Germany that exports to France). Indeed, Euro-
According to Eq. (5), domestic production x can be expressed as
a function of the final demand for domestic products. Consequently stat Input–Output Tables are used in the following, instead of
the imports can be calculated as a function of the final demand for existing multiregional tables, in order to enable the construction
both domestic and imported products: of tables of waste multipliers for French import suppliers (see
1
details in Section 2.2). Considering the multiregional unilateral
xm ¼ Am ðI  Ad Þ yd þ ym ð10Þ approach in this study implies poorer accuracy of the results (the
multiregional multilateral approach better represents the true
Considering that these imports are supplied by one country
value of wastes induced by the final demand), but at the same time
only as an approximation, the corresponding production abroad
enables a better precision of the results (considering the statistics
to provide them can be rewritten as:
of waste generation currently available, waste multipliers are more
1
X m ¼ ðI  Ac Þ1 ½Am ðI  Ad Þ yd þ ym  ð11Þ precise in the chosen unilateral framework than they would have
been in the multilateral one).
On the one hand, a share of this production is required as a
response to the direct final demand for imports (term 2.1.2. Calculation framework for waste treatment embodied in the
ðI  Ac Þ1 ym ). On the other, the remaining share corresponds to final demand
the content in imports of the final demand for domestic products, Waste Input–Output Analysis additionally distinguishes waste
via the intermediary consumptions of imports by economic activ- treatment activities with accounting for their different technolog-
ities. Hence the generation of waste abroad as induced by the total ical performances and environmental impacts as a function of
final demand in domestic products and imports is: the nature of waste fractions (Nakamura and Kondo, 2002;
508 A. Beylot et al. / Waste Management 49 (2016) 505–515

Takase et al., 2005). As an extension to the conventional Waste products and imports (respectively yd and ym ) are drawn from
Input–Output Analysis framework, in which the ‘‘allocation pat- Eurostat in monetary terms. In this study household consumption
tern” of wastes to treatments is defined with respect to each speci- is additionally considered in terms of the COICOP nomenclature
fic waste fraction independently of sectors of the economy, Beylot (‘‘Classification of Individual Consumption According to Purpose”),
et al. (2015) additionally considered the rates of treatment and so that the results of waste generation and treatments embodied in
recycling to be both waste and activity-specific. In the following, the household consumption are expressed not only in a ‘‘nomen-
the latter approach is implemented, considering only the quanti- clature of activities” but also in a ‘‘nomenclature of consumption”.
ties of wastes to treatments but not their relative environmental The correspondence from CPA to COICOP is performed using allo-
impacts. Consequently, waste quantities to treatments are dealt cation keys, as extracted from (BIO Intelligence Service, 2012). It
with in a similar manner as with waste generation as an approxi- should be noted that the latter allocation keys are only an approx-
mation in this study (i.e. as an extension to the technology matrix). imation based on the EIPRO study at the scale of the EU-25 (JRC,
Waste treatment (including recycling) is not fully integrated 2006), meaning that the results presented in the following in COI-
within Input–Output Tables, meaning that the model considered COP nomenclature involve greater uncertainty than those pre-
is a limited version of standard Waste Input–Output Analysis: it sented in CPA nomenclature.
only enables gross waste generation to be accounted for, as Fourteen waste fractions (mostly non-hazardous) for which the
opposed to net waste generation (i.e., net of recycling), and does nomenclature is defined in the Eurostat manual on waste statistics
not consider waste transformation in treatment processes (e.g. (Eurostat, 2011) are considered in the study (Table 1). These
organic waste into bottom and fly ashes through incineration). include dry recyclables, equipment, discarded vehicles, batteries
The total quantities of waste to treatments as induced by the total and accumulators wastes, mineral wastes, wood wastes, animal,
final demand in domestic products and imports are accordingly vegetal and mixed food wastes, household and similar wastes
calculated as follows: and mixed and undifferentiated materials. These waste fractions
1
are targeted by French policies relating to waste prevention and
T ¼ ðS  Dwd ÞðI  Ad Þ hyd i treatment, generally at an aggregated level (e.g. Municipal Solid
h 1
i Waste). The matrix of domestic multipliers of waste generation
þ ðS  Dwc ÞðI  Ac Þ1 Am ðI  Ad Þ hyd i þ hym i ð17Þ
wd is derived from data of waste generation by economic activities
The activity-specific matrix S of dimensions f waste fractions
t in France in 2008, as drawn from three statistical sources:
treatment methods (in line) by n industries (in column), represents
the share of waste fraction p which is treated using treatment – firstly, statistics of waste generation from industrial activities
method k, as a function of the industry j. The shares of treatment are taken from (INSEE, 2010) for the year 2008, as provided
methods must add up to unity for each type of waste, at the scale regarding 14 distinct economic activities considering the NACE
of each industry. That is, for each industry j 2 f1; . . . ; ng and each nomenclature revision 2;
waste fraction p 2 f1; . . . ; f g: – secondly, statistics of waste generation from commercial activ-
ities are taken from (INSEE, 2008) for the year 2006, as an
X
t approximation for the year 2008;
Sððp  1Þ
t þ m; jÞ ¼ 1 ð18Þ – finally, Eurostat statistics are used regarding all other economic
m¼1
activities. In nine activity-cases, the correspondence between
Matrix D, of dimensions f waste fractions
t treatment meth- the nomenclature of activities used in Eurostat waste statistics
ods (in line) and f waste fractions (in column) is finally chosen and the NACE nomenclature makes it possible to calculate the
for calculation purposes in order to allocate the waste fractions corresponding waste multipliers directly. In all the other cases
to their potential treatments (e.g. plastic to recycling of plastic, of activities, data are provided in a format which is too aggre-
incineration of plastic or landfilling of plastic). gated for their direct implementation in this study. These raw
0 1 data of waste generation are accordingly disaggregated
It
1 0   0 between the producing activities. The key for the disaggregation
B .. ..C
B 0 . .C is set to be the intermediary consumptions of products poten-
B C
B .. C
.. tially generating the corresponding waste fraction. For example,
D¼B
B . It
1 C
.
C ð19Þ considering the case of textile waste generation from a given
B C
B .. .. C activity that should be disaggregated into sub-activities, the
@ . . 0 A
total amount of textile waste is allocated to each of these sub-
0   0 It
1 activities as a function of their intermediary consumptions of
‘‘Textiles, wearing apparel and leather products”.

2.2. Construction of Waste Input–Output Tables

2.2.2. Waste management in French Input–Output Tables
2.2.1. French Input–Output Table extended to waste Table S of allocation of wastes to treatments, both specific to
At present, the European supply and use tables are square tables waste fractions and to activities generating the waste, is con-
with the same number of products and industries. The classifica- structed in the specific case of France for the year 2008 based on
tion used for industries is the ‘‘General Industrial Classification of statistical data. Four activities of waste treatment are distinguished
Economic Activities within the European Communities” (NACE) in the study: valorization (including material and organic recy-
and the classification used for products is the ‘‘Classification of cling), incineration respectively with and without energy recovery,
Products by Activity” (CPA; Eurostat, 2008). These classifications and landfilling. Thirteen waste fractions out of the 14 focused on in
are fully aligned with each other. The monetary product-by- the extension to wastes described above are allocated to treat-
product matrices of domestic and imported input coefficients ments through Table S (batteries and accumulators are the only
(respectively Ad and Am ) relative to France in the year 2008 are waste fraction to be excluded, in the absence of any available
drawn from Eurostat (2014b) in a 64
64 framework according activity-specific data regarding their rates of treatment). Four
to the CPA nomenclature (2008 version). Similarly the monetary sources of statistical data on the treatment of waste generated by
vectors describing the household consumption of domestic French economic activities are used to compile Table S:
 A. Beylot et al. / Waste Management 49 (2016) 505–515 509

Table 1
Wastes from economic activities induced by French household consumption in 2008, taking into consideration two calculation approaches and 14 waste fractions.

Waste fractions Approach 1: hypothesis of production of imports Sensitivity analysis: Differences in masses of wastes between the two
identical to domestic production differentiation of imports in a approaches of calculations (Approach 1 vs
multiregional approach sensitivity analysis) (%)
Mass (in Proportion Contribution to the Mass (in ktonnes wet wastes)
ktonnes wet produced total generation of
wastes) abroad (%) waste (%)
Metallic wastes 6781 41 10 6424 5
Glass wastes 328 38 1 442 35
Paper and 3759 29 6 3853 2
cardboard
wastes
Rubber wastes 222 33 0 224 1
Plastic wastes 733 42 1 806 10
Wood wastes 4651 53 7 3986 14
Textile wastes 210 25 0 369 76
Equipment 1210 27 2 1143 5
Discarded vehicles 751 28 1 675 10
Batteries and 158 33 0 139 12
accumulators
wastes
Mineral wastesa 37,664 33 57 189,569 403
Household and 2294 18 4 2912 27
similar wastes
Mixed and 4721 33 7 4122 13
undifferentiated
materials
Animal and mixed 2683 24 4 4972 85
food waste;
vegetal wastes
Total 66,165 34 100 219,636 232
a
Except combustion wastes, contaminated soils and polluted dredging spoils.

– firstly, statistics for treatment of wastes from industrial activi- distinguished from coefficients relative to domestic production
ties are taken from (INSEE, 2010) for the year 2008; (Eq. (16)). Five different countries supplying products to France
– secondly, statistics for treatment of wastes from commercial are considered in this analysis: Germany, Belgium, Italy, the United
activities are taken from (INSEE, 2008) for the year 2006, as Kingdom and the Netherlands, all together representing 48.1% of
an approximation for 2008; the total imports to France (as calculated from the World Input–
– thirdly, statistics for residual Municipal Solid Waste manage- Output Database, WIOD; Timmer, 2012). Considering the remain-
ment in France in 2007 (ADEME, 2009) are used here as an ing share of the imports, coefficients relative to the EU27 are
approximation for the management of household and similar implemented as a proxy. The respective column vectors /i stand-
wastes in the year 2008. These data are global and therefore ing for the share of imports from each country i over the total of
not specific to a given economic activity producing wastes; French imports are also calculated from the WIOD. The monetary
– finally global waste statistics relative to the orientation of matrices of input coefficients relative to these five countries and
wastes from sorting centers toward recycling, incineration with the EU27 are drawn from Eurostat (2014b) in a 64
64 framework
or without recovery of energy, and landfilling for the year 2010 according to the CPA-2008. These tables correspond to the year
are used as an approximation for 2008 (ADEME, 2012). 2008, except regarding Belgium, Italy and the United Kingdom,
for which data for the year 2010 are used as a proxy. Their respec-
Data on waste management are unavailable so far regarding tive matrices of multipliers of waste generation wc are calculated
wastes from primary and service activities. In these cases, the aver- using a two-step process. A matrix of waste multipliers wc18 is first
age treatment rates observed in industrial activities with respect to calculated by direct use of the Eurostat data of waste generation,
each waste fraction are used as a proxy. considering 14 waste fractions and 18 economic activities (wd18
The use of data for 2006 or 2010 as an approximation for 2008 standing for the corresponding matrix relative to France). In a sec-
regarding waste generation and treatment is expected to involve ond step, this matrix is disaggregated into 64 economic activities,
relatively little uncertainty, and accordingly to cause only minor assuming that the disaggregation of waste multipliers for French
alteration in the interpretation of the results. On the contrary, import suppliers is identical to that of the French table. Consider-
the proxies implemented to compensate for the excessively aggre- ing c, a single country from among the different suppliers studied,
gated data of waste generation and treatment are expected to i a given waste fraction, J one of the 18 economic activities of
induce greater uncertainty. The assessment of input data uncer- matrix wc18 , and j one of the 64 economic activities of matrix wc
tainties and of their propagation to the output results lies outside aggregated within activity J:
the scope of this study, but should be retained as an important field
for future research on the issue.
wc18 ði; JÞ
wc ði; jÞ ¼
wd ði; jÞ: ð20Þ
wd18 ði; JÞ
2.2.3. Input–Output Tables extended to wastes for French imports
suppliers (for sensitivity analysis) The use of Input–Output Tables extended to wastes for French
In a second approach, referred to in the following as a sensitiv- import suppliers makes it possible to account more accurately
ity analysis, the coefficients of intermediary consumptions and for wastes from imports than the approach assuming production
waste generation relative to production of imports (abroad) are abroad to be identical to domestic production. However, the tables
510 A. Beylot et al. / Waste Management 49 (2016) 505–515

of waste multipliers for French import suppliers are compiled together 44% of the total. Likewise these four product categories,
based on data available at a lower degree of disaggregation than in addition to ‘‘Motor vehicles, trailers and semi-trailers”, display
the data used for compiling the French tables, implying their lower the largest contributions to the generation of dry recyclables
precision. Further research should be performed to quantify prop- (49% of the total). Moreover, regarding the generation of mineral
erly the uncertainty associated with each calculation approach. In and organic wastes, the first two product categories (respectively
this context, and in the absence of data at a sufficient level of dis- ‘‘Constructions and construction works” and ‘‘Coke and refined
aggregation for the allocation of wastes to treatments regarding petroleum products” on the one hand, and ‘‘Food products, bever-
French import suppliers, the implementation of the multiregional ages and tobacco products” and ‘‘Accommodation and food ser-
approach is considered only as a sensitivity analysis in this study. vices” on the other hand) contribute respectively 42% and 64% to
the total wastes induced by French household consumption.
Finally, all these product categories observed as a key with respect
3. Results and discussion
to the different aggregated waste categories are accordingly pre-
dominant with respect to the total generation of wastes induced
3.1. Wastes induced by French household consumption
by French household consumption. In particular, the category
‘‘Constructions and construction works”, which contributes sub-
Wastes from economic activities are reallocated to the product
stantially to mineral waste generation, also accounts for the largest
categories of household consumption that induce their generation,
contribution to the total wastes embodied in French household
using Eqs. (6) and (12). In a first section, the total amounts of waste
consumption.
are described fraction by fraction. Then, the ‘‘priority products”, i.e.,
Wastes induced by the household demand for a given product
those products that induce the largest contributions to waste gen-
category are generated either by this product’s own economic
eration, are ranked and discussed. Finally, the drivers for these lar-
activity (i.e., by the activity supplying this product) or by other eco-
gest contributions are depicted.
nomic activities (indirectly involved along the supply chain; Fig. 1).
Several product categories responsible for the most important
3.1.1. Total results according to waste fraction waste generation embodied in French household consumption pri-
Considering the 14 waste fractions under study, and regarding marily cause wastes from their own economic activity. In particu-
the year 2008, French household consumption induces the genera- lar, the demand for ‘‘Constructions and construction works”
tion of 66.2 million tonnes of wet wastes from economic activities primarily induces mineral wastes from the corresponding activity
(Table 1). Mineral wastes represent the largest share of this total ‘‘Constructions and construction works” (accounting for 99% of
amount (57%). Of lesser importance, mixed wastes (including the total; Table 2). Similarly, the demands for ‘‘Food products, bev-
‘‘household and similar wastes” and ‘‘mixed and undifferentiated erages and tobacco products” and ‘‘Accommodation and food ser-
materials”) and metallic wastes both represent 10% of the total, fol- vices” primarily induce organic wastes from their own economic
lowed by wood wastes (7%) and paper and cardboards (6%). activities (respectively 67% and 80%; Table 2). On the contrary, sev-
Although all these wastes are generated to supply the consumption eral other product categories are identified as mainly inducing
of French households, approximately 1/3 is produced abroad, while wastes along the supply chain, in other activities. In particular,
the remainder is produced in France. Indeed, a share of the French only 1% of the total mineral wastes induced by the demand for
household consumption in the first place directly requires imports ‘‘Coke and refined petroleum products” are generated in the eco-
(13% at a global scale, with discrepancies from one product to nomic activity ‘‘Coke and refined petroleum products” (26% regard-
another). By definition, these imports need to be produced – and ing dry recyclables from ‘‘Food products, beverages and tobacco
accordingly imply the generation of wastes – abroad. Secondly products”).
the consumption of domestic products by French households addi- The generation of wastes from economic activities induced by
tionally induces a demand for imports as intermediary consump- French household consumption can be further analyzed in line
tions of economic activities, which subsequently generate wastes with the COICOP nomenclature (Table 3). ‘‘Food and non-
abroad (as accounted for through Eq. (12)). The location of waste alcoholic beverages”, ‘‘Housing, water, electricity, gas and other
generation (either in France or abroad) is relatively similar from fuels”, ‘‘Transport” and ‘‘Recreation and culture” present the largest
one fraction to another, still ranging from 47% generated in France contributions to the generation of mixed wastes and dry recy-
regarding wood wastes up to 82% regarding household and similar clables (all together contributing to 68% of the total). A single cat-
wastes. egory of consumption in the case of mineral wastes (‘‘Housing,
water, electricity, gas and other fuels”) and two categories in the
3.1.2. Contribution analysis case of organic wastes (‘‘Food and non-alcoholic beverages” and
The contributions of the 64 distinct product categories of French ‘‘Restaurants and hotels”) are responsible respectively for over
household consumption (in CPA nomenclature) to the latter’s 58% and 70% of the total wastes embodied in French household
induced generation of wastes can be ranked as a function of waste consumption.
fractions. Five waste categories, aggregating the 14 waste fractions
under study, are considered in the following: dry recyclable wastes 3.1.3. Waste intensity and volume of expenses: the drivers for waste
(encompassing metallic, glass, paper and cardboard, rubber, plastic generation
and textile wastes); mixed wastes; mineral wastes; animal and The amounts of wastes induced by each product category are
mixed food wastes, and vegetal wastes (hereinafter, so-called ‘‘or- driven by two main characteristics of household consumption. In
ganic wastes”); and total wastes. Considering these five waste cat- the first place, the wastes embodied are a function of the ‘‘waste
egories, the ten product categories accounting for the highest intensity” specific to each product category, expressed in ktonnes
contributions represent from 64% to 86% of the total generation waste/Meuro of final demand. On the one hand, a first group of
of wastes induced by French household consumption (Fig. 2). The products includes those categories with large waste intensities
consumptions of ‘‘Retail trade services, except of motor vehicles and, subsequently, large contributions to the total waste embodied
and motorcycles”, ‘‘Food products, beverages and tobacco prod- in household consumption. For example, as to mixed wastes (char-
ucts”, ‘‘Accommodation and food services” and ‘‘Wholesale trade acterized by relatively high rates of disposal compared to source-
services, except of motor vehicles and motorcycles” are the largest segregated waste fractions and set as one of the waste categories
contributors to the generation of mixed wastes, representing all targeted by French policies on waste prevention and reduction of
 A. Beylot et al. / Waste Management 49 (2016) 505–515 511

Fig. 2. Ten product categories (CPA-2008 nomenclature) that are the highest contributors to the generation of wastes from economic activities induced by French household
consumption, considering four aggregated indicators of waste generation.

Table 2
Wastes from economic activities induced by the demand for seven products: share generated by the own producing activity, considering four aggregated indicators of waste
generation.

Retail trade Motor Food products and Accommodation and Wholesale trade Constructions and Coke and refined
services (%) vehicles beverages (%) food services (%) services (%) construction works (%) petroleum products (%)
(%)
Dry recyclables 79 51 26 54 46 42 1
Organic wastes 26 0 67 80 23 78 0
Mixed wastes 87 20 47 48 52 89 2
Mineral wastes 4 4 2 2 2 99 1

waste landfilling), ‘‘Retail trade services” and ‘‘Constructions and waste generation (3.2 ktonnes/Meuro) and, at the same time, a rel-
construction works” both display a relatively high intensity in atively low contribution to the total waste embodied in French
waste generation (respectively 20.1 ktonnes/Meuro and 33.7 kton- household consumption (<0.1%).
nes/Meuro) and, at the same time, a relatively large contribution to Secondly the wastes induced by each product category of
the total waste embodied in French household consumption household consumption are correlated with the volume of
(respectively 16.9% and 5.7%; Fig. 3). On the other hand, a second expenses associated with the corresponding product category.
group of products includes those product categories with low Consequently, a third product-group includes those product cate-
intensities in waste generation and, subsequently, low contribu- gories with relatively low intensities in waste generation but, at
tions to the total waste embodied in household consumption. For the same time, accounting for relatively large contributions to
example, still considering mixed wastes, ‘‘Products of forestry, log- the total wastes embodied in household consumption. For example
ging and related services” display a relatively low intensity in ‘‘Real estate services” contribute to 4.9% of the total amount of
512 A. Beylot et al. / Waste Management 49 (2016) 505–515

Table 3
Contribution of five product categories (COICOP nomenclature) to the generation of wastes from economic activities induced by French household consumption, considering five
aggregated categories of wastes.

Food and non- Housing, water, electricity, Transport Recreation and Restaurants Contribution of these five product
alcoholic beverages (%) gas and other fuels (%) (%) culture (%) and hotels (%) categories to the total (%)
Dry recyclables 22 16 19 11 8 76
Organic wastes 40 13 4 4 30 91
Mixed wastes 26 19 12 11 8 75
Mineral wastes 9 58 12 6 3 87
Total wastes from 16 41 13 7 5 83
economic activities

Fig. 3. Contribution to the total generation of mixed wastes (in % of the total) versus intensity in mixed waste generation, considering the 64 product categories of French
household consumption.

mixed wastes embodied in French household consumption, need for waste treatment as a response to French household con-
whereas their corresponding intensity in mixed waste generation sumption account, respectively, for 70% (in case of valorization)
‘‘only” amounts to 1.8 ktonnes/Meuro. Conversely a final group of and 74% (in case of disposal) of the total (Fig. 4). Those activities
products includes those product categories with relatively large with relatively large contributions to the total generation of wastes
intensities in waste generation, but accounting for a minor share also induce the largest contributions to waste treatment require-
of the total wastes embodied in household consumption. This is ments. Only four product categories in particular (‘‘Constructions
the case for ‘‘Printing and recording services”, of intensity in mixed and construction works”, ‘‘Food products, beverages and tobacco
waste generation amounting to 14.2 ktonnes/Meuro, but resulting products”, ‘‘Coke and refined petroleum products” and ‘‘Real estate
in no contribution to the total amount of mixed wastes embodied services”) are responsible for 43% of the total amount of waste
in French household consumption. embodied in household consumption, while at the same time
accounting for 48% of the total amount of waste to be disposed
3.2. Waste treatment induced by French household consumption of. Several discrepancies, however, can be observed between the
rankings of product categories respectively versus waste genera-
Similarly, the consumption approach adopted to characterize tion and versus waste to treatments. For example, ‘‘Motor vehicles,
the amounts of waste embodied in household consumption can trailers and semi-trailers” are ranked in seventh place with respect
be applied to the amounts of waste according to treatments. Still to their contribution to the total wastes embodied in household
considering the 14 waste fractions under study, and regarding consumption (5.3%), in fifth place regarding their contribution to
the year 2008, the consumption of French households induces the recycling needs (5.8%), but in ninth place considering their contri-
incineration without energy recovery and the landfilling (i.e., the bution to the needs for disposal (3.9%). Such differences are the
disposal, at the bottom of the ‘‘waste management hierarchy” as result of differing rates of treatment (both activity- and waste-
defined by the EU Waste Framework Directive) of 17.6 million specific) as accounted for through Table S in Eqs. (17) and (18).
tonnes of wet wastes from economic activities. The recycling and
valorization of 44.3 million tonnes of wet wastes is at the same 3.3. Sensitivity analysis to the hypothesis on imports
time required. Extrapolating from the content in metals of mixed
residual Municipal Solid Waste (ADEME, 2010), French household When considering the differentiation of imports in a multire-
consumption accordingly induces the incineration of 224 ktonnes gional approach as a sensitivity analysis, the consumption of
of metals and metal-containing products (including metallic French households is seen to induce the generation of 219.6 mil-
wastes, discarded equipment and vehicles, and metals from lion tonnes of wet wastes from economic activities, i.e., 3.3 times
MSW). Similarly, extrapolating from the composition of residual greater than when the hypothesis of identical production between
MSW in metals, papers, cardboards and organic matters, 2.7 mil- imports and domestic products is considered (Table 1). Such a dis-
lion tonnes of metal, paper, cardboard and organic containing crepancy is mainly explained by the relatively large difference in
products are induced to be landfilled. embodied mineral wastes as calculated from one approach to the
The 64 distinct product categories of French household con- other. Mineral wastes do indeed represent the largest share of total
sumption (in CPA nomenclature) can be ranked as a function of wastes embodied in French household consumption (57% of the
their contribution to the latter’s induced quantities of wastes to total in the initial calculation, 86% in this sensitivity analysis),
treatments. The ten product categories contributing most to the while the amount of mineral wastes embodied in household
 A. Beylot et al. / Waste Management 49 (2016) 505–515 513

Fig. 4. Ten product categories (CPA-2008 nomenclature) that are the highest contributors to waste treatment requirements induced by French household consumption.

consumption is calculated to be 5.0 times higher in this sensitivity differences in rates of organic waste source segregation and the
analysis. The difference between the amounts of mineral wastes distinct efficiencies of production versus waste generation.
embodied in consumption as calculated using each approach is pri- On the contrary considering the 11 other waste fractions
marily due to the larger multipliers of mineral waste generation embodied in French household consumption, the differences
relative to the ‘‘Mining and quarrying activities” in foreign coun- observed are not as great, in the range [1–35%] (in absolute value;
tries as compared to the French multiplier. This multiplier in par- Table 1). Considering the aggregated amounts of dry recyclables
ticular amounts to 3423 tonnes mineral wastes/Meuro of product and mixed wastes, the differences observed from one fraction to
from ‘‘Mining and quarrying activities” on average in the EU27, another even compensate for each other when summed up. This
1874 tonnes/Meuro in Germany and 1689 tonnes/Meuro in the means that a difference comprised between [0–1%] is observed
United Kingdom, to be placed in perspective with the French mul- when comparing the amounts of dry recyclables and mixed wastes
tiplier amounting to only 129 tonnes/Meuro. The specific nature of embodied in household consumption as calculated from the two
‘‘Mining and quarrying activities” in France, with limited extrac- different approaches. Similarly the differentiation of imports in a
tion of metals and fossil fuels compared to other EU27 countries multiregional approach in this sensitivity analysis reveals only
such as Great-Britain, Germany or Poland (Eurostat, 2014a), is to minor effects regarding the identified ‘‘priority products” as to
be seen one of the main reasons for such a difference between their waste generation and treatment. In particular, it will be noted that,
corresponding multipliers of mineral waste generation. The subse- of the ten product categories contributing the most to waste gen-
quent larger amounts of mineral wastes calculated as being gener- eration (in terms of dry recyclables, organic wastes, mixed wastes
ated by ‘‘Mining and quarrying activities” imply 98% of the increase and mineral wastes), nine to ten – depending on the waste cate-
in the amount of mineral wastes embodied in French household gory – are similar from one calculation approach to the other,
consumption observed when performing this sensitivity analysis. despite slight changes in their ranking.
Conversely, the differences in input coefficients of intermediary
consumptions from one country to another, i.e., differences in pro- 3.4. Implications for waste policies
duction structures, show only a small contribution to the gap in
mineral waste generation. These results, depicting the links between French household
Similarly, relatively large differences are observed regarding the consumption on the one hand, and waste generation from eco-
amounts of organic wastes and textile wastes generated (respec- nomic activities and their management on the other, may be fur-
tively 85% and 76%). Considering organic wastes, the difference ther employed in a general context of Integrated Product Policies
between the amounts of waste embodied in French household con- planning. Indeed, this study first delineates the product categories
sumption as calculated using each approach is almost entirely due contributing most to waste generation and waste treatments
to the larger multipliers of organic waste generation associated embodied in French household consumption. A few products are
with the foreign activities of ‘‘Products of agriculture, hunting shown to be responsible for the largest share of the waste genera-
and related services” and ‘‘Food products, beverages and tobacco tion and waste to treatments (e.g., two product categories induce
products”. These multipliers in particular amount to 54.6 tonnes 64% of organic wastes generation, ten product categories induce
organic wastes/Meuro of ‘‘Products of agriculture, hunting and 74% of waste to disposal, etc.). This hierarchy of contributions
related services” and to 36.0 tonnes organic wastes/Meuro of ‘‘Food may be considered in the framework of the orientation of policies
products, beverages and tobacco products” as an average in the on consumption and production from economic activities in the
EU27, to be placed in perspective with the French multipliers, sense of the waste hierarchy, i.e. favoring waste prevention and
amounting respectively to only 7.2 and 5.2 tonnes/Meuro. Such the reduction of wastes toward treatments at the bottom of the
differences in organic waste multipliers are of several origins, waste hierarchy. Firstly, in a context of constant or increasing total
including the distinct production mixes from one country to volume of expenses by French households, policies on household
another (expected to be the main driver, as in the case of mineral consumption may include extending product eco-labeling to the
waste generation from ‘‘Mining and quarrying activities”), the ‘‘waste footprint”. The latter may for example be introduced as a
514 A. Beylot et al. / Waste Management 49 (2016) 505–515

complement to the product environmental footprint developed in Moreover, the Waste Input–Output Tables constructed, specific to
Europe in order to support the assessment and labeling of products France and to some of its suppliers, can be reused in other studies,
(European Commission, 2013) and, for the sake of efficiency, could regarding, for example, the amount of waste generated domesti-
be restricted to those product categories identified through this cally in order to satisfy external final demands or regarding the
study as the largest contributors in terms of waste generation influence of changes in consumption patterns on waste generation
and disposal (e.g. ‘‘Food products, beverages and tobacco prod- and treatment.
ucts”, ‘‘Motor vehicles, trailers and semi-trailers” or ‘‘Textiles, However, this study also highlights the fact that further efforts
wearing apparel and leather products”). The waste footprint of are required to improve the reliability of Input–Output Analyses
products, further calculated using process-based Life Cycle Assess- applied to waste generation and treatment. We first need to bear
ment, would help consumers in their decision-making, potentially in mind that this study focused solely on wastes from economic
contributing to the diversion of household consumption toward activities induced by French household consumption. A major per-
those product categories with the lowest intensities in waste gen- spective for improvements will be to simultaneously account for
eration. Moreover, the identified ‘‘priority-products” may be tar- the amounts of wastes originating directly from the consumption
geted when defining policies relative to production in the most of households as post-consumer wastes. Moreover, another major
efficient way as to global waste reduction. On the one hand, area for improvements regards the construction of multiregional
regarding the products that mainly induce wastes from the eco- Input–Output Tables extended to waste generation and treat-
nomic activities involved in producing them (e.g. organic wastes ments. These multiregional Waste Input–Output Tables would
from ‘‘Accommodation and food services”), decision-makers enable more precise account to be taken of imports than what
should favor measures of waste prevention targeting these specific has been achieved in the scope of the present study, limited to a
producing economic activities. On the other hand, eco-design multiregional unilateral approach including only a few foreign sup-
(including mass reduction or materials substitution, for example) pliers. At present, technology matrices of input coefficients are
should be encouraged regarding the products that mainly give rise often available for multi-regions in a disaggregated framework
to wastes along their supply-chain (e.g. ‘‘Food products, beverages (e.g. within WIOD). However, in the case of most countries, the
and tobacco products” as to dry recyclables). statistics of waste generation from economic activities and treat-
Furthermore, this study highlights the relative importance of ment, necessary for the calculation of waste multipliers, are not
wastes produced abroad as compared to those produced in France, available at a sufficiently disaggregated level, whether it be in
regarding the total amount of wastes induced by French household terms of economic activities, products or wastes. In the few cases
consumption. These results may prompt reconsideration of target of countries for which a Waste Input–Output Table is available
set concerning waste generation and treatment. Up to now, these (e.g. regarding Japan; Nakamura and Kondo, 2009), the differences
targets, either through the EU Waste Directive or through specific in terminology of economic activities and wastes as compared to
French legislations, have focused solely on waste produced domes- the French table have prevented their use in this study. A common
tically. These accordingly do not account for the potential delocal- framework for compiling waste statistics from economic activities
ization of waste generation and treatment abroad, through the on a global scale, e.g., similar to what is already available on the
increase of importations of products with more unfavorable waste scale of the European Union but considering a more disaggregated
generation intensities. Consequently, a non-quantified risk exists level regarding waste-generating economic activities, would sig-
today that the reduction in waste generation and in waste quanti- nificantly improve the reliability of such Input–Output Analyses.
ties routed toward undesired treatments in France might simulta- The direct inputs waste estimation methodology to create a
neously result in increased generation of wastes abroad. This detailed estimate of Municipal and industrial Solid Waste for an
means that objectives set by European or French legislations might economy (including sectoral and waste-type disaggregation), and
well be attained, while, at the same time, opposing trends (e.g. the its potential applications for Waste Input–Output Analysis, also
increase in the amounts of waste generated) would be observed on appears promising to compensate for the gaps in waste statistics
a global scale. A similar issue has already been the subject of inten- (Reynolds et al., 2014, 2015).
sive discussion regarding greenhouse gas emissions accounted for Attention should be called to the fact that the assumption made
in a ‘‘consumer perspective”, leading to question the ‘‘production regarding imports assumed to be produced identically to domestic
approach” considered in the Kyoto protocol (Davis and Caldeira, products has long been integrated into Input–Output Analyses
2010; Peters et al., 2011; Helm, 2012; Boitier, 2012). Similarly, focusing on energy or emissions of pollutants (see e.g. Palm
our study calls for opening the discussion regarding the potential et al., 2006; Kondo et al., 1998). After significant research efforts,
accounting, within waste policy targets, of foreign wastes embod- multiregional approaches have now emerged as the standard in
ied in domestic consumption. This may further lead to taking issue this kind of study (e.g., considering greenhouse gas emissions
with the structure of trade balances, not only in terms of monetary embodied in French household consumption; Lenglart et al.,
or carbon flows, but also in those of waste generation and treat- 2010). One can therefore be optimistic as to the possibility of
ment in a global perspective. improving in future years the reliability of Waste Input–Output
Analyses in a similar manner thanks to multiregional databases.

4. Conclusions and perspectives Acknowledgements

This study has therefore allowed us to assess waste generation This study was partly funded by the French Environment and
and treatment embodied in French household consumption. Firstly Energy Management Agency (ADEME) in the framework of the
those products which induce the largest amounts of waste gener- Reactivity project. The authors wish to thank Laurent Meunier
ation in economic activities, both in France and abroad, were iden- (ADEME), Elisabeth Poncelet (ADEME) and Alexandre Godzinski
tified. Moreover, the products inducing the largest amounts of (CGDD) for their contributions to this work.
waste landfilling and incineration without energy recovery, i.e.,
those treatments at the bottom of the ‘‘waste management hierar-
References
chy”, were identified. These results may further be of use in a gen-
eral context of Integrated Product Policies planning, and provide ADEME, 2009. La collecte des déchets par le service public en France. Résultats
the basis for a discussion regarding current waste policy targets. année 2007, Juin 2009 (in French).
 A. Beylot et al. / Waste Management 49 (2016) 505–515 515

ADEME, 2010. La composition des ordures ménagères et assimilées en France. Helm, D., 2012. Climate policy: the Kyoto approach has failed. Nature 491
Campagne nationale de caractérisation 2007. ADEME Editions (in French). (November), 663–665. http://dx.doi.org/10.1038/491663a.
ADEME, 2012. Etat des lieux des centres de tri de Déchets Non Dangereux des INSEE, 2008. Les déchets des grands établissements commerciaux en 2006. No 1200
Activités Economiques en France – Données 2010, Février 2012 (in French). – Juillet 2008 (in French).
ADEME, 2014. Alléger l’empreinte environnementale de la consommation des INSEE, 2010. La production de déchets non dangereux dans l’industrie en 2008. Mise
Français en 2030. Vers une évolution profonde des modes de production et de à jour Mai 2010. <http://www.insee.fr/fr/themes/detail.asp?reg_id=99&ref_id=
consommation, Novembre 2014. Réf. 8356, 8357 (in French). dechets> (in French).
Assemblée Nationale, 2014. Projet de Loi relatif à la transition énergétique pour la Joshi, S., 1999. Product environmental life-cycle assessment using input–output
croissance verte, adopté par l’Assemblée Nationale en première lecture. <http:// techniques. J. Ind. Ecol. 3 (2–3), 95–120.
www.assemblee-nationale.fr/14/ta/ta0412.asp> (in French). JRC, 2006. EIPRO Environmental Impact of Products, Analysis of the life Cycle
Barata, E.J.G., 2002. Solid waste generation and management in Portugal: an Environmental Impacts Related to the Final Consumption of the EU 25, 2006,
environmental input–output modelling approach. In: Paper for the 7th Biennial JRC.
Conference of the International Society for Ecological Economics, ‘‘Environment Kondo, Y., Moriguchi, Y., Schimizu, H., 1998. CO2 emissions in Japan: influences of
and Development: Globalisation & the Challenges for Local & International imports and exports. Appl. Energy 59, 163–174.
Governance”, Sousse (Tunisia), 6–9 March 2002. Lave, L.B., Cobas-Flores, E., Hendrickson, C.T., McMichael, F.C., 1995. Using input–
Beylot, A., Vaxelaire, S., Villeneuve, J., 2015. Reducing gaseous emissions and output analysis to estimate economy-wide discharges. Environ. Sci. Technol. 29
resource consumption embodied in french final demand: how much can waste (9), 420A–426A.
policies contribute? J. Ind. Ecol. http://dx.doi.org/10.1111/jiec.12318. Lenglart, F., Lesieur, C., Pasquier, J.-L., 2010. Les émissions de CO2 du circuit
BIO Intelligence Service, 2012. Analyse des impacts environnementaux de la économique en France. L’économie française, édition 2010. <http://www.insee.
consommation des ménages et des marges de manœuvre pour réduire ces fr/fr/ffc/docs_ffc/ref/ecofra10e.PDF> (in French).
impacts. Rapport de l’étude « Outils économiques d’incitation à la Leontief, W., 1970. Environmental repercussions and the economic structure: an
consommation durable: étude de positionnement (environnementale, input–output approach. Rev. Econ. Stat. 52, 262–277.
économique, sociologique) et élaboration de propositions, Janvier 2012. Étude Liao, M., Chen, P., Ma, H., Nakamura, S., 2015. Identification of the driving force of
réalisée pour le compte de l’ADEME, Service Économie et Prospective (in waste generation using a high-resolution waste input–output table. J. Cleaner
French). Prod. 94 (2015), 294–303.
Boitier, B., 2012. CO2 Emissions Production-based Accounting Vs. Consumption: Munksgaard, J., Pedersen, K.A., 2001. CO2 accounts for open economies: producer or
Insights from the WIOD Database. ERASME, Ecole Centrale, Paris. consumer responsibility? Energy Policy 29 (4), 327–334.
Cadarso, M., López, L., Gómez, N., Tobarra, M., 2012. International trade and shared Nakamura, S., Kondo, Y., 2002. Input–output analysis of waste management. J. Ind.
environmental responsibility by sector. An application to the Spanish economy. Ecol. 6 (1), 39–64.
Ecol. Econ. 83 (2012), 221–235. Nakamura, S., Kondo, Y., 2009. Waste Input–Output Analysis Concepts and
CAS, 2011. Pour une consommation durable. Rapports et Documents. 2011, no 33. Application to Industrial Ecology Series: Eco-Efficiency in Industry and
La documentation Française. Science, vol. 26, XVIII, 294p.
CGDD, 2011. Consommation des ménages et environnement. Édition 2011. Service Palm, V., Finnveden, G., Wadeskog, A., 2006. Swedish experience using
de l’observation et des statistiques. environmental accounts data for integrated product policy issues. J. Ind. Ecol.
Court, C.D., 2012. Enhancing U.S. hazardous waste accounting through economic 10 (3), 57–72.
modeling. Ecol. Econ. 83, 79–89. Peters, G.P., Minx, J.C., Weber, C.L., Edenhofer, O., 2011. Growth in emission
Court, C.D., Munday, M., Roberts, A., Turner, K., 2015. Can hazardous waste supply transfers via international trade from 1990 to 2008. Proc. Natl. Acad. Sci. 108,
chain ‘‘hotspots” be identified using an input–output framework? Eur. J. Oper. 8903–8908.
Res. 241 (1), 177–187. Reynolds, C., Geschke, A., Piantadosi, J., Boland, J., 2015. Estimating industrial solid
Davis, S.J., Caldeira, K., 2010. Consumption-based accounting of CO2 emissions. Proc. waste and municipal solid waste data at high resolution using economic
Natl. Acad. Sci. 107 (12), 5687–5692. accounts: an input–output approach with Australian case study. J. Mater. Cycles
European Commission, 2013. 2013/179/EU: Commission Recommendation of 9 Waste Manage. http://dx.doi.org/10.1007/s10163-015-0363-1.
April 2013 on the use of Common Methods to Measure and Communicate the Reynolds, C.J., Piantadosi, J., Boland, J., 2014. A waste supply-use analysis of
Life Cycle Environmental Performance of Products and Organisations. Australian waste flows. J. Econ. Struct. 3, 5. http://dx.doi.org/10.1186/s40008-
European Commission, 2014. IPP Toolbox. <http://ec.europa.eu/environment/ 014-0005-0.
ipp/toolbox.htm> (Last updated 30/10/2014). Takase, K., Kondo, Y., Washizu, A., 2005. An analysis of sustainable consumption by
Eurostat, 2011. Manual on waste statistics. A handbook for data collection on waste the waste input–output model. J. Ind. Ecol. 9 (1–2), 201–219.
generation and treatment. Eurostat Methodologies and Working Papers, 2010 Timmer, M.P., 2012. The World Input–Output Database (WIOD): Contents, Sources
edition. Publications Office of the European Union, Luxembourg. and Methods. WIOD Working Paper Number 10, downloadable at <http://www.
Eurostat, 2008. Eurostat Manual of Supply, Use and Input–Output Tables, 2008 wiod.org/publications/papers/wiod10.pdf>.
edition. Office for Official Publications of the European Communities, Tukker, A., Huppes, G., Suh, S., de Koning, A., Geerken, T., van Holderbeke, M.,
Luxembourg. Nielsen, P., 2004. Evaluation of the environmental impact of products. Working
Eurostat, 2014a. Mining and Quarrying Statistics – NACE Rev.2. Available at: Paper Covering Steps 1 and 2 of the EIPRO Project. Institute for Prospective
<http://ec.europa.eu/eurostat/statistics-explained/> (Last visit: August 2015). Technological Studies, Seville.
Eurostat, 2014b. Statistics by Theme. Available at: <http://epp.eurostat.ec.europa. Weidema, B.P., Suh, S., Notten, P., 2006. Setting priorities within product-oriented
eu/portal/page/portal/statistics/themes> (Last visit: December 2014). environmental policy. J. Ind. Ecol. 10 (3), 73–87.